BORKf\lM2285

M02
11111'lli11 ]lllli11lrillllllllllll'" i~li{fl
 BORKNM2285
SECOf\JO EOITIO~ M02

Collect~; a
and
Interprt:.ting
. ualltatlve
ateri31s

ZJ\PADOCESKJ\
UNIVERZITA v Plzni
 SECOND EDITIOI\I
INTERNATIONAL ADVISORY BOARD

Arthur P. Bochner
COlllllllllliciitiOll,
Unil'crsil)' of South Florida
Patti Lather
EdllCdtioll,
OIJio Statl! Ulliuersit)' Collecting
Ivan Brady
Anthropology, Stale Ulliucrsi/),
of Nell/ Yor"- ill Oswego

juliJllne Cheek
!Vlorten Levin
Sociology and Pofitical Science,

or
Norwegian fJl/iuersity
Scicllcl! IIJld Tec/JIlology
and -
f-IC<I!tfJ,
VIIII/CrSII)' of South Australia

Patrici>l Clough
Sociology illld \1/011I211'$ Studies,
JVlcaghan l'vlorris
Hwnilllities ,md Social Sciellce,
University of Teclmo/of:,'J', Sydney
Intetpre,ting
-ualltltlve
Nlichael A. Olivas
the Graduate Celllcl;
Lmv,
Cit)' Ulliucrsity of New YOTh
University of HOI/stOll
Yen Le Espiritu
Ethnic Studies,
Shlllamit Reinharz
Ulliucrsil)' of Cali(ofllia,
Sail Diego

Orlando F<11s Bore\;]

Sociology,
N'Iliolla/ !Ju;ucrsil}' or Colombia
Soei%t,'}',
Bralldcis Ulliui'rsity

James J. Scheurich
tIl/cation,
Ulliuersity o(Texils at AI/still
1at ri;lls
I'vlichelle Fille David Silverman
Sociall'sydJOlog)', Sociology,
City Uniuersi!), of NeU' 'lori:. Goldsmith Coffegc, IJlliut!rsity 0/ Loudoll
Davydd Greenwood Linda Srnircich
Allthro/m!ogy, tVlJlIwgelllCII! IIlId Orgalli:;:"ltioll Studies, eclitol'S
Cornell Un/uersit)' Ulliuersil), 0/ Massacfl1lsctts~Alllherst

Jaber F. Gubrium
NORMAN K, DENZIN
Sociology,
Roben E. Stake University of Illinois at Ui'bana-Cilaillpaign
Edlleatin/I.
Unir'iTsity of Florida Ullil'crsity ullllil/ois, Urllilll<i-Chilll1!Jaigll
YVONNA S, LlNCOLl'J
ROS<lnll;l Hertz
Barbara Tcdlock
Texas A&M Uiliversity
Sociology ,wd \VOlllell'S Stl/dies,
Allthropology,
Wellesley Coflcgc
Stllte Ullil'ersily of Nelli Yorl:. ilt DII/falo
Ueln Kelle
[nstilllie (or IlItcrdisciplilhJr)' Gerontology. \X1iHiam G. Tierney
University of Vt!chta. Fell/GUion,
GcmWII)' Unil'ersity 01 SOllthem California

Allton Kuzel Harrv \Volcott

Fall/ily fdcdicilll!, EdllcatirJl/ a;ld Anthropology,
Virginia COll1ll11l1l1l'mfth Ulliwrsity Ulliucrsity of Orcgoll
Copyright (Q) 2003 br Sage Publications, Inc.

All rights reserved. No part of this book may be rf~rodllced or ur~lized in any
form
or by any means, electronic or mechanic:!l, incJucltng photocorYlJ?g~ reC?rdlll?,. or
by any information storage and retrieval system, without permiSSIon li1 wntmg Contents
from the publisher.

For information:

Sage f'ublicati01l5, Inc.

1455 Tellcr Road
ThouS:llld Oaks, California 91310
E-mail: ordcr@sagepub.colll
Sage Publications Ltd.
G Bonhill Street
London EC2A 4PU
United Kingdom
Preface
IX

Sage Publications India Pv('. LtJ.

!",)orman t\. Denzin and Yvonna S. Lincoln

B-A1 Panchsheel Encbve

POSt Box 4109
New Delhi 110-017 India 1. Introduction: The Discipline and Practice
of Qualitative Resarch
Prillted ill the Ullitcd Stdtes of America f'Jorman 1<' Denzln and Yvonnd S Uncoln
Library of Congress Cataloging-in-Publication Data

Collecting and imerprering qualitative materi:lb,/ Norrmm K. Dcnzin,

Part 1. Methods or Collecting
Yvonna S. Lincoln, editors.- 2nd ed. and Analyzing Empirical Materials 47
p. em.
Includes bibliographic<ll references and inde:.:.
ISBN 0-7619-2687-9 (Paper) 2. The Interview: From Structured Qllestions
1. Social 5ciences-Rescarch-tvlcthodology. 2. Qualit:Hive reasoning.
to Negotiated Text 61
1. Denzin, Norman K. II. Lincoln, Yvonna S.
H62 .C566 2003 /\ndrecl Fonlana and James H Frey
300'.7'23-clc21 20021566]3

3. Rethinking Observation: From Method to Context 107

Primed on acid-free paper
fV\ichdcl V Angroslno dnd Kimberly .P,. fv\d,/S de PCri:"Z
05 06 07 08 09 10 9 8 7 6 5 4 J 2

4. The Interpretation of Documents and Material Culture 155

ldn Hodder
Acquiring Edilor; lvbrg:m.:! H. S~3wtll
['mlilielioll Eililo,-: Cbudia A. Hoffm;m
Tl'peSl'ller: Chris!ill;\ Hill
/;,dcxcr: I\'lolly Hall KN070002425
Couer Di!SlgIIer: !vlichdk l~e Jnd Ibvi Ibb';uri)'il 5. Reirnagining Visual Methods: GaWeo to Neuromancer 176
Cnl'er Ph%gmph: C. A. !-loffm.m Douglas Harper
6. Autoethnography, Personal Narrative, Reflexivity: Suggested Readings
645
R,-,,,c<>.,,-,A~e< ,~" ~"1~;,,,,,.1 199
Carolyn Ellis dnd /\rthur P, Bochner Namt; Index
651

7. Data Management and Analysis Methods 259 Subject Index

665
Guy \Xl Pyan and l-I. Russell BCflldrd
About the Authors
675
8. Software and Qualitative Research 310
Eben /\. \X!cll:rnan

9. Analyzing Talk and Text 340

David Silverman

10. Focus Groups in Feminist Research 363

Esther {Vladflz

11. Applied Ethnography 389

Ervc Chdl1lbc,s

Part U. The Art and Practices of Interpretation,

Evaluation, and Representation 419

12. The Problem of Criteria in the Age of Relativism 427

John 1: SiT'lith and Debordh I::. Deemer

'j 3. The Practices and Politics of Interpretation 458

!'lo'iTIJn 1<. Dcnzin

14. \Xlriting: A Method of Inquiry 499

Laurel Richa,clson

15. Anthropological Poetics 542

IVdll Brach'

16. Understanding Social Programs Through Evaluation 590

Jcnnilcr C, (:rrccnc

17. Influencing the Policy Process \Vith Qualitative Rese3rch 619

C Pist
 Preface

For over than three decades, a quiet methodological rcvolurion has

been taking place in the social sciences. A blurring of disciplinary
boundaries has occurred. The social scil~nces and humanities have drawn
closer together in a I11UtL18! focus on an :,nrerprcrive, qualitative approach
to research and theory. Although these trends are !l0!' new, the extent to
which the "qualitative revolution" has l)vcrrakcn the social sciences and
relateel professional fields has been nothing short of amazing.
Reflecting this revolution, a host Jf textbooks, journals, research
monographs, and readers have been published in recem years. In 1994 ViC
published the first edition of the Halldboo/;:. o(Qllalitatiue l~ese'-lrch in an
attempt to represent the field in its emir,=ty, to rake srock of how far it had
come and how far it might yet go. The immediate success of the first edi-
tion suggested the need ro offer the Handbaa/::. in terms of three sep;:lI~Cltc
volumes. So in "1998 we published a three-volume SCt, TIJe Landscape 0/
Qllalilatiuc Research: Theories and Issues; Strategies a/Inquiry; and Co/-
lecting and Interpreting Q1talitatiue kIdterials. In 2003 we offer a ne\v
three-volume set, based on the second edition of the handbook.
In 2000 we published the second edition of the Handbook. Although it
became abundantly clear that the "fiel d" of qualitative research is still
defined primarily by tensions, contradictions, and hesitations-and thar
they exist in a less-than-unified arena--we believed that the handbook
could and would be valuable for solidifying, interpreting, and orgclllizing
the field in spite of the essential differences that characterize it.
The first edition attempted to define the field of qua!tiative research.
The second eclition went one step further. Building 011 theilles ill the first
 COllECTIHG AHD IHTERPRETII'IG QUALITATIVE MATERLb,LS
edition, vve asked how the practices of qU:.llitative inquiry could be used to
~JJress ;,,::;Ul.'S of cqu;ry and ot ~oci~l justice.
We have been enormously gratified and heartened by the response to
the I-hl1ldhoo!~ since its publication. Especia1!y grJtifying has been that it
has been used and adapted by such a \vicle variety of scholars and graduate
students in precisely the way we had hoped: as 2 starring point, a spring-
board for new thought and new work.
'" The Paperback Project
The second edition of the Landscape Series of the Halldbook ofQflalitatiue
Research is virtually all new. Over half of the authors from the first edition
have been replaced by new contributors. Indeed, there 3fe 33 new chapter
authors or co-authors. There are six totally new chapter topics, including
contributions on queer theory, performance cthnography, testimonio,
focus groups in feminist research, applied ethnography, and anthropologi-
cal poetics. All returning authors have subswotially revised their original
contributions, in many cases producing totally nev\' chapters.
The second edition of thc j-]alldbooh. OfQlwlilatiue Research continues
where the first edirion ended. \YJith Thomas Schw'lJ1dt (Chapter 7, Volume
1), we Illay observe that qualitative inquiry, among other things, is the
name for a "reformist movement that began in the early 1970s in the acad-
emy." The interpretive and critical paradigms, in their multiple forms, are
central to this movement. Indeed, Schwandt argues that this movement
encompasses multiple paf<ldigmatic formulations. It also includes com-
plex epistemological and ethical criticisms of traditional social science re-
search. The movement now has its OWI1 journals, scientific associations,
conferences, and faculty positions.
The transformations in the field of qualitative research that were raking
place in the early 1990s continued ro gain momentum as we entered the
ne\v century. Today, fevl' in the interpretive c01l1munity look back \vith
skepticism on the narrative turn. The turn has been taken, and that is all
there is to say about iLl'v'la.l1V have now told their tales from the field. Fur-
ther, today \~'e know that n~en and women write culture differenrly, and
that writing itself is not ;1n innocent practice.
Experimental ways of writing firsr-person ethnographic texts are now
commonplace. Sociologists and anrhropologists continue to explore new
Preface
ways of composing ethnography, and many write fiction, drama, perfor-
n1nllCe texts, and ethnographic poetry_ S(,cial science journals hold fiction
comests. Civic journalism shapes calls for a civic, or public, ethnography.
There is a pressing need ro shO\v how the practices of qualitative re-
search can heIp change the world in positive ways. So, at the beginning of
the twenty-first century, it is necessary rooe-engage the promise of qualita-
tive research as a generative form of inguTy (Peshkin, 1993) and as a form
of radical democratic practice. This is tht: agenda of the second edition of
the Landscape Series, as it is for the se':ond edition of the Handbook;
!lamely ro show how the discourses of qualitative research can be used to
help imagine and create a free, democratic society. Each of the chapters in
the three-volume set takes up this project, in one way or another.
A handbook, we were told by our publisher, should ideally represent
the distillation of knowledge of a field, a benchmark volume that synthe-
sizes an existing literature, helping to ddine and shape the presenr and
future of that discipline. This mandate organized the second edition. In
metaphoric terms, if you were co take one book on qualitative research
with you to a desert island (or for a comprehensive graduate examination),
a handbook would be the book.
\YJe decided that the part structure of the Handbook could serve as 3
lIseful point of departure for the organization of the paperbacks. Thus
Volume 1, titled The Landscape of QIU!litatiue Research: Theories alld
Issues, rakes a look at the field from a breadly theoretical perspective and
is composed of the Halldbook's Parts I ("Locating the Field"), II ("Para-
digms and Perspectives in 1i'ansition"), and VI ("The Future of Qualitative
Research"). Volume 2, titled Strategies ol Qlwlitatiue Inquiry, focuses
on just that and consists of Pan HI of the Handbook. Volume 3, tided Col-
lecting and hlterpreting Qualitatiue A1aferials, considers the tasks of col-
lecting, analyzing, <md interpreting empirical materials and comprises the
Handbook's Pans IV (,'lVlethods of Collecting and Analyzing Empirical
IV1aterials") and V ("The Art and Practices of Interpretation, Evaluation,
and Representation").
As with the first edition of the LandsC(!pe series, we decided that noth-
ing should be Cut from the original Handhook. Nearly everyone we spoke
ro \\'110 used the Handbook bad his or her own way of using it, leaning
heavily on certain chapters and skipping others altogether. Bur there was
consensus that this reorganization made a great deal of Sense both peda-
gogically 30d economically. We and Sage are committed to making this
 COLLECTlHG AHD IHTERPRETIHG QUALITATIVE MATERIALS Preface

iteration of the Handboo/(. accessible for classroom usc. This commitment
:, ,efLceed ;n tl,e "70, org.'lIliZ:1tioll, dnO nrice of [he paperbacks, as well as
in the addition of end-of-book bibliographies.
It also became clear in our conversations with colleagues who used the
Hmldbool:.. that the single-volume, harel-cover version has a distinct place
and value, and Sage will keep the original version ~1Vai1able until a revised
edition is published.
'" Organization of This Volume
Collecting Lind Illterpreting Qualitative J\ilateril1/s introduces rhe researcher
to basic methods of gathering, analyzing, and interprering qualitative
empirical materials. Parr 1moves from intervicwing to observing; to the use
of artifacts, documcnts, and records from thc past; to visual and auto-
ethnographic methods. It then rakes up analysis methods, including com-
puter-assisted mcthodologics, as \vell as strategies for analyzing talk and
text. Esther Madriz reads focus groups through critical feminist inquiry,
and Erve Chambers discusses applied ethnography.
methodologists around the world. Claudia Hoffman was essential in mov-
ing the series through production; we arc also grateful to the copy editor,
Judy Selhorsr, and to those whose proolreading and indexing skills were
so cemral to the publication of the Handbook. On which these volumes are
based. Finally, as ever, we thank our spouses, Katherine Ryan and Egan
Guba, for their forbearance and constaLt support.
The idea for this three~volume paperl::ack version of the Handbook did
not arise in a vacuum, and we are grateful for the feedback we received
from countless teachers and students.
-Norman K. Denzin
University of Illinois at Urbana-Champaign
- Yvonna S. Lincoln
Texas A&iVr Ul1ivers;ty

<$' Acknowledgments

Of course, this book would not exist without its authors or the editorial
board mcmbers for the Halldbooh on which it is based. These individuals
were able to offer barh long-rerm sustained commitments to the project
and shorr-term emergency assistance.
In addition, we would like to thank the following individuals and insti-
tutions for their clssistance, support, insights, and patience: our respective
universities and depanmellls, as \'\Jell as Jack Bratich, Ben Scott, Ruoyun
Bai, and Francyne Huckaby, our respective graduate students. \Vithour
them, we could never have kept this project on COllrse. There are also sev-
eral people to thank at Sage Publications. \YJe thank Ivlargaret Seawell, our
new editor; this three-volume version of the I-1a/ldbooh would not have
been possible \vithout lvlargaret's wisdom, S\1pporr, humor, and grasp of
the field in all its C\1rrem diversity.
As always, we apprccbre the efforts of Greg Daurelle, the director of
books marketing at Sage, along with his staff, for their indehtigable efforts
in getting the w~1rd our about the Hondboot: to teclchers, researchers, and
 1
Introduction
The Discipline and
Practice ofQualitative Research

Norman J<. Denzin and Yvonna S. Lincoln

Qu::l1itarive research has a long, distinguished, and sOl1lcl'lmcs ~m

gllished history in the human disciplines. In sociology, the vvork of
the "Chicago school" in the 19205 and 19305 established the impor-
tance of qualitative inquiry for the sruc1/ of human group life. In anthro-
pology, during the same rime period, 1"I1C discipline-defining studies of
Boas, JVlead, Benedict, Bateson, Evans-Pritchard, Radcliffe-Brown, and
Malinovvski chancel the omlines of the fieldwork method (sec Gupta &
Ferguson, 1997; Stocking, 1986, 1989). The agenda was clear-cut: The
observer wcnr to a foreign setting to t:tudy the customs clOd habits of
another society and culrure (see in Volume 1, Vidich & Lyman, Chapter 2;
Tedlock, Volume 1, Chapter 6; see also Rosaldo, 1989, pp. 25-45, for criti-
cisms of this tradition). Soon, qualitativ,~ research would be employed in
other social and behavioral science disciplines, including education (espe-
cially the work of Dewey), history, political science, business, medicine,
nursing, social work, and communicatie.ns.
In the opening chapter in Part I of Volume 1, Vidich and Lyman chan
many key features of this history. In thi~; now classic analysis, rhey note,
 COLLECTlt,IG NW It>.JTERPRETIHG QUALITATIVE MATERIALS
with some irony, that qualitative research in sociology and anthropology
"',,:.11:: "born our of COl1cel'n to t1nder~t~md the ;other.~ ;, Furthermore, this
other was the exotic other, a primitive, nonwhite person from a foreign
culture judged to be less civilized than th,n of the researcher. Of course,
there were colonialists long before there were anthropologists. Nonethe-
less, there would be no colonial, and now no postcolonial, history were it
not for this investigative mentality that turned the darl{~skinned other into
the object of the ethnographer's gaze.
Thus does bell hooks (1990, pp. 126-128) read the famous photo that
appears on the cover of W!riting Culture (Clifford & IVlarcus, 1986) as an
instance of this menraliry (see also Behar, 1995, p. 8; Gordon, 1988). The
photo depicts Stephen '~vler doing fieldwork in India. Tyler is seated some
distance from three dark~skinned persons. 1\ child is poking his or her
head out of a ll3.sker. 1\ woman is hidden in the shadows of a hur. A man, a
checkered white-and-black shaw! across his shoulder, clbow propped On
his knee, hand resting along the side of his face, is staring at 'T)'ler. Tyler is
writing in a field journaL A piece of white doth is attached to his glasses,
perhaps shielding him from the sun. This patch of whiteness l11<lrks lyler as
the white male \vriter studying these passive brown and black persons.
Indeed, the brown male's gaze signals some desire, or some attachment to
Tyler. In contrast, the female's gaze is complctely hiddcn by the shadows
,mel by the words of the book's title, which cross het face (hooks, 1990,
p. 127). And so this cover photo of perhaps the most influential book 011
ethnography in the last half of the 20th cemury reproduces "two ideas that
are quite fresh in the racist imagination: the notion of the white male as
writer/authority ... and the idea of the passive bnnvn/bJack man [and
woman and child] who is doing nothing, merely looking on" (hooks,
1990, p, 127),
In this imroductory chapter, we will define the field of qualitative re-
search and then navigate, chart, and review the history of qualitative re-
search in the human disciplines. This \vill allow us to locate this vol-
ume and its coments within their historical moments. (These historical
moments are somewhat artificial; they arc socially constructed, quasi-
historical, and overlapping conventions. Nevertheless, they permit a ';per-
formancc" of developing ideas. They also facilitate an increasing sensitiv-
ity to and sophisticarion about the pitfalls ;mo promises of ethnography
and qualitative rcscc1rch.) \Y/e will present a conceptual framework for
reading the qualitative research act as a multicultural, gendered pro-
cess, ,mel then provide <1 brief introduction ro the chapters thar follow.
Introduction
Returning to the observations of Vidich and Lyman as well as those of
books;, we will conclude with Do brief discussion of qualitative research and
critical race theory (see also in Volume 1, Ladson-Billings, Chapter 9; and
in this volume, Dcnzin, Chapter 13). As we indicate in our preface, we use
the metaphor of the bridge to structure wh~;t follows. \Ve see this volume
as a bridge connecting historical moments, research methods, paradigms,
and communities of interpretive scholars.
'" Definitional Issues
Qualitative research is a field of inquiry in its own right. It crosscuts dis-
ciplines, fields, and subjecr matters. I A complex, interconnected family of
terms, concepts, and assumptions surround the term qualitative research.
These include the traditions associated with foundationalism, positivism,
postfoundationalism, postpositivism, poststrucruralisrn, and the many
qualitative research perspectives, and/or methods, connected to cultural
and interpretive studies (the chapters in Parr II of Volume 1 take up these
paradigms).:>' There are separate and detailed literatures on the many meth-
ods and approaches that fall under the cai'l~gory of qualitative research,
such as case study, politics and ethics, participatory inquiry, interviewing,
participant observation, visual methods, and interpretive analysis.
In North America, qualitative research operates in <1 complex historical
field that crosscuts seven historical momems (we discuss these moments
in detail below). These seven moments overlap and simultaneously oper-
a[c in rhe present. 3 \"X'e define them as the rraditional (1900-1950); the
modernist or golden age (1950-1970); blurred genres (1970-1986); the
crisis of representation (1986-1990); the postmodern, a period of experi-
mental and new ethnographies (1990-199.i); posrexperimental inquiry
(1995-2000); and the future, which is now (2000-). The future, the sev-
enth moment, is concerned with moral discourse, with thc development of
sacred rextualities. The sevenrh moment asks that the social sciences and
the humanities become sites for critical conversations about democracy,
race, gender, class, nation-states, globalization, freedom, and cornmunitl-,.
The postmodern moment was defined in part by a concern for litera"rv
and rhetorical tropes and the narrative turn,.} concern for storytelling, fo'r
composing ethnographies in new ways (ElLs & Bochner, 1996). Laurel
Richardson (1997) observes that this moment was shaped by a new sen-
sibility, by doubt, by a refusal to privilege any method or theory (p. 173).
 COLLECT1HG AHD l\,lTERPRETIHG QUALITATIVE MATERIALS
But now, at the beginning of the 21st cenrury, the narrative turn has been
",hn. Many [,ovo learned how to write differently, including how to
locate themselves in their (CXts. \YJe now struggle to connect qualitative
research to the hopes, needs, goals, ;:mel promises of a free democratic
society.
Successive W3ves of epistemological theorizing move across these seven
moments. The traditional period is associated with the positivist, foun-
dational paradigrn. The modernist or golden 3ge and blurred genres
moments are connected to the appearance of postpositivist arguments. At
the same time, a variety of new interpretive, qualitative perspectives were
taken up, including hermeneutics, structuralism, semiotics, phenomenol-
ogy, cultural studies, and feminisl11.- 1 In the blurred genres .phase, the
humanities became cemr;]1 resources for critical, interpretlve theory,
and for the qualitative research project broadly conceived. The rescm:cher
became a brieolclIr (see below), learning how to borrow from many dIffer-
ent disciplines. . .
The blurred genres phase produced the next stage, the CrISIS of repre-
sentation. Here rcseMchers struggled with how to locate themselves and
their subjccts in reflexive texts. A kind of methodological diaspora rook
place, a two-way exodus. Humanists migrated to the social scienc~s,
searching for new social theory, new ways to study popular culture a~~ ItS
local, ethnographic contexts. Social scientists turned ro the humant~les,
hoping to learn how to do complex structural and poststructural readings
of social texts. From the humanities, social scientists also learned ho,:' to
produce texts thar refused lO be read in simplistic, linear, incontrovertIble
tcrms. The line between text and context blurred. In the postJ11odern
experimental moment researchers continued to move away from .foun~
dational and quasi-founcbtional criteria (see in this volume, Sl11lth &
Deemer. Chapter J 2. and Richardson, Chapter 14; and in Volume 1,
Gergen' & Gergen, Chapter 13). Alternative evaluative criteria we~e
sought, criteria that might prove evocative, moral, critical, and rooted 111
local understandings.
Any definition l;r qualitative research must work within this complex
historical field. Ollafitatiue research means different things in each of these
moments. N()n~'heless, an initial, generic definition can be offered: Quali-
tative research is a silUated activity thJt locates the observer in the world.
lt consists of a set of interpretive, material practices that make the world
visible. These praC[ices transform the "vodel. They turn the world i~ro a
series of reprcsc!1wtions. including field nares, interviews, conversatIOns,
Introductio.1
phorographs, recordings, and memos to the self. At this level, qualitative
research involves an interpretive, naturalistic approach to the world. This
means that qualitative resean.:her:::; :;tuuy things in their natural settings,
attempting to make sense of, Or to interpret, phenomena in terms of the
meanings people bring to them. s
Qualitative research involves the studied use and collection of a variety
of empirical materials-case study; perwnal experience; introspection;
life story; interview; artifacts; cultural texts and productions; observa-
tional, historical, interactional, and visnal texts-that describe routine
and problematic moments and meanings,n individuals' lives. Accordingly,
qualitative researchers deploy a wide ran:s:e of interconnected interpretive
pnlCtices, hoping always to get a better understanding of the subject mat-
ter at hane!' It is understood, however, that each practice makes the world
visible in a different way. Hence there is fTquently a commitment to using
more than one interpretive practice in any study.
The Qualitative Researcher
as Bricoleur and Quilt Maker
The qualitative researcher may take on multiple and gendered im-
ages: scientist, naturalist, field-worker, journalist, social critic, artist, per-
former, jazz musician, filmmaker, quilt maker, essayist. The many meth-
odological practices of qualitative research may be viewed as soft science,
journalism, ethnography, bricolage, quilt making, or montage. The re-
searcher, in turn, may be seen as a brieoleur, as a maker of quilts, or, as in
filmmaking, a person who assembles images into montages. (On montage,
See the discussion below as well as Cook, 1981, pp. 171-177; Monaco,
1981, pp. 322-328. On quilting, see hooks, 1990, pp. 115-122; Wolcott,
1995, pp, 31-33.)
Nelson, Treichler, and Grossberg (J992), Levi-Strauss (1966), and
Weinstein and Weinstein (1991) clarify the meanings of brieolage and bri-
6
eo/eur. A brieoleur is a "Jack of all trades or a kind of professional do-it-
yourself person" (Levi-Strauss, 1966, p. 17). There are many kinds of
brieo/curs-interpretive, narrative, theoretical, political (see below). The
in_terpretive bricoleur produces a brieoh~ge-that is, a pieced-together set
ot representations that are fitted to the specifics of a complex situation.
"The solution [bricolage) which is the re~:lIlt of the brieolcur's method is an
[emergent] construction" (Weinstein S: Weinstein, 1991, p. 161) that
changes and takes new forms as different tools, methods, and techniques
 COLlECflt'IG /.\1'-ID It'lTERPRETIHG QUALITATtVE MATERtALS
But now, at the beginning of the 21st century, the narrative turn has been
taken. lVIanv hJve learned how to write differently, including hm\' ro
locare them~e1ves in their texts. \\le now struggle to connect qualitative
research to the hopes, needs, gO<lls, and promises of a free democratic
soclery.
Successive waves of epistemological theorizing move JCross these seven
moments. The traditional period is associated with the positivist, foun-
darional panldigm. The modernist or golden Jge and blurred genres
moments arc connccted to the appearance of postpositivist arguments. At
the same time, 'J vmiety of new interpretive, qualitative perspectives wcre
takcn up, including hermeneutics, structuralism, semiotics, phenomenol-
ogy, cultural studies, and feITlinism.~ In the blurred genres .rInse, the
humanities becamc central resources for cririeal, interprenve theory,
and for the qualitative rese.Jrch project broadly conceived. The resea~cher
became a bricolellr (see below), learning how ro borrmv from many dIffer-
ent disciplines. . .
The blurred genres phase produced the next stage, the cnsls of repre-
sentation. Here researchers struggled with how to locate themselves i.ll1d
their subjects in reflexive ttxts. A kind of methodologicJI diaspora rook
place, a two-way exodus. Humanists migtated to the social scienc~s,
searching for new social theory, new ways ro study popubr culture a~d. Its
local, ethnographic contexts. Social scientists turned to the hUmaI1I~leS,
hoping to learn I1mv to do complex structural and poststrucrural readlllgs
of social texts. From the humanities, social scientists also learned how to
produce textS thar refused to be re:ld in simplistic, linear, incontrovertible
terms. The line between text and context blurred. In the postmoclern
experimental moment rese:u-chcrs continued to move away from foun-
dational and quasi~four\(.bti(Jnal criteria (see in this volume, Smith &
Deemer, Chalner 12, and Ricll",lfdson, Chapter 14; and in Volume 1,
Gergen & Gergen, Chapter 13). Alternative evaluative criteria wc:e
sought, criteria that might prove evocative, moral, critical, ,1l1d rooted In
local understandings.
Any definition of qualitative research must work within this complex
histor'ic:II field. Qlfl.dit.tfiue research means different things in each of these
moments. Nonetheless, all initial, generic definition Gin be offered: Quali-
tative research is a situated activiry th:It locates the observer in the world.
Tr consists of a set of interpretive, material practices that make the world
visible. These pracrices transform the world. They turn the v>-'OrId illl'O a
series of representations, including field notes, interviews, conversations,
Introduction
photographs, recordings, and memos to the self. At this level, qualitative
research involves an interpretive, naturalistic approach to the world. This
means that qualitative researchers study things in their natural settings,
attempting to make sense of, or to interpret, phenomena in terms of the
meanings people bring to them..1
Qualitative research involves the studied use and collection of a variety
of empirical materials-case study; personal experience; introspection;
life srory; interview; artifacts; cultural texts and productions; observa-
tional, historical, interactional, and visual texts-that describe routine
and problematic moments and meanings in individuals' lives. Accordingly,
qUJlitative researchers deploy a wide range of interconnecred interpretive
practices, hoping always to get a better understanding of the subject mat-
ter at hand. It is understood, however, that each practice makes the \vorld
visible in a different way. Hence there is frequently a commitment to using
more than one interpretive practice in any study.
The Qualitative Researcher
as Bricoleur and Quilt Maker
The qUJlitative researcher may rake on multiple and gendered im-
ages: scientist, naturalist, field-worker, journalist, social critic, artist, per-
former, jazz musician, filmmaker, quilt maker, essayist. The many meth-
odological pracrices of qualirativc rescarch may be viewed as 50ft science
journalism, ethnography, bricoJage, quilt mal~ing, or montage. The re~
searcher, in tllrn, may be seen as a brico/c1tI; as a maker of quilts, or, as in
fjJmmaking, a person who assembles images into montages. (On montage,
see the discussion below as well as Cook, 1981, pp. 171-177; lVlonaco,
1981, pp. 312-328. On quilting, see hooks, 1990, pp. 115-122; \Vo1cott,
[995, pp. 31-33.)
Nelson, Treichler, and Grossberg (1992), Levi-Strauss (1966), and
Weins rein and \'7einstein (1991) clarify the meanings of brieo/age and bri-
colelfl:(, A brieo/ellT is a "]ack of all trades or a kind of professional do-it-
yourself person" (Levi-Strauss, 1966, p. 17). There are many kinds of
brieolems-interpretive, narrative, theoretical, political (see below). The
interpretive bricoleur produces a brieo/age-thar is, a pieced-together set
of representations thar Me fitted to the specifics of a complex situation.
"The solution [bricolageJ which is the result of tho:: brico/cur's mcr110d is:m
[emergent] construction" (\X1einstein & Weinstein, 1991, p. 161) that
changes and takes new forms as different tools, methods, and techniques
 COllECT!I'lG AHD !I'ITERPRETIHG QUALITATIVE MATERIALS
of representation and interpretation are added to the puzzle. Nelson et a1.
(1992) describe the methodology of cultural studies "as a bricolage. Its
choice of practice, thar is, is pragmatic, strategic and self-reflexive" (p. 2).
This understanding em be applied, with qualifications, to qualitative
research,
The qualitative researcher as bricolellr or maker of quilts uses the aes-
thetic and material rools of his or her craft, deploying whatever strategies,
methods, or empirical materials are at hand (Becker, 1998, p. 1). If new
tools or techniques have to be invented, or pieced together, then the re-
searcher will do this. The choices as to which interpretive practices to
employ are not necessarily set in advance. The "choice of research prac-
tices depends upon the questions that are asked, and the questions depend
on their context" (Nelson et ai., 1991, p.1), what is available in the con-
tCXI, and what the researcher can do in that setting.
These interpretive practices involve aesthetic issues, an aesthetics of
representation that goes beyond the pragmatic, or the practical. Here
the concept of mOl/lage is useful (see Cook, 1981, p. 313; 1'v10naco, 1981,
pp. 171-172). rVIontage is a method of editing cinematic images. In the his-
tory of cinematography, montage is associated \vith the work of Sergei
Eisenstein, especially his film The Battleship PotemhiTl (1925). In mon-
tage, sever;]l different in13ges are superimposed onto one another to create
a picture. In a sense, mont;:1ge is like pemimento, in which something that
has been paimed out of a picturc (an image the paimer "repented," or
denied) becomes visible again, creating something new. \Vhat is new is
what had been obscured by a previous image.
lvionrage and pentimetlto, like jazz, which is improvisation, create the
sense thar images, sounds, and understandings 3re blending together,
overlapping, forming a composite, a new creation. The images seem to
shape and define one ;mother, and an emotional, gestalt effect is pro~
duced. Often these images are combined in a swiftly run filmic sequence
that produces a dizzily revolving collection of several images around a cen-
tral or focused picture or sequence; such effects arc often Llsed to signify
the passage of time.
Perhaps thc most famous instance of montage is the Odessa Steps
sequcnce in The Battleship Potemhll. In the climax of the film, the citi-
zens of Odessa arc being massacred by czarist troops on the stone steps
leading down ro the harbor. Eisenstein cuts to a young mother as she
pushes her baby in a carriage across the landing in front of the firing
troops. Citizens rush P;1st her, jolting the c.arriage, which she is ;:1fraid to
Introduction
push dO\vn to the next flight of stairs. The troops are above her firing at
tbe citizens. She is trapped berween the troops and the steps. She screams.
A line of rifles pointing to the sky erupt in smoke. The mother's hcad
sways back. The wheels of the carriage teeter on the edge of the steps. The
mother's hand clutches the silver buckle of her belt. Below her people are
being beaten by soldiers. Blood drips over tbe mother's white gloves. The
baby's hand reaches our of the carriage. The mother sways back and forth.
The troops advance. The mother falls back against the carriage. A woman
watches in borror as the rear \vheels of tbe carriage roll off the edge of the
landing. With accelerating speed the carriage bounces down the steps, past
the de'ld citizens. The baby is jostled [rom side to side inside the carriage.
The soldiers fire their rifles into a group of wounded cirizens. A student
scrC;lnlS as the carriage leaps across the steps, till'S, and overmrns (Cook,
1981, p. 167)."
IVlonrage uses brief images to create a clearly defined sense of urgency
and complexity.1vlonrage invites viewers 1O construct interpretations that
build on one anodler as the scene unfolds. These interpretations are built
on :lssociations based on the contr;1sting images that blend into one
another. The underlying assumption of montage is that viewers perceive
and interpret the shots in a "montage sequence not sequentially, or one
at a time, bur rarher simultaneollsly" (Cook, 1981, p. 171). The viewer
puts dIe sequences together imo a meaningful emotional INhale, as if in a
glance, all at once.
The qualirmive researcher who uses montage is like ;-1 quilt maker or a
jazz improviser. The quilter stitches, edits, and puts slices of reality to-
gether. This process creatcs and brings psychological and emotional ~li1ity
to an interpretive experience. There are many'--'examples of montage i;l
current qualitative research (see Diversi, 1998; Jones, 1999; Lather &
Smithies, 1997; Ronai, 1998). Using multiple voices, different textual for-
mats, and various typefaces, Lather and Smithies (1997) weave a complex
text abolit women who arc HIV positive and \vomen with AIDS. Jones
(1999) creates a performance text using lyrics from the blues songs sung by
Billie Holiday.
In texts based on the metaphors of montage, quilt making, and jazz
improvisation, many different rhings are going on at the same time-
different voices, different perspectives, points of views, angles of vision.
Like performance texts, works th,u usc 1lI0J1hlS'>: tiimulrul1coU::ily [rC/irc
and enact moral meaning. They move from the personal to the political,
the local to the historic}l and the cultural. These are dialogical texts. They
 LULLt:L I !I'll:J AI'IU II'll CKt"'I,C! II'j\;J ,,<UM\-! 1M!' V,- 'V'r\' '-"'r\'---'
presume an active audience. They create spaces for give-and-rake between
reader and writer. They do more than rurn the other into the object of the
social science gaze (see IVlcCall, Chapter 4, Volume2).
QU~llitative research is inherenrly multimethod in focus (Flick, 1998,
p. 119). However, the use of multiple methods, or triangulation, reflects
an attempr to secure an in-depth understanding of the phenomenon in
question. Objective reality can never be caprured. \\Te can know a thing
only through its represenrations. Triangulation is not a tool or a strmegy of
validation, but an alternative to validmion (Flick, 1998, p. 230). The com-
bination of multiple mcrhodological practices, empirical materials, per-
spectives, and observers in a single study is best understood, then, as a
s~rategy that adds rigor, breadth, complexity, richness, and depth to any
inquiry (see Flick, 1998, p. 231).
In Chapter 14 of this volume, Richardson disputes 1I,e concept of tri-
;:lllgulatio~, asserting that the central image for qualitative inquiry is tbe
cr);s[;1l, nor the trianglc. Mixed-gcnre texts in the posrexperi111cntal mo-
mcnt have more than three sides. Like crystals, Eisenstein's montage,
the jazz solo, or the pieces that make up a quilt, the mixed-genre tcxt, as
Richardson notes, "combines symmetry and substance with ~111 infinite
v3riety of shapes, substances, transmutations.... Crystals grmv, change,
alter. ... Crystals are prisms thar reflect externalities and refract within
themselves, crearing different colors, patterns, and arrays, c3sring off in
different directions."
In the crystallization process, the writer tells the same tale from differ-
ent points of view. For example, in Ii Thrice-Told ~D11e (J991), I'I/largery
\X1olf uses fiction, field notes, and a scientific article to give an accounting
of the same set of experiences in a native village. Similarly, in her play Fires
ill the Mirror (1993), Anna Deavere Smith presents a series of performance
nieces based on interviews with people involved in a racial conflict in
Crown Heights, Brooklyn, on August, 19, 1991 (sec Denzin, Chapter 13,
this volume). The play has multiple speaking parts, including conversa-
tions \vith gang members, police officers, and anonymous young girls and
boys. There is no "correct" telling of this event. Each telling, like light hit-
ting a crystal, reflects a differenrperspective on this incident.
Viewed as a crystalline form, 3$ a montage, or as a creative performance
around a central theme) triangulation as;] form of) or a.lternmive to, valid
ity thus can be extended. Tria.ngulation is the display of multiple, refracted
realities simu Itaneously_ Each of the metaphors "vmrks" to create simulta-
neity rather than the sequential or linear. Readers and audiences are then
8 9
"'t) UUULLlUI'
invited ro explore cornpeting visions of the context, to become immersed
in and merge with new realities to comprehend.
The methodological bricole1/r is adept at performing a large number of
diverse tasks, ranging from intervie\ving to intensive self-reflection and
introspection. The theoretical bricoleur reads widely and is knowledge-
able about the many interpretive paradigms (feminism, ivlarxism, cultural
srudies, constructivism, queer theory) that Gill be brought to any particu-
lar problem. He or she may not, however, feel that paradigms can be min-
gled or synrhesized. That is, one cannot easily move between paradigms 3S
overarching philosophical systems denoting particular ontologies, episte-
mologies, and methodologies. They represent belief systems that attach
users to particular worldviews. Perspectives, in contrast, are less well
developed systems, and one can more easily move between them. The
researcher-as-bricoleltr-theorist works between and within competing ;md
overlapping perspectives and paradigms.
The interpretive bl'icoleltr understands that research is an interactive
process shaped by his or her personal history, biography, gender, social
class, race, and ethnicity, and by those of the people in the setting. The
political bricoleur knows that science is power, for all research find-
ings have political implications. There is no value-free science. A civic
social science based on a politics of hope is sought (Lincoln, 1999). The
gendered, narrative bricolellr also knows that researchers all teB Stories
about the worlds they have studied. Thus the narratives, or stories, scien-
tists tell arc aCcounts couched and framed within specific storytelling
traditions, often defined as paradigms (e.g., positivism, postpositivism,
constructivism).
The produc[ of the interpretive bricolellr's labor is a complex, quiltlike
~)ricolage, a reflexive collage or montage-a set of fluid, interconnected
Images and representations. This interpretive structure is like a quilr, a
performance text, a sequence of representations connecting the parts to
the whole.
Qualitative Research as a
Site of Multiple Interpretive Practices
. Qualitative research) as a set of interpretive acrivities, privileges no
Single methodological practice over another. As a site of discussion or
discourse, qualitative research is difficult ro define dearly. It has no the'ory
Or paradigm that is distinctly its own. As the contributions to Part II of
Volume 'I reveal, multiple theoretical paradigms claim use of qualitative
rese~lfcb methods and strategies, from constructivist to cultural studies,
feminism, lVbrxism, and cdmic models of study. Qualitative research is
used in many separate disciplines, as we will discuss below. It does not
belong to a single discipline.
Nor docs qualitative research have;] distiner set of methods or pfilCtices
that 3fe entirely its o\""n. Qualitative rescarchers use semiotics, narrative,
content, discourse, Jrchival and phonemic analysis, even statistics, tables,
graphs, and numbers. They also draw upon and utilize the approachcs,
methods, and techniques of cthnomethodology, phenomenology, herrne-
neutics, feminism, rhizomatics, deconstrucrionism, ethnography, inter-
vic\vs, psychoanalysis, cultural studies, survcy research, and participant
observation, among others. 9 All of these research practices "call provide
important insights and knowledge" (Nelson etal., ] 992, p. 2). No specific
method or practice can be privileged over any other.
IvLlllY of these methods, or research pr'1etices, are used in other con-
texts in the human disciplines. Each bears the traces of its own disciplinary
history. Thus there is an extensive history of the uses and meanings of eth~
nography and ethnology in education (see Fine, \Xleis, \X1eseen, & \X1ong,
Volume 1, Chapter 4); of participant observation and ethnography in
'H1thropology (see Tedlock, Volume 2, Clupter 6; Ryan & Bernard, this
volume, Chapter 7; Brady, this volume, Chapter 15), sociology (sec
Gubrium & Holstein, Volume 2, Chapter 7; and in this volumc, Harper,
Chapter 5; Fontana & Frey, Chapter 1; Silverman, Cbapter 9), communi-
cation (see Ellis & Bochner, this volume, Chapter 6), and cultural studies
(see Frow & Ivlorris, Volume 1, Chapter 11); of textual, hermeneutic,
feminist, psychoanalytic, semiotic, and narrative analysis in cinema and
literary studies (see Olesen, Volume 1, Chapter 8; BrJdy, this volume,
Chapter 15); of archival, material culture, historical, and document analy-
sis in history, biography, and archacology (see Hodder, this volume, Chap-
ter 4; Tierney, Volume 2, Chapter 9); and of discourse and conversational
analysis in mcdicine, communications, and education (scc Ivliller & Crab-
tree, Volume 2, Chapter 12; Silverman, this volume, Chapter 9).
The many histories that surround each method or rese:lrch strategy
reveal how multiple uscs and meanings are brought ro each practice. Tex-
tual .lllalyses in lirerary swdies, for example, oftcn treat tCXtS as self-
contained system~, On rhe other hand, a researcher raking:] culrural stud-
ies or feminist perspective will read a text in terms of its loc:ltion within a
10
hiswriol moment marked by a particular gender, race, or cl ass ideology. A
cultural studies use of ethnography would bring a set of understandings
from feminism, postmodernism, and poststrueturalisrn to the project.
These understandings would not be shared by mainstream postpositivist
sociologists. Similarly, postpositi\'ist and poststrucrur<1list historians bring
different understandings and uses to the methods and findings of his-
torical research (see Tierney, Volumc 2, Chapter 9). These tensions ;1lld
contradictions 3re all evident in the chapters in this volume.
These separate anclmultipJc uses and meanings of the merhods of quali-
tative research make it difficult for researchers to agree on ilny essential
definition of the field, for it is never just: one thing. lo Still, \ve must estab-
lish a definition for our purposcs here. \X7e borrow from, ;ll1d paraphrase,
Nelson et al.'s (1992, p. 4) attempt to define cultUf:.l! studies:
Qualitative research is an interdisciplinary, transdisciplillary, and some-
times counrerdisciplinary field. It crosscuts the hum:mitics and rhe social
and physical sciences. Qualirative research is mall)' things at the S:l[lle time.
it is multi paradigmatic in focus. Its praetirioncrs are sensitive to the v;l!lIe of
the Olulrimerhod 'lppro:lCh. They 3re commitred ro the naturalisric perspec-
rive and to the interpretive understanding of human experience. Ar the
same time, the field is inherently polirical and shaped by multiple ethical
dnd political positions.
Qualitarivc research embraces two rensions at rhe same time. On the one
hand, it is drawn to a broad, inrerpretive, postexperirnenwl, P05tl1lOdcrn,
feminist, ::md crirical sensibility. On the other h;l\1d, ir is drawn to more nar-
rowly defined positivist, posrpositivisr, humanisric, and naturalisric COll-
ceptions of human experience ;l11d irs ::l11alysis. Further, these tensions can
be combined in the same project, bringing both posunodcrn and n:Hur;!lisric
or borh critical and humanistic perspecrives to bear.
This rather complex statement means that qualitative research, as a set
of practices, embraces within its own multiple disciplinary histories con-
stant tensions and contradictions over the project itself, including irs
methods and til e f orms Its iln dmgs ancI Interpretatlons
. ,. . . , freld
take. TI1C
~pra\~'ls bet\veen and croSSCutS all of the human di,;ciplines, even indud-
J~.sl In. some cases, the physical seiene-eo;. Its prClct;tioncr5 Llrc v;.triv\l:;ly
I.:ommmed to modern, postmodern, and postexperirnental sensibilities
and the apI'( oac
. Iles to
. .SOCla
. I rcseare I1 t I1.l[ t Ilese senstbdltles
. ... Imply.
.
II
 ...... VLLL'-' \l>'U r'I'-'l.-' "" '-,,, ,,,.. ,,, ..... "<:'-" ,_."

Volume I reveal, multiple theoretical paradigms claim lise of qualitative
research methods <ind strategies, from constructivist to cultural studies,
feminism, IVlarxisl1l, and ethnic models of study. Qualitative research is
used in many separate disciplines, as we "vill discuss below. It does not
belong to a single discipline.
Nor docs qualitative research have a distinct set of methods or practices
that are emirelv its own. Qualitative researchers use semiotics, narrative,
content, discOl;rse, archival and phonemic analysis, even statistics, tables,
graphs, and numbers. They also draw upon anclurilize the approaches,
methods, and techniques of ethnomethodology, phenomenology, herme-
neutics, feminism, rhizomatics, deconstrucrionism, ethnography, inter-
views, psychoanalysis, cultural studies, survey research, and participant
observation, among others. ') All of these research practices "can provide
important insights and knowledge" (Nelson et aI., 1992, p.l). No specific
method or practice can be privileged over any other.
Ivlanv of these rnethods, Or research practices, are used in other con-
texts in ~he human disciplines. E::lch bears the tr::lces of its own disciplinary
history. Thus there is an extensive history of the uses and meanings of eth-
nography and ethnology in education (sec Fine, \X!eis, \X/cscen, & \Vong,
Volume 1, Chapter 4); of participant observation and ethnography in
anthropology (see Tedlock, Volume 2, Chapter 6; Ry,1I1 & Bernard, this
volume, Chapter 7; Brady, this volume, Chapter 15), sociology (sce
Gubrium & Holstein, Volume 2, Chapter 7; and in this volume, Harper,
Chapter 5; Fontana & Frey, Chapter 2; Silverman, Chapter 9), communi-
cation (see Ellis & Bochner, this volume, Chapter 6), and cultural studies
(see Frow & I\iforris, Volume I, Chapter 11); of textual, hermeneutic,
feminist, psychoanalytic, semiotic, and narrative analysis in cinema and
literary studies (see Olesen, Volume 1, Chapter 8; Brady, this volume,
Chapter 15); of archival, material culture, historical, and document analy-
sis in history, biography, and archaeology (see Hodder, this volume, Chap-
ter 4; Tierney, Volume 2, Chapter 9); and of discourse and conversational
analysis in medicine, communications, and education (see Miller & Crab-
tree, Volume 2, Chapter 12; Silverman, this volume, Chapter 9).
The Illany histories that surround each method or research strategy
reveal how multiple uses and meanings are brought to each practice. Tex-
tual analyses in literary studies, for example, often treM texts :1S self-
historical moment marked by a particular gender, race, or class ideology. A
culrural studies use of cthnography would bring a set of understandings
from feminism, postlllodernism, and poststrucruralisl1l to the project.
These undersrandings \\'oulcl not be sharcd by mainsrream postpositivist
sociologists. Silllilarly, postposirivist and poststrueturalist historians bring
differenr understandings and uses to the methods and findings of his-
torical research (see Tierney, Volume 2, Chaprer 9). These tensions and
conrradictions are al1 evident in the chapters in this volume.
These scparate and multiple uses and meanings of rhe methods of quali-
tative research make it difficult for researchers to 3gree 011 all\! essential
definition of the field, for it is never just one thing. If! Srill, we n~ltlSt estab-
lish a dcfinition for our purposes herc. \'Xle borrmv [rom, and parapbr3se,
Nelson et 0.1. 's (1992, p. 4) 3trernpt to define cultural studies:
Qualitative research is an interdisciplinary, tr:J.nsdisciplillary. ~llld some-
times counterdisciplin;:Hy field. It crOSSCuts the humanities and the soci;ll
;lnd physical sciences. Quallt,Hive research is many things ,H the S;l1ne time.
It is multi p,u;ldigmatic in focus. Its practitioners arc sensitive to the value of
rhe l11ultimcthod approach. They are committed to the naturalistic perspec-
tive and to the imcrprerive understanding of human experience. At the
same time, rhe field is inherently political and shaped by llluiriplc ethical
and political positions.
Qualitative research embraces two tensions at the same time. On the one
hand, it is drawn to a broad, irnerprerive, postexpcrimental, postmodern,
feminist, and critical sensibility. On the other hand, it is drawn to more nar-
rowly defined positivist, postpositivist, humanistic, and naturalistic con-
ceptions of hunufl experience and irs analysis. Funher. these tensions can
be combined in the same project, bringing both poslmodern and naturalistic
or both critical and humanistic perspectives (0 bear.
. This rather complex statement means that qualitative research, as a set
ot practices, embraces within its own multiple disciplinary histOries con-
Stant tensions and cOntradicrions over the project itself, including its
methods and tIl corms
< J
f ' ItS
' , .m-mgs
f' d' 'I
an d rntctpreutlons
' fa (e. The field '
sprawls
- . between
- and
- Cl , - II t' 1 I I" I'
.osscuts a 0 [le lUman (rSCIp Incs, even l11CU(- , I I
l!lg In Some (;1
)
hi' [ . .. "
Sf'S:. t e p ly,;tC'l SCIenCeS. Its praCl.l"'OJ1~or'j "c'", ... an<.,,_,,>!y
<

contained systems_ On the other hand, a researcher taking a cultural stud- Committed d. "" " " ,'
. . to rno ern, postmodern, and postcxpenmemal sensrbrlltles
ies or feminist perspective "vill read a tcxt in terms of its location "vi thin a dndtheal,proa-I_' _"I .....
-. C les to SOCIa research that these senslbJ1ltlcs !lnply.
<

10 11
 l..ULLCl..l Il'l\.;l 1-\1'1\..1 11~ I.Lr\r l\'L I ""\,;) ",:,v,..,,_, ,,..., 'Y L ""-,, .... ,",-,~..,
Resistances to Qualitative Studies
'The academic Jnd disciplinary resistances to qualitative research illus-
trate the politics embedded in this field of discourse, The challenges to
qualitative research are many. Qualitative researchers are called journal-
ists, or soft scientists. Their \vode is termed unscientific, or only explor-
atory, or subjective. It is called criticism and nor theory, or it is inrerpreted
politically, as ;] disguised version of IVIarxism or secular humanism (see
Huber, 1995; see also Denzin, 1997, pp. 258-261).
These resistances reflect an uneasy awareness that the traditions of
qualitative research commit the researcher to a critique of the positivist or
posrpositivist project. Bur the positivist resistance to qualitative research
goes beyond the "ever-present desire to maintain a distinction between
hard science and soft scholarship" (Carey, 1989, p. 99; see also in Volume
1, Schwandt, Chapter 7; in this volume, Smith & Deemer, Chapter 12).
The experimental (positivist) sciences (physics, chemistry, economics, and
psychology, for example) are often seen as the crovvning achievements of
\Xlesrern civilization, and in their practices it is assumed that "truth" can
transcend opinion and personal bias (Carey, 1989, p. 99; Schwandt,
1997b, p. 309). Qualitative research is seen as an assault on this tradition,
whose adherents often retreat into a "value-free objectivist science"
(Carey, 1989, p. 104) model to defend their position. They seldom
attempt to make explicit, or to critique, the "moral and political com-
mitments in their own contingent work" (Carey, 1989, p. 104; sec also
Lincoln & Guba, Chapter 6, Volume 1).
Positivists further allege that the so-called new experimental qualitative
researchers write fiction, not science, and that these researchers have no
\vay of verifying their truth statements. Ethnographic poetry and fiction
signal the cleath of empirical science, and there is little to be gained by
attempting to engage in moral criticism. These critics presume a stable,
unch:mging reality that can be studied using the empirical methods of
objective social science (see Huber, 1995). The province of qualitative re-
search, accordingly, is the vi,iQrld of lived experience, for this is where indi-
viclual belief and action intersect with culture. Under this model there is no
preoccupation with discourse and method as material interpretive prac-
tices tbat constitute representation and description. Thus is the texrual,
narrative turn rejected by the positivists_
12
J,/tlUUUl..UUfl
The opposition to positive science by the postposil'ivlsts (see below)
and the poststructur8lists is seen, then, as an arrack on reason and trmh. At
the same time, the positivist science attack on qualitative research is
regarded as an attempt to legislate one version of truth over another.
This complex political terrain defines the many traditions and strands
of qualitative research: the British tradition and its presence in other
national contexts; the American pragmatic, naturalistic, and interpretive
traditions in sociology, anthropology, communication, and education; the
German and French phenomenological, hennencutic, semiotic, L\/larxist,
structural, and poststrucrural perspectives; feminist studies, African
American stlldies, Latino studies, queer studies, studies of indigenous and
aboriginal cultures. The politics of qualitative research create a tension
thar informs each of the above traditions. This tension itself is constantly
being reexamined and interrogated, as qualitative research confronrs ~
changing historical world, Ilew intellectual positions, and its o\\'n institu-
tional and academic conditions.
To sU~ll11arize: Qualitative research is many things ro many people. Its
essence IS twofold: a commirlTIent to some version of the nawralistic,
int~r?retive approach to its subject matter and an ongoing critique of the
pohncs and methods of postposjtivism. \X!e turn now to a brief discussion
of the major differences between qualitative and quantitative approaches
~o research. \'(/c then discuss ongoing differences and tensions within qual-
Itative inquiry.
Qualitative Versus Quantitative Research
The word qualitative implies an emphasis on the qU<llitics of {'-
and on processes and meanings that are not experimentally"
measured (if measured 3t all) in terms of quantity, arr
frequency. Qualitative researchers stress the soeir ,-
reality, the intimate relationship between ..'
ied, and the situational constr~inr"
emphasize the value-laden nature G
lions that Stress how social expcrien
Contrast, quantitative studies emphasi'i.
causal relationships benvcen variables, l.
Studies claim that their work is clone from,
13
 LULL!:\.. I II'!\,;! AI'H..! II" I Lf'." f'.L , "''-J '""Ur. ... , , ..... , " ... ",n, ~''',..,~~
" " ' UUULlIUI,

----------~-~.. ~-._-~

Resistances to Qualitative Studies The opposition to positive science by the postpositivists (sec belm\')
and the POststflleturalists is seen, then, as an arrack on reason ;:md truth. Ar
The academic and disciplinary resistances to qualitative research illus- rhe same time, the positivist science arrack on qualitative research is
trate the politics embedded in this field of discourse. The challenges to regarded as an attempt ro legislate one version of truth Over another.
qualitative research are many. Qualitative researchers are called journal- This complex political terrain defines the many traditions and strands
ists, or soft scientists. Their work is termed unscientific, or only explor- of qualitative research: the British tradition and its presence in other
atory, or subjective. It is called criticism and not theory, or it is interpreted national contexts; the American pragmatic, naturalistic, and interpretive
politically, as a disguised version of tvlarxism or secular humanism (see traditions in sociology, anthropology, communication, and education' the
I-luber, 1995; see also Denzin, 1997, pp. 25S-261). German and French phenomenological, hermeneutic, semiotic, I\Ilar~ist,
These resistances reflee[ an uneasy awareness that the traditions of structural, and POStstrucrural perspectives; feminist studies, African
qualitative research commit the researcher to a critiqtle of the positivist or American studies, Latino studies, ClLleer studies, swdies of indigenolls and
postpositivist project. Bur the positivist resistance to qualitative research aboriginal cultures. The politics of qualitative research create a tension
goes beyond the "ever-present desire to maintain a distinction between dl;:~t informs c,ach of the, above traditions. This tension itself is constantly
hard science and solt scholarship" (Carey, 1989, p. 99; see also in Volume bemg reexammed and Interrogared, as qualitative research confronts a
1, Schwandt, Chapter 7; in this volume, Smith & Deemer, Chapter 12). changing historical world, new imellecnwl positions, and irs own institu-
The experimental (positivist) sciences (physics, chemistry, economics, and tional and academic conditions.
psychology, for example) are often seen as the crowning achievements of To sUI,llmarize: Qualitative research is mallY things to many people. Its
\Vestern civilization, and in their practices it is assumed that "truth" can essence IS twofold: a coml11irment to SOl11e version of the naturalistic
transcend opinion and personal bias (Carey, 1989, p. 99; Schwandt, int~r~retive approach to its subject matter and an ongoing critique of th~
1997b, p. 309). Qualitative research is seen as an assault on this tradition, polmc5 and methods of postpositivism. \\1e rurn no\\' to J brief discLission
v,Ihose adherents often retreat into a "value-free objectivist science" of the major differences between qualitative and quantitative approaches
(Carey, 1989, p. 104) model to defend their position. They seldom ~'o r~se~rch',Wethen discuss ongoing differences and tensions within qual-
Itative mqll1ry.
attempt to make explicit, or to critique, the "moral and political com-
mitments in their own contingent work" (Carey, 1989, p. 104; see also
Lincoln & Guba, Chapter 6, Voilime 1).
Qualitative Versus Quantitative Research
Positivists further allege that the so-called new experimental qualitative
researchers write fiction, not science, and that these researchers have no
The word qualitative implies an emphasis on the qualities of entities
\vay of verifying their truth statements. Ethnographic poetry and fiction
and on processes and meanings that are not experimentally examined or
signal the death of empirical science, and there is little to be gained by
measured (if measured at all) in terms of quantity, amount, intensity, or
attempting to engage in moral criticism. These critics presume a stable,
[req.uency. Qualitative researchers stress the socially constructed nature of
unchanging reality that can be studied llsing the empirical methods of realIty
, .' the intIma tere I
aLluns'1
11p between t I1C researcher and v,1hat is stud-
objective social science (see Huber, 1995). The province of qualitative re~ !Cd and tlI . I .
. '. ' e SttuatlOna COl1Strall1ts that shape inquiry. Such researchers
search, accordingly, is the \vorld of lived experience, for this is where indi- emphaSize
~'
the v"'IL]e-l~deJ
'" d.1 n,~tlIfe () ftnqUlry.
' -flley see I( answers to qucs-
vidual belief and action intersect with culture. Under this model there is no ~lon5 that stre I ' I . .
'_ ,S5 JOW SOCIa experIence 1$ created and given meaning. In
preoccupation with discol1r~e and method as m:1terial interpretive prac- Contrast, QU:10tit;1!:ivc stlld;es c,npb,,-s;LC tbc ITl",aSUrClllcnl "nd ..n .. l(::;~s of
tices that constitute reprcsentation and description. Thus is the textual, causal
_'
reiationl I . . bl
,. SlipS JC[Ween vana es, not processes. Proponents of such
narrative tllrn rejected by the positivists. Studlesclalmtb'Hth--',
_c
1"1 f 1
- ell \\or \.IS L one rom "'I'll' 11ll a value-free framework.

12 13
 LULLt'_III'I\:l AI'lU II'll cr<.rr:.c, u''..:1 \,:IVM,-' 'M'" '- ",,...., ~.u,-,~ ""uu ........ '-'u,,

Research Styles: Doing the Same Things Differently~ Flick (1998, pp. 2-3) llsefully sllmmarizes the differences between these
t\\lO approaches to inquiry. He observes that the quantitative approach has
Of course, both qualitative and quantitative researchers "think they been used for purposes of isolating "causes and dfects ... opcrational-
kno\v something about society worth telling to others, and they use a vari- izing theoretical relations ... [and] measuring and ... quantifying phe-
ety of forms, medi::t and means to communicate their ideas and findings" nomena ... allowing the generalization of findings" (p. 3). Bur today
(Becker, 1986, p. 122). Qualitative research differs from quantitative re- doubt is cast on such projects, because "Rapid social change and the re-
search in five significant ways (Becker, 1996). These points of difference sulting diversification of life worlds arc increasingly confronting social
turn on different ways of addressing the samc set of issues. They return researchers \..\Iith new social COntexts and perspectives.... traditional de-
aiways to the politic~ of research, and to who has the power to legislate ductive methodologies ... are failing.... thus research is increasingly
correct solutions to these problems. forccd to make use of inductive strategies instead of starring from theories
and testing them.... knmvledge and practice 3re studied as local knowl-
Uses of positivism alld postpositivisJII. First, both perspectives are shaped edge and practice" (p. 2).
by the positivist and postpositivist traditions in the physical and social sci- Spindler and Spindler (1992) summarize their qualitative approach to
ences (see the discussion below). These two positivist science traditions quantitative materials: "Instrumentation and quanrification are simply
hold to na'ive and critical realist positions concerning reality and its per- procedures employed to extcnd and reinforce certain kinds of data, inter-
ception. In the positivist version it is contended that there is a reality out pretations and test hypotheses aCross samples. Both must be kept in their
there to be studied, captured, and understood, whereas the postpositivists place. One must avoid thei r premature or over! V extensive use as a securitv
argue that reality can never be fully apprehended, only approximated mechanism" (p. 69). . .
(Guba, 1990, p. 22). Postpositivisll1 relies on multiple methods as a way of Although many qualitative researchers in the postpositivist tradition
capturing as much of reality as possible. At the same time, emphasis is will use statistical measures, methods, and documems as a way of locating
placed on the discovery and verification of theories. Traditional evalua- groups of subjects within larger populations, they will seldom report their
tion criteria, such as internal and external validity, are stressed, as is the use findings in terms of the kinds of complex statistical measures or methods
of qualitative procedures that lend themselves to structured (sometimes to which quantitative researchers arc drawn (i.e., path, regression, or log~
statistical) analysis. Computer~assisted methods of analysis that permit linear analyses). '-
frequency counts, tabulations, and low-level statistical analyses may also
be employed. Acceptance of postmodem sensibilities. The use of quantitative, positivist
The positivist and postpositivist traditions linger like long shadows me.thods and assumptions has been rejected by a nC\\I generation of quali-
over the qualitative research project. Historically, qualitative research was tatIve researchers who are attached to poststructural and/or postmodern
defined within the positivist paradigm, where qualitative researchers sensibilities (see below; see also in Volume 1, Vidich & Lyman, Chapter 2;
attempted to do good positivist research with less rigorous methods and and.i~ ~his volume, Richardson, Chapter 14). These researchers argue that
procedures. Some mid-20th-century qualitative researchers (e.g., Becker, POSItiVISt methods are but one way of telling stories about society or the
Geer, Hughes, & Strauss, 1961) reported participanrobservation findings social world. These methods may be no better or 110 worse than any other
in terms of quasi-statistics. As recemly as 1998, Strauss and Corbin, tWO methods; they JUSt tell differenr kinds of stories.
leaders of the grounded theory approach to qualitative research, at- This toleram view is not shared by everyone (ITuber, 1995). Nlany
tempted to modify the usual canons of good (positivist) science to fit their members - of t Ile' Cfmca
- .. -j t Ileory, constructIvIst, .. poststrucrural, and post-
o\,.;n postpositivist conception of rigorous resc:1fch (but see CI13rrnaz, modern schools, of r Ilong h t reject .- POsltlVl~t::in
.-' . d postpOSltlVlst
. .. cntena
"
Chapter 8, Volume 2; see also Glaser, 1992). Some applied rese3rchers, When . evalu' 1
o' I I ' "
anng tklf 0\'1'11 \\lor (. T ley see these CrIterIa as Irrelevant to
\vhile claiming to be atheoretical, often fit within the positivist or postpOS- rhelr
. _.
worl"I. and COntenc
- . I t Ilat sue' 1jctlrena
" .
reproduce onlv a certam .
lond of
itivist framework by default. SCience a . - -I . _ _ _ ....
, SCIence t 1m silences too many VOlees. [hesc researchers seek
14 15
 LULLtL 111'1\..:1 AI'IU 11"-11 t:r~n~1: 111'1\.:1 \.!UMLlIKII
alternative methods for evaluating their work, including verisimilitude,
emotionality, personal responsibility, an ethic of caring, political praxis,
multivoiccd texts, and dialogues with subjects. In response, positivists and
postposirivisrs argue that what they do is good science, free of individual
bias and subjectivity. As noted above, they see postmodernism and post-
structuralism as attacks on reason and truth.
Capturing the individual's point ofview. Both qualitative and quantitative
researchers are concerned with the individual's point of view. Hmvever,
qualitative investigators think they can get closer to the actor's perspective
through detailed interviewing and observation. They argue that quantita-
tive researchers arc seldom able to capture their subjects' perspectives
because they have to rely on more remote, inferential empirical methods
and materials. The empirical materials produced by interpretive methods
8fe regarded by mallY quantitative researchers as unreliable, impressionis-
tic, and not objective.
Exanli71ing the constraints of eueryda)' Iile. Qualitative researchers are
more likely to confront and come up against the constraints of the every-
day social world. They see this world in action and embed thcir findings in
it. Quantitative researchers abstract from this world and seldom study it
directly. Thcy seek a nomothetic or etic science based on probabilities
derived from the study of large numbers of randomly selected cases. Thcse
kinds of statements stand above and outside the constraints of everyday
life. Qualitative researchers, on the other hand, are committed to an ernic,
idiographic, case-based position, which directs their attention to the spe-
cifics of particular cases.
Securing rich descr;ptiotls. Qualitative researchers believe that rich de-
scriptions of the social world are valuable, whereas quantitative research-
ers, with their etic, nomothetic commitments, are less concerned with
such detail. Quantitative researchers are deliberately unconcerned with
rich descriptions because such detail interrupts the process of developing
generalizarions.
The five points of difference described above (uses of positivism and
postpositivism, postmodernism, capturing the individual's point of view,
examining the constraints of everyday life, securing thick descriptions)
reflect commitments to different styles of research, different epistemolo-
gies, and different forms of representation. Euch ,York tradition i::. gOY-
16
v ... IVlKI "",1'"''''-'
erned by its own set of genres; each has its O\vn classics, its own pre-
ferred forms of representation, inrerpretarion, trustworthiness, and
textual evaluation (see Becker, 1986, pro 134-135). QUJlitative research-
ers lise ethnographic prose, historical n;lrr;Jtivcs, first-person aCCOUIHS
still photographs, life histories, fictionalized "facts," and biographical and
autobiographical materials, among others. Quantitative researchers use
mathematical models, statistical tables, and graphs, and usually write
about their research in impersonal, third-person prose.
Tensions Within Qualitative Research
It is erroneous to presume thea all qualitative researchers share the same
assumptions about the five points of difference described above. As the
discussion below will reveal, positivist, postpositivist, <lnd poststrucrural
differences define and shape the discourses of qualitative research. Real-
ists and postpositivists within the interpretive qualitative research tr;:!cli-
tion criticize POststructuralists for taking the texrual, narrative turn, These
critics contend that such work is navel gazing. It produces conditions "for
a dialogue of the deaf between itself and the community" (Silverman,
1997, p. 1.40). Those who attempt to capture the point of view of the
interacting subject in the world are accused of naive humanism, of repro-
ducing "a Romantic impulse which elevates the experiential to the level of
the authentic" (Silverman, 1997, p. 248).
Still others argue that lived experience is ignored by those who take the
textual, performance turn. Snow and 1\110rri1l (1995) argue thar "this per-
formance turn, like the preoccupation with discourse and storytelling, will
take us further from the field of social action and the real dramas of every-
day life and thus signal the death knell of ethnography as an empirical'ly
groUJ:dcd enterprise" (p. 361). Of course, \ve disagree.
\X1ltn these differences wirhin and benvecn the t\\'o traditions nell\' in
hand, we must now briefly discuss the history of qualitative rcse3tch. \"X1e
brcak this history into seven historical moments, mindful that anv history is
olwavs
I somew Ilat ar t
1ltrary an d always m I '
east part1311y .
a SOCi~lj construction.
.. The History of Qualitative Research
The. hisro f} 0 f qua j"ltatlve research reveals, as Vlchch
. . .
and Lvman rcmmd
Us 1Il Ch8pter 2 of Volume 1, thal Lhe modern social science clisciplines
J7
 l.,.VLl...L'-..' <l~U rll'v II" '-'" " .... " , ' ..... "<:,-".,~.",, . - . " ','uv .... u>.,.uv"

have taken as their mission "the analysis and understanding of the pat-
terned conduct and socia! processes of society." The notion that this task
could he carried out presupposed that social scientists had the ability to
observe this world objecrively. Qualitative methods \vere a major too! of
such observations.
11
proliferated from the earlr 19005 to the 1.9605 ~md included the \vorlc of
E. Franklin FraZier, Roberr Park, and Roben Redfield and their stuclenr5,
as weli as \X1illiam Foote \Vhyrc, the Lynds, August Hollingshead, Herbert
j
Gans, Stanford Lyman, Arthur Vidic:h, and Joseph Bensm:::m. The POS[-

Throughout the history of qualitative research, investigators have

1960 ctbnicity studies challenged the "melting pot" hypothesis of Park
:lnd his follO\vers and corresponded to the emergence of ethnic studies
j
ahvavs defined their work in terms of hopes and values, "religious faiths, programs that saw' Native Americans, Latinos, Asian Amcrlcms, and j\fri-
occupational and professional ideologies" (Viclich & Lyman, Chapter 2,
Volume I). Qualitative resc3rch (like all research) has always been judged
can Americans attempting co take control over the study of their own
peoples.
j
on the "standard of whethcr the work communicates or "says' something The postmodern and poststrucrurai challenge ernerged in the mid~
to us" (Vidich LI...[ Lyman, Chapter 2, Volume 1), based on how we concep-
tualize our reality and our images of the \vorld. Epistemology is the word
19805. It questioned the assumptions that had organized this enrlicr his~
rory in each of irs colonializing moments. Qualttati-ve research that crosses
j
that has historically defined these standards of evaluation. [n the contcm- the "'postl11odern divide" requires one, Vidich Jnd Lyman :lfgue in Volume
1, Chapter 2, to "'abandon aU established and preconceived values, theo~
porary period, as we have argued above, many received discourses on epis-
temology afC no\\! being reevaluated. ries, perspectives ... and prejudices as resources for erhnographic study." j
Vidich and Lyman's history covers the following (somewh<lt) overlap- In this ne\\, era, the qualitative researcher does more th:m observc history;
ping stages: early ethnography (to the 17th century); colonial ethnogra- he or she plays a part in it. Nc\v tales from the field will now be written,
phy (17tb-, 18th- 1 and 19th-century explorers); the ethnography of the and they \vill reflect the researcher's direct and personal eng3.gemem with j
American Indian as "other" (larc-19th- and carly-20th~ccJ1turyanthropol- this historical period.
ogy); the ethnography of the "civic other," or community studies, and Vidich and Lyman's analysis covers rhe fnll sweep of ethnographic his~
ethnographies of American immigrants (carly 10th cemur)' through the tory. Ours is confined to the 20th century and complements mJny of their j
19605); studies of tchnicity and assimilation (midcentury through the divisions. \YJe begin with the eady foundational work of the British and
19805); and the present, which ,ve call the seventh 1Tloment. French as well the Chicago, Columbia, Harvard, Berkeley, dnd British
In each of these eras, researchers were and have been influenced by schools of liOciology and amhropotogy. This early foundational period j
their political hopes Jnd ideologies, discovering findings in their research established [he norms of classical qualitative and erhnographic research
that confirmed prior theories or bel ids. Early ethnographers confirmed (see Gupta & Ferguson, 1997; Rosaldo, 1989; Stocking, 1989).
the racial and cultural diversity of peoples throughout the globe and j
attempted to fit this diversity into a theary about the origins of history, the
races, and civilizarions. Colonial ethnographers, before the profcssional- .... The Seven Moments of Qualitative Research
ization of ed1l1ography in the 20th century, fostered a colonial pluralism
that left natives on their own as long as their leaders could be co-opted by As,SUGgested
b "',0V
... l e, our'Illsrory
' 0 f qua I"lwnve researe I1lf1
'N orth Amenca
' In
,
j
[he colonial administration. thIS century divides into seven phases, each of which we describe in turn
below,
European ethnographers studied Africans, Asians, and other Third
\World peoples of color. Early American erhnographers studied the Ameri~
j
can indian from the perspective of the conqueror, who saw the life vvorld The Traditional Period
of the primitive as a windo.. N. to the prehistoric past. The Calvinist mission
to save the Indian W;:1S soon transferred to the mission of saving the , We call the first [1'10 '
ment t !le tra d"ltlona I PCflO(
' ! (t I115
' covers VIC, I'lC11 j
"hordes" of immigrants who entered the United SW[CS v'/ith the beginnings and Lvman's
.. seconu-J ancI Iih'lrd phases). Ie begms , in the carll' 19005 and
COntinues un-it
~
\11/ ld"
of industriaiizarion. Qualitative community studies of the ethnic other \ or \\1ar II. In this period, qualitative researchers wrote
j
18 19

j
j
 '-'-'<"L.'-'-'"''-'''''~.'''''-''' ,.~, .. , - "'~ .. _ ..
"objective," colonializing accounts of field experiences that were reflec-
tive of the positivist scientist paradigm. They were concerned with offer-
ing valid, reliable, and objective interpretations in their \vritings. The
"other" who was swelied was alien, foreign, and strange.
Here is lvialinowski (1967) discussing his field experiences in New
Guinea and the Trobriand Islands in the years 1914~1915 and 1917-
1918. He is bartering his way into field data:
Nothing whatever draws me ro ethnographic studies.... On the whole the
village struck me rather unfavorably. There is;l certain disorganization ...
the rowdiness and persistence of the people who laugh and SGlre and lie dis-
couraged me somewhat. ... Went ro rhe village hoping ro photograph a few
stages of the bara dance. I h<lnded out half-Slicks o( tobacco, then w3rched a
few d3!1ces; lhen took picrures-bur results were poor.... they would not
pose long enough for lime exposures. At moments I was furious ar them,
particularly beC;l\lSe afterl gave them their portions of tobacco they all went
away. (quoted in Geertz, 1988, pp. 73-74)
fn another work, this lonely, frustrated, isolated field-worker describes
his methods in the following words:
1n the field one has to face;] chaos of faets .... in this crude form they ate not
scientific facts at all; they arc absolutely elusive, and Gln only be fixed by
interpretation.... Ollly laws ilIul gellerali:::.atiolls ,Ire scieJltific (acts, and
field work consists only and exclusively in the interpretation of the chaotic
SOCial rea lit)', in subordinating it to general rules. (Malinowski, 1916/1948,
p. 328; quoted in Geertz, 1988, p. 81)
lvlalinowski's remarks are provocative. On the one hand they disparage
fieldwork, bur on the other they speak of itwithin the glorified language of
science, with laws and generalizations fashioned om of this selfsame
experience.
The field-worker during this period was lionized, made into a larger~
than-life figure who went iuro and then returned from the field with sto-
ries about strange people. Rosaldo (1989, p. 30) describes this as the peri-
od of the Lone Ethnographer, the story of the man-scientist who went off
in search of his native in <1 distant land. There this figure "encountered the
object of his quest ... {and] underwent his rite of passage by enduring the
ultimare ordeal of 'field\vork'" (p. 30). Returning home with his dara, the
Lone Ethnographer wrote up an objective account of the culture studied.
20
These accounts were structured by the norms of classical ethnography.
This sacred bundle of terms (Rosaldo, 1989, p. 31) organized ethno-
graphic texts in terms of four beliefs and commitments: a commitment to
objectivism, a complicity with imperialism, a belief in monumentalism
(the ethnography would create a museum like picture of the culture stud-
ied), and a belief in timelessness (what was studied would ncver change).
The other was an "object" to be archived. This model of the researcher,
who could also write complex, dense theorics abour what was studied,
holds co the present day.
The myth of the Lone Ethnographer depicts the birth of classic ethnog-
raphy. The tcxrs of .Malinowski, Radcliffe-Brown, Ivlargaret lVlead, and
Gregory Bateson are still carefully studied for what they can tell the novice
about conducting fieldwork, raking field notes, and writing theory. Today
this image has been shattered. The works of the classic ethnographers arc
seen by many as relics from the colonial past (Rosaldo, 1989, p. 44).
Although many feel nostalgia for this past, others celebrate its passing.
Rosaldo (1989) quotes Cora Du Bois, a retired Harvard anthropology
professor, who lamented this passing at a conference in 1980, reflecting
on the crisis in amhropology: "[I feel a distance] from the complexity
and disarray of what I once found a justifiable and challenging discipline..
. . It has been like moving from a distinguished art museum into a garage
sale" (p" 44)"
Du Bois regards the classic ethnographies as pieces of timeless artwork
contained in a inuseulll. She feels uncomfortable in the chaos of the garage
sale. In comrast, Rosaldo (1989) is drawn to this metaphor: "[The garage
sale] provides a precise image of the postcolonial situation where cultural
artifacts flow between unlikely pbces, and nothing is sacred, permanent,
or sealed off. The image of anthropology as a garage sale depicts our pres M
ent global situation" (p. 44). Indeed, many valuable treasures may be
found if one is willing to look long and hard, in unexpected places. Old
:tandards no longer hold. Ethnographies do not produce timeless truths.
,fhe commitment to objectivism is now in doubt. The complicity with
~mperialiS!1l is openly challenged today, and the belief in l110numcntalism
IS a thing of the past,
The legacies of this first period begin at the end of the 19th century,
When the novel ::md the social sciences had become distinguished <1S sepa-
rate( .,' ems 0 f d"lscourse (CJ ough, 1992, pp. 21-22; see also Clough,
sYSt
199H). However, the Chicago school, with its emphasis 011 the life story
ond [he ""I"' ," J"f
s lLe-or- 1 e "' approach to ethnographiC
" "
matenals, sought to
21
 \...VI..-I..-l.:.\...ll'~\J r l l ' ..... ' . ' , 1.-", ,~
develop an lnterpretive methodology that maintained the centrality of the
narrated life history approach. This led to the production of texts that
gave the researcher-as-author the power to represent the subject's story.
\'{7ritten under the mantle of srraightforward, sentiment-free social real-
ism, these texts used the language of ordinary people. They articulated a
social science version of literary naturalism, \"hich often produced the
sympathetic 11lusion that a solution to a social problem had been found.
Like the Depression~era juvenile delinquent and other "social problems"
films (Roffman & Purdy, 1981), these accounts romanticized the subject.
They turned the deviant into a sociological version of a screen hero. These
sociological stories, like their film counterparts, usually had happy end-
ings as the v followed individuals through the three stages of the classic
m~r~lity tale: being in a state of grace, being seduced by evil and falling,
and finally achieving redemption through suffering.
Modernist Phose
The modernist phase, or second moment, builds on the canonical
\vorks from the traditional period. Social realism, naturalism, and slice-of-
life ethnographies 3re still valued. This phase extended through the post-
war vcars to the 1970s and is still present in the work of many (for reviews,
see Wolcott, 1990, 1992, 1995; see also Tedlock, Chapter 6, Volume 2). In
this period many texts sought to formalize qualitative methods (see, for
example, Bogdan & Taylor, 1975; Cicourel, 1964; Filstead, 1970; Glaser
& Strauss, 1967; Lot1and, 1971, 1995; Lot1and & Lofland, 1984, 1995;
Taylor & Bogdan, 1998)." The modernist ethnographer and sociological
participant observer attempted rigorous qualitative studies of important
~ocial processes, including deviance and social control in the classroom
and society. This was a moment of creative ferment.
A ncw generation of graduate students across the human disciplines
encountered nevI-' interpretive theories (ethnomethodology, phenomenol-
ogy, critical theory, feminism). They were drawn to qualitative research
practices that would let them give a voice to society's underc1ass. Postposi-
tivism fUflctioncd ;;\5 u powerful epistemological paradigm. Researchers
attempted ro fit Campbell and Stanley's (1963) model of internal and
external validity to constructionist and interactionist conceptions of the
research act. They returned to the texts of the Chicago school as sources of
inspiration (see Denzin, 1970, 1978).
... ,,, ''-' ..., 'n' ..... """ _.,..
~...,,..... ,_~
A canonical text from this moment remains Boys ill While (Becker et aI.,
1961; see also Becker, 1998). Firmlv entrenched in mid-10th-century
methodological discourse, this work attempted to make qualit3.tlve re-
search JS rigorous as its quantitative counterpart. Causal narratives wcre
central to this project. This rnultirnethod work combined open-cnded and
quasi-structured interviev.,ring \vith participant observation and the careful
analysis of such materials in standardized, statistical form. In a classic arti-
cle, "Problcms of Inference and Proof in Participant Observation,"
Howard S. Becker (1958/1970) describes the usc of quasi-statistics:
Participanr observations have occlsionally been gathered in stalldardized
form capable of being transformed into legitimate statistical data. But the
exigencies of the field usually preveIH the collection of data in such a form
to llleet the assumptions of statistical tests, so that the observer deals in \vbat
h3VC been called "quasi-sr3tistics. " H is conclusions, while imp licitly numc[-
iul, do nor require precise quantification. (p. J l)
In the analysis of (bta, Becker notes, the qualitative researcher takes a
cue fro III statistical colleagues. The researcher looks for probabilities or
suPpOrt for argumel1ts concerning thc likelihood th,1[, or frequency with
which, a conclusion in fact applies in a specific situation (see also Becker,
1998, pp. 166-170). Thus did work in the modernist period clothc itself in
the language and rhcrovic of positivist and postpositivist discourse.
This was the golden age of rigorous qualitative analysis, bracketed in
sociology by Boys ill While (Becker et aI., 1961) at one end and 17Je Dis-
COvery of Grollllcled Theory (Glaser & Strauss, 19(7) at the other. In ed-
ucation, qualitative rescarch in this period was defined by George and
~o.llise .S:in(~ler~ Jules Henr~', ~arry Wolcott, and John Singleton. This
.01 111 ot qualitative research IS snU present in the \vork of such pcrsons as
Strauss and Corbin (1998) and Ryan and Bernard (see Chapter 7, this
volume).
The "golden age" reinforced the picture of qualitative researchers as
cultu'al
. - 10 mantles.
' "1 1ll1ne
I .d Wlt'I1 Promet Ilean IlU111an powers, they valo-
~lzed villains and outsiders as heroes to mainstream society. They embod-
'ed
. ..
a belief in "ll
l
~'
C L-Onnl1gcncy or.. se If. an d '
socIety, ::!nCIIle Id to emanClpatory
'
:",1..::.1-:5 fO,r '\\-'l1icl1 olle lives amj dies." They put ll1 pbce a tr:lBic and oft.:-n
IrOl1lC View .
-" ~. 03.11
of SClClt.:ty , d se If, ancI .lomed
' aI
ong' , cultura I
1111c of leftist
rOlllantlcs that' ,I I I E - ~.
1Vl. ' c , .. ilK tiC ec ~ll1erSOIl, Ivlarx, James, Dewey, (;r<.1111S(1, and
. ,lrtln Luthcr King, .Ir. (\X1est, 1989, chap. 6).
22 23
 LV'-'-'-'-' ,"'U '-" ~'-' ,,' , ,-", " , ",,-, "<:'-'''''''''' ,r" ,
As this moment came to an end, the Vietnam \'(1ar was everywhere pres-
cnt in American society. In 1969, alongside these political currents,
Hcrben Blumer and Everctt Hughes met with a group of young sociolo-
gists called the "'Chicago Irregulars" at the American Sociological Associa-
tion meetings held in San Francisco and shared their memories of the
"Chicago years." Lyn Lofland (1980) describes the 1969 meetings as a
moment of creative ferment-scholarly and political. The San Francisco
meetings witnessed nor simply the Blumer-Hughes event bllt a "counter-
revolution." .. _a group first came to ... talk about the problems of being a
sociologist and a female .... the discipline seemed literally to be bursting
\virh new _ .. ideas: Libelling theory, ethnomethodology, conflict theory,
phenomenology, dr~l.1naturgical analysis. (p. 253)
Thus did the modernist phase come to an end.
Blurred Genres
By the beginning of the third stage (1970-1986), which we call the
moment of blurred genres, qualitative researchers had a full comple-
ment of paradigms, methods, and strategies to employ in their research.
Theories ranged from symbolic inreracrionism to constructivism, natural-
istic inquiry, positivism and postpositivism, phenomenology, ethnometh-
odology, critical theory, neo-lvlarxist theory, semiotics, structuralism,
feminism, ancl various racial/ethnic paradigms. Applied qualitative re-
search ,vas gaining in stature, and the politics and ethics of qualitative
research-implicated as they were in various applications of this work-
were topics of considerable concern. Research strategies and formats for
reporting research ranged from grounded theory to the case study, to
methods of historical, biographical, ethnographic, action, and clinical re-
search. Diverse ways of collecting and analyzing empirical materials were
also available) including qualitative interviewing (open-ended and quasi-
structured) and observational, visual, personal experience, and documen-
tary methods. Computers were entering the situation, to be fully devel-
oped as aids in the analysis of qualitative data in the next decade, along
\vith narrative, content, and semiotic methocls of reading interviews and
culrlltal texts.
Two books by Geenz, The Interpretatioll of Culture (1973) and Local
Knowledge (1983), defined the beginning and end of this moment. In
24
" ,,,. ,~~
"HI UUU<.,UU' I
these tWO works, Geertz argued that the old functional) positivist, behav-
ioral) totalizing approaches to the human disciplines wcre giving way to a
more pluralistic, interpretivc, opcn-ended perspective. This new per-
spective took cultural representations and their meanings as its point of
departure. Calling for "thick descriptions" of particular events, rituals,
and customs, Geertz suggested that all anthropological writings are inter-
pretations of interpretations. U The observer has no privileged voice in the
interpretations that are written. The central task of theory is to make sense
out of a local situation.
Geertz went on to propose that the boundaries between the social sci-
ences and the humanities had become blurted. Social scientists were now
turning to the humanities for models, theories, and methods of analysis
(semiotics, hermeneutics). A form of genre diaspora was occurring: docu-
mentaries that read like fiction (I'vlailer), parables posing as ethnographies
(Castaneda), theoretical treatises that look like travelogues (Levi-Strauss).
At the same time, other new approaches were emerging: postsrructuralism
Barthes), neopositivisl11 (Philips), neo-ivlarxisI1l (Althusser), micro-macro
dcscriptivism (Geertz}) ritual theories of drama and culture (\Z Turner),
deconstructionism (Derrida), ethnomethodology (Garfinkel). The golden
age of the social sciences "vas over, and a new age of blurred, interpretive
genres was upon us. The essay as an an form was replacing the scientific
article. At issue now is the author's presence in tbe interpretive text
(Geerrz, 1988). How can the researcher speak with authority in an age
when there are no longer any firm rules concerning the text, including the
author's place in it, its standards of evalliation, and its subject matter?
The naturalistic, postpositivist, and constructionist paradigms gained
power in this period, especially in education, in the works of Harry
\\lo lcott) Frederick Erickson, Egan Guba, Yvonna Lincoln, Roben Stake,
and Elliot Eisner. By rhe end of the 19705, several qualitative journals
were in place, including Urban Life and Culture (nO\v]oumal of Contcm-
fJ?l'ary Ethnography), Cultural Anthropology, Anthropology alld Educa-
tton Quarterly, Qualitative Sociology, and Symbolic Interactioll, as well as
the book series Studies ill Symbolic Interaction.
Crisis of Representation
A profound rupture occurred in the micl-1980s. \Vhat we call rhe
o,. .. 01~ representation,
fOUrth m Ol11em, or t I1e CflSiS . . Allthropol-
3ppearecl With
ogy as Clfltural Critique (jvlarcus & Fischer, 1986), The AJithropolog}' of
25
 '--VLI.-,-'--' ""U MI 'u 1'" '-,,, ,,,.. ,,,;v ,,!'-IFl'-'
Ext1crieJJce (Turner & Brunerl 19 86), \~Ij'itjllg Culture (Clifford & 1\181'c115 1
1986), WorJ2s mId Lives (Geerrz, 1988), and The Predicament of Culture
(CtifforJ, 1988). These works made research and \-\Tiring more reflexive
and caUed inro quesrion the issues of gender, class, and race. They anicu-
bred the consequences of Geertz's "blurred genres" interpretation of the
field in the early 19805.1'1
New models of truth, method, and representation were sought
(Rosaldo, 1989). The erosion of classic norms in anthropology (objec~
ti\i"m, complicity vvith colonialism, social life structurcd by fixed rituals
and customs, ethnographies as monuments to a culture) \-vas complete
(Rosaido., 1989, pp. 44-45; see ~l1so Jackson, 1998., pp. 7-8), eritical l fem-
inist~ and episremologies of color nmv competed for ;:memion in chis
arena. Issues such as validity., reliability., and objectivity, previously
believed setded, Vi/cre once more problematic. Pattern and interpretive
theories, as opposed to causo.!, linear theories, were now more common l
as writers continued to challenge older models of truth and meaning
(Rosaklo, 1989).
Stoller and Otkcs (1987, pp. 227~229) describe how the crisis of repre~
semation WJS felt in their fieldwork among the Songhay of Niger. Stoller
observes: "\'\lhen I began to write <ll1thropological texts, r followed the
conventions of my training. [ 'gathered data,' and once the 'dara' were
arranged in neat piles) I 'wrote them up.' In one case f reduced Songbay
insults to a series of neat logical formulas" (p. 227). Stoller became dissat-
isfied with this form of writing, in part because he learned that "everyone
had lied to me and ... the data 1 had so painstakingly coUected \verc
\vorthless. I learned a lesson: Informants routinely iie to their anthropolo-
gists" (Stoller & Olkes, 1987, p. 9), This discovery led to a second-thar
he had, in foHO\ving the conventions of ethnographic realism., edited him-
self om of his text. This led StoBer to produce a different type of text, a
memoir) in \vhich he became a central character in the story he told. This
story, an account of his experiences in the SongJuy world, became an
analysis of the dash bCf\veen his wodd and the world of Sanghay sorcery.
Thus Stoller's journey represents an attempt to confront the crisis of rep-
resemation in the fourth rnorncnr.
Clough (1992) eJabofarcs this crisis and criticizes those \\'ho would ar-
gue [hat new forms of writing represent;] \vay ant of the crisis. She argues:
While 1113ny sociologists no\\' commenting on the criticism of ethnogrJ~
ph]' view \vriring as "downright central to the ethnographic enterprise"
26
'n, , ..... ""--,, L..""....
~
, <fvuu .... uu,J
IV811 IVh:men, J 988, p. xi], the problems of writing arc still viewed as differ-
ent from rhe problems of method or fieldwork irself. Thus the solurion llStl-
:lily offered is cxperl11lcJHs in writing, that is :l self~consciollsncss about
j
wrii'ing. (p. 136)
It is rhis insistence on the difference bccween writing and fieldwork that
j
must be analyzed. (Richardson is quite articulate about this issue in Chap-
ter 14 of this volume.)
In wridng l the field-worker makes a cbim to rnot;]l and scienrific
authority. This claim allo\.vs rhe realist and experimental ethnographic
j
texts to function as sources of v~llidatl0n for an empirical science. They
show that the world of real lived experience can still be caprured, if only in
the \Ilriter's memoirs, or fictional experimentations., or dramatic read- j
ings. Bur these works b~ve the danger of directing attention away from the
ways in which the text constructs sexually situated individu315 in J field of
social difference. They also perpetuate "empirical science's hegemony"
(Cluugh) 1992, p. 8), for chese nnv writing technologies of rhe subject
j
become the site "for the production of kno\vtedge/power .... [aligned]
with .. , the capit:ll/stJti.': axis" (Aronowitz, 1988, p. 30G; quoted in
Clough, 1992, p. 8). Such experimcnts come up against, and then back j
away from., rhe difference between empirical science ::lnci social criticism.
Too often they f3!l to engage fully a new politics of texruality that would
"rduse the idenrity of empirical science" (Clough, 1992, p. 135). This
ne\v socia! criticism "would intervene in the relarionship of information
j
economics, nation~statc politics, and technologies of mass communica-
tion, especially in terms of the empirical sciences" (Clough, 1992 p. 16). 1
This) of course, is the terrain occupied by cultural studies.
In rhis series, Richardson (this volume, Chapter 14), Ted/ock (Volume
j
2, Chapter 6), Brady (this volume, Chapter 15), Jnd Ellis and Bochner
(this volume, Chapter 6) develop the above arguments, viewing writing
35;] method of inquiry that moves through sllccessive stages of self- j
ren~~Ion.. A-s a series
- 0 f wntren
- - the
representarIOns, - fleld-l,vorker's texts
v
flo \" from the field experience, through inrermediare works., to later
work, and finally ro the resc::lrch text, which is the publjc presentation of
the ~thnographic and narrative experience. Thus fieldwork and writing
j
bh~r.l11to one another. There is, in the finJI analysis, no difference between
I\Tltl11g -, d r- Id I I -
n Ie wur..:. T 1cse t\vo perspect)ves inform on(' another
j
<
througham every chapter in these volumes, In these v)/ays the crisis of rep-
resc t .
. natIon moves qualitative research in new and critical directions,
27
j
j
 LVLLI:LIII'I'-.:l f-'ll'l'-' 11'11 L",.' " ' - , " , ' - ' ~Vr-\"" Ir-\I"
A Triple Crisis
The ethnographer's authority remains under assault today (Behar,
1995, p, 3; Gupta & Ferguson, 1997, p, 16; Jackson, 1998; Ortner, 1997,
p. 2). A triple crisis of representation, legitimation, and praxis confronts
qualitativc researchers in the human disciplines. Embedded in the dis-
courses of poststructuralism and postmodcrnism (sce Vidich & Lyman,
Volume 1, Chapter 2; and Richardson, Chapter 14, this volume), thesc
three crises are coded in multiple terms, variously called and associated
with the critical, interpretive, linguistic, feminist, and rhetorical turns in
social theory. These new turns make problematic two key assumptions of
qualitative ;esearch. The first is that qualitative researchers can no longer
directly capture lived experience. Such cxperience, it is argued, is created
in the social textwrittcn by the researcher. This is the representational cri-
sis. It confronts the inescapable problem of representation, bur does so
\.Ilitbin a framework that makes The direct link beTween experience and
text problematic.
The second assumption makes problematic the traditional criteria for
evaluating and interpreting qualitative research. This is the legitimation
crisis. It involves a serious rethinking of such terms as ualidity, gell-
eralizability, and reliability, terms already retheorized in posrpositivist
(Hammersley, 1992), constructionist-naturalistic (Guba & Lincoln, 1989,
pp. 163-183), feminist (Olesen, Chaprer 8, Volume 1), interpretive
(Denzin, 1997), poststrucmral (Lather, 1993; Lather & Smithies, 1997),
and critical (Kincheloe & l\t1cLaren, Chapter 10, Volume 1) discourses.
This crisis asks, How are qualitative studies to be evaluated in the contem-
porary, poststrucrural moment? The first rwo crises shape the third, which
asks, Is it possible to effect change in the world if society is only and always
a text? Clearlv these crises intersect and blur, as do the answers to the ques~
tions they g~nerate (see in Volume 1, Schwandt, Chapter 7; Ladson-
Billings, Chapter 9; and in this volume, Smith & Deemer, Chapter 12).
The fifth moment, the postmodern period of experimental ethno-
graphic writing, struggled to make sense of these crises. New ways of com-
posing ethnography were explored (Ellis & Bochner, 1996). Theories
vvcn: rCild U:l cabs from the field. Writer>: struggled with differem ways to
[t.:prcGcnt the "other/, ::dthol1gh they were now joined by new representa-
tional concerns (see Fine et aI., Chapter 4, Volume 1). Epistemologies
from previously silenced groups emerged to offer solutions to these prob-
lems. The concept of the aloof observer has been abandoned. More action,
28
L. ",'" ... , , , , , .....,
IflLfuuucuon
participatory, and activist-oriented research is on the horizon. The search
for grand narratives is being replaced b~1 more local, small-scale theories
fitted to specific problems and particular situations.
The sixth (postcxperil11emal) and seventh (the future) moments are
upon us. Fictional cthnographies, edmographic poetry, and multimedia
texts are today taken for granred. Postexperimental writers seek to con-
nect their writings to the nceds of a free democratic society. The demands
of a moral and sacred qualitL1tive social science are actively being explored
by a host of new writers from many different disciplines (see .Jackson,
1998; Lincoln & Denzin, Chapter 6, Volume 1).
Reading History
\~/e draw four conclusions from this brief history, noring that it is, like
all histories, somewhat arbirrary. First, each of the earlier historical mo-
ments is still operating in the present, either as legacy or as a set of prac-
tices that researchers continue to Follow or argue against. The multiple
and fractured histories of qual itati ve research nmv make it possible for any
given researcher to attach a project to a canonical text from any of the
above-described historical moments. ivlulripic criteria of evaluation COI11-
pete for attel1tion in this field (Lincoln, ill press). Second, an embarrass-
ment of choiccs now characterizes the field of qualitative research. There
have never been so many paradigms, strategies of inquiry, or methods
of analysis for researchers to draw upon and utilize. Third, we are in a
:nomen[ of discovery and rediscovery, as new ways of looking, imerpret-
1~1g, arguing, and writing arc debated and discussed. Fourth, the qualita~
t:ve research act can no longer be viewed from within a neutral or objec-
nve positivist perspective. Class, race, gender, and ethnicity shape the
process of inquiry, making research a multicultural process. It is to this
topic that we now turn.
.. Qualitative Research as Process
Three inter~c d - . . ,. d C_lJ1e
L 1l111ecte , gcncnc acrtvltles
f' t 1le qua IItatlve
' research
pro<';<;;;jij, They go by a. vJricry of' tliLfcn::n[ !abc1;)j incluLlillt, tflfiUI]'J m",'rho:.x!J
HihilY5is
. ' ant a Iogy, epIstemology,
. and methodology. Hehind these terms
"ands. [be pers,ona1,.11l0grapl1Y
. 'f h i
0 L e rescare ler, \\'ho speaks tram a partic-
,
ular class" "d , . ' . I . ,
ben er, LJ.CJaI, cultura , and ethnIC communIty perspective. The .
29
 COLLECTII-lG AHD IN I tl<f-'I<t Ilr~G \.lUAU IAllVe IVIAI el\l ...... w InUOOUCHon

(tendered , multiculturally situated researcher approaches the world with a

b this search for a method has leel to a perennial focLls in the human dis~
set of ideas, a framework (theory, ontology) that specifies a set of questions
(epistemology) that he or she then examines in specific \\lays (met~lOdology,
analysis). That is, the researcher collects empirical materials bearIng on the
question and then analyzes and writes about them. Every researcher speaks
from within a distinct interpretive community that configures, in its special
way, the multicultural, gendered components of the research act.
In this volume \-\re ([eat these generic activities under five headings, or
phases: the researcher and the researched as multicultural subjects, major
paradigms and interpretive perspectives, research strategies, methods of
~ollecting and analyzing empirical materials, and the art, practices, and
politics of interpretation. Behind and within each of rhese phases stands
the biographically situated researcher. This individual enters the research
process from inside an interpretive community. This community has its
~\vn historical research tradirions, which constitute a distinct point of
view. This perspective leads the researcher to adopt particular vievvs of
the "other" who is studied. At the same time, the politics and the ethics of
research must also be considered, for these concerns permeate every phase
of the research process.
<> The Other as Research Subject
ciplines on qualitative, interpretive methods.
Recently, as noted above, this position and its beliefs have come under
assault. Poststructuralists and postmodernists have contributed to the
understanding that there is no clear window into the inner life of an indi-
vicll131. Any gaze is always filtered through the lenses of language, gencler,
social class, race, and cthnicity. There are no objective observarions, only
observations socially situated in the worlds of-and between-the ob-
server and the observed. Subjects, or individuals, arc seldom able to give
full explanations of rheir actions or intentions; all they can offer are
accounts, or stories, about what they did and why. No single method can
grasp all of the subrle variations in ongoing human experience. Conse-
quently, qualitative researchers deploy a wide range of interconnected
interpretive methods, ahvays seeking better ways ro make more under-
standable the worlds of experience they have swdied.
Table L 1 depicts the relationships \ve see among the five phases that
define the research process. Behind all but one of these phases stands rhe
biographicaHy situated researcher. These five levels of activity, or practice,
work their way through the biography of the researcher. \X'c rake them up
briefly in order here; we discuss these phases more fully in the introduc-
tions to dle individual parts of this volume.

Since its early-20th-century birth in modern, interpretive form, qualitative

Phose 1: The Researcher
research has been haunted by a double-faced ghost. On the one hand, quali-
tative researchers have assumed that qualified, competent observers can,
with objectivity, clarity, and precision, report on rheir own observations of Our remarks above indicate the depth and complexity of the traditional
the social world, including the experiences of others. Second, researchers and applied qualitative research perspectives into which a sociallv situated
have held to the belief in a real subject, or real individual, who is present researcher emers. These traditions locate the researcher in history, si111ul~
in the world and able, in some form, to rcport on his or her experiences. taneoilsly guiding and constraining work that will be done in any specific
So armed, researchers could blend their own observations with the self- St,udy. This field has been characterized consrantly by diversity and con-
reports provided by subjects through interviews and life story, personal flict , and the s'e are Its
. most en d ' tra d'ltions
unng . (see Grreenwoo d & I eVlll
.
experience, case study, and other documents. Chapter 3, Volume 1). As a carrier of this complex and conrradictor-y his~
These t\VO beliefs have led qualitative researchers across disciplines to tory, the researcher mUst also confronr the ethics and politics of research
seek a method that would allO\v them to record accurately their own (see Chris'il ",l1S,
Cl1apter.),
- "vo_wne
I 'I . .
1). l1C age of value-free lllqUiry for
observ~ltion~ \vhile nko llncovering the rrIcanings their subjects bring to ~h=an d
J
l 'lS over (sec .III Volume .
15Cl11 1l1e>: 1, "
VldlCh & Lyman, Chapter
their life experiences. This method would rely upon the subjective verbal -jandFineetal
. . <.., CIlapler.,..
' I)"wc1ay researchers struggle to develop sltua-
.
tlol1alandt"'_,,~ . I .' .
and written expressions of meaning given by the individuals studied a5 , ra!JSSiLU3tlona etlucs that apply to all torms of the research act
windlJws into the inner lives of these persons. Since Dilthey (1900/1976), and Its hum' . 1 . . '
an-to- luman relatlOl1shlpS.

30 31
 LULU:L liNG I\NU.!I'l I t.Kr'Kt.lll'll.:J \,JUALlIAI IVC !VIAl CI";'IP.L.:l IntroaUCflon

"rhe I<.escarch Process Phose 2: Interpretive Paradigms

TABLE 1.1

All qualitative researchers are philosophers in that "universal sense in

Phose 1: The Researcher as a Multicultural Subject
which all human bcings .. _ are guided by highly abstract principles"
history and research traditions
(Bateson, 1972, p. 320). These principles combine beliefs about ontology
conceptions of self and the other
ethics and politics of research (\'ZIhat kind of being is the human being? \Vhat is the nature of reality?),
PhtlSe 2: Theoretical Paradigms and Perspectives epistemology (\XThat is the relationship between the inquirer and the
positivism, Pos\positivism known?), and mcthodology (How do we know the world, or gain knowl-
interpretivism, constructivism, hermeneutics edge of it') (sec Gubo, 1991J, p. 18; Lincoln & Gubo, 1985, pp. 14-15; sec
feminism(sl also Lincoln & Guba, Chapter 6, Volume 1). These beliefs shape how the
raciCilized discourses qualitative researcher sees the world and acts in it. The researcher is
critiCCII theory and Marxist models
"bound within a net of epistemological and ontological premises which-
cultural studies models
regardless of ultimate truth or falsity-become partially self-validating"
queer theory
Phose 3: Research Strategies
(Bateson, 1972, p. 314).
The net that contains the researchet's epistemological, ontological, and
study design
case study methodological premises may be termed a paradigm, or an interpretive
ethnography, participant observation, performance ethnography ftamework, a "basic set of beliefs that guides action" (Guba, 1990, p. 17).
phenomenology, ethnomethodology All research is interpretive; it is guided by a set of beliefs and feelings abom
grounded theory the \vorld and how it should be understood and studied. Some beliefs may
life history, testimonio
be taken for granted, invisible, only assumed, whereas others are highly
historical method
action and applied research problematic and controversial. Each interpretive paradigm makes particu-
clinical research lar demands on the researcher, including the questions he or she asks and
Phose .1". Methods of Collection and Analysis the interprcrations the researcher brings to them.
interviewing At the most general level, four major interpretive paradigms structure
observing qualitative research: positivist ;md postpositivist, constructivist-interpre-
artifacts, documents, and records tive, critical (l\llarxist, emancip,=1tory), and feminist-poststructural. These
visual methods four abstract paradigms become more complicated at the level of concrete
uutoethnogropllY
data management methods
specific interpretive communities. At this level it is possible to identify not
computer~assistedanalysis only the constructivist, bur also multiple versions of fe.minism (Afrocentric
textual analysis ~nd POStstrucnrral)l:i as weI! as specific ethnic, .Nlarxist, and cultural stud-
focus groups Ies paradigms. These perspectives, or paradigms, are examined in Part II of
applied ethnography Volume 1.
Phase 5: The Art, Practices, and Politics of Interpretation and Presentation
. The paradigms examined in Parr 11 of Volume 1 work against and along-
criteria for judging adequacy 5HJC (and some within) thc positivist and postpositivist models. They all
practices and politics of interpretation
\:'ork within relativist ontologies (mu lriple constructed realities), inrerpre-
writing as interpretation
nvc tpisrenl0Iop;;es (rIle k.n~n:"er ",nd Lnv~..n ;nrcr",ct al'<.1 :;harc Qnc
policy analv~;"
evaluation traditions another), and interpretive, naturalistic methods.
applied research . Table 1.2 presenrs these paradigms and their assumptions, includ-
lena for eva
IIlg' their cri-' . I
uatli1g . I f
[cseare I1, ancI t I1C typlcaorm that an

32 33
 InrroaUC[lon
COLLECTING AND INTI:I{I-'I<1: liNG \JUALItAIIVt. MAl t.KIA~

TABLE 1.2 Interpretive Paradigms
Paradigml
Form of Theory
Theory Criteria
Positivist! internal, external logical~deductive,
postpositivist validitv grounded
trustworthiness, credi~ substantive-formal
Cons'iructivio;t
bility, transferability,
confirmebili!y
critical, standpoint
Feminist Afrocentric, lived
experience, dialogue,
coring, accountability,
race, doss, gender,
reflexivity, praxis,
emotion, concrete
grounding
Ethnic Afrocentric, lived
experience, diCllogue,
caring, accountability,
race, doss, gender
emancipatory theory,
Marxist
falsifiable, dialogical,
race, doss, gender
Cultural studies cultural practices,
praxis, social te:<ts,
subjectivities
Queer theory reflexivitv,
deconstruction
Frow and Morris (Chapter 11), and Gamson (Chapter 11). We have dis-
cussed the positivist and postpositivist paradigms above. They work from
within a realist and critical realist ontology and objective epistemologies,
Type of Narration and rely upon experimental, quasi-experimental, survey, and rigorously
defined qualitative methodologies. Ryan and Bernard (Chapter 7, rhis
scientific report volume) develop elements of this paradigm.
The constructivist paradigm assumes a relativist ontology (there are
multiple realities), ;] subjectivist epistemology (knower and respondent
interpretive cocreare understandings), and a naturalistic (in the natural world) set of
case studies,
ethnographic fiction
methodological procedures. Findings arc usually presented in terms of the
criteria of grounded theory or pattern theories (see in Volume 1, Lincoln
essays, stories, & Cuba, Chapter 6; in Volume 2, Charmaz, Chapter 8; and in this volume,
experimental writing Ryan & Bernard, ClMpter 7). Terms such as credibility, tmnsferabilit)',
depeildability, and cOllfirl71ability replace the usual positivist criteria of
internal and external validity, reliability, and objectivity.
Feminist, ethnic, IVlarxist, and cultural studies and queer theory models
privilege a materialist-realist onrology; that is, the real world makes a
materia! difference in terms of race, class, and gender. Subjectivist episte-
standpoint, essays, fables, mologies and naturalistic methodologies (usually ethnographies) are also
critical, historical dramas employed. Empirical materials and theoretical arguments are evaluated in
terms of their emancipatory implications. Criteria from gender and racial
communities (e.g., African American) may be applied (emotionality and
feeling, caring, personal accountability, dialogue).
critical, historical, historical,
Poststrucrural feminist theories emphasize problems with the social
economic economic,
sociocultural text, its logic, and its inability ever to represent the world of lived experi-
analyses ence fully. Positivist and postpositivist criteria of evaluation are replaced
by other terms, including the reflexive, multi voiced text thar is grounded
social criticism cultural theory in the experiences of oppressed people.
as criticism . The cultural studies and queer theory paradigms are multi focused, with
many different strands drawing from Marxism, feminism, and the post-
modern sensibility (see in Volume 1, Prow & Morris, Chapter 11;
theory as crilieism,
social criticism, his-
autobiography
~alTIson, Chapter 12; and in this volume, Richardson, Chapter 14). There
torical analysis
IS a t~nsion between a humanistic cultural studies, which stresses lived

ex~efIences (meaning), and a more structural cultural studies project,

whtch stresses the Structural and material determinants (race, class, gen-
interpretive or theoretical statement assumes in each paradigm. 16 These der) and effects of experience. Of course, there are tWO side,.; to cYl:ry coin,
paradigms arc explored in considerable detail in Volume 1, Part II by iUld both sides ate needed and arc indeed critical. The cultural studies and

Lincoln and Gulla (Chapter 6), Schwandt (Chapter 7), Olesen (Chapter ?ue er theory paradigms use methods strategically-that is, as resources
8), Ladson-Billings (Chapter 9), Kincheloe and Iv1cLaren (Chapter 10), Or understanding and for producing resistances to local structures of

34 35
 LULLt:I.III'I\.:l AI'lU II' J t:t-:'YI-:.C. t 1t'l'U \dUALI Il-\I 1V r. IV\l-\/ CI'.'l-\LJ
InrrOauetlon

domination. Scholars may do close textual rC:J.dings and discourse analyses complex literature, anel each has a separate history, exemplary works, and
of cultural tcxts (see in Volume 1, Olesen, Chapter 8; Frow & 1\'lorris, preferred ways {or putting the strategy into motion.
Chanter 11; and Silverman, this volume, Chapter 9) as \vell as conducting
locai ethnographies, open-ended imcrvie\ving, and participant observa- Phose 4: Methods of Collecting
tiCH\. The focus is on how race, class, and gender are produced and enacted and Analyzing Empirical Materials
in historically specific situations.
Paradigm and personal history in hand, focused on a concrete empirical The researcher has several methods for collecting empirical materials. J7
problem [Q examinc, the researcher now moves to the next stage of the re- These methods are taken up in Part J of this volume. They range from the
search process-namely, working with a specific strategy of inquiry. interview to direct observation, the analysis of artifacts, documems, and
cultural records, and the use of visual materials or personal experience.
Phose 3: Strategies of The researcher may also use a v,lriety of different methods of reading and
Inquiry and Interpretive Paradigms analyzing inrerviews or cultllral texts, including contcnt, narrative, and
semiotic strategies. Faced with large amoums of qualitative materials, the
"Elble 1.1 presents somc of the major strategies of inquiry a researcher investigator seeks \vays of managing and interpreting these documents,
rnav use. Phase 3 beg-ins with research design, which, bro:ldly conceived, and here data managemem methods and computer-assisted models of
inv~)lves a clear focu: on the research question, the purposes of the stud}\ analysis may be of usc. Ryan and Bernard (this volume, Chapter 7) and
'''vhat information most appropriately will answcr specific research ques- Weitzman (this volume, Chaptcr 8) discuss these techniques.
tions, and which strategies arc most effective for obtaining it" (LeCompte
8-:: Preissle, 1993, p. 30; see also in Volume l,]anesick, Chapter 2; Cheek, Phase 5: The Art and Politics
Chapter 3). A research design describes a flexible set of guidelines that of Interpretation and Evaluation
connect theoretical paradigms first to srr3tegics of inquiry and second to
methods for collecting empirical l113tcrial. A research design situates Qualitative research is endlessly creative and interpretive. The re-
researchers in the empirical \vorld and connects them to specific sites, per- searcher does nor just leave the field with lllounrains of empirical materials
sons, groups, institutions, and bodies of relevant interpretive material, and then easily write up his or her findings. Qualitative imerpretations are
including documents and archives. A rescarch design also specifies how constructed. The researcher first creates a field tcXt consisting of field
the investigator will address the [WO crirical issues of representation and nOtes and documents from the field, what Roger Sanjek (1990, p. 386)
legitimation. calls "indexing" and David PLuh (1990, p. 374) calls "filework." The
A strategy of inquiry comprises a bundle of skills, assumptions, and ~vriter-asiI1terpretermoves from this text to a research text: notes and
practices that the researcher employs as he or she moves from paradigm to mterpretations based on the field text. This text is then re-created a5 a
the empirical \vorld. Strategies of inquiry put paradigms of interpret:ltion Working interpretive documem that contains the writer's initial attempts
into motion. At the same time, strategies of inquiry also connect the re- to make sense of what he or she has learned. Finally the writer produces
searcher to specific methods of collecting and analyzing empirical materi- the public text that comes to the reader. This final tale from the field may
als. For example, the case study relies on interviewing, observing, and doc- assume several forms: confessional, realist, impressionistic, critical, for-
ument analysis. Research strategics implement and anchor paradigms in mal, literary, analytic, grounded theory, and so on (see Van lvlaanen
specific empirical sites, or in specific methodological practices, such as 1988)" " ,
l~"1;;tl.;;ino""1 <': .."1:;<': ;::.n obj<.:<.:r of :;ruJ)"_ Thc:;c so".... tcSi",,, il,duJ", ti,,,, <;'-''''''' :;tu,Jy , . :I~c ;nkrpccrivc pri.l.Lticc 01 rnali:illS sense vf ont:':511mling:5 is bor11 ar-
phenomenological and erhnomerhodological techniques, and the use of tiStiC and
- I Ica I. d'[ II Itip
pol"t" " " for eva Iuat1l1g
"I e Crl[efla " qua I""
ltatlve researc h
grounded theory, as well as biographical, autoethnographic, historical, no\\, exist, and those that we emphasize stress the situated, relational,
action, and clinical methods. Each of these strategies is connected to a and textLnl s"r
l llctures 0 I"' t Ile et I1flograp IlIe
L -
" cxpenence_
" TIlere IS
" no S1l1g
"Ie

36 37
 LUlU:L! !NG 1"I'lU It'll t.1<'F'I\t.! l!'lll \.dUALlIAllVt MAl t:KIAL:)
interpretive truth. As \ve argued earlier, there are multipie inrerpretive
communities, each with its own crireria for evaluating an interpretation.
Program evaluation is a major site of qualiurive research, and qualita-
tive researchers can influence social policy in imponant \vays. The contri-
butions by Greenwood and Levin (Volume 1, Chapter 3), Kemmis and
Mctaggart (Volume 2, Chaprer 11), Miller and Crabrree (Vo!ume2, Chap-
ter 12), Chambers (this volume, Chapter 11), Greene (this volume, Chap-
ter 16), and Rist (this volume, Chapter 17) trace and discuss the rich his-
tory of applied qualitative research in the social sciences. This is the
critical site where theory, method, praxis, action, anel policy all come
together. Qualitative researchers can isolate target populations, show
the immediate effects of certain programs on such groups, and isolate the
constraints thm operate against policy changes in such settings. Acrion-
oriented and clinically oriented qualitative researchcrs can also create
spaces for those who are studied (the other) to speak. The evaluator be-
comes the conduit through which such voices call be heard. Clumbers,
Greene, and Rist ex:p!icidy develop these topics in their chapters.
Bridging the Historical Moments: What Comes ~ley.t?
Ellis and Bochner (this volume, Chapter 6), Gergen and Gergen (Volume 1,
Chapter 13), and Richardson (this volume, Chapter 14) argue that we are
already in the post "POSt" period-post-poststructuralist, post-postmod-
ernist, post-postexperimenral. \'{That this means for interpretive ethno-
graphic practices is still not clear, but it is certain that things will never
again bc the samc. \'{Te are in a new age where mcssy, uncertain, multivoiced
texts, cultural criticism, and new experimcmal works will become more
common, as will more reflexive forms of fieldwork, analysis, and inter-
textual representation. \'{/e take as the subject of our final essay in this vol-
ume these fifth, sixth, and seventh moments. It is true that, as the poet said,
the center no longer holds. We can reflect on what should be at the ne\v
center.
Thus we come full circle. Returning to our bridge metaphor, the chap-
ters that follow take the researcher back and forth through every phase of
the research act. Like a good bridge, the chapters provide for two-\vay
traffic, coming and going between moments, formations, and interpretive
communities. Each chapter examines the relevant histories, controversies,
and current practices that are associated with each paradigm, strategy, and
38
introductiOn
method. Each chapter also offers projections for the fucure, where a spe-
cific paradigm, strategYl or mcthod \vill bc 10 years from now, decp into
the formative years of rhe 21st ccntury.
In reading the chapters that follow, it is imporram to remember that the
field of qualitative research is defined by a series of tensions, contraclic~
tions, and hesitations. This tension works back and forth between the
broad, doubting postmoclern sensibility and the more certain, more tradi-
tional positivist, posrposirivist, and naturalistic conceptions of rhis proj-
ect. All of rhe chapters that foHow are caught in and articulate this tension.
Notes
1. Qu;llitarive research hilS separate and distinguished histories in education, social
work, communicltiom, psychology, history, org;ll1izational studies, medical science,
anthropology, and sociology.
2. Some definitions are in ordcr here. l'osili[!ISIJI asserts that objective aCCOUiHS of the
real world can be given. l'ostpositiuislI1 holds that only parrially objective accounts of the
world can be produced, because all methods for examining rhem are flawed. According to
j!}1/lJd,1tiollil!iSIII, we can have an ultim;He grounding for our knowledge claims about the
world, and rhis involves the use of empiricist ;mel positivist epistemologies (Schw;lodt,
1997a, p. 103). NOIl(ol/lldaliolla!islJl holds that we can make staterneiHS abour rhe world
wirhout "recourse to ultimate proof or foundations for that knowing" (p. 102). Quasi-
(1JII1ldatiOllil!iSIII holds that certain knowledge claims can be made about rhe world based on
lleoreaJist crireria, including the correspondence concept of truth; rhere is an independem
reality that Girl be m;lpped (see Smith & Deemer, Chapter 12, this volume).
3. JllTteson (1991, pp. J--i) reminds tiS that any periudizarion hypothesis is always sus-
pect, cven one rhat rejects linear, stagelike models. Ir is nevet clear to whal' reality a stage
nJers, and what divides one stage from another is always debat<lble. Our seven moments are
!lleall[ to mark discernible shifts in style, gcnre, epistemology, ethics, polirics, and aestherics.
4. Somc fUrther definitions arc in order. Structuralism holds rhat an\' svsrem is made
llP o.r a set of opposition";).l C"Jtegories embedded in language. Semiotics is th~ s~ience of signs
or sign systcms_a structuralist project. According co {JoststmctllmlislJJ, languagc is an
lWitabl e system of referents, thus it is impossible ever ro capture completely the meaning of
a~ action, text, or imenriotl. Postlllodcmism is a contemporary sensibility, developing since
World War Il
<. ,
t'Jat prtVI..,eges no srng
. Ie aut I
mer h 0 d , or para dIgm. HerlllclJellflCS
10flf)',
. IS
. an
approach to the analysis of texts that stresses how prior understandings and prejudices
;;hape the i n r - process. PIJCIIOll1el1O I ' ' . .
I erprC!lve 01,')' IS;t complex system of Ideas aSSOCIated with
t 1e.works of HusserJ, I-lciJegger, Sanre, i'viericaU-POlltV, am! Alfred Schurz. Cllitural swd-
l/!SI~ a complex, imcrdbciplillary field r!tu merge.': c;'iticn! theory, feminism, ~nd post-
structuralism .
.5. Of course, all scnings ;lre tlarut;Jl~that is, places whete everyday experiences rake
place. Qualitative researchets study people doing things together ill rhe places where these
39
 LULLl::LIII'1u AI'lU 1I'11l:IH.. t~l: 111'IU ,<UMLllf-'lIIVl: IVIHI t::r<.lf-'lLJ
things are done (Becker, 1986). There is no field site or natural place where one goes to do
rhis kind of work (see also Gnpta & Ferguson, 1997, p. B). The site is constituted through
the researcher's interpretive prJctices. l-lisroricallr, analysts have distinguished between
experilTlelHal (laboratory) :md field (natural) reseJrch settings, hence the argument that
qtulitative research is nanlralistic. Activity thenry erases this distinction (Keller I..\:: Keller,
1996, p. 20; Vygotsky, 197b').
6. According to Weinstein and Weinstein (1991), "The meaning of bricoleur in French
popular speech is 'someone who works with his (Ot her) hands and uses de\'ious means com-
pared !<) those !If the craftsman.' ... the brienlel/r is practical and gets the job done" (p. 1(1).
These authors provide a history of the term, connecting it to the works of the German soci-
ologist and social theorist Georg Simrliel and, by implication, Ib.udelaire. Hammersley (in
prt:ss) disputes our me of this t"nll. Following L~viStratls5, he reads the bricoleut as a
lllythrt1~lker. He suggeSts the term he replaced \vith the notion of the bOiltbuilder.
H:lrIllllers1cy also quarrels with our "lllomenrs" model of qualitative research, contending
that it implies some sense of progress.
7. Brian De Palma reproduced this baby carriage scene in his 1987 film The
UlltOl/clwbles.
8. In the harbor, the muzzles uf the Potellll.. . ill'S twO huge gUllS swing slowly toward the
camera. \\'ords unscrccn inform us, "The brutal military power ans\vered by gUllS of the bat-
tleship." A final famous three-shot lllolltage sequence shows first a sculptured sleeping lion,
then a lion rising from hiss!et:p, and finally the lion roaring, symbolizing the rJge of the Rus-
sian people (Cook, 19[\ I, p. 1(7). In this sequence Eisenstein uses montage to expand time,
crt:ating a psychological duration for this horrible event. I3y drawing out this sequence, by
showing the baby in the carriage, the soldiers firing On the citizens, the blood on the
mother's glove, the descending carriage on the steps, he suggests a levi.'! of destruction of
great magnitude.
.9. Here it is rdevant to make a distinction between techniques thar are used across dis-
ciplines and med1Uds that are used Within disciplines. Ethnomethodologists, for example,
employ their approach as a method, whereas others selectively borrow that method 3S:l
technique for their own applications. Harry \'i/olcott (personal communication, 1993) sug-
gests this distinctiOn. It is also relevant to make distinCtions among topic, method, and
resource. Methods can be studied as topics uf inquiry; that is how a case study gets done.ln
this ironic, ethnomethodological setl5e, method is both a resource and a topic of inquiry.
to. Indeed, any attempt to give an essential definition of quaIitati\'e research requires a
qualit3tive ;lll;llysis of' the circumstances rh3t produce such a definition.
11. In this sense all research is qualitative, because "the observer is at the center of the
l-cseatch process" (Vidich & Lyman, Chapter 2, Volumt: 1).
12. Sec Lincoln and Guba (1985) for an extension and elaboration of tbis tradition in
the mid-19HOs, and for mort: recem e:qensions see Taylor and Bogdan (19.9S) and Creswell
(1997).
lJ. Greenblatt (1997, pp. 15-1S) offets a useful deconstructivL' reading of the many
meanings ;md praCtices Ceert2. brillg" to the term thicl::. descriptiO/I-
lA. These works marginalized and minimized the comributiot1s uf standpoint femini:;t
theory and research to this discourse (see Behat, 1995, p. 3; Gordon, 1995, p. 432).
is. Olesen (Chapter 8, Volume 1) identifies three strands of feminist research: main-
stream empirical, standpoint and cultural studies, and poststructllral, postmoJern. She
40
places Afrocentric and Other modds of color under the cultural studies and postlllodern
categories.
16. These, of course, are our intt:rprctations of these paradigms and interpretive styles.
17. Empirical lIl1teri<1ls is the preterred term for what are traditionally desctibed as
data.
.. References
Aronowitz, S. (1988). Science as pOlucr: Discourse and ideology illll10dem society.
lVlinneapolis: University of IVlinnesota Press.
Bateson, G. (1972). Steps to t.1I1 ecology of mind. New York: Ballantine.
Becker, H. S. (1970). Problems of inference and proof in participant observation. In
H. S. Becker, Sociological Il!orl~: Method and substance. Chicago: Aldine.
(Reprinted from American Sociological Review, 1958,23,652-660)
Becker, I-I. S. (1986). Doing things together. Evanston: Northwestern University
Press.
Becker, H. S. (1996). The epistemology of qualitative research. In R. Jessor,
A. Colby, & R. A. Shweder (Eds.), Etlmogra/Jhy alld hI/malI development:
Call text and meaning in social illquiry (pp. 53-71). Chicago: University of
Chicago Press.
Becker, H. S. (1998). Tiicl~s of the trade: How to think aboIlt your research while
YO/l're doilIg it. Chicago: University of Chicago Press.
Becker, H. S., Ceer, E., Hughes, E. c., & Strauss, A. L. (1961). Boys ill white: Stll~
dent wltIlre ill medical school. Chicago: University of Chicago Press.
Behar, R. (l995). Introduction: Out of exile. In R. Behar & D. A. Gordon (Eds.),
\'(Iomell !Uritiug wltllre (pp. 1-29). Berkeley: University of California Press.
Bogdan, R. c., & Taylor, S. .J. (1975). Introduction to qualitative research methods:
A phellomenological approach to the social sciences. New York: John Wiley.
Campbell, D. T., & Stanley, j. C. (1963). E.."Cperimental and quasi-experimental
designs for research. Chicago: Rand IVlcNally.
Carey, .J. WI. (1989). COll1l1IlIIIicatioll as wltltre: Essays all media and society.
Boston: Unwin Hvman,
Geourd, A. V. (1964). 'Method and meaSllrell1ellt in sociology. New York: Free
Press.
Clifford, J. (1988). The predicament ofwlture: Twentieth-cenlItry ethnography, lit-
eratltre, and art. Cambridge, l\.1.A: Harvard University Press.
Clifford,j.\ lX Marcus, G. E. (Ed.s.). (1 ~}t)6). V1/rlting culture: The paccic3 arll polifir;~
of ethnography. Berkeley: University of Califomin Press.
Clough, P. T. (1992). The end(s) of ethnography, From realism to social criticism.
Newbury Park, CA: Sage.
41
 LULLt:\"IIl'IIJ HI'ILi 1\'11 CKr'I'\.C 11I'l1.;l \,lVMLI'MI' ~ I-. 'V'M'I-."'MI-....
Clough, E T. (1998). The eJld(s) of cthl/ogrtlplJy: Frolll realislII to social criticislll
(2nd ed.). New York: Petcr Lang.
Cook, D. A. (1981). A history of }wlTatil'(! lillll. New lode \V. W Norton.
Creswell,.J. W (1997). QlIahlatiue il/quiry and research desigl/: Choosing among
five !raditio/Is. Thousand O:::lks, CA: Sage.
Denzin, N. K. (1970). The research det. Chicago: Aldine.
Demin, N. K. (1978). The research act (2nd cd.). New York: lvlcGraw-Hill.
Denzin, N. K. (1997). Illterpretive ethnography. Thousand Oaks, CA: Sage.
Dilthey, \XZ L (1976). Selected writillg,:. Clmbridge: Cambridge University Press.
(Original work published 1900)
Divcrsi, jVL (1993). Glimpses of sneer [i fe: Representing lived experience through
shorr stories. QI/alitatiue !Jlqfliry, 4, 131-137.
Ellis, c., & Bodmer, A. P. (Eds.). (1996). Composilfg ethnography: Altenwtiue
forms of qualitative writing. Walnut Creek, CA: Altatvlira.
Filstead, \Y./. J. (Ed.). (1970). Qualitative lIIethodo/ogy. Chicago: .tvlarkham.
Flick, U. (1998). An introdllclic!IIto qlliditd!ive research: Theory, method alld dppli-
wtio/ls. London: Sage.
Geenz, C. (1973). The illter/lrelatioJI fJ(Cl/ftures: Selected essays. New York: Basic
Books,
Geenz, C. (1983). Lon7llwOIuledge: Further essays ill illterpretilie mztIJropo/ogy.
New York: Basic Books.
Geerrz, C. (l988). Worhs </JId liues: T/Je dllthropologist as illlthOl: Stanford, CA:
Stanford University Press.
Glaser, B. G. (1992). Emergence I/S. forcillg: Basics ofgrounded them}'. Mill Valley,
CA: Sociology Press.
Glaser, B. G., & Strauss, A. L. (1967). The discovery ofgroullded the{)/}!: Strategies
for qualitative resedrcIJ. Chicago: Aldine.
Gordon, D. A. (1995). Culture writing women: Inscribing feminist anthropology.
In R. Behar & D. A. Gordon (Ed:;.), WOlJlen writing Cl/ltJlre (pp. 419-441).
Berkeley: University of California Press.
Gordon, D. A. (1988). \Vriting culture, writing feminism: The poetics and politics
of experimemal ethnography. InscriptioJls, 3/4(8), 21-31.
Greenblatt, S. (1997). The touch of tbe real. In S. 13. Ortner (Ed.), The fate of "cul-
ture": Geenz and beyond lSpecial issue]. Re{Jresel1tatiolls, 59, 14-29.
Guba, E. G. (1990). The alternative paradigm dialog. In E. G. Guba (Ed.), 'l7JC pal"<.l-
digm dialog (pp. 17-30). Ne\vbury Park, CA: Sage.
Guba, E. G., & Lincoln, Y. S. ("1989). Fourth generatioll eua!lwtioll. Newbury Park,
CA: Sage.
Gupta, A., L\:: Ferguson, J. (1997). Discipline and practice: "The field" as site, meth
od, and location in anthropology. In A. Gupta & J. Fetguson (Eds.), Anthro-
pological IOCLllirJlls: BOlIl/darics alld grounds ola field sciellce (pp. 1-46).
Berl,eley: University of California Press.
42
JlIUUUULUUII
Hammersley, tvI. (1991). What's lUrolJg luith ethllography? lvfethodological explora-
tiolls. London: Routledge.
Hammersley, J'v1. (in press). Not bricolagc but boarbuilding.JolIl'lwl uleol/tempo-
rary Ethl/ography.
hooks, b. (1990). YeamilIg: Race. gender, {llld wltllral politics. Boston: South End.
Huber,.J. (1995). Cemennial essay: Insritutional perspectives on sociology. AlIleri-
cml JOIl/"ll<11 of Sociology, /OJ, 194-216.
Jackson, Iv1. (1998). !'dill ima etlJJ/ogra phiw: [II ters1lbjeetivity aJId the alltl.!ro!Jologi-
cal project. Chicago: University of Chicago Press.
Jameson; F. (1991). Postlllodemism; 01; The wlt1lral logic of lilte Ci.lpitalisll1. Dur-
ham, NC: Duke University Press.
Jones, S. H. (1999). Torch. Qlwlitatiuc Inqlliry, 5, 235-250.
Keller, C. ivl., & Keller, J. D. (1996). Cogllition aJId tool1lse: The b1achsmith It
worh. New York: Cambridge University Press.
Lather, E (1993). Fertile obsession: V~J1idity after poststructuralism. Sociological
Quarterly, 35, 673694.
Lather, 1\ & Smithies, C. (1997). '1I-01tblillg the angels: \Vowell liuillg with I-IIVI
AIDS. Boulder, CO: Westview.
LeCompte, )\,,1. D., & Preissle, J. (with Tesch, R.). (1993). Ethl10gmphy and qualita-
tive design ill edllCdtiOJwl research (2nd ed.). New York: Academic Press.
LeviStrauss, C. 09(6). The sill.'lge lI1il1d (2nd cd.). Chicago: University of Chicago
Press.
Lincoln, Y. S. (1999; June). Courage, vulnerability and truth. Keynote address
delivered at the conference "Reclaiming Voice II: Ethnographic inquiry and
Qualitative Research in a Postmodern Age," University of California, Irvine.
Lincoln, Y. S. (in press). V~lfieties of validity: Quality in qualitative research. In J. S.
Smart & C. Ethington (Eds.), Higher education: HOlldhooh oftheory al1d rc-
search. New York: Agathon Press.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic il1quiry. Bevetly Hills, CA: Sage.
Lofland, J. (1971). Analyzing SOCi11 settings. Belmont, CA: \"X!adswunh.
Lofland,]. (1995). Analytic ethnography: Features, failings; and futures.Jolll"lwl of
COlltc/llpomry Etlmogr<1{1hy, 24, 30-67.
Lofland, j., & Lofland, L H. (1984). Analyzing social settings: A gllide to qllalita-
tive ohseruatiotl mId lJ!1lysis (2nd ed.). Belmont, CA: \X'adsworth.
Lofland, j., & Lofland, L H. (1995). Analyzing social settings: A guide to qllalitt1-
tive observatiol1 and analysis (Jrd ed.). Belmont, CA: \"X1adswonh.
Lofland, L. (1930). The 1969 [Humer~HllghesT:llk. UrbaJl Lile alld Cllltllrc, 8, 243-
260.
~1alilloWski, Ii (1948) . .M:7gic, Sd81lct! t11Ui I'8ligioll, and a/her essays. New lorle:
. Natural HistOry Press. (Original work published 1916)
Mahnowski, B. (967). A diary ill the strict sellse ofthe tel'll! (N, Gmerman, Trans.).
New lorle: Harcourt, Brace L\:: \Vodd.
43
 I....ULLCI....III'lQ MI'lU !I'll CKr"r<;C 111'1\:1 '.dUAL! lMllVC IVIMJ Cr<;IML.J
tvIarcus, G. E., & Fischer, lVL 1vl. J. (1986). Allthropo/ogy as wltural critique: All
e.yperinlelltallllolllent in the hWlliIllsciwccs. Chicago: University of Chicago
Press.
Ivlonaco,]. (1981). How to read a (illll: The art, tedmoIogy, language, history and
theory o( (iIm (Rev. ed.). New York: Oxford University Press.
Nelson, c., Treichler, P. A., & Grussberg, L. (1992). Cultural studies: An introduc-
tion. In L. Grossberg, C. Nelson, & P. A. Treichler (Eds.), CuItllral studies
(pp. 1-16). New York: Roudedge.
Ortner, S. B. (1997). Introduction. In S. B. Ortner (Ed.), The fate of "culture":
Geenz and beyond [Special issue]. Represelltations, 59,1-13.
Plath, D. W (1990). Ficldnotes, filed notes, and the conferring of note. In R. Sanjele
(Ed.), FieIdllotes: The makings of anthropology (pp. 371-384). Ithaca, NY:
Cornell University Press.
Richardson, L. (1997). Fields of play: COllstmctillg an academic life. New Bruns-
wick, N]: Rutgers University Press.
Roffman, I~, & Purdy,]. (1981). The HolI}.wood soci,71 problel/l film. Bloomington:
Indiana University Press.
Ronai, C R. (1998). Sketching with Derrida: All ethnography of a researcher/erotic
dancer. Qualitative Jnquiry, 4, 405-420.
Itosaldo, R. (1989). Culture and trtfth: The rellwkiJlg of social al/dysis. Boston:
Beacon.
Sanjek, R. (Ed.). (1990). Eddl/otes: The m{/f~illgs of al/thro!}ology. Ithaca, NY:
Cornell University Press.
Schwandt, T. A. (1997a). Qualitative illquiry: A dictionary o( terms. Thousand
Oaks, CA: Sage.
Schwandt, T. A. (1997b). Textual gymnastics, ethics and angst. In \Y/. G. Tierney &
Y. S. Lincoln (Eds.), RepresCl/!ation alld the text: Re~(ralllillg the /wrratiue
voice (pp. 305-311). Albany: State University of Nc\v York Press.
Silverman, D. (1997). Towards an aes[hetics of research. In D. Silverman (Ed.),
Qualitative research: Theory, method mul {mIctice (pp. 239~253). London:
Sage.
Smith, A. D. (1993). Fires ill the liIirror: CroWII Heights, BrooklYI/, a1/d other idellli~
ties. Garden City, Nl': Anchor.
Snow, D., & 1\'10rri1l, C. (1995). Ironies, puzzles, and contradictions in Denzin and
Lincoln's vision of qualitative research. Jot/mal of Contemporary Etlmogra-
ph}', 22, 358-362.
Spindler, G., & Spindler, L. (1991). Cultural process and ethnography: An anthro~
po logical perspective. Jn 1'v1. D. LeCompte, \Y..Z L. Ivlillroy, & J. Preil;de (Eds.),
The hmnluook. of qlUzlitatiue resenrch ill education (pp. 53~92). New York:
Academic Press.
Stocking, G. V/..z, Jt. (1986). Anthropology and the science of the irrational:
Ivialinowski's encounter with Freudian psychoanalysis. In G. WI. Stocking,.1 r.
44
fllllLJUULULJI!
(Ed.), I\tlalinowski, Rivers, Benedict and others: Essays all cultltre alld per-
sonality (pp. 13-49). Madison: University of \'(/isconsin Press.
Stocking, G. W,]t. (1989). The cthnogr.1phic sensibility of the 1920s and the dual-
ism of the anthropological tradition. In G. WI. Stocking, Jr. (Ed.), Romantic
lIIotives: Essa)'s all iII/thropoIogical sellsibility (pp. 208276). Madison: Uni-
versity of Wisconsin Press.
Stoller, I~, & Olkes, C. (1987). III sorcery's slwdow: A memoir o( apprenticeship
(/mollg the Sangha)' o( Nigel: Chicago: University of Chicago Press.
Strauss, A. L., & Corbin, J. (1998). Basics o( qllalitatiue research: Techlliques alld
procedures (or deuefopillg grollllded theory (2nd eel.). Thousand Oaks, CA:
Sagc.
Taylor, S.]., & Bogdan, R. (1998). Illtroduction to qllalit{/tive research methods: A
gllidebooh and resOllrce (3rd cd.). New York: John Wiley.
Turner, \z, & Bruner, E. (Eds.). (1986). The anthropology of experiellce. Urbana:
University of Illinois Press.
Van I'vlaanen, J. (1988). Tales of the field: 011 writing ethnography. Chicago: Uni-
versity of Chicago Press.
Vygotsky, L. S. (1978). Mind ilz society: The developmel/t of higher psychological
processes (IvL Cole, V John~Steiner, S. Scribner, & E. Souberman, Eds.).
Cambridge, IviA: Harvard University Press.
\X!einstein, D., & \X'einstcin, ivl. A. (1991). Georg Simmel: Sociological flaneur bti-
colem. Theory, Cllltltre 6 Society, S, 151-168.
West, C. (1989). The Americall evasioll of philosophy: A gelled{o!:,')' of pmgmatis1l1.
Ivbdison: University of Wisconsin Press.
Wolcott, H. F. (1990). \\/j-itillg lip qualitalilie research. Newbury Park, CA: Sage.
Wolcott, H. F. (1992). Posturing in qualitative inquiry. In M. D. LeCompte, \VI. L.
iVlillroy, & J. Prcisslc (Eds.), Tbe hmldhoo,~ of qualitative research iI/ educa-
tion (pp. 3-52). New York: Academic Ptess.
\'(!olcott, H. F. (1995). The art of (ieldworl~. Walnut Creek, CA: AltaMira.
\'\'olf, M. A. (1992). A. thrice-told tale: Fell/iI/ism, postmodemism, alld ethnographic
respOllsibility. Stanford, CA: Stanford University Press.
45
 PART}_ _

Methods of
Collecting and
Analyzing Empirical
Materials

The socially situated researcher creates through interaccion the realities

that constitute the places where empirical materials arc collected and ana-
lyzed. In such sites, the interpretive practices of qualitative research afC
implemented. These methodological practices represent different ways of
generating empirical materials grounded in the everyday world. The con-
tributions to Part I of this volume examine the multiple practices and
methods of analysis that qualitative researchers-as-methodological-brico-
leurs now employ.

4> The Interview

\Y/e live in an interview sociery, in a society whose members seem to believe

that interviews generate useful information about lived experience and its
meanings. The interview has become a taken-for-granted feature of our

47
 - '"<. - - ' .... 'L ,. " ,... " ' ....... .,, v, ,-VI<C''''L''''d U"U """U'r.:..II,!:! I..II'j-'''''''U'
mediated, mass culture. But the interview is a negotiated text, a site w-here
pm'>'er, gender, race, and class inteL,ect. In Chapter 2, Andrea Fontana and
James Frey review the hislory of the interview in the social sciences, noting
its three major forms-structured, unsttuctured, and open-ended-and
showing how the tool is modified and changed during lISC. They also dis-
cuss group (or focused) interviews (see also lvladriz, Chapter 10), oral his-
tory interviews., creative interviewing, and gendered, feminist, and post-
modern, or l11ultivoiccd, interviewing.
The interview is a convers"ltion, the art of asking questions and listen-
ing. It is not;1 neutral mol, for 3t least two people create the reality of the
imervievv situation. In this siru::ttion answers are given. Thus the interview
produces situated understandings grounded in specific interactional epi-
~()des. This method is influenced by the personal characteristics of the
interviewer, induding race, class, ethnicity, and gender.
Fontana and Frey review the important \vork of feminist scholars on
tbe interview, cspecially the argumcnts of Beh3r, Reinharz, Hertz,
Rich3rdson, Clough, Collins, Smith, and Oakley. British sociologist
Oakley (1981) and mher feminist scholars have identified a major contra-
diction between scientific, positivistic research, which requires objectivity
and detachment, and feminist-based interviewing, \"/hich requires open-
ness, emotional engagement, and the development of a potentially long-
term., trusting relationship between the interviewer and the subject.
A feminist interviewing ethic, as Fontana and Frey suggest, redefines
the interview situ3tion_ This ethic transforms interviewer and respondent
into coequals who are carrying on a conversation about mutually relevant,
often biographically critic31, issues. This narrative, storyrelling framework
challenges the informed consent and deception models of inquiry dis-
cussed by Christians in Volume 1, Ch3pter 5. This ethic changes the inter-
view into an important tool for both applied action research (see Kemmis
& IvteTaggart, Volume 2, Chapter 11) 3nd clinical research (see Miller &
Crabtree, Volume 2, Chapter 11).
4 Observational Methods
Going into 3 social situation and looking is another important way of gath~
ering marerials about the social world. In Chapter 3, Micl13c1 Angrosino
and Kimberly lVlays de Perez fundamemally rc\vrite the methods and
48
/YIU~"'IIU'::'
practices of naturalistic observarion. All observation involves the
observer's participation in the world being studied. There is no pure, objec-
tive, detached observation; the effects of the observer's presence can never
be erased. Further, the colonial concept of the subject (the object of the
observer's gaze) is no longer appropriate. Observers now function as col-
Iabor3tive participants in action inquiry settings. Angrosino and Perez
argue that observational interaction is a tentative, situational process. It is
shaped by shifts in gendered identity as well as by existing structures
of power. As relationships unfOld, participants validate the cues gener-
ated by others in the setting. Finally, during the observational process peo-
ple assume situational identities that may not be socially or culturally
normative.
Like Christians in Volume 1, Chapter 5, Angrosino and Perez offer
compelling criticisms of insritutional review boarels (IRBs), noting that
positivistic, experimental social scienrists seldom recognize the needs of
observational ethnographers. In many universities, the IREs are tied to the
experimemal, hypothesis-testing, so-called scientific paradigm. This para-
digm creates problems for postmodern observers, for scholars \vho be-
come part of rhe worlds they study. In order to get approval for their re-
search, scholars 1113Y have to engage in deception (in this instance of the
IRE). This leads some ethnographers to claim that their research will not
be intrusive and hence will nor cause harm. Yet interactive observers are by
definition intrusive. \X1hcn collaborative inquiry is undertaken, subjects
become stakeholders, persons who shape the inquiry itself. \X1hat this
means for consent forms-and forms of participatory inquiry more
broadlY-is not clear. Alternative forms of ethnographic writing, includ-
ing the use of fictionalized stories, represent one avenue for addressing
this ethical quandary.
Angrosino and Perez offer an ethic of "proponion3te reason." This
utilitarian ethic attempts to balance the benefits, costs, and consequences
?f actions in the field, asking if the means to an end arc justified by the
IlTIportance and value of the goals attained. These authors demystify
the observation method. Observation is no longer the key to some grand
ana.lysis of culture or society. Instead, observational research is a method
that focuses on differences, 011 the lives of particular people in con-
crere, but constantly changing, human relationships. The relevance and
nccd fo r .a fe1l1tmst
.. et lllCS
. of care and commItment
. become even more
apParent.
49
 LULLCLIII'l'-..:l MI'lU 11'11 CKF('\.C t I!-'\..J ,-<,UMl..' ,,-.,, 'L ,n,-.'
.;p Reading Material Culture and Its Records
Iviuce evidence-that is, written texts and cultural artifacts-endures phys-
ically and leaves its traces on the material past. It is impossible to talk to and
with these materials. Researchers must interpret them, for in them arc
found important meanings about the human shape of lived cultures.
Archaeologists study material culture. In an essay that moves with ease
across Jnd within the postpositivist and postmodern sensibilities, Ian
Hodder (Chapter 4) shows how this is done. Central to his position is the
constructionist (and constructivist) argument that researchers create,
through a set of interpretive practices, the 1l1Jtcriais and evidence they then
theoretically analyze. Today itis understood that material culture, in all its
forms, is a gendered, social, and political construction. Previous theories of
culture, evolution, and the material past are being rewritten. How the past
is reconstructed and interpreted very much determines hmv it will be con-
stituted in the present and remembered in the future.
~, Reimagining Visual Methods
Today, visual sociologists and anthropologists usc photography, motion
pictu~es, the \Xlorld \XTide Web, interactive CDs, CD-ROl\,ls, and virtual
reality as ways of forging connections between human existence and visual
perception. These forms of visual representation constitute different ways
ot recording and documenting what passes as social life. OEren called the
mirror with a memory, photography takes the researcher into the everyday
world, where the issues of observer identity, the subject's point or view, and
\'lhat to photograph become problematic. In Chapter 5, Douglas Harper
presents a history of this method and brings it up against postmodern
developments in virtual and real ethnography.
HistoricaUy, visual sociology began within the postpositivist tradition;
researchers provided visual information to support the realist tales of
traditional ethnography. Photographs were a part of the unproblematic
"facts" that constituted the "truth" of these tales. Now, visual sociology,
like ethnography, is in a period of deep questioning and great change.
Visual sociology, I-Iarpt:r contend::;, must fiod il pL..cc in rhi:> 0<;\'1/ ethnoc;r n -
phy. He engages this ne'.\' turn through a close, storied reading of a series of
photographs he made on the streets of Bologna. This bicycling sequence is
a visual narrative. It tells many different stories at the same time as it mixes
50
L,,,,,,L"; rul < I. ly,t::U 'uu~ U, I...UII!,;;'UH ly Uf/U p,nUlYZlng r::mpmcO! /VlatenalS
and combines multiple images, cultural meanings, points of view, geo-
graphic spaces, interactional sequences, and shifting, gendered forms of
the gaze. Harper also analyzes the ideological aspects of representation,
[he social construction of images, the authority of visual knowledge, the
mechanical capabilities of the camera, framing (point of view), printing
techniques, editing, and image sequencing.
As Harper's bicycle shots indicate, sequences of photos can be con-
nected through visual narratives, stories [hat connect images to first-
person accounts, and cultural stories that unfold through time and space.
Of course, every image tells a story, but visual narratives attempt to tell
the stories of a culture, of individuals, and of institutions, and their inter-
relationships. Photo elicitation is one method used to elaborate these
meanings.
\Y./e need to learn how to experiment with visual (and nonvisual) ways
of thinking. \Ve need to develop a critica!, visual sensibility, a sensibil-
ity rhat will allow us to bring the gendered material world into play in crit-
ically different ways. \XTe need to interrogate cri[ically the hyperlogics
of cyberspace and its virrual realities. \'XIe also need to understand more
fully the rules and methods for establishing truth that hold these worlds
together.
+ Autoethnography and the Researcher as Subject
Personal experience reflects the flow of thoughts and meanings that per-
sons have in their immediate situations. These experiences can be routine
or problematic. They occur within the life of a person. \X'hen they are
talked about, they assume the shape of a story, or a narrative. \Ve cannot
s~udy lived experience directlYl because language, speech, and systems of
discourse mediate and define the very experience we attempt to describe.
\X'e study the representations of experience, not experience itself. We
examine the stories people tell one another about the experiences they
hav~ had. These stories may be personal experience narratives or self-
Stones , intc
- rpretatIons. rna de up as t I1e person goes along.
lvIany now argue that v,'e can study only our O\\ln experiences. The re~
searcher hecome,; rl~c r<.:':>carcb :;;ubjcc"t. This 1:; til\,:; tupi\,; of autoctill10lf
raph
'h. y In Ch apter"6 Caro Iyn Ellis . and Arthur Bochner reflexively present
t e argume t I .. I I ' . .
' , n s or wntmg re eXlve personal narratives. Indeed theIr
dlalogic ['V . '. ' . ,. . "
e..,.r IS an example of such \\'ntlllg; It performs Its O\vn narratIve
51
 L-UI..LI..L-""U ,...,~ ..... "" L'" " L " ' ...... ~~,-,~""',. ~
reflexivity. Ellis Jnd Bochner masrerfully review rhe arguments for srudy-
ing personal experience narratives, anchoring their text in the discourses
of poststructuralism and postrnodernislll, especially the works of Rony
and Richardson.
Thev review the hiswry or this writing form, starring with David
Havan~'s introduction of the term alltoethnography in 1979, A variety of
ter~s and methodological strategies are associated v.lith the meanings and
uses of alltoethnographies, including personal narratives, narratives of the
self, writing stories, self-stories, auto-observation, personal ethnography,
literary tales, critical alltobiography, radical empiricism, evocative narra-
tives, reflexive ethnography, biographical method, co-construcred narra-
tive, indigenous anthropology, anthropological poetics, and performance
ethnography. Ellis and Bochner usc the case of Sylvia Smith, Ph.D. camli-
date in psychology, to illustrate the value of this form of writing.
They thcn turn to thcir own intellectual biographies, showing how they
came to their current understandings concerning the need to write about
the researcher as subject. They show that the commitment to this style of
writing docs not comc easily. Ie involves learning how to write diffcrently,
including how to use personal experience and the first-person voice as
vehicles for authorizing claims to lrllth and knowledge, And there are
many critics, including those who \vonder about murative trllth, emo-
tional recall, and layered texts, who qucstion rhe point of a storied life and
worry as \vell about such traditional issues as reliability and validity.
Of course, this aucoethnography can be read as a variation on the tes-
timonio and the first-person life hiscory. Thus Ellis and Bochner's chapter
complements Tierney's (Volume 2, Chapter 9) and Beverley's (Volume 2,
Chapter 10) treatmenrs of these narrative forms.
" Data Management and Analytic Techniques
The management analysis and interpretation of qualitative empirical
materials is a co:n~plex ~ro~css involving highly technical languages and
systems of discourse. It also entails the mastery of a special set of inter~
pretive practices and nartative techniques. In Chapter 7, Gery Ryan and
l-l. Russell Bernard advunce perhaps the most sophisticated and compre-
hensive model of this process and its discourses.
The management and analysis of empirical matcrials involves argu-
ments concerning the differences between empiricism (there is a real
52
,.0> .... ,. , . , ,~ ,~~ ........ , '--'-"''-'L'" '':1 U, 'u ...." ,u,y"'-" '':1 L' "/J"'LU' "I~'lCIIU';:'
world out there) and constructivism (the world is constructed). It also
involves disputes between empiricists and nonempiricists-between those
who would tran.slate their observations into words and those who would
translate their observations into numbers. Like others, Ryan and Bernard
also distinguish two approaches to texts, the narrative or the linguistic and
the sociological. The narrative approach to texts treats them as objects of
narrative, conversation, performance, or formal analysis. The sociological
approach treats texts as windows into experience and includes both texts
generated by the analyst and free-flovving texts, or narratives.
Ryan and Bernard focus on methods used in the sociological tradition,
that is, methods for collecting such materials (free lists, pile sorts, frame
elicitations, and triad tests) and techniques for their analysis (taxonomies,
mental maps, componential analysis), They also discuss methods for ana-
lyzing free-flowing texts, starting \virh texts that use raw text input (key-
words-in-context, \vord counts, semantic network analysis, cognitive
maps). They then take up methods that reduce texts to codes: grounded
theory, schema analysis, classic content analysis, content dictionaries,
analytic induction, and ethnographic decision models.
This is a far-reaching, encyclopedic, elegant, and systematic postposi-
tivist approach to the issues surrounding the rigorous analysis of empirical
materials. Ryan and Bernard's treatmenr of grounded theory, conversation
analysis, and computer-assisted models of analysis should be read in con-
junction with the treatment of these topics by, r~spectively, Charmaz (Vol-
ume 2, Chapter 8), Silverman (Chapter 9), and Weitzman (Chapter 8).
Computer-Assisted Gualitative Analysis
It is now becoming relatively commonplace for researchers to use COI11-
puter softwarc programs to assist them in their analysis of qualitative
~mpirical materials. Lee and Fielding (1998, p. 1) c~I1 such programs
~omputer-assistcd qualitative data analysis software," or CAQDAS, In
Chaprer 8, Eben \X!eitzman presents a comprehensive and user-friendly
s~rvey of a wide array of CAQDAS currently available to support qualita-
nvc analysis. He reviews computer-based tools dur can help researchers to
record , !-:tot'E', In
- d ex,_ <;:l.u",,-,ndcx
_.. _ ' .
, o.;:,;,dc] o-vrl , and l.nrcrcon.nc<;t; rcxt;-ba,;""d
materials. Of course, these tools are not ideologicall}! neutral (Schwandt,
1997
1 p. 18). They structure the work of interpretation and presume a par-
tlcular gend cre d stance tmvard the matenal
. world. They frequently lmpose
.
53
<1 rational, hierarchical, linc3f Of quasi-Ii near, and scqucnrial framework on
the world and its empirical materials. This can creare the impression that
meaningful patterns acrually exist in the d3ta, when in fact they are created
by the software and analytic frameworks being used (in the case of analysis
that is not assisted by computer, of course, it is the researcher who creates
the seeming <'order"). These tools can also distance researchers from their
fieldwork and their empirical materials. These methods presume an
objectivist, realist, foundational epistemology, and their use too often takes
for granted the interpretive procedures and assumptions that transform
field notes into text-based materials.
\\leitzman divides the most frequently used programs into five main
software families: rextbase managers, code programs, retrieve programs,
code-based theory builders, and conceptual network builders. These pro-
grams have multiple text management uses, such as helping researchers to
locate and retrieve key materials based on phrases and words, build con-
ceptual models, sort categories, attach key words and codes to text seg-
ments, isolate negative otdcviant cases, and create indices. IVlultimeclia
software programs are JUSt now appearing that allow researchers to use
audio and video as well as text-based data. CD-ROIv'ls arc also functioning
as sites where field nores and other versions of ethnographies are stored
3nd made accessible for hypenextual analysis (see Coffey & Atkinson,
1996, p. 186).
Such powerful tools of graphic, visual, and audio representation allow
researchers to consolidate, and establish patterns of consistency in their
materials. However, they can also create negative effects, including the
false hope that such programs can actually write a theory (or a case) [or
researchers. They may even encourage quick-and-dirty, or "blitzkrieg,"
research. Coding and retrieval schemes can lead to an overemphasis on
the discovery of categories and indicators, with a corresponding under-
emphasis on the multiple meanings of experience in concrete situations
(sec Fielding & Lee, 1998, pp. 120-121). The search for grounded theory
can shift attention away from the theories of interpretation that operate in
the social world.
Software programs for the qualitative tesearcher need to be interactive,
allowing for many different interpretive spaces to emerge, spaces that con-
nect patterns vvith mC:lnings :lnd experience. Nonetheless, it is impor~
tant that the researcher avoid letting the computer (and the software)
determine the form and content of interpretive activity. An emphasis on
codes and categories can produce endless variable analyses that fail to rake
54
account of important situational and contextual factors. I There is fre~
quently a tendency among researchers doing computer-assisted analysis to
reduce field materials to onlv those data that are codable. There is also the
danger that researchers will turn over the transcription of their field notes
to persons who lack intimate familiarity with the field setting and the pro-
cesses being studied (Lee & Fielding, 1991, p. 12).
Seidel (1991) speaks of a form of Jnalytic madness that can accompany
the use of these methods. This madness can lead researchers to an infatua-
tion with the large volumes of dara the computer allows them to deal with.
In addition, researchers may develop understandings based on misunder-
standings; that is, patterns identified in the dara may be "artefacts of a rela-
tionship [they] have with the dara" (Seidel, 1991, p. 114). Fin3lly, re-
searchers may focus only on those aspects of their research that can be
helped by computer methods (Agar, 1991, p. 193). They then select and
ready particular software for use in analyzing the m3teriais they gather.
Thus the problem arises: Frequently, researchers conduct research that
fits the available software and then report that the software constituted
their methodology. This is the methodological tail wagging the ethno-
graphic dog.
Finally, ethical problems may arise from the use of such programs.
Akeroyd (1991) isolates the crux of the matter: the potenrialloss of per-
sonal privacy that can occur when a personal, confidential database is
developed on an individual or group. \X!hcn such materials are entered
into a computer, the problem of secutity is immediately created. In
multiuser systems, privacy cannot be guaranteed (Akeroyd, 1991, p. 100).
Nothing is any longer completely private or completely secure.
Fielding and Lee (1998, pp. 186~189) have speculated on the future of
CAQDAS, and they suggest that the field is entering a period of "winnow-
ing OUt," with some soft\'Ilare packages becoming more sophisticated and
others remaining undeveloped since their initial release. Some developers
have left the field. The \X!indows operating system seems to have "caused a
major shake-ollt of those willing to keep up with its programming and
development requirements" (Fielding & Lee, 1998, p. 186). Software that
permits the direct transcription of speech stored on CD-ROIvI continues
to be developed. In some programs the speech is actually heard 35 it is
~cing transcrihed. This ]nls0.<; issues :1.hout the difference" involved in
1l1terpreting heard versus written words (Fielding & Lee, 1998, p. 188).
The Internet has also produced changes in CAQDAS. Large-scale proj-
ects can now be located on \X1eb sites. Such use of the Internet is not
55
 ,-,-,,--,-,--,-,,, "--' r--., ...... "" ,--n, ",--,,, "--' ,-" ...;r,,--'
\virhollt problems, including rhe commodification of information, [he
control of encryption devices, electronic privacy, and the developmenr of
ethical protocols to produce subjects (Fielding L~ Lee, 1998, p. 188).
Clearly, computer technology as a whole continues to transform and com-
plicate qualitative research.
'!> Analyzing Talk and Text
Qualitative researchers study spoken and written records of human expe-
rience, including transcribed talk, films, novels, and photographs. His-
torically, there have been three major social science and Iitcrary ~lpproaches
to textual-discourse analysis. Each is associated with a long theorerica! and
research tradirion: content analysis with rhe quantitative approach to
media srudies; semiotics with the suucrural tradirion in literary criticism;
and narr"nive, discourse analysis with the recent posrsfructural devel-
opment in interpretive theory (see Lieblich, Tuval-IVlashiach, & Zilber,
1998, p. 18).
David Silverman contends that the world's business gets done in ralk
and in conversation. Hence field data are always linguisric, and in Chapter
9 he analyzes three kinds of linguistically mediated data: inrerviews, texts,
and transcripts. \Y/ith Fontana and Frey (Chapter 2), and Guhrium and
Holstein (Volume 2, Chapter 7), Silverman treats inrerview marcrials as
narrative accounts rather than as true pictures of realiry. l:-lc poses five
questions for interview researchers~ inel uding how they usc their narrative
data to make rheoretical claims about the world.
Texts are based on transcriptions of interviews and other forms of talk.
These texts arc social facts; they are produced, shared, and used in socialty
organized ways. Silverman objects to those forms of text-based analyses
that use the methods of content analysis. Content analyses reify the taken~
for-granted understandings person~, bring to words, terms, or experiences.
Coorent analyses, he conrends, obscure the interpretive processes that
turn talk into text.
It is important that researchers not use rext~based documentary materi-
als as stand-ins for other kinds of evidence. These documents are social
routines, or of decision-making processes. They are situared constntC~
tions, particular kinds of representarions shaped by cerwin conventions
and understandings (Arkinson & Coffey, 1997, p. 47). Such documentS
56
,r--.", '-- ".r--., ,--",r--.,---, rUI< I. "'t::LIIVU;:' VI \..uOIe.... Wly UfJU MIIUIYLJJJY CIIIJ-JlrJLUI /V\ULf;HIUlS
are properly studied through the methods of semiotics, narrative, and dis-
course analysis. 11embership categoriz::uion analysis (MCA) is a less famil~
iar form of narrative analysis. Drawing on the work of Harvey Sacks,
Silverman illustrates the logic of MCA (on Sacks, sec Silverman, 1998).
\"X?ith this method, the researcher asks how persons use everyday terms and
categories in their interactions with others. Silverman turns next to tran~
scripts of ralle There are two main social science traditions that inform the
analysis of transcripts; convcrsation analysis (CA) and discourse analysis
(DA). Silverman revie\'.'s and offers examples of both traditions. He con-
cludes his chaptcr with four arguments, contending that qualitative re-
search (a) is not based on ;] set of freestanding techniques, (b) has special
strengths for revealing how social imeracrions are routinely enacted, (c)
shows us how people do things, and (d) has uses thar extend far beyond
exploratory purposes.
To summarize: Text-based documents of experience are complex. BlIt if
rark constirLltes much of what we have, then the [arms of analysis that
Silverman outlines represent significant ways of making the world and its
words more visible.
1) Focus Groups in Feminist Research
In Chapter 2, Fonrana and Frey note that the group, or focus group, inter-
view relics upon the systematic quesrioning of several individuals simulta~
neousl)' in a formal or informal setting. In Chapter 10, Esther tvIadriz sig-
nificantly advances the discourse on this method by showing how focus
groups are used in feminist research with women of color. 2 Using a femi-
nist/postmodern approach, she offers a model of focus group interviewing
that emphasizes a feminist cthic of empowerment, moral community, emo-
tional engagement, and the development of long-term, trusting relation-
ships. This method gives a voice to women of color who have long been
silenced. Focus groups facilitate women writing culture together. As a
~atina feminist, i\!ladriz places focus grOL!pS within the context of collec-
t~\'c tc:;timonies Jnd group resistance narratives (sec Beverley, Volume 2]
Lh!1Phc.-r 1 OJ T.icnH;Y, Volume Z, ClIaplCf ':l) . .focus groups reduce [he clis-
tanee between the researcher and the researched. The rnultivocaJity of the
P,3rticipanrs limits the control of the researcher over the research process.
1 he unstructured nature of focus group conversations also reduces rhe
57
 \...VI..I-.-I..'-.-, " ' ' - ' ,....." ..... '" ,1-.-,,, " .... ,,, , .....
researcher's (omral over the interview process. Madriz illustrates these
points with examples drawn from her srudy of lower-class women of color.
Drawing on recent developments in critical race theory and feminist
theory (see, respectively, Ladson-Billings, Volume 1, Chapter 9; Olesen,
Volume 1, Chapter 8), JVfadriz reminds us that women of color experiencc
a triple subjugation based on class, race, and gender oppression. Focus
groups create the condirions for the emergence of a critical racc conscious-
ness, a consciousness focl!Sed on socia! change. It sccms that with focus
groups, critical race theory has found its mcthodology.
" Applied Ethnography
The applied, action themes of Peirt I arc continucd in Ervc Chambers's
comprehensive analysis of the history, forms, and uses of applied ethnogra-
phy in Chapter 11. Applied research is inquiry intentionally developed
w-ithin a context of dccision making and directed toward the interests of
one or more clients. So framed, applied (or action) ethnography is about
research that advocates social change and increased cultural understand-
ings berween different social groups.
Chambers discusses three distinct traditions \vithin applied ethnogra-
phy: cognitive, semioric, and sem::lntic approaches; micro/macro analyses;
and action and clinical models. Cognitive approaches <lttempt to map the
native point of view in particular situations, to isolate the language, cate-
gories, and terms used in specific locales.lvlicro/macro analyses examine
]ocal contexts with an eye to generalizing to larger, more macro srruc-
rures-for example, moving frol11 the economy of a local community to
the national economy. Action and clinical approaches follow an advocacy
model. Researchers build collaborative relations with a variety of different
types of persons in the local community, from indigenous expens to in-
formed insiders, leaders in churches, schools, and local government, and
representatives of state and federal bureaucracies.
Chambers notes thac applied research places great ethical responsibility
on the shoulders of the researcher (see also Trotter & Schensul, 1998,
p. 692). It carrics human, social, and ecological consequences that are im-
rl:1c..:!i.cli"c Clnd s()rnctil1~es c ..it;cal to tl-,c I;[c
change oriented and can be very disruptive.
Chambers also takes up the issue of professional ethics, He indicates
that the principle of informed consent has proven to be particularly diffi~
58
~'-',... ...,""" ~ .,,,
cult for applied ethnographers. lvL1intaining the confidentiality of re-
search subjects can also be problematic. At tile same time, applied cth-
nogrJphers confront moral issues, such as questions of for whom they
should advocate and \vhether or nor their services should be avaibble
without discrimination.
.. Conclusion
The researcher-as-mcthodological-bricoleur should have a working famil-
iarity with all of rhe methods of collecting and analyzing empiricalmateri-
als presented in this section of the Handbook. This familiarity should
include an understanding of rhe history of each mcthod and tcchniquc as
well as hands-on experience with each. Only in this \vay can a researcher
fully appreciate the limitations and stfellgths of the various methods and, at
the same time, see clearly how each, as J set of practices, creates its own
subject matter.
In addition, the researcher must understand thar each paradigm and
perspective, as prescnted in Part II, has a distinct history \vlth each of these
methods of research, Although methods-as-tools 3re somcvvhat universal
in application, they are not uniformly used by researchers from all para-
digms. And when they are used, they are fittcd and adapted to the particu-
larities of the paradigm in question.
Of the six specific methods and techniques addressed in Pan I of this
volume (interviews, observarion, cultural anibcrs, visualmcrhods, auto-
ethnography, focus groups), positivists and postpositivists arc most likely
to make use of structured intcrviews and those cultural artifacts that lend
themselves to formal analysis. Constructionists and critical theorists also
have histories of using ea~h of the methods, as do feminists, queer theo-
rists, ethnic rcsearchers, and cultural stltdies investigators. Similarly,
researchers from all paradigms and perspectives can profitably make use
of the data management and analysis methods, as \veJl as the computer-
assisted models discllssed.
or" corntnun;t:r_ Tts ,-cs'.tlts "tC
" 1. The cominued shifr towilrd variable ilnalysis h,]s moved cornpmef-assistcd methods
hrml)' in II le'I' 'f
(lfeCtion ' , , mode 15 0 I" IIHcrprctation.
0 pOStpOSltJl'ISt
59
 \....'Jl..l..l..\....' ""0..:1 M"' ..... "" , '-", I''-'"~''''' ,-,:VM'-' 'M' ,. '- ,.,'""', '"-",r> ..........

2. Also rcc;'!ll Fint.:, \Vt.:is, \Vt.:st.:t.:Il, and Wong's discussion of focus groups ill Volume 1,
Chapter .:1.

+ References

Agar, i'vI. (1991). The right brain strikes back. In N. G. Fielding & R. 1v1. Lee (Eds.),
The Intervievv
Using computers in qualitative research (pp. 181-] 94). London: Sage.
Akeroyd, A. V. (J99J). Personal information and qUillit;uive research darel: Some
practical and ethical problems arising from d<Ha protection legislation. In
Franz Structured
N. G. Fielding &.c R. ivi. Lee (Eds.), Using computers in qualitative research
(pp. 89-106). London: Sage. Questions to Negotiated ll.'xt
Atkinson, P., & Coffey, A. (J 997). Analysing documentary realities. In D. Silverman
(Ed.), Qualit<Hive rese8rch: Theory, method and practice (pp. 45-62). Lon-
don: Sage.
Coffey, A., & Atkinson, P. (1996). Iviaking scnse of qUillitative clara: Complemcn- Andrea Fontana and James H. Frey
tary research strategies. Thousand Oaks, CA: Sage.
Fielding, N. G., & Lee, R. M. (1998). Computer analysis and qualitative research.
London: Sage. Hamlet: Do yOll see yonder cloud that's almost in shape of a c:1lllel?
HaY,lno, D. M. (1979). Auto-ethnography: P;lradigms, problems, and prospects. Polollius: By the mass, and 'tis like a camel, indeed.
Human Organization, 38,113-120. Hamlet: IVlethink it is like a weasel.
Lee, R. lvI., & Fielding, N. G. (1991). Computing for qualitative rese:1fch: Options, POIOllills: It is backed like a weasel.
problems and pmemial. In N. G. Fielding & R. tv!. Lee (Eds.), Using comput- Hamlet: Or like a whale?
ers in qualitative research (pp. 1-13). London: Sage. POIOlliIfS: Very like a whale.
Lieblich, A., Tuval-Mashiach, R., & Zilber, T. (1998). Narrative rese:l!"ch. Thou-
sand Oaks, CA: Sage. Hamlet's illterl'iew . .. approximates the threefold ideal of being il1tel1lrctcd, vaJi~
Oakley, A. (l981). Imerviewing women: A contradiction in terms. In H. Roberts dated and com/llIwicated. ...
(Ed.), Doing feminist research (pp. 30-61). London: Routledge & Kegan The il1terview appears as a display of the 110/ver relations at a royal ((IlIrt . ..
Paul. Hamlet's interuiCiu may ... be seen as all illustration of a perl'asft'e doubt abollt
Schwandt, T. A. (1997). Qualitative inquiry: A dictionary of terms. Thousand Oaks, the appearance of the world. [01; we /{/Ould Jihe to add, the illtcrvieu.-' can elllClge as
CA, Sage. an exm/lple of a negotiated text.]
Seidel,]. (1991). Method and madness in the application of computer technology
to qualitative data analysis. In N. G. Fielding & R. lvI. Lee (Eds.), Using com-
puters in qualitative research (pp. 107-J 16). London: Sage.
Silverman, D. (1998). Harvey Sacks: Social science and conversation analysis.
Cambridge: Polity. Asking questions and getting answers is a much harder task than it
Trotter, R. T., & Schensul,J..J. (1998).lvIcthods in applied anthropology. In H. R. may seem at first. The spoken or written word has always a residue
of "'-I
::1mb'IgUlty,
. .
no matter how carefully \ye \yord the questIOnS ;1I1d how
Bernard (Ed.), Handbook of cultural methods in cultural anthropology
(pp. 691-736). \l(7alnu t Creek, CA: AltalVlira. carefully we rep on or code the answers. Yer interviewing is one of the

60 61
 NIt I HUll:' Ur LULLl:L 111'1\;1 MI'H.J f-\I"KLI 1..-11 "'-' '-"" ""'-"~ , .. " ~"
most common and powerful \vays in \vhich we try to understand our fel-
low human beings. Interviewing includes a wide variety of forms and a
multiplicitv of lIses. The most common form of interviewing involves inch-
vidual, face-to-face verbal interchange, but imervie\:ving can also rake rhe
form of face-to-face group interchange, mailed or self-administered ques-
tionnaires, and telephone surveys. It can be structured, semistructured, or
unstructured. Interviewing can be used for marketing research, political
opinion polling, therapeutic reasons, or academic analysis. It can be used
for the purpose of measurement or its scope can be the understanding
of an individual or a group perspective. An interview Can be a one-time,
brief event-say, 5 minutes over the telephone-or it can take place over
multiple, lengthy sessions, at times spanning days, as in life history
interviewing.
The use of interviewing to acquire information is so extensive today
that it has been said that we live in an "interview society" (Atkinson L~
Silverman, 1997; Silverman, 1993). Increasingly, qualitarivc researchers
are realizing that interviews are not neutral tools of data gathering but
active interactions bet\veen two (or more) people leading to negotiated,
contextually based results. Thus rhe focus of interviews is moving to
encompass the haws of people's lives (the constructive work involved in
producing order in everyday life) as well as the rraditional whats (the
activities of everyday life) (Cicourel, 1964; Dingwall, 1997; GubriulTl &
Holstein, 1997, 1998; Holstein & Gubrium, 1995; Kvale, 1996; Sarup,
1996; Seidman, 1991; Silverman, 1993, 1997a).
In this chapter, after discussing the interview society, we examine inter-
views by beginning with strucrured methods of interviewing and gradually
moving to more qualitative types, ending with interviews as negotiated
texts. \YJe begin by briefly outlining the history of interviewing, then we
turn to a discussion of the academic uses of interviewing. Although the
focus of this volume is qualitative research, in order to demonstrate the
full import of interviewing, we need to discuss the major types of inter-
viewing (structured, group, and unstructured) as well as other ways to
conduct interviews. A caveat: In discussing rhe various interview methods,
we use the language and r3tionales employed by practitioners of these
methods; we note our differences with these practitioners and our criti-
cisms later in the chapter, in our discussion of genclered and other neW
types of qualirative interviewing. Following our examination of st[[lC~
tured interviewing, we address in derail the various clements of qualitative
interviewing. \Ve then discuss the problems related to gendered intervic\v-
62
ing as well as issues of interpretation and reporting, and we broach some
considerations related to ethical issues. Finally, we note some of the ncw
trends in qualitative interviewing.
+ The Interview Society
Before embarking on our journey through interviewing per se, we want to
comment briefly on the tremendous reliance on interviewing in U.S. soci-
ety today, which has reached such a level that a number of scholars h;we
referred to the United States as "tbe interview socicty" (Atkinson &
Silverman, 1997; Silverman, 1993). Both qualitative alld quantitative
researchers tend to rely on the inrervievv as the basic method of dara gather-
ing, whether the purpose is to obtain a rich, in-depth experienri;:ll accoum
of an evem or episode in the life of the respondent or to garner a simple
poim on a scale of 2 to "10 dimensions. There is inherent faith that the
results arc trustworthy and accurate and that the relation of the interviewer
to respondent that evolves in the inrerview process has not unduly biased
the accollnt (Atkinson & Silverman, 1997; Silverman, 1993). The cOlllmit-
ment to and reliance on thc interview to produce narrative experience
reflects and reinforces the view of the United States as an interview society.
It seems that everyone, not just social researchers, relies on the inter-
view as a SOurce of information, with the assumption that interviewing
results in true and accurate pictures of respondents' selves and lives. One
cannot escape being interviewed; inrervie\vs are everywhere, in the forms
of political polls, questionnaires about doctor's visits, housing applica-
tions, forms regarding social service eligibility, college applications, talk
shows, news programs-the list goes on and all. The interview as a means
of data gathering is no longer limited to use by social science re:;earcbers or
police detectives; it is a "universal mode of systematic inquiry" (Holstein
& Gubriu111, 1995, p. 1). It seems thar almost any type of qucstion-
p~rsonal, sensitive, probing, upsetting, accllsatory-is fair game and per-
01lssible in the interview setting. Almost all interviews, no matter their
purposes (and these can be varied-to describe, to interrogate, to assist, to
test, to evaluate), seek various forms of biographical description. As
Gubrium and Holstein (1998) have noted, the interview h2.s become a
~leans of Contemporary storytelling, where persons divulge life accounts
1I1.res ponse to interview inquiries. The media have been especially adept at
USing
. -I~ liS
. tecIlJ11que.
63
 IV''- I, ''-''-'..J '-'I '-V'-'-'-'-, ,,' ..... r" , ..... ,.." ,,...,'-,~,,,
As a society we rely on the interview and by and large take it for
granted. The interview and the norms surrounding the enactment of the
rt5Pondent and re,earcher role, have evolved to the point where they are
institutionalized and no longer require extensive training; rules and roles
arc known and shared. However, there is a growing group of individuals
who increasingly question the traditional assumptions of the interview-
we address their concerns in our later discussion of gendcrcd interviewing
and new trends in interview. Many practitioners continue to use and take
for granted traditional intervie\ving techniques. It is as if interviewing is
now part of the mass culture, so that it has acnmHy become the most feasi-
ble mechanism for obtaining information about individuals, groups, and
organizations in a society characterized by individuation, diversity, and
specialized role relations. Thus, many feel that it is not necessary to rc-
invent the wheel for each interview siruarion, as "interviewing has become
a routine technical practice and a pervasive, taken-for-granted activity in
our culture" (lvEshler, 1986, p. 23).
This is not to say, however, that the interview is so technical and the
procedures so standardized that interviewers can ignore contextual, soci~
etal, and interpersonal elements. Each interview context is one of interac~
rion and relation; the result is as much a product of this social dynamic as it
is a product of accurate accounts and replies. The interview has become a
routine, almost unnoticed, part of everyday life. Yet response rates con-
tinue to decline, indicating that fewer people are willing to disclose their
"selves" or that they are so overburdened by requests for interviews that
they are becoming more selective regarding which interviews to grant.
Social scientists arc more likely to recognize l however, that interviews are
interactional encounters and that the nature of the social dynamic of the
interview can shape the nature of the knowledge generated. Interviewers
v'lith less training and experience than social scientists may not recognize
that interview participants are "actively" constructing knowledge around
questions and responses (Holstein & Gubrium, 1995).
\VIe turn now to a brief history of interviewing to frame its roots and
development.
" The History of Interviewing
Onc form of interviewing or another has been with us for a vcry long time.
Even ancient Egyptians conducted population censuses (Babbie, 1992). In
64
..... ,-"" ",,_, ,'- ""
more recent times, the tradition of interviewing evolved from two trends.
First, intervic\ving found great popularity and widespread lise in clinicli
diagnosis and counseling, where the concern WJ.S with the quality of re-
sponses. Second, during \'x/orld \Vnr I intervicwing came to be '.videly em-
ployed in psychological testing; here the emphasis was on measurement
(Maccoby & .tvlaccobYl 1954).
The individual generally credited \vith being rhe first to develop a social
survey relying on interviewing was Charles Booth (Converse., 1987). In
1886, Booth embarked on a comprehensive survey of the economic and
social conditions of the people of London, published as Life awl Labour of
the Pco/J!e in Londoll (1902-1903). In his early study, Booth embodied
whar were to become separate interviewing methods, because he not: only
implemented survey research but triangulated his work by relying on
unstructured interviews and cthnographic observations:
The dara were checked <lnd supplemented by visits to many neighborhoods,
streets and homes, and by conferences with various wei f.lre ,lod co III 111 U ni ty
leaders. From time to time Booth lived as;1 lodger in districts where he was
nor known, so that he could become more intimately 3cqu3iIHcd with the
lives and habits of the poorer classes (Parten, 1950, pp. 6-7)
Many other surveys of London and other English cities follmved, pat-
terned after Booth's example. In the United States a similar pattern en-
sucd. Among others, an 1895 study attcmpted to do in Chicago \vhat
Booth had done in London (see Conversc, 1987), and in 1896, self-
admittedly following Booth's lead, the American sociologi~:t \YJ. E. B.
Du Bois Studied the black population of Philadelphia (see Du Bois, 1899).
Surveys of cities and small towns followed, most notable among them R. S.
Lynd and H. Ivt Lynd's MiddletOWN (1929) and Middletolll/1 iJlTi'<1llsitir)ll
(1937).
Opinion polling was another early form of interviewing. Some polling
~ook place well before the start of the 20th century, bllt it really came into
lts OWn in 1935 with the formation of the American Instinlt"c of Public
o .
P:nJon by George Gallup. Preceding G~lllup, in both psychology and
SOCIOlogy in the 19205 there was a movement toward the study (andusu-
ally measurement) of attitudes. \"X!. I. Thomas and Florian Zna~iecki used
the documentary method to introduce rhe study of attitudes in social psy-
chology. Thomas's influence, along with that of Robert Park, a former
rCpOrter who believed sociology was to be found ou[ in the field, sparked a
65
 /VIC I nvl..J.) vr ..... VLl..X ...... I II~'-.) M'~l.-' ...... , ~"""'--I "-II ''-' ... " " " " .............. ,,,,..., ... ' " ......... -' ",'- "oc,-' " ' - " . ' " .."" '.'<''''''''.''C-\./ ":'WC-..:>UVII..:> lU IH:yULtWll':U .'C,\l

number of community studies at the University of Chicago that came to be taking with him his market research and other applied grants, and became
known collectively as the \vorks of the Chicago schooL .Many other instrumental in the directing of the Bureau of Applied Sociai Research.
re5earchm wtre also grearly influenrial, such as Albion Small, George H. Tvw other "survey organizations 1' were ;llso formed: one In 1941, by
Mead, E. \XI. Ilurgess, Everett C. Hughes, Louis Wirth, \XI. Lloyd Warner, Harry Field, the National Opinion Research Center, first at Denver and
and Anselm Strauss (for a recent discussion of the relations and influence then at Chicago; and one in 1946, by Liken and his group, the Survey Rc~
of various Chicagoans, see Becker, 1999). search Center at ~/1ichigan.
Although the members of the Chicago school arc reputed to have used Academia at the time was dominated by theoretical concerns: and there
the ethnographic method ill their inquiries, some disagree, and have noted was some resistance toward this applied, numbers-based kind of sociology.
that many of the Chicago school studies lacked the analytic component of Sociologists and other humanists were critical of Lazarsfeld and the other
modern-day ethnography, and so were, at best, "firsthand descriptive survey researchers. Herbert Blumer, C. Wright lVlills, Arthur Schlesinger,
studies" (Harvey, 1987, p. 50). Regardless of the correct 13bel for the Chi- Jr., and Pitirin Sorokin, among others, voiced their displeasure. Accord-
cagoans' fieldwork, they clearly relied on a combination of observation, ing to Converse (1987), Sorokin felt that "the new cmphJsis on quan-
personal documents, and informal interviews in their studies. Interviews titative work was obsessive, and he called the IlC\V practitioners
were especiaIly in evidence in the \vork of Thrasher (1927/1963), who in 'quanrophrenics'-with special reference to Stouffer and Lazarsfeld"
his study of gang members relied primarily on about 130 qualitative inter- (p. 253). And Converse quotes .Mills: "Those in the grip of the method~
views, and in that of Nels Anderson (1923), whose c13ssic study of hoboes ological inhibition often refuse to say anything about modern society
relied on informal, in-depth conversations. unless it has been through the fine little mill of the Statisrical Ritual"
It was left to Herbert Blumer and his former student Howard Becker to (p. 252). Schlesinger, Converse notes, called the survey researchers "social
formalize and give impetus to sociological ethnography in the 1950s and relations hucksters" (p. 253).
1960s, and interviewing began to lose both the eclectic flavor given to it by But the survey researchers had powerful allies also, such as [vlerton,
Charles Booth and the qualitative accent of the Chicagoans. Understand- who joined the Survey Center at Columbia in 1943, and government mon~
ing gang members or hoboes through interviews lost importance; what eys were becoming increasing available for survey research. The 1950s
became relevant was the use of interviewing in survey research as a tool to saw a grO\vth of survey research in the universities and a proliferation of
quantify data. This waS not new, as opinion polls and market research had survey research texts. Gradually, survey research increased its domain
been doing it for years. But during \X1orld \Var 11 there \vaS a tremendous over sociology, culminating in 1960 with the election of Lazarsfeld to the
increase in survey research as the U.S. armed forces hired great numbers of presidency of the American Sociological Association. The methodological
sociologists as survey researchers. l\1ore than half a million American sol- dominance of survey research continued unabated through rhe 1970s,
diers \vere interviewed in one manner or another (Young, 1966), and their 19805, and 1990s, although other methods began to erode the promi-
mental and emorionallives were reported in a four-volume survey titled nence of survey methods.
Studies in Social Psychology ill \'{Iorid Wlar II, the first two volumes of \vhich Qualitative intervievving continued to be practiced, hand in hand with
were directed by Samuel Stouffer and titled The Americall Soldie1: This participanr observation methods, but it too assumed some of the quanrifi-
\\lork had tremendous impact and led the way to widespread use of sys~ a~le scientific rigor that so preoccupied survey research. This was espe-
tematic survey research. clal,ly visible in grounded theory (Glaser & Strauss, 1967), with its pains-
\X1hat was new, hmvcver, waS that quantitative survey research moved takmg emphasis on coding data, and in ethnomethodology, with its
into academia and came to dominate sociology as the method of choice for quest for invariant properties of social action (Cicourel, 1970). Other
the next three decades. An Austrian immigrant, Paul LazarsfeId, spear- G.ualitative researchers suggested variations. John Lofland (1971) criti-
ed grounded theory for paying little atten~ion to data gathering tech-
headed this move. He welcomed The American Soldier with great enthusi- Ci.Z

asm. In fact, Robert lvlerton and Lazarsfeld (1950) edited a book of reflec- ~Ique~) Jack Douglas (1985) suggested lengthy, existential one~on-one
JntcrvlewS 1 '
tions on The American Soldier. Lazarsfeld moved to Columbia in 1940, .astmg one or more days, and James Spradley (1980) tried to

66 67
 IV, ... " , ........... .,) V ' ..... '-' .............. , " '''''''-'' ,'-'
clarify the difference between ethnographic observation and ethno-
graphic interviewing.
Rccemly, posrmoderniH erhnogr~phm h~ve
witll some of the assumptions present in llltcrvlewlllg and with rhe con-
trolling role of the interviewer. These concerns have led to new ditections
in qualitative interviewing focusing on increased attention to the voices of
the respondents (Ivlarcus & Fischer, 1986), the interviewer-respondenr
relationship (Crapanzano, 1980), the importance of rhe researcher's gen-
der in interviewing (Gluck & Patai, 1991), and the roles of other elements,
such as race, social status, and age (Seidman, 1991).
Structured Interviewing
In structured interviewing, the interviewer asks aU respondents the same
series of preestablished questions with a limited set of response categories.
There is generally little room for variation in responses except where open-
ended questions (vvhich are infrequent) may be used. The interviewer
records the responses according to a coding scheme that has already been
established by the project director or research supervisor. The interviewer
controls the pace of the interview by treating the questionnaire as if it were
a theatrical script to be followed in a standardized and straightforward
manner. Thus all respondents receive the same set of questions asked in
the same order or sequence by an interviewer who has been trained to treat
all intervievv situations in a like manner. There is very little flexibility in
the way questions are asked or answered in the structured interview set-
ting. Instructions to interviewers often include some of the following
guidelines:
- Never get involved in long explanations of the study; use the standard ex-
planation provided by the supervisor.
+ Never deviate from rhe study introduction, sequence of questions, or ques-
tion wording.
'> Never let another person interrupt the interview; do not let another person
answer for the respondent or offer his or her opinions all the question.
<r Never suggest an answer or agree or disagree with an answer. Do not give the
respondent any idea of your personal views on the topic of rhe question or
the survey.
n, "., ..... , ... " ........... "" ,.,,_<.~
* Never imerpret the meaning of a question; just repeat the que~tion :mel give
instructions or cbrificalions th:il are provided in training or by the sllper-
concerned. themselves visors.
+ Never improvise, such as by adding answer cnegories or m<lking wording
changes.
Inrerviev.'s by telephone, f<lce-to-face interviews in respondents' house-
holds, intercept interviews in malls and parks, and interviews generally
associated with survey research are most likely to be included in the struc-
tured interview category.
This interview context calis for the interviewer to pi.Jy a neutral rolc,
never interjecting his or her opinion of a respondent's 'U1S\Ver. The inter-
viewer must establish what has been called "balanced rappon"; he or she
must be casual and friendly all the one hand, bur directive and impersonal
on the other. The interviewer must perfect a style of "interestcd listening';
that rewards the respondent's participation but does not evaluate the
responses (Converse & Schuman, 1974).
In a structured interview, hopefu liy, nothi ng is left to chance. However,
response effects, or nonsampling errors, that can be attribUted to the qucs-
tionnaire administration process commonly evolve from three Sources.
The first of these is res pond ent behavior. The respondent may deliberately
try to please the interviewer or to prevent the interviewer from learning
something abom the respondent. In order to do this, the respondent may
embellish a response, give what is described as a "socially desirable" re-
sponse, or omit certain relevant information (Bradburn, 1983, p. 191).
The respondent may also err due to faulty memory. The second SOurce of
error is found in the nature of the task: the method of questionnaire
Jdministration (face-to-face or telephone) or the sequence or wording of
the questions. The third source of errOt is the interviewer, whose charac-
teristics or questioning techniques can impede proper communication of
:he questions (Bradburn, 1983). It is the degree of error assigned to the
Interviewer that is of greatest concern.
. Most Structured interviews leave little room for the inre:vicwer to
Improvise or exercise independent judgment, bur even in the moSt struc-
tured interview situation not every contingency can be anticipated, and
~Ot every interviewer behaves according to the script (Bradburn, 1983;
hey, 1989). In fact, one study of interviewer effects found that intervie\v-
Crs Changed the wording to as many as one-third of the questions
(Bradburn, Sudman, & Associates, 1979).
68 69
 ", " ,'-''-'-.' '-" ," ,,,.~, ~. -<>.- _..
In general, research on interviewer effects has shown interviewer char-
acteristics such as age, gender, and interviewing expericnce to have rel-
atively 5mall impact on re5ponses (Singer & Presser, 1989). However,
there -is some evidence that student interviewers produce larger response
effects than do nonstudents, higher-status interviewers produce larger
response effects than do lower-status interviewers, and the race of an
interviewer makes a difference only on questions specifically related
to race (Bradburn, 1983; Hyman, 1954i Singer, Frankel, & Glassman,
1983).
The relatively minor impact of the interviewer on response quality in
structured interview settings is directly attributable to the inflexible, stan-
dardized, and predetermined nature of this type of interviewing. There is
simply little room for error. However, those who are advocates of struc-
tured interviewing are not unaware that the interview takcs place in 3
social interaction context and that it is influenced by that context. Good
interviewers recognize this faet and are sensitive to how interaction can
influence responses, Converse and Schuman (1974) observe, "There is no
single interview style that fits every occasion or all respondents" (p. 53).
This means that interviewers must be aware of respondent differences and
must be able to make the proper adjustments called for by unanticipated
developments. As Raymond Gorden (1992) states, "Interviewing skills arc
not simple motor skil!s like riding a bicycle: ratber, they involve a high-
order combination of observation, empathic sensitivity, and inteJlecrual
judgment" (p. 7).
It is not enough to understand the mechanics of intervievving, it is also
important to understand the respondent's world and forces that might
stimulate or retard response (Kahn & Cannel, 1957). Still, the structured
interview proceeds under a stimulus-response format, assuming that the
respondent will truthfully answer questions previously determined to
reveal adequate indicators of the variable in question, as long as those
questions are properly phrased. This kind of interview often elicits ratio-
nal responses, but it overlooks or inadequately assesses the emotional
dimension.
"* G roup Interviews
The group interview is essentially a qualitative data gathering techniqL:c
(see Madriz, Chapter 10, this volume) that relies upon the systematic
70
~ ~, "~'" "<'>'" ~., ............"" .... '-' "':: ..........."'v",, 'v, '''':1v"uc"",--, ,,,.,,,
questioning of several individuals simultaneously in a formal or informal
setTing, Thus this techniquc straddles the line between formal and inror-
11131 interviewing.
The use of the group interview h<ls ordinarily been associated with mar-
keting research under [he label of foclls group, where the purpose is to
gather consumer opinion on product characteristics, advertising rhemes,
or service delivery. This format has also been used to a considerable ex-
tent by polirical parties and candidates who are inreres[ed in voter re-
action to issues and policies. The group interview has also been llsed in
sociological research. Bog8rdus llsed it to rest his social distance scale in
1926, Zuckerman (1972) interviewed Nobel laureates, Thompson and
Demerath (1952) looked at management problems in the military,
1vlorgan and Spanish (1984) studied health issues, we investig~lted older-
worker labor force reentry (Fontana & Frey, 1990), and l\'Icnon and his
associates studied the impact of propaganda using group interviews (see
Frey & Fontana, 1991). In fact, 1V1enon, Fiske, and Kendall (1956) coined
the term foclIs grollfJ to apply ro a si[uation in \vhich the rescarcher!
interviewcr asks very specific questions about a topic after having already
completed considerable research. There is also some evidence rhar estab-
lished anthropologists such as l'vlalinO\vski used this technique, although
they did not report it (Frey & Fontana, 1991). Today, all group interviews
are often generically designated foclls group imerviews, even though therc
are considerable variations in the natures and types of group interviews.
In a group interview, the interviewer/modercnor directs the inquiry and
the interaction among respondents in a very structured fashion or in;] very
unstructured manner, depending on the interview's purpose. The purpose
may be exploratorYi for exam pIc, the researcher may bring se vet::l! persons
together to test 3. methodological technique, to try our a definition of a re-
search problem, or to identify key informants. An extension of this explor-
atory intent is the use of the group intervie\v for the purpose of pretesting
questionnaire wording, measurement scales, or other elemenrs of a survey
d~sign. This is now quite common in survey research (Desvousges& Frey,
] 989). Group interviews can also be used sllccessfully to aid respondents'
recall of specific events or to stimulate embellished descripriom of events
(e.g., a disaster or a celebration) or experiences shared by members of a
group. Group interviews can also be used for triangulation purposes or
can be used in conjunction with other dara gathering techniques. For
('~;lmple, group interviews could be helpful in rhe process of "indefinite
tnanguhtio n, "b y plltt1I1g
L.
. .III d"d
IVI U8 I responses '
IntoC
a context ('Icoure, I
71
 Ml:! HUU~ Ul- LVLLtL III'H::J AI'ILJ Al'l ...... LT t..11'!\.:l !:IV\F1r\".... MI.. IV,,...., 1..''''"''--'
"'''' "'<""Y'''''Y. 'JU'" -'UUl.lUIt:U ",,!ul:.'~tlurl~ tU rveyUUUleu leX[

1974). Finally, phenomenological purposes may be served whether group TABLE 2.1 Types of Group Interviews and Dimensions
interviews arc the sole basis for gathering data or they arc used in associa-
tion with other techniques. Role of QUEStion
Group interviews can talee different forms depending on their pur- Type Setting Interviewer Fornlot Purpo~e
poses. They can be brainstorming sessions with little or no structure or
direction from the interviewer, or they can be very structured, as in nomi- Focus group formal- directive structured exploratory
nal, Delphi, and marketing focus groups. In the latter cases the role of the preset pretest
interviewer is very prominent and directive. Fieldwork settings provide
Brainstorming formal or nondirective very exploratory
both formal and informal occasions for group interviews. The field re- informal structured
searchet can bring respondents into a formal setting in the field context
and ask very directed questions, or a natural field setting, such as a street Hominol! formal directive structured pretest
corner or a neighborhood tavern, can be a conducive setting for casual but delphi
e>:plorotory
purposive inquiries.
Field, informal moderately very
Group interviews can be compared on several dimensions. First, the exploratory
natural spontaneous nondireclive structured
intervie\ver can be very formal, taking a very directive and controlling phenomenological

posture, guiding discussion strictly, and not permitting digression or varia~ Field, preset, but somewhat semi~ phenomenological
tion from topic or agenda. This is the mode of focus alld nominal/Delphi formal in field direclive structured
groups. In the latter case participants arc physically isolated but share
views through a coordinator/interviev,'er. The nondirective approach is
more likely to be implemented in naturally established field settings, such
as a street corner, or in controlled settings (e.g., research labs) where the problems: First, the interviewer must keep one person or small coalition
research purpose is phenomenologil;:al, to establish the \videst range of of petsons from dominating the group; second, the interviewer must
meaning and interpretation for the topic. Groups can also be differenti- e~courage recalcitrant respondents to participate; and third, the inter-
ated by question format and purpose, which in the case of group inter~ Viewer must obtain responses from the entire group to ensure the fullest
vie"ws usually means exploration, pretest, Or phenomenological. Explor- Coverage of the topic. In addition, the interviewer must balance the direc-
atory interviews are designed to establish familiarity with a topic or tive, interviewer role with the role of moderator, which calls for the man-
setting; the interviewer can be very directive (or the opposite), but the a~cl11ent of the dynamics of the group being interviewed; the group inter-
questions are usually unstructured or open-ended. The same format is viewer must simultaneously worry about the script of questions and be
used in interviews with phenomenological purposes, where the intent is to Sensitive to the evolving patterns of group interaction.
tap intersubjeetive meaning with depth and diversity. Pretest intervie\;vs Group intervievi1s have some advantages over individual interViews.
are generally structured in question format and the interviewer is directive Ihev are '[ . I . . c
. re atlve y lllexpensive to conduct and often produc,~ rich data
in style. Table 2.1 compares the types of group interviews on various ~hat are. cumulative and elaborative; they can be stimulating for respon-
dimensions. hems, aldil1.g recall; and the format is flexible. Group interviews are nor,
The skills that are required to conduct the group interviev./ arc not . O\vever, Without problems: The results cannot be generaiized' the emerg-
significantly different from those needed for individual interviews. The Ing group culture may inrerfere with individual expression, an~l the gro~:l
l11ayb d . ,
interviewer must be flexible, objective, empathic, persuasive, a good li5- I[' e ~ITIlnated by one person; and "groupthink" is a possibl,~ Outcome.
rener, and so on. But the group format does present some problems not 1e reqUlre f' . .
"d . "rnents or llltervievver skills are greater than those for illdi-
found in thc individual interview. Merton et al. (1956) note three specific \1 unl Interv ". I f [ .
. "Iewmg lecause 0 t 1C group dynamtcs that are present. In

72 73
 IVle I nVlJ.) vr- I....vl...l...l...l....lll',D I-\'~'-' 1-\"'1-\1...1 L " " D L.lV"-
addition, it is difficult to research sensitive topics using this rechnique.
Nevertheless, the group interview is a viable option for both qualitative
and quanritariYE [E'Earch.
+ Unstructured Interviewing
Unstructured interviewing can provide a greater breadth of data than the
other types, given its qualitative natllre. In this section we discuss the tradi-
tional type of unstructured interview: t'he open-ended, ethnographic (il1-
depth) imerview. ?\Ilany qualitative researchers differentiate between in-
depth (or ethnographic) interviewing and parricip;]m observation. Yet, as
Lofland (1971) points out, the two go hand in hand, and many of the data
gathered in participant observation come from informal interviewing in
the field. Consider the following report, from Malinovi-'ski's (1967/1989)
diary:
Saturday 8 [December 1917]. GO[ up latc, fclr rotten, rook enema. Ar about
1 1 went our; I heard cries; [people from] Kapw~lpll were bringing uri to
Teyava. I sar wirh the n3tives, talked, look pictures. \Vent b3ck. Billy cor-
rected 3nd supplemented my notes about wasi. ArTey;]va, an old Illan talked
a great de31 about fishes, bur I did not undersrand him ton welL Then we
moved to his bluayallw. Talked about fjfj'lI. They kept quesrioning me 3bour
the war~In the evening 1talked to the policeman about bluogo'/f, lifi'll and
)'oyou.1. I was irritated by their laughing. Billy again told me <l number of in-
teresting things. Took quinine and calomel. (p. 145)
IVlalinowski's "day in the field" shO\vs how very importanr unstruc-
tured imervie-wing is in the conduct of fieldwork and clearly illustrates
the difference between structured and unstructured interviewing.
1vlalinowski has some general topics he wishes to know about, but he does
not use closed-ended questions or a formal approach to interviewing.
\\That's more, he commits (as most field-workers do) what structured
interviewers would see as two "capital offenscs": (a) He answers questions
asked by the respondents, and (b) he lets his personal feelings influence
him (as all field-workers do), thus he deviates from the "ideal" of a cool,
distant, and rational intervie-wer.
1Vlalinowski's example captures the differenccs bcnvccn structured and
unstructured interviewing: The former aims at capturing precise data of;]
74
"~''-ML !VIM' '-"'M'---'
codable nature in order to explain behavior \vithin preestablished catego-
ries, \vhereas the latter attempts to understand the complex behavior of
members of society without imposing any apriori categorization that may
limit the field of inquiry.
In a way, lvlalinowski's interviewing is still structured to some degree-
tint is, there is a setting, there are identified informants, and the respon-
dents are clearly discernible. In other types of intcrviewing there may bc
no setting; for instance, Rosanna Hertz (1995, 1997b, 1997c) focused on
locating women in a historic moment rather than in a place. Additionally,
in their study of single mothers, Henz and Ferguson (1997) interviewed
women who did not know each other, who were not part of a single group
or village. At times, informants are not readily accessible or identifiable,
bur anyone the researcher meets may become a valuable source of infor-
mation. Hertz and Ferguson relied on tradespeople and friends to identify
single mothers for their study. Fontana and Smith (1989) found that
respondents are not always readily identifiable. In studying Alzheimer's
disease patients, they discovered it was often possible to confuse care-
givers and patients in the early stages of the disease. Also, in Fontana's
(1977) research on poor elderly, he had no fixed setting at all; he simply
wandered from bench to bench in the park where the old folks were sit-
ting, ralking to any disheveled old person who would talk back.
Spradley (1979) aptly differemiates among various types of interview-
ing. He describes the following interviewer-respondent interaction, which
would be unthinkable in traditional sociological circles yet is d1e very
essence of unstructured interviewing-the establishment of a human-to-
human relation with the respondent and the desire to understand rather
~han to explain:
Presently she smiled, pressed her hand t'O her chest, and said: "Tsetchwc." It
was her mme. "Elizabeth," I said, poiming to myseiL "Nisabe," she an-
swered.... Then, having surely sllspecred rhm I was a \voman, she puther
hand on my breast gravely, and, finding our dut I was, she touched her own
breast. Many Bushmen do this; to them all Europeans look alike. "l:lsl1 si"
(women), she said. Then 3frer a moment's pause Tsetchwe began to teach
me. (pp. 3-4)
~pradley goes on to discuss all the things an interviewer learns rom the
~~tJves about them, their culture, their language, their ways of life.
1 though each and every study is different, these are some of the basic
75
 IVle I nULl.) u r '-Ul..LL,-1 ""\.:l MI'I'-' M'''MI..' ",-Il'IU l..!V1r '''''-Ml.. ",,....,, l..",,...,l..-'
elements of unstructured interviewing. These dements have been dis-
cussed in details already, and we need not elaborate upon them roo much
(for dtuiled JCCDUnLc of um:rruc[[lI'ed intt;lrvil?wil1Q;, !=:cc, :1lnong otlwrs,
Adams & Preiss, 1960; Denzin, 1989b; Lofland, 1971; Spradley, 1979).
Here we provide brief synopses. Please remember that these arc presented
only as heurisric devices; every study uses slightly different elements and
often in different combinations.
Later in this chapter, in discussing ne\v trends, we will deconstruct these
notions as we frame the interview as an activc, emergent process. We con-
tend that our interview society gives people instructions on how to comply
with thesc heuristics (see Silverman, 1993, 1997.1, 1997b). Similarly,
James Scheurich (1997) is openly critical of both positivistic and interpre-
tive interviewing, as they afe both based on modernist assumptions.
Rather than being a process "by the numbers," for Scheurich, inrervie\Ning
(and its language) are "pefsistently slippery, unstablc, and ambiguous from
person to person, from situation to situation, from time to time" (p. 62).
Accessing the setting. How do we '''get in"? That, of course, varies accord-
ing to the group one is attempting ro study. One may have to disrobe and
casually stroll in the nude if doing a study of nude beaches (Douglas &
Rasmussen, 1977), or one may have to buy a huge motorbike and frequent
seedy bars in certain locations if attempting to befriend and study the
Hell's Angels (Thompson, 1985). The different ways and attempts to "get
in" vary tremendously, but they all share the common goal of gaining
access to the setting. Sometimes there is no setting per se, as when Fontana
(1977) attempted to study poor elderly on the streets and had to gain
access anew with each and every interviewee.
Understanding the language and clilture of tbe respondents. Rosalie Wax
(1960) gives perhaps the most poignant description of learning the lan-
guage and culture of the respondents in her study of "disloyal" Japanese in
concentration camps in the United Stares between 1943 and 1945. Wax
had to overcome a number of language and cultural problems in her study.
Although respondents may be fluent in the language or the interviewer,
there are different ways of saying things and, indeed, certain things that
should not be said at all, linking language and cultural manifestations. \XTax
makes this point:
76
I remarked rhat I would like to sec the letter. The silence th.1t fdlon the
chatting group was almost p~1Ip;]ble, and rhe embarrassmenr of rhc hosrs \\',15
p:linful to see. Tlw /IIL-': /-',15 WilS II'..Jt ilsking, LV jet il letter. for letters were
passed abom rarher freely. It rested all the {ncr th;lt one did not givc n Cluca-
sian a lerter in which rhe "disloyal" sratemcnt o{ ;1 friend might be ex-
pressed. (p. 172)
Some researchers, especially in anrhropological imervie\vs, ret'ld to rely
on interpreters, and thus become vulncr3ble to added byers of meanings,
biases, and interprew['ions, \\'hich illay lead to dis3srrous misunderstand-
ings (Freeman, 1983). At times, specific jargon, such as the medical meta-
language of pbysiciJns, may be ;] code rlut is hard for nonmembers 1:0
understand.
Deciding all how to present ollesel/ Do we present ourselves as represcnta-
tives from academia smdying medical studems (Becker, 1956)? Do we
approach the intervinv as a \Voman-to~Wo111an discussion (Spradley,
J 979)? Do we "dress down" to look like the respondems (Fonuna, 1977;
Thompson, 1985)? Do \ve represent the colonial culture (lVlalinowsky,
1922), or do we humbly present' ourselves as "learners" (\\lax, 19(0)? This
decision is very important, because once the inrerviewer's preselHational
self is "cast," it leaves a profound impression on the respondents and has
great influence over the success (or lack of it) of the study. Sometimes,
inadvertently, the researcher's presentational self may be misrepresentcd,
as John Johnson (1976) discovered in studying ;:\ welfare office, when
SOme of the employees assumed he was a "spy" for management despite
his best efforts to the contrary.
Locating an informallt. The researcher must find an insider, a rnember of
the group studied, who is willing to be an informant and act as a guide and
a translator of cultural mores and, at times, jargon or language. Although
the researcher can conduct interviews witham an informant, he or she can
saVe much time and avoid mist3kes if a good informant becomes available.
The "classic" sociological illformanr is Doc in \\!iUiam Foote \vhytc's
Street Comer Society (1943). \X1ithollt Doc's help and guidance, it is
dOubtful that \vhyte would have been able to access his subjects Jt the level
he.did. Very instruccive is Paul RabinO\v's (1977) discussion of hi:; relation-
slup with his main informant, Abel al-IvIalik ben Lahcen. lvlalik ~eted as a
77
 METHODS OF COLLECTlNG AND ANAlY LIN\.;> tMr'II'\lLP.L /VIAl CI'IHLJ
translator bur also provided Rabinovv with access to the cultural ways of
the subjects, and by his actions provided Rabinow with insights into the
V;;l.,,;t: ,J;({ercnces l"wrwccn :l Univ(>t'siry of ChiclgO researcher and a native
Jvloroccan.
Gaining tmst. Survey researchers asking respondents whether they would
or would not favor the establishment of a nuclear dump in their state (Frey,
1993) do not have too much vvork to do in the way of gaining trust;
respondents have opinions about nuclear dumps and are very wil1ing to
express them, sometimes forcefully. Bm it is clearly a different story if one
\\'ants to ask about a person's frequency of sexual intercourse or preferred
method of birth control. The interviewer needs to establish some trust
with the respondents (Cicourcl, 1974). Paul Rasmussen (1989) had to
spcnd monrhs as a "wallflO\vcr" in the waiting room of a massage parlor
before any of the masseuses gained enough trust in him to divulge to him,
in unstructured interviews, the nature of their "massage" relations with
clients. Gaining trust is essential ro the success of the interviews and, once
gained, trust can still be very fragile. Any faux pas by the researcher may
destroy days, weeks, or monrhs of painfully gaincd trust.
ESfablishhzg rapport. Because the goal of unstructured interviewing is
IwderstLlnding, it is paramount that the researcher establish rapport with
respondents; that is, the researcher must be able to take the role of the
respondents and attempt to see the situation from their viewpoint, rather
than superimpose his or her world of academia and preconceptions upon
them. Although a close rapport with the respondents opens the doors to
more informed research, it may create problems as the researcher may
become a spokesperson for the group studied, losing his or her distance
and objectivity, or may "go n:1tive" and become a member of the group and
forgo his or her academic role. At times, what the researcher may fe~l is
good rapport turns out not to be, as Thompson (1985) found out 111 a
nightmarish way when he was subjected to a brutal beating by the Hell's
Angels just as his study of them was coming to a close. At the other end of
the spectrum, some researchers may never feel they have established rap-
parr with their subjects. Ivlalinowski (1967/1989), for example, aJwa}:s
mistrusted the motives of the natives and at times was troubled by their
brutish sensuality or angered by their- outfight
- 1ymg
- or deceptlons:
- "fVter
lunch I [carried] yellow calico and spoke about the balama. I made a small
sagali, Navavile. I was fed liP with the niggers" (p. 154).
78
",e- ",eC' .JCVV. "U"J .JlJULlUJt:U \,fut::>lIUII:> {U I'/eyuuarea leXT
Collecting empirical malerials. Being Ollt in the field does not afford
researchers the luxury of video cameras, soundproof rooms, ;;ond high-
quality recording equipment. Lofland (J 971) provides detailed informa-
tion on doing and writing up interviews and on the types of field notes
researchers ought to take and how to organize them. Yet field-workers
often must make do; their "tales" of their methods range from holding a
miniature tape recorder as inconspicuously as possible to taking mental
notes and then rushing to the privacy of a bathroom to jot notes dovvn, on
toilet papers at times. W/c agree with Lofland that regardless of the circum-
stances, researchers ought to (a) take notes regularly and promptly: (b)
write everything dO\vn, no matter how unimportant it may seem at the
time; (c) try to be as inconspicuous as possible in note raking; and (d) ana-
lyze their notes frequently.
Other Types of Unstructured Interviewing
\Y,/e consider the issues of interpreting and reporting empirical material
later in this chapter. In this section, we briefly outline some different types
of unstructured interviews.
Oral History
The oral history differs from other unstructured interviews in purpose,
but not methodologically. The oral collection of historical materials goes
back to ancient times, but its modern-day formal organization can be
t: aced to 1948, when Allan Nevins began the Oral History Project at
Columbia University (Starr, 1984, p. 4). Oral history captures a variety of
forms of life, from common folks talking about their jobs in Studs Terkel's
Wlo,-hi"" - I reco II ectlOns 0 f preSident
.::, (J975) to t1le h-lstonca -
. Harry Truman
C -
~~rIVlerle 1'vli.l1er's Plain Speak!ng (1974; see Starr, 1984). Often, oral his-
__ j y transcrIpts are not pubiJshed, but many may be found in libraries,
5\ cntme .. f or someone to rummage through them and bring
. mOlrs wJmng
h
tCIrtesri lcR ecem I y, oral hIstory
mOllY to .IIC.
- has found great popularinr
anton C ' . _ .
and g.lemlDlsts (Gluck & Patai, 1991), who see it as a \va}' to understand
, bnng forth the history of women in a culture that has traditionally
on masculine intcrpretation: "Refusing to be rendered historically
any longer, "vomen arc creating a new history-using our own
and experiences" (Gluck, 1984, p. 222)_
79

Relevant to the study of oral history (and, in fact, to all intervic\ving) is
the smell' of memory and it~ relation lO recall. For instance, Barrv
Schwanz (1999) has examined rhe ages at \vhich \ve recall critical episode's
in our lives; he concludes thar "biographical memory ... is better under-
slOod as :::I social process" and that "as we look back, we find ourselves
remembering our lives in terms of our experience v,rith others" (p. 15; see
also Schwartz, 1996). Carolyn Ellis (1991) has resorted lO the usc of
"sociological introspection" to reconstruct biographical episodes of her
past life. Norable among Ellis's works in this genre is her reconstruction of
her 9-l'ear relationship with her parmer, Gene \Xleinstein, in which she
describes the emotional negQ[i~~tions the two of them went through as they
coped with his dowI1ward-spir:::.ling health, until the final negotiation with
dearh (Ellis, 1995).
Creative Interviewing
Close to oral hislOry, bur wied more conventionally as a sociological
tool, is .lack Douglas's (1985) "creative interviewing." Douglas argues
against "how-to" guides to conducring interviews because unstructured
interviews rake place in the largely situational everyday worlds of mem-
bers of society. Thus interviewing and interviewers must necessarily be
crearive, forget how-to rules, and adapr themselves to the ever-changing
situations they face. Similar to oral historians, Douglas sees interviewing
as collecting oral reports from the members of society. In creative inter-
vievving, these reports go well beyond the length of conventional unstruc-
tured interviews and may become "life histories," with interviewing rak-
ing place in multiple sessions over many days with the subjecr(s).
Postmodern Interviewing
Douglas's concern with the important role played by thc interviewcr
qua human being, which is also shared by feminist oral historians, became
a paramount element in the interviewing approaches of postmodern
anthropologists and sociologists in the mid-19S0s. lVlarcus and Fischer
(1986) address ethnography at large, but their discussion is germane to
unS1Tuctured interviewing becltlse, a~ ,Vi;' h8ve seen, such intcl-vic,ving
constitutes the major way of collecting dara in fieldwork. Marcus and
Fischer voice reflexive concerns abollt the ways in which rhe researcher
influences the srudy, both in the methods of (bta collection and in the
80
I "t:;;" ""C' Y'<:o"_ ',U", ..JLIW ..... 'u'<:ou '""U<:O..>UV,,-' <v, ,<.::::/ul.,u,-,:.:u ,"',,'
rechniqucs of reporting findings; this concern leads to new ways to con-
duct interviews, in rhe hope of minimizing, if not eliminating, interviewer
influence. One such way is polyphonic interviewing, in \vhich the voices of
rhe subjects are recorded with minimal influence from the researcher and
are nor collapsed together and reporred as one, th rough the intcrpretation
of rhe researcher. Instead, the multiple perspecrivcs of the various subjects
are reported and differences and problems encountered are discussed,
rather rhan glosscd over (see Krieger, 1983). Illterpretiue interactionism
follows in the foorsreps of creative and polyphonic interviewing, but, bor-
ro\ving from James Joyce, adds a new element, that of epiphanies, which
Denzin (1989a) describes ~lS "those interactional moments thar leave
marks on people's lives [and] have the potential for creating transfor-
marional experiences for the person" (p. 15). Thus rhe topic of inquiry
becomes dramatized by the focus on exisrenrial moments in people's lives,
hopefully producing, richer and more meaningful dara. Finally, as post-
modernists seek new ways of understanding and reporting data, \ve \vish
to note the concept of oralysis, which refers "ro the \vays in which oral
forms, derived from everyday life, are, with the recording powers of
video, applied ro rhe analyrical tasks associated wirh literate forms"
(Ulmer, 1989, p. xi). In oralysis, rhe rradirional product of interviewing,
talk, is coupled with rhe visual, providing, according to Ulmer (1989),
a product consonant with a society thar is dominated by the medium of
television.
.. Gendered Interviews
The housewife goes into 3. well-stocked store to look for 3. frying ]J3.n. Her
thinking probably does not proceed exactly this W3Y, bur it is hdpful to
think of rhe many possible two-way choices she mighr l11:1ke: Cast iron or
aluminum? Thick or thin? lVletal or wooden handle? Covered or not? Deep
or shallow? L3rge or small? This brand or that? Reasonable or toO high in
price? To buy or not? Cash or charge? I-lave it delivered or carry it. .. , The
two-way question is simplicity itself when it comes to recording answers
and tabulating them. (Payne, ] 951, pp. 55-56)
The above quote represents the prevalent paternalistic attitude toward
Women in interviewing (see Oakley, 1981, p. 39) as well as the paradig-
matic concern with coding answers and therefore with presenting limited,
81
 IV1!: I nUl.)-J U , \..Ul..l..l..\.., lI"\..J MI~L.J M'~Ml..1 '<""-''-.J LlV" '''''-...'"'1... IV',...., I...''''''''I.....J I fit: IflLt:rVlew: rrom .)rructureo \.duestlons to Negotiated "{ext

dichotomous choices. Apart from a tendency to be condescending to with the women \vhile the lllen gathered in the parlors or in front of the
women, the traditional interview paradigm does not account for gendered compound.... I never entered any of the places where men sat around to
differences. In fact, Babbie's classic text The Practice of Social Research drink beer or palm wine and to chat" (Sudarkasa, 1986; quoted in \V'arren,
(1992) briefly references gender only three times and says nothing about 1988, p. 16).
the influence of gcnder on imerviews. As Ann Oakley (1981) cogently Solutions to the problem have been to view the female anthropologist
points out, bmh the interviewers and the respondents are considered face- as androgyne or to grant her honorary male status for the duration of her
less and invisible, and they must be if the paradigmatic assumption of gath~ research. \X'arren (1988) also poinrs to some advantages of a researcher's
ering value-free data is (Q be maintained. Yet, as Denzin (1989a, p. 116) being female and therefore seen as harmless or invisible. Other problems
tells us, "gender filters knowledge"; that is, the sex of the interviewer and are associated with the researcher's status and race and with the context of
that of the respondent do make a difference, as the interview takes place the interview; again, these problems are magnified for female researchers
\vithin the cultural boundaries of a paternalistic social system in which in a paternalistic worlel. Female inrervievvers at times [ace the added bur-
masculine identities are differentiated from feminine ones. den of sexual overtures or covert sexual hassle (\Varren, 1988, p. 33).
In the typical interview there exists a hierarchical relation, with the Feminist researchers have suggested \vays to circumvent the traditional
respondcnt being in the subordinate position. Thc interviewer is in- interviewing paradigm. Oakley (J 981) notes that interviewing is a mascu-
structcd to be courteous, friendly, and pleasant: line paradigm, embedded in <1 masculine culture and stressing masculine
traits while at the same time excluding traits such as sensitivity, emotional-
The interviewer's manner should be friendly, courteous, conversational and ity, and others that are culturally viewed as feminine rraits. There is, how-
unbiased. He should be neither too grim nor too effusive; neithcr toO talk- ever, a growing reluctance, especially among female researchers (Oakley,
ative nor too timid. The idea should be to pur the respondent at ease, so that 1981; Reinharz, 1992; Smith, 1987), to continue interviewing women as
he will lath (reely and fIllly. (Sel1tiz, laboda, Dcutsch, & Cook, 1965, "objects," with little or no regard [or them as individuals. Although this
p. 576; emphasis added)
reluctance stems from mora! and ethical reasons, it is also relevant meth-
odologically. As Oakley (1981) points out, in interviewing there is "no
Yet, as the last above-quoted line shows, this demeanor is a ruse to gain the intimacy withom reciprocity" (p. 49). Thus the cmphasis is shifting to
trust and confidence of the re:;pondent without reciprocating those fce!~ allow the development of a closer relation between interviewer and re-
ings in any way. Interviewers a:::e not to give their own opinions and are to spondent; researchers arc attempting to minimize status differences and
evade direct questions. \X!hat ;;cems ro be a convcrsation is really a one~ are doing away v-lirh the traditional hierarchical situation in interviewing.
\vay pseudoconversation, raising the ethical dilemma (Fine, 1983-1984) Inre.rvie\vers can show their human side and answer questions and express
inherent in the study of people for opportunistic reasons. When the re- fcchngs. 1vlcthodologically, this new approach provides a greater spec-
spondent is female, the interview presents added problems, because the t.rum of responses and greater insight into the lives of respondents-or
preestablished format directed at information relevant for the study tends "participants," to avoid the hierarchical pitfall (Reinharz, 1992, p.22)-
both to ignore the respondent's own concerns and to curtail any attemptS . it encourages them to control the sequencing and the language of
to digress and elaborate. This format also stymies any revelarion of per~ the 1 m d I
erVlewan a so allows them the freedom of open-ended responses
sonal feelings and emotions. \VaKlev. 1981; Rcinharz., 1992; Smith, 1987). To wit: "\'{fo men were
\Varren (1988) discusses problems of gender in both anthropological ... encouraged to 'digress' into details of the iT personal histories
and sociological fieldwork,arld many of these are found as well in the to reCount anecdotes of their working lives. 1vluch important infor-
ethnogr~phic intcrvie'\v. gOlne ,::>t J:l~e"c p,oblcrns arc dIe J:r~lditionalonC~ of We' ,p,hcccd in llll' WUy" (Ycandlt, DB'!; quored in Kelll]mz,
entree and trust, which may be heightened by the sex of the interviewer, P25).
especially in highly scx~segregated societies: "I never witnessed any cere~ Rosanna Hertz (1997a) makes the self of the researcher visible and sug-
monies that were barred to vvomen. \Vhenever I visited compounds I sat that it is only one of many selves the researcher takes to the field. She
82 83
 METHUD:::, Ul- LULL!:!... liNG .t-\1'lU AI'~ALY LII'lu tlVWlKlLAL jV\H! Cr<;lML.:l
8.sscrts that interviewers needto be reflexive; that is, they necd "to have an
ongoing conversation abom experience while simult:mcously living in the
moment" (p_ viii). By doing so, they 'vvill hcighten the understanding of
differences of ideologies, cultures, and politics between intcrviewers and
interviewees.
Hertz also underscores the importance of "voices"-how we, as
authors, express 8.nd write our stories, which data we include and which
we exclude, whose voices we choose to represent and which we do nor.
The concern with voices is also found, very powerfully, in a volume edited
by Kim Marie Vaz tided Gral Narrative Research \Y!ith Blach \Vonlell
(1997). One of the contributors, Christine Obbo (1997), StattS:
This chapter is a modest exercise in giving expression to women's voices
and in rescuing their perceptions and experiences from being mere mur-
murs or backdrop ro politic<ll, socia! nne! cultural happenings_ \'{lomen 's
voices have been devalued by male chronicles of cllltliral history even when
the men acknowledge female infonn:ll1ts; they are overshadowed by the
voice of male authority and as(cndance in society. (pp. 42-43)
This commitment to m8.inraining the integrity of the phenomena and
preserving the vie\vpoinr of the subjects, as expressed in their everyday
language, is akin to the stand taken by phenomenological and existential
sociologies (Douglas & Johnson, 1977; Kotarba & Fontana, 1984) and
also reflects the concerns of postlllodern ethnographers (lvlarclls &
Fischer, 1986). The difference:; :uc (3) the heightened moral concern for
subjects/participants, (b) the attempt to redress rhe male/female hierarchy
and existing paternalistic power structure, and (c) the paramount impor-
tance placed upon membership, because the effectiveness of male research-
ers in interviewing female subj~cts has been largely discredited.
Ruth Behar (1996) addresses the ambiguous nature of the enterprise of
interviewing by asking: Where do we locate the researcher in the field?
How much do vve reveal about ourselves? How do we reconcile our differ-
ent roles and Dositions? Behm makes us see that interviewer, v,niter,
respondenr, and interview are not clearly distinct entities; rather, they arc
intertwined in a deeply problematic way.
SVl11.C Cluil1i:;;t: :;;ociol0l:5i::;t::; hayc S01lC bC)'Olld CVllCCTll:> widl i111:"''-''''';''''''-
iog or fieldwork in itself. Lamel Richardson (1992a) is striving for new
forms of expression to reporr her findings and has presented some of her
fieldwork in the form of poe!:ry. Patricia Clough (1998) questions the
84
I ne mIen/lew: rrom :JIrucWrea \.duestlons to l\legotJoted le;..t
whole enterprise of fieldwork under the current paradigm 8.ncl calls for a
reassessment of rhe whole sociological enterprise and for a rereading of
existing sociological texrs in a light th8.t is not marred by a paternalistic
bias. Their voices echo the concern of Dorothy Smith (1987), who elo-
quently states:
The problem [of a rese3!"ch project] and its particular solution are analogolls
to those by which fresco painters solved the problems of representing the
different remporal moments of a srory in the singular sp3ce of the wall. The
problem is to produce in a [\vo~dimellsion;]1 space framed as a wall a world
of action and movement in time. (p. 2B 1)
A growing number of reseatchers feel that wc cannot isolate gender
from other import31lt clements that also "filter knowledge." Among oth-
ers, Patricia Hill Collins (1990) has written eloquently about the filtering
of kno\vledge th rough membershi ps-of being black. and female in Ameri-
can culture, in hcr casco Kath \X!eStofl (1998) makes just ~lS pmverful a case
for sexuality, which, she contends, should not be treated as a compartmen-
talized subspecialty, because it underlies and is integral to the whole of the
social sciences. Clearly, gender, sexualilY, and race Cannot be considered in
isolation; race, class, hierarchy, status, and age (Seidman, 1991) are all
parr of the complex, yet often ignored, elements that shape interviewing.
... Framing and Interpreting Interviews
Aside from the problem of framing real-life events in a two-dimensional
SPJC~, we face the added problems of how the framing is being done and
.who is'. doi ng t-I1e f famLllg.
- In 50ClO
. Ioglca
. I terms, thiS. means that the type of
InterViewing
. - Ieete
se - d, tIle teeIlnlques
' use d-, and t Ile ways of rccordlllg
.
lnfOrrnatio n a II come to I)ear on t I1e resu I ts at- the study. Addmana!!v
- .. data. .
b . ' . '
e Interpreted, and the researcher has a gre8.t deal of influence on
part of the data will be reported and how ir will be reponed.
Interviews
. volumes have been published on the techniques of Struc~
Intervie' ( .
, W1l1g see, among others, Babble, 1992; Bradburn et aLl
Gorden, 1980; Kahn & Cannel, 1957). There is also a voluminous
85
 METHODS OF COLLECTlt-1G AND ANALYZING tMl-'IKILAL !VIAl t.KIf.l,U
f r H:' II 1t!::'1 VI!::'W: rrom .)rrUCWrea Vuesuons to Negotiated le>..1

literature Oil group interviewing, especially in marketing and survey re-

NOl1 v.er bal
techniques are also import"nt ,'n .
search (for a comprehensive review of literature in this area, see Stewart L'c '" Interviewing. There are
four baSIC modes of nonverbal communication:
Shamdasani, 1990). The uses of group interviewing have also been linked
to qualitative sociology (1Vlorgan, 1988). Unstructured intervie\ving tech-
Proxemic commullication is the use of Interpc,son"1 Sp"ce to .
niques have been covered thoroughly (Denzin, 1989b; Lofland, 1971; . d' ' " ConlrnUnJcate
atntu es, chrollcl1llCS communication is til . f . f
," _ e use 0 pacing a speech and
Lofland & Lofland, 1984; Spradley, (979). length of s,dence Hi conversatlon, bllesic communication includes any body
As we have noted, unstructured interviews vary widely, given their mo:c~lent~ or pOStures, and paralillguistic commLInication includes all tl
informal nature and depending on the nature of the setting, and some vanatlons111 volume '[ I d I" le
, pL Clan qua Ity of VOice. (Gorden, 19S(),p. 3"-)
.J)
eschew the use of any preestablished set of techniques (Douglas, 1985).
Yet there are techniques involved in intervie\ving whether the interviewer ~1l ~~~r.af these modes reprcsem imporranr techniques for the researcher'
is just being "a nice person" or is following a formar. Techniques can be ma mon, the researcher should carefully note and record d:
u ' f h _ ' . - respon cnts
varied to meet various situaticns, and varying one's techniques is knmvn ses 0 ,t ese modes, for Interview dara are more than verbal records and
as using taerics. Traditionally, the researcher is involved in an informal sl:ould Include, as much as possible, nonverbal features of the ',nte t.
Fm II r t ,1 ' _ , - rac Ian.
conversation with the respondent, thus he or she must maintain a wne of , Hllques vary With the groups being intcrvie\ved', f or .Instance
. , a ,.}, ec
"friendly" chat while trying to remain close to the guidelines of the topic ~~lt:ra\le,wI,ngr~ g~oup ,of chil~ren re~1Uires a different approach from th~
of inquiry he or she has in mind. The researcher begins by "breJking the n lntenle\\,er lmght use when Interviewing a group of ~lde I 'd
ows (Lopata, 1980). e r y WI -
ice" with general questions and gradually moves on to more specific ones,
while also-as inconspicuously as possible-asking questions intended to
check the veracity of the respondent's statemcnts. The researcher should Interpreting Interviews
avoid getting involved in a ''"rcal'' conversation in which he or shc ansvvers
questions asked by the respondent or provides personal opinions on the 'b l\t1a~y ~tlldies u~il1g unstructured interviews arc not reflexive enougl
matters discussed. A researcher can avoid '"'"getting trapped" by shrugging a Out the Interprctlng p .. ' I . 1
S I' h rocess, common p 3utudes proclaim that the data
, pea ( tor t emsclves rl t I . I'
hi e."' The d a t ' 1a t le rcsearc ler IS neutral unbi"sed
of[ the relevance of his or her opinions (''"It docsn't matter how 1 feel, it's
a, ' <l, an d ,","'InVISI-
,.
your opinion that's important") or by feigning ignorance ('"I really don't dat _ d re~orted tend to flow nicely, there are no contradictory
know enough about this to say anything; you're the expert"), Of course, as , ' a an no mention of what data were exduded and/or why 1m .
cues neVer ha en _ d ,I . . . propn-
we have seen in the case of gcndered interviewing, rhe researcher may l1nreflexive fT
pp an tIe mam concern seems t-o be rl-
_ 1 ' ' , le proper. I,
'f
reject these techniques and '"'"come down" to the level of the respondent to engaged . 'f,llln g, anI;] yZl1lg, and reporting of events. But anyone who 11a5
l
In Ie (Wor ~ knows bert . I .
engage in a ''"real'' conversation, with give~~1llcl-take and shared empathiC searcher '_ r b er, no matter lOW orga11Jzed the re-
understanding. Old) e, he or she slowlv be b' d
!l1OLlntain of field not- . . '.' comes llr~e .under an increasing
The usc of language, particularly the use of specific terms, is important Traditionall r , des, tJanscnpls, newspaper chpPll1gs, and audiotapes.
in the creation of a '"'"sharedness of meanings" in which both interviewer of the d }, rle a ers were presenred with the researcher's interpretation
ata, c eaned and st, I' d d II
and respondent understand the contextual nature of specific referents. For coorrad' , r e a m me an co apsed in rational 0011-
ICtory accounts. Ivlore rc' 'I -' I . '
instance, in studying nude b,~acbes~ Douglas and Rasmussen (1977) dis~ the r fl " ' cent y, SOLIO OglStS have come to grips
e CXl\re problcn ,t',- I .
covered thelt the term llude beach virgin had nothing to do with chastity; witl h _ ' 12 L, ane, at times, contradictory nature of data
1 t t: tremend .f l '
rather1 it n:fcrn:u to tht fact thaL a pcr:::;uu's Gutta,;::k,; ,,,,,,,r'e ""vb;,,,,, rl-,us in- W1lult Van l'vICI Ol~S, 1 ~ t~I~SP,O (~~: lI1flue~ce of the researcher as
dicating to others that he or :'ihe was a newcomer to the nude beach. Lan~ in the 19-0 .. "'len (]J~~) callo contesslOnal slyle" began in ear-
. I S (see Johnson 19 7 6) d .
guage is also important in del ineating the type of question (broad, 11[\r1'O\\; II] a sou! clea ' ' , ,) an COntll1UCS unabated to OUr day
, , nSll1g by researchers f r bl 'f' <c> ..'
leading, instructive, and so on). Hons in the field AI, 0 p_o cmatlc eelmgs and stICky sltua-
. though perhaps somewhat overdone at times, these
86
87
 . '''' """"' .,"' . 111.1111 ..J"ULlUleu \,!llt:sllurls [0 f'legotlorea lext
At'-lALY LING tMt~IKIl...P,L IV\I-\l Lr\.!Ml..-..J
METHDDS DF CDLLECTIt,G AH D

concerns have revolved around the topics of in(onned consent (receiving

_ " ",' _ t I' as they make the readers ;]'vvare of the
1 'bi
"contesslO llS aft: 'vCt} \a U"l C, f int'ervicwing people in their natural consenr by the subject after having carefully and truthfully informed him or
complex and cumbersome na~u~'~l~)and veracity ~o studies. For example: her about the research), right to privacy (protecting the identity of the sub-
settings and lend;] rune of rea Ii.> ' . j :10 The day before 1 had ject), and protection (rom harm (physical, emotional, or any other kind).
,j'l[e Got up afOUl1l - -
"Yesterday I s1cpt vcr) " . " " " A'" If -U No sociologist or other social scientist would dismiss these three ethical
,T d a few others. [hey (helo t camt:. game
engaged Omaga, ,-oupa, .an
[
< _ ) concerns. Yet, there arc other concerns that arc less unanimously upheld.
, "(Malinowski 1967/1989, p, 67), The controversy concerning overt/covert fieldwork is more germane to
Into a ra~e \ 1 . "J f the researcher and the problemaries of
Showing t 1.C 1ll1~1~:1l1. 51 C 0 taken new forms in deconstruetionislll participant observation, bur could include the surreptitious use of tape-
unstructured lllterVlcwlOg h a s . '. I' d _ -ru- recording devices, Warwick (1973) and Douglas (1985) argue for (he use
D rricb 1976). Here the influence ot the author IS b[~~g 1t un cr St.: . of covert methods, because they mirror the deceitfulness of everyday~life
( e ,- d I ' [I,e I'ese'ucher's rendmon of events IS
. TI' I text create T\ . , ~. . reality, whereas others, including Kai Erikson (1967), arc vehemently
n~y. 1lIS t lC " " 1 " biases and taken-far-granted nonons are
"deconstructed ; the aut lor s.. . _ I 1-. t the delta arc introduced opposed to the study of uninformed subjects.
exposed, and, at times, alternatlve ways to 00'- 3. Another problematic issue stems from the researcher's degree of
(Clough, 1998). . . . ' r lnvc seen attempt to expose and involvement with the group under study. Whyte (1943) was asked to vote
Postmodern social rescarchos, as \\e f "1 -I 'I+er and qua author.
< more than once during the same local elections (i.e., to vote illegally) by
_ '.' . e the role of the rese:ucher qua Ie l-WO ~, .,. _ . members of the group he had gained access to, and befriended, gaining
iIlllllml~ . . . ' (1980) reports fuhaml s accounts,
no
Thus, lor msral1ce, Crapanza '.. J" outright lies their trust. He used "situational ethics," judging the legal infraction to be
I l . I' tical rcndmons, l reams, or ,
\vhether t ley Je SO~IO liS or. fl' IVlorrocan Arab subject's sense of minor in comparison to the toss of his fieldwork if he refused to vote.
because they all constitute a ~::lrt 0. t l~S 11 I. . Crapanzano learns not Thompson (1985) was faced with a more serious possible legal breach. He
self and personal history. In lI1tervI.eWll1~ Ulaml, \"'as terrified of having to \vitness one of the alleged rapes for which the
only about his subject but about h1ll1selt: Hell's Angels had become notorious, but, as he reports, none rook place
during his research. The most famous, and widely discussed, case of
As Tohallli's imerlocuwr, I became an active. pa:ticipatl t ,itl hi,S l.ife hi,st~:~<: questionable ethics in qualitative sociology took place during Laud
I II I pear di"'ectly in his tCcltatlons. Not onl) dId Ill) P
even t lOUg 1 rare _~ap. '. 'l;'e him for the text he was to produce, but they Humphreys's research for Tearoom Ti-ade (1970). Humphreys stlldied
CtlCC, and my quesnons, prep. " . I' They produced a homosexual encounters in public rest rooms in parks ("tearooms") by act-
ha1ge of consCIousness 111 Hm.
prOdl1CClj w Iut I rea d 3S a c ,- . b I' I I f am cur :lssump- ing as a lookout ("watchqueen"). Although this fact in itself may be seen as
. . , toO \YJe were at 1 lost cc r J
change of conSCIOusness 111 m~ j 1_ '. :I 'roped for ethically incorrect, it is the following one that has raised many academic
, 1 I- of the evervday world am nurse yes am g
nonS a )Oltt t le n:lture - -. ( 11)
, ['rcilce points within this limbo at Interchange. p. eyebrows. Humphreys, unable ro interview the men in the "tearoom,"
com1110n re e
recorded their cars' license-plate numbers, which led him ro find their
_ _. 1: ," faceless subject and invisible researcb~rl residences with the help of police files. He then interviewed many of
No longer pletendlllg to 1_ d ' d' 'd al human beings wltb the men in their homes without being recognized as having been their
, de' portraye as 10 IV1 u
Tubal111 an rapanzano are
'd' . . d I ' dors learn
es an we t le lea t: , <
their own personal histories ;ll1CI I losyncrasl, '
,Another ethical problem is raised by the veracity of the repons made by
about two people and tWO cultures.
For example, Whyte's (1943) famous study of Italian street
men in Boston has come under severe scrutiny (Boelen, 1992), as
have alleged that Whyte purtrayed the men in demeaning waY5 111ilL
"\)" Ethical Comidcrat:ions . reflect their visions of themselves. \X/hyte's case is stiJi unresolved,
. . ., . ' . g 'Ire human beings, rese;:lrch~ It does illUstrate the delicate issue of ethical decisions in the field and in
Because the objects of mqwry In mtcrvleWlll II< _I ical ,-----0' > /',f"P'Jrtinp field notes, even more than 50 years later (Richardson, 1992b).
ers mllst t:lkc extreme care to avol'd'<In}, Ila[l11 '0
I them. lradltlona y, et 1
89
88
 MI: 1HUU::J Ul- LULLtL lll'l\.."l /\I'lU J-\1'1ALl Lit'.\..) l:.ly,r ll\,''-M. .... 1VlM.1 .... ,"...., ........ J ilL "JeLl Ylt::W. I lUlU JlIULtUH;'U \.!ueSHons to l'JegoHotea le.\t

A growing number of scholar5, as we have seen (Oakley, 1981), fed that
most of traditional in-depth interviewing is unethical, whether wittingly
or ull\viningly. The rechniques amI tactics of intcrviewing, they say, are
reall}' ways of manipulating the respondents while trearing them as objects
or numbers rarher rhan individual human beings. Should the qucst for
objectivity supersede the human side of those we study? Consider rhe
following:
One day while doing research at rhe convalescent cenrer, I was ralking (0
one of the aides while she was beginning to change the bedding of one of rhe
p~Hients who had urinated and soaked the bed. He was the old, blind, ex-
vvresder confined in the emerge,lCY r00111. Suddenly, the \vresrler decided he
was not going to cooperate \\!ith the aide and began striking violenrly at the
air about him, fortunately missing the aide. Since nobody else \vas around, I
had no choice but to hold the patienr pinned down to the bed while the aide
proceeded to change the bedding. It: was not pleasant: The p~ltient was
squirming and yelling horrible :hre;1ts at' the top of his voice; the acid smell
of urine was Iwuse~1tit1g; I was slowly loosing my grip on the much Stronger
patient, wbile;Ill along feeling horribly like Chief Bromden when he suffo-
Gites the lobotomized Ivlac IVlurphy in Ken Kesey's novel. Blltthere /Vas 110
choice, one just could /lot sit bad~ alld talIe 1l0teSll.'hile the patiellt tore apl1rt
the aide. (Fontana, 1977, p. lE:7; emphasis added)
Clearly, as \ve move forvvard with sociology, we cannot, to paraphrase
what Herbert Blumer said so many years ago, ler the methods dictate our
images of human beings. As Punch (1986) suggests, as field-workers we
need to exercise common sense and responsibility, and, we would like to
add, to our subjects first, rothe study next, and to ourselves last.
.fJo New Trends in Interviewing
The latest trends in interviewing have come some distance from structured
respondents, and interviews arc seen as negOtiated accomplishments of
both interviewers and respondenrs that are shaped by the contextS and situ-
ations in which they take place. As Sclnvandt (1997) notes, "It has become
increasingly common in qualitative studies to view the interview as a form
of discourse between two or more speakers or as a linguistic event in which
the meanings of questions and responses are contextually grounded and
jointly constructed by interviewer and respondent" (p. 79). \'\!e are begin-
ning to realize that we cannot lift the results of interviews oUt' of the COI1-
tcxts in \vhich rhey \verc gathered and claim them as objective claw with no
strings attached.
Interview as Negotiated Accomplishment
Lct us briefly recap the two traditional approaches to the interview,
following Holstein and Gubriul11 (1995, 1997). These amhors usc Jean
Converse and Howard Schuman's COllversaliollsat Random (1974) as an
exemplar of the interview as used in survey research. In this context the
interviewer is carefully instructed to remain as passive as possiblc, so as to
reduce his or her influence-the scope of the interviewer's function is to
access respondents' answcrs. This is a ratiollal type of interviewing; it
assumes that there is an objective knO\vledge our there and that if one is
skilled enough one can access it, JUSt as a skilled surgeon can remove a kid-
~ey from a donor and use it in a different context (e.g., for a patient await-
Ing rransplant).
Holstein and Gubrium (] 995,1997) regard Jack Douglas's (1985) cre-
a~ive. interviewing as a romanticist type of interviewing. Creative Inter-
VIeWing is based on feelillgs; it assumes that rescarchers, qua imervie\vers,
need to "get to know" respondents beneath their rational facades, and that
rescaI'cl lers can reac I1 respon d," ems deep wells of emotion . by engaglI1g .
them, by sharing feclings and thoughts with them. Douglas' 5 interviewer is
certainlv. I 110re ' d
acnve an f ar Iess neurra I t han Converse and Schuman's '
but the assu mp tIOns' , are st!'II t I1C same: t I
lat'It ,IS t I1C S;:J
I 'II5 0 f' , '
. lI1tervlcwers

questions; we have reached th.;: point of inrervicw as negotiated text. Et.h- that will P'10\1'd e access ro Imow IecIge an d that t.h ere'IS a core lmowledge
nographers have realized forql1ite some time that researchers are not inVIS- that researchers can access.
ible, neurral entities; rather, they are part of the interactions they seck to H~,lsteiI1 and Gubrium finally consider the ne\v type of interviewing-
studY and inGuencc rhose intcractions. At last, intt:rvicwill.g is being; nev,,n isn't <.:xactly accuran:, given that thdr rdr.::rtl1cc [or tlli;; ia r11C
work of 11 ' IdSe 0 1a Pool, published in 1957. To wit: "Every inrer-
brOl~ght in line with ethnography. There is a growing realization that inter- .t lie
viewers are not the mythical, neutral tools envisioned by survey research, . . 1S an jmcrpersonal drama with a developing plot" (Pool 1057
193 . . . .
Interviewers are incre~singly sl::en as active participants in interactions with
<. , - ,

, quoted 111 Holstem & Gubnum, 1995, p. 14). Holstein and

90 91
 Me I t-IUlJ:::' Ur- LULLtL 111'l0 }.\I'IU AI"lALT LH'11.;l CIVW!r\I ....ML IVIM, Lf\,"',L..J
Gubrium (1995) go on to note that thus far \-ve have focused on the luhals
of the interview, the substantive findings, and it is time that we pay arten-
tion to the haws of the inrervlcw-thc comens, particular siruations,
nuances, manners, people involved, and so on in which interview interac-
tions take place. This concept harks back to ethnomethodology, according
to Holstein and Gubrium (1995): "To say that the interview is an interper-
sonal drama with a developing plot is part of a broader claim that reality is
an ongoing, interpretive accomplishment" (p. 16). Garfinkel, Sacks, and
others clearly stated in the late 1960s that reality is an ever-changing,
ongoing accomplishment based on the practical reasoning of the members
of societv. It is time to consider the interview as a pracrical production, the
meanin~ of which is accomplished at the intersection of the interaction of
interviewer and respondent.
In a later essay, Gubrium and Holstein (1998) continue their argument
by looking at in;ervicws as storytelling, which they see as a practical pro-
duction used by members of society to accomplish coherence in (heir
accounts. Once more they encourage LIS to examine the /JOws as well as the
luhats of storytelling. Similarly, .Madan Sarup (1996) tells us:
Each narrative has two parts; a SlOry (iJistoire) and a discourse (discourse).
The story is the content, or chain of events. The stOry is the "what" in a nar-
rative, t1~e discourse is the "how." The discourse is rathet like a plm, how the
teader becomes aware of what happened, the order of appearance of the
events. (p. 17)
Gubriu111 and Holstein are not alone in advocating this reflexive
approach to inrcrvie\vs. Both David Silverman (1993) and Robert
Dingwall (1997) credit Cicourel's classic work Method and lvfeasllremellt
ill Sociology (1964) with pointlTIg to the interviev.,r as a social encounter.
Dingwall (1997) notes:
If the inrervie\v is a social encounter, then, logically, it must be analyzed in
the same way as any other social encounter. The products of an inrerview
are the outcome of a sociallv situated activity where the responses afe
passed through the roie-pbyi~lg and impressioll management of both the
intcl'vicwel' :1nd the le<:pondent. (p_ S(,;)
1. E. Seidman (1991) discuss::s interviewing as a relationship by relying
upon a principal intellectual amecedem of the ethnomcthodologist Alfred
92
I II~ 1/ /leI Vlt'VV. num .)truetureo "'duestlons to I'legotloted lext
Schutz (1967). Seidman analyzes the interviewer-respondent relation in
terms of Schutz's "I-thou" relation, in which the two share a reciprocity of
perspective and, by both being "thou" oriented, create a '\ve" relation-
ship. Thus the respondent is no longer "an object or a eype" (Seidman,
1991, p. 73) but becomes an equal participant in the interaction.
The Problematics af New Approaches
Some of the proponents of the ethnomethodologically informed inter-
view are critical of interactionist as well as positivist intervie\v methods.
Dingwall (1997), as well as others, speaks of the romantic movement in
ethnography (and interviewing)-the idea that the nearer we come to the
respondent, the closer we are to apprehending the "real self." This
assumption negleCtS the faCt that the self is a process, ever negotiated and
accomplished in the interaction. Dingwall also faults the "postmodern"
turn; that is, if there is no real self, therc is no real world and 1 can create
one of l11y own. Finally, he is troubled by the "crusading" nature of the
romantics and asks, "\'V'hat is the value of a scholarly enterprise that is
more concerned with being 'right on' than with being right?" (p. 64).
In a similar vein, Atkinson and Silverman (1997) reject the postmodern
notion of "polyphonic voices," corrcctly noting that interviewer and
respondent collaborate together to create an essentially monologic view
of reality. This same rejection could be made using Schutz's (1967) argu-
n.lent-that is, "I" and "thou" create a unified "we," not t\vo separate ver-
Stons of it.
Etbnomethodologically informed interviewing is not, however, im-
mune from criticism itself. Schutz assumes a reciprocity of perspective that
may not exist. Granted, in our interview society we all know the common-
sense routines and ground rules of interviewing, but in othcr societies this
may not be the case. Isabel Bowlet (1997) anempred to interview Pakistani
women about their experience with maternity services and found a total
lack of understanding of the value of socia! research and interviewing: "I
had told -h I 1 .. h I f d
L em t lat was \vntmg a 00;:: on my In Ings. Yarns, who spoke
the bettet E I h i d h . I I I f b ehd
. on her face and
h. . ng ts , trans ate t IS \Vlt 1;.1 00 <:: 0' dIS
Tt~. ,I,c)" bo<h d',"olvcu into lilUIJI1[Cr. TIle 11O,pirals were very good.
lete weren't any problems. AlI was well" (p. 72). Bowler \vas forced to
condude t l ..
. h lat mtervIeWIng mav not work where there is no "shared notion
ott e Process 0 f research" (p.. 66).
93

METHODS OF COLLt:.L 11r'~G j\NU AI'IALY LII'<\..:l t.!VIt'IKI\..J-\L IYIMI 1:.1,1""''--'
silverman (1993) envisions a different problem. He seems to feel
that some ethnomethodologists have suspended their interest in substan-
rive concerns of everyday life, claiming that they call1lQ[ address them
until thev know mor~ abour th,= ways (methods) in which these realities
arc acco~lplished. He notes, "Pur simply, according to one rea,di,n g of
Cicourel, we would focus on the conversational skills of the parnClpams
rather th-an on the content of what they are saying and its relation to the
\vorld outside the interview" (fl 98).
Cicourel (1970) states that sociologists need to outline a wQ[k~lble
model of the actor before engaging in the srudy of self and society.
Garfinkel held similar beliefs. For instance, in his famous sruely of a trans-
sexual, Agnes, Garfinkel (196 7 ) was examining the routines. by whi~h
socictalmcmbers pass as maLes or females; he had little or no mttrest H1
issues of transsexuality per se. Thus it \vOldcl follow that, according
to Silverman's reading of ethnomethodology, \ve should learn the con-
versational methods before attempting to learn substantive matters in
lnrervicvving.
Future Directions
~[o borrow from Gubrlum and Holstein (1997): "\Xlhere do we go from
here?" (p. 97). \YJe share with these two authors a concern with appreci~t
ing the new horizons of posrInodernism while simu~ta~eously rem::lln-
log conservatively committed to the empirical deSC~lptlOn of evc~rd~Y
life. Gubrium and Holstein (1998) introduce a techI1lque they call al1,l-
lytic bracketing" to deal with the multiple levels of interviewing (and
ethnography):
We mav focus, for example, on holU a story is being told, while tempornrily
deferrj~g our concern for the various Iuhats that are involvec.l-:for e~an~
pIe, the substance, structure, or plot of the story, the context wlthm which It
is told, or the audience to which it is accountable. \\!e can later retLIrn to
these issues. (p. 165)
The use of this analytic bracketing allows the authors to analyze intervie\~'-
;lle;n its coherence nnd J;"c-r,,:t)' "lS nr-l e"enr- c<_~IL,I:>or-nt;"cly,",cl~;ev,,-~J. In
wbich product and process arc mutually constituted.
A pressing problem in imerviewing concernS the kinds of standards we
should apply to these new and different types of interviews. To assume
94
, " .... "" ..... , ., ....... ) 'V), I ..JtI U'-lU) 'CU "'!lIC",UUII:> LU I Yt::yuuuteu !t::XL
absolute relativism is not the solution, for it would lead, in Silverman's
(l997b) words, to the "sociology of navel-gazing" (p. 240). Silverman
proposes an aesthetics for research; he rejects attempts to use literary
forms in sociology: "If I want to read a good poem, why on earth should
I turn to a social science journal?" (p. 240). Silverman's critique of inter-
actionist sociology and proposal for aesthetic values seems to focus on the
following three points: (a) He arr-acks the grandiose, political theorizing of
British sociology and invokes a rerurn to more modest, more minute goals;
(b) he rejects the romanticist notion of equating experience (from the
members' vicvvpoint) with authenticity; and (c) he notes that in sociology
we mimic the mass media of the interview society, thus succumbing to the
trivial, the kitschy, the gossipy, and the melodramatic and ignoring sim-
plicity and profundity.
Silverman's notion that we should pay attention to minute derails in
sociological studies, rather than embarking on grandiose, abstract proj-
ects, in a way is not dissimilar to Lyotarel's appeal for a return to local ele-
ments and away from metathcorizing. For Silverman, the "minute" are the
small details that go on in from of our eyes in our everyday lives-very
similar to Garfinkel's mundane routines, which allow us to sustain the
world and interact with each other.
We are in agreement with Silverman that we need to stop deluding our-
selves that in our particular method (whichever it may be) we have the key
to the understanding of the self. \Y/e also agree that it is imperative that we
look for new standards, as we are quickly digressing into a new form of the
theater of the absurd (without the lirerary flair, we fear). But we cannot
wait to find a model of the methods used by participants in interviews or in
everyday life before we proceed; Cicourc!'s (1970) invariant properties of
interaction turned out to be 50 general as to be of little use to sociological
inqUiry.
\\Te need to proceed by looking at the substanrive concerns of the mem-
bers of society while simultaneously examining the construcrive activities
used to produce order in everyday life, and, all along, remaining reflexive
about how interviews are accomplished (see Gubrium & Holstein, 1997,
1998). For instance, as Carolyn Baker (1997) points out, a researcher's
H:l~ing d re.<;pond",'nt: "T ~n-l '" '"od~c,- ol threc" ,,<or-bUS to,;;l1iuti her '1 "Ill "
University prof essor ,.- accesses dff 1 erent categorIes. an d eI KitS dff
I erent
aCCOunts. \,(Ie need to move on with sociological inquiry, even though we
realize ConLj'mons
,
afe 1ess than perfect. To paraphrase Robert 5010\\', as
95
 IVll: J MULJ,) ur \....VLLl:\""III'ID MI~L-' M,-'''''''I..-' '-"''-' ...... 11 ",,<-.-r> ...
cited by Geertz (1973): Just because complete asepsis is impossible it
doesn't mean we may just as well perform surgery in a sewer.
A different kind of future direction for interviewing stems largely from
the new feminist interviewing practices. Traditional interview has pains-
takingly attempted to maintain neutrality and achieve objectivity, and has
kept the role of the inrervie",,-er as invisible as possible. Feminists, instead,
arc rebelling at the practice of exploiting respondents and wish to use
interviewing for ameliorative purposes. To wit: "As researchers with a
commitment to change, we must decenrer ourselves from the 'ivory tower'
and construct more partici patory, democratic practices. \\"e must keep peo-
ple iJlld politics at the ccnter of 0111' research" (Benmayor, 1991, pp. 172-
173; emphasis added). Denzin (1989a) refers to rhis approach as the "fem-
inist, communitarian ethical model" (see also Lincoln, 1995) and tells us:
The feminist, cUlllmunitarian researcher does not invade thc privacy of oth-
ers, use informed consent forms, select 5ubjens randomly, or measure re-
search designs in term of (heir validity. This framework presumes a
rescarcher who builds collaborative, rcciprocal, trusting, and friendly rela-
tions with those studied.... It is also understood that those studied have
claims of O\vnership over any ll1aterial that are produced in the rese;uch
process, including field noteS.
Combining the roles of the scholar and the feminist may be problematic
and, at times) may lead to conflict if the researcher has a different political
orientation than that of the_ people studied (\Vasserfall, 1993), but this
approach may also be very rewarding in aHO\ving the researcher to see pos-
itive results stemming from the research (see Gluck, 1991).
Electronic Interviewing
Another direction currenely being taken in inrerviewing is related to the
clunging technologies available. The reliance on the interview as a means
of information gathering has most recently expanded to electronic outlets,
\vith qucstionnJires being administered vb fax, electronic mail, and \'(leb
site. Estimates suggest that nearly 509--0 of all households have computers)
and n,;:arl)' half of these urili,.;c the Internet. $o(nvClre is no,v available rho:lt
allows researchers to schedule and archive interview data gathered via
chat-room interviews. The limited population of potential respondentS
96
,.'r">' ... ",r,,_-, "'~ "'"~' .,~, . ,'w", -'" .... '-, ... ,'-u ,<,-,"'.a, ..."." I.V, ""'::'VI.IV'C'" 'CAl.
with access to computers makes general-population surveys infeasible, but
electronic interviewing can reach 100 0AJ of some specialized populations
(Schaefer & Dillman, 1998).
It is now possible to engage in "vinual interviewing," in which Internet
connections arc used synchronousl y or asynchronously to obtain informa-
tion. The advantages include low cost, as the result of no telephone or
interviewer charges, and speed of return, Of course, face-to-face inter-
action is eliminated, as is the possibility, for both interviewer and re-
spondent, of reading nonverbal behavior or of cuing from gender, race,
age, class, and other personal characteristics. Thus, establishing an
interviewer-interviewee "relationship" and "living the moment" while
gathering information (Henz, 1997a) is difficult, if not impossible.
Internet surveys make it easy for respondents to manufacture ficcional
social realities without anyone knowing the difference (J\'1arkham, 1998).
Of course, interviewers can also deceive respondents by claiming experi-
ences or characteristics they do not have in hopes of establishing better
[3pporr. They can feign responses for the same purpose by claiming "false
nonverbals," such as telling respondents that they "laughed" or "were
pained" by particular comments. IVIarkham (1998), in her autoethnog-
raphy of Intcrnet intcrviewing, reports that electron ic interviews take lon-
ger than their traditional counterparts and that responses arc marc cryptic
and less in depth, but the interviewer has more time to phrase follow-up
questions or probes properly.
It is also virtually impossible to preserve anonymity in Internet e-mail
surveys, but chat rooms and similar sites do permit the usc of pseudonyms.
Although electronic interviews are currently used primarily for quantita-
tive research and usually employ structured questionnaires, it is only a
matter of time before researchers adapt these techniques to qualitative
work, just as they have adapted electronic techniques of data analysis. For
example, IVlarkham immersed herself in the process of engaging with vari-
ous electronic or Internet formats (i.e.) chat rooms, listservs) to interview
:Jther participants and to document her journey in the virtual world, learn-
Ing the experience of cyberspace and the meaning participants attached to
their on~line lifestyles. She asks an intriguing question: "Can I have a self
where my body does not exist?" (p. 8).
T~1e [urun; 1lIi.ll" ~cc cOllsidcntblc ctl1no5riJ.pll)' b} IIH.:amj 01 campuu:r-
medIated communication, where virtual space radler than a living room
Or a Workplace is the setting of the interview. It remains to be seen whether
97
 Mt 1,IUU) ur- LULLtL 111'11.:3 , ... I'il) MI',nLI ,_""U L'V" "" ...............
electronic interviewing will allow researchers to obtain "thick descrip-
tions" or accounts of subjective experiences, and whether such inter-
viewing \vi11 provide the "pro.:ess context" so imponam to qualitative
interviews. In addition, rese,-l'~chers conducting such inrerviev'iing can
never be sure they are receiving answers from desired or eligible respon-
dents. Interviewing via the Internet is so prominent today that researchers
are smdying its effects on response quality. Schaefer and Dillman (1998),
for example, found that e-mail surveys achieved similar response rates to
mail surveys but yielded better-quality clata in terms of item completion
and more detailed responses w open-ended questions.
There arc clearly many unanswered questions and problems related to
the usc of electronic interviewing. This mode of intetviewing will obvi-
ously increase in the forthcoming millennium, as people rely increasingly on
electronic modes of communication. But just how much internet commu-
nication \,vill displace face-co-race interviewing is a matter that only timc
\vill tell.
'' Conclusion
in this chapter we have examined the interview, from structured types to
interview as negotiated text. \'i/e have outlined the history of interviewing,
\vith its qualitative and quantitative origins. We have looked at structured,
group, and various types of unstructured interviewing. \YJe have examined
the importance of gender in imerviewing and the ways in which framing
and interpreting affect intervievl,'s. We have examined the importance of
ethics in interviewing, and, finaily we have discussed the latest trcnds in
interviewing.
Wle have included discussion of the whole gamut of interviews, despite
the bct that this book is ccm:erned with qualitative research, because we
believe that researchers must be cognizant of all the various types of inter~
views if they are to gain a de3r understanding of interviewing. Clearly, cer-
tain types of interviewing are better suited to particular kinds of situations,
and researchers nwst be aU.'tJre of the implications; pitfal!s, and problems
of the ty!-ws of interviews they choose. If we \vish to find OLLt how [TIany
people 0PPO::iC elH; ci'itilbliGluuem of a nucIe'll' l'epo<;itory in their area, a
structured type of interview, such as that used in survey research, is the
best tool; we can qUcll1tify and code the responses and use mathematical
models to explain our findings. If we arc interested in opinions abolIt a
98
'>\"" ..... "r>...- l ,,'.... "",,:,;, .,"'.... , 'V'" -lc'w..... w'Cu "":,"u,",,,,"u,,,, 'u ,vcyUUUtCU ,,,,,,,,
givcn product, a focus group interview will provide us with the most effi-
cient results) whereas if we wish to know 3bour the lives of Palestinian
women in the resistance (Gluck, 1991), we need to intervicw them at
length and in depth in an unstructured way. In the first example used
above, and perhaps in the second, we can speak in the formal language of
scientific rigor and verifiability of findings; in rhe third example we can
speak of understanding a negotiated way of life.
lVIore scholars are realizing thar to pit one type of interviewing against
another is futile, a leftover from the paradigmatic quanticative/qualitative
hostility 0 f past generations. Thus an increasing number of researchers are
using multimethod approaches to achieve broader and often better results.
This multimethod approach, referred was triangulation (Denzin, 1989b;
Flick, 1998), allows researchers to usc different methods in different com-
binations. For instance, group interviewing has long been used to comple-
ment survey research and is now being used w complemcnt participant
obscrvation (lvlorgan) 1988). I-Iuman beings are complex, and their lives
afe cver changing; the marc methods \'Ve use to study them, the better our
chances to gain some understanding of how they construct their lives and
the swries they tell us about them. .
The brief journey we have taken through the \'Vorld of interviewing
sbould allow us to be bctter informed and perhaps more sensitized to the
problematics of asking questions for sociological reasons. We must re-
member that each individual has his or her own social history and an indi-
vidual perspecti ve on the world. Thus v,re cannot take our task for granted.
As Oakley (1981) notes, "Interviewing is rather like a marriage: every-
body knO\vs wbat it is, an awful lot of people do it, and yet behind each
closed front door thete is a world of secrets" (p. 41). She is quite correct-
We
a II t I" I \ve know how to ask questions
lin" " and talk to people, from com-
mOn, everyday folks to highly qualified qllantophrenic experts. Yet to
learn about people we must treat them as people) and they will work with
~s t~ help us create accounts of their lives. As long as many researchers
l:OntllllIe to treat respon dents as 11l1lmpOrranr,acc
" f 1ess mdlvl " ""d 1I31S W I10se
only Cal1 t"b" " f"
n Utlon IS to tll one more boxed response, thc answers wc, as
researchers ' get WI"II be commensllra)II e Wit" I1 the questions we ask and the
u.
Ways Wt -151
C
1lem, WI
\
'f
1'1 c arc no di -ferent from Genrudt Stein, who) on her
t
tarhbed, ~lskcd her lifelong companion) Alice B. Toklas, II What is the
And \vhen Alice could nor brin u herself to speak Gertrude
I
as(Cd"f . . . . 0 ,
, _11 that case, what tS the question?"
99
"9 References
}-\cbms, R. N., & Preiss, J. J. (Eds.). (1960). Hlllllall orgillfi:::ational research: Field
relatiolls alfd techniques. Homewood, IL: Dorsey.
Anderson, N. (1923). The hobo: The sociology of the homeless lJlall. Chicago: Uni-
versity of Chicc1go Press.
Atkinson, I~, & Silverman, D. (1997). Kundera's Immortality: The interview soci-
ety and dlc ilwcmion of self. Qualilative Inquiry, 3, 304-315.
~abbie, E. (1992). The practice Df social research (6th ed.). Belmont, CA:
Wadsworth.
IL,ker, C. (1997). IVlembership:ategorization and interview accounts. In
D. Silverman (Ed.), Qfwlitaiiue research: Theory, method and practice
(pp. 130143). London: Sage.
Becker, H. S. (1956). Interviewing medical students. AlJlericall}eJltrlwl ofSociology,
62, 199201.
ecckcr, H. S. (1999). The Chicago school, so-called. Qlwiitatiue Sociology, 22, 3-
12.
Behar, R. (1996). TI!e 1111111erable ubserver: Anthropology that hrea/~s your heart.
BostOn: Beacon.
Benmayot, R. (l991). Tesrimony, action research, and empowerment: Puerto Rican
women :lIld popubr education. Tn S. B. Gluck & D. Patai (Eds.), Womell's
words: The felllillist prdctice of oral history (pp. 159-174). New York:
Routledge.
Boden, W A. 1'\'1. (1992). Street comer society: Corncrville revisited. jOlln1(/1 of
Contemporary EtlJIlogra/Jhy, ~~ 1, 11-5-1.
Bogardus, E. S. (1926). The group in terview.joumal ofApplied Sociology, 10, 372-
382.
Booth, C. (1902-1903). Life ,md labOllr of the people ill LOlldoll. London:
IVlacrnillan.
Bowler, I. (J 997). Problems with interviewing: Experiences with service providers
,mel c1iems. In G. J\ililler L\: R. Dingwall (Eds.), COlltext and method ill qW1/i-
tetive research (pp. 66-76). Thousand Oaks, CA: Sage.
Bradburn, N. M. (1983). Response effects. In E H. Rossi, J. D. Wright, & A. B.
Anderson (Eck), Handbook of survey research (pp. 289-328). New York:
Academic Press.
Bradburn, N. M., Sud man, S., & As~;ociates. (1979). 1111/Jrovil1g interview method
alld qllcstiolllliJire design. San Francisco: Jossey-Bass.
Cicoure!, A. (1964). kfethod alld measurelllent ill sociology. New York: Free Press,
CicOlltc1, A. (1970). The acquisition of social structure: Toward .1 dcvclopmcl1 tJl
sociology of language and meaning. In J. D. Douglas (Ed.), UllderstandiNg
euery(idy Ii(e: Toward a reCOllslTllctioll o( social !wololedge (pp. 136-168).
~hicag(J: Aicline.
100
Cicourd, A. (1974). Theory mId method ill it study ofA/gentille (ertilily. New York:
John Wiley.
Clough, P. T. (1998). The clld(s) of etlmograph)': Prom realism to social criticism
(2nd cd.). New York: Petcr Lang.
Collins, I~ f-1. (199U). Black feminist thought: Knowledge, consciousness, and the
politics of empowerlllent. New York: Routledge, Chapman & Hall.
Converse, J. M. (1987). SlIl'LIe)' research ill the United States: Roots and emergence
1890-1960. Berkeley: University of California Press.
Converse,]. lVI., & Schuman, H. (1974). Conversations at randolll: SlIrvey research
as interviewers sec it. New York: John Wiley.
Cr<lpanzano, V. (1980). TlIhallli: Porlrait of a Moroccall. Chicago: University of
Chicago Press.
Denzin, N. K. (l989a).1Iltelpretiuc iJltemctiollis1Il. Newbury Park, CA: Sage.
Dcnzin, N. K. (l989b). The research aet: A theoretical introduction to sociological
methods (3rd cd.). Englewood Cliffs, NJ: Prentice Hall.
Derrida, J. (1976). O( grallllllatology (G. C. Spivak, Trans.). Baltimore: Johns
Hopkins University Press.
Dcsvousges, W H., & Frey, J. H. (1989). Integrat'ing focus groups and surveys:
Examples from environmental risk surveys. jOflnzal of O((icial Statistics, 5,
349-363.
Dingwall, R. (1997). Accollnts, interviews, and observations. In G. IVliller &
R. Dingw<lll (Eds.), COlltext and method ill qualitative research (pp. 51-65).
Thousand Oaks, CA: Sage.
Douglas, J. D. (1985). Creative illteruieruillg. Beverly Hills, CA: Sage.
Douglas, J. D., & Johnson, J. lVI. (1977). Existential sociology. Cambridge: Cam-
bridge University Press.
Douglas, J. D., & Rasmussen, P. (with Flanagan, C. A.). (1977). The /llule beach.
Beverly Hills, CA: Sage.
Du Bois, W E. B. (1899). The Philadelphia Negro: A social study. Philadelphia:
Ginn.
Ellis, C. (991). Sociological introspection and emotional experience. Symbolic
Illteraction, 14,23-50.
Ellis, C. (1995). Final l1egotiatioJls: A story of love, loss, and chrolfic ilhress. Phila-
delphia: Temple University Press.
Erikson, 1(. T. (1967). A comment on disguised observation in sociology. Social
Problems, 14, 366-373.
hoe, M (19"" - 0.)- _
~. Wit
1984) LOplllg . h rape: C nnca
.' 1 perspectives
. .
on conSCIousness.
. Imaginatioll, Cogl1itiol1 and Perso/lality, 3, 249-267.
Fltck, U. (1998). An illlrodllctioll to qualitative research: Theory, method alId appli-
r CGliol/s. london: Sage.
'olltana~ A. (1977). The last frO/ftier: The SOci/Ill11caning of gmlui/lg old. Beverly
Hdls, CA, Sage.
101
 IVII: I MULJ.:l ur \...VLLL\...' "'u "',,'-' "" ' r u.. ' .... " ' .... ~"" , ... _ _ . . _
. _ . _ . ' .... , -''' u,-u CV ,<vCJtlU' 'J lV J ~t::':JVUuteu It:'_~{

Fomana, A., & Frey,J, H. (1.990). POSITctiremem workers in the labor force. Worl~
Hertz, ~ ..( I 997c~. A ~'~'pology of approaches to child care: The ccruerpiece of orga-
11lld Occupations, 17, 355 -3 (-i I.
fllZlllg Lundy hIc for dUell-earner couples. jOltrlwl 0/ Filll/ilv Issues. 18 355-
Fontana, A., & Smith, R. (1989). fdz.heimer\ diseasc victims: The "unbecoming" 385. . ...
of self and the normallzario:1 of competence. Socio/ogiuzl Perspectives, 32,
Hertz, R., & Ferguson, F. (1997). Kinship strategies and self-sufficiency among sin-
35-46.
gle mothers by choice: Posrmodern family ties. Qlwlitatiue Sociolo'Sv 20(2)
Freeman, D. (1983). Margaret .I.V[e"d alld Samoa: The mahilIg alld /tIlI/lakillg of i.lIl 13-.17. ~-' ,
mtthrolJOlogical 1lI)'lh. Cambridge, 1v11\: Harvard University Ptess.
Holstein, J. A., & Gubrium, J. F. (1995). The actiue intcruielu. Thousand Oaks CA.
Ftey, J, H. (1989). STiruey research by tclelJholle (2nd cd.). Newbury Park, CA: Sage. Sage. ' .
Frey, J. I-I. (1993). Risk perception :lssociated \vith a high-level nudear waste repos-
Holst'ein,J, ~., ~ Gubrium,j. F. (1997). Active interviewing. In D. Silverman (Ed.),
itory. Sociological Spectl"lllll, 13, 139-151.
Qllalztalzue research: Theory, method alld practice (pp. 113-129). London:
Frey, J. H., & Fonwna, A. (1991). The group interview in soci,ll research. Social Sage.
SciCllcejoltrlwl, 28, 175-IH7.
Hllrnphrc~'s, L. (1970). Tearooll//rildc: IlIlpersorwl sex ill public places. Chicago:
Garfinkel, H. (1967). Studies in ctlJllomelhodology. Englewood Cliffs, NJ: Prentice
Aldmc.
Hall.
Hyman, H. H. (1954). Interviewing ill social research. Chicago: Universiry of
Geenz, C. (J 973). Thiele dcscriptinn: Toward an interpretive theory of culture. In
Chicago Press.
C. Geenz, The illtcrpretatic'n of Cltlturcs: Selected essdys (pp. 3-30). New
johnson,]. (1976). Doing (ieid research. New i'ork: Free Prcss.
York: Basic Books.
Glaser, B. G., & Srrauss, A. L. (1967). The discuI!cry o( gronllded theory: Strategies Kahn, R., & Cannell, C. F. (1957). The dYl/amics olil1leruiewillg. New Yorlc: John
\Viley.
(or qllalitative research. Chiclgo: Aldine.
Gluck, S. B. (1984). What's so special about women: Women's oral hislOry. [n Kotarba,~. A.,.& Fon:a~a, A. (Eels.). (1984). The existelllialse!fill society. Chicago:
D. Dllnmvay & \XZ K. Baulll (Eels.), Oml history: All ill/erdisciplinary <1Ilthol- UnIversity of ChICago Press.
og)' (pp. 221-237). Nashville, TN: American Association for Stare ;md Local Krieger, ~. _ _. TI-
S (1981) .., (/-/l/Ci!:
Je lIll/TOr 1 1
(ell/ ' lit a womell's cOllllI/unity. Philadel-
Illy
History. pIlla: femple University Press.
1,'IU-\ . '1 -I'
Gluck, S. It (1991). Advocacy oul history: Palestinian women in resistance. In S. B. Kvale , S.
_ (1996) .11I1-'rl
."" s. r 1I Il/lOi IlL lUll to. qua I"Ita/we rese'lrch IIltewiewillg.
.
Gluck & D. Parai (Eds.), \'(1omClI's words: Thc (emil/ist practice oforal history 1 housand Oaks, CA: Sage.
(pp. 205-220). New York: I~ourledge. Lincoln , Y.S . (199-)
J.
Tl' ... I . '.. .
le Slxr 1 lllomt:nl: Emcrgl11g problems 111 qualitative re-
Gluck, S. B., & Parai, D. (Eds.). (1991). Womell's words: The fell/illist practice of search. In N .K . Demi n (E-d) S', d
. , .1I I. I'
les III syll1uo .
Ie lItteractlOlI: A research
amI history. New York: Routledge. aIlIIl/al (Vol. 19, pp. 37-55). Greenwich, CT: .fAl.
Gorden, R. L (J 980). InteruicwiJl,s: Stn.ltcg)', techniques, alld tactics. Homc\vood, Lofl,nd
.
J (1'l7!)
,. - .
A I' . /. I
I1i1 yzmg SOC/i.l setttllgs. Be mom, CA: \Vads\vonh.
IL: Dorsey. Lotlan(~, ,J., & Lofland, L H. (19B4). Analyzing social settings: A guide to qllalitl1-
Gorden, R. L. (1992). Basic interl'icwing shills. Itasca, II..: F. E. Peacock. tlue observatio/1 al1d analysis (2nd ed.). Belmont, CA:\X!adsworrh.
Gubrium, J. F., & Holstein, J. A. (1997). The lIew language of qualitative l/Ie/hod. LOpata H Z (1980) I .' . . . .
, -. _.. . ntcrvlewmg Amencan WIdows. In \X( Shaffir, R. Stcbbins, &
New York: Oxford Univers'xy Press. A. Turowerz (Eds.), Fieldwork experience: Quali/ative approachcs 10 social
Gubrillm, j. F, & Holstein, J. A. 1)998). Narrative pmctice and the coht:rence of L research (pp. 68-81). New York: Sr. lVIartin's.
personal stories. SociologiCilI Quarterly, 39, 163~ 137. ynd,R.S.
,,'
&Lynd'H.JV[ (I')"') \rI/1 1 d' .
. . "'-/ J. f( {, etolUn:.r 511/ y til AmCnCi1l1 culture. Ncw
Harvey, L. (1987). Myths o( the Chicago school of sociolob'y. Aldershot, England: lOrk: Harcourt, Brace.
Avebury. LYnd, R. 5.,. & L"nd H M . (19") 11(1 '///' ..
, -, I [ CIOlUl1 III tml1slllOl1: A stl/dl' ill cllltural
j'

Hertz, R. (1995). Separatc bur simukmeous interviewing of husbands and \vives: conflrcts. New York: Harcourt, Bracc. .
J'vlaking sense of their srories. QllaIit/-1tiue Inqlliry, .1, 429-451. Njacencby, ~. E., & i\Llccoby, N. (1954). The interview: A lool of :iocJul ::iCicncc. In
Hertz, R. (1997a). Introduction: Reflexivity ,md voice. In R. Hertz (Ed.), Re(lex~ '. Lllldzc.y (Ed.), l-Illldbook ofsocial psvcholorrll: Fol. 1. Thenr]! Till/III "I'D i
iuity and voice. Thousand Oaks, CA: Sage. (pp 449 4 - .. - oj ... ' . t
.. - 8/). Reaellllg, ~'1A: Addison-\X1esley.
J [

Iv!.'I.
Hertz, R. (Ed.). (1997b). Reflexivily and mice. Thollsand Oaks, CA: Sage. ,. rno\\,sk! B (1977 . .
,. --). Algollallts of the lUestem Pacific. New York: E. r~ Dulton.
102
103
 ", ... ",e..., .,...... , ,,-,,,, .... ,,'-''-,'-','''1.1 ,,<!'-'C'::'UU"'::' LU 1~""YU(JUl""U It:'Xt

IVblinowski, B. (1989). A diLlr)' if,' the strict sense oj" the terlll. 5t3nrord, (A: 5t:1O- Sarup, NT. (1996). [dmtity, C/Ilture LInd the flostmodern Ilor/d. Athens: University
ford Universiry Press. (Original work published t9(7) of Georgi(l Press.
l\'larcus. G. E., & Fischer, M. IV!.]. (1986). Allfhropology as ([tllllI'cll critique: All Schaefer, D. R., & Dillman, D. A. (1998). Development of a standard e-mailmeth-
e.~perimelltaI1l1omel/t ill tlJ/! hllllWIl sciences. Chiclgo: University of Chiclgo odology. Public Opinioll Qtlarterly, 62, 378-397.
Press. Scheurich, 1-], (1997). Research method ill the fJost1l1odem. London: Falmer.
lViarkhal11, A. N. (1998). Life onlille: Resemching real e.\pericltce ill uirtuill SPilce. Schutz, A. (1967). The phenomello/ogy of the social IUorld. Evanston, IL: Norrh-
Walnut Creek, C!\: AltalVlira. wesrcrn University Press.
IVlcnon, R. K., Fislee, lVI., & Kendall, t L. (1956). The fOCllsed interview. Glencoe, Schwandt, T A. (1997). Qualitative il/quiry: A dictiollary of terll/s. Thousand
II.: Free Press. Oaks, CA: Sage.
J'vlerton, R. K., & Lazarsfdd, l~ F. (Ecis.). (1950). COlltinuilies ill social research: Schwartz, B. (Ed.). (1996). Colkcrive mcmory [Special issue]. Qualitative Sociol-
Studies ill/he scope aud method of "The Ameriwll soldie/:" Glencoe, rL: Free ngy, 19(3).
Press. Schwartz, 13. (1999). }'/ltmocy and the practice of commitment. In B. Glassner &
Iv1i1ie r, Ivt (1974). Plain speaking: An oml biogwphy olHilrry S /illl/hm. New ''{ode R. Hertz (Eds.), Qllalitative sociology as everyday life. Thousand Q,lks, CA:
Putnam. Sage.
Mishler, E. G. (1986). Resef1rch i,/ten!iewillg: COlltext ,7IId Ilarrati!)e. Cambridge, Seidman, L E. (1991). Interviewing as qualitative research. New York: Teachers
MA: Harvard University Prcss. College Press.
j'dorgan, D. (1938). Focllsgrollpsllsqualitatiue rese(7rch. Ne\vbury Park, CA: Sage. Sdltiz, C, Jahoda, 1\/1., Deutsch, M., L\: Cook, S. W. (1965). Research methods ill
iV1organ, D., & Spanish, IV1. T. (l984). Focus groups: A new 1001 for qualitatIve social relations. London: iVIcthuen.
rescarch. Qualitative Sociology, 7, 253-270. Silvcrtlun, D. (1993). Illterpretillg qllolil<1til'e data: Slrategies (or iIlwlysing talk,
Oakley, A. (1981). Interviewing women: A contradiction in terms. In H. RobertS text altd interaction. London: Sage.
(Ed.), Doing feminist research (pp. 30-(1). London: Roudedge & J(egan Silverman, D. (Ed.). (1997a). Qlla/itatiue research: Theory. method ./lId pr.lctice.
Paul. London: Sage.
Obbo, C. (1997). What do women know? .' . As I was saying! In K. l\t. V,n (Ed.), Silverman, D. (1997b). Towards <In aesthetics of research. In D. Silverman (Ed.),
Omlnarrative research /oith bl<1d~ U/OJ/lell (pp. 41-(3). Thousand Oaks, C1\: Qlfalitatiue research: TheOl)!, method alld practice (pp. 139-253). London:
Sage. S<lge.
Parten, M. (1950). Stlrueys, polls, and samples. New York: Harper. Singer, E., Frankel, M., & Glassman, M. B. (1983). The effect of interviewer
Payne, S. L. (1951). The art ofasking qtlCstioJls. Princeton, NJ: Princeron University characteristics and expectations on response. Public Opillioll Quarterly, 47,
Press. 68-83.
Pool, 1. de S. (1957). A critiquc of the twentieth anniversary issue. Public Opinion Singer, E., & Presser, S. (1989). Slime)' research tllethods. Chicago: University of
Quarterly, 21, 190-'198. Chicllgo Prcss.
Punch,1v1. (1986). The politics mid ethics o{ fieldll./ork. Newbury Park, CA: Sage. Smith, D. E. (1987). The eueryd,;1J' world as jJroblematic: A femillist sociology.
Rabinow, l~ (1977). Reflectiolls Cll fieldLUor/~ i1l lVIorocco. Berkeley: Universiry of Boston: Northeastern Univcrsity Press.
California Press. Spradley, ]. P. (1979). The ethllogtap/~ic interview. New York: Holt" Rinehart &
H.aslllussen, p. (1989). Massage parlor prostitution. New York: Irvington. Winston.
Reinharz, S. (1992). Ferninist methods ill social research. New York: Oxford Univer- Spradley, J. E (1980). Participant obseruation. Ncw York: Holt, Rinehart &
Winston.
sitv Press.
Richard;on,1. (1992a). The paeric representation of lives: \X!riting a postmodern Starr, L. (1984). Oral history. In D. Dunaway & \\Z K. Ballin (Eds.), Oral history: An
E;ocio]o5Y' Tn N. K. Den",;n (J:J.), St"Ll;e,;;" s)'n,l>olic illteraction; A rcsc,,,-cIJ interdiSCiplilwry HlIlfJO[Ogy (pp. 3-26). N3shville, TN: American Association
rlIIJllli11 (Vol. 13, pp. 19-18). Greenwich, CT: JAr. for State and Local History
Richardson, L. (1992b). Trash on the corner: Ethics and technography.joltr1wl of
Stc\varr, D. , & 51lam d aS3m,. P. (1990).
. ' . Foclls groups: Theory and practice.
. Newbury

Contemporary Ethnography, 21, 103-119. Park, CA: Sagc.

104 105
Sudarkasa, N. (1986). In a world ofwolllen: Field worl, in a Yoruba community. In
P. Golde (Ed.), Women ill the fi'e1d: Allthropological experi2llces (pp. J 67-
191). Berkeley: Universiry of California Press.
Terkel, S. (1975). \\"larking. New "l'url.;:: Avon.
Thompson, H. (J 985). Hell's Aoge/s. New York: Ballantine.
Thompson, J., & Dernerath, 1,1.]. (1952). Some experiences with the group inter-
3
view. Social Forces, 31, 148-154,
Thr<1sher, F. M. (1963). Thegallf: A stlldy of 1,313 gal1gs ill Chicago. Chicago: Uni-
Rethinlcing Observation
versiry of Chicago Press. (Original work published 1927)
Ulmer, G. (1989). Te!etheory: Gralll1l1ilfology ill nil age of l'ideo. New York:
Routledge. Franz Method to Context
Van I'vlaanen,.1. (1988). Tales of the ti'eld: OllwritiJlg ellmogrl/phy. Chicago: Uni-
versity of Chicago Press.
Vaz, K, 11,11. (Ed.). (1997), Orallwrmtit'e research luith blach Il-'Dmell. Thousand
Oaks, C1\: Sage.
Michael V Angrosino and
\X/arren, C. A. B. (1988). Gender Issues ill field research. Newbury Park, CA: Sage.
Warwick, D. I~ (1973). Tearooll1:raeJc: l\ileans and ends in social research. J-Iastillgs Kimberly A. Mays de Perez
Celller Studies, 1,27-38,
\\?asserf<1li, R. (1993), Reflexivit:/, feminism and difference. Qlli.llitatiue Sociology,
16,23-011.
\V:Jx, R. (1960). T\velve years !arer: An analysis of field experiences. In R. N. ;\cbms
& J,.J. Preiss (Eds.), Hllliuttl orgallizi./tiollol research: Field relatiolls lind lee/;-
l1iques (pp, 166-178). Homewood, IL: Dorsey. .. Observation: Basic Assumptions
Weston, K. (1998). Long slow bllnt: Sexuality and soeLll science. New '{ork:
Routledge. Observation has been characterized as "the fundamental base of all re-
Whyte, \X~ F. (1943). Street comer society: The social stmetllre ofdllltaliall slulJI. search methods" in the social and behavior'll sciences (Adler & Adler,
Chic~go: Universiry of Chicago Press. 1994, p. 389) and as "the mainstay of rhe ed1l1ographic enterprise"
Yeandlc, S. (1984). \'(fomell's llF)rI~illg liues: J{llterrlS and strategies, New '{ork: (\Verner & Schoepfle, 1987, p. 257). Even studies based on direct inter-
Tavistoclc. views employ observational techniques to note body language and other
\'oung, P. (1966). Scienlific social surveys {[/ld research (4th cd.). Englewood Cliffs, gestural cues that lend meaning to the words of rhe persons being inter-
NJ: Prenrice Hall. viewed. Social scientists art observers both of human iKtiviries and of the
Zuckerman, H. (1972). Interviewing In ultra-elire. Public Opinion, 36, 159-175.
physical settings in which such activities take place. Some such observation
may take place in a lab or clinic, in which case the activity may be the result
of a controlled experiment. On the other hand, it is also possible to con-
duct observations in settings that are the "natura!" loci of those activities.
Some schobrs have criticized the very concept of the "namral" setting,

,'u IJlOIIJ' NOTE: \lie thank Alvin IV Wolfe Gnd I(athryn BOI'n",n fOl' shoring theil' in,ighr<
~IH.ll'xpericncl's with us during the prepafiltion of this chapter. \\Ie ,1150 gratefully acknowl"
<:~gl' the advice and cOll5rrucrive criticism of Arthur Bochner, Norm;l1\ Denzin, YVUIln;l
Llllcoln, and l-hrbarOi Tcdlodc.

106 107
 IV\C I n V U J '..II ' - u ......... '- I ",,-, '"'I ,...., ,",' ,,-,,-, ;L."
particularly when fieldwork is conducted in Third \X1orld locations (or in
domestic inner-city sites) that are the products of inherently "unnatural"
colonial relationships (Gupta & Ferguson, 1996c, p. 6), but the designa-
tion is still prevalent throughout the literature. In that case, it is proper to
speak of "naturalistic observation," or fieldwork, which is the focus of this
chapter.
Observations i11 natural setting~; can be rendered as descriptions either
through open-ended narr<.ltive or through the use of published checklists
or field guides (Rossman & Rallis, 1998, p. 137; see Stocking, 19833, for a
historical overview of this dichotomy). In either case, it has generally been
assumed that naturalistic observation does not in tederc with the people or
activities under observation. i\lost social scientists have long recognized
the possibility of the observer's affecting what he or she observes, but care-
ful resei:lrchers arc nonethelesssc.pposed to adhere to rigorous standards
of objective reponing designed [(J overcome that potential bias. Even cul-
tural anthropologists, who havc usually thought of themsdves as "partici-
pam observers" and who have deliberately set out to achieve a degree of
subjective immersion in the cultures they study (Cole, 1983, p. 50;
\Vokott, 1995, p, 66), still claim to be able [0 maintain their scientific
objectivity. Failure to cia so woulclmean that they had "gone native," with
their \-vork consequet1tly rendered suspect as scientific data (Pelto & Pelto,
1978, p. 69). The achievement of the delicate balance between participa-
tion and observation remains the ideal of anthropologists (Stocking,
1983b, p. 8), evcn though it is n,) longer "fetishized" (Gupta & Ferguson,
1996c, p. 37). Objectivity remains central [0 the self-images of most prac-
titioners of the social and behclVioral sciences. Objective rigor has most
often been associated \"lith quantitative research mcthods, and so impor-
tam has been the harmonization of empathy and detachment that even
those dedicated to qualitative methods have devoted considerable effort
to organizing their observational data in the most nearly objective form
(i.e., the form that looks most quamitative) for analysis (see, e.g., Altheide
& Johnson, 1994; Bernard, 1988; rdiles & Huberman, 1994; Silverman,
1993l.
Adler and Adler (1994) have, in fact, suggested that in the future, obser-
vational research will be found as "pan of a Il1t:thoJolO oical spectru11l.,"
but that in that spectrum, it will serve as "the most powerful source of valt-
dation" (p. 389). Observation, they claim, rests on "something researchers
can find constant," by which they mean "their own direct knowledge and
their ovvn judgment" (p. 389). In social science research, as inlega\ cases,
108
.~ ~,." ,,~, .- ' .~""'''.,,'~ '-" ....... , ...... v,'. "v'" ",ct, ''-'''' 'v ,-u,,,c,o,,
eyewitness testimony from trust\vonhy observers has been seen as a par-
ticularly convincing form of verificJ.tion (Pelto LX Pelto, 1978, p. 69). In
actuality, the production of a convincing narrative report of the research
has most often servcd as de facto validation, evcn if the only thing it vali-
dates is the ethnographer's writing skill <mel nor his or her observational
capacities (Kuklick, 1996, p. 60l.
\Vhatever else may be said aboLlt the postmodernist turn in contempo-
rary studics of society and culture, its critique of assumptions about the
objectivity of science and its presllmed authorit3ti ve voice has raised issues
that all qualitative rcsearchers need to address. t Earlier criticism might
have been directed at particular researchers, with rhe question being
whether they had lived up to the expected standards of objecrive scholar-
ship. In rhe postmodcrnist milieu, by contrast, the criticism is directed at
the standards themselves. In effect, it is now possible to question whether
observational objectivity is either desirable or fC<lsible as a goal. James
Clifford (1983a), who has written extensively and criticl1ly ::lbotlt the
study of culture and society, has called into question even the work of the
revered Bronislaw rVlalinowski, the archetype of the scientific participant
observer, who, according to Stocki ng (19838), is the scholar most directly
respOllsible for the "shift in the conccption of the etlll1ogrJ.pher's role,
from that of inquirer to that of participant 'in a \\'.1Y' in village lifc" (p. 93).
Perhaps more surprisingly, Clifford has also questioned the research of the
very influential comemporary imerprerivist Clifford Geertz; he takes
Geertz to task for suggesting that through empathy, the ethnographer can
describe a culture in terms of the meanings specific ro members of that cul-
~ure. In other \ovords, the ethnographer, as a distinct person, disappcars-
lUSt as he or she was supposed to do in l\'1alino\\'ski's more openly
positivistic world. This assessment is echoed by Sewell (1997), who points
OUt that Geenz did not expect ficld-workers to "achieve some miracle of
empathy with the people whose lives they briefly and incompletely share;
they acquite no preternatural capacity to think, feel, and perceive like a
. (p. 40). The problem is not that Gcertz failed to achieve some sort
Idealizcd empathic state; rather, the question is whether such a sratc is
relevant to ethnographic resc8rch, and wherher it is desirable to
aesn'ib, and/or interpret cultures as if those depictions could exist with-
the ethnographer's being part of the ~c[ion.
1 he posr Ino dernlst
. c[Jnque,
., l. . ,
w 11Ch emphaslzcs the Importance of uo-
s,c,cs"lndiino the ethnographer's '''situation'' (e.g., his or hcr gender, class,
as part of interpreting the erhnographic product, is particularly
109
 J'ill.. l E EVl..'-' VE ..... '-'L. ........... ' ",~,,.,, ,_ ... "
salient because the temote, traditional fOlk societies that were the anthro-
pologists' stock-in-trade have virtually disappeared; most cultural an-
~-hropology is now carried our in literate societies that are pan of global
communication and transportation networks. Like sociologists, anthro-
pologists now '\rudy up" (i.e., conduct research among clites) almost as
often as they study the poor and the marginalized. Doing so overcomes
some of rhe problems associac:ed with the lingering COlonialist bias of tra~
ditional ethnography (D. L \X1olf, 1996, p. 37), bur it raises new issues
regarding the position and Status of the observational researcher. For one
thing, ethnographers can no longer claim to be the sole arbiters of knowl-
edge about the societies and cultures they study, because they arc in a posi-
tion to have their analyses read and contested by those for whom they pre-
sUIne t:o speak (Bell &.Iankowiak, 1992; Larcom, 1983, p. 191). In eHect,
objective truth about a society or a culture cannot be established, because
there arc inevitably going to be conflicting versions of what h'-lppened.
Sociologists and other social scientists were \vorking in such settings long
beforc the anthropologists G,me on the scenc, and werc ;:liready beginning
to be aware of the problems inherent in claiming the privilege of obiective,
authoritative knowledge \vh,~n there arc all toO lTlany "natives" ready and
able to chaltcnge them. As lviargery \X1olf (1992) \vryly comment,;;: "\Ve
can no longer assume that an isolated village will not within an amazingly
short period of time move into the circuit of rapid social and economic
change. A barefoot village kid who used to trail along after you luifl onc
day show up on your doorstep with an Oxford degree and your book in
hand" (p. 137). The validity of the traditional assumption, that the truth
can be established through cardul cross-checking of ethnographers' and
insiders' rcpofts, is no longer universally granted, as contemporary social
and behavioral scientists arc increasingly inclined to expect diHerences in
testimony groundcd in gender, class, ethnicity, and other factors that are
not easy to mix into a COllSt:nsus. Ethnographic truth has come to be seen
as a thing of many pans, and no one perspective can claim exclusive privi-
lege in the rcpresentation thereof. Indeed, the result of ethnographic r~~
search "is nevcr rcducible t:J a form of knowledge that can be packaged III
the monologic voice of the erhnographer alone" (1\1arcus, 1997, p. 92):
Some erhnogr:1phers (of various disciplines) have responded to thiS
new situation by revising the ways in which they conduct obscrv:1tio w
based research and present their analyses. No longer can it be taken for
granted that ethnographers operate at a distance from their subject~.
Indeed, the very term subject, with its implicit colonialist connotations, is
110
._. _ . ~ _ .. . . . . . . . . ., ~~~~. '~"~".' '-"'" " ".. CE'U-" 'u ,-UII,,:;-.U
no longer appropriate. Rather, there is said to be a dia/ogIle between
researchers and those whose cultures/societies are to be described...'. Dis..
cllssions of ethnographers' own inter3cticHls, relationships, and emotional
states while in the field have as a result been moved from their traditional
discreet place in acknowledgments or forewords to the centers of the
ethnographies thcmselves. Although this practice has ccrtainly opened up
new horizons in ethnographic reportage, ir raises furrher issues of its own.
For example, because it is likely to be the ethnographers who write up (or
at least collate or edit) the results of the fidd studies, do they not continue
to claim the implicit status of arbiters/mediators of sociallcultural knowl-
edge (\X7olf, 1992, p. 12G)? Ethnographers may assert th<lt they represent
the m3ny voices involved in the research, but we still have only their 3ssm-
ance that such is the case.
Nonctheless, we 1l0\V [unction in a conre:t of "collaborati ve" research,
Collaboratioll no longer refers only to the conduct of multidisciplinary
teams of professional researchers; it often means the presumably equal
participation of professional researchers and their erstwhile "sllbjccts"
(Kuhlmann, 1992; D. L. Wolf, 1996, p. 26), Matsumoto (1996), for exam-
ple, sent a prepared list of questions to the people she was intercsted in
interviewing for an oral history project. She assured them all th3t any
qucstions to which they objected would be eliminatcd. The potential
re.spondcl1ts reactcd favorably to this invitation to participate in the for-
mulation of the research design. As such situations become more common l
L
It IS Important dut we rethink our received notions :lbollt "observation"-
what it is, how it is done, what role it plays in the gcneration of ethno-
graphic knowledge. To that end, it might bc useful to shift from a concen-
t:ario n on observation as a "method" per se to a pcrspectivc that empha-
Sizes observation as a COntcxt for interaction among those involved in the
research collaboration.
+ Observation: The Classic Tradition
As ~ prelude to an exploration of observation-as-context, we \vill brieOv
reView the traditions of observation-as-method thm form the basis of our
~~ercisc ill ""rl;thinking." CO[I::Jcicntiouo I;thIlQsr~lphn~ havc) i.n bet] lono
OLen aW'lr tl
. . natura
e lat 111.
<
- \"IStlc sert1ngs,
, h' ,
t e ll1tcro.ctlOI1 of researcher and
SU l)JCCts of d \ . ,
stu y C<tn c 1ange behavlOrs 111 wavs that would nor have
OCCurred in the absence of such interaction. The}; have believed, however,
111
 /VIr: t nul,.../~ ur I....UI...I...I...I.... t ,,~v ,...,,~
that it is both possible and desirable to develop standardized procedures
thac can "maximize observational efficacy, minimize investigator bias, and
allow for replication and/or verification to check our the degree to \vhich
these procedures have enabled the investigator to produce valid, reliable
data that, when incorporated into his or her published report, will be
regarded by peers as objective findings" (Gold, 1997, p. 397). True ob-
jectivity has been held to be the result of agreement between participants
and observers as to what is really going on in a given situation. Such agree-
ment has been thought to be attained through the elicitation of feedback
[rom those whose behaviors w:-re being reported. Ethnography's "self-
correcting investigative process" has typically included adequate and ap-
propriate sampling procedures, systematic techniques for gathering and
analyzing data, validation of data, avoidance of observer bias, and docu-
mentation of findings (Clifford, 1983b, p. 129; Gold, 1997, p. 399). The
lllJin difference between sociolc,gical and anthropological practitioners of
ethnography seems to have been that the former have generally felr the
need to validate their eyewitne~;s accounts through other fotlTIS of docu-
mentation, whereas the latter !l;lve tended to usc participant observation,
"relativelv unsvstematized" though it may be, as the uhimate reality check
on "aU th~ oth~r, morc refined, research techniques" (Pelto & Pelto, 1978,
p. (9).
The possibility of "observer bias" looms large in the thinking of both
sociologists and anthropologist~in the ethnographic tradition (\V'erner &
Schoepfle, 1987, p. 259). Even setting aside the expected distortion of
ethnocentrism (which can presumably be controlled for as long as the eth-
nographer is consciolls of it), the plain fact is that each pcrson \vho con-
ducts observational research brings his or her distinctive talents and limi~
tations to the enterprise; therefore, the quality of what is recorded
becomes the measure of llSable observational dara (because it can be moni-
tored and replicated) rather than the quality of the observation itself
(which is by definition idiosyncratic and not subject to replication).
Although' th~oretical or 'conceptual frames of analysis inevitably direct
observers' observations, it W8S traditionally assumed that researchers
could keep these in the background when recording basic observational
data. for this reason, the emphasis was placed on observarionalmcthods,
[be bw,,;c; d-,c;n,c; of "\'o'],;d, ~Ya"; a" 01<<':
puts it, "Primary reporting of concrete events and things in field work
should proceed at as Iowa level of abstraction as possible" (Pelto & pelto,
..... .-., ,,....... ' /...." ''-'
in'porrOlnr n'Olnu""]
112
~ ...... ,,~. ,~ ....
1978, p. 70). Theorerical analysis was therefore an epiphenomenon to the
process of observation.
According to Gold (1958), the sociological ethnographers of the first
half of the 20th century often made implicit reference to a typology of
roles that might characterize naturalistic research: the complete pattici-
pam (a highly subjective stance whose scientific validity was suspect), the
parricipanr-as-observer (only slightly less problematic), the observchls-
participant, and the complete observer. The complete observer was one
who was to all intents and purposes rcrnoved from the setting, (lnd who
functioned without interacting in any way with rhose being observed.
Because of the difficulty of maintaining the purity of such a stance (\\lerner
& Schoepfle, 1987, p. 259), and because such research was sometimes
conducted without the informed consent of the observed (an ethical lanse
that is no longer tolerated by responsible social researchers), the obsen:er-
as-participant role was considered an acceptable compromise, allowing
the researcher to interact "casually and nondirectively" with subjects; the
rese;ucher remained a researcher1 however, and did nor cross over the line
into friendship (Adler & Adler1 1994, p. 380). Perhaps the most irnpolTant
COntemporary use of this role is in classroom observ;nional studies con-
ducted by educational researchers (Rossman & Rallis, 1998, p. 137).
Ethnographers trained in sociology are nowadays more inclilled than
were their predecessors to accepr participation as a legitimate base from
which to conduct observation. Adler and Adler (1987) have therefore pro-
~osed a modification of Gold's familiar typology in recognirion of the
mcreasing emphasis in contemporary ethnographic research on "member-
ship ro Ies" "as oppose d"LO to Ies grounded III " pure observation.
" In other
w~rds, the older assumption thar "participation" (which bmhered sociol-
~glSts more than it did anthropologists) seriously compromises the valid-
Ityof observational data has given way to the realities of contemporary re-
~earch, which is often conducted with a greater degree of rese;u~het
Immersion (deliberate or otherwise) in the culture under study than \vas
~nce. considered desirable. Adler and Adler describe, for- example,
penpheral-member researchers" as those \vho believe that they can
develop a des," 'abl " "d ' .. I ".".
. I e illS! er s perspective WIt lOut participating in those
cOnst[ruting [he core of group membership. By contrast, "active-
of f;clcl pro~co:,dllre. ~ 10~c-=>,d,,,,r,," are t:ho,,,,, vviIv bCCQ1l1C iuvu!vcu 'witll lIn:: ct:lIlrul
of the group, sometimes even assuming responsibilities that
the group; they do not, however, necessarily fully commit them-
to members' values and goals. A third cmego;y, th;t of "complete-
J J3
 JVIL , , lV'-'..l V' \...Vi..i..L.\...'" ~'-' H' , r--., ,r--.L.,
member researchers," is composed of those who study serrings in which
they are already members or wich which they become fully affiliated in the
course of research. Even though practitioners in this category celebrate
the "subjectively lived experience," they still strive to use their member-
ship "so as not to alter the flow of interaction unnaturally" (Adler & Adler,
1994, p. 380).
Traditional anthropological ethnographers did not question the utility
of participation or membership as a base for observation, bur they often
worried abour the unsystematic nature of their observational methods.
\Verner and Schoepfle (1987, pp. 262-264) have addressed this concern
by suggesting a typology of observation undertaken in naturalistic settings
that focuses on process rather than on role. In this system, there are three
types of observational proces:" reprcseming increasingly deep under-
standing of the social group under study. First, there is "descriptive obser-
vation," which is, to all intents and purposes, the observation of every-
thing. The ethnographer assumes a childlike attitude, assuming that he or
she knows nothing about what is going on and raking nothing for granted.
Such an approach quickly leads to a morass of "irrelevant minutiae,"
although it is only with increased exposure to the culture that the ethnog-
rapher begins to l111derstand what is and is not irrelevant. At that point, he
or she moves into "focused obs,:::rvation," in which certain things, defined
as irrelevant, can be ignored. Focused observation necessarily entails
interviewing, because the in:;ights gleaned from the experience of
"natives" guide the ethnographer in his or her decisions about what is
more or less important in that culture. Focused observations usually con-
centrate on well-defined typeE; of group activity (e.g., religious rituals,
classroom instruction, political elections). Finally, and most systemaci-
cally, there is "selective observation," in which the ethnographer concen-
trates on the Jrrributes of different types of activities (e.g., apart from the
obvious difference in content, what makes instructing a class in language
arts different from instructing a class in social studies?)
-:? Rethinking Observation as C:onte}:t of Interaction
Contempor:1ry soci[\l resemch m:1Y be ch:1r.:lcterized by (:1) the incl'c:1sing
willingness of ethnographers to affirm or develop a "membership" role
in rhe communities they study, (b) the recognition of rhe possibility that it
may be neither feasible nor possible to harmonize observer and "insider"
114
",,-, i..,." ",.~, '" ~
"""""''''''''::1 ............,,"', 'ucoU". "U'" 'Y'"","UU 'v '-~"'LCA,
perspectives so as to achieve a consensus ,1hoLlt "ethnoguphic truth," and
(c) the transformation of the ersnvhile "subjects" of research into erhnog-
raphers' cO!!8borative partners. The traditional concern with process 8nd
method has therefore been supplemented with (bur by no means sup-
planted by) an interest in the vliays in which ethnographic observers inter-
act vvith or enter into a dialogic relationship with members of the group
being studied. In this secrion, we discuss severa! selected works by contem-
porary ethnographers in order to illustrate these supplemental factors in
the contemporary interactive context of observational research. \'i"!e use
five very general principles of social interaction to organize rhe lollo\ving
review of rhis otherwise quite disparate body of theoretical, methodologi-
cal, and substantive literature.
The Conscious Adoption of 0 Situotionol Identity
The first principle is as follows: The basis of sucial interaction is the
decision (which may be spontaneo1fs or pLlrt ofa care/It! plan) to take part
ill a social setting rather thall react passiuely to a positioll assigned by oth-
ers. In some of the older sociologicalliterarure, this process is referred to
as "role making," as opposed to "role taking." In the conrext of this discus-
sion, this principle animates those ethnographers who actively seek out
situational identitics based on "membership" rather than on "observa-
tion" as traditionally understood.
For example, Angrosino has conducted a long-term study of adults with
mental rctardation and/or chronic mcntal illness who are served by com-
munity-based agencies in the United States. The question at the hean of
this research project concerned hovv these adults, who had been socialized
as Youths in large-scale institutions, adapted to life in the community in
the wake of the move to deinstiturionalize all but the mosr seriously dis-
turbed individuals. Answering such a question required an immersion into
the Jives of these people, because they would not likely respond ade-
quately to questionnaires or clinical survey instruments. Angrosino also
expressed a desire to understand what it might feci like to be mentally
"disabled" in a society that places high value on technical competence. To
Jnvesrigate this issues, it would not be reasonable to "observe" people
s~rved by the sclCClcJ ag<;;tlcic:> in the older, ncutfc,ii5l, objective 111anner
dllScussed above, beclUse the ethnographer could nor presume to be
a. lie to" rea d" t Ile atntu
. des and responses 0 f- peoplc \\'hose behavlOra. I
CUes Wete , by cIe fIIlltlOn,
.. "
not norma. I "0 n t le or ler lao ,engagl11g.111
I I I d .
115

intcnsive imervie\ving in and 0:: itself vmuld not work very well, because
thc clicnts \"/ould not !ileely trUSt someone with \vhom they were not
already familiar outside the im<~rview setting.
Angrosino therefore actively sought out a membership role in the world
in which the cliems lived and vvorked. He did not want to adopt one of the
recognized professional roles that would have been familiar to the diems
(e.g., therapist, social worker, teacher, parole officer) because the very
familiarity \vould have resulted in stereotypical responses. On the other
hand, he could not just "hang ,)ut." Unlike other kinds of communities,
where strangers do often show~lp and stay to become friends, no one juSt
"shov/s up" in a sheltered \vor,.::shop or at a group home. So Angrosino
opted for a role as a "volunteer." He assisted the teacher 35 a rutor in rhe
classroom, he occasionally drove cliems co and frolll 3ppoimmenrs, he
clerked at the thri ft shop run by one of the agencies as a fund-raising effort
and that was staffed by the dieneS, and he helped out at special events (e.g.,
he helped co organize a charity softball game). By assuming these duties,
he made it clear rhat he fit no preconceived model of what someone did in
this community, and yet he was able co demonstrate that he did indeed
have a meaningful function (other than simply "researcher," which would
not have explained anything as far as the clients were concerned). He was
thus able to spend a considerabl,= amount of rime making derailed observa-
tions of the settings in which the cliems lived and worked, because his
presence after a while ceased w be novel cnough to be disruptive, and he
was able to conduct intervicws with the cliems, who had already learned
thar he was someone who could be trusted. (For further details about this
project, see Angrosino, 1992, :l994, 1995a, 1997b, 1998; Angrosino &
Zagnoli, 1992.)
Behar's (1993) study of Esperanza illustrates the ways in which an eth-
nographer \vas lcd, by the force of her collaborator's personality, to adopt
more of a membership role than she had originally expected. Indeed, that
study could fairly be described as an account by a feminist ethnographer
\vho realized only in the course of writing the book that she was a feminist.
In relating the story of a poor .Mexican Indian woman \vho has defined
herself through her life's struggles, Behar comes to understand more abOln:
herself as a Cuban immigrant to the United States who had always felt Ollr-
<:ide the <;oci:l! '1nd '1C''1dCnli.c s;;stcnlS ;,-1 VI.>!l;"'!l sl1e sousl,t rncrnl,c1,Jl;p.
Behar first encountered Esperanza \vhen the latter was selling flowers on a
street corner. Behar asked the woman permission to photograph her,
expecting some son of deferential acquiescence. Instead, Esperanza asked
116
,<0",,,,'3 ~~~~, .~"...",., ''-'''' ,.''-',,'v.... _u .......,' ......,,,
her (in a "haughty" manner) \vhy she needed the picture. Behar (1993)
admits, "I jumped on her as an alluring image of Ivlexican womanhood,
ready to create my own exotic portrait of her, but the image turned around
and spoke back to me, questioning my project and daring me to carry it
our" (p. 4). Behar responded to Esperanza's challenge by questioning her
o\""n assumptions about the power relationship in ethnographic research
(see also D. L. Wolf, 1996, p. 2; M. A. Wolf, 1992, p. 5); in this case, Behar
felt herself to be directed by the more assertive woman she wished to
study. Esperanza could in no sense be described as a "subject" of research.
If she \""ished to understand Esperanza and the world in which she lived,
Behar would ultimately have to become parr of Esperanza's family net-
\\'orl.::; she did so, becoming comadre co Espcranza's daughter in rhe
process.
The decision to insert oneself in a social sctting othcr rhan one's own
has emotiollal consequences, which Behar (1996) discusses at some
length. She translates the old anthropological problem of establishing rap-
pan without '"going native" (a qucstion of methodology, with strong ethi-
cal overtones) into a problem of allowing oneself to be vulnerable withollt
being "too" vulnerable (a question of personal psychology, with strong
moral overtones). She is wary of using the language of theory and analysis;
it is her only tool for "making sense" of new expcriences, bur it is also a
way of distancing herself from an emotionally affecring (and perhaps pain-
ful) encounter.
An even more emotionally affecting (and definitely more painful)
encounter is reported by Eva Ivlorcno (a pseudonym), \vho writes abom
being raped vvhile conducting ficldwork in Ethiopia..Moreno (1995,
p.146) admits that ir is neither feasible nor desirable to "maintain;] fiction
of a genderless self" while in the field, which means thar when an ethnog-
rapher chooses how to express her own sexuality, she musr always be
a\vare of the degree to which she thereby makes herself the object of atten-
rion of others who may see her as a target of (unwanted) sexual advances.
[via reno suspects that the sexual violence she suffered was directed as
mucb against her professional" identity 3S against ber "private" self-
there was, at the time of her research, a generalized hostility aimed at for-
eigners, panicularly dlOse who presented themselves as "experts" and
\\!I~o \'I.'<OICC blull1Cd 1'01: dll.: d vii di~OI:dcr that had ovcrLakcn tile COUl1lry. SlIt:
had heard, for example, that at least one other foreign woman had been
raped by the very police to whom she had gone to report an assault perpe-
trated by local men.lvloreno (1995) concludes that "women must always,
J 17

everywhere, deal with the spectre of sexual violence" (p. lA8). It is diffi-
cult to believe that males are not also victimized in this way-although it is
much less likely that they weuld discuss ir openly-but 1vloreno is
undoubtedly correct in her assumption that as a generalized pattern, sex~
L1al violence is most often, in most situations, directed against women. It
would therefore be a painfully :la'ive female ethnographer \vho was not
prepared to factor this possibility into her plan for her observations, as
"reasonable precautions" should almost certainly affect what, where,
how, and with whom one conducts research.
In SLIm, "making a role" may mean assuming a quasi-professional
stance, becoming part of a family nerwork, or becoming hyperconscious
of one's sexuality-or some combination of them all. In no case is it advan-
tageous ror the ethnographer to be passive in the face of the assumptions
of the community he or she is studying.
The Perception of Power
The second principle is as foHO\vs: In 1l1Ost social interactions, people
assess behavior not ill terms ofits cOllfonnity to social or cultIlralnorms in
the abstract, b1ft in regard to its cOllsistency.. which is a perceived pattem
that somehow makes sense to others ill a given social situation. This princi-
ple is related to the traditional anthropological distinction between
"ideal" and "real" culture. An f:thnographer who took the observer-as-
panici pant role was largely concerned with the ideal culture and took
stcps not to transgress general norms of propriety. But an ethnographer
who actively makes a membership role must be more familiar with behav-
ior as it is lived. Ivlernbers of :1 social group typically work their way
through given situations in ways that do not necessarily conform to thc
principles enshrined in ideal tradition.
\"X7e often function in terms of an ideology that leads us to expect (and,
therefore, possibly also to see) power working downward from whitc,
\Vestern institutions (and their representatives, such as ethnographers) to
various subordinated or margindized peoples. Yet the litcrature is increas-
ingly filled with examples of "how people in subordinate positions mal1-
c-'occl 10o oppose ::lncl eV::lc]e ,:1-,,,, F'IO",~t't":on~ of 1-.i 5 11clC 1""0"",'c,'s" (7\... L-.cl~lc>x.
1996, p. 277). We also have a lingering bias in favor of conceptualizing
both culture and society as unined, cohesive wholes. Yet ethnographers
increasingly find themselves studying "communities" that are defined as
118
"eu '" "'" "d '-'1..1:>"" ~U"V' '. I IV' I' '~Ieu 'uu lU I...UIIlt:'Xt
------~-----------~
much by their conflicts, factions, and divisions as they arc by their com-
monalities (Hubbard, 1997; I'v1cCall, Ngeva, & I'vlbcbe, 1997).
For example, Angrosino (199"1) compiled an oral hisrory of a Bcnedic-
tine monastery in Florida on the occasion of the centennial of its founding.
He spenr a month living at the monastery while conducting the inrervic\vs,
during which time he adhered to the round of tllily prayer, work, and
reflective leisure that is prescribed by the Rule of Sr. Benedict. The re-
sC3fch was approved by the abbot (to whom all the monks in the commu-
nity have vowed obedicnce), and 3lthough Angrosino \vas not a vowed
membcr of the community, the abbot made it clem rhat he expected the
same deference from the ethnographer th3t he received from the monks.
The abbot \vas always very cordi~ll, bllt he demarcated his position vcry
dearly in ways both subtle (sitting behind a desk when it came time for his
interview) and blatant (reserving the final S3Y as to which members of the
communit)' Angrosino could approach for interviews). The other mClll-
bcrs of rhe community claimed to be very supportive of the research, bllt
Ang,rosino found that a fair number of them gave vcry trunGlted inter-
views, explaining that it is nor seemly for a monk to spe3k roo much abou t
his own experiences-doing so smacks of vanity. In the course of living in
and observing the community, Angrosino came to realize that this humble
reticence, although sincere to a point, covered other motives. for one
thing, it expressed a quiet rebellion against the authority of the abbot, who
had mandated their cooperation. In the ordinary course of things, they
would not have dared to be seen as Jess than eager to carry our the abbot's
wishes, but because the ethnographer was, for all his temporarv immer-
Sion in the life of the monastery, an outsider, he could be disobliged in a
~va~' that would have spelled trouble h,]d it been directcd against any
lllslder, much less the abbot himself. Moreover, this tactic allmved some of
the monks to make an oblique criticism of those who had been more fully
cooperative; their own humility was a kind of symbolic indictment of their
brothers, who could be seen as either toadying to the abbot or preening in
their O\vn vanity by "telling all" to rhe researcher. In any evcnr, itWelS clear
that the ideal arrangements of monastic life-with its formal, even codi
lied
. SY' s'tem 0 f I' - IlY aneI d eerence-were
11el.arC [ not what happened HI , real
Ilr:~~. n"'",n ill >;uch a lliob1r circu1H~cribcd cU1LUrc] people I,:Q~lhi V2\pt:riHleIlt
.With St},les of'InteractIOn
, an d lllVO
' Ive t I1e visItor
" , su bt]e, yet very reveal-
In
Ingly subversive, power games, games thar inevitably sha"ped b~t'h what
the ethnographer observed and hOIU he interpreted what he saw.
119

Behar (1993) discusses her inclination to fit Esperanza's story into the
prevailing mode! of feminist studies of Latin American women. She real-
ized, however, rhar Espcranza could not fit the pan of the "exemplary
feminist heroine." The reality of Esperanza was no less admirable and
heroic, for her life was a kind 0:: epic of female struggle, rage, and defiance
of the patriarchal institutions of her culture; but she was a flesh-and-blood
woman capable of "misbehavior," and not a stereotypical Third \\Torld
feminist plaster saint. Behar concludes that an ethnographer's desire to
produce stories that empower [he people she studies must be grounded in
an allowance for the way women in other cultures "misbehave." There
must be respect for their "diff,:::rem ways of making sense," even if their
sense docs nor conform to the European or North American expectations
of the feminist ideal (p. 270). fn a similar vein, Hirsh and Olson (1995,
p. 23) cite Sandra Harrling to the effect that feminist scholarship in general
has sought to surmount the e~;t3.blished categories of social knowledge,
\vhich have been developed from a male point of view (even when applied
in the past by female scholars). lvlargery \'Volf (1992), however, asks
"whether by studying our subjects we are also exploiting them and
whether by attempting to improve \vomen's living situations we are
imposing another (powerful) society's values" (p. 2).lVloreover, it may be
misleading to conceptualize the "power relationship" as that obtaining
between researcher and "subject." In fact, Hsiung (1996) claims that
this standard binary view overlooks the patriarchal context in \vhich both
the (female) ethnographer and her female informants are situated. It may
be more useful to think in terms of a "multidimensional power relation~
ship, of which the patriarchal/capitalist system, individual agents of the
system, female informants, and female feminist researcher are the key con-
stituents" (p. 123).
The injunction to pay attention to what makes sense in a given setting
takes on particular importance when, as is now so often the case, ethnog-
raphy is conducted "witham the ethnos" (Gupta & Ferguson, 1996b,
p. 2). In other words, few ethnographers function within the circum-
scribed communities that lent coherence to the cultures or societies that
figured so prominently in the conceptual frameworks of earlier genera-
tions of observational researchers. It is no longer possible to assume that
"the cultural object of study i.s fully accessible within a particular site"
(lvIarcus, 1997, p. 96).lvluch of the contemporary ethnographic field con-
sists of studies or those who inhabit the "borders bet\-vecn culture areas,"
of localities that demonstrate a diversity of bchavioral and attirudinal
120
'~<=CI""""'~ <JU"CI'UcnJ',. "V'" I.,CU'VC' 'v ,-V, "c_",-,
patterns, of "postcolonial hybrid cultures," and of the social changes and
cultural transformations that typically are found '\vithin interconnected
spaces" (Gupta & Ferguson, 1996a, p. 35). People, after all, "live in differ-
ent overlapping bur not always overdetermining spaces and times: domes-
tic spaces; national spaces; broadcasting and narrowcasting spaces; bio-
graphical times; daily times; scheduled, spontaneous, but also socio-
geological times" (Abu-Lughod, 1997, p. 112).1vlalkki (1996b), for exam-
ple, describes working in "accidental communities of mcmory," which
include "people who have experienced war together ... ; people who \vere
bombed in Hiroshima or Nagasaki; people who all fled a particular revo-
lution; people who are stricken by a particular illness; or people who
worked together on a particular humanitarian or developmellt project"
(p. 92). In all of these cases, "it is the communities that are accidental, nor
the happenings" (p. 92). The ethnographer therefore no longer enjoys the
luxury of assuming that the local scene he or she is observing is somehow
typical or representative of "a" culture of "a" society. It is ~lnexus of inter-
actions defined by "interstitiality and hybridity" (Gupta & Ferguson,
1996a, p. 48), factors of "the globalizing discourses and images of the
media" (Peters, 1996, p. 81), that the ethnographer, the classic neither-
here-nor-there person, helps to define. In some cases, the ethnographer
may even be said to create a community simply by virtue of studying cer-
tain people and by implying that the links he or she has perceived among
them constitute a society. The "street corner society" studied by \X1 hyte
(1955), or Liebow (1967), or Hannerz (1969) became a "society" only
because an ethnographer chose to treat that "nexus of interaction" as a
site. Oral historians are often in the position of creating virtual communi-
ties by linking several persona! experiences around a cemral theme of their
OWn choosing (Hareven, 1996).
The principle is confounded when gender-that "enormous, extreme"
question, in the words of]ean-Frans:ois Lyotard (quoted in Olson, 1995a,
p. 186)-and sexual orientation enter the picture, because "differing sex-
ualized perspectives of 'the field' influence the kind of relationship that
tbe ethnographer has with the field and this, in turn, affects interpreta-
tion" (\\7illson, 1995, p. 253). Gender and sexual orientation are ex-
tremely meaningful elements in defining an ethnographer's personal iden-
tity; they can also btC01l1t filter:; tluuu1:5h vvhid! LIlt: obst:rv<:1tion of
communities is mediated. The problem is that the meanings shift from one
community to another. The observer cannot assume a universal, let alone
an ideal, symbolism of gender and/or sexual orientation. The cues of
121
 j" ... j , IVV'" VI ..... V'--........... I , , " .... ....... ,'V . . . . , ........... , ':""'v ... ,n, "" .....'"' ... ''''"'' '-'" ............J ",-"'-' ' ' ' ' ' ' ' ' ' ' ; } '-'OJ'>'-' ..... "'v,,. "v., ....... ", ..................", ....,..

personal identity must always be interpreted in the context of the reality of The process of open acknowledgment may be hindered in the case of
a given social setting. lesbian or gay ethnographers, who may be habituated to a degree of con-
Dubisch (1995, p. 34), for l~xample, notes that female amhropologists, cealmcnt in both their personal and professional lives. Goodman (1996,
simply by virtue of being female, have not been gr~mted the indulgcnce to p. 50) notes that lesbian and gay male ethnographers extJeci to eng3ge in
engage in casual scx, whethcr their sexual encounters occur at home or in subterfuge whilc in the field, but Burkhart (1996) believes that his initial
the field. IvIale anthropologists, by contrast, have long been assumed to cfforts at concealment (rationalized as an effort to achieve the "idcal of ob-
have had casual flings while in the field (Ncwton, 1993, p. 5). There was a server neutrality") led only to "spells of incrtia and depression" (p. 34).
t3cit 3SSulllption that such male behavior did not matter because it was Williams (1996, p. 74) suggests a compromise: being completely honest
expected and approved by all partics, whereas analogous behavior by with people in the community although less so \\lith granting agencies. He
women \vas always disruptive because it was neither expected nor ap- claims to have had positive experiences v'lith people to \\lhom he divulged
proved. This assumption I11W;t now be called into question, because the his sexual oricntation in the communities in which he conductcd research,
decision of allY ethnographer to insert him~ or herself imo the social set- but he has found it prudent to apply for funding by stressing other research
topics, and thcll studying homosexual behavior once in the field. Even
ting in a sexual mannet must be seen to have repercussions with respect to
AIDS research, now a reasonably well-funded area for social scientists,
what he or she is able to e'bserve. \'\7hether or not such behavior is
was initially not something funding agencies \\Tamed to hear abom,
approved is less important dun the recognition that it will make a differ-
because it was assumed that AIDS was a purely homosexual concern
ence to the entire set of relationships initiated by the ethnographer, and
(BollOn, 1996, p. 157).
hence t'O the type and quality of observations he or she is able to conduct.
Lesbians and gay men arc used to "constant, and conscious, idclltity
In recognition of this reality, Killick (1995) seems t'O counsel abstinence,
managcment" 3nd have typically carried this mind-set from their personal
noting thar "while in the field, the fear of upsetting the delicate balance of
lives into their rescarch settings (Lewin & Leap, 1996, p. 13). A bir num-
relationships with informants is likely to be a significant curb on the
ber of homosexual ethnographers have chosen to study homosexual
libido" (p. 81). On the other hand, KiJlick advises those whose libidos are
behavior in the field (apparently on the assumption that they have a ready
unrcstrained by such methodological niceties to "keep quiet about it if
point of reference), but it is easy to be disappointed if one assumes that the
their behavior is likely to be seen as either uninteresting (a possibility we
understanding of and manifestation of homosexuality is the same in 3lJ
should not discount) or repn::hensible" (p. 81). Altork (1995), however, communities, just as it is easy to be misled into assuming that female
cugucs against both repression and concealment. She believes that "insread ethnographers are in an advantaged position when it comes to undcr-
of blocking out [the] wealth of sensory (and sensual) inpllt, or relegating it standing women in all cultures (Lewin & Leap, 1996, p. 17). The point is
to private field journals, we night consider making room for our sensual that one's gender, as well as one's sexual orientation, are matters that must
responses in our work" (p. 116). In her view, 'vvhether or not one "did it" in be taken into conscious account when one endeavors to conduct observa-
the field is less important than whether or not one is able to be honest in tional ethnography; but neither factor can be considered a source of privi-
3cknowlcdging what did or did not happen and why, because such admis- leged knowledge in and of itself. As Lang (1996) notes, "Quite obviously,
sions leave the ethnographer "open to the fertile possibilities for dialogue there is no 'universal gay community'" (p. 103). At the very least, homo-
about the ways in which 'it' changed, enhanced, or detracted from what sexuality does not "override the social hierarchies of the contemporary
we felt, \vitncssed, and interpreted in the field" (p. 121). In an ironic twist world" (Kenncdy & Davis, 1996, p. 193); it is still necessary to investigate
on this old (but only recently public) dilemma, Altork suggests that "per~ the impact of "the hierarchies of class and race" even within a presumed
haps by aclmo\vledging our own feelings and desires, we might actLl811y "gay community."
look at other people and places more objectively, by being able to ferret In any case, it is clear that "no longer is it generally acceptable for
om our O\\'n biases and distortions as we do our work" (p. 132; emphasis [ethnographic researchers] to conceal or deny the significance of their
added). gtnder identity, age, class, or ethnicity. (Sexual identity represents a sort of

122 123

final frontier in this regard.) Instead, contemporary ethnographic writing
tends to acknowledge these attributes as factors that shape an [ethnogra-
pher's] interprctations of what she or he observed in the field" (\X7eston,
1996, p. 276). For example, Edelman (1996) discusses the varying impacts
his Jewish erhnicity had in tl1rel~ different field sites. Othcr factors, less
\vdl established as demographic categories, may also playa parr in how an
ethnographer relates to what is studied-for example, the ethnographer's
personal struggle with bulimia (Tillman-Healy, 1996) or breast cancer
(Kolker, 1996), or the ethnographer's having survived a nonmainstream
childhood (Fox, 1996; Ronai, 1996) or having undergone detoxification
therapy (lv1ienczakO\vski, 1996). In all of the cases just cited, the eth-
nographers' personal experiences were the main focus of both observation
and interpretation. But it is clear that if these ethnographers should go on
to study other people with those same characteristics, they would have to
shift from a perspective that implicitly elevates the personal to the norma-
tive in order to observe what is going on in a natural setting.
I~egotioting a Situational Identity
The third principle is as follows: Interaction is ahuays a tentatiue pro~
cess that iJllJolues the cOlltinuous testing by all participants of the concep-
tions they have of the roles of others. In other words, ethnographers and
their collaborators do not step into fixed and fully defined positions;
rather, their behaviors and expectations of each other are part of a
dynamic process that continues to grow (one hopes in healthy ways,
although the outcome is sometirnes problematic) throughout the course of
single research projects or as they move from one project to another
(Wolcott, 1995, p. 77). Giroux (1995) speaks of the need for "intellectu-
als" in general (and ethnographers in particular) to "reinvent themselves
in diverse sites" (p. 197). Denzin (1997a) discusses the "mobile conscious-
ness" of an ethnographer who is aware of his or her "relationship to an
ever-changing external world" (p. 46).
For example, Angrosino (1997a) conducted an oral history of the
Southern Anthropological Society on the occasion of the 30th anniversary
or its founding. He had been a member of the SAS almost since its estab-
lishment ::.md h~ld served over the years in both elective and appointive
offices. As a professional anrhr'Jpologist affiliated with a department at a
university in the South, and whJse research often dealt with aspects of life
in the contemporary South, he was in terms of status a fully integwted
124
r,t;:'Lf III 11\11 lid VU:>t:1 vuuu/ J. nurtl tYlt;'U IUU [U I...UrI[f;;'XI
member of the institutional culture of the organization.1vforeover, he had
long-term personal and professional ties \vith all of the people who were
scheduled w be interviewed. The bulk of the interviews vvere conducted
during the special anniversary meetings of the society, and Angrosino
brought along three graduate students w work as his assistants-and l
more important, to serve as "reality checks" to make sure that he did not
act like too much of an "insider" and thereby miss important cues or take
for granted too many items that outsiders would find in need of clarifica-
tion. The interviews, however, began awkwardly, as many of the partici-
pams seemed annoyed at being questioned by someone they assumed
already knew the answers. "Oh, you remember what happened in New
Orleans in '70 _.. " someone would say. "\\7ell, why don't you tell about it
in your own \vords?" Angrosino would respond. They usually sighed in
frustration at that suggestion_ I twas vcry difficult for professional anthro-
pologists to act as informants, particularly \vhen the interviewer was
already assumed to be in the know. Some others decided to shorr~circuitan
uncomfortable situation and, in effeer, to hijack the interview, carrying on
in lecture/monologue fashion widlO11t paying attention to the inter-
viewer's questions. Still others demanded to be interviewed by one of the
graduate assistants; "I can't talk to you with a straight face," one of them
told Angrosino.
AIrer a while, the awkwardness wore off, preslImably as members began
to share with one another their reactions to having been interviewed. They
reaffirmed all the reasons they had thought of collecting an oral history in
the first place, primarily because doing so would be a good idea "for pos-
terity." 'The later interviews went much more smoothly, as participants had
clearly made a tacit decision to treat the overly familiar Angrosino as sim-
ply a n:live outsider to whom euerythillg needed to be explained. The eth-
nographer came to think of himself as if he were onc of his students, so
that he would remember ro ask all the questions thar someone who had
nor been in on the action would want ask. Although most ethnographers
seek to move from outsider status to a status of participant/member,
Angrosino in this case (abetted by his collaborators) reinvented himself
~rom complete insider to interested-bur-ignorant bystander. \X/ithin the
Interactive context of observational research, roles mutate in response to
chantiiuS circumstances and arc ncvcr ddined )ViLh firwlity,
Behar describes this process in terms oEher own evoillti~n from "femi-
nist anthropologist" to "feminist ethnographer," by which she means a
researcher who is attentive to the "reflexiveness about the politics of
125
 .- - - -, ~~---~'" .....' rn ...... r" ......... , L." ''-'
practicing feminism and exper.:mental cultural writing." The focus of such
reflexiveness must be '\vomen's relationships to other women" (Behar,
1993, p. 301) rather than a scientific observer's relationship to a "subject"
of research. For Behar (1996, p. 5), rhe very term participant observation
is an "oxymoron." The ethnographer must, Behar suggests, be defined by
the creative tension in the role of member/observer, not by some finit~
quantity of information gathcn.:d by one who plays that role. According to
Ferguson (1996, p. 153), the exploration of how shifting connections
frame experiences of place, community, and society among "the partially
c:nd .provisionally dislocated" may well represent the most fertile ground
tor future ethnographic research.
Biackwood (1995, p. 53) therefore speaks of the identities assumed bv
the ethnographer in the field in terms of the many different ways he or sh~
is perceived by "others." As a woman, she has "continually t~lcked back
and forth be(\veen various assigned and consrructed identities: researcher,
friend, daughter, professional, American" (p. 58), and she concludes that
"idenrities (in the field] are nev,:r stable, never simply defined" (p. 70). At
rhis point in tbe psychosocial history of \Vestern culrure, ir is probably
clem that our "identities" in the existential sense are always in a process of
evolurion and never achieve a fixed, final poinr. The naIve assumption thar
ethnographers' identities in rhe field should be clearly defined and finite
is, perhaps, the last vesrige of the old belief thar "who!eness" means per-
sonal :1uronomy and fixity of identity. Nevertheless, there is a very srrong
sense in which erhnographers continue to believe, as they have since rhe
days of the ascendancy of the Freudian perspective in social analysis, rhat
doing fieldwork is a way in which they can come to rerms with themselves.
"Sharing a different lifestyle," according to Barnett (1983), "has a mirror
dfecr, providing glimpses of an observer's foibles as well as his dignity"
(I'. 169).
\'{ialters (1996) expresses this vie\v with an interesting figure of speech:
In rhe matter of establishing one's identity, one must "constantly ... pivot
rhe cenrer" (p. 63). She was led ro this perspective when it became neces-
sary for her ro deal with the problem of bringing her partner to the field-
the Yemen Arab Republic. She a~:sumed that a conservative Islamic society
\vould not be a friendly place for t\'vo openly homosexual women, and she
considered various strategies of concealment (e.g., claiming to be rela-
tives, going through a process ot adoption); she finally decided to refer to
rhc other \voman:1S her "companion," which, while perfecrly true, seemed
vague enough ro avoid the suspicion of the authorities. (It even had a whiff
126
'-'.Ir "\.II....J'-\L IVIMI C"I"'IL) I\.'="'''''''-'''':::1 '-.-'u"",,;:, 'UCo,-", "'-"" "'~.,. .. _ ..
of Vicrorian propriety abour it-a young woman trclVcling in the company
of an older female companion has long been a well-known im:1gc in many
parts of the world.)
Despite such occasionally sLlccessful srrategies, it remains true that
\,vomcn researchers ofren feel pressured to conform to the gender behav-
ior norms of rhe cultures they study, even if those norms are not the ones
the V would freely choose for themselves, and even if they have ro resorr to
a c~rtain amount of deception so as to appear ro conform. According ro
Diane \X'olf (1996), "Feminisr fieldworkers have lied abour their marital
sums, ... about rheir national identity or ethnic/reI igious background ... ,
about divorce and former marriages ... , and about their class back-
ground" (p. 11). Even in rhose caseS where the deception did no real harm
to the people the researcher was observing, the ethnographers often felt
guilty, in parr because the very act of deception "directly contr.::lclicts
attempts at a more feminist approach to fieldwork, which includes
attempts to equalize a relationship and create more of a friendship" (Berile,
1996, p. 56; see also D. L. \\lolf, 1996, p. 12). On the othcr hanel,
the refusal to deceive-in eHecr, ro defy the norms of the community be-
ing studied in order to make a principled stand for what one actually
believes-can sometimes have uninrenclecl negative consequences. Berik
(1996, p. 65), for example, :1c!mits rhar her openly feminisr stance whilc
conducting research in a Turkish village unwirtingly led ro onc of her
female informants' being beaten by an outraged husband, \'I'ho assumed
she was being led asrray by the insufficienrly submissive ethnographer.
Lang (1996) conducred a research project focusing on Native American
lesbians. Although she is openly gay herself, she found it unacceptable ro
locate potential informants in bars or other obvious meeting places. She
Was concerned lest tbe other women assume, because she was hanging om
in a bar, that she was therefore interested in finding scxual, rarher than re-
search, partners. She decided that it would be unethical to pose as a poten-
tial sexual partner in order ro dicit information (Lang, 1996, p. 94), and
so she had to seek her "community" in less symbolically charged environ-
ments. In this case, lang decided that her status as "ethnographer" would
take precedence over her sexual orienrarion as a way of defining herself ro
the people she intended to study, despite rhe facr thar sexual orientation
Was thc thcmatic focus of her study. Her decision was a matter of strategy
dictated by her reading of the nature of the particular community.
IV[uch ~f the recent literature bearing on the creation, mainrenance, and
creative evolution of observers' identities (and on the pros and cons of
127
 Ml: J HUU::. Vr '-..VLLC'-..I fl'll..:l f-\rH.) f\1'If-\Ll LII'11..:l
deception and disclosure) has dcJlt \vitb issues particular to women and
lesbians/gay men, as shown abov,~. It is worth mcnrioning, however, that
there are other issues or identity that arc of concern to researchers who
stud v silUations of political unrest and who come to be identified \vith
poli~ically proscribed groups (Hammond, 1996; Mahmood, 1996; Sluka,
1990), or who work with groups that are defined by their need for decep-
tive concealmem, such as iHeg81 migrants (Chavez, Flores, & Lopez-
Garza, 1990; Stcpick & Stepick,J990) or those involved in criminal activ-
ities (Agar & Feldman, 1980; Brewer, 1991; Dembo, Hughes,Jackson, &
IVlieczkowski, 1993; Koestcr, 1994; van Gelder & Kaplan, 1992).
Criteria for Validation
The fourth principle is as follows: Participmlts validate the cues gel/er-
Llted by others ill the settillg by illterJlal and/or external criteria. Internal
criteria are d105e by which mcmbers of a community check their behavior
against the prevailing norms of their own group. External criteria are
those by which members of a conIn1Unity check their behavior in termS of
presumably universal stanclards. In other words, panicipams in the inrcr-
action ask, "Does it work?" or, f:,erhaps less nobly, "Can I get away with
it?" (that is, "Docs my inrerpretarion help me and Illy potential collabora~
tor work our a vtable relationship?") rather than, "Is it correct?" (That is,
"Does 'the culmre' somehow 'require' people to act in a certain manner?")
For example, Angrosino (1995b) led a team of graduatc students in a
study of local tesponses to the AIDS epidemic. The research centered on a
particular agency that provided limited direct service (mostly testing) bur
\".'as far more important as an information and referral network. The
agency had been founded by partners and relatives or people who had con-
tracted AIDS; these people believed that their personal involvement in-
indeed, their emotional commitment to-the cause was a primary reason
for the success of their project. They were somewhat put out by the appar-
cnt transformation of the agency into a more professional outfit, with
leadership positions being increasingly taken by human service managers
and development specialists with no special concern for the particular
characteristics of the AIDS crisis except for the recognition thar it was a
t11:ljor public hC:11th concern. Th~ profcss;onal
the researchers, but in a rather distant fashion; by contrast, the "01 d-I'me "
founders were eager t:o draw the leam members into a kind of social circle,
the better for them to learn "wlnt it's really all about."
128
CIVlrlr',.I\...ML Ivl,.....' L"-'MLJ ,,<:t' '" ''''I 'y LlL)::>t::1 VUl-'UI ,. I Iu,r I IVI(::l1IUU tU \.-VII(t',>;(
This situation was one in which internal and external criteria seemed [0
work at cross-purposes. On the one hand, the two factions within the
agency were reaccing to very di Herent criteria for validation. One side val-
idated its activities and sense of mission with reference to the quality of
interpersonal interactions (an internal criterion) and to genera! humani-
tarian concerns (an cxternal criterion). The other group sought validation
in the efficiellcy of interpersonal interactions (an internal criterion) and in
"objective" standards of professional conduct (an external criterion).
There \vas clearly no "correer" corporate culture in SlIeh an agency or in
the client community it served; the question \vas, "\'{!hat \vorked?" and the
answer was, "It depends on \vhat you wam to accomplish." The founders
vigorously sought to convince the ethnographers that the key to success in
the crisis was to "work from the hean, not from thc head." People \vith
AIDS and their caretakers needed emotional support more than sound
fiscal management from an agency like theirs. They did not deny that the
clients appreciated good management, bur believed that a coolly compe-
tent accountant was nor the "face" potential clients wanted to see \vhen
they contacted the agency for help. They sought empathy, having already
gotten qllite enough unfeeling "competence" in clinical settings. The eth-
nographers, for their parr, tended to agree, although they certainly appre-
ciated the way in \vhich the "competent" managers facilitated the research
process for them.
It may be useful to characterize this aspect of the interactive COntCXt in
tcrms of the ways in which personal experience serves as an organizing
principle in the process of mediating internal and external criteria in social
settings. For Denzin (1997a), "the starring point is experience" (p. 55),
which leads to a c1i5course between the ethnographer and other members
of the community, (] discourse that"" often begins from the painful autobio-
graphical experiences of the writer" (p. 57). Indeed, there is increasing
tolerance for a discourse that ends with those same cxperiences as well
(Quinney, 1996), "Liie in the iield," Hinsley (1983) points out, "is an indi-
vidual experience" (p. 55), and "ethnography" (despite the traditional
connotation of the term as the study of "a people") is seen in certain quar-
ters as a species of autobiography, the "personal ethnography" (Quinney,
199 6), Olson (1995b) cites Donn:1 Harawav to the effect that eth-
m.anager<; v.'c'c l,elpfu! to nograph;c ohscrvation lTltlst be tr<JllsL:Hl.:d in;u ),TiLtGlI rcprGit::I1WLiOllj
that place "the writer's own situatedness in history" in the foreground
(~. 46). Denzin's (1997a)-and, more obliquely, Olson's-remarks are
directed mainly to the production of "standpoint texts" that flow from the
129

particular experiences of those who have been excluded from "the domi-
nam discourses in the human disciplines" (p. 55). Stocking (1983b) refers
to the same trend in somewhat less favorable tetms as the proliferation of
"adjectival anthropologies" (p. 4).
It is nevertheless certainly possible to apply the same perspecrive to
anyone engaging in ethnographic research. Even those who come from
traditionally "dominant" ~;ocial groups must engage in a process of con-
sciousness~raising about the nature and effects of their interaction with
orhers. For rhem, as for tbose previously marginalized, the starring point
of observational research is experience, for theit own existential immer-
sion in the "cultural displacement" of people, things, and cultural prod-
ucts is a defining quality of the state of the world roday (lvlalkki, 1996a,
p. 53). According to l.vlary Belenky: "We all need to understand how writ-
ing the same material for different audiences changes the voice. This is
very empowering knowledge to have" (quoted in Ashton-Jones &
Th~mas:1995, p. 86). 0:1 the other hand, one cannot be "preoccupied"
with one's atldience, because such a focus can lead to "self-censorship,"
according to bell hooks (quoted in Olson & Hirsh, 1995, p. 110). In
effect, the ethnographer who is a member/observer is an artifact of the
very situation of cultural displacement that he or she intends to study. It
may not, in fact, be possible to resolve the tension between what the eth-
nographer "is" and what n.e or she must "become" in the field; rather than
fret about that tension, it may now be time to '''find some practical use" for
it in our analysis (M. A. Wolf, 1996, p. 217).
Bolton (1995) points out that when the rapic of an ethnographic srudy
is sexuality, the ethnographer is limited by his or her inability to "observe"
the behavior in the strict, traditional sense of the term. h1uch of the social
scientific discourse on sexual behavior has been recorded via hearsay
rather than evewitness testimony, although it has often been conducted
within the cOl~pass of supposedly observational research designs. The best
way for the ethnographer to overcome this limitation, Bolron suggests, is
through participation, if the ethnographer feels comfortable doing so. I-Ie
admits that "I learned m'Jre through participation than by simple observa-
tion or direct intcrviewi::l.g" (p. 148). This solution raises some additional
issue~ of all interactional nature: Is the ethnographer who "participates"
in this manner really learning about the norms of sexual practice in [he
community he or she is studying, or is the ethnographer importing atti~
tudes and emotions from his or her own culture into the field setting? Is
the ethnographer, in effect, confusing internal and external criteria? Ivlor c
130
Kernml\Jng VDServatlon: ,rom MetllOQ 10 Lontext
provocatively, is he or she participating in the creation of a new set of
norms or standards that are specific to this parricular interaction, and not
of either the host community or the home community? If the larter, is the
ethnographer still doing social research, or has a new field for observation
been introduced, requiring at minimum the much-discussed "blurring of
genres" in reporting and at maximum a blurring of traditional academic/
disciplinary boundaries in order to conceive of a new topic of discourse?
As 1vlurray (1996) narcs: "Having sex with the natives is nO[ a royal road
to insight about alien sexuality.... In answering questions or inscribing
life histories at a researcher's behest, as in having sex with them, the per-
son whose sexuality is being studied is likely to be guessing what the re~
searcher wants to hear rather than representing his Or her most fundamen-
tal desires and identities" (p. 250).
A female variant of Bolton's point of view is provided by Gearing
(1995), who fell in love with and married her "best informant" while con~
ducting field research. Ethnographic research, she conrends, is always a
"joint endeavor" between the would-be observer and those he or she
would observe; it is therefore dependent on the "quality of our personal
relationships" (p. 207). Gearing advocates abandoning the "model of the
dispassionate parricipant observer" and adopting instead the persona of
"an emotionally aware inrer-actor engaged with other actors" (p. 211).
The validation of the individual experience of the ethnographer has
traditionally been bound up in the ethnography (usually written) pro-
duced as the result of observational research. The only audience that really
matters has been the academic, although there have been recent attempts
to write for an audicnce composed (at least in part) of rhe "subjects" of re-
search. Nevertheless, there arc now many formats in which a reporr can be
generated (Polkinghorne, 1997), reflecting the variety of constituencies to
which the ethnographer is now responsible. Thus the ethnographic ob-
server must be concerned with the different "voices" in which he or she
presents material. Traditional ethnographic reportage favored the sup-
posedly objective third-person voice, emanating from the "omniscient
~arrator," as Tierney (1997, p. 27) notes. The move toward greater parric-
lpation allowed the ethnographer to acknowledge his or her own pres
tnce, although this was often done via circumlocution, with the ethnog-
'"pher referring [Q llim- or herself, [or example, d' ""[he [men[ewcr"
(Tierney, 1997, p. 26). The once-banned "f" is now much more common
as ~ubjective experience comes to the fore (Tierney, 1997, p. 25; see also
ElliS, 1997; Lather, 1997; Tanaka, 1997). Wolcott (1995) declares his
131
 IYIC r nVlJ.) Vr- \",ULLt::\... I WID Mr'llJ MI'lMLT LlI'l\.;:J I:NWII(I\...,\L MAl tl(IAL::'
preference for "an approach that keeps humans always visibly present, re-
searcher as well as the researched" (p. 15). JVlargery \\lolf (1992, p. 52)
suggc,\[j rim ir W3S women dud orher, rreviolldy morginalized by the aca-
demic world who first dared challenge orthodoxy by writing in the first
person, a trend that she believes has now entered the mainstream-now
that male academics are also doing so. These shifts in reference are not
irrelevant matters of style; they reflect evolving self-images of the
ethnographic observer, changing relcltions between the observer and the
observed, and new perceptions about the diverse (and possiblv even con-
tradictory) audiences to whom ethnographic research must ;lOW bc acl-
dressed. Certain kinds of ethnographic texts can have professional, partic-
ipatory, lay, and aesthetic audiences (Denzin, 1997b, p. 188). It: is de3f
that validating what "I" say is a very different matter-philosophically as
well as scientifically-from validating \vhat "the interviewer" says.
Perhaps the most widely cited case study of the subtle interplay of inter-
nal and cxternal criteria is the "thrice-told tale" of !vIargery \\lolf (1992).
In her hook, \\1olf embarks on a personal, reflexive journey by revisiting a
fictional shorr story she had written some 30 years earlier, when she was
the \vife/assistant of an 3mh ropologist conducting his first field reseclrch in
Taiwan. Since that time, she had become an anthropologist in her own
righr, as weJ] as a feminist. The original story (which was based on real
events) is reprinted first, followed by the field notes and journal entries
referring to the same events. The third part of the book is a formal eth-
nographic article that \vas originally published in Americall Ethnologist, a
mainstream academic journal. Each section of the book is followed by
commenrary in which Wolf explains what she remembers about the events
as represenred in cach of the three written accounts and what she now
rhinks about those same events from the perspective of three decades. In
the process shc sees changes in herself as both a woman and an ethnogra-
pher. For example, she notes, "\"X!here once I \vas satisfied to describe what
I thought I saw and heard as accurately as possible, to the point of trying to
resolve differences of opinion among my informants, I have comc to real-
ize the importance of retaining these 'contested meanings' " (p. 4).
\X10lf's point is that no ethnographic research, including supposedly
objective naturalistic observation, can be considered complete and valid
until it has undergone what Pollcinghorne (1997) describes as the transfor-
mation of a "list or sequence of disconnected research events into a unified
story with a thematic point" (p. 14). Agood observer can develop the skill
of catching cultural meanings as members of the community themselves
132
r\t;'u'''lI\r,ry \Ju"",r vUe,v".' 'v", "''='''''vv ev ,-v',ec",
underst3nd them, but equally important is the skill of writing up the report
in such a way as "to convcy that meaning to an interested reader from
another culture" (Wolf, 1992, p, 5).
Contextualizing Meaning
The fifth principle is as follows: People corne into interactions by
assuming situational identities that enhance their Olun self-collccpt;olls or
SC1"UC their oWl111eeds, which may be context specific rather than socially or
clflturally nonnative. IVlembers of the community arc reacting ro this par-
t;cu/ar ethnographer and the cues he or she generates, not to "an outsider"
in 3. generic sensc. Some of those cues are matters over which the ethnogra-
pher can exercise some control if he or she is made aware of [hem (e.g.,
improving language facility, dressing in an "appropriate" W<lY), although
many others are simply parr of the package (e.g., gender, race/erhniciry,
relarive age). In the latter case, the ethnographer n13)' need to realize that
what he or she observes is conditioned by \vho he or she is, and that dif-
ferent ethnographers-equally well trained and well versed in 1:heory and
method bm of different gender, race, or age-might well stimulmc a very
different set of intcractions, and hence a different set of observations lead-
ing to a different set of conclusions.
Angrosino, for example, has been involved \vith a long-term project
documenting the patterns and impacts of inter-island labor migration in
the Netherlands Antilles. One of his informants was an elderly woman
now living on Saba, the smalJest of the islands, and he published her life
history in an anthology devoted to the Saba parr of the project. Shonly
thereafter, the same woman was interviewed by a Saba-borr: folklorist
(and political leader) who was publishing a collection dedicated to "the
island's treasures" (i.e., the accumulated wisdom of its senior citizens).
The general outlines of the woman's life story were the samc in both
accounrs, but there were clear differences as well. As might be expected,
the story she told Angrosino had many more explanatory details than
the one she told her fellow islander. Angrosino had obviously asked her
l1lany questions to clarify matters about which a nonnative would have
no knowledge. But there were more subtle, yet telling differ,~nces. The
Woman had lived a life of great hmdship, and yet she had survived to
raise (Virtually single-handedly) a large family; all of her children had
gone On to become pillars of the community, and she herself was recog~
nizecl as a person of the utmost integrity. \\7hcn she told her story to the
133

ethnographer, she allmved herself a bit of pride in recounting how she had
surmounted all her travails; she comes across in that account as a humble,
yet definitely heroic figure, In the story she told her fellow 5abian, she is
considerably more sc!f~deprecating; the island culture is nOt very cordial
to those who "try to get above themselves." On the other hand, she in-
cluded anecdotes in her discourse with the Sabian about her defiance of
the white establishment, incidents she suppressed when talking with the
(white) ethnographer. The point is that Angrosino, who is white, from
another country, and of the same generation as this woman's grandsons,
evoked a qualitatively different story from the one she told the bbck
Sabian of her own generation. There is little evidence of consciolls dissim-
ulation; she merely responded to cues both obviolls and covert in her two
"audiences," ;md, like any good performer, she engaged her interlocutors
in terms that resonated most clearly \vith them and their personal circum-
stances. (For a more detailed comparison of the two life stories, sec
Angrosino, 1989.) As Behar (1996) notes, citing George Devereux, the ob-
server "never observes the behavioral event which 'would have taken
place' in his absence, nor hears an account identical with rhar v1lhich rhe
same narrator would give to another person" (p. 6).
Denzin (1997a) points out that it is now imporwnt to be ,nvare of cldss,
race, gender, and ethnicitYl and of how these factors "shape the process of
inquiry, thcreby making research a multicultural process" (p. 19). This
insight is not in and of itself new; what is new and important for the pur-
poses of this discussion is the implication that the ethnographer must
become aware of these factors not to minimize them or "hold them con-
stant," as cbssic observers were taught to do, bur to integrate them cre-
<Hively into both the proccss of observation and the production of a writ-
ten representation of the fruits of that observation. Diane \\Tolf (1996)
echoes this position in a feminist context; she advises ethnographers to
analyze their field research in terms that use, rather than deny, their "intLl-
ition, feelings, and viewpoint" (p. 5).
lvlorron (1995), for example, conducted research in Tonga, a "seduc-
tive," "exotic" culture. On her first visit, she "nearly" succumbed to the
scduction; 10 years later, she was pregnant while conducting research, and
she WdS able to wear her pregnancy as a "chastity belt" to dvoid sexual pur-
suit (p. 168). The reader assumes that lvlorron had other Illatives for her
pregnancYl bur it certainly helped her out of an undesirable situation (that
of being sexuaIJy active in the field); it probably also meant chat she waS
134
rcefnlnl(lny VlJ:.ervuuur I. nUIl1 IVlt'LIIUU LV ,-UI HeM
able to hear a significanrly different side of the stor)' of Tongan culture
from the one to which she was privy as a single, childless woman.
<So The Ethical Dimension of Observational Research
Observation was once thought of as a data collection technique employed
primarily by ethnographers who thought of themselves as e,bjective re-
searchers extrinsic to the social settings they studied. 1t has become a con-
text in which researchers who define themselves dS members of those social
settings interact in dialogic fashion with other members of th=>se settings.
This ~allsition has also effected a shift in the parameters of our ongoing
reflections on the ethics of social research.
Institutional Structures
For good or ill 1 virtually all social research in our time is governed by
the structure of institutional review boards (1RBs), which grew out of fed-
eral regulations beginning in the 1960s that mandated informed consent
for all those participating in federally funded research. Th~ ~erc.eiv~d
threat was from "intrusive" research (usually biomedical), partiCIpatIOn III
which was to be under the control of the "subjects," who h,ld a right to
know what was going to happen to them and to agree formally to all provi-
sions of the research. The right of informed consent, and the review
boards that were eventually created to enforce it at each institution receiv-
ing federal moneys (assuming a function originally carried out by the fed-
eral Office of Ivlanagement and Budget), radically altered the power rela-
tionship between researcher and "subject," allowing both parties to have a
say in the conduct and character of research. (For more deraikd reviev/s of
this history, see Fluehr-Lobban 1 1994; Wa..x & Cassell, 1979.)
Ethnogr3phic researchcrs, however, have 31ways been uncomfortable
with this situation-not, of course, because they wanted to conduct
covert, harmful research, but because they did not believe chat their re-
search \Vas "intrusive." Such a claim was of a piece ,vith the assumptions
typical of the "observer-as-participant" role, although it is certainly possi-
ble to imerpret it as a relic of the "paternalism" that traditional researchers
often adopted with regard to their "subjects" (Fluehr-Lobban, 1994, p. 8).
Ethnographers \vere also concerned that the proposals sent to I~Bs had
to be fairly complete 1 so that all possibilities for doing harm mtght be
135
 ~-~....-. ~""-""''-'-'--'''''''''''''''''-'"""""L, ,-".,\,;l
adequately assessed. Their research, they argued, often grew and changed
as it went along cll1d could not always be set our with the kind of predeter-
mined 5pccificiry [har rhE lEgal expem seemed to expect. They hmlw,
pointed our that the statements of professional ethics promulgated by the
relevant disciplinary associations already provided for informed consent,
such that IRBs were merely being redundant in their oversighr.
In the 1980s, social scientists won from the federal Department of
Health and I-Iuman Services an exemption from review for <Ill social re-
se8rch except tbat dealing v.lith children, people with disabilities, and oth-
ers defined as members of "vulner<lble" populations. Nevertheless, legal
advisers at many universities (including the University of Sourh Florid3,
"",here we are both based) have opted for caution and have been very reluc-
tant to allow this near-blanket exemption to be applied. Indeed, at USF it is
possible for a research proposal to undergo "expedited" (or "parrial")
rcview if it seems ro meet the federal criteria for excmption, and even
those that are judged worthy of ful! exemption must still be on file. USF
ntnv has two IRBs-one for biomedical resc;.1fch and one for "behavioral
research." Because the Ianer is dominated by psychologists (by far the
largest department in the social sciences division of the College of Arts and
Sciences), this separate status rarely works to the satisfaction of ethno-
graphic researchers. The psychologists, used to de<lling with hypothesis-
testing, experimental, or clinical or lab-based research, h3ve been reluc-
tant to recognize a subcategory of "observational" research design. As a
result, the form currently required by the behavioral research IRB is
couched in terms of the individual subject rather than in terms of popula-
tions or communities, <lnd it mandates the statement of a hypothesis ro be
tested and a "prorocoI for the experiment." Concerned ethnogr<lphers at
USF have discovered that some other institutions have developed forms
more congenial to their particular needs, bur as of this writing they have
had no success in convincing the USF authorities to adopt any of them as
an alternative to the current "behavioral research" form for review.
It is interesting to note that the only kind of "observational" rese<lfch
that is explicitly mentioned and routinely placed in rhe "exempt" category
at USF is that defined as "public"-for example, studying patterns of
vvhere people sit in airport waiting rooms, one of the rare remaining cla5~
sic "pure observer" types of ethnography. The exemption, however, disar~
pears if the researcher intends to publish photos or othenvise identify the
people who make up "the public."
136
LJVI,Irl.Il.... ...... L IVI...... I t:l\)!-\.LJ /I.I;:U III 11\11 Iy vU.:>C'Jvut/UJJ. I rUIIII.,C'uruu <1.1 '-U"CC'AC
Issues for Contemporary Observoflonal Researchers
P-rhiC:ll cthnogr;1phclg \vho adopt more cle:1rl y "mcmbership'1-ot;cntcd
identities-certainly <I very strong trend, 3S this review has demon-
strated-are therefore cauglu between t\vo equally untenable models of
research. On the one hand is the official IRB, which is tied to the experi-
mental, hypothesis-testing, clinical model. On the other hand are those
ethnographers who, in their zeal to win exemption from irrelevant and
time-consuming strictures, appear to be claiming that their research is not,
should not be, "intrusive" at all. Yet the inreractive, membership-oriented
researchers arc, by definition, intrusive-not in the negative sense of the
word, to be sure, bur they are still deeply involved in the live~; and activities
of the communities they study, a stance fraught \\lith all sorts of possibili-
ties for "harm." There are ethnographers with an "applied" orientation
(i.e., those who seek to use their research to effect social or institutional
change), those interested in using their research as parr of a project [or
social criticism, and those who advocate for "universalistic" values (e.g.,
women's rights, ecological justice) even when the local corr,-muniti~s they
happen to be studying act in ways inimical to those value~. All ot these
researchers may do "harm" in the strict sense of the term, bur it has not
been satisfactorily determined whether such "harm" is ll,xessarily and
inevitably to be avoided by the ethical researcher. It is difficult to prepare
an informed consent form when one cannot even begin to anticipate the
possibilities that might flow from personalized interaction. In principle, <It
least, it might" be possible to say that because research collaborators arc no
longer subjects, by definition they have as much power as do researchers in
shaping the research agenda; they do not need to be warned or protected.
Bur in reality, the researcher is still in a privileged position, at least insofar
as actually doing the research and disseminating its results are concerned.
The researcher probably does not want to retreat to the objective cold of
the classic observer role, but neither does he or she want to shirk the
responsibility for doing everything possible to avoid hurting or embarrass-
ing people who have been trusting partners in the research endeavor. A_s
Fluehr-Lobban (1994) concludes:
Openness and disclosure; reference in social studies to participants instead
of informams; models of collaborative research rhar incorporate informed
consent; all are componenrs of (ethnographic] research, whether academic
or applied, federally or privately funded, that is fully current with develop-
137
 Nil: I MUU.) Vr- \...VLLt:\...III'H..:;l f-\1'IU f-\1'1I-\l.. 11.1I~\.:1 CIVlr-Ir\I\...HI. IVIMII::r\IKl.-..l
ments taking place in the world we study and the professions that study it.
Informed consent may only be a convenient summary term for what has
t3ken place in biomedical and social science research, but when its spirit h
l'''ph;'\1<':ll~'''d l~ rc:,ulr:, In bcnct LC"Ciudu::r::; dill! lICttcr ITJCiUC!J. (p. 0)
All Interim Solution
This ethical dilemma would seem to be thc pivot on \vhich further
developments in observational research will rum, although there havc
been only provisional efforts to resolve it:. One example of such 3nattempr
is Angrosino's Opportllnity House (1998), his summ3tive report on the
study, discussed above, of ncarly two decades' duration of cOlll1l1unity-
based agencies serving adult clients wirh mental disabilities. Inforlllcd
consent was secured from those clients \\'ho were classified as "legally
competent" and from the legal guardians of those who were not. Never-
theless, Angrosino ncver felt confident thar the people with whom he
vvorked fu!!y understood the ramifiGltions of thcir consent, particu-
larly given that much of the ethnographic research was conducted in the
form of extended life history intervie\vs that often went off in directions
that could not have been predicted at the time the original study \vas pro-
posed and approved. Various interim publications about the project were
\vrirren in the standard authoritative voice of the obiecrive scientist, with
aggregared observations and limited excerpts from interview dma (attrib-
uted to pseudonymous informants) as illusrrations. When ir came time to
\vrite an overall analysis of the entire project, Angrosino found that such a
strategy seemed inadequate. It was necessary to draw the reader into borh
the experiences of people with mental disabilities (people \vho arc sO
much like us, and yet with a critical difference somewhat beyond our
capacities to imagine) and the experiences of a researcher trying to figure
out the p3tterns of the communities in which those people interact. But
doing so by means of an implicitly distancing language of expository scien-
tific \vriting and <l blurring of individual differences was not an attractive
optIon.
Angrosino therefore decided to try a form of "alternative ethnographic
\vriting" and to present his material in the form of fictionalized stories that
preserved the truth of individ ual experience without making explicit iden-
tificarions of particular people with specific situations. There are many
valid reasons for experimenting with nonexposirory presentations of
ethnographic material, but it may also be llseful to think of such alrern a-
138
"'-<''''''''''~ .......... .>'-, ''--'P ..... " . , ''-'''' "''-'''V'--' cv ..... vr'";:'''
tive genres as one response to the erhical quandary of obscf'lational re-
search in transirion.
Steps Toward an Ethic of
Proportionate Reason in Observational Research
Because observational research, as it has evolved in rccem rimes, is
essentially a matter of interpersonal interaction and not a matter of objec-
tive hypothesis testing, it \vould seem that a standard for the making of
ethical judgments appropriate to the analysis of "the morality of human
action" (Cula, 1989, p. 272) is in order. Human action must always be
interpretcd in situational context, and not in terms of objecti'1c "codes."
As Gula (1989) has pointed out, "No one enjoys an ahistori,:al vantage
point which will givc absolutc certitude on moral matrcrs" (p. 275). The
norion of "proportionate reason" is the key to such an inr,:;rpretation
(Cahill, 198]; Curran, 1979; Hoose, 1987; Walter, 198'1), Propor-
liollalisnl can sometimes refer to a strictly utilitarian cost~hencfit: analysis,
but it is more properly thought of in this contexr as that which gives an
Jetion its moral meaning. ln that: sense, "'proportionate' refers to rhe rela-
tion between the specific value at stake and the ... limitations, the harm,
or the inconvenience which will inevitably come about in trying to achieve
th;lt value" (Gula, 1989, p. 273). In other words, ir is certainly important
to "weigh rhe consequences" of an action, but consequences afe only one
part of the total meaning of the action. From this perspective, proportion-
ate reason defines \vhat a person is doing in all acrion (as an ethnographer
engaged in an observational context) and not something merely added to
the action alrcady defined (i.e., the old notion of the ethnographic ob-
server as extrinsic to the "action" he or she was recording).
There are three criteria that help us decide whether a proper relation-
ship exists between the specific value and the other elementt; of the aet
(McCormick, 1973; lvlcCormick & Ramsey, 1978). First, the means used
will IlOt cause more harm th(71l11ccessary to achieve the value. In traditional
moral terms, the ends cannot be said to justify the mans. If we take "rhe
value" to refer ro the production of some form of ethnogr3phy, then we
mUSt be careful ro assure that "the means used" (e.g., inserting oneself into
a social network, using photographs or other personal records) do not
cause disproportionate harm. \XTe might all agree that serving as cOllladrc
to an "informant's" child is sufficiently proportionate; we might well
argue about \vhether becoming the lover of an "informant" (particularly
139

if that sexual liaison is not intended to last beyond the time of the re-
search) does more harm than an ethnographic book, paper, or presenta-
tion might be \vorth. Volunreerins as a dQSSfOOm tutor in a program Set
Vlll b duului witll mtlHal fnarLlarJon WllOfll one is interested in ob~en'ing
and interviewing is probably sufficiently proportionate; becoming a bill-
paying benefactor to induce cooperation would, by contrast, be morally
questionable.
The second criterion is that 110 lcss harm/It! way exists at present to pro-
tect the valuc. Some might argue that observational research always and
inevitably compromises personal privacy, such that no form of research
can ethically protect that cherished value. Bur most researchers (and oth-
ers) would probably reject such an extreme view and take the position that
there is real value in disseminating the fruits of ethnographic research so as
to increase our knowledge and understanding of cultural diversity, or the
nature of coping strategies, or any number of currently salient social jus-
tice issues. Granced that all methods have the potential to harm, we must
be sure to choose those that do the least amount of harm, but that still
enable us to come up with the SOrt of product that will be effective in com-
municating the valuable message. The strategy of \vriting ethnographic
fiction, for example, is certainly not foolproof, as anyone with a knowl-
edge of the population with vvhich the ethnographer worked would be
able to identify the "characters." But there is far less chance that an outside
reader would be able to do so than would be the case with a reporr based
on "objective" materials that are on the public record.
The third criterion is that the /Ilcans 1/sed to achieve the value will not
wzclermine it. If one sets our, for example, to usc research in order to pro-
mote the dignity of people defined as mencaHy disabled, one must make
sure that the research techniques do not subject such people to ridicule.
Videotaping a group of people with mental retardation as they playa game
of softball might conceivably result in viewers' concluding that such peo-
ple are gallantly trying their best, but more likely it will result in confirm-
ing the popular stereotypes of such people as clumsy and inept, objects of
pity (at best) or of scorn (at worst) rather than dignified individuals. Vid~
eotaping as an ethnographic method is ethically neutral; its appropri:lte~
ness must be evaluated in this proportionate context.
1v1cCormick (1973) suggests three modes of knowing whether there is a
proporrionate reason to carry out a suggested action. First, we know that a
proper relation exists between a specific value and all other elements of ;:til
aCt through experience, which sometimes amounts to plain common sense.
140
'''",C' "'''''' ":1 ,-,,-,-,~, .~ .. ~ .
For example, although we may think that it is important to encourage indi-
vidual expression, we know from experience that doing so in the context
of acommunity (such as amonastery) in which the individual is, by tradi-
tion, subordinate to the group '.vill do real violence to the precepts by
\,>,hich the people we are intent on studying have historically formed them-
selves into a community. Experience might suggest that we rethink a deci-
sion to collect pei~sonallife histories of people in such groups in favor of
focusing on the collective reconstruction of remembered comnon activi-
ties or events.
Second, \\'e may know that a proper relationship exists through our
own intuition that some actions arc inherently disproportionate, even if
we do not have personal experience of their being so. Janss1;ns (1979,
p. 63) asserts that we can discover disproportion through ''"feellllgs of dis-
unity" within the self. For example, we should intuitively know that pub-
lishing personal material collected from people living in an oppressive,
totalitarian society might ultimately result in that material being used
against them, even in the absence of a direct or explicit threat. Our righ-
teous goal of exposing the tyrannical regime might well backfire on the
very people we arc trying to help. Our intuition might warn us that an
otherwise praiseworthy research proposal (e.g., to collect life histories or
genealogies, or to observe the daily activities at the local market) could
have harmful consequences if the product of the research were to fall into
the wrong hands. A perception of what could happen (the result of intu-
ition) is, of course, different from a perception of what will happen (the
result of experience), and we are clearly not well served by dreaming up
every conceivable disaster. It serves no purpose to allow our~elves to be
paralyzed beforehand by overactive guilty consciences. Bur there is cer-
tainly a commonsensical hierarchy of plausibility that obtai.ns in such
cases-some things that could happen are more likely to come about than
others.
Third, we knllw through trial a71d errOl: This is a mode of knowing that
would be completely impossible under current institutional ethical guide-
lines. But the fact is that we do not and cannot know all possible elements
in any given human social interaction, and the idea that we can predict-
and thereby forestall-all harm is na'lve in the extreme. An ethical research
design wm;ld omit (or seek to modify) that\vhich experience and intuition
tell us is most likely to do harm; we can then proceed, bur only on the
understanding that the plan \\'i11 be modified in the midst 01: the action
when it becomes clear what is feasible and desirable in the real-life
141

situation. For those uncomfortable with the indeterminacy of the term
trial alld error, W8lrer (1984) suggests "rational analysis and argumenr"
(p. 38). By gathering evidence and formulating logical arguments, we try
to give reasons to support our choices for certain actions over others. But
the plain fact is that this way of knowing does, indeed, involve the possibil-
ity of committing an "error," perhaps one that may have unexpected
harmful consequences. It is nonctheless disingenuous to hold that all pos-
sibility of harm can be anticipated and that any human action (including a
research project based on interpersonal interaction) can be made risk-free.
The moral advantage of the proportionate reasoning strategy is that it
encourages the researcher to 3dmit to an error once it has occurred, to cor-
rect it as far as possible, and to move on; the "objective" mode of research
ethics, by contrast, encourages researchers to believe that they have elimi-
nated all such problems, so that they arc disinclined to own up ro thosc
problems that (perhaps inevitably) crop up and hence are less capable of
repairing the damage. Those who work with people with developmentJl
disabilities are familiar with the expression "the dignity of risk"; it is used
to describe the "habilitation" of clients for full participation in the com-
munity. To deny the cI ients the possibi lity of making mistakes (by assu ming
that all risk can be eliminated beforehand and by failing to provide train-
ing in rcasonable problem-solving techniques) is to deny them one of the
fundamental characteristics of responsible adult living. One either lives in
a shelter, protected from risk by objectified codes," or one lives reallifc.
The ethical paradigm suggested here does nothing more than allow the
observational researcher the dignity of risk,
The logic of proportionate reason as a foundation for an ethical praC-
tice of social research might seem, at first glance, to slide into subjective
relativism. Indeed, the conscience of the individual researcher plays a very
large part in detcrmining the lllorality of a given interaction. Bur proper
proportionalism cannot be reduced to a proposition that an action can
mean anything an individual wants it to mean, or that ethics is simply a
lllatter of personal soul-searching. The strategy, rather, is based on a sense
of community-the individual making the ethical decision must ulti-
mately be guidcd by a kind of "communal disccrnmenr" (Gub, 1989,
p. 278). \Vhen we speak of "experience," for example, Vle refer not just to
personal experience, bur to the "wisdom of the past" as it is embodied in
the community's traditions. As such, it "demands broad consultation to
seek the experience and reflection of others in order to prevent the influ-
ence of self interest from biasing perception and judgment. Using propor-
142
l'\.ellllllKlIlY vu:>ervullurr: rrurn /VlernOQ to 1...0"ne.X"[
tionalism requires more moral consultation with th,~ community than
would ever be required if the morality of actions were based on only one
aspect ... apart from its relation [Q all the ... features of the action" (GuIa,
1989, p. 278). That being the case, the ideal IRE would not be content
with a utilitarian checklist of presumed consequences; it would constitute
a circle of "wise" peers with whom the researcher could discuss and work
out the somctimes conflicting demands of experience, intuition, and the
potential for rational analysis and argument. The essential problem vvith
current ethical codes, from the standpoint of qualitative observational
researchers, is that they set up an arbitrary-and quite unnecessary-
adversarial relationship between researchers and the [est of the schobrly
community. The framework of proportionate reason ::mplies that ethical
research is the product of shared discourse, not of a species of prosecuro-
rial inquisition.
oC> Prospects for Observational Research
As Adler and Adler (1994) remark in their chapter appearing in the first
edition of this Handbook, "Forecasting the wax and wc:ne of social science
research methods is ahvays uncertain" (p. 389), although they were able to
do so by extrapolating from existing trends. In that sam,~ cautious spirit, we
suggest that the future of observational research will most likely be in the
direction of what Barrett (1996) refers to as '"qualitative investigation with
a difference" (p. 237), Barrett refers to the '"demysti fication" of methodol-
ogy; whereas once ethnographers spoke in a vague way about "rapport" or
"empathy," they now publish and lecture extensively in the soul-baring
manner suggested by the preceding literature review. One important result
of that de mystification is that observation can no longer be said to be a key
to those grand, but somewhat opaque, units of analysis, "culture" and
"society." Abu-Lughod (1991) has, indeed, urged qualitative researchers to
use their techniques to undermine those concepts, which, she feels, have
become the contemporary equivalents of "race"-categories that separate
people, arrange them into hierarchies, and freeze the system so that institu-
tionalized inequality prevails. To speak in such terms ~eifjes the treatment
of difference and hierarchy as somehow "natural." Observational research,
by COntrast, has the pote~tial to tufn our attention to \vhat Abu-Lughocl
(1991) calls "the ethnography of the particular" (p_ L'4)_ Rather than
attempting to describe the composite culture of a group or analyze the full
143
 Mt I HUU::> U)- LULLt.L I WI\.;] AI'IU AI'IALI LII'I'J CIVlrlrl.l\...MI.. IV'M, L"'ML-l
range of institutions that supposedly constitute the society, the observa-
tional ethnographer will be able to provide a rounded aCCOunt of the lives
nf n'lrticll br nennle, the focm heing on ind ividuals and their ever-changing
relationships rather than on the supposedly homogeneous, coherent, pat-
terned, and (particularly in the case of traditional anthropologists) timeless
na[Ure of the supposed "group."
Abu-Lughod's position was foreshadowed by Grenz (1973) more than
two decades ago (an indication of how slowly it takes some preclictions to
come to pass). Geertz advocated setting aside the traditional social science
concern for "complexes of concrete behavior patteflls" in favor of a "con-
cern with the particular" based on the interpretation of "significant sym-
bols" (p. 44). The <lbovc literature review clearly indicates that this shifr is
already taking place, in the interest of feminists and postmodernists of all
persuasions in life history and "meaning." At present, the type of social sci-
ence represented by this <lpproach to the observation of the parricubr
coexists uneasily with more quantitative and positivistic schools of sociol-
ogy, Jnthropology, and social psychology. There IS, however, considerable
doubt as to how long that link can survive, given the vcry different aims
and approaches of the diverging branches of the once epistcmologically
unified social sciences. It seems not unlikely that observational techniques
will find a home in a redefined genre of cultural studies (composed of the
qualitative elements of the older disciplines), leaving their positivist col-
leagues to carryon in a redefined social science discipline.
Humanists though they may be, observational researchers are 3S depen-
dent on the evolution of technology as their quantitative colleagues.
Obserllation once implied a notebook and pencil, and perhaps a sketch
pad; first still and then motion picrures were bter added to the ethnogra-
pher's resources. Tape recorders have bcen supplemcnted (and, in some
cases, even supplanted) by video recorders. Note taking has been en~
hanced by the advent of the laptop computer, and computer programs for
the analysis of narrative data are being developed at a brisk pace. Obscrva-
tion-based ethnographers are, as a consequence, being pulled in twO direc~
tl0ns. On the one hand, they speak the theoretical language of "siw-
atedness," indeterminacy, and relativism; bur on the other hand, they rely
more and more on technology that suggests the capturing of reality" in
ways that could be said to transcend the individual researcher's relatively
limited capacity to interpret. The technology makes it possible for the eth-
nographer to record and analyze people and events with a degree of partic-
ularity that would have been impossible just a decade ago, but it also haS
144
the potential to privilege what is capwred on the record at the expense of
the lived experience as the ethnographer has personally known it. It would
be foolish to suggest that for rhe sake of consistency, observation-based
ethnogr:1phers should eschew further tr:1ffic with sophisticated r<:.:cordino-
and analyzing technology. But it would be equally foolish to assume th~~
the current very strong trend in the direction of individualized parti-
cularization can continue witham significant modification in the face of
technology that has the perceived power to objectify and turn into "data"
everything it encounters. Perhaps it will become necessary for us to turn
our observational powers on the very process of observation, to under-
stand ourselves not only as psychosoci<ll creatures (which is the current
tendency) but as users of technology. As Postman (1993) has pointed our,
technological change is never merely additive or subtractive, never simply
an aid to doing what has always been done. It is, rather, "ecologicll" in the
sense that <I cl13nge in one aspect of behavior has rami fications throul) hou t
the enrire system of \vhich that behavior is a part. "Surrounding ~very
technology are institutions \vhose organization ... reflects the wedd-vicw
promoted by that technology" (p. 18). Under those circumstances, per-
haps the most effective use we can make of observational techniques in the
near fllture will be to discern the ethos of the technology that we can no
longer afford to think of as a neutral adjunct to Our business-as-usual. It is a
technology that itself has the capacity to define our business. \Ve need to
turn our observational powers to vvhat happens not just when "we" en-
COUnter "them," but when "we" do so v,'ith a particular kind of totalizing
technology.
Nevertheless, it seems quite clear that the once-unquestioned hege-
mony of positivistic epistemology that encompassed even so fundamen-
tally humanistic <I research technique as observation has now been shaken
to its roots. One telling indication of the power of that transition-and a
challenging indication of things to come-is a recent comment by Stephen
Jay Gould (1998), the renowned paleontologist and historian 0:: science,
who has ruefully admitted:
No faith can be more misleading than an unqucsrio11cd personal C01l\'icrion
thar the ilpparcnt testimony of one's eyes must provide a purely objective
::lecount, scarcely requiring any validarion beyond the claim itself. Utterly
Lmbiased observation must rank as a primary myth and shibboleth of sci-
ence, for we em only see what fits into our ment<ll space, and all description
includes interpretation as well as sensory reponing. (p. 71)
145

'" Hotes
1. The critique of objenivitT W;lS certainly not invented by the posrmodernists, Indeed,
" \.'pod c"~,, can he- IY""I,, th:H ,he ~n"5cicnt, dio;l.:ip]iIlL":.pannillg s<.:lJol;J[
Gregory Bateson
(1972) was conrributing rigorous ,md systematic analyses of rhe place of observers in the
field long before it was considered possible to question the rationale of striving to eliminate
"observer bias." His "cybernetic" theory suggested that the observer is inevirablv tied to
what is observed. Nevertheless, the deb:ttc that currently commands rhe ;mc;uion of
ethnographic researchers was jump-started by crjtics of the cultural srudies persl!<1sion,
and the debate is most deeply informed by their vocabubry as well as by their specific
epistemological concerns.
2. Dialogue need not be taken lirerally to mean:l conversation between two ranies; ill
practice, it often consists of multiple-even conrr:Jdictory~voice5.
References
Abu-Lughod, L (1991). Writing against culture. In R. G. Fox (Ed.), ReCillJtlfrillg
L1l1throJiOlogy: \'i/orkillg ill the J7resCllt (pp. 137-162). Sama Fe, NIVl: School of
American Research Press.
Abu-Lughod, L (1997). The inrerpretarion of culrure(s) after television. Reflcc-
tiOllS, 59, 109-134.
Adler, P. A., & Adler, P. C1987). Membership roles in {reid rescarch. Newbury P<lrk,
CA: Sage.
Adler, EA., & Adler, P. (1994). Obsenrational techniques. In N. K. Denzin & Y. S.
Lincoln (Eels.), Handbook of qualitative research (pp. 377-392). Thousand
Oaks, CA: Sage.
Agar, .M., & Feldman, H. (1980). A four-city study of PCP users: JVIethodology and
findings. In C. Akins & G. Beschner (Eels.), Ethnography: A research tool {or
policymolwrs in the dnrg and alcohol fields (pp. 80-146). Rockville, 1vlD:
National Institute on Drug Abuse.
Altheide, D. L., & Johnson,]. M. (1994). Criteria for assessing interprerive validit)'
in qualitative research. In N. K. Denzin & Y S. lincoln (Eds.), Handbook. o{
qualitative research (pp. 485A99). Thollsand Oaks, CA: Sage.
Altark, K. (1995). Walking the fire line: The erotic dimension of the fieldwork
experience. In D. Kulick & M. \VtlJson (Ecls.), Taboo: Sex, identityamf erotic
subjectivity i11 anthropological fieldworh (pp. 107-139). London: Routledge.
Angrosino, 1\'1. \~ (1989). The twO lives of Rebecca Levenstolle: Symbolic intenlC-
rion in the generation of the life hisrory.jottmal ofAllthropological ResCilrch,
45, 315~326.
Angrosino, M. V. (1991). Conversations in a monastery. Oral History Relliew. 19,
55-73.
146
Angrosino, Iv!. V. (J 991). Ivletaphors of stigma: How deinstitutionaliud mentally
retarded adults see themselves. JOIlt/wl of Contemporary Ethllography, 21,
171-199.
Angrosino, [vt. V. (1994). On the bus wirh Vonnie Lee: Explorations in lifc history
and metaphor. ]oltnwl of COlllemporary Ethnography, 23, I4~28.
Angrosino, I'vl. V. (1995a). Eutaw Jack: The man born blind. In B. Grinclal &
F. Salamone (Eds.), Bridges to hult/anity: Narratives 011 anthropology alld
friel1dship (pp. 7~22). Prospect Heights, IL: \'{laveland.
Angrosino, 1v1. \~ ("1993 b). Thc TallllJa IUDS Netlvork: All oml history. Tampa: Uni-
versity of South Florida, Center for Applied Anthropology.
Angrosino,1V1. V. (1997a). Among the savage Anrhros: Reflections on the SAS oral
histOry project. SOlllhem Anthropologist, 24, 25-32.
Angrosino, IvL \Z (l997b). The ethnography of mental retardation: An :lpplicd per-
spective. JOHnIal of Conte1l1fJOrm)' Et/mograIJhy, 16, 98-109.
Angrosino, M. \Z (1998). Opportl/Jlit)' Hotlse: Ethnographic stories O(IJ/elltai retar-
datioll. Walnut Creek, CA: AltaMira.
Angrosino,lv1. V, & Zagnoli, L. J. (1992). Gender constructs and socia! identiry:
Implications for community-based care of retarded adults. In T. WhitcheJd
& B. Reid (Eds.), Gender COl1stmcts {/ltd social issues (pp. 40-(9). Urbana:
University of Illinois Press.
Ashton-Jones, E., & Thomas, D. K. (1995). Composition, collaboration, and
women's ways of knowing: A conversation with Mary Belenky. In G. A.
Olson & E. Hirsh (Eds.), \'{Iomell writing wltllre (pp. 81-104). Albany: State
University of New l'Drk Press.
Barnett, H. G. (1983). Learning about culture: Reconstruaion, participation, admin-
istration, 1934-1954. In G. \Y./. Stocking, Jr. (Ed.), Observers ohsci"ucd: Essays
on ethl10gmphic (ieldlVork (pp. 157-174). Madison: University of Wisconsin
Press.
Barren, S. R. (1996). AnthrofJology: A strident's guide to the01Y LInd method.
Toronro: University of Toronto Press.
Bateson, G. (1972), Steps to all ecolog)' of mind: Collccted essays ill anthropology,
IJsychiatr)', evoltltion, and epistemology. San Francisco: Chandl~r.
Behar, R. (1993). Ti"(lllslated woman: Crossing the border with EsperGllza's slol)'.
Boston: Beacon.
Behar, R. (1996). The vlIlncrable observer: Anthropology that breab }'Olll' heart.
Bosron: Beacon.
Bell, ]., & Jankowiak, \Y,Z R. (1992). The ctlmographer vs. the folk expert: Pithlils
of contract ethnography. Human Olgani:zntioll, 51, 412-417.
Baik, G. (1996). Understanding the gender system in tural Turkey: Fieldwork
dilemmas of conformity and intervention. In D. L. Wolf (Ed.), Feminist
dilemmas in fie/e/II/ork (pp. 56-71). Boulder, CO: \Vcstview.
147

Bcrn:lrcl, H. R. (l9HH). Research methods ill Cltltflral al1thropology. Newbury Park,
CA: SJgc.
Blackwood, E. (1995). FJ!Ilng in love with an-Other lesbian: Reflections on iden-
tit)" in f;dd~vod~. In D. Kulick & 1\'1. \Vilbun (Ed:;;.), Taboo; Sex, it/cIICit}' wid
erotic slIbjectiuity ill anthropological (ieldworf:. (pp. 5'1-75). London:
Roudedgc.
Bolmn, R. (1995). Tricks, friends, and lovers: Eroric cncounrers in the field. In
D. Kulick & i\'L Willson (Eds.), Taboo: Sex, idel1tity and erotic subjectivity ill
mtthropn/ogical (ieldtl/orh (pp. 140-167). London: Routledge.
Bolton, R. CJ 996). Coming home: The journey of a gay ethnographer in the years of
rhe pbguc. In E. Lewin & \Y!. L. Leap (Eds.), Ollt in the (ield: Rej7cctiOlzs o(
Icsbian alld gay dlzthropologists (pp. 147-170). Urbana: University of Illinois
Press.
Brewer, D. D. (1992). Hip hop graffiti writers' evaluations of strmegies to conrrol
illegal graffiti. HI/malt Otgmzizatioll, 51, 188-196.
Burkhart, G. (1996). Not given to personal disclosure. In E. Lewin & \'1. L Leap
(Eels.), 0111 ill the (ield: Reflections o( lesbiall ,mdgay anthropologists (pp. 31-
4S). Urbana: University of Illinois Press.
CJhill, L. S. (1981), Teleology, utilitarianism, Jnd Christian ethics. Thealogie/II
Studies, 41, 601-619.
Chavez, L. R., Flores, E. T., & Lopez-GJrza, Iv1. (1990). Here tOday, gone tomor-
row? Undocumenred settlers and immigration reform. Human Org[J/ziza-
tioll,49,193-205.
Clifford, J. (1983a). On ethnographic authoriry. Representatiolls, 1, 118-146.
Clifford, J. (l983b). I'mver and dialogue in ethnography: Marcel Griaule's initia-
tion. In G. \Y./. Srocking, Jr. (Ed.), Obseruers observed: Essays 011 etlmogmphic
(ieldzuorh (pp. 121-156). !vIadison: University of\\Tisconsin Press.
Cole, D. (1983). "The nille of a person lies in his f-lerzenbildllllg": Franz Boas'
Baffin IslJndlerrer-diary, 1883-1884, In G. W Stocking, Jr. (Ed.), Observers
obserued: Essays on ethnographic fieldwork (pp. 13-52). !vladison: University
of Wisconsin Press.
Curran, C. E. (1979). UtilitariJnisl11 and contemporary moral rheology: Situ,lting
the debates. In C. E. Curran & R. A. l'vlcCormick (Eels.), Readillgs i/l1/loml
theology (pr. 341-3(';2). Ramsey, NJ: Paulist Press.
Dembo, R., Hughes, I~, Jackson, L., & lVlieczkowski, T. (1993). Crack cocaine dealing
by adolescenrs in twO public housing projects: A pilot srudy. HZflllan 01galli-
zalioJl, 51, 89-96.
Denzin, N. K. (1997a). Interpretive ethnography: Ethnographic practices (ar the
21st century. Thousand OJks, CA: Sage.
Del12in, N. K. (1997b). Performance texts. In \\1: G. Tierney & Y. S. Lincoln (Eds.),
Represelltation {llld the tcxt: Re-framilfg the narrative uoice (pp. 179-218).
Albany: Star:e University of New York Press.
148
'~"'LO""""'~ ............ '- '-'.' ....., . ,' .....", , ~".'-'~ ...... '-'...... ,._..
Dubisch, J. (1995). Lovers in the field: Sex, dominance, and the female anthropolo
gist. In D. Kulick & Iv!. \"XTillson (Eds.), Taboo: Sex, identit)' and c'Dtic slIbjec-
tivity ill IIl1thropological fieldwork (pp. 29-50). London: Routledge.
Edelman, 1\1, (1~J96). Devil, not-qulte-white, rootless cosmopolitan: "li1lTis in L1tin
America, the Bronx, and the USSR. In C. Ellis & A. P. Bochner (Eds.), COIII-
jJosing ethnograph)': Altematiue (On/IS o( qualitativc writing (pp. 267-300).
Walnut Creek, CA: Alta1vlira.
Ellis, C. (1997). Evocative C1utoethnogr;lphy: \Vriting emotionally about our lives.
In \YJ. G. Tierney & Y. S. Lincoln (Eds.), Representation and the text: Re-
framing the lIarmtiue uoice (pp. 115-142). Albany: State University of New
York Press.
Ferguson,.J. (1996). The country and the city on the copperbelt. In A. Gupt:1 &
J. Ferguson (Eds.), Cllltllre, !-1Ower, IJlacc: E:-.:plomtiolls ilz critical antbropol-
ogy (pp, 137-154). Durham, NC: Duke University Press.
Fluehr-Lobban, C. (1994). Informed consent in anthropological research: We are
nO[ exempt. Hlllmm Organization, 53, 1-10.
Fox, K. V. (1996). Silent \-'oices: A subversive reading of child sexual <1buse. In
C. Ellis & A. I~ Bochner (Eds.), Composing ethnography: Altenzatiue (arms o(
qtUllitatiuc writillg (pp. 330-356). Walnut Creek, CA: AltaMir~;.
Gearing,]. (1995). Fear and loving in the \Vest Indies: Research from the heart (as
weil as the head). In D. Kulick & M. \\Tillson (Eds,), 7;1boo: Sex, identityalld
erotic subjectivity ill anthropological fieldworh (pp. 186-21:.l). London:
Routledge.
Geenz, C. (1973). The illterpretation of Cltltures: Selected essays. New'{ork: Basic
Books.
Giroux, H. A. (1995). Writing the space of the public intellectual. In G. A. Olson &
E. Hirsh (Eds.), Womcn writing ctllttlre (pp. 195-198). Albany: ~;tate Univer-
sity of New York Press.
Gold, R. L. (1958). Roles in sociological field observation, Social For,~es, 36, 217-
223.
Gold, R. L. (1997). The ethllogr<1phic method in sociology. Qzwlitatie.-'{! lllqlliry, 3,
388-402.
Goodman, L. (1996). Rites of passing. In E. Lewin & \\1. L Leap (Eels,), Out in the
field: Reflections of lesbian alld gay al1thropologists (pp. 49-57). Urbana:
University of Illinois Press.
Gould, S. .J. (19'98). The sharp-eyed lynx, outfoxed by nature (parr rNo). Natllral
History, 107,23-27,69-73.
Cuia, R. }.,/I. (1989). Reasol1 informed by faith. Nc\v York: Paulist Press.
Gupta, A., & Ferguson, J. (1996a). Beyond "culture": Space, identity, and the poli-
tics of difference. In A. Gupta &]. Fergllson (Eds.), Culture, power, place:
Explorations in critical anthropology (pp. 33-52). Durham, NC: Duke Uni-
versity Press.
149
Gupta, A., & Ferguson, J. CI996b). Culture, power, place: Ethnography at the end
of all era. In A. CUpt'c1 & J. Ferguson (Eels.), Culturc, powel; Illace: .-..:plo1"<1-
tirJ/ls ill critical i1IltlJ1"opology (pp. 1-31). Durham} NC: Duke University
Pro,,;,,. '
Gupra, A., & Ferguson,.J. (1996c). Discipline and practice: "The field" as sire
method, and location in anthropology. Tn A. Gupt'a & ./. Ferguson (E~ls.):
Allthropological lowtions: Bou11daries alld grounds o( a (ield science (pp. 1-
46). Berkeley; Universit'y of C<lliforniCl Press.
Hammond, J. L (1996). Popular education in t'he SalvCleioran guerrilla army.
HUll/ail Orgallizatirm, 5S, 43(}~445.
Hannerz, U. (1969). Soulside. New York; ColumbiCl Universiry Press.
Hareven, T. ("1996). The seClrch for genefiuional memory. In D. K. Dunawav &
\'{Z K. [hum (Eds.), Oral history: Illl illterdisciplillary anthology (2nd ed.,
pp. 241-256). Walnut Creek, CA; AltaIVlira.
I-linsley, C. (1983). Ethnographic charisma and scientific routine; Cushing and
Fcwkes in rhe American Southwest, 1879-1893. In G. \XZ Stocking,.It. '(r~d.),
Obserucrs observed: Essays 011 ethnographic jieldumrk (pp. 53-(9).lvlaelison:
University of \X1iscomin Press.
Hirsh, E., & Olson, G. A. (1995). Starting fromll1arginalized lives; A conversation
with Sandra I-lareling. In G. A. Olson Lx E. Hirsh (Eels.), \X/OII/CIl writing C111-
tllre (pr 3-44). Alb<1ny; State University of New York Press.
Hoose, B. (1987). ProportiollalislI1. \Vashington; Georgetown University Press.
Hsiung, p-c. (1996). Between bosses and workers: The dilemma of a kec~ observer
and a vocal feminist'. In D. L Wolf (Ed.), Feminist dilemmas ill fieldwork
(pp. 111-137). Boulder, CO: Westview.
Hubbard, A. S. (1997). Face-to-face at arm's length; Conflict norlllS amI CXtGl-
group relations in grassroots di;l!ogue groups. Htlman OrgaJlizlltiolt, 56,
165-174.
J;lIlssens, L (1979). Ontic evil and monl evil. In C. E. Curran & R. A. IV1cCormick
(Eds.), Readings illll1oralthcolo::''-')1 (pp. 49-71). Ramsey, NJ: Paulist Press.
Kennedy, E. L, & Davis, M. (1996). Constructing an ethnohisrory of rhe Buf-
falo lesbian comlllunity; Reflexivity, dialogue, and politics. !l~ E. Lewin &
\YJ. L. Leap (Eds.), alit ilT the (ield: Re(lectiolls of gay and lesbian an-
thropologists (pr. 171--199). Urbana; University of Illinois Press.
Killick, A. P. (1995). The penetrating intellect: On being whire, straight, and male
in Korea. In D. Kulick & IvI. \X1illson (Eds.), Taboo: Sex, identit)' aud
erotic subial il/ity ill anthropological fic!dworh (pp. 76-106). London: Rout-
ledge.
Koester, S. K. (1994). Copping, running, and paraphernalia bws: Comexnml vari-
ables and needle risk behavior among injection drug users in Denver. Human
Orgallization, 53, 187-195.
150
Kolker, A. (1996). Thrown overboard; The hum,m cosrs of health carc rationing. In
C. Ellis & 1\. P. Bochner (Eds.), COlllposingetfJ!lography: Altefl/6tive forllls of
qIlalitatiue lurilillg (pp. 132-159). \'X/alnut Creek, CA: Alralvlir;l.
Kuhlmann, A. (19';)2.). Collabordtive rt:~can::h among the l'\ickdpoo uibc of Okld'
homa. [-IlIIIUlll Or,---f!,ilIrization, 51, 274-183.
Kuklick, H. (1996). After Ishmael: The fieldwork tt::ldit'ion and its future. In
A. Cupra & J. Ferguson (Eds.), Anthropologicalloca/iolls: Bo.'mdaries alld
grouuds of a field science (pp. 47-(5). Berkcley: University of California
Press.
Lang, S. (1996). 'Ti'aveling wom;:l1l: Conducting a fieldwork projecr Oil gender vari
anct: and homosexuality among North American IndiallS. In E. Lewin &
\Y/ L Le,lp (Eds.), Ont ill the (ield: Reflcctiolls of lesbiml a/ld gay anthropolo
gists (pp. 8(;~ I J(J). Urbana; University of Illinois Press.
Larcom, J. (l983). Following Deacon: Thc problem of erhnographic reanalysis,
1916-J98"1. In G. \Y.!. Stocking, Jr. (Ed.), Observers obsenwl: Essays 011
etlmograplJic fieldworh (pp. 17S-195). Madison; Universiry of Wisconsin
Press.
Lather, I: (1997). Creating a multilayered text: Womcll, AIDS, and ansels. In \'\Z C.
Tierney & Y. S. Lincoln (Eds.), Representatioll and the tc.'\.-t: Rc-fmlllillg the
narrative voice (pp. 233-258). Albany: State University of NeVll'ork Press.
Lewin, E., & Leap, W L. (J 996). Introduction. In E. Lewin & \Y,Z L Leap (Eds.),
alit ill the field: Reflectiolls of lesbiall alld gay illtthropologists (pp. l-30).
UrllClna: University of lIIinois Press.
Liebow, E. (1967). 7izlly's comer: A study ofNegro street comer IIICII. Boston; Linle,
Brown.
lVladdox, R. (1996). Bombs, bikinis, and the popes of rock 'n' roll: Reflections on
resisrance, the pby of subordinations, and liberalism in Andalusin and ncade-
mia, 19S3~1995. In A. Gupta &]. Ferguson (Eds.), Ctlltttre, POWCI~ j)lace:
EX!Jlol'i.1tiolls ill critical allthmlJOlogy (pp. 277-290). Durham, NC: Duke
University Press.
I'VIahmood, C. K. (1996). Why Sikhs fight. In A. Wolfe & H. Yang (Ed~.),A1IthrofJo~
logical col/tributions to conflicl fesollllioll (pp. 730). Athens: UniversitT of
Georgia Press.
I\lalkki, L H. (1996a). National geographic: The rooting of peoples and the
territorialization of narional identity among scholars and refugees. In
A. Gupta &]. Ferguson (Eds.), ClIltIlTC, !)owcr, place: EXjJloraticJ1Is ill critical
aWhro[Jology (pp. 53-74). Durham, NC: Duke Universiry Pre~;s.
Malkki, L H. (1996b). News and culture: Transitory phenomena and -:he fieldwork
tradition. In A. Cupta & J. Ferguson (Eds.), Antlnopological locatio1ls:
Boundaries alld grollluls of a field sciellce (pp. 86-101). Berkeley: Universit'y
of California Press.
151

i'vL:!rcns, C. E. (1997). The uses of complicity in the changing mise-en-scene of
anthropological fieldwork. Reflectiolls, 59, 85-108.
Matsumoto, V. (1996). Reflections on oral history: Research in aJapanese Ameri-
Lim LUlIlHlUllity. ln D. L. \Volf (EeL), Fcmillf:j( dile.!lIflW5 ill fieldwork
(pp. 160-169). Boulder, co: \Xlestview.
IVlcCall, G. .I., Ngeva, ./., & Mbebe, i\II. (1997).1vlapping conflicr cultures: Interper-
sonal disputing in a Somh African black rownship. Human Or...r;;allizatioll, 56,
71-78.
i\'kCorrnick, R. A. (1973). Ambiguity alld moml choice. IVrilwaukee, WI: TvIar-
quette University Press.
IV1cCormick, R. A., & Ramsey, P. (1978). Doillg euil to achieve good. Chicago:
Loyola Universiry Press.
J\'1ienczakowslci, J. (I ~}96). An ethnographic .1(t: The construction of consensual
theater. In C. Ellis & A. P. Bochner (Eds.), Composillg etlmogmphy: Aftema-
I iuc forms o( qlliditatiue writing (pp. 244-266). \\laln ur Creek, CA: AltaMir;:1.
Ivliles, [,,'1. B., & Huberman, A. M. (1994). Qualitatiue data alia lysis: All expanded
sOllrcebooh (2nd ed.). Thousand Oaks, CA: Sage.
ivloreno, E. (1995). Rape in the field: RefieClions from a survivor. In D. Kulick &
l'\it Willson (Eds.), Taboo: Sex, idelltity alld erotic subjectivity i1l a/lthropo-
logical fieldwork (pp. 219-250). London: Routledge.
.Morton, H. (1995). JVly "chastity belt"; Avoiding seduction in Tonga. In D. Kulick
& ;V1. \"X1illson (Eels.), Ttlboo: Sex, identity alld erotic slfbjectiuity ill aJlthropo-
logical fieldluork (pp. 168-185). London: Roudedge.
tvlurray, S. E (1996). Male homosexuality in Guatemala: Possible insights and cer-
tain confusions from sleeping with the natives. In E. Lewin & \Y,f. L. Leap
(Eds.), Qut in the (ield: Reflectiolls of lesbian and gay anthropologists
(pp. 236-260). Urbana: University of JIIinois Press.
Newton, E. (1993). NI)' best informant's dress: The erotic equation in fieldwork.
ClIltllral Allthropolog)', S, 3-23.
Olson, G, A. (1995a). Resisting a discourse of mastery: A conversation with Jean-
Fram;ois Lyotard. In G. A. Olson & E. Hirsh (Eds.), \'lomen writing wlwrc
(pp. 169-194). Albany: State University of New York Press.
Olson, G. A. (1995b). Writing, liter<lcy and technology: Toward a cyborg writing. In
G. A. Olson & E. Hirsh (Eds.), \'\Iomelf writillg wltllre (pp_ 45~80). Albany:
State University of New York Press.
Olson, G. A., & Hirsh, E. (1995). Feminist praxis and rhe politics of literacy: A con-
versation with bell hooks. In G. A. Olson & E. Hirsh (Eds.), Women ruritiNg
wltul'e (pp. 105-140). Albany: State University of New York Press.
Pelto, Pol, & Pelto, G. H. (l978).Allthropological research: Thestmcturco(illqlliry
(2nd ed.)_ New York: Clmbridge University Press.
152
"ce'",'''''':! ,-,LJ,>C, "-,c,v,,., 'V'J' J""""'VV cv '-v',<.""/\<.
Peters,]. D. (1996). Seeing bifocally: rVIeelia, place, culrure. In A. Gupra &j. Fergu-
son (Eds.), Clllture, power; place: Explomtiolls ill critical anthropology
(pp, 75-92). Durham, NC: Duke University Press.
Polkinghorne, D. E. (1997). Reporting quahtative research as ptactice. In \\1. G.
Tierney & Y. S. Lincoln (Eels.), Represelltation alld the text: Re-fral1ling the
narratiull uoice (pp. 3-22). Albany: State University of New York Press.
Postman, N. (1993). Techno/wI)': The sltrrel1dero(cltltlne to technology. New 'fork:
Vintage.
Quinney, R. (l996). Once my father rraveled west to California. In C. Ellis & A. P.
Bochner (Eds.), Composing ethnography: Altematiue (arms o( qualitllliue
writing (pp. 357-382). \X1alnut Creek, CA: AltaMita.
Ronai, C. R. 0996). My mother is mentally rerareled. In C. Ellis & A. P. Boc,mer
(Eds.), Composing ethnography: Altematiuc forms of qualitativc writing
(pp. 109-131). W;:dllut Creek, CA: Alt<lMira.
lZossman, G. B., & Rallis, S. F. (1998). Leaming ill the field: All intl'Odtictiol1 to
qlwlittltil'e research. Thousand Oaks, CA: Sage.
Sewell, \Xl: H. (1997). Geertz and history: From synchrony to transforma::ion.
f{c/IeclioJls, 59, 35-55.
Silverman, D. (1993). Interpretillg qualitative data: Stmtegies fol' allalysing lalk,
text and il/temeliol/. London: Sage.
Sluka, .I .1\. (1990). Participant observation in violent social contexts. [-[lIman Orga-
nization, 49, 114-116.
Stepick. A., & Stepick, C. D. (J 990). People in the shadows: Survey research among
Haitialls in Miami. Hllman Organization, 49, 64-77.
Stocking, G. \Y,f., Jr. (1983a). The ethnographer's magic: Fieldwork in British
anthropology from Tylor to Malinowski. In G. \"'(/. Srocking, Jr- (Ed.),
Obseruel's observed: Essays all ethnographic (ieldwork (pp. 70-120)_ Madi-
son: University of Wisconsin Press.
Stocking, G. \"\7., Jr. (l983b). History of anthropology: Whence/whither. In G. W
Stocking, Jr. (Ed.), Observers oDsel"/lcd: Essays 011 ethnographic (ieldwork
(pp. 3~11). Madison: University of\Visconsin Press.
Tan<lka, G. (1997). Pico College. In \-0. G. Tierney &): S. Lincoln (Eds.),Repl"csell-
tatioll alld the tcxt: Re-fmmillg the narrative voice (pp. 259-304). Albny:
State University of New York Press.
Tierney, \'\( G. (1997). Lost in translation: Time and voice in qualitative research. In
W. G. Tierney & Y. S. Lincoln (Eds.), Representation and the text: Re-(rallling
the l1t.1rrative uoice (pp. 23-36). Albany: State University of New York Press.
Tillman-Healy, L M. (J 996). A secret life in a culture of thinness: RefleetioLs on
body, food, ,mel bulimia. In C. Ellis & A. P. Bochner (Eds.), Composing eth-
lIography: Alternalive fOl'1IIs o( qlwlitatfl!e u..,,.iting (pp. 76-108). Walnut
Creek, CA: AlraMira.
153
van Gelder, I~ .J., & Kapbn, C. D. Cl991). The finishing moment: Temporal ,md
sp;nial featUres of sexual imeranions between streetwalkers and Cill' clients.
HUllhl!1 Olg,miziniol1, 5 J, 253-163.

J. (1 QR4). ProPol'r;On::ltc rc"'"n anJ ;t.~. '[.''-'_'.<-' lev"l, of i"<-l',i,-y' Scr"cturinEr

UJ~llrlll',
the ongoing debate. LOIwaili Studies, 10,30-40.
\Xlalters, D. IVI. (1996). Cast among outcastes: Interpreting sexual orientarion,
racial, <lnd gender identity in the Yemen Arab Republic. In E. Lewin & W L
Leap (Eds.), Out ill the field: Re(lections of lesbialt alld gay '1IIthrotJO/ogists The Interpretation
(pr. 58-69). Urbana: University of lIlinois Press.
\'(lax, M. L., & Cassell, j. (1979). federal regulatious: Ethical issues I1nd social re-
search. Boulder, CO: Westview.
of DOCUlnents and
\Xlerner, 0., & Schocpfle, G.lVl. (1987). Systematic fieidlUorh: Vol. 1. FOlIl/datiolls
ol etlmogmphy and il/terlliewing. Newbury P:lrk, CA: Sage.
\\k:ston, K. (1996). Requiem for a street fighter. In E. Lewin & \X'. L Leap (Eds.),
lVlaterial Culture
Ollt ilt the (ield: Reflectiolls oflesbiall alldga), anthropologists (pp. 274-286).
Urbana: University of Illinois Press.
\'(.Ihytt, \'(,~ E (1955). Street comer society: The social stmctllre of ilIt Italitll/ slllm
Ian Hodder
(2nd ed.). Chicago: University of Chicago Press.
\Xlilliams, \X: L. (1996). Being gay and doing fieldwork. In E. Lewin & \\Z L. Leap
(Eds.), Gilt in the licld: Reflectiolts ollcsbian alldgay anthropologists (pp. 70~
85). Urbana: University of l!Iinois Press.
Willson, .LvI. (1995). Perspective and difference: Sexualization, the field, and the
ethnographer. fn D. Kulick & [vI. Wil!son (Eds.), Taboo: Sex, idcntity a/ld
erotic slIbjectivity in allfhropological fieldwork (pr. 251-275). London:
Routledge.
WlolcolT, H. F. (1995). The art ol fieldwork. Walnut Creek, C1\: Alta./'vIira.
\X'olf, D. L. (1996). Situating feminist dilemmas in fieldwork. In D. L. Wolf (Ed.), This chapter is concerned \vith the interpretation or mute
Feminist dilelllllUls ilt lieldwork (pr. 1-55). Boulder, CO: Westview. evidence-that is, with written texts and artifacts. Such cvidence,
Wolf, IVI. A. (1992). A thrice-told tale: Femiltism, postlllodemislll, altd ethltographic unlike the spoken word, cndures physically and thus can be s~paratcd
responsibility_ Swnford, CA: Stanford University Press. across space and time from its author, producer, or user. Ivlatend traces
Wolf, IVI. A. (1996). Afterword: IVlusings from an old gray wolf. In D. L. \"Xlolf (Ed.), thus often have to be interpreted without the benefit of indigenous cOl~
J--elllinist dilclItlItils inlicldwork (pp. 215-222). Boulder, CO: Westview. mentarv. There is often no possibility of interaction with spoken CIl1IC
"inside~" as opposed to etic "outsider" perspectives. Even when such
interaction is possible, actors often seem curiously inarticulate ab~ut the
reasons thev dress in particular ways, choose particular pottery deSigns, or
discard du~g in particular locations. IVIaterial traces and resid~les thus
pose special problems for qualitative research. The main disciplLles th.at
have tried to develop appropriate theory and method are history, art hls-
torv.' archaeolog)'
c , antllfopologv, ,-,.. sociology, cognitive psychology,
. . tech-
_ f
nolog)', and modern material culture studies, and it is from thIS range 0
disciplines that my account is drawn.

154 155
"1' Written Documents and Records
Lincoln and Guba (1985, p. 277) distinguish documents and records on the
basis of whether the text was prepared to attest to some formal transaction.
Thus records include marriage certificates, driving licenses, building con-
tracts, and banking statements. Documents, on the other hand, are pre-
pared for personal rather than official reasons and include diaries, memos,
letters, field notes, and so on. In fact, the two terms 3rc often used inter-
changeably, although the distinction is an important one and has some par-
allels with the distinction bet\veen writing and speech, to be discussed
below. Documents, closer to speech, require more contexrualized interpre-
tation. Records, 011 the other hand, may havc local uses that bccome very
distant from officially sanctioned meanings. Documents involve a personal
technology, <md records a full state technology of power. The distinction
is also relevant for qualitative research, in that researchers may often be
clhle to get access to documents, whereas access to records may be restricted
by laws regarding privacy, confidentiality, and anonymity.
Despite the utility of the distinction between documents and records,
my concern here is more the problems of inrerprewtion of written texts of
all kinds. Such texts arc of importance for qualitative research because, in
general terms, access call be easy and low cost, because the information
provided may differ from and may not be available in spoken form, and
because texts endure and thus give historical insight.
It has often been assumed, for example, in the archaeology of historical
periods, that written texts provide a "truer" indication of original mean-
ings than do other types of evidence (to be considered below). Indeed,
\Vestern social science has long privileged the spoken over the written and
the "vritten over the nonverbal (Derrida, 1978). Somehow it is assumed
that words get us closer to minds. But as Derrida has sho"m, meaning does
not reside in a text but in the writing and reading of it. As the text is reread
in different contexts it is given new meanings, often contradictory and
3hvays socially embedded. Thus there is no "original" or "true" meaning
of a ttxt outside specific historical contexts. Historical archaeologists
have come to accept that historical documents and records give nor a bet~
tel' but simply a different picture from that provided by artifacts and arch i-
tecrure. Texts can be used alongside other forms of evidence so that the
particular biases of each can be understood and compared.
Egu311y, different types of texts have to be understood in the contextS of
their conditions of production and reading. For example, the analyst will
156
be concerned \vith whether a text W,-15 \vritten as a result of firsth:md expe-
rience or from secondary sources, whether it was solicited or un:;;olicited,
edited or unedited, ;111onymoLls or signl:J, and so 011 (\Y'cbb, Campbell,
Schwartz, & Sechrest, 1966). As Ricoeur (1971) demonstrates, concrete
texts differ from the abstract structures of language in that they are written
to do something. They can be understood only as wh3t they are-a form of
artifact produced under certain material conditions (not everyone can
write, or write in a certain way, or have access to relevant technologies of
reproduction) embedded within social and ideological systems.
\vords are, of course, spoken to do things as \vell as to say things-they
have practical and social impact as well as communication function. Once
words are transformed into a written text, the gap bet\veen the "author"
and the "reader" vvidens and the possibility of multiple reinterpretations
increases. The text can "say" many different things in different conte.xts.
But also the written text is an anihcr., capable of transmission, manipula-
tion, and alteration, of being used and discarded, reused and re,:ycled-
"doing" cliffcrem things contextually through time. The \vriting down of
words often allows language and meanings to be comrolled more effec-
tively, and to be linked to strategies of centralization and codification. The
word, concretized or "made flesh" in the artifact, can transcend con-
text and gather through time extended symbolic connotations. Thc \vord
made enduring in artifacts has an important role to play in both secular
and religious processes of the legitimation of pmver. Yet there i:; often a
tension between the concrete nature of the written word, its enduring
nature, and the continuous potential for rereading mcanings in new con-
texts, undermining the authority of the worcI. Text and context are in a
continual state of tension, each defining and redefining the other, saying
and doing things differently through time.
In a related way, the wrirren texts of anthropologist's and archaeologists
are increasingly coming under scrutiny as employing rhetorical scrategies
in order to establish positions of authority (e.g., Tilley, 1989). Archaeolo-
gists are used to the idea that their sciemific activities leave traces and
transform the worlds they study. Excavations cannot be repeated: and the
residues of trenches, spoil tips, and old beer GillS remain as specific expres-
sions of a particular way of looking at the world. Thc past has been trans-
formed into a present product, including tbe field notes and site reports.
Ethnographic field notes (Sanjck, 1990) also transform the objeer of study
into a historically situated product, "capturing" the "other" within a
157
familiar routine. The field text has to be contexrualizecl within specific his-
torical momcms.
I shall in this chapter treat wl"irrcn texts as special cases of artifacts, sub-
ject to similar interpretive proo~dures. In both texts and artifacts the prob-
lem is one of situaring mare rial cuJrure within varying contexts while at the
same rime entering into '-1 dialectic relationship between those contexts
and the comext of the analyst. This hermeneutical exercise, in which the
lived experience surrounding the material culture is translated into a dif-
ferent context of interpretation, is common for both texts and other forms
of material culrure. I will note various differences between language and
material culture in \vhat foHows, bur the interpretive parallels have been
widely discllssed in the consideration of material culrure as text (e.g.,
Hodder, 1991; Moore, 19H6; Tilley, 1990).
~ Artifact Analysis and Its Importance
for the Interpretation of Sodal Experience
Ancienr and modern buildings ;md artifacts, the intended and unintended
residues of human activity, give alternative insights imo the ways in which
people perceived and fashioned their lives. Shortcuts aCross lawns indicate
preferred traffic patterns, foreign-language signs indicate the degree of
integration of a neighborhood, the number of cigarettes in an ashtray
betrays a nervous tension, and the amounr of paperwork in an "in" tray is J
measure of workload or of work eHiciency and priority (Lincoln & Guba,
1985, p. 280). Despite the inferential problems surrounding such evi-
dence, I wish to establish at the outset that material traces of behavior give
an important and different insight from that provided by any number of
questionnaires.
"\X!hat people say" is often very different from "what people do." This
point has perhaps been most successfully established over recent years by
research stemming from the \vorle of Bill Rathje (Rathje & lvlurphy, 1992;
Rathje & Thompson, 1981). In studies in Tucson, Arizona, and elsewhere,
Rathje and his colleagues collected domestic garbage bags and itemized
the coments. It became clear that, for example, people's estimates about
the amounts of garbage they produced were \vildly incorrect, rhJr dis-
carded beer cans indicated a higher level of alcohol consumption than waS
admitted to, and thar in rimes of meat shortage people threw away morc
meat than usual 2S a result of overhoarding. Thus a full sociological analy~
158
sis Cannot be restricted to interview clara. It musr also consider the material
tfaces.
In another series of studies, the decoration of roOlllS as well as pots and
other containers has been interpreted as a form of silent discourse con-
ducted by \vomen, whose voice has been silenced by dominant male inter-
esrs. Decoration may be used to mark our, silently, and ro draw <lncnrion
to, tacitly, areas of female control, such as female areas of houses and the
prepararion and provision of food in containers. The decoration may at
one level provide protection from female pollution, but at another level
it expresses female power (Brairlnvairc, 1982; Donley, 1982; Hodder,
1991).
The study of !TImerial culture is rhus of importance for qualitative
researchers who \vish to explore multiple and conflicting voices, differ-
ing and interacring interpretations. l\1any areas of experience are hidden
from language, particularly SLlborclinare experience. Ferguson (1991) has
shown how srudy or rhe material traces of food and potS can provide
insighr into how slaves on planr:uions in the American South made sense
of and reacted to their domination. The members of this normally silenced
grollp expressed their own perspective in the lllLlndJne activities of every-
day life.
Analysis of such traces is not a trivial pursuit, as the mundane and the
everyclay, because unimportant to dominant interests, may be of great
importance for the expression of alternative perspectives. The material
expression of power (parades, regalia, tOmbs, and arr) can be set against
the expression of reSlsrance. The importance of such analysis is increased
by the realization that material culture is not simply a passive by-product
of other areas of life. Rmher, materia! culture is active (Hodder, 1982). By
this I mean that artifacts arc produced so as to transform, materially,
socially, and ideologically. It is the exchange of artifacts themselves that
constructs social relationshipsj it is the style of spear thar creares a feeling
of common identitYj it is the badge of authority that itself confers author-
ity. Material culture is thus necessary for most social constructs. An adeM
quate study of social inrer3etion thus depends on the incorpor3tion of
!TIute mareri3l evidence.
Toward a Theory of Matorial <:;"1\""
Having established that the study of m3terial culture can be an impor-
tallt rool for sociological and amhropological analysis, \ve must attempr to
159
build a theory on which the interpretation of material culture can be based.
A difficulty here has been the diversiry of the category "material culture,"
ranging from written texts to material symbols surrounding death, drama,
and ritual to shopping behavior and to the construction of roads and air-
planes. As a result, theoretical directions have often taken rather differ-
ent paths, as one can see by comparing arrempts to build a comprehensive
theory for technological behavior (Lemonnier, 1986) and attempts to con-
sider material culrure as text (Tilley, 1990).
Ultimately, material culture always has to be interpreted in relation to a
situated context of production, use, discard, and reuse. In working toward
that contextual interprctation, it may be helpful to distinguish some gcn~
eral characteristics and analogies for the different types of material cul-
ture. In this attempt to build a g.eneral theory, recent research in a range of
disciplines has begun to separate two areas of material meaning.
Some material culmre is designed specifically to be communicative and
representational. The clearest example is a \vritten text, but this category
extends, for example, to the budge and uniform of certain professions, to
red and green stop and go traffic lights, to smoke signals, to the images of
Christ on the cross. Because this category includes y,rritten texts, it is to be
expected that meaning in this category might be organized in ways similar
to language. Thus, as with words in a language, the material symbols are,
outside a historical context, ofr:en arbitrary. For example, any design on a
flag could be used as long as it differs from the designs on other flags and is
recognizable with its own identity. Thus the system of meanings in the case
of flags is constructcd through similarities and differences in a semiotic
code. Iviiller (1982) has show::1 how dress is organized both syntagmat-
ieally and paradigmatically. The choice of hat tic shirt trousers shoes
and so on for a particular occasion is informed b; a syn'tax that ;Ilows ~
particular set of clothes to be pm together. On the other hand, the distinc~
tions among different types of hats (bowler, straw, cloth, baseball) or jack-
ets constitute paradigmatic choices.
The three broad areas of theory that have been applied to this first type
or material meaning derive from information technology, Marxism, and
structuralism. In the first, the aim has been to account for the v,/ays in
which material symboling can provide adaptive advantage to social
groups. Thus the development of complex symboling systems allows more
information to be processed more efficiently (e.g., \X1obst, 1977). This
type of approach is of limited value to qualitative research because it is not
concerned \vith the interpretation and experience of meaningful symbols.
160
In the second, rhe ideological componcnt of symbols is identified within
relations of power and domination (Leone, 1984; [viitler & Tilley, 1984)
and increasingly power and systems of value and prestige are seen as multi-
ple and dialectical ([vEUer, RO\vlands, & Tilley, 1939; Shanks & Tilley,
1987). The aim of structuralist analysis has been to ex,-lmine design (e.g.,
\Xfashburn, 1983) or spatial relationships (e.g., Gbssie, 1975; :l'vlcGhec,
1977) in terms of underlying codes, although here roo the tendency has
been to emphasize multiple meanings contested within active social con-
texts as the various directions of poststructuralisr thought luve becn
debated (Tilley, 1990).
In much of this \vork the meraphor of language has been applied to
material culture relatively unproblcmatically. The pur appears to "mean"
in the same VITaI' as the word pot. Recent work has begun to draw attention
to the limitations of this analogy between material culture and language,
as will become clear in my consideration of the second type of material
culture meaning. One can begin to explore the limit.Hions of the analogy
by considering that many examples of matcrial culture arc not produced
to "mean" at all. In other words, they arc not produced with symbolic
functions as primary. Thus the madeleine cookie discussed in Proust's A fa
recherche du temps perdu (SUHlIlI1 's \XZ7Y) was produced as an entici ng food
made in a shape representing a fluted scallop. Bur Proust describes its
meaning as quite different rrom this symbolic representation. Rather, the
meaning was the evocation of a whole series of childhood memories,
sounds, tastes, smells surrounding Proust's having tea with his mother in
winter.
Many if not most material symbols do nor work through rules of repre-
sentation, using a language-like syntax. Rather, they work through the
evocation of sets of practices within individual experience. It vl'Ould be rel-
atively difficult to construct a grammar or dictionary of material s}rmbols
except in the case of deliberately representational or symbolic items, such
as flags and road signs. This is because most material symbols cia not mean
in the same way as language. Rather, they come to have abstract meaning
through association and practice. Insofar as members of society experi-
ence common practices, material symbols can come to have common evo-
catiom nnd comn1On ll1cnnings. Thus, for example, the ways in \vhich cer-
tain types of food, drink, music, and sport are experienced are embedded
within social convention anel thus come to have common meaning. A gar~
lic crusher may nor be used overtly in Britain to represent or symbolize
161
class, bur through a complex set of pracrices surrounding food and its
preparation, the crLlsher has ':Ol11e to mean class through evocation.
Because objects endure, have their own traces, their o\vn grain, individ-
ual objects with unique evoG~tions can be recognized. The specific mem-
ory traces associated with any particular object (a particular garlic crusher)
will vary from individual to icdividual. The particularity of material expe-
rience and meaning derives not only from the diversity of human life bur
also from the identifiability of material objects. The identifiable particu-
larity of material experience always has the potential to work against and
transform societywide conventions through practice. Because of this dia-
lectic betvveen structure and practice, and because of the multiple local
meanings that can be given to things, it would be difficult to construct dic-
tionaries and grammars for most material culture meanings.
Another reason for the inability to produce dictionaries of material cul-
ture returns us to the difficulty with \vhich people give discursive accounts
of material symbolism. The meanings often remain tacit and implicit. A
smell or taste of a madeleine cookie may awake strong feelings, but it is
notoriously difficult to describe a tastc or a feel or to pin down the cmo-
tions evoked. \vc may know that in practice this or that item of clothing
"looks good," "works well," 'Jr "is stylish," but we would be at a loss to say
v;/hat it "means" because the item does not mean-rather, it is embedded
in a set of practices that include class, status, goals, aesthetics. We may not
know much about art, but we know what we like. On the basis of a set of
practical associations, we build up an implicit knowledge about the associ-
ations and evocations of particular artifacts or styles. This type of embed-
ded, practical experience seems to be different from the manipulation of
rules of representation and from conscious analytic thought. IVlaterial
symbolic meanings may get LIS close to lived experience, but they cannot
easily be articulated.
The importance of practice for the social and symbolic meanings of
artifacts has been emphasized in recent work on technology (Schlanger,
1990). Each technical operation is linked to others in operational chains
(Leroi-Gourhan, 1964) involving materials, energy, and gestures. For ex-
ample, some clays are better for throwing than others, so the type of clay
constrains whether a manufacturer can make thrown pots or hand-built
statuettes. Quality of clay is related to types of temper that should be used.
All such operational chains are nondeterministic, and some degree of
social choice is involved (Lemonnier, 1986; NEller, 1985). All operational
chains involve aspects of production, exchange, and consumption, and 50
162
are part of a nenvork of relations incorporating the marcrial, the eco-
nomic, the social, and the cOI1Ceprual.
The practical operational chains often have implications that extend
into not only social but also mora! realms. For example, LatoLIr (1988) dis-
cusses hydraulic door closers, devices that automatically close a door after
someone has opened it. The material door closer thus takes the place of, or
delegates, the role of a porter, someone who stands there and makes SLIre
that the door stays shut after people have gone through. But use of this
particular delegate has various implications, one of which is that very
young or infirm people have difficulty getting through the door. A social
distinction is unwittingly implied by this technology. In another example,
Latour discusses a key used by some inhabitants of Berlin. This double-
ended key forces the user to lock the door in order to get tbe key ant. The
key delegates for staff or signs that might order a person t:o "relock the
door behind you." Staff or signs would be unreliable-d1ey could be Out-
witted or ignored. The key enforces a morality. In the same way, "sleeping
policemen" (speed bumps) force the driver of a car to be moral and to slav\'
down in front of a school, but this morality is not socially encoded. That
would be too unreliable. The morality is embedded \vithin the practical
consequences of breaking up one's car by driving too fast over the bumps.
The social and moralmcanings of the claar closer, the Berlin key, or the
speed bump are thoroughly embedded in the implic3tions of material
practices.
I have suggested that in developing a theory of material culture, the first
task is to distinguish at least two different ways in which material culture
has abstract meaning beyond primary utilitarian concerns. The first is
through rules of representation. The second is through practice and evo-
cation-through the networking, interconnection, and mutual implica-
tion of material and nonmateriaL Whereas it may be the case that written
language is the prime example of the first category and tools the prime
example of the second, language also has [Q be v.,rorked out in practices
from which it derives much of its meaning. Equally, we have seen that
material items can be placed within language-like codes. Blit there is some
sUPPOrt from cognitive psychology for a general difference between the
two types of knowledge. For example, Bechtel (1990, p. 264) argues that
ruJe~bnsed IT1odel,; of cognition ~rc n::1turally good ~t quite diff",_"ent types
of activity trom connectionist models. Where the first is appropriate for
problem solving, the second is best at tasks such as pattern recognition and
1110tor COntrol. It seems likely, then, that the skills involved in material
163

practice and the social, symbolic, and moral me:.ll1ings dlat are implicated
in such practices might involve diffctent cognitive systems fro111 those
involved in rules and representations.
Bloch (1991) argues that practical knovv'ledge is fundamcntally differ-
ent from linguistic knowledge in the way it is organized in the mind. Prac-
tical knowledge is "chunked" into highly conrextualized information
about how to "get on" in specific domains of action.lVluch cultural knowl-
edge is nonlinear and purpose dedicated, formed through the practice of
closely related activities. I have argued here that even the practical world
in"\'olves social and symbolic meanings that arc not organized represen~
tational codes but that are chunked or contextually organized realms of
activity in which emotions, desires, morals, and social relations are in~
volved at the level of implicit taken-far-granted skill or know-how.
It should perhaps be emphasized th8t the two types of material
symbolism-the representational and the evocative or implicative-often
\vork in close relation to each other. Thus a set of pracrices Illay associate
men and women with different pans of houses or times of day, but in cer-
tain social contexts these asso<::iarions might be built upon to construct
symbolic rules of separation and exclusion and to build an abstract repre-
sentational scheme in which mythology and cosmology playa part (e,g.,
Yates, 1989). Such schemes 3!:;0 have ideological components that feed
back to constrain the practices. Thus practice, evocation, and representa-
tion interpenetrate and feed off each other in many if nor all areas of life.
Structure and practice are recursively related in the "strucruration" of
material life (Giddens, 1979; see also Bourdieu, 1977).
1} Material Meanings in Time
It appears that people both experience and "read" material culture mean-
ings. There is much more that could be said about how material culture
works in the social context. For instance, some examples work by direct
and explicit metaphor, \vhere similarities in form refer to historical ante-
cedents, \vhereas others work by being ambiguous and abstract, by using
spectacle or dramatic effect, by controlling the approach of the onlooker.
by controlling perspective. Although there is not space here to explore the
fuil range of material strategie:;, it is important to establish the temporal
dimension of lived experience.
164
lilt:: HICC'!"''''''''''''"''' "" .... '-''- ...",.. " ......." ... , ..... '-.,~, ...
~~,,~
As already noted, material culture is durable and can be given new
meanings as it is separated from its primary producer. This temporal varia-
tion in meaning is often related to changes in meaning across space and
culture. Archaeological or ethnographic artifacts are continually being
raken out of their contexts and reinterpreted within museums within dif-
ferent social and cultural contexts. The Elgin Marbles housed in the Brit-
ish .Nluseum take on new meanings that are in tLIrn reinterpreted antago-
nistically in some circles in Greece. American Indian human and artifact
remains may have a scientific meaning for archaeologists and biological
anthropologists, bur they have important emotive and identity meanings
for indigenous peoples.
l'vlaterial items arc continually being reinterpreted in new contexts.
Also, material culture can be added to or removed from, leaving the traces
of reuses and reinterpretations. In some cases, the sequence of use can give
insight into the thought processes of an individual, as when flint flakes that
have been struck off a core in early prehistory are refitted by archaeolo-
gists roday (e.g., Pelegrin, 1990) in order to rebuild the flint core and to
follow the decisions made by the original flint knapper in producing flakes
and tools. In other cases, longer frames of time are involved, as when a
monument such as Stonehenge is adapted, rebuilt, and reused for diver-
gem purposes over millennia lip to the present day (Chippindale, 1983).
In such an example, the narrative held within traces on the artifact has an
overall form that has been produced by multiple individuals and groups,
often unaware of earlier intentions and meanings. Few people today,
although lmowledgeable about Christmas practices, are aware of the his-
torical reasons behind the choice of Christmas tree, Santa Claus, reel coats,
and flying reindeer.
There are many trajectories that material items can take through shift-
ing meanings. For example, many are made initiaHy to refer to or evoke
metaphorically, whereas through time the original meaning becomes lost
or the item becomes a cliche, having lost its novelty. An artifact may start as
a focus but become simply a frame, part of an appropriate background. In
the skeuomorphic process a functional component becomes decorative, as
when a gas fire depicts burning wood or C031. In other cases the load of
meaning invested in an artifact increases through time) as in the case of a
tnlis n1 <lIl or holy relic. l\.'1atl:tla! ill:IllS are often ccntx"i in tt.c b<lcl"........ :->.rd-
lOoking invention of tradition, as when the Italian fascist movement ele-
vated the Roman symbol of authority-a bundle of rods-to provide
authority for a new form of centralized power.
165
 This brief discussion of the temporal dimension emphasizes the con~
tcxtuality of material culturE: meaning. As is clear from some of the exam-
ples given l changing meanin:~s through time are often involved in antago-
nistic relations between groups. Past and present meanings are continually
being contested and reinterpreted as part of social and political strategies.
Such conflict over material meanings is of particular interest to qualitative
research in that it expresses and focuses alternative views and inrerests.
The reburial of American Injian and Australian Aboriginal remains is an
iSSllC that has expressed, but perhaps also helped to construct, a new sense
of indigenous rights in Nortb America and Australia. As "ethnic cleansing"
reappcars in Europe, so too do attempts to reinterpret documents, monu-
ments, and artifacts in ethnic terms. But past artifacts can also be used to
help local communities in pwductive and practical ways. One example of
the active use of the past in the present is provided by the work of Erickson
(1988) in thc area around Lake Titicaca in Peru. Information from the
archaeological study of raised fields was used to reconstruct agricultural
systems on the ancient model, with the participation and to the benefit of
local tanners.
<P Method
The interpretation of mutc material evidence puts the inreractionist view
under pressure. How can an approach that gives considerable importance
to interaction with speaking wbjects (e.g., Denzin, 1989) deal with mate-
rial traces for which informants are long dead or about which informants
are not articulate?
I have already noted the iraportance of material evidence in providing
insight into other components of lived experience. The methodological
issues that are raised are nor, however, uniquc. In all types of interac-
tive research the analyst has to decide whether or not to take commentary
at face value and how to evaluate spoken or unspoken responses. HO\v
docs what is said fit into mote general understanding? Analysts of mate-
rial culture may nor have much spoken commentary to work with, but
they do have patterned evidence that has to be evaluated in relation co
the fnll range of -'lvrribhle inforlTl:1t;on. Tl,ey to", 1,av<;: ro [it di[{crent
aspects of the evidence into a hermeneutical whole (Hodder, 1992;
Shanks & Tilley, 1987). They ask, How does what is done fit into more
general understanding?
166
In general terms, the interpreter of material culture works between past
and present or betvveell diHerent examples of material culture, making
analogies between thcm. The material evidence always has the porential to
be patterned in unexpected ways. Thus it provides an "other" ::lgainst
which the analyst's own experience of the world has to be evaluated and
can be enlarged. Although the evidence cannot "speak back," it can con-
front the interpreter in ways that enforce self-reappraisal. At least when a
researcher is dealing with prehisroric remains, there arc no "member
checks" because the artifacts arc themselves mure. On the other hand,
material culture is the product of and is embedded in "internal" experi-
ence. Incleed, it could be argued that some material culture, precisely
because it is not overt, .'lei f-conscious speech, may give deeper insights into
the internal meanings according to which people lived their lives. I noted
above some examples of material culture being used to express covert
meanings. Thus the lack of spoken member checks is counteracted by the
checks provided by unspoken material patterning that remain able to COI1-
front and undermine interpretation.
An important initial assumption made by those interpreting materia!
culture is thar belief, idea, and intention are important to anion and prac-
tice (see above). It follows that the conceptual has some impact on the pat-
terning of material remains. The ideational component of materiel! pat-
terning is not opposed to bur is integrated with its material functioning. It
is possible therefore to infer barh utilitarian and conceptual meaning from
the patterning of material evidence.
The interpreter is faced with material data that are patterned along a
number of different dimensions simultaneousl y. 1\/lini mally, archaeologists
distinguish technology, function, and style, and they lise such attributes to
form typologies and to seek spatial and temporal patterning. In practice,
however, as the discussion above has shown, it has become increasingly
difficult to separate technology from style or to separate types from their
spatial and temporal contexts. In other words, the analytic or pattcrn-
recognition stage has itself been identified as interpretive.
Thus at all stages, from the identification of classes and attributes to the
understanding of high-level social proccsses, the interpreter has to deal
simultaneouslY with three areas of evaluation. First, the interpreter h8s to
;c!crItiCy die ~ontcxt5 wlLhili vdlidl Lld.uS::' IIdU ::.inlil~lr nlcanin::;. The
boundaries of the context arc never "given"; they h~l\'e to be interpreted.
Of coursc, physical traces and separations might assist the definition of
c011textual boundaries, such as the boundaries around a village or the
167
 lYle J nvu.) v r \...VI..L.I:I...., ,,,,1.;:1 """'1'-' """'IML.' "-II -,u
separation in time bet\vcen s,,;ts of events. Ritual contexts might be more
formalized than or may invert mundane contexts. But despite such clues
there is an infinity of possibl,~ contexts that might have been constructed
by indigenous actors. The notion of context is always relevant when dif-
ferent sets of data are being compared and where a primary question is
\vhether the different examples are comparable, whether the apparent
similarities arc real.
Second, in conjunction with and inseparable from tbe identification of
context is the recognition of similarities and differences. The interpreter
argues for a context by showing that things are done similarly, that people
respond similarly to similar Situations, within its boundaries. The assump~
tion is made that within the context similar events or things had similar
meaning. But this is true onl r' if the boundaries of the context have been
correctly identified. lVlany ;}nifacts initially identified as ritual or cuI tic
have bter been shown to come from entirely utilitarian contexts. Equally,
claimed cross-cultural similexities always have to be evaluated to see if
their contexts are comparable. Thus the interpretations of context and of
meaningful similarities and differences are mutually dependent.
The identification of contexts, similarities, and differences \vithin pat-
terned materials depends on the application of appropriate social and
material culture theories. The third evaluation that has to be made by
the interpreter is of thc relevance of general or specific historical theories
to the data at hand. Observation and interpretation are theory laden,
although theories can be changed in confrontation with material evidence
in a di~1.lectical fashion. Some of the appropriatc types of genera! theory for
material culture have been identified above. The more specific theories
include the intentions and social goals of participants, or the nature of rit~
ual or culric as opposed to sl:cular or utilitarian behavior.
In terms of the two types of material meaning identified earlier, rules of
representation are built up from patrerns of association and exclusion. For
example, if a pin type is exclusively associated with women in a wide vari-
ety of contexts, then it might: be interpreted as representing women in all
situations. The aspect of womanhood that is represented by this associa-
tion with pins is derived from other associations of the pins-perhaps with
(,:,rc;or" n",nlnc,,} "rf-;L,~,r,,, (f:orcH5cn, 19f17)_ TLe r<lUre r:cLl r <lcr.,.vo,["",d
the associations that can be followed by the interpreter, and the thicker the
description (Denzin, 1989) that can be produced, the subtler the interpre-
tations that can be made.
168
,-'v," ,,">-'-M'- ''','"'' '-,,,,.............,
For the other type of material meaning, grounded in practice, the ini-
tial task of the interpreter is to understand all the social and material
implications of particular practices. This is greatly enhanced by studies
of modern material culture, including ethnoarchaeology (Orme, 1981).
Experimental archaeologists (Coles, 1979) are now \'Vell experienced in
reconstructing past practices, fro111 storage of cereals in pits to flaking flim
tools. Such reconstructions, always unavoidably artificial to some degree,
allow some direct insight into another lived experience. On the basis of
such knowledge the implications of material practices, extending into the
social and the moral, em be theorized. But again it is detailed thick
description of associations and contexts that allO\vs the material practices
to be set within specific historical situations and the particular evocations
to be understood.
An example of the application of these methods is provided by
Ivferriman's (1987) inrerpretation of the imenrions behind the building of
a wall around the elite settlement of Hellneberg, Germany, in the sixth
century R.C. (an example similar to that provided by Collingwood, 1956).
In cultural terms, the Hallstart context in central Europe, including Ger~
many, can be separated from other cultural areas such as the Aegean at this
time. And yet the \valls are made of mud brick and they have bastions, both
of which have parallels only in the Aegean. In practice, mud brick would
nor have been an effective long-term form of defense in the German cli-
mate. Thus some purpose other than defense is supposed, The walls are
different from other contemporary walls in Germany and yet they are sim-
ilar to walls found in the Aegean context. Other similarities and differ-
ences that seem relevant are the examples of prestige exchange-valuable
objects such as wine flagons traded from the Aegean to Germany. This
trade seems relevant because of a theory that elites in central Europe based
their power on the control of prestige exchange with the Ivlediterranean.
It seems likely, in the context of such prestige exchange, that the walls built
in a 1V1editcrranean form were also designed to confer prestige on the elites
who organized their construction. In this example the intention of the wall
~Uilding is interpreted as being for prestige rather than for defense. The
lnterpretation is based on the simultaneous evaluation of similarities and
differences, Contex[ and theory. Bmh representational symbolism (confer-
rin.Q; prc-,',r:s",) and 1"',-"",';<.;al n'l.canin5" {dlC buikliuO vI ,vulb b)" cUtc;:, in i.l
nOl1~Mediterranean climate) are considered. For other examples of the
1hethod applied to modern material culture, see Hodder (1991) and
Moore (1986).
169

' Confirmation
How is it possible to confirm mch hyporheses about the meanings of mute
l1l;lterial and written culture? \X/hy arc some interpretations more plausible
than others? The ans\vers to such questions are unlikely to differ radically
from the procedures followed in orher areas of inrerpreration, and so I \vil!
discuss them relatively briefly here (sec Denzin, 1989; Lincoln & Guba,
1985). However, there are wme differences in confirming hypotheses
regarding material objects. Perhaps rhe major difficulty is that material cul-
ture, by its very nature, straddles rhe divide between a universal, natural
science approach to n1<1terials and a historical, inrerpretive approach to cul-
mte. There is thus a particularly marked lack of agreement in the scientific
community abour the appropriate basis for confirmation procedures. in
my view, an interpretive position Can and should accommodate scientific
information abour, [or example, natura! processes of transformation ;md
decay of artifacts. It is thus an interpretive posirion that I describe here.
The twill struts of confirmation arc coherence and correspondence.
Coherence is produced if the parts of the argument do nor contradict each
orher and if the conclusions follow from the premises. There is a partial
autonomy of different types of theory, from the observational to the
global, and a coherent interpretation is one in which these different levels
do nor produce contradictory results. The partial autonomy of different
types of theory is especially dear in relation to materia! culture. Because
material evidence endures, it can continually be reobserved, reanalyzed,
and reinterpreted. The observations made in earlier excavations arc con-
tinually being reconsidered within new inrerpretive frameworks. It is de3r
from these reconsiderations of earlier work that earlier observations can
be used to allow different intapretations-the different levels of theory
are parriaHy autonomous. The internal coherence between different levels
of theory is continually being renegotiared.
As well as internal coherence there is external coherence-the degree
to which the interpretation fits theories accepted in and outside the disci-
pline. Of course, the evaluatic,n of a coherent argument itself depends on
rhe application of theoretical criteria, and I have already noted the lack of
agreement in studies of material culture about foundational issues such a5
the importance of a T1atLlrJI science or hum:1nif:tic :1ppro~ch. gut wh.'ltever
their views on such issues, most of those working with material culture
seem [0 accept implicitly the importance of simplicity and elegance. An
argument in which toO much special pleading is required in order to claim
170
'""t" l/"t;;"'/J' t;;",.... WJI' VI ...... ~,'--w"'o;;;,O<.> """'W ,,'W,o;;;"WI ,-U"UIC
coherence is less likely to be adopted than is a simple or elegant theory.
The notion of coherence could also be extended to social and political
issues within and beyond disciplines, but I shall here treat these questions
separately.
The notion of correspondence between theory and data does not imply
absolute objectivity and independence, bur rather embeds the fit of data
and rheory within coherence. The data are made to cohere by being linked
\vithin theoretical argumcms. Similarly, the coherence of the arguments is
supported by the fit to data. On the other hand, data can confront theory,
as already noted. Correspondence with the data is thus an essential parr of
arguments of coherence. There are many aspects of correspondence argu-
ments that might be used. One is the exactness of fit, perhaps measured in
s13tistical terms, between theoretical expectation and data, and this is a
p~lrticularly important aspect of arguments exploiting the mure aspects of
material culture. Other arguments of correspondence include the number
of cases that are accounted for, their range in space and time, and the v;lfi-
ety of different classes of data that are explained. However, such numeri-
Gil indications of correspondence always have to be evaluated against con-
textual relevance and thiclc descri ption to determine whether the di fferent
examples of fit are relevant to each other. In ethnographic and historical
COntexts correspondence with indigenous accounts can be part of the
argument that supports contexrual relevance.
Other criteria that affeer the success of theories about material culture
meaning include fruitfulness-how many new directions, new lines of in-
quiry, new perspectives Jre opened up. Reproducibility concerns whether
other people, perhaps with different perspectives, come to similar results.
Perhaps different arguments, based on different starring points, produce
similar results. I have already noted that one of the advantages of materia!
evidence is that it can continually be returned to, unexcavated parts of
sites excavated and old trenches dug out and reexamined. Intersubjective
agreement is of considerable importance although of particular difficulty
in an area that so completely bridges the science-humanity divide. The
SUCcess of interpretations depends on peer review (either informal or for-
mally in journals) and on the number of people \\'ho believe, cite, and build
On them.
gUt rlluch depL'Iu..I.::; toO on the tru::;tworrhint:5s, prQh;::;::;iunal crcdcnlia15,
~lld status of the author and supporters of an interpretation. Issues here
1l1clude bow long the interpreter spent in the fidd and hm\' well she or he
knows the data: their biases, problems, and unusual examples. Has the
171
 ~ , , ,'-''-' ..... '-', .... '-' .............. , " ''-' ,..." ,..., ,...." ,,..., ... ' "-,, ,'-' L"" "~' ....,...,L ,"1'"'1' 1..'''MI..-' I l l e !!Hel/-lfetuliVII VI tJULUI','="';> VIIU f',UlC"UO '-""lW,-

author obtained appropriate degrees and been admitted into professional The ilUerpre(er Icarns from the experience of rnatcri~ll remains-the dena
societies? Is (he individual ;1n established and consistem writer, or has he and the interpreter bring each other into existence in dialecticll fash-
or she yec to prove her- or himself? Does the author keep changing her or ion. The interpretations can be confirmed or made more or less plausible
his mind? than mhers llsing a fairly standard range of inrernal and external (social)
In fact, the audience docs not respond directly to an interpretation but criteria.
to an interpretation vvrittcn or staged as an article or presentation. The
audience thus responds to and reinterprets a material artifact or event.
The persuasiveness or the agument is closely tied to the rhetoric vvithin <r- References
which it is couched (Gero, 1991; Hodder, 1989; Spector, 1991; Tilley,
1989). The rhetoric determines how the differem components ofthe dis- Bechtel, WI. (1990). Connectionism and rhe philosophy of rn"llld: An overview. In
cipline talk about and define problems and their solutions. W G. Lycan (Ed.), Mimi and cogllition: A redde]: Oxford: Basil Blackwell.
Bloch, ;\/1. (1991). Language, anthropology and cognitive science, Ali/II, 26, 183-
198.
Conclusion Bourdicu, E (1977). Olltlille of a theory of practice. Cambridge: C:llllbridge Univer-
sirv Press.
Draith\v~ite, ;\/1. (1982). Decoration as rlwal symbol. In I. Hodder (Ed.), Symholic
I\ilarerial culture, including written texts, poses a challenge for interprerive
and structural arclweology (pp. 80-88). Cambridge: Cambridge University
Jpproaches that often stress the importance of dialogue with and spoken
Press.
critical comment from participants. Material culture evidence, on the Chippindale, C. (1983). Stonehenge complete. London: Thames &. Hudson.
other hand, may have no living participants who can respond to its inter- Coles, j. M. (1979). Experill1elttal archaeology. London: Academic Press.
pretation. Even if such participants do exist, they may often be unable to be Collingwood, R. (1956). The idea of history. Oxford: Oxford University Press.
articulate about material cc:lture meanings. In any case, material culture Dcnzin, N. K. (1989). Interpretive illteractiollislll. Newbury Park, CA: Sage.
endures, and so the original makers and users may be able to give only a Derrida, j. (1978)_ m-itillg alld difference. London: Routledge & Kegan Paul.
p;lrtlal picture of the full hiscory of meanings given to an object as it is Llsed Donley, L. (1982). House power: SW;lhili space and symbolic mJrkers. [n L Hodder
and reinterpreted through time. (Eel.), Symbolic <lod structural archaeolo!::ry (pp. 63-73). Cambridge: Cambridge
The challenge posed by material culture is important for anthropologi- University Press.
cal and sociological analysis because material culture is often a medium in Erickson, C. L.' (1988). RJised field agriculture in the Lake Titicaca Basin: Putting
v\.'hich alternative and oftf:n muted voices can be expressed. But the ancient agriculture back to work. E.ypeditioll, 30(3), 8-16.
"reader" or material culture must recognize that only some aspects of Ferguson, L. (1991). Struggling with pots in colonial South Carolina. In R. Mc~
Cuire & R. Paynter (Eds.), The archaeology of II/equality (pp. 28~39).
material culture meaning afe language-like. The meaning of much ImHe~
Oxford: Basil Blackwell.
rial culture comes about through use, ancl material culture knowledge is
Cero, j. (1991). Who experienced whar in prehistory? A narrative expbnJtion
often highly chunked and conrexrualized. Technical operations implicate
from Queyash, Peru. In R. Preucd (Ed.), Processual dlld PostlJrocessllal
a wide ner-work of material, social, and symbolic resources and the archaeologies (pp. 126-189). Carbondale: Southern Illinois University.
abstract meanings that resu:"t are closely tied in with the material. Giddens, A. (1979). Central problems ill soci,;z1 theory. London: iVlacmilbn.
The methods of interpretation of material culture center on the simul- Classic, H. (1975). Folk hal/sing ill middle Virgilfia. Knoxville: University of
taneous hermeneutical procedures of context definirion, the consrruction Tenncssee Prcss.
of p3tterned ~imil:.1ritie_o:: Olnd differences, and tl~c usc of relevant" so",;",1 "nd I~loddt:rl 1. (1SJS1). Symbols ill action, Cambridge: Cambridge UnJVersHy I'ress.
material culture theory. The material culture may not be able directly to Hodder, L (1989). \'Vriting archaeology: Site repo[(s in conrex[. Al/tiquIty, 63, 268-
"speak back," but if appropriate procedures are followed, there is roo 01 274.
for the data and for different levels of (heory to confront interpretations. Hodder, 1. (1991). ReadiNg the past. Cambridge: Cambridge University Press.

172 173
 ,,, ... , , ''-''-'-.' '-" ..... VLL.L.'-'" '''-' M'''..' M"M'-' LII'\.:) l:J~\FIK''''''rI,- JVll-\ll:t\ll-\L) ",~ "'._'/-" ~.~ .. _,. _. --~_ ... _... - _.. - ... _.. _.. _. -_ .. _.. ~

Hodder, 1. (1992). Theory i71fd omclia in archaeology. London: Routledge. Sorensen, M.-L. (1987). I'vlateri31 order and cultural classificarion. In 1. Hodder
Larour, B. (1988). IVlixing humans and tlonhumans wgether: The sociology of a (Ed.), The archaeology of coJltcxtllalmeallil1gs (pp. 90-101). Cambridge: Cam-
door closer. .','odal Problems, 35, 198~310. bridge University Press.
Lemonnier, p. (19S6). The study of marerial culture today: Towards an anthropol- Specror, J. (1991). What this awl means: TOW;1fd 3 feminist 3rchaeology. Tn]. l'v'l.
ogy of technical systems. jOllmal of Anthmpological Archaeology, 5, 147- Cera & Ivl. \'\Z Conkey (Eds.), Ellgendering archaeology (pp. 388-406).
186. Oxford: Basil Blackwell.
Leone,lvl, (1984). Imerprcring ideology in historical archaeology. In D. Miller & Tilley, C. (1989). Discourse and power: The genre of the Cambridge inaugural. In
C. Tilley (Eds.), Ideology, power alld prehistory (pp. 25-36). Cambridge: D. f\![iller, M.. RO\vlands, & C. Tilley (Eds.), Domillation alld resistance
Cambridge Universiry Press. (pp. 41~(2). London: Unwin Hyman.
Leroi~Gourhan, A. (1964). Le g':5te el la parole. Paris: Iviichel. Tilley, C. (Ed.), (1990). Reading IllaleriLl! ellltllrc. Oxford: Basil Blackwell.
\\lashburn, D. (1983). Stmctmc mfd coglliti(m ill art. Cambridge: Cambridge Uni-
Lincoln, '{ 5., & Gub3, E. G. (1935). Natllralislic illquiry. Beverly Hills, CA: Sage.
versity Press.
IvleChee, R.. (1977). Ivary for rile sea woman. Cmwdiall jot/mal ofArchaeology, I,
Webb, E. J., Campbell, D. T., Schwartz, R. c., & Sechrest, L. (1966). Ullobtmsive
141-159.
measures: Nonreactiue research ill the social sciences. Chicago: Universiry of
J\'1erriman, N. (1987). Value and motivation in prehistory: The evidence for "Celtic Chicago Press.
spirit." In I. Hodder (Ed.), Jhe archaeology of contextual meanings (pp. 111~ Wohst, 1\.'1. (1977). S[ylistic behavior ;}Ild information exchange. Universil')' ol
116). London: Unwin i-lyman.
Michigall MuseulII o(A.ntlnopo/ogy, AllllmJ/Jological Pape/; 61, 317-342.
l\1iJler, D.. (1982). Artifacts as producrs of human c3tegorisation processes. In Yates, T. (1989). Habitus and social sp:lCe. In l. Hodder (Ed.), The meanings of
I. Hodder (Ed.), S'ymbolic alld stmctl/ml archaeology (pp. 89-98). Cam- things (pp. 248-2(2). London: Unwin Hyman.
bridge: Cambridge University Press.
IvIiller, D. (1985).Artifactsascategories. Cambridge: Cambridge University Press.
.rAiller, D., Ro\vlands, lVI., & Till,:y, C. (1989). Dominatiolt alld resistance.. London: Suggested Further Readings
Unwin Hymll1.
lvIiller, D., & Tilley, C. (1984). ideology, power and prehistory. Cambridge: Cam- Hodder, 1. (1999). 11]c arc!weologiol! process. Oxford: Basil BbckwdL
bridge University Press. Hodder, 1., & Shanks, [vI. (19~)5). Interpreting archaeology. l.ondon: Routledge.
Moore, H. (1986). Space, text altdgcndel: Cambridge: Cambridge University Press. Miller, D. (Ed.). (1997). M'l/eria! wllures. London: University College Press.
Orme, B. (1981). A.ntlJropalogy (ar l.1J'chaeolagists. London: Duckworth. Tilley, C. (1994). The phenoll1e11ology ol landscape. London; Berg.
PeJegrin,]. (1990). Prehistoric lirhic rechnology. ArclJl1eological Review From Cam- Tilley, C. (1999). Mefa!Jhor alld 1/U1!erii.11 wlture. Oxford: Basil Blackwell.
bridge, 9, IJ 6-125.
Rathje, W, & Ivlurphy, C. (1992:,. Rubbish! The archaeology ofgarbage. New lor!e:
!-I3rperCollins.
Rathje, \"XI:, & Thompson, B. (19:~1). The Milwaukee Garbage Project. \\7ashington,
DC: American Paper Insri ,ute, Solid Waste Council of the Paper Industry.
Ricoeur, P. (1971). The model of rhe text: Meaningful action considered as rex[.
Social Research, 38, 529-562.
Sanjek, R. (EeL). (l990). Fieldl/Cltes: The makings of anthropology. Albany: State
Universiry of New York Press.
Schlanger, N. (1990). Techniqlle~; as human action; Two perspecrives. Archaeologi-
cal Reviezu From Cambridge, 9, 18-26.
Shanks,M., &:.: Tilley, C. (1987). RecoJlstntcting archaeology. C:Jmbridge: Cam-
bridge University Press.

174 175

ReiJmagining
------- -=----=--------
Visual Methods
Galileo to Neuroma71.cer
Douglos Harper
This chapter is based 011111)' sense that the chapter I prepared for the
first edition of this Halldbook might help a researcher understand
the relationship between visual and other methods, bm it would not be
very useful to a student or researcher trying to LIse visual methods, or to a
researcher looking for a framework in which to understand visual meth-
ods. I In this chapter I rake::1 different approach. First, I suggest a context in
which to sec photography and social research, this being the history of
recorded perception. NeXT, I present visual sociology as fieldwork pho-
tography guided by several research traditions. Third, I describe the social
influences around \vhich "'picture making" has wizen place, noting how
the social power involved il1111aking images redefines institlltions, groupS,
and individuals. Finally, 1 suggest that visual sociology is, above all, a pro-
cess of seeing guided by theory. Because visual sociology is '-1. grab b'"lo or r2-
search approaches and perspectives on understanding images in society, I
aim to make several attenuated arguments and to we,-lve them inw ;:l
\vhole.
176
f\~IIIIUIdIlIJlIY VI~UUI IV1~1IIUU". I.;lUIII~U tU 1'i~UIUIIIUIIL~1
" Visual Methods and the History of Recorded Perception
Logical positivism assumed its modern form in the 16th cemury, \vhen
Roger Bacon suggested that observable d::Ha are the basis of knowledge.
Bacon also argued that knmvledge ought to be practically applied to solv-
ing social problems, anticipating Comte's mandate for sociology in the
early 19th century. I nare that Bacon's insights traced to the rationalist phi-
losophers of 6th-century B.C. Greece to remind us that the rootS of this per-
spective are nearly as old JS recorded thought.
Bacon and logical positiv ism were contemporary to the invention of the
telescope and the microscope, t()Dls that showed that the world observed
with human eyes was nor complete or even correct. In the 16th century
this was such a challenging idea that philosophers at the University of
Padua, where Galileo was <1 professor, refused to look through the tele-
scope. Galileo, censured and threatened with death, recanted the discov-
eries his telescope had revealed. But, of course, the scientific revolution
was not stopped by the violence of the Church, and the revolution inspired
by Galileo and Bacon flourished in the 17th century and became the basis
of the modern scientific world.
In the 19th century, the camera became p:ut of the revolurion in seeing
and understanding that W35 the scientific revolurion, and several subse-
quent instruments and tools, which I \vill describe below, further rede-
fined the relationship bet\veen seeing and knowing.
There arc two important implications to the changes brought by Gali-
leo's telescope. The firSt is that the ne\v, instrument-based perception was
treared as more real than was the world based on faith and belief. It became
understood that to see through an instrument (such as the telescope or
microscope, and, eventually, the camera) was to see a morc profound real-
ity than could be observed by the eye. The second implication of these
changes is th3t the legitimacy of science carne ro be based in large part on
its claim to describe a world in visual terms. In this way the eye became the
privileged sense of science, and of modernism.
The hegemony of science and the image did not last, due a succession of
tools and instruments that redefined both image making and the social
role of images. The first and perhaps most important of these is the camera
itself. Tho,;: (:dJl1Cl-i:i5 u5cd today iU"t: tt;t;tlltially :Jimi1~U" ~Q c~uncra:J I1c515110;.:.;.\
150 years ago, although they have the ability to see farther, closer, fastet,
and in lower light. In surprising ways, the public use of cameras has
remained constant. The middle-class public enthusiastically adopted
177
 I"\~ I nVl,...IJ VI" '-..V~~t::'-..I ""u MI~l,...I A"'A~r L l l " U CJVlrll\l\...M~ IV\MI t:rl.IAL')
phorography in the 19th cemury and continues to do so. Different social
classes and groups, of course, use photography differently (Bourdieu,
1990), bur the 19th-cenwqr family snapshoorer would have no trouble
[merpreting the modern-day family album.
In the mcamime, marion pictures, video, the World Wide \\leb, and vir-
tual reality have offered new connections between human existence and
visual perception. These new instruments influence the meanings of
images, their relationship to spoken words and sounds, and the emergence
and development of visual wciology.
By the end of the 19th century, it became possible ro link individual
frames of film together, giving a succession of images;] verisimilitude the
still image could never achieve. Suddenly it seemed that the process of life
itself could be recorded. But at fitst this record was limited to images alone
(synchronized sound could not be recorded ,\'ith the images), and because
cameras were bulky and heavy, filming the flow of natural life was nearly
impossible. The earliest documemary films) such as Nanook of the Nort;}
(filmed during the 1920s), were thus extraordinary technical achieve-
ments. Because there was no sound track (except for a piano score for a
live musician), the film comained frames of typed words between se-
quences of images. These h2-d to be brief statements, quickly readable by
an audience impatient to rejoin the visual narrative of the film. These texts
"vere what we nO\\, call "captions)" and they performed the same function,
which was to summarize the visual text.
By the 1930s it became possible to add sound to film, but it was nearly
impossible to do so outsid~ of a studio. Documentaries such as Pare
~orentz's The ~{iver and The PIOlU That Bro/~e the Plains (see Snyder, 1968)
ll1troduced vOice-over commentary. This commentary functioned as did
the captions of the silent film., in that it defined and provided closure for
the images. In this sense it ,:an be thought of as the confident voice of
(modernist) science.
The era of television brought documentaries in this form to the living
rooms of America in the weekly Omnibus series shown during the 19505.
Again, the voice-over-image narrator supplied the logic that tied the film
statements together. As a result, these films are linear and narrative in
organization, and unambiguous in structure and message.
Portable c~mer:1>: thnt could siruultoneollsly record sound were JcYcl~
oped in the early 1960s. These capabilities potentially eliminated the
hegemony of image and scieuific authority. Film could now explain itself;
it leaped closer to reality, and the results resembled the messy flow of life.
178
Interestingly enough, the movement \vas called cinema vetite, or ttLlth-
cinema. Jean Rouch and Edgar l\llorin's (a French sociologist) Chronique
d'lIn Ete (Chronicle of d Summer; 1960), one of the first cinema verite
films, took cameras to the streets of Paris to measure the public reaction to
the Algerian \\'ar and to comment on several cultural themes. In a de par-
wre from previous film structute, in Chrolliqlle d'Ull Ete the subjects of the
film are shown watching their earlier filmed interviews, adding a layer of
intetpretation to the already ambiguous film statement.
In the United States, the early cinema verite movement (see i\1amber,
1974) coincided \vith the emcrgence of experimental ethnography,
although only a handful of sociological filmmakers and sociologists and
anthropologists made the connection. Several films of this cr3 offered
bridges between ethnography and experimental filmmaking. Th~se
included the .Maysles brothers' documentary The Salesma/l (1969), which
follows the day~to-day routines of a Bible salesman; Robert Drew's Pri-
mary (1960), which shows the mundane events of a political primary; and
Richard Leacock's Happy lvlother's Day (1963), which explores how a
small town experienced and marketed the bitth of quintuplets. Frederick
\X1iseman's Titicut Follies (1967), which depicts the day-to-day routines of
a prison for the criminally insane, VI/as the first in a series of more than 20
cinema verite studies of mostly American institutions that offered an
extraordinary rcsource for experimental sociology and anthropology.
Cinema verite showed sociological dimensions to everyday life, and the
movement embraced a technology that redefined what, in fact, could be
filmed. Sadly for the social sciences, the potemial of this technology has
been little used or even recognized.
Although cinem3 verite redefined the relationship between film and
reality, its adoption, for example, as a common research technique was
limited by its expense. By the early 1980s a typical budget for a documen-
tary film, not including salaries) was more than $1,000 per minute. The
cameras could record onlv a few minutes of simultaneous sound and
images, 3nd generally there' needed to be a sound recorder working in tan-
dem with the filmmaker, which added to the intrusion of the recorders of
realitv 011 reality itself.
Advances in ~idco technology) dating to the early 1980s, made it possi-
6lt [oJ: videogwphers to record hours of synchronized visl1:l1 and Judio
information at a fraction of the cost of film. Video camer;1S ll:1ve become
smaller and yet produce higher-quality images and sound (most recently in
the Hi S format), and they have also become simpler to operate J.nd cheap
179
 IVIC I nvu,J vr l....VLLCl....III'l\.] f-\I'IU f-\I'lf-\LT LlI'l\.] CIVIOII"I\..-f-\L N\!-\.I CKIf-\L::J
enough to be a routine home expense. Video wrested image making from
the monopoly of weU-heeled expcrts. It is reasonable to say that with this
change, visual representations of the wodd lost their close connection
with the authority of science. Yet the implications of this change are
ambiguous. For the most part, tbe video revolution has not worked h;:md
in hand with experimental ethnography to redefine the social science.
Indeed, the social effects have been trivialized inco "funniest home videos"
television programs (now \\'orldwidc), in which the extraordinary power
of the camera is llsed to present the lowest Common denominator of public
life.
Contemporary electronic technologies continue to redefine the image
and its social meaning. Nov\' that images can be created and/or changed
digira\ly, the connection between image and "truth" has been forever sev-
ered. \X7hat was once disturbing (such as the addirion of Hitler-like visual
cues to a portrait of Sadd3m Hussein on the cover of a national news-
magazine) is now commonplace. Electronically altered images have
become parr of our enrertainmcnr (in Zelig, Woody Allen appears along-
side Lenin; in Forres! GllIl1P, Tom Hanks is shown in conversation with
John Kennedy); advertising juxtaposes the images of people \\'ho have
nothing to do with each other, 'Nho in fact do not even live in the same era,
presenting them as enthusiastk product spokespersons.
Bur the electronic revolution is a great deal more than the ability to alter
photographic or video images. Because of what is now called hytJcrlogic,
there is an alternative to linear or narrative form of visual presentation.
I-Iyperlogic, \vhich began in the HyperC<lrd program (packaged \ovith early
lvlacinrosh computers) is the basis of the \\1orld \\1ide \X'eb. As is common
knowledge today, the Web is organized so that viewers can create the it
ovm paths through text, images, and even film or video clips. The most
sllccessful current example is Peter Bielb, Napoleon Chagnon, and Gary
Seaman's (1997) interacrive CD-RO"M of the anthropological film The Ax
Fight, by Timothy Asch, and additional hyperlinked materials. The inter-
active CD allows a vie\ver to vIew the actual film in any of several possible
ways (in real time, backward as well as forward, frame by frame, in .slow
motion, or keyed to significant moments as identified by the anthropolo-
gists). The viewer can also link (0 scene-by-sccnc: descriptions of the film,
or can link to any individual shown in the film to get information on that
person's age, sex, spouses, children, birthplace, lineage, residence, year of
death, place in the kin systems (presented in kin chans), ancl other anthro-
pological details. The CD cClmains complete footage and edited versions
180
r<..t!I!IIUYI/II/IIj VI~UUI IVIt::'lI'UU~. \..Ju",cv tU J'CtIlV"IU"Lt:"
of the film, hundreds of photographs, and several full-length essays. The
viewer can access any part of the film and digress to any of several
analyses.
The CD is a curious step in the evolution I have described. On one
hand, it represents the scientific expert whose claims are reinforced visu-
ally; on the other, it de constructs the authority of the scientist and makes
the viewer the author of the viewing and learning experience.
The last step in the process that began with Galileo's telescope is the
jumbling up of our senses through electronic manipulation, called virtual
reality. Giovanni Artieri (1996) suggests that the "technological image
is the real marker of the postmodern, able to trace out subjects' experi~
ence as border lines between natural and artificial" (p. 56). By wearing
machines that transform nerve stimuli into visual ~1tlcl tactile sensations,
users alter the boundaries separating the body and society, and experience
itself loses its link to authenticity. Now it is no longer what we see (or hear
and feel) that is real, as in the case of J science based on unchallenged
claims to represent the world. Rather, we choose to immerse ourselves in a
fictional perceptional reality-that is, a perceptual world that is the result
of our imagination and a machine. In the next step, still fictional (best rep-
resented in cyberpunk novels such as \\1illi:1m Gibson's NellJ'OIJU1IICel;
1985), computer chips implanted into rhe brain allow II person to jack
directly into tbe Web, to experience viscerally \vhat is a fictional, yet oper-
ative, space.
Because visual sociology comprises images and science, it is appropriate
that \ve study the relationship between these clcments. The images that
visual sociologists make are also part of these issues; we should study our
work as part of the study of visual society. As new technologies alter what
and how we see (even changing the nature of sight, reality, and imagina-
tion), the issues will become both more complex and more important.
+ A Visual Social Science Through Research Photography
The above comments offer a broad view. 1now shi ft to the mundane opera-
tion of visual sociology, showing the operation of several paradigms of
visual methodology in photos of a bike ride in Italy.;'
I made the photographs that accompany this text while I was teaching
visual sociology at the University of Bologna, experiencing a culture I was
pretty much new to. lVly students and I explored the idea of seeing (and
181
 Me I HUU::' (jr I....ULLt:L Ill'll.:! AI'IU AI'IALY LINl.:! r.lYlnIQ\..AL IV\J-l,l r.KlI-\.W
photographing) forms Jnd levels of social control. In aU cultures, the rela-
tionship between formal and informal soci~1t control is complex, and in
italy it seems extremely so. It \\;ould be only fair, I reasoncd, For me to
explore this assignment myself. Participating in traffic from the vantage
point of a bicycle scemed sufficiently complex and visual to be a suitable
photographic project. Thus, armed with my intrepid Leica, 1 phoro-
graphcd the samc route, a maie street that bisects the circle of the inner
city of Bologna, on several occasions.
From a personal standpoint, I cert~lin!y found it intimidating to mount
my beatcn-up {pink} bike and take on Italian traffic, with or without a
CGmera. 1 was a guest of my colleague Patrizia Faccioli, who told me to sim-
ply follow her through the moving maze of dangerous vehicles and vulner-
able, darting pedestrians. Of course, this is the point: Things:takcn for
granted by a cultural insider (which rules are followed, which norms guide
beh3vior that is not regulated hy rules, and what areas of social life lie pri~
marily outside the perusal or rules) arc not obvious to cultural olltsiders. In
the case of negotiating the street, not knowing is dangerous.
I had already photographed a different European city from a bicycle,
and I'd become rather adept ar the process.Ivll' camera is not automatic, so
1steered with my left hand, holding the camera in my right. 1focused and
set apertures and shutter speeds before starting off. 1 framed the images as I
biked, advancing the film with the thumb of my right band. The resulting
photos can be used to illustrate complementary paradigms of visual re-
search, which is to say that their sociological meanings can be organized
within the logics of at least four research strategies.
The first of these strategies can be referred to (cautiously!) as empirical.
This orientation toward photography recognizes that a photographic
image is created when light leaves its trace on an dement that has a mem-
ory. For the image to exist, there had to be tight reflected off a subject; thus
the photograph is a record of the subiect at a particular moment.
From this perspective, the photographs document several levels of
social life. The simplest is a bicycle commute through [talian traffic. The
photographs record such mundane information as how many vehicles of
what types share the ltalic.n streets at a certain time of day. The pho tO -
grapbs suggest some of the jockeying that rep_tlates a complex mix of
human interaction mediated by machines. If they are read carefully, with
the help of a cultural insider, they begin to offer evidence of normarive
behavior. This is simple information, but it is precisely the daily occur~
rcnces of life that are often the basis of sociological analysis. Above nil, the
182
K!:;'I/f/UY'fllrJg V!;,-UOJ IVlCtnOGS: \.:JOWeo to l'.Jeuromancer
visual cataloging of life is efficient; it would take J lot of words to convey
the information in the photographs-.
Although these images produce what 1 consider to be empirical daD., I
do not claim that these images represent "objective truth." The very act of
observing is ~nterpretive, for to observe is to choose a point of view. To
elaborate: As a photographer, [choose a point of view by aiming the cam-
era and by choosing the toea/length of the lens. Camera lenses do not see
precisely what the human eye sees-some magnify a small visual area, likc
a tdescope; some see wider than the normal vision but move the fore-
ground of the subject farther away; others act as microscopes. For the bike
photographs, 1 used a 35 mm focal-length lens because it sees neady what
the eye sees (but Cllts off the edges or normal vision); thus when 1 hold the
camera in front of my bce I need not look through the range finder to
imaglne the image I'll make. The person looking at the phocograph sees
the frame as familiar because it is rather close to how the eye records infor-
mation. Thus the decisions of the image maker have profound effects on
the kinds of sociological statements that result from their images. J
In this project 1 used 400 ASA (black-and-white) film, which is properly
exposed with a small amount of light. Thus 1 could use a fast shutter speed
(freezing the motion in the frame), and [ could also use a small aperture,
which created a deep depth of field (in lay terms, the images are in focus
from the front nearly to the back). These are decisions that rendered the
photographs a certain way. Different film, shutter speeds, apertures, and
lenses would have created different visual statements. These statements
would not be "betrer" or "worse," but they would be decidedly different.
For example, Photo 5 freez.es the action of about 10 individuals on differ~
ent kinds of two-wheeled vehiclcs in an extremely dangerous confluence.
The image, crisp and frozen, suggests order. Likewise, the elderly man in
Photo 8 is not a passive spectator, as he appears, but a pedestrian about to
make a decision as to whether or not to venture into my path. If I had used
different shutter speeds and even diffcrent angles of the frame (point of
viev.,')l the photographs would have communicated more of the chaos and
uncertaintics in the negotiation of the public space.
It is easy to sec that empirical evidence is both constructed and real.
~ecoming a visual cthnographer means becoming conscious of the potcn-
tlal to make visual statements by knowing how the camera interprets socia!
r~ality. This means learning how cameras work, making the technical deci-
s;ons that, in fact, create the photograph self-consciously, and relegating
t 1e automatic camera to the wastebasket.
183
METHULJ':J UI- LULLtL lin\.;] AI'lU ANALY LII'Il:l t:NWlt"'..ll..AL /VIAl t:t"'..IAL:) ",,~rrlluYlfjjIlY VI:'UU/!VII;;'LfIUOS: \.:JOIlleo [0 l'leUrOmoncer

8 9

G~
~a
".:.... It:/2'l~, ... '

10 11

13
6

14 15

5 7 All photogr"phs (i) by Douglas Harper

184 185
 Heimaglnlng VISUOIlVlernoas: l.JOllleo to l'leWUlrlurIL~(
METHODS OF COLLECTIHG AHD A",ALYZ,HG EMPIRICAL MATERIALS

ingly larger tower in the phoros is a way to nate the progress of the
.z,. Visual Narratives sequence. Seeing the context of the photographs from the bird's-eye view
of the tower, one realizes that the path of the bicycle trip is through one of
_ 1d the idea of empirical data in single images to.J seq,uene of pl~~~
e
T the main drags of the inner city of Bologna. The overview is a visual sum-
o e.xpal. 'once, ( of visual narrative. SoclOloglsrs use vcr a
lOS IS to tntroduce r\:c c . f '.
b th as first-person accountS and as
narratives to tell sociOlog IC2. storleS'1 0. _', d Sp'lce On the level of
mary of the images that precede it.
The second image in the series frames the series from the perspective of
. I' f I-I throng 1 time an ~.
cultural stones t lat un t .
.
?
. ,._ lsistcnt with symbolic interaction,
I ' -h'" naff"ltiVe VICW \S l-Ol .
a spectator rather than :1 participant. \Y/e gJ.in the sense that cars, motor
mlcroana ySlS, l t: < "" t . etion based on interpre- scooters, and bicycles arc waiting for the signJ.1 to begin a race. In fact, that
: -' ~ to hov,I \\'c process 111 CI a c
which 111a1(cs us scnSIUVt: . .. . _. : .. '. ~t11bodies the flow of is exactly what the moment is like, as recorded in Photo 3. At the intersec-
.
[aOons.-111reraction bv
, I dehnttlon, IS naflcltlVe, It e tion where we wait, a slTIall street suddenly becomes a much larger street,
human experience." .' [_ The most common visual and oncoming traffic crosses immediately to the right. Photo 4 shows the
1 . 1 - - tC'11 n'uotlve Il1 leH m. '
Vis~la ~-la~ena s a.re ~-:s de'scrLibed above: single images taken sequen- moment just out of the "starting gate," with several different kinds of bikes
narranve IS fIlm or vid~ _' d) that \vhen viewed in rapiel succession, seem
L
and scooters roughly following the white lines marking a place for t\'o'O-
(iaUy (ohen many per secOn. _ L,: " But stitl photographs can also cre- wheelers. Noteworthy is the presence of a "sportsman" on a touring bike,
ta re-create the moven:ent thec)e ~e;~;:51 198') For example, a succes- seemingly out of place in the urban stop-and-go traffic. The two buses pic-
::ltc sociological na[[<1nve~, (Harpet'l '-'. (~. n from'l drunk board- tured are going opposite directions, one leading traffic to the right and the
f h \1S~ )f'l tramp wor (er COl11mg ow ' , other bearing toward the dispersing traffic. The remaining images show
~iono p .orograp: / "1 _~ SOO-mile trip, re<1sscmbling his identity as.:1
mg a frclght tr~li1 or _' -1 ' _b tells the story of a culture, a story th3.t IS different points on the roure, as recorded on different days (note the pres-
worker, and t<1kmg on anot leI' l~ __ ._d photo'\Ollrnalists hnve enCe of shadows on some images, and disappearing buses and cars). Pharo
.1 bI
rcpe'lted WIt 1 su t e vana 1
. -t' ons . fhc most rcspecre _ S \ d '\llene 5 shows the close reckoning between the scooters that are turning left im-
, _ hantage Euuene mit 1 an [
used this narrati~e form to ~r~at /1~ ~ociLal i~lov:menr surrounding the mediately after the scooters moving ahead (a crash seems imminent, as is
Smith's (1975) Visual n:l[[J.n,vc a ~ _ .. n describes the scenar~ often the case). Photo 11 records a moment in a game of "chicken" where I
lethal poisoning of a fishing vtllage by a corporano . 'fl Bc)a!' o(Beth did not faint at the onrush of a cab (and a woman on:1 scooter), causing the
, ' T VI' 1" . SU'l\ narrative JC ~
ios of several social groups. l\.ent '- IC 1 S Vdl I' (. ',n addict who is also a cabbie to slam all his brakes halfway through his turn into my path, Several
. \, Efe of a Swe 151 nero 1
(1989) traces events 111 t 1 e , ~.l _ . dividua\ to severa images show how pedestrians and motorized traffic interact. The elderly
prostitute; here the sac, narratlve moves trom t le m man in Photo 8 appears alone halfway across the street; several groups in
institutionat connections. 1 'l)icveling doV','ll the Photo 12 are crossing the street using differentsrraregies. Finally, in Photo
. stu d
Nil' narrative ' I 'Iltut'll
Y (ll t lC Cl. '- -
p 1en011lenon Ol. J
.- d in seV~ 71am observed photographing and the onlookers seem surprised, perhaps
- . . f 1 Trer sc>vera\ days IS constructe
same street at the same tlme o. c a) 0\ . c '- . - tt e same route, perturbed. This is such usual bell<1vior it is by itself a norm violation. In any
_ _' ' . :ie at dIfferent ttiTIes on 1 '
eral ways. It nllxes Images l~laL fl' htandshJ.do\",rsthatappear it is a record of the impact of the researcher on the reality he studies.
" , . a CQ'UCHUb'
\v hich is indicated by the varymgblpatterns
d' \ .
0 19
ay the sequence \S
.choosing images from several bike trips along the same path, I was able
in different images, Evcn assem e - tn t 11S ~v'-'.
.:J'
. the narrative lnclude several events that occurred at different times.
cal narrative, although the relationship ot these Images to Every visual narrative involves a decision concerning how much in for-
'Ia\\' of time is arbirrary. .. \ b y .Images tlla. r Sll'Jges t , to include per time unit. A film or video shows so much informa-
I d .1 C>~
Tl b' ~vcle narrati'/c is trame at e1t1er ene . ' ,-, It appears to reproduce reality. Thus it is not surprising that video and
1e lC, . d 1 1- - -. t1 e narratiVe 111 '"'-
cultural definition of tht: blCyd~ ~n .t 1at UC\3tc 1. ~ s (he sequence used in the analysis of L:lCC-lo-fao;,;l; interaction. The exan.plc ben'
. .., '~\ -, aclvcrtIs1l1g Image t 1at open ., 1
geo?,raphlc context. .- It: . . d ~. 1-' on The twa a small number of images over several minutes of social life. It
gest~c; that the bicycle represents both utility an ias ~~ rs 'in PhotoS 9- .b,e in the sequence that questions of social nOrms can be approached,
was taken from a tower (the same rower th~t aPIpea _. 'el"' The It 15 clear that one would need more visual information to explore this
1.cl the blcyc 1St tl,1\' ~.
looking clown at the street a \ong \V 11 1 "
187
186

subject in satisfying depth. Probably video, with simultaneous sound,
\'vould yield more meaningful data.
The visual narrative, like the ',ndividual frames from which it is made, is
a result of choices and decisions. If the researcher is conscious of these
choices, the visual narrative may become a useful way to study certain
kinds of social pattcrns. The methods used will, of course, influence the
questions asked.
.,y Eliciting Cultural Explanation
Phoro elicitation has become a familiar, if underurilizcd, qualitative
method..; Simply described, in using this technique, rather than asking a
subject, [or eXQmplc, "\Vhar j~; the role of the bicycle in Italian culture?"
a researcher might present a sct of images such as those in this chapter to
an Italian or a group of Italians to dicit their explanations. Of course,
responses to thc images could well vary according to whether the respon-
dents participate in the pictured reality as bicyclists, scooter or car driv-
ers, or pedestrians. As such, the images function as a kind of ';culrural
Rorschach" test.
To demonstrate how photo elicitation works, I presented these images
to my Italian colleague Patrizia Faccioli (PF), who appears in Photos 9 and
13 as lead driver in our convoy. Because she is a visual sociologist, the pho-
to-elicitation interview provided a natural way [or her to share her re-
sponses to the photographs. Her responses to these few images filled sev-
eral pages; the edited excerpts presented below show the range of topics
and the depth of the information produced but do not reflect the full range
of her responses. As a demonstration, this is the tip of a larger cultural
study iceberg.
The photos led PF to comment generally on the overallmcanings of the
bike in Italian culture:
The bicycle stands for: rapidity, conveniellCe, physical exercise, danger,
risk, challenge, reduction of stress, increase of stress, cancer in the lungs,
cold in the winter, hot in tile Slimmer ... fleXibility of the [trAficl rules,
aggressiveness, attention 1 ncgoti~1tion, social conscience, no pollution,
autonomy.
The bike increases your stress; you must have 1,000 eyes open (behind
you, in from of you, at right, at left, above, under ... ) because every 10
188
"C"JP, IU~IO ,,, 'y y I::;UUI IYIt'tlllJU~" \JUIIIC'U LU l'leUrOnlanCer
seconds you risk your life. Nobody in the street cares about cyclists; they
seem not to exist: pedestrians who suddenly cross the road, ignoring rh;t
yOll arc coming; motorbikes which pass you on right and left and almost
touch you; cars which don't give you the way even if you have the right of
way. And the buses ... It seems that among the bus drivers in Bologna,
there is a sort of bet on who can pass the most closely to.'l bike wirhout
touching it.
So, using the bike n1C<1l1S p.'lying much :mention, bur also being aggtessive
and brutal. You have to stare at the car or motor scooter driver (thc bus is
too high) telling him, through rhe eyes, that you are going to pass first
and that you will not stop.
In reference to Pharos 5 and 6, PF commcllts:
The light has turned green and the scooters and the bikes weIH quickly in
front of the cars, because the light is green also for those coming from the
other side who wam ro turn left. The challenge is to pass first, before the
others cross. You can see this in Photo 5; those going toward the tWO towers
havc the right of way, but' if those arriving from the direction of the two toW-
ers can arrive first to turn left, they will turn and the others will have to Stop.
That is especially the case if you arc on a bike (tecall what I said about the
use of the eyes to conullunicDte you CIre going to pass anyway). This is one of
the most dangerous moments in my everyday experience of going to work.
Understanding the experience of the bicyclist, PF tells us, leads LIS to
understand what Italians call "minor dcviance":
Using the bike means ... not being oblig;:ned to follow rules. You are half
1llachine and half pedestrian, and you :lre not very d,ll1gerous for the others,
you can glance around and, if possible, pass when the light is red. When
there is a traffic jam yOll can enter the portico [the covered sidewa lks of Bo-
logna, which weave 42 kilometers rhrough the city, visible in Pharos 5' and
11] and O1ove ,llllong pedestrians. YOtl can go the o'pposite ditection in one-
\V~IY streets, paying attention to the cars, of course. As an ArnericlIl you
lnJght say: "The rule is the rule, always and anyway!" Butyotl arc wrong! ...
Certaill types of rules have to be evaluated in order to decide if it is right to
.foll -
nw ,h,
- em ur not. TI llS . ;lIUlU(
. Ie 15 very dff
1 use
dIl1 It:1 Iy, espeCIally
. regard-
Ing the rules of the road. Of course it's not the same for the penal code ...
bUtn;:C:-lr-\'
,
-. "... -f I I \. \. I h
<.1' l:.lllg L-<::rtalll typC5 u ru es we La lans t 1111: t ::It we h:lve the right
t~ ]lidg the rule and behave accordingly. For example, Italian people de-
e
cidedl\VItl1 t Ilelr . b-l . I I .
t: laVlor) t lat t le safety belts are Important and useful on
189
 IJlt.l HUu:J V, \...ULU:''-' "',U '""'" .... ",", '-' -"

~ Experience and Image

, _, WI1 The police have tried for a long time [()
the (reeways, but not 111 tht: t o . . I'll. .' tlll-' 'lbout. So now the
'I '\ l orhll1 o COUll 11I1l1g ~.
have the rute tol o\\,el, lUt n ~ 0, ~ \" _ "ec"e_ an ul1\.. . ritten rule tbat I next address a fourth way to look at these images, which 1 call the
1. b,\ 'WlO" as 1\ t lere \, ,
,Jolice l1'1ve aceepttx.\ tllat 1;.1< .,' .' I' . le',wn- \\Ie can't phenomenological mOLle. The vantage point from this view is the self, in
[l I s when people l nve In ~....
to\er~lteS the absence 0 )e L - _ j, . 'I-.d th'lt the ru\e is right. rhis instance a self who happens to be driving a bicycle through crowded
. I,' . l ecaU5e someone l eelL t;
accept any sort 0 [- tu e lust 1
, ' .
,T" k" leo Ie who live every day III streets of an alien culture. From the phenomenological perspective, photo-
People who n1;1ke rules can also l11,t\.-e In\SW es, l . ~
the rO:1d can understand which i" the best behavlo\. graphs express the artistic, emotional, or experiential intent of the photog-
rapher. Here sociology borrows from art; the phenomenological mode
draws from early photographic art movements, as photographers sought to
. . _'
\a, with female models is a di fficu\t cultural define themselves as more than technicians and to assert that photographic
A bicycle In a \\'tndow dlSP ) , . . ' . p\ ' l PF cxpbins
svmbol for an olltsider to understand. ~efernng to . _10~O , expression is equivalent to artistic expression in other media.)
, \. f seven\ forms ot transportatlun. Treating photographs phenomenologicall); is not common in sociology.
the culmra meamng 0 <

Perhaps the best examples are found in the work of Richard Quinney (e.g.,
. \: 'l-~ I a is the prize drawn by lot a\110ng those 1991, 1995), who uses images to interprer and render his own experi-
This perfume shop shows all '-- t 1,. t T '. ,t~ment tr~Hlsbtes "The :lU- ences, reflections, and memories. In the case of the Italian bicycle photog-
, ie -, the perlul11e. 'I1C St,l e .
who buy a Ccrl,lIn pro uLt, . j' _ . ' f how much the bike raphy, m), intent was to photograph what was new to me, but routine to
. . c.
rogrtlphcd bILl: cou )C Y
Id \' ours." That's an ll1lll.;ltOI 0
'. j 1'[-. \'\/)111en \vho well' ,1 (.1-
:,
~I'\"' u\t:lreandevt.'lyeayll:. l .
natives.
has bccotn. cpa. rt ol t".hll1 L - I' _ I .. I' in Ihe I)oster IS wearing There was a mix of euphoria and fear in the experience of biking to
' \ ' 1. . ve')'lhv lite: tleWOl\1,l
mOllS pertl\\l1c use 1\ ,-esll1 e c 1 h' S'd bil'es were pco-
. . Ol1e- In the '605 mose \V 0 1I. e .. - work. To photograph the experience was to look as I was doing-a harder
elegant dresses, nor sport~. \ d' j )t \,ave money to buy (or to
, I r k' g chss w 10 Ie nl ' . thing than one might imagine. If you look at the photographs with this in
pic bdongmg to t ,e wo. ln~ " d I , . _ I, n SundaY with their
\ I f leople usc tnc L1f on ) 0 " mind, you see the photographer choosing points of view that illuminate
lise cverv cbv) a car. i or 0 I .I (r . J :11\ people
. ", \ '_ , 0': the increased standaru 0 \Vll1g -' .. different aspects of the unfolding social reality. The sentiments behind the
familv. Nowadays, 1e",ause . ,\ 11 _. tel'S -lre crowded.
. , _ d the towns, 111,lluly tlte OIL Len ,< image making must be understood through another modc of expression: a
have one or more e,Us a n , -I' \.,1 [0 soci,-11 cbss or
_ , " al'-ernatlve to the car, not In ~t:l, ._ .
So the bllce has become ,\0. , . < I ' _. ~rt)SS town. I hl~
ro diary, a poem, or othcr expressive form. The photographs by themselves
L _ I t 'lplC W'lYS to mOVL: ",--
muney. It has become one at t le mos _r, . \, t' youn a IJCoP\e, but abo seem curiously detached, even quiet and subdued, yet they emerged from
. . 'c\'" (0- rhe scouters. Sllll\t::11 s, . 1:> ,,-I
is a suntlar conSI erat. Ion., \_\' _). __ b.wc scooters rnon; all l an adrenaline-saturated experience. Here I question whether the same
I ,- (teachers I\CC me ,elL.,' -
employees, lawyers, (octot., " ' I :-;;; 000 So the scooter means phOtographs serve both the empirical and the phenomenological mode
. j ~. _ -,)snng as Inuc 1 as "J, . ..:1
marc hne am expcn~lve, L . . . 1-' r' different; uscl equally well. The clinical distance in rhe crisp images of the empirical ren-
. . ~" s 'l11bol. 1. he motoreyc es a
both f'1pldlty and a ~tatu .. )
l;: [
. .coo ter " (or dering does not communicate the experience of the moment.
,. .. -e" on the road. In a town you can sec s _~ , ,
lluinl\' tor vacation or rae ~ \) b'k', (for both the ve r) Photography can produce data that enlarge our understanding of socio-
, I [t . -lOor peop e or I t:. .
the rich people) or mopCl s " .at Ii . ") Th esc arc I'd L" '0\ t)'pes of course.
poor people and rhe urbnn CCll O~l~t
.., processes, from the formation of one's own definition of the sim-
The reality is more complex and mlxccL to the negotiation of actors with different machines. Photographs
details that may engage viewers to reflect upon larger cultural reali-
Using photos as sequences allows us to see how social actions rake
I I tmages elictt cu\tural
These interview excerpts 5\lOW 10\\ <., -.' l 'l"tioo) to
, ( . 've negonanon at sOCIa c l.-
that ranaes from the \ntcro normat~. . ' , IIv '1 completiOn example above shows how the most mundane of events, biking
~ ' \ . ["-'tation ll1tervleW 15 rea. <
mra\ definition. 1 he p lotO-e \d < \ "\ ' t bein F that the a city strCCt, may be thought about in several sociological ways.
, d . - - Horts t le cntlC:l pom t> l four approaches do not begin to exhaust the possibilities for using
'llC empincalal1 l1arranve e " .' t,),tV','lJ(
, \. . I ~rl' '1 -ec.:Onl1ltlOll I d
,mg supp
. I'Ie d ,', '([0[11 a culrura 1IlSl(ler. 1\5 \5 11 ~ t> in the research process.
_:> c

.
sntutes ,~" ,'s L"[['t[I1"'II\)'
. a ",--r:1U I
deftned,
191
190
 'V\E..' I lV'-'-J VI ,--V'-'-'-'--' ,,~'-- M'~"'" r-<';M'-' "-Ol''-' ' - " " " ' ' '.... ru_
The Social Construction of Photography and Visual Sociology
It is not enough to describe visual research in the terms offered above. Like
all research, visual research depends upon and redistributes social pO\ver.
In the case of visual research, these issues me compounded by the power
associated \vith photography. Thus we speak of the "social construction of
the photograph" as part of the discussion of the politics of visual sociology
itself.
The photograph is socially .:onstructed in the sense that the social posi-
tions of the photographer and the subject come into play when a photo-
graph is made. It rakes social power to make photographs (1~1gg, 1988),
panly because making photographs defines identities (Spence, 1988),
institutional relationships Uackson, 1977), and histories (Copeland,
1969; Rieger, 1996). A father may photograph his children in ridiculous
poses, but the children do not generally have the social power to photo-
graph their parents arguing (ur making love) and to present those images
as the "official" family story (Chalfen, 1987). Sociologists and anthropol-
ogists have assumed that it i~ their right to photograph the people they
study, and thus to present them as academic subjects (and in ways in which
the ideological bases of thclr relationships are disguised). Edwards's
(1992) collection of essays 011 early British anthropological photography
demonstrates exactly this point: The images are not objective renderings
of objects of scientific studit:s, as interpreted by early-20th-century an-
thropologists; rather, they are markers of colonial relationships. Anthro-
pology was a science of the c:::llonizer, and the images made in the service
of anthropology defined the Ilative in ways that rei fied the relationships of
superiority and inferiority endemic to colonialism. There are no photo-
graphic records made by nati'/es of colonialism (or colonials), but in recent
decades prior "natives" have assumed the right to make their own images
and tell their own visual stories. This has, of course, called the relation-
ships of traditional anthropology into question.
The social construction of photography is also a matter of gender. For
example, in my current rcseach I am using about 110 documentary pho-
tographs made just after \X!orld War II in a study of the evolution of dairy
farming in the northeastern United States. The phorographs are part of
a collection of nearly 70,000 images produced under the sponsorship of
Standard Oil of New Jersey and directed by Roy Stryker. 1v1y farming
collection includes about 40 images made by a female photographer,
Charlotte Brooks, and about 70 by several male photographers. \X!hcn I
192
".r" .... "'n......., "~"" ... ~"""~ .,J....... , '.,~ .. '........ J ..., .... " , .... '-J CV, .""v' v,,'v' ''-'"''
compared the phOtographs on the basis of the gender of the photogra~
phers, I was startlcd by what I found. For the men, the farm womcn's work
was largely invisible. They did not photograph women as productive parts
of the farm, nor did they photograph the work of maimaining and
provisioning the house (and taking care of children). Brooks's photo-
graphs, however, cover many of these excluded topics. The difference is
significant because it shows clearly that if we regard these photographs as a
document of farm life during the \Vorld \Var II era, we accept an incom-
plete portrait as a full record. Indeed, the photos, in this case, are a resulr
of the social constnlcrion of "maleness" and "femaleness" typical of 1940s
America. The role of gender in creating photographic meaning is yet
another largely unexplored subject. 6
Social scientists should be aware of the social construction of the image
for several reasons. Wle need to acknowledge that photography embodies
the unequal relationships that are part of most research activities. I can
enter into the worlds of the poor by living temporarily on the street, and I
can photograph the worlds I encounter there, but a homeless person can-
not infiltratc and photograph the life of my university president. This real-
ization has led many social scientists to abandon the use of photography.
For others it is a cautioning awareness that should help us to overcome the
inevitable power differentials of subject and researcher. Some sociologists
have confronted the issue by giving up their own photography and instead
teaching their subjects to use photography and writing to investigate their
own culwres and, perhaps, to empower themselves. \Vendy Ewald is a
leader in this field, for she takes on the asymmetry of adult/child as well as
First World/Third World power differences. Her first published project on
Appalachian children is a good introduction to this approach. The photo-
graphs and \vritings of her students contradict the stereotypes long associ-
ated with the imernal colony of Appalachia (Ewald, 1985). In this method\
photographs (often accompanied by text that expands lipan the images)
represent inspired refleetion~they are not social science in and of them-
selves, but data that sociologists and others should put to use.
+ The Essence of Visual Sociology; and, Where Are We Going?
A~suming we are talking about research methods (given that this is a hand-
book of qualitative methods), and assuming we are speaking about the pho-
tographic end of the movement, the simplest way to do visual sociology is
193
 Mt II"1UU::" U , \..ULLC\"III'j\.:l HI'jU HI'IMI..I !..'""'..:J I..IV,, "~I"""'''... "or.... , '-,,,,,'--'
to photograph with sociological consciousness. Howard Becker (1974)
was the first to make this argument and the point has nOt been made more
elegantly since then.
Becker suggests that we think theoretically when we do photography.
\\lhm does this mean, exactly? For Becker, all photography is done from a
theoretical perspective, but lit6e of the theory is sociological. Our normal
vie\vs of the world (which Becker calls "lay" theory) tell us where to point
the camera and how to use the camera (speaking technically) to make
images. Thus, when we photograph, we re-create our unexamined, taken-
for-granted perceptions, \Y/e are interpreting sociological topics in our
unexamined theorizing, and our photographs are our conclusions. If we
are photojournalists (sec Hag~.man, 1996), we learn to present the theo-
ries of our nevvspaper editors and the recent conventions of photojournal-
ism. \V'/e do this not only by choosing topics and specific images of those
wpics, but by using particular lenses (frames), apertures, and shutter
speeds, If we are sociologists, we presumably have theoretical knO\vledge
of our subject, the way Bateson and :tv1cad (1942) kncw (he Balincse before
they began photographing their stillullequaled study of Bali culture. This
prior knowledge will tell the researcher: "There is the enactment of a rit~
ual my subjects have described . . . . It lasts 20 minutes and has four
stages. , .. I will photograph it to highlight the transitions and inter-
actions among actors." From this perspective, the work of visual sociology
is straightforward: \Y/e bring it into the research process to extend our
knowledge of the subject.
But not all field-workers have the kind of preexisting knowledge that
Bateson and :Nlead had. Thef(~ arc at least two alternatives for visual soci-
ologists. The first is to use the camera as an information-gathering ins[ru~
ment, to discover what Glaser and Strauss (1967) call "grounded thcory."
Photographs made during the research experience concretize the observa~
tions that field-workers use continually to redefine their theories. In this
way photographs help build theory, In fact, the need to make photographs
in the field requires that the field-worker look at something, and these
beginning observations can be the starting point for making theory.
The second alternative is w use photographs to confirm and develoP
existing theory through photo elicitation, as described above. Researcbers
may use photos of events people experienced in the past to draw OU[ mel11-
aries of their history. l\!largolis (1998) studied the political conscious~ess
of coal miners, using decades-old photos of mine \vork to interVieW
elderly miners about events and their interpretations. In this case, one
194
'''''''"'U~'''''"~ "';'UUI J"t:"LIIUU". \..JUJlJt'U LU I'leururnuncer
senses that the intervening years betwceI1 when the photographs were
taken and their interpretation led to deeper reflection than would nor~
mally be associated with the photo-elicitation process,
Other researchers have engaged subjects in the photography as wel1
as tbe interviews in photo-elicitation research. Van der Dos Gooskens
Liefting, and van.Mierlo (1992) photographed a multiethnic Dutch neigh~
borhood under the direction of five subjects. The researchers then inter-
viewed their subjects using the images made on their earlier photo tour."
Finally, the researchers interviewed other informants llsing photos made
by their neighbors, \vho were of a different age, gender, and ethnic back-
ground. The result is that as researchers we understand that a neighbor-
hood is made up of people who share a material space bur define it dif-
ferently; but in this case, the five subjects also Came to understand and
appreciate the perspectives of their neighbors.
In all examples of photo-elicitation research, the photograph loses its
claim to objectivity. Indeed, the power of the photo lies in its ability to
unlock the subjectivity of those who see the image differently from the
researcher.
A final thought: One can say that visual sociologists seek to find a way
to intcgrate seeing into the research process. A sensitive field~worker is
already nearly equipped to do visual sociology. It helps a great deal to un-
derstand how the camera records information, and it is important to
understand the impact of photography on the research process. Finally, it
is important to understand how various constructions (technical and
Social) influence how the photograph is made and interpreted,
\11e are living in a time of electronic revolution, which, as I have pointed
our, has affected how images are made, distribured, experienced, and
understood. Surely these changes should have an impact on visual sociol-
ogy, Yet we are in a curious place. Sociology remains essentially uninter~
ested in visual culture (Denzin, 1995, being an exception) or visual meth-
Ods. Our textbooks contribute to the problem rather than solve it. The
bOOks most professors lise for first courses in sociology are filled with col-
orfUl, uncomplicated images (chosen by art directors primarily to spiff up
pages of text). The photogrJphs restate the most obvious themes of the
already watered-down text. Given that these images are the entirety of
whot most sociolosists cxpcrit:ncl;; d::l '"Yi::lual ;mciologYI" it io p\;rlmpo IWL
surpris; h h . 1 . 1 . .
. ng t at t e VIsua SOCIO ogy movement remams tangential rather
than, ~n the mainstream of the discipline. In 3mhropology, with its tong
traditIOn of anthropological filmmaking and examples like Bateson and
195
 IVICI nULl.) u r ..... UI...I...I... ..... ,II"'.:l M"',-, M"'M ... ' '-II''""' ... " " "",\..,,.., ... ,.,,..,, ... ",,.., ...-'
ivIcad's still-admired visual etlnography, the Case is a bit bencr. Still, most
an-ides published in the nvo journals of visual anthropology are primarily
words-;lbour-images rather dEn repons of experiments and demonstra-
tions of visual thinking.
\X!h;l[ is the future? One p:Jssibility has visual sensibility leading an
energized social science thar is experimentally ethnographic and theo-
retically interdisciplinary. The new technologies promise myriad WJys to
bring changing visual experience into the production of social science and
the understanding of visual dimensions of society. Another possible future
hZlS social science largely unconnected with;l [;lpidly dunging technologi-
cal \vorld, only mildly imeresr,~d in studying society as an observed phe-
nomenon. In this scenario, those of us in the snullmovement of visually
inspired thinkers continue L'O wave our banners frolll the sidelines. [sup-
pose the ::lcw::ll future will be some mix of the two.
';p t'--!otes
I. !n my ch~lpter for the first editicn of this book, I suggest that the area in visual sociDI~
ogy lllO;;t relevanl to l.jualit;lr"lve methods is photographic fieldwork, or visual ethnogrJphy;
! note that visu~l1 ethnogr,1phy, howe','er, is the focI! puint of several criticisms (Harper,
19(4). These lnclude a postlllodern critique of scientific ethnography, a postlnodern cri"
tique of documentary photography, and a criticism that the ponrayal of subjects inevitahle
in rllllch visld ethnography vio!dtes ethical principles of fieldwork. These critiques r:lise
substantial and serious issues, but I argue that it is bener to reform visual methods than to
discard visual ethnogrJ.phy in a sweeping rejection o[ modernism, scieuCt:, ethnography, and
dOCllllJe!lt:lrr photography. Because that paper w:!s subsequently updmed (Harper, 19(8),1
have soughr new ground for this chapter.
2. [ first suggested these "rypes" of visual sociology in a paper that W;JS inspired by
t'-licllOls's (I nl) discussion of film thory (Harper, 1988). For a broader view of visl!"JI soci-
ology, sec also John Grady's (1996) arf;Ullll'1lt for an enbrged scope [or visual sociology :llld
Charles Suchar's (1997) plea [or an "interrogatory principle" for visual sociology. A collec-
tion edited by Chris Jenks (1995b) explores several dimensions of the sociology of visual
perception on this gl'neral theme. Jenks (1995 a) notes that" 'idea' derives from the Greek
verb meaning 'to see.' This lexical etY:l101ogy reminds us that the way that we think about
the war that we think in Western culture is guided by a visllal p<lf3digm" (p. I). Jenks later
amicip,nes some ideas I have offered in this chapter with the statement: "The modern world
is very l1luch a 'seen' phenomenon. Suciology, however, itself in many senses the emergent
discourse of !11o(!l'rnity, Ius been rathc: neglectful o[ addressing cultural ocular conl'ClltiO llS
dllll lld:i :itlb:ieqlltmly become somewllal inarticulJ.te in relation to the vist1:11 dimensions of
social rebtions" (p. 2).
3. Steiger's (1995) analysis of relationship between photogr::Iphl'rs' choices .mel
tllt_'
sociologieJ! st~1tell1ents is the best place to become acquainted with this topic.
196
""''''' ..... ~''''''~ .'-'..,u, ,.,o.;;"'",u". '_H-''''':CV c..... ,":Cu,v",..",..."",
4. Collier (1967; Collier & Collier, 1986) first described photo elicitat'lon, a process
llsed in the 19505 in Cornell University rese3rch on rural :lssimibtion 10 urban center, in
Canada. The researchers C:lme to the realization that without photographs to indicate com
mUll understandings, their imcrl'lews did not make a great deal of sense. Photographs made
in the research became a bridge: between the subjects and rhe researchers.
5. A good place to start on this tOpic is with Dorothy Norman's (1960) study of Alfred
Stieglitz, [or whom photography was inregral to the emergence o[ modern <lrt.
6. After;\ thurough search (and on-line discussion), I found no references to the influ-
ence of gender in documetltarr expression. There have been many stndies of the illlpaCl of
gender on artiStiC phorography. Thus although my eX:lmple is limited because of the diffi-
culty of comparing the work of a single fcmale to the vmrk of several men, it is an interesting
first study.
<P References
Artieri, G. (1996). The virtual image: Technology, media, and construction of the
visual rcaliry. \liS/hI! Sociology, 11(2),56-61.
Bateson, G., & l\!lead, M. (1941.). Balinese character: Ii photographic illlalysis. New
York: New York Academy of Sciences.
Becker, H. S. (1974). PhcHography and sociology. Studies ill thelillthropology 0/
Visual COlllnmllieatioll, 1(1), "I ~"l9.
Bidla, P.., Chagnon, N., & Seaman, G. (1997). )~1IfOllUllIIO illteractiue: The ,J.Y fight
[CD-ROI'vl and text]. New York: Harcourt Brace Jovanovich.
P,ourdieu, p. (1990). Photography: A lIIiddle-brolu art. Stanford, C1\: Stanford Uni-
versity Press.
Chalfen, R. 0987). Snapshot' versions of life. Bowling Green, OH: Bowling Green
Popular Press.
Collier, J., Jr. (1967). Visual I1llthrofJology: Photography as a research method. New
York: Holt, Rinehart & Wil1sLOtl.
Collier,]., Jr., & Collier, 11,'1. (1986). Visllal anthropology: Photography as a research
method (Rev. ed.). Albuquerque: University of New lvlexico Press.
Cope!:J.nd, A. (Ed.). (1969). People's ParJ~. New York: Ballantine.
Denzitl, N. K. (1995). The cine/IUttic society: The uoye/lf's gaze. Thousand Oaks, CA:
Sage.
Edwards, E. (Ed.). (1992). Allthropology alld photography 1860-1920. New
Haven, CT: Yale University Press.
Ewald, \XZ (1935). Portraits alld dreaJIls: Photographs (/luI stories by children 0/ the
Appalachians. New York: \X'riters & Readers.
Gib5Ull, W. (1985). Nellrouwllcc/: New York: Ace.
Glaser, B. G., & Strauss, A. L. (1967). The discoVel}I 0/groullded theory: Strategies
(or q!laliti.1tiu resedrch. Chicago: Aldine.
Grady, j. (1996). The scope of visual sociology. Visual Sociology, 11 (2), 10-1.4.
197
H:1gaman, D. (J 996). HolU I teamed w>t to be a photojournalist. Lexington: Uni~
\.'crsity Press of Kentucky. j
Harper, D. (J 982). Good C01l1jJL11I)'. Chicago: University of Chicago Press.
Harper, D. (1987). The visual ethnographic narrative. Vis/wI AlltiJropology, 1(1), J-:J 9.
Harper, D, (1988). Visual sociology: Expanding sociological vision, A/neried/{ Soci~
olt.Jgisl, 19(1),54-70, j
I-Luper, D, (1994). On the authority of the image: Visual mcthods at the cross~
roads. [n N, K_ Denzin & y, S. 'I'voI1I1;1 Lincoln (Eds,), Hl1wJboo/~ of qllillita~
Autoethnography,
tiue research (pp, 403-412). Thousand Oaks, CA: Sagc.
!-L:!rpcr, D, (1998), An :ugument for visual sociology, in l Prosser (Ed,), ll11age~ Personal Narrative, j
based resedrch: A somce(JfJOI:. (;Jr qlfd/ihltiuc researchers (pp, 24-4 oJ). London:
Falmtr.
J:Kkson, It (1977). Killillg time. lduea, NY: Cornell University Press.
Reflexivity
-"----------- j
Jenks, c. (1995:\). The ctntrality of Jhe eye in \X'estern culture; An introduction. In
C. Jenks (Ed,), Visual culture (pp. 1-25"), London: Rourleclge_
Jenks, c. (Ed,), (199513), \!i$lIdl whim:, London: Routledge.
Researcher as Subject
Klich, K, (1989). The book olLkt(J. tvlillertowl1, NY: Aperture. j
IVlamber, S. (1974). Cinemi/ uJ6/t; ill America: StudiES iJ1llllcolitrolled dOCllnlCJ1-
Iilry. C;1mbridgc: ivlIT Press.
Carolyn Ellis and Arthur P Bochner
Margolis, E, (1998), Picturing labor: A visual ethnography of the coal mine labor
process. Visual Sociology, U(l), 5-37,
j
Nichols, B, (] 981). ideology' i1!ld the fllWgi'. Bloomington: Indinna Univcrsity PtCs-s.
Norlllan, 0, (1960), Alfred 51 ieglit::: All American sea New York: Random House,
Quinney, R. (1991 ).10111"ll(,Y tOil/~lI" place. Philadelphia: Temple University Press. j
Quinney, R. (1~)9 5). A sense sublime: Visual sociologY' as a fine art. Visl/al Sociology, + Overview: Extending the Handbook Genre
W(J-l),61-84_
Rieger, J, (1996). Photographing. social change, Visual Sociology, 11(1), 5-49, "Hi, glad it's YOlll" 1say, relieved to hear An's voice on the other end of the
Smith, WE" & Smith, A. (1975). Milld/rWlil. New York: Holt, funehan & \VinsroJ1. tine, j
Snyder, R, L (1968), Pare Lor{,JIi"z ilnd the docl/menlary film. Norman: Okbhoma "You sound upser. \X7har's the matter?" Art asks. .
University Prcss. "Oh, It's been a zoo in the office today-long distance calls, forms to fill
Spence, J. (1988), [J1I((illg fflysdf in the picture: A political, jJersollal and pholv-
grapbic cl11tol}iography. Seattle: Real Comet,
Olit, a barrage of e-mail, one student after ~mo[her. I'm feeling that end-of-
the-semester panic. ('d hoped to finish reading class papers and turn in
j
Steiger; R, (1995), First childten and [<lIndy dynamics, Visual Sociology> 10(1-2),
28-49.
Suchar, C. (1997), Grounding '/isual sociology research in shooting scripts. QlILlIi- ---
AUn-[ORS' NOTE: Th:m);;s ro our j!;t,ldtl;ltC 5!Udents at University of South Florida and to
Mitch Allen, Norm,lll Dem.in, Ros;mna Hertz., Yvonna Lincoln, L:llJrd Richardson, ,"nd
j
h1til'/2 Sociology, 20, 33-55,
\'\lilliam Tierney for cOl11menting on the chapter m,1nuscripL Carolyn Ellis gave"" short
TJgg:,1. (1988). The bllrdett of re!JrescJJlatioll: f".:.ssays Oil photographics mid histories.
partloa of this article as a keynote address at The International Advances in Qualitative
Bnsingsroke, Hampshire, England: Macmillan,
van der Dos, S, E., Gooskens, L, Licfting, lVI., (I.,,: van IVIierlo, IV!. (1992). Reading
lV!cthQds Conference held in Edmonton, Albena, Cmada, Fcbruilry 1999, <1nd published ir
in Qllalitath'L' }--]c;.1/t{J Rsi'arcIJ (VoL 9, no. 5, Septcmber 1999, pp, 653-667) as "Hcartful j
images: A 5wdy of a Dutch neighborhood. Visual Sociology, 7(1),4-68, Autoclhnography."

198 199
j

j
 Mel HULJ:J ur- \...ULLC\... \ Il'-l\.:l HI'-ll-/KI'If"U.. ! LIl">.:! '-Iv,r "", ....",- ,"'" '-''''''--'
semester gnldcs, but I haven't gotten to them yet. How's everything at
home? Are the dogs okay?"
"The~"re fine," Art replies cnickly, "but I'm not. I started working on
our chapter [or the I-LlIldbook of Ql1i1litative Research. The marc I think
about it, the more frustrated I become."
"I thought yOll really wanted to do this."
"\'\!ell, the first edition of the Hmldbook didn't sufficiently highlight
autoethnography and personal narrative. So I initially dlOUght it would be
a good opportunity to show how important it is ro make the researcher's
own experiencc a topic of investigation in its ovm right."
"\X1cH, the timing is right now to do that," I respond. "\X'hen the first
edition was published a lot of academics were trying to figure out how to
\vrite their way oUt of the crisis of representation, but there wercn't l1l~ll1Y
examples of authors who made themselves and their personal experience a
central focus of their research. Over the past 5 years, however, that's
changed significantly, what with the beginning of the Ethnographic Alter-
natives series, Dcnz.in's cmphasis on personal writing in InterfJretiue Etb-
nography, Beh;]x's TfJe Vlllnerable Obsel1luJ: .. "
"Granted, there's been a wave of interest in more pcrsonal, imimatc 1
and embodied writing," Art assents.
"So what's the problem?"
"I think we've underestimated the constraints imposed by the genre of
the handbook chapter as a form of writing," Art continues, apparently
deep in thought.
'"Hut you've wriucn many handbook chapters before," I point out.
'"Why is this different?"
'"Because those chap ten conformed to the conventions of thc hand-
book genre. They were essays, not stories. But in this piece \ve want to
ShOiU, not just tell abollt 3U':oethnography. Look at any handbook on your
shelf and what you'll find i~, that most chapters are written in third-persall,
passive voice. It's as if they're written from nowhere by nobody. The co n -
~7entions militate against personal and passionate writing. These books arc
filled 'vyith dry) dist::ll1t, ab~tracr, propositional essays."
"That's called academic 'Nrtting, darling. \1 When Art doesn't li.ll1lJh 1 I
continue in a more seriou:; tone, "But some of the authors in the first edi-
tion of the Handbook o( Qllolitatiue B..eseaJ'ch wrote in first person."
"Yes, but thc T usually disappeared after the introduction and then
reappeared ~lbruptly in [be conclusion," Art replies.
200
"And the T usually was a 'we,' and an ambiguous 'we' at best, which
sometimes referred to the authors as writers of the chaptcrs and somc-
times included all of us, whoever wc might be," I add.
"And the authors almost never became characters in the stories they
wrote ... "
"They couldn't," I interrupt, now immersed in the conversation, "be-
C[luse their chapters weren't really stories. They included little in the way
of dialogue, dramatic tension, or plotline, for [hat matter."
"'But, look, handbooks do provide a service," I continue, fearing that
An will decide not to collaborate on the chapter. "They provide citations
and sources, a sense of history, 'lod arguments others can use as justifica-
tions for their own work."
'"l don't question that they serve an important purpose. Hundreds of
students have been inspired to do qualitative research by the first edition
of the Halldbook ur QlliTlitatiue ResearcfJ, for example.7'l
"That's true, and we can't criticize handbook writers for failing to do
what they're not asked to do."
'"But we can ask why authors aren't encouraged to write academic arti-
des in the first person, " Art rctorts. "\"'\Thy should we take it for granted
that :In author's personal feelings and thoughts should be omitted in a
handbook chapter? After all, who is the person collecting the eVidence,
drawing the inferences, and reaching the conclusions? By not insisting on
some SOrt of personal accountability, our academic publications reinforce
thc third-person, passive voice as the standard, which gives morc weight to
abstract and categorical knowledge than to the direct testimony of per-
sonal narrative and the first-person voice. It doesn't even occur to most
authors that writing in the first person is an option. They've been shaped
by the prevailing norms of scholarly discourse within which they opcrate.
Once the anonymous essay became the norm, then the personal, autobio-
graphical story became a delinquent form of expression."
JUSt as 1'111 beginning to doubt that we can do this project, Art says,
"This morning I wrote out some of my concerns in conventional social sci-
ence prose.lvlaybe this will give us a place to start. If you have a minute, I'd
like to read it to you."
Relieved, 1S3Y,' "Sure, go ahead." I glance toward the large stack of term
papers. They'll have to wait.
An reads:
Lilw most social scicntists educated ill the 19605 and 1970s, I was
socil-1liz ed illto tlJe legacy of cln/Jiricis111. I developed all appetite (or
201
 NIt: I MUU.) vr ..... Vl..u... '--, "~u , " , ..... ,...,., .r. ... , L.." ,~ ~".,

generalizable ilbstractions alld unified /:;:Ilowledge. The first social sciencoe Tn the lua/z.e of tbese deue!opments, 1 doubt /uhether a handbooh chapter
lJi1ndDool.!.s wcre published when I was in graclztdte school, al1~i they (cd :/ns call help guide the lUnrl:;: ofthose who haue turned towLIrd autoetlmography
My professors pressed th.e pomt .that SClell-
/nllH;c r {or receiued l.!.11O/u/edge. and personal narnltiue if it holds to the uoice and dllthol'ity of a forI/! o{
tifi~-/(JlOw!edge is Clllllltlative and iiI/em; so eL:cry Ollce ~Jl a w/nle scholLlrs writing {hat this lUor/~ sec!(.s to transgress. How help/it/would it be to lis!
haue to step bad:_ alld assess th[ state of the {ie/d. JroJIlcally, Ihese assess~ references, define terllls, abstract {roll! imel critique exemplars, formulate
me/Its sometimes luere referred to as "stale~of-the-art" essays (all [[.rt t:h1t criteria for evaluation, or theorize the perspective of the "1," so readers can
was sllpfJ(Jsedly science). That's iuhat a hal1dboo/::. did-it gaue LIU ob/ectlue, mahe our lmowledge theirs? No, we need a form that will allow readers to
neutra! read Oil the euidenC/!. The authors were the experts, bIll they wrote (eel the moral dilemmas, think with Ollr story instead o{ abont it, jail!
(IS if they were (lilOIl)'IIJOtlS. Becm/se it WdSll't i,npor~allt luho gathered the actil/ely ill the decision poillts that define all mttoetlJJlographic project, and
euidellce or who judged alul weighed it, ha/l(fbooh wntas followed the COll- consider hOlU their OW!! liues call be made a story worth telling.
uentiullS of using a passiue ucice that erases subjectiuit)' and personal \X!hen An scops reading, I sa)', "\X!el!, dDt's clever. Although you started
vi
ta Iitv.
,1 CCD/lJ I
with the'I,' you quickly fell inw using the handbookgenrc to argue against
. A/ter [ ear;leel Ph.D., [ became increasingly circffJl1spect aDolft the
Ill)' writing in the handbook genre." I can't stop laughing. "Rcminds me of
, / I ' I II "/1970' nlW hm\-' so many of our textS argue in postITlodern abstract jargon for greater
lJOssibilities and lilllitations o{ i Je JllllWJ1 sC/eHces.. Jt Je 11,:!L -)~. ,." ~'.o-_,
of i/ly colleagllUs, lulJO /['as teaching a gr[~dlla~e seminar l!Je .~/.. r;to~ II- u~ 0::, accessibility and expCrill1CnLl1 forms."
, c , " SI/(fOcs,tl,d tl1at 1 <;tltd)1 tbe fJrolUlll~ Ittcnlture Oil the ClIS/S o{ COli "See how pmverful the conventions 3re?" An agrees, now chuckling as
SClCll C, ~,::, ,,- ." - C> -f
C _ " ,

lidellce" ill social science. [begml by reading I'-/lim (1962), who shmueL
that the lmildjllg-bloch model o{ science lac/::.eL! fOlflld.ltiolls,o thell Rorly
(/982), TOlflmil~ (1969), aile! (,ther philosophers luho illlfstrated how the
"facts" scientists see are illextrialbly COllllected to the uocablih~ry~hey us~
to cxpress or represent them; L:mtard (1984) deblm/::.ed the behe(1Il (J It!ll~
ficd totality of hi/ow/edge, i]uestiOlling whether master llarralwes :ucn:
either possible or desirable; POSj~stntctlll"alist alld decollstn/ctionist /U!"11_erS ,
/ ,p ,I (1977) Derrilh/ (1978 1981) alld FOllcault (1970), elfee'
sue 7 a::; J([/ r Jes ' , . ' 0' I ! " holU lUG
lfUely obliterated the moderlllst conceptIOn o{ tlJe aut Jar, a terl1lg
understand the connections LlllthcJ/; text, and :eaders,o ltlld~ll;::~
mTlollg
in{lllence of Bah!Jtin (1981.), the interlJretiue space auadable to the le~ _
WIS bro'ldened encoul"ap;im!, IImltiple perspectiues, lillscttled lI1eml11:g:.,
I I- I- , L L 0
lumlvoices, and local and illeglt/l1ldte ImolUledges tho.t tmllsgJoess 1..1'::'(1
P - r I tI 'lstS suC J as
the c1aillls o{a unitary body c:/theo ry ; {e/lll/11st C: lt1Cl-l Jeol.,
well. Then he adds more seriously, "Let's just vvrite to Norman and
Yvonna and bow our. Think of the time we could spend on the beClch
instead. Get some immediate gratification for a change."
"It's tempting, especially given how we've been fceling lately that our
life is too dominated by our work-but no way. Not ;lFtcr I've alrcady ago-
nized over writing the section on 'whar is autoethnography?' You know
how I resist doing this kind of writing. At the same time, I know it's impor-
tant," I reply, as I hear a knock at my door.
. "I don't think rJl bc satisfied, nor will you, unless \ve find a way to
transgress the conventions. \X1hat if we were to create a story that would
work within the handbook genre bur also outside it, showing what we do
a Illlst , " as we tell about it? That could be fun/' Art suggests playfully.
"How delightfully paradoxical," I 8l.id in a mischievous tone. "But we'd
O

1990, have to be careful not to give the impression that we're being oppositional
Cloligh (1994), Hardillg (19i1), Hortsoch (19IU), alid 811llth ( ,',
, i '1'/ dp' t all {partlot/Olf and advocating tbar everyone should write the way we do," I warn, ex-
1992) 11mlnoted the lftllqlle (lJIL tnargl!1L1 lzeL stan om 5 I
~ressing something we hear often and try hard to dispel. Then I get back
L 0"

, . -' ~ 1 t /chas TiIllJ

ties of /uomell,o and standpomt bOlfl1Cl(lry-clOs~lllgtextlhl IS 5 Sl .~ ,
" dB/ (199" '19%) op'lied 0111 ,VIS 1Ilto the irony of Art's idea. "\\1ol1'r our critics love it-yOll know, the ones
(1989,1992), A../lzaldlta (19.8 1 ), an e w,... ..J, - - ) . (; ""eie; oj"
o already accuse us of being irreverent, self-absorbed, sentimental, and
aile! em's to the necessity oj exposmg how the complex COlllli/gc_ ... Jf
' / / .,_Ot u'Qt'CIl infO ~l"cl"l1"f[(
-Ice c{aS5 SCXft(i!if)' efi;;.11!
t / (tl'J <lfh r:t IfU .... ' ) .lIe .
0' -"

'" 1 J - ., ".

" , f rel7ex/Vei Before An can answer, I tell him I have to go. "Someone's ar the door.
cO/ierete, persoualliued expenences, clwmplOnll1g the calise 0
in," I say in one breath, in response to the third knock.
experimental, illftobiographicizl, and uulnerable texts.

202 203
 !V\C I nuu,) vr ,-ULLI::\... \ 1\'1>..:1 n'~L.J n " n ..... ' ,-",,-, ....... " " .... n .......-" ... ', .. , ... ~
..tJ- Introduction to Autoethnography
A \vornan in her mid-40s opens the door and hesitates in the entry way. A
lJrge-brimmed, floppy straW hat covered with purple bangles hides her
fac~. A matching scarf hangs loosely around her neck. "Professor Ellis?" I
nod. ;'r.l1y name is Sylvia Smith. I'm a Ph.D. student in the Psychology
Department. ['m planning to do 111Y dissertation on breast cancer, and your
name was given to me as a so..-:ial scientist interested in research on illness.
I'd like you to be on my dis3crtation committee. Three members of my
committee are fr0111 the Psydwlogy Department and the (ourch is a re-
search oncologist."
"Hold it," I say, my hands extended in (ront of me to slow down her
monologue. "Back up. [-{ave a scat and Ict's talk about your proiect."
Sylvia removes her scarf lncl hat with a s\vecping crisscross motion of
borl~ hands and continues sneaking as rapidly as before. "I want to inter-
view breast cancer survivo::s to understand how they're adiusting after
cancer. 1 hope to combine qualitative and quantitative approaches. Send
out a survey and then interview ... oh, maybe 30 women and include Afri-
can Amcri~ans and tesbians, older and young women, professional and
working-class women. 'Ihn W3Y I can generalize ..."
"I-Iow'd you get interested in this ropic?" I interrupt.
"Welt, uh," she says, now slowing down and looking at me quizzically,
"{'ve had breast cancer." TLlen, going back to her rapid-fire, assertive style,
"'But I won't let that bias mv research. You can count on that."
"Of course you will," I s~y, and she immediately assumes a downcast,
defeated posture, before I add, "as you should."
"\Vhat do you mean?" 3he asks, looking straight at me with penetrating
eyes. "1 thou~ht I had to keep my personal experience out of my research.
If I Want my study to be valid, I can't mention to my participants that I've
had cancer, can I?"
"Hold that question," I say again, and move my chair closer to hers.
"\\!olllcl yOll be willing to tell me a little about your breast cancer first? It'U
help me understand more about your academic interest in the topic. Are
you okay ralking about your own experience?" . .
"Of course," she resp'Jnds, "but I didn't think anybody at the U11lverslry
would be interested in my personal experience." She breathes deeply and
slowly begins her story about the lump she discovered 7 years befO[~'
her mastectomy, and follow-up C1lcmot Ilerapy. Tl1en,''A n d'1t' s had
~
a big'-'
204
rtL'CVC""'V!d'UIJ"Y, rt;:I:>UIIUI ''1UIIULlV!:.', KeneXJVrry
impact on my family, especially my re\atiollship with my daughter, and
how I see myself- ... ," she says, her voice trailing orf.
"How has it impacted your relationship wir.-h your daughter?" I ask
quietly.
"She has to worry about getting cancer as well now. You know, the
genetic link, and we seem to have trouble talking openly about the risks
and about our feelings."
Sylvia continues to talk about her daughter, and after a while, I ask,
"And your self-image?"
"'I could write a book about that," she says, shaking her head back and
fonh. "\'ou know, I'm a therapist. I thought 1 could deal with it all. But it's
hard to feel like a whole person. I dOll 't mean because I lost a breast. Good
ridd~U1ce, 1 say to that. They were always toO big anyway. I had breast
reduction on the other one when 1 had reconstruction.lt's JUSt ... well ...
my life has changed so drasticaUy, except the day-tn-day, well actmllly
that's not all that different ... "
She becomes animated as she tells her story. Sensing that she is comfort-
able and desires to keep going, I continue asking questions. Her story
inspires thoughts about myself. How would I feel if I had a breast re-
moved? As she talks, I glance at her small breasts, then casually glide my
hands across my own large ones. lean't imagine their not being there.
\Xlou ldn't I feel incomplete, desexualized? Did she really feel "good rid-
dance" or is that a cover?
'" ... And the h3ir," I hear her say through my thoughts. "Just look at my
fuz?. It never really grev,' back like before. Shaving it was the most difficult
yet exhilarating thing I've ever done." The thin, inch-long brown and gray
strands don't move as she casually tosses her head from side to side. l\1y
fingers reach for my fine-textured, shoulder~length brown hair-I'd feel
llaked without it. I even resist pulling my hair back from my face. I wonder
why she curs hers so shorr now, as if she's dravving attention to having had
cancer. But what about the hat and scarf? Docs she use them in case she
Wants to "pass"? 1 wonder.
Sylvia and I arc about the same age. This could happen to me. No, it
couldn't. I get an annual mammogram.
"... I'd had a mammogram just a few months before I found the lump,"
he r VOice
'.'Intrudes into my thoughts.
But I do self-examinations every month, I argue back fr0111 inside my
head.
205
 lYle I nuu.;> u r \..Vl..l..C\..III'1..J H''I../ H'',....'-' "-"'''''' '-,V<, ""~,.~
"I found it during my monthly self-exam," she continues, shaking the
false predictability of my world. I listen intently, understanding that Sylvia
has a lot to teach me,
"Anyway, I'm interested in orher women's experience," she says, adding
hesitantly, "you know, how it cum pares ro mine. That's not something I've
admitted before, the personal part, I mean."
I nod. \Xfhat do I do now? I don't want ro wean another student off the
science model and deal v.,rith a ~;cience-oriented committee. And I'm wary
of getting involved in anorhcr study tbat simplifies, categorizes, slices and
dices the illness process. Bur S'{lvia is a therapist and fonhcoming about
her fcelings and what happened ro her. IVlaybe her stlldy could cxplore the
feelings associated with breast cancer and be useful for other women. The
pain on Sylvia's face, in spite of the casualness of her words, also makes me
think that this study might be a useful exploration for her. And I know it
could be a valuable experience for me as well. Bur what am I getting into?
"Do you have any idea what I do?" I ask.
"J ustthat you stu~ly illness and do qualitative work. Nobody does quali-
tative research in illy departmcnt. Bur I've taken a qU'-llitative course in
edllCation and I think I could get my committec to accept grounded theory
for my dissertation research."
"I don't use grounded theory lTluch anymore," I say. "!V'Iost of what I do
is autoethnography."
"\X1hat's that?" she asks, writing the word dlltoethnography on her
notepad as she looks at me,
"'I start with my personal life. I pay attention to my physical feelings,
thoughts, and emotions. I use v/hat I call systematic sociological introspec-
tion and emotional recall to try to understand an expericnce I've lived
through. Then I write my experience 3S a story. By exploring a particular
life, I hope to understand a way of life, as Reed-Danahay says."
"\Who?" she asks, pen poised in the air.
"Reed-Danahay, an anrhrcpologist who wrote a book on auroeth-
nography."
"How do I get a copy?"
"Don't worry 3bom that yet. There's plenty of timc to rcad about
autoethnography. I \vant you ro eXjJerience auroethnography firsL" ..
I L5uore SrlYi;,~'" conf<-l"c.J.I~:,ok~"'-" I diD t:hruuDI, '''r {ile caL-,:,.,eL "bn 11" r
understand you correctly, the E;oa 1"IS to use your I"I fe expenence " to gen (;0[01-
ize to a larger group or culture," Sylvia speaks to my back.
206
"",,-
"i:es, but that's not aIL The goal is also ro emer and documcnt the
moment-to-moment, concrete derails of a life. Thar's an important way or
knowing as well."
"So, you just \vrite about your lifc? That doesn't sound too difficult,"
Sylvia says casually.
I turn around, stare at her for a moment, as though 1'][ get a sign as to
whether r should promote <lmoethnography to Sylvia. When no sign is
forthcoming, I say, "Oil, it's amazingly difficult. It's certainly not some-
thing that most people can do well. lvlost social scientists don't write well
enough to carry ir off. Or they're not sufficiently introspective about rheir
feelings or motives, or the cOlHradictions they experience. Ironically,
many aren't observant enough of the world around them. Thc se!f-
questioning auroethnography demands is extremely difficult. So is con-
fronting things about yourself that are less than flattering. Believe me,
honest autoethnographic explOL:nion gencr,-Ites a lot of fears and doubrs-
and emotional pain. JUSt when yOll think you can'l stand the pain any~
more, well, that's when the real work has only begun. Then there's the vul-
nerability of revealing yourself, not being able to take back what vOll've
written or having :emy comrol over bow readers interpret it. It's hard not to
feel YOllr life is being critiqued as well as your work. It can be humiliating.
And the ethical issues," I warn, "jUSt \vait until you're writing about family
members and loved ones who are pan of your story."
Sylvia balds on to her chair, hcr eyes wide. I smile and let our the breath
I've been holding. "I'm sorry. I get really passionate abom all tbis," I sav
more gently. "Of course, there are rewards too-for example, yOll come t~
understand yourself in deeper ways. And witb understanding yourself
Comes understanding others. Autocthnography provides an avenue for
doing something mcaningful for yoursclf and the world ... "
"Ah, here thcy are," I interrupt myself as I pull two stapled papers from
my autoethnography file. "Thc one on top is 'Survivors,' a paper I wrote
about my brother's death. The other one's a chapter from Butler and
Rosenblum's book Cancer in DUO Voices, a co-consrructed narrative about
a woman with breast cancer and her lesbian lover who takes care of her."
"Co-constructed?"
"W'II
. e . talk about rhM later. For now, just sec how you respond to lbl:'~c
r lhill!>. thul after you"ve read llJeHl, what rve been saying will be
c. If yOll ' re sn"II "lnterestecI r1len, 1cave me a note and I'll matI" vou
dearor
Some orher materials." .
207
 Mt I MULJ::J UI- LULLt:L 11I'1,=, ~\I'-Il.J f-\.I'IALI t..1I'l1.,;l CJVlrlrl.I\...KL IVIM' LrI.'ML.J
"One more thing," I add, pointing to the syllabi on my desk. "I'll want
to meet often and you'll have to read the assignments from my classes on
'i Il ness narratives' and 'communicating emotion.' Also, I want you to meet
\\lith An Bochner, my coauthor, who teaches courses on narrative and, by
the way, ;::liso happens to be my husband." Her down-turned mouth
changes to a smile for a brief moment, until I add, "These are minimum
requirements if I'm going to be: on your committee."
"Oh, my. I don't knovv if I'll have time, given my program," she says. "1
still have to take 'Tests and IVl~asuremcnt' and <Advanced Experimental
Resemch Design.' I hope to finish my course work during this coming fall,
and then rake my prelims in early spring and finish my proposal by the
beginning of next summer."
1 shrug my shoulders as I stand and open the door. .LviI' exuberance, the
\varnings, all rhe requirements--any of these could scare her off. Oh, \vell,
better if it happens nO\v than bter. Suspecting this will be the last I see of
her, I'm glad I've given her an easy way au[. Sylvia winds her scarf around
her neck, throws her hat along with the papers I've given her into her large
open bag, says good-bye, and quickly scurries from view.
T\\lO days bter, I arrive at school and find a faxed message from Sylvia.
Dear Professor Ellis:
This is some ofthe lJIost powerful writing I've ever read. I idelltified with
)'ollr grief over losing your brother so sllddenly. You remilIded me of holU I
{elt n-helI1 found Ollt I had can(el; So did Butler and Rosenblum. I recall ex-
periencing that kind o{ turlJloir confusioJl, aud meanillglesslless. This work
violales everything I'ue been taught aboltt social science research, but 1'111
{clsclnated and wallt to !mow more. Will you mail s01l1e materials to help
clarify the origillS alld practices ofdutoetlmograjJhy? Maybe some "-illd ofa
literature review would s1lffice. While you're at it, do YOti mind il1cludil/g (/
lew lIlore i.wloetlJllogrllfJhies?
I smile and pull alit articles from my auto ethnography file. Jag o ,
Kiesinger, Kolker, Ronai, Tillmann-Hcaly-that ought to do it-and a
seerion on defining autocthnography I have just written as part of a chap-
[cr [or Dcnz.in and Lincoln's ,;L:conJ L:Jitiun uf the I-Iandbook of Q"dJ-
itatiue Research. I pause to read the draft, which is titled "\'\7hat Is Auto-
ethnography?"
208
<P What Is Autoethnograph y ?
Alftoet!mography is an autobiographical genre of writing ami research that
distJ!ays multiple layers of consciousness, cOllnectillg the pers(JIlal to the
cllltural. Back and forth aIftoetlmographers gaze, first through an eth-
nographic ruide-angle lells, focusing outward 011 social and cultural aspects
of their personal experience; then, they loo/? illluard, exposing a vulnerable
sell that is moued by and may 1l1Oue through, refract, alld resist cultura!
inlerpretations (see Decl;:., 1990; Neumann, 1996; Reed-Danahay, 1997).
As they Z00111 bad':.luard and forward, illluard ami olltward, distinctions be-
lIucel/ the personal and cultl/ral become blurred, sometimes beyond distillct
recognition. Usually written ill lirst-persoll uoice, alltoetlJllographic texts
appear in a variet)' of forms-short stories, poetry, jictiOll, nouels, photo-
graphic essays, personal essays, journals, (ragnwllied and layered writing,
alld social science prose. In these texts, call crete actioll, dialogue, ernotion,
embodilllent, spiritliality, and self-consciousness are fealured, appearing as
relL1tiollal alld illslitltticmal stories affected by history, social structure, and
cliiture, luhiclJ themse/ues are dialectically reuealed through action, (eeling,
thought, and language.
The term (1litoetlmography has been in circulation for at least two
decades. Although anthropologist Karl Heider referred in 1975 to the
Dani's own account ofwhat people do as alltoetlmography, David Hayano
(1979) usually is credited as the originator ofthe term. Hayano limited the
term to cultural-leur I studies by anthropologists of their "own people," ill
which the researcher is a (//1/ insider by uirtlle of being "native," acquiring
[/// illtilllate famil iari!)! with the group, or achieving (1l11 lnembership in the
group being studied (p. 100).
Like llwny terms used by social scientists, the meanings and applications
o( olftoellmography haue euolued in a Jlli1nller that makes precise defini-
tiOll and application difficlilt. It seems appropriate now to incllide under
the broad mbric of alltoetlmography those studies that have beell referred
to by other similarly situated ternIs, such as personal narratives (Personal
Narratiues Group, 1989), narratives of the self (Richardson, 1994b)., per-
S?,wl experience narratives (Denzill, 1989), self-stories (Denzin, 1989),
flrst-perSOIl accounts (Ellis, 1998a), personal essays (Kriegel; 1991), etlmo-
g,..l!,INC "'!:l(..n t :storir.::s (Ellis, 1995d), wririflS-jtorit:s (l'i..iCfJardSOIl; 1';1')7))'
cOl1lplete-member research (Adler & Adlel; 1987), auto-obseruation
(Adler G~ Adlel; 1994), opportuJlistic research (RiemC/; 1977), personal
209
 NIt: I nvl.}.) VI"" ..... VLLI:: ..... ' "'!\.:l ...... "H-' M'~M'-' .:..,'~W
ethnography (Cntn{ord, 1996), literary tales (Van NIaancll, 1988), liued
experience (Vall }.f[aaJlell, 1990), critical autobiography (Church, 1995),
self-ethl/ography (VitJl Maallell, 1995"), radical e111piricism (jacholl, 1989),
socioL1utobiogrllphy (Zola, 19 S2), illlto/Jathography (Hawkins, 1993),
euocative narratiues (Eodmel; Ellis, & Tillma1l11-Healy, 1997), persollLzl
writillg (DeValilt, 1997), reflexh'e ethllography (Ellis & Boc/mel; 1996a),
cOllfessiollal tales (Vall Maanell, 1988), ethnographic memoir (Tedlocl(,
1991), etlmobiogmphy (Leje1lIle, 1989), a1ltobiology (Pay1le, 1996), col-
laboratiue autobiography (Goldman, .1993), ethnographic autobiography
(Brandes, 1982), emotionalism (Gllbrium & I-Iolstein, 1997), experiential
texts (f)enzin, 1997), narratiue ethnography (Abu-Llfghod, 1993), auto-
biographical ethnography (Reed-Danahay, 1997), ethnographic poetics
(A'larclfs 6' Fischel~ 1986), natiue ethuography (Olmlthi-Tienzey, 1984),
indigenolfs ethnography (Gonzalez & Krizeh, 1994), and ethnic alftobiog~
rafllry (Reed-Dallahay, 1997). Nevertheless, social scientists often discuss
autocthnography as a subtype ofsome other forms, such as impressionistic
aCCOlfJits (\ZlIl ]'vlaiTllell, 1988), narrative ethnography (Tedloc/(, 1991),
interpretiue biography (Denzin, 1989), new or experimental ethnography
(Ellis & Boch"el; 1996b), socioe'oetics (Ellis & Bochllel; 1996,,), or post-
ilIoderJl ethnography (Tylel~ 1986).
Various methodological strategies haue hee1t deueloped ill c01lJ1ectiou
luith LllltoetlmogralJhic projects, although they may be applied to other
forms ofqltalit.atiue research as well. These include systematic sociological
introspection (Ellis, 1991b), biogra!Jhical method (Denzin, 1989), per~
sonal experience methods (Clondinill 6' Connelly, 1994), femillist Ineth-
ods (Reinharz, 1992), experient.ial alwlysis (Reinharz, 1979)J narrative
inquiry (Bodme1; ~l994), COll5ciou511ess-raising methods (Hollway, 1989),
co-constructed llarratiue (Bochner & Ellis, 1992), and interactiue inter-
viewing (Ellis, KiesingeI; 0~ Tilllllallll-I-Iealy, 1997). In some disciplines,
terms endemic to a particular fieM have evolved sllch as in sociology, pcr-
sO/wI sociology (Higgins C,.., ]olmsoll, 1988), autobiographical sociology
(Friedman, 1990)J sociological autobiography (lvlertoll, 1972/1988), pri~
vote sociology (Shostak, 1996), alld emotiol/al sociology (Ellis, 1991a); ill
onthro{JOlogy, t.mLlJropological 31lfobiography (Brandes, 1982), natiue
anthropology (Narayan, 1993), 1Ildigel1olts anthrolJology (Yedloch, 1991),
LtutoalltlJrojJoloJ;:v (Strathem, 1987).. .,elf-com;cious flnthrnjJO/ngy (Cohen,
1991), anthropology of the self (Kondo, 1990), anthropology at home
(jackson, 1987), anthropological poetics (Brady, 1991), and autoetlmol-
ogy (Lejeune, 1.989).; alld ill COll.'l1Ittllication, rhetorical autoetlmography
210
'-IVI! "",-,",,,,, IT,,",, .... ,,,.-. ....-' rUHVCU" ,vy' up"y, ,- <::, O>U"U' '"UIIU~IVt", !\'C'IIt"AIVlty
(French, 1998), per(ormi11lCe autobiography (Miller & Taylor, 1997), and
aJftoetlmographic performance (Park-FIlllel; 1998). Increasingly, hmueuel;
autoetlmography has become the term of choice ill describing studies and
procedures that COIl11ect the personal to the cultural, frequently appearing
in titles of boo/(s, theses, sections of boob, articles, special issues of jour-
nals, dlnl booh series (for example, Cloligh, 1997; Dech, 1990; Ellis, 1997,
1998a; Ellis & BoclmC/; 1996a; Gravel, 1997; Herl/(ion 1993; Li01l11et, J
1989; Pratt, 1994; Reed-Dmwhay, 1997; Trotter, 1992).
Autoetlmographers vary in their emphasis 011 the research process
(graphy), Oil culture (etlmos), and on self (an to) (see Reed-Danahay, 1997,
p. 2). Different exel1lplars o(alft.oethnography fall at different places along
the cOllti11U1I11I of each of these three axes. Researchers disagree all the
boundaries o( each category and OJ[ the precise defInitions of the types of
a1/toetlmography. Indeed, tlWIlY writers move bacl( and forth umo7lgterms
and meanings euen in the same articles. Recogl1izing this li11litatioll, I wi/!
mellficIIl, for heuristic p1ll1JOses, a few widely used expressiolls that prouide
a sense of the range of approaches associated with flutoetlmography.
Although reflexive etlmographies primarily foclls Oll a clliture or subcul-
ture, anthors lise their owu experiences in the cll/ture reflexively to bend
bac"- Oil self and look more deeply at self-other interactions. In native
ethnographies, researchers loho are natiues of cultures that haue been
marginalized or exoticized by others write about alld intelpret their OlOll
cllltures (or others. hl texts by "complete-member researchers," resea1"ch~
ers eX1Jlore groups of which they already are members or ill luhich, during
the research process, they haue become full members with complete identi-
(ication and acceptance. In pcrs01wlllarratiues social scientists talw on
J
the dual identities ofacademic and persollal selves to tell autobiographical
stories abollt sOlne aspect of tIJeir experience ill daily life. In literary auto-
etlmograjJbies, an author's prirnary identificatio1l is as all autobiographi-
J cal writer mthcr than a social scientist and tbe text focuses as much 011
J
examining a selfautobiographically as on illterpreting a culture for a 1I0n-
Hatiue audience (see Deck, 1990).
III reflexiue etlmograpbies the researcher's personal experience be-
J
comes important primarily in holU it illuminates the culture under study.
Reflexiue ethllogmplJies range along a cOl1til11wm fro111 starting research
{I'OIi! one's otU/l experience to etln/ographies where the researcher's experi-
ence is actually studied a/aug with other participants, to confessional tales
where the researcher's experiences of doing the study become the foclfs of
t/l uestigatiou.
211

Fetnillism has contributed significantly to legitimating the autobio-
graphical [loiee associated wi!/J reflexive ethnography ((or example, Dehm;
1996; Behar & Gordon, 1995; Kriegel; 1991, 1996; Personal Narratives
Group, 1989; Richardson, 1~J97). iVTall)' feminist writers have advocated
slarting research from one's O~UJ1 experience (e.g., Smith, 1979). Tlms, to a
rzrcafer or lesser extent, researchers incorporate their personal experiences
:;nd standpoints i11 their research by startiug Iuith a story abollt thel1l-
selues, explaining their personal COlll/cetion to the project, or by lIsing per-
sonal IVlOwledge to help them in the research process ((or examples, see
lOHcs, 1998; Lindell, 1992: for a sumJlh1ry of reflexiue stltdies, see
Reillharz, 1992, pp. 2S8-263).
Jackson (1989) Hses the term radical empiricism to refer to a process
that i/leil/des the ethllographer"'s eX!Jeriences and interaction with other
participants as uital parts of what is being studied. Reflexive etlJlfographers
ideal h' use all their sellses, their bodies, movement, feeling, alld their whole
beinoo --thel'_ use the "sel!" to learn about the other (Cohen, 1991; Jack-
SOil, 1989; O/::.elv, 1991; 'I,trller 0" Bmllel; 1986). Particularly controuer-
sial is the notion'of the role o(sextlality illlearl1ing about the other (Kulich
0:.0 \\lillson, 1995: Lewin & Leat}, 1996).
In slIllllnarizing reflexive ethnography and tracing its history thor-
oughly, Tedloeli'. (1991) distinguishes between ethnographic memoir (also
called confessional tales by \1(,)1 lvfaanell, 1. 98 S), in which the etlJllograpl}(!I;
who is the focus of the story, tells a personal tale of what luellt 0/1 ill the
backstage oldoing research, c/ld /larratiue ethnography, where the etlmog-
mpher's experiences are incorporated into the ethnographic description
alld analysis oj" others and the emphasis is on the "ethnographic dialoglle or
encounter" between the llarrJtor and members of the f:.,rroup being studied
(tJ. 78). The etlmographic lnemoir is rooted historically in the persollal dia-
ries and journals IWjJt by hldillolUs!{i (1967). Standing 011 his sholtlders,
many ethnographers who fol/owed wrote confessional tales aboltt their re-
search in volumes separate (rom their research documents (e.g., DUll/ont,
1976, 1978,. Rabillow, J975, 1977); some wrote wlder penllames in order
to [woid losing academic credibility (e.g., BOluen, 1954). The development
of this hilld of reflexillc turit irrg is connected, according to Tedlock (1991),
to a shift i11 the 19705 (rom a11 emphasis all participant obseruatioll to t~Je
l'o}'SLorlNtf;on of l>"lrl;'~;l,.71;o"JJ,,,,,,J 10 ."" <,?np},cls;S on fl,c prOC'L'S_~ of !/!nt~
ing. This shift tuas inspired by the epistemological doubt associated wit/; the
crisis ofrepresentation and the changing cornposition ofthose who become
ethnographers, with more tUGmen, lower-class, ethnic alId racidl groUpS,
212
"~''"'''''''''''''=J,.... ,...""" .... ,-'v"u, "U"U"''=', '~'="'t"A'Vlty
and Third and FOIlrth \Forld scholars /lOW represented (Bochner &, Ellis,
199')).
This changing composition also is associated with COllcerns aboIlt power
and praxis ami with more ethnographers turiting about their own people.
Native etlmography, for example, is written by researchers from the Third
and Fourth VI/orlds who share a history of colonialism or economic sub-
ordination, including subjugation by ethnographers who have made them
subjects of their work. Now as bicIlltIlral insiders/outsiders, llatiue etlmog-
raphers construct their own cultural stories (oftenlocffsing Oil their own
i1l/tobiogralJhies; for example, see Kinkaid, 1988; Rodriguez, 1983), raise
seriolls questions abollt the interpretations ofothers who write about them,
and use their dual positionalit}' to problematize the distinction between ob-
sauer and observed, insider and oI/tsider (see, for example, ldotza(i-HalIcl;
1997; 1hnh, 1989; (or more detailed discussions, see Neltl1W11lI, 1996;
Reed-Danahay, 1997; Tedlocl:., 1991.).
Complete-member researchers is a term coined by Adler and Adler
(1987) to refer to researchers who are (ully cOlJImitted to and imlJIersed i11
the groups they study. During the research process, the "collvert" researcher
identifies with the group and "becomes the phenomenon" (ldehan &
\Y/ood, 1.975) being studied. For example, JIIles-Rosette (1975) becmne a
baptized true belieuer in the Aj"rican Apostolic chIlrch she studied. The
""opportunistic" researcher (R.icmel; 1977,. sometimes called an indigenous
researcher in anthropology-see Tedlock, 1991) studies settings ol which
he or she is already a member (slich as Haymzo's 1982 study of poker or
Krieger's 1983 stItdy o( a lesbian comm1l71ity).
In contrast to complete-mernber research, where the emphasis is on the
research process and the group being studied, social scientists recently haue
begUJI to view thelliselues as the phenomenon and to write cuocatille per~
SOllalllarmtives specifically j"ocllsed on their academic as well as their per-
s~lIalli[)cs. 'Their primary purpose is to linderstand a sellor some aspect ofa
"fe lived in a cultItral context. In personal narrative texts, authors become
"I," readers become "YOH," subjects become 'us." R1rticipants are ellcoItr-
aged to participate ill {/ personal relationship with the allthor!researchCl; to
be treated as coresearchers, to share mtthoritv, and to allthor their OWII liues
in
.their
u0 n / '
UOlces. l'",cac1ers, too, ta k e a more . m 1e as t 1Jey are mUlled
- flctwe ..
lJ1to fl,,, alftl~or':s u_'urld, evuked to (/ feelillg fcuel about the euellts beillfj
described, and stimulated to use what they learn there to rellect all, 1l1ldel~
Sialld' alld co pe lU/t). 1 t IJetr
' oum I'lves. T i l 'IS to /Unte
)c goa . meanlJ1gtidly
. r mul
eVOcatively abol1t topics that matter and lllay ma/:w a difference, to illclude
213
 1',/\(.1 nULl..) v , \...VI..I..L.\...II'~1.) M'~"" M'~MI..' "-"''-' ....... "~''-'''''''' ",,...,, ...
sensory and enlOtional experience (Shelton, 1995), and to write from dl1
ethic 0/ care alld concern (Denziu, 1997; Noddillgs, 1984; Riclwrdsoll,
1997).
Literary and cultural critics ofteH join social scientists i/1 ell'Iployi1lg the
terill al1toetlmogmphy il1 refere,llce to autobiographies that self-con-
scioHsly explore the interplay of the introspective, personally e1lgaged self
u,'itfJ clIltural descriptions Inecnated through language, history, ami
ethnographic explanation (see DEch-, 1990; Lio1111et, 1989; Pratt, 1994).
For example, LiollllCt (1989) ad Deck (1990) both label alld explore
Hurston's (.1942/1991) memoirs as autoet/mography, il1 which the tradi-
tiol/al historical /rmne and specific: dates and events associated with autobi-
ogral_ilry arc minimized and the attempt to demonstrate the lived experi-
ence {//1(1 Iwmallity of authors c,:nd their peoples to olltside audiellces is
mLlximized. As Hurston (1942Il991) explains about the (011<.. S01lgs she
gathered ill lJerown research, "The words do /lot C01/11t. ... The tfllle is the
ilnity of the thing" (p. 144). Deck compares literary alltoetlmogmplJies to
self-reflexiLie (ieldworl<.. accot/ilts, sllch as Shostah's Nisa: The Life and
\YJords of a !Kung Woman (198:1) and Crapanzano's Tuhami: Portrait of a
Ivloroccan (1980), ill wfJiclJ the aut.hors ground themselves in their (ield
cX/Jeriences, reference otlJer social scientists wlJO serve to validate the chal'-
deters ill their stories, keep the autobiographical components maillly in the
introductiolls and epilogues, L1}u.l (OCIIS personal revelations directly 011 the
(ieldworh. [/t hand rather than OIL their OWJI personal development.
,Social scientists [/Iso write iiterm'y and poetic etlmogralJhy. Dart Rose
(1991), (or example, distinguishes between his own personal poetry, whielJ
is not con/lected to his GllthrolJOJogy, and the poetry of other il1Ithropolo-
gists slIch as StL11Iley Diamond (1982), wfJich focuses 011 the etlmogmphic
experience of i.1Ilthropologists as obseruers. MallY anthropologists, sllch as
Edluard Sapir and Ruth Benedict, have published realist ethnography in
il1aitlstrearn i.111throlJOJogy jottmals and personal poetry in literary olltlets
(Bnlllf!l; 1993). Now Anthropology and Humanism publislJCS {ictioll and
poetry by a11thropologists. Iii sociology, Laurel Richardson, (or example,
has published essays inlitera,y (1995) and social science journals (199 6)
i.7nd poetry as et.lmograp(Jy (1994a).
Alftoetfmography, /Ult.ive ethnography, self-et(71lOgraphy, I1Ie1lloiJ; aHto~
biography, eVen fiction, have become blurred genres. 111 man')' cases,
whether a social science worh is called an autoetlmograplJ)' or an etl]/logra~
ph'v depends on the claims I;wde by those who write and those who write
i.1b~lIt the wor/;:. W/hetfJer a worh is called fiction or fact, autoethllography
214
",,...,l-l
or IIWl1lOiJ; is cO/l1Iated to II/riting praetices-socii.ll science illftoetlmog-
raphies uSlfally contain citcltions to otfJc1' ilci.1demics and JfSC all academic,
disciplinary vocabulary; publishing jJraclices-wlJO jJublishes the bool::.,
how it is promoted (for eXaJnjJlc, the field identified all the olftside couer)
d/IClldbeled (ISBN /lumber), and wfJO the targeted il1/dience is; and reuieul~
ing practices-who endorses it, wfJO reviews it, and who writes ab01lt it.
Literary critics treat sOl1le alftobiograjJIJies as L1lftoetlmogmphies and HOt
others; Hurston, who sees herself as essayist, i.1/lllJro/Jologist, and (ietioII
H/riter (Lion net, J989), prouides [/ good exam/Jle 0/ tlJe messiness amI ouer-
lap. Ivlainstrcal11 social science tends to classify ~1/!toetIJllogrt/phies (lor
cXdlnple, Ellis, 1995 b) Lwd life histories about {lcadelll ic coreas (lor exam-
ple, 13ergel~ 1990; Goetting & FellsterrllalwJ; 1995; Eilcy, 19S5') into the
genre o(lIlcrnoiroralltobiogra/Jhy (see Zussmall, 1996). Perhaps the loose
i.1pplicatioll 0/ the term atttoetfmography ollly signifies a greater tolerance
110W for the diuerse goals of ethnography and a better lIJ1derstam:lillg 0/ the
{.l11ibility alld il1determillac}' of lallguage LIlld cOllcepts.
Smiling at the social sciencc prose, I place the capy of the Halldbook
draft in the pack~1gc with the stories I'm sending Sylvi~1. I try to imagine
how she will take it in.
A weck later, Sylvia again appears at 111)' door. "Okay, 1read everything
yOll sent me. \YJaw, those personal narratives just blew me away. Your
3moethnography piece was interesting, but hard to get through. It'l! be
more helpful later, I'm sure," she reassures, then continues qUIckly, "but
now I'm very confused."
"Listen, I only have 5 minutes," 1 say. "I'm going to a department
colloquium."
"Oh, I'm sorry.I'l! come back another time," Sylvia responds, retreat-
ing through the doorway.
"No, wair. \'V'hat confuscs you?" I ask.
"\'{fell, in my methods classes I was taught that I had to protect against
tnl' own biases imcdering 'with my observations and that my research
should produce general knovvledge and theory. But the articles you ga\'e
ll1e emphasize concrete expressions over abstractions. So, I'm confused
abOUt what my objecrives would be if I do an autocthnography. Why
Would others be interested? How could I prove that what 1 have to say
~bout my experience is tfue? Auroethnography isn't feally social science,
IS it?"
215
 ""y'our timing is perfect. Come with me," I say, grabbing my keys and
\valking down the halL "I want you to hear somebody."
\YJe entcr a crowded roon~, v..rhere a talk is about to begin. "Thac's Art,
my partner, the good-looking guy sitting at the rabIe," I whisper to Sylvia,
as we take seats in rhe back.
"\X elcome to another session of our Interdisciplinary Colloquium
'
Series on Interprctive Research in the Social Sciences," says Jim Spiro, a
departmental colleague who organizes rhe talks. "Today's speaker is Art
Bochner, who reaches a Ph.D. seminar on 'narrative inquiry.' "
"\Ve have these talks ever:r week," I whisper to Sylvia. "They're pretty
informaL"
"An will present his remarks for about 15 minutes and then turn to
questions fr0111 the audiene::," Jim announces. "I've asked him to talk
about what some writers have called 'the narrative turn in the human sci-
ences' and r.o focus specifica1.ly on personal narratives. His talk is entitled
'\\llly Personal Narrative .Matters.' Please welcome Art Bochner."
Art st:1l10S with his right foot hooked behind his left leg, runs his fingers
through his hair, and begins.
<> Why Personal Narrative Matters
It's Ill}' pleasure to be here today and to have this opportunity to speak on a
topic abollt which I feel so p,lssiol1ately..As mall}' of yOll know, I was edu-
cated as d traditional empiricist and spent most ofthe first decade of 111y aca-
demic life plying the trade T had learned as a graduate student. In the late
1970s 1 began to feel uneasy aboHt the political, philosophical, ethical, and
ideological foundations of social science research (Boc/mel; 1981). hI my
chosen field, commIll/icatioll research, empiricism rested largely on the
premise that comml/llicatiolf between humans could be described as {/II
object. But l]lflrtan cOJnmImicatioll is lIOt all object, or a discipline sttldyil1g
objects. Communication is a process COllsisting ofsequences of i1lteractiol1s
alld the dynamic ImJrlatl activity of studying them. hloreove/; as coml/l1m/-
eating humalls stue/ying hllllHlllS comnnmicating, we are inside what we are
studying. The re/7exiuc qualdies of human COl1l1JHt1licatioll should /lot be
ur.-r,,-kc(cd "in thc ((0'1111"- v( "ci,,"r;cr::. '.' Tln_"y slJvuld be "<ccunutto,},,,t.ul ,'1J1d
integrated into research and its products.
Like manv other social scientists who tool? these mal-ters seriol/sly, my
conlidence i~l orthodox, so::;ial science methodology was shal?ell by the
216
critiques 0/ poststmctllralist, postlllodemist, alld felninist rvriters. [turned
to rwrratiue as a mode of inquiry because 1 was persuaded that social sci-
ence texts needed to construct a dilferent relationship between researchers
and subjects and between authors and readers. [wan led a 1/l0re fJersoHol,
collaborative, and interactive relationship, one that cel/tered on the ques-
tion ofhow human experience is endowed with meanillg and on the lItoral
and ethical choices we face as human beings who live in ({nUllcertail! and
changing world. I also wanted to Iwderstalld the conl/clltions that C071-
strain which stories we can tell and how we can tell them, ami to shOl/! how
people call and do resist the fa rills of social cOIlt.rol that marginalize or si-
lence cO//lltemarratiues, stories that deviate from or transgress the calloni-
cal ones. The texts produced under the ntbric of what 1 eall narrative
illquiry wOllld be stories that create the effect ofreality, sholl/ing characters
embedded in the complexities of lived moments of stmggle, resisting the
intrusions of chaos, discollflection, fraglltentatioll, marginalization, and
incoherence, trying to preserve or restore the continuity and coherence of
life's lInity in the face of Ill1expected blows of fate that call one's l1le~1Ilillgs
and ualues into question.
I refer to these pasollal stories as evocatiue narratives (Boc/mel; Ellis, &
Tillmawl-Healy, 1997, 1998). The word euocative contrasts the expressive
and dialogic goals of this work with the lnore traditional orientations of
mainstrea1/l, representatioJlal social sciellce. Usually the author ofall evo-
cative narrative writes in the first person, Ina!?ing herself the object of re-
search and thus breaching the conVellti01101 separation of researcher alld
subjects (Jackson, 1989); the story ofteH focuses all a single case and thus
breaches the traditio/wI concenzs of research from generalization across
cases to generalization within a case (Geertz, 1973); the mode ofstorytell-
ing is akin to the novel or biography and thus fractures the boundaries that
normally separate social science fromliteratttre; the accessibility and read-
ability of the text repositions the reader as a coparticip!1nt in dialogue and
thIls rejects the orthodox uiew of the reader as a passive receiuer of Iwowl-
edge; the disclosure of hidden details of private life highlights emotional
experience and thus challenges the rational actor lnode! of social pel-(or-
liIallce; the narrative text refuses the impulse to abstract and explain,
stressing the journey over the destination, and thilS ecli!Jses the scientific
or
;11",,,'0" ur
<;:ulTlrul nrnl nW:5cc:ry; (,(let trle cpt;jor.tic purtrayal {he dJU a/ll:l
(low of relatiOllshifJ experience dramatizes the rnotioll of connected lives
~cross the curve of time, and thus resists the standard practice of portray-
Illg social life dlld relationships as a sHapshot. Evocative stories activate
217

subjectiuity and cOllzpel elllOti.J1WI respollse. They long to be Hsed rather
thall mwlyzed; to be told amI tetold rather thall theotized and settled; to
offer lessolls for (ltrther conuersatioll rather thLIJl fl1ldebatable conclusions;
and to substitute the c01J1tJQllio'fship of intimate detail (or the lonefiness of
abstracted facts.
PersonLII writing ahill to euo::ative narmtiue has recently proliferated ilz
the mainstream press, iu llew jcurnalisl/f, ill creative l1onfiction, and iu the
genres of literary mel/lOll; autobiography, and autopathography (BlIford,
1996; Harringtoll, 1997; Hawhins, 1993; Parini, 1998; "Thte Confes-
sirms," 1996). /-\fl o{the lire writing genres (Tierney, 1.998; see also Chapter
9, Volume 2) seem to hLIue tllmed toward rnore intimate, !Jcrsonal, and self-
cOl/sciolls writillg. [thinh the moue ill the social sciellces tOluard less L1Ifony-
mCJltS, more personal writing pi/ral/els the sanze trend ill literature and jour-
nalism (Dellzin, 1997; Nellmanll, 1996). \\{lhatever the reasons, I see ample
evidence of L1 bIflgerJlfing interest i1mollg diverse fields of social scielfce in
the genres ofpersolfalllarratiue and alltoetlmogra!Jhy. Thc examples 1 have
in mind illclude the realft speCl~I! issues of such journals I1S journal of Con-
temporary Ethnography (Elfis .j" Boc/mel; 1.996b) aud Qualitative Sociol-
ogy (Gli1ssllel; 1997; Hertz_, 1996); the boo!::' series Ethnographic Altenw-
tiues, published by AltaA'1ira Press; the edited collections by anthropologists
(Benson, 1993; Brady, 1991; 01wly & Callaway, 1992), sociologists (Ellis
c," Flaherty, 1992; Hertz, 1997), alld educators (fierney & Lincoln, 1997);
the JJwny articles and 171onogI'LIphs published in academic journals slich as
Amcrican Anthropologist, Anth"opology and Humanism Quarterly, rellfi~
nist Studies, journal of Personal and Interpersonal Loss, Qualitative
Inquiry, Sociological Quarterly, Symbolic Interactioll, Text ami PeTlor~
iIlance Quarterly, Western ]ollrllal of COllllllunicatiol1, and Women's
Studies Illtenwtio1lal Forum.
By UhlY of exam!Jle, let me briefly mention three published euocatiue
narratives. Each highlights t.he t:ol1111umicative practices through which the
author's identity evolues, is di.-;playcd, and put to use (BrtlJlcl; 1990). In
particfflm; these writers illustrate how certail1metaphors and meanings are
narratiuized into their tiues. lvIlIlwia (1989) sholUs the lived experience of
"olforexia (rom luithin," expre5sing the ways i/1 which food aud starr:ation
LIre emplotted into her identity; lZorzai (1992) presents a layered stOTY ill
which she jJcr!orl11s her sitttatid, lI1ultiple selves, expressing IJer tortI/red
O/nbiualence in assuming tbe dual identities ofsocial science researcher and
erotic dancer; and Ellis (1993) navigates the emotional maze olshocl~ and
grief as she copes with conflicting academic and family personas in the
218
,"",V.I.,,,,",," IVy' Uj-Jlly, rt'I.:>UIIUi I 'fUIIUUV/;" ,,/;,Ij/;,x/Vny
afternwth o( her brother's sudden death ill all airplalfe crash. Each is a first-
person account, furiUeH ilS a stOlT, that expresses uivid details about the
author's OW1I experience. 'The "-research text" is the story, complete (but
opell) in itself, largely free of academic jargon and abstracted theory. The
allthors privilege stories over analysis, allOluhzg amI encouraging alter-
native readings and multiple intCl"/Jretatiol1s. They ask their readers to feel
the tmth of their stories and to become coparticipL11lts, engaging the story
line morally, cmotio/flIlly, aesthetically, and intellectually (Richardson,
1994b).
The question that I'm tlsffally aslwd is, "To what kind of truth do these
stories aspire?" Often this question is aslwd in a tone that expresses skep-
ticislll, doubt, and even hostility. Some critics (e.g., Mhlh, 1969-1970;
Shottel; 1987) l1Iglle that stories give life a structure it does not have and,
thus, stories ficlionalize life. Sillce the experiences all which narmtiues are
based ]}lay be uague and flllcertailZ, the stories they aro1lse ({Ill neuer be
determinale or complete (e.g. ShotteJ; 1987). Giuen the distortiolls of
lIfemory and tbe mcdiatioll of language, lfal'ratilJe is always il story dhoftt
the past ami 110t the past itself.
A second criticism is that persolfal narratiue reflects or adlJaJfces a
"romantic constructiolf of the self" (At1:z.insolf, 1997) flJUuorthy of being
cfassified as part of social science. If yOft are "a storyteller rather thall a
story analyst," argues At/~iJlsolf (1997, p. 335), thell YOllr goal becomes
therapeutic rather thall analytic. At1:z.ills011 believes that a text that acts as
alf agent olself-discouery or self-creation-precisely the narratiue chal-
lel1ge one faces when an expected life story is interrupted bv illlless, uio-
lence, or accidellt-ca11llOt. be an academic text. Presllmabl):, if you don't
subject narratiuc to sociological, cultural, or some other form of analysis,
treating stories as "-social facts," then yaH are not doing social science.
\\fhile passionately protesting the ways in which some writers wan! "to
priui/ege certain I:z.inds and occasions of narm/ive performance, Athinson H
(1997) aims to redeem (and privilege) the standard version of representa-
tirmal social science by triuiafizing 07" dismissing any Iuor!? t!Jat docs not, in
his words, "use narrative to achieve serious social analysis" (pp. 338-339).
Let me briefly address the reservations expressed by these critics. First,
there is the question of narrative tmth. \Vhat is the {lOillt of a storied life?
Na",.alivc truth seeks to l~eep the past aliue ill the present. Stories show 115
that the meanings and signi(icaJlce of the past are incomplete, tentative,
and revisable according to contingencies of aliI' preseJlt life circllllfstallCes,
the present from luhich we narrate. Doesn't this //lean that the stories we
219
 M t I MUU.) vr \...ULLC\...I II'lI.;l Al'H.J 1-\1 '<MI... 1 "-II~'-' ..... " ",,'-,.-, ... "''-'' ... ,,,, 'h~
'".-"-vcu" ,v':1' U/-" 'Y, re, ;:,u, lui I ~u, IUlIve, r-eIH;:,x!Vny

tell ahuays nm the rish. of distorting the past? Of course, it does. After all, what is experienced, lind what is desired (Freeman, 1993, 1998; Kerby,
stories rearrange, redescribe, inuent, omit, and revise. They call be wrong in 1991),
1111ll1erous ways-tone, detail, :;llbstallce, etc. Does this attribute of story- So the ,~z/Cstiol~ is not, "Does my story reflect Illy past accurately?" as if 1
telling threaten the project ofpersonal narrative? Not at all, because a story were hold11lg a mIrror to my past. Rather I must ash., "\v/hat are the conse-
is not a neutral attempt to mirror the facts of one's life; it does lIOt seek to quences Iny story produces? \Vhat hind of a person does it shape me into?
recoua already cOllstitl/ted meallhlgs. 0111y within the memoro-politics ~'?hat nCJ{/ possibilities does it introduce for Iil/ing my life?" The cmcial
SllITOlfJUlillg the accuracy of Tfcouered lnemories, which emelged within ISSlfes are what Juzrratiues do, w/1l1t consequences they /mue, to what uses
the context of positiuist psychology, would such a criticism be threatening they can be Pllt. These consequences often precede rather thall follolii the
(Haching, 1995), story because ~hey are enmeshed hz the act oftelling. "The story ofOllr lives
The tntt/; o(lforratiue is not ld:.in to correspondence with lJ1"ior meanings bec_om~s Ollr hues," writes Adrienne Rich (1978, p. 34). Thus /Jersollal nar-
assumed to be located in some sort of prellarrative experience. aile lUlrra- :.-,:twe JS part o( the human, existential struggle to mOlle life forward.
tiue illterprehltieJl! of cuel/ts can be judged against a11otheJ~ but there is 1/0 through the narrative actiuit)' ofself-creation we seef::. to become identical
stillnlard by which to lIIeasure ,my narrative agaillst the meaning of euents t~ the story we tell. Anais Nin underscores this desire {or self-created, uarra-
themselves, because the meanj;'zg of fJrenarrative experience is constillited Itu~ meaning when she t.11l1Z01IllCes, "[ could not liuc ill any of the worlds
iiI its IWITdtiue expressioll. Li(t' and rwrratiue are inextricably c01Jnected. offered to me. ... I belieue one lurites because one has to create a world in
Life both anticipates telling and draws meaning from it. Narrative is both which to liue" (quoted ill Oak.ley, 1984).
about liuillg and part of it. I get imp(.~tie}}t luith writers luho belittle ordimillish the thempeutic COIt-
I titled this little tall::. "T(!hy Personal Narrative lvIatters" to emphasize scquences of stories. They fcnd to draw a hard-mzd-fast distinction between
that zue liue luithill the tensions constituted by 01/1' memories ofthe past and therapy and social research, implying that narratives are useful ollly insofar
anticipations ofthe future. Personal 11arratiue, the project oftellillg a life, is as they {:c!.uance sociological, anthrofJOlogical, or psychological theory. For
a response to the human problenz of authorship, the desire to mahe sense these. cntlcs, J~arratiue threatens the whole project o{ science. They reply
and preserue coherence over the course of our lives. Our personal identities iII,lgnly, sholltmg the canol/ically giue11, professional respOllse: "Ifyolf call 't
seem largely contingent all how well we bridge the remembered past with Pzt.ch ': theory, then yOll Gan't play ill the big leagues." The most important
the {/llticipated flltllre to provide what Stephen Crites (1971) calls "a conti- tI:mg JS to be smart, dcuel; analytical; that's what it meallS to be academic.
ll1/ity ofexperience ouer time. "The /lan-ative challenge that we face as nar- \'{Ihat they oppose is what they equate with the thera!Jelltic: the senti-
rators is the desire for continuity, to malw sense of aliI' lives as a whole. "The lnel/,tal, the mushy, the populm: T/ms they engage surreptitious/v in what
femmist critic.!ane Tomphil/s (1989) calls "the trashing of eJJloti~n, " a war
present of things past and the present of things future," says Crites (1971),
"are the tension ofevery I1lOment ofexperience, both united in that prcsel/t
u:l-~ged ceas~lessly by academic intellectuals "against feeling, against
nomen, agamst what is personal" (fJ. 138).
and qualitatively differentiated by it" (p. 302). The lUorl(. o( self-1Jarratiol1
.4 text that (zmctieJ1ls as an agent ofself-discouery or self-creation, for the
is to produce this sense o( contilli/ity: to make a life that sometinzcs seems
author as well as (01' those who read and engage the text, is only threat-citing
to be falling apart come together agaiu, by retelling and restoryillg the ullder a narrow . def'1I1I.JOn
'1' a f SOCia ' { mqI/lrv,
" Olle t /Jot eschews a social sci-
euellts of one's life. Thus, narmtiuc matters to us because, as David Carr ence with a 1110' I - t' . ~ d' /. .. IV'I - I / '
.). I a t-cn.o dn l-l Jearl. w, Jy S JOll d carmg and empathy be
(1986) observes, ,ccoherence seems to be a Need i'mposed upon /IS whether secondaf)! to coo t' II' j { , - \"1/
- . If 10 mg mIL ;;.nowmg:' '\ J)' must academics be condi-
we seeh it or not" (p. 97). At stafw in Ollr narrative attenIpts to achieve L1 {zoIJcd to b-'I tl . 1 . ..
to leve hi. a text IS llllfJortallt only to the extent it llloues bevond
coherent sense of ourselves are the very integrity and intelligibility o( oW t IJe mcrelv p.' -- .. {' IV!. ... I ' ,
. (;1.:;01f([ :ve llee( to question Ollr ~1&Slllllrtio115J thL! /l1t:tantlr:;.;
se/fhood, which rest so tenderly and fallibly on the story we use to huh birth tr{Jal."ROvei'll II]e" IJJS 1'1 ' I { , f ' ,
IlttJOlla worwzgs a SOCIal sCtence-arp,ulllellts aI/a
to life to death (IvfacIntyre, 1981). III the (inal analysis, the self is indistill- t ee llllgs th - _., ...". I ' ,
c .' eOI JeS oue1 staT les, a Jstractzons ouer concrete euents, sophisti-
guishable from the life story it ':onslmcts for itself out of what is inherited, aled jm'aOll
6 Over ac aSSJ > . . \'7/
bl't .PJose. ,1 ])1 S /JOlt /1
(we b e ashamea' 1(.. our /.uorl;;.

220 221
 MI: I HUU::, Ul- LULLtL 111'1\..:1 AI'lU AI'IALT LII'll;l CIVlnKl\...AL IVlt-\1 C .... 'ML..J
has therc1pcutic 01' personal [la!He? Besides, hauen't 01lr fJerso1tal stories
always been embedded in OfF research monographs? The question is
luhether lue should express Ollr /lull/crability and subjectivity opellly ill the
text or hide them behind "socie! analysis."
SOilletimes I thil/k: Art, ifolliy you could do a better job c01111mmicatilzg
the important differences bettlJeen a representational and an evocative
social science. \'Vhy is it so hard to grasp that personal narrative is moral
IUork and ethical practice? WihcJl the narrator is the illvestigat01; to a cer~
tdin extent she is always asking what it is right to do and good to be. At its
iifOst extreme, those who want "to put narrative ill its place" (Atkinson,
1997, fJ. 343) seem to think the:~e is only one right place to lJut it. They seek
to preserve luhat already has been lost (Gergen, 1994; Schwandt, 1996).
They thin!? that if these personal voices can be silenced, then perhaps they
can return to business as usual in the social sciences, protected agaillst the
cOlltingellcies of IJlltlli.1ll experience, restored in their traditional belief i11 a
trl-1Ilscelldent positioll fmlll which to SfJeah (ami illtelpret) with authority,
(reed of moml choices and cmotiollal dilemmas, and inspired to champion
cOlltrol ouer late, ~lcls over Jlu'anillgs, and rigor Duer peace of milld.
"\'\IeU, [guess this is a good place to stop and throw this session open for
comments or questions," Art invites. People in the audience shuffle in
their seats anxiously, then several hands go up. "Yes, Billy," Art says, point-
ing toward a philosophy professor I recognize.
"An, you mentioned that your turn toward narrative was provoked by
postmodernism. Could you ebborate on that?"
"I had read Thomas Kuhn's The Stmcture of Scientific Revolutiolls and
\vas impressed by his argllmert that there was no way to distinguish un-
equivocally what's in our minds from what's am there in the world. About
the same time, I was introduced to the writings of \X'ittgenstein (1953),
Heidegger (1971), Cadamer (1989), and Derrida (1978), and to speech
act theory. In quite diverse ways, all of this work stood in opposition (0
the view-that now seems incredibly na'ive-that language could be a nell-
tral or transparent medium of communication. \Vhether we apply lan~
guage to ourselves or to the \vorlcl there always is slippage, inexacrnes s,
indeterminacy. Then along carne Richard Rorry's Philosophy arrd the .Mir-
ror Qr N"lul"<.' (1979), v\,hich provided .:I po;verful :srnrhc:si:s of tl1e ch!ll-
ienges to our most venerable :lotions about trmll and knowledge. It was
hard to rcad Ron)' withollt feeling totally shaken. I came away convinced
that the foundations of traditional epistemology were fallible. No strong
222
case could be made that human knowledge was independent of rhe human
mind. All twths were contingent on the describing activities of human
beings. No sharp distinctions could be made between faets and values. If
you couldn't eliminate the influence of the observer on rhe observed, then
no theories or findings could ever be completely free of human values. The
investigator would always be implicated in rhe product. So wh}' not ob-
serve the observer, focus on turning our observations back on ourselves?
And why not write more directly, frolll the source of your own experi-
ence? Narratively. Poetically. Evocatively. No longer was there any deep
re{lSOn to believe that social science is closer to physics than to literature or
poetry. Besides, I became a social scientist because I thought it \vas a way to
address deep and troubling questions about how to live a meaningful, use-
ful, and ethical life. Somewhere along the way these ql!estions took a
backsear to methodological rigor. Nmv I felt liberated to grapple with
these questions again, more dialogically, through personal narrative."
A woman [ don't recognize stands and shouts from the back of the
room, "I've always found the postmodernists depressing and cynical.
They seem to be saying yOll can't know anything. It all seems so destruc-
tive." Laughter circulates through the audience and I notice a number of
people nodding in agreement.
Art responds, "\X'ell there's an affirming strain of postmodernism too.
At least I read it that way. In the writings of certain postmodernists and
particularly vvithin feminist and queer theory you see a renewed apprecia-
tion for emotion, intuition, personal experience, embodiment, and spiri-
tuality. They've helped uS' cross some 0 f the boundaries separating the arts
and the sciences and to foclls attemion on diversity and difference insread
of unity and similarity. [don't regard these move~ as negative or depress-
ing. Perhaps, like you, I find them unsertling, even painful at times. But
that's where the learning is. \Y/e lose our innocence and our lost innocence
validates some good values. \"X7e gain tolerance and humility. Sometimes
\ve're ashamed of how much we've excluded from our experience, tried
nat to see, hidden from. And we should be. \Y/e don't need to run from the
fear or anxiety we feel. \X'e need to learn from it. Racism, sexism, poverty,
homophobia, disability-these issues touch all of us. \'iJe can't hide from
them. \Ve're all cornplicit in some way. No one's immune, invulnerable. So
it',> irnport:.:Int to l:5<.:r ...o:"'po:s<.:o:.1 ro local stories tl.at b.-;n& us into ..,rorlds ....-,f
experience that are unknO\vn to us, show us the concrete daily derails of
people \vhose lives have been underrepresented or not represented at all,
help LIS reduce their marginal ization, show us how partial and situated our
223
 METHOD~ UI- LULLtL 111'1\.:1 AI'lU AI'lAU Llt'l\.:l CIVlr''''''-H'- '''H' ... ,,,,..,.......
undcrstanding of thc world is.1Vlaybe that's depressing to some of yOll, but
I think it's enlightening and possibly transforming."
"1 think you lllisunders[{Jod her," a man in the front row interjects. r
rccognize the voice of a colleague wedded to mainstream social science
methods. "The resistance a:ld political dimensions are clear enough, but
some of us still want to kno'N how we can teU when we're right, when our
representations afC accurat,= and we can generalize."
Art sighs in frustration ;ond continues, "\XTe may have to agree to dis-
agree. 1 take the crisis of representation more seriously than you do. For
me, it necessitates a radical transformation in the goals of our \\'ork-from
description to communicarion. That's the inspiration for the narrativc
Ultn. As I sec it, thc practio:'S of human communication-dlc negotiation
and performance of acts of mcaning-should become our Illodel for hm\'
we tell about the empirical world (Bochner & \X!augh, 1995). Then, we
\vouJd feel compelled to ptoduce narrative, evoGltive, dialogic texts that
show human beings, including ourselves, in the process of creating, nego-
tiating, and performing m';aning in 3. world of others, making our W3Y
through a world that pose.:; obstacles, interruptions, contingencies, turn-
ing points, epiphanies, and moral choices."
"So what are the goals? [don't quite follow," the same lllan continues.
"Could you be more preci:;e?"
"The 'goal is to encouraf;e compassion and promote dialogue. Actually,
I \vould be pleased if we understood our whole endeavor as a search for
better conversation in the face of all the barriers and boundaries that make
conversation difficult. The stories we write pur uS' inro convers~ltion with
ourselves as well as with our readers. In conversation with ourselves, we
expose our vulnerabilities, conflicts, choices, and values. We rake measure
of our uncertainties, our mixed emorions, and the multiple layers of our
experience. Our accoums :;eek to express the complexities and difficulties
of coping and feeling resolved, shov'ling how we changed over time as we
struggled to make sense (,f our experience, Often OUf accounts of our-
selves arc unflattering and imperfect, but human and believable. The text
is used, then, as an agent of self-understanding and ethical discussion..
"In conversation with our readers, we use storytelling as a method tOt
inviting them ta pur themselves in aur place. Our dialogue centers on
moral choices, que~tions ilavins l11un: to do, ~l.S Ivfichad Jnd::son ('1 q95)
observes, with how to live than with how to know. The usefulness of
these stories is their capacity to inspire conversation from the point of
, of the readers, who C':"lter from the perspecnve
view ' of t I'
lelr own I'Ives. The
224
. '~.~_L'" '~';J' "'I~' 'r, ' .... ' .......,,'-', "'-" ,~,uve, ',e"t=A,vlly
narrative rises or falls on its capacity to provoke readers to broaden theit
horizons, reflect critically Oil their own experience, enter empathically
into worlds of experience different from their own, and actively engage in
dialogue regarding the social and moral implications of the different per-
spectives and standpoints encountered. Invited to take the story in and use
it for themselves, readers become coperformers, examining 'themselves
through the evocative power of the narrative text."
Art pauses to take a sip of water. Jim, the coHoquium organizer, turns
to\vard him and says, "I liked your anempr to enter into dialogue with the
critics of narrative inquiry, bur you left out one of my main reserv:nions.
How do you react to critics who say thar petsonalnarrarives simulate rcal-
ity TV? Aren't these nartatives teflectivc of the culture of confession and
victimization and don't they end up as spectacles thar sentimentalize,
humiliate, and take pleasure in revealing anguish and pain? Personalnar~
rJtives remind me of victim art. Thcy play on your sympathics and mJnip-
ubte your emotions."
"I've heard that one before, Jim. IVfy first response is to consider the
source. That's a particular teading, by a particular person. So it's always
the case, in Illy vie\v, that the criticism speaks the critic's life roo. Tl~e
text's meanings are never transparent. Thete is ahvavs a connection be-
ing made between the reader's consciousness and wh;t is being read. So I
want to know something about the reader-her interests, desires, values,
premises, and what she resists and why.
:'50, Jim, where are you in this pictllre?" Art tcases. He pauses and
smlles gently as the audience chuckles and Jim looks around quizzic311y,
shrugging his shoulders.
, "Seriously, Jim," Art continues. "I didn't mean to put you on the spor.
1
\X ell, maybe I did. But, as a critic, I don't think it's your job to condemn
something categorically. I think you have to look at the merits of each case.
Ir's hard for me to respond in terms of some general principle. If yOll take a
genre of stories that might be called 'illness narratives,' for example, the
SOrt 0 f
5. "I
stones flat A rflur
I Frank (199)) " has analyzed, well, f think the
goal IS to reduce the stigma and marginalization of illness and disability.
Mastofd
. lese stones ' arc
.Wfltten
' I1Y peop
' Ie who don't want to surrender to .
tbe, victimization and marginal identities promoted b}' the canonical nar-
rJtlVe
1 ,. () f n~".J''<.ell"".
_' . ;vfeUl)"
~. .
<..>1 t:hc'n H} ~Q WrlU.:: thcrn~d..,c:; i:li5 jUrvtvOf:i, c\iij-
PaYing their embodiment as a source of knOlvledge. It's hard to under-
stand h ' Ij
, 0\\ anyone cou c read Anatole Broyarcl (1992) or Nancy lvlairs
(l986, 1990, 1998) or Audre Lorde (19S0)-and j could name dozens
225
 IVIC J n u u . ) u r \....ULLC\.... J 1J'1\..J f-\J'JU f-\!"f-\LJ LII'iU LIVlrll'\lLML IVIM' LI'\'MLJ
more-as victim confessionals, They aren't seeking pity and they don't
ponray themselves as pathetic~ helpless, downtrodden characters. If any-
thing, they use narrative as a sc,urce of empowerment and a form of resis-
tancc to COUnfer the domination and authority of canonical discourses. I
think that Couscr (1997) thoroughly discredits the 'victim art' argument
by showing that the vast majority of narratives focused on the 'recovering
body' 'are much more likely to devictimize their subjects and others like
them' (p, 291), Their m8in fun,::tion is to confirm and humanize the expe-
rience of illness by bearing \'vitness to what it means ro live with bodily dys-
function and to gain agency through testimony. So, what are the choices,
Jim? Erasure? Silence? Surrender? I think you have to understand some of
the identity politics that are involved here too, Whose stories get told? By
\vhom? And for what purpose? I know you're interested in cultural and
political implications of narrative. Don't you think these stories help us
understand hmv culture and p,]litics are written on the body?"
"1 sce your point," .lim says, "but I stiH worry about voycurism and the
way these personnl stories indulge our culture's perverse curiosity about
the private, peeking in on dam:1ged selves. How do we judge the merits of
these stories? \X!hcn do vve know they're reliable and telling?"
"I think it's the same judgment we make about any author or any char-
"letrr. Is the work honest or did1onest? Docs the author take the measure
of herself, her limitations, her cDl1fusion, ambivalence, mixed feelings? Do
you gain a sense of emotional reliability? Do you sense a passage through
emotional cpiphany to some c1l11ll1llnicated truth, not resolution per se,
bUl: some rransformation frorn an old self to a new one (Rhett, 1997)?
Does the story enable }'OU to understand and feel the experience it seeks to
convey? There is complexity, multiplicity, uncertainty, desire. Phillip
Lopate (1994) refers to the pcrsonal essay as something akin to basic re-
search on thc self that ends up as 'a mode of being' (p. xliv). It's nm sci-
ence; it's not philosophy. The same can be said for the evocative, personal
story. It's an existential struggle for honesty and expansion in an uncertain
\vorld. "
r tap Sylvia's shoulder and whisper, "Notice how Art dodges questions
that try to get him to stipulate categorical criteria. He always wants to b:1l-
:lI1ce rigor ::Inc! imngin:.1tioTI. H,~ thinl.:-s if you're too bound up '\vith rules,
you probably v/On't do anything interesting. Anyway, I've got to meet
another student in my office for a makeup exam. I'll meet you back here
:Jfter the talk."
226
f-I.uroernnograplly, f-'ersonw l'iUlIUL!Vt:", """f"'-"lVlLy
\Vhen I return, Sylvia is standing alone watching the students gathered
around Art talking passionately about their writing projects. "The woman
facing us is Lisa Tillmann-Healy," I tell Sylvia. "She's published a story
about her mvn eating disorder, and she's recently finished her dissertation
on straight couples' relationships with gay men, teUing the story of her
own friendships. The woman to her left is Deborah Austin, who writes lyr-
ical poems and did her dissertation on African American marriages in the
aftermath of the .iYiillion 1\11an 1v1arch. For her dissertation defense, she
performed a script she wrote based on focus groups she studied. Christine
Kiesinger, the woman talking [0 Art, has published several Stories from her
dissertation on women with caring disorders. You might be interested in
looking at her dissertation to see how she weaves her story with the story
of one of her participants. Over there, that's Laura Ellingson," I say, nod-
ding JUSt to the left of the group. '''She published an article recently in
Qualitative Inquiry on how her own illness affected her understanding of
other cancer patients and the organizational environment at the cancer
hospital she is studying, and hO\\I, in rurn, this experience helped her rein-
terpret her own illness. I'll give you a reprint. Laura is talking with Leigh
Berger, who published a story about her relationship with her hearing-
impaired sister and another about her father who was institutionalized for
mental illness. She's studying .Messianic Judaism now, obserVing her own
transformation as she panicipares in a religious group. Come, I'll intro-
duce you to them."
"I read the articles by Lisa and Christine," Sylvia reminds me. "Interest-
ing that they are all \vomen," she says thoughtfully, and then exclaims,
"\X'ow! This is exciting!" I smile, but before I can say anything, Sylvia
blurts Out, "I want to write my story. But 1 haven't been keeping notes or
anything. How would 1 do it? \\7here would I stan?"
Doing Autoethnography: Considerations
"Answering your questions will take a while. Let's go get a cup of coffee.
You can meet the other smdents later," I decide, \vaving to them oyer my
shuulder. "There are a nurl1ber of ways to go about writing auto-
e.thnograpby," I say as we walle "It really depends on where along the con-
tinuum of art and science you want to laGue yourself. What claims do you
Want to make? If you \vanr to claim vau're following traditional rules of
ethnographic method, then it \vould 'be best if you had kept notes on the
227
 METHODS OF COLLECTII'lG AND AHALYZING EMPIRICAL MAl tRIAL~
experience as it happened. The notes would serve as field notes and you'd
write from those."
"If you didn't have notes, how v.,rQuld you remember what actually hap-
pened?" Sylvia asks.
"Do yOll think the notes would tell you what actually happened? Aren't
they partial interpretations as wdl?"
"\Xl ell, yes, but then how wl)uld I make sure that what I said was
truthful?"
"The truth is that we can never capture experience. As An said 1 <Narra-
tive is always a story abom the plst,' and that's really all field notes are-
one selective story abom what happened written from a particular point of
view for a particular purpose. But if representation is your goal, it's best to
have as many sources and levels eJf story recorded at different times as pos-
sible. Even so, realize that every story is partial and situated."
I take four quarters from my pocket and insert them into the coffee
machine. ''I'm buying," I say. "Cream and sugar?"
"Oh 1 no. Let me pay," she imists, opening her purse.
"Next time. Okay?"
"Okay. Just black for me." \Xl':;. take our coffees outside and sit under a
tree to enjoy the perfect Florida;pring day. "Is there a way other than rep-
resentation to think about personal narrative?" Sylvia asks.
"\Xlell, yes 1 if you viewed your project as closer to art than science, tben
your goal would not be so much to portray the (acts of what happened to
you accurately, but instead to convey the meanings you attached to the
experience. You'd want to tell ;] story th::lt readers could enter and feel a
part of. You'd vvrite in a way to ("'voke readers to feel and think about your
life, and their lives in relation to yours. You'd Want them to experience the
experience you're writing about-in your case, breast cancer."
"If these \vere your goals," I continue, "writing notes at the time the
experience occurred would ha\'c been helpful, but not absolutely neces-
sary. If you're ,vriting about an epiphany, which you usually are in this kind
of research, you may be too caught up in living it to write about it."
"But then how do you remember all the dialogue and details later?"
"\X1hen I wrote Final Negotiations, about the chronic illness and death
of my first hllSband, I didn't Clctllally remember everything I wrote about',
certainly not the exact words we spoke, anyway. I had nares for lTlllch of
,vhat I described, but I still had to construct scenes and dialogue from the
partial descriptions in my notes. And I hadn't kept immediate notes for
everything I wrote about, though I consttucred them later. Bur it's amazing
228
l-\utUernnograpny, t-'ersonaJ /'JarratlVe, !<etJexivity
whm you can recall, and for how long, if the eVent was emotionally evoca-
tive. Another story I wrote, about race relations in a small town , \~as con-
structed without notes more than 25 years aher the event occurred."
"Bur how can thar be valid?"
"It depends On your definition of validity. I start from the position that
language is not transparent and there's no single standard of trllth. To me
validity means that our work seeks verisimilitude; it evokes in readers a
feeling that the experience described is lifelike, believable, and possible.
You might also judge validity by whether it helps readers communicate
with others different from themselves, or offers a way to improve the lives
of participants and readers or even your own. Take a look at Larher's dis-
cUE-sion of validity and couiHerpraetices of authority in the Sociological
Quarterly, 1993,1 believe it is."
Sylvia looks up from her note taking and grimaces, "\\lhat about
reliability?"
"Since we always create our personal narrative from a situated location.
trying to make our present, imagined future, and remembered past co~
here, there's no such thing as Orthodox reliability inautoethnographic re-
search. However, we can do reliability checks. \X/hen other people are in-
volved, you might take your \vork back to them and give them a chance to
comment, add materials, change their minds, and offer their interpre-
tations. "
"Genefi.llizability? Is that a concern?"
"Of course, though again not in the usual sense. Our lives are particular,
?utthey also are typical and generalizable, since we all participate in a lim-
ned number of cultures and institutions. \X1e wam to convey both in our
stories. A story's generalizability is constantly being test-ed by readers as
they determine if it speaks to them about their experience or abom the
lives of others they know. Likewise, does it tell them abollt unfamiliar peo-
~le or lives? Does a work have what Stake calls 'naturalistic generaliza-
tlon,' meaning that it brings'felt' news from One world to another and
provides opportunities for the reader to have vicarious experience of the
tbings told?"
."That's sure different from \Vh3t I've learned, bur I think 1 understand.
Still 1 don't know where to start my own project."
. \~\\D'hr don't )'ou ~tan br writins u clru[L of yvur Jwry, TIHnK of 1r a~ l11a1(-
1l1g retro~pective field notes on your life. Include all the derails you can
recall. I find it I I f l .. I . . .
tb. .' . . le p u to organize my wrJt!l1g c 1rol1olog1cally first, usmg
e ma1l1 events to srructure the tale. I try to write daily, rereading what I
229
 Mt j HUU::' Ur- LULU:LIINV.I P.1'lU P.1'IALl LII'l1.;l CIVIFIr<.I'-..ML rVIMICI>IML.)
wrote the day before, then filling in new memories. Remember, you are
creating this ~tory; it is not tht:re waiting to be found. Your final story will
be crafted from these notes. "
"But how will I know when I'm writing from my perspective then and
\vhen my current perspective i~ clouding my memory of what happened?"
"Well, you won't really. lvlemory doesn't work in a linear way, nor does
life for that matter. As Denzin and also Ronai say in InvestigatiJlg Subjec-
tivi~y, the book I edited with Ivlichael Flaherty, thoughts and feelings circle
around us, flash back, then f:)[ward, the topical is interwoven with the
chronological, thouglHs and feelings merge, drop from our grasp, then
reappear in ::mother context. [n real Efe, \ve don't always know when we
know something. Remember An's talk-events in the past are always
interpreted from our current position. Yet that doesn't mean there's no
valLIe in trying ro disentangle now from then, as long as you realize it's not
J project you'll ever complete or get completely right; instead, you strive
to get it 'differently contoured and nuanced' in a meaningful \vay, as Rich-
ardson says in her Hmufbooh chapter."
"\\7hat do you mean? How do you do that?"
"1 use a process of emotional recall in which I imagine being back in the
scene emotionally and physiolly. If you can revisit the scene emotionally,
then yOll remember other details. The advantage of writing close to the
time of the event is that it dc,esn't take much effon to access lived emo-
tions-thev're often there whether you want them to be or not. The disad-
vantage is ~hat being so involved in the scene emotionally means that ir's
difficult to get outside of it to analyze from a cultural perspective. Yet both
of these processes, moving in and moving our, are necessary to produce an
effective autocthnography. That's why it's good to write about an event
while your feelings are still intense, and then to go back to it when you're
emotionally distant. I've had students who were great at getting il1side
e~10tional 'experience, but they had tunnel vision. They couldn't move
around in the experience. They were unable to see it as it might appear to
others. They had trouble analyzing their thoughts and feelings as socially
constructed processes. I'll give you my articlc on systematic sociolog icJI
introspection, \\'hlch talks more about introspection as a social process.".
''I'd like that. But I'm nor sure I'd wam to fed all those emOtions again.
Jt
t\nu JOHlC of lllt feditl);5 rye had and ::;till have abvltt lll} \,;~lll\,;(;r- I \'I!uulcln
want to share. I'd feel so vuh.erable."
"\X1ell, that's your call. But if you're not \villing to become a vulnerable
observer, then maybe you ought to reconsider doing aurocthnography. If
230
Allroernnograpl1y, r'ersonal Narrative, HetJexivity
you let yourself be vulnerable, then your readers are more likely to
respond vulnerably, and that's what you want, vulnerable readers. 1 agree
v>/ith Ruth Behar, who wrote in The Vulnerable Obseruer that social science
'that doesn't break your heart just isn't worth doing.' IvIy goal is the same
as Dorothy Allison 's-'to take the reader by the throat, break her heart,
and heal it again.' Vulnerability can be scary, but it also can be the source of
growth and understanding."
''I've always assumed my task as a social scientist was to deliver knowl-
edge and stay invulnerable," Sylvia responds. "I didn't know 1 had a
choice.
"So, suppose I am willing to be vulnerable," she continues slowly.
"Hmv do I get from field notes to writing in a way that opens up myself
and readers to being vulner~lble?"
"Do you ever rcad fiction?" \\1hen she nods, I continue, "\\1ell, think
about how a good novel makes you feel. It does make you feel, right?" She
nods again, waiting for what 1 will say next. "\X1hat provokes these
feelings?"
"Sometimes I identify with the characters. I feel for them. Or I think
about being in the situations they're in, doing what they're doing, or imag-
ine what I'd do in the same situation. And sometimes 1 stop reading to
think about hmv my life is different or similar."
"Exactly. Good fiction \vriters make you feel the feelings of the charac-
ters, smell the smells, see the sights, hear the sounds, as though you were
there. They do this with devices of fictional writing such as internal mono-
logue, dialogue among the characters, dramatic recall, strong imagery,
things like scene setting, charaner development, flashbacks, suspense, and
action. You enter the reality of the novel through a dramatic plorline,
which is developed through the specific actions of specific characters with
specific bodies doing specific things."
"Then how is \vhat you do different from fiction writing?"
"A number of soci<ll scientists 11dve addressed your question. Take a
look at Denzin's discussion of the relationship of social science writers to
the new journalists in Interpretive Ethnography. Susan Krieger's early
piece on fiction and social science ;md Richardson and Lockridge's new
\vork on fiction ;:ll1d ethnography also might be helpful."
'"The two 3t:llrc::; arc more similar tl1al1lIiHcrClH," 1cominuc. '110 WaIf
Harrington says about intimate journalism, in autoethnography you try to
write from inside the heads of participants and evoke the tone of their felt
lives. "
231
 METHODS OF COLLECTING .!\ND ANALYZING EMPII{ILAL MAIl:.I<IAL~
"Of course, writing and publishing convenrions are different," I add,
now s\vitching gears. "You're a ~;ocial sciemist, so that probably will affect
what you look at ,md how you see. And, among social scientists, auto-
ethnography often has more of an overt analytic purpose and an analytic
rrame. Remcmbcr how Carol Ronai in thc piece I gave you layers analysis
through her personal narrative? But in Final Negotiations, I cmphasized
thar analysis can come through story and dialogue too. Arthur Frank says
in The WOl/nded Storyteller that it is important to think with a story, not
just about a story. Thinking with a story means allowing yourself to reso-
nate with the swry, reflect all it, become a part of it.
'Td suggest you read some cxemplars of this work and note the differ-
ern ways authors intersect story and analytic frame. Look at some of the
books in the Alralvlira Ethnogc1phic Alternatives series edited by Art and
me. For example, Jones's Kaleidosco!Je Notes uses conversation, songs,
poetry, stories, performance, a-:ld amoethnography w examine women's
music, a folk music club, and ethnography; Angrosino's Opportlfllity
HOlfse is made up of fictional stories of adults with mental illness that are
based on his decade of participant observation \",ork; and Ivlarkham'sLife
Online uses her own experiences to study life on the Imernet. Our Com-
posing Ethnography and Fictio;'l and Social Research by Banks and Banks
both shO\vcase a multitude of creative forms of narrative writing."
"Aren't decisions social scienrists make di Herem rrom fiction writers?"
"\XTell, genetally, autoethnographers limit themselves, unlike fiction
writers, to what they remember acruaHy happened. Or at least they don't
tel1 something they kl1LWV to bE false. \Vell, even that's not so clear-cut. It
depends ... JJ
"On what?"
"\XJell, say you want to prott~ct the privacy of a character in your story.
Then you might use composites or change some identifying information.
Or you might collapse events to write a more engaging story, which might
be more truthful in a narrative sense though not in a historical one."
\Vhen Sylvia looks at me questioningly, I say, "You know-the story
evokes in readers the feeling that the tale is true. The story is coherent. It
connects readers to writers and provides continuity in their lives." \Vhen I
see a look of recognition on Sylvia's face, I continue, "Even realist
e1l1l1ograpllers, wila claim to fullow tIlt:: rules [or dOing science, usc de-
vices such as composites or collapsing events to tell bener stories and pro-
tect their participants. Yet they worship <accuracy' in description. A friend
of mine, Sherry] Kleinman, says, If it didn't happen, don't tell it. Thar's
232
f-IUlUl:!trlnugrapny, t"'ersonal ""arrotlVe, l<etlexlVity
another version or 'Don't pur words in participants' mourhs if they didn't
say them.' Bur, of course, ethnographers do put \\lords in participants'
mouths all the time."
"Really? How can they get away with that?"
"By relying on memory, editing, and selecting verbatim prose out of
conteXt and then surrounding it with their own constructed analytic con-
texts. When it comes to analysis, most traditional ethnographers have no
problems reaching beyond description for all kinds of interpretation."
"Give me an example."
"Oh, from limited time and 3ccess in the field, they create the 'typical'
person or day, the 'common' event. They use ambiguous and qualifying
descriptors like most, some, frequent, and few. And, of course, they reify
concepts such as soci3l structure and organizational climate. I did this too
in my first study of two fishing villages. Let me tell you, when community
members read what I wrotc-vvcll, \vhat 1 saw as typical was certainly not
what they saw as typical. \X!hat I wrote told you more about how I organize
my world than how they organized theirs."
"Don't believe the propaganda," Art says, suddenly walking toward us
with a stacie of books piled in his arms.
I laugh and ask, "Hi, where have you been?"
"In the library, retrieving Some sources for our Handbook paper."
"They look pretty heavy to me," I say, smiling as I eye the titles on the
spines of the books. "An, this is my student Sylvia, the one who is studying
breast cancer."
"Oh, yes, hi. I noticed you sitting next to Carolyn at my talk today."
"Yes, I found it very interesting," Sylvia replies.
"We \vere just talking abour how auroerhnography differs from fic-
tion," I explain.
"Oh, was Carolyn giving you her rap on how you have to be systematic
and stick ro the facts?" Art asks, turning to Svlvia. "Just the facts ma'am"
he mimics. . , ,
"Ah ... ," Sylvia stalls.
''Art, stop it," I say playfully. Then turning to Sylvia, I explain, "An and
I have this running commentary on writing autoethnography. I argue
that you try to construct the story as close to the experience as you can
.-eHlcmbcl- it, e.spccially in the iUitial version. Tllat doIng 50 llelps yOll WOrlf
through the meaning and purpose of the story. He likes to argue that
wh a.t',' .' h ' I
S Important IS t e useru ness of the story. Of course, I agree that our
Stones should have therapeutic value ... "
233
 METHODS OF COLLECTIHG I\ND AI'lALY Llr'llJ tMf-'II~II.AL MAl tK.IAL:)
"Therapeutic v,-llue?" Sylvia swmmers.
"'{cs,l (hink of i( as action research for (he individual. Though therapy
might nm be the major objective in our research, it often is a useful result
of good writing," I respond.
"That reminds me, An ... uh, may I call you An?" Sylvia inquires.
\"XIhen Art nods, she continues, "\X1hat you said in your talk abollt the focus
of stories, well, 1 thought therapy and research were separate entities. 1
mean, I'm a therapist, but 1 ass'J111ed 1 had (0 keep (hat role separate from
my interviewer identity, because if I acted as a therapist It might bias the
data. And wouldn't it be unethic::l1?"
Art and 1 look at each othEr and try not to smile. Her questions and
concerns help us realize how dose together our positions are. I quickly
interject, "Bur you told me you hoped your research would provide under-
standing of whar happened ((0 you and help others who face similar cir-
cumstances cope. So what will you do if an interviewee breaks down or if
you see a place \vhcre you could be of help?"
She looks at me, waiting for the answer, then murmurs, ''I'm not sure. "
"\Xlhat would you want someone in a similar situation to do for you if
you \vcre a research panicipanr?"
"\\lell, I'd want thcm W GTe about me and rry to understand where I
was coming from, ,. she responds softly. "Otherwise 1 wouldn't want to
share my life stories wi th them."
"And wouldn't it be unethical for a researcher not to help or empathize
with you if you were in need?"
''I've never thought of it tbat way hefore, bur I would want my subjects
to feel that I care about them. \\lhat good would my research be if it does-
n't help others who are going through this experience, especially my
subjects?"
"Participants," I say quietly.
"Participants," she repeats, her face tuming red. "But isn't it true that
nor everybody can do good therapy? I mean most academics aren't trained
therapists."
"Being able to do therapy and being a trained therapist are not synony-
mous," 1 respond, and Sylvia nods in agreement. "In fact, ethnographic
training migl~t be just:IS il~-Il)ortanr for a thtrupist us thcmpeutic training."
"And therapeutic training probably should be <'1 prercguisite W being an
ethnographer," Art adds, laughing.
1 smile and continue, "But you're right, not everybody is comfortable
or capable of dealing with emotionality. Those who aren't probablY
234
MUtU\::'UIl Juyru[Jr ry, ,t!r~onOI l'IOrrOtlve, l<etlexJV!ty
shouldn't be doing this kind of research in [he first place, or directing
students who arc."
"Perhaps you should give her citations to articles on some of these
issues, I ike the ethic of caring and personal accoumability, maybe Collins,"
Art suggests, as several of his books fa\! to the ground.
"I will," I say, helping to retrieve the books. "Other feminist writers
would be helpful too. Let's sec, Lieblich,MiHer, Cook and Fonow, and
Oakley. They'd be a good stan," I say, marking them off as I return each
book to Arr's stack.
"And there arc good summaries in Reinharz and also in Denzin's lnter-
pretiue Etlmogl"aphy," Art adds. He then rurns to Sylvia, "But enough liter-
ature, 1 want to know more about how you'd respond if yOU were an inter-
viewee. What would make you comfortable enough to'tell your story?"
"To know the other person was listening, really listening. I'd want
someone I could cry in from of, actually who might cry with me. A person
who might tell me some of her stOIy if she had been through a similar
experience."
"So are you going to share your story with your participants?" I ask.
"Ah ... I think ..."
"Go on."
"\X/ell, I was going to say that my stOty would contaminate rheirs, but
I'm not so sure anymore."
Art and I smile. "This is probably enough for now," I sav. "We've come a
l_ong way. \XJhy don't you think about how this conversati~n provides clues
tor .how you might want to do your own interviews and let's pick up this
toPIC next time we meec."
\X/e say our good-byes, and Art and I make our way to our car. "Are you
sure this is the right move?" Art asks. "Is she ready to write her story?"
"Oh, I think she's ready. I sense she wants to tell her story."
"\'{1hat if it opens up things for her that are just too painful?"
''I'll keep in close contact with her, just in case. Bur in my experience
with personal narrative, people pretty quickly find their ~wn comfort
Zone. They know when the time's right. Bur I'I! make sure to provide
Opportunities for her to pull bJck or change gears in the project, if she
~eeds to. She can always do that survey," I add, playfully tugging at his arm
ane! then skipping ahead.
1 '~l admire how much you're willing to risle with your students," Art says
OVlngly when I return to help him pick up the books that once again have
toppled to the ground. ''And hm"" much you care aboUT them."
235

"Same \\'ith you," I say.
''It's not easy being vulnerable, especially in the academy, where you're
expected to be in control and keep your prIvate life removed from your
professional life. That's what I tried to say in 'It's About Time.' "
"It's scary, when you think about the professor at Colby who asked stu-
dents to write personal narrativ,;:s and ended up being charged with sexual
harassment. Of course, we don't know what really happened there; we
have only Ruth 5balit's report," 1 add. "Ivlaybe his private and professional
lives did become too entwined."
"That's certainly a possibility. But what about the article in the Chroni-
cle of Higher Education that described how some of Jane Tompkins's col-
leagues attacked her for sllggesi:ing that the emotional and spiritual lives of
university students are just as '_mporta~t as their intellects?"
"Maybe we should JUSt wri1:e fiction," 1 offer.
"Nmv wait a minute," Art r':-primands. "1'l1u know everything we write
is fiction ... "
<> Doing Autoethnography: Method and Form
Two weeks later, Sylvia appears in my office. "Hi, I've written most of my
story about my past now, and I waited until 1 was almost finished before I
began reading other personal narratives of breast cancer. It's been very
therapeutic," she says, "to write and to react. Bur I'm not sure I'm getting
anywhere on my dissertation. I have so many questions."
"Like \:vhat?" I ask.
"\\7hy would anybody want to read my story? How does my story differ
from what's already published? And how will my story fit with the inter-
views I want to do of other women?"
"Slow down. Are you learning anything?"
~'Oh, yes, at every turn."
"Tell me what you're learning."
"Well, that I have a lot in common with other women's breast C<loa f
SruriC5. For cxampl~, mo~t women tell uf their discovery of the lump-
that'S nJw~y a traUmatic CYGUG-dlCTl dl."-' J;a5nos;s :md Qs~es~ment. of
treatment options, then they describe waking up from the surgery, gOll1g
through the follow-up treatment, and finally there's recovery and some
kind of resolution at the end."
236
...... UlV<:'(.''''UyIUj-Jlly, Ft.::I:::'UfIUJ ('jurrUI/ve , Kt;!/It::AI'''l
"Interesting, that's almosr exactly how Couser summarized breast can-
cer narratives in his book on illness narratives," I respond, pleased with
how much reading Sylvia has done.
"IvIost survivors describe making decisions about reconstructive sur-
gery, shopping for a prosthesis-if they decide to vvcar one-their hair
falling out, and seeking alternative treatment," she continues without
skipping a bear. "I wrote about these things as' well, and ... "
"And have you learned anything new from vi/riring yOUl" story? Sorry, I
didn'r mean to cut you off, but I'm curious."
"Yes, rhat cancer is more than a medical storY it's a feeling story. I
learned how scared I am even though I've been a sl;;vivor nOW for 7 years.
And that's the interesting thing-there's little about long-term survivors in
stories or in social science research. IvIost survivors tell their stories soon
after recovery from treanncnt and the},'re usually pretty optimistic about
recovery and often claim to be better off at the end than the beginning.
"1 felt that too, the optimism I mean, immediately after my treatment
was over, that is," Sylvia continues passionatelv. "But I don't feel that way
novv. I try so hard to pretend that I'm an upbea~, optimistic person with no
worries, a v-mrrior who has learned from her experiences. But what I had
to face as I wrote my story is that I'm scared all the time that the cancer will
come back. I've had carpal runnel syndrome and it's probably from the
chemo. And now I have sweats at night, and I don't know if it's early
menopause-another gift of chemo-or signs of the cancer returning. I'm
sorry, bur cancer has not improved illy life and I can't make it into a gift.
Holding in these feelings, all these years, has been difficult and 1 think it's
had negative effects on my psychological and physical well-being and on
my famil)~"
Sylvia begins to cry.lmuch her shoulder and hand her a Kleenex. \YJe sit
silently for a while, sadness connecting us. Needing to stay in the role of
adviser, 1 hold back my tears. "Do you still want to continue this project?"
I ask gently. "Or is it too painful?"
"Oh, no, I have to continue it," she responds forcefully, although her
Voice shakes. "What I'm experiencing is important to D1e. It was hard
pretending; sometimes 1 thought I \\-'a5 going crazy. NoW I realize I don't
have to pretend. There are othcr stories to live and \vrite. Ivlaybc through
W,it"lng an d ta II" " I1 other )vomen about their expt:fll,;PCes,
ung vYlt " I can f"19ure
OUt another story to live, one that might help me cope bcttt:r and not tnke
so much Out of me. 1Vlaybe I can write myself as a survivor in a deeper,
more meaningful way, like Art was talking about. You kno\V, I can't help
237
 METHODS OF COLLl:.L \ lNG ANU ANALY LI Nb tMr'IKll...AL IV\/-\I t.t'(.l/-\LJ

\vondtring how other women fcc', years after their treatment. That's what personal realm. I'll also give you a piece I wrote with Art o,n. co-
I want to know-how it feels to them, how they cope ... or don't," she constrUCte d narratIvc,' w Ille J '
' h (escflbes a two-part process 0 f indlVldu- .I
adds, the tears starting up again. "Does the experience continue to be as ally writing stories that arc then shared and co-constructed by severa
fresh and scary to them as it stillls to me? Maybe t can both contribute to participants." ,
knowledge and hclp others-and myself-write a story we can live wlth. "I guess there's no interview schedule then?" Sylvia asks, but since she 5
How I'm living now, denylng my feelings-\vell, this is no way to live." smiling, I don't respond. "How will the chapters' look and where wii! my
"Okay, we'rc getting somewhere now," I say softly. "I think j'ou have story be?" she asks, this time seriouslv.
your topic. I imagine that other women share your sense of vulnerabiHty "Tl1e f arm W1'II evo 1vc d ' t h e research
unng ' process. You m1g, I1t start
~ the
and loss of control over their liyes. I think 1 would," [ add, involuntarily dissertation with a shorr personal storj' to position j10ursdf for the reader,
shivering as I imagine how difficult it would be to have cancer hanging , four
or tell your longer story as a chapter. Or you might integrate parts 0 Y
over me in such an intrusive way. "Nmv hmv do we find out how other experience into each participant's story, each of which could form. sepa-
long-term survivors experience cancer?" ratc cl1apters. 0 r wnte 'your
'1n comparison to one a f t 11C pa rtiopantS
story . ' T
''I'd like to do intensive imerviews with survivors of more than S who is similar to you, as Christine Kicsinger did in her study of eatlng
yeats," Sylvia responds energetically, "and include an African American disorders.
woman-there's so little on their experience of breast cancer-and maybe "P erh aps you W1' Iwnte
I ' eac h i '10 a unique form to re fleer the dif-
c 1npter
evcn a lesbian womall, because I think their experiences might be differ- '
ferent expenences you I1a d'111 each 111terv1ew,"
" , " or to refleer
r contmue, .
ent. Hmv many panicipants viOuld f need? l.\venty-five?" ' b I I
samet hmg a out t le C 1aracter of each woman's story. For - exam pie, 1 a
c

"Oh no," I laugh. "If you're going to do intensive interviews, you'd partici~ant tcl~s h~r story without much input or qu~stiOning.fr01~~ yao:ci
nced only a few, maybe five or six including yourself. You'1I want to inter- you mIght Wrtte 111 the thougbts vou had as you ltstened to he .
vicw each woman a number of times to build trust in the relationship, and , . ' . 1 t wf1tC
reflected on your hfe. If another interview is interactive, you m 1g:l
also so they can read and respond to each transcript before you follow up d'IaI ague to sIlOW the process of communication and interpreta tion that I"
\vith thc next interview." Occurre dtletween you. 1.you "re successful you should not on ly'unm as ,
"How much will I participate?" them and yourself'for others, but, as Har;ld Rosen says, yOU should also
"Given that yOll share aspects of their experience, tbe interviews should discover the face under rhe mask. ,
be an interactive conversation, 1 would think. But you have to play that by "0 r, "I conttnue
' J '
1esltanr Iy,"you could write the dissertatlO
. n , 'IS
co
Elhat
ear. Rather than overlay method onto experience, you want to relate your E lTI the form of a novel. The plot would conSIst
lsner suggests, . a four Y
re -
approach to each wom~n's lifc and think about what would help her to tell search journey. You'd lct readers experience with you your search for
her storv. In some cases, participants wiH feel comfortable having a COI1- u n derstan d'tng, the quesnons
' YOU ask how the \vomen res po nd , what
versntio~, if you setit up th~,t way. Bur as a society, we're so accustomed to , "
thelt answers open up for you, new questions that arise~ and 10W
! you
the authoritative interview situation that some women stilI wiH expect yoU int erpret tl ' stones.
1e1r ' In t Ilat case, you might end by 5 h oWll1g
' hoW. your d
to be the authority and ask all the questions. Some might inquire about ~rories compare and finally how your story changed as yOll rook In ~n t
youI' story; others witt be t'JO glad for an opportunity to tell their o\vo to lllteracted with the other women's stories. You'd have to be carehI ,
pay attention to yours. They'll want you to be tbe researcher and therapist. though, that your story didn't overshadow theirs."
Perhaps a ftvv of the wom,~n will Want to write their stories. Remind me "Yoes, an d It
' would probably bc hard to get my comm1ttee , to bnY . a
next time to give you an article on interactive interviewing that I wrote novel "
with Christine Kiesingct and Lisa TiEmann-Healy, where we had conver- J 'd' u tell the
no 111 agreement, and then remind her "No matter hoW yo . 1
s:ltions over dinner abom eating disorders. It'll get you thinking abat1t Story, t 1le wntmg
" h as to be engaging and evocative.
' That's not 11 oW SOCIa

form and the problems of doing interactive interviews-the time involved scie'llt1Sts have been taught to write. You'll essentially have to Ie arn hoW..
and the emotional commitment and ethical issues of dealing in such a to writ e b y rea d'mg nove.s,
1 an d yb ' .
wnnng ..
and rcwntlng <lOci actnng
I:'

238 239
 METHOD) Ur LULLtL IlillJ Af'lU Al"lALY Lll'1lJ l:JVW1KIl...I-"L MAl CI"IAU
feedback. Of course, I'll provide response, but you might want to consider
joining a writing group as welL"
"That's a good idea," Sylvia responds, jotting down notes as she talks.
"\\lon't I also have to do traditional writing? \Xlhat about analysis, (or
example? \Vill I do grounded theory?"
"\l(1cll, your committee will demand an analytic chapter, you can bet on
tbat. I also think you need one. The article I wrote on stigma convinced me
of the benefits of moving between narrative and categorical knmvledge,
though I don't think that is necessary in every study."
f continue hesitantly, "You could do a straight grounded theory analf-
sis. Then you'd divide chapters by concepts that emerge, or types, or some
kind of category. Or each chapter might represent a stage in the illness
process, like David Karp did in his study of depression. If you choose
grounded theory, you'd need to pay a lot of attention to coding your mate-
rials and comparing and an:1lyzing your data along the way, and you'd
\vritc in an authoritative voice about the patterns you saw. If you choose
this strategy, I'd recommend yOll follow the procedures th:H Kathy
Charmaz describes in the new HalJdbook of Qualitative Research."
"\l(1hat vmuld happen dL~n to the women's stories? And my story?"
"\X1eH, you'd LIse snippet~ from all the stories where they applied in each
chapter. "
Sylvia pauses (or a moment, jots down some notes, and then says
thoughtfully, "1 don't think so. It seems to me that would take away from
the evocative nature of the stories as a whole, which is the value of my
study. Besides, the women deserve to tell their own stories, though [know
I'll influence hm\' they gee told ... "
"I agree," I inrerrupt,celievecl, '"given the nature of your project and
your goals. But just becau~ie we decide to do analysis doesn't mean we have
~o do it traditionally." Sylvia's eyes open wide. "What about inviting all
your participants to read each other's stories and then meet together and
tape-record the discllssion? This could serve as the basis for your analy-
sis-you'd 'ground' the analysis in your participants' understandings, as
well as your own. You might provide your own interpretations for them La
respond to."
"Okay," Sylvia says, leaning forward, speaking passionately. 'Of really
like this idea. I'll invite my participants over for dinner one night. It'll be
my way of doing something for them. Before they come, I'll send them the
stories r wrote about e.xh of the women. Then ... "
"As tong as you get permission," I caution.
240
~'Oh, yes, I know that's important. wlaybe I'll only send them their own
stories." She pauses, then suddenly blurts, "\X!hat if somebody \vams me to
leave out something?"
'"Then you might omit it, or ask your participant to help you rewrite it.
Or you could fictionalize a detail in J way that camouflages the actual
event bur still conveys the meaning you want to get across. Or use pseud-
onyms or composite characters, if that helps."
"I'd also want them to listen and respond to my interpretations. Bur
what if they disagree with my analysis?" she asks suddenly, frowning.
"That can happen, so you have to have some understanding up from
about how you'll handle that. Perhaps you'll put alternative interpreta-
tions, yours and theirs, into the text. Or you could listen to their interpre-
tations without giving them yours."
To provide an example, [ say, "Susan Chase, a sociologist, chose not to
give her analysis to participants to read before publication, though she
asked for permission to use their words and gave them an opportunity to
amend their nartatives. She makes a distinction between what she wanted
to communicate in her analysis-how culture shapes narrative process-
and what her participants wanted to communicate in thci'r narrations-
their life experiences.
"In any case," I continue, "you'llnced to explain in your dissertation
the kinds of decisions you made and on what grounds you made them.yoli
owe that to readers.
"It's a hard balance," I continue, suddenly reminded of readers, "giving
readers the information they expect without betraying the trust of partici-
pants, I mean. As Ruth]osselson says, when we get to the writing stage, we
tend to rake ourselves out of relationship with our participants to form_ a
relationship with teaders. How can wc help then but have feelings of
betraying our participants?
"Oh, and it gets even more complicated," I say to Sylvia, whose hand
covers her open mouth as she shakes her head in disbelief. "We haven't
evcn talked about your family members yet. They may become central
characters in your very personal story. Say your husband or daughter
doesn't want you to reveal things about them or your relationship to them.
What do you do then?"
"Oh, my, I hadn't thought of that," she says quietly. "But I'd have to talk
about my family in order to penetrate the depths of my experience. How
could I ask my participants to do this, if I couldn't?"
241
 "This is one of the most important ethical problems in this kind of re-
1'~,1 abo~t to get started again, when Sylvia says with a twinkle in her
search. Because now \ve're not just talking about faceless, nameless, eye, "So will you be on my committee?"
unidentifiable subjects-if we ever were. Your intimates are idenrifiablc
. "Oniy if you're still planning to do that survey," I say, both of us chuck-
individuals with names. Don't they deserve the same consideration as your ling as we V-lave good-bye.
participants who have given yOll permission to write about them?"
"\X1ell, of course ..."
"Are there any situations in \\-hich the 'greater good' outweighs individ- 4> De/ending and Expanding Autoethnography (One Year Later)
uals' rights to priv3cy, in which you h3ve a right to teU your story even if
other characters in it object?" "Hi, Art, I just had to call."
\X1hen I see the look of defeat on Sylvia's bee, I realiz.e that I am trans- "\X1hy? \X1hat's the matter?"
ferring too many of my own concerns to her roo quickly. "Hey, these issues "I just gOt O~t of Sylvia's proposal defense. Actually it wasn't as bad as I
don't all have to be resolved today. I just wanted you to know that they will expected. I dunk she held her ground. Hey, would vou quiet the dogs)
come up. \X1e'li discuss each aile as it arises and try to make good, ethical I can hardly hear you." . .
decisions." " :'Oh, y~ah. I guess that amibark contraption you bought for Christnlas
Then, before Sylvia has a c'nance to be too relieved, I add, "But by the lsn t workll1g any better than all the other ones we've purchased. Likker
next time we talk, we do have to consider how to get your proposal past Traf, Ande, Sunya-quiet, your mom's on the phone," Art vells l and I'l~
the IRB committee. You'll want to read Michael Angrosino :111d Kimberly am,~lz~_d when t.he y actually .sto~ barking. "So what happen~d?"
IVlays de perez.'s discussion of IRBs in the new Handbaa/::.. You'll have to be \Xell, I started the questIoning and at first it went very well. Commit-
strategic in writing your proposal, because the first thing the comminee tee m~mbers seemed to understand what we were proposing. But then
""",ill ask is about your independent and dependent variables. Then thq"'U .\~rhen It \vasthei.r turn to ask questions, suddenly we moved from talking
want to see a copy of your interview schedule. All the talk of risk and ced~ a.)~l~t the .ex~encnce of breast cancer to talking about bias, validity eli-
ing of responsibility by the university that they'll want you to put into your gIbIlIty ~n~e[la, operationalization, control v3riables, confounding' fac-
consent form, well, that will likely scare away some participants. But \VC tors, bl1l.l~tng models, replicability, and objectivity. In response, I found
myself
.. glVlng long
. sp eec. Iles peppere d Wtt
"11 words lIke
" literature, literar}'
have to go through the pro.::ess to protect the university and ourselves, /ICC/l<;e evoclllV'" I "bl
t:, Vlf lle/a
"
e, lIarmtlue tmth, verisimilitude, interactive
especially since you're dealing with an at-risk population. The board will - , L
and thera/Jentic. " ,
be concerned \vith how you're protecting your p3rticipants-their iJenti~
ties and their well-being. At the least, you'll have to provide the name of a h "Nothing like these forays OUt into the other world to make YOU realize
therapist your participants can see. I don't know how the board would nOw fortunate we are to have created what we have in the Com~unicarion
respond to your telling them that you'\! be the therapist," I laugh. Then I ,eparrmenr, where we take the significance of this work for granted" Art
responds. '
add more seriously, "But redly protecting the participants and your family "That's for" ". Tl "
members-well, in the end that's left up to you and me." I pick up the book d . sure. 1e experIence gave me a lot of empathy for what stu-
ems and .Valing. f-a~u 1ty memb ers"tn other universities may have to go
I was reading when Sylvia Jrrived to indicate our time is up.
" to do tIllS kmd of work.
"1 think I'm ready for those syllabi now. You know, from the courses But," I contint '. 1" "
you've taught." I smile and hand Sylvia the syllabi \vaiting on my desk. In tl I. f-
0 te, samet lIng very Interesting happened near the end
kaecnse I '1' . .
turn, she hands me a folded piece of paper. "It's a poem I v.,rrore about [os~ "h . was lstel1lng to the oncologIst talk about IJrediction and
W en I b I" J" b "
h" d . egan t lln ung a Ollt how tmportant rhese goals lTIU~t seen1-
1-'.WO r,-Ins "
ing my breast. I know it isn't research, but ..."
o" a>lv t c. a Instead of giving yet another speech, I asked
"Of course it's research. Think about including it as part of ~'Our nco ug1St \ 't .. I"k
"II . \ 13t It IS t Te to have to tell women the bad news to deal
Have you read any or Laurel Richardson's ethnographic poetry?" 1 ness and d ' 1 "11 1 ". Before I knew " " "a
eat 1 a t le tIme. tt, we were haVing
242
243
 MUlveU 1IIU!:Jrupny, ,.-ersonal Narrative, Hetlexivity
METHODS 01'- LULLtL III'H..:J Al'll-J hl'lM.L' L.II ' ..... , - , . " " , , - , , -

'., . ~_. ' . s how emotionally difficult his job is, and how "That's vcry dose to how I see the private geriatric care managers I've
conversation about Ie. eJmg, d h f' I "V upser he was yesrerday, been studying for the past 2 years," Art responds. "/\.S they work between
. b' H' t I t e story 0 10 -
he'd like [0 do It etter. e 0 '1 . . 110ther of two young children, long-distanced families and their elderly relatives, they become ethnog-
.
I d to Ie a.) ,- Jl. '4 y'C'lr-o 1L w0111en, a I
when Ilela < I I' d how bad he fdt when raphers of aging_ They aren't academics. They don't do academic research
I ,I I 6 mont lS to lye an
that she probab 1y us ess t 1an I fl' . t' le He had tears in his eyes and they don't write articles. Yet in every other respect they think and act
, df '1-' . up W0l11UC1 0 11S lIT
she apologize or ra " l I 1 g . d R bert Cole's work and he told as ethnographers. In each case they manage, they function as a channel
\I' (] I . ry I mentlone 0 '
\vhen he \vas te lOb t le sto _'.' d I ' d the writings of \X1illiam through which pass the emotional, economic, medical, and social crises
, d TI C if oj Stones an ae mire
j11e he had rea Je a h' I I oals of Sylvia's work relate that must be negotiated by families coping with the contingencies of aging.
. I ' 1 ,I W 1111 lOW r 1e g , .
Carlos \xriliIams. - tflCC to s 10 I . tories to try to figure OLlt how They occupy a unique, dynamic, holistic, and engaged perspecrive. They
' . ' r ClbOLlt lOW \ve lise s
to what Co 1es.. \vas saymg. are participants and observers and their private lives arc deeply affected by
L

fill d i d him where hc lived, at the

' 'fully 1 -e [ 18. reac 1e
to live oLir Ilves meal1lng .' " ljJ 1'1' Ie first pan of the meet- their public and professional services. As sroryrellers and auroerhnog-
. . . . , r. d deep feelmgs, 11 he t 1 - ,
site of hiS subJectI vI [) an 'd r Js and were commUl1Icat- rap hers, they have as much, if not more, to teach us abour the concrete,
"l'X it seemed bot I1 a us la. f l d ,ct own our gUJ L
11 b' b' " everyday details associated with aging as Jo scholars of aging."
ing with each other as human elOgs. " "\Ve like to think we have a lor to teach people in the public sector, bur
L _ . hr. I: ee'l some moment.
'\Y./ow, that must 3\ e 1 _ . f i t wo. do To figure out they have a lot to reach us as \vell, if \ve just listen," I add. Yet I know,
"ft was. You know, I think this is the hll[Ll~e to 'tvl1:apra:tiC~1 con~exts of JS Elliot Eisner discusses, that it will be difficult to wean scholars and the
, I - -. . I c,thnograp 1Y m 0 -
how to tntroc.uce persona - I _ Id b enhanced bv it like doctors, American public from a vie\v that measuring, comparison, and olltcomes
'f - - I vhosc'wor"wo u
everyday II e, to peup e \ . -. _ ,~ ,I" " . '
e '- arc all that marter."
'I . IT . admll11strators, and teae letS.
nurses, SOCia wor "ers, '- . d I fight lately'" Art responds. "Bur I think we're slowly knocking down some of the waJls," Arts says
" 1 '. e a great ea 0 t lOU ,
'Tve been glvmg t lat ISSll , . 1 l ~'ond rhe encouragingly. "We've opened a space to write between traditional social
. from exrending all of cthnograp lY Je)
"I th'mk rhere's a lot to gam .. - f, l ' 'elY"m academic practice. science prose and literature and to stimulate more discussion of working
1 'n-mg 0' It as exc USI\ '-
academy so rhat we stop t 11"- . I " '. d helping professions the spaces between subjectivity and objectiviry, passion and intellect, and
I f people In t le serViCe an
Couldn't the \vor ( 0 many d I ' , I, effectively service work~. autobiography and culture. Look at all the book manuscripts we've
h '11'/;1 To 0 t 1elr wor" , .
be rhought of as et nogr~p '.' I ' d ' . 'n contexts that crosS the received to review for our series, I rake that as strong evidence that more
. . , bjectlve une erstan mg I I
erS have to gaIl1 mterSU I d rlllliciry I mean any time t 1C and more academics think it's possible to write from the heart, to bring the
, f . bT t nce c ass an c ' I
boundaries 0 age, a I I y, <' '1' , 'legr ec of intercultur,., first-person voice into their work, and to merge art and science. I don't
~ \ d ends on deve 0Pll1g some c .I
success ot your war.;: ep I ' 1 1-' II we associate WIt 1 think there's any danger of going back to the way it used to be, not in our
'I have to use t le SOCia 5 d S . )
understanding, t len y o u . I -hotherapist s do thiS. lifetimes anyway."
, I ic empJ.th)'. Wouldn't you say nat psyc
et Illlog r ap 1 (;I" "'Yes, we've encouraged writers to make ethnography readable, evoca-
"ren't they ethnographers of the scI. . . 1 to" I add. tive, engaging, and personally meaningful. And it's working, Autoethnog-
n - \ d d :"eaching involves ethnograp ly 0 , .' . ,
"Of course. j- n goo - '- h I I b rriers of unfamI!tant), ~aphy is being read widely by graduate and undergraduate students. Now
I' l'our way t roug 1 t 1e a d' 0
"Over time you trY. to war , , - .d ' '. f I IaboratIon ' u 11clerstal1 In!:" :t':; time [0 show its usefulness in the public realm. It's interesting that this
distance, and differen c.e towar d3 SPlrJ.t ~ c~ 1 w/hen \:re learn hoW [0 IS the same argument being made in some of the mainstream sociology
- . nee an partlClpatIOl. V\ d P
and openness to expene 1. .-. 1 it easier to rO Have you seen the symposium in Contemporary Sociology on
open ourselves to ourse ve.., a 1 ". nd to each ot ler, we Ilne
. I' 1 [r.hi<: 1S wor (illf..!
I' "
II put:>l;c" ;11 :;Qciul clialut5uc';"' 1 il:;lc
resistance \ ' [[~..
crcIlt .'lL-t t:d::i.
.- ' I. like to tun '- 0
some a f Ollf ' "
[0 LI
, '1 1 11 that IS
. Yes , 1rca d'It yesterday. Perhaps our purposes are coming together for a
hic consciousness m t lC C assroo]
tOW areI an etllnograp . '- After all, that's what we are trying to do in the Ethnographic
intimate, and empathic."

245
244
 METHODS OF COLLECTIHG AND Af'JALYZII~G EMPlHlLAL !v\AIl:I\.IAl~ MULU\;,lIIIIUIjIUPIIY, ,f:HSOnOl l'IOrrOtlve, KenexlVlty

Alternatives series, to publish books that say something meaningful and Atkinson, I~ (1997). Narrative turn in a blind aUev? Ol/alilotiue Health Resem-cfJ 7
325-344, , - " ,
:J.ttraCt a wide audience."
"Like l\/likc Angrosino's stories of adult mental illness. Now we just Austin, D. (1996). Kaleidoscope: The s~une anel different. In C. Ellis & A. T:
need to get the book into the hands of those who work with and make poli~ Bochner (Eds.), COI1l!Josil1g ethnography: Aftemative (arms of qllalilative
cies aboUt mental illness." writing (pp. 106-230). Walnut Creek, CA: Altalvlira.
"1 don't know if this Handbook piece will help with that," Art says. "But Austin, D. (1998). Understanding close relatiolfshijJs .lJIlOng Afrin11l Americans ill
it may encourage more people I:Q do autoethnography and help legitimate the context of the Million AJan March. Unpublished doctoral dissertation,
University of South Florida.
this approach for those swden:s and young faculty members you're wor-
ried about. Those arc important goals we've tried to achieve in our chap- Bakhtin, M. M. (1981). The dialogic imagination: FOllr essays (M. Holquist, Ed.;
fer. I'm glad \ve've written it." 11,'1. I-1olquist & C. Emerson, Trans.). Austin: University of Texas Press.
"[vIe, too," I say, smiling as I think of Art's initial resistance to writing Banks, A., & Banks, S. l~ (Eels.). (] 998). Fiction and soci.J! rcse'lrcJJ: By ice or /Ire.
Walnut Cteek, CA: Alralv'lira.
for the Handboo!:.:..
Banhes, R. (1977). Image, lI11/sic, text (5. Heath, Trans.). New York: Hill & Wang.
"So, then, we're finished with this piece?" Art asks.
"Looks like it," l reply. "Let's reward ourselves and go to the beach for Behar, R. (1993). Tim/slated WOman: Crossing the border wit/; Esperm/::a's storv.
B05ton: Beacon. -
the vveekend. \'('e need to gct out of our offices and engage in some other
Behar, R. (1996). The vulnerable observer: Anthro!Jology thaI brca/.:.s YOllr heart.
life cxperiences, or else the only thing we're going to be able to write about
Bosl'On: Beacon.
is writing,"
Behar, It, & Gordon, D. A. (Eels.). (1995). \Y/olllell writing Cltllnre. Bcrkelev: Uni-
"But what about the chapter for the Handbook of Loss alId IhllllJUl and
versity of California Press. .
the one for the I-landbook of Iilterpersondl Commllllicdtioll we just C0111-
Benson, P. (Ed.). (1993). AnthrojJology ill/d literature. Urbana: University of Illillois
mitted to do? \x!c really should get started," An says. Press.
"An!" I yell, as I hear simultaneously his laughter and a knock on my
Berger, B. (Ed.). (1990). Authors of their Olun lives: Imellectllaf mllobiogmphies 61'
office door.
ftventy American sociologists. Berkeley: University of California Press. .
Berger, L. (1997). Between the candy store and rhe mall: The spiritual loss of a
fathet.]ollrnal of Persol/al and Interpersonal Loss, 2, 397-409.
<>- References Berger, L (1998). Silent movies: Scenes from a life. InA. Banks & S. I~ Banks (Eds.),
Fiction and SOci,1l research: By ice or fire (pp. 137-146). Walnut Creek, CA:
Abu-Lughod, L. (1993). Wlritfng l/I01l1ell'S worlds: Bedouin stories. Berkeley: Uni- AltaMira.
versity of California PreS5. Bochner, A. l~ (1981). Forming warm ideas. In C. \XTilder-Mort &]. H. Weakland
Adler, EA.; & Adler, P. (1987). iH'!/IIbership roles ill field research. Newbury Park, (Eds.), Rigor alld imaginatioll: Essays (rom the legaC)' o( Gregory Brltesoll
CA: Sage. (pp. 65-81). New York: Praeger.
Adler; l~ A., & Adier, P. (1994). Observational techniques. In N. K. Denzin &)~ S. Bochner, A I~ (1994). Perspectives 011 inquiry II: Theories and stOries. In M. L.
Lincoln (Eds.), Hmulbool:. of ql/alitaliue rese{/r~h (pp. 377-392). Thousand Knapp & G. R. l'\ililler (Eds.), Hondbook of interpersonal COl1llll11llicatioll
Oaks, CA: Sage. (2nd ed., pp. 21-41). Thousand Oaks, CA: Sage.
Allison, D. (1994). Shill: Tal!~ing about sex, class, and literature. Ithaca, NY: Bochner, A. P. (1997). It's about time: Narrative and the divided self. Qualitative
Firebrand. IlIquiry, 3, 4'18-438.
Angrosino, M. (1990). apportlllll,) HOllse: f..l1mograplJic stories or melltal n:/iIrda- P."'d
lll
t::
L
,1\. P., LX. ellis, C. (1:.-':.-'2). Personal narrative as a social approach to inrer-
tiCJI/, \Valnllt Creek, CA: AltaMira. personal communication. COlll/llllllicatioll Theory, 2, 165 -172.
Anzakhia, G. (1987). Borderlands/fa frolltem: The new mestiza. San Fr,mcisco: Bochner. ' A r , l:S:: Ell'I SC
,.
(19')9)
.
'"'I' I -'
. W HC 1 way to tllrn~ Jotlmal of COJltelllpOl'dl)1
Aunt Lute. Ethnography, 18, 500-509.

246 247
 Mt! t-IUUJ UI- l.ULLl:l.lll'l\.;l Pd'lLi P.1'lI-\Ll LII'I\.:I CI~\r'''''-f"\l..- ,"M' l..-"'f"\l..--'
Bochner, A. l~, Ellis, c., & Tillmann-Fealy, L (1997). Relationships as st'Ories. In
S. Duck (Ed.), Handbook of persoT/al relatiollships: Theory, research il/ld
illlerucllliolls (2nd ed., pp. 307-324). Nc\". York: John Wiley.
Bochner, A. P., Ellis, C, & Tillmann-Healy, L. (1998). Mucking around looking for
truth. In B. Iv1. .1vlomgomcry & L. A. Baxter (Eds.). DialectiC/I! approaches
10 studying personal relationships (pp. 41-62). rVlahwah, NJ: Lawrence
Erlba 11 Ill.
Bochner, A. P., & \Vaugh, J. B. (1995). Tulking-with as a model for writing-ubout:
Implications of Rortyean pragmatism. In L. Langsdorf & A. R. Smith (Eds.),
Recovering pragl1/dtism's uoice: 'The c1assicol tradition, norty, il/ld the philos-
ophy of COJllmllniwtioll (pp. 211-233). Albany: State University of New York
Press.
Bowen, E. S. (1954). Retllm to lal/ghter: An anthropological novel. New 'York:
!--I:uper & l~ow.
Brady, I. (Ed.). (1991). Anthropological poetics. Savage, !\'1D: Rowman &
Littlefield.
Brandes, S. (1982). Edmographic <1llt'Jbiographics in American anthropology. In
E. A. Huebel, R. Currier, & S. Kaiser (Eds.), Crisis ill artthropology: View
frolll Sprillg Hill, 1980 (pp. 181'-202). Ne\v York: Garland.
Brny::trJ, A. (1992). fJltoxicated by /Il} illlIess alId other writillgs Oil /ife alld death.
New York: Fawcett~Columbine.
Bruner, J. (1990). Aels of meanillg. Clmbridge, MA: Harvard University Press.
Bruner, J. (1993). Introduction: The ethnographic self and the personal self. In
[~ Benson (Ed.), Anthropology alld literature (pp. 2-26). Urbana: University of
Illinois Press.
Buford, B. (1996,]une 24). The seduC[ions of storytelling. Nelu Yor/{.cl; PP' 11-12.
Butler, S., & Rosenblum, B. (1991). ('ulIcer ill tlUO voices. San Francisco: Spinster.
Carr, D. (1986). Tillie, lIarrative, md history. Bloomington: Indiana University
Press.
Charmaz, K. (1991). Good days, bad days: The self in chrolIic iIIlless and tiJl/e. New
Brunswick, NJ: Rutgers University Press.
Chase, S. (1996). Personal vulnerability and interpretive authority in narrative.re~
search. In R. Josselson (Ed.), Ethics alld process itl the /wrratiue study of hues
(pp. 45-59). Thousand Oaks, CA: Sage.
Church, K. (1995). Forbidde1/ lIarratives: Critical autobiography as social scicnce.
Newark, NJ: Gordon & Breach. . Dumom, J-I~ (1978). The headl1ulIl a}/d I: Ambiguity al1d ambiualel1ce il1 the field-
Clandinin, D. J., & Connelly, F.1v1. (1994). Personal experience meth~ds. In N/~'
Dem:in IX)~ S. LlI1co[n (td5,h J-imu/lJook or'luafffulItJc lc:.,c<!IJ.--h (pl-'. -\
427). Thousand Oaks, CA: Saf;c.
Clough, P. T. (1994). Fem1tlist thought: Desire, power (Iud academic discourse. Cam-
bridge, IvlA.: Blackwell.
248
r;Ulut:tlII 'U':::JI UIJI Iy, ,I;:I",UI lUi l'IUrrU[lve, Kenexlvny
Clough, E (1997). AutotelecoIl1ll1unication and alltoethnography: A reading of
Carolyn Ellis's Fillalnegotiatiolls. Sociological Quarterly, 38, 95-110.
Cohen, A. (1992). Self-conscious ':mhropolo!:,'Y. In J. Gkely & H. Callaway (Eds.),
Amhropology I1ml cllltobiography (pp. 221-241). London: Routledge.
Coles, R. ("1989). "l1Je call of stories: Teaching alld the moral imaginatioll. BostOn:
Houghton Ivlifflin.
Collins, l~ H. (1986). Learning from tbe outsider within: The sociological signifi-
cance of black fcminist thought. Social Problellls, 33, 14~32.
Cook, J. A., & ronow, M. M. (1986). Knowledge and women's interests: Issues of
epistemology and methodology in feminist sociological research. Sociologi-
cal lllqllil}'. 56, 2-27.
Couser, G. T. (1997). Recovering bodies: Illness, disability, al/d li{e luriling. IvIadi-
son: Univcrsity of Wisconsin Press.
Crap;::mzano, V. (1980). nJlwllli: Por/rilit of a Moroccall. Chicago: University of
Chicago Press.
Cr;nvford, L (1996). Person;:!1 ethnography. COlllmunication lvlOllograp!Js, 63,
158-170.
Crites, S. (1971). The narrative quality of experience. JOllrnal of the American
ACi1deJll)' of Religion, 39,291-311.
Deck, A. (1990). Amocthnography: Zora Neale Hurston, NoniJabavu, and crosS-
disciplinary discourse. Black American Literature For;I111, 24, 237-256.
Denzin, N. K. (1989). Interpretiue biography. Newbury Park, CA: Sage.
Denzin, N. K. (1997). IJlterpreth'e etlmogmphy: EtlJllogmphic practices for the 21st
eellfllry. Thousand Oaks, CA: Sage.
Dcnzin, N. K., & Lincoln, Y. S. (Eds.). (1994). Halldbook of qualitative research.
Thousand Oaks, CA: Sage.
Dtrrida, J (1978). Whtillg and difference (A. Bass, Trans.). Chicago: University of
Chicago Press.
Derricb,]. (1981). Positiolls (A. Bass, Truns.). Chicago: University of Chicago Press.
DeVault, 1\'1. (1997). Personal wriring in social science: Issues of production and
interpretation. In R. Hcnz (Ed.). Reflexivity and voice (pp. 216-228). Thou-
s.1nd Oaks, CA: Sage.
Diamond, S. (1982). Totems. Barrytown, NY: Open Book/Sration Hill.
Dumont, J.-I~ (1976). Ullder the raillbow: Natllre and supematllre among the
H1Ilare India!1s. Austin: University of Texas Press.
,. lUorking eX!Jeriellce. Austin: University of Texas Press.
y b';"cr. E. \1"'0'0). 51loulu a novel counr as a disstnarion in euucaUOl1: l(CJeCm;11 in
.. the Teaching of English, 30, 403-427.
hiller, .E (1997) "fllC new trontlcr
' . . .111 qualitative
. . research methodology. Qua /"Ita-
lIve Inquiry, 3, 259-273.
249
 METHODS Ur LULLtL I Ir'll.J ANU /.l,NALl LlI'Iu t:.IVWIKI\...HI...IV1MI Lf'II"1I....!
1'.11 . L. (199)
,mgson, ... ~ <) "Then \rou know how 1 feel": Empathy, identification, and
reHexivirv in fieldwork. Qllalitative Illquiry, 4, 491-514.
Ellis, C. (1986).' Fisher folk: 'Two (.olJlI1l1t1lities Oil Chesapeake Bay. Lexington: Uni-
versirv Press of Kentucky.
Ellis, C. (19'91a). Emotional soci,)logy. In N. K. Denzin (Ed.), Stlld~i!s ill.symbolic
illteractioll: A rese1rdJ all/llwl (Vol. 11, pp. 123-145). Greemvlch, CT: J1\1.
Ellis, C. (1991b). Sociological in:rospecrion and emorional experience. Symbolic
Illtcractioll, 14, 23-50.
Ellis. C. (1993). "There arc survivors": Telling a story of sudden death. Sociologic.11
Qlfarterlv, 34, 71 1-730.
Ellis, C~ (1995;). Emotional and ethical quagmires in returning to the field. jOl/mal
ofCollterll!JOfi.11)' EtJmogri1plry, 14, 711-713.
Ellis, C. (1995b). Fillalnegotiatiol/s: A story of 101'e, loss, and cfJrullic illJ/ess. Phila
delphia: Temple University Press.
Ellis, C. (1995c). The other side of the fence: Seeing black and \vhite in a small,
southern town. OIli.J!itatiu? [Ilquiry, 1, 147-167.
Ellis, C. (1995d). Speaking o[ dying: An ethnogr~phic short story. Symbolic Illtcr-
actioll, 1S, 73-81.
Ellis, C. (1997). Ewcative autoe::hnography: Writing emotionally about our lives.
In \X!. G. Tierney & Y. S. Lincoln (Eds.), RejJrcselltatioll and the te.':l: Re-
ji'mllillg the lIan':1tiue 1'0iCI' (pp. 115-142). Albany: Srate University o[ New
Yorlc Press.
Ellis, C. (l998a). Exploring loss through autoethnographic inquiry: Auroethno-
graphic stories, co-constructed 113rratives, and interactive inrervie~vs. InJ. ~.
Harvey (Ed.), PerslJectiues OIl loss: A sourcebook (pp. 49-62). Plllladdphw:
"ElVlor & Francis.
Ellis, C. (1998b). "1 hate my voice"; Coming to terms with minor bodily stigmas.
Sociological Quarterly, 39,517-537.
Ellis, c., & Bochner, A. P. (1992) Telling and performing personal stories: The con-
str~ints of choice in aborcion. in C. Ellis & M. G, Flaherty (Eels.), [nues-
tigatil/g subjectiuily: Research Oil lived experiel1cc (pp. 79-101). Newbury
Park, CA: Sage. , at rhe annual meeting of the \Vestern Stares Communication Association, San
Ellis, c.,
& Bochner, A. P. (Eels.). (1996a). ComjJosing et.lmography: Alternall1Jc
forms of qualitative rurif.ing. Walnut Creek, CA: Alra~vlira. . Gravel, A. ("1997). The ideological fUJlctiolls of I1Iltoetlmography. Unpublished
Ellis. c., & Bochner, A. P. (Eels.). (1996b). Taking ethnography into the twenty-hrst
. century rSpccial issueJ.Jo!mral ofCOl1tcmpomry EtlmogralJhy, 25(1).
Ellis, c., & Flaherty, Iv1. G. (Ed.. ). (1991). bwcstigating subjectivity: RB.search 011
(Ivn( cXpCnCIIGG, Ncwuuq filrh.}'""\; :;ii.ln\;' . '''O"h''O' L \l//j), l\eWTllIlIg
Ellis, c., Kiesinger, c., & Tillmann-Healy, L. (1997). Interactive interviewll~g~
.
Tllking abOlIt emottona 1experience.
. I11 R H ertz
. (Ed) ~rl"xil'ity
. , l'\(o/Jr;. ~ ami I/OIC e
(pp. 119-149). Tholl sand Oaks, CA: Sage.
250
,-,UlUt't1I1IUY'UtJ1'Y, r'l:fS(JrJU! I'lanot/ve , KelleXlvlty
Engaging publics in sociological dialogue [Symposium]. (1998). COl/temporary
Sociology, 27, 435-462.
FOllclUlr, J\A. (1970). The order of thillgs: Au archaeology ol the J.IIIIIWII sciellces.
New York: Random Huuse.
Frank, 1\. (1991). At the wi!! of the Dody: Reflections 011 illness. Buston: Houghton
Mimin.
Frank, A. (1994). Reclaiming an orphan genre: The first-person n:ur:uive of illness.
Litertltlfre alld Medicine, 13, 1-21.
Frank, A. (1995). The uJollluled stol)Jteller: Body, illness, alld ethics. ChicJgo: Uni-
versity of Chicago Press.
Freeman, IvI. (1993). Rewriting the self: History, lIIell1ory, Ilarmtiue. London:
Routledge.
Freeman, I'v1. (j 998). 1Vlythical time, historical time, and the n~lrrative fabric of self.
Narrative Inquiry, 8, 17-50.
French, D. (1998). Through the eyes of the comic l11i1sh: AlI etlmogralJhic eX/110m.
tiolI of the identity of a stand-lip colllediall. Unpublished docloral disseru-
tion, University of South Florida.
Friedman, N. (1990, Spring). Autobiographical sociology. Americall Sociologist,
21, 60-66.
Gadamer, H.-G. (1989). Thlt/; alId method (2nd rev. ed.; J. Weinsheimer & D. G.
IVlarshall, Eds. & Trans.). New York: Crossroad.
G,~ertz, C. (1973). l1Je iJlferpretation of wltmes: Selected essays. New T()flc: l3<lslc
Books.
Gergen, K. J. (1994). Realities i1!u! relatiollships: Soulldings iJlsocial co/lstmctioll.
Cambridge, iv1A: Harvard University Press.
Glassner, B. (Ed.). (1997). Qualirative sociology as everyday life [Special issueJ.
Qualitative Sociology, 20(4).
Goetting, 1\., & Fensrermaker, s. (1995).lndiuidllal voices, collective visiolls: Fifty
years of IUOl!tel1 iII sociology. PhilndeiphiJ: 1emplc University Press.
Goldman, A. (1993, Fall). Is that what she said? The politics of collaborative auro-
biography. Cllltllwi Critique, pp. 177-204.
Gonzalez, M. c., & Krizek, R. L (1994).1IldigeJlolIs ellmography, Paper presenred
Jose, CA.
manuscript, University of \X'isconsin-Whirewarer.
Gubrill!l1, J. E, & Holstein, J. A. (1997). The new language of qlh1litatiue method.
Nt:\V York: Oxford Univt:f::iiry Frei,;:l,
We 50111: MIIltiPie pmolloll!y ilnd the scienLes of
memory. Princeton, Nj: Princeton University Press,
Harding, S. (1991). \\7hose science? Whose hllowledge? ThillJ"illg ti-oTll women's
lives. ithaca, NY: Cornell University Press.
251
 Iv\!: IHUU::. Ut- l.ULLtL 11I'lb AI'lU f-\I'O-\L1 LII'I\.:l l:.IVI,Ir\I ....r\L IVI,....' 1..-1'... ,.."-.. ..>
HarringlOn, \\~ (1997). IntillJa/1.! ;ollmalislll: The art and craft of reporting everyday
life. Thousand Oaks, Ci\: Sage.
I-hrtsock, N. C. J'vL (1983). The feminist standpoint: Developing the ground for a
specifically feminist historical materialism. In S. Harding & M. B. Himikka
(Eds.), Discouering reaIi.~y (pp. 283-310). Amsterdam: D. Reidel.
Harvey, J (Ed.). (in press). Handbao/<. of loss and trallma. London: Taylor &
Francis.
Hawkins, A. H. (1993). RecOilstrJIcting illness: Studies ill pathogmphy. \Xlest La-
fayette, IN: Purduc University Press.
Hayano, D. 1''/1. (1979). AUlO-ethnography: Paradigms, problems, and prospects.
[-[tllltaH Organization, 38, 113-120.
I-byann, D. Iv1. (1982). Poher faces. Berkeley: University of California Press.
l-leidegger,lvI. (1971). On the toay to language (P. D. Hertz & J. Stambaugh,
Trans.). New York: Harper & Row.
Herndon, C. (1993). GClldered fictiolls of selfand com/mlllity: Autobiography and
iwtoetlmography ill Cmibbemlluomen's writing. Unpublished doctoral dis-
sertation, University of Texas at Austin.
Hertz, R. (Ed.). (1996). Ethics, reflexivity and voice [Special issue], Qlwlitatiuc
Sociolog)',19(l).
Hertz, R. (Ed.). (1997). Refle:-:iuity and unice. Thollsand Oaks, CA: Sage.
Higgins, P., & johnson,]. (Ecis.). (1988). Personal sociology. New York: Praeger.
Hallway, \Y./. (1989). Subjectivity alId method in psychology: Gende/; meaning and
science. London: Sage.
Hurston, Z. N. (1991). Dust l'achs 0/1 a road. New York: HarperColiins. (Original
work published 1942)
jackson, A. (Ed.). (1987). Anthropology at home. London: Tavistock.
Jackson, i'v1. (1989). Paths toward a clearing: Radical empiricism alld ethl10gmphic
illquiry. Bloomington: ::ndiana University Press.
Jackson, Ivl. (1995). At home in the world. Durham, NC: Duke University Press.
jago, B. (1996). Postcards, gh,)sts, and fathers: Revising family stories. QlIalitatiue
If/quil}l, 2, 495-516.
jones, S. H. (1998). Kaleidoscape noles: Writing womcn's m1lsic and orgal1izatioll al
wItllre. Walnut Creek, CA: Altal\1ira.
jossdson, R. (1996). On writing other people's lives: Self-analytic reflections of a
narrative researcher. In R. josselson (Ed.), Ethics alld proccss ill the /wrrdtiuC
stlldy of lilies (pp. 60-71). Thousand Oaks, CA: Sage.
Jules-Rosette, B. (1975). Afrirml apostles. Ithaca, NY: Cornell University Press.
KdrF'l D. (19~)0). Sf"""kiI<8 o(;:;,ldI1O:::;:;. New York; OXfOl-d Univcn;ity P,-",,,,,.
"
Kerby, A. (1991). Narrative arl d t IJe se If.. Bloomington: I n d"lana U" " y Press
TllverSlt ..
Kiesinger, C. (1995). Anorexi: alld bulimic lives: Mak.ing sellse of food alld eatmg
Unpublished doeroral dissertation, University of South Florida.
252
''U'-VC:;UJJ,V'd'Vp"y, I t::1;:'UI lUI I~UIIUllVt:::, 1\~IIf;'XIV/[Y
Kiesinger, C. (1998a). From inrerviewing to story: Writing Abbie's life. QIlt1Iihrtiue
Illquiry, 4, 71-95.
Kicsingcr, C. (1998b). Portf3.it of an anorexic life. In A. Banks & S.I~ Banks (Eds.),
Fictioll and social research: By ice or fire (pp. 115-136). \X1alnur Creek, CA:
AltalVlira.
Kinkaid,]. (1988). A sll/all place. New York: Penguin.
Kleinman, S. (1997). Essaying the personal: Making sociological stories stick.
Qualitative Sociology, 10, 553-564.
Kolker, A. (1996). Thrown overboard: The human costs of health C:Irc rationing. In
C. Ellis & A. P. Bochner (Eds.), Composing etlmography: Altenwtiue forms 0/
qUillitative IvritilIg (pp. 132.. 159). Walnut Creek, CA: AltaMira.
Kondo, D. (1990). Crafting selves: POIuel; gendcl; and discourses ofidcntity ill aJap-
anese worl~/J!ace. Chicago: University of Chicago Press.
Kricger, S. (1933). The mirror dance: Identit)' ill a women's commlmity. PhiLldcl-
phia: Temple University Press.
Krieger, S. (1984). Fiction and social science. In N. K. Denzin (Ed.), Studies ill sym-
bolic interaction: A research (lImlta! (Vol. 5, pp. 269-286). Greenwich, CT:
JAI.
Krieger, S. (1991). Social sciellce and thescI( Persona! essavs 011 i..1lI Lut form. New
Brunswick, N.f: Rutgers University Press. -
Krieger, S. (1996). The fall/ily silver: Essays 011 relatiollships among womell. Berke-
ley: University of California Press.
Kuhn, T. (1962). The stl'1lctltrc ofscientific reuolutions. Chicago: University of Chi-
cago Press.
Kulick, D., & Willson, M. (Eds.). (1995). Taboo: Sex, identity and erotic subjectiv-
ity illlllIthrofJological fie!dworh. London: Routledge.
Lather, r>. (1986). Research as praxis. I-laruard Educational Reuien', 56, 257-177.
Lather, p. (1993). Fertile obsession: Validity after poststructuralism. Sociological
Quarterly, 34, 673-693.
Lcjeune, p. (1989). On autobiography (K. Leary, Trans.). Minneapolis: University of
Ivlinnesota Press.
Lnvin, E., & Leap, \Y./. L (Eds.). (1996). Out in the field: Reflections of lesbian and
gay I1IllhrofJOlogists. Urbana: University of Illinois Press.
Lieblich~ A. (1996). Some unforeseen outcomes of conducting narrative resear~h
With people of one's own culture. In R. Josselson (Ed.). Ethics and process m
the l1i..lrmtiue study of lives (pp. 151-184). Thousand Oaks, CA: Sage.
Linden, R. R. (1991). lHahllS stories, lI/i.1king sclues: Eeminist reflectioll s all the
. l-lulocmrsr. COILll11bu:;; Uhio 5rare University Press,
L1onnet, F. (1989). Auroedmography: The an-archic style of Dust tracks 011 a road.
In E Lionnet, Autobiogl"ll/Jhical voices: Race, gender, self..portmitllre (pp. 97-
129). Ithaca, N'{: Cornell University Press.
253
 MI: I HUU::' Ul- LULLl:Ll Ir~1.:J AI'<U f\1'lJ-\LI LWH..:J C1V\,'''-' .... ,....,L
Lopate, E (1994). Introduction. In e Lopate (Ed.), The art oj"the personal essay: All
i-Il1tIJO!of!,Y from the c!t1ssica! en' to the present. Garden City, NY: Anchor.
Lorelc, A. (1980). The callcer jOIJrlwL. Argyle, Mr': Spinsters.
Lyorard, J.-F. (1984). The {lostlllodem condition: A report 011 !mowledge (G. Ben-
nington & B. Ivbssumi, Trans.). I'vlinneapolis: University of Ivlinnesota Press.
i\;Iadnryre, A. (1981). After virtue: A study in mom! theory. Notre Dame, IN: Uni
versity of Notre Dame Press.
l\'1airs, N. (1986). Plain text. Tucson: University of Arizona Press.
1viairs, N. (1990). CIrIIII! acts: ESSdy';, New York: Harper.
1v13i,s, N. (1998). W!elist-high ill the lI,orld: A life iII/long the lIondisdlJled. Boston:
Beacon.
lVlalinowski, B. (1967). A diary in Ihe strict sellse of the term. (N. Guterman,
Trans.). New Yotk: Harcourt, Brace & \X!odd.
Ivlarcus, G. E., & Fischer, 1\,;1. M. J. (1986). Anthropology as ClIltlln11 criliqlle: An
e.yperimelltallllOmenl in the hll'lIan sciel1ces. Chicago: University of Chicago
Press.
tviarkham, A. (1998). Life (1IIIille: Researching real e.yperience ill uirtllol space. \X'al-
nut Creek. CA: AlraMira.
[vlehan, H., & \\;ood, H. (1975). Thc "cality oletimolllethodology. New York: Juhn
\Vilcy.
1\'lert'on, R. (1988). Some thoughts all the concept of sociological autobiography. In
J'v1. Riley (Ed.), Sociological rues (pp. 17-21). Newbury Park, CA: Sage.
(Original work published 197:~)
Miller, L, & Taylor,.J. (1997). Performing autobiography: Editors' introduction.
Text and Pcrforl1uwce Quarterly, 4, v~vi.
Jvlillcr,1V1. (1996). Ethics and unders:anding through interrelationship: I and thOll
in dialogue. Jn R. Jossclson (Eel.), Ethics and process ill Ihe Ilarmtiue stltdy of
lives (pp. 129-147). Thousand Oaks, CA: Sage.
L\,link, L (1969-1970). HislOry and fiction as modes of comprehension. New Liter-
dll' Hislory, 1,541-558.
1Vlorzafi~HaHer, P. (1997). Writing birthright: On native anthropologists and the
politics of representation. In D. Reed-Danahay, Alftolctlmogmphy: 1{e-
writing the self and thc social ;:pp. 195-222). Oxford: Berg.
1Vlubia, T. (1989). A call for our language: Anorexia from within. \Y/omen's Studles
IJllerl/alional Forlflll, 12,613-638.
Narayan, K. (1993). How native is a "native" anthropologist? American A/lthmpol-
oC;isl, 95, 671-686.
Neun13:1t1, M. (1996). Collecting ourselves at the end of the century. In C. Ellis &
1\. t Docllllcr \Eds.h COIHposill8 ctlJllograplJy: f\(rcnratil/c {un II:;; o('I'IC,{;tct-
tiue writing (pp. 172-198). \'(!alnue Creek, CA: AltaMira. . \\7. \'IZ Norron.
Noddings, N. (1984). Caring: A {ell/inine a!Jproach to ethics al1d moral edllCtltlOl I.
Cerkeley: University of California Press.
'VI,...." '-'~E""''-..J
f-\uroernnograpfly, fJersona! Narrative, !,;etlexivlty
-------------------
Oakley, A. (1981). Interviewing women: A contradiction in terms. In H. Roberts
(Ed.), Doing (ell/illist research (pp. 30-61). London: Routledge & Keg an
Paul.
Oakley, A. (1984). Tahing it like a tuolllalt. New York: Random House.
Ohnuki-Tierney, E. ('1984). Native anthropologists. American Etlmologisl, 2, 584-
586.
Okely,j. (1992). Participatory experience and embodied knmvledge. In]. Okely &
H. Calbw<lY (Eds.), AII/hropology al1d alftobiography (pp. 1-28). London:
Routledge.
Okely, j., & Callaway, H. (Eds.). (1992). Allthmllology .71/(1 autobiography. Lon-
don: Routledge.
Parilli, J. (1998, July 10). The memoir versus the novel in a time of transition.
Clmmicle of Higher Educatio/!, p. 1\40.
Park-Fuller, L (1998). IlItmdlfctioJf to session all mguing the shifting slJ1pes all d
, r
spew(/ng tOllgucs a IwtoetimograplJlc (Jerforlllallcc. Paper presenre d 'at the
annual meeting of the National Communication Associmion, Ncw York.
Payne, D. (1996). Amobioiogy. In C. Ellis & A. P. Bochncr (Eds.), Composing efh-
. IJy: .f\1tematwe
IlOglt1p . forms
r r o qllil IItalwe . . (pp. 4')_ -1_
WrIt/J/g -,-) . \XI"I"u[
Creek, CA: Alta Mira.
Personal Narratives Group. (1989). III/erpl'etillg women's lives: Feminist theory
and persollalnarmliues. Bloomington: Indiana University Press.
Pran, M. L (1994). Transcu!turation and auroethnography: P~ru 1615/1980. ~:l~
F. Barker, I~ Holme, & IvI. Iverson (Eds.), Colonial discollrse/postco{Ol/lll
Iheory (pp. 24-46). IVIanchester: l'vIanchester University Press.
Quinney, R. (1991). jOllme}' to a far place: AUlobiographicai reflectiolls. philadel-
phia: Temple University Press. .
Rabinow, p. (1975). Symbolic domillalion: Cultllral (arm mid historical change til
Morocco. Chicago: University of Chicago Press
I'\.<1 I)1Il0W,
. p.. (1977). Re,lectioJlS
( all {ieldll/Ork ill ivlorocco. Berkeley: U I11versit) of
California Press.
Reed-Danahay, D. (1997). Alftoletlmogmphy: Rewritillg the self and the social.
Oxford: Berg.
. R.einharz, S. (1979). all becoming a social scientist: Fi'om survey resem'ch and par-
ticipation to eXI}erielltial analysis. San Francisco: Jossey-Bass. .
Reinharz, S. (1992). Feminist methods i1l social rescarch. New York: Oxford UOI-
versitv Press
Rhett, K. (Ed.). (~997). SlIruiual stories: Memoirs of crisis. Garden Cit)" NY:
Doubleday.
Rich, A. Il~7o),Onlies, secrets and silence: Selecled prose 1966-]978. New yod<'
Richardson, L (1990a). Narrative and sociology.jollmal ofCOl/temporary EtlJHOg-
l"ilphy, 19, 116-135.

254 255
 METHODS OF COLLECTING ANU ANALY LIND tMt-'lt<ILAL MAl tt~IAW
Richnrdson, L. (1990b). \Ylriti'lg strategies: Reaching diuerse audiences. Newbury
Park, CA: Sage.
H.ichardson, L. (1994a). Nine Doems: Marriage and the family. journal ofColltclI1-
porar)' Ethnography, 23, 3-14.
Richardson, L. (1994b). \"XTricng: A method of inquiry. In N. K. Deozin & Y. S.
Lincoln (Eds.), Hmldbooh o( qualitotiue research (pp. 516-529). Thousand
Oaks, CA: Sage.
Richardson, L. (1995). Vespel-s. Chicago Reuiew, 41,129-146.
Richardson, L. (1996). Speecn lessons. In C. Ellis & A. P. Bochner (Eds.), COIll-
posillg ethnography: Alternatiue forms of qualitative writing (pp. 231-239).
\\laln ur Creek, CA: Altal'vlira.
Richardson, L. (l997) . Fields of play: Constructing ml academic li(e. New Bruns~
wick, NJ: Rutgers University Press.
Richardson, L., & Lockridge, E. (1998). Fiction and ethnography: A conversation.
QlIalilatiLle Inquiry, '1, 328-336.
Rieillcr,j. \'1/. (1977). Varieties of opportunistic research. Urban Life, S, 467-477.
Riley, LvI. \XI. (1988). Sociological lives. Newbury Parle, CA: Sage.
Rodriguez, R. (1983). Hunger of memory: The educatioll o( Richard Rodriguez.
New York: Bantam.
Ronai, C. R. (1991.). The ref\.:,xive self through narrative: A night in the life of an
erotic dancer/researcher. In C. Ellis & 1\11. G. Flaherty (Eds.), !Jwestigatillg
subjectivity: ResearclJ 0'1 hued experience (pp. 102-124). Newbury Parle, CA:
Sage.
Ronai, C. R. (1995). !'-/luhiple reflections of child sex abuse: An argument for a lay-
ered account. JOllmal cof Contemporary Ethnography, 23, 395 -426.
Rony, R. (1979). Philosophy and the mirror of nature. Princeton, NJ: Princeton
University Press.
Rorty, R. (1981) . COl1sequclll:es of pragmatism (essays 1972-1980). l'viinneapolis:
University of L\:1innesot:a Press.
Rose, D. (1991). In search of experience: The anthropological poetics of Stanley
Diamond. In I. Brady (Ed.), Anthropological poetics (pp. 219-133). Savage,
IVID: Rowman & Littldield.
Rosen, H. (1988). The autobiographical impulse. In D. Tannen (Ed.), Linguistics ill
cOllte.".'t: COllllccting olJservatioll and understanding (pp. 69-88). Norwood,
Njo Ablcx.
Schneider, A. (1998, July 10;. Jane Tompkins's message to academe: Nurture the
individual, not just the i mellect. Chl'Ollicle ofHigher Education, pp. AS-AIO.
Schwandt, T. f\. (1996). farewell to criteriology. Qlwlitative IlIql/iry, 2, 58-1--
-,
Shalit, R. (1998, February). The man who knew too much. Lingua Franca, pp. 31-4~.
Shelton, A. (1995). FOllcault'3 Madonna: The secret life of Carolyn Ellis. SYil1bo!Jc
Intcmctioll, 1 S, 83-87
256
Shoslale, A. (Ed.). (1996). Primte sociology: UllsfHlring reflections, lIJlCOlll1l1011
gail1s. Dix Hills, Nl-': General j-bll.
Shostak, ivI. (1981). Nisa: The hfe alld words ofd !KlIllg WU/Ifilll. Cambridge, ;VIA:
Harvard University Press. ~
Shorter, j. (1987). The soci81 construction of an "us": Problems of accoullt3bilitv
and narrarology. In R. Burnett, P.McGee, & D. Clarke (Eds.), AccollntiJlg fa')'
relatiol1ships: Explmwtioll, representatiolf, ulfd !moluledge (pp. 125-247).
London: I'vlethuen.
Smith, D. (1979). A sociology for women. In]. Sherman & E. Becle (Eds.), The
IJrislll of sex: Essays in the sociology of Iwoluledge. 1vladison: University of
\Xlisconsi n Press. .
Smith, D. (1990). The CO/lceptual practices ofpolucr: A lell1inist sociology ofl"-/Jowl-
edge. Boston: Northeastern University Press.
Smith, D. (1992). Sociology from women's perspective: A reaffirmation. Sociologi-
cal T'heory, 10, 88-97.
Sontag, S. (1978). IIlncss as a II/etaphor. New ''{ork: Farrar, Straus & Giroux.
St8ke, R. E. (1994). Case studies. In N. K. Denzin & Y. S.l..incoln (Eels.). The IJdlld~
book of ql/olitatiue resmrch (pp. 236-247). Thollsand O,lks, C1\: Sage.
Strarhern, 1v1. (1987). The limits of auto-anthropology. In A. Jackson (Ed.), Anthro-
pology at home (pp. 59-67). London: Tavistock.
Tedlock, B. (1991). From participant observation to the observation of participa-
tion: The emergence of narrative ethnogr<lphy. JOl/ma! of AllthrofJological
Research, 41> 69-94.
Tierney, \\1. Ci 998). Life history's history: Subjects foretold. Qualitative Illquiry, 4,
49-70.
Tierney, \XZ G., & Lincoln, Y. S. (Eds.). (1997). Relm!sC/1tatioll <lIId the text: Re-
framing the llarratiuc 1/oice. Albany: State University of New York Press.
Tillmann-Hcaly, L. (1996). A secret life in a culture of tbinness: Reflections on
body, food, and bulimia. In C. Ellis & A. E Bochner (Elis.), Composil1g eth-
nography: Alterlll1tiue forms of qualitative writing (pp. 76-108). \'{!alnLlt
Creek, CA: Altalvlira.
Tillmann- Healy, L. (1999). Life projects: A 1l11rrative ethnography of gay-straight
(rielldship. Unpublished doctoral dissertation, University of South Florida.
Tompkins,]. (1989). ]Vle and my shadow. In 1. Kauffman (Ed.), Gmdermld thearv:
Dialogues Oll (eminist criticism (pp. 121139). Cambridge, 1vIA: BlJckweil.
Toulmin, S. (1969). Concepts and the expl::mation of human behavior. In
T_ 1vr:,,~~l,cl (Qel.), rl'''''d1'1 ,_<ct;o1'1 (pp. 7-i--l0\). Ne-..v -','urI<, "'""den.ie_ P'-cso..
Trinh T. 1v1. (1989). VVollwn, notiue, other: Writing pastcoloniality and feminism.
Bloomington: Indiana University Press.
Trinh T. 1'\/1. (1992). Friltner (ramed. N-ew York: Routledge.
257
 'VII.:: I I IVU-, V ' ..... V ......... ' _ " , ,,-,,-,, H .... r" '''~,~" ' ..... _,." .... ~ .. _ .. _ .. _

TroneI', IV1. (J 992). Ufe writing: Explorillg the practice of alftoet!mography ill
dllthropo!o!::,'J" Unpublished master's thesis, University of Illinois, Urbana+
Ch3mpaign.
True confessions: The age of the literary memoir [Special issue]. (1996, ]day 12).
New York Times Magazine.
Turner, V, & Bruner, E. (Eds.) . (1986). The anthropology of experience. Urbana:
Universitv of l!linois Press.
7
Tyler, S. (1986).' Post-modem ethnography: From document of the occult to occult Data Management
. document. In J. Clifford and G. E.lvlarcus (Eds.), \T!riting culture: l1Je poetics
alld politics ofet!mognrphy (pp. 122-140). Berkeley: University of California
Press.
and Analysis Methods
Van ~ilaanen, J. (1988). Tales of the field: On writing et!mogratJhy. Chicago: Uni-
versitv of Chicago Press.
Van I\'1aane~, J. (1995). An e:ld to innocence: The ethnography of ethnography. [n
J. Van IVlaanen (Ed.. ), Representation ill ethnography (pp. 1-35). Thousand Gery W Ryan and H. Russell Bernard
Oaks, CA: Sage.
Van I\1aanen, 1V1. (1990). Res,~i1rchi!lgiil/ed experience: HUlIlall science for an action
sClIsitive pedagogy. Alhany: State University of New York Press.
\Virrgenstein, L (1953). PhilOsophical il1liestigatiolls (G. Anscombe, Trans.). New
York: rVlacmilian.
Zob, L K. (1982). Missing pieces: A. chro/licle of living with a disability. Philadel-
<1> Texts Are Us
phia: Temple University Press.
Zussman, Ie (1996). Autobiographical occasions. Contemporary Sociology, 25, This chapter is about methods for managing and analyzing qualitative data,
143-148. By qualitative data we mean text: newspapers, movies, sitcoms, e-mail traf-
fic, folktales, life histories. \X!e also mean narratives-narratives about get-
ting divorced, about being sick, about surviving hand-to~hand combat,
abom selling sex, about trying to quit smoking. In fact, most of the archaeo-
logically recoverable information about human thought and human behav-
ior is text, the "good stuff" of social science.
Scholars in content analysis began using computers in the 19505 to do
statistical analysis of texts (Pool, 1959), but recent advances in technology
are changing the economics of the social sciences. Optical scanning today
makes light work of converting written texts to machine-readable form.
\1?ithin a few years, voice-recognition software will make light work of
transcribing open-ended interviews. These technologies are blind to
epistemological differences. Interpretivists and positivists alike are using
these technologies for the analysis of texts, and will do so more and more.
Like Tesch (1990), \ve distinguish between the linguistic tn:rditiol1,
which treats text as an object of analysis itself, and the sociological tradi~
tiolt, which treats text as a window into human experience (see Figure
7.1). The linguistic tradition includes narrative analysis, conversation (or

258 259
 l./UtU IVllIIIUYl;!/IIt::11t UlllI MIIUIY::'/::' IVlt'lI/UU"

discourse) analysis, performance analysis, and formallinguisric analysis.

Methods for analyses in this tradition are covered elsc\vhere in this J-Iand-
huoh. \'\/e focus here 011 methods used in the sociological tradition, which
\VC take to include work across the social sciences.

There are two kinds of written texts in the sociological tradition: (3)
.~ ~
words or phrases generated by techniques for systematic elicitation and (b)
eo ">
.Q
S1
'',;j ";;" );" free-flowing texes, such as narratives, discourse, and responses to open-
0
<J) "i! c;..;
0
" aU}
E
~/~ 5
u ended interview questions. In the next section, we describe some methods
, 'U
.~. ~ ~
/
for collecting and analyzing words or phrases. Techniques for data collec-
> C; -_..+
.if]
1:'
~
u
u
tion include free lists, pile sons, frame elicitations, and triad tests. Tech-
"
:<? oj
A
c
"
E
niques for the analysis of these kinds of data include componential analy-
C) "
<t:
.~ cEu sis, taxonomies, and mental maps.

-""'"
Q)
f.:; U
"> '"
u

;S
We then turn ro the analysis of free-flowing texts. \Ve look first at meth-
ods that use raw text as their input-mcthods such as key-words-in-
context, word CGlints, scmamic network analysis, and cognitive maps. \X!e
then describe methods that require the reduction of text to codes. These
include grounded theory, schema analysis, classical content analysis, con-
cd
~
cd
tent dictionaries, analytic induction, and ethnographic decision models.
Q Each of these mcthods of analysis has advantages and disadvantages. Some
<J)
arc appropriate for exploring data, others for making comparisons, and
>- -,... "",
~

.~
others for building and testing models. Nothing c10es it all.

""'"
;~ <J)
cd
~ E-'
~ .,...
cd
~
a '" .. Collecting and Analyzing Words or Phrases
""
c::l
~ Techniques for Systematic Elicitation
i:;\
s:i
0:; Researchers lise techniques for systematic elicitation to identify lists of

-"'"
Q
f.:;
items that belong in a cultural domain and to assess the relationships
lI
,: llong these items (for detailed reviews of these methods, set Bernard,

1994; Borgatti, 1998; Weller, 1998; Weller & Romney, 1988). Cultural
0
.~

'U
domains comprise lists of words in a language that somehow "belong
~ together." Some domains (such as animals, illnesses, things to eat) arc very
-< large and inclusive, whereas others (animals you can keep at home, ill-
nesses that children get, brands of beer) are relatively small. Some lists
(such as the list of terms for members of a family or the names of all the
Major League Baseball teams) are agreed on by all native speakers of a !an-
~l1age; others (such as the list of carpenters' tools) represent highly special-
Ized knOWledge, and still others (like the list of great left-handed baseball
260
261
 Mt 1 HUU,) Ut" \",ULLl:L 111'f1.:! ;'>'1'lU ;'>'1'11-\1..1 LIl'.'.:] L.IVlrlf'l.,,-rU.. '.''''''' '-'~'''''''-''''
oirchcrs of the 20th century) are matters of heated debate. Below we
~evie\v some of the most common systematic elicitation techniques and
discuss how researchers analyze the data they generate.
Free Lists
Frec lists arc particularly useful for identifying the items in a cultural
domain. To clicit domains~ researchers might ask, "What kinds of illnesses
do you know?" Some short, opcn-ended questions on surveys can be con-
sidered rree lists, as can some responses generated from in-depth ethno-
graphic imervic\\'s and focus groups. Investigators interprct the frequency
of mention and the order in which items are mentioned in the lists as indi-
cators of items' salience (for measures of salience, see Robbins & Nolan,
1997; Smith, 1993; Smith & Borgatti, 1998). The co-occurrence of items
across lists and the proximity with which items appear in lists may be used
as measures of similarity among items (Borgatti, 1998; Henley, 1969; for a
clear example, see Fleisher & Harrington, 1998).
Paired Comparisons, Pile Sorts, Triad Tests
Researchers use paired comparisons, pile sorts, and triads tests to ex-
plore the relatioJ1ships among items. Here are two questions we might ask
someone in a paired comparison test about a list offruits: (a) "On a scale of
1 to 5 how similar are lemons and watermelons with regard to sweet-
ness?"'(b) "Wlhich is sweeter, watermelons or lemons?" The first question
produces a set of fruit-by-fruit matrices, one for each respondent, the
entries of which are scale values on the similarity of sweetness among all
pairs of fruits. The second question produces, for each respondent, a per-
fect rank ordering of the ~et of fruits.
In a pile sort, the researcher asks each respondent to sort a set of cards
or objects into piles. Item similarity is the number of times each pair of
items is placed in the same pile (for examples, see Boster, 1994; RODS,
1998). In a triad test, the researcher presents sets of three items and asks
each respon d ent elt . Iler to "'c Iloose t IlC tvvo nlost SImI
. '1'
ar Items " or to " p icl.::
- tures to distinguish four itenl~ (2 2 = 4, in othcr . . ' lords)] rhTl:c fcaLun:s LO
the item that is the mosrdifferent." The similarity among pairs of items is
the number of times people choose to keep pairs of items together (for
some good examples, see Albert, 1991; Harman, 1998).
262
UUlU tV'UIIUyl;'lllt'llt UIIU rlrJWY-"IS IVlernoos
Frome Substitut'ion
In the frame substirution task (D'Andrade, 1995; D'Andrade, Quinn,
Nerlovc, & Romney, 1972; Frake, 1964; IVletzger & \X!ilJiams, 1966), the
researcher asks the respondent to link each item in a list of items with a list
of attributes. D'Andrade et al. (1972) gave people a list of 30 illness terms
and asked them to fill in the blanks in frames such as "'You can carch
from other people," "You can have _ _ and never know it," and "Most
people get _ _ ar one time or other" (p. 12; for other examples of frame
substitution, see Furbee & Benfer, 1983; Young, 1978).
Techniques for Analyzing Data About Cultural Domains
Researchers use these kinds of dara to build several kinds of models
about how people think. Componential allalysis produces formalmodc!s
of the e1emcnrs in a cultural domain, and taxonomies display hierarchical
associations among the dements in a domain. hIelltall/lajJs are best for
displaying fuzzy constructs and dimensions. \'Ve rreat these in turn.
Componential Analysis
As we have outlined else\vhere, componential analysis (or feature anal-
ysis) is a formal, qualitative technique for studying the coment of meaning
(Bernard, 1994; Bernard & Ryan, 1998). Developed by linguists to iden-
tify the features and rules that distinguish one sound from another
Uakobson & Halle, 1956), the technique was elaborated by anthropolo-
gists in the 1950s and 1960s (Conklin, 1955; D'Andrade, 1995; Frake,
1962; Goodenough, 1956; Rushfonh, 1982; Wallace, 1962), (For a par-
ticularly good description of how to apply the method, sec Spradley, 1979,
pp. 173-184.)
Componential analysis is based on the principle of distinctive features.
Any tWo items (sounds, kinship terms, names of plants, names of animals,
anel so on) can be distinguished by some minimal set (2Jl) of binary fea-
tures-that is, features that either occur or do nor occur. It t;lkes two fea-
distinguish eight items (2.1 8), and so on. The trick is to identify the
:0:;:
smallest set of features that best describes the domain of interest. Table 7.1
shows that just three features are needed to describe kinds of horses.
263
 'Y''''''' ,v ...... ..} V, .... v ................ , " ,u ..... , , ........... ' .......... , L.." ''-' .... "" ,,,' ............. >Y'..... , ""'~J""'''''..J

TABLE 7.1 A Componential Analysis of Six Kinds of Horses

~'"o
Z
Nome Female Neuter Adult

More + +
Stallion +
Gelding + +
Fool +
Filly +
Colt
SOURCE: Adapted from D' Andrade :1995).

Componcmial analysis produces models based on logical relationships

among fcawres. The models do not account for variations in the meanings
of terms across individuals. For example, when we tried to do a com-
ponential analysis on the terms for cattle (bul/, COW, heife7; calf, steel; and
ox), we found that native speakers of English in the United States (even
farmers) disagreed about the differences between cow and heifel~ and be-
nvecn steer and ox. \X/hen the relationships among items are less well
defined, taxonomies or mental models may be llSefU1. Nor is there any inti-
mation that componential analyses reflect how "people really think."

Taxonomies

Folk taxonomies are meant to capture the hierarchical structure in setS

of terms and are commonly displayed as branching tree diagrams. Figure
7.1 presents a taxonomy of our own understanding of qualitative analysis
techniques. Figure 7.2 depicts a taxonomy we have adapted from Pamela
Erickson's (1997) study of the perceptions among clinicians and adoles-
cents of methods of conrraceprion. Researchers can elicit folk taXono-
mies directly by using successive pile sorts (Boster, 1994; Perchonock &
/]'
\Verner, 1969). This im-olves asking people to continually subdivide the
piles of a free pile sort until each item is in its own individual pile. Taxo~
nomic models can also be crented with duster 3nalysis on the .similarity
dJla from pn.ired comparisons, pile sons, and triad tests. Hierarchical clus-
}\Zb
a.>
5
ter alIa lysis (Johnson, 1967) builds a taxonomic tree where each item " 0
Old;
appears in only one group.

264 265
 1Y1t: I nULJ.':! ur \""ULLtl...l '1'11.;:1 I,,"'LJ MI'IML' LII"U LIVlr """"''"'L IV',",I L''',",~

lnterinformant variation is common in folk taxonomies. That is, differ- blintlne55

ell[ people may use different words to refer to the same category of things.
Some of Erickson's (J 997) clinician informants referred to the "highly Jain! pain
swolleJl feci
effective" group of methods as "safe," "more reliable," and "sure bets."
Category labeLs need not be simple words, bur may be complex phrases; kJlee pain
Cbronic
for eX<lmple, see the category in Fig me 7.2 comprising contraceptive
l~pro5Y
methods in vvhich you "have to pay attention [Q timing." Sometimes,
car pain toOl hache
people have no labels at all for particular categories-at least none that Sian hilek pain matlness
abc= cancer-l1p'(IH ~piJcpsy
they em dredge up easilY-;ll1d categories, even when named, may be smallpo:< hC1l(lach~

fuzzy and may overlap with other cltegories. Overlapping cluster l-1ualysis ilching
measle',
(Hartigan, 1975) identifies groups of items where a single item may ap~ psoriasis Head
rinb'wilml
pear in multiple groups. "rubies high fever
boils
gniler fever
Mental Maps
yaws
Chest/Side
IVfental maps arc visual displays of the similarities among items,
b,calhing-crugh side paill-l,,,ynda
whether or llot those items are organi7-ed hierarchically. One popular
;;ide p,lin-ichi!J-i-
method for making these maps is by collening daw about the cognitive
Stomach
similarity or dissimilarity among a set of objects and then applying multi- ","m~h~I."
diarrht:a mugh
dimensional scaling, or lVIDS, to the similarities (Kruskal & \'{!ish, 1978). h))]lDfrhca aolhma
( poison gil';!rilis
Cognitive maps are meant to be directly analogous to physical maps.
Consider a table of distances Jetween all pairs of cities on a map. Objects ~
(cities) that are very dissimil<:.r have high mileage between them and arc
placed far apart on the map; objects that are less dissimilar have low mile- Figure 7.3. MemalMap of Kom Illness Terms
age bet\veen them and are placed closer together. Pile sorts, triad tests, and
paired comparison tests are measures of cognitive distance. For example,
Ryan (1995) asked 11 hterate Kom speakers in Cameroon to perform suc- by their surviving friends and family as active, intelligent, ontstanding,
cessive pile SOrts on Kom illness terms. Figure 7.3 presents an MDS plot of conscientiolls, and eX!JericJlced experts. Although the managers who died
the collective mental map of these terms. The five major illness categories, in 1986 were still respected, they were more likely to be described as entre-
circled, were identified by hierarchical cluster analysis of the same matrix preneurs, opiniol1 leaders, and decision mahers. Perceptions of female
used to produce the MDS plot. I managers also changed, bur they did not become more like their male
Data from frame substitution tasks can be displayed with correspon- counterparts. In 1974 and 1980, female managers were remembered for
dence analysis (\"X'eller & Romney, 1990).2 Correspondence analysis scales being nice people. They were described as hilld, li/whle, and adorable. By
both the ro\vs and the columns into the same space. For example, Kirchler 1986, \vomen were remembered for their COl/rage and commitment.
(1992) analyzed 562 obituaries of managers who had died in 1974, 1980, Kirchler interpreted these data to mean that gender stereotypes changed
and 198(1. He identified 31 ckscriptive categories from adjectives used in in the early 1980s. By 196, both 1TI'<11e and female manager~ were per-
the obituaries and then llsed c.Jrrespondence analysis to display how these ceived as working for success, bur men impressed their colleagues through
categories were associated with men and women managers over time. Fig- their knowledge and expertise, whereas women impressed their col-
ure 7.4 shows that male man<:.gers vdw died in 1974 and 1980 were seen leagues with motivation and engagement.
266 267
 Mt) HUU~ Ur l.ULL!::\.... IIIY-.:I ....J\lU AI'I ....Ll 41 1'1\.::J L.'V"I'~I'-r-'l"'" '''r-'ll '-,,,''''',--,
L.JUtU IVlunuyeme()[ ana AnOlYSIS JV\ernoas

I I I I , , , , -
115 - dficicnt Analyzing Words

Techniques for word analysis include key-words-in-context, word

.7i r pioneer likable - coU!us, strucrural analysis, and cognitive maps. \Ve review each below.
kind FEMALE 'SO adorable
comruuelike
conscientiOU5 FEMALE '74
active MALE '800Ul.'llanding
Key-Words-in-Context
.28 -intelligent
experienced
l\-ZALE '74 .
expert supportlVe
. I[ I
[alt 1 U
-
D honest admirable Researchers create key-words-in-conrext (KWIC) lists by finding all the
m respected places in a rext where a panicular word or phrase appears and prinring it
e -16 [
- our in the contex[ of some Ilumber of words (say, 30) before and after it.
work oriented
f! o:;::~~:~cund instruclJr unselfish This produces a ccmcordLlllce. \X!ell-known concordances have been done
i SPlnt 1\-1 \LE '36
o opinIOn Jcader on sacred texts~ such as [he Old and New Testamems (Darton~ 1976;
Harch & Redpath, 1.954) Jnd the Koran (Kassis~ 1983), and on famous
n _ 59 sociable amiable -
decision maker farsighted
2 caring works of literature from Euripides (Allen & Italie, 1954) to Homer
f:m
(PrenclergJst, 1971), to Beowulf (Bessinger, 1969), to Dylan Thomas
.'OJ - - (Farringclon & Farringdon, 1980). (On the usc of concordances in modern
highly committed JiterJry smdies, see Burton, 1981a, 1981b, 1982; !\IIcKinnon) 1993.)

-1.016 Word Counts

courageous

FEMALE '116 \Vord coums are useful for discovering patterns of ideas in any body of
I I
-.08
text, from field notes to responses to open-ended questions. Students of
-.18 1'.91
mass media have used use word counts to [race the ebb and flow of support
Dimension 1
for political figures over rime (Danielson & Lasorsa, 1997; Pool, 1952).
Differences in the use of vvords common to rhe writings ofjamcsMaciison
Figure 7.4. Correspondence Analysis of the Frequencies of 31 Disruptive
Obituary Categories by Gender and Year of Publication and Alexander Hamilml1 led IVlostelier and \X!allace (1964) ro conclude
SOURCE: Erich Kirchler, "Adorable Woman, Expert Man: Changing Gender Images of clur lvIadison and not Hamilton had written 11 of the federalist Papers.
Women and Men in Management," European IOl/mal of Social Psychology, 22 (1992),~. 371. (For other examples of aurhorship studies, see .Nlartindale & 1'vlcKenzie,
Copyright 1992 by John Wiley & Sons Limited. Reproduced by permission of John Wdey & 1995; Yule 194411968.)
Sons Limited.
\Vord analysis (like constant comparison) memoing, and other tech-
niques) can help researchers to discover themes in texts. Ryan and \Weisner
'" Methods for Analyzing FreeFlowing Text (1996) inst"rucred fathers and mothers of adolescents in Los Angeles:
"Describe your children. In your own words, just tell us about them."
AI[hough taxonomies) MDS maps, and the like are useful for analyzing Ryan and \X/eisner identified ::111 the unique words in the ansvvers they got
short phrases or words, most qualitative data come in the form of free- co that grand-rour question and noted the number of times each word was
flmving texts. There arc two major types of analysis. In one, the [ext is scg~ Used by mothers anel by fathers.1V[othcrs) [or example, \vere more likely to
Use words like friends, cred!iue, time, and honest, farhers were more likely
rnented imo its most basic meaninghii components: words. In the other,
meanings are found in large blocks of text. to Use words like 5choo/, good, lack, student, enjoys, independent, and
extrelnely. This suggests that mothers, on first mention) express concern
268
269
 Mt 1HUU:::' Ur LULLtL 11I'1l:l HI'lU ANALT Lll'lt.:l l:!Vlnr<;Il.-Al. IVlMI Crl.II'-'IL-l _ .... ,.... ,. ,,,, ,,,':/'--,, ''''''' '-" ''-' ,-" ''''r-''-' ,. ''-- ,~~~

over interpersonal issues, wherc3s fathers appear to prioritize achieve-

ment-oriented and individualistic issues. This kind of analysis considers
neither the contexts in which the words occur nor whether the words are 0'
0
used negatively or positivel/, but distillations like these can help research- 0'
ers to identify important constructs and can provide data for systematic
comparisons across groups. 0' 0'
0' 0' 0'

Structural Analysis and Semantic Networks 0'0'

o' o' 0'
0' 0

Net\vork, or structural, analysis examines the properties that emetge

from relations among things. As early as 1959, Charles Osgood created
word co-occurrence 1TI3trices :md applied factor analysis and dimensional
plotting to describe the relarions among words. Today, semantic network 0'

analysis is a growing field (Earnerr & Danowski, 1992; Danowski, 1982,

1993). For example, Nolan c.nd Ryan (1999) asked 59 undergraduates (30
vvomen and 29 men) to describe their "most memorable horror film." The
researchers identified the 45 most common adjectives, verbs, and nouns
Llsed across the descriptions of the films. They produced a 45 (word)-by-
59(person) matrix, the cells ,)f which indicated whether each student had
used each key word in his or her description. Finally, Nolan and Ryan cre-
ated a 59(person)~by-59(per:;on) similarity matrix of people based on the
co-occurrence of the words :n their descriptions.
Figure 7.5 shows the IVIDS of Nolan and Ryan's data. Although there is
some overlap, it is pretty clear that the men and women in their study llsed
different sets of words to de.<;cribe horror films. Men were more likely to
use words such as teowgel; disturbing, violel1ce~ rural, darll, COlflltry, and
hillbilly, whereas women were more likely to use words such as boy, little,
dellil, yOllng, hOlTm; t~1thel; and euil. Nolan and Ryan interpreted these
results to mean that the men had a fear of rural people and places, whereas
the women were marc afraid of betrayed intimacy and spiritual posses~
sian. (For other examples of the use of word-by-word matrices, seeJang &.
Barnett, 1994; Schnegg & Bernard, 1996.) This example makes abun-
dantly clcar rhe value of turning qualitative data inro quantitative data:
Doing so can produce information that engenders deeper interpretations
of the mcanings in the original corpus of qualitative data. Just as in any
mass of numbers, it is hard tc sec patterns in words unless one firsr does
some kind of dara reduction. l\.10re abom this below.
As in word analysis, one appeal of semantic network analysis is that the
data processing is done by computer. The only investigator bias intro-
Q
GENDER
a'" Male
9 Female
Figure 7.5. Muitidil1lensioll;l! Scaling of Informants Based on Words Used in
Descri ptions of Horfor Films
ducecl in the process is the decision to include words that Occur at least 10
times or 5 times or whatever. (For discussion of computer programs that
produce word-by-text and word-by-word co-occurrence matrices see
Borgatti, 1992; Doerfel & Barnett, i996.) There is, however, no gu~ran
tee that the output of any word co-occurrence matrix will be meaningful,
and it is notoriously easy to read patterns (and thus meanings) into any set
of items.
Cognitive Mops
Cognitive map analysis combines the intuition of human coders \vith
the quantitative methods of network analysis. Carley's work with this
teclmiclut,; is in~[fllC[ive. C.arley argues that if cognitive models or sche-
Ihata exisr, rhey arc expressed in the texts of people's speech and can
be represented as networks of concepts (see Carley & Palmquist, 1992,

270 271
 Mt I HUU::' Ur LULLtL 11I'1\.:l ANU Al'IALf Lil'I\.:! t:/VIrIf\I\...ML IVIMI Lr.'ML..J
UOW IVlonuyernenr uno f\IlOlYSIS IVlernoas

p. 6(2), an approach also suggested by D'Andrade (1991). To the extent

thea cognitive models are widely shared, Carley asserts, even a very small STUDENT A STUDENTB
set of texTS will comain the inform,ltiol1 required for describing the mod-
els, especially for narrO\vly defined arenas of life.
In one study, Carley (1993) asked studcms some questions about the
work of scientists. Here are two examples she collected: \
Student A: J found that scientists engage in research in order to make discov-
eries and generate new icle:ls. Such research by scientists is hard work and
"'"f'' \
rcsellfclt

often involves collaboratiOlI with other scientists which IC3ds to discoveries

which make the scientists famolls. Such coUaboration may be inform;ll, such new idens collnboration
as when they share new ideas over IUllch, or formal, stich as when they are
coauthors of a p3per. formal

Studellt B: Ir was hard work to research famous scientists engaged in collab~

...",....~,J couuthors
oration and I madc many informal discoveries. ;Vly rcse3rch showed that
scientists engaged in colbb,Jration wirh other scientists are coaurhors of ar
least one paper containing their ncw ideas. Some scientists make forma! dis- Shurcd Concepts II
cO\'cries and have new ideas. (p. 89) Shnred Stutcments (1 bidirectionnl = 2 relations) 5
Shared Concepts given Shared Helationshifls 5
Concepts Student A Dnly 0
Cuncepts Student n Only.............................................. 0
Carley comp8[cd the students' teXTS by analyzing 11 concepts: I, scielltists) Statements Student A Only............................................ 13
Statements Student B Onl.'1' 9
research, hard wor/::!., collaboratioJI, discoueries, Hew ideas) formal, infor-
mal, coauthors, paper. She coded the concepts for their strength, sign (pos-
itive or negative), and direcion (whether one concept is logically prior to
others), not just for their existence. She found that although students used -=m posHI've
relationship
the same concepts in their C.'xts, the concepts clearly had differenr mean-
~ nega.tive
ings. To display the differences in understandings, Carley advocates the relationship
use of maps that show the rdations between and among concepts. Figure
7.6 shows Carley's maps of two of the texes.
Carley's appro:1Ch is promising because it combines the automation of
word counts with the sensiti vity of human intuition and interpretation. As
Figure 7.6. Coded IvIaps of Two Swclems' Texts
Carley recognizes, however, a lot depends on who does the coding. Differ-
SOURCE: Kathleen Carley, "Coding Choices for TextL1nl Annlysis: A Comparison of Con-
ent coders \vill produce different maps by making different coding lent Anillysis and I'vlap Analysis," in P. IvL:lfsden (Ed.), Sociological Methodology (Oxford:
choices. [n the end, native-I;::.nguage competence is one of the fundamental Blackwell, 1993), p. 104. Copyright 1993 by the American Sociological Association. Repro-
methodological rcquiremens for analysis (see also C<.lrley, 1997; Carley duced by permission of the American Sociological Association.
& Kaufer, 1993; Carley & Palmquist, 1992j Palmquist, Carley, & Dale,
1997). make comparisons <.lcross texts. \',/ith the except:ion of K\X!lC 7 however]
Key-words-in-context, v..-ord COllnts, structural analysis, and cognitive these techniques remove words from the contexts in \vhich they occur.
maps all reduce text to the fundamental meanings of specific words. These Subtle nuances are likely to be lost-which brings us to the analysis of
reductions make it easy for researchers to identify general patterns and whole texts.

272 273
 Mt I HUU:J Ur l.ULltl.) II'll;) AI'IU Al'lALI LII'H.;J CJII\rlKIL. ..... L IVI..... 1 Cr\I ..... LJ
Analyzing Chunks of Texi: Ceding
Coding is the hcart and soul of whole-text {/1wlysis. Coding forces the
researcher to make judgments abour the meanings of contiguous blocks of
texr. The fundamental rasks aswciated with coding are sampling, identify-
ing themes, building codebooks, marking texts, constructing models (rela-
tionships among codes), and t,~sting these models against empirical data.
\X/e outline each task below. \Ve then describe some of the major coding
traditions: grounded theory, Khema analysis, classic content analysis,
conreIl[ dictionaries, analytic induction, and ethnographic decision trees.
\YJe want to emphasize that no particular tradition, whether humanistic or
positivistic, has a monopoly 011 text analysis.
Sampling
Investigators must first identify a corpus of texts, and then select the
units of analysis within the texts. Selection can be either random or pur-
posive, but the choice is not a matter of cleaving to one epistemological
tradition or another. Waitzkin and Britt (1993) did a thoroughgoing inter-
pretive analysis of encounters between patients and doctors by selecting
50 texts at random from 336 audiotaped encounters. Trost (1986) used
classical conrem analysis to test how the relationships between teen-
agers and their families might be affected by five different dichotomous
variables. He intentionally selected five cases from each of the 32 possi-
ble combinations of the five yariables and conducted 32 x 5 = 160
interviews.
Samples may also be based on extreme or deviant cases, cases that illus-
trate maximum variety on variables, cases that are somehow typical of a
phenomenon, or cases that confirm or disconfirm a hypothesis. (For re-
views of nonrandom sampling !;trategies, see Patton, 1990, pp. 169-186;
Sandelowski, 1995b.) A single case may be sufficient to display something
of subscantivc importance, bur 1vlorse (1994) suggests using at least six
participants in studies where one is trying to understand the essence of
experience. lvlorse also suggest:; 30-50 interviews for ethnographies and
grounded theory studies. Finding themes and building theory may require
fewer cases than comp:lring :1:::roo;s 8roups and testins hrpodlC5C5 or
models.
Once the researcher has established a sample of texts, the next step is to
idemify the basic units of analy"is. The units may be entire textS (books,
274
UULU IY1U1IUljerfU;!f!r uno p,nOlYSIS fYlernOaS
interviews, responses to an open-ended qucstion on a survey), grammati-
cal segments (words, word senses, sentences, themes, paragraphs), for-
matting units (rO\vs, columns, or pages), or simply chunks of text that
reflect a single theme-what Krippendorf (1980, p. 62) calls thematic
units. In general, where the objective is to compare across texts (as in the
case of classical content analysis), the units of analysis need to be non-
overlapping. (For discussion of additional kinds of units of analysis, see
Krippendorf. 1980, pp. 57-64; Tesch, 1990.)
Finding Themes
Themes arc abstract (and often fuzzy) constructs that investigators
identify beforc, during, and after data collection. Literature reviews are
rich sources for themes, as arc investigators' own expcriences ,vith subject
matter. Ivlore oftcn than Ilot, however, researchers induce themes from
the text itself.
There is more than one way to induce themes. Grounded theorists sug-
gest a careful, line-by~line reading of the text while looking for processes,
actions, assumptions, and consequences. Schema analysts suggest looking
for metaphors, for repetitions of \vords, and for shifts in content (Agar &
Hobbs, 1985). Content analysts have used KWIC to identify different
meanings. Spradley (1979, pp. 199-201) suggests looking for evidence of
social conflict, culrura] contradictions, informal methods of social con-
trol, things that people do in managing impersonal social relationships,
methods by which people acquire and maintain achieved and ascribed sta~
tus, and information about how people solve problems. Each of these are-
11as is likely to yield major themes in cultures. Barkin, Ryan, and Gelberg
(1999) had multiple coders independently sort informants' statements
into thematic piles. They then used multidimensional scaling and cluster
analysis on the pile-son data to identify subthemes shared across coders.
(For another example, see Patterson, Bettini, & Nussbaum, 1993.)
Willms et al. (1990) and Miles and Hoberman (1994) suggest rhat
researchers start with some general themes derived from reading the liter-
ature and add more themes and subthemes as they go. Shelley (1992) fol-
lowed this ad\rice in her study of how social net\vorks affect people with
end~St~lge kidney disc"!s",. Shr; llllcd (he OWliT!f! of ClIltw'f1l Nf./tfJt;aI5
(Murdock, 1971) as the basis of her coding scheme and then added addi-
tional themes based on a close reading of the text. Bulmer (1979) lists
10 different sources of themes, including literature reviews, professional
275
 METHOD::' Ur LUlltL Ilnl:! ANU j.\I'IALT LII'H.,;l !:JVWI!\lLML
definitions, local coml11ansen:;e constructs, and researchers' values and
prior experiences. He ,,[so nortS that investigators' general theoretical ori-
entations, the richness of the existing literature, and the characteristics of
the phenomena being studied :nfluence the themes researchers arc likely
to find.
No matter how the researchet actually does inductive coding, by the
rime he or she has identified [he themes and refined them to the paim
\vhere they can be applied co an entire corpus of texts, a lot of interpretive
analysis has already been done. rVliles and Huberman (1994) say simply,
'''Coding is analysis" (p. 56).
Building Codeboo!<s
Codebooks are simply organized lists of codes (often in hierarchies).
How a researcher can develop a codebook is covered in derail by Dey
(1993, pp. 95-151), Crabtre" and Miller (1992), and Miles and
Huberman (1994, pp. 55-72). l'vlacQueen, IVlcLellan, Kay, and .Milstein
(1998) suggest that a good cockbook should include a detailed description
of each code, inclusion and exdusion criteria, and exemplars of real text
ror each theme. If a theme is particularly abstract, we suggest that d1e re-
searcher also provide example~; of the theme's boundaries and even some
cases that arc closely related but not included within the theme. Coding is
supposed (0 be data reduction, not proliferation (.~.liles, 1979, pp. 593-
594). The codes themselves ::Ire mnemonic devices used to identify or
mark the specific themes in a text. They can be either words or numbers-
v,'hatcver the researchcr finds I~asiesr co remember and to apply.
Qualitative researchers working as a team need co agree up front on
\vhat to include in their codeb;Jok. Ivforse (1994) suggests beginning the
process with a group meeting. I\ilacQueen et a1. (1998) suggest that a single
team member should be designated "Keeper of the Codebook"-\ve
strongly agree.
Good codebooks 3re develoJed and refined as the research goes along.
Kurasaki (1997) interviewed 20 sausei-third-gcneration Japanese Ame~~
ieans-and used a grounded theory Jpproach to do her analysis of ethOiC
identity. She started with seven major themes. As the analysis progressed,
o;;he split the n"l<1jor rhen"les into SllbthenlCS. EVCnt\..l~ll1y, she
of the major themes and wounclup with six major themes and a [Ural ~{J.g
subthemes. (Richards 8-:: Richards, 1991, discuss the theoretical p[lnCI~
. co d'mg structures th at emerge our
pies related to hierarchical . OCI the datJ.
276
,V1J-\Il..-l"/'"ll..-..> ......... " ... I" .... , ''-'~'-''''-' ''- -.., ,'-' , "''-''r~'~ ".~'-' ' ...... ....,.:>
Araujo, 1995, lIses an example from his own research on the traditional
British manufaclUring industry to describe the process of designing and
refining hierarchical codes.)
The development and refinement of coding categories have long been
central tasks in classical content analysis (see Bcrelson, 1952, pp. 147-
168; Holsti, 1969, pp. 95-126) and are particularly importamin the con-
struction of concept dictionaries (Deese, 1969; Stone, Dunphy, Smith, &
Ogilvie, 1966, pp. L14-168). Krippendorf (1980, pp. 71-84) and Carey,
lvforgan, and Oxroby (1996) note that much of cadebook refineme~t
comes during the training of coders to mark [he [ext and in the act of
checking for inrcrcoder agreement. Disagreement among multiple cod-
ers shows when the codebook is ambiguous and confusing. The first run
also allows the researcher to identify good examples to include in the
codebook.
Marking Texts
The act of coding involves the assigning of codes to contiguous units of
text. Coding serves two disrinet purposes in qualitative analysis. First,
codes act as tags to mark off text in a corpus for brer retrieval or indexing.
Tags are nor associated with any fixed units of text; they can mark simple
phrases or extend across multiple pages. Second, codes act as vallies
assigned to fixed nnits (see Bernard, 1991, 1994; Seidel & Kelle, 1995).
Here, codes are nominal, ordinal, or ratio scale values that are applied to
fixed, nonoverlapping units of analysis. The nonoverlapping units can be
texts (such as paragraphs, pages, documents), episodes, cases, or persons.
Codes. as tags arc associated vvirh grounded theory and schema analysis
(reviewed below). Codes as uallles are associated with classic content
analysis and contenr dictionaries, The twO types of codes are nor mlituallv
exclusive, but the use of one gloss-code-for both concepts can b~
misleading.
Analyzing Chunks of Texts:
Building Conceptual Models
c0I11bincd t\vo Once the researcher idcnrifies a set of thing~ (themes, conccp~-", b.,..licL,
behaviors), the next step is to identify how these things are linked to each
other in a theoretical model (.I'Ailes & Huberman, 1994, pp. 134-137).
l'vlodels arc sets of abstract constructs and the relationships among them
277

METHODS OF COLLECTING At\lD At,lALYZING l::.MI-'It<ILAL NIHI t:KII-\(_.)
matrix is designed to help investigators to be more sensitive to conditions,
actions/interactions, and consequences of a phenomenon and ro order
these conditions and consequences into theories. ..
IV!emoing is one of the principal techniques for recording reiatlOnshlps
c d '18 T-7"
among themes. Strauss an -'Corb'III (1990 , pp.", ..) '7, 109-129 ' 197-
219) discuss three kinds .)f memos: code notes, theory n.otes,. and oper~
tional notes. Code notes describe the concepts that are bemg d,scover~d 111
"the discovery of grounded theory." In theory notes, the researcher tr~es to
summarize his or her ide~ls about what is going on in the text. Operatlonal
notes are about practical matterS. .
Once a model starts \0 take shape, the researcher uses negative case
analysis to identify problems and make appropriate revisions. Th~ end
results of grounded theory are often displayed through the presentation of:
segments of text-verbatim quotes fr0111 informants-as ~xem~lars ~j
concepts and theories. These illustrations may be prorotYPlcal examples
of central tendencies (r~ they may represent exceptions to the. norm.
Grounded theon' researchers also display their theoretical results 1l1111aps
of the major cat~gories and the relationships among themy,earne y et ::d.,~
1995; J:Vliles & Huberman, 1994, pp. 134-137). The~e 'concept. maps
are similar to the pcrsonal semantic networks descrtbed by Le1l1h;:lrdt
(1987,1989), Strauss (1992), and D'Andrade (1991) (see below).
Schema Analysis
Schema analysis combines elements of the linguistic a~~ so~lolo.g~~~l
traditions. It is based on the idea that people must use cogTIlnv~ sllnphflc",~
tions to help make sen~e of the complex information to WhlCh they a~e
constantly exposed (CaiSon, 1983, p. 430). Schank and Abelson (197;~
postulate that schemata-o.r sCl~ipts, as they call th_~m~~nable C1~lrU[~;
skilled people to fill in details at a swry or event. It IS, sa) s Wodak (19 .'
. "1 L" - Ie 'lS eVI~
p . .J;;;"1... ..}") , our schemata that lead us to Interpret dona Isa s S1111 '-
dence of her perplexity or her desperation. . _
From a methodolog:_cal view, schema analysis is simtlar to ground~,
theory. Both begin with a careful reading of verbatim texts an~ seek to ~IS-
.' 0 f artlC I es, QUinn
cover.and link themes ir,to theoretical . mode Is.n I a senes
\l:7Bl 1/()! 1;;2 J IJ~'6J 11'17) hilS dl1dlp:;cllllLlw.:!rcdti of hv m ,; "f ,nter
J 1 . . ,_ _ d to show
views to discover concepts underlymg Amencan marnage an
. . (1997) me [hod IS [0
how these concepts are tieel together. Q LHl1lYS
280
- ---, -,- -
"exploit clues in ordinary discourse for vo/hat they tell us about shared cog-
nition-to glean VI/hat people must have in mind in order to say the things
they do" (p. 140). She begins by looking at patterns of speech and the repe-
tition of key words and phrases, paying particular attention to informants'
use of metaphors and the commonalities in their reasoning about mar-
riage. Quinn found rhat the hundreds of metaphors in her corpus of texts
fit into just eight linked classes, which she calls lastingncss, sharedness,
compatibility, mutual benefit, difficulty, effort, success (or failure), and
risk of failure.
Metaphors and proverbs are not the only linguistic features used to
infer meaning from text. D'Andrade (1991) notes that "perhaps the sim-
plest and most direct indication of schematic organization in naturalistic
discourse is the repetition of associative linkages" (p. 294). He observes
that "indeed, anyone who has listencd to long stretches of talk-whether
generated by a friend, spouse, workmatc, informant, or patiem-knmvs
how frequemly people circlc through the same network of ideas" (p. 287).
In a study of blue-collar workers in Rhode Island, Claudia Strauss
(1992) refers to these idcas as "personal semantic networks." She de-
scribes such a network from one of her informants. On rereading her
intensive interviews with one of the workers, Strauss found that her infor-
mant repeatedly referred to ideas associated with greed, money, business-
men, siblings, and "being different." She displays the relationships among
these ideas by writing the concepts on a page of paper and connecting
them with lines and explanations.
Price (1987) observes that when people tell stories, they assume that
their listeners share \vith them many assumptions about how the world
works, and 50 they leave out information that"everyone knows." Thus she
looks for what is not said in order to identify underlying cultural assump-
tions (p. 314).
For more examples of the search for cultural schemata in texts, see Hol-
land's (1985) study of the reasoning that Americans apply to interpersonal
__d problems, Kempton's (1987) study of ordinary Americans' theories of
home heat control, Claudia Strauss's (1997) srudy of what chemical plant
workers and their neighbors think about the free entetprise system, and
. Ag~r ~nd I-Iobb",'s (19Q_(~ 8n81ys;s ot l-,o,v an ;nfonnant b<::c,-unc a b\.u:Blar.
\Y/e next turn to the two other methods used across the social sciences for
,
analYZing text: classical content analysis and content dictionaries.
281
 !Vlt: I nvl..I.J VI ..... '--' .............. ,., ....... ".,~
Displaying Concepts one! Models
Visual displays are an imponanr part of qualitative analysis. Selecting
key quotes as exemplars, building matrices or forms, and laying theories
out in the form of flowcharts or maps arc all potent ways to communicate
ideas visually to others. IvIodels are typically displayed using boxes and
arrows, witl; the boxes concaining themes and the arrows representing the
relationships among them. Lines can be unidirectional or bidirectional.
For example, taxonomies are models in which the lines represent the
super- ~lI1d subordinate relationships among items. Relationships can
include causalitv, association, choices, and time, to name a few.
A widely us~d method for describing themes is the presentation of
direct quotes from re$pondents~quotes that lead the reader to und.~r
stand quickly what it may IHve taken the researcher momhs or years to hg-
ure out. The researcher chooses segments of text~verbatim quotes from
respondents~asexe111plar~ of concepts, of theories, and of negative cases.
Rvan (1999) has used multiple coders to iclemify typical quotes. He asks
10 coders to mark the same corpus of text for three themes. Ryan argues
that the tcxt marked by all the coders represents the central tendency or
typical examples of the abstract constructs, whereas text marked by only
some of the coders represcnts less typical examplcs and is more typical of
the "'edges" of the construct.
Tables can be used to organize and display raw text or can be llsed to
summarize qualitative dan along multiple dimensions (rmvs and col-
umns). The cells can be fiEed with verbatim quotes (Bernard & Ashton-
Voyoucalos, 1976; Leinhar:lt & Smith) 1985, p. 254; Iv'Iiles & Huberman,
1994, p. 130), summary statements (Yoder, 1995), or symbols (Fjellman &
Gladwin) 1935; Van 1Vlaanen, tvliller) & Johnson, 1982). (For a range of
presentation formats, see Bernard, 1994; lVIi1cs & Huberman, 1994;
\Verncr & Schoepflc) 1987.)
Classical Content Analys,:s
\\?hereas grounded theory is concerned with the discovery of dara-
induced hypotheses, classical contcnt analysis comprises techniques f~r
reducing texts to a l1nir-by-variable matrix and ~nalyzing tInt ma[r!X
quantit;tively to test hypotheses. The researcher can produce a marrix by
applying a set of codcs to a set of qualit::1tive data (including written textS
as \\'eU as audio and video media). Unlike grounded theory or schema
,--" .. ,~, ~" ,~ _ ..
--------
analysis, content analysis assumes that the codes of interest have already
been discovered and described.
Once the researcher has selected a sample of texts, the next step in clas-
sical content analysis is to code each unit for each of the themes or vari-
ables in the coclebook. This produces a unit-by-variable matrix that can be
analyzed llsing a variety of statistical techniques. For example, Cowan and
O'Brien (1990) tested \vhether males or females are more likely to be sur-
vivors in slasher films. Conventional wisdom about such films suggests
thar victims are mostly women and slashers 8re mosdy men. Cowan and
O'Brien selected a corpus of 56 slasher films and identified 474 victims.
They coded each victim for gender and surviv:ll. They found that slashers
are mostly men, but it turned out that victims arc equally likely to be male
or female. \X!omen who survive arc less likely to be shO\vn eng8ging in sex-
ual behavior and are less likely to be physically attractive than their
nonsurviving counterparts. IVlale victims are cynical, egotistical, and dic-
tatorial. Cowan and O'Brien conclude that, in slasher films, sexually pure
women survive and "unmicigated masculinity" leads to death (p. 195).
The coding of texts is usuall y assigned to Inul tiple coders so that the re-
searcher can see whether the constructs being investigated are shared and
whether multiple coders can reliably apply the same codes. Typically, in-
vestigators first calculate the percentage of agreement among coders for
each variable or thcme. They then apply a correction formula to t8ke
account of the fact that some fraction of agrecmcm will always occllr by
chance. The amount of that fraction depends on the number of coders and
the precision of measurement [or each code. If tWO people code a theme
present or abscnt, they could agrec, ceteris paribus, on any answer 25qilJ of
the time by chance. If a theme, such as wealth, is measured ordinally (low,
medium, highL then the likelihood of chance agreement changes accord-
ingly. Cohen's (1960) kappa, or J(, is a popular measure for taking these
chances into account. \Vhcn K is zero, agreement is what might be ex-
pected by chance. \X!hen J( is negative, the observed level of agreement is
less than one "vould expect by chance. How much inrercoder agreement is
enough? The standards are still ad hoc, but Krippendorf (1980, pp. 147-
148) advocates agreemcm of at least .70 and notes that some scholars
(e.g., Brouwer] Clark, Gcrbner, & Krippe.ndorf, -1969) use:1 cutoff of .RO.
neiss (1971) and Light (1971) expand kappa to handle multiple coders.
For other measures of intercoder agreement, see Krippendorf (1980,
pp. 147-154) and Craig (1981).
282 283
 METHODS OF CULU:l..IINI.:l AI'lU I-\I'll-\Ll LII"1\.:l l-.-Ivn "",-.-.~ , ..,,, ~
Reliability "concerns the extent to which an experiment, test, or any
measuring procedure yields the same results on repeated trials" (Carmines
8.: Zeller, 1979, p. 11). A high level of intercoder agreement is evidence
that a theme has some external validity and is not just a figment of the
investigator's imagination (~/1itchell, 1979). Not surprisingly, investiga-
tors have suggested many wa)'s to assess validity (for reviews of key issues,
see Campbell, 1957; Campbell & Stanley, 1963; Cook & Campbell,
1979; DeOlin, 1997; Fielding & Fielding, 1936; Guba, 1931; Guba &
Lincoln, 1982; Hammersley, 1992; Kirk & lvEller, 1986; Lincoln &
Guba, 1985). Bernard (1994) argues that, ultimately, the validity of a con-
cept depends on the utility of the device that measures it and the collective
judgmem of the scientific community that a construct and its measu~e are
v3lid. "'In the end," he says, "we are left to deal \vith the effcers of our judg-
ments, which is just as it should be. Valid measurement makes valid data,
bmvalidity itself depends on the collective opinion of researchers" (p. 43).
Generalizabilit}' refers to th(' degree to which the findings are applicable
to othcr populations or samples. It draws on the degree to which the origi-
nal data were representativE- of a larger population.
For reviews of work in content analysis, see Pool (1959); Gerbner,
Holsti, Krippendorf, Pai~ley, and Stone (1969); Holsti (1969);
Krippendorf (1980); Weber (1990); and Roberts (1997), Examples of
classical coo tent analysis can be fouod in mcdia studies (Hirsch man, 1987;
Kolbe & Albanese, 1996; Spiggle, 1936), political rhetoric (Kaid,
Tedesco, 8.: lv1cKinnon, 1996), folklore Uohnson & Price-\''{1iHiams,
"1997), business relations (Spears, lvlowen, & Chakraborty, 1996), health
care delivery (Potts, Runyan, Zerger, & Marchetti, 1996; Sleath, Svarstad,
& Rater, 1997), and law (1m rich, Ivlullin, & Linz, 1995). Classical content
analysis is also the fundar.lcnral means by which anthropologists tcst
cross~cultural hypotheses (Bradley, lvloore, Burton, &r mh' w ltc,
1990'
" ,
Ember & Ember, 1992; White & Burton, 1988). For early, but fundamen-
tal, criticisms of the appro:::.ch, see Kracauer (1953) and George (1959).
Content Dictionaries
1
COlufutcr~bas",,-I, 6",ne,"" -purp':""" enn~cnt an .. }'s,s '- ,cbonnr1 ' ,1'
, ' toI
Investlgators automate t le co d'mg 0 " T10 bUl'Id suc11 dictionn nes .
I texLS.
, ,( here afC
res~archers assign wor~s, hy hand, to ~ne or more categ~n~~ _ ~ 1aries)
typIcally 50-60 categones J1 computerIzed contcnt analySIS dlLDOI
284
,. '~"" .... ,. ''-=<1 'UU""
according to a set of rules. The rules are part of a computer program that
parses new texts, assigning words to categories.
\Vork on conrcnr dictionaries began in the 1960s with the General
Inquircr and conrinues to this day (Kelly & Stone 1975' S I
"'. . " rone et a .,
1966; Zllell, Weber, & Mohler, 1939), The General Inquirer is a com-
p.mer program that uses a dictionary (the Harvard Psychosocial Dic-
U01wry) to parse and assign text to coded categories. Over time the dic-
tionary has been updated. The latest version (Harvard IV) conra'ins more
than 10,000 words and can distinguish among multiple meanings of
\:,ords.(Rosenberg, Schnurr, & Oxman, 1990, p. 303). Because such dic-
tlonanes d~ not contain all the words in the English language, invesrig~l
tors can assign unrecognized words to categorl'es, <as ti,e}' sec f'It. a process
0 _
of further modifying the "codebook." '
~ How effec~ive are computer-based dictionaries? An early version of the
CeneralinquIrer was tested on 66 suicide notes-33 written by men who
had actually raken their O\vn lives and 33 written by men \vho were asked
tC,l produce simulated suicide notes. The program parsed the texts and
picked the actual suicide notes 91q{1 of the time (Ogilvie S~ Q"
5 I .d
0'
' ' Lone, LX.
c mel man, 1966). Content dictionaries do OOt need to be vcry big to be
useful. Colby (1~66) created a simple dictionary to distinguish between
Navaho and ZUI11 responses to rhematic apperception tests. For additional
examples of special-purpose dictionaries in Content analysis sec Fan and
Shaffer (1990), Furbee (1996), Holsti (1966), ]ellll and' W:rner (1993),
Laflal (1990,1995), McTavish and Pirro (1990), and Schnurr, Rosenberg,
Oxman, and Tucker (1986),
Contenr dictionaries are artractive because they arc entirely reliable
and auto.mate~, .but, as Shapiro (1997) argues, this may be offset by a
dec~ease In valJ.dIty. For the time being, only humans can parse certain sub-
t!~tl:S of ~neal11ng re~lected in context (Viney, 1983), but computer-based
c!ICtlOnanes are gettlng better all the time. For example, texts are now
~co.red by ~ompllter for the Gottschalk-Gieser psychological scales (mea-
:u~l~g varIOUS forms of anxiety and hostility) with greater than .80 reli-
..bdlty (Gottschalk & Bechtel, 1993),
Analytic Induction and Boolean Tests
ip, 'lila\\,
0' ,-
.
Analyt' -' d ' ' I I
. IC 111 llCtlOn IS a onna, nonquamitative method for building up
causal expl " I I
\Vas . anatlOns 0 p lcn.omena from a close examination of cases. It
, proposed as an alternanve to statistical analysis by Znaniecki (1934,
235
 METHODS OF COLLECTING ANu AI'1ALT LII'I\.:l CIV\rll'.' .... r\>- """ h

.. '_.!~.~ , ...... " , .... '-'"

PI'. 249-331), modified by Lindcsmith (1947/1968) and Cressey (1953/

read and code text and then roduc., .
1971), and is discnssed by Denzin (1978), Bulmer (1979), Manning object of the analy h p. e an e\ cm-by-vanable matrix. The
SIS, owever Is not to sho I tl I .
(1982), and Seeker (l99S), among others. (For critiqucs of the approach, codes. but ro find tl _ " I' , \ \ le re atlOnshlps among all
, lC m1l11l11a set of logIcal rell[' 1.
see Robinson, 1951.) The n"ethod is a formal kind of negative case cepts that aCCOUnts f . I d ' < tons lipS among the COll-
an.1lysis. ar a Slllg e ependenr vanabl wr I h
variables the analysis b I e. It 1 more t an three
, eCOl11es mue 11110re difficult C
The technique can be described in a series of steps: First, define a phe- such as QCA (Dnlss 1980). I A . omputer programs
nomenon that requires expla:1ation and propose an explanation. Next, possible multivariate' hypath:::s ~T;-j~~PAC (Borgarti, 1992) resr all
examine a case to see if the explanation fits. If it does, then examine reviewed in \'{feitzman & lvlil~s, 1~n95.;n tle Optimal SOlution. (QCA is
another case. An explanation is accepted until a new case falsifies it. \\lhen
a case is found that doesn't fit: then, under the rules of analytic induction,
Ethnographic Decision Models
the alternatives arc to change the explanation (so thar you can include the
new case) or redefine the phellOmenon (so that you exclude the nuisance
Ethnographic decision models (EDM ,) . ..
case). Ideally, the process continues umil a universal explanation for all that predict bellavlo . I 1 . s are qualltatlve, causal analyses
~ ra c Wlees und'" -'f .
known cases of a phenomell0n is attained. Explaining cases by declar- often referred to a' d '.. cr SpeCl.lC CIrCUmstances. An EDlvl,
s a eCISlOn tree or fl . I .
ing them all unique is a tempt!i1g but illegitill1ate option. Classic examples nested if-then statem ~ r h 1. I, . . ow.c lan, comprises a series of
of analytic induction includl: Lindesmith's (1947/1968) study of drug en stat 1111.. cntena (a d b .
to the behavior of I t (F' < n COm !nations of crireria)
addicts, Cressey's (1953/1971) study of embezzlers, and IvteCleary's < nere" 'lgure7.7).EDM1 b .
how fishermen decide wi 0 f I Slave een used to explalll
(1978) study of how parole officers decide when one of their charges is in lr:re to .151 (Gatewood 198") I .
pIe decide to place l' d ,~l , W 1m pnces peo-
violation of parole. For a particularly clear example of the technique, see < on UClr pro UCts (Gladwin 1971' Q . ,
wl11(h trearments people 1 f . ' ,u1lln, 1978), and
Bloor's (1976, 1978) analysis of how doctors decide \vhether or not ro Ryan & MarrfIlez 1996 ~ woseor an Illness (Mathews & Hill, 1990;
- ' ; loung, 1980).
remove children's ronsils.
EDlvls combine many of the techni u I .
Ragin (1987, 1994) formdized the logic of analytic induction, using;1 and classic COlltent I . .. GI d . g es emp oyed In grounded theory
ana YSIS. a Will (1989) I
Boolean approach, and R01Tlme (1995) applies the approach to textual steps for building an tl 1< . ,. ays Out the fundamental
data. Boolean algebra involv,~s just two states (true and false, present and e lnograp llC deCISIOn t d I
descriptions of the steps tnl 1 ree nlO e. (For other clear
~lbsellt), bur even with such simple inputs, things Cln get very complicated, EDM . ., see I , 998; Ryan & Marrinez 1996)
s reqUIre exploraro d II. . ,.
very quickly. \X!ith just three ,::J.ichoromous causal conditions (A and not A, ing, and model testin F' ry ata co e~t1on, preliminary model build-
B and not B, and C and not C) and one outcome variable (D and not D),
g. Irst, researchers Ide 'fy I d ..
to explore and ti,e alt . I. nn t le eClSlOns they want
ernatlves tlat 'I bl .
there are 16 possible eases: A, B, C, D; A, not B, C, D; A, B, not C, 0; and done on simple yesln t f b hare aval a e. TYPically, EDMs are
so 011. Boolean analysis involves setting up what is known as a truth tablc, a ypes a e aVlOrs The b d h
predict multiple behaviors ('1 I ) ' , y can elise, owever, to
or a matrix of the actual versus the possible outcomes. (For more on truth '" ar lews & HIli 1990. Y
as the order of multiple b h ( , , oung, 1980) as well
tables and how they are related to negative case analysis, see Becker, 1998, N I e aVlOrs Ryan & Martinez 1996)
ext, r Ie researchers conduct d . _', .
pp.146-214.) the criteria people u. I open-en ed Il1terVIC\VS to discover
se co se eet among It .
Schweizer (1991, 1996) ,lpplied this method in his analysis of conflict first ask people to re II h a ernatJves. The researchers
ea t e most recent ex I f
and social statuS alTIong residents of Chen Village, China. (For a discussion hypothetical beh' . d < amp e a an actual-not a
of Schweizer's data eollecrien and analysis methods, see Bernard & Ryan, behavior. Here is aa~rlO~ an to recall why they did or did not do the
J998.) All the dara i-lbour th." actors in this politic::ll drnm:1 were extracted about the last time ~xahmdPle from a study we've done recently: "Think
1. d .
from a historical narrative about Chen Village. Like classic content an,-:dy~ soda I . byou a' a can of '
somet lIng to nnk In ,our hnnd-
, illee, water, eer whatever D' d
sis and cognitive mapping, <:malytic induction requires that human coders nor]'" TI .. I. d f ' . . I yOll recycle the can! Why [Wh),
. liS un 0 questlo. .
n generatcs a lIst of decision criteria. To

286
287
 understand how these criteria might be linked, EDM researchers ask peo-
ple to compare the latest decision with other similar decisions made in the
past, Some researchers have used vignettes ro elicit the relationships
among criteria (e.g., Weller, Ruebush, & Klein, 1997; Young, 1980).
W/ith a list of decision criteria in hand, the researchers' flext step is to
systematically collect data, preferably from 3 new group of people, about
how each criterion applies or does not apply ro a recent example of the
behavior. "Was a recycling bin handy?" and "Do you normally recycle cans
at home?" are 2 of the 30 questions we've asked people in our srudy of
recycling behavior. The data from this stage are used to build a preliminary
model of rhe decision process for the behavior under sCfminy. Cases that
do nor fit the model afe examined closely and the model is modified.
Researchers tweak, or tune, the mndel until they achieve a satisfactory
level of postdicrive accuracy-understood to be at least 80% among EDM
researchers. Parsimonious modeLs are favored over more complicated
Ones. (For automated \vays of building and pruning decisioll trees, see
lvlingers, 1989a, 1989b.)

The process doesn't end there-rhe same dato ore used in building a
preliminary model and in testing its postdicrive accuracy. \X7hen ED!v!
'"
o
o 0'
0 researchers feel confident in their model, they test it On an independent
5 ~
0
0
" sample to see if it predicts as well as it postdiets. Tvpically, EDlvTs predict
o ~
o more than 809'0 of whatever behavior is being modeled, far above what we
~
o -5 -5" ~
expect by chance. (For more detailed arguments on hm\' to calculate accu-
"
~
.S c ~

" 0
-s ," racy in EDIVTs, see H..yan & 1vlartfnez, 1996; \Xleller et aL, 1997.)
~
0
> .f 0
Because of the iIltensive labor involved, EDMs have beeu necessarily
'2
~
0
~ 0
0 2 restricted to relatively simple decisions in relatively small ond homoge-
0
~
C'
0 u ? neous Populations. Recently, however, we found we could effectively tesr,
" on a nationally representative sal11ple, Our ethnographically derived deci-
sion models for whether or not to recycle cans and whether or nor to ask
for paper Or plastic bags at the grocery store (Bernard, Ryan, & Borgarti,
1999).
'5-
o
,e EDMs Can be displayed as decision trees (e.g., Gladwin, 1989), as deci-
o
sion tabb (Mathews & Hill, 1990; eJung, 1980), or as sets of rules in the
form of if-thell statements (Ryan & Martinez, 1996). Like componential
analysis, folk taxonomies, and schema analys;s, EDM, ,ep,e,en' On "00'>-
8atc deCiSion Jlrocess ond do not necessarily represent What is going on
inSide people's heads (Garro, 1998).

288
289
 METHODS OF COllEC !(\JG Al'lU f-\I'lf-\Ll LII-; ..... l..'''' ""~..-
4- Breaking Down the Boundaries
Text analysis as a research strategy permeates the social scicnces, a.nd the
range of methods for conducting text analysis is inspiring. ~nvest1gators
examine words, sentence:;, paragraphs, pages, documents, Ideas, mean~
ings, paralinguistic features, and even what is missing from th~ teXL TI~ey
interpret, mark, retrieve, and count. By rums, they apply Interpretlve
analysis and numerical analysis. They use text analysis for ex.ploraror y and
confirmatory purposes. Researchers identify themes, descnbe them, and
compare them across cast's and groups. Finally, they combine themes into
conceptual models and theories to explain and predict social phcnomena.
Figure 7.1 depicts a broad range of analysis techniques found across ~he
social sciences. To confor:n our presentation with thc literature on qualtta-
tive methods, we have organizcd these techniques according to the gO:lls
of the investigators and the kinds of texts to which the techniques are typi-
cally applied,
In this chapter, we foclls on the sociological tradition that uses text as a
"window into experience" rather than the linguistic tradition that de-
scribes how texcs are developed and structured. Texts such as conversa-
tions, performances, and narratives are analyzed by investigators from
both the sociological ancllinguistic traditions. Although the agendas of the
investigators may differ, we see no reason why many ~f ~hc soc.i~logical
techniques we desctibe:ould not be useful in the lingUlstlc rradmon and
vice versa.
We also distinguish between those analyses associated with systemati-
cally elicited data and those associated with free-flowing textS. ~/e arguc,
however, that these data-analytic pairings are ones of convcntlon r~1thcr
than necessity. Investigators want to (a) identify the range and salie~ce of
key items and concepts, (b) discover the relationships among these Item~
and concepts, and (c) build and test models linking these conceptS
together. They use free-listing tasks, ICW"IC, word counts, and the explor~
atory phases of grounded theory, schema analysis, and EDM to discovd
potentiallY useful themes and concepts.
,
Researchets "
use pile- sorts, palred ,
compansons, tna d s tests, framc::.
' "lib"
"I
stttution ras (5, semantic ' networ 1zs, cognltlve
. . map,5 cont'm e analysis
~ .
content dictionaries, and the modeling phases of grounded theory, <;chern~
analysis, and EDl\1 to discover how abstract concepts arc relatcd to ead
L1ey d'ISP Iay t:le
ot h.er. "I ' re IatlOnS
' 111pS
' as mo d e sior
e 1r,-,111
" e'\'orl's These
\...'
290
f~at1leworks include formal models that rely on Boolean logic (componen-
tlal analysis and analytic induction), hierarchical models (taxonomies
and Ct':1nogrJphic decision models), probabilistic models (cbssic content
analYSIS and content dictionaries), and more abstract models such as those
prod.ueed by grounded theory and schema analysis. Below we describe
two lmpOrtant examples of studies in vvhich researchers combined meth-
ods to understand their data more fully.
Jelm and Doucet (1996, 1997) used word counts, classical content
analysis, and mental mapping to examine conflicts among Chil1ese and
~.S. business associates. They asked 76 U.S. managers who had worked in
Smo-American joint ventures to describe recent interpersonal conflicts
with business partners. Each person described a situation with a samc-
culture manager and a differem-cultLlral manager. The researchers made
sure that each manager interviewed included information about his or her
relarionship to the other person, who was involved, \vhat the conflict was
about, what caused the conflict, and how the conflict was resolved.
Aftcr collecting the narratives, Jc11I1 and Doucet asked their informants
to help identify the ernie themes in the narratives. First, they generated
separate lists of words from the intercultural and inrraculrural conflict
~arra.tives. They asked three expatriate managers to act as judges and to
IclentIfy all the words that were related to conflict. They settled on a list of
542 conflict words from the interculrurallist and 242 conflict words from
the inrraculrurallist. Jcl1l1 and Doucet then asked the three judges to sort
the wor~ls into piles or categories. The experts identified 15 subcategories
for the ll1~ercultLl[al data (things like cOllflict~ expectations, rules, palUel;
and [lalattle) and 15 categories for the inrracultural data (things like can-
fl~ct, lleeds, sta/ldards, paluel; contentious, and lose). Taking into consider-
atIOn ~h~ toral number of words in each corpus, conflict words were used
more 1l11nrracuhural interviews and resolution terms \vere more likely to
be llsed in intercultural interviews.
~ Jelm and Doucet also llsed traditional content analysis on their data.
[he had two coders read the 152 conflict scenarios (76 intraculrural and
76 '
. Intercultural) and evaluate (on a 5-point scale) each on 27 different
themes they had identified from the literature, This produced t\vo 76 X 27
3!1d scenario-by-thel11e profile matrices-one for the intraeultural conflicts
~nd one for the intel'cllltur:11 conl1;cts. TI... c [;r<:;t" three factors from the
Intercultural matrix reflect (a) inrerpersonal animos;ty artd hottllty, (b)
aggravation, and (c) the volatile nature of the conflict. The first two
291
factors from the imraculrural matrix reflect (3) han"cd and. ani~1osity with
'I t 11,1 (b) L'ont'licrs condllcred calmly with lltrle verbal
a vo Ian e na Uft.' a '- -
intensity. . . . ,
Finally, Jchn and Doucet identified the 30 !ntracl1lwral.aJ~d the ,)0
intercultural scenarios that they felt were the clearest and pithlesr. They
recruited 50 more expatriate managers to assesS the similarities (on a 5-
, I) f 60 1') 0 randomly selected pairs of scenarios. \X1hen (0111-
POlllt sea eo. 1 - - < '
bined across informants, the managers' judgments produced tWO aggre-
g;ltC, scenario-by-scenario, similarity matrices-one for the ~nrracllltu.ral
'1' do for ti,e I'ntercultural conflicts IVlu!ridimcnslOnal scaling
con'1 lets an ne .
'til" "Intercultural similarity data identified four dimensions: (a) open
.~ I '
~.'crsus resistant to change, (b) situational causes versuS individua tr~ltS,
ell I .
(c) high- versUS low~resolution potential based all trus~, an~ (d) ,hlgb-
versus low~resolution potential based on patience. Scalmg ot the mt~a
cultural simibrity data identJied four different dime~sions: (a) 11Igh
versUS low cooperation, (b) h:gh versuS low confrontation, (c) problem
solving versuS accepting, and (d) resolved versuS ongoing. .
The work or Jehn and Doucet is impressive because the analysls ~f the
y pressed toward a dominant socia! other when the social other possessed,
data from these tasks produced different sets of themes. All three en1lcall.
induced theme sets have some intuitive appeal, and all three yield analync
results that are useful. The researchers could have also llsed the techniques
of grounded theory or schema analysis to discover c.ven mar.e the.mes .
Jelm and Doucet are not thl:' only researchers ever [0 combme (iJf:eren,~
analytic techniques. In a seri,~s of articles on young adult "occasional
79 19SC 1983) used grounded theory methods to
(rug
I users, 1\ ga r (19 , ' , . .
build models of behavior. He then used classical content analySIS to test hiS
hypotheses. Agar conducted and transcribed three ~nterv~e~v~~itl: eac,h. 0:
his three informants. In his 1979 article, Agar deSCribes hiS 111wal, mt~wve
~1l1alysis. He pulled all the S(8.rementS that pertained to informants' lllter~
actions or assessments of oth(~r people. He then looked at the state,mcnts
and sorted them into piles based on their content. He named each pde as ~
theme and assessed how the themes interacted. He found that he had three
iles. The first contained stat,:,:ments in which the inf,ormant was ,exPI:rc'l's-
P
ing negative feelings toward 1 person in a ~~minant soci~ll poSitlO.[~.~ .le
<;ccond w~s m~lde up of ~t'ltenent~ emnln~azmg the other 5 knowledgt:, or
~wareness. The statements in the third small cluster emphasized the nn-
ponance of change or openness to new experiences.
292
From this intuitive analysis, Agar felt that his informams were telling
him that those in authority were only interested in displaying their author-
ity unless they had knowleclge or awareness; thar knowledge or awareness
comes through openness to new experience; and that most in authority are
closed to new experience or change.
To test his intuitive understanding of the data, Agar (1983) used all
the statements from a single informant and coded the statements for their
role type (kin, friend/acquaintance, educarional, occuparional, or other),
power (dominant, symmetrical, subordinate, or undetermined), and af-
fect (positive, negative, ambivalent, or absent). Agar was particularly in~
teres ted in whether negative sentiments \vere expressed to\vard rho~e in
dominant social roles. For one informant, Agar found that our of 40 state~
menrs coded as dominant, 32 were coded negative and 8 were coded posi~
tive. For the 36 statements coded as symmetrical, 20 were coded positive
and 16 negative, lending support to his original theory.
Next, Agar looked closely at the deviant cases-the S statements where
the informant expressed positive :Iffeer toward a person in a dominant
role. These counterexamples suggested that the positive affect was ex-
or was communicating to the informant, knowledge that the informant
valued.
Finally, Agar (1980) developed a more systematic questionnaire to test
his hypothesis further. He selected 12 statements, 4 from each of the con-
trol, knowledge, and change themes identified earlier. He matched these
Statements with eight roles from the informant's transcript (father,
mother, employer, teacher, friend, wife, cO\vorker, and teammate). Agar
then returned to his informant and asked if rhe resulting statements were
true, false, or irrelevant. (In no case did the informant report "irrelevant.")
Agar then compared the informant's responses to his original hyporheses.
I~Ie found that on balance his hypotheses were correct, but discrep3n~
Cies between his expectations and his results suggested areas for further
research.
Thcs~ examples show that investigators can apply one technique to dif-
ferem bods of data and they can apply multiple techniques to the same
rl~r1. "et_ Text 3""1>",,;,, ;s Ll"cd by avv\y~d pmiitiYi5l.) and interprerlviSrs
alrlee.
As we have argued elsewhere (Bernard, 1993; Bernard & Ryan,
1998), methods are simply tools that belong to everyone.
293
 METHODS OF COlLECTIN(:l AI'IU Al'IA\....1 LII'(U ,-,." " " - ' >_.

(lb' I o~e pI'fograms ofter researchers the

grams do, would also be helpf I S '
0. ption of recording. the milrl,'
<i> What's Next? . ' \mg e laVlQr 0' mll] t' 1 d
dIrect way to measure imerc d . Ip e co ers, yet offer no
. 0 er agreemenr.
\i}e do nor want co minimize the profound intellectI/al differences in the
ago, spell checkers thesallru~es 'I d ,] bll JlflSt beglllnJng. Some 15 years
The evolutIon of text analYsis software" s ." .
epistemological positions of positivists and inrerpretivists. We think, how- ~ d l ' , a 1 sca a e 'onts were 11' ld
ever, that when researchers C;in move easily and cheaply between qualira- .10 ay, these fLl11ctions,regrarec
are j'n 1.111[Q all J. II f a dso separatelY. '
Illg packages. Just 1 0 ' . U - eaturc word-pro cess-
tive ;:;ll1d quanritative data collection and analysis, the distinctions between L

rhe twO epistemological positions will become of less practical importance. .f

yeals ago, graphiCS progr <;. ~
rom programs that do st r' .. ] I'
a IStlCa ana YSIS 11 I
am~ were h'
ld
so. - separately
'
l
That is, as researchers recogn)ze the full array of tools at their disposal and Integrated into all fulJ-f' r d ' oc ay, grap JCS functions arc
ea ure packages for 't- . ] ,
~lS these tools become easier ':0 use, the pragmatics of research will lessen grammers of text anal)lsis f
,
. : atlStlC1 analYSIS. As pro-
so tware compere t i l
will become more incl . , ' . or mar eet s lare, packages
the distinction between qualitative and quantitative data and analysis. , llSlve, Illcorporatlllg med -I' f b' . .
The process is under way--and is moving fa5t~with the development epistemological divide It 'b
. can tappen roo soon.
lOt S rom mh Sides Of the

of increasingiy useful softw~.re tools for qualitative dara analysis. Useful

tools create markets, and market needs create increasingly useful tools.
Qualitarive dara analysis packages (ATLAS!ri, NUD oIST, Code-A-Text, > Notes
the Ethnograph, AnS\'VR, a':1d others) have improved dramatical1y over
the past few years (Fischer, 1994; Kelle, 1995; \'Veirlma n & :tvtiles, 1995).
Illlerprcte.d. \X1hy arc some illnesse~ ~t } f ~.tu be llltcrpreted. In faCt, they II1I1St be
, I, [vlDS displays arr highly C\'UGltlvc. The' he .' .'
These products, and others., make it easier and C<lsier for researchers to t"
identify themes, build codebooks, lllark text, create memos, and develop
I I I
t lin (t le dlne~ses i' '
. J,.. 1(' tup O' hgure 7 J
at the top', re more' a t 1e chromc variet ' I, ' d
an some
- , at the butwmt We
theoretical models. Based loosely on ~1 grounded theory type of approach
lllure acme. We also think that tI' 'l! ... " }, \V 1cre.1S t lose at the bottom afe
e Lin test Ideas hke these 11\;- ~I"
\X . _ . " " Ie I nesses on t 1e left are le~s ' I.
to qualitative analysis, l11an~' program suites have recently folded in tech-
_ . . . '
at' the illnesses In the IvlOS pi t iF
I. . f
1 a" dng ,ry 1ll Oflnants to] ,]
- 5CflOllS
H.' P us understand the artan~emeilt
. , ~> .
t 1.111 those on the rip/it

' '\ o. or more examples [ , ,I -

niques from classical contenr analysis, Several programs, for example, . Ildrade et :d. 197 , E . 1- 0 metlta maps, see Albert 19" 1
D , _,
7 tlC ,-son 1997) (Tl. ' , ' ' :J,

allow researchcrs to exporr data to matrices that they can then analyze analysis, or PROFIT, for testing ,'d'" . b' ',ere IS a i,arm:!! method, called property fitling
, ...as d om r le dlSt b ' ".
met wd is based on linear te " S. tl mlOn (1 Items in an ivlDS map. This
)I \ 1 . ' gresSlotl. ee "'-tusk:!] & \Vis!l197X ) -
using ochcr programs.
Investigators, however, remain constrained by program-defined units ~~ t ternatlvely, profile mattices (the usual th'
tous In the social sciences) c,,'I'"'' d ' ~ng-by-van;lble anr.dmte mattix ubilJui-
' " ,. .
of analysis-usually marked blocks of text or informants. Researchers . , ' '"' ,..... onverre ro s I' ,
mwhlch the cells contain measure of ' " . Iml atlty matrices (thmg-b}'-thing matrices
Irs 0 f!t lings
need the flexibility to creat,;: matrices on demand, whether they be word- wltI 11iVIDS,(for step-by-step instructi
'. s 511111 anty
B
among
.
pa' ' ) and then- analyzed
OilS, see orgattl, 1999), .
by-theme or word-by-informant matrices for word analysis and sentence-
by-code or paragraph-by-code matrices for coneene analysis, A series of
word analysis functions would greatly enhance the automated coding fea-
o References
tures found in programs that are geared to the interests of scholars in the
grounded rheory school. Investigators should be able to code a secrion of
Agar,M. (1979) . TI1el11eS revisited: Some problem" .'
rext using grounded theory, then identify the key words associared wirh Course Processes, 2, 11-31 s t11 cognlllve anthropology. Dis-
each theme. They should be able to usc key words to search for additional Ag oc, M.
, .. (1980)
' . bener
. Gc.ttlllg . ' qUJlitv stuff: IvIc I d I '., .-
occurrences of the theme in large eorpuses of teXt. Illterdlsciplinarv niche U'b J:{ t 10 0 oglea! cOlllpennon in an

\'V'hen programs make it easy to use multiple coders and to identify Ag;rr, i\/l. (19~1 .. ,- . I all ~lle, 9, 34-50.
_ , _,).
17,19-26. l\.I,<.:rocornputcr5 as field tool
s. C ompwcrsill1d tile I-ll1lt1i1l1Uie~J
Intercvucr a5,cerne"" "Hd d""ococmcne> "y"cm",;coll y, cesearcbers will
be better able to describe rhemes and to train assistants, Adding a variety of
o Agar, M. (1996) . .Speakillg uf et/mogmphy (2nd ed.). TIlOusand Oaks, CA: Sage.
measures for calculating intercoder agreement, which only some pr -

295
294

Agar, Ivl., & Hobbs,]. (1985). How to grow schemata out of imerviews. In]. W D.
Dougherty (Ed.), Directions in cognitive anthropology (pp. 413-431).
Urbana: UniversitY of J1linoi, Press.
Albert, s. M. (1991). C~gnition of caregiving tasks: J\:lultidimensional scaling of
the caregiver task domain. The Gerontologist, 31, 716-734.
Allen, J. T., & Italic, G. (1954). A C(lncordance to Ellri!Jides. Berkeley: University ot
California Press.
Araujo, L (1995). Designing and refining hierarchical coding frames. In U. Kdle
(Ed.), Computer-aided qllali,~ative data analysis: Theory, methods and IJrdC-
tice (pp. 96-104). London: ~;age.
t'Jarkin, 5., Ryan, G. \Y/., & Gelberg, L. (1999). \V'hat dinici3ns can do to further
youth violence primary prev'~ntion: A qualitative study. Injury Prevention, 5,
53-58.
Barnett, G. A., & Danowski, J. f,. (1992). The strucLUre of comnlunication: A
network analysis of the International Communication Association. HllllhlJI
COIIIJl11lllicatioll Research, 19, 164-285.
Becker, H. S. (1998). 1iichs of the trade: HolV to thinh abollt )'0111' research while
YOII're doing it. Chicago: University of Chicago Press.
Becker, H. 5., Geer, B., Hughes, E, c., & Strauss, A. L. (1961). Boys ill white: Stl/-
dcnt wltttre in medical school. Chicago: University of Chicago Press.
Berelson, B. (1952). Content analysis in colmml1licatioll research. Glencoe, IL: Free
Press.
Bernard, H. R. (1991). About ten management and computers. CAM Net{)slellel;
3(1), H, 7, 12.
Bernard, H. R. (1993). Methods belong to all of us. In R. Borofsky (Eel.), Assessing
Cllfttlml allthropology (pp. 168-178). New ""rork: McGraw-Hill.
Bernard, H. R. (1994). Research n:etfJods in anthropology: Qllalihllive and qUilnti-
talive approaches (2nd ed.). \xralnut Creek, CA: Altaivlira.
Bernard, H. R., & Ashton-Voyouc:tlos, S. (1976). Return migration to Greece.jOllr-
/lal of the Steward Anthropological Society, 8(1}, 31-51.
Bernard, H. R., & Ryan, G. \Y./. (1998). Qualitative aod quantitative methods of text
analysis. In H. R. Bernard (Ed.), Halldboo/~ of methods in wltJlml /.1Itthropul-
ogy (pp, 595-646). Walnut Creek, CA: Alta.1vlira.
Bernard, H. R., Ryan, G. W, & B(lrgatti, S. (1999). Grecn cognition and behaviors.
Report submitted to Ford Research Laboratories, Dearborn, MI.
Bessinger, J. B. (1969). A concora'ance to Beaw1Ilf. Ithaca, NJ:': Cornell Universiry
Press.
Bloor, M. (1976). Bishop Berkeley <lod the adenotonsillectomy dilemma. Sociol-
og:V., 10, 43-6l.
Bloor,lv1. (1978). On the analysis of observational data: A discussion of the worth
and uses of inductive techniques and respondent validation. Sociology, 11,
545-557.
296
-~,- ",...." .... ~c"'''''''' UIIU t-\IIUlYSIS IVler/rOGS
-----
Bogdall, R. c., & BikJen, 5. K. (1992). Qua/itatiue research foreducatiol1: A11 !lllro-
dllclion to theory mid methods (2nd ed.), Boston: Allyn & Bacon.
Borgatti, S. P. (1992). ANT[-lROPAC 4.0. Columbi;J, SC: Analytic Technologies.
Available Interner: lmp:/lwwvv.analytictech.com
. Borgatti, S. P. (1998). The methods. In V. C. De Munck & E. J. Sobo (Eels.), Using
methods in the field: A practical introduction and caseboof~ (pp. 249-252).
Walnut Creek, CA: AltaMira.
Borg;Jtti, S. P. (1999). Elicitation merhods for cultural domain analysis. In
]. Schensul, Ivt LeCompte, S. Horgani, & B, Nastasi (Eds.), The etlmogra-
pher's toolkit: Vol. 3. Enhanced etfmogril!Jhic methods (pr. 115-151). Walnut
Creek, CA: AltaIVIira.
Boster,]. (1994). The successive pile sort. Cult1ll'al Allthropology Methods jOl1l"1lJI,
6(2),1[-12.
Bradley, C, l\--1oore, C. C, Burton, M. L., & White, D. R. (1990). A cross-cultural
historical analysis of subsistence clwnge. American Anthropologist, 92,447-
457.
Brouwer, IVl., Cbrk, C. C, Gerbner, G., & Krippendorf, K. (l969). The television
world of violence. In R. K. Baker & S. j. Ball (Eels.), lHass lIIedii? ,md uiolmce:
I~. report to the National Colltmission 011 the Causes and Preuelltioll o(
VIOlence (pp. 311-339, 519-591). \Vashington, DC: Government Printing
Office.
Bulmer, 1v1. (1979). Concepts in the analysis of qualirative data. Sociological
Review, 27, 651-677.
Burton, D. IvL (1981a). Automated concordances and word indexes: The early six-
ties and the early cemers. Computers alld the Htl/1wf1ities, IS, 83-100.
Burton, D. M. (1981b). Auwmated concordances and word indexes: The process,
the programs, and the products. Computers I.1IId the Huml.1nities, 15, 139-
154.
Burton, D.l'vL (1982). Automated concordances and word indexes: Machine deci-
sions ;md editorial revisions. Computers and the Humanities, 16, 195-218.
Call1 p b,ell, D. T (1957). Facwrs relevant to the validity of experiments in social set-
tings. Psychological Bulletin, 54, 297- 31l.
Campbell ~ D' .T, & Sra n Iey, .J.C. (196') .). E'xpenlllenta
. {atl/./.
J q1lasl-ex/Jerimelltid
.
deSIgns (or research. Chicago: Rand McNally.
Carey,). W, Morgan, lVI., & Oxwby, M.]. (1996). Intercoder <lgreemcm in analysis
of responses to open-ended interview questions: Examples from tuberculosis
. research. Cultllral Anthropology l'dethads joumal, 8(3), 1-5.
Carley,K. (1991) , . C0 d'lng CI '
lOlces f l
or textua ' A comparison of content
analYSIS;
'''''''pi, dB" map analy'1'. In r. Marselen (tel.), Sociologlcn! !JIathodQIQ&v
(pp. 75-126). Oxford: BlackwelL -
Carley, K...(1907)
~
N I I '
. etwo!".( text ana YSIS: The network position of concepts. In
C. \XZ Robens (Ed.), 1(!xt analysis (or the social sciCllces: Methods for drawing
297
 -"'_ . . _ ... -,'~, "' .... '1"'" "ICU,UU';'

statistjCid inferences {raPI texts alld transcripts (pp. 79~lOO). i'vlahwah, NJ: statistical inferences fram texts and transcripts (pp. 103-116). i'vbhwah, NJ:
Lawrence Erlbaum. Lawrence Erlbaum.
Carley, K., & Kaufer, D. S. (1993). Semamicconnectivity: An approach for analyz- Danowski, .f. A. (1981).
.
A network-based Cl.JntCnt analvsis .
metlwdoloHvb'
for
inIT semantic networks. Comlllllllicationl1Jeory, 3. 181~113. c.ompurer-rncdlated cOllllllunication: An illustration with a compmer bulle-
Carley, I~., & Palmquist, P. (1991). Extracting, representing, and analyzing mental nn board. In 1v1. Burgoon (Ed.), COlllll1ll11icatiol! yearboolc 6 (pr. 904-925).
models. Socia! Forccs, 7'), 601-636. Beverly Hills, CA: Sage.
Carmines, E. C., & Zeller, R. A. (1979). Reliability alld validity assessment. Bcverly Danowski,,r. A. (1993). Network analysis of message content. In \Xi: D. Richards &
Hills, CA, Sage, G. A. Barnen (Eds.), Progress ill c01lllJllfllicLltiol! science (Vol. 12, pp. 197~
Casson, R. (1983). Schemata in cultural anthropology. Anll/w! Reuielv olAllt!no- 21]). Norwood, NJ: Ablex.
po!ogy, 11,429-462. . Darton, I'v1. (1976). Modem concordallce to the Nell' Testall/cllt. Garden City, N't':
Clurnr,lz, K. (1990). "Discovering" chronic illness: Using, grounded theory. Socw! Doubleday_
Science dlld .Medicille, 30, 1161-1172. Deese,]. (1969). Conceptual cJtegories in the study of contem. 1n C. Gerbner,
Cohen, J. (1960). A coefficient of agreement for nominal scales. Edt/catiollal and O. R. Holst!, K. Krippendorf, \X( J. PaisleY'"' & E j. Stone (Eds.), The Lllla/vsis
Psychologiedl fvlcasllremellt, 10, 37-48. ofcOII.mllll!icatioll C01l ten t: Deuelopl1lems in sciCliI iti"c theories alld COlllp;lter
Colby, B. N. (1966). The an:llysis of culture content and the panerning of natrative techmqucs (pp. 39-56). New torle John \X/iley.
concern in texts. Allleril;m/ Al1thropologist, 68, 374-388. Denz!n , N .K
.
(1978) . 7'I,C
I,
r>5"al"cl'
to r;
t "I)eOretlcil
"dC: 11
" "I 1I!lrc)(
" 1ttctlOll
" to sociological
Conklin, H. (1955). Hanun6n color caregories. SOlllhwest ]ol/rJlill of AllthrofJOI- lI1ethods (2nd ed.). New '1'orlc NJcGraw-Hill.
ogy, 11, 339~344. Dl.'nzin, N. K. CJ 997). Interpretiue ethllography: Ethnographic practices (or the 11s1
Cook, T D., & Campbell, D. T. (1979). Qllasi~ex(Jel'illlclltillioll:Desigll i1Julllllaly- century. Thousand Oaks, CA: Sage.
sis isslles {!)r field settillSs. Chicago: Rand rVIcNally. Dev-' ~.J (199")
,).
Q)hl"I", ", I" I" ""
I i1tlUc (ata dlW YSIS: A user-fnendly gllide for social scielltists.
Cowan, G., & O'Brien, M. (1990). Gender and survival versuS death in slasher London: Rourledge & Kegan Paul.
films: A content analysis. Sex Roles, 13(3-4),187-196. Doerfel, 1\11. L., & Barnetr, G. A. (1996). The Lise of Cupac for text an<llvsis. Cul-
Crabtree, B. E, & IVliller, \X: L. (1991). A template approach to text an,llysis: tllral Allthropology Methods]oltnlal, 8(2),4-7. .
Developing and using codebooks. In B. E Crabtree & W/ L. Ivl.iller (Eds.), Drass, K. (1980). The analysis of guaJit<lrive dat<l: A computer progr<lm. Urban Li(e,
Doing q!wlitative resc{[;"ch (pp. 93~109). Newbury P,lrk, CA: Sage. . 9,332-353.
Craig, R. T. (l981). Genera!izltion of Scott's index of intercoder agreement. Pllbhc Ember, C. R., & Ember, ?vl. (1992). Resource unpredictability, misrrust, and war: A
Opinioll Quarterly, 45, 160-264. , '. cross-cultural study.]ollr1wl ofColl(fict Resolution, 36, 242-262.
Cressey, D. R. (1971). Other peop!e's mOlley: A sl1Id)' i/1 the social psycholab""Y at End'so
\. 11,.
P (199-) C . -, .' .
! . omracepuve methods: Do Hispanic adolescents and their
~lIIbez:::.lelllel!t. .Nlontdair, N]: Patterson Smith. (Original work published fami!y phu.Hling care providers rhink about contraceprive methods rhe same
1953) way:, Medical Anthropology, 17, 65-82.
D'Andrade, R. (1991). The identification of schemas in naturalistic data. In lvl..J. Fan, D. P., &.Shaffer, C. L. (1990). Use of open-ended essays and computer contem
Horowitz (Ed.), PerSO/l schell/as alld maladaptive illLUlpl!fsollal (Jattel'lls analYSIS to survey college students' knowledge of AIDS. College Health 38
(pp. 279-301). Chicagc: University of Chicago Press. 221-229. ' ,
D'Andrade, R. (1995). The develofJllleI!t of cognitive anthrot.JOlogy. Cambridge: lvI '-':x- Farrmgc Ion, jM G (1 9 ~O).
J ~',
Farringdon ' .. > ' A cOllcorda/lce al!d lUord-lists 10
Cambridge University Press. . . ,the poellls of Dylan Thomas. Swansea, England: Ariel House.
D'Anclrade, R., Quinn, N., N~rlove, S., & Romney, A. K. (1971). Categories of (I!s- FIelding
'. ,N. .G, L:S:: F'ld"
Ie mg,..
r L . ( 1986).
, " " dma. Beverly Hills, C1\: Sag-e.
Lltlk.mg
ease in American-Engiish and lvlexican-Spanish. In A. K. Romney cr ,11.
FIscher, .M. D (1994) A
.
r . .' .
r . ~
. (JP IcatlOllS m COIII(JlItlJlg {or socwf c7llthrofJOlogists. Lon-
(Eels.), Alfltltidimellsiollul scaling: Theory alld applicatiol1s ill the behavioral don: Routledge.
sci<.."c:c.o; (\'01. 2, pp. 9_.5~i). New "\'urlc.: 5"'1";";1'". Fjellm;ln .s i'v! &: Gj t 1 I-I (1"85) hI "" "
, '. ". - .Ie W1l1 ./ l Jitlan tJmily pmrerns of migrarion [0
D:lllielsotl, \VI. A., & Lasorsa, D. L. (1997). Perceptions of social change: 100 years n' South Flonda. HlIlIWI! Olgallhatioll, 44, 301-312.
of front-page content in the New York Tillles and the Los Angeles Times. In : NI. S" & I-hrri
. els!ll.'r, ,n;:,r"01 r\
1,_. f
(1998) f" I" ""
( . "ree lsung: j"Ianagcment ar a women's
C. W Roberts (Ed.), Text analysis lor the social scicnces: Methods (or drl1Lcillg lederal prison canlp_. In V." C 0 e j"1
\ LIne ~
r
-I' & E'." SI) b0 (Ec
-Is.),"
Usmg /1/ctho(Is III
"

298 299
 - - - - ....... , ''--'":;}1:;;'' Ie-, 0< UIIU MIIUIYSIS 1V1er1lO(/S

the field: A jJractical introductio/1 alld Cl-1sebooh (pp. 69-84). Walnut Creek,
Cuba, E. G., ~ ~il:coJn: Y S. (1982). Epistemological and methodological bases for
Ci\: AlwIVlira. naturalistic mqUlry Educational C _ . .
~,
, 2.=,_)-1)2.
_ . . 011l/Jllfl1laItIO/IS <md Tec!J/1olof{)'
Fleiss, J. L. (1971). !'vleasuring nominal scale agreement among many raters. Psy- ..)0, '- -{oumal
- ,
chological BI/lletin, 76, 378-381. Hammerslejr,Iv1 (1997) \\7/ .,' .
hake, C. O. (1962). The ethnographic study of cognitive systems. In T. Gladwin & .
, Ions. L .d e JO( i0 {oglca
- . - 11 - SWl"OlIg WIth etlJJlor>m{Jh),:? NIl/ . / exp {om-
(> -
on on: Routledge. .
\\Z C. Sturtevant (Eels. I, Allthropology and Int/llan behavior (pp. 72-85).
Harm~n, R. C. (1998). Triad questionnaires: Old age in Karen and IvIava culrures.
\X!ashingtol1, DC: Andwopology Association of \Xlashington.
.11 :- c. D~ IVlunck & E. j. Sobo (Eds.), Using methods ill the field:'A practical
Frake, C. O. (1964). Notes 0'1 queries in ethnography. American Anthropologist, mtlodllctlOlliJ11dcL1sebook (pp 171-1'8) W I ~ - I' .
66, 132-145. Harti 'a f A - l - . ' ~ - .) a nut Cree (, CA: Alt3tv1lra.
g n,.. . (Ln.)). Clllsterlllg algontlJllls New voll .. J 1 HTI
H _- I F - _ _ _ . l~. 0111 \\ ley.
Furbee, L. (:1996). The religion of politics ill Chiapas: Foltnding a wit of com mll/li- atc 1'_" & Re~lpath, H. (1954). A cOllcordallce to the Sept1lagint <md the other
catillg saillts. Paper prejcmcd at the 96th Annual Meeting of the American C reek UeI'SIO/IS of the Old TestalJ1eJ1/ G A . 1 .
. r.I7., UStrl3: A cadenllsche Druck~u.
Anrhropologic<ll Association, San Francisco. VerJagsClnstalr.
Furbee, L., &;: Benfer, R. (19~n). Cognitive and geographic maps: Study of individ- Henley N 1\1 (1969) A I I . I I
.' - ',_ _ J, . _ p~yc 10 oglca stU( y of the semantics of animalterms.jollr_
ual vari,-uion alllong Tc'iolabaIMayans. Americoll Anthropologist, 85, 305- . na! of \wbal Le,mlll1g and Verbal Behavior, S, 176-184.
Hdl, C. E. (1998). Decision modeling: Its use in medical anthropology. In V. C. De
C;1rfO, L. (1998). On the rationality of decision-making studies: Part 1. Decision ~'fLlnck & E.j. Subo (Eds.), UsiJIg methods ill the fidd: A practical imrodllc-
models of treatment chnice.l'dedic,11 Anthropolo!,'}' Qllarterly, 12, J 19-340. _ ."_ t'O/l all~ Ci-1seboo!<. (pp. 139-164). W.::dnut Creek, CA: AltaI'vlira.
Gatewood,j. B. (1933). Deciding where to fish: The skipper's dilemma in South~ (Imehman, E. C. (1987). Pea Ie as d _ . _ , ..
~ ._, P pro HctS. Anal}sls of a complex marketing
east Alaskan salmon seining. Coastal Zone Managemellt jOIl/"llal, 10, 347- f- _ r::xehange: j~1/mal of ~/Iarl{(!ti!lg, 51, 98-108. -
367. rolland, D. (198)). From SituatIon to i . . H .
! ' mptesslon: env Americans get to knO\v
<

George, A. L. (1959). Quanti-:ative and qualitative approaches to content analysis. tlcmselves and one another. InJ. \\Z D Dougherty (Ed) D' 1 " ' .
lille {//lthrop I , ( , _ -, '. ., lIec 10115111 cog1l1-
In 1. de S. Pool (Ed.), Tr-cllds ilt cOlltent analysis (pp. 7-32). Urbana: University H -I . . _-a ~g) _ pp. J89~41",-). Urbana: UnIversity of Illinois Press.
of lHil10is Press. usn, O. R. (196b). External conflict and internal con<:ellSIIS' '[1, S' S .
case. In P. J ' .< ~ e ltlO- OVlet
~., .. " S~one, D. C. Dunphy, M. S. Smith, & D. M. Ogilvie (Eds.), The

Gerbner, G., Holsti, O. R., r::'rippendorf, K., Paisley, \Y/. j., & Stone, l~ J. (Eds.).
(1969). The mwlysis o( comlll1mication cOlltent: Developmellts in scientific Gt'nela.l luqUireI': A cOll1puter 1!Jproach to cOlltent mudvsis (pp. 14 1_ ~ 5 S)
theories fIlld COI/l/JIIter techniques New York: John Wiley. Cambndge:MIT Press. - - . _1 L
Glad\vin, C. (1989). Ethllogra!J!Jic decision tree modeling. Ne\\'bury Park, CA: Holsti: 0. R. (1969) .. COllteM al1alysis for the socialsciellces and hl/J1Ul/lities. Read-
Sage. mg, rvlA: Adchson-Wesl ey.
Gladwin, I-I. (1971). Decisiollmakillg ill the Cape Coast (Fmlle) fishillg and fislJ Hum, j. C & R A (1995) Iv I
_ . "_ opo,. . . [u tl-level leadership: Grounded theory and
lI1arketillg system. Unpublished doctoral dissertation, Stanford University. l1la{lnstream theory apphed to the case of Generall'vlotors. Leadershi{J 01111'-
tel' y, 6, 3 79-411. ~ ,
Gbser, B. G. (1978). Theorerical seusitivity. Mill Valley, CA: Sociology Press.
hnrichD.j.,lvlullin C &Lim D (199-) Iv .
Glaser, B. G., & Strauss, A. L. (1967). The discovery ofgroullded theory: Strategies . I _,., ,. .). [easunng the extemofprejudicial pre
for qualitatille research. Chicago: Aldine. ~Ia pUbli~ity.in major American newsp<1pers: A content analysis. Iotlmal
. OT/1/lIIl/llCatIO/l, 45(3), 94-117.
or
Goodenough, \'V. H. (1956). Componential analysis and the study of meaning. Lall~
IrUtlta, V. F. (1996). Hidden dimensions r" I d P .
gllage, 32, 195-116. d ' . e\ca e: rogresslve grounded theorv
Gorrschalk, L. A., & Eechtd, R. ]. (1993). Psychological and IlCIIrupsychialric ~t:9.y of qualny care 11l the hospital. Qualitative Health Research, 6, 331'-
asseSS/IIeI!t afJfJlyiJlg the Gottschalh-Glesercontent analysis method to verbal Jakobso n , R., '" 'I II
"-.-lflfF/", "'i..,/J"i~- ,,;;i"6 .-h<? Cou.-sch...,lh--Lkchrel ""-''''r''ler """01""3 s).sl..-,,-,-,. P"ln l'.IQl.HUl1. e r a c J 1\-'1. (1956). fllfldal1lf1ltals nlll111QI111{lP. Tht> W:1l~ue,
Alto, CA: :tVlind Garden.
lang, H.,Y., & Barnett, G. A. (1994 Se b) C 1 ., .
Gub3, E. G. (1981). Criteria for assessing the trustworthiness of naturalistiC inqui- . 1 .. ' .ptem er. u tural dlfH'rences J[J organiza-
nona COmI1lUHlC:1tlOn A _ . I
ries. Edllcational COIli/lllmications and Technology jOll1"11ll, 29, 75-92. ,., 5" : semantlc networ ( analysis. BIIIIetilT de Methodolo-
gle oCIologh]lIe, 44, 31-59.

300
301
 . _.,~,~ ,,".... 'vv..}

METHODS OF c.:Ulll:L Ilfll.:l AI'IU AI'IHLI !.."~<.;J '-"" "",-C'~ .'"

Kruskal, J. B., & Wish, tv!. (1978). MlIl/.idil/lcllsiollal scaling. Beverly Hills, CA:
" 96
Develo iog categories from interview data: Part
Jehn, k.. A., & DOUC tt.,.L.
1. Text ,lI'la IYS[s ane,
(1:
[r)l~ltidimel~sioIl31 scc1ling. ClIlttlral Allthropology
.
Sage.
Kurasaki, K. S. (1997). Ethllic idcntity and its deveIopmel1t among third-generation
Mcthodsjollrllill, 8(2), 1~:-16_ . ' , " d'" P' 7 Japanese Americans. Unpublished doctoral disscrtation, DePaul University_
", ,r- r Doucet L. (1997). Developing caregones for mter~'lew at.1. alt __
Jehn, k. A., & ' , ' .' d' ,I nalysis strategies m undetstandmg Laffai, J. (1990). A concept dictiOlulry of English, with compllter programs for COIl-
Consequences of different co 109 anc a. . , ,- tent analysis. Essex, CT: Gallery.
text. CIIIWri11 Allthropology Ivle/.1Jods joomal, 9( 1)'.-1-7., '" . and Laffal, J. (1995). A concept analysis of Jonathan Swift's A tale of a tub and GlIl-
" r VI" 0 (1993). Hapax Legomenon 11: 1 heorj-, a [hes.wrlls,.
Jehn k A & \verner, . I -(I) 8 IU liver's tmvels. Computers and the [-[1I11U1IIities, 29, 339-361.
, '." , C It . r AlItIJropolob'V Methods jOllrna ,) ,'- .
word frequcncy. U lila' - .' TI f, "I, _ Leinhardr, G. (1937). Development of an expert explanation: An analysis of a
. W'II" 'D R (1997). Ocdipus UblqultO IlS : JC 1<-11/11) COlli
Johnson. A., & Pnce- I lams" . . ~, f d U . , .'t' Press sequence of subtraction lessons. Cognition d/ld InstructioJ1, 4, 215-282.
)/~x in world (olh litemwre. Stanford, LA: Stan or 11Iver~1 j-" ,7' 741~
Leinharclt, G. (1989). lvlath lessons: A contrast of novice and expert competence.
1 5 C (1967) Hietarchical clustering schemes. Psyc!JO/lIClnka, 3_, -
101nso11,
I . . ' . jOlfrllal (or Research inlvlathellh1tics Educatio/l, 20(1), 52-75.
253. I 1( L 11.'1 (1996)_ Presidenrialads as nightly Leinhardt, G., & Smith, D. A. (1985). Expertise in mathematics instruction: Sub-
"c! L L T,d' 0] C. &1\c [nnon, . n . ' .
kat , . . , e esc,," 'I ' [1988 111997 televised Adwatchcs.jollrl/alo! ject matter knowledge. JOllrllal of Educational Psychology, 77, 147-271.
newS: 1\ content ana YSIS 0 , al l -
Light, R. J. (1971).lvleasures of response agreement for qualitative dara: Some gen-
Broadcasting Electronic Media, 40, 2~7-308: " ". Univcrsir' of California eralizations and alrernatives. Psychological Bulletill, 76,365-377.
"... ' I-I (1983). A cOllcordance o{ thc QUI' <-111. BetLele). }
I,-aSSIS, .. lincoln, Y. S., & Cuba, E. G. (1985). Naturalistic i/lquiry. Beverly Hills, CA: Sage.
Press. i\1 (199')) S'lh"lgingse1f' A Lindesmith, A. R. (1968). Addiction and opidtcs. Chicago: Aldine. (Original work
"." . '1 H Ivlur hv S.,lr"\'in,K.,&RosenbJum,J' . . ," . < " "
[,-carn e y,1\. -', - p., I" __ . e Nllrsillf; Resc[lrch, 44, 208- published 1947)
grounded theory of pregnancy on crac ~ cocall1 . <-

lofland, J., & Lofland, L. H. (1995). Analyzing social settings (3rd cd.). Belmolll,
213. _ ". '" b. ' M (1994). Motheting on crack CA: \X'ads\Vorrh.
r. .. M H Murphv S 0: Rosen aU l1l , . . - -
kearney, .' A' u"d~d tl~:'orv analvsis. Social Science and Medicine, ]8, 3)1- Lonkila, IV1. (1995). Grounded theory as an emerging paradigm for compUter-
cocame: gro . .' .
assisted qualitative dam analysis. In U. Kelle (Ed.), COlllplIter-aided qualitd-
361. . _ c'd d ethods in qualitiltivc research. tive data 1111alysis: Theol)I, methods and practicc (pp. 41-51). London: Sage.
-II U (1995) An overVieW of COl11puter-<l1 e I n , 1- {
I,-e e,. - . I' . -j t ' 1-vsis' Theory me! JU( 5
In U. Kelle (Ed.), COIJI!JlItcr-dided q1la Ita{lue L a a dllLl - ' , IVlacQueen, K.Jv!., IV1cLellan, E., Kay, K., & lVlilstcin, B. (1998). Codebook devel-
opment for team-based qualitative research. ClIltllral AntlJtojJO/Of:,I)-r .!.\,I[cthods
alld /m1ctice (pp. 1-18). London: Sage. '. 1'1 I . / senses
.' P J (I (175) COn/jJllter recognitioll o{ fog IS J II 01 ( . JOlin/aI, 10(2),31-36.
l\dly, E. F., & Stone, . . -," . .
iVf;uming, P. K. (1982). Analytic induction. In R. Smith & P. K. Manning (Eds.),
Amsterdam: NOrth-Holland.. D Holland &
. 'rVI (1987) Two theories of home heat c.'omrol. In ' . I') Handbook of social science methods: Vol. 2. QllalitatilJe methods (pp. 273-
l\..empwn, w. . d I It (p 177_7L.:-.
N. Quinn (Eds.), Cltltural 1IIodels ill langllagc dll t )Ollg). p. --- - 302). New York: Harper.
Clmbridge: Cambridge University Press. . d' " f tVlartindale, c., & IV1cKenzie, D. (1995). On the utility of content" <lI1alysis in author
Chang,ng "en er Images U
r- attribution: The Federalist. Computers alld the HflIllanities, 29, 259-270.
C

' l i E (19"')
I\.lrCler :/_. Adorable
. woman, expert m<l11: . . b i
IP I amI "
" d . gemem ElIro/.lCml j01lrllal 0 ~~OCIi1 5}'C)0.
\vomen an men L11 mara .
tl.J' - - ,
tVIathews, H. E, & Hill, C. (1990). Applying cognitive decision theory to the study
of regional patterns of illness treannent choice. American Anthropologist, 91,
C' 363&-3~,:J['''11 !VI L (1986). Rcliubilit}' dnd uo/idit)' ill qualili/tiuc rescdrch. 155-170,
I\.[rk, j., ,I lV [ cr, - .. .
Beverly Hills, CA: S;1gC. , , ,1 ...- t Jle-ITl<lle IVfcCleary, R. (1978). Dallgerous /llell: The sociology of pi/role. Beverly Hills, CA:
_. r'lb "J P(1996).lvLmtoman:Acontentdll.1):>bo,S( Sage.
Kolbe, R. H., &A al1cSt:,... ~ Ad ." 75(4) 1-20.
. , ,. :lIe-audience maoa7.ines. JOIll'1wl of vel u5mg, - , .' lVlcKinnon, A. (1993). Tl ,e multi-dimen,;ioll'-d cOllcordance; f\ new tool lor llterary
lm<lges [n 111. '=' " '.,. "'In anah."i". fl"l,/;(o O'"'/10Jl
_ , (1'05-' "J'I'" L-Il'll\;lll':t: vi ~lualltanYc ,-UHl research. Computers aml the Hlt1l1t1llities, 27, 165-183.
K.racauer, u. .J,J. " 0

Oll1rterlv 16, 631-64L

- ( .'
i 1
\ ,,Itrodllctioll to its metlloL 0 og)'
McTavish, D. G., & Pirro, E. B. (1990). Contextual conrclH analysis. Qllality alld
- .. -d .[ K
[\.f\ppcn o r , . (1980) ('O;ltellt
.' LlIwl)lsis:
. ,< II Quantity, 24~ 44-63.
Beverly Hills, CA: SaR':
303
302
i'vIerzger, D., & Williams, G. (1966). Some procedures and results in the study of
native categories: Tz~ltal "firewood." American Anthropologist, 68, 389-407.
[vliles, lvi. B. (1979). Quajitative dam as an am'active nuisance: The problem of
~lt1alysis. Administra.:ive Sciellce Quarterly, 24, 590-601.
lVli1es, LvI. B., & Huberman, A. M. (1994). Qualitative data mtalysis: All expallded
sOllTcebooh (2nd cd.). Thousand Oaks, CA: Sage.
lvlingers, J. (1989a). An empirical comparison of pruning methods for decision ttee
induction. Machine Leamillg, 4,227-243.
Ming:ers,]. (1989b). An empirical comparison of selection measures for decision-
, tree induction. MaciJine Learning, 3, 319-342.
Mitchell, S. K. (1979). interobserver agreement, reliability, and generalizability of
dara collected in obs,~rvational studies. Psychological Blllletill, 86, 3 76~ 390.
r.
1\/Iorse, M. (1994). Designing funded qualitative research. In N. K. Denzin & Y. S.
L"incoln (Eds.), Handbook of qualitative research (pp. 210-235). Thousand
Oaks, CA: Sage.
IV1osteller, E, & Walbce, D. L. (1964). ]lffercllce alld diS/JIlted authorship: The Fed-
eralist Pailers. Reading, MA: Addison-\'\lesley.
IVlurdock, G. r: (1971). Outline ofCflltllml materials (4th rev. ed.). New I-laven, CT:
Human Relations Area Files.
Nolan,]., & Ryan, G. \' (in press). rear and loathillgat thecinelllaplex: Differences
ill gender perceptiolls and descriptions ofslasher films. Manuscript submitted
for publication.
Ogilvie, D. M., Stone, P. J., & Schneidman, E. S. (1966). Some characteristics of
genuine versus siml.Jated suicide notes. In P. ]. Stone, D. C. Dunphy, M. S.
Smith, & D.lvi. Ogilvie (Eds.), The General Inquirer: A COIII[Jllterapproach to
contellt illlalysis (pr. 527-535). Cambridge: MIT Press.
Osgood, C. (1959). The re presentational model and relevant research med~ods..In
1. de S. Pool (Eels.), ],'ends ill collten! analysis (pp. 33-88). Urbana: Utllverslty
of Illinois Press.
Palmquist, lVI., Carley, K., & Dale, T. (1997). Applications of computer-aidedrext
analysis: Analyzing literary and nonliterary texts. In C. \Y/. Robens (Ed.), Text
(/lIalvsis for the social sciences: Methods for drawing statistical inferellccs
froll; texts alld tmllS,;ripts (pp. 171-189). Mahwah, NJ: Lawrence Erlbaum.
Patterson, B. R., Bettini, L, & Nussbaum, J. F. (1993). The meaning of friendship
across the life-span: Two studies. Communicatioll Quarterly, 41,145-160.
Patton, lvi. Q. (1990). Qualitative evaluation and research metlJOds (2nd ed.).
Newbury Park, CA: Sage.
Pcrchonocl" N., & W'crne,', O. (1969). N8.v8.ho systems of cbssification. Ethltology,
8, 129-242.
Pool, L de S. (1952). Symbols of detl'lOcracy. Stanford, CA: Stanford University
Press.
304
Pool, L de S. (Ed.). (I 959). u'el1ds ill COl/tent mwlysis. Urbana: University of Illinois
Press.
Pons R., Runyan, D., Zcrger, A., & IVIarcherti, K. (1996). A content analysis of
safery behaviors of television characters: lmplications for children's s;:lfcty
and injury.jollmal Of Pediatric Psychology, 21, 517-518.
Prendergast, G. L (1971). A complete concordallcc to the Iliad 0/ Homel: Hilde-
shelm, Germany: G. Olms.
Price, L (1987). Ecuadorian illness stories. in D. Holland L":. N. Quinn (Eds.), Cul-
tlfri.lImodels iI/language alld thollght (pp. 313~341). Cambridge: Cambridge
University Press.
Quinn, N. (1978). Do IvlfalHse fish sellers esrimate probabilities in their heads?
Americall Etlmologist, 5, 106-126.
Quinn, N. (1982). "Commitment" in American marriage: A cultural analysis.
American Ethllologist, 9, 755-798.
Quinn, N. (1987). Convergent evidence for a cultural model of Americll1 marriage.
in D. Holland & N. Quinn (Eds.), Cultfll"almodels inli./IIgllifge ,-wd thought
(pp. J 73-191). Cambridge: Cambridge University Press.
Quinn, N. (1991). The motivational force of self-understanding: Evidence from
wives' inner conflicts. In R. D'Andrade & C. Strauss (Eds.), }-JIIII/dllmotives
<llld clIltllralmodels (pp. 90-126). New 'rode Cambridge University Press.
Quinn, N. (1996). Culture and contradiction: The case of Americans reasoning
about marriage. Ethos, 24, 391-425.
Quinn, N. (1997). Research on shared task solutions. In C. Strauss & N. Quinn
(Eds.), A cognitive theory ofcllltllral meauing (pp. 137-188). Cambridge: Cam-
bridge University Press.
Ragin, C. C. (1987). The colllparatiue method: Moving beyond qualitative and
qlfi./JItihltiue strategies. Berkeley: University of California Press.
[{agin, C C (1994). Introduction l'O qualitative comparative analysis. In
T. Janowski L~ A. M. Hicks (Ecls.), The wmpamtive political ecol/omy of the
,(.Ie/fare state (pp. 299-317). Cambridge: Cambridge University Press.
Ridurds, T J., & Richards, L. (I991). The NUD GIST qualitative daw analysis syS-
tem. Qtlalitatiue Sociology, 14,307-325. .
Robbins,Iv1. C, & Nolan,j.1vL (1997). A measure of dichot'Omous category bias in
free listing tasks. Cultural Anthropology IVfethods jormlal, 9(3}, 8-12.
Robens, C. W (1997). A theoretical map for selecting among text analysis methods.
In C. \\1: Robens (Ed.), Text analysis for the sociili sciences: Methods for draw-
ing statistical inferences (rom texts alld transcripts (pp. 275-283). i\lahwah, NJ:
Lawrence Erlbaum.
Robinson, \1:/. S. (195-1). Tht: logical structurc of analytic induction. Americlln
Sociological ReuielU, 16, 812-818. .
mme
Ro , A. G. L (1995). Boolean comparative analysis of qualitative data: A meth-
odological note. Quality alld Qualltity, 29, 317-329.
305
 .. _ .. ~, "'~"r.ol'" 1."=tlIUU~

ROlls, G. (1998). Pile sorting: "Kids like candy." In V. C. De Munck & E. J. Saba Shapiro, G. (19"97). The furure of coders: Human judgments in a \vorld of sophisti-
(Eels.), Using IIIcthiJds i/1 the field: A practical introduction alld casebook cated sottware. In C. W Roberts (Ed.), Text dJwlysis (or the social sciences:
(pr. 97-110). Walnut Creek, CA: AlralVlira. A'lcthods (ordmwilIg stahsIical illferences fmlll texts and /ralrscripts (pp. 225~
Rosenberg, S. D" Schnurr, I~ E, & Oxman, T. E. (1990). Content analysis: A com- 238). l'vIalnvah, NJ: Lawrence Erlbaul1l.
parison of manual and computerized systems. journal of Pers01wlil)1 Asscss- Shelley, G. A. (1991). The social networhs of people with end-stage rcnal disease:
54, 298-310
lIICllt, COll/parilIg hemodialysis and peritoneal dialysis patients. Unpublished doc~
Rushfonh, S. (1982). A snuetural semantic analysis of Bear Lake Athapaskan kin- tora! dissertation, University of Florida.
ship dassificiltion. American Ethnologist, 9, 559-577. Sleath, B., Svarsrad, D., & Rorer, D. (1997). Physician versus patient initiation of
Ryan, G. \Y,~ (1995). AJfedical decision making among the Kom of Camerool1: psY~hotropic prescribing in primary care setrings: A content analysis of
Mode/il/g holU c/J[iracteristics of illnesses, patients, caretakers, alld COIJ1- audIOtapes. Socil11 Science and lVledicine, 44, 541-548.
[lOll/Ids arlee! treall/ZeIll choice ill a ruml commzJIlity. Unpublished doctoral Smith,! . .1. (-:993). Using A...NTHROPAC 3.5 and a spreadsheet to compute a free~
dissertation, University of Florida, Gainesville. !1st sCllJence index. Cllltural Anthropology klethodsjollmal, 5(3), 1~3.
RV;lt1, G. \XZ (1999). rvle~lsuring the typicality of text: Using multiple coders for Smith,J. j., & Borgatti, S. P. (1998). S::dience counts-and so does accuracy; Cor-
more than just reliJbility and validity checks. NII/Harl Orgmlizl1tioll, 58(3), r~eting and updating a measure for free-list-item sCllience. jOltrlwl of Linguis-
313<122. tiC Anthropology, 7, 208-209.
Rvan, G. \X,:, L~ IVIartinez, H. (1996). Can we predict what mothers do? Modeling Soh;"
-
R (1993) F"I"
'.'
1 "A 1
J la reconstruction: ' t leary on development through adver-
childhood diarrhea in rural TVlexico. Human Orgmlizatioll, 55, 47-57. my. Qlwlitative Health Research, 3, 465-492.
Rvan, G. \"X/., & Weisner, T. (1996). Analyzing words in brief descriptions: Fnthers Spears,~. E, tVfowen,j. c., & Chakraborty, G. (1996). Symbolic role of animals in
and mothers describe their children. Cultural Anthropology Me/hods jOlll'- pn~r Jdvertising: Content analysis and conceptual developmenr.jotll'llal of
lIill, 8(3), 13-16. Bllsmess Research, 37(2), 87-95.
Sandelowski,l'v1. (1995;1). Qualitative analysis: What it is and how to begin. Re- Spiggle, S. (1986). J\.tleasuring social values: A content analysis of Sunday comics
search ill Nursing ,md Heal/h, 18,371-375. . and underground comix. joumal of COl/sumer Research, 13, 100-113.
Sandelo\Vski,lvL (1995bl. Sample size in qualitative research. Research ill Nursing Spr:lelley, j. I~ (1979). The ethnographic interuiew. New York: Holt, Rinehart &
ami Health. 18, 179-183. Winsron.
Schank, R. c., & AbelsoIl, R. P. (1977). Scripts, plalls, goals, and 1I11derstl1l1dirlg: All Stone, P..f., Dunphy, D. c., Smith, M. S., & Ogilvie, D. M. (Eels.). (1966). The Ge!l~
enquiry illlo 11IIII1a;'1 k.nowledge structures. Hillsdale, N]: Lawrence Erlbau~n. eml l/lqllirer: A computer a/}proach to COli tent murlysis. Cambridge: IVIIT
Schnegg, lvL, & Bern::uJ, H. R. ("1996). \'\fords as actors: A method for d01l1g Press.
semantic network analysis. Cllitural An/hropology Me/hods jOllmal, 8(2), Strauss., A
_ . I_. (198-)
- ') 'ItatIVe
I . ~lfa " " analYSIS " for social scielltists. New York: Cam-
7-10" bndge Universitv Press.
Schnurr, E P., Rosenberg, S. D., Oxman, T. E., & Tucker, G. (1986). A methodolo~i Strauss, A" L., & C or b:In,..f (1990)" ""
_ . Bastcs ofql/aMatIVe research: Groullded /heor)'
Gli note on content analysis: Estimates of reliability. JO/lf/wI 0/ Personaht}' procedures alld techniques. Ne\vbury Park, CA: Sage.
Assessment, 50, 6o-I~609. Srrauss A L ~ C b
, . ., c
r
d "
or Ill", (E s.). (1997). Grounded theol}' ill pmctice. Thousand
Schweizer, T (1991). The power struggle in a Chinese community, 1950-1980: ~ , Oaks, CA: Sage.
social network amJysis of the duality of actors and events.jotlmal o/Qrwnll- StrJUSS, C. (1997) "'I . . _ .
, - . w 13r make!) Tony runt Schemas as mauve reconsidered. In
tative Anthropology, 3, 19-44. , R_ 0 Andrade & C . 5, rauss . (Ed 5.,
) II " and cultllml models
-lllIUI1I mO/IVes
Sd1\veizer, T. (1996). Ac::or Jnd event orderings across time: Lattice representat1~n - (pp. 191~214). New York: C::linbridge University Press.
Strauss C (149-)
and Boolean analYSIS " ot- t 1lC po 1""
Itlca 1d"ISputCS 10
"CIlen VII
I age, ChitH Socul!
c.
',' ~ I . Research on cultural discontinuities. In C. Strauss & N Quinn
1 (Eds) 4 "" _" "
Ndu'ol"b, 1.9, 247-2(;(;. " C -. ': cog",t",,, tlJwry or cultural mel-millS IPp. ZlV-ZJ1), Cambridge:
Seidel,]" & Kelle, U. (15'95). DiHerent functions of coding in the analysis at textua, . :lmbnd ge University Press.
larlo r S J '\: n "
data. In U. Kelle (Ed.), COmplltehJU"Ied qua /"l/a/rue " LIaatmla
/ ysr.
"5" TheOl},
. . , . ., c-" Dog'dan R C (19" 84) I lIira d " I""
7-1.Jc se-1JcJ I. '
" ltC/lOll to qua rtairue research methods:
methods ilnd practice (pp. 52-61). London: Sage. l- ] or lIIem1lJrgs (2nd cd.). New )()rk: John \\:tiley.

306 307
Tesch, R. (1990). QIJillitatiue research: Analysis types and software lools. New Yoder, S. (1995). Examining ethnomedical diagnoses and treatment choices for
York: Falmer. diarrheal disorders in Lubumbashi Swahili ..MedicI11 Anthropology, 16 211-
Trost, j. E. (1986). Statisric:dly nonrepresentative stratified sampling: A sampling 248. - ,
technique for qualita:ive swdies. Qualitative Sociology, 9, 54-57. "y'oung, J C. (1978). lIlncss categories and <lction straregies in a Tarascan town.
Van IvTaanen,j., Miller, M., &Johnson,j. (1982). An occupation in transition: Tra- America!! Ethnologist, 5, 81-97.
ditional and modern forms of commercial fishing. Yf/orL1IId Occupatiolls, 9, Young,]. C (1980). A model of illness rreatmem decisions in a Tarascan town.
193-216. American Ethltologist, 7, 106-151.
Viney, L. L. (1983). The aSSEssment of psychological states through content analysis Yule, G. U. (1968). The statisticd study of literary vocabulary. Hamden, CT:
of verbal communications. Psychological Bulletin, 94, 542-56"3. Archon. (Original work published 1944)
Waitzkin, H., & Britt, T. (1993). Processing narratives of self-destructive behavior Zn311iecki, F. (1934). The method of sociology. New York: Farrar & Rineh3rr.
in routine medical encounters: Health promotion, disease prevemion, and Zuell, C, Weber, R. P., & l'v1ohler, P. (Eds.). (1989). Com/HIler-assisted te.yt allalysis
the discourse of health carc. Social Science and ll'Jedicille, 36, 1121-1136. for the social sciences: The General Inquirer III. IVlannheim, German\-': Ccmer
\';taHacc, A. F. C. (J962). C.dture and cognition. Scie1lce, 135,352-357. for Surveys, l'vlerhods, and Analysis (ZUlvIA). .
\X!cber, R. (1990). Basic COlltClft analysis (2nd ed.). Newbury Park, CA: Sage.
Weitzman, E. A., & i'vliles, Iv1. B. (1995). Compltter programs forq!h1litatiue data
dl/idysis: A. softluare :oflrcebooh. Thousand Oaks, CA: Sage.
\X/eller, S. C. (1998). Srnlclured interviewing and quesrionnaire construction. In
H. R. Bernard (Ed.), Hmdbooh of methods ill cllltltral (/l1thropolo'lj}'
(pp. 365-4(9). \X!alnllt Creek, CA: AltaMira.
\Veller, S. C, & Romney, A. K. (1988). Systelllalic data collection. Newbury Park,
CA: Sage.
\\leller, S. C, & Romney, J\. K. (1 ~)90). Metric scaling: Correspondence illli1lysis.
Newbury Park, CA: ~;age_
\"\/eller, S. C, Ruebusb, T. K., II, & Klein, R. E. (1997). Predicting treannent-
seeking behavior in Cuatemala: A comparison of the health services researcb
and decision-theoretic approaches. Medical Allthmpolof,-')' Quarterly, 11,
224-245.
Werner, 0., & Schoepfle, G.lvl. (1987). Systematic fieldworh: Vol. 1. Ethnographic
ilnalysis il!U.! data management. Newbury Park, CA: S3ge.
\X/hite, D. R., & Burton, .M. L (1988). Causes of polygyny: Ecology, economy,
kinship, and warfare. American Anthropologist, 90, 871-887.
Willms, D. G., Best, J. A., Taylor, D. \XZ, Gilben,]. R., \XTilson, D. lVI. c., Lindsay,
E. A., & Singer, J. (1 1)90). A systematic approach for using qualitative meth-
ods in prim3ry prevention research. JVledical Anthropology QUilrterly, 4,
391-409.
\\lilson, H. S., & I-Imchinson, S. A. (1996). Iv1erhodological mistakes in grounded
tbeory. Nursing Rescdrch, 45, 122-114.
\\!odak, R. (1992), Strategies In text production and text comprehension: A new
pcrspcct;"e. I" D. g,.~ ;jj (R.I.); C"''''f-".>r.d;''S <L;,1, "-'rillen texts (pp .~9J-S.:!.!n-

New York: l\llouron de Gruner.

\'I'/right, K. B. (1997). Shared id;ology in Alcoholics Anonymous: A grounded
[heory approach.joumal ofHealtlJ COll1ll1ll1ticatioll, 2(2), 83-99.

308 309
 "",UI[VVUfe ana yuoMa/we /":eseorch

(d) an indication of future directions for both scholarship on the use of

8
QDA software Zlnd development of such softw;lrc.

<> A Minihistory of theUse of

Software and Computers in Qualitative Research

Qualitative Research Traditionally, qU<1litative researchers have carried OUt the mechanics of
analysis by hand: typing up field notes and interviews, photocopying them,
"coding" by marking them up with 1118rkers or pencils, cutting Jnd pJsting
the m<1rked segments Onto file c<1rds, sorting ':l11d shuffling cards, and typ~
ing up their analyses. This picture h8S been slO\vly changing :;ince thc early
Eben A Weitzman to mid-1980s. At that point, some researchers Were beginning to use word
ptocessors for the typing work, and just a fe\v were beginning to experi-
ment with database programs for storing and accessing the:r texts. ivfost
qualitative methods textbooks at the timc (c.g., Bogdan &.. Bifden, 1.982;
Goetz & LeCompte, 1984; Lofland & Lofland, 19.'.-:4: l'diles C':: Huberman;
1984) m<1de little, if any, reference 1'0 the usc of compurcrs.
.' "3rc available to suppOrt the work of qualitari:re In the early 19805, a couple of programs designed specificl!!y for the
The arnlY of soft\\
.
. . f . f I tools 'lre now avatl-
A wIde vanety (} use II <
analysis of qualitative dara began to appear (Drass, 1980; Seide! & Clark,
researchers IS maturIng. I" litative research. Ivlosr 1984; Sbel1y & Sibert, 1985). Early programs like QUALOG and the first
rHerem approac Ies to qua .
able to support
. many1 enl f 'III d so f.
now [war e tl,at
'- is ap\Jfopriatc to thelf versions of The Erhnograph and NUD"JST reflected rhe staL: of comput-
qualitatIve researe lers ea f l ' d . . nd tlleir ease-of-usc and cost ing at that time. Researchers typically 8ccomplished the coding of texts
. -1 [ re 0 t 1elf <lra, a '-"
analysIs plans, LIe strue u. I . 'ate m'nch still requires system- (ragging chunks of text vvith labels-cades-that inclicmc rl-k conccptual
1 kmg t lac appropn
categories the rescarcher \vanrs to son them inro) hy typing in line nUll1~
<-
preferences. 1- owever, rna . d I -I (s) and careful com-
. {I ,d f the project an tIe reseafl.. ler ,
atic analysIs 01 t Ie nee 5 (}, . '1 bl t the time of purchase with an eye bers 8nd code nalnes at a command prompt, JncJ there \vas little or no bcil-
. f I ft r'ue optlons aVaI a e a
panson (} t.lC so \\ '- Tl 'C is still no one best program. ity far memoing ar other annotation or markup of tex!:. In comparison to
kept fixed fIrmly on those needs. leI f and cannot do to
d d vhat so [ware can
m8rking up texr with colored pencils, this felt awkvvard to many research-
To help researchers un erstan "\ d" b th the potential benefits and ers. And computer support for the analysis of video or Judio data was at
sUI'port their research efforts, un d ~rst.an 0 I 'ccts and find sofr- best a famasy.
comput~rs dU~
. htatlve researc 1 pro] , <

pirfalls of usin.g 111 vide in this chapter (a) an introduc- Bur the landscape has changed dramaricJ.lJy, in terms of berh sofrware
ware that is SUited to their nee IS, fPro . 'In qualitative research, (b) a and the lirerature devoted to it, Bv the time the late lvratt j\;fiJes and J wrote
f tl e ro e 0 so f tware "
tion to and overview o. 1 . . I f' Id about the u11- ComPuter Programs (or Qualita;ilJe Data Aual)'sis (Weirzman & Miles,
.. I d 1.: t . d concerns In t le Ie <
discussion of the CfInca e 1a es an I' . d t analysis (QDA) 50ft- 1995b), We reviewed no fewer tlun 24 different programs that WCte useful
pact and appropnateness 0 ~s
f 'ing qua Itatlve a a d
f t match individual needs, an lor arlalyzing qualirative dara. Half of those progtams had beer developed
ware, (c) guidelines for choOSIng so tware 0
Specifically for qi,wlirative dar<1 an<1lysis, whcre8s the other half hJd been
developed for more general-purpose applications, such as text search and
.- r. t Ilan~"
h to Norman Denzm, . I F"Ie ld"lllg, Uda Kelle, Iby 1_I:'1: " and
. NIge
AUTHOR'S NO
IVlortl:n Levin forI E: r:t
their comments 0
nan earliet draft of this chapter. Since then, the field has continued to grow rapidly. PrJgrams 8re
revised at a regular ratc, new programs appear on tbe s,:ene elf rhe
310
311

rarc of one or two a year, and programs that don't find users disappear.
There has been some convergence as good features in one program are
imitated by the developers of others. And there has also been divergence,
as developers look for new and different ways to conceptualize suppOrt
for analysis.
There afC now tools available that can help researchers who are using a
wide variety of research and analysis methodologies, from grounded
theory to textual analysis to narrative analysis to interpretive inter-
actio~ism. It is important to emphasize that software is not now, if it ever
\vas, something that is relevam only to "positivist" or "quasi-positivist"
approaches to qualitative research. If you see language in this chapter that
does not match your approach, you may find it helpful to do some specula-
tive translation. For example, if the discussion is about "verification" or
"hypothesis testing" and your approach is postmodern, the discussion
may seem irrelevant. But it may be that there is a way to understand the
cOI~cept that makes sense from your perspective, such as "looking to see
whether there is more material supporting, or contradicting, a certain
assumption or interpretation." The same software tools that someone else
might usc for classical hypothesis testing might be very useful for youI'
purposes.
IVIany programs now allow the researcher to specify relationships
among codes and use these relationships in analysis, and to write memos
and link them to text and codes. Some programs allow the researcher to
create lillks between different points in the text (hypertext), and a small
but grovving handful allow the usc of audio and video in place of, or in
addition to, text. And there are a variety of approaches to linking cate-
gorical and quantitative data (e.g., demographics, test scores, quantitative
ratings) to text and for exporting categorical and quantitative dara (e.r;.,
word frequencies or coding summaries) to quantitative analysis programs
for statistical analysis. Finally, there arc now some free programs avail-
able, notably two from the U.S. Centers for Disease Control: EZ-Text,
which focuses on qualitative surveys, and AnSWR, intended for a marc
general range of qualitative data. The software continues to vary \v.idely,
and it remains very much the case that there is no one best program for all
needs. . d
In parallel with the growth in software, literature reporting swdles an
commenting on the software has begun to appear regularly. Tbere has
been an outpouring, 0 f Journa
' I 'f
' Ies, a senes
artlc ' ! confer-
0 'mternatlona
ences on computers and qualitative methodology, thoughtful books on the
312
..Jl./llVVUlt' UlIU \,!UOllfOrlVe Kesearcn
topic (Fielding & Lee, 1991, 1998; Kclle, 1995; Tesch, 1990; Weitzman
& IVliles, 1995b), and special journal issues (IVfangabeira, 1996; Tesch.
1991), '
Periodically, commentators have raised concerns about whether the
range of available software is dominated by a particular approach, meth-
odology, or epistemology (see, e.g., Coffey, Holbrook, & Atkinson, 1996;
Lonkila, 1995). Although there is certainly room for further development
to support certain specific analytic processes (I offer s.ome sLlggestions
larer in this chapter, and the list appearing in the chaptEr titled "Reflec-
tions and Hopes" in \\Teitzman & lVliles, 1995b, has (,nly begun to be
addressed), these concerns arc clearly missing the marIe In this chapter, I
suggest a wide variety of types of programs that arc available to support a
wide variety of research approaches. Qualitative researcbers arc not lim-
ited only to coding-oriented programs, or even to pre grams expJicirly
marketed to qualitative researchers. For example, as Fielding and Le~
(1998) point Ollt, there arc a variety of options for those wishing to follow
the suggestion of Coffey and Atkinson (J 996) that text retrievers may be
Illore helpful for discourse analysis than code-and-retrieve programs.
Fielding and Lee go on to argue that
developers of CAQDAS [computer-aided qualitative data analysis software]
programs have increasingly included facilities for proximity searching,
which might be llseful for narrative analysis, and for "auro:oding" which
could be adapted to some kinds of semiotic analysis. The pt:lvision of ne\.\'
features in CAQDAS programs reflects the generally close rdationship be-
tween users and developers char3cteristic of the field, and r1-e general will-
ingness of developers to incorporate fearures desired by users even if these
do nOt always accord with the epistemological preferences of the developer.
Since packages increasingly suppOrt procedures, rolltine~ and features
which are new to qualitative analysis or make procedures possible that wcre
not practicable without the power of tbe computcr, it is less wei less plausi-
ble either to arguc that the software is merelv an aid to code<nd-retricve or
to argue tbat code-and-tetrieve is tbe silfe ~Ita nOll of qualilative analysis.
(P,175)
. I address these issues at more length through much of this chapter, par-
tICU! I ' h
H ar~: 111 t e subsections below headed "False I-lopes and Fears," "Real
r,oIPes,
'Ie d."
and "Real Fears," and in the later section headed "Debates in the
313
 Mt I HUU.':l ur I....VL.L.l::I....' H ,U ,...." ''-' ,.." ,,,.... ~ ,~
<> What Software Can and Cannot Do
Simply put, sofnvare can provide tools to help you analyze qualitative dara,
but it cannot do the analyst::; fur you, not in the ::lame sense in which a statis-
tical package like SPSS or SAS can do, S:1)\ multiple regression. NIan)'
researchers have had the hope-for others it is :1 fear-that the computer
could somehow read the text and decide what it all means. That is, gener-
ally speaking, not the case. 1 Thus it is particularly il~lpOrtant ro en~phasize
that using software cannot be ;:l substitute for leaflung data an<~lysls met~l
ods: The researcher must knO\v ,vhat needs to be done, and do It. The soft-
ware provides tools to do it with.
The following arc some of the things computers call be used for to facil-
itate the analysis process:
1. lVl11killg 1I0les in rhe field;
') \Y/ritil/g lip or cranscribing, field notes;
_J. Editing: cnrreccing, extending, or revising field notes;
.1 Coding: attaching key words or rags ro segments of text, graphics, audio,
or video to permit later retrieval;
5. Storage: keeping text in <In organized datab'lse;
6. Search /lIfd retricual: [oelling relevant segments of text and making them
available for inspection;
7. Data "linkil/g": connccting relevant dara segments to each other, forming
Gltegories, clusters, or networks of infonn,nion;
S. A1cmoillg: writing reflective commentaries on some aspect of rhe (bt,l,
theory, or method as a basis for deeper analysis;
9. Contellt analysis: counring frequencies, sequences, or 10Glt'ions of words
and phrases;
10. Data displa)': placing selected or reduced data in a condensed, organized
format, such ;15 a matrix or network, for inspection;
11. COI/c1usioll drawiltg alld vcri(icatioll: aiding in the inrerpret<ltion of dis-
playcd data and the resting or confirmation of findings;
12. Theon' buildil1g: developing systematic, conceptually coherent explana-
tions ~f findings; testing hypotheses;
13. Graphic IIw/J!Jil/g: creating diagrams th,l( depict findings or theories;
14. Report writing: interim and final (adapted from Miles
1994, p. 44).
314
.
Obviously, many of these arc things that researchers can do with a word
processor. Other software, v\rhich is rhe focus of this chapter1 helps with
the other tasks. The developments seen in recent years have made it possi-
ble for researchers to do these things more and more easily, and more and
more powerfully. In the section headed "Types and Functions of Software
for QDA," below, you will find more specific details about what software
can do. Blit first, consider some of the hopes and fears, both real and false,
that people have concerning QDA software.
Folse Hopes and Fears
In \X'eitzman and lVIiles (1995b), we argue:
As Pfaffenberger ... points out, it'5 equally nnlve to believe thac a progr<lm is
(a) <l neutral technical tool or (b) ,In overdctermining monster. The iS~ille is
understanding a program's properties and presuppositions, and ha"\' rhey
call support or constrain your thinking to produce unanricipated eh:cts.
(p.330)
As already mentioned, many people apparently continue to believe that
QDA software intends to do the data analysis. Skeptical researchers raise
challenges to the notion of "dumping my text into a program and see-
ing what comes out." Others express this more as a hope that if they buy
the right program, they wil! not have to engage in the often verv time-
consuming process of analyzing ail that text themselves. QDA softlvarc
provides tools that help you do these things; it does not do them for you.
In an extension of this concern, many researchers have worried about
the soft\vare going yet a step further and "building theory." Bm, a5 ]vhles
and I also argued in 1995, "Software will never 'do' theory building for
you ... 1 but it can explicitly support your inteBectual efforts, making
it easier for you to think coherently about the meaning of your data"
(Weitzman & Miles, 1995b, p. 330).
This situation may change in the coming years. There are some current
efforts to use artificial intelligence (AI) approaches to get computers to
interpret text. For example, the SPSS module TextSmart uses information
about frequency of occurrence of words and proximiry of words to each
other to categorize text responses automatically. Microsoft Word97 has an
"Auto Summarize" feature that aims to identify the most important <"con-
[I{: Huberman,
cepts" in a document according to word frequency. Other develop~rs are
315
 NIt: I MUU2> ur \...v'-'-I..:. ...... , ,,~u r\"'-' r\' . ".... , +..-" ...... _ ....

_ _ _ _ _ _ _ _ _Jomvare and \.dual/ralive Research

thinking about using AI techniques to gct software to participatc ill thc

theory-building process wirh researchers. These approaches rely on things encourages the researcher to revise the analysis and the thinking nbout it
like frequency to indicare importance and proximity in the text to indicate ~\'he~ever ne.ccssa? B~il1g able to quickly pull rogether all the text (or cells
~n a ,col1~plex_l11.atnx display enables and encourages the researcher to ru~
0\\11 PWVOLilllye leads and new ide ' II .
tions, but for many others they are not, and for such researchers the results . as-JS we as Wornes [hat (he current
Cone 1uSlOns may be W'lY off tn, + I .
of these approaches do not yield useful interpretations of text. ~ C '-111UC 1 more often and with much less
L.

cost.

Real Hopes f ~.J~ exa;l1 p le from Cju:lI1titative research 1118}' be instructive. In the days
0" s, ~ e ~Iubels, adnd. eve_ll of handheld ca!cubtors, before statistical softwa~e
\\ as ava! a e olllg bctor an I .
\"'\lhat can we really expect to gain from the use of soft\Nare? QDA soft- f '. ..' <. < ~ YSIS was a monrhs-lo ng enterprise. Now a
ware provides tools for searching, marking up, linking, and reorganiz- ac~ortnalY5lS ~an L~e run III m III utes or seconds on;] desktop computer. As
a resu t, researchers can run factor analyses much more often, as art of
ing the data, and representing and storing your own reflections, ideas,
otht~ atlalyses rather th~n only as major undenakings of their own ~nd on
and theorizing. Some of it gives you tools for further exploration-
which in some cases might amount to hypothesis testing or conclusion ,mu t~p ,e S~ts of scores In dlC same project. The speed of the
alone Cdn change what researchers " -, _ ., I
c;m _ p
mer
verification-based on your theorizing and interpretive work. L L . eVen contemp 3te llndertakmg.

Representation. Software that allows e1}" -, . I . .

COilsistency. Software can help with consistency. If I can search for all the . . I ' . . namIe, red -rIme representatIOn of
a researc ler s thmklng can be a sllbst'lnr, I . I I, '.' S
places a given key word appears, or all the places where a given code or -f
T' . L a dlC to [leonzlI1g.. ohware that
< .

combination of codes was applied, or always see the relationship between P,ro~ Il es a graphIC map of rel.arionships among codes, text seg:11ents, or
t\VO features of the data that I have recorded, it becomes possible for me to ~ases canlhel p researchers to VIsualize ;:md extend their thinking 1bout' the
Cata or t leorv at hand Researche f d. .
be more consistent in a couple of ways. I can be much more careful about .. .. " . I.S a ten use raWlIlgs to depict these
re IatlOnshlps, butsofrware can keel .'-d ... 1 .. c

not missing the data that contradict my brilliant, but wrong, new hypothe- I, maps tle to t le under/vlwr project
so natI cIlanges to the linh in t! d . I - h - )
sis. I can easily review all the data I assigned to a given conceptual category obe " h' " . le rawmg c li1nge the links among the
J cts 1ll t e dat3base, 3nd VICe versa.
or theme and check to sec if they (a) all belong together and (b) still seem to
suPPOrt the interpretation 1started out with; if not, I can easily reorganize.
(Note that the problem of my making bad interpretations has not been ~Ollso1idatioll. Finally, allowing the researcher to record field nOll'S ll1ter-
VIC\VS codes m ' . " '
removed. But the kinds of facilities mentioned here can be tremendously : ,e.mos, annOtatlO1ls, reflectIve remarks) diagrams auelio
and Video recordings ,I n.. 1 . L ,

helpful to competent researchers in checking their own work, as well as d I ---') t. emobldp HC variables, and structural maps of the
ata anc the theory 3U m on ~ I I
in allowing colleagues or research participants to check it and provide _ . e p. dce can le a tremendously powerful sup-
POrt to t Ile analYSIS 'If J I .
feedback.) 'l11o\v. I . I' process. t le c eSlgn of the program is such that it
, . s t le researc ler to move f .' II
witl '. I . rom one Lnrc ecrual activity to another
Speed. The speed of computers is a critical issue in making QDA software atl "1 m~nllna effort) and carryover the results of one sort of thinking to
lers, It can both free LIp lOll' J
helpful. First, a camion: It can take time to learn to use a program, ar':1d help [I I ' <ge an.lOunrs at energy for the critical tasks anel
le researc ler to see 'm :11 p . I - .
once you have, it can take some time to prepare and set up the data for \vise .1 f II I. - < l .eee trac.: at connectlons that might other-
eaSI y a t lJ'ough the cracks.
analysis. But once that is done, the speed of the computer quickly pays for
that investment. Being able to search and re-search almost instantaneously Real Fears
encourages the researcher to conduct multiple searches to zero in on
the data that really apply to a particular question. Being able to quickly re~
son a database, redefine codes, and reassign chunks of text enables and tag~:ltU)t dOd wte realJly ha~re to be wo . rried about? lVlany of the ad van-
<. ute a love lave fl. Jp SIC
. Ies. TIle very ease, speed, and pmver of the
316
317
 rv'll::! HVU) VI- \...VLLl:\...Ilr-'1'.J P.l'lU P.1'V\Ll Lll'll:1 CIVWll\II....P.L IV\MI Ct'-IMLJ
Software and Qualitative Research

software have the potential to encourage the kind of thinking I have '" Types and Functions of Software for aDA
referred to as "false hopes and fears" above. Although the software will
oar figure aU(' \vhat a complex account of a childhood traum3 really means
~11 this seerion I. offer ~1 rough sorting of :1v:1ibbJe software imo types. There
in the context of the currt:llt study, the of 5earching for key ,\YOrd5
IS na[UraIJ~ qulte a. bIt of overlap among categories, with hdividual pro-
ell::;C

and "autocoding" them may encourage the researcher to take shortcuts.

grams hav~n~ func~lOns that would seem to belong to more than one type.
\'l./e mav fail to check to see what passages were actually coded in the
~~wev.er,l~ IS pOSSIble to focus on the "heart :md soul" of a program: What
aurocoding process, and to use their own intelligence to analyze whether
It_IS. maInly Intended for. This categorization scheme was fir.":t presented in
they fir. There is the real potential that we will get lazy. As Lee and Fielding Weitzman and Miles (l995b).
(19"91) have noted, "There is the possibility that the use of computers may
tempt qualitative researchers inro 'quick and dirty' research 'with its atten- Text Retrievers
dant danger of premature theoretical closure" (p" 8).
I[ is also possible that the availability of software may tempt researchers
Text retrievers specialize in finding all the instances (If vvords and
to skip over the process of learning properly about research. Again from
phrases in text, in one or several files. They typically also allow YOU to
Lee and Fielding (1991):
search for pbces where two or more words or phrases coincide within a
specified distance (a number of words, sentences, pages, ;:mc! so on) and
Of course, rhe ultimate fear here is of Frankenstein's monster. It is suscepti- allow you to sort the resulting passages into different Output files and
ble to the same caveats, too. Like the monster, the programs are misunder- r.cport~. They ill,ay do. other things as \vell, such as content <:'nalysis func-
srood. The programs arc innocent of guile. It is their misapplication which tions like countIng, dIsplaying key words in context or crez.ting concor-
poses the threat. It was exposure to human depravity which made a threat of
dances (organized lists of all words and phrases in their comexts), or ther
Fr~mkenstein's creation. Equally, the untutored use of analysis programs can
may allm:v y.ou to attach annotations or even variable valucs (hr things Jik~
cerrainly produce banal, unedifying and off-target analyses. But the fault
would lie \vith the user. This is why tcaching the use of the programs to nov- demogr~pll1cs or SOurce information) to points in the text. Examples of
ice researchers has to be embedded in a pedagogy which I1<IS a sense of the text retnevers are ~onar Professional, the Text Collector, anc. ZyINDEX;
exemplars of qualitative analysis, rather than as skills and techniques to be there arc also a vanery of free (but hard to usc) GREP tools available on the
World Wide Web.
mechanically applied. (p. 8)

Textbnse Managers
The final rear that has some truth to it is that the conceptual assump-
tions behind the program-for example, that the relationships among
Textbase managers are database programs specialized for storing text in
codes are ;:tlways strictly hierarchical-will shape the analysis. This fear re
m.o or less organized fashion. They are good ar holding text together
both has truth to it and is often overstated. For example, jf the program With infor t b . d II . ' c
. .rna IOn a our It, an a OWIng you to quickly organize and SOrt
allows you to directly represent hierarchical relationships among codes,
~our data In a variety of ways and retrieve it accorcling to different criteria.
but not nonhierarchical relationships, such as circular loops or unstrLlC~
"ome are better suited to highly structured data that can be on;anized into
rured netviTOrks, it will probably encourage you to thm . I(pnman
1 y or on Iv_
rccords" (that is, specific cases) and "fields" (variables-info ;.n"<:ltion that
in terms of hierarchical relationships among your codes/concepts. If YOl!o
iPcars for each case), whereas others easily manage "free-;:orm" text.
are aware of the assumptions behind a program, you have a couple ot cy
dh may allow you to define fields in the fixed manner of a traditional
options: You can choose aoot Ilet program or you canIne f I a way to work .ata.b~se such as Microsoft Access or File1vIaker Pro@, or they may allow
around the assumptions in the program-for example, by keeping an ev e.r- SlgnlfIcantl ' II hl I . .
'
h'vo dIll } more eXI I Ity, or example, alJowlllg dIfferent records to
.
changing, nonhieratchlC1 . I code map pll1l1e
. dto i
t lCl
wal . 1\
'1 are. on rhis IJ1 f l .
Or So erent
' .Ie d structures. TheIr search operations may
. be as good ~as ,
the section headed "Debates in the Field," below.
iTIetlIlles even better than, those of some text retrievers. Examples of
318
319
research with low-income minority populations. In her swdy of low-
income African American women, she visited Head Start programs in poor
neighborhoods throughout Chicago and talked to women directly about
her study in order to gain their participation.
lVly sources of recruitment were personal networks such as friends,
students, community leaders, and friends of friends who \vorked in non-
profit, community organizations. The first contact with the recruiter was
usually face-to-face; in a few instances, first contact took place over the
telephone. In cases INhere I did not know the recruiter personally, I asked
my fric11ds and students to initiate the contact. They would mention my
name to the recruiter, indicating that I was a professor doing a study on
fear of crime and informing him or her that I would call at a later date.
They usually informed the person that I was a professor involved in social
justice and social change. As one of my students said, "I told the director
of the organization that y'JU are okay." This personalistic approach was
extremely useful given the reluctance many people of color, and particu-
brly undocumented women, feel about participating in research (Zinn,
1(79).
Although this personal approach works better than other methods for
reaching populations of color, it is still a challenge to gain the participation
of members of some lower socioeconomic groups, such as lower~class
Latinas. Some feel apprehensive about panicipating in group discussions,
especially if they are recent immigrants or undocumented women, live on
welfare or engage in any nonnormative behavior (such as alternative fam-
ily living arrangements), or work in the underground economy. I found
that economic incentives definitely increased the likelihood of participa-
tion in the project, and I believe participants should be offered such incen-
tives as a reward for giving their time.
Some social scientists claim that Latinos are more willing to participate
in research projects (Marin, Perez-Stable, & VanOss Marin, 1989). How-
ever, my experience in doing focus groups with low-income Latinas shows
otherwise. Several factor;:; influence the participation of Latinas in re-
search projects. First, ma,ly Latinas, especially those of lower socioeco~
nomic status, fulfill traditional roles within the family. Indeed, most of
them carry full responsibility for the care of the children and the house.
Second, traditional Latino families are hierarchical institutions in which
men afC considered the cabeza de familia (head of the family). Often,
Latinas exercise their power in the family par debajo (behind the scenes;
see Heyde, 1994). Frequently, in traditional Latino families-even in tWO-
378
wage-earner families-women still perform more of the household work
(Pesquera, 1993). Latinas fulfill many functions as mothers, caretakers,
cooks, and keepers of the house. They often feel that they should be fully
available to their spouses and to their children, making it difficult for some
low-socioeconomic-status Latinas to attend activities outside of the house.
Third, often these women have more unforeseen demands on their time
and less conrrol over their schedules than do other women (Krinitzky,
1990; Stack, 1(74). Hence it is my experience that often Latinas fail to
arrive at interviews or come late. In my experience, in spite of over-
recruiting, only a fevv women will show lip out of 12 to 15 who had con~
firmed their participation. Providing transportation for participants,
either by picking them up or by sending drivers to bring them to the meet-
ing place, or just offering to reimburse them for transportation costs
improves attendance. But it is even better to meet them at their most con-
venient place, such as the SOlIP kitchen where they have dinner or the
church or English-as-a-sccond-Ianguage school they attend.
One reason for the high rate of nonattendance among lower-
socioeconomic-status \vomen is that they may feel uncomfortable saying
no and turning down an invitation. In the case of Latinas, Latinasr'm/Jatfa
emphasizes cordial and affable relationships. However, it may interfere
with recruitment because it makes it more socially acceptable to say yes
and accept an invitation. Thus a commitment to participate in a group may
be made, but later the promise may not be fulfilled (Marin & VanOss
Marin, 1991, p. 13).
In addition, I have found that it is not unusual for these women partici-
pants to encounter emergencies at the last minute that prevent them from
participating in the group-a child gets sick or gets into trouble at school,
a relative needs help with child care or needs to be accompanied to some
appoimment, or a basic service breaks down at home, such as water or
electricity.
It was especially important for my study to have homogeneous groups
in terms of age, class, and race because it was evident that participants felt
freer to express their ideas about images and cultural representations of
crime, criminals, and victims when disclIssing them with other women
from their own backgrounds. Explaining this need to the recruiter, how-
ever, became a very sensitive issue. On several occasions, I was asked ques-
tions such as "\'\lhy do you want a group of all Latinas?" Or "\\lhy only
Black girls?" "\\lhy can't \ve mix them up with \X1hite girls?" In most cases,
379
 METHOD') 01- LULLtL I I I'll;> AI'IU f\1'If\Ll LIi'It.:l t::l~WIl"\.I'-f-\l... IVIM' L''''""'\L-J
---
texrbasc managcrs are askSam, Folio Views, Idealist, InfoTree32 XT, and
TEXTBASE ALPHA.
Code-and-Retrieve Programs
Codc-and-retrieve programs arc often developed by qualitative re-
searchers specifically for the purpose of qualitative data analysis. The pro-
grams in this category specialize in allowing you to apply category tags
(codes) ro passages of text and later retrieve and displa)' the text according
to your coding. These programs have at least some search capacity, allmv-
in~ you to search cither for codes or words and phrases in the text. They
may have a capacity to store memos. Even the weakest of these pro-
grams represent :1 quantum leap forward from the old scissors-and-papcr
approach: they're more systematic, more thorough) less likely to miss
things, more flexible, and much, much faster. EX;1ll1ples of code-;1nd-
retrieve programs are HyperQual2, ](waliran, QUALPRO, l'vlartin, and
the Data Collector.
Code-Bosed Theory Builders
l\10st of the code-based theory-building programs are also based on a
code-and-retrieve model, but th~y go beyond the functions of code-and-
retrieve programs. They do not, nor would you want them to, build theory
for vou. Rather, they have special features or romincs that go beyond
tho~e of codc<l11d-retrieve programs in supporting your tbeory-building
efforts. For example, they may allow yOll to represent relations among
codes, build bigher-order classifications and categories, or formulate and
test thcorctical propositions about thc data. They Illay have more power-
ful memoing features (allowing you, for example, to categorize or coele
vour memos) or more sophisticatcd scarch-and-retrieval functions than
~ode-and-retrieve programs. They may have extended and sophisticated
hyperlinking features, allowing you to link segments of tcxt together or to
create links among segments of text, graphics, pharos, video, audio, \X1e::
sitcs, and more. They may also offer capabilities for "system closure,
allmving you to fced results of your analyses (such as search results or
memos) back into tbe system as data. Examples of code-bascd theory
builders are AFTER, AnSWR, AQUAD, ATLAS!ti, Code-A-Text, Hypcr-
RESEARCH NUD 1ST, NVivo, QCA, the Ethnograph, and winMAX.
0
Two of these' programs, AQUAD and QCA, support cross-case configu ral
320
..JUllwure ana \.!ualltatlve I<esearch
analysis (Ragin, 1987), QCA being dedicated wholly to thi" method and
not having any text-coding capabilities. .
Conceptual Network BUilders
Conceptual network builders are programs that emphasize the creation
and analysis of nenvork displays. Some of them are focused on allowing
you to create network drawings: graphic representations of the relation-
ships ~J~ong concepts. Examples of these arc Inspiration)VletaDesign,
and \1IS1O. Others are focuscd on the analysis of cognitive or semantic net-
works, for example, the program IvlECA. Still others offer SOl11e cOl11bina-
ti?n of the nvo approaches, for example, SemNet and Decisi<m Explorer.
FlIl <1 lly, ATLAS/ti, <1 program also mentioned above under code-based
theory builders, also has a fine graphical network builder connected to the
analytic \vork you do with your text and codes.
Summary
In concluding this discussion of the five main software family types, I
want to emphasize that functions often cross type boundarie~. For exam-
ple, Folio VIE\VS can code and retrievc, and has an excelicllI text search
faciliry. ATLAS!ti, NUD 1ST, NVivo, the Ethllograph, and wiIlNI.A)(
0
graphically represent the relationships among codcs, although among
~hese, only ATLAS/ti alloW5 you to work with and manipulate the draw-
Ing':'- The Ethnograph and \vinlvLAX both have systems for atuching vari-
able valucs (text, date, numeric, and so on) to text files and/or cases.
Sphi.I1x ~urvey allows you to work with survey data consisting of a mix of
qualitatIve and quantitative data. The implication: Do not decide too earlv
d ' l l '1 .
~ lIe 1 ami y you want to choose from. Inste8d, stay focused ('11 the func-
tIons you need.
lv~l~ltimedia. lvlultimedia capabilities 8fe just beginning to emege as a sig-
11lflCant is . 1 I' I
,sue 111 so hvare c lOIce. T lere are now several programs in the
Code-based theory builder categorv that allow Vall to lise audio and video
as welJ as text, as data: AFTER, ATLAS/ti, and Code-A-Texr aJ! allO\v YO~
to code and annotate audio and video files, and search and retrieve f;om
them in - . . 'I h
' \\'ays quIte SImI 8f to t e ways they let you manipulatc text, as
~oes version 2 ..of HyperRESEARCH,
time 01 tl . I I
which is under develop[n.ent at the
11S WrItmg. n t lese programs, you can playa media fili~ (audio or
321
 Mt I HUU::' UI- I.ULLtl.llI\l\..:J AI'H) AI'lAL! LI!'11.;l l:;IVlFlI,''-M~ ,,,.-., 1..''''"''--'
:Jarnvare and Qualitative Research

video), mark the beginning and ending points of segments, and then treat 3. \'(lilat kind of proj"-'ct(') d i b
- - 1; 5;]0 (;]ta ase(s) wiIJ I be \yorking 0 ,
those segments much like segments of text. A program now in beta testing d WI I' d IL
lat WI s of analyses am I pbnning to do?
called InterClipper is designed primarily for audio files, \vith the assump-
tion that you only bother to transcribe the segments that you find most In addition to these fou . J. .
important (it is targeted:Jt focus group researchers in commercial environ- ' I11m':
be ar m 'd I ,-ey questIOIls, there are two C.it-across isslle, to
- .
ments ..vho need to be able to generate analyses and repofts quickly). This
program will probably fall in the code-and-retrieve family when it is ready " Em\' important i" _"
'H'ij . __ c f' Sltt~youtom;]lIlt;]lnaseoseof"closeness"toYourdat;]~
for release. There is also a growing field of software dedicated exclusively n lat clre \'uur lllancJal . j b " c.
, d" constraints W len HYing software 'md the 1 j
to managing video. It nee s to run on? " 1 3 f l Ware

\Vit~1 these basic. issues clear, you wiJ! be able to 100" at specific pfO'
grams Jl1 a more actlve d I"b .
~ How to Mal<e Intelligent, your needs CrT ' f C I .crate way, secIng \vhat does or does not meet
Individualized Software Choices questioIls ~;l a :r~r~::ee~nd It helpful to organize your lnswers to these
199 .lcb ,wI'
HC
hh ' such as the one proposed in \Veitzman & l'vIiles
as rows f ~ J 0 f 1 ,
1 have emphasized from the beginning of this chapter that there is no one
best sofnvare program for analyzing qualitative data. Furthermore, there is o~~s~~.cn'd c:n]~ ~~;:t~~:~l~:~~\~ ~~~r~(o~~::~s\\:,o]yr
'f"flloswer" i mplica;ions/ll i
m answenng quest', h' . . '<
no one best program for a particular type of research or analytic method. gram choice to ' d' lO~s, to t e Implications of those answers for pro"-
Researchers will sometimes ask, What's the best program for a school eth- complex e\,.'1 c.dn ldadte pro.grams. For example, if you arc \'i'orking on a
" d uatlon stu y wuh . b"
nography? or, \Vhat's the best program for doing grounded theory? or, focu d , . a com matIOn of Structured interviews
\\lhat's the best program for analyzing focus groups? None of these ques- thro~gglf'OdUIPffse'an tCd studies, you will need strong tools for tracking case~
ase
ren ocuments Yo . 1 f' d
tions has a good answer. Instead, analysts need to approach choice based Prog, " I . u mIg It'm good suppOrt for this in a
ram S coc e structures or thro I 1 . .. '
on the structure of the data, the specific things they will want to do as part track individual tl J' h ug 1 t Ie use of speaket IdentIfIers that
of the analysis, and their needs around issues such as ease of use, cost, time , s lroug lOut t e database (see Question i below) S 'I
suggestIons are elaborated beJow. ., . . uc 1
available, and collaboration.
Researchers can ask themselves four broad questions, as \vell as con- Question I;
sider twO cut-across issues, to help guide their choices (\Veitzman & 1'diles,
What Kind of Computer User Are You'
1995a, 1995b). These guidelines for choice have seen wide use in practice
since their original formulation, and have proven to be effective for guid-
YOur present level of .'.
ing researchers to appropriate choices. Because this approach to cllOice Pfogra If. . cOmputel use 15 an lJnporranr factor in choice of a
m. youarenewtoco b' '
emphasizes matching functions, rather than specific programs, to partic- a word' mpurers, your est bet IS probably to choose
ular needs, these guidelines can continue to be uscfullong after the pro- learning,prtoocessmg program with advice from friends and begin using it
use your co , . ,
grams referenced here as examples have evolved into new versions and \V,. d
Win ows or lvi ) d mplltef
. g MS -DOS
s operatlllg Sj'stem (e ..,
new programs have arrived on the scene.} moving "f' 'd ~c ,an gettlI1g comfortable \vith the idea of creating text'
a. oun I1l It and r .. '. TI . ,
Specifically, there arc four key questions you need to ask and ans\ver as calJ Levell 0 '' eVISll1g It. . lat would bring yOL to what we'jJ
you move toward choosing one or more soft\vare packages: program . r you maY.have gotten acquainted with se'reral different
s, use your operatl '1
idea of expl' d ng system eaSl Y, and feel comfortable with the
. orIng an lea .
1. What kind of computer user am I? person witI] ',' rnl~g new programs (Level 2). Or you may be a
actIve ll1terest In t l d
2. Am I choosing for one project or for the next few years? (level 3) and f I " le lnS an outs of how plograms work
ee easy WIth cu to ' . . .
S ffilZatlOn, WfltIng macros. and the like.

322
323
 Mt I HUU::' Ur- LULL!:L 111'1\.:1 """lL.J J-\1'1J-\Ll LII'H;J Llvlrll"\...Ml.. /VIMI L''',..., .........
(l \villnot deal here with the "hacker," a Level 4 person who lives and
breathes computing.)
Being marc of a novice does nOt mean you have to choose a "baby" pro-
gram, or even that you shouldn't choose a very complex program. It does
mean, however, allowing for extra learning time, perhaps placing more
emphasis on user-friendliness, and finding sources of support for your
Ica~ning, such as friends or collc8gues, or on-line discussion groups on the
Internet. People at different levels seem to have quite different reactions to
the same programs. So, for example, a person at Level 2 or 3 might like a
program that purs the m8ximum information on one screen because this
allows her to find what she wants quickly, and she might learn the program
vcry quickly. A person at Levell might find all that information ovcr-
whelming at first, and might take a little longer to learn that program
because of it. Bm, once he has lcmned the program, our Level 1 perSall
\vould probably benefit from the layout in the same way as the morc ad-
v:Jnced compurer user.
Ouestion 2:
Are You Choosing for One Project or for the l'iext Few Years?
A word processor does not care what you arc vvriting about, so most
people pick one and stick with it until something better comes along and
they feel motivated to learn it. But particular qualitative analysis programs
tend to be good for certain types of analyses. S\vitching will cost you learn-
ing time and money. Think about whether you should choose the best pro-
gr;m for this projcct or the program that best covers the kinds of projects
you are considering over the next few years. For cxample l a particular
~ode-alld-retrieve program might look adequate for the current project
and be cheaper or look easier to learn than some other program. But if
you are likely to need a more fully featured code-based theory builder
~lown the ro~d, it might make more sense to get started with one of those
nmv (assuming you choose one that includes good code~and-retrieve
capabilities).
Ouestion 3:
Who! l(ind of Database and Project Will You Be Working On?
Here the questions begin to get a bit more specific. As you look at
detailed software features, you need to play them against a series of de-
324
..JUI {ware ana VUO!ltot/ve h!eseorch
------
railed issues. Because of the nature of computers, it becomes essential to
give careful anention to the issue of understanding the naturc and struc-
ture of qualitJtive data sets. The issues here have to do \vith the phv.t:ic;11
and logical form of the data: how strucrured and how consistent it is: 11O\v
c1ata about a case arc organized, and so on. In terms of the isslle~ presented
below, there may be great \rariation from project to project, even within a
given ~naJytic approach (say, grounded theory, ethnography, o:~ narrative
analySIS). Epistemological issues, such as the interpretive nature of obser-
vationalnotes, coding, or memos l or the social construction of interviC\v
~ara, although very Important for research methodology, do not come
llltQ play here; thc question is whether the program you choose provides
the organizational roo Is for the text, graphics, audio, or video you want to
put into it.
Data sources jJercase: single versus multiple. IOU may be collecting data on
a case from many different sources (say your case is defined as;l student,
an~ Y.Oll ralk with several tcachers, rhe student's parents and f[jends, the
~nnclpal, and the student herself). Some programs arc specifically de-
SIgned to handle data organized like this, others are not designed this way
but can handlc multiple sources pretty \ve[l, and some really do not hav~
the flexibility ~ou'!l need. As mentioned abovc, you should l~ok for strong
t~ols for tracJ.'::lng cases through different documents. Some programs pro-
v~de good suPpOrt for this in code structures (particularly progr,ll11s with
hIghly structured code systems, like NUD" IST and NVivo, and to lesser
degrees in programs with flexible code systems like ATLAS/ti) or through
tbe usc of speaker identifiers in programs like AFTER l the Ethnograph, or
Code-A-T~xt. Also, look fOf programs that afC good at making liIlks, such
as thos~ \~lth hy~crt~xt capability, and tbat attach "source tags" teUing you
where l11Jormatlon IS coming from"
Single verSl(S multiple cases. If you have multiple cases, vou uSllallv will
want to sort them out according to different patterns or configura"tions
and/or vvork \vith only some of the cases, and/or do cross-case compari~
so.ns. l'v1ulticase studies can get complicated. For example, your cases
Jnlgbt be students (and you might have data from multiple sources for each
stu~lent)..Your stude11ts might all be "nested" in (grouped by) c1as,;;rooms,
\~hlCb mIght be nested within schools, which in turn might be n,:;:sted in
dlstricts. Look for software that will easily select different porrions of
the database
, ,an djor d a can f19uratlOna
. J ana IYSIS
(Raglll,
. 1987) across your
325
 'VIC 1 1 IVI-'-' V" <...VLLL<...'" ~'-' ,...,., , ..... ,..." ,,"",L' "--"''-' L"n ''''-''-'''''L ",,...,, L' ",-,~..,

Softvva re and Qualitative Research

cases; software that can help you crcate multiple-casc m3trix displays,
Size ofdatabase. A program's d . I .
ata 1asc capacity may b - __ .
usually by gathering together the data that correspond to tbe different of numbers of cases n!lml . f I
< ,
d . . e expre,sed 10 terms
cells of the matrix, is also llSeful. . ' . 1elS a (;]fa ocument (f l~). ' ..
flJes, and/or IOtal datab'1s' f . s I l'S , sIze of. IndIVidual
<, e SIze, 0 ten expresse 1
L k' J b ,-
byres (MB). (Roughly can J'd I ." - C 111 J 0 ytes II\.) or mega-
Fixed records versus reuised. \"'\Till you be working with data that are fixed -, s er Ilat a smgle-spac d I'
about J to 3K ) E ,.. e page a pnnted text is
. ,- . stImate your total size in wha ' . ,
(such as official documents or survey responses) or data thar will be re- 1.lrnlts arc expressed and at leastd iii' 1<. tcver terms the program's
vised (with corrections, added codes, annotations, memos, and so on)? , ' c au e It Jvlost prog j
eraus In terms of toni dat I . 'I . rams tal ay are gen-
Some programs make database revision easy, whereas others are quite , a 1ase SlZe. 11. ew arc st"lI l' <.
the size of indivichLJl texts F . . . I . ' I stl11gy \,\'ler, 11: COmes to
rigid, so that revising can use up a lot of time and energy. Some will let you 01 examp e 111 some P Ii
mellt goes beyond about 10 ' h' .ro~rams w, en a dOCLI-
revise annotations and coding easily, but not the underlying text, and some . pages, I e program WIJ1 I . t l 1.
mro smaller chunks or will . I' '- nSlS on 1rea {l11g it
will let you revise both. Although this has been a constraining issue up to - open It on y 1I1 a "read-only mode" browser..~
now, the trend in new programs and upcoming revisions or existing oncs is Question 4:
toward programs that allow you to edit underlying text easily.
What Kind 01 Analyses Are Yau Planning ta Da'
Stmctured versus open. Arc your data strictly organized (for example,
As mcmioned abovc idcntif'in -1 J
responses to a standard questionnaire or inrervie'vv) or free-form (running ogy realJy won'r cia tb~ t . J- h) g ~le name. of your analysis method 01-
field notes, panicipant observation, and so on)? Highly organized data 1 fie I.. ere tOur cholCe of s Ft J
10\v you expeer to go about analYsi~ Th. d . 0, Ware (':pends on
can usually be more easily, quickly, Jnd powerfully managed in programs sis plan, but a general sense of tl' . I IS oes not mean a deta:Jed an31y-
set up to accommodate them-for example, those with well-defined "rec- . . Ie sty c anel approach yo '
W1llCh In turn will tell YOU the 1' d
C ,

f J. II are expectlI1g,
ords" for each case and "fields" (or variables) with data for each record. \vith the data Fa .' _, II I..1n s ~ t llI1gs yOll will need to be able to do
Structured surveys may benefit from survey-oriented programs like . c. ran exce ent overview of a ran f I
tatlvc data analysis and, d ' . . ge 0,' approac Ies to qua/i-
Sphinx Surveyor EZ-Text, which take advantage of predictable structure ., a ISCUSSIOn of Some of cl d
with them, see Fielding and L (1998' le proce ures cssociated
to provide good data-manipulation tools. Free-form text demands a more 11. ' e e , chap, 2),
you \v111 be coding your dlta'Ot d
flcxible program. Therc are programs that specialize in one or the other code the way rOur metllod I e , ) . I nee a program that will let you
. . 0 ogy reqUires If ' d '
type of elata, and some that work fairly well with either. sis you may need to track tern 1 . ~ au are OIng ~larrative analy-
Or you may be focusing OJl bPolrd'~ or narrative structures 111 cert;Iin ways.
Uniform uerSllS diverse entries. Your data may all come from interviews, or understanding, tI J ~
UJ. mg a weh 01 h)'] t I I
. . Jer ext III (s as a way of
Ie p lenoillena In Your data d d
you may have information of many sorts: archival documents, field obser- served by one or another wa ' of . . , an your nee s may ;Je better
vations, questionnaires, pictures, audiotapes, videotapes (this issue over- Web. Coding is prol . bI' I )b"'L creatmg and representing that hypertext
0

laps "vith single versus multiple sources, above). Some programs handle
.
lIlg, and m , .
101 ) tIt: t:sl-supporred
,I .
h I
approac at t le current writ-
an) researc lers who Use tl 1
diverse data types easily, and others are narrow and stern in their require- best option is to Use a "cod'" 0 ler approac les may find that their
g
ments. If you wili have diverse entries, look for software designed to pie, to mark up the narrati\:;l system for their own purposes-fer exam-
handle multiple sources and types of data, \vith good source tags and good headed "'The Future," below~trUctureof a text.l'vIore on this in the section
linking features in a hypertext mode. The ability to handle "off-line" The subsections below Ia,' h. ,
data-referring you to material not actually loaded into yoor program-is you move be d . , ) lllg out t e parts of QuestlOn 4. sholild help
a plus. Ivlany programs can be tricked into doing this if you are clever You identify t~~l~p~~l.sf~ the njan.le of your methodology. The; shou'Jd help
about it. If you want to be able to code, and then retrieve, audio or video,
o CI IC ana Ync moves )'ou']J d I
perations YOII' '11 d . WI nee to rna (ej the ::pecific
. . WI nee to perform J l' d I ,
look lor programs like AFTER, ATLAS/ti, Code-A-Text, and ]nterClipper, Hlle'lJrc'ta;tions . '. , t le (lll so lI1s1ghts inferences a d
t you \VJ1J need to reco d d I .' , n
\vhich let you treat these media much like text. o record them I I r ,an t Ie manner Jll which you plan
, n oner words the) I Id 1 I '
, ' S lOll Ie p you to get specific: about
326
327
 ... ~, "'v, C" U'ru I.!UQllronve Kesearch

the things you would do if working with paper and to tr:mslate these into
coding, are flexible, invite repea d .
the functions you will need from software. good search and autocoding f t~e ~unlsl' make codIng revision easy, have
ea urcs a ow you to t J '
tween different parts of th, > >h' h . rac.;: connectlons be-
Exploratory verSllS confirmawry. Arc yOLl mainl y planning to poke around e text WIt, ypenxt" d I
your Work :15 yOli go (Sec also t _. c., an can rna (e .1 log of
in your dara to see \-vhat they are like, evolving your ideas inductively? Or fixed or revisable dllr> >,ng I . he) queStIOn of whether yOU! records are
ana YSIS.
do you have some specific hypotheses in mind linked to an existing theory
to test deductively? If the former, it is especially important that you have
Fineness ofanalysis. \X7ill Vour analysis focus 'f d
features of fast and powerful search and retrieval, easy coding and revi- tcXt? Or sentences? Paragraphs~ Pa es;J \\7h ~n s.pec.I lC war s? Or Jines of
sion, along vvith good text and/or graphic display. program permits (or requires, ~r f~rbids) ~ e fIles:' Look to ~,ee.wh3.t r!le
If, on the other hand, yOll have a beginning theory and want to test Can you look at uaryilIg sizes of chunks i }au ~o d~. How flcXl~le IS It?
some specific hypotheses, programs with strong theory-building and -test- form segments \"lith ease;J S n your 3ta. Can you define (ree-
ing features are bener bets. Look for programs that test propositions, or Vour cOdOI)le se I' orne programs make you choose the sizc of
. . ~,. gments Wlcn yo f . .
those that help you develop and extend conceptual networks. you mix and match ell I.. u Irst Import the data, where;:s others let
un \. SIzes as you go.
Coding scheme (inn at start versus euoluillg. Does your study have a fairly
Interest in context of data. \Vhen t1 .
well defined a priori scheme fot codes (categories, key words), perhaps response to yo I Ie program pulls out chunks of text in
ur searc 1 requests how n h d>
theoretically derived, that you will apply to your data? Or will such a do you want to have' D >d>' I mc Slitroun lllg Information
. a vou nee on y the d I j..
scheme evolve as you go, in a grounded theory style, using the "constant you want the precedin a~d. '" war, p uase, or lnl~ ltself? Do
comparative" method (Glaser, 1965; Glaser & Strauss, 1967; Strauss & 'vant to see the entiregf>IIe) fDOlloWlllg lindes/sentences/paragraphs? Do you
Corbin, "199S)? If the latter, it is especially important dut you have casy . a you nee to b bl . .
place in the file and do som I . e a e to Jump nght to that
on-screen coding (rather than being required to code on hard copy, or hav- e war ( on It (e g d d
you want the inform' t" I ..., co e, e It, annotate)? Do
ing to deal with cumbersome on-screen coding procedures) and features a IOn to Je marked v'lIth a " " 1
where it came from (e g I t ' 3 "I . source tag t lC;t tells yOll
supporting easy or automated revision of codes. Hypertext linking capa- Or do > >1 ., n erVIew ~ \VltlJanrce Chang, page 22 line 6)'
you Just want tIe Sal . f .. ,.
bilities are helpful here too. "Automated" coding (in \vhich the program gr aonIS vary 'VI.d eI y on this. lrce 111 ormatIOn WIthout the text irselP
.'.
Pro
-
applies a code according to a rule you set up, such as when a certain phrase
or a combination of other codes exists) can be helpful in either case.
~;~:~ti;!lS for d~sPdl~Yfs, Anal.ysis,goes much better when you can sec orga-
, omprcsse In ormatlol1ln I 1
Multi/Jle uerSl/S single coding. Some programs let you assign several di ffer- of unred d > S one p ace rat ler than in page after page
uce text. ome program d ' . '--
ent codes to the same segment of text, including higher-order codes, and segments hits cod . d s pro Uce Output 111 lIst form (li,ts of text
may lct you overlap or nest coded "chunks" (the ranges of text you apply pia,s Tl' " . es, an so on). Some can help you produce matrix dis-
codes (0). Others are stern: one chunk, one code. Still other programs will tholU~h y~el;\~:~~l~l~: :~~tc::~;;;~::a~~~~~es for each cell of a m a[,ix, aI-
let you apply more than one code to a chunk, but will not "know" dut
there are multiple codes on the chunk-they'll treat it like two chunks,
one for each code.
for programs that let
them into a text-fill r ou
d' em 111 a matrIX for dIsplay. Look
e. ~t, ~educe, or summarize hits before you put
Can give l'Oti quantl>te'tmatlflx With your word processor. Some programs
, a lve (ata (generalll' f > ). .
can give you networh J. . requencles 111 a matnx. Others
Iteratiue verS1!S aile pass. Do you want-and do you have the time-to data d> I '- or lIcrarclucal diagrams, the other maJ'or form of
ISP ay.
keep v\lalking through your data several times, taking different and revised
cuts? Or will you limit yourself to one pass? An iterative intent should Qualitarve ll I
point you toward programs that give you J good display of your previous ses, inel ld I y or :blt.fl!lbers included. If your data, and/or your anJly-
U e ne POSSl dlt}' of nu l, I.
m kt crunc ung, look to see whether the
328
329
 1V1l.' nVL.J..J v ' '-..'Jt...t...t...'-,,, ...... ' , , , - ' .. "
program will count things and/or whether it can share inform~tion with
qtwntitarive analysis programs such as SPSS or SAS, Think carehllly about
what kind of quantitative analysis you'll be doing, and make sure the pro-
uraJ1l vou
b .
are t!linking abour can arrange the data apprnpri:uely. Consider,
roo, \vhether the program can link qualitative and quantitative data in a
meaningful way (in terms of the analytic approach you are raking). For
example, do you need to be able to select subsets of your qualitative data
based on quantitative scores or demographics? Or do you need to use your
qualitative coding to generate scaled variables for statistical analysis in
SPSS? Or do you w8nt to be able to generate word arcade frequency tables
for statistical analysis?
Collaboratioll. If you will be working with a tcam and more than one of
yOll will be working on data analysis, look to see how the program sup-
~orts collaboration. Some are fine if you just want to divide IIp the work
with each of you coding different parts of the data and then combining the
work. Othe;s will support comparing multiple researchers' interpreta-
tions of the same data. Some programs will allow multiple users to access a
shared database over a network; others will allow you to merge periodi-
cally separate copies of the database that different researchers have been
wo~king on. Programs diffet in how much control they givc you over .th,e
merge process. Somc allow you to specify what the program will do If It
finds, say, codes or memos with the same name in each of the copies being
merged, whereas others follow a fixed rule. Some progr3111S are ?ood
at lenin?; you tell which copy a code, memo, or other object came from;
others l~se all identifying information so you have [Q use trid::s like using
different names in each copy (e.g., I might start the names of all codes I
create with my initials, and you start yours with your initials). Some pro-
~rams offer specific features for letting you compare the coding of two dif-
ferent researchers, for example, by showing you J table in which you can
see the coding done by each.
Cut-Across Issues
The two m3in cut-across issues are closeness to the (bta and financial
resources. Let's dispense quickly with the latter question first. Software
varies dramatically in price. The range of prices for the programs \ve r~
viewed ill Weitzman and Miles (1995b) was $0 to $1,644 per user. That IS
330
._.~ ... - - -, ~ ~ ~ ,<UUHLU<lVC ,'\.CJt;;'UILII
~,ti,1! th~ range ...(Look fo~ ~Iiscounrs for educational users and multiple-user
slte. !Jcenses..) In addmon, programs vary a lot in the hardware thci'
reqUlre. to run effic!ently. You obviollsly cannot use a program if it is to~
expenSlVC for you, If it reLluires a machine you cannot afford, or if it funs
on the wrong platform-say, PC instead of IVlac. Happily, the U.S. Centers
for Disease Control now distributes two programs free via the Web: EZ~
T~xt for qt181itative survcys and AnS\VR for unstructured text. Also hJp-
pIly, reports from the field are that the 1Vlacintosh computers being sold
today (th~ G3 is t~day,'s top !vbc) run even the most powerful new PC pro-
grams sansfactonly WIth PC "emulation" soft\vare. This will, presumably,
continue to be the case with future generations of ivfacs. .
The remaining issuc, closeness to the data, is more complex, and I also
address it below in the section headed "Deb8tes in the Field." For choice
purposes, remember to think about \vhat kind of closeness to the data is
important to you. iVfJny researchers fear that working \:I/ith qualitative
data on a computer will have the effect of "distancing" them from their
data. This can in fJct be the case. You may wind up looking at only small
chunks of text at a time, or maybe even just line-number refere;1Ces to
\vhere the text is. This is a far cry from the feeling of deep immersion in the
datJ that comes from reading and flipping through piles of paper.
. But other programs minimize this effect. They typically keep your data
~lles onscreen in front of you at all times; show you search results by scroll-
109 to the hit, so that you see it in its full context; and allow you to ~xeCLlte
most or all actions from the same data-viewing screen. Programs thJt
aJJo~' you to b~ild in hypertext links between different points in your data,
P~ovldc good facilities for keeping track of where yOll are in the database,
cbspl ay YOllr coding and memoing, and allow you to pull together related
data quickly can in some ways help you get even closer to the data than you
Can with paper transcripts. If you choose with this consideration in mind
software can help, fJther than hinder, your work at staying ~lose to th~
data. It can, in fact, help keep you from drowning in those piles of paper.
However, having software that enhances the sense of closeness to the
data. . Ilssue f or everyone. Some researchers do not mind
may not bC J crucla
rehe 11 . d .
) Illg 1eaVl yon prmte transcripts to get a feeling of closeness, \Vhere8s
Others think such heavy reliance defeats the purpose of QDA sofhvare.
FUrthc rmore, some prOjects .. 1
SImp y do not require intense closeness CO the
data. You may be doing more abstract work, and in fact may wallt to move
aWay from the raw data. .
331
 , , , , - , I , ' - ' ........ ..." _ ~ ~
'" Debates in the Field
A number of debates have taken place over the past two decades in the
qUGlitative research comnluniry about whether the use of so[tw,ue is ,1
good idea and, if so, what kinds of software are a good idea. I will address
four of the debated issues here: closeness to the clata, whether software
drives methodology, whether new researchers should start off doing analy-
sis by hand, and whether software really affects rigor, consistency, and
thoroughness,'>
Closeness to the Data
The issue of closeness to the data, which I have just discussed in terms of
program choice, has been one of the big concerns raised by qualitative
researchers over the years. Experienced researchers have often found that
as difficult as it was, the process of spending endless hours sitting on the
floor surrounded by piles and piles of paper led them, by necessity, to a
very rich and thorough familiarity with their data. But as I have tried to
arg~e above, softwa;e need nor cut down on this familiarity. Software
neither makes it better nor worse, it simply changes it. Although some pro-
grams still create the sense that you are staring at just a small window of
text with no sense of what lies around it, there are now many programs
available that provide rich contextual information (such as source infor~
mation, graphical maps of hypertext links, navigable outlines, and linked
lists of codes, documents, and text segments), and may in fact help you get
to know your data better than ever before.
Does Software Drive Methodology'
Another concern has been that researchers might wind up adapting
their research to the software tbey use, rather than the other way around
(Coffey et aI., 1996; Kelle, 1997; Lonkila, 1995)-that is, that the soft-
ware will impose a methodological or conceptual approach. In fact, soft-
ware developers bring assumptions, conceptual framevvorks, and sonlC-
times even methodological and theoretical ideologies to the development
of their products. These have important implications for the impact that
using a particular program will have on your analyses. Ho\~tever, as I have
argued else\vhere, you need not, and in fact should not, be trapped by
.... (. b r 8-
these assumptions and frameworks; there arc often ways o~ enc InC'
(T
332
... ,-" '''UIl:' UIIU \dUOllfat/ve t<esearC/l
program to your own purposes (sce \\1eitzman, in press). For example., a
program may allow you to define only hierarchical relations among cocles.
You might work around this by creating redundant codes in different port,
of the hierarchy, or by keeping track of the extra relationships )'OU want to
define with memos and network diagrams.
The fact that developers bring conceptual assumptions to their work is
in fact one of the strengths of the field. Many of the developers, particu-
larly of code-and-retrieve and code-based tbeory builder programs, arc
~es~arc.hers themselves. They have invested enormous intellectual energy
ll1 fmdmg the right tools for analyses of different types, and the user Gill
benefit greatly from their investment.
It is also true that the design of the soft-ware can have an impact on anal-
ysis. For example, different programs work with different "metaphors"-
that is, different ways of presenting the relationships among codes, and
between codes and text. Kwalitan, NUD" 1ST, the Ethnograph, and win-
l'vl.AX all allow hierarchical relations among codes. For studies in which
you arc organizing your conceptual categories hierarchically, these pro-
grams offer significant strengths. If you want to represent nonhierarchical
relationships, even if you choose to try to work around this metaphor, it
may be less comfortable than using a program like ATLAS/ti that explicitly
supports more flexible networks. HyperRESEARCH emphasizes the rela-
tionship between codes and cases, rather than codes and chunks of text.
:X 1
hen you code a chunk of text, you create an entry on what looks like an
Index card for a particular case. This strongly supports and encourages
thinking that stresses casewise and cross-case phenomena, but makes it
harder to look for and think about relationships among codes within a text
(although version 2, under development at the time of this writing, ap~
pears to be solving this problem). Finally, Code-A-Tex-r offers a quite dif-
fer~nt set of coding metaphors: (a) codes arranged into scales" (you can
assign only onc code from each scale to a segment of text, useful if you
wa nr.to code your text chunks by making mutually exclusive judgments on
~ vanety of factors), (b) codes automatically assigned according to "words
In the text, and (c) open-ended "interpretations" you write about each text
chunk. \"X!orking \vith this collection of coding metaphors could be ex-
P~cted to lead you to consider your text in somewhat different ways than
WIth one of the other metaphors described above.
1: Similar issues exist in rhe choice of other types of software, such as
~~tbasc managers. InfoTree32 XT allows YOU to arrange texts in a hierar-
c 11cal tree and lets you drag texts around ~o rearrange them. Folio Views
333

also al1O\vs yOll to create <-1 hierarchical outline, but gives you the addi-
tional capability of creating multiple, nonsequentiall.'groupings" of texts
that you can .1ctivare any time you wish. Folio Views and askSam let you
whereas InfoTree32 XT requires that you use a standard field set (of your
O\vn design) throughout the whole database. Idealist not only lets you cus-
tomize the ficld set for each record, you can also create a variety of record
types, each with its own set of fields, and mix them in the same database.
Fina!!y, whereas most of these programs show you just one record at a
time, Folio Vinvs shows you a word processor-like view in which records
appear one right after the othcr as paragraphs. Clearly, each of these pro-
grams shows you a quite different view of your data, and so each may
encourage different ways of thinking about your data. The different pro-
grams all have different strengths and weaknesses. It is also true that a
clever user will be able to bend each of these flexible packages to a wide
varicty of different tasks, overcoming many of the differences between
them.
Each of these assumptions is both a benefit for some modes of analysis
<md a constraint. The key, then l is not to get trapped by the assumptions of
the program. If you arc aware of what they are, you call be clever and work
around them. The program should serve )lOllr analytic needs, goals, and
assumptions, not the ocher way around. Researchers interested in empiri-
cal work on the impacts of different programs on research are again en-
couraged to refer to Fielding and Lee (1998).
Should New Researchers
Start Off Doing Analysis by Hand?
There is no clear-cut answer to this question. Certainly, it is important
that new researchers begin by learning about how to do good analysis,
ratber than just how to use a program. Whether that means doing a first
project by hand or learning about analysis and software to do it with in the
same course is a question best left to teachers. I have taught both ways, and
in my experience students benefit from having some experience with man-
ual methods, if only a few coding exercises, so that they can get the feel of
. . v hat is happening analytically before they stan worrying about using the
software.
334
- - " " " " ' - UI,U ,<"UUlIlUlfl'to' r;;to's(;;'urcn
Does Software Really Affect
Rigor? Consistency? Thoroughness?
Some r,-,sc3rcll<.;r~ a.rc Ul.::dicattd to the notion that software mnkes for
more rigorous research. There are even rumors floating around of federal
funding agencies requiring the use of software in grant proposals. Yet, as I
have argued above, software will not pull good work out of a poor re-
sear.cher. On the other hand, for all the reasons outlined above in rhe sub-
SeCtlOn headed "Real Hopes, 'l software can in fact help competent re-
searchers do more rigorous consistent, and thorough analysis -than they
l
othenvise might. The issue should be conceptualized not as whether th~
software makes the work more rigorous, but whether the researcher uses
the software to do more rigorous \vork than he or she could without it.
" The Future
~t is my hope that the future will see a continuation of current trencls both l
Inscholarship and in software development. Some of my specific hopes are
outlined below.
Needs for Scholarship on the Topic
Ollgoillg review wor/?. In addition to books like \\leitzman and IVliles
(~99~b) and its upcoming revision (which I am coauthoring with Nigel
FIelding and Ray Lce), which offer comprehensive comparative reviews of
the range of software available at a particular time, there is a need for regu-
:arly appearing reviews of new and revised programs as they appear. The
lournal Field klethods (formerly Cllltllral Anthropology Methods) offers
:eg ular software reviev,ls (of quantitative as well as qualitative programs)
~n the same way that many journals feature regular book reviews. l\tlore
Journals that serve qualitative research audiences should follow this lead.
Debate all methodological questions. The kind of controversial issues
~ddressed in this chapter need to be subjected to continued debate in the
~Iterature and among researchers. We need to be both wary of unintended
Influences of softwarc and actively participating in sl1a~ing the future
developm ent of so f tware b y argUl11g
ab . (constructively) with developers
Our what we need and what \ve do not like.
335

IVlore e111phical war/(. The kind of empirical work on the impact of soft-
ware on analysis that has been pioneered by Fielding and Lee (1998),
Weaver and Arkinson (1995), Horney and Healey (1991), and Walker
(1993) needs to be continued. Opinions about rhe impacr of ."io(rware are
nice, but we also need to continue to subject Ollf hyporheses to empirical
research.
Heeds for Software Development
\Ve can at this point identify some of the needs of researchers that arc
not vet met. For example, the field is still lagging in its support for case-
orie~tcd \\iorlc A fevv programs have features built in for explicitly track-
ing individual cases through multiple documents, but few programs are set
up with a strong case-oriented structure.
Display building, especia!1y of matrices, still needs much development.
A product newly released at the time of this writing, NViyo (from the
developers of NUD <> 1ST), allows you to build ell1 interactive matrix in
which you can dick on cells to call up the corresponding text. Ivlatt IvIiles's
dream of a program that would combine this sort of functionality with the
ability" to actuallv" compose the summary text for the output matrix (rather
than switching to a word processor) is still one step away.
Tools for narrative and discourse analysis arc still lagging as well.
Researchers using these approaches continue to call for features that let
them flexibly describe the structure of text and discourse, and longitudinal
researchers do not yet have much in the way of tools built explicitly for
tracking cases over time, though NVivo has an "attributes" feature that
allows you to attach date values to codes or documents. G In each of these
cases, researchers can either adapt coding systems to their needs or look
for yet other kinds of software (such as hypertext authoring programs,
project schedulers, and so on) that they can adapt to their needs.
Finally, because no one program will ever do it all best, researchers need
developers to create the possibility of importing and exporting marked-
up, coded, annotated data from one program to another. At this writing,
there is just a little of this beginning to happen. The developers of Code-A-
Text, the Ethnograph, and winlvI.A.,.X have agreed to work on a common
structure, partially realized at this point. And ATLAS/ti has become the
first program to support export of fully developed projects in X1tlL, a neW
markup language rhar may succeed HTML, rhe World Wide Web fm"mar-
ting standard. A COlllmon standard like this, if adopted by other develop-
336
_.. - "",-_.'--"'~ ., ....... ~ ... , .... "
ers, would allow researchers [0 move fu1!y developed projects easl1y from
one program ro anorher, JUSt as we can now move tabular data among mul-
tiple spreadsheet and dat8base programs. '-
~ Conclusion
Unlike the situation JUSt a decade or so ago, qlJ;lJitative researchers now
have available to them an 8rray of very good software tools to assist in their
research, and the use of software-incl~lding,bur not limited to, word pro-
cessors-seems more and more to be a regular parr of the qualitative re-
search process. There is stil1 no one "best" program, not even for;l p8rticu-
Jar methodology, and that's good. It means that researchers have to think
through their methods and choose programs that fit, which should keep
them from becoming reliant on the software to lead them. As researchers
continue to hullt: around for programs that will do the things they want,
and do them better, SOftw3re developers will likely continue to respond by
making their programs more and more useful. .
\Vhat else can we hope will come out of this collaboration between
users and developers in the ncar fmure? Ivlore and better rools for sharing
analyses and [a\v clara, perhaps by allowing posting of project databases,
with analytic markups, links, and memos, to the \Vorld \\lide \\leb, as
ATLASjti allows, or on CDs; roo Is for building complex reports that
include analyses and data right in the report itself; :md more and bet-
ter tools for supporring coJJaboration among research teams, and for in~
voIving informants in the research process without intensive computer
training.
<1> Notes
1. 1discuss some exceptions in the subsection below headed "false Hopes and Fears."
2. The first release ofNVivo lets you draw diagr3ms, but any connections you draw are
represented only in rhe diagram, they arc l10t representations of the defined rcbtiomhips
~mon.g codes and other objects, as in ATLAS/ti. You scc (he 3crnal rcbtionships among codes
1:1a 111er~lrchical "explorer" with expanclahle and colbpsible branches, as in NUD ~ rST, the
Ethnograph, and winiVIA..X.
3. This section does not cOnl3in much in [he way of references to specific 50fn\'3re, both
~('callse the lanJsc3pe changes every few ye3rs and because a single chaprer docs not 311aw
Or reSPonsible comparisons ~mlOng programs.
337
 4. Foranywarnillg likc this, check at the time you arc choosing to sce if the progral1111Il'
der considerarion presenrs this problem. This type of problem is worked at so regllbrly by
developers thar it would be unfair and unhelpful for me to name particular programs.
Things change.
5. Thc reader interesteu in pursuin b thest.' <'luestions further is rderred to Fieldins "md
Lee (1998) for reports of users' experiences of these and orher issues when using different
programs.
6. You can, in fact, ;mach not only date values, but text or numericll values as well.
'' References
Bogdan, R. c., & Biklcn, S. K. (1982). QIlalitatillc research foredlfcotioll: All intro-
duction to thear"v and methods. Boston: Allyn & B<J.con.
Coffey, A., 1..": Atkinso~, P. (1996). lvIahillg sense of qualitativc data: COIIIIJleIt/CIl-
tary research strategies. Thollsand Oaks, CA: Sage.
Coffey, A., Holbrook, B., & Atkinson, P. (1996). Qualiwtiye data analysis: Technol-
ogies and representations. Sociological Rcsmrch Ouline, 1(1). Available
Internet: lmp://www. socresonline.org.ulJsocn::sonline/1/1/4.hlml
Drass, K. A. (1980). The analysis of qualitative data: A complHer program. Urball
Life, 9, 322-353.
Fielding, N. G., &-~ Lee, R. M. (Eds.). (1991). Using computers ill qualitatiue re-
search. London: Sage.
Fielding, N. G., & Lee, R. M. (1998). Comptlter analysis alld qrwlitatiuc research.
London: Sage.
Glaset, B. G. (1965). The constant comparative method of qualitative analysis.
Social Problems, 12,436-445.
Glaser, B. G., & Strauss, A. L. (1967). The discouery ofgrounded theory: Strategies
for qltalitative research. Chicago: Aldine.
Goetz,J. P., & LeCompte, Iv1. D. (1984). Ethnography and qualitative design ill edu-
cational research. New York: Academic Press.
Horney, M. A., & Healey, D. (1991, April). Hypertext dlld database tools forqua/i-
tative research. Paper presented at the annual meeting of the American Edu-
cational Research Association, Chicago.
Kelle, U. (Ed.). (1995). Computer-aided qualitative data allal-ysis: Theory, methods
and practice. London: Sage.
Ktlle, U. (1997). Theory building in qualitative research and computer programs
for the management of textual data. Sociological Research Online, 2(2).
Available Internet: http://www.socresonline.org.uk/socresonlinc/2/2/1.html
Lee, R.lv1., & Fielding, N. G. (1991). Computing for qualitative research: Options,
problems and potential. In N. G. Fielding & R. Iv1. Lee (Eds.), UsingcomJ!ltf-
ei'S ill qualitatiue research (pp. 1-13). London: Sage.
338
Lofland, j., & Lofland, L H. (1984). Analyzing social settillgs: A guide to qualita-
fiue obseruatioll alld analysis (2nd ed.). Belmont, CA: Wadsworrh.
Lonki!a, IVI. (1995). Grounded theory as an emerging paradigm for computcr-
assisted {ll1~lit:.1tive dat;J. an;)l y!':;.,;. Tn II I,(c:.>lJe (rod.), Cump:ae'-"id"d fpfHffw-
tiue data an<J!ysis: l1Jcory, methods alief practice (pp. 41-51). London: Sage.
Mangabeira, \Y/. (Ed.). (1996). Qualitative sociology and computer programs:
Advem and diffusion ofCAQDAS [Special issue]. Currellt Sociology, 44(1).
Miles, j"I. B., & Huberman, A. Ivr. (1984). Qualitative data altalysis: A s~ltrceboo/~
of new methods. Beverly Hills, CA: Sage.
IVli1es, M. B., & Huberman, A.1'v1. (1994). Qualitative data allalysis: An e.Yfh1l1ded
sourcebook (2nd ed.). Thollsand Oaks, CA: Sage.
lbgin, C. C. (1987). The comparative method: MOiling beyond qualitatilJe and
qualftitatiuc strategies. Berkeley: University of Ca!iforni3 Press.
Seidel,.J. \~, & Clark,]. A. (1984). The Ethnogr~ph; A computer progr;:J.!TI for rhe
analysis of qualitative d<lta. Qualitative SOciology, 7, 110-125.
Shelly, A., & Sibert, E. (1985). The QUALOG users' IIwlllwl. Syracuse, NY: Syra~
cuse Universiry, Schoo! of Computer and Information Science.
Strauss, A. L., & Corbin, J. (1998). Basics ofqllalitatilJe research: Techniques and
procedures for del/cloping grou/lded theory (2nd ed.). Thousand Oaks, CA:
Sage.
Tesch, R. (1990). Qualitative research: Alfalysis tYIJes alld software tools. New
York: Falmer.
Tesch, R. (1991). Computers and qualitarive data II. Qualitative Sociology, 14(3).
\Valker, B. L. (1993). Com purer analysis of qU31itative data: A comparis~n of three
packages. Qlfalitatiuc Health Research, 3(1), 91-111.
Weaver, A., & Atkinson, P. (1995). From coding to hypertext: Strategies fat
microcompuring and qualitative data analysis. In R. G. Burgess (Ed.), Studies
in qlfalitatiuc methodology. Greenwich, CT: JAL
Weitzman, E. A. (1999). Analyzing qU;:J.liratiYe data \~'ith computer software. Health
Seruices Research, 34(5), 1241-1263.
Weitzn lan, E. A., oC
o '1'1 'I B (1995a). Choosing software for qualitative data
lV I es, n
l
" 3nalysis: An overview. ClIltural Allthropology Methods, 7(1), 1-5.
Weitzman, E. A., & Ivliles, M. B. (1995b). Computer programs (or qualitatiue data
a/{alysis: A software sourcebook. Thousand Oaks, CA: Sage.
339

Analyzing
Talk and Text
David Silverman
The linguistic character of field d.3r3. is most obvioLls in the
texts and interviews. Even if our 3111115 to search for s~p~o~edl;.
" I' .' ( class gender power) our raw matenal15 mc\ Itab )
terna 1 rca ItlCS c.g.) < " , " . d'
the words written in documents or spoken by inrervlc.w respoll. en.ts.
lYIoreover, although observational clata sho.uld properl~' Il1clude c1esc:~iJ~
[ions of contextual aspects of social interactIOn (V\That Stimson, 1~8~"1
"the sociology of space and place"), mud: of what \ve o~serve 111 Drm;]
and informal settings ,viB inevitably conSIst o~ converS3tlO ns .
Of (ourse if our clata arc transcripts of audIOtapes, then \\:e come foa
to-face with h~w [aIle organizes the world. Although talk 15 ~omct1m~s
l
< _ " . 1 ("n1e"e" talk) it has increasingly become recogmzed as t e
seen as [nvu -, . I I 1 se-
primary medium through which social interaction t~kcs p ace. ~ll0~1 .
holds ~nd in more "public" settings, families a~d f[lends a~sem e ~ ~~::
-I- I converse wah onc another anel 1
activities through ta IIc. At \'','011.., \r.,e . _. (1984)
accounts of our acrivities placed in dossiers and fdes. As Hentage
. f N rm<lll Denzin Jaber Gnbrinrn,
AUTHOR'S NOTE: r <1m most gr<lteflll fo~ the commentS 0 0 , ,
and Yvonna Lincoln on a first draft of tlm chapter.
340
.-., 'l.JlyL.llly IOI/e ana tex[
argues, "The social \..vorlel is a pervasively conversational one in which an
overwhelming proportion of the \'vorld~s business is conducted through
the medium of spoken interaction" (p. 239). Indeed, whar Heritage calls
".-1"" world's businc-::;,," indulic::i 5ucll basic features as telling news, decid-
ing if one should commi[ suicide, and children's learning how to converse
\-vith their mothers (sec Sacks, 1992a; Silverman, 1998b, chap. 1).
Yet a curiously prelinguistic sensibility pervades much social research.
This is most obvious in quantitative researchers' preference for "oper-
ational definitions" that arbitrarily define the meaning of (linguistically
mediated) phenomena. Less apparently, an inattention to participants'
talk-in-interaction is shown by those qualitative researchers \\'ho claim to
have direct access to the external "realities" mentioned earlier. These "re-
alities" are often simply "read off" interview respondents' answers or
transcripts of talk, with little or no reference to whether (and how) they
are made reference to by the speakers. So, for instance, our "knowledge"
that the speaker is a womal1 or that the setting is, say, a medical consulta-
tion is of no analytic relevance unless we can demonstrate the relevance of
these features in the actions of the parries concerned (see Schegloff, 1991).
By contrast, 20th-century thought has resisted such researchers' as-
c~s,~ ~: sumptions that words are simply a transparent medium to "reality." From
e~, Saussure (1974), the S\viss linguist, we learn that signs derive meaning
from their relation to other signs. From Wirrgenstein (1968), the philoso-
pher, we understand that the meaning of a word derives largely from its
use. Consequently, as \Vittgenstein puts it,
when philosophers use a word-"knowledge," "being," "object" (etc.)
_ __
ce ., . -and try ro gr3sp the essence of the tbing, one must first ask oneself: is
the word ever ,1Ctu3l!y used in this \V3Y in the bngw:lge-g3me which is its
original home?
What we do is to bring words back from their met3physlcal to [heir
everyd3Y llse. (pard. 116)
\\7ittgenstein's critique of some philosophers can, of course, be turned
upon those quantitative researchers who arbitrarily construct "opera-
tional definitions" of phenomena without ever studying the "languagc-
game" in which a phenomenon has its everyday home. Ho\vever, what
\Vittgensrein is saying constitutes an equally relevant critique of quali-
341
 IV''-' , ,,-'v.... VI '-V'-'-'-'-' II ' V ' " ' ..... ' " 'r>.'-' L.. , ' - ' .... . , . . . " ' - , , ... , . ,n ..... _" .~~

rative research that claims to discover social "realities" unaddressed by Interviews

participants.
As G3rfinkel (1967) implies, both qU3ntitative researchers' scien- For the qualitative-minded researcherl the open-ended inrerview appar-
0pF'0rn1l1j~r for <;tIl uuthcIlric gaze intu tbe soul of another,
ent!}" offers rl-,e
deeply commonsensical, because both trade off the c3pacity of societal or even for a politically correct dialogue in which researcher and re-
members to "see through" appearances to an underlying reality. In this searched offer mutual understanding and support. The rhetoric of inter-
sense, there is a strong similarity between social researchers who claim to viewing "in deprh" repeatedly hims at such a collection of assumptions.
be able to access some social structure or emotion "behind" their data and, Here we see a srubbornly persistenr romanric impulse in cOlltemporary so-
say, TV sports commentators who tell their audiences what sporrspeople ciology: the elevation of the experiential as rhe authentic-the selfsame
are "feeling." Both parries use (as a tacit resource) what Garfinkel calls the gambit that can make the TV talk-show or news interview so appealing.
"documentary method of interpretation" to produce thei r "findings" (see Such qualitative researchers share survey researchers' assumption that
Gubrium & Holstein, Chapter 7, Volume 2). interview responses index some external reality ("facts" or "events" for
This link between social research and society is hardly surprising. Such the hurer group and "feelings" Or "meanings" for the former).Borh groups
activities as observation and interviewing are not unique to social re- build into their research designs variuus devices to ensure the accuracy of
searchers. For instance, as Foucmlt (1977) has noted, the observation of their interpretarions. So you can rry to ensure thar you have accurately de-
the prisoner has becn at the heart of modern prison reform, and thc mcth- picted such realities and experiences by such measures as imercoder agree-
ad of questioning used in the interview rcproduces lllany of the fearures of ment and computcr-assisrcd qualitative dena programs. And you can check
the Catholic confessional or the psychoanalytic consultation. Its perva- the accuracy of what your respondenrs tell you through other observa-
siveness is reflected by the centrality of the interview srudy in so much con- tions. Let us call this a realist approach to interview data.
temponiry social research. Think l for instance, of how much interviews An alternative approach treats intervinv data as accessing various sto-
are a central (and popular) feature of mass-media products, from "talk ries or narratives through which people describe their worlds (see Hol-
shows" to "celebrity interviews." Perhaps we all live in \vhat might be stein & Gubrium, 1995, 1997). This narrative approach claims that, by
called an "interview society," in which interviews seem central to making abandoning the attempt to treat respondents' accoullts as potentially
sense of our lives (see Atkinson & Silverman, 1997). "true" pictures of "reality," we open up for analysis the culturally rich
This broader societal context may explain qualitative researchers' methods through which intervic\vers and interviewees, in concert, gener-
temptation to gloss their methodology as "empathic understanding" and ate plausible accoums of rhe world (e.g., Gubrium, 1993; Voysey, 1975).
to use methods such as the interview. Of course, such a link between I am aware that many readers of this volume will favor rhe former ap-
culture and method should be an opportunity to question ourselves proach. In this light, I want to give an example of how one realisr interview
abollt our methodological preferences. However, such self-questioning srudy \vas eventually driven in a narrative direction. JodI' lvIiller and Barry
(sometimes-mistakenlYl I think-referred to as reflexivity) does not itself Glassner (1997) describe a study involving in-deprh, open-ended inter-
provide a warrant for the choices we make. I As I argue below, such a war- views with young \vomen (aged 13 to 18) \vho claim affiliation wirh youth
rant depends on our preparedness to describe societal members' actual gangs in their communities (lv1iller, 1996). These intervie\vs folJO\ved the
methods for achieving \vhatever they do achieve. completion of survey interviews administered by lvIiller. Here is how
These large-scale questions need, of course, to be embedded in our data Miller and Glassner (997) describe the purposes of each form of data:
analysis. I now rum to such issues, examining three kinds of linguistically
mediated data: inrerviews, texts, and transcripts. It should be apparent While the smvey inrerview gathers information about a wide range of rop-
that here, as elsevv'here, I am concerned with data analysis rather than the ics, including the individual, her school , friends , famill'. ' neiohborhood d"-
0 , '-
mechanics of data gathering. linquenr involvcmcm, arrest history, sexual history, and vicrimization, in

342 343

addition to information about the gang, the in-depth interview is concerned
exclusively with the roles and activities of young women in yuurh gangs, and
the meanings they describe as emerging from their g;mg affiliation. (p. 105)
let us focus on the data that 1-Ailler obtained in her in~depth interviews.
This is one example:
Describing why she joined her gang, one young women told Miller, "well, I
didn't get any respect' at home. f wanted to get some love and respect from
somebody somewhere clse." (p. 107)
Here is another respondent's explanation of \vhy she joined a gang:
r didn't have 110 family.... I had nothin' else. (p.I07)
Another young woman, whcn asked to speculate on \-vhy young people
join gangs, suggested:
Some of 'em are like me, don't have, don't really have a basic horne or sready
home to go to, you know, and they don't have as mueh love and respect' in
the home so they want to get it elsewhere. And, ;md, like we get, have family
members in gangs or that were in gangs, stuff like thar. (p. 107)
Let us assume that you have gathered thcse data and now want to begin
analysis. Put at its starkest, what are you to do with them?
In line with the realist approach, using programs such as the
Ethnograph or NUD" 1ST (see Weitzman, Chapter 8, rhis volume), you
may start by coding respondents' ans\vers into the different sets of reasons
they give for participation in gangs. From these data, two reasonS seem to
predominate: "push" factors (unsupportive families) and "pull" factors
(supportive gangs).
lvloreover, given the availability of survey data on the same respon-
dents, you are now in a position to correlate each factor with various back-
ground characteristics that they have. This seems to set up your research
in good shape. Not only can you search for the "subjective" meanings of
adolescent gangs, you can relate these meanings to "objectivc" social
structures.
The realist approach thus has a high degree of plausibility to social sci-
entists who theorize the world in terms of the impact of (objective) social
structures upon (subjective) dispositions.lvloreover, the kind of research
344
'--"''''''Y"'-'''':1 JUlio,. VIlli ICAL
outputs that this approach secks to deliver arc precisely those dcmandcd
by "users" in the community, who seek immediate practical payoffs from
social science resc8rch.
C"lli"/j th<.--;r ClITn-"~l.ch <.I mcLl!uUulog,y for Ii::ill:ning," l\!Iiller :lnd
Glassner (1997) ~1re thus centrally concerned with '''seeing the world from
the perspective of our subjects" (Glassner & Loughlin, 1987, p. 37). In
this respect, they share the samc assumptions about the "authenticity" of
"experience" as other realists and, therefore, fail to detect culturally (and
locally) specific elemcnts in several "personal" tales (see Gubrium, 1993;
Voysey, 1975).
Howcver, say we are not entirely satisfied by the apparent plausibil-
ity of realism. How Can thc narrative approach kick-start data analysis?
l'v1illcr and Glassner (1997, pp. 103~104) suggest that one way to begin is
to think about how respondents 8fe using culturally 8vailable resources in
order to construct their stories. They refer to Richardson's (1990) sugges-
tion that "participation in a culturc includes p8fticip3tion in the narratives
of that culture, a general undcrstanding of the stock of meanings and their
rclationships to each other" (p. 24).
How, then, can the dat;} above be read in thcse tcrms? The idea is to see
respondents' ans\vers as Cit/if/raJ stories. Th is means examining the rhetor-
ical force of what inrerviC\vces say as "intervic\vees deploy these narratives
to m8ke their actions explainable and understandable (0 those who other~
\vise may not undcrsrand" (1vliller & Glassner, 1997, p. 107).
In the data alrcady presented, I\'1iller 8nd Glassner note that respon-
dents make their actions understandable in two ways. First, they do not
attempt to challenge public views of gangs as bad. But, second, they do
challenge the notion that thc intcrviewee herself is bad. However, lvliller
and Glassner (1997) notc that not all their respondents glibly recycle con-
ventional cultural stories. As they put it, "Some of thc young women go
farther and describe their gang involvement in ways that directly challenge
prevailing stercotypes about gangs as groups that are inherently bad or
antisocial and about females' rolcs within gangs" (p. 108). Here are some
of the respondents' accounts that they have in mind:
It was really, it was just normal life, the only difference was, is, that we had
meetings. (p. 109)
[We] play cards, smoke bud, play dominoes, play video games. That's
basically all we do is play. You would be surprised. This is a bunch of big
kids. It's a bunch of big olel kids in my set'. (p. 109)
345

In accounts like these, JVfilJer and Gbssner argue that there is an explicit
chaJJenge to what the inrcrvievvees know to be popular beliefs about youth
g;mgs. Instcad of accepting the conventional definition of their behavior
as "devi;Jrlt," the girls attempt to convey the normalcy of their activities.
These n;Jfrarivcs directly challenge stereorypic31 cultural stories of
the gang. Following Richardson, Ivli II er 3nd Glassner (1997) refer to such
accounts as "collective stories" that "resist the cultural narratives about
groups of people and rell altern3tive stories" (Richardson, 1990, p. 25).
IVliller and Glassner's sensitive address of the narrative forms from which
perspectives arise suggests an alternative path for interview analysis (for
a more developed version of the narrative approach, see Gubrium &
Holstein, 1997).
Summary
In light of the discllssion above, I suggest below five questions th3t
interview researchers might ask themselves.
,yrhat StatllS do )lOll attach
to )'Ollr data? I\!.LJI1Y interview studies are used to
elicit respondents' perceptions. How far is it appropriate to think that
people attach single meanings to their experiences? IVby there not be mul-
tiple me3nings of a situation (e.g., living in a community home) or of an
activity (e.g., being 3 male football fan) represemed by what people say to
the researcher, to e3ch other, to caregivers l and so on (Gubrium, 1975/
1997)?
This raises the important methodologic~ll issue of \vhetber interview
responses are to be treated as giving direcr access to "experience" or
as activelv constructed "narratives" involving activities that themselves
demand ~lJ1alysis (Holstein & Gubrium, 1995; Silverman, 1993). Both
positions arc entirely legitimate, bur yOli need to justify and explain the
position you t3lce.
Is l'our analvtic position appropriate to YOllr practical concer1ls? Some
al~bitious a~alytic positions (e.g' l hermencutics, discourse analysis) may
actually cloud the issue if your aim is simply to respond to a given social
problem (e.g., living and coping in a community of elderly people, stu-
dents' views of evaluation and feedback). If so, it might be simpler to
acknowledge that there arc more complex ways of addressing your data
346
Analyzing Talk and Text.
bur to settle on presenting your research as a descriptil'e study based upon
a clear social problem.
Do interuiew data really help ill addressilrg YOlfT research topic?If vou are
interested in, say, what h3ppens in school classrooms, should yOll be using
interviews as your major source of data? Think about exactly \'\dlY you
have settled on an intervicw study. Certainly, it can be relatively quick to
gather interview data, but not as quick as, say, gathering texts and docu-
ments. HO\v far arc you being influenced by the prominence of imervie"ws
in the media (see Atkinson & Silvernwn, 1997)?
In the case of the classroom, couldn't: you observe what people do there
instead of asking them what they think about it? Or gather documents that
routinely arisc in schools, such as pupils' repons, mission statements, and
so on? Of course, you may still want' to do ;]n imerview study. But, what~
ever your mcthoG l you \villneed ro justify it and show you have thought
through the practical and 3nalytic issues involved in your choice.
Are YOH mal?iJlg too~/arge claims about your research? It always helps to
make limited claims about your own research. Grandiose claims about
originality, scope, or applicability to social problems arc all hostages to
fortune. Be careful in how yOll specify the claims of your 3pproach. Show
that yOll understJnd that it constitutes one \vay of "slicing the cake" and
thJt other 3pproaches, llsing other forms of dara, may not be directly
competitive.
Does Y01lr allalysis go beyond a mere list? Idcnri Eying the main clements in
your data according to some theoretical scheme should be only the first
stage of your data analysis. By examining how these elemenrs are linked
together, you can bring out the activc work of both inrerviewer and inter-
viewee and, like them, say something lively and original.
<} Texts
To introduce a separate scction on "texts'" now begins to look a little artifi-
cia1. After all, to treat an interview as a narrative GlIl mean looking for the
S3me textual features as researchers working with pri nted material. Indeed,
the mere act of transcription of an interview turns it into a written text. In
347

this section, I use text as 3 heuristic device to identify dara consisting of
words and images that have become recorded without the intervention of a
researcher (e.g., through an interview).
(1992, p. 459) points out, one of the disadvantages of the coding schemes
used in both interview and text-based analysis is that, because they are
based upon given sets of categories, they furnish "a powerful conceptual
grieP' from which it is difficult to escape. Although this grid is very helpful
for organizing the data analysis, it also deflects attention away from
unc<ltegorized activities.
In part, Atkinson's critiquc vitiates thc claims of many quantitative
researchers in their attcmpts to produce reliable evidence about large sam-
ples of texts. Their favored method is "content analysis," in which the
researchers establish a sct of categories and then count the number of
instances that fall into cach category. The crucial requirement is that the
categories arc sufficiently precise to enable different coders to arrive at the
same results when the same body of material (e.g., newspaper headlines)
is examined (see Berelson, 1952).
The meat of the problem with content analysis (and its relatives) is not
simply Atkinson's point about overlooked categories, but how analysts
usually simply trade off their tacit members' knowledge in coining and
applying whatever categories they do use. For instance, in a lecture given
in the 1960s, Harvey Sacks compared the social psychologist Bales's
(1950) tendency to produce immediate categories of "interaction pro-
cess" 'with the relatively long time taken by experienced physicians to read
the output of electroencephalographs. According to Sacks (1992b), one
should not "categorize ... as it comes out" (p. 28). Indeed, as we shall see
shortly, our ability to categorize quickly is properly treated as a research
topic rather than a research resource.
By contrast, in some qualitative research, small numbers of texts and
documents may be analyzed for a very different purpose. The aim is to
understand the participants' categories and to see how these are used in
concrete activities such as telling stories (Propp, 1968; Sacks, 1974),
assembling files (Cicourel, 1968; Gllbrillm & Buckhaldt, 1982), and de-
scribing "family life" (Gubrium, 1992). The theoretical orientation of
these qualitative researchers makes them more concerned with the pro~
cesses ~hrough which texts depict "reality" than with whether such texts
contain true or false statements. As Atkinson and Coffey (1997) put it:
348
",,,,,.-")'-'-"'y IUII\ UIIU J(;'xt
In paying due <mention to such materials, however, one must be quite clear
about what they cm and cannot be used [or. They are "soci:J! bets," in that
they arc produced, shared and L1sed in socially org::mised ways. They arc
not. however, tr:lnsp:lrent l'epr'~5enLl~ion"of or o::l";7.",;0",,1 rCJ",inc~l Jt:ci-
SiOn~lll<lking processes, or professional diagnoses. They construct partiClllar
kinds of representations with their own conventions. (p. 47)
The implicJtions of this are clear:
We should not lise documentary sources as surrogates for other kinds of
dara. \'Ve Clnnor, for instance, learn rhrough records alone how an
org:Jnizarion Jctuaily operates day~by~day. Equally, we cannot treat re~
cords~howevcr"officiar'-cls firm evidence of what they report .... That
strong reservation does nor mean that we should ignore or downgrade doc-
umentary dara. On the contrary, our recognition of their existence:Js social
f<lets alerts us to the necessity to trear rhem very seriollsly indeed. \X1c have to
approach them [or what they are and whar rhey are used to accomplish.
(p.47)
What does it mean to approach teXTS "for what they are"? Potential
examples are semiotics, ethnographically oriemed narra~ive analysis, and
discourse analysis. In crude terms, semiotics treats texts as systems of signs
on the basis that no meaning every resides in a single term (see Silverman,
1993, pp. 71-80), narrative analysis focuses on how accounts artfully use
local cultural resources (e.g., Atkinson & Coffey, 1997; Denzin, 1990;
Holstein & Gubrium, 1997), and discourse analvsis focuses on how differ-
ent versions of the world are produced through ~he use of interpretive rep-
ertoires, claims to "stakes" in an aCCOunt (see Potter, 1997), and construc-
tions of knowing subjects (Prior, 1997).
Following my interest in ethnomethodology's focus on members'
methods, I will sketch out a less familiar exa~pJe-Sacks's account of
membership categorization analysis (see also Silverman 1998b'Watson
1997). . '"
Membership Cotegorizotion Analysis
Like some contemporary ethnographers concerned with narratives,
Sacks (1992b) believes the issue is not to second-guess societal members,
but to try to work out "how it is that people can produce sets of actions

that provide that others can see such things ... [as] persons doing imi-
maC}: ... persons lying, etc." (p. 119). Given rhelt many categories can be
Llsed to describe the same person or an, Sacks's task \vas "to find our how
rhflv rt11~1"h~i'~1 0;0 "bou!: c1,0o<;;nQ :11110n(': tIle ::l.V::l.;bl,l", 'Ocr" or
c,",l"er: C>1';e<;
for grasping somc cvem" (p. 41). So Sacks does nor mean to imply that
"society" determines which category one chooses. Instead, he wants to
show the active interpretive work involved in rendering any description
and the local implications of choosing any particular Glregory. \\7hether or
not we choose to use Sacks's precise method, he offers an inspiring way to
begin [Q amllyze the productivities of allY texL
As we have already seell, "coding" is not the preserve of research scien-
tists. All of us "code" what we hear and see in the world 3wund us. This is
what Garfinkel (1967) and Sacks (1992a) mean when they say that ~ocietal
members, like social scientists, make the world observable and reporr-
;Jble. Put at irs simplest, this means that researchers must he very careful
how they use categories. POl' instance, Sacks (1992b) quotes from tWO lin-
guists \vho appear to have no problem characterizing particular (invented)
utterances as "simple," "complex," "casual," or "ceremonial." For Sacks
(1992a), such rapid characterizations of data assume "chat ",ve can know
that [such categories are accur"itte] without an analysis of what it is [mem-
bers] are doing" (p. 429).
At this point, the experienced researcher might respond that Sacks has
characterized conventional research as overly naiVe. In particular, most
researchers are ,nvare of the danger of assuming any one-to-one corre-
spondence between their categories and the aspects of "reality" that they
purport to describe. Instead, following \X!cber (1949), many researchers
claim that they are simply using hyporhetical constructs (or "ideal types")
that are t'O be judged only in relation to whether they are useful, not
whether they are "accurate" or "true." However, S;lCks (1992a) was aware
of this argument:
It is (l very conventional way to proceed ill the social sciences to propose that
the machinery YOllllse to analyze some data yOLl have is acceptable if it is not
intendedly the ;1I1alysis of rca 1 phenomen<l. That is, you can h;:\ve machinery
which is a "v,llid hypotheticlI consuuCl," and it can analyze something for
you. (p. 315)
By contrast, tbe "machinery" in \vhich Sacks is interested is not a sec
of "hypothetical constructs." Instead, Sacks's (1992a) ambitious cbim
350
MIIUIYLIIIY IUlI( una text
throughout is "to be dealing with the real world" (p. 316). The "machin-
cry" he sets out, then, is nor to be seen 3S a set of more or less useful catego-
ries, bur as the actual categories and mechanisms that members LISt.
LeI: us w_k", ~l <..:vncn;:tc cll:amplc.
In lWO of j-Iarvey Sacks's lectures, he
refers to a Nell-' Yorl:. Times story about an interview with a U.S. Navy pilot
about the pilot's missions in the Vietnam \"X'ar (see Sacks, 1992b, pp. 205-
222,306-311). Sacks is especially imerested in the story's report of rhe
pilot's reported answer to a question:
How did he feelabour knowing that even with all the care he took in aiming
only at military targets someone was probably being killed by his bombs?
"1 certainly don't like the idea that 1 might be killing anybody," he rc~
plied. "But I don't lose any sleep over it. You have to be illlpersonal in this
business. Over North Vietnam I condition myself to think that f'ma military
man being shot at by another military man like myself." (Sacks, 1992b,
p.205)
Sacks invites us ro see how the pilot's immediate reply ("I certainly
don't like the idea ") shows his commitment to the evaluational scheme
offered by the journalist's question. For instance, if the pilot had instead
said, "\X!hy do you ask?" he would have shown that he did not necessarily
subscribe to the same moral universe as the reporter (and, by implication,
the readers of the article) (Sacks, 1992b, p. 211).
Having accepted this moral schema, the pilot, as Sacks shows, now
builds an answer that helps us to see him in a favorable light. The category
"military man" works to defend his bombing as a category-bound activity
that reminds us this is, after all, what military pilots do. The effect of this is
magnified by the pilot's idenrification of his coparticipant as '"'"another mil-
itary man like myself." In this way, the pilot creates a pair (military man!
military man) with recognizable mutual obligations (bombing/shooting at
the other). In terms of this pair, the other parry cannot properly complain,
or, as Sacks (1992b) puts it, ''"there are no complaints to be offered on their
part about the error of his \vays, except if he happens to violate the norms
that, given the device used, are operative" (p. 206).
Notice also that the pilot suggests, "You have to be impersonal in this
business." Note how the category "this business" sets up the terrain on
which the specific pair of military men will shortly be used. So this account
could be offered by either pair-parr. However, as Sacks (1992b) argues,
the implication is that "this business" is one of many where impersonality
351
is required, for "if it were the case that, that you had to be impersonal in
this business held only for this business, then it might be that doing this
business would be \vrong in the first instance" (p. 206).
1\1oreovec rhe impul'.QI1IBlit:y involved ;" of a "pedal son. S"d.-s points
out that we hear the pilot as saying not that it is unfortunate that he cannot
kill "personally," but rather that being involved in this "business" fielDS
that one must not consider that one is killing persons (p. 209). However,
the pilot is only proposing a pair of military man/military man. In that
sense, he is inviting the North Vietnamese to "pIay the game" in the same
way a child might say to another, "I'll be third base." However, as Sacks
(1992b) notes, in children's baseball, such proposals can be rejected: "If
you say 'I'l! be third base,' unless someone else says 'and I'll be ... ' another
position, and the others say they'll be the other positions, then you're not
that thing. You can'r play" (p. 307). . we have to work to sbO\v how the (theoretically defined) elemc11ts we have
Of course, the North Vietnamese indeed did reject the pilot's proposal.
Instead, they proposed the identification of the pilot as a "criminal" and
defined themselves as "doing police action." As Sacks notes, these compet-
ing definitions had implications that went beyond mere propaganda. For
instance, if the navy pilot were shot down, then the Geneva Conventions
about his subsequent treatment would be properly applied only if he
indeed were a "military man" rather than a "criminal" (p. 307).
Unlike morc formalistic accounts of action (l'vlead, 1934; Parsons,
1937), Sacks's analysis shO\vs us the nitty-gritty mechanisms through
which we construct moral universes "involving appropriate kinds of
action and particular actors with motives, desires, feelings, aspirations and
sense of justice" 0. F. Gubrium, personal communication, January 1997).
Like Garfinkel (1967), Sacks wants to avoid treating people as "cultural
dopes," representing the wodd in ways that some culture demanded.
Instead, Sacks approaches "culture" as an "inference-making machine": a
descriptive apparatus, administered and used in specific co~texts,
Summary
I \vill conclude my discussion of texts with a further summary state-
ment. Here T will not use questions, but take the risk of offering three
pieces of advice that emerge Out of the prcceding discussion (for a devel-
opmcnt of this argument, see Silverman, 2000).
352
Hove a clear aJwlytic approach, Successful textual studies recognize the
value of working with a clearly defined approach. I-laving chosen your
approach (c. g., Foucauldian discourse analysis, Saussurian sernio[ics J
.c;"cl~s\; anc-,l)',,;s of nl<:rnbership clltcsorildutjuna)l [rCdl j( a::! a ""luulllOX"
providing a set of concepts and methods to select your data and to illlllni-
nate your analysis.
Recognize that successful allalysis goes beyond a list. I make no apology for
rcpeating a point that I made above in my discussion of interview studies.
It seems to me that the distinctive contribution qualitarive researchers can
make is in utilizing their theoretical resources in the deep anJlysis of small
bodies of publicly shareable daLl. This means that, unlike many quantita-
tive researchers, we are not satisfied with a simple coding of data. Instead,
identified are assembled or mutually laminated.
Limit )'Olll" data. Like many other qualitari ve approaches~ textual analysis
depends upon very detailed data analysis. To make such analysis effec-
tive, it is imperative that you have a limited body of data \vith 'vvhich to
work. So, although it may be useful initially to explore different kinds
of data (c. g., newspaper reports, scientific textbooks, magazine advice
pages), you should usually do this only to establish the clara ser \vith which
you can most effectively work. Having chosen your clata set, you should
limit your material further by taking only a few texts or parts of texts (c.g.,
headlines).
<> Transcripts
The three types of qualitJtive clata cliscllssed here all end up in the form of
some kind of text. In interviews, researchers usually work with \vritten
transcripts. Similarly, audiotapes of naturally occurring interaction are
usually transcribed prior to (and as part of) the analysis.
The t\\lO main social science traditions that inform the analysis of rran-
scripts of tapes arc conversation analysis (CA) and discourse analysis (DA).
For an introduction to CA, see ten Have (1998); on DA, see Potter and
Wetherell (1987) and Potter (1997)-' In the rest of this chapter, I will deal
\vith (\VO more practical issues: (8) the advantages of working \vith tapes
353
and transcripcs of naturally occurring calk and (b) che elemencs of how to
do analysis of such tapes.
The kind of phenomena I deal with are always transcriptions of acrual oc-
currences in their actual sequence. (Sacks, 1984, p. 25)
In contemporary philosophy, Sacks was attracted to speech-act theory,
which, like him, treats talk as an activity. However, Austin (1962) and
Searle (1969) did not study actual talk but worked with invented examples
and their own intuitions about what it makes sense to say. Sacks (1992b),
on the contrary, notes:
One Cannot invent new sequences of conversation and fee! happy with
them. You may be 3ble to take "a quesrion and answer," but if we have to ex~
rend it very far, then the issue of whether somebody would really say that,
after, S<lY, the fjfth utterance, is one which we could not confidently argue.
One doesn'r h3ve a strong intuition for sequencing in conversation. (p. 5)
The earlier ethnographers had generally relied on recording their
observations through ficld notes. \Vby did Sacks prefer to use an audio re-
corder? Sacks's answer is that we canIlot rely on our recollections of con-
versations. Certainly, depending on our memories, we can usually summa-
rize what different people said. But it is simply impossible to remember (or
even to note at the time) such matters as pauses, overlaps, and inbreaths.
Novv, whether you think these kinds of things are important or not will
depend upon what you can shovv with or without them. Indeed, you may
not even be convinced that conversarion itself is a particularly interesting
topic. But, at least by studying tapes of conversations, you are able to focus
on the "actua] details" of one aspect of social life. As Sacks (1992b) puts it:
My research is abour conversation only in this incidental way, thac \ve can
ger the actual happenings on tape and transcribe them more or less, and
therefore have something to begin with. If you can't deal with the actual de-
t:::!il of actual events then you can't have a science of social life. (p. 26)
Tapes and transcripts also offer more than just "something to begin
with." In the first place, they are a public record, available to the scientific
community, in a \vay that field notes are not. Second, they can be replayed
354
and transcriptions can be improved, and analyses can take off on differ-
ent tacks unlimited by the original transcript. As Sacks (l992b) told his
students:
I started to p1<ly around with tape recorded conversations, for the single vir-
rue thar I could replay them; that I could type them our somewhat, and
study them extendedly, who knew how long it might take.... It wasn't from
any large inrerest in language, or from some theoretical formulation of what
should be studied, but simply by virtue of that; r could get my hands on it,
and I could study ir again and again. And also, consequentially, others could
look::lt wh::lr I h.1d studied, and make of it what they could, if they wanted ro
disagree with me. (p. 622)
A third advantage of detailed transcripts is that, if you want to, yOll can
inspect sequences of utterances without being limited to the extracts cho-
sen by the first researcher. For it is within these sequences, rather than in
single turns of talk, that we make sense of conversation. As S;:.Icks (1992b)
pOints alit:
Having available for any given utterance other utterances around it, is ex~
tremel)' importallt for determining what was said. If you have available only
the snatch of talk rhat you're now transcribing, you're in rough shape for de-
termining what it is. (p. 729)
There remains the potential charge that data based mainly on audio
recordings is incomplete. \,(Ie see Sacks's response to this issue when a stu-
dent asks a question about "le8ving out things like facial expressions" from
his analysis (1992b, p. 26). Sacks at once concedes that "it would be great to
study them [sllch things]. It's an absence." Nonetheless, he constructs a
two~part defense of his data.
First, the idea of "completeness" may itself be an illusion. Surely, there
cannot be totaHy "complete" data any more than there can be a "perfect'"
transcript? Second, Sacks (l992b, pp. 26-27) recognized some of the un-
doubted technical problems involved in camera positioning and the like if
one were to use videos. These are the very issues that have been addressed,
if nor resolved, by more recent work based on video-recorded data (e.g.,
Heath, 1986, 1997; Heath & Luff, 1992). Rather, as always in science,
everything will depend on what you are trying to do and where it seems
that yOll may be able to make progress. As Sacks (1992b) puts it, "One gets
starred where you Can maybe get somewhere" (p. 26).
355

It should nor be assumed that the preparation of transcripts is simply a
technical detail prior to the main business of the analysis. The convenience
of transcripts for presentational purposes is no more rh8Jl an added bonus.
As Atkinson and I-lerltagc Cl984) poinr our, rhe producrion ::lnd w;e of
transcripts are essentially "research activities." They involve close, re-
peated listcnings to recordings that often reveal previousl y ullllotcd recur-
ring features of the organization of talk. Such listeni ngs call most fru irfully
be done in group clara scssions. As described by Paul tcn Have (1998),
\vork in such groups usually begins with the members listening to an
extracr from a rape with a draft transcript and agreeing upon improve-
ments to the transcript. Then
rhe participalHs are invited to proffer some observations on the data, to se-
lect all episode which they find "interesting" {or whatever reason, and for-
Illulate their undcrsrandlng or puzzlcmelH, regarding dut episode. Then
anyone can corne In to re::tCi to these remarks, offering alt'ern,ltives, raising
doubts, or wha[ever. (p. J14)
However, as ten Have makes clear, such group dara sessions should be
rather more th211 anarchic free-for-::dls:
Participants ;:He, on rhe one hand, free [0 bring in anything they like, but, on
[he orher hand, required [0 ground their observations In the dara at hand, al-
though they m;lY also support them with reference to rheir own dar-J-based
findings or those published in the literature. (p. 114)
Analyzing Tapes
As \vith any kind of data, the amdysis of tapes and transcripts depends
upon the generation of some research problem out of a particular theoreti-
cal orientation. Like the writing of field notes, the preparation of a tran-
script from an audio- or videotape is a theoretically saturated activity.
\Vhere there is more than one researcher, debate about what yOll are see-
ing and hearing is never just about coUating daw-it is data analysis.
But hmv do you push the analysis beyond an agreed transcript? The
temptation is to start at line 1 of your transcript and work your way down
the page, making observations as you go. However, the danger of proceed~
ing in this way is that your observations arc likely to be ad hoc and
commonsensical. IVloreover, if you are committed to em approelCh (like CA
356
. "''''''~'''::J ' .."" u,'u 'CAe
TABLE 9.1 How to Do Conversation Analysis
1. Always try to identify sequences of related talk.
2. Tryto examinE' how speaker:; loke on c:cr1oin role5 or identities tllrougll
their talk (e.g., questioner/answerer or client-professional).
3. Look for particular outcomes in the talk (e.g., a request for clarification, a
repair, laughter) and work backward to trace the trajectory through which
a particular outcome was produced.
SOURCE: Silverman (1998a).
or DA) that looks at how tbe participants coproduce some meaning, then
beginning with a single utterance gets you off on the wrong foot.
How else can you proceed? Jennifer Mason (]996) suggests that you
can formulate a research ropic in terms of different kinds of puzzles. Iden-
tifying a puzzle can also be a way to kick-start the analysis of J transcript.
Once yOll have found your puzzle, the best method is often to luorh bach
and forth through your transcript to see how the puzzle arises and is
resolved. This implies a strongly inductive bent to this kind of research. It
follows that any research claims need to be identified in precise analyses of
detailed transcripts. It is therefore necessary to avoid premature theory
construction and the "idealization" of research materials that uses only
general, nondetailed characterizations. Heritage (1984) sums up these
assumptions as follows:
Specifically, analySIS is srrongly "daca-drivcn"-developcd from phenomena
which are in various ways evidenced in the data of interaction. Correspond-
ingly, there is a strong bias against a priori speculation about dle orientations
and motives of speakers and in favour of detailed examination of conversa-
tiona!lsts' actual actions. Thus the empirical conduct of speakers is treated
as the central resource alit of which analysis may develop. (p. 143)
In practice, Heritage adds, this means that it must be demonstrated that
the regularities described can be shown to be produced by the participants
and attended to by them as grounds for their own inferences and actions.
Further, deviant cases, in which such regularities are absent, must be iden-
tified and anal yzed.
However, the way in which CA obtains its results is rather different
from how we mighr intuitively try to analyze talk. It may be helpful,
357

TABLE 9. 7 Common Errors in Conversation Analysis
i. Explaining C1 turn at talk by reference to the speaker's intentions (except
insofar as such intentions ore topicalized in -the conversation\.
2. Explaining a turn ot tolk by reference to a speaker's role or status (e.g., as
doctor or as 0 man or woman).
3. Trying to make sense of 0 single line of transcript or utterance in isolation
from the surrounding tollc
SOURCE: Silverman (1998<1).
therefore, if I conclude this section by offering a crude set of prescriptions
about how to do CA (Table 9.1) and a list of things to avoid in doing CA
(Table 9.2). If we follow these rules, the analysis of conversations does not
require exceptional skills. As Schegloff (1992) puts it in his introduction to
Sacks's collected lectures, all we need to do is "begin with some observa-
tions, then find the problem for which these observations could serve
as ... the solution" (p. xlviii).
This mcans that doing the kind of systematic data analysis that CA
demands is not an impossibly difficult activity. As Harvey Sacks (1992a)
once pointed out, in doing CA wc are only reminding ourselves about
things we already know:
I take it that lots of the results laffer, people can see for themselves. And
they needn't be afraid to. And they needn't figure that the results are wrong
because thev can see them.... [It is] as if we found a new pbnt.It may have
been a plan~ in your garden, bur nOw you see it's different than something
else. And you can look at it to see how it's different, and whether it's differ-
ent in the way rhat somebody has said, (p. 488)
\'X!ittgenstein (although mentioned only nvice in Sacks's lectures), and
his concern for assembling reminders of what we know already, is dearly
relevant. Wittgenstein (1968) writes: "The aspects of things that are most
important for us are hidden because of thcir simplicity and familiarity"
(para. 129). Now Wittgenstein, of course, is referring to what is hidden
from philosophers. But the same issue often arises for social scientists-to
"vhom things can be "hidden because of their simplicity and familiarity."
358
, "'-'/h"':::J ,~". ,_",~ ''-'"
' Conclusion
In a sense, nearly all qualitative research touches upon talk and text. In this
chapter, hcnvcver, I have n:si:m,:::d such inclUSiveness in favor of a much
0
more strictly defined version of appropriate ways of responding to the lin-
guistically mediated character of qualitative data.
Although my own approach derives from Garfinkel's and Sacks's con-
cern \vith members' methods, I have nonetheless tried to avoid adopting a
"take it or leave it" approach. In particular, I have highlighted points of
conract with a wide range of other approaches, including narrative-based
ethnography, discourse analysis, and semiotics.
Above all, r have endeavored to offer pr3ctical advice ro novice
researchers vvho may be considering setting om in this direction. How-
ever, rather than offering a simple cookbook of techniqucs, this chaprer,
I hope, derives from a coherellt set of principles. At the risk of restating
what is <liI'eady obvious, r conclude with a statement of four of those
principles:
(> Qualitative research is best viewed not as a set of freestanding techniques
but as based on some analytically defined perspective.
. From my perspective, the particular strength of qu,llitative research, for both
researchers and practitioners, is its ability to fOCllS on actLlal practice in situ,
looking at how social imer3etions are routinely eJ/acted.
4> The fashionable identification of qualitative method with an analysis of how
people "see things" ignores the importance of how people "do things." This
means thaI' the apparent identification of 110nquantitative social science with
the open-ended interview needs to be reexamined. This docs IlOtmean thar
we should never interview, but that, as a minimum, we should first think
through the alternatives.
'- Qualitarive researchers ollght to question the conventional wisdom that their
kind of research can only be "exploratory" or "anecdotal." Case study meth-
ods can be applied to large data setS and standard issues of "reliability" can, in
part, be addressed by systematic transcription of dara (see Per:ikyHi, 1997;
Sllverman, 1993, pp. 144-170).
~ Notes
1. In Garfinkel's (1967) sense, rc(lexi!!ity refers not to self-ql1esri0ning but to how (011-
rext is constitllted through interaction (see Gubril1m & Holstein, Chapter 7. Volume 2).
359

1 The diHcrence between DA and CA is <l 11l3rter for deb<lte. Some DA resc;lfchers find
eNs refus;l] to cngage Uifccdy with cultural and political comexr disconcerting (see
\'{'etherell, 1998). Equally, CA specialists question the v3lidiry of some DA researchers'
,ppeal, w [heir own .It1m: of cofllm (.\tI: ')[hegloff, J998).
{J;> References
Atkinson, J. M., LV: Heritage, J. C. (Eds.). (1984). Structures ofsocial action. Cam-
bridge: Cambridge University Press.
Atkinson, P. A. (1992). The ethnography of a medical setting: Reading, writing and
rhetoric. Qflalitatiuc Health Rescarch, 2, 451-474.
Atkinson, EA., & Coffey, A. (1997). Analysing documentary realiries. In
D. Siiverman (Ed.), Qualitative research: Theory, method and practice
(pp. 45-61). London: S3ge.
Atkinson, P. 1\., &~ Silverman, D. (1997). Kundera's Immortality: The inrerview
societ')' and the invention of self. Qllalilatiue Inquiry, 3, 304-325.
Austin,.J. L. (1962). HolU to do things l&'ith words. Oxford: Clarendon.
Bales, R. F. (1950). Illteractioll!nocess analysis. Cambridge, Iv1A: Addison-\Vesley.
Berelson, B. (1952). COl/tellt analysis in coml11ll1licative research. New York: Free
Press.
Cicourcl, A. (1968). The social orgallization of juvenile justice. New York: John
Wiley.
Demin, N. K. (1990). Harold and Agnes: A feminist narrative's undoing. Sociologi-
cal Theory, S, 198-216.
Foucault, M. (1977). Disciplil1e and tJlfllish: The birth of the prison (A. Sheridan,
Trans.). New York: Pamheon.
Garfinkel, H. (1967). Studies in etlJllomethodology. Englewood Cliffs, NJ: Prentice
Hall.
Glassner, B., & Loughlin, J. (1987). Dmgs ill adolcscent worlds: Bt/nlotlts to
st"ilight~. New York: Sr.l\1artin's.
Gubrium, ]. F. (l992). QlIt of control: Family therapy and domestic disorder.
Newbury Park, CA: Sage.
Gubrium, J. F. (1993). Speaking of life: Horizons of meaning for nursiug home resi-
dents. New)"ork: Aldine de Gruyter.
Gllbrium, J. F. (1997). Living and dying at Murray A'lal101: Charlottesville: Univer-
siry Press of Virginiil. (Original work published 1975)
Gubrium, J. F., & Buckholdt, D. (1982). Describing care: Image and practice i11
rehabilitation. Cambridge, lVLA.: Oelschlager, Gunn & Hain.
Gubrium, j. E, & Holstein, J. A. (1997). The llew language of qualitative method.
Ne\v York: Oxford Universiry Press.
360
.... ~"~ .. ,~ ,....", .... ,,"" ,""..,<
Heath, C. C. (1986). Body lJIoucmellf and speech in medical illteractiolT. Cam-
bridge: Cambridge University Press.
Heath, C. C. (J 997). Using video: Analysing activities in face to face interaction. In
D. Si]verm;m (Ed.), Qu,ditathc: n:sc..ncll' 1"YJ;:UI)I, lIu:rlJUd and practice
(pp. 183-200). London: Sage.
Heath, C. c., & Luff, P (] 992). Collaboration and comrol: Crisis management and
multimedia technology in London underground line control rooms.fotlmaf
of Computer Supported Cooperatille \V/ork, 1(1-2), 69-94.
Heritage, J C. (1984). Garfinkel and etfmolllethodology. Cambridge: Polity.
Holsrein, J. A., LV: Gubriul1l,.f. F. (1995). The active interview. Thousand Oaks, CA:
Sage.
Holstein, J. A., & Gubrium, J. F. (1997). Active inrerviewing. In D. Silverman (Ed.),
Qllalitl1tille research: TheOI)I, method and practice (pp. 113-129). London:
Sage.
Nbson, J. ("1996). Qualilatiue researching. London: Sage.
IV!ead, G. H. (1934 ).lvIilld, self, alld society: From the stLmdjmi1lt o(a social hehau-
ion'SI (c. \Yo Morris, Ed.). Chicago: University of Chicago Press.
Miller,]. (1996). f'cmale gang i/ll/oll/emCilt ill the ivIidwcst: A two-city com/Jarisoll.
Unpublished docroral dissertation, University of Southern California.
:tvIiller, J., & Glassner, B. (1997). The "inside" and the "outside": Finding realiries
in interviews. In D. Silverman (Ed.), Qualitative research: TheOl}', method and
practice (pp. 99112). London: Sage.
PJrsons, T. (1937). The stmctme of social actiolt. Ne,,'! York: 1v1cGraw-Hill.
PerakyIa, A. (1997). Reliability and validity in research based upon transcripts.
In D. Silverman (Ed.), Qualitatiue research: ThcOl}!, method and practice
(pp.101-210). London: Sage.
Poner,]. (1997). Discourse analysis a::. a way of analysing naturally-occurring rail"
In D. Silverman (Ed.), Qualitatiue research: Thcory, method mid practice
(pp. 144+160). London: Sage.
Potter, J., & Wetherell, IvL (1987). Discourse and social psychology: Beyond alti
tudes and behal/iolf/: London: Sage.
Prior, L. (1997). Following in Foucault's footsteps: Text and context in qualitative
research. In D. Silverman (Ed.), Qualitative research: Thcory, method alld
practice (pp. 63-79). London: Sage.
Propp, \~ 1. (1968). The morphology ofthe folktale (L. A. \V3gner, Ed.; 2nd rev. ed.).
Austin: University of Texas Press.
Richardson, L. (1990). Writillg strategies: Reachillg diverse alldimces. Newbury
Park, CA: Sage.
Sacks, H. (1974). On the analyzability of stories by children. In R. Turner (Ed.),
Etlmomethodology (pr. 216-232). Harmondsworrh: Penguin.
361
Sacks, H. (1984). Notes on methodology. In J. iVr. Atkinson & J. Heritage (Eds.),
Stmctures of social act iOll: StluJies ill COll/'ersa/ iOIl dlwlysis (pp. 21-27). Cam-
bridge: Cambridge University Press.
5ackl, H. (]~nJ). LWlIFes 011 CfIllVersalllll1
Ulacll.wcll.
(G.lefferson. Ed.; Vol. Il. Oxford,
Sacks, H. (1992b). LcctllTi!S Oll COII{'ersatiO/l (G. Jefferson, Ed.; Vol. 1). OxIord:
Blacbvell.
10
Focus Groups in
Saussure, F. de. (1974). COllrse ill geltcrallillgllistics. London: Fontana.
Schegloff, E. A. (1991). Reflections on ralk and social srructure. In D. Boden &
D. Zimmerman (Eds.), TaI!~ (flld socialstmetflre: Studies ill etimolllethadol-
Db')! and COlwcrsiltioll /Iwlysis (pp . .:.14-70). Cambridge: Polity.
Feminist Research
Schegloff, E. A. (1991). Introduction. In H. S:lCks, Lcct/ll"cs 011 colH-'crs{1/ioll
(G.,Jefferson, Ed.; Vol. 1). Oxford: Blackwell.
Schegloff, E. A. (1998). Reply ro \'(!ethcreIL Discourse alld Socie!'',', 9, 413-416.
Searle,]. R. (1%9). Speech acts. Cambridge: CJlTlhridge Univer;ity Press.
Esther Madriz
Silverman, D. (1993). IlIterfJreting i-ju<Tlill1/il.'c dat<1: Strategies for analysing talh.
text alld interactioll. London: Sage.
Sih'erman, D. (1998:1). Analysing conversation. InC. Seale (Ed.), Researchillg soci- !He gusta mas cOl/versar asf, COli Ill! grllpo de lIlu;eres. ...
ety and wIt lire (pp. 261-174). London: S;:lge. CWllldo es(o)' sola COll 1111 clltrevistl1d01; me siento
Silverman, D. (199311). Harl/e:1' Sdchs: Social sciellce lind conversatioll I1Iwlysis. illtimidada, {[sus/ada. ): si elias mc /lall/all pOI' eI te/crollo,
Cambridge: Polity. yo 11Il/lca les call testa S/lS prcglllltas. 2COll/O plledo saber
Silverman, D. (2000). Doillg qlfalita/iuc research: l\ rlTilctical hi1J/{lbooh. London: quiclles 5011 (J que qllierell?
Sage.
Stimson, G. (1986). Place and space in sociological fieldwork. Sociological Reuiew, [I'd rather talk this way, with a group of women.
34, 641~656. When I am alone with an interviewer, I feel inrimidated,
ten I-lave, E (1998). Doing col/llersatioll allLZl}'sis: A practical guide. Thousand scared. And if rhey call me over the telcphone, I ncver
Oaks, CA: Sage. answer thcir questions. How call I ImoIU luhat thc)' really
Voysey, iVL (1975). A COllstallt burdell. London: Routledge. luallt or who the)' are?]
\Vatson, R. (1997). Ethnomethodology and textual analysis. In D. Silverman (Ed.), -f\/lorid FCfIl,indcz r 25yNroId Dominicdn womdn

Qth1litative research: Theory, method ,md practice (pp. 80-98). London: (pd,ticipdnt in d focus group lid in \YJdshington

Sage. He;sht'i, ('-Jew Yo~. Cit'll sprrng 1995)

\'';/eber, M. (1949). The methodology of the social sciellces (E. A. Shils & H. A.
Finch, Eels. & Trans.). New 'y'ork: Free Press.
Wetherell, l\tl. (1998). Positioning anel interpretative repertoires: Conversation
analysis and post-srructuralism in dialogue. Discourse and Society, 9, 387~
412.
Wingenstein, L (1968). Philosophical illl'cst(r:;'ltiolls. Oxford: Basil Blackwell.
+- The two major techniques used by researchers to collect qualitative
data are participant observation and individual interviews. Focus
groups, or group interviews, possess elements of both techniques while
maintaining their own uniqueness as a distinctive research method
(Morgan, 1988). Fundamentally, rhey are "a way of listening ro people

362 363
and learning from them" (lvIorgan, 1998, p. 9). As I\1arfCl Fermlndez's
vmrds suggest, focus groups allmv access to research participants who m8Y
finel one-on-one, face-to-face inreraction "scary" or "intimidating." By
ClTiHin~ JTllllriplt: linu!:; of comm.lmic::d:;on, d"lc srOl-lp Interview offers par-
ticipants such as Ivlarfa a safe environment where they can share ideas,
beliefs, and attitudes in the company of people from the same socioeco-
nomic, ethnic, and gender backgrounds. Some of the studies that have
been conducted on focus groups show that group participants find the
experience more gratifying and stimulating than individual imervic\vs
(Morgan, 1988; Wilkinson, 1998).
This chapter deals specifically with the usc of focus groups from a
feminist/postmodernist framework. The major concerns of feminist/post-
modernist ethnographers are the moral dilemmas present in the process of
intcrvicwing and thc role of the interviewer in this process. These con-
cerns have given particular significance to the voices and feelings of the
participants (Fontana &..: Frey, 1994). ivloreover, by acknowledging the
absence and invisibiliry that have surrounded certain population groups in
the social sciences, and more specifically women of color, feminist/post-
modernist researchers have made the race and ethnicity of the researchers
and the respondents focal points in their studies.
For years, the voices of women of color have been silenced in most re-
search projects. Focus groups may facilitate women of color "writing
culture together" by exposing not only the layers of oppression that have
suppressed these women's expressions, bm the forms of resistance that
they use every day to deal with such oppressions. In this regard, I argue
that focllS groups can be an important element in the advancement of an
agenda of social justice for women, because they can serve to expose and
validate women's everyday experiences of subjugation and their individ-
ual and collective survival and resistance strategies.
The focus group is a collectivistic rather than an individualistic research
method that focuses on the multivocality of participants' attitudes, experi-
ences, and beliefs. In the social sciences, however, individualistic research
methods have been prevalent for several reasons. First, the dominance of
positivistic, quantitative studies, particularly in the United States, has led
to a preference for the individual questionnaire as the favored 8nd more
accepted data gathering method. Second, even among qualitative re-
searchers, the one~to-one, face-to-face interview is the most widely used
research tool. It is only recently that collective testimonies and narra-
tives have gained limited ascendancy among qualitative researchers. Even
364
among feminist scholars, who-par<:ldoxically-emphasize the communal
~ll1d collectivist nature of women's lives, the most important research
method is still the individual interview (Finch, ] 984; Oakley, 1931;
Y\lilkinson, J 998). As oJ. Latina feminist, 1 plan; iocm; groups within the
context of collective testimonies and gtou p resistance narratives. They can
be used by women in general and by women of color in particular to unveil
specific and little-researched aspects of women's daily existences, their
feelings, attitudes, hopes, and dreams.
The singularity of focus groups is that they allow social scientists to ob-
serve the most important sociological process-collective human interac-
tion. Furthermore, they enable researchers to gather large amounts of
information about such interactions in limited periods of time. Curiousl)\
although focus groups have been used extensively in market research, it
has taken some time for qualitative and ethnographic social researchers
to accept them and to get acquainted with the method. Hence tbe exist-
ing information regarding focus groups is not only scarce but unsystem-
atic. As Morgan and Krueger (1993) suggest, "Social science and evalua-
tion research are still at a stage at which most of our knowledge about
focus groups comes from personal experience rather than sysrematic in-
vestigation" (p. 3). During the past 5 yeats, howevet, group interviews
have gained in popularity among a few feminist and postmodernist social
researchers_
Compared with participant observation, focus groups have the disad-
vantage of-sometimes-taking place outside of the settings where social
interaction typically OCcurs. Therefore, the range of behavioral informa-
tion that can be gathered through group interviews is narrower and is-
with some exceptions-limited to verbal communication, body language,
and self-report dat;:]. In addirion, given the necessary presence of a facilita-
tor, it is difficult to discern how "authentic" the social interaction in a
focus group really is. This last limitation, however, is also shared vi/ith par-
ticipant observation, for it has been argued that the presence of the re-
searcher may also alter the behavior of those he or she observes.
Compared with individual interviews, the clear advantage of focus
groups is that they make it possible for researchers to observe rhe inter-
active processes occurring among panicipants. Often these processes
include spolltaneous responses from the members of the group that ease
their involvement and participatioll in the discussion ..Moreover, the
interaction among group participants often decreases the amount of inter-
action between the facilitator and the individual members of the group.
365
 "U.. I I IV'-'.,) VI '-V'-'-'-'-,,,.V,..,,, ...... ,..,,, .,..,,'-, '---" ........... "
This gives more weight to the participants' opinions, decreasing the influ-
ence the researcher has over rhe interview process.
~ History of Focus Groups
Ivlorgan (1998) explains how focus groups developed in three phases:
First, e1uring rhe 19205, social scientists used the technique for a variety of
purposes, one of the most important being the development of survey
questionnaires. Second, between \X1orld \X1ar II and the 1970s, focus groups
were used mainly by market researchers to understand people's wants and
needs. Finally, from the 19805 to the present, focus group interviews have
been used by various professionals to do research on issues dealing with
health, sexual behavior, and other social issues. Indeed, in recent years,
social scientists have begun [0 consider the focus group to be an important
qualitative research technique.
Although some carly field researchers acknowledged using group inter-
views, few made explicit reference to them as a distinctive methodology.
Bronislaw Malinowski, for example, one of the major figures in cultural
anthropology, acknowledged in his diaries conversations among groups of
native Trobriand Islanders. Nevertheless, he did not explicitly describe the
particularities of these group interviews in his repofts (Frey & Fontana,
1993). William Foote Whyte, author of Street Comer Society (1943),
made use of group interviews with gang members in Boston, bur he did not
explicitly acknowledge the use of group interviews as a unique research
technique.
During the early 1940s, Robert: lvlerton and Paul Lazarsfeld introduced
the method of group interviewing into the social sciences. In their pioneer-
ing work, they llsed focused group interviews ro evaluate the reaction of a
group of subjects to wartime radio programs at the Office of Radio Re-
search at Columbia University (lvlerton, 1987; Stewart & Shamdasani,
1990). Since then, focus group interviews have been used consistently,
especially by market researchers, among whom they are one of the favored
qualitative research techniques (lvlorgan, 1988).
For about 40 years after J\tlerron and Lazarsfeld's work, there was little
acceptance of group interviews as a method of social science inquiry. Since
the late 1980s, hO\vever, there has been a renewed interest in promoting
the use of group intervie\vs as a recognized social science research method.
This interest is not limited to sociologists and anthropologists. Indeed, a
366
" " .... n ..... ,n' ... '''n......
diversity of social scientists and other professionals are finding this quali-
tative techniqUE useful. Political scientists, for example, are using focus
groups to assess the public perceptions of political candidates and their
views on specific issues. Stewart and Shamdasani (1990) document hO\v
focus groups were used in the 1980s during President Ronald Reagan's
administration to understand the perception of u.s, and Soviet citizens
regarding relations between the two cOllntries. Similarly, they il1ustrate
their own use of this technique to learn aboLlt consumer practices in the
purchase of new automobiles. In sum, foclls groups have become a valu-
able rool for social researchers and other professiomds, regardless of their
particular fields of inquiry (Frey & Fontana, 1993; l'v'lorgan, 1998).
In the social sciences, group interviews developed as reservations con-
cerning the effectiveness of individual information gathering techniques
grew. Such reservations focused on the influence of the intervicwer on re-
search participams and on the limitations imposed by closed-cnded qucs-
tions.. Traditional interview techniques, which used highly structured
questionnaires, had a major disadvanuge-the interviewer's framework,
viewpoint, and beliefs, consciously or not, influenced the nature of the
questions asked and the choices given to research participants (Krueger,
1994). Less direcred methods were based on the premise that they were
more appropriate to elicit responses that better reflected the social reality
of the interviewee.
As Krueger (1994) argues, the acceptance of qualitative techniques,
especiall y focus group interviews, has been slow due to the positivistic leg-
acy of reliance on numbers as the most accepted way to measure social
reality and due to the emphasis on quantitarive methods, considered more
"scientific" than qualitative methods. Although quantitative methods
have dominated the social sciences, particularly in the United States, many
Contemporary social researchers arc interested in learning more about re-
search participants' opinions, atritudes, and everyday interactions, Femi-
nist and postmodernist researchers, particularly, believe that traditional
methods are alien to population groups who have been traditionally
marginalized. These researchers also consider such methods inappropri-
ate to recover the voices of members of marginalized groups because,
among other reasons, those methods force upon participants an agenda
that it is not their own but the researcher's (lVlaynard & Purvis, 1994).
Even more important, feminist and postJ1lod~rn researchers believe
that traditional research methods do not result in high-quality data
(Wilkinson, (998).
367

Focus GroUPSI Postmodernism, and Feminism
Influenced by feminist ;md postll1odernisr trhnogrJphic srudics) a few
.'iocialre,c~lrcher!,; hs.v(l "redic:covCIC'd" focus groups and arc nu\v increas-
ingly urilizing them in their specific areas of research. This technique is par-
ticulady useful to postmodernist ethnographers, who attempt to remain as
dose as possible to accounts of everyday life while trying to minimize the
distance between themselves and their research participants. It is believed
that the group situation may reduce the influence of the interviewer on the
research subjects by tilting the bJlance of power towelrd the group. Because
focus groups emphasize the collective, rather than the individual, they fos-
ter free expression of ideas, encouraging the members of the group to
speak up (Del1zin, 198(~; Frey & Fontana, 1993).
Feminist scholars, c1isenchanred with the disengJgement and aloofness
of positivistic research and with its inability to explore women's experi-
ences and life siw8tions, h3ve also begun ro .:ldvocate a more integr;ltive,
experiential approach to research. Such an approach regards women's
everyday experiences as an imponanr area of study that necessitates al-
ternative methods of scrutiny. Among these methods, ethnography has
been regarded as "particularly suited to feminist research" (Stacey, 1991,
p. 112). Feminists have advanced many arguments ro justify their prefer-
ence for ethnographic research over traditional survey research methods.
First, they have pointed to the potential for one-to-one interviewing to
reproduce power relationships between the researcher and the partici-
pants (Finch, 1984; Oakley, 1981; \X'ilkinson, 1998). Their argument is
that the researcher usually domin;ltes the whole research process, frOlT! the
selection of the topic to the choice of the method and the questions asked,
to the imposition of her own framework on the research findings. Focus
groups minimize the control the researcher bas during the data gathering
process by decreasing the power of the researcher over rese3rch partici-
pants. The collective nature of the group interview empowers the partici-
pants and validates their voices and experiences.
Second, feminist research attempts to lessen the dichotomy that tradi-
tional research imposes between thought Jnd feeling, bel"\veen the per-
sonal and the political, between the observed and the observer, between
"dispassionate" or "objective" research and "passionate" or "subjective"
knowledge. Giving voice to the subjective experiences of women becomes
the focus of inquiry (Stacey, 1991). Rather than seeing research as a linear,
368
'U,-U-" >.;l,uUjJ-" H' I t'1/IIIU-"t ['I;f.:;',:,,t'U/\..11
one-way process, feminist researchers emphasize the contradictions and
complexities encountered in their work.
Ic should be nored, however, thac feminisc researchers hold several and
conflicting vieWSj they do not talce one unified approach. Olesen (1994)
characterizes three types of feminist research. First is standpoint research,
which emphasizes the need to focus on women's experiences in everyday
life as it is familiar to them. Those experiences are constantly shaped, cre-
ated, and re-created by women. Second is feminist empiricism, which
adheres to "the standards of the current norms of qualitative inquiry"
(p. 163). This also creates new and strict research pr3ctices to make
research findings believable. Third is postmodernist feminism, which
focuses on stories and narratives and on the construction and reproduc-
tion of knowledge. In spite of their differences, however, feminist re-
searchers share the common need of centering and "problematizing"
women's diversity of views and life experiences. Group interviews, I
believe, offer a viable qualitative research method that satisfies the re-
quirements of all three models of feminist research. Group interviews are
particularly suited for uncovering women's daily experience through
collective stories and resistance narratives that are filled with culrural sym-
bols, words, signs, and ideologicl1 representations that reflect the differ-
ent dimensions of power and domination that frame women's quotidian
experiences.
However, in order to enhance the requirements and standards of quali-
tative inquiry (Olesen, 1994), as feminist empiricists demand, \ve must
pay greater attention to the follO\ving questions: How can a social scientist
conduct ethical research that uses subjects not only as providers of infor-
mation but as human agents with potential to exert social change? \X!hat is
the potential of group interviews and of qualitative research projects in
general for fostering such change? Can the fOCllS group be a data gathering
technique as well as parr of a consciousness-raising process through which
human sharing and interaction take place? Is the focus group a valid re-
search method that minimizes the voice and influence of the researcher?
Black feminists (c.g., Collins, 1986; hooks, 1990), Latina fcminists
(e.g., Benmayor, 1991; Garcia, 1989; 1vladriz, 1997), and Asian feminists
(e.g., Chow, 1987; Espiritu, 1997) remind us that women of color experi-
ence a triple subjugation based on class, race, ~111d gender oppression.
Researchers should take such subjugation into account in their selection
of methods for studying women's lives. Not all methods are suitable for
369

interviewing women Jnd much less women of color, who, understandably,
may feel apprehensive about talking with an interviewer abour their lives.
As Collins (1998) argues, the contemporary debate about quantitative ver~
::iU::i qualildllvC researcl1 reneClS lhe polarization that exists benveen what
is male, \1V'hite, and scientific on one side and what is considered soft, sub-
jective, female, and possibly Black, Latino, or Asian on the other.
Some researchers have documented the reluctance of some Latinas and
African American women to participate in survey research and other stud-
ies that use positivistic methodologies. This reluctance is especially strong
among undocumented women, women who do not speak or are not fluent
in English, and \vomen who engage in nontraditional activities or activities
involving the informal or the underground economy (.NIadriz, 1998). As a
response, a few qualitative feminist researchers are advocating the use of
focus groups for interviewing women (Fine, 1994;Jarrett, 1993; lviadriz,
1997, 1998; Wilkinson, 1998). Jarrett (1993) has emphasized rhe impor-
tance of the use of focus groups in research with women of lower socio-
economic status, particularly vmmen of color (see also .Madriz, 1998).
Focus groups not only encourage researchers to listen to the voices of
those who have been subjugated, they also represent a methodology that is
consistent with the particularities and everyday experiences of women of
color. \Vomen have historically used conversation with other women as a
way to deal with their oppression. For African Amcrican, Latina, and
Asian American \'lomen, for examplc, sharing with other women has been
an important way to confront and endure their marginality. Historically,
women have gathered to talk about issues important to them and to get
involved in political activism. Referring to African American women, for
example, Cheryl Townsend Gilkes (1994, p. 235) documents how, after
slavery ended, churclnvomen and teachers gathered to organize through-
out the South. Women's dubs were also important places where women
learned from each other how to challenge male domination in the existing
male-controlled organizations. Similarly, Mexican women have been
instrumental in keeping traditional practices through getting together to
cook, to organize birthday and quiJIceaiios parties, and to talk and keep
alive their rich oral traditions (Dill, 1994). Finally, Yen Le Espiritu (1997,
p. 39) explains how, in 1937, Chinese women employed in San Francisco's
Chinatown organized themselves and struck against the garment factory
owned by the National Dollar Store. All these examples illustrate how
,vomen's' gathering and sharing with other women has the potential to
result in actions and movements for social change.
370
t-ocus Groups in Feminist Research
'I$- Shattering Otherization as a
Way to Promote Social Change
Focus groups are similar to other research methods in that they enable
researchers to have access to the opinions, viewpoints, attitudes, and expe-
riences of individuals. They also provide social scientists access to individ~
llal and collective life stories. However, as a qualitative methodology, focus
groups have a lot to offer social researchers interested in building new para-
digms of social research and promoting social change. Obviollsly, no re-
search method per se can facilitate such change. Iv!y point, however, is that
some methodologies are more suitable than others for shattering J coloniz~
ing discourse in which images of research subjects as the Other are con-
srantly reproduced (Fine, 1994; Madriz, 1998).
Several femi !list researchers who have worked with focus groups have
reported how participants begin to discuss issues of interest to them with-
out even waiting for questions from the moderator. Griffin (1986), for
example, found in her research dealing with young women's experiences
that some of her participants' focus group conversations did not even fall
within the areas she had anticipated asking about. Thus group interviews
heighten the opportunities for participants to decide the direction and
content of the discussion.
As .Michel Fine (1994) remarks, qualitative researchers collaborate in
the construction of the self-other hyphen. In fact, it has been asserted that
the relationship between researcher and participants reproduces colonial
and postcolonial structures. I argue that group interviews minimize some
of the self-other distance in various ways. First, the multivocality of
the participants limits the control of the moderator, who has less power
over a group than over a single individual (Holstein & Gubrium, 1995;
Wilkinson, 1998). Second, the unstructured character of the focus group
interview guide (the instrumenr favored by feminist ethnographers) de-
creases the control of the researcher over the interview process. Struc-
tured questionnaires, \vhich consist of closed- or open-ended questions,
maximize the influence of the researcher over the direction of the con-
versation and introduce the researcher's preconceived notions and opin-
ions and even words and concepts. Indeed, it is possible to have a focus
group without an interview guide, entirely eliminating the research-
er's prejudices from the interaction (Morgan, 1988; \"X'ilkinson, 1998).
Finally, focus groups involve not only "vertical interaction," or interac-
tion between the moderator and the interviewees, but also "horizontal
371
 Mt I HUU:=' UI- LULLtL liNG AI'1U AI'IAlY LIf'll:J l:Ml-'II{ILAL MAl t:1{IAL)
interaction" among the group participants, Although it can be argued that
there is a potential for power rclations co surface among the participants,
these relations] if they arise, arc the p'lrticipants' OWll power relations, in
their Dum constructed hierarchies, Indeed, observing and documenting
the development of these hierarchies may provide the researcher v,Iith
some very important data.
Rather than giving voice to the other, or knm'ving the other, focus
groups open possibilities of listening to the plural voices of Others "as con-
structors and agents of knowledge" (Fine, 1994, p. 75) and as agents of
social change. Jenny Kitzinger (1994) explains hmv the group siruation
cnsures that prccedence is given to the participants' hierarchies of impor-
rancc, their own words and language, and the frame\vorks they use to
describe their own experiences. Language is of particular importance
because a sensitive understanding of people's lives requires shared sym-
bols, meanings, and voc8bularies. Therefore, the pmver that research par-
ticipants hold at the data collection stage of focus groups is not only a mat-
ter of ethics, IVIany researchers have also reponed hm\' the interaction
among participants leads to the gathering of high-qu8lity dara (\Vilkinson,
1998). By asking questions, participants contribute to challenge each
ocher's contradictions and responses. In my mvn use of focus groups v.Thile
doing research on fear of crime, participants confronted each other in this
manner:
M.aria: No, I tell you, de verdad, I am not afraid of crime.
JI/ana: \Vell, you are saying now that you arc not afraid of crime, So,
then \vhy did you say before that you don't like to go Ollt alone at
night?
Rosa: Yes, that's true, You did say that before,
l\tIaria: Okay, I am talking more in general. I try not to be afraid.
In this and many similar instances, participants challenged each other on
responses that seemed to reflect contradictions between whM they s8id
and the ways they behaved \vhen confronted with particu13r situations.
Group interviews are especially significant in that they allow the re-
searcher to \vitness one of the most important processes for the social sci-
ences-social interaction (Berg, 1998). Unlike more traditional research
techniques, such as face-to-face interviews, in group interviews research-
ers observe participants engaging in dialogue, sharing ideas, opinions, and
372
rocus l.:Jroups m remmlst kesearC/l
experiences, and even debating with each other. Indeed, it is this sharing
that creates the socially constructed interactional experiences necessary
for what Denzin (1989) C:1l1s "interpretive interactionism," The plurality
of actors involved in the focus groups makes the process of interviewing
more active and dynamic, facilitating the social construction of meaning
(Holstein & Gubrium, 1995). Therefore, the group interaction becomes a
very important 8specr of the research, contributing to "the development
of shared stocks of knowledge" (Holstein & Gubrium, 1995, p. 71). This
shared stock of knowledge is essential to the process of writing history and
culture together. For women of color, whose experiences-as researchers
as well as subjects of rese8rch-have often been neglected in the social sci~
ences, this writing is especially important bec8use it contributes to tearing
down the walls of silence that have hidden women of color's triple and
overlapping marginality: being female, being of color, and, usually, being
poor.
Obviously, focus groups arc not a solution to the reproduction of the
Other. However, at the data gathering stage, they facilitate researchers'
hearing the plural uo;ces of the participants (Fine, 1994, p. 75). The multi-
vocality of the group situation validates the subjects' experiences with
other subjects of similar socioeconomic, gender, and racial/ethnic back-
grounds. This validation empowers participants, conrribllting to the con-
struction of a research 8genda embedded in the slTuggles for social justice
(\'{1est, 1988). In a culture that highlights individualism and separ8tion,
shifting the research agenda in the direction of commonality and together-
ness is, in itself, subversive.
\Vhen engaged in participant observation, the researcher is able to ob-
serve the unfolding of social processes in their actual social environment.
It could be argued that this does not occur in the environment of the focus
group, because the pbce where the group meets is decided by the re-
searcher. As Frey and Fontana (1993) argue, qualitative research methods
can be placed on a continuum; focus groups use more familiar settings
than other research techniques and less familiar settings than field re-
search or ethnographic studies. Evidently, there is a wide vari8tion in hmv
researchers use focus groups, with some even llsing special facilities, onc-
Way mirrors, and Videotaping in an attempt to create a more unfamiliClr en-
vironment. Among feminist and postmodernist researchers, however, the
tendency is to use the participants' own settings to carry our "in the field"
group interviews, In my work, I have purposefully used participants'
familiar settings, such as their living rooms and kitchens, a senior citizen's
373
 Mt I HUUJ Ur I.ULLtl.lll'll.:! AI'IU AI'iALT LII'Il.:! t:NWIKI\...AL !VIAl t:KIAL.)
dining area, a church basement, a classroom, and even a teenager's bed-
room. Using the participants' familiar spaces further diffuses rhe power of
the researcher, decreasing the possibilities of "Othcrizarion."
<> The Use of Focus Groups as Collective Testimony
Feminist researchers have objected to the use of conventional, positivistic
methodologies in the study of women and, more specifically, women of
color and women of 10\ver socioeconomic sw.tus (Etter-Lewis, ] 991;
Fonow & Cook, 1991; Reinharz, 1992). Some of the most compelling
issues addressed by feminists deal with research imo women's everyday
lives and their particularities, and thus with gaining access into their lives.
I-lence feminists are atrempting to lise and develop research methods
geared toward facilitating forms of communication with women and
among women. Some the most common questions raised by feminist
researchers include these: Doing research for whom and by whom? How is
it conducted? \Vhose words arc privileged (Benmayor1 "J 991)? One of the
principles of feminist research has been expressed in the phrase "research
by, about1 and for women" (Acker, Barry, & Essevelcl, 1983; Harding,
1987). This has resulted in a large amount of literature on women based on
oral histories, in-depth interviews, and testimonies aimed at narrating the
everyday reality of women's lives. Interestingly, very few books dealing
with feminist research methods have included discussions on the use of
focus groups (Wilkinson, 1998).
Since the 19605, the entrance of African American 1 Larina, Asian Amer-
ican, and Native American women into the social sciences in small but
unparalleled numbers has led to mounting criticism of the absence of
women of color from many research projects and of the colonizing nature
of many of those research projects. Third \Vorld feminists h8ve joined
these criticisms of the postcolonial nature of research practices.
As a Latina feminist researcher, I regard the usc of focus groups as 3.
form of collective testimony. "Multivocal conversations" have been used
by women for generations in the form of exchanges with thelr mothers,
sisters, and female neighbors and friends. In a male-centered culture,
some of these conversations have been caricatured as "idle talk" or even
'gossip." However, these dialogues have traditionally been a major wa}'
in which "'lomen have faced their social isolation and their oppression.
Thus testimonies, individual or collective, become a vchicle for capturing
374
lOCUS I.:Jroups In t-eminist P-esearch
tbe socioeconomic, political, and human challenges that women face
(Randall, J980).
The interaction occurring within the group accentuates empathy rind
commonality of experiences Jnd fosters self-disclosure and self-
v8lidation. Communication among women can be an a\vakening expe-
rience and an important element in the consciousness-raising process. It
asserts women's right to validate their own experience, and it allows them
to build on each other's opinions and thoughts (Oakley, 1981). The
awareness that other women experience similar problems or share analo-
gous ideas is important in that it contributes to \vomen's realization that
their opinions are legitimate and valid. This awareness may contribute to
raising consciousness among women that their problems are not JUSt indi-
vidual but structural and that these problems are shared by other \vomen.
This 8wareness is also consistcnt vVlth a social research agenda inserted
into movements for social change.
Latina women, for example, belong to a culture in which llli familia y
mi comflllidad (my family and my community) take precedence (Heyde,
1994; lVlarin & VanOss Marin, 1991; rVIoore, 1970). Because extended
kinship groups are common, sharing is a customary practice, but sharing
with other women is especially important for Latin8s. In many Latina fam-
ilies, women sit around a kitchen table to plan d,e meals, drink coffee, and
share stories and concerns. Hence communication exchange through the
use of focus groups can be considered a practice \vith which many Latina
women are accustomed. Similar claims can be made about African Ameri-
Can women, \vho have faced centuries of oppression and the legacy of slav-
ery. Sharing with other women has been an important way in which Afri-
can American women have dealt with their oppression. Moreover, as
Glenn (1986) notes, Asian American women "kept many customs alive by
cooking ethnic foods, organizing traditional celebrations, practicing folk
hC8ling methods, anel transmitting folklore" (p. 196). They performed
most of these activities while engaging in dialogue and convers8tions \vith
Other women.
Rina Benmayor (1991), a Puerto Rican scholar, is one of the fc\v to
have pointed our tbe transformative experience of collective oral history.
Benmayor asserts that group testimonies empower people, either individ-
uaJJy or collectively: "Social empowerment enables people to speak and
speaking empowers" (p. 159). Focus groups, as a form of collective testi-
mony, can become an empowering experience for women in general, and
for Women of color in particular.
375

Using Focus Groups to Study
Lower-Socioeconomic-Class Women of Color
l\,,-r~<.::~a I I:! I CoW.,s (100 A) ~'rgl-lC5 I.llL'~ I1b.o.:;.l.king slltncc: enables individual
African-American women to reclaim humanity in a system that gains parr
of its strength by objectifying Black women" (p. 47). Similar words can
be applied to Latina, Native American, and Asian .A.merican women, es-
pecially those of lower socioeconomic status. By speaking collectively,
\vomen of color not only reclaim their humanity but, at the same time,
cmpovver themselves by making sense of their experience of vulnerability
and subjugation.
I decided to use foclls groups based on several theoretical and method-
ological considerations. First, because fear of crime has been related to
social vulnerability, I believed it was particularly important for me to gain
access to women who, by virtue of their race and class, occupy socially
devalued positions. Ivlore specifically, for my research goals it was crucial
that I intervic\v Latina and African American 'Nomen of lower socioeco-
nomic background. Second, I was concerned that \vomen of color, who
are usually more afraid of crime than members of other groups, ,vould feel
intimidated bv face-to-face, one-an-one interviews. The fact that several
women were ~resent 8t the same time in a room definitely e3sed some of
these concerns. Third, it was necessary to include undocumented women,
who face unique situations of socioeconomic and political vulnerability. I
believe that because the undocumented women involved in my project
\vere interviewed in the company of other undocumented and docu-
mented women, rather than individually, they were more willing to partic-
ipate in the study and were marc open to emering into the discussion.
Collins (1990) has written about systems of inequality or matrices of
domination. In the case of lower-socioeconomic-status women of color,
the intersection of race, class, and gender puts them in an extremely vul-
nerable position, making it difficult in maIlY cases for researchers to gain
access into their lives. In the case of undocumented Latinas and Asian
American women, who are considered "illegal" or "criminal" by virtue of
their not having legal resident status in the United States, their situation
places them in a particularly defenseless position. Indeed, the multiple and
overlapping vulnerabilities of undocumented, lower-class Latinas and
Asian American women make them one of the most extreme cases of
"Otberization." Aura, a 16-year-old Puerto Rican who participated in onc
of the focus groups for my study and who lives in 3 Latino neighborhood
376
, ULU:, \.;1fUUPS In rem,nJ$! /(esearcl1
in Brooklyn, New York, expressed her feelings of multiple subjugation in a
vivid manner;
Like my modH:r said, you are born with t\VO strikes already: you are Puerto
Rican and you are a woman. So, if snmedling happens to you nobody is
going to help. I think it is worse for us Latinas, because \ve are in the middle
between Black and whire.... so, it's like, we get it from both ways, yo~
know? (quored in Madriz, 1997, p. 54)
In sum, I believed that traditional methodologies were not appropriate
for my study. As a feminist, I wanted to minimize as much as possible
th~ separation between myself and the participants, although my being a
Duddle-class professor contributed to this separation. r believed that the
focus group method would Jimit the imposition of my own ideas and
beliefs onto the women and amplify the power of the participants whiJe
also diminishing my own influence on the conversations. It was also
extremely imp.orrant for me to observe the interaction among partici-
pants and to witness how they were building on each other's words 1 ideas,
and feelings. The following is an example of a conversation on the subject
of carrying some form of protection. The participants are all Latina
teenagers:
CanHen: I usually carry something on me.
Isabel: Like a weapon or something
Gloria: I also carry a weapon with me ... like keys or other sharp things
to protect mysel f.
Carmen: Come onl Tell her the truth.
Gloria: Yes, I carry a knife.
Carmel1: And she carries a small gun. (quoted in Ivladriz, 1997, p. 141)
One important consideration for my study was how I might gain access
to lower-socioeconomic-status women of color. As a Latina researcher I
~e~ieve that impersonal recruiting strategies-often preferred by positi~
IStIc researchers-do not work with lower-socioeconomic-status women
of C010f, and especially with undocumented women. Other social scien-
tists who have done research with women of color have reported similar
experiences (Cannon, Higginbotham, & Leung, 1988; Jarrett, 1993).
Jarrett (1993) considers impersonal recruiting strategies inappropriate for
377
 MI: I HUU:' Uf- LULLtL III'I~ Pd'IU AI'fALl LlI'l\.;l CJVI,Ir\f\"'Ml.. 'VIM' ,-",n .....,
,'-'L. .... " ,-"uup" III rt:'rUffll.<;t r;;eseorcn

research with low-income minority populations. In her study of lo\\'- wage-earner families-women still perform more of the household \vork
income African American VofOmen, she visited Head Start programs in poor (Pesquera, 1993). LHinas fulfill many functions as mothers, caretakers,
neighborhoods throughom Chicago and talked to women directly abollt
cooks, and keepers of rhe hOllse. They often feel thar they should he fully
her study in order to gain their panicip~ltion. _.
available to their spouses and to their children, making it difficult for some
lvly s~l1l"Ces of recruitment were personal networks such as ~flends,
low-socioeconomic-status Latinas to attend activities outside of the house.
students, community leaders, and friends of friends who worked.1n non-
Third) often these women have more unforeseen demands on their time
profit, community organizations. The first contact with the reCrLllter was
and less control over their schedules than do other women (Krinitzky,
usually face-to-face; in a few instcJnces) first contact took place over the
1990; St<lck, 1974). Hence it is my experience that often Latinas fail to
telephone. In cases where 1 did not know the recruiter personally, asked ! arrive at interviews or come late. In my experience, in spite of Over-
mv friends and students to initiate the conuct. They would mentIOn my
recruiting, only a few vvomen will show up our of 12 to 15 who had con-
n~me to the recruiter, indicating that I was a professor doing a study on
firmed their participation. Providing transportation (or participants,
fear of crime and informing him or her that r would call ar a later da~e.
either by picking them up or by sending drivers to bring them to the meet-
They usually informed the person that I was a professor involved iT: sOCIal
ing place, or JUSt offering to reimburse them for transporration costs
justice and social change. As one of my studenrs said, '.'1 ~old the cltrector
improves attendance. But it is even better to meet them at their most con-
of the organization that you are okay." This personalistlC ,-1pproach ~V,-lS
extremely useful given the reluctance many people of color, and pa~t.lcll venient place, such as the soup kitchen where they have dinner or the
church or English-as-a-second-Ianguage school they attend.
larly undocuI1lented women, feel about participating in research (Zl!lT1,
1979). One reason for the high rate of nonattendance among lower-
Although this personal approach works berter than other l11e~h~ds.for socioeconomic-status women is that they may feeluncornfort;lble saying
reaching populations of color, it is still a challenge to gain the particIpatiOn no and turning down an invitation. In the case of Larinas, Latina silllpatfa
of members of some 100".rer socioeconomic groups, such as l~wer-.dass emphasizes cordial and affable relationships. However, it may interfere
Latinas. Some feel apprehensive about panicipating in group cltscUS.SlOIlS, with recruitment because it makes it more socially acceptable to say yes
especially if they are recent immigrants or undocumented wOll1en~ l~ve all and accept an invitation. Thus a commitment l"O participate in a group may
welfare or engage in any non normative behavior (such as alternatIve fan~ be made) but later the promise may not be fulfilled (Marin & VanOss
Marin, 1991, p. 13). -
ily living arrangements), or work in the underground econoll: Y. I r:O~ln
tl~at economic incentives definitely increased the likelihood ot par~lC1p;J~ In addition, I have found that it is not unusual for these women partici-
rion in the project, and I believe participants should be offered such lI1ccn~ pants to encounter emergencies at the last minute thm prevent them from
tives as a reward for giving their time. .. ~ participating in the group-a child gets sick or gets into trouble at school)
Some social scientists claim. that Latlllos
, arc more WI'II'1I1g to part! Clp'Ht:
' a relative needs help with child care or needs to be accompanied to some
in research projects (1\/1ar1l1,
. Perez-Stable, LX.'- Van 0 ss M ann,.
' 1989) . How- aPPoinn nent) or a basic service breaks down 3t home, such as water or
, 1I ' ch:ctricity.
ever, my experience in doin~ fOCllS groups Wit 1. ?w-I.ncom~ L,tnasshows
a.1 .: 'in rc-
othenvise. Several factors 1I1fluence the panlclpanon 01 Lann ....s It Was especially important for my study to have homogeneous groups
search projects. First) many Latinas, especially those of lower socioeco~ of age, class, and race because it was evident that participants felt
nomic status fulfill traditional roles within the family. Indeed, most ~ to express rheir ideas about images and cultllr31 representations of
, responslbtIrry
them carry full . " - tI1e care 0 f t 1le c1'I
tor 11 cIren an d Ihe . house..1 criminals, and victims when discussing them with other women
- , , 1- I . . 111 whJC 1
Second, traditional Latino fatmlres are ~I~rarc 11C3 II1StJtUtIOJ:s, often. their OWn backgrounds. Explaining this need to the recruiter, huw-
men are considered the cabeza de (anu/w (head of the bmtl}). >
a very sensitive issue. On several occasions, 1 was asked gues-
, the fam1-I y /Jar deLh1/U
Latinas exercise their power 111 '(b e l'line1rhe :sceneS.'0- sllch as "\\1hy do you wam a group of all Latinas?" Or "\\lhy only
, tradmona
see Beyck, 1994). Frequently, 111 -, I L
atlno t.lcl1111Tles - e\'en In t\\ "\\fhy can't we mix them up with \X/hite girls?" In most cases,
378 379
 Mi::.l HUUJ Ul- LULLtLI lNl...:J ANU AP1Alr L1J'lU J::NIYIKII....f\L.IVV-'-.! t:1\1..... \-J
after I carefully explained the nature of the topic, the recruiter understood
the need for homogeneous focus groups.
I held 18 focus groups involving between 5 and 12 \Vhite, Black, and
Latina women. I avoided using larger groups because of thc difficulties
they would pose for handling tbe discussion ;:lOd keeptng the conversation
around the topic of research. Moreover, larger groups make it more dir
ficult for all the participants to have their opinions heard. I acted as the
facilitator in all the focus groups, except for one that was led by a student
acting as a research assiswnt. The discussions typically lasted from 90 min-
utes to 2 hams. With the permission of the participants, the discussions
were raped aod later transcribed (Ivladriz, 1997).
One of the major dilemmas in data gdthering concerns the question of
the race/ethnicit}' of the facilitator versus that of the participants. I\.1y
experience with focus groups has been similar to that found in other stud-
ies: A facilitator of the same race/ethnicity as participants usually enhances
rappon and increases the willingness of participants to respond (Jarrett,
1993). A facilitator of the same racial or ethnic background comribures to
participants' feelings that the faedi t~ltor shares with them common experi-
ences. Although I believe this to be true with many other methodologies, it
is especially important in the case of focns groups, where establishing rap-
pon with the participants is key to eliciting high-quality information. Al-
though research on Latinas and the race of the interviewer is scarce, stud-
ies with African Americans show dun the race of the interviewer has an
effect on responses (Jarrett, 1993). For example, a study on race-related
attitudes of African Americans showed that when they were interviewed
by \Vhites, they were more likely to express closeness toward \Vhites than
when they werc intcrviewed by African Americans (Anderson, Silver, &
Abramson, 1988).
For my focus groups, I developed an unstructured interview guide with
some introductory remarks. I first asked a very general question cleating
with the parricipants' opinions about the incidence of crime in the place
where rhey lived. Only after framing the topic did I begin to ask more spe-
cific questions about their fears and images of crime and the cultural repn>
scntarions associated with those fears. Because the topic of fear of crime
was very relevant to the participants, they easily became wilting partiCi-
pants in the discussion. On many occasions the participants moved away
from the interview guide, tapping into areaS of the topic that 1had not pre-
viously considered. This process added a weahh of information to 111y re~
search ::mel gave me new insights into the topic of fear or crime.
380
rULlJ~ l..:!roups In remmlS[ !'"{eseorC/l
One problem with the use of group interviews is that some lower~
j
socioeconomic-status women have been socialized to reserve thdr opin-
ions. Thus some may be less likely to participate in the discussion than
others. Under these circumstances, it can be a challenge to elicit responses
j
from particular women. They may feel uncomfortable in a group, espe~
ciaily l,vhen asked to express thejr opinions in front of strangers. More-
over, some women may feel uneasy disagreeing with others, and that may
j
prevent them from expressing theif opinions and feelings in the group.
\Vhen there are a few timid or reticent participants in a groupl the dynamic
changes and interaction becomes less spontaneous and more directed by
j
the facilitator. This problem lTIay be minimized in groups that arc homog;-
neous not only in terms of race, gender1 and class, but in terms of ethnic
culrures and languages. To include in the same group Latinas of different
j
ethnic backgrounds-such as Caribbean and Mexican women-may be
a mistake, bec:ause there are many differences among different Latina
groups. For cX3mplc. language differences may creare problems of com-
j
munication. Rules of interaction afe also different among various Latina
groups, with some being more comfortable when addressed '.vith the for-
ma! pronoun 1Jsted and others being more comfortable with the use of the
j
informal pronoun fu,
I found it very Llseful to let the groups know that the opinions of all the
parricipants were important for the research project and were valued.
j
Also, 1 emphasized that there could be no "right" or "wrong" answers to
my questions. Because one of the dangers of the group discussion is that
people may feel pressured to agree with others, it is important to let par- j
ticipants know th:n it is acceptable-and in fact desirable-for them to
disagree on issues. Disagreements allow the researcher to underst:md the
range of opinions on a topic. This is especially important in the Case of j
fOCliS groups \vith low-socioeconomic-status women, some of whom hnve
been socialized to agree Olltvl,rardly with others' opinions and to reserve
their ov,m. j
Finally, it is important to note that conducting focus groups with people
who speak Spanish, or Chinese, or any other language different from Eng-
llsh l can result in additional expenses for sllch things as hiring bilingual j
f~Cil!tators~ research assistants, and transcribers. In addition, given the
tllne constraints and multiple demands experienced by lower socioeco-
nomic: women, researchersStlldying this group may find that cancellations j
afe common. Group cancellations that lead to the need for replacement
[oells groups mean that the researcher incurs additional costs. In most
j
381
j
j
 " ".. , ' ,'Jl../ ..... '--', ,-,--,,-'-'-'-""'-''-'''L.' ,.." '''"''-' ... ",'-' '-"" "",-rn_ ",,--. , ,.. ",,....,'--'
social research, where funds are scarce, any unexpected expense can put a
strain on the srudy budget. Researchers must take such likely expenses into
consideration \vhen applying for research grants. Because many social
n;starchers arc, likc lHe, also professors with lirI111Cd tirIlc [or their re-
search, cancellations impose additional hardships on their tight schedules.
<> Focus Groups and the Ethics of
Empowerment for Women of Color
Some scho18.rs have poinred out the role of social theories in the reproduc-
tion of social inequality (Collins, 1998; van Dijk, 1993). According to
Patricia Hill Collins (] 998), elite discourses "represent the interest of those
privileged by hierarchical power relations of race, economic class, gender,
sexuality, and nationality" (pp. 44-45). For centuries, "scientific" theories
jusrifying racial and gender inferiority have been commonplace. For exam-
ple, in my ;:JrC<l of research, criminology, biological theories have related
physical characteristics of Blacks and other non-\"Xlhire groups ro a predis-
position to engage in criminal behavior. Similarly, theories of female crimi-
nality usually present women offenders as inferior beings who emulate
male criminality because of "penis envy/' or else as social misfits who do
nor follow mainstream gender expectations. Therefore, female offenders
are often regarded as "bad girls" (Klein, 1995).
In the same way that elitist social theories have served to perpetuate
inequalities, traditional methodologies have played an important role in
the reproduction of race, class, and gender inequalities. Aside from per-
petuating fallacious and insubstantial information about certain groups,
the knowledge generated by positivistic research tends to reproduce dis-
crimination and prejudice against those groups whose members do not
"perform" according to social expectations or do not conform to the ste-
reotypical ideas of researchers.
The predominant use of quantitative techniques and one-to-one inter-
viewing among social scientists in the United States has resulted in an array
or individualistic research methods. This individualism has even penne-
ated ethnographic studies and feminist research. HO\vever, as Sue
Wilkinson (1998) points Olit, "research methods that isolate individuals
from their social context should clearly be viewed as inappropriate" for
feminist researchers (p. 111). This observation is even more valid in the
case of research with women of color and \vomen of lower socioeconomic
382
rocus l:rroups m reminist P-esearch
status. In the case of many women of color who belong to communitarian
cultures, individualistic research methods place them in artificial, unfamil-
iar, and even "unsafe" environments.
As Rina Benmayor (1~~l) has stated, collective testimonies have the
potential for "impacting direccly on individual and collective empower-
ment" (p. 159). This, I believe, is the potential of focus groups for research
into women's lives and particularly the lives of women of color. The
shared dialogues, stories, and knowledge generated by the group inter-
vin\! have the potential to help such women to develop a sense of identity,
self-validation, bonding, and commonality of experiences. Focus groups
tend to create environments in which participants feel open toreHing their
stories and to giving their testimonies in front of other '''lomen like
themselves.
Although the use of focus groups has plenty of challenges, this method
is particularly \Nell suited to research concerning low-socioeconomic-
status women of color. Focus groups have great potential for uncovering
the complexity of layers that shape women's collecrive, and individual, life
experiences. Therefore, rhe properties inherent in focus groups allow
them to be a culturally sensitive data gathering method. The experience of
women of color in the United States is shaped by the intersection of multi-
ple and overlapping hierarchies-gender, race, and class together with
culture, language, and legal status. Some \-vomen of color, such as Asian
and Latina immigrants in the United States, are trying to adapt to an alien
culture and occupy socially, economically, and politically vulnerable posi-
tions. They are often part of cultures in which women's opinions and con-
cerns are systematically devalued. In this context, I regard focus group
interviewing as an appropriate methodology for women in general, and
particularly for low-socioeconomic-status women of color. Focus groups
offer a way of listening to multivocal conversations on topics relevant to
the understanding of these women's lives, feelings, and thoughts. The col-
lective experience of the focus group empowers participants to take con-
trol of the discussion process, moving the conversation toward areas of the
topic relevant to them, sometimes encouraging and even compelling the
researcher to reconsider her vic\vs on a certain subject. Focus group meth-
ods can contribute to correcting the individualistic bias existing in social
research by offering a unique opportunity to sUld}' individuals in their
social contexts, by generating high-quality interactive dara, by contribut-
ing to the social construction of meaning, and by accessing women's
shared, and often ignored, stocks of knowledge.
383

'" The Future of Focus Groups
Although focus groups represent a sound method of inquiry with much
potential for the study of women of color, their usc in the future depends
on researchers' abilities to understand when and where it becomes appro-
priate to usc them. Some researchers may be lured into using the technique
for many of the reasons developed in rhis chapter: the face that participants
enjoy the opportunity to take part in the discussion, the quality and quan-
tity of information gathered, the possibility of interviewing several people
at once and to witness theif interaction, the empowerment of participants,
and the enthusiasm generated toward the subject matter. However, it is
important for researchers to recognize that in some instances other re-
search techniques, such as individual interviews, may be more appropriate
for the goals of the research. For instance, in a situation where a researcher
needs participants to share very inrimme details abollt their lives, a focus
group would not be the most appropri:Jtc technique. Furthermore,
researchers should avoid the usc of focus groups in cases where partici-
pams may not feel comfortable v,,rith each other. For example, in some
work situations, individuals may not feel comfortable opening up and pre-
senting their views in front of their supervisors or other superiors (l\!lorgan,
1998). Researchers should also avoid bringing together in a group individ-
uals who have strong disagreements or who are hostile toward each other.
Finally, focus groups are not appropriate when the researcher needs to be
able to gcneralize from the rescarch results. When generalizability is a
requirement, the researcher must employ quantitative research techniques
and adequate sampling methods.
Focus groups have become an imporrant technique because they offer a
way for researchers to listen to the plural voices of others. They are espe-
cially important for making audible the voices of oppressed people who
afe demanding to be heard. Many people want politicians, policy makers,
academics, and others in position of pmver to listen to them (Krueger,
1994). Focus groups, although not a solmion to the silencing of the
oppressed, may help to facilitate this listening. For those social scientists
interested in social change, the hope embedded in the use of focus groupS
is that they may contribute to some individuals' recognition and aware-
ness of their own subjugation. Using their own words and their own
framework, this awareness may lead to the participants' involvement as
change agents in the affairs that affect their neighborhoods and their
communities.
384
,uLU" ~IUUjJ:> III rl:!rrurllS[ I\esearcn
'" References
Acker, J., Barry, K., & Esseveld, J. (1983). Objectivity and truth: Problems in doing
feminist n::;carch. \\;;UIlICII'~ Sfudies Illlemariolla{ FOrI/ill, 6, 423-4jS.
Anderson, B. A., Silver, B. D., & Abramson, P. R. (1988). The effects of the race of
thc interviewcr on race-relatcd attitudes of black respondents in SRClCPS
narion;:J! election srudies. Public Opinion Quarterly, 52, 289-324.
Benmayor, R. (1991). Testimon y, action research, and empowerment: Puerto Rican
women and popubr education. In S. B. Gluck & D. Pawi (Eck), \Voll1en's
words: The feminist prr1ctice of amI history (pp. 159-174). New York:
Routledge.
Berg, B. L. CJ 998). Q/talitatiu('. research methods for the social sciences. Boston: Allyn
& Bacon.
Cannon, L. \'v., Higginbotham, E., & Leung, 1\11. L A. (1988). Race and class bias in
qualiwtivc research on women. Gender C;w Society, 2. 449-462.
Chow, E. N. (1987). The development of feminist consciousness ,lmong Asian
American women. Gender G'" Society, 1,284-299.
Collins, P. H. (1986). Learning from the outsider within: The sociological signifi~
cance of black feminist thought. Social Problems, 33, 14-32.
Collins, P. H. (1990). Black feminist thought: Knowledge, consciolfsness, alld the
politics of elll!)(JU'emlCJlt. New York: Rourlcdge, Ch<lpman & Hall.
Collins, r H. (1998). Fighting [llords: Black women and the search (or jllStice. Min-
neapolis: University of l\linnesota Press.
Denzin, N. K. (986). A postmodern social theory. Sociologial Theory, 4, 194-
204.
Denzin, N. K. (1989). Illterprctiue interactionism. Newbury Park, CA: Sage.
Dill, B. T. (1994). Fictive kin, paper sons, and compJdrazgo: \Vomen of color and
the struggle for fJmily survival. In Iv1. B. Zinn & B. T. Dill (Eds.), \\lomen of
color in U.S. society (pp. 149-1(9). Philadelphia: Temple University Press.
Espiritu, Y L. (1997). Asian American womclt and melt: LauOl; latliS, mId loue.
Thousand Oaks, CA: Sage.
Etter-Lewis, G. (1991). Black women's life stories: Reclaiming self in narrative
texts. In S. B. Gluck & D. P,nai (Eds.), \\lomen's /vords: The femil/ist practice
of oml history (pp. 43-58). New York: Routledge.
Finch, j. (1984). "It's great to have someone to talk to": The ethics anel politics of
interviewing women. In C. Bell & H. Roberts (Eds.), Social researching: Poli-
tics, problems, !Jrac{ice (pp. 70-87). London: Routledge & Kegan Paul.
Fine, M. (1994). \Vorking the hyphens: Reinventing self and other in qualitative
research. In N. K. Denzin & Y. S. lincoln (Eds.), Handbook of qllalitr7live
research (pp. 70-82). Thousands Oaks, CA: Sage.
Fonow, J'v1.1vl., & Cook,]. A. (Eels.). (1991). Beyond methodology: Feminist schol-
arship as /iued resct1rch. Bloomington: Indiana University Press.
385
 IV1!: I nvu.;> vr \...VI_l..!:\...11I'1\.:l MI1 .... MI"""l..l t..11"\.;> I:IVlrlf''.I''-..Ml.. IVIM, L["\.I'",LJ
Fontana, A., & Frey,.J. H. (199'i). Interviewing: The an of sciencc.ln N. K. Denzin
& Y. S. L.incoln (Eels.), Ha/ldbooh of qualitative research (pp. 361-376).
Thousands Oaks, CA: Sage.
r."\,, J. J--L. &
f.'nnr"n", A. l1 QQ.J\. TI,e I';roup ;""",'-";''"'-'-';'' 50c;,,1 ,",,,,,,,,rclL T" r:L r __
Ivforg;m (Ed.), Successful (OCIIS grolJps: Advancillg the state of/he art (pp. 20-
34). Ne\vbury Park, CA: Sage.
Garcia, A. M. (1989). Thc development of Chicana feminist discourse 1970-1980.
GCllder&Society, 3, 217-238.
Gilkes, C. T. (1994). "If it wasn't for the women ... "; African Amcrican women,
community work and social change. In M. B. Zinn & B. T. Dill (Eds.),
Womell o( color in U.S. society (pp. 229-246). Philadelphia: Temple Univer-
sity Ptess.
Glenn, E. N. (1986). Issei, /lisei, IINzr bride: Jhree generations o{japanese American
WOniell ill domestic service. Philadelphia: Temple University Press.
Griffin, C. (1986). Qualitative methods and female experience: Young women
from school to the job marker. In S. Wilkinson (Ed.),FclI1inist50cial ps)'chol-
0 6')': Developing theory alld IJractice (pp. 173-19 J ).Miltol1 Keynes, England:
Open University Press.
Harding, S. (1987). Introduction: Is there a feminist method? In S. Harding (Ed.),
Feminism alld methodology: Social science issues (pp. 1-J4). Bloomington:
Indiana University Press.
Heyck, D. L. D. (1994). Barrios and borderlands: Cultures o(LatillOs and L1til/as ill
the U/lited States. New York: Routledge.
Holstein, J. A., & Gubrium, J. F. (J99 5). The active interview. Thousand Oaks, CA:
Sage.
hooks, b. (1990). The poli[ics of radical black subjectivi[y. In b. hooks, 'leamil1g:
Race, gelUlc/; and Cllltllral politics (pp. 15-22). Boston: Somh End.
Jarrett, R. L. (1993). Focus group interviewing with low-income minority popuJa-
[ions: A research experience. In D. L. Morgan (Ed.), Success(ul locus groups:
Advancing the state o{ the art (pp. 184-20:1). Newbury Park, CA: Sage.
Ki[zinger, J. (1994). The methodology of focus groups: The importance of imerac-
60n bcnveen research participams. Sociology ofHealth alld Illness, 16, 103-
121.
Klein, D. (1995). The etiology of female crime: A review of tile literature. In B. R.
Price & N. J. Sokoloff (Eds.), The criminal justice s)'stem alld tuomell:
O{{el/ders, uictims, and lUorhers (pp. 30-53). New Ymk: _McGraw-Hill.
Krinitzky, N. (1990). W!elfarc dependence and psychological distress: A st.udy of
Puerto Rican luomen in Neill Yorh City. Unpublished doctoral dissertation,
New York University.
Krueger, R. A. (1994). !-"OCltS grolljJs: A practical guide to aN/lied resc{?rch. Thou-
sand Oaks, CA: Sage.
386
rULU~ ~ruups In remlnlsr KeSearCI1
IVladriz, E. (1997). Nothing bad hap/Jells to good girls: The impact o( fear o( crime
011 womell's lives. Berkeley: University of C;)lifornia Press.
lViadriz, E. (1998). Using focus groups with lowcr socioeconomic stams Latina
Marin, G., Perez-Stable, E. ]., & VanOss Marin, B. (1989). Cigarette smoking
among San Francisco Hispanics: The role of acculmration and gender. Amer-
iaw joltrllal o{ Public Health, 79, 196+198.
Marin, G., & VanOss Ivlarin, B. (1991). Research with Hispa/lic populations.
Newbury Park, CA: Sage.
l\:laynard, lvI., & Purvis,.J. (1994). Doing feminist research. In l\'L Maynard &
J. Purvis (Eds.), Rcsearching wail/ell's livcs (rom a (eminist pcrsjJectiue (pp. 1-
9). London: Taylor & Francis.
:;Vlenon, R. (I 987). Focuscd inrervic'INs and focus groups: Con[inui[ies and discon-
tinuities. Public Opinion Quartcrly, 51, 550-566.
Ivloore,./. \XZ (1970). tl'fcxicaJz-Americ<1lrs. Englewood Cliffs, NJ: Prentice Hall.
IVlorgan, D. L. (1988). Fows groups as qualit.ative research. Newbury Park, CA:
Sage.
I'\'forgan, D. L. (J 998). The (ows group gtridebooh. Thousand Oaks, CA: Sage,
1'dorgan, D. L., & Krueger, R. A. (1993). When to use focus groups and why. In
D. L. IVforgan (Ed.), Success{ul focfls groups: Adual1cing the state o{ the art
(pp. 3-19). Newbury Park, CA: Sage.
Oakley, A. (1981). Interviewing \vomcn: A contradictions in terms. In H. Robens
(Ed.), Doing {eminist research (pp. 30-61). London: Routledge & Kegan
Paul.
Olesen, \~ (1994). Feminisms and models of qualitative research. In N. K. Denzin &
Y S. Lincoln (Eds.), Handbooh o{ qualit.atiue research (pp. 158-174). Thou-
sand Oaks, CA: Sage.
Pesquera, B. lv1. (1993). In the beginning he wouldn't even lift a spoon: The division
of household labor. In A. de I;> Torre & B. 1'v1. Pesquera (Eds.), Building with
Ollr haltds: New directions i/1 Chicana studies (pp. 181-195). Berkeley: Uni-
versity of California Press.
Randall,]VL (1980). Todas es{amos dcsfJiertas. Mexico City: Editorial 5iglo XXI.
Reinharz, S. (1992). Feminislmethods in socia! research. New York: Oxford Uni-
versity Press.
Stacey,]. (1991). Can there be a feminist ethnography? In S. B. Gluck & D. Patai
(Eels.), \'i,!ol11en's ll.'ords: The (eminist practice of ora! history (pp. 111-119).
NC\v'tork: Routledge.
Stack, C. B. (1974). All Ollr I~ill: Strategies (01" sUTuival ilf a black co/lf1111I1lity. New
York: Harper & Row.
Stewart, D. \17., & Shamdasani, P. N. (1990). Focus groups: Theor)' and pracIice.
Newbury Park, CA: Sage.
Van Dijk, T. A. (1993), Elit.e discollrse and racism. Newbury Park, CA: Sage.
387
West, C. (1988). IVlarxist theory and the specificity of Afro-American oppression.
In C. Nelson & L. Grossberg (Ecls.), Mdrxistll mul the illterjJrctatio ll of ClI/-
lfIre (pp. 17-29). Urban;]: Universiry of Illinois Press.
WfiIlLim!ol1. ( ('/ 99.~ LPoem.; UrflU[ll; in fllminic:r l'lI~ef11'ch,
co-construction of meaning.
125.
\\'OJllUII '5
PnHIt'lI', inrt'lrncrlon. [lnd rhe
Stlli..-lies llltematiOlwl Fomlll] 21, 111- 11
\\/hne, \XI. F. (1943). Street comer society: The social stmctllre of ilil ltalian Sllllll.
Chicago: University of Chicago Press. Applied Ethnography
Zinn, Ivi. B. (1979). Field research in minority com1llunities: Ethical, methodologi~
cal and political observations by an insider. Social Problems, 27, 209-219.

frye Chambers

There is no general agreement \vithin the social sciences as to the

definition or value of either of the words that make up the title of this
chapter. Both words are imbued with assumptions of C0111mon meaning
and yet subject to many conflicting and largely unexpressed distinctions.
Accordingly, the subject of this chapter needs to be qualified in both of its
terms. \Vhat is it that makes a piece of research or a kind of practice specifi-
cally applied, and what do we mean by the term etlmogralJhy?
The question of how social inquiry becomes applied is intriguing and
can include considerations of intent as well as of research design and prod-
uct. I opt here for a broad definition. Applied \vorlc helps people make de-
cisions and is generally directed toward informing others of the possible
consequences of policy options or of programs of directed change. These
consequences may be anticipated (as in impact studies and forecasting), or
they may be determined in retrospect (as in evaluation research). In this
chapter I consider a variety of approaches to applied ethnography, focus-
ing primarily on cognitive approaches, micro/macro analyses, and action
Or clinical models. A major difference between basic and applied research
rests with the criteria used to judge their significance. For basic research,
the ultimate measure of significance is a research product's contribution to
theory and disciplinary knowledge. The most immediate meaSure of the
significance of applied research is its contribution to decision making.
This is not to sa y that a piece of research cannot be judged by both criteria,

388 389
bur only that it need not be. At any rate, the close Z1ssociation with decision
making accounts for much of the v8riation in approaches ro applied eth-
nography. It seems inevitable that the kinds of settings in which ethnog-
r~phei~ \.v()rl.", rnnp;:ng; {lon... (edc-rc.. ll..,llre"ucl'3cies 10 local "ctiv;s{: ot&'-'ni-
zations, should help determine the kinds of problems they are likely to
address, as well as their methodological approaches to trying to under-
stand those problems.
I reserve the term applied researclJ for inquiry that is intentionally
developed within a context of decision 1l18king and that is directed toward
the interests of onc or more clients. Basic research can certainly be usefully
applied, bur in order for the transition to occur betwecn its potential for
usefulness and its actual use, there must be a deliberate act of transfer from
the one realm to the other. This can be accomplished through various
kinds of practice in which individuals bring cthnographic knowledge to
bear on particular human problems. Although this latter role might be
realized in several ways (E. Chambers, 1985), I will discuss it here primar-
ily in the context of what are sometimes referred to as clinical modes of
practice.
In contrast to the above, my definition of etlmogra!Jhy wil! seem fairly
narrow to some readers. This is because in many quarters the term has
come to be regarded as virtually synonymous with qualitative research.
Follovving \Vo1cotr's (1995) clear but frequently unheeded counsel, I
restrict my use of the term here to those varieties of inquiry that aim to
describe or interpret the placc of culture in human affairs. In other words,
ethnography is principally defined by its subjcct marter, which is etlmos,
or culture, and not by its methodology, \vhich is often bur not invariably
qualitative.
The tcrm clliture is itself ambiguous and has becn subject to a variety of
interpretations, each implying somewh8t different methodological ap-
proaches. In my view, culture is composed of those undcrstandings and
ways of understanding that are judged to be characteristic of a discernible
group. As such, culture is an abstraction that has been applied, with vari-
ous degrees of success, to small groups (i.e., the "culture" of the class-
room) and to large groups (i.e., nation-states); an operational definition
might be that culture is apparent when distinctly shared meanings are dis-
covered to be present in any given group. \V'ithin most definitions of cul-
ture there is also an expectation that its features must h8ve some lasting
quality-durable enough, at least, to be acquired by newcomers to a
group. It is important to recognize that cultural understclllding can be
390
acquired from a variety of dara. It might, for example, be extracted our of
observable patterns of human behavior, drawn from "native" accounts of
explicit standards of valuc and behavior, discovered through the discern-
rn.cnt of I:ulcs or p,.nlcrn::; of intcr-prcm.tion that an; particular to d group, or
realized through the study of the more tacit dimensions of cognitive or
symbolic processes. In practice, ethnographers often combine these data
in ways that they feci will be productive given particular research aims.
The problem with the concept of culture as an abstraction that is attrib-
uted to particular societies or groups is that it has not been helpful for the
anticipation of actual behaviors, which seem almost invariably to result
from interaction between cultural groups. It has taken most anthropolo-
gists some time to realize that the concept is marc uscful and representa-
tional when applied to actual proccsses of cultural negotiation. Accord-
ingly, the attribution of culture to shared meanings \vithi n groups does not
necessarily imply that cthnographers are solely interesred in describing
such groups in isolation; in faCt, they are becoming less so interested. Nei-
ther does the expectation of some degree of lastingness suggest that a
group's culture is not subject to change. Earl y ethnographers generally did
cmphasize the uniquencss of neatly bounded cultures, and often assumed
that cultural patterns were deeply rooted in human consciousness and
resistant to change. They were more interested in describing durable
and distinct cultural units than they were in discerning the processes by
which culture becomes meaningful. lvlore recently, however, many eth-
nographers have shifted their attention to these issues of process, rrying to
understand how culture is constructed and negotiated, particularly as a
result of interactions between groups (or, if you will, between cultures). It
is not yet entirely clear what this shift might mean to the special interests
apparent in an applied ethnography, which has built most of its relation-
ships with clients and with the general public on the basis of advocating for
the significance of cultural understanding, in its earlier incarnation.
<$> Early Applications of the Ethnographic Approach
Although ethnography is no longer the sole province of anthropologists, it
clearly developed with a close relationship to the discipline, and its first
applications follow closely on anthropology's early commitment to the
study of presumably traditional and predominantly non-\Vestern societies.
By tbe time antbropology became establisbed as a discipline in its own
391
right, most such peoples were in one \Va y or another pol itically or econom-
ically subject to Western nations. It is not surprising that many of the earli-
est opportunities to apply ethnography \-vere closely related to problems
that represent3tives Dr coionin! or quasi-colon;'ll systems o::ncounrercd in
administering these subject peoples. l"i.vo early defenses of an applied
"anthropology" (i.e., erhnography) that is directed to the problems of
administering subject peoples stand Ollt, borh for the reasonableness of
their arguments and for their contrasting views of the kinds of data applied
ethnographers might collect.
The first of rhese is Bronislaw .Malinowski's essay "Practical Anthropol-
ogy" (1929), which offers his views of how ethnographic understanding
mighr assist British colonial ;'ldministrators in Africa. IVlalinowski's argu-
ment for the usefulness of ethnographic data in colonial administration
rests principally on the anthropologist's presumed grasp of what is some-
times called "the native point of vin\'." Because the anthropologist has
(more or less) lived within native communities and often participated in as
well as observed their pracrices l he or she is in 3 better position to interpret
the meanings and possible consequences of their behaviors in a cult-ural
context. The promise of a distinctly applied ethnography rests, howeverl
upon ethnographers' being willing 1'0 direct their research interests to mat-
ters of particular concern to administrators. IVfalinowski chided his col-
leagues for paying too little attention to understanding the subtleties of
language usc, Jnd for ignoring such vital administrative issues as local
practices related to land tenure, labor, and justice. He did not challenge
the premises of colonial rule l but argued that a more practical ethnogra-
phy vi'Oulcl provide administrators \virh the kinds of information they
needed to make more effective as well as more humane decisions regard-
ing native peoples.
lVlalinowski's defense of applied ethnography rests primarily on the
ethnographerls ability to observe behaviors and to explain their signifi-
cance in relation to rheir functions in a larger institurional and cultural
context. Another rationale for applied ethnography was offered by Clyde
Kluckhohn in his article "Covert Culture and Administrative Problems"
(1947). \X7riting on the basis of his experience with the U.S. Indian Service,
Kluckhohn noted that the agency had made considerable strides in ime-
grating cultural concepts and ethnographic data as part of their policy and
program development. He argued that further progress might be made if
ethnographers were to pay more attention to the COvert dimensions of cul-
ture, or to those culruraJ configurations that lie somewhere beyond a
392
people's conscious awareness of their own culture. This is a different justi-
fication for applied ethnography from one that relies solely on the discern-
ment of the narive point of view. Kluckhnhn suggested that ethnography
could uniqudy serve to make explicit those cultural configurations that in-
form human behavior, even though they may not be explicitly recognized
or easily verbalized by the "native." This kind of interpretation could be
particularly useful in instances where there were srriking differences be-
tween the configurations of particular Indian groups and those of the pre-
dominantly \vhite administrators-t\vo examples provided by Kluclchohn
included Navajo configurarions that relate sexual rights closely with prop-
erty rights and the relative lack of "guilt" as a motivating or sanctioning
factor in Navajo society. Kluckhohn \vas vague as to the kinds of ethno-
graphic methods rhat mighr provide a better understanding of covert cul-
ture. He encouraged ethnographers not ro limit themsCives to those fea-
tures of culture that could be "measured or quantitatively validated," and
suggested that informed speculation based on long~term familiarity with
another culture might weI! be appropriate.
A third distinct approach to applied ethnography was offered by Sol
Tax (1958) as a result of his work with the Fox Indians in Iowa. Tax
described his research team's efforts both as "acrion anthropology" and as
"participant interference," playing on the familiarity of the term partici-
parrt obscTlJatioll as a mainstay of ethnographic technique. Although
action anthropology did not suggest any new methodological approaches,
it did indicate a strikingly different way of applying ethnographic ex-
pertise. The Fox Project began in 1948, at a time when most of the eth-
nography devoted to American Indian communities was focused on doc-
umenting cultural traditions. Tax's team directed its interest instead to
the problems that the Fox were then facing, most notably the effects of
internal factionalism 2nd difficulties in their relations with whites. The
researchers committed themselves to a style of research and action that
would permit them to "learn while helping." In terms of innovations in
applying ethnographYl the most notable characteristic of Tax's efforts was
the extent to which the research team surrendered much of their control
over the direction of their inquiries, \vhich became increasingly dependent
Upon decisions madc by the Fox in dctermining how they would respond
to rheir own problcms. The action anthropology proposed with the Fox
Project (along with the work of sociologist Kurt Lewin) provided the
impetus for much of the collaborative research to be discussed later in this
chapter.
393

Tn the United States, much of the groullchvork for all applied ethnogra*
phy \vas established during and within the first cl.ecade after \X1orld \X1ar II.
In addition to the contributions mentioned above, ethnographers of this
era were involved in 3 wide variety of applications, including \.vorlz in in-
dustrial relations and organizational behavior (e.g., \X1hyte, 1984); large-
scale nutritional studies (IvIolltgomery & Bennett, 1979); research with
the War Relocation Authority related to the internment of the Japanese
(Leighton, J 945; Spicer, 1979); ethnographic research in the Pacific Trust
Territories, newly acquired by the United Stares (Barnett, 1956); and
numerous projects in rural and international development (e.g., Foster,
1969; Holmberg, 1958; Spicer, 1952). At least some of the impews for an
increased interest in applied ethnography at this time can be attributed to
the emergence of the United States as a global power and the recognition
of <I practical need for an improved understanding of cultural differences.
The research clients of the time appear not to have been particularly inter-
ested in ethnography as a specific method; it just came along with the
methodological preferences of most of the "cultural experts" of the time.
'*' From Stasis to Process
As noted <lbove, the development of ethnographic methods and strategies
has been closely associated with anthropology. Anthropologists continue
to take the lead in insisting on linking their inquiries to the understanding
of culturally based behaviors and values, thereby distinguishing ethnogra-
phy from the broader category of qualitative research methods. By the
same token, applied ethnography has never been defined solely as a prop-
erty of anthropology. For example, the Society for Applied Anthropology
was founded in 1941 as an organization devoted to encouraging interdisci-
plinary contriblltion5 to culturally based research and practice. Unlike the
leaders of other professional organizations for anthropologists of the time,
the founders of the Society for Applied Anthropology seemed less con-
cerned with drawing disciplinary boundaries or maintaining methodologi-
cal distinctions between disciplines than they were with establishing the
status of culture as an important variable in matters of human change and
adaptation.
Over the past few decades, a major shift in the way anthropologists, and
some others, think about culture and its significance in human relations
has led to a variety of innovations in ethnographic research. The kinds of
394
'I~I""'-'-' ... ,. .. ''J~' '-'1-" 'r
inquiries discussed above were based on an assumption thatrhe practical
value of ethnography was to prepare cultural profiles of differelH human
groups that could inform decision makers \vho were responsible for poli-
cies or programs that affected the lives of these people. 501 Tax's action
anthropology was an exception to this trend, although it shared with the
others a strong focus on viewing subjects in terms of the integrity of their
"traditional," usually non~\'{1estern, cultural base.
Early applied ethnography was broadly influenced by anthropology's
focus on peoples who had beelll11arginalized, gener<llly as;] result of\X!est-
ern expansion, and by the discipline's prevailing imerest in describing
these peoples' traditional cultures before they disappeared altogether
(another manifestation of early applied anthropology, sometimes called
"salvage anthropology"). The consequence, if not the intent, of this re-
search was to describe a people in terms of cultural ideals and behaviors
that were presumed to be enduring and homogeneous. These "cultures"
were often presumed to have changed litcle prior to \Vestern conwct, and
their standards of behavior wete generally described as being uniform
within the prescribed statuses and roles of each culture. The usual ethno-
graphic monograph of the time \vas written in \vhat has been called the
"ethnographic presem," which was in reality an aet:empt to describe an
idealized ethnographic past, largely out of time and with little or no refer-
ence to the present conditions of cultures that had been "adulterated" by
\Vestern influence (see, for example, Rosaldo, 1989). The impression \vas
one of cultures that did not change in terms of any kind of internal
dynamic, and that generally resisted change from the outside. Accord-
ingly, much of rhe applied anthropology of the time was devoted to pro-
viding the kinds of cultural information that might: help change agents
respond to the resistance they almost invariably faced in the field.
So long as anthropologists focused their attention on small-scale societ:~
ies and maintained the (generally erroneolls) presumprions that the cul-
tures they studied had been homogeneolls, isolated from other major
cultural influences, and had experienced little change prior to \X!est:ern
intrusion, their methods of research were based principally on their com~
mitmcnt to being there and on their ability to observe behaviors systemati-
cally in the cultural settings in which they naturally occurred. Anthropol-
ogy emphasized, and undoubtedly romanticized, the position of the lone
field-worker immersed in another culture and confident' in her or his abil-
ity to discover the salient patterns of that culture through sustained partic-
ipant observation. As a rule, anthropology students received little training
395
in methods prior to entering the field, and fieldwork instructions were
directed as much to practical matters of how to survive in anorher society
as they \vere to methods of inquiry. The methods were much like, and fre-
(JI.H:::ntly likened to, those of the natur<llisl, who assumed little significant
variation in the patterns of species behavior that he or she observed in the
wild.
Early ethnographic research was also influenced by a secondary mis-
sion within anthropology. A major goal of anthropology has been to serve
as an antidote to ethnocentric values, especially when those values lead
to misunderstandings about other cultures. This mission has led many an-
thropologists to emphasize the uniqueness of a culture and to focus on
rationalizing and defending differences rather than similarities betvveen
cultures. The role of the anthropologist as an investigator came to play
an interesting part in this quest. Ethnographies were frequently written
as narratives in which the anthropologist validated his or her discovery
of the rationale of human difference by recounting her or his own
nc(ivete, misunderstandings, and eventual enlightenment during the pro-
ceSS of doing fieldwork. Again, the methods that seemed best suited to such;]
perspective were those that were the most informal, experiential, and
reflexive.
Although early anthropologists did on occasion apply ethnographic
methods to a variety of situations outside the preferences described above,
it seems fair to assert that this model did predominate until the late 1960s.
At that time, several events served as stimuli to a fairly rapid transforma-
tion in the development of ethnographic methods and in the use of those
methods in applied settings. One of these events was the advent of more
systematic approaches to ethnographic inquiry, including ethnoscientific,
or cognitive, research methods, and increased attention to the quantifi-
cation of field data (Poggie, DeWalt, & Dressler 1992). During the same
period of time, anthropology experienced a rather profound shift in ori-
entation from its focus on small-scale societies in relative isolation to an
increased interest in seemingly more complex and heterogeneous social
groupings. This \vas not an entirely new situation for anthropologists, but
its increascd legitimation within in the fjeld helped transform the disci~
pline in at least three major ways. First, as anthropologists attempted to
deal with larger, more stratified, and more diverse populations, the need
for a systematic and somewhat more formal approach to field research and
ethnographic methods became apparent. These new research intercsts
396
also put anthropologists in closer association with practitioners of other
social science disciplines-sometimes in cooperation, and occasionally in
direcr competition. Closer association with specific applied research 5[[<1[-
cgic5, such as evaluation research and social impact assessment, encour~
aged some applied cthnographers to adapt their inquiries to these models.
Finally, and importantly, the experience of doing research with popula-
tions in \I"'hich heterogeneity was obvious led some anthropologists to
challenge many of their earlier assumptions concerning the relative "sim-
plicity" and seeming passivity of the smallerscale societies in which they
had been doing research.
These transformations were especially apparent in the development of
applied ethnographic methods. Here, the greater and most app:ucnt op~
portunity for anthropologists was still to try to assist change agents in
interpreting the behaviors of marginal groups. It no longer seemed suffi-
cient, however, to devote so much atrention to traditional culture, at least
not at the expense of understanding how these groups did indeed interact
With, adapt to, and participate in the larger political and economic systems
of which they WEre clearly a part. In ethnography devoted to agricultural
development, for example, the original questions had focused on under-
standing what it was in a people's traditional culture that prevented them
from participating fully in a newly introduced development scheme. The
new research questions were framed in a context in \,."hich research sub~
jects could be seen as rational actors in a development process (e.g.,
Bartlett, 1980; \X!arren, Slikkerveer, & Brokensha, 1995).
All these factors led to a rather rapid shift of focus. For many applied
ethnographers, the object of research changed from the study of particular
cultures or social groups in place to the study of the cultural processes that
occur in efforts to respond to particular human problems. Applied eth-
nographers have become increasingly interested in such matters as how
social groups collide and mix in situations of change and in the cultural
meanings that result from the interactions of interested groups or stake-
holders. They have become less convinced of the value of viewing culture
as a quality that people possess (or, in rum, a quality that possesses them)
and more interested in discovering how cultural meanings might be ex-
changed and negotiated as a result of intracultural attempts to find solu-
tions to problems. In this sense, the culrure described by many applied
ethnographers has shifted from being a durable repository of a people's
traditions to an unstable and mutable process by which people actively
397
strive to derive meaning from their continuall I' changing relatiollsh ips and
circumstances.
~ Varieties of Applied Ethnography
Ivluch of the value of ethnography lies in its narrative-in the telling of a
story tbat is based in cultural representations. \X!hereas traditional ethnog-
raphy has focused almost exclusively on telling stories about "orher" and
generally distant peoples, many contemporary ethnographic ~lpproaches
tend to focus on the ways in which people fashion culrurally meaningful
expressions from fields of experience in which meaning is routinely con-
tested, and where culture is perennizllly under constructioll. This has been
particularly true of applied ethnography, which is by its very nature inter-
ventionist and culturally intrusive. As noted above, the problem of the
8pplied ethnographer 1135 largely shifted from telling stories about others
to describing what happens when cultural systems overlap as a resulr of
some sort of deliberate, recent, or anticipated intervention.
Cognitive Approaches
Cognitive, semiotic, and semantic approaches have played a significant
role in the development of applied etlmogrJphy. These approaches follmv
a variety of language or discourse models and ethnographic methodolo-
gies in order to provide, much as Ivlalinowski espoused, "tbe native point
of vievv" in relation to particubr kinds of interventions. Classic ethnogn.-
phies in this vein include James Spradley's YOll Owe Yourselfa Drt/n/;:,: All
Ethnography of Urban Nomads (1975), IVlichael Agar's R.ipl}illg and RUH-
ning: A Formal EtlmogratJhy of Urban Heroin Addicts (1973), and Peter
Manning's Police \Y/ork: The Social Orgallization ofPolicillg (1977). These
and similar contributions differ from standard (i.e., nonapplied) eth-
nographies in that they are focused on a research population that is at least
in part defined by some larger social "problem." Spradley's study is
grounded in public perception of the problem of street J.lcoholics and in
police approaches ro solving the problem, largely through incarceration.
Agar's study was conducted in a treatment facility for heroin addicts.
Other ethnographies, such as 1\1anni ng's study of police \vork and l\1ichael
Agar's more recentll1dependents Declared: The Dilemmas of J/ldepeJldent
398
Th/ckillg (l986a), focus on particular occupational groups that are subject
ro significant public interest or institutional intervention-Agar's study is
concerned, for eX3mple, with the impacts of federal transportation tegula-
liOll Oil iIldcpcndclll lfucli:cLs.
In an applied sense, the value of cognitive approaches to ethnography is
that they often reveal patterns of cultural construction within a target
group that vary considerably from the interpretations of that group that
are made by outsiders, and most particularly by those who have some
authority over the group under study. The ethnographies often reveal con-
siderable disparities between policies, programs, or treatments directed ro
the group and the real-world exigencies that provide a cultural landscape
and help determine behaviors within the group. To my mind, the most
appealing of these ethnographies treat culture as a consequence of practi-
cal instances of intervention (including, interestingly enough, the inrer-
venrions of research). As Agar (1994) notes in his discussion of the role of
the intercultural practitioner (ICP), culrure should nor be thought of as
an abstract set of principles that belong to a parricular group, but rather
as an orientation toward resolving differences:
Culture is something the ICP creates, a story he or she tells that highlights
and explains the differences that cause breakdmvns. Culture is not some-
thing people have; it is something that fills the spaces betwccllthem. And
culture is not an exhaustive description of anything; it focuses on differ-
ences, rich points, differences that can vary from task to task and g,roup to
group. (p. 236)
Cognitive approaches to applied ethnography derive much of their
practical value fro111 tbe fact that they tend to focus on failures of commu-
nication, or cultural "breakdowns" (see, e.g., Agar, 1986b; Briggs, 1986).
Such failures arc easily translated to applied situations, and cognitive erh-
nographic methods seem readily adaptable to the particular situations and
communicative dilemmas that arise when one group attempts to intervene
upon another. Although the approaches seem to hark back most dosely to
IVIalillO\vski's defense of applied edmography in terms of eliciting a
"native point of view," there are also parallels to Kluckhohn's attempt to
describe a unique role for applied ethnographers in their ability to ex-
POse more tacit dimensions of culture. Cognitive and language-based
ethnographies tend to be built on those communicative breakdowns that
399
 ,.lL" "J'-'..J V, L.VLLL'- I " " " " MI~U MI~MLI .1...,,',>.;>
arc directly experienced by the researcher and that are to be resolved by
tbe researcher's attempts to understand what m8de the breakdowns occur.
This is different from an applied appro8ch that repons problems as ex-
p,..,,>s<c:c[ l-,y ..,!,ang", "lSent:s or h y t1-.", s,-oups t:hcy
There is also an assumption, seldom made entirely clear, that the ethnog-
rapher is better positioned to unravel communicative disorders between
groups than 3rc members of the groups themselves. This is not simply
because researchers might have less at stake in anticipated <lctions, but also
because they initially understand much less of the situation and are there-
fore more likely to experience firsthand the kinds of blunders and break-
downs that yield rich data and point toward communicative resolutions.
Stakeholders, \vho are more closely associated with the problem at hand,
are likely to have already developed cognitive defenses that insulate them
from direct experience of the kinds of breakdowns that yield significant
data or understanding. James Spradley (1979) has, for example, suggested
that his own techniques of ethnoscmanric elicitation seem to work best
\vhen the ethnographer comes to a cultural situation with little prior
knowledge.
Approaches of IV\icro/Macro Analysis
Although ethnography has traditionally dwelled lIpon the local and
relied heavily upon qualitative research techniques and methods, a variety
of current approaches to applied ethnography have served to expand
upon both the subjects and the methods of inquiry. Two factors have
played major roles in this transition. One has been the recent tendency
within the social sciences to resituate the local within the larger contexts of
regional, national, and even global events. This focus seems particularly
relevant to applied research and action, where deliberate efforts to inter-
vene and bring about change invariably pbce the populations subject to
such efforts in a relationship with a larger sphere of influence. Increas-
ingly, applied ethnography is about these relationships, rarher than about
the experiences of particular groups or populations in isolation. The other
factor has been the tendency for applied ethnography to expand its meth-
odologic81 reach, not only with the use of increasingly sophisticated quali-
tative methods, but also by adopting or at leasr responding to more quanti-
tative research methods. Some of this shift can be attributed to changes in
the ways in \vhich social scientists and practitioners have become involved
in applled work. (Again, I am speaking here primarily on the basis of Illy
400
~IV\rlJ'\I'-f-""_ 'VI""", CJ'\!Kl-.l
experience with applied anthropology, although I suspect these comments
apply to the social sciences more generally.) In the past, anthropologists
normally became involved in applied research when they were called upon
~-'r", ,-ntcn'ptitl5 to chanse. lO proy ide r.;x pcnibc 1.Ju.0cd Oll LlIcir [(!lOW leuge of a panicu!;u people. -rIley
were most likely to become involved in a project after something had
already gone wrong, and their involvement in the overall project or re-
search activity was minimal. In other words, theif ethnography \vas largely
independent of the larger intervention, and their role was primaril y that of
expert witness or troubleshooter.
There have 8tways been exceprions to this practice, and nOVi' the excep-
tion is often the rule. Applied ethnographers are more likely, for example,
ro be involved in a project or research effon from its inception, and to play
a role in research design. They are also more likely [Q be a parr of the ;m<lly-
sis or interpretation stage of applied research efforts. This greater degree
of integration of applied ethnography within larger res,::arch projects and
inrcrvemiol1s has meant thar a considerable amounr of [he qlwlitative,
field-based applied ethnography being clone today has been developed in
studied relation to other research approaches. One way in which this
has occurred is when the applied ethnographer becomes a p8rticipcl11t ill
a larger, more quantitative research activity, generally as a qualitative
researcher. The other approach has been when applied ethnographers
themselves adopt quantitative methods as a part of rheir own ethnog-
raphic research plan-this has occurred to the extent that some applied
ethnography now relies predominantly upon quantitative rather than
qualitative methods. (Here it is again important to keep in mind that eth-
nography is being defined in this chapter in relation to its subject matter,
rarher than as a particular kind of methodology.)
In terms of methodological innovativeness, the most imeresting exam-
ples of ethnographic inquiry that include qualitative and quantirative
approaches, as well as varied levels of analysis, are those in which rhe eth-
nographer plays an active if not leading role in research design and is
thereby able to maintain a primary interest in cultural analysis. \"X'ithin
anthropology, the qualitative/quantitative approaches to ethnography,
Wlth a focus on application, arc perhaps most rhoroughly described in
Penti Pelto and Gretel Pelto's Anthropological Research: The Structure of
Inquiry (1978), H. Russell Bernard's Research Jvlethods in Anthropol-
Db')!: Qualitative and Qualltitative Approaches (1994), and Bernard's
Halldbook. of Research ldethods ill Cultural Anthropology (1998). The
shift within anthropology from local or community levels of analysis to
401
approaches that include "macro" analyses in several dimensions (space,
causal, :md time) is evidenced by the contributions to be found in Billie
De \Valt Jnd Penti Pelw's.Micro al1d A[aero Levels ofAl1alysis il1 Allthmpo/-
oU]!: I.rHfl!5: ill T1Jlwl'~' and R~_~,MI'cl, Cl9QS} and ;'1 Possie <.or a1. 's Alltlnopo~
logical Research: Process ami Application (1992).
These kinds of research experiences have provided opportunities for
applied ethnographers to articulate ways in which the qualitative aspects
of their work can be inregrated with larger research activities. Qualitative
approaches have, for example, proven particularly useful at the beginning
of a research effort, often to help define the parameters of survey research
(e.g., Kempton, Boster, & Hardey, 1995). In other work, qualitative field
research has proven to be particularly useful at the analysis stage of re-
search, in helping to explain anomalies in the dara, especially in research
that relies heavily upon reponed data from a number of sites, where there
might be unaccounted-for variation in how the sites interpret reporting
requirements (Trend, 1978).
\Vhcre research requires accurate portrayals of stakeholder values or
opinions, qualitative ethnographic data have often proven superior to sur-
vey data, particularly in cases that involve long-term field exposure and in
situations where informants might feel at risk or have other reasons to pro-
vide incorrect responses, or where their "truer" responses might develop
over time. In some cases, as I will discuss later in this chapter, qualitative
ethnographic dara have also proven to be more credible to particular re-
search clients and stakeholders.
It is rare to find repofts that explicitly compare qualitative and quan-
titative data obtained in the same applied research project. Susan
Scrimshaw (1985) offers one such instance, related to a study of practices
and attitudes concerning induced abortion in Ecuador. She notes that the
ethnographic (i.e., field-based) data were in most cases validated through
the comparison of these results with tbose obtained as a result of the larger
and more rigorously defined sampling procedures developed for the sur-
veys. On the other hand, Scrimshaw cautiously suggests tbat the ethno-
graphic dara might have been more accurate in those few instances in
which significant anomalies betvveen the t\VO data sets did exist. This
appears to have been the case, for example, where research subjects were
asked to repoft practices or attitudes that were particularly sensitive or
potentially embarrassing. In these instances, the longer-term familiaritY
gained as a result of ethnographic experience enabled the researchers to
402
gain a deeper and more accurate grounding from which to discern the sub-
jeers' attitudes.
Adion and Clinical Approoclles
A number of current approaches to applied ethnography follow the
early action or advocacy models offered by Sol Tax (1958) and, in another
context, discussed by psychologist Kurt Lewin (1948), who emphasized
participatory research based on client-oriented attempts to resolve partic-
ular social problems. Some of the models clerive directly from these earlier
examples, whereas others have been encouraged by recent interest on the
part of research funding agencies, particularly some federal agencies, in
sponsoring applied research that is participatory and that involves the
"subjects" of research or intervention in one or several stages of inquiry.
The actual extent of participation required for these activities varies con-
siderably from one research effort to another, and in some cases mandated
"participatory" research has earned a reputation for superficiality that is
not dissimilar from earlier mandates that called for "citizen participation"
in public decision making.
There is nothing necessarily ethnographic about action or advocacy re-
search, although some researchers have suggested that ethnographic re-
search strategies may be more accessible than other strategies, and, at rhe
stage of interpretation, might provide kinds of data that are more convinc-
ing to those participants and community members who have not been
trained as social scientists. As in the other approaches to applied research
discussed here, much of what has been identified as applied "ethnogra-
phy" would not conform to the definition offered above-it may be good
and informative qualitative research, but it lacks the association with
understanding cultural processes that would make it ethnographic.
Not\'vithstanding these provisos, there remain plenty of instances in
which applied ethnographic methods have been incorporated into action
and advocacy research strategies. Jean Schensl1l (1985) has described
action-oriented research within the context of a collaborative model in
which researchers and community activists form "policy research clus-
ters" that are focused on important community problems. She notes that,
in these settings, the ethnography is shaped in part by "the constraints of
field situations including the social and political realities of the dissemina-
tion/utilization context," as weil as by considerations that the research not
403

violate "cultural principles" within the community (p. 193). \X1illiam
Foote \X/hyte (1984) has offered a similar model, which he calls "participa-
tory action research." Dianne Argyris (1990) has provided perhaps
thL: most dt:Li:likd accouming uf applied ethnography within [he action
context, and she suggests that a major goal of action research is to encour-
age participants to test their own "theorics~in-l1se"as they relate to partic-
ular social problems.
The major innovation to applied ethnography resulting from action
models has been the tendency to include individuals who arc not pro-
fessiona! social scientists in various stages of the research, including proj-
ect selection and design, fieldwork, and research analysis. Conversely,
and probably more than any other approach, advocacy research has also
resulted in the professionalism of any number of traditional research sub-
jects as social scientists in their own right-a practice extending back again
to Sol Tax, who devoted consider;lblc time during the latcr years of his
career to facilitating the education of Native American students in the
socialscicl1ces and in other fields. Davydd Grccnv\iood and Ivloften Levin
(1998) have offered a survey of action research that focuses on its criteria
of scientific evidence and commitment to "democratizing" the research
process.
Allied in many respects with action and advocacy research is the clinical
approach to applied ethnography, which in the simplest of terms seeks to
train people to use ethnographic strategics to gain a better understanding
of their own cultural situations, or to understand more fully those cultural
processes that influence others with \VhOlll they are involved. The clinical
model has referred most often to those activities in which professionals of
various kinds arc encouraged to think about their practice in ethnographic
terms. It has most often been applied to settings in which the professional
is likely to encoumer considerable cultural diversity, as in medicine and
health (e.g., Shimkin & Golde, 1983), education (e.g., Heath, 1982),
and social work (e.g., Green, 1982). Howard Stein (1982) describes the
clinicalmoelel as one of "ethnographic teaching," in which the profession-
als (in this case, medical school students) are encouraged to use an
ethnographic perspective and at least some methods of ethnographiC
inquiry to first conduct "self-ethnography," and later to apply principles
of ethnography to understanding cultural diversity \vithin the clinical
setting.
404
MfJfJllf::::U crnnograpny
Edgar Schein (1987) has argued the clinical perspective somewhat dif-
ferently. His clinicialls are organizational researchers and advisets who
rely on fieldwork approaches that Schein claims ate distinct in several
respects [rum those of ethnographers. I t is difficult, however, to determine
how Schein derives his model of ethnography-wh3C he describes as the
opposing clinical model seems in actuality to be quite similar to the
approaches of applied ethnography described in this chapter.
lr seems likely that attention to ethnography as a way of thinking and
problem solving will increase as social scientists learn that applied work is
nor simply a mattcr of affixing existing rescarch paradigms to human
problems. In the examples cited above, an ethnographic perspective is
offered to aid practitioners in a variety of fields, presumably to increase
their sensitivity toward issues related to cultural differences and cultural
process. It is equa!ly important to recognize thm practitioners do havc
theit own models of change and intervention, and that these models might
conflict with an ethnographic perspective and militate against the effec-
tive use of ethnography in clinical or applied settings. Peter Rigby and
Peter Sevareid (1992) h3ve, for example, contrasted the ways in which
lawyers and ethnographers deal with "facts" and "evidence," and Dr8ke
(1988) has contrasted ethnographic thinking with that of the governmem
bureaucrat.
~ Major Issues in Applied Ethnography
The increased use of ethnographic methods and perspectives in applied re-
search has leel to a variety of emergent issues, some of which are unique to
the ethnographic approach and others of which are more general and relate
primarily to differences between applied and basic research.
Criticisms of Ethnography in Applied Work
Issues of reliability and internal validity remain significant concerns
:e1ated to the usc of ethnographic approaches in applied research. These
ISSues have become more prominent as ethnographers increase their in-
volvement"with research teams that include social scientists who approach
their work from a more quantitative perspective. By the same token, such
eXperiences have contribmed to an increased sophistication on the part of
405
 Mt IliUU) Ul- LULLtL 11I'1l.:> AI'JU Al'lALT LIJ'H.:J tNIYJl<.:ILAL IY\AI CKIAL:>
many ethnographers, both in terms of expanding their methodological
repertoires and in terms of their being better able to articulate measures of
reliability and validity within the more qlidlit3tive ~spects of their inqui~
ries (e.g., Bernard, 1994; Kirk & Miller, 1986), although some ethnog
raphers have argued that ethnographic research requires distinct criteria
for judging the validity of its methods (e.g., Guba & Lincoln, 1989). As
Hammersley (1992) notes, criteria of validity ultimately have their point
of reference in concepts of '"truth." This allows for the possibility that the
"truths" derived from attempts to understand and interpret cultural pro-
cesses are different from those of other phenomena (e.g., Gcenz, 1983). It
also allows for the certainty that the "truths" of applied research arc gen~
crally more contingent and subject to varied criteria of utility (see below)
than are those drawn from the framework of "pure" and discipline-bound
research. If validity is not entirely in the eye of the beholder, it clearly does
stand in relation to the needs and judgments of those who have in a stake in
any particular applied research activity.
Another criticism of ethnography has related to the wealth and richness
of the data that applied ethnographers typically retrieve, particularly
when they are involved in large-scale research efforts that might involve
collecting data over periods of several years, and that include a number of
ethnographers or qualitative researchers producing data from several dif-
ferent research sites. Early efforts of this kind v{ere oEren initiated without
adequate attention to how the field data would later be analyzed or the
extent to which data collected by different field-workers at different sites
would be comparable. l\1ore recently, the adoption of computer-based
ethnographic programs, such as NUD" 1ST and the Ethnograph, has
proven useful, in terms of both managing ethnographic data and facilitat-
ing preliminary data analyses. The usc of ethnography in applied research
has also been criticized as taking too long to serve as an effective rool for
decision making, and as being too broadly focused (ethnographers do not
always stick to the "problem" specified by their clients). In regard to time-
liness, ethnographers have made considerable progress in developing
"rapid assessment" procedures, particularly in areas of development re-
search and resource management. In defense of their usually broad or
holistic focus, ethnographers have countered that their methods are often
fundamental to exploring the extent to which the "problem" identified in
applied research efforts might have been misconstrued or might fall off the
406
mark-an argument that may have considerable merit but hcls not proved
particularly comforting to some research diems.
The potential for perception~ of rese;ucher bias in applied ethnog-
raphic research nnd pr3crice extl'nds beyond issues of reliability and valid-
ity. \Vhether their efforts are applied or not, ethnographers have often
been regarded as advocates for the people they study, and at: least some
ethnographers would themselves not dispute such a claim. This seems par-
ticularly true of work in anthropology, \vbere a claim ro representing and
protecting the "best" interests of those people typic:dly studied by anthro-
pologists has long been a part of the disciplinary standard. Although
ethnographers certainly need to be attentive to the extent to which their
traditions of advocacy contribute to biases, it is equally apparent that criti-
cisms concerning possible bias can come about simply as a result of the
kinds of conclusions for vvhich ethnographers aim. In an applied setting,
the srrength of ethnography is its capacity to identify culwral parrerns tbat
provide reason and meaningfulness to human values ~!l1d behaviors.
Change agenrs and other potential clients for applied research are them-
selves usually close enough to the problem that they have their own inter-
pretations of the val ues and behaviors that arc being studied. Ethnography
has the potential to provide alternative interpretations that might be
greeted with enthusiasm by some research clients, or that might be so
incompatible with client expectations as to be dismissed JS biased and
rejected outright.
Defenses of Ethnography in Applied Work
NIany of the characteristics that have given risc to criticisms of tbe value
of ethnography for applied research can also be offered as strengths pZlrtic-
ular to the method. The long-term and relatively intimate acqu<linrance
with research subjects that is characteristic of much ethnography provides
rich, contextual information that can increase the depth of our knowledge
of particular subjects. Ethnographic and qualiwtive research approaches
have been successfully applied to research situations in which subjects are
not likely to be candid in response to such instruments as survey qucstion-
I~aires, or where there are likely to be significanr differences of inrerpret<l-
tlOn regarding the appropriate responses to direct questions, cultural dif-
ferences in the etiquette of inquiry, or even in tbe meaning of particular
questions or responses. In terms of analysis and interpretation of research
407

results, ethnography adds a cultural dimension that is likely to be absent in
other approaches.
Recent approaches to applied ethnography have also scrved the useful
purpose of simply helping people (clients, research subjects, and so on)
think about the idea of culture, how culture "works," and culture's conse-
quences. In some instances, this might require ethnographers to help oth-
ers unlearn the popuhlr, more static concepts of culture taught by earlier
ethnographers, and most particularly by anthropologists (Chambers,
1986).
Research Clients and Applied Ethnographic Methods
\\lherc3s basic rese3rch methods arc generally developed within disci-
plinary boundaries and subject almost solely to peer review, applied re-
search invariably enta.ils convincing a client of the appropriateness and
effectiveness of a research strategy. This may require the ethnographer to
adapt his or her methods to the particular needs and constraints expressed
by the client. Approaches to rapid ethnographic methods have, for exam-
ple, derived specifically from the need of particular dients to receive data
in a timely fashion (e.g., R. Chambers, 1985; Harris, Jerome, & Fa\vcett,
1997; van \\lilligan & Finan, 1991). Similarly, ethnographic methods have
been adapted to the special needs of clients who feel the need to conduct
assessments of the potential social or cultural impacts of their interven-
tions (e.g., Cernea, 1985; Cochrane, 1979).
Rapid research might include a variety of techniques, but those that
seem to be most commonly associated with ethnography arc (a) the use of
focus group interviews; (b) stepwise" research, in which long-term field
presence is replaced by brief ethnographic "visits" to solve particular r~~
search problems posed by an ongoing research project; and (c) participa-
rory research strategies that involve those at the research site in data col-
lection efforts. These approaches, as they relate to applied ethnographic
research, are discllssed in greater detail by the contributors to van Willigan
and Finan's (1991) edited volume.
Applied ethnographers ate not of a single mind as to the extent to which
their research approaches should be adapted to the declared needs or
expectations of research clients. For example, \X'olcott (1994) has
expressed concern about the influence of evaluation research models on
ethnographic research in education. \"X7olcott argues for a more descriptive
408
MpfJlI!::U ctrlrluyrupny
and nonevaluative approach. In contrast, others have built their ap-
proaches to applied ethnography direcdy upon evaluJtion research mod-
els (e,g" Fetterman 1984; Ryan, Greene, Lincoln, Mathison, & Mertens,
1995 ).
Criteria of Utility in Applied Ethnography
By and large, applied ethnographers have nm been attentive to the task
of proving the value of their inquiries. Although there h3:; been consider-
able discussion of ways in which ethnographic research might be useful,
there has been relatively little effort made to establish actual uses. \"X7ithin
anthropology, the usual genre for discussing the practical uses of ethnog-
raphy has been the case study, which relies heavily upon establishing the
reasonableness of an ethnographic interpretation, bur seldom offers clues
as to whether those interpretations were actlI::1Jly incorporated into c1icnt
decision making, policy construction, or program intervention. A major
exception is Mahillg Our Research Useful: Case Studies in the Utilization
of Allthro[JOlogical Knowledge (van \\lilligan, Rylko-Bauer, & IV1cElroy,
1989). Although the authors in this volume identify numerous factors that
appear to contribute to effective utilization, many of them cluster around
issues of collaboration and communication. Thc effective application of
ethnography seems related to the extent to which the research client is
actively involved in the research effort and the extent to which the eth-
nographers are willing to serve as advocates of their tesearch and to com-
municate their findings in different ways to a variety of stakeholders.
In a similar vein, I have argued that criteria of utility are as vital for
effective applied research as might be thc more usual and variable criteria
for establishing scientific reliability and validity (E. Chambers, 1985). I
have identified five such criteria. The accessibility of research findings
refers to the criterion that knowledge be available in an appropriate man-
ner to those who have a stake in a program of change. Applied edmogra-
phy should also be releuant to the goals and prescribed activities of stake-
holders and clients. It also needs to be responsive to different claims upon
the sigllificance of a course of action (understanding, for example, that the
claims of the significance of maintaining historic structures might well be
different among historic preservationists and among community members
who more acutely feel the need for a modern shopping ceorer). Fourth,
applied ethnography should meet a criterion of credibility in terllls of
409
heing responsive to those stand;:lfcls of evidence and proof that 3ft Lwored
b\' clients and stakeholders. Finally, applied research needs to address m,H-
r;rs of prospect and judgment (or, in other words, to understand that
stakeholders and clients are often more interested in what could be, or
even in what should be, than they arc in what currently is).
What Kind of Science'
As fm as I can determine, there has never been agreement 3mong
applied ethnographers as to rhe kind of science they practice. I am certain
this is true within anthropology. Early accounts range from the interpre-
tive and even speculative sciences (e.g., Kluckhohn, ] 947) to models that
emulate the natur;:d sciences in their attention to precision and predictabil-
ity (e.g., Chapple, 1941).lvlost current applied ethnography follows one
or the other of these models, with some researchers hJsing their work on a
narural science paradigm that places emphasis on the search for testable
and causal statements related La cultural processes (e.g., Bernard, 1994;
Poggie et al., 1992) Jnd others focusing on interpretive models that, La an
extent at least, emphasize the uniqueness of particular cultural circum-
stances and events (e.g., Agar, 1996; \\lolcott, 1994). In this regard,
\'V'olcott has gone so far as to question whether a preoccupation with valid-
ity in applied ethnography might actually detract from an ethnographer's
quest for understanding.
A major difference in the \vays qualitative ethnography plays out in
these twO models is in the manner in which the method is integrated imo J
research project. For the natural science model, qualitative ethnography is
often accompanied by, and sometimes dominated by, more quantitative
methods. Interpretive researchers and practitioners tend La rely more fully
on qualitative approaches as the basis for their inquiries.
- In contrast to developments in general anthropology and elsewhere in
the social sciences, postmodernisl11 h3s played a relatively minor role in
the further development of applied ethnography, and in some quarters
there is open antagonism toward the approach (e.g., Young, 1998). Other
have suggested that some aspects of the postmodern critique might con-
tribute positivelv to a more critical understanding of applied ethnog-
raphers'- own po~itions in the processes of inquiry, as well as to an explora-
tion of alternative approaches to textual rf'presentation (e.g., Johannsen,
1992).
410
Issues of Ethics and Morality
Applied research often has immediate consequences for [hose who
become subject to its gaze. In many cases, where applied ethnography is
employed as a pan of a knowledge base from which to make decisions
about the fate of communities and their environments, the ethical and
moral considerations can be daunting. Not infrequently, applied ethnog-
raphy is conducted under conditions of planned or unplanned change, in
which communities arc experiencing the pressures of disruption and
manipulation of their lives. Ethnographers vary considerably in the ex-rem
to which they feel their primary role in such situations is to try to under-
stand wlnt is going on or to try to help alieviJte human suffering.
Beyond those issues that ~lffect social science researchers in general,
applied cthnogr<1phcrs have encountered a number of ethical and moral
dilemmas that arise as a result of the researchers' unique relarionships to
research diems. One of the more intriguing of these is the question of
whether or not ethnographers should be engaged in resemch with clients
whose policies or actions might not conform to professional or personal
standards of mor3lity. In putting forth the idea that anthropologists could
serve useful purposes related to British colonial administration in Africa,
I\ilalinowski (1929) walked a thin line between complicity and arguing
that the presence of anthropology would help humanize the colonial mis-
sion. Similarly, social scientists involved as community analysts in the re-
location camps established for Japanese Americans during \Vorld \Var II
have sought ro distance themselves from the policy and have described
themselves primarily as "cultural brokers" between the Japanese internees
and the \XTar Relocation Authority (Spicer, 1979). This idea that involve-
ment is better than boycott pervades much of current applied research,
although others have argued that association is the equivalent of complic-
ity and have recommended that applied researchers excuse themselves
from work with a wide variety of cliems (e.g., Berreman, 1991; Escobar,
199] ).
Three issues of professional ethics have proven particularly difficult
for applied ethnographers. One of these is the principle of informed
consent. \Vithin anthropology, Fluehr-Lobban (1994) has suggesred
that the tendency of field-workers to resist informed consent may result
in part from their assumption that they are in a better position than
their research subjecrs to determine and to mitigate any potentially harI11
ful consequences of their research. She decries this attitude as being
411
 ,. ,.... , , ,'-''-''-' ..... ' ...................... __ , " ..... ,-., ..... ,..." .,... .... , ,-" ''-' ....'.n ,,,,'-,..,.... ,,,,..., .... '",... ....-' MIJIJ/lI:;'U crnnograpny

inappropriately paternalistic. On the other hand, Murray \"X7ax (1995) has ~ "native point of view," in uncovering the more tacit dimensions of CUlture,
argued that there are instances in which strict adherence to principles and in facilitating collaborative or "acrion" research str<ltt:gies. On the
of informed consent would make fieldwork difficult if not impossible- other hand, a major change in the ethnographic perspective has been the
espec;,..,lly in C<1.SC". ..vl~",re etl~nogr<>.pl~,"r"nre inve"r; ontine; illec;o.l ..~ctivity <.n- S["ClduJJ shift frvmvinving, cultures as relatively closed and geographically
are investigating elites (i.e., "studying up") who are likely to resist their bounded systems to focusing attention upon those cultural processes that
inquiries. emerge as a result of programs of intervention and changc. In this respect,
Applied ethnographers also face ethical dilemmas related to the protec- applied ethnographic research has kept pace with more general trends in
tion of the confidentiality of research subjects. This problem can relate to tracing the interplay of local and global ,lSpeets of cultural expression.
research subjects in general, as it does in cases in which research clients The past two decades have seen changes and improvements in the
have or request access to records that identify research subjects. Jt can also development of applied ethnographic research methods :.md techniqucs.
relate to particular research subjects whose identities cannot be concealed These changes have included the following:
by vinue of their unique positions in an institution. The professional codes
of both the American Anthropological Associarion and the Society for ~ Increased sophisticmion in developing cultural (and cognitive) models in
Applied Anthropology, which at one time specified that the confidentiality response to issues of inrervemion and problem solving;
of research subjects should be maintained in all cases, have recently been v A greater diversification of research methodologies and techniques to in-
revised to indicate rhat subjects should at least be informed of those cases clude both qualitative and quantitative approaches, and the use of compurer-
assisted ethnographic research programs;
in which confidentiality cannot be assured.
~ Ivlore complete involvement of ethnographers in the v,lriotl5 stages of applied
In general, the social science professions have encouraged open and
research (project design, analysis, disseminmioll. and 50 on), resulting in gre<Her
public dissemination of the results of research. During the 1960s, rhe
understanding of specifically applied and policy-oriemcd research strategies;
American AnthropologicalAssociation went so far as to proscribe "secret"
+ A greater tendency for ethnogr<lpbers to work. 35 members of rese,lrch teams,
research altogether. This prohibition seemed unreasonable to some ap- either within their own disciplinary framework. or as p<lrticipants in imcr-
plied ethnographers, who found themselves engaged in research in which disciplimry research;
their clients appeared to have legitimate proprietary interests over the * A tendency for ethnographers to pay greater anemion to the importance of
data. Such interests included, for example, research devoted to the test matching research strategies to specific diem needs for information and in-
marketing of new producrs, evaluations of marketable techniques for sight, and to pay heed to "criteria of utility";
counseling and consultation, and long-term social experiment research + Increased experience in applying ethnographic methods to participatory and
in \vhich the early release of preliminary results might compromise tbe "action" research strategies.
experiment. These claims have by and large been held to have merit, and
most professional associations, including the American Anthropological \\!ithin anthropology as a \vhole, this same period 113s seen increlsed
Association in its 1998 revision of its principles of professional responsi- acceptance of specifically applied research as a legitimate ane! important
bility, have revised their codes to allow for the withholding of research enterprise. Increased support and institutionalization of applied ethnog-
results in cases in which there appear to be legitimate proprietary interests. raphy from within the discipline, as well as increased interest from outside
the discipline, suggest thar the progress made to date will continue well
into the future.
.. Summary Are there other areas in which applied ethnography has not pro-
gressed? I think there are. One of these is the failure of many ethnog-
Early attempts to apply ethnographic research to problem solving an(~ raphers to follow up on the actual uses of their contributions. Except in
social intervention introduced three major rationales that retain much at Occasional asides and anecdotes, no clear conceptualizations of utilization
their influence. These point to the strengths of ethnography in eliciting the have emerged from the practice of applied ethnography. As in rhe past,

412 413
 1V1t: I MUU:;, U, LULLt:L III'H.:l P,1'IV P,j'jP,1..1 LII'I\.:l !:!V\rl"'ILMI.. !VIMI L!"IML-'
ethnographers seem content to stress the supposed usefulness of their
inquiries rather than to explore empirically what happens when their
kllO\vledge enters the fra)' of public recognition, intervention, and deci-
sion making. In this rcspccr, much applied ethnography seems locked in
the '(positivist" assumption that good knowledge will find good uses with-
out much effoft or commitment on the part of the investigator. IVlore criti-
cal analyses of the actual uses (and misuses) of ethnographic knowledge
arc badly needed.
Related to this is the ethical and moral stance of applied ethnography.
Although guided in a general way by a preference for advocating for those
who often have the least voice in public affairs, applied ethnographers
have not resolved [or themselves the vexing issue of the proper role for
their inquiries. Are they specialists in the production of knowledge \vhose
services should be available without discrimination, n.uch as physicians
and lawyers endeavor to respond equal!y to the needs of all potential cli-
ents? Or do they limit their participation to those clients they feel are most
deserving? Although it might be that this issue \\fill never be resolved
beyond
. the level of individual choices.. a higher level of discussion of the
issues would be helpful.
Another area in which we have made few advances is in the translation
of the ethnographic perspective to clinical modes of practice other than
research. Although there is precedence for such an interest, which I have
discussed above, our efforts to understand how ethnographic inquiry can
better inform the practices of individuals who enter careers related to the
human services, planning, and decision making fall short of the situation
in which we find ourselves. In anthropology, for example, the dominant
role of academic practice has begun to yield to a much broader sense of the
kinds of careers that are appropriate to the discipline. \"'('e need to pay
more attention to the \vays in which such a major and distinct mode of
inquiry as ethnography might prepare individuals for these new career
modes-increasing our appreciation for the possibility that the "method"
has salience beyond its more formal applications.
If I can spe~ulate as to where the field will be 10 years from noW, I
would certainly hope for improvements in the areas just mentioned. Be-
yond this, I would place my bets on (a) increased sophistication in researc,h
methods and techniques, and advances in integrating ethnographic inqul~
ries with other approaches to problem solving; and (b) in terms of research
directions, greater emphasis upon convincing decision makers and rc-
scmch cliems of the importance of paying heed to "indigenouslcnowledge
414
rl.ppllea ctnnograp11Y
systems," nor only as means to understanding more fuJly tll'::': needs of peo-
ple and their communities, but also as a vital source of potential solutions
to communities' problems. Too often, '(top-down" re5pon:sc5 in at[empt~
to solve social problems have served to erode if not destroy those very
sociaj and cultural resources and survival strategies that have helped keep
troubled communities alive.
+ References
Agar,lVI. H. (1973). Ripping and runl1ing: A formal ethnography of urball heroin
([ddie/s. New York: Seminar.
Agar, M. H. (1986a). Independents declared: The dilell1llws of illde/JclUlent tmch-
illg. Washington, DC: Smirhsonian Insriturion.
Agar, M. H. (1936b). S!Jeahillg of ethnography. Beverly Hills, CA: S:J.gc.
Agar, 1v1. H. (1994). The inrerculturaJ frame. International JOl!mal ollnleJ"wllllral
Relatio/1s) 18) 221-237.
Agar,lv1. H. (1996). The professional stranger: All iJlfomwl illtrodllctionto ctlmog-
ra/}hy. San Diego, CA: Academic Press,
Argyris, D. (1990). The ethnographic approach to intervention and fundamenwl
change. In C. Argyris, R. Putnam, & D. IV[, Smith (Eds.), Aetioll science
(pp. 158-189). San Francisco: lossey-Bass.
Barnett, H. G. (1956). Anthropology ill admillistratioll. Evanston, IL: Row,
Peterson.
Bartlett, P. F. (Ed.). (1980). Agriwltural decisio/1111ahing: Allthro/)ological cOlliribll-
tialls to ntml deuelopment. New York: Academic Press.
Bernard, H. R. (1994). Research methods in allthropology: Qualitative /md quanti-
tative approaches (2nd cd.). Walnut Creek, CA: AltaMira.
Bernard, H. R. (Ed.). (1998). Hmldbooh ofresearch methods ilt cultural anthropo/~
og}'. Walnut Creek, CA: Altalvlira.
Berrcman, G. D. (1991). Anthropological ethics in the 19805: A positive approach.
In C. Fluehr-lobban (Ed.), EtlJies and the professioll o( alltlnojJology: A dia-
logue for a new era (pp. 36-71). Philadelphia: University of Pennsylvania
Press.
Briggs, C. 1. (1986). Learnillg hall) to ask: A sociologllist a/ltnaisal o(the role of the
hlterl'iew in social science research. Cambridge: Cambridge Universiry Press.
Cernea, M. 1v1. (Ed.). (1985). Putting peo!Jle (irst: Sociologicaluariables in ruml
development. New 'I'orlc: Oxford University Press.
Chambers, E. (1985). Applied illlthrotJOlogy: A !Jmclical guide. Prospecr Heighrs,
IL: Waveland.
415
 ,.u.. , , 'V"'''> V' \...VE-U.. \...' " ~u h ' "'" ,...." ~"""E-' L" 0''.:;)
Chambers, E. (1986). The cultures o( science and policy. In \Y!. N. Dunn (Ed.), Po/~
ic), il/wlysis: Perspectiues, concepts, alld II1cthods (pp. 93-110). Greemvich,
CT: JA!.
Cl~"mk:l:';,n. (19HS). shortcut amI participatory J!lcrhocj:) for gaining S()Ci3] infor-
mation for proje-ct5. In h1. 1'1"1. CCfllca (El1.), I',,((iug people lirst: Sociological
uariables in ntml dcuelopmellt (pr. 515-537). New York: Oxford University
Press.
Chapple, E. D. (1941). Anrhropological engineering: Its use lO 3ClministralOrs.
AjJplied Anthro!Jology, 2(1), 23 .. 32.
Cochrane, G. (1979). The wltural a!Jpraisal of dcuclo/JllIcllt projects. New York:
Pr<leger.
DeWalt, B. R., & Pelto, E j. (Eds.). (1985). Micro alld macro lcvels o( alwlysis ill
'111thI'OIJOlogy: Isslfes ill theory and research. Boulder, CO: Wesrview.
Drake, H. 1\11. (1988). Mail/streaming antlnopolog)': Expel'imccs in gouerlllllel/t
elllploY/ltc/lt. \\'as hingron, DC: AlllericJtl Anthropoiogicli Association.
Escobar, A. (1991). Anthropology and the development encounter: The making
and lll8fketing of developmenr anrhropology. Al11erican Ethnologist, 18,
658~682.
Ferrerman, D. M. (Ed.). (1984). Ethnography ill edllcatiolli11 cualllatioll. Beverly
Hills, CA: Sage.
Fluchr~Lobban, C. (1994). Informed consent in anrhropological research: \Vc ,He
not exempt. [-fill/ta/l OrgallizaticJ1l, .'i3, 1-10.
Foster, G. M. (1969). Applied anthropology. Boston: Little, Brown.
Geertz, C. (1983). Loeal Iwowledge: rltrther essays ill illterpretil'c al1thropology.
New York: BaSIC Books.
Green,]. W. (1982). Cultural awareness ill the hilI/lim seruices. Englewood Cliffs,
N]: Prentice Hall.
Greenwood, D. J., & Levin, 1\11. (1998). 1l1trodf/ctiollto action research: Social re~
search for social challge. Thousand Oaks, CA: S<lge.
Guba, E. G., & Lincoln, Y. S. (1989). FOlfrth gelleration eUiJhwlioll. Newbury Park,
CA, Sage.
Hammersley, lVL 0992). What's WI'OJ1g luit;; etlJJ1ogmphy? Methodologic<11 cr/Jlora-
tiol/s. London: Routledge.
Harris, K. j., Jcrome, N. \\~, & Fawcett, S. B. (1997). Rapid assessment procedures:
1\ review and critique. HI/II/all O'gt1lrizatioll, S6, 375-378.
Heath, S. B. (1981). Questioning at home -and at schoo!: A comparative study. III
G. Spindler (Ed.), Doil1g the ethllography 0/ schooling. New )'otk: Holt,
Rinehan & Winston.
Holmberg, A. R. (1958). The research and developmcnt approach to the study o(
change. HI/mall Orgil1/i;;atirJII, 17, 12~16.
Joh<lnnsen, A.1\11. (1992). Applied anthropology and post-modernist erlmography.
HlI//h1ll Org{///i'::atiolt, 51, 71-S-I.
416
l.-Iv""~''-hl.- IV'M' l.-I\IMLJ ?'pplleo crnnograpny
Kcmpton, \X~, Boster, J. S., & Hanley,]. A. (1995). Envil'OJ/llle/ltai valrlcs ill AJIlcri-
CiI1l wltl/re. Cambridge: rvlIT Prcss.
Kirk, j., & Miller, 11,11. L. C19~U~). Reliability '71I<1 /JoliAil)' i11 '71Ur/itative rt:~wrcll,
Reverly l-lills, Ct\: ~asr.;.
Kluckhohn, C. (1947). Covert culture and administrative problems. Americall
Anthropologist, 45, 213-229.
Leighton, A. H. (1945). The governillg of mel/. Princeton, N]: Princeton University
Press.
Lewin, K. (1948). Resolving social conflicts. New York: Harper &: Row.
IVIalinowski, B. (1929). Practical anthropology. Africa, 2, 23 .. 38.
Manning, l~ K. (1977). Police wod,-: The social orgmli;;atiol1 ofpolic!lIg. Cambridge:
IvlIT Press.
rvlomgomery, E., & Bennett, J. \\Y. (1979). Anthropological studies of food and
nutrition: The 1940s and the 1970s.II1 \X~ Goldschmidt (Ed.), The /lses 0/
m1fhl'O!Joiogy (pp. 124-144). \'\!ashington, DC: Americun Anthropological
Association.
Pelto, P.]., & Pelto, G. H. (1978 ).AlltlJropologicall'escarch: The structure o(ilrqlliry
(2nd ed.). New York: Cambridge University Press.
Poggie,J.].,Jr., De\Valt, B. R., & Dressler, \Y/. \Y/. (Eds.). (1992).Allthrb!Joiogical re ..
se'1rch: Process alld applicatioll. Albany: State University of N~w York Press.
Rigby, r., & Sevareid, P. (1992). Lawyers, anthropologists, and the knowledge of
facts. In R. F. Kandel (Ed.), Double vision: Anthropologists at 1m! (pp. 5~21).
Washington, DC: American Anthropological Association.
Rosaldo, R. (1989). Cultme and tmth: The remaking of social mwl)'sis. Boston:
Beacon.
Ryan, K. E., Greene, ]. c., LincDln, Y. S., Nbthison, S., & Ivlertel1s, D. (1998).
Advamages and challenges of using inclusive evaluation approaches in evalu-
ation practice. Amcriwl/ jOllmal of Evaluation, 19, 101-122.
Schein, E. H. (1987). The c1illical perspective in fieldwork. Newbury Park, CA:
Sage.
Schensul,].]. (1985). Systems consistency in field research, dissemination, and
social change. American Behavioral Scientist, 29, 186-104.
Scrimshaw, S. C. _M. (1985). Bringing the period down: Government and squatter
settlement confront induced abortion in Ecuador. In B. R. De\\!alt & I~ J.
Pelto (Eds.), _Micro alld lIIacro leuels of analysis in anthropology: Issues ill
theor)' and research (pp. 121-146). Boulder, CO: W1estview.
Shim kin, D. B., & Golde, E (Eds.). (1983). Clinical anthropolo!:.')': A /lew approach
to AmericaJ/ health. Lanham, MD: University Press of America.
Spicer, E. H. (1952). HUll/an problems in tec!mological chal/ge. New York: Jobn
Wiley.
417
 IVll: I nvu.) vr \..VI_LC\"III'II.;J AI'lU AI'lALT LlI'll.;J CIYWIK1\..AL 1V1,",,1 CKI,""L.)

Spicer, E. H. (1979). Anthropologists ;md the War Relocation Authority. In

W Goldschmidt (Ed.), The uses o( unthrotJO/ogy (pp. 217-237). \'\!ashingron,
DC: American Anthropological Association.
f:pr'-ldlcy, J. P. (1975). 1'01/ owe yrJIITsdfa dmnk; All ethnography of urbal1110mads.
Bo,;ton: Little, Brown.
Spradley,]. P. (1979). The ethnographic interview. New York: Holt, Rinehart &
PART 11-
\"X!insron.
Stein, H. F. (1982). The ethnographic mode of reaching clinical behavioral science.
In N. j. Chrisman & T. \X'. IvIaretzki (Eels.), Clioieall)' applied ullthl'Opolog)':
Anthropologists ill health science settings (pp. 61-82). Bosron: D. Reidel.
van \X!illigan, j., & finan, T. L. (Eds.). (1991). Sotlndings: Rapid and reliable re-
search methods (or practicing onthrotJologists. Washington, DC: American The Art
Amhropological Association.
van \Villigan, ]., Rylka-Bauer, B., & IvIeElroy, A. (Eels.). (1989). Mahing our re-
search useful: Case sttldies in the Iltilization o( anthropologicBf JUlOwledge.
and Practices
Boulder, CO: Wesrvie'w.
Tax, S. (1958). The Fox ProjeC[. Human 0lgallizatiol1, 17, 17-19. of Interpretation,)
- -
Trend, 11,11. G. (1978). On the rcconcili;:ltion of qualitative and quantitative analysis:
A case study. HIIIIWIt Organization, 37, 345-354.
\X1arren.- D. 1\11., S'likkerveer, L. -J., & Brokensha, D. (Eds.). (1995). The wltllmf
Evaluation 5 and
dimension o( dcuefo!JllIent: Indigenous knowledge systems. London: Inter-
mediate Technology. Representation
Wax,lv1. L. (1995). Informed consent in applied research: A comment. Homan
Organization, 54, 330-331.
Whyte, W. F. (with Whyte, K. K.). (1984). Learning (rom the (ield: A gtlide frolll
experience. Beverly Hills, CA: Sage.
\'\Tolcott, H. F. (1994). 7hllls(onning qtlalitatiue data: Description, analysis, and
illtel1m:latiolt. Thollsand Oaks, CA: Sage. In conventional terms, Part II of this volume signals the terminal phase of
\\tolcott, H. F. (1995). Making a study "more ethnographic." In J. Van l\'laanen qualitative inquiry. The researcher or evaluator now assesses, analyzes,
(Ed.), Re/JTesentatioJl in ethnography (pp. 44-72). Thousand Oaks, CA: Sage. and interprets the empirical matcrials that have been collccted. This pro-
Young,]. (1998). SfAA president's letter. Society (or Anthropology Nelus!etter, 9(4),
cess, conventionally conceived, involves a set of analytic procedures that
1-2.
produce interpretations, which are then integrated into a theory or put
forward as a set of policy recommendations. The resulting interpretations
are assessed in terms of a set of criteria, from the positivist or postpositivist
tradition, that include validity, reliability, and objectivity. Those interpre-
tations that stand up to scrutiny arc put forward as the findings of the
research.
The contributors to Parr II explore the art, practices, and politics of
interpretation, evaluation, and representation. In so doing, they return
to the themes raised by the authors in Part I of Volume i-asking, that is,
HolV call we use the discourses of qualitative research to help create and
418 419
 I.ULLtl.llI'll:l Ai'll) II'l I tl-<I-'I<t liNG VUJ:I,U IAIIVI: MAl tl~IAL.::>
imagille a (ree democratic society? It is understood tbat the processes of
analysis, evaluation, and interpretation arc neither terminal nor mechani-
caL They are like a dance, to invoke the metaphor used by Volerie JmecicL
in Chapter 12 of Volume 2. This dance is informed at every step of the way
by a commitment to this civic agenda. The processes cInt define rhe prac-
tices of interpretation and representation are always ongoing, emergent,
unpredictable, and unfinished.
\'{ie begin by assessing a number of criteria that have been used tradi-
tionally (as weI[ as recemly) to judge the adequacy of qualitative research.
These criteria flmv from the major p3radigms no\\' operating in this field
(in Chapter 12,] ohn Smith and Deborah Deemer explore these criteria in
detail).
'f Criteria for Evaluating Qualitative Research
Smith and Deemer remind us that we live in an age of relativism. They note
that in the social sciences today there is no longer a Gocl's-eye view that
guarantees absolute methodological certainty; to assert such is to court
embarrassment_Indeed, as Lincoln and Guba discuss in detJil in Volume 1,
Chapter 6, there is considerable debate ovcr wh~]t constitLHcs good inter-
pretation in qualitative research. Nonetheless, there seems to be an emerg-
ing consensus that all inquiry reflects the standpoint of the inquirer, that 3It
observation is theory laden, 3nd that there is no possibility of theory-free
knmvledge. \"X7e can no longer think of ourselves as neutn!l spectators of the
social world.
Consequently, as Smith and Deemer observe, in the current discourses
of qualitative inquiry it is no longer possible to speak in terms of a foun-
dationJI epistemology and a direct ontological realism. No method is a
neutral tool of inquiry, and hence the notion of procedural objectivity can~
not be sustained. The days of naYve realism and na-ive positivism are over.
In their place stand critical and historical realism and various versions of
relativism. The criteria for evaluating research Jre nmv relative.
Extending Smith and Deemer's argument, there are rhree basic posi-
tions on the issue of evaluative criteria: foundational, quasi-foundational,
and non foundational. There are still those who think in terms of a fouw
dational epistemology. They would apply the same criteria to qualitative
research as are employed in quantitative inquiry, contending that is there
is nothing special about qualitative research that demands a special set of
420
r-arr II: 1 ne Art ana r-ractlces ot Interpretation, tvafuation, and f(epresentation
evaluative criteria. As we indicate in our introduction to Part II of Volume
], the positivist and postpositivist paradigms apply four standard criteria
to disciplined inquiry: internal validity, external v;J1idity, reliability, and
objectivity. The use of these criteria, or their varianrs, is consistent with the
(oundational position.
In contrast, quasi-(olflldationalists approach the criteria issue from the
standpoint of a nonna"ive neorealism or subtle realism. They comend that
the discussion of criteria must take place within the conto:r of an ontologi-
cal neorealism and a constructivist epistemology. They believe in a real
world thar is independent of our fallible knowledge of it. Thcir construc-
tivism commits them to the position that there can be no theory-free
knowledge.
Proponents of the quasi-foundational position argue for the develop-
ment of a set of criteria unique to qualitative rcsearch. Hammersley (] 992,
p. 64; also 1995, p. 18; see also Wolcott, 1999, p. 194) is a leading propo-
nent of this position. He wants to maintain the correspondence theory of
truth while suggesting that researchers assess a work in terms of its ability
(a) to generate generic/formal theory; (b) to be cmpirically grounded and
scientifically credible; (c) to produce findings that can be generalized, or
transferred to otber settings; and (d) to be internally reflexive in terms of
taking account of the effects of the researcher and the research strategy on
the findings that have been produced.
Hammersley reduces his criteria to three essential terms; plausibility (Is
a claim plausible?), credibility (Is the claim based on credible cvidence?),
and relevance (\"'\1hElt is the claim's relevance for knowledge about the
world?). Of course, these criteria require social judgments. They cannot
be assessed in terms of any set of external or foundational criteria. Their
meanings are arrived at through consensus and discussion in the scientific
community. Smith and Deemer note that within Hammersley's model
there is no satisfactory method for resolving this issue of how to evaluate
an empirical claim.
For the l1on(olllldatioJlalists, relativism is not an issllc. They accept the
argument that there is no theory-free knowledge. As Smith and Deemer
note, relativism, or uncertainty, is the inevitable consequence of the fact
that as human beings \ve have finite knowledge of ourselves and the world
We live in. Nonfoundationalists contend that the injunction to pursue
knowledge cannot be given epistemologically; rather, the injunction is
moral and political.
421

Accordingly, the criteria for evaluating qualitative work arc also moral
and fitted to the pragmatic, ethical, and pol itica! contingencies of concrete
situations. Good or bad inquiry in any Biven context is assessed in terms of
C!llch ['l'itt't'i:t::l.'" lcho,,"c "utl;ned by Gn:cn ...vood and Levin (Volume 1, Chap-
ter 3), rine, \Veis, \Veseen, and \Vong (Volume 1, Chapter 4), Christians
(Volume 1, Chapter 5), Lincoln and Guba (Volume 1, Chapter 6),
Schwandt (Volume 1, Chapter 7), Kemmis and IvlcTaggarr (Volume 2,
Chapter 11), Angrosino and Perez (Chapter 3, this volume), and l'dadriz
(Chapter 10, this volume). These are the criteria that flow from a feminist,
cornmllnitarian moral ethic of empowerment, community, and moral soli-
darity. Returning to Christians, this moral ethic calls for rcse8rch rooted in
the concepts of care, shated governance, neighborliness, love, and kind-
ness. Further, this work should provide the fo undations for social criticism
and social action.
" The Practices and Politics of Interpretation
In Chapter 13, Norman Denzin proposes to reengage the methods and
promises of qualitative inquiry as forms of radical democratic practice. I-Ie
explores new and older writing forms, working back ancl forth among the
new, civic, and intimate journalism, calls for performance-based ethnogra-
phies, and variations on a Chicano/a and African American aesthetic. He
explores the relationships between these practices and critical race theory.
Like Ladson-Billings (Volume 1, Chapter 9), Denzin seeks to align critical
theory with the posrstrucrural turn in ethnography.
" Writing as Inquiry
\X1riters interpret as they write, so writing is a form of inquiry, a "way of mak-
ing sense of the world. In Chapter 14, Laurel Richardson explores new
writing and interpretive styles that fol low from the narrative, literary turn
in the social sciences. She calls these different forms of writing CAP (cre~
ative analytic practices) ethnography. These new forms include autO-
ethnography, fiction stories, poetry, drama, performance texts, polyvocal
texts, readers' theater, responsive readings, aphorisms, comedy Jnd satire,
visual presentations, conversation, layered accounts, writing-stories, and
mi..xed genres. Richardson discusses in detail Doe class of experimcn[31
422
run II: I ne r-\rf ana ,..racnces or interpretation, tvaluation, and k:epresentation
genre that she calls evocative representations. \Vorle in this genre includes
narratives of the self, microprocess writing-stories, ethnographic fictional
repre.~ent:1tions~ poetic representations, ethnographic dramas, and mixed
genre",.
The cry~tal is a central image in Richardson's text; she contrasts the
crystal with the triangle. Traditional postpositivist research has relied
upon triangulation, including thc use of multiple methods, as a method of
validation. This model implies a fixed point of reference that can be trian-
gulated. NIixed-genre texts, in contrast, do not triangulate. The central
image is the crystal, which "combines symmetry <md substance with an
infinite variety of shapes, substances, transmutations, ... and angles of
approach." Crystals are prisms that reflect and refract, creating ever-
changing images and pictures of reality. Crystallization deconstfllcrs the
traditional idea of validity, ror now there can be no single, or triangulatecI,
truth.
Richardson offers five criteria for evaluating CAP ethnography: sub-
stantive contribution, aesthetic merit, reflexivity, impactfulness, and 8bi!-
ity [Q evoke lived expcrience. She concludes with a list of writing practices,
\vays of using writing as a method of knowing.
Anthropological Poetics
Anthropologists have been writing experimental, literary, and poetic
ethnographic texts for at least 40 years. In Chapter 15, Ivan Brady analyzcs
three subcategories of this writing: ethnopoetics, literary anthropology,
and anthropological poetry. Ethnopoetics focuses on the poetic produc-
tions, proverbs, riddles, laments, prayers, and oral performances of indige-
nous verbal artists. Literary anthropology includes the fiction in realistic
ethnographies, ethnographic novels, and the many different CAP eth-
nographies discussed by Richardson. Anthropological poetry is just that,
poetry written by anthropologists.
In the literary, poetic form, anthropologists enact a moral aesthetic, an
aesthetic that allows them to say things they could not otherwise say. In so
doing, they push the boundaries of artful ethnographic discourse. Thus
3re the boundaries between the humanities and the human sciences
blurred. In this blurring, our mowl sensibilities are enlivened. \"Xie are able
to imagine new ways of being ourselves in this bewilderingly complex
World called the postmodern.
423
 '-...Vl.l.(:'-...III'l'.;l MI'iLJ Wi I (:r\rru:: 111'1\:1 '.!UMl.l 11-\1 lYE:: !VIHI (:1'1.11-\1...) t-'an11: I ne An ana t-'raalces ot Interpretation, tva/uation, and f<epresentation

<+ The Practice and Promise ~ Conclusion

of OualitativeProgram Evaluation
rrubranl CyalUiHivn, vf \;VLlDC, iCi d lllajvr Ji~c vf lJUdliwlivc rC5carch
(the contributions to this Handbook. by Greenwood & Levin, Stake,
Kemmis & iv1cTaggart, 1vEller & Crabtree, and Chambers help to establish
this fact), Evaluators are interpreters. Their texts tell stories. These stories,
Jennifer Greene argues in Chapter 16, are inherently moral ;md political.
She examines four contemporary genres of evaluation work: postpositiv-
ism, utilitarian pragmatism, interpretivism/constructivislll, and critical
social science. She reviews the work of the major figures in this area,
including Guba and Lincoln, Schwandt, Greenwood and Levin, and Stake.
She presents evaluation as a narrative craft, asking, How good is this story?
She calls for a morally engaged, constructivist, qualitative genre of evalua-
tion practice.
~ Influencing Policy
With Oualitative Research
The readings in Part II affirm our position that qualitative [esearch has
cume of age.lVlultiple discourses 1100V surround topics that in earlier histor-
ical moments where contained within the broad grasp of the positivist and
postpositivist epistemologies. There are nmv many ways to write, read,
assess, evaluate, and apply qualitative research texts. This complex field
invites reflexive appraisal, which is the topic of Part II of this volume.
~ References
Hammersley,1v1. (1992). \Vhat's wrollg with etimogratJhy? Methodological eX!Jlora-
tiolls. London: Routledge.
Hammersley, IvI. (1995). The politics ofsocial research. Thollsand Oaks, CA: Sage.
Ryan, K. E. (1995). Evaluation ethics and issues of social justice: Contributions
from female moral rhinking. Tn N. K. Denzin (Ed.), Studies in symbolic inter-
action: A research anllual (Vol. 19, pp. 143-151). Greenwich, CT: ]AJ.
Wolcott, H. F. (1999). Ethnography: A way ofseeilIg. \\Talnur Creek, CA: AlraMira.

Qualitative researchers can influence social policy, and in Chapter 17,

Ray Rist shows how they do this. He shows how qualitative research has
pivotal relevance in each stage of the policy cycle, from the problem for-
mulation to the implementation and accountability stages. Qualitative
researchers can isolate target populations, show the immediate effects of
certain programs on such groups, and isolate the constraints thar operate
against policy changes in such settings. Unlike othcr models that relegate
qualitative research to a secondary posirion in this process, Rist's model
gives this research a central part in the shaping of social policy.
Greene's and Rise's contributions here connect back to the applied re~
search traditions identified by Greenwood and Levin, Stake, Chambers,
and Kemmis and IV1cTaggart. They show how the interpretive, qualitative
text can morally empower citizens and shape government policies. At the
same time, thcy chart new lines of action for evaluators who arc them-
selves part of the ruling apparatuses of society (Ryan, 1995). Rist and
Greene reclaim a new moral authority for the evaluator. This claim can
also empower the qualitative researcher who does not engage in direct
program evaluation.

424 425
 12
The Problem
of Criteria in the
Age of Relativism

John 1<' Smith and Deborah 1<' Deemer

In the introduction to the third volume in his series on the nature of

inquiry, Hazelrigg (1995) talks about "issues of relativism and its
contrary" (p. IX). He then goes on to say about the latter, '\vhich is what,
now that we no longer speak the name of absolutism with om embarrass-
ment?" (p. IX). His point, which has been noted by many others, such as
Bernstein (1983), is rharrhere no longer is a "contrary" ro place in opposi-
tion to relativism. 1 this is the case, rhen it is clear that in this age when the
God's-eye view is no longer a realizable hope, relativism, in some form or
another, is a consequence that is inescapable (Hesse, 1980, p. xiv). Thus
any discussion of criteria for judging social and educational inquiry must
confront the issue of relativism-but, of course, as we shall see, how rela-
tiVism is understood or conceptualized matters greatly.
\YJe begin this chapter with a brief historical review of how ,ve have
arrived at this point where we must accept relativism, nor only as the cru~
ciaI fearure in any discussion of criteria, bur also as the central condition of
our very being in the world. At the core of this journey srands the realiza-
tion, which has been expressed in innumerable ways, that all observation
is theory laden or that there is no possibility of theory-free observation or

427
 "" ,1." ... ,",... 1- , ..... , 'VI', ............V..... ' 'V", .....,,'-' " ... , ""'-F"", ,..... , 'V',
knowledge. Although this idea well predates the work of people such as
Hanson (1958) and Kuhn (1962), one can argue it is their writings that
have forced upon us the recognition that as knowing subjects we are inri-
or of any claim we make to knowledge. \X1ith the end of the possibility that
we could think of ourselves as neutral spectators at the game of knowl-
edge, the central problem that has preoccupied the thoughts of numerous
researchers for the past few decades is that of "Now \-vhat are \-vc going to
do with us?" This is a preoccupation that bas been expressed with great
frequency and in all manner of ways: Gallagher's (1995) revisiting of her
role in her dissertation research, Sparkes's (1993) concerns over reciproc~
ity, various aspects ofWolcotr's (1995) discussion of the art of fieldwork,
Punch's (1994) discussion of the politics and ethics of field inquiry, and on
and on. The discussions that focus on rhe position or role of the inquirer
seemingly have become endless.
Following this brief historical review, we examine some of the ways the
issue of criteria has been addressed in this age of relativism. One of the
most prominent of these is what we call the quas;-(otl1uiat;ollalist re-
sponse. At the center of the quasi-foundationa!ist project is the desire to
deny the relativism that must accompany the realization that there can be
no theory-free knowledge or observation. This group has dravo/11 much of
its intellectual sustenance from various neorealist philosophers (neoreahst
is a general term used to distinguish this group from "what are often
referred to, especially by the neorealists themselves, as na"ive, direct, or
unsophisticated realists). From at least the time of Popper's (1959) rejec-
tion of positivism in favor of a sophisticated or scientific realism, the
neorealists have attempted to thread a line between their acceptance of a
constructivist epistemology on the one side and, on the other, their adop-
tion of a neorealist ontology. The latter commits them to the position that
there is a reality out there that, because it can be known, at least in princi-
ple, as it really is, acts as a constraint on any claim to knowledge. The for-
mer announces their recognition that there can be no theory-free knowl-
edge or theory-free observation.
In the second half of the chapter we address how our ideas about crite-
ria must change when we realize that the epistemological project is over
and relativism must be accepted. This examination begins with the very
important point that relativism need not and must not be seen in terms of
"anything goes." Rather, relativism is nothing more or less than our condi-
tion in the world-it announces that as human beings \ve are, and can be
428
,,,,= "WUlo:;::lI, WI <""/ILt'flU If! [fie ;.;ge or r;:elonVlsm
nothing but, finite. This understanding of relativism is then coupled with a
concern that we must change our imageries and metaphors from those of
discovery and finding to those of constructing and making. Based on this
Ir,,"sili O rl, t:hc cennul bsw.::~ bCo;,;onlC 111V"c: v[ hvw UV V'll: a:i indiViduals
make judgments, which we all must, and the extent to which we allow our
judgments to move into a public space, which to a certain degree \ve also
must, to engage the judgments of others.
As is always the case, before we begin at least two caveats are necessary.
Firsr, rhe issues we address are complex. Given the space available and,
mllch more important, our limited capacities) there is no way vve can do
them full justice. Second, contrary to what is generally the case for others
who have addressed this topic, we offer no particular criteria and instruc-
tions on ho\'v they might be applied. Our intent is to discuss the limits and
possibilities of judgment-both individual and within social spaces-in
this age of relativism.
<> Historical Background
The long-standing problem empiricists constructed for themsdves was
how to reconnect that which they had separated-the knowing subject
from the object of knowing. After having established a dualism of subject
and object, empiricists of all varieties were compelled to spend consider-
able time and energy attempting to rework or get around this duality in a
way that would allow them to claim the knowing subject could have access
to reality in such a way that it could be depicted accurately. Put differently,
empiricists had developed a situation in vvhich the referent point for judg~
ing claims to knowledge was, and could only be, that which was outside)
and independently so, of the knower. Knowledge was a matter of accurate
representation, and thus was born the correspondence theory of truth. The
solution of choice to make good on this correspondence theory \vas, as has
long been announced in our research textbooks, a methodical one. The
idea was quite straightforward: If the proper procedures or methods were
properly applied, the subjectivities of the knO\ving subject would be con-
tained and the knower would thereby gain an accurate, objective depiction
of reality. This is what Kerlinger (1979) meant when he said rhat "the pro-
cedures of science are objective-not the scientists. Scientists, like all men
and women are opinionated, dogmatic, ideological ... that is the very
429
 ""'-,,, ",-,M"'-"', '-.M'-'-'M"'-"',M" ..... ,,'-, ",--.',-"'M"'-'"
reason for insisting on procedural objectivity; to get the whole business
outside of ourselves" (p. 264).
Given this line of thought, a judgmenr about the quality of research, at
procedures had been properly applied. The methods were seen as neutral
and hence established that neutral arena within which judgments could be
made about the goodness or badness of any individual research. Of course,
there still remained a second level of interest-that of the significance of
any particular study. It has long been recognized by the proceduralists that
even the most appropriately conducted study could still be judged insigni f-
icant because it focused on a trivial problem. 1n any event, for empiricists
the issue can be summarized as follows: Because method allmvecl for
undistorted contact with reality, it was the basis for distinguishing good
from bad inquiry bur not important from unimportant research.
The attempts by empiricists over the past few hundred years to recon-
nect subject and object-from Locke's "blank slate" to the ostensive refer-
ence of the logical empiricists-need not be discllssed. These moves are
quite well-knO\vn and, more important, such a discussion would take us
far afield from the main topic at hand (for a brief recounting of this history,
see Smith, 1993). A few brief comments, however, must be made with
regare! to \vhat has happened over the past 30 or 40 years as traditional
empiricism has unraveled and we have witnessed the demise of the
methodical solution to the problem of criteria. The problems associated
with the dualism of subject and object were apparent to many philoso-
phers prior to the middle of the 20th century. However, within Anglo-
American philosophical circles, a good case can be made that it was
Hanson (1958) and then, most definitely, Kuhn (1962) who brought
the problematic associated with the dualism of subject and object to the
forefront.
At the core of Hanson's (1958) arguments was the now seemingly obvi-
ous point that "the theory, hypothesis, framework, or background knowl-
edge held by an investigator call strongly influence what is observed"
(p. 7). A few shorr years!ater Kuhn (1962) followed up on this line of rea-
s~i1ing \vith his talk of incommensurable paradigms, paradigm shifts, that
all knO\'vledge is framework dependent, and so on. By the mid-to-late
19805, the work of numerous people made it apparent that any claim
of or hope for "theory-free knowledge" was untenable on all counts. To
citt just a few examples, there were Putnam's (1981) arguments about
"no God's-cye point of view," Nagel's (1986) claim that there \vas no
430
I lie r, UUlt"lrl UI .....n[eno In rne Age or KeionVIsn1
possibility of a "view from nowhere," Bernstein's (1983) discussion of no
"Archimedean point," Taylor's (1971) undermining of the hope for access
to "brute data," Gadamer's (1995) discussion of "effective history," and
Goodmarl'5 (1978) elaboration of the ways we make the world.
All of these arguments, and others, combinecl with another series of
arguments that focused directly on, and cast grave doubt upon, the claim
that method itself was neutral and could thereby be the repository of pro-
cedural objectivity. This certainly was one of the main messages that could
be taken from the work of Hesse (1980), Giddens (1976), IVlacKenzie
(1981), Cherryholmes (1988), Smirh (1985), and so on. In rhe end, the
result of all of this intellectual ferment was the elaboration of a number of
points of great consequence for any discussion of criteria: There is nO pos-
sibility of theory-free observation and knowledge, the duality of subject
and object is untenable, no special epistemic privilege can be attached to
any particular method or set of methods, and we cannot have the kind of
objective access to an external, extralinguistic referent that would allO\v us
to adjudicate from among different knowledge claims.
The legacy of all of this, of course, is that we no longcr can talk in terms
of a foundational epistcmology and a direct ontological realism. To the
contrary, if the discussions of the past few decades have made anything
clear to us, it is that we cannot adopt a God's-eye point of vic\\'; all wc can
have arc "the various points of vie,\' of actual pcrsons reflecting various
interests and purposes that their descriptions and theories subserve"
(Putnam, 1981, p. 50). Under these conditions, with the demise of empiri M
cism and the methodical stance, any discussion of criteria must come to
terms, in one form or another or in one way or another, with the issue of
relativism.
<l> The Quasi-Foundationalist Response
The quasi-foundationaJists, who have attempted to address the criteria
issue from the perspective of some form of nonna'ive realism) have had to
comend with a very interesting problem. Any elaboration of criteria must
take place within the context of their commitment to ontological realism
on the one side and, on the other, their realization that they are obligated to
accept a constructivist epistemology. The former announces a commitment
to the proposition that there is a real world out there independent of our
interest in, or knowledge of, it. The latter announces 3 commitment to the
431

proposition that we can never know if we have depicted this real world as it
really is. In shorr, nonn~live realists or neorealists "assert a belief in a real
world independent of our knowledge while also making it clear that our
I..::no\.v!cdgc n( tl~;" n~<'_hH::_ogn;t:~v,--, "vn,ld ;s <-lu;t,--, L~11;ble" (L'--'8'Y7 :l98.-I J
1'.918).
In light of these dual commitments, many neorealist philosophers have
felt compelled to express their positions with the use of the term relativ-
ism. Bhaskar (1983), for example, says that his transcendental realism pre-
supposes "ontological realism and epistemological relativism" (p. 259).
Similarly, House (1991) notes that the world can be "knmvn only under
particular descriptions and is, in that limited sense, epistemologically rela-
tive" (p. 5). Even more directly, lvlanicas (1987) has written that "knowl-
edge is a social product and the standards of inquiry are generated in the
course of inquiry. This is a relativism. But it is not irrationalism, because it
presupposes a realism" (p. 261).
There is no question, then, that neorealist acceptance of epistemo-
logical constructivism entails the realization thar there can be no "God's-
eye view" in that any claim to knowledge must take into account the per-
spective of the person making the claim. lvIaxweli (1992) notes quite
clearly the key aspects of the neorea!ist response to the consequences of
this realization. First, he recognizes that "as observers and interpreters ...
\ve cannot step outside of our own experience to obtain some observer-
independent account of what we experience. Thus, it is always possible for
there to be different, equally valid accounts from different perspectives"
(p. 283). That said, however, it is also equally clear that he, like the other
neD realists, does not accept a situation in which "all possible accounts ...
3re equally useful, credible, or legitimate" (p. 283). Thus "understanding
is relative," there is no account that can stand "independent of any partic-
ular perspective" (p. 284), but this does not mean that our only option is to
descend into a relativism of all-accounrs-are-equal or "anything goes."
Finally, in that the neorealists acknowledge that the methodical stallce of
their empiricist predecessors is not possible, the way to avoid relativism is
to call upon a "realist conception of validity that sees the validity of an ac-
count as inherent, not in the procedures llsed to produce and validate it,
but in its relationship to those things that it is intended to be an account
of" (I'. 281).
To examine how well they have done with this quasi-foundarionalist
project, with its obvious truth-as~correspondence overtones, we turn
to exemplars-beginning with a particular emphasis on the \vork of
432
.. ~~.~"' ........,,'-'--,'u ", '-'"" r->';}o::: "', ,,0:::''-''-'"''''''
Hammersley, in that he has presented one of the clearest and most sophis-
ticated attempts to develop criteria from a neorealist perspective. In one
very well arrencled piece of \vork, Hammersley (1990) begins with the
assertion that "\.ve do not need to "lbanJon the concept of truth as corre-
spondence to an independent reality. \YJe can retain this concept of truth by
adopting a more subtle realism" (p. 61). Following this assertion that we
can retain the empiricist concept of truth but must reject nai"ve interpreta-
tions of it, he then attempts to elaborate criteria in order to prevent a slide
into what is, for bim, the void of relativism. These are criteria that must be
worked out in light of a constructivist epistemology and a realist, however
subtle, ontology. Hammersley frames this task in terIllS of how researchers
can validate their accounts and argues that the only basis for validation
is to engage in judgments of the "likelihood of error" (p. 61). This is a posi-
tion that has distinct Popperian overtones and, 8ccordingly, shares in the
problems attendant to Popper's account of falsification. We return to this
point shortly.
H8mmersley (1990) proposes tlut the HI/O key elements of validity are
plausibility and credibility. The former issue is one of whether or not a
claim seems plausible; "that is, whether we judge it as likely to be true
given our existing knowledge" (p. 61). He adds that some claims are so
plausible we can immediately accept them at face value, where8s others
require the presentation of evidence. In the latter instance, ;;1 judgment
about the credibility of a claim must be taken "given the nature of the phe-
nomena concerned, the circumstances of the research" (p. 61L and so on.
In both instances, then, Hammersley recognizes that when a claim has nei-
ther face plausibility nor face credibility, evidence is required. However,
he further recognizes that the particular evidence presemed by a re-
searcher in support of the plausibility and credibility of a study must itsdf
be assessed as to its own plausibility and credibility. And, as he continues,
"we may require further evidence to support that evidence, which we shall
judge in terms of plausibility and credibility" (I'. 62).
It is clear Hammersley fully ackuowledges tbat plausibility and credibil-
ity arc social judgments, and, not surprisingly, his attempt to establish cri-
teria has become entangled in an infinite regress-if not a hermeneutic
circle. At this point, it would seem prudent to call upon ontological real-
ism and make it do some serious work-in particular, the work of making
COntact with reality in such a way as to blunt the infinite regress of social
judgments or, to put it differently, to blunt the relativism that would seem
to lie at tbe end of it all.
433

This is not the case, however, and the notion of correspondence and
realism, no matter how subtle or sophisticated, plays no role of conse-
quence for the balance of his discussion. Hammersley's (1990) movement
:J.W3V {rom nC>Orf,,,!:qm lWf';:nq wilen he f>3YS tll:J.t "r>I:1u_,,;l.. ;/;ty :1nd ct"('cl;l.. ;l~
ity form a relatively weak basis for judging claims, compared to the idea
that we can asseSs claims directly according to their correspondence with
reality or by relying on some body of knowledge whose validity is certain"
(p. 62). He continues this movement when he reinvokes a construcrivist
epistemology and thereby acknowledges that judgments \vill not "neces-
sarily be consensual, since there may be different views about what is plau-
sible and credible" (p. 62).
Ultimately, Hammersley's line of argument ends in a discussion of the
norms that should govern discourse among members of a scientific com-
munity. He lists what he considers to be three key features of this scientific
community. First, all findings should be subject to "communal assess-
ment," with an effort to "resolve disagreements by seeking common
grounds ... and trying to work back to a resolution of the dispute relying
only on what is accepted as valid by all disputants" (p. 63). Second, al1
involved should not only attempt to persuade, bur be willing to be per-
suaded and to change their views. Finally, the community is open to those
who are "willing to operate on the basis of the first tvvo rules" (p. 63). In
the end, then, judgments about plausibility and credibility are social judg-
ments, and they should be made based upon as free and open a dialogue
and discussion as possible. There is nothing in Hammersley's disclls-
sian that shows us how we can make contact with an external referent-a
reality-that would give the criteria of plausibility and credibility a stand-
ing beyond the social process and the inevitable relativism that lurks
behind this process.
A similar point can be made about the well-known work of 1vlanicas
and Secord (1983). They admit that "knowledge is a social and historical
product" (p. 401), there is "no preinterpreted 'given' and the test of trllth
cannot be 'correspondence.' Epistemologically, there can be nothing
known to which our ideas (sentences, theories) can correspond" (p. 401).
But, if that is so, then, as Leary (1984) notes, "Exactly how 1VIanic3s and
Secord propose to link experiences-which they admit to be culturally
and historically mediated-with reality indepenclent from experience is
an issue I shall leave for them to clarify" (p. 918). They do not do well vvith
this task, for as Leary further notes, their best response is the negative
assertion that there is no theory-free observation, which "does not elimi~
434
,.~ '-''-' ..... " , '-', ' - ' o<C' IV '" "'0;;: r"l!:lC VI "CIVlIVI.;o"1
narc the possibility of objectivity, construed here as \v:uranred assert-
ibility" (p. 410). Bm, of course, the problem they face here is the same one
Hammersley (1990) faces; the warrants one brings to judgments arc them~
<;<.:Ivc,o; ,o;"c;",1I)' ,~ncl hi"lOl-icOlll y co"Jirioned---,-,,; '-'("c die Wi.llTo:Illta rJldl Wilr-
ram the warrants, and on and 011. In both of these instances, and generally
for neorealists, any attempt to elaborate on the issue of criteria in a way
that gives criteria force beyond socially and hisrorically condirioned judg-
ments must involve an explanation of exactly hO\v their ontological real-
ism does some serious work and is not merely an assertion.
A further example of the same situation occurs in 1Vlaxwell's (1992) dis-
cussion of five aspects of validity: descriptive, interpretive, theoretical,
generalizability, and evaluative. \Ve focus only on his discussion of descrip-
tive validity, as this is the one that seems to make the most use of the realist
commirment. lvlaxwell argues that this is rhe first concern for qualitative
inquirers in that it focuses on the factual accuracy of an account-that
researchers arc "not making up or distorting the things they saw and
heard" (p. 285). On the face of it, this concern seems to be very un-
problematic, especially if one confines description to the level of did the
researcher actually speak with the person she said she did or did the re-
searcher actually enter the classroom to observe the goings on. However,
as soon as one moves that short distance to the level of "description" on
the order of did or did not the student "throw an eraser on a specific occa-
sion" (p. 286) or did she raise her arm or did she thrust her arm in the air,
the prospects for a viable concept of descriptive validity disappear. Put dif-
ferently, as soon as we move past the most minimal level of the depiction of
the movement of physical bodies in time and space, the coercive or con-
straining quality that is desired from the real or from reality itself has been
left behind long ago in the name of the need for context.
1vraxwell (1992) recognizes this issue in that he further notes even
descriptive validity is "by no means independent of theory ... even if this
theory is implicit or common sense" (p. 287). This understanding then
leads to the comment that although there can be disagreement about the
descriptive validity of an account, this can be resolved "in principle" by the
data. Further, however, he brings this line of argument to its inevitable
Outcome \vhen he notes, "Of course, the theory could be made problem-
atic by one of the participants in the discussion-for example, by challeng-
ing the applicability of 'throwing' to what the student did with the eraser"
(p. 287). One might coumer at this point that to challenge the applicability
of "throwing" is not only to make theory problematic, it is to make the
435

description problematic. For example, one might say that "throwing" can-
not be applied here because the student "tossed" the eraser. If someone
were to raise this point, then what would be at stake \volJld not be just a
m:lrrer of m:1"=:ino: the tl,e()n' Dl'oblel113.tic: to the contl~3.ry, it \.vould I..,,. :l
fundamental chaHenge to the very accuracy of depiction, with each accus-
ing the other of an incorrect or inaccurate accounting of the faGS.
lvIaxweli (1992) responds to this complication with a shift of validity
categories and says that this then is no longer an issue of descriptive valid-
ity, but rather one of interpretive validity. But this will not do because the
whole point of the idea that there can be no theory-free observation is that
it is virtually impossible to disentangle the descriptive from the interpre-
tive. How we describe the world is an interpretation-or, put dif-ferently,
interpretation cuts very, very deep indeed. That much of the time we can
agree on "thrm:ving" versus "tossing" is not because reality stands over
against us, but beGlUse we happen to share a theoretical or pre theoretical
disposition and a language for depicting movement in time and space. In
his next paragr;lph, Max\vell recognizes this point when he says, "Descrip-
tive understanding pertains, by definition, to matters for which we have a
framework for resolving such disagreement, a framev.,rork provided in
large part by taken-far-granted ideas ;lbout time, space, physic;ll objects,
behavior, and our perception of these" (p. 286). In other words, if we start
from the same perspective, sharing a language and so on, we will tend to
describe/interpret things in basicaJly the same ways. If we start from dif-
ferent theoretical or pretheoretical perspectives, our descriptions/inter-
pretations of events and actions will differ. This is exactly the point made
by Hanson (1958), Kuhn (1962), ;lnd many others, including the neo-
r;alists. Accordingly, just like Hammersley (1990) and lVlanicas and
Secord (1983), IVlaxwcll is unable to show us how to get reality to do some
serious work-work that would give his discussion of validity, and by
extension criteria, the kind of force th;lt will allow it to stand over and
against or beyond a process of socially and historically constrained
judgments.
"'\Ve dose this section by noting that the problems faced by the neO-
realists are little different from the ones faced by one of their major intel-
lectual precursors, Popper (1959, 1972), who, although he as much as
anyone else contributed to the retreat from logical empiricism, still at-
tempted to salvage something from the remains of that project. popper
freely accepted that there could be no theory-free observation. That point
granted, he still wamed to preserve the capacity to test thcories against
436
JItt;:flU
I l i e rIUUJ",'1I VI I... In rne f-\ge ur r;;eluuvism
observational statements. In order to do so, he appealed to a version of
realism as the key element in our ability to falsify hypotheses. The problem
with his appeal is that if falsification is a matter of a disjuncture between a
tl-lC<.H7J or a h)'puthc5i5 derived [wma lheory, and the facts, [he [arrer must
be independent of the former. This is precisely what is not possible if all
observation is theory laden. That is, it is impossible to think of the compar-
ison of a hypothesis with theory-mediated facts as the same ;lS resting a
hypothesis against an independently existing reality. Hindess (1977)
dearly notes the incoherence present in this line of thought: "If theory is
incscapably implicated in observation then testing cannot be a rational
procedure. If testing is a rational procedure then there must be an
atheoretical mode of observation governed by a preestablished harmony
between language and the real. To maintain ... both the rationality of test-
ing and the thesis that observation is an interpretation in light of theory
is ... [a] contradiction" (p. 186).
The problems with Popper's realism arc further evident in his claims
abom inquiry and approximations to truth. This is a claim that
Hammersley (1990) reiterates in slightly different terms when he men-
tions the possibility of research producing "cumulative knowledge"
(p. 62). The basic idea sponsoring these claims is that over timc our re-
search gets us closer and closer to truth or reality. Unfortunately, this
whole line of rhought, as is well noted by Hazelrigg (1989), lacks cogency
because the "notion of approximation requires an existent referent to
which one can achieve proximity, yet how can we know that existent?
According to Popper we cannot; but if we cannot, then hmv are we possi-
bly to judge the question of proximity?" (p. 78). Again, the neorealist
claim that it is reality itself that can serve as the bottom line for judgments
cannot be made good.
In summary, then, it is clear the neorealists cannot grant us the kind of
purchase on the criteria issue that they desire. This lack is evident in the
work of Hammersley (1990), Maxwell (1992), and others. In the cases of
the aforementioned, both begin with, in the first instance, the claim that
we can retain the concept of "truth as correspondence to an independent
reality" (p. 61) and, in the second, that a judgment about the validity of a
study can be taken with regard to its "relationship to those things that it is
intended to be an account of" (p. 281). Howevcr, in both cases this is as far
as their neorealism goes, and both eventually turn to judgment as a social
process. For Hammersley this move appears with full force when he fails
to show us how to make contact with an external referent and, in response
437

to this failure, he turns his full attention in the only direction possible-to
the issue of the norms that should guide discourse among researchers. For
iVlaxwell, the signal that he should abandon his neorcalism comes when
he nott:'~ th!1t [IV(l11 :1 ;t1d~111C111' "he.nt cL~<;ct:pt;.... c vc-.l:<~;';r', ......I,;,J-. 1-.~, say"
about the "factual accuracy of [anJ account" (p. 285), requires a shared
framework for resolving disagreements (p. 287) and, accordingly, "valid-
ity is relative in this sense because understanding is relative" (p. 284).
+ A Transition
\'\lith the failure of the neoreJlists to salvage the last remnants of the em-
piricist project, it is time to make a transition and begin to change the
conversation. In othenvords, it is time to move beyond any lingering hopes
for a foundational or quasi-foundational epistemology, to change our
metaphors and imageries from those of discoverers/finders to those of
constructors/makers, and to accept that relativism is our inescapable condi-
tion as finite beings. There is no question that we must, as both inquirers
and laypersons, learn to live with uncertainty and contingency, and forgo
the hope that something will come along that will enable us to transcend
our human finitude. But, that said, the relativism that arrives with this per-
spective does not mean we must give ourselves over to the neorealist fear
that all judgments are equally valid, or that "anything goes" in our assess-
ments of the quality of inquiry. Schwandt (1996) summarizes the non-
foundationalist situation quite succinctly: "\X!e must Jearn to live with un-
certainty, with the absence of final vindications, without the hope of
solutions in the form of epistemological guarantees. Contingency, falli-
bilism, dialogue, and deliberation mark our way of being in the world. But
these ontological conditions are not the equivalent to eternal ambiguity,
thc Jack of commitment, the inability to act in the face of uncertainty"
(p. 59). In shorr, the problematic for nonfoundationalists, which is pro-
foundly unsettling for us as individuals and as a community of inquirers (if
such a sense of an overall community is even possible), is that of how to
lllake and defend judgments \\'hen there can be no appeal to foundations or
to something outside of the social processes of knowledge construction.
Discussion of this problematic, and of what it means for the judgments
~ve make about inquiry, must begin by clearing some ground and establish-
1l1g some points. Ivlost immediately, a response is required to the ofc-
expressed charge that relativism is self-refuting. The refrain here is quite
438
... ~ ~~.~ ... ", ....'n ...... '" '" u,'" ,...,~c <-', " ..... n.'''.'''",
well-known: To say that all things are relative is to make a nonrelative
statement, As such, relativism is self-refuting and an untenable position
to adopt. \X!e must address two issues in order to demonstrate why this
i,; ..::hCl,q;,'" is o[ 110 "::OIlSCLJucnct::.
First, is it the case that relativism is self-rcfuring? Given the form of
logic we normally bring to such questions, the answer, of course; must be
yes. But this then leads to the second point: Docs it matter, again given this
form of logic, that relativism is self-refuting? In this instancc, the answer is
no. This second point obviously requires elaboration.
One could attempt to argue that this point about being self-refuting
should be a crippling blow to nonfoundationalisrs and their approach to
criteria, if only because other schools of philosophical thought have met
their demise in similar circumstances. For example, logical empiricism
certainly unraveled over the verification issue. As is well-known, logical
empiricists said that any statement that cannot be empirically vaified is
meaningless. But, of course, that statement itself cannot be empirically
verified and so, logically, it is meaningless. As Putnam (1981) so clearly
points out, it was this problem as much as anything else that led to the end
of logical empiricism. So, why should not the same stricrures be applied ro
relativism?
The initial answers to this self-refuting situation came from Gadamer
and Rorty. Gadamer's (1995) response is to agree that relativism is self-
refming, but then to shrug it off by saying that to make this point is to
make a point of no interest, or one that does "not express any superior
insight of value" (p. 344). Rorty's (1985) tactic is to argue that it is a mis-
take to think of relativism as another theory of knowledge. Quite to the
contrary, he dispenses with the self-refuting problem by saying that
because his type of pragmatist is not interested in advancing any "episte-
mology, a fortiori he does not have a relativistic one" (p. 6).
Although we agree with much of how both Gadamer (1995) and Rorty
(1985) have approached this issue, we feel a further comment remains
necessary. One reason the self-refuting claim is and should be devastating
to logical empiricism, and, for that matter, to other schools of thought, but
not to relativism, depends on the aspirations expressed by these philo-
sophical dispositions. \X!e would argue that empiricism in any form, for
example, was based on the claim and the hope that we could deny our
human finitude and adopt a God's-eye point of view. That is, the claim was
that we could "bracket" ourselvcs in some way or another, predominantly
through submission to method, such that it would become possible for us
439
to seize upon some external referent (i.e., reality knovm as it rcally is) to
adjudicate different knowledge claims and resolve oLir differences. Given
this claim, it is quite appropriate to apply the logic of self-refuting-the
::;rrengrh of the claim and rhe aspirarions advanced musr be directly relared
to the strength of the argument brought forth in supporr of those claims
and aspirations.
Relativism as understood here, quite to the contrary, is not a theory of
knO\vledge and advances no pretense that we can escape our finite condi-
tion of being in the world. As such, the issue of self-refuting, even though it
is a logically correct position given a standard form of logic, is unimpor-
rant. Taking a page from Godel's idea of incompleteness (sec Hofswdter,
1979), this situation is only what one would expect from any human,
social and historical, construction. Relativism is not something to be tran-
scended, it is merely something with which we, as finite beings, must [earn
to live.
Does this understanding of relativism leave nonfoundationalists
enmeshed in a situation of "anything goes"-a situation where all judg-
ments about inquiry, and, for that matter, everything else, 3fe equal?
There are two intimately related reasons why relativism should not and
need not be understood in these terms. First, the charge of relativism as
anything goes does not make sense because this charge requires for its
vitality a viable concept of the absolute. As previously noted with refer-
ence to Hazelrigg (1995), if we no longer can speak of the absolute with-
out embarrassment, then we must realize we cannot speak of "anything
goes" without embarrassment.
Second, and more tangibly, it is very clear ""ve all already do make judg-
ments, and as far as anyone can see, we will continue to do so for a long
time to come. To say these judgments cannor be grounded extralinguis-
tically does not mean we are exempt from engaging in as open and uncon-
strained dialogue as possible in order to attempt to justify our assessments.
But to attempt to persuade another further requires that one is open to
being persuaded. All relativism brings to the table with regard to the issue
of criteria is rhat to be a finite human being who must live with and make
judgments in concert with other finite human beings can be, with some
frequency, very tough work indeed. This is, in effect, \vhat Schwandt
(1996) means when he says, as noted above, that learning ro live with
uncertainty and the impossibility of final vindication does not mean that
vve must abandon commitment and our ability to make judgments.
440
\X1hat, then, does this understanding of relativism as a recognition and
acceptance of Out human finitude mean for social and educational
inquiry-especially for the issue of criteria for judging the quality of our
studies? One vcry important way to frame this question is to introduce the
difference between depictions of inquiry with the use of the metaphors
and imagery of discovery or finding versus the metaphors and imagery of
making or constructing. If one accepts a constructivist stance, this means
that, as Rony (1979) so well notes, the metaphors and imageries of discov-
ery and finding no longer can be made good. However, Raft)' also goes on
to add that we might as well stay with the language of a found world. Parr
of his justification for this injunction is that "nothing deep turns on the
choice between these twO phrases-between the imagery of waking and of
finding" (p. 344). However, Hazelrigg (1989) has been persuasive in help~
ing us understand that J great deal of consequence turns on this disrinc~
tion. This is so because, as he putS it, "the difference in consequence is not
merely of imageries and stories but of practices-of who we are, of what
we do, of \vhat world \ve make" (p. 167). The problem is that to continue
to employ the language of a "discovered world" is to continue a "passivity
in regard to responsibility for the world" (p. 168). And, as Hazelrigg con-
tinues, to think in terms of finding is to place the world "outside the
domain of human will. That is an enormously dangerous consequence of
any retention of the 'found world' language of storytelling" (p. 168).
To illustrate this point, we note that in the histories, especially more
recent, of both educational practice and research there has been a strong
tendency to categorize children as this or that-as learning disabled (LD),
for example. For those who operate with the image of researchers as "dis-
coverers," LD is a natural category-there always were LD children, and
the reason we have recently discovered them is that our inquiries have
become increasingly sophisticated. For those who use the language of
making or constructing, LD is a constructed category-a \vay \\ie have
chosen, for \vhatever social/historical reasons, to categorize children. The
point is not that we dispense with categorization, which in any case is
impossible for the finite human mind. To the contrary, the point is to
examine and fully discuss why we construct the world 3S we do. This is a
discussion that is practical and moral, framed by contingent social and his-
torical circumstances, and certainly not epistemological in any theory-of-
knowledge sense of the word. Put differently, for those possessed by the
metaphors of discovery, issues of moral responsibiliry are at best contin-
gent because "discovery" is only a matter of contact with reality and what
441

reality is telling us. For those who accept the metaphors of constructing,
moral responsibility is central because one must be morally responsible for
\vhar onc constructs or makes (for a funher example and discllssion, see
Smith, 19S9. e'neci"lly chnn. 4).
ft is in this sense that the relativism of the nonfoundationalists requires
that inquiry be seen as an act of construction that is practical and moral
and not epistemological. Likewise, as must follow from this requirement,
any judgments about the goodness or badness of research must themselves
be practical and moral judgments and not epistemological ones. For the
nonfoundationalists, to move away from epistemology is to recognize
inquiry as a social process in which we construct reality as we go along
and as a social process in which we, at one and the same time, construct
our criteria for judging inquiries as we go :llong.
+ Changing the Conversation
Relativism is nothing more or less than the expression of our human fmi-
tuele; we must see ourselves as practical and moral beings, and abandon
hope for knowledge that is not embedded within our historical, cultural,
and engendered ways of being. If this so, then to continue to talk in the
vocabularies of theories of knowledge is to continue to talk in the shadows
of the unrealizable desire that something will corne along to allow us to get
off of Momaigne's wheel or out of the hermeneutic circle. Chisholm's
('1973) paraphrase of IVlomaigne clearly summarizes the unresolved and
unresolvable epistemological problematic:
To know whether things really are as they seem to bc, we must have a proce-
dmc for distinguishing appearances that are true from appearances thar are
false. But to know whether our procedure is a good procedure, we have to
know whether it really sllcceeds in distinguishing appearances that are true
from appcarances that arc false. And wc cannot know whether it does really
succeed unless we already know which appearances are tme and which oncs
arc false. And so we are caught in a circle. (po 3)
The neorealists have been unsuccessful in their attempt to cling to an or-
ganizing narrative of epistemology free of time and place. They have not
been able to shmv us a way to get off of the wheel or out of the circle. \Xle
sllspect it is time to move on; it is time to change the conversation by
442
....".",,u "' '0:;: ,"'y<;, UJ ,\t:'JUlIVI::>rll
asking what making judgments is about when we can no longer be epi-
stemological. Although we are cerrainly able to urk about knmvledge,
especially its social production, there is little point in talking about theo-
r;cs of L:nm.\!l~,Jsc "nd rursuins the iHnbitiuIl to solve the problel11:1tic.
That we can no longer play the game of philosophy as epistemology and
find solace in a foundatiomll narrative can be nothing but profoundly
unsettling to us. This is a break, a rupture, that appears to leave social :lnd
cducational inquiry "undone"-to use a vvord from the title of a reccnt
book by Stronach and IVlaciurc (1997). Put di Herend)', all of this talk abou t
judging can no longer take place within that seemingly abstracted, disem-
bodied public space that was fostered by the desire to be epistemological
and to thereby specify the context of judgment apart from the actual judg-
ment process. In the age of relativism thc issue of who is making judg-
ments, about what inquiries, for what purposes, and with whom One
shares these judgments is of critical importance. As inc1ividu:lls \ve must
make judgments, and <:IS members of social groups, ho\vever loosely orga~
nized, we must be witness to situations in which our individual judgments
are played our with the judgments of other individuals.
But, we must immcdiately notc, there is 8t least one very app<:lrent Jnd
often pracriced way to avoid the difficulty of judgment. Cert:8inly one
could turn to power, in one or another of its numerous guises, to render
thc social context and process of judgment very specifi8ble indeed. There
is the use of power in a rather raw or open application-for example, the
ali-tao-common situation where a reviewer rejects an article \vith little or
no explanation. Or there is the power of the shrug of the shoulder rll<:1t
seems so popular today with various people when they are questioned.
The questioner just does not get the point and the situation is so hopeless
that there is no reason to attempt to engage any further. In these instances
and many, many others of similar nature \vith which we are all familiar. an
individual has made a judgment, criteria have been applied, but eng~ge
lTlenr with the judgment and criteria has been deniee!. To engage this
defensive stance, based on the claim of theoretical, standpoint, or what-
eVer differences, works to narrow the public space or social group with
whom one shares judgments-all others are read out of the exchange. In
any event, although the possibilities for the application of power to solve
~he problem of criteria seem 8lmost endless, they also seem very unsatisfy-
Ing and undesirable.
This should not be taken to mean that \YC arc so nai"ve as ro hold that
POWer should and can be eliminated from judgments as they are plJyec1 out
443
 ,,~ , LOU ,,,. ,I"""'.' 'V, ~, '- 'r>L.'-'r> , ' ..... ' "
socially. \X!e have no intention ro embrace some sort of romanticized
"inrellecrualist flight from power" (Hazelrigg, 1995, p. 202). Rather, the
issue for uS ,enter, on the responsible use of power while "voiding ite
excesses. But we also are well aware that it is impossible to specify in some
permanent way exactly when one has crossed the line from responsible to
excessive. Likewise, it would seem there are often good reasons to narrow
the socia! group. Wben marginalized groups seek to bave their voices
heard, solidarity Gm be very important. In fact, we would argue that this
was tbe case for many qualitative inquirers a few years ago. But, again, the
question arises as to how to honor this need while avoiding its excesses-
fully realizing that we are unable to specify exactly when the line has been
crossed. In short, then, "ve recognize that power is ever and always pres-
ent, but we also bold that tbe temptation (also ever and always present?) to
use power as a way to avoid critique and tbe risk of one's prejudices denies
the value of and need for public discourse. So tbe issue remains: What can
be SJid aboLlt judgment when relativism demands attention co context and
forgoes the possibility of appeal ro an abstract, disembodied (epistemo-
logical) public space?
One response to this question from some people is the claim that any
discussion of criteria is pointless. Although we will not engage Rosenau's
(1992) rendering of affirmative versus skeptical postmodern and post-
structural perspectives, we must acknowledge that some of ber comments
successfully surface issues that have been troubling many within the acad-
emy. Those \vbom Rosenau calls skeptic postmodernists argue that we can
only be "nonjudgmental" (p. 55) because "good/bad criteria are unavail-
able" (p. 175) and all interpretations are equally interesting or uninterest-
ing. In the absence of established criteria for sorting the good from the less
so, tbere is no choice but to simply throw up our hands and leave the field
of judgment behind. Certainly this sentiment is not unknown, at least to
us. There are times as we review articles for publication, grade papers,
accept the dissertation even though it disappointed, tbat the thought
arises: Who are we to judge? The field should be left open to the play of
differences, and so on. A similar arena of self-doubt, given the fallibility of
human judgment, seems to be the driving force behind the often agonized
and agonizing discussions of voice, writing the other, the crisis of repre~
semation, and so on.
But to begin from this point or to pretend (even hope?) that we can lead
a life free of judgment is to immediately go astray, and obviously so. Judg-
ments must be made and must be argued and justified, unless, in the latter
444
'" , ..... ,,'-, '''_-'L.' ,
,~, ,'-'. ,
instance, one simply resorts to raw power. As Schwandt (1996) has noted
for us, that we all must live with uncertainty and contingency does not
mean that we can dismiss commitment and abandon judgment (p. 59).
There is no question that we all think some things are better than others
and that we make judgments accordingly, often even when we say \ve do
not. This process of judgment making has been going on for quite some
time now, and will continue to do so for as far as can be seen.
It is impossible to imagine a person leading a life without making judg-
ments or without making discriminations. To attempt to do so evolces the
image of Ebenezer in Barth's novel The Sot-\v/eed Factor (1960), who,
because he found all things equally interesting, "threw up his hands at
choice" and "sat immobile in the window scat in his nightshirt and stared
at the activit}' in the street below" (p. 11). The same immobility must be
the fate of all those who would claim to find everything equally uninterest-
ing. Taylor (1989) has expressed this situation and its consequences vcry
clearly: "To knm'V who you are is to be oriented in moral space, a space in
which questions arise about what is good or bad, VI/hat is vmnh doing and
what not, yvhat has meaning and importance for you and what is trivial
and secondary" (p. 28). To not make judgments is to lose sight of one's
orientation in moral space, which is to lose one's grounding as a human
being.
In our roles as inquirers, educators, evaluators, we arc always making
judgments about papers for publication, presentations, books, disserta-
tions, srudent papers, and so on. As we approach judgment in any given
case, we have in mind or bring to the task a list, for lack of a better term, of
characteristics that we use to judge the quality of that production. The use
of the term list should not be taken to mean that we are referring to some-
thing like an enclosed and precisely specified or specifiable shopping or
laundry list. Plit differently, to talk of a list in this sense is not at all to talk
about, for example, an accumulation of 20 items, scaled 1 to 5, wbere
everyone's presentation proposal is then numerically scored with a curoff
point for acceptance. Obviously, to think of a list in these terms is to miss
the entire point, so much so that it makes a mockery of the idea of qualita-
tive inquiry, of what relativism means, and of ,,\That it means to realize that
We can no longer be epistemological.
To the contrary, for us a list of characteristics must be seen as always
open~ended, in part un articulated, and, even when a characteristic is mo~e
Or less articulated l it is always and ever subject to constant reinterpreta-
tion. IVIoreover, the items on the list can never be the distillation of some
445

abstracted epistemology, they must inevitably be rooted in one's stand-
point or, to usc Gadamer's (1995) term, they must inevitably evolve out of
and reflect one's "effective history" (pp. 301-302). These points require
The lists we bring to judgment are and can only be open-ended in that
\ve have the permanenc capacity to add items to and subtract items from
the lists-although, as \ve shall soon note, doing so involves risk. Should
we accept the risk, then the limits for recasting our lists derive not from
theoretical labor, but from the use to which the lists are put. The limits on
modification are a practical matter; they are worked and reworked within
the context of actual practices/applications and cannot be set dovm in
abstracted formulas.
Second, ;my lists we bring to judgment are only partly articulated. Some
items can be more or less specified, whereas others seem to resist such
specification. Polanyi's (1962) idea of tacit knowledge applies here very
\vell indeed. \Vhen we make judgments we morc or less can specify some
of the reasons, bllt other things seem to be out there-there is surplus that
seems to stand just beyond our grasp, just beyond our language. This does
not mean that we should nor and do not attempt to bring this surplus to
fuller conscioLls articulation; it only means that it can never be done com-
pletely. That this is so should not be surprising, because as finite beings we
can never be fully transparent, not only to others, but even to ourselves.
Similarly, even those items on our lists that are more or less well articulated
can be reinterpreted, even if subtly and gradually over time. And, finally,
the importance we assign to items on our lists may easily change over time.
A characteristic of research that we thought important at one time and in
one place may take on diminished importance at another time and place
(D. Gallagher, personal communication, 1998).
But what is most important to reiterate here is that our lists are chal-
lenged, changed, and modified not through abstracted discussions of the
lists and items in and of themselves, but in application to actual inquiries.
That is, very often something "new" comes along, such as Sparkes's
(1996) self-rei1ective narrative of the fragile body-self, the developing
genre of ethnographic fiction, or readers' theater. The new does not fit
well with one's existing list of characteristics. To read these "new" things
that come along or to attend the challenge seriously immediately opens up
the possibility that one must reformulate one's list and possibl y replace the
exemplars one calls upon in the never-ending process of making judg-
ments. Ho\,vever, the key here is possibility, because inquirers may choose
446
,,'t: r'vu,t:',/ v/ '-'/l\::'/U 1"lI!\:: MY\:: v' /""",Ul'V'::>/"
to preserve their lists, as many often seem to do, and judge the production
as not even qualifying ro be considered research. The latter is.] comment,
by the way, that was often heatd about qualitative inquiry in years past.
The liSIS VI'e hold are part of us-they are expressions of OUf own partic-
ular standpoints Ot effective histories. In other words, to refer again to
Gadamer (1995), a list is an expression of our prejudices and, as he notes,
it is the prejudices of an individual that "constitute the historical reality of
his being" (p. 277). In any encounter with a production, especially some-
thing "new," as noted above, one must be willing to risk one's prejudices.
.lust as in the process of judgment one asks questions of a text or a person,
the person or a text must be allowed to ask questions in rerum. As per the
example of Sparkes's (1996) work mentioned above, to approach this
inquiry requires that one be open, that one be willing to allow the text to
challenge one's prejudices and possibly change one's list and one's idea
J-bout what is and is not good inquiry. But twO important points must be
made here. First, to reiterate, the word possibly is crucial here because to
be open does not mean automatically to accept. One may still offer reasons
for not accepting something new. Second, there is no method for engaging
in the risking of prejudices. If anything, to risk one's prejudices is a matter
of disposition-or, better said, moral obligation-that requires one to
accept that if one wishes to persuade others, one must be equally open
to being persuaded.
1t is time to turn to an exemplar to illustrate these points, The work of
Lather (1986a, 1986b, 1993, 1998a, 1998b) over the past lOot 12 years,
in various articles and presentations and in her book with Smithies (1997),
stands as an almost paradigmatic example of the fact that we all approach a
piece of work with something in mind. Lather's discussions both partici-
pate in and trouble our discourse about criteria for judging inquiry. In the
process, Lather struggles with lists of characteristics that are open, evolv-
ing over time, and always in flux, and that resist full articulation. The items
Lather has brought forth under the banner of validity have clearly evolved
over the years. In 1986, Lather was talking about validity \vith refer-
ence to triangulation, face validity, catalytic validity, and so on. By 1993
this list had evolved to a discussion of transgressive validity, with her in-
terest focused on such concepts as simulacra/ironic validity, p8falogy/neo-
pragmatic validity, rhizomatic validity, and voluptuous/situated validity.
There are two important points to be noted here. First, Lather is well
a\vare that her lists are not composed of highly specified items that arc
enclosed, complete, can be numerically scored, and so on. In fact, when
447

she develops a "checldist" it is only [Q mimic and thereby disrupt the very
ideel or possibility of a checklist. Second, she also is well aware that her
theorizing is and Gm only be grounded in practice or in the actual process
of applying the lists and making judgments. This is a point she noted in her
work in 1986 and that she has reiterated, to one degree or another, in her
more recent efforts (1998a, 1998b).
Lather sees the task of judgment not only as a practical engagement, but
also as a mora! onc. Although we must acknowledge her resistance to the
appropriation of poststructuralist perspectives within unproblematized
concerns of critical theorists, we also mllst acknowledge that she clearly is
<md has been concerned deeply with such things as oppression, exploi-
tation, and domination. There is no question that she holds a normative
frame of reference in which some definitive preferences are expressed.
Nor surprisingly, then, Lather holds, as \ve assume do most othcr inquir-
ers, thin socia! :md educational inquiry can, should, and must have an
ameliorative purpose-that \.vhat we do as inquirers has the purpose of
contributing to making people's lives and futures better (again, with full
recognition that \vhatever better might mean in any given context cannot
be precisely specified).
However, at this point a version of the problem of Ebenezer's paralysis
(as noted above) seems to be lurking in the shadows. Lather (1993,
pp. 684-685) poses <l number of worrisome questions that arose as sbe par-
ticipated in a project involving women with HIV/AIDS. These are prob-
lems that are, as she puts it, "grounded in the crisis of representation"
(p. 684). Her concerns range widely over issues about voice, the other, the
methodological interest she advances, her goals as a researcher, and so on.
These concerns seem to be echoed, even if expressed in a different sense
and in different terminology, by Scheurich's (1997) fears about the
"resourcefulness of the Same to reappear with new masks that only seem
to be the Other" (p. 90) and about the "anonymous imperial violence tbat
slips quietly and invisibly into our best intentions and practices" (p. 90). In
both instances, we argue that at the core of the problem is these authors'
concern that any judgment must close off other possibilities, that any writ-
ing discloses in its silences as much as it reveals in its articulations, that to
write the other is to appropriate the other, that any judgment is quietly but
immediately turned on itself, and so on.
\YJe have no doubt that the questions are important and the concerns are
genuine, and \ve agree that we must always question the judgments we
make and guard against the reappearance of the Same. In this sense, both
448
. - -- _. '._, ' - .. , ,~ '::1~ ~ '~'~L" ,~'"
Llther and Scheurich have raised impon:mt" issues that have moved us at
lcast partly "nv;]}' from the epistemological project. Bur still n'e must" ask:
Can this questioning be taken too far? Is the quest framed by J desire for
answers that can never be had and guarantees that can never be given? In a
text that presents the voices of the women who ii\re vvith the reality of
being HIV-posirive, Lather and Smitbies (1997) demonstrate that they un-
derstand what might be called the imperialist position of the author and
O\\'n their particular judgments to do one thing as opposed to another in
telling the stories of tbese women. However, in later discussions abour her
judgments, Lather (19983) 1998b) seemingly \vorks to undo the responsi-
bility she has enacted, reinvoking a concern that she is still seeking a WJy to
aJIeviate human vulnerability and contingency in judgment.
These points also CJll be notcd in tbe responses of Lather (1993) and
Scheurich (1997) to the questions she asks and the concerns he expresses.
\'{/e argue that because the answers sought renuin unavailable, as ('hey
must, and the concerns resist relief, as they must, they have turned too far
to the other side with tbe call for a "loud clamor of a polyphonous, open,
tumultuous, subversive conversation on validity as the wild, unknowable
play of differences" (Scheurich, 1997, p. 90) and "a disjunctive affirma-
tion oE incommensurates" (Lather, 1993, p. 687). But if these injunctions
that everyrhing should be an unknowable play of differences and that all
contributions are incommensurable are taken seriously or taken to conclu-
sion, have they arrived at the point of "everything goes" such that nothing
is left for judgment? Their lingering concern with the epistemological
project has left them in an untenable situation iNhere, intellectually or con-
ceptually, they have rendered judgment pointless if nor impossible,
whereas practically and moraUy they do and must make judgments. A
hypothetical example is necessary to make our point as dear a:~ possible.
Y\ie are confident that if Lather Jnd Scheurich were to review for publica-
tion an article with racist overtones, as a practical and moral affair ther
would reject that papet for just that reason. However, at another level, 0;'
what basis could this be done? Given their comments above, why should
not this paper be seen as simply another element in the play of differences
or as another affirnlJtion of incommensurates?
For us, it 1S time to accept our vulnerability and contingency, drop the
last traces of the epistemological project, and thus change the conversa-
tion. There is no way off of the wheel or out of the circle and , in what may.
seem an odd rwist, in a strange way there is no epistemologicat crisis of
representation, but only a practical and moral problem of representation.
449

\X1c are finite human beings who must learn to accept, for example, that
anything we write must always and inevitably leave silences, dlat to speak
at all must always and inevitably be to speak for the someone else, and that
...ve cannot TI1Clkc iudgrncnr~ Clnd at the SaITle tinlt: have a Uconstandy nIOV-
ing speaking position that fixes neither subject or object" (Lather, 1993,
p. 684). To lament this condition and to search for a solution to these
"problems" is actually to lament and search for a solution to our human
finitude. Bur that we are finite is something we can do nothing about.
Hazelrigg (1995) puts it very well: These are our "deficiencies" as human
beings, if you \vi!1. But they are deficiencies "proper to the fields of human
knowing and doing" because "what the human creature is not, no matter
what the strength of will, is omniscient and omnipotent" (p. 102).
Bur a caution must be reiterated lest we be misinterpreted. There is no
doubt that there is a problem of representation. Silences must be ques-
tioned, \ve must be careful of what we speak in that vve speak for others,
and we must be camious in our judgments and be willing to risk our preju-
dices as we share and justify our judgments in a public space. But these are
not epistemological issues; to the contrary, they are practical and moral
issues. \Y/e inquire, we make judgments about inquiries, we must give rea-
sons for our judgments, offer up these reasons to others and simply
attempt to do the best we can. Knowledge is a human social production. As
finite beings, all we can do is construct social and educational worlds,
social and educational constructed realities for which we are morally
responsible. To realize that we are finite human beings is to realize that
there may be little more to say than this about judgment, criteria, and
validity.
Be that as it may, our remaining comments must turn more directly
toward the fact that judgment obviously is not a solitary engagement. As
teachers and researchers, \-\le are members of social groups and, as briefly
indicatcd at various timcs above, our individual judgments inevitably must
be moved into a public space wherc they are placed in concert Y'lith the
judgments of others. To a certain extent the public space for sharing judg-
ments, certainly the initial and immediate public space, has been predeter-
mined because of institutional and organizational arrangements. For
example, grading papers must at least involve a public space with a stu-
dent; dissertation approval/disapproval must involve the student, the
committee, and whoever comes to the defense; reviewing a paper for jour-
nal publication must necessarily involve the author and the editor. In ;l
broader sense, however, there is room for choice as to the public space
450
. _. ~, ..... " ... n, "''''' r"'\::1"" VI '~""'U<'V'''''''
within which or with whom one's judgments will be allowed play and the
extent to which one allows one's prejudices to be tested and challenged.
This is possible because, although the ideal of community i" imporram,
no overarching social and/or educational research com-
jn praclice then.: i::i
munity. Educational researchers, for example, are dispersed along numer-
ous lines from what are referred to as subfields, to theoretical 2nd method-
ological differences, to differences based on characteristics like gender,
race, and sexual orientation, and so on. That we are dispersed in this way is
what allows for some degree of individual choice in terms of with whom
one will share judgments and risk one's prejudices.
There is no question that over the past few years the rejection of
meta narratives and appeals to final vindications have prommed the con-
cern to open new spaces of inquiry and increase the number of possible
groups, however loosely formed, vvith which one can choose to share or
not share judgments. This is a recognition that has appeared in the work of
numerous people, including, for example, Denzin (1997) and Lincoln
(1995). For the latter, one of the key points is that all texts "are always par-
tial and incomplete; socially, culturally, historically, racially, and sexually
located; and can therefore never represent any truth except those truths
that exhibit the same characteristics" (p. 280). This is further coupled with
the recognition "that research takes place in, and is addressed a com- t;,
munity; it is also accurately labeled because of the desire of those who dis-
cuss such research to have it serve the purposes of the community in which
it was carried out" (p. 280). And, very important, Lincoln says 'that these
ideas must be understood within the Context of ethics and soci2.l ameliora-
tion. About the latter she says we must have a "vision of research tbat
enables and promotes social justice, community, diversity, civic discourse,
and caring" (pp. 277-278).
For Denzin (1997), we have entered the era of the new ethnography,
one rhat has left science behind, that finds no "break between empirical
activity ... theorizing and social criticism" (p. 86), and where we will
enCounter the continuing development of "standpoint epistemologies of
race, class, nation, gender, and sexuality" (p. 87). Denzin continues by
noting three shared commitments that stand behind the advent of stami-
point epistemologies. First, our research must be undertaken from the
"point of view of the historically and culturally situated individual"
(p. 87). Second, "ethnographers will continue to work outward from their
OWn biographies to the worlds of experience that surround them" (p. 87).
Third, the desire is to produce inquiry that will "speak clearly and
451
 It', '-I" , , ' - 1'->1 ' ' - ' ' ' , '- .n<- ..... '-" , ..... , . , ' ' ' ' ' ' -
powerfully abour these worlds [of experience)" (p. 87). Thus, in the end,
"the stories ethnographers tell to one another will change, and the criteria
for reading stories will also change" (p. 87).
1"0 the extent that this new ethnography brings with it an increasing
pluralism, multiplicity, and the play of differences to open new spaces and
new possible futures, we think the move must be applauded. Hmvever, as
alluded to above, the issue of honoring pluralism and Illultiplicity while
avoiding its excesses is ever and 3lways present. Schrag (1992) frames the
possibilities quite well. On the one side, he says that it is possible that a
multiplicity of stories C3n be "assembled in such a manner that lines of
similitude remain operative and a binding of multiple discourses 3nd prac-
tices remain possible" (p. 33). If this can be accomplished, then we could
offer up our judgments to others in the name of the '<possibilities for ratio-
nal critique, articubtion, and disclosure as these are geared to an under-
standing of shared experience, evaluation, and cmancipation" (p. 49).
Howev~r, on the other side, he also notes that if this multiplicity and plu-
rality engage nothing beyond an endless play of local and localized storics,
vocabularies, and judgments, then \ve may be ensnared in a never-ending
process of "dissensus, incommensurability, irretrievable conflicts of inter-
pretation, and hermeneutical nihilism" (p. 33).
As a practical and moral matter, Schrag (1992) centers a number of
important questions: How does one preservc diversity and multiplicity
while avoiding its excesses? \Vith whom and under what conditions am I
obligated to risk my prejudices by offering up my work and my judgments
for commem and criticism? Or, put differently, What is there that \vilJ
allow us to put faith in Denzin's (1997) claim that in the era of the ncw eth-
nography inquircrs will work Ollt from their own standpoints? But two
points arise immediately to madden any discussion of these questions.
First, what does it even mcan to say that we have reached an exccss of
diversity; when and how does one know when the line of excess has been
crossed? Second, it is na"ive to hold that one should constantly offer vmrk
and judgments to a total diversity of localized communities or social
spaces. At times there may be very good reasons, physical and mental
cxhaustion aside, for not reaching out and engaging in extended dialogue
\vith others. For example, there are times when dialogue has been en-
gaged, progress has not been made, and there seems no reason to continue.
As has been so throughout this change of conversation, the issues are iJlu-
sive indeed and resist easy solution.
452
.,~, ~-'-
",,=,, 1UUI'='1 'I vr \...trH::t,u If' tnt:: ,..",ye or KE:'IQrIVISm
That said, \ve begin by noting that we value Schrag's (1992) hope that
lines of similirude can remain open and effective across a range of dis-
COI1]'f;e~ :md practices. The reason for this is framed by the fact that our
inguirics do not discover realJry, bm rather construct reality-a COI1-
structed social and educational reality for which we are morally responsi-
ble, as individuals and collectively. And this point must be coupled v'lith
the injunction from Ivlarx that our task is not merely to study the world,
but to help change it. \Ve thereby have a moral obligation to maximize our
collectivc influence on acwaJ policies and practices. To do this requires
that we engage across lines, that we move our from our individual stand-
points and risk our individual prejudices to keep lines of connection open
to others.
If these lines cannot be kept open, then there may well be a Balkan-
ization of, or a deep heterogeneity among, inquirer-based standpoints,
theoretical differences, or whatever, and we could well end up with deep
dissensus and a permanent conflict of interpretations. The play of dif-
ferences to open spaces could easily tip over into little more than the
announcement of differences, and the affirmation of incornmensurates
easily could become exactly that. This would seem to lead to inquirers'
talking past one another, or not talking to each other at all, as they con-
struct their localized communities, realities, and criteria. Under these con-
ditions, it is difficult to imagine how our work can have moral purpose and
social influence. The point is straightforward: If we cannot talk and listen
to each other, it is difficult to imagine why anyone else would want to talk
with us, listen to us, or attend to our judgments.
But the main problem is that there is no method or process to which we
can appeal to energize Denzin's claim (hope?) that in this era of the new
ethnography inquirers will work ourward from their own biographies to
engage the biographies of others. If inquirers are to do this, then it can
only be a matter of disposition, not method. This is a disposition-or, as
noted above, a moral obligation-that enjoins one to risk one's prejudices
and defend one's judgments in a social space with others, even with those
whose standpoints may seem seriously different from one's own. At the
core of this moral obligation is the requirement that we accepr that just as
We attempt to persuade others to accept our judgments of good versus bad
inquiries, we must in turn be equally willing to be persuaded by the judg-
ments of others, And, to reiterate, it is only with this obligation in mind
that we can maximize the possibility that our inquiries will assist in the
construction of more just social and educational worlds.
453

{& Summary
The quasi-foundJ.tionalists have attempted to establish criteri:1 within the
context of an epistemological constructivism on the one side and, on the
otht:r, an ontological realism. The latter announces their claim that there is
a reality independent of us that can be known as it is-at least in principle;
the fanner assumption directs their attention to the idea that knowledge is
sociaUy constructed and always fallible. This has led them to talk about cri-
teria with the use of terms on the order of plausibility, credibility, and
descrifJliue ualidily. Ivloreover, because the neorealists claim they afe able
to have contact \vith reality, however subtle this contact may be, they fur-
ther claim that these criteria have force beyond particular time and place.
Unfortunately, the ncorcalists have not been able to present a con-
vincing case, or an adequate justification, for their position. Neither
Hammersley (1990) nor Maxwell (1992), for example, is able to demon-
strate how his realist commitment does any real work in terms of moving
us beyond a continual process of social judgments. In fact, were the neo-
realists to drop their realist ontology and their attempt to salvage some-
thing from the remains of empiricism, they would look very much like
relativists of a Gadamerian variety.
Nonfoundationalists have fully accepted the relativist implications of
the fact that there can be no theory-free observation or knowledge. HO\v~
ever, for rhern relativism is not a problem, it is just the inevitable result of
the fact that we, as human beings, are finite-a finitude we should learn to
live with and not lament. For nonfoundationalists, the issue of criteria for
judging inquiry is a practical and moral affair, not an epistemological one.
They take very seriously Rorty's (1985) injunction that our pursuit of
knowledge, as is the case for all human social activities, "has only an ethi-
cal base, not an epistemological or metaphysical one" (p. 6). Accordingly,
criteria should not be thought of in abstraction, but as a list of features that
we think, or more or less agree at any given time and place, characterize
good versus bad inquiry. This is a list that can be challenged, added to, sub-
tracted from, modified, and so on, as it is applied in actual practice-in
actual application to actual inquiries. Relativists also recognize the need
for and value of plurality, multiplicity, the acceptance and celebration of
differences, and so on. The essential problem, however, is to honor this
need without giving over to excesses-to inquiry that is so fragmented that
lines of connection have been lost and the social amelioration possibilities
of our work have been rendered moot.
454
" ..... 'vv,...... " v, ..... ','c"v ", "'C,.,!:!'" v' '~""U"Y''::''''
So, we have come to the end With, as promised, no discussion of par-
ticular criteria and how to apply them. Rather, we have attempted to clab-
01':11"': SOI11<: o( [he conditiuns under which we must think abollt the possi-
bility of critcl"i::J and the possibility of individual and collective judgment in
this age of relativism.
' References
Barth,]. (J 960). The sot-wced /actO!: Garden City, Nl': Doubleday.
Bernstein, R. (1983). BcyoJld objectil'islll ilnd relativis/JI. Philadelphi.:I: University
of Pennsy!v:lIli.:I Press.
Bh.:Iskar, R. (1983). jReview of the book Scicntific rcvolutioJls]. Isis, 74,259.
Cherryholmes, C. (198 S). Power and criticism. New York: Te.:Ichers College Press.
Chisholm, R. (1973). The problem ofthe criterioll. I'vlilwaukee, WI: lVbrqucttc Uni-
versity Press.
Denzin, N. K. (1997). Interpretive eth/lography: Ethnographic practices for the lIs!
century. Thousand Oaks, CA: Sage.
Gadamer, H.-G. (1995). Thtlh a/ld method (2nd rev. ed.;]. \X!einsheilller L,,: D. G.
iVlarshall, Trans.). New York: Crossroad.
Gallagher, D. (1995). In search ofrhe rightful role of method: Reflections on can-
ducring a qualitative disserration. In T. Tiller, A. Sparkes, S. Karhus, &
F. Dowling-Nacss (Eds.), The qualitativc challenge (pp. 17-35). Oslo, Nor-
way: Casper.
Giddens, A. (1976). Nov rIIlcs of sociological method: A positive critique of inter-
pretativc sociologies_ New Yotk: Basic Books.
Goodman, N. (l978). Ways of wor/dmaking. Indianapolis: Hacketl:.
Hammersley, 1\'1. (1990). Reading ethnographic research: A critical guide. London:
Longman.
I--f8nson, N. (1958). nlttemso(discouery. Cambridge: Cambridge University Press.
Hazelrigg, L. (l9B9). Claims of l:.ItOlVledge. Tullahassee: Florida State University
Press.
Hazelrigg, L. (1995). Cultures afttailire. Tallahassee: Florida State Uni\,~tsity Press.
Hesse, M. (1980). Revolt/tions and reco/lstrtfct.iotls in the philosophy of science.
Brighton: Harvester.
Hindess, B. (1977). Philosophy and methodology in lhe social sciences. J\danric
Highlands, N.J: Humanities Press.
Hofsradter, D. (1979). Code!, Escher, Bach: All etema! golden bwid. New York:
Basic Booles.
House, E, (1991). Realism in rese:1fch. Educational Reseal'che}; 20(6), 2~9.
Kerlinger, F. (1979). Behavioral research. New York: Holt, Rinehart & \\'inston.
455
 " , , _ . 'v"-",-,,, "-,, '-"<"""1.1 '" e""" My"" 1.11 1~""IUUyl"""

Kuhn, T. S. 0962). The stmctllte of scientific rel/oll/tioIlS. Chicago: University of Scheurich,.J.]. (1997). Rescarch method ill the postmodem. London: Ellmer.
Chicago Press. Schrag, C. (1992). The resources of n1tiollality. Bloomington: Indiana University
Lather, P. (1986a). Issues of validity in openly ideological research: Between a rock Press,
,".,J "- ~'C)(l pl..tcc. Ifllenhallse, 17(4),63-8'1-. Schwandt, 1. A. (199{;). Farewell to critcriology. Qualitative Inquiry, 1, 58-72.
L~thcr, I~ (l98Gb), Research as praxis. NmHlrd Educational Review, 56, 257-177. Smith,]. (1985). Social re;:tlity a5 mind~dependent versus mind-independent and
Lather, P. (1993). Fenile obsession: Validity after poststructuralism. Sociological the interpretation of test validity. louwal of Research and Development ill
Quarterly, 34, 673-693. Educatiol1, 1) 1-9.
Larher, P. (1998a, April). Against empathy, voice, and authenticity. Paper presented Smith,]. (1989). The nature of social alld educational inquiry: E1I1j/iricis/Il uerSlls
at the annual meeting of the American Educational Research Association, illtelpretaiioll. Norwood, NJ: Ablex.
San Diego, CI\, Smith, J. (1993). After the demise of empiricism: The problem of judging social alld
Lather, P. (199Sb, April). v'ollbling IJTaxis: The worh of mourning. Paper presented educatiol1al inqlfil}l. Norwood, N]: Ablex.
at the annual meeting of the American Educational Research Association, Sparkes, A. (1993). Reciprocity in criticnl research? Some unsettling rhoughts. In
San Diego, CA. G. Shacldock &]. Smyth (Eds.), Beilrg reflexive in critical edllciltionalllild
Lad1l.'r, P., & Smithies, C. (1997). Tro1lbling the angels: "f/omel! livillg with HN/ social research (pr. 67-82). London: Falmer.
AIDS. Boulder, CO: Westview. Sparkes, A. (1996). The fatal flaw: A narrative of the fragile body-self. Qualitative
Leary, D. (1984). Philosophy, psychology, and reality. American Ps}'chologist, 39, Inquiry, 2, 463-494.
917-919. Stronach, I., & J'vlaclure, tvL CJ 997). Educatiollal research lIudollc. Ivlilron Keynes,
Lincoln, Y. S. (1995). Emerging criteria for quality in qualitative and interpretive England: Open University Press. .
inquiry. Ql/alitative Inquiry, 1, 275-189. Taylor, C. (1971). Interpretation and the sciences of man. RevielL' of Metaphysics,
MacKenzie, D. (1981). Stat.istics in Great Britain: 1885-1930. Edinburgh: Edin- 25, 3~51.
burgh University Press. Taylor, C. (1989). SOlfrces of the self. Cambridge: Cambridge University Press.
IVlanicas, P. (1987). A history al1d philosophy of the social sciences. Oxford: Basil Wolcott, H. F. (1995). The art of fieldworh. \Valour Creek, CA: AltaiVlira.
Blackwell.
IVlanicas, E, & Secord, P. (1983). Implications for psychology of the new philoso-
phy of science. American Psychologist, 38, 399-413.
Ivlaxwell, ]. (1992). Understanding and validity in qualitative research. Harvard
Educational RClIiew, 62, 279-300.
Nagel, T. (1986). The view {rolll /lowhere. New York: Oxford University Press.
Polanyi, M. (1961). Personal kl1owledge. Chicago: University of Chicago Press.
Popper, K. (1959). The logic of scielltific discovery. London: Hutchinson.
Popper, K. (1971). Objectiue browledge. Oxford: Clarendon.
Punch, 1\'1. (1994). Politics and ethics in qualitative research. In N. K. Denzin &Y. S.
Lincoln (Eds.), Handbook of qualitative research (pp. 83-97). Thousand
Oaks, CA: Sage.
Putnam, H. (1981). Reason, trllth and history. Cambridge: Cambridge University
Press.
Rotty, R. (1979). Philosophy and the mirror of !lature. Princeton, N]: Princeton
University Press.
Rorty, R. (1985). Solidarity or objectivity? In]. Rajchman & C. \'Vest (Eds.), Post-
analytic philosophy (pp. 3-19). New York: Columbia University Press.
Rosenau, P. M. (1992). Post-modemism (/ltd the social sciel1ces: Insights, inroads,
and hltrtlsioJls. Princeton, NJ: Princeton University Press.

456 457
 J Tie rTOctices ana {-'Dllt/CS ot Interpretation

I reckon it'll be

13
IVle myself!
-Lcng,ton H\I~h2'" ")'c" It'll Bc:: tA:,"j 940/1994'

have seen it
The Practices and and like it: The blood
the way like Sand Creek
even irs name brings fear,
Politics of Interpretation because I am an American
Indian and have learned
words are another kind of violence.

Norman f( Denzin

A marginalized group needs to be wary of the seductive

power of realism, of accepting al! that a realistic At the beginning of the end of the sixth moment it is necessary to
representation implies. reengage the promise of qualitative research and interpretive eth-
-\X1dhnccma Lubiano, "G\1l, Cornpd/;:d to W/hat 7 i\~ddin3 nography as forms of radical democratic practice. I The narrative turn in
RcaJism,Repl;?Senlrltioll, did Em:ntidbn in School Dd:C, tbe social sciences bas been taken, we have told our tales from the field,
Do the Right Tnlng, dnd the Spil:c Lee Discourse," 1997 and \Ne understand today that we write culture (Brady, 1998; R:charclson,
1998). \\lriting is not an innocent practice, although in tbe social sciences
and the humanities there is only interpretation (Rinehart, 1998). None-
You've taken my blues and gone- theless, 1Vlarx (1888/1983, p. 158) continues to remind LIS that' we are in
You sing 'em on Broadway the business of not just interpreting bllt changing the world.
... And yOlI fixed 'em In this chapter I explore new (and old) forms of writing, forms that are
So they don't sound like me. intended to forward the project of interpreting and changing the world-
\'es, yOli done taken my bl ues and gone. and this is the global world, not just the world as it is known in North
America. Specifically, I work back and forth among three interpretive
You also rook my spiritual and gone.
practices: the new civic, intimate, and literary journalisll1s (Charity, 1995;
... But someday somebody'll
Dash, 1997; Harrington, 1992, 19970, 1997b; [(ramer, 1995; Sims, 1995);
Stand up and write abour me,
And write about me- caUs for critical, performance-based ethnographies (Ceglowski, 1997;
B!<Jck and Beauriful- Cohen-Cruz, 1998; Degh, 1995; Denison, 1996; Donzin, 1997, 1999a,
And sing about mc, ] 999b, 1999c; Diversi, 1998; Dunbar, 1999;Jackson, 1998; Jones, ] 999;
And pur on plays about me! Jordan, 1998; Lincoln, 1997; Rinehart, 1998; Ronai, 1998; Smith, 1993,
1994); and variations on a Chicano/a (Gonzalez, 1998; Pizarro, 1998) and
African American aesthetic (Davis, 1998; hooks, 1990, 1996) and
AUTHOR'S NOTE: ! would like to thank Meaghan Morris, Ivan Brady, Yvanna Lincoln, J;JCk
Br;1l'ich, LJ.nrel Richudson, <lnd Karherine Ryan for rheir comments on e3rlier versions of rhis the relationship between these practices and critical race theory (Ladson-
chaprer. All <lsterisked marerialis used by permission. Billings, ]998; Parker, 1998).

458 459
 II'l I CIU-I\C IMI IVI'l, CYMLUMI IVI'!, AI'lLJ KCr-KC,:)CI'lIAIIUI'l
Although there have been effons to bring critical race theory into quali-
tative research, few have merged this theory with the poststructural turn
in ethnography (see hooks, 1990, pp. 123-134; also Ladson-Billings,
Chapter 9, Volume 1). Nor have critical race theory and qualitative
inquiry been connected to the radical performance texts stemming from
the black arts movement of the 1960s and 1970s (Baker, 1997; Baraka,
1997; Harris, 1998). These interconnections are now being established
in the various black cultural studies projects of the new black public in-
teHecruals and cultural critics (Hall, Gilroy, hooks, Gates, West, Reed,
IvIorrison, Wallace, Steele). A current generation of blues, rap, hip-hop,
and popular singers, jazz performers, poets (Angelou, Dove, Jordan,
Knight, Correz), novelists (\X1alker, Morrison, Bambara), playvvrights
(\vilson, Shange, Smith), and filmmakers (Lee, Singleton, Burnett, Dash)
are also making these links (see Christian, 1997, pp. 2019-2020; Harris,
1998, pp. 1344-1345).2
This chapter is a utopian project in which I attempt to bring these multi-
ple discourses together into a unified framework. I present examples of
writing from each these frameworks. In so doing, I assume that words and
language have a material presence in the world-that words have effects
on people. Amir; Saraka (1969/1998, p. 1502") puts ir this way:
we wam poems that wrestle cops into alleys
and take their weapons ...
We want a black poem. And a Black \\7orld.
Let the world be a Black Poem.
And Let All BLJck People Speak This Poem
Silently
Or LOUD.
\X!ords matter.
I imagine J world where race, ethniciry, class, gender, and sexual orien-
ration intersect; a world where language empowers and humans are free to
become who they can be, free of prejudice, repression, and discrimination
Gackson, 1998, p. 21; see also Parker, Deyhle, Vlilenas, & Nebeker, 1998,
p. 5). Those who write culture must learn to use language in a way that
brings people together. The goal is to create sacred, loving texts that
"demonstrate a strong fondness ... for freedom and an affEcrion~1te [nn
cern for the lives of people" Goyce, 1987, p. 344). This writing addresses
and demonstrates tbe benevolence and kindness that people should feel
to\vard one another Uoyce, 1987, p. 344).
460
I tit;' rrUl.-lIl.eJ UlIU ,UJI[ICS OT /nrerpreraflon
Thus I examine new ways of writing culture, new ways of making quali-
CHive research cemral to the workings of a free democratic society. I begin
with rhe civic, public affairs, Jnd intimate journalists.
An Intimate, Civic Journalism
As qualitative researchers engage experimental \vriring forms, a paraJIeI
movement is occurring in journalism, and there is much to be learned from
these developments. Building on earlier calls for a new journalism (\Volfe,
1973), a current generation of journalists (Harrington, 1997a, 1997b;
Kramer, 1995; Sims, 1995) is producing a new writing genre variously
termed literary, intimate, or creative nonfiction journalism (Harrington,
1997a, p. xv). This intimate journalism extends the project of the new jour-
nalism of the 19705. That project was based on seven understandings. The
new writers of \XJolfe's generation treared facts as social constructions.
They blurred writing genres and combined literary and investigative jour-
nalism with the realist novel, the confession, rhe travel report, and the
autobiography. They Llsed the scenic method to shm"" rather than tell. They
wrote about real people and created composite characters. They lIsed mul-
tiple points of vie\v, including third-person narration to establish authorial
presence, and deployed multiple narrative strategies (flashbacks, foreshad-
owing, interior monologues, parallel plots) to build dramatic tension. They
positioned themselves as moral witnesses to the radical changes going on in
U.S. society (Denzin, 1997, p. 131). These writers understood thar social
life and the reports about it are social constructions. Journalists do not
map, or report on, an objective reality.
[have no desire to reproduce arguments that maintain some distinction
between fictional (literary) and nonfictional (journalism, ethnography)
texts. Nor do I distinguish literary, nonliterary, fictional, and nonfictional
textual forms. These are socially and politically constructed categories.
They are too often used to police certain transgressive writing forms, such
as fictional ethnographies. There is only narrative-that is, only different
genrc-defined ways of represeming and writing about experiences and
their multiple realities. The discourses of the postmodern worlel coo-
R sr~ntly intel'n,ingle lite relIT, poet;.:;:, iourn,,!;st;C, Gct;onal, c;n",,_,-ua~ic, .JOCLt
mentary, factual, and ethnographic writing and representation. No form is
privileged over others. Each simply performs a different function for a
writer and an interpretive community.
461
 11'11 CKr-'KC IAIIUI'l, !:V/\LUAIIUN, AI'lU 1'(!:t"t'(!:':;CI'lIA1IUI'l
These practices and understandings shape the work of the intimate
journalists. Writers such as Harrington (1992) use the methods of descrip-
tive realism [0 produce in-depth, narrative accounts of everyday life, lived
up close. They use real-life dialogue, intimate first- and third-person
voice, multiple points of view, interior monologues, scene-by-scene narra-
tion, and a plain, spare style (Harrington, 1997b, pp. xlii-xlv; Kramer,
1995, p. 24). The writer may be invisible in the text or present as narrator
and participant. Here is Harrington (1992) talking about himself; the
swry is "Family Portrait in Black and \X1hite":
1'\i1)' journey begins in the dentist's chair. The nurse ... <md the doctor are
[tellingl funny stories about their kids, when in walks another demist....
"I've got a good one," he says cheerfully, and then he tells a racist joke. I
can't recall the joke, only that it ends with a black man who is stupid. Dead
silence. It's just us white folks here in the room, bur my de mist and his nurse
know my wife, who is black, and they know my son and daughter, who are,
as they describe themselves, tan and bright tan. How many racist jokes have
I heard in my life? .. for rhe first time ... I am struck with a deep sharp pain.
! look ar this man, with his pasty face, pale hair and weak lips, and I think:
This idiot is talking abour my children! (p. 1)
Comp3re this telling, with its first-person narration, to Leon Dash's
(1997) description of Rosa Lee:
Rosa Lee Cunningham is thankful thac she doesn't have to get up early this
illorning. She is doz.ing, floating back and forth betwecn sleep and drowsi-
ness. OccasiOl13l1y she hears thc muted conversations of the nurses and doc~
tors puttering around the nurse's station.... She's tired and worn down....
A full night's sleep and daylong quiet are rare luxuries in her life. This is the
closest she cver comes to having a vacation.... Rosa Lee ... is fifty-two
years old, a longtime heroin addict ... a member of the urban undetdass....
(She] has nO intention of ending her heroin use. (p. 3)
Harrington speaks only for himself. He is fully present in his text. Dash is
invisible. He is the all-knowing observer. He is the fly on the wall narrating
an unfolding scene. Dash describes a world, whereas Harrington talks
abom how it feels to be present in a world. Each writer cre:1tes a scene.
Each penetrates the images that surround a situation. Harrington and
Dash both use sparse, dean prose. Each creates a vivid image of his sub~
ject, I-L:trrington of himself, Dash of Rosa Lee. On the other hand, Dash
462
I IIC rlULLJ\..~~ UIIU "'U!lLlL~ Uf HlLe/jJfelUUUfl
presumes to know what Rosa Lee is feeling and thinking. Hers is a scory
\vaiting to be told, and he will tell it. In contrast, Harrington's text suggests
that scories are not waiting to be told; rather1 they are constructed by the
writer, who attempts to impose order on some set of experiences or per-
cei ved events.
Both writers ground their prose in facts and their meanings. Dash, how-
ever, works with so-called verifiable, factually accurate facts, whereas
Harrington writes of impressions and truths that, although not necessarily
factually accurate, are aesthetically and emotionally truc. If something did
not happen, it could have happened, and it will happen in Harrington's
text.
Accounts like Harrington's and Dash's illvoke the felt life. The goal
of such writers is to understand "ocher people's worlds from the inside
out, to understand and portray people as they understand themselves"
(Harrington, 1997b, p. xxv). The intent is to build an emotional relation-
ship joining the writer, the life told about, and the reader.
A year later, Harrington (1992) returns to his experience in the den-
tist's chair:
What r discovered while waiting in the dentist's chair marc than a year
ago ... still remains the greatest insight I have to share: The idiot was talking
abollt Ill)' kids!
I remember a time when my son was a baby. It was late at night ... I sar in
the dark of my son's room.... I watched his face grimace ... in thc shadows.
And then, in time so short it passed only in the mind, my son was gone and I
was the boy ... and my Cather was me.... just as suddenly, I was gone again
and the light was falling across the knees of my son, who was grown, who
was a father, who "vas me.... this kind of understanding changes every~
thing. Only when I became black by proxy-rhrough my son, through my
daughter-could I see the racism I had been willing to tolerate. Becoming
black, even [or a fraction of an instant, created an urgency for justice thar I
couldn't feel as only a whitc man, no matter how good-hearted.... no white
mall in his or her right mind would yet volunteer to rrade places, become
black, in America today. (p. 447)
Such writing connect,; readers to their neWRp:::tperR hy prodllc;ng l1.::l.rr::'l-
rives a60m people in extreme and ordinary situations. These stories, or
journalistic case studies, politicize the everyday world, illuminating the
structures and processes tIut shape individuals' lives and their relations
463
 11'111::1\,1\1:: 1/-\1 IVI'I, L. Y/-\I..UMIIUI'l, /-\I'ILJ 1\ I::FI\ 1::')1::!'1 IMI IVI'l
with orhers, In so doing, they "nmrure civic transformation" {Harrington,
1997a, p. xiv; see also Hatringron, 1997b, p, xviii).3
Civic Transformations
At the moment of civic transformation, intimate journalism joins with
the call for public journalism, a critical ethnographic journalism that fuses
persons and their troubles with public issues and the public arena. A prag-
matic, civic journalism invites readers to become participants, not mere
spectators, in the public dramas that define meaningful, engaged life in
society roday. Public journalism creates the space for local ethnographies
of problematic community and personal experiences. This is a socially
responsible civic journalism. It advocates participatory democracy. It gives
a public voice to the biographically meaningful, epiphanic experiences
that occur within the confines of the local moral comnmnity:l This form of
journalistic ethnography speaks to the morally committed reader. This is a
reader who is a coparticipant in a public project that demands democratic
solutions to personal and public problems (Charity, 1995, p. 146).5
Taken to the next level, transformed into public-journalism-as-ethnog-
r3phy, this writing 3nswers to the follO\ving goals. Critical, intimate 1 pub-
lic ethnography does the following things:
4' It presents the public with in-depth, intimate stories of problematic every-
day life, lived up close. These stories create moral compassion and help citi-
zens make intelligent decisions and take public 3ction on private troubles
that have become public issues, including helping to get these action pro-
posals carried om (Charity, 1995, p. 1; Mills, 1959, p. 8).
<} It promotcs interpretivc works that raise public and private consciousness.
These works help persons collectively \vorl( through the decision-making
process. They help isolate chOices, core values, utilize expert and local sys-
tems of knowledge, and facilitate deliberative, civic discourse (Charity, 1995,
pp.48).
* It rejects the classic model of investigative journalism, where the reporter ex-
poses corruption, goes on crusades, roots "om the inside story, tells the brave
truth, faces down the Joseph l\!icCarthy's and Richard Nixon's ... comforts
the afflicted and afflicts the comfortable" (Charity, 1995, p. 9).
~ It sceks the ethnographer and journalist who is an expert in the history and
public life of the local community, who knows how to listcn to and talk [0 citi-
zens, and how to hear and preselH consensus when it emerges, and who is also
464
J /le I-'raetlces and Politics of Interpretation
a full-time citizen and committed to the belief that public life can be made to
work (Charity, 1995, p. 10).
+ It sees the writer as a watchdog for the local community, a person who writes
stories that contribute to deliberative, participatory discoursc, thereby main-
taining the public's awareness of its own voice (Charity, 1995, pp. 104-105,
127).
' It values writing that moves a public to meaningful judgment and meaning-
ful action (Charity, 1995, p. 50). A cenrml goal is civic transformation
(Christians, Ferre, & Fackler, 1993, p. 14).
~ It exposes complacency, bigotry, and wishful thinking (Charity, 1995, p. 146)
while "attempting to strengthen the political community's capacity to under-
stand itself, converse well, and make choices" (Roscn, 1994, p. 381).
c> Ir seeks dramatic stories, narrativcs that separate facts from stories, telling
moving accounts that join private troubles with public issues (Charity, 1995,
p. 72; Mills, 1959, p. 8).
+ It promotes a form of texwalit)' that turns citizcns into readers and readers
into pcrsons who rake democratic action in the world (Charity, 1995, pp. 19,
83-84).
These are goals, ideals, ways of merging critical ethnography with applied
action research, with the new public journalism, and with qualitative re-
search in the seventh moment. (They presume the feminist, comll1Llni-
tarian ethical model discussed by Christians in Chapter 5 of Volume 1.)
These goals assume an ethnographer \vho functions and writes like a lit-
erary and intimate public journalist. This means that ethnography as a per-
former-centered form of storytelling will be given greater emphasis
(Degh, 1995 1 p. 8). A shared public consciousness is sought, a common
awareness of troubles that have become issues in the public arena. This
consciousness is shaped by a form of writing that merges the personal, the
biographical, with the public. Janet Cooke's (l980) fictional story
"Jimmy's \Votld" is an instance of such writing. Such stories expose com-
placency and bigotry in the public sphere.
Writing Norms
Afeminist, communirarian ethical model produces a series of norms for
the public ethnogr3 p hic writing project.r, Th~$e norrr>'; bulld on ",-nd cb.6o-
rate the four nonnegotiable journalistic norms of accuracy, nonmal-
eficence, the right to know, and making one's 1110ral position public. 7 The
ethnographer's mora! tales arc not \vritten to produce harm for those \vho
465
 INlI:Rl-'H.1: fA! IUN, I:VALUAIIUI'I, ANU Ktl-'Kt~tl'i lAIIUI'l
luve been oppressed by the culture's systems of domination and repres-
sion (the principle of nonmaleficence). The identities of those written
J.bollt should always be protected. These tales are factually and fictionally
correct. H \Vhen fiction, or imaginative narrative, is written, or when com-
posite cases are molded into a single story, the writer is under an obligation
to report this to the reader (see Christians et aL, 1993, p. 55).
The reader has tbe right to read what the ethnographer has learned,
but the right to know should be balanced against the principle of
nonmaleficence. Accounts should exhibit "interpretive sufficiency"
(Christians et aI., 1993, p. 120), that is, they should possess depth, detail,
emotionality, nuance, and coherence. These qualities assist the reader in
forming a critical interpretive consciousness. Such texts should also
exhibit representational adequacy, including the absence of racial, class,
and gender stereotyping.'}
The wtiter must be honest \vith the reader. lO The text must be realistic,
concrete as to character, setting, atmosphere~ and dialogue. The text
should provide a forum for the search for moral truths about the self. This
forum may explore the unpresentable in the culture; the discontents and
violence or contemporary life are documented and placed in narrative
form. This writer stirs up the world, and the writer's story ("mystory")
becomes parr of rhe tale that is told. The writer has a theory about how the
world works, and this theory is never far from the surrace of the text. Self-
reflexive readers are presumed-readers who seek honest but reflexive
works that draw rhem into the many experiences of daily life.
There remJins the struggle to find a narrarive voice tharwrites against a
long tradirion rhat favors autobiography and lived experience as the sires
for reflexivity and self-hood (Clough, 1994, p. 157). This form of subjec-
tive reflexivity can be a trap. It too easily reproduces sad, celebratory, and
melodramatic conceptions of self, agency, gender, desire, and sexuality.
There is a pressing need to invent a reflexive form of writing thar turns eth-
nography and experimenralliterary texts back "onto each other" (Clough,
[994, p. 162; 1998, p. 134).
Always a skeptic, this new writer is suspicious of conspiracies, align-
ments of pO\vcr and desire that turn segments of the public imo victims. So
thest' works trouble traditional, realist notions of truth and verification,
asking always who stands to benefit from a particular version of the truth.
The intimate journalist as public ethnographer enacts an ethics of practice
that privileges the cliem-public relationship. The ethnographer is a moral
advocate for the public, although their own personal moral codes may
466
r r I~ r-rucnces ana {-'OlrtiCS at Interpretation
lead individual researchers to work against the so-called best interests of
their cliems or particular segments of the public.
The ethnographer's tale is always allegorical, a symbolic tale, a parable
that is not just a record of human experience. This tale is a means of expe-
rience, a method of empowerment for the reader. It is a vehicle through
which readers may discover moral truths about themselves. lviore deeply,
the tale is a utopian story of self and social redemption, a tale that brings a
moral compass back into the reader's (and the writer's) life. The ethnogra-
pher discovers the multiple "truths" that operate in the social world, the
stories people tell one another about the things that matter to them. The
intimate journalist writes stories that stimulate critical public discourse.
Thus these stories enable transformations in the public and private spheres
of everyday life.
<l> Performing Ethnography
I turn next to the concept of the performance text (Conquergood, 1992j
Turner, 1986), illustrating my arguments with materials drawn from an
ongoing interpretive ethnography or a small Ivlontana town (Denzin,
1999a, 1999b, 1999c).1l I seek a set of writing practices that turn notes
from ~he field into texts that are performed. A single, yet complex thesis
orgamzes my argument.
. \"X7e inhabit a performance-based, dramaturgical culture. The dividing
hne behveen performer and audience blurs, and culture itself becomes a
dramatic performance. Performance ethnography enters a gendered
culture with nearly invisible boundaries separating everyday theatrical
performances from formal theater, dance, music, MT\!, video, and film
(Birringer, 1993, p. 182; Butler, 1990, p. 25; 1997, p. 159; 1999, p. 19).
But the matter goes even deeper then blurred boundaries. The- perfor-
mance has become reality. Of this, speaking of gender and personal iden-
tity, Butler (1990) is certain. Gender is perfonnative, gender is always
doing, "though not a doing by a subject \vho might be said to preexist the
deed.... there is no being behind doing.... the deed is everything....
there is no gender identity behind the expressions of Bender.... identit;, is
perforrnatively constituted by the very 'expressions' that are said to be its
results" (p. 25). Further, the linguistic act is performative, and words can
hurr (Butler, 1997, p. 4).
467
 I,~TERPRETATIOH, EVALUAIIUI\I, AI"'lLJ I\Crr,t...J .... "
._---
,,..,, .~

to ride the bike ...

Performance textS arc situated in complex systems of discourse, where
traditional, everyday, and :IVant-garcle meanings of theater, film, video,
ethnography, cinema, performance, text, and audience all circulate an(~ in-
form one another. As Collins (1990, p. 210) has suggested, the meanmgs
of lived experience are inscribed and sometimes made visible in these per-
formances (sec also Brady, 1999, p. 245).
Anna Deavcrc Smith's Fires in the Mirror (1993) is an example. In this
play, Smith offers a series of performance pieces based on interviews with
people involved in a racial conflict in Crown Heights, Brooklyn, on
August 19,1991. The conflict was set in motion when a young black Guy-
e
anese boy, Gavin Cato, was accidentally killed by an aura in a polic -
escorted entourage carrying Lubavitcher Grand Rebbe Ivlenachem
Schneerson. Later that day, a group of black men fatally stabbed Yanke!
Rosenbaum, a 29-year-old Hasidic scholar from Australia. This killing was
followed by a racial conflict that lasted 3 days and involved many members
of the community. A jury acquitted Yankel Rosenbaum's accused mur-
derer, causing considerable pain for his family as well as feelings of victim-
ization for the Lubavitchers (Smith, 1993, p. Xiv).12 Smith's play has
speaking parts for gang members, police officers, anonymous young ~irls
and boys, mothers, fathers, rabbis, the Reverend Al Sharpton, playwnght
Ntozakc Shange, and African American cultural critic Angela Davis.
Cornel \':(!est (1993) observes that Fires ill the lvIirror is a "grand exam-
ple of how art can constitute a public space that is perceived by people
as empO\vering rather than disempowering" (p. xix). Thus blacks, gang me~
bers, the police, and the Jewish community all come together and talk 111
this play. The dram:J crosses racial boundaries. Smith's te:t shows that
"American character lives not on one place or the other, but 111 the gaps be-
tween places, and in our struggle to be together in our differences" (p. xi~).
An Anonymous Young Nian #1 Wa \Va \X1a, a Caribbean American wlth
dreadlocks, describes the auto accident:
\'x/hat I saw was
she was pushin'
her brother on dlC bike like
this,
right?
She \VD.S pushin'
him
and he keep dippin' around
like he did~1't know how
468
So she was already runnin'
when the car was com in' ..
we was w;Jtchin' the car
weavin',
and we was gain'
"Oh, yo
it's a Jew man.
He broke the Stop light, they neVer get
arrested." (Smith, 1993, pp. 79-80)"
And so in performing this young man's words, Smith contexrualizes this
drama, showing how it looked from the standpoint of a person \\'ho
watched the accident unfold.
Performance ethnography simultaneously crt;ates and enacts moral
texts, texts that move from thc personal to the political, the local to the
historical and the cultural. Following Conquergood (1985), these
dialogical works create sp:1Ces for give and take, doing more than turning
~h~ other into the object of a voyeuristic, fetishistic, custodial, or paternal-
IStiC gaze.
Texts turned into radicJl street performances act to question and "re-
envision ingrained social arrangemems of power" (Cohen-Cruz, 1998,
p. 1). Such works, in the forms of rallies, puppet shows, marches, vigils,
choruses, clown shows, and ritual performances, transport spectators and
performers out of evcryd3Y reality into idealized spaces where the taken
for ~ranted is co~tested (Cohen-Cruz, 1998, p. 3). Street, or public place,
per.ton~1allces oHer members of the culture alternative scripts or ways of
actIng in and hence of changing the \vorld (Cohen-Cruz, 1.998, p. 1.).
Cohen-Cruz (1998, p. 5) suggests that these performances can take
several overlapping forms, including agit-prop, or attcmpts to mobilize
people around a partisan view; witnessing, or making a spectacle out of
an act that perhaps cannot be changed; can(ralltatiolt, or inserting a per-
formance into people's everyday life, thereby asking them to confront a
sc.enario that is otherwise distant; utopia, or enacting an idealized version
ot ~eality; and tradition, that is, honoring a set of culturally shared beliefs,
as 10 Founh of July parades in small-town America. -
use1n my performance pr~jeet I seek l11il~imalist social scie~ceJ one that
S few concepts. TIllS IS a dramaturgical (Branaman, lYY7. p. xliX:
U
Goffman 1959 Lcmert, 1997, p. XXIV
,~., . ) or per formatlve
. 3nthropology
.
ol1
acks , 1998; Turner, 1986) that attempts to stay close to how people
469
 Ii'll tlI-'KI: lAllUl'l, I:VALUAIIUf'l, AI'fU 1,(1:1-'1\1:.)I:I'lIAI IUI'l
represent everyday life experiences. A performative edmography simulta-
neously writes and studies performances, showing how people enact cul-
rural meanings in thcir daily lives.
Shaped by the sociological imagination (lvlills, 1959), this version of
qualitative inquiry attempts to show how terms such as biography, gcndcl;
race, etlmicity, (amdy, and history interact and shape one another in con-
crete social situations. These works are usually written in the first-persall
voice, from the point of view of the sociologist doing the observing and
the writing. A minimalist, performative social science is also about stories,
performances, and storytelling. \X!hen performed well, these stories create
a ritual space ''Ivhere people gather to listen, to experience, to better un-
derstand the \vorld and their place in it" Uenkins, 1999, p. 19).
The Performance Turn
The performance turn in thc human disciplines (Bochner & Ellis, 1996;
Conquergood, 1992) poses three closely interrelated problems for a criti-
cal, interdisciplinary interpretive project-namely, how to construct, per-
form, and critically analyze performance texts (see Stern & Henderson,
1993). Glossing the issues involved in construction and critical analysis, I
vvill privilege performance and coperformance (audience-performer)
texts in contrast to single performer, text-centered approaches to intcr-
pretation (see Denzin, 1997, p. 96). Through the act of coparticipation,
these works bring audiences back into the text, creating a field of shared
emotional experience. The phenomenon being described is created
through the act of represemation. A resistance model of textual perfor~
mance and imerpretation is foregrounded. A good performance text must
be more than ea-thartic, it must be political, moving people to action and
reflection.
The attention to performance is interdisciplinary; sociologists
(Bochner & Ellis, 1996; Clough, 1994; Denzin, 1997; Ellis, 1997; Ellis &
Bochner, 1992, 1996; Ellis & Flaherty, 1992; Kotarba, 1998; Richardson,
1997), anthropologists (Behar, 1996; Brady, 1999; Bruner, 1986, 1989,
1996; Cruikshank, 1997jJackson, 1998; Turner, 1986), communication
scholars (Conquergood, 1992; Hill, 1997), and education theoristS
(Lather, 1993; Lather & Smithies, 1997; Lincoln. 1995:1, -199Sb, 1997;
-[lerney, 1997) are calling for texts that move beyond the purely represen-
Lltional and toward the presentational. At the same time, action (Stringer,
1996), communitarian (Christians et al., 1993), feminist (Lather, 1.993),
470
'''~ nULlILl::'~ UflU ,UIIUCS OT mrerprerouon
constructivist (Lincoln, 1995a, 1995b, 1997), cooperative inquiry
(Reason, 1993, 1994), and participant researchers (Carspecken, 1996)
are exploring nontraditional presentational performance formats. Such
works allow community researchers and community members to co-
construct meaning through ;Ierion-based performance projects (Stringer,
1996; see also Conquergood, 1998; Schwandt, 1997, p. 307). This call
merges a feminist, coml11unitarian ethic with a moral ethnography that
presumes a researcher who builds collaborative, reciprocal, trusting,
friendly relations widl those studied (Lincoln, 1995a).
Performed texts "have nan'arors, drama, action, shifting points of
view ... [and] make experience concrete, anchoring it in the here and
now" (Paget, 1993 p. 27; see also Donmoyer & Yennie-Donmoyer, 1995;
Mienczakowski, 1995). Centered in the audience-researcher nexus, these
texts are the site for "mystories" (Ulmer, 1989); that is, teflexive, critical
stories that feel the sting of memoty, stories that enact liminal experiences.
These are storied teteHings that seek the truth of life's fictions via evoca-
tion rather than explnnation Ot analysis. In them ethnographers, audi-
ences, and performers meet in a shared field of experience, emotion, and
action.
Such performances return to memory, not lived experience, as the site
of criticism, interpretation, and action. It is understood that experience
exists only in its represenration, it does not stand outside memory or per-
ception. The meanings of bets are always reconstituted in the telling,
as they are remembered and connected to other events. Hence the appeal
of the performance text docs not lie in its offer of the certainty of the fac-
tual. The appeal is more complicated than that. Working from the site of
memory, the reflexive, performed text asks readers as viewers (or co~
performers) to relive the experience through the writer's or performer's
eyes. Readers thus move through the re-created experience with the per-
former. This allows them to relive the experience for themselves.
Thus we can share in Harrington's experiences when he tells us about
remembering the night he held his youngson in his arms. The writer acting
in this manner fe-creates in the mind's eye a series of emotional moments.
Life is then retraced through that moment, interpreting the past from the
point of vie\v of the present. Here is Susan Krieger (1996):
I have just come back from a trip to Florida to settle the affairs of my lover's
aunt [Maxine], who died suddenly at the age of seventy. She was carrying
471
 !I'll t:1"..Vt"..t: 1P'IIUl'l, t: VP.LUP.IIUI'I, HI'ILI 1-,LrI'\LJL1~ 1M' lV"
her groceries up the stairs to her apartment when she dropped dead of a
heart dttack...
. . . it was an Olherworldly experience: going to Florida ... ro clean out
the house of a woman I did not know-:;orting through her clothes and jew-
elry, finding snapshots she recently took, using her bathroom, meeting her
friends. (pr. 65, 68).
On her last day in Florida, Krieger finds Aunt Iv1axine's silver flatware in-
side an old ac~ordion case in the back corner of a greasy kitchen cabinet.
Ivlaxine's silver is cheap. It had been a replacement set. It is tarnished, not
the real thing like the family silver her mother had gotten from her mother.
Krieger ask:s, "\\?hat determines the value of a person's life, is value differ-
ent if you are a woman, how do you separace a woman from the things she
owns, leaves behind, her clothes, the cheap family silver?" (p. 70). This
cheap set of silver flatware is not an adequate measure of Aunt 1Vlaxine.
L.1lIrei Richardson's mother died of breast cancer in 1Vllami Beach on
June 8, 1968. Nearly 30 years later, Richardson (1997) wrote about her
mother's death:
On June 8, 1 <lwoke determined to drive to Key West for the day. I sponge-
bathed Mother, greeted the nursc, kisscd my parents good-bye, and drove
off in Father's Dodge Dart. When I got to Long Key, I was overwhelmed
\vith the need to phone my parents. Mother had just died-less than a min-
ute ago-Father said. (p. 234)
Of this ncws she obsetves, ''"I was grateful to have not witnessed her pass-
ing-on. I was thirty years old" (p. 235).
Ten years after her mother's death, Richardson wrote a poem, "Last
Conversation" C1997, p. 234):
r want to hold your
weightless body to my
breasts, cradle yOll,
rock you to sleep.
The truth of:1 person's and a culture'~ W:1Y~ are given in texts like [hese.
Such v.,rorks, when performed or read, become symbolic representations of
"vhat the culture and the person values. In their performances, performers
embody these valucs.
472
I '!O;:' n ULLIL/'O':' unu ,ol/ncs aT mrerprerQnon
Novo[ listen to Angela Davis. In her recent study of the female blues sing-
ers .Ma Rainey, Bessie Smith, and Billie Holiday, Davis (1998) observes
that within African American culture the "blues marked the advent of a
popular culture of performance, with the borders of performer and audi-
ence becoming increasing differentiated.... this ... mode of presenting
popular music crystallized into a performance culture that has had an
enduring influence on African-American music" (p. 5).
In their performanccs, 1\11<1 Rainey, Bessie Smith, and Billie Holiday pre-
sented a set of black feminist understandings concerning class, race,
gender, violence, sexuality, marriage, men, and intimacy (Collins, 1990,
p. 209). This blues legacy is deeply entrcnched in an indigenous, c1ass-
conscious, black feminism. The female blues singers created performance
spaces for black "vomen, spaces to sing and live the blues, black female
voices talking about and doing black culture on stage, a public, critical
performance art (see Jones, 1999).
The blues are improvised songs sung from the heart about love and
about women and men gone \vrong (Collins, 1990, p. 210). Rainey, Smith,
and Holiday sang in ways that went beyond the written text. They turned
the blues into a living art form, a form that would with Holiday move into
the spaces of jazz. And jazz, like the blues and culture, is an improvisa-
tional, not static, art form. Bill Evans, the great bebop pianist, put it this
way, "Jazz is not a what ... it is a how. If it were a what, it would be static,
never growing. The how is that the music comes from the moment, it is
spontaneous, it exists at the time it is created" (quoted in Lee, 1996,
p. 426). The improvised co performance text, the jazz solo, like Billie
Holiday singing the blues, is a spontaneous production, it lives in the
moment of creation. The how of culture as it connects people to one
another in Joving, conflictual, and disempowering ways is what perfor-
mance ethnography seeks to capture (see Jackson, 1998, p. 21).
~ Red Lodger Montana: E"periences and Performances
\ViIliam Kittredge (1996, p. 97) says that the West is an enormous empty
and innocent stage waiting for a performance (see also Kittredge, 1987,
1994). He conti'1ues, "We Gee tl,c !,;story of OL,t pcr(orn.ancc;-. ev",ry_
where ... inscribed on the landscape (fences, roads, canals, pmver lines,
city plans, bomb ranges" (p. 97). 1v10reover, the West is comained in the
stories people tell about it. l.vlontana is both a performance and a place for
473
 11'11 l:KYKl: IAIIUN, t:VI-'.LUHIIUI'I, HI'<U Kt:r'I'<t:2It:f'IIHI IUI'l
According to local history, the festival started shortly after \Vorld \Var
II. Local community leaders decided to build a civic center. They \vanted a
place ror community activities, for an annual summer music festival, for
an and craft exhibits and the display of flags and cultural items from each
of the nations represented in the town. Thus \vas born the Festival of
l'\btions, and Red Lodge was soon transformed into a "wurist wwn, a
place that offered good scenery, fine fishing, the I-Ii-Road, the rodeo, cool
summers" (Red Lodge Chamber of Commerce, 1982, p. 4). But the town
had more to offer: "The inhabitJots themselves were a resource" (p. 4).
For 9 days every summer, Red Lodge puts its version of local ethnic culture
on display, mming everyday people into performers of their respective
ethnic herit:lges.
In these performances, residents only had their own histories to go on,
so they made it up as they went along, one day for each ethnic group, but
each group vvould do pretty much the same thing: a parade down Broad-
\vay with national flags from the country of origin, people in native cos-
tume, an afternoon performance of some sort (singing, storytelling, rug
making), ethnic rood in the Labor Temple (every day from 3:00 to 5:00
p.m.), an evening of music and dance in the civic center. This is improvised
ethnicity connected to the performance of made-up rituals handed down
from one generation to the next, white privilege, white cultural memory,
lvfonrana style (see also Hill, 1997, p. 245).
Montana Jazz
I like lvlomana Day, especially that part that involves ranch women
reading their stories about being Montana wives. This is improvised
theater, cowgirl women singing their version of the blues-lvlomana jazz.
A tent is set up over the dance pavilion in Lions Park, just next to the Depot
Gallery, which is housed in a converted red train caboose, across the street
from the Carnegie Library. \'\lhite plastic chairs are lined up in rows on the
lmvn. Loudspeakers are on each side of the stage. An old-fashioned lvIon-
tan a rancher meal is served afterward: barbecued beef, baked beans, cole-
slaw, baked potatoes, ]ell-o salad, brownies for dessert, coffee and ice rea
for beverages.
1:111 and shan middle-aged suntanned women take the stage. -rhese are
h:1flhvorking women, mothers, daughters, and grandmothers who live in
the ranches in the Beartooth Valley, along the East Rosebud, Rock Creek,
\Villmv Creek, and Stillwater River south of Columbus and Absarokee.
476
111e' rIUl..Ul..t:'~ Uf!U r-UII[/CS or Interpretation
Some have on cowboy boots, others wear Nike sneakers. They are dressed
in blue jeans and decorated red, white, and blue cowboy shirts and skirts,
and have red bandannas around their necks. They have shan and long
hair, ponytails and curls. They read cowboy poetry, tell shorr scories aboU[
hard \vinters and horses that freeze to death in snowbanks. Some sing
songs. Country music plays quietly, and people come and go, some stop-
ping and listening for a \vhile. As (hese women perform, old ranchers in
wide-brimmed cowboy hats dose their eyes and tap their feet to the music.
Young children run around through the crowd, and husbands proudly
watch their wives read their poerry. Like .NIa Rainev Bessie Smith and
Billie Holiday, these women sing their version of th~ 'cowgirl blues.' And
there is a certain firm trurh in the way they do this.
Dead Indians
Fiedler (1988) argues thar IvIonrana as a white territory became psycho-
logically possible only after the Native Americans, the Nez Perce, Black-
feet, Sioux, Assinboine, Gros Ventre, Cheyenne, Chipewayan, Cree, and
Crow were killed, driven away, or forced onto reservations. He asserrs
thar the struggle to rid the \X!est of the "noble savage" and the "redskin"
\vas integral to the myth of the .Nlomana frontier as a wild wilderness
(p. 745). According co Fiedler, the Indian was lVIontana's Negro, an OU[-
cast living in an openMair ghetto (p. 752). \'\lith the passing of the redskin
came trappers, mountain men, explorers, General Custer, Chief Joseph,
and then ranchers and homesteaders. Indian sites were marked with
names such as Dead I ndian Pass.
Entering Red Lodge, you drive past a \vooden statue of a Native Ameri-
can, a male Indian face sitting on a big stone. It was carved by a non-Native
American. The history of lvlomana's relations with the Native American is
folded into mountain man festivals and events like lvlontana Day in Red
Lodge. It is a hiscory that simultaneously honors the dead Indian while
denying the violent past that is so central to white supremacy.
Two summers ago, \ve drove over Dead Indian Pass, taking the Chief
Joseph Highway back from Yellowstone Parle Chief Joseph led rhe Nez
Perce over this pass just momhs before Custer's last stand. At the top of the
mounD.in rhere i~:l tllrnoff \VhC'l:"'_ you CCln stop .and lool<- hack nerO"" d~",
valley, 10,000 feet below. Although this place is called Dead Indian Pass,
there is no mention of Chief Joseph on the little plaque at this viewing
spor. Instead, there is a stOry about the ranch families who settled the
477
 t fit' r'raetlces and Politics of Interpretation

valley while fighting off the Indians-hence tl~e name Dead Indi~n Pass. rights Chicano/a and black arts cultural movemenrs into the next century
The ~vhites, anunot the Indians, are honored III popular mellOr). (sec Harrington, 1992, p. 208). These practices serve to implemenr criti-
cal race theory (Anzaldlla, 1987; Collins, 1990; Davis, 1998, p. 155;
Performing Montana Gonzalez, 1998; hooks, 1990, p. 105; Joyce, 1987; Ladson-Billings, 1998;
Parker, 1998; Parker et aL, 1998; Scheurich, 1997, pp. 144-158; Smith,
f M t na lvlontana is a place where
There are many \vays to per Ort~ on a . O' thrted fly fisher- 1993, p. xxvi; Smith, 1994, p. xxiii Trinh, 1992), Critical race theory
' I ' 11lmgle W l1cre rVls-ou
lo'c-'lls '111d toUflstS constant y c a n , 1'1 1 II 'ds wllo "seeks to decloak the seemingly race-neutral, and color-blind ways ... of
. , .. , I, F I 1'- I ze oca Cl
men from Connecticut connect wIth Hue... 1IlI1 00 ... a ' es mix constructing and administering race-based appraisals ... of the law, admin-
I-I ~ ~ p ales These lvlontana pel. forllanc - istrative policy, electoral politics ... polidcal discourse [and education] in
fish with worms and 0 L lamloo .. off 'd 'tOes differem
\.1D manv different things 3t the same tlme-dl erent I cntt I,. ' _ _ the USA" (Parker et aI., 1998, p. 5).
'II . b. 'leas chssical music, antique hunters, skiers, mou.n Thus Collins's (1990, 1998) Afrocentric feminist agenda for the 1990s
se ves: cow oys, rm , " . 1 .' wives who write
' men, F'lllill'sh
t3Ul , women who make rugs, ranc lers is moved into the next centurYi that is, theorists and practitioners enact a
poetry, fi~hing for .trout~_,
standpoint epistemology that sees the world from the point of view of
t of the lViontana sclf. .ivly wife and I enact
Being In nature IS a major par f b d' 1 T down and oppressed persons of color. Representative sociopoetic and interpretive
_. b' .' to ourselves through the very act a en 11 g works in this tradition include those of Baraka (1969/19981,Jones (19661,
naturc; \Vc flng It . I ,I . This contact with the
' 'ld'I'" f wallong a ong t le flver. Shange (1977), Joyce (1987), Neal (1988), and Jordan (1998), as well as
smelllllg a WI lower, a _ , rIo a iEt" (Kittredge,
"natural world is an experience that comes t~ us l ... e d ~Iace a hole at rhe more recent arguments of hooks (1990,1996), Smith (1993, 1994),
1996 D. 108). And so our corner of lvlonrana IS a sacre d D " ('1978') a Davis (1998), Anzaldua (1987), Noriega (1996), and others. This aes-
- ". I" d (1994) an Olg "
house of sky, to use phrases from \.Ittre ge - n when we perform thetic is also informed by the sllccessive waves of activism among Asian
place where wonderful things happen, and they happe Americans and Native Americans, women, and gays, lesbians, and bisexu-
als who "usc their art as a weapon for political activism" (Harris, 1998,
them.
~ll re the kinds of things a mllllma ,. I'1St, storyte 11'109, P erformative
p. 1384; also Nero, 1998, p. 1973).
.1ese a . . . 1 I ll' s the world comes
form of qualitative inqUIry makes vlSlb e. In t leselte ling . gs avail- Theorists critically engage and interrogate the anti-civil rights agendas
alive. The qualitative researc I1cr attempts.. . to rna ...'e, t lese
'f ffieJIlln
[ iological of the New Right (see Jordan, 1998), but this is not a protest or inte-
able to the reader, hoping to show how thIS versIOn o. t le soc. ut grationist initiative aimed solely at informing a white audience of racial
~ImagmatIon
. '. -'~s some of the things that matter in everyday hfe. B
engage: . l i t ' sruo- injustice. It dismisses these narrow agendas. In so doing it rejects classical
'.
sometimes war d's felil
... , for in its naturalness 1Vlonrana IS a pace t 1a IS Eurocenrric and postpositivist standards for evaluating literary, artistic,
nin T in its beauty, a world of images beyond \~~rcls. . . lore and research work.
g 1 .' f color l5 and CrItIcal race theory. I exp
I turn now to an aest letlC o f ' to criti- The following understandings shape this complex project:
' '- I race
I10W cnnca "
' - tlleory can use expeumel l~al
l~ forlTIs
. , 0 ' narratlve

cize and resisr racist cultural practices.

<) Ethics, aesthetics, political praxis and epistemology are joined; every act of
representation, arristic or research, is a political and ethical statement (see
Neal, 1988/1998, p. 1451). There is not a separate aesthetic or epistemo~
{, An Aesthetic of Color and Critical Race Theory
logical rcalm regulated by ttanscendent ideals, although an ethics of care
should always be paramOUnt.
b d - I (c uses ,In,
A feminist. Chicano/a, anti black pe~fo.[manlce- ase. aestlepelrformance 4 CL:,m ,'-' tnnh i.l1ll1 !mowlcdge are assessed in terms of mulriple eriter]::!,
. d
photography, mUSiC, ance, poe 1 , P ty amung 1
tleater (merna,
'. d 'IClnguag including asking if a text (a) interrogates existing cultural, sexist, and ~a~i;]l
.
'e' X'~S 'llltobio1rraphv, . st oryTelling-~, and poenc, rarnanC
narrative, L I '" t-cl'vile .
stereotypes, especlallr those connected to faml'I y, f emmml},.
.. ' t ' masculinity'
L "L " 0 , l b . d g t lC pas
to create a critical race conSCIousness, t lere y exten In ,
marnage, I 1988/1998, p. 1~157)'
an d"intimacy (l"Jea, _, , (b) gives pnmacy' ro
478 ~t79
 "" ... ", ,,~,n' '''''''"1 ... " , ... _, . ' __ L'~~~ " " ' U ' .... "u'-" VI "'"O::'fJ'CU.J",VII

concrete lived experience; (c) uses dialogue and an ethics of personal respon- The difference between poetry :1lld rhetoric
sibility, values beamy, spiriLUality, and love of others; (d) implements an is being .
emancipatory agenda committed to equality, freedom, socia! justice, and par- ready to kill
ticipatory democratic practices; and (e) emphasizes community, collective yourself
action, solidarity, and group empowerment (Dcm.in, 1997, p. 65; hooks, instead of your children.
1990, p. I! I; Piz,:mo, 1998, pp. 63-65).
{,o No tOpic is taboo, including sexuality, sexual abuse, death, and violence. In lin~s such as these, critical race theory is connected to :.1 heightened
{> This project presumes an artist and social researcher who is part of, and a refleXive and moral sense of race consciousness. Lorde (1973/19983,
spokesperson for, a local moral community, a community with its own symbol- pp. 1626-1627r' again, this time from "Coal":
ism, mythology, and heroic figures.
<} This projecr J.sb that the writer-artist draw upon vernacular, folk, and popu-
lar cultllre forms of representation, including proverbs, work songs, spiritu- is the total black, being spoken
from the earth's inside.
als, sermons, prayers, poems, choreopoems (Shange, 1977), folktales, blues
There are many kinds of open
{Davis, 1998), jazz, rap, film, paintings, theater, movies, photographs, per-
how a diamond comes into a knot of flame
formance art, murals, and corridos (see Fregoso, 1993; Gates & l\'lcKay, how a sound comes into a word colouted
1997, p. xxvii; Hill, 1998; Noriega, 1996; Pizarro, 1998, p. (5). by who pays wh:lt for speaking _..
<~ This project includes a search for texts that speak to women and children of
color and to persons who suffcr from violencc, ra~e, and racist and sexist Love is a word, another kind of open.
injustice. As the diamond comes into 3 knot of flJ.mc
4- This project' seeks anists-researchers-\vriters who produce works that speak I am black because I come from rhe c3rth's inside
to and represem the needs of the community (e.g., concerning drug addic- Now talce my word for jewel in the open light.
tion, teenage pregnJ.ney, murder, gang warfare, AIDS, school dropout).
9 It is understood, of course, that no single representation or work can speak to Lorde (1986/1998(, pp. 1630-1(31)"" once more, from her poem "Sta-
the collective needs of the community. Rather, local communities are often tions":
divided along tacial, ethnic, gender, residential, age, and class lines.
Some women love
to wait
for life for a ring
Thus this project seeks cmallcipawry, utopian texts grounded in the dis-
for a touch ... for another
tinctive styles, rhythms, idioms, and personal identities of local folk and
W01TI3n'S voice to make them whole ...
vernacllbr culture. As historical documents, these texts record the histo- Some women wait [or something
ries of injustices experienced by the members of oppressed groups. They to change and nothing
show how members of local groups have struggled to find places of dignity does change
so they d13ngc
and respect in a violent, racist, and sexist civil society (see Gates & 1v1cKay,
themselves.
1997, p. xxvii).
These texts are sites of resistance. They are places where meanings, pol-
In "Taking in Wash," Rita Dove (1993/1998, pp. 1966-1967)* speaks
itics, and identities are negotiated. They transform and challenge all forms
to drinking, drunkenncss, and violencc in rhe house:
of cultural representation: white, black, Chicano/a, Asian Americ3J1,
Native American, gay or straight.
Pap3 called her Pe3rl when he came home
In her poem "Power," black lesbian poet Audre Lorde (197S/1998b, drunk, sWJying as if the wind touched
p. 1627r talks about t3king action in the world: only him ... Mama never changed:

480 481
 lI'l I l::rl.rrl.l:. IMI IU"I,. L. VKL.UMI 'U'~, ,...." ~'--' '~L.' ,~'--''-'' ,,....,, ''""'"
, '''.. , 'u'-co,-",,, UIIU rUlltlL.'> UI Irrt~rfJrf:!tunon

when the dog crawled under the stove lady in brown

:llld the back gate slammed, !Vlama hid & this is for colored girls who have considered
the laundry ... Papa is making the hankies suicide/bur are mavin to the ends of their own
sail. Her foot upon a silk rainbows
stitched rose, she wailS
he turns, his smile sliding over.
II/lama a tight dark fist. And so the world is taken back, and we dream of another day of lighmess
TOl/ch that child and being free, of making our own rainbow.
And I'll CIt! yOIl dOll./II
jt/S! like the Cedar of Leballol1. Aesthetics Gnd Cinemotic Practices

Ntozake Shange's (1977r powerful, Obie Award-winning choreo- \Vithin the contemporary black and Chicano/a film communities, there
poem For Colored Girls 1X/ho Have Considered Suicide W/hen the Rainbow are specific sets of film practices associated with this aesthetic project. Jf,
Is EllUl encls this way. Six of the seven women in the company, the ladies in These practices inform and shape the narrative and visual content of these
red, purple, orange, blue, green, and yellow, have spoken. The lady in red experimental texts. They include the following:
summarizes (pp. 60, 63-64):
Experiments with narrative forms, folk ballads, and corridos that honor
bdy in red long-standing Chicano/a discourse traditions (Fregoso, 1993, pp. 70-76;
i waz missin somethin .. Noriega, 1991, pp. 151- 153);
i sat up one nite walkin a boardin house $ The use of improvisation, lllise~en-scene, and montage to fill the screen with
screamin/cryin/the ghost of another woman multiracial images and to manipulate bicultural visual and linguistic codes;
who was missin what I waz miss in
,z. The use of personal testimonials, life stories, voice-overs, and offscreen
i wanted to jump up outta my bones
narration to provide overall narrative unity to texts (Noriega, 1992, pp. 156-
& be done wit myself
159);
leave me alone ...
i fell into a numbness t Celebration of key elements in Chicano/a culture, especially the themes of
til the only tree I cd see ... resistance, maintenance, affirmation, and neoindigenism, or mestizaie
held me in the breeze (Nuriega, 1991, p.iSe), thereby challenging assimilation and melting~pot
made me dcnvn dew narratives;
that chill at daybreak t Production of texts that deconstruct machismo, the masculine identity, and
the sun wrapped me up s\vingin rose light everywhere the celebration of works that give the Chicana subject an active parr in the
the sky laid over me like a miUion men text while criticizing such timeworn stereotypes as the virgin, whore, sup~
i \Vaz cold/I \vaz burnin up/a child porrive wife, and homegirJ (Fregoso, 1993, pp. 29, 93~94);
& endlessly v'leaving garments for the moon
l' Rejection of essentializing approaches to identity and emphasis on a pro-
wit me teats
cessual, gendered, performance view of self and the location of identity
within, nor outside of, systems of cultural and media representation;
i found god in myself ' Refusal to accept the official race-relations narrative of the culture, which
LV:. i loved her/i loved her fiercely
privileges the ideology of assimilation while comending that black and His-
panic youth pose grave threats to white society (Fregoso, 1993, p. 29).
All of rhe ladies repe:1t to themselve~ wftly the line~ 'j found !lod in
myself & Iloved her.' It soon becomes Q song of joy, srarted by the lady in Return to Anna Deavere Smith's Pires ill t!,e Mirror (1993)'>, and lis-
blue. The ladies sing first to each other, then gradually to the audience. ten now to An Anonymous Young Ivlan #2 Bad Boy. The time is evening;
After the song peaks the ladies enter a closed tight circle. the season, spring. The setting is the same recreation room INhere the
482 483
 , "t::" , I Ul.-"'.. <:':> UIIU rulltlL'S UI Interpretation

llllcrview with Anonvmous Young l\.1an #1 took place. Young lvlan #2 has Simone, Nancy \\7ilson, Billie Holiday, Bessie Smith, John Coltrane, or
dreadlocks. On his h~ad is J "very odd-shaped multicolored hat" (p. 100). l\1iles Davis.
He is wearing a black jacket over his dorhes. He has a gold toorh. He is But rhen,JuneJordan (1998)'C might put it differently. In her jazz prose
soEt-spoken. poem "A Good News Blues," she pays homage to Billie and Louis, Nina,
and Bessie, to anyone who sings the blues. In the following lines she praises
That youth,
Billie Holiday (pp. 199-200):
that sixtccn-year-old
didn't murder that Jew ... Since the blues left my sky
mostly the Black youth in Crown Heights have two things to do- I'm runnin out on l'vlonday
You c"lther To chase down all my Sundays
DJ, bc a l\tC, a rapper Nowa man must be pushing and shoving
or Jall1Clican rappet, But a \",0013n'5 born to strut and stretch
ragamuffin, To go outside (for Icatch some) lovin
or you be a bad boy Get over past regrets
you sell drugs or you rob people. I'm IiErin' weights and wearin sweats_
\X'lut do you do? And thrillin through the night ...
[ sell drugs And if 1 want to rewrire
\Vhat do you do? all the sorry-assl
[ rap. victim/passi velfeminine/
That's how it is in Crown Heights. (pp. 100, 102) traditional
propaganda
spinnin oUt here
In this speech Smith catches the language of youth, its rhythm, syntax,
ain' Ilobody's business
,mJ semantics. She uses the personal testimonial as a way of bringing if I do
:mother voice into the text, as weD as providing narrative unity to the play. I been lost
Her young man rejects essentializing views of identity, bue notes that ill bur
Crovm Heights yOLmg people have fev." choices or options in life. Young r been found
I am bound
Jvlan #2 resists these interpretations.
for Billie's land
Artistic representations such as those presented here from Lorde, Dove,
that place of longing
and Smith (and fordan-see below) are based on the notion of a radical where the angels learn
and constantly ~hanging set of aesthetic practices. As hooks (1990) ob- to sing and play
serves: "There can never be one critical paradigm for the evaluation of the syncopated music
artistic work.... a radical aesthetic acknmvledges that we are constantly of my soul
changing positions, locations, that our needs and concerns vary, that
these diverse directions must correspond with shifts in critical thinking" And brilliant rainbows cut through the blue-and-white clouds, connecting
(p.ll1). distant hills and mOllntains to the places where I live.
At this level, there is nO preferred aesthetic. For example, realistic art is Each text, each performance should be valued in terms of the collective
nor necessarily better than abstract, expressionist, or impressionist art. In and individual reflection and critical aerion it produces, including conver-
the \vorlcls of jazz, ragtime, New Orleans, classic, and swing are not neces- sations that cross the boundaries of race, class, gender, and narion. \l{7e <'sic
sarilv more or less politically correct or aesthetically better than bebop, how each work of art and each instance of qualitative inquiry promotes
cool: h8.rd bop, Latin, avant-garde, and fusion. Nor is Charlie Parker less the development of human agency, "resistance ... and critical conscious-
polittcally correct then Lester Young, or Ben Webster, Ella Fitzgerald, Nina ness" (hooks, 1990, p. 111).

484 485
 This aesthetic also seeks and values beauty, and looks to find beauty in
the everyday, especially in "the lives of poor people" (hooks, 1990,
p. 111). Here is an illustration from bell hooks, who recalls the homes of
her childhood, especially the house of Baba, her grandmother. Looking
back into her childhood, hooks observes that she now sees how this black
woman was struggling to create, in spite of poverty and hardship, an
oppositional world of beauty. Baba had a clean house crowded \'-lith many
precious objects. Baba was also a quilt maker. She turned worn-our every-
day clothing into bealltiful works of art, and her quilts were present in
every room of her small house.
Baba's house \vas like an art gallery. Lare at night, hooks would sit alone
in an upstairs room in the house, and light from the moon \vould send
crisscrossing patterns and shadows across the floor and the wall. In the
stillness of the night, in the reflections from the moon's light, hooks came
to see darkness and beauty in different ways.
Now, in a different time, late at night, she and her sisters "think about
our skin as a dark room, a place of shadO\vs. \YJe talk often abour color pol-
itics, and the ways racism has created an aesthetic that wounds us, \ve talk
about the need to see darkness differently.... in that space of shadows we
long for an aesthetic of b1ackness~strange and oppositional" (hooks,
1990, p. 113). Baba, the quilt maker, shows her how ro do this.
Aesthetics, art, performance, history, culture, and politics are thus
intertwined, for in the artful, interpretive production, cultural heroes,
heroines, mythic pasts, and senses of moral community are created. It
remains to chart the future~to return to the beginning, to reimagine the
\-vays in which qualitative inquiry and interpretive ethnography can ad-
vance the agendas of radical democratic practice, to ask where these prac-
tices will take us next.
.~ Into the Future
Of course, persons who do interpretations feel uncomfortable making pre-
dictions. Bur where the fields of interpretation, qualitative inquiry, and the
practices and politics of telling stories will be in 10 years must be addressed.
If the past predicts the future, and if the decade of the 1990s is to be taken
seriously, then interpretation is moving more and more deeply into the
regions of the postmodern, multicultural sensibility. A new postinter-
prcrive, posrfoundational paradigm is emerging. This framework is attach-
486
ing itself to new and less foundational interpretive criteria. A more expan-
sive framework shaped by an aesthetics of color and critical race theory
principles informs these criteria.
Epistemologies and aesthetics of color will proliferate, building on
Afrocentric, Chicano/a, Native American, Asian, and Third \X1orld per-
spectives. IVlore elaborated epistemologies of gender (and class) wiII
appear, including queer theory (see Gamson, Chapter 12, Volume 1) and
feminisms of color (see Ladson-Billings, Chapter 9, Volume 1). These
interpretive communities and their scholars will draw on their group
experiences as the basis of the texts rhey write, and they will seek texts that
speak ro the logic and cultures of these communities. They will be commit-
ted to advancing the political, economic, cultural, and educarional prac-
tices of critical race theory.
These race-, etl1l1ic-, and gender-specific interpretive communities will
fashion interpretive criteria our of their interactions with postpositivist,
constructivist, critical theory, and poststructuralsensibilities. These crite-
ria will be local, aesthetic, elllic, existential, political, and emotional. They
\vill push the personal to the forefront of the political, where the social
text becomes the vehicle for the expression of politics.
This projected proliferation of interpretive communities does not
mean that the field of qualitative research will splinter into warring fac-
tions, or into groups that cannot speak to one another. Underneath the
complexities and contradictions that define this field rest four common
commitments. The first reflects the belief that the world of human experi-
ence must be studied from the point of view of the historically and cultur-
ally situated individual. Second, qualitative researchers will persist in
working outward from their own biographies to the worlds of experience
that surround them. Third, scholars will continue to value and seek to
produce works that speak clearly and powerfully about these worlds. As
Raymond Carver (1989, p. 24) observes, the real experimenters will
always be those who make it new, who find things out for themselves, and
who want to carry this news from their world to ours. Thus will qualitative
researchers find new and different ways of joining their interpretive work
with constantly changing fotms of radical democratic practice.
Fourth, these texts \vill be committed nor just to describing the world,
but to changing it. These texts will be performance based and informed by
the prdcrices of civic, intimate, and public journalism. They will be COI11-
mitted to creating civic transformations and to using minimalist social
theory. They \vill inscribe and perform utopian dreams, dreams shaped by
487
 "" ... ,,, ,,,... ,r,, ,...... " _ ... _~ .... _,.,.
, "e ," Ul..l/{.. !.:'S una t-'OJltlcs ot Interpretation

critical race theory, dreams of a world where all are frec to be who they personal diaries nature wrt' d f
choose to be, free of gender, class, race, religiolls, at' ethnic prejudice or I, ld ~ -~ , .1 mg, an per orrnance texts anchored in the
natllr3 wor (see Denzin, 1999a; sec also Lincoln & Denzin Chapter 14 "01
discrimination. The seventh moment will be one in which the practices of tunc j ). ,< ,Vl ...

qualitative inquiry finally move, without hesitation or encumbrance, from

the personal to the politicaL This move is a spiraling process, It builds on
previous levels of expression, It does nor abandon the personal, for exam-
ple, in the name of the political. l ? 'l> Notes
So the stories we tel1 one another will change, and the criteria for read..
ing stories \vill also change, And this is how it should be. The good stories ,;- I am indebted to Sage Publications dl I "
supplying the funds to pay the er . _ . an. / Ie nstltute of Communications Research [or
afC always told by those \\'ho have learned well the stories of the past bur 1 11 _ P llIlSSlons ees for quoted material used in this chapter
0 repeat, as Yvonn;l Lincoln and r assert in CI -, f l ' _. .
are unable to tell them any longer. This is so because the stories from the moments of qU3!ir;)rive imillit" ~ d.. 1 lapter 1 0 [liS senes, the seven
.' "re Ira lUana (1900-1950)' d . ,
past no longer speak to them or to us. blurred genres (1970-1CJS6)' .. r . ' rna erniSI (l950-1970);
- ) ,cnS1SO represcmanon(1986 1990) _
In the end, then, to summarize, I seek an existential, interpretive eth- experilllCntn] ilnd new ethllogra I' 1(. (_ - ;.pustnlOdern, a period of
2000) 'md .h, f . I I' plies ( 990-199,:,); postexpcrlmcnral inquiry (1995-
nography, an ethnography that offers a blueprint for cultural criticism. ,. , t o lltUle,WllCII Sno w(2000_j. - .
This criticism is grounded in the specific worlds made visible in the eth.. 2. t\ parallel mOvemenl (see below) is occ . . .
Chicano!a :lIld black [ilmlllal.'" '. . . urrlllg 111 the worlds of cJl1ema, where
nography, It undersr::mds that all ethnography is theory and value laden. . ' . ,,,f.'i .Ire uSing VOIce-Over' f - - d ff _ .
tllISt-etl-sct:-ne, and monmge '15 wavs f -: . s, IrSt person an 0 screen tlarrMlon,
There can be no objective account of a culture and its ways (see Smith & subject (Fregoso 1991 pep 7~ 7~'- N ~Isturbmg traditional gendered images of the r:lcial
Deemer, Chapter 12, this volume), Taking a lead from midcentury A..frican movement is a ~onse~l~len~e, :n ;art~r~~g~,/::S2'J~' 152~ l~D; ste also I-.bll, 1989). This
American cultural critics (Du Bois, Hurston, Ellison, Wright, Baldwin, funded program called the Ne\ C . ,5, Office of EconomIc Opponunlty-
~ v OlllmulllCators "dcsigl I .' . .. -
mem in the film i d. "(N . , l e e [Q tr.lIn mlllOrltles tor employ-
Hines), vve no\\' Icnmv that the ethnographic, the aesthetic, and the politi.. . n ustr)' onega, 1996, p, 7),
cal can never be nearly separated. Ethnography, like art, is always political. ~. [he fo.llowin. g se~ti~n draws from Denzin (1997, pp. 280-231).
Accordingly, after Ford (1950/1998), a critical, civic, literary ethnogra.. Atthe5amellme,ltlsundl'rstoodrl " '-' . . .. ""
up a pollured hmbor isn . I> I. _ I 1 lat p~rtlclpanng111 a CItizen s lllltlative to clean
n toSS po ItIca t Ian debann'
phy is one that must meet four criteria, It must evidence a mastery of lit- J . I. I. ..
sentation of cert'lin gro' f g III Cll tura jOurna 5 the pejoranve pre-
erary craftsmanship, the art of good writing. It should present a well.. ... '. ' . ups In terms 0 stereotypical images" (Benhabib 1992 _ 104)
.). PubliC Journahs m is not without its crirics i I d' 1 '.':'
plotred, compelling, but minimalist narrative. This narrative should be [OCIISgrol rl. ..... I'. ' nc tl lllg tlOse who Sd}' It 15 news by
lp, Ut It IS a mar cetlng deVice to sell new I' ,
b3sed on realistic, natural conversation, with a focus on memorable, 1Il0vcmell[ 'limed .' _ . ' 1 spapers, t lat It IS a conscn':ltive refotm
. :n IIKreaslng t lC power of f" I' .
recognizable characters. These characters should be located in well- do nor understand th' I' pro eSSlOna journ<lllsts, and that its advoc:ltes
' t; rea meamngs of communl[)' hl"]"f. d" .. .
essays In Cral)'" 'I QI d _ ,pu lC I e, an CiVIC discourse. Sec the
described, "unforgettable scenes" (Ford, 1950/1998, p. 1112). Second, (995).
, , .... r, IV C ual all - Norr' (1998)
'IS an
d GI
C1sser (1999), but see also Carey
the work should present clearly identifiable cultural and political issues,
6. Thisethicvaluescomrnulliry sol' I . , .
including injustices based on the structures and meanings of race, class, observers and,' 1-' _ . ' IC atlq, care, love, empowerment, morally involved
.. " c;:ltIng re anons with the communit}' (see Denzin, 1997, p. 275j.
gender, and sexual orientation, Third, the work should articulate a politics 7. [hese dre extensions of tI I ..
or hope, It should criticize how things are and imagine how they could be opcr;ulng for journalists. Ie norms r ldt Chl"l5tlans et al. 0993, pp. 55-57) sec as
different. Finally, it should 'do these things through direct and indirect 8. lllings :Ire known only through thei " . .
regulated b. f' r representations. Each represematlOnal form is
symbolic ~J.nd rhetorical means. \Vriters who do these things are fully y.1 set a conventions Faetu:l1 t I h ld b . .
certain rules of, f _. F. . , :I es s ou e objecnve and should conform to
immersed in the oppressions and injustices of their time. They direct their \ en lcatlon. !CtlOnal nle' . . 1 db
emotional ... '1 j . , s arc regu :lte y understandings connected to
erhnographic energies ro higher, uropian, morally sacred goals. q . VenS-lmJ lfUl e, emotlOn31 rcalism, and so on.
1 -. I d,,,,,,lo Cli((o,,j ClubLian5 fur t'lli:; pnu\..-iplC,
Finally, this performative ethnography searches for new ways to locale
O. The rules in rhis para'" I 1-' . ,
and represent the gendered, sacred self in its ethical relationships to !V1vst'''' . S ol:lp 1 p .'glatlze Ravnllllld Cbandl!;:r s 'Twelve NOles on the
.... ry tory" (1995). -
nature. An exploration of other forms of writing is sought, including
vs. f10m an d reworks
11. This section dra\ .
portloll~;
. C1999b),
of Denzln
488
489
 - -_. - - - ~,.- '"'" ........ '-', ",,,.:::,/-', ccu",-",

12. As l'v!e:lghall i\lorris has indicated ro me in conversation, the Australians involved in

this else fdt that iustice was nor served, that they had to deal with the loss of a loved one
strictly within American political terms. Rosenbaum became a cipher in these accounts.
13. There are various stories about how the town got its name. "The generally accepted
theory is that the Crow Indians who inhabited the area colored their lodges with red clay"
(Graetz, 1997, p. 23).
H. The rown has a Web site. Using a search engine such as '(,"lhoo, you can simply type
"Red Lodge, /l.ilontana" and within seconds yOll can be looking at a Illap of downtown Red
Lodge.
15. I borrow this phrase from bell hooks's (1990) essay "An Aesthetic of Blackness."
16. The implementation ofthel968 New Communicators program (see note 2, above)
at the University of Sourhern California and the University of California at Los Angeles pro-
Juud first- and second-generation minority filmmakers called by some the black and brown
Los Angeles schools (see Oiawara, 1993; Fregoso, 1993, pp. 31, 129; rvtasilela, l L!:!3;
Noriega, 1992, p. 141; Noriega, 1996, pp. 7-H). Among this group of filmmakers are such
names as Burnett, Dash, Van Peebles, Rich, Duke, Trevino, Vasque7., Martinez, Valdez, Nair,
\Vang, hooks, and NaVel. N!ore recent names include Lee, Singleton, and Hughes. These
post-civil rights filmmakers are implementing a cinematic version of critical race theory, and
their works should be ITJd alongside those of bonks, Wesr, and others. Funding for this proj-
ect was Clll b~lCk under the Reagan administration in the 19805.
17. !v,ln Brady clarified this point for me.
<> References
Alexic, S. (1993). aIel shirts mId Hew shillS. Los Angeles: University of California,
American Indian Smdies Center. (Used by permission.)
AnzaldLia, G. (1987). Borderlands/la ji'olltem: The new mestiza. San Francisco:
Aunt Lure.
Saker, H.,.!r. (1997). The blackarrs movement. In H. L Gates,]r., &N. Y.lv1cKay
(Eds_), The Norton illtthology of African Americal1 litemtllre (pp. 1791-
IS06). Ne\v York: \X( WI. Narron.
Baraka, A. (1997). The autobiography of Leroi jones. Chicago: Lawrence I--lill.
Baraka, A. (1998). Black an. In P. L Hill (Ed.), Call a/ld respollse: The Riverside an-
thology of the African A1Ilericallliterary tradition (pp. 1501-1502). Boston:
Houghton .Mifflin. (Original work published 1969. Reprinted by permission
of Sterling Lord Lireristic, Inc. Copyright by Amiri Baraka.)
Behar, R. (1996). The vlll/lerable obseruer: Anthropology that brea/~s your hmrt.
Boston: Beacon.
nenhabib, S. (1992). Situating the self: Gendel; comlllI/nity (lnd postmodemislIl ill
C071tempormy ethICS. New York: Routledge.
Birringer,]. (1993). Theatre, history, postmodemislll. Bloomington: Indiana Uni-
versity Press.
Bochner, A. P., & E!lis, C. (1996). Taking ethnography imo the twenty-first cemurv.
jOllmal of Contell1porary Ethllography, 25, 3~5. -
Brady,1. (1998). A gift of rhe journey. Qualitative II/quiry, 4, 463.
Urady, L (1999)_ Review essily: Ritual as cognitive process, performance as history.
Current Anthropology, 40, 243-248.
Branaman, A. (1997). Goffman's social theory. In C. Lernerr & A. Branaman (Eels.),
The Coffman reader (pp. xvi-Jxxxii). Malden, [vLA..: Blackwell.
Bruner, E. M. (1986). Experience and its expressions. In V. Turner & E.1\-1. Bruner
(Eds.), The al/thropology of eX/Jeriellce (pp. 3~30). Urbana: Universitv of illi-
nois Press. .
Bruner, E.l\-r. (1939). Tourism, creativity, and authenticity. In N. K. Denzin (Ed.),
Studies ill symbolic interaction: A research ilI1llllal (Vol. 10, pp. 109-114).
Greenwich, CT: JAr.
Bruner, E. 11/1. (1996). Abraham Lincoln as authentic reproduction: A critique of
postmodernislll. Americcllf Allthropologist, 96, 397~415.
Buder,.1. (1990). Gender trouble: femiltism and the slIbuersioll ol idclltitv. New
lork: Rourledge. -
Butler, J. (1997). Excitable spec(h: A /Jolitics of the perfol"llliltiue. New York:
Routledge.
Butler,]. (1999). Revisiting bodies and pleasures. Theory, Culture 6'" Society, 16,
11-20.
Carey, J. \X~ (1995). The press, public opinion, and public discourse. In T L Glasser
& C. T. 5almon (Eds.), Fu/;/ic opinion alld the COlIlJllflllicatioll of COI/sellt
(pp. 380-402). New York: Guilford.
Carspecken, P. F. (1996). Critical etlmogra/J!Jy in educational research: A theoretical
aile! practical guide. Ne\v York: Routledge.
Carver, R. Cl989). Fires. New Yorlc: Vantage.
Ceglowslci, D. (1997). That's n good story, but is it really research? QtWlihIlive
Inquiry,3,188-10l.
Chandler, R. (1995). Twelve notes on the mystery story. In R. Chandler, L.Iter IlOll-
elsl1l1d other /urithlgs (pr. 1004-1011). New York: Penguin.
Charity, A. (1995). Doing public joumalislll. New York: Guilford.
Christian, B. T. (1997). Literature since 1970. In H. L Gates, Jr., & N. Y. McKav
(Eels.), The Nortoll L1I1thology of African A.merican literatllre (pp. 2011'-
lOlO). New York: \Y/. \Y/. Norton.
Christians, C. G., Ferre,]. P., & Fackler, P. M. (1993). GoOi.! netus: Social ethics and
the press. New York: Oxford University Press.
Clough, P. T. (1994). FCII/inist tl,otlgl'f' J),.s;rL'. !,c"-,.,,,,,.,,,d ,'c"c!ct"ic dis,oo'''SL-. Cacn-
bridge, MA: Blackwell.
Clough, I~ T. (1998). The eltd(s)
of ethnography: From realism to social criticism
(2nd cd.). New York: Peter Lang.

490
491
 " ... l.-,~, '''_ ,...... "_"', ~"''--'-'''' ._. '.'
'fie nULUces ana I-'olltlcs a/Interpretation

Cl.Jhen-Cruz, ]. (1998). General introduction. In j. Cohen-Cruz (Ed.), Radical

Donmoyer, R., &Yennie-Dollmovcr J (199~) D ~
strcet perfon/wllce: An internatiollal anthology (pp. 1-6). New lork: , f . . ,. .). ata as llral1l~l: Reflections on the
Routledge. use ~ rcadets theater as a mode of qualitmive dara dispIn\'
1l1qllfry, 1, 402-428. - .' QI/alitatiue
Collins, P. H. (1990). Black felllinist thought: {(nowledge, cOllsciousness, alId the
Dove, R. (1993). Taking in wash In P L Hill (Ed) C II d _ "
{Iolitics of emtJOwermcnt. New York: Routledge, Chapman & Hall. mrtholo ' of I '1 f... . ". ." a all response: 7 he Riuerside
Collins, P. H. (1998). Fighting words: Black women and the search for justice. Min- g) t Je 'JUCal! tlmcrtca1l li/erar)' traditiol1 (PP. 1966-1967)
neapolis: University of Minnesota Press. Bjoston: II-Ioughton ,Mifflin. (Reprinted from R. Dove, Selected /)Oe1ll5 1993'
N- ew Yor (: Vantage) , -,
Conq uergood, D. (1985). Performing as a moral act: Ethical dimensions of the eth-
Du Bois, W E. B. (1920). D I \1 ,
nogr3phy of performance, Literatllre in PerfonlU11Icc, 5, 1-13. aUIl/ater: oices {mill within the veil. Nt'\v "orl".'
Schocken. l .
Conquergood, D. (1992). Ethnography, rhetoric and performance. Quarterly jOllr-
nal ofSpee[h, 7S, 80-97. Dunbar C Jr (1999) TI '1 " .
. '. ".. .. . lfee 510rt stones. Qualttatiue Inquiry, S, 130-140
Conquergood, D. (1998). Health theatre ill a Hmong refugee camp: Performance, ElliS, C ~~;G97)T.EvOcative auroethnography: \V'riting emotionally aboUt" our' lives
communication and culture. In J. Cohen-Cruz (Ed.), Radical street pellor- In w lerney ScY S L' I (Ed ..
. ",' _ . l: .' ~. mea n ' s.), Representation and the text: Re-fralll-
I/lilllce: Al1 ilIlenwtiO/wl anthology (pp. 220-229). New York: Routledge. 1Il,." t Je Ih1JTallve vOice (PIJ. 115~ 139) Alb "S ' II' .
Press. -- . . all}. talC l11verslty of New tork
Cooke,]. (1980, September 23). Jimmy's world. Washington Post, p. AI.
Cruikshank, J. (1997). Negotiating with narrative: Establishing cultural identity at Ellis, c., & Bochner A P (1997) T lJ" d ' ,
, ... _: ' .... - -. c mg an PCrfofllllflg personal stories: The con-
the Yukon International Storytelling Festival. Americal/ Anthropologist, 99,
s~r,lJJ~ts ot c.hol~e. III abortion. In C. Ellis & M. G. Fbherty (Eels.) blves~
56-69. IIgatlllg Sllbect/vlt" R . / ,. i '
! .1' \/:SUIICJ 011 we( cX/JcrieJIcc (PP 79-1(1) N b
Dash. L (1997). Rosa Lee: A mother and /;er family ill urban America. New York: Park, CA: Sage. . - . ew ury
Penguin. Ellis, c.,, &Bochner A P. (Ed ) (199(j) C '
Davis, A. Y. (1998). Bllles legacies and black femillism: Gertmde ivia R1-1illC}', Bessie . ' .' . .. s. . J. o/llposmg etlmography: Altctlli1tiue
fO/lllS ofquaiitatwc luntll/O
O'
\'\'alllllt
<
Creel'
~,:
C/\ AI tal\/
' j 'Ira
Srl/ith, and Billie Holiday. New York: Panrheoo_
Ellis, C, & Flaherty NI G (Ed,') (1997) " "
Dawson, P. (1996). Not another fish story from occupied Monraoa_ In R. Ne\vby & 1" d ','" .l. .. Investfgiltmg sub;ectilJitv: Research all

we. expenence. Newbury Park, CA: Sage. _.

S, l-lunger (Eds.), \'(!ritil/g MOlltalla: Literature ullder the big sky (pp. 10-23).
Fiedler L (1988) Th M
Helena: iviomana Cenrer for the Book. ,. ' . e omana face. In W Kittredge & A. Smith (Eds) TIp I t
best {Jlace: A Montalla anthology (pp 744 75 7 ) S i u. -." e as
Degh, L. (1995). Narratives ill society: A performer-eel/tered study of lIarratiol/. ington Press. . - -. cart e: mverslty of\'Vash-
Bloomington: Indiana University Press.
Denison, J. (1996). Sports narratives. Qualitative Illqlliry, 2, 351~362. Fine, J\-t, Powell, L. c., Weis, L., &\Vong, L Jv1. (1997). Preface 1 M F' l C
Powell L \\leis ~ L Ivi \"0 i . n. . Ine, .. .
Dtnzin, N. K. (1997). Illterpretive ethnography: Ethllographic practices for the 21st I '.' , l : . 1. ong (Et s,), Off white, Readings all raCe power
centllty. Thollsand Oaks, CA: Sage. all( sonety (pp. vii-xii). New York: Routledge '
Denzin, N. K. (1999a). An interpretive ethnography for the next ceotury.]oltflldl of Ford, N. A. (1998) A hi . i ' .
.,. .. e,' . ue.pnnt or Negro authors. In P. L. Hill (Ed.), Call and
CCJlltem/Jorary Ethnography, 28, 510-519. It;;S/JOllse: 111e Rwcrsld-, aJ t! I f / AI;' ,
( " .' . JO ogy 0 t Je fl'lcan Amencan literal'" 1,.,,1,'
1011 pp 1117 111. 1 ) B . -' cL-
Denzin, N. K. (1999b). Performing lv10ntana. In B. Glassner & R. Hertz (Eds.), "1950) .. -- "T. os ton; Houghton Mifflin. (Original work published
Qualitative sociology as eueryday life (pp. 147-158). Thousand Oaks, C1\:
Sage. Fregoso, R. L: (199.3). The brow;:e screcn: Chicana and Chicallo film c1l1t1ll"" Min-
Demin, N. K. (1999c). Performing Ivlomana, part II. Symbolic Interactioll, 23, 76~ neapo j IS: Umversity of Minnesora Press. c.

89. Gatcs,H.L.,Jr.,&Iv1cKa}~N.y'(1997).Preface:T1Ikjngbooks
InH L Gr J'
Diawara, Ivl. (1993). Black American cinema: The new realism. In M. Diawara & N. )~ McKay (Eds) TI N . / / ' . . ,;} cs, I.,
, . ,.. I' ' .' Je. Olt011 alit JO ogy ofAfrican American literawre
(d.), Hl'lc1z. AlII 1,ticl-11l cinema (pp. 3~lS). New York~ R.outledge. (pp. XXlll-X I). New)ork! 'i.Xl. W Norton
Di versi, M. (1998). Glimpses of street life: Representing lived experience through Glasser T L (Ed) (1999) '"/ 'J .
. ' " . . . 1,Je I ea of public journalism. New York: Guilford.
short stories. Qualitative Inquiry, 4, 131-137. GoHman , E. (1959) TI '
- . Je /nesel1tatlOll of self ill euerydav life. Garden City, NI'~
Doig, I. (1978). This h01lse of shy. New York: Harcourt Brace. Doubleday. -

492
493
 1"'
C.ronzcl l:L, F. E (1998).
, , L Formarions of lvlexicananess: Trenzas de idemidades. 1 Joyce, J. A. (1987). The black canon: Reconstructing black American litefJry criti-
multiples/Growing lip lvlcxicana: Braids of multiple idclHjries.llttcmatwJli1 cism. New L.itemry History, 18,335-344.
jOll/"llal of QuaIitalive Studies ilt EdllCdlioll, 1.1, .s 1~ 102.
Kittredge, W (1987). OWllilfg it all. San Francisco: Murray House.
Graber, D., j'v1cQuail, D., & Norris, P. (Eels.). (1998). The politics of Hews: The 1/11/5 Kittredge, W (1994). Hole in the s'~y: A mCJlfoit: San Francisco: IVIurray House.
of politics. \'{lasbingwn, DC: Congressiol1<.l1 Quarterly Press. Kittredge, \XZ (]996). 'VIJO owns the Wiest? San Francisco: Murray House.
Graetz, 11... (1997, July-August), Sojourn to rhe sky: The Beartooth Highway. Kotarba,J. A. (1998). Black men, bbck voices: The role of the producer in srmhetic
/vIOl/tal/a tI'Ii7f;ilZiIlC, pp. 18-26. performance ethnography. Qualitative Inquiry, 4, 389-404.
H:JI!, S. (1989). Ct;!rural identity and cinematic representation. FI'i1lJ1ell./OrJ~_, 36, Kramer, M. (1995). Breakable rules for literary journalists. In N. Sims &
(-;8-81. M. Kramer (Eds.), Litemry jOllntalism:A new collection ofthe best American
l-.arnngwll,
"f . - '"1
n. (199') _. em-shies'
__ ",." Llwhite
_ . i1lto blackAlllerica. New
_ ;OllnlCV
1I1i111'S nOlffictioll (pp. 21-34). New York: Ballantine.
York: HarpcrCollins. . Krieger, S. (1996). The family silver: Essays OJ! relationships amollg womclI. Berke-
, ~. -'to,1, H'1
H .hrlng n. (1997'1).
~- - , Prologue: The job of remembering for . thef mbe. ' in ley: University of California Press.
- \'{Z Harrington (Ed.), intimate jounwlism: The art alld crajt 0 re/-lOrtlllg ladson-Billings, G. (1998). JUSt what is critical race theory and what is it doing in a
eueryday life (pp. vii-xvi). Thollsand Oaks; CA: Sage. .. . "nice" field like education? International jOllntal of QllalitatiLtc Studies ill
1_"-larrll1glOn, \1.'1 u
V\. (19 ./ ""b)
I . ,A,. writer's essel\-"
Seeking the exrraordl1lary 111
_ the __ ordl-
.
Education, 11,7-24.
nary. In \Y.,Z Harrington (Ed.), illtillUlte journalism: The art and craft of r2p01 l- lather, P. (1993). Fertile obsession: Validity after POststructuralism. Sociological
ing 'cl/erydL/11 life (pp. xvii-xlvi). Thousand Oaks, CA: Sage. _ Quarterly, 34, 673-693.
[-Lmis, \'\Z J. (199~~). Cross roads blues: African American history a~~-1 c~ll.turc, 1%(~ Lather, P., & Smithies, C. (1997). v'oIlbling the angels: \\Iomcll living with Hi\!/
to t 11e present. I n L ~ . , e,11
n L. H',II (Ed) , alTd respollSe: The 1\1021Stde .mthol fUDS. Boulder, CO: Westvie\v.
ogy of the African American literary traditioll (pp. 1343-1385). Boston: Lee, G. (l996). The poet: Bill Evans. In R. Gottlieb (Ed.), Reddillg jazz: A gathering
~!oughtonMimin. . o/alftobiography, reportage, 1!1d criticism from 1919 to flOW (pp. 419-444).
Hill, P. L. (Ed.). (1998). Call and respolfse: The J{il!ersid~ aI:thology of the Afnct1/1 New York: Pantheon.
Americal/ literar!! tradition. Boston: Houghton Ivll[[Jlll. Lemen, C. (1997). Goffman. In C. Lemen & A. Branaman (Eds.), The Go(fmalz
Hill, R. 1'. G. (1997). "Performance and t1le " po I"mca 1 ,matomy
. " of pioneer Colo- reader (pp. ix-xliii). l'vIalden, IvLA: Blackwell.
rado. Text and Perfomwlfce Quarterly, 17, 236-255. Limerick, P. N. (1997). The shadows of heaven itself. In WE. Reibsame (EJ.),Atlas
I.100(S,.
1 b (19 'JO) of the tWtL' West (pp. 151- 178). New York: \"X( \"X( Norton.
- .\'e1mitlfJ'
, 6Race (Tellder
"': > ' alldclfltllral politics. Boston: Somh End.
r I
hooks. b. (1996). Reel to reil!: Race, sex, and class at the mouics. New lOr(: Lincoln, l~ S. (1995a). Emerging criteria for quality in qualitative and interpretive
inquiry. Qualitative Inquiry, 1, 275~289.
Routledge.
-,-I ug,Iles,. L. (.196 1 ). The wavs of white folks. New York: Vintage. Lincoln, 1'. 5, (1995b). The sixth moment: Emerging problems in qualitative re-
-- - - __ _ If d \
1, 'I I (199") Collected poems by Lallgston_ HlIgh13s. NewI York: A re. 1, search. In N. K. Denzin (Ed.), Studies in symbolic illterac!ioll: A research
~llglt;S, -,. R- . - cltv
~ Knopf. (Permission granted by the Estate of Langston Hug 1es. eprlUte 1. ammal (Vol. 19, pp, 37-55). Greenwich, CT: JAJ.
"T.

permission of Alfred A. Knopf, a Division of Random House, Inc.)
Jackson,lv1. (1998). Minima etlJJlographica: Intersllbjectivity and the anthropologt-
. cal project. Chicago: University of Chicago Press.
. w 1'1
Jenkins. J. E. (1999, May 2). Narrating the actIOn 11 e ta I'
(lUg part .
In.It. NewYorh
-
. 7~;/les, Arts and Leisure sec., pp. 9, ]9.
Jones, L. (1966). HOIne: Social essays. Hopewell, NJ; Ecco.
jones, S, I-I. (1999),_Torch. Qllalita~ive Il1CfUfl)/ 5, 23j-Z50.
Jordan..[. (1998), AffirmaTIve aets, Garden City NY~ Ancl1oL(COPYllgl1t IQ 1~l:usl
.June Jon-lan. Used by permission of Doubleday, a c!lvlslOn of Random 1-
inc.)
. Lincoln, Y. S. (1997). Self, subject, audience, text: Living at the edge, writing in the
margins. In \"X( G. Tierney & Y. 5. Lincoln (Eds.), Representi1tion and the text:
Re-(ramilfg the narrcttiue voice (pp. 37-56). Albany: State University of New
York Press.
Lincoln, y. 5. (1998, November). Whell research is not enough: Commllllity, care,
alld love. Presidential Address delivered;It the annual meeting of the Associa-
tion for the Study of Higher Education, Miami, fL
. C II ~ b' Larde, A. (199RGl. CO:11. In P. L. IT:ll ([;'.,J_>, 0,,11 and r"_,p"",,-,c,, Th" Ri"."-o"<1,,, .m
th%gy of flu. Af';CG/T Allferrcl1/1 literal)! traditioll
(pp. 1626-1627). Boston:
, Houghton Ntifflin. (Copyright 1986 by Audre Lorde, from Collected Poems by
Audre Larde. Used by permission of \Xi: \"X( Narron & Company, Inc.)

494 495
 Il'l I L.f\.r f\.L. 1M' ,'--'. '1 , ... "-''''W/-'' ,'-',,,, ,...., ....... " ..., ,,,_ ...... ' , " ., . _. ,
Lorde, A. (1998b). Power. In E L. Hill (Ed.), Call alld response: The Riverside
antholog)! of the Afriwll American literary tradition (p. 1627). Boston:
HouglHon Mifflin. (Reprinted from A. Lorde, The black unicom: Poems,
1978, Nc\v York: \X( \YJ. Norton)
Lorde, A. (l998c). Stations. In P. L. Hill (Ed.), Call and response: The Riverside
anthology of the Africall AII1erict1l1 literary tradition (pp. 1630-1631).
Boston: Houghton rVlifflin. (Copyright @ 1988 by Audre Lorde, from Col-
lected Poems by Audre Lorde. Used by permission of W. \Y/. Norton & Com-
pany, Inc.)
Lubiano, \X'. (1997). But, compared to what? Reading realism, representation, and
essentialism in School Daze, Do the IVght Thing, and the Spike Lee discourse.
In \( Smith (Ed.), Representing blacklless: Issues ill fillll and video (pp. 1-12)_
New Brunswick, N]: Rutgers University Press.
A'Iarx, K. (1983). Theses on Feuerbach. In E. Kamenka (Ed.), The portable Km1
Ivlarx (pp. 155-158). New York: Penguin. (Original work published 1888)
rVbsilela, N. (1993). The Los Angeles school of black filmmakers. In 1v1. Diawara
(Ed.), Black American cinema (pp. 107-117). New York: Routledge.
lVIilis, C. W (1959). The sociological imagillt1tioll. New York: Oxford University
Press.
lvlienczakowski, J. (1995). The reconstruction of ethnography into theater wirh
emancipator)' potential. Qualitative IlIquiry, 1, 360-375.
Neal, L. (1988). Visiolls of a liberated futllre. New York: Thunder's Mouth.
Neal, L (1998)_ The black arts movement. In P. L Hill (Ed.), Cal! alld response: The
Riverside al1thology of the African American literary tradition (pp. 1450-
1458). Bosron: Houghton Mifflin. (Reprinted from L. Neal, Visions ofa lib-
erated fl/tl/re, 1988, New York: Thunder's Mouth)
Nero, C. L. (1998). Toward a gay black aesthetic: Signifying in contemporary black
gay literature. In P. L. Hill (Ed.), Cal! and respol1se: The IVverside anthology
of the Africall JlmeriCail liteI'm}' tradition (pp. 1973-1987). Boston: Houghton
Mifflin.
Noriega, C. A. (1992). Between a weapon and a formula: Chicano cinema and itS
contexts. In C. A. Noriega (Ed.), Chicanos and film: Representation and resis-
tmlce (pp. 141-167).1Vlinneapolis: University of 1vlinnesota Press.
Noriega, C. A. (1996). Imagined borders: Locating Chicano cinema in America/
America. In C. A. Noriega & A. M. Lopez (Eds.), The ethnic eye: Lati110
media arts (pp. 3-21). Ivlinneapolis: University of Minnesota Press.
Paget, 1v1. A. (1993). A complex sorrow (M. L. DeVault, Ed.)_ Philadelphia: Temple
University Press.
Parker, L. (1998). Race is ... race ain't: An exploration of the utility of critical r:lCe
theory in qualitative research in education. Intemational jOllmal of Qualitd-
five Studies ill Edllcation, 11, 43-55.
496
. . - _ ~-~ ~ ~ VII."....} VI" ,.c, t-" clUUUl1
Parker, L., Deyhle, D., Villenas, S., & Nebeker, K. C. (l99B). Guest editors' intro-
duction: Critical race l'heary ;ltld qualital'ive studies in education. Il1ter~
national JOltrlwl ofQllalitatilJc Studies ill Education, 11,5-6.
Pizarro,1v!. (1998). "Chicana/a power": Epistemology and methodology for social
justice and empowerment in Chicana/a communities. lntemational Journal
of Qualitative Studies ill Education, 11, 57-79.
Red Lodge Chamber of Commerce. (1982). Pestiual of Natiolls, Red Lodge,
fdolltalta [brochure]. Red Lodge, J'dT: Author.
Reason, l~ (1993). Sacred experience and sacred science. JOllrtfal of Mallagemelll
Inquiry, 2, 10-27.
Reason, P. (Ed.). (1994). Participation ill/mlilall inql/iry. London: Sage.
Richardson, L. (1997). Fields of play: COllstntcting an academic life. New Bruns-
wick, NJ: Rmgers University Press.
Richardson,lVL (1998). Poetics in the field and on the page. Qualiti1tive [Ilquiry, 4,
451-462.
Rinehart, R. (1998). Fictional methods in ethnography: Believnbility, specks of
glass, and Chekhov. QlIalitative Inqlliry, 4, 100-124.
Ronai, C. R. (1998). Sketching widl Derrida: An ethnography of a researcher/erotic
dancer. Qualitative inquiry, 4, 405-420.
Rony, R. (1979). Philosophy alld the mirror of !wllIre. Princeton, NJ: Princeton
University Press.
Rosen,]. (1994). Making things more public: On the political responsibility of the
media intellectual. Critical Sludies ill Mass COl/llIlfIllicatioll, 11,362-388.
Scheurich,]. J. (1997). Research method iff the /Jostmodem. London: Falmer.
Schwandt, T. A. (1997). Textual gymnastics, ethics and angst. In W G. Tiernev &
t': S. Lincoln (Eds.), RetJresel1tatioll and the text: Re-framitzg the Iwrmtive
paice (pp. 305-311). Albany: State University of New York Press.
Shange, N. (1977). For colored girls who have considered Silicide when the rainbow
is eml(. New York: Simon & Schuster. (Reprinted with the permission of
Scribner, a Division of Simon & Schuster. Copyright ([) 1975, 1976, 1977 by
Ntozake Shange.)
Sims, N. (1995). The art of literary journalism. In N. Sims & lvL Kramer (Eels.), Lit-
erary jOlll'!/aIism: A /lew collection of the best Americall nonfictioll (pr. 3-
19). New York: Ballantine.
Smith, A. D. (1993). Fires ill the mirror: Crown Heights Broo/dvlI and other idwti-
tics. Garden City, N'r': Anchor. (Copyright iQ 19'93 by f~1l1'a De<lvere Smith.
Used by permission of Doubleday, a division of Random House, Inc.)
Smith, l\. D. (1994). TIlJiIigl'f: L,--,,: .!\"g"le,;. 1002_ Q",.c!"" C;,y, 'W'..'o A"el,,-,,".
Stern, C. ~., & Henderson, B. (1993). Pcrfo/"llwnce texts mId conte.""-!s. New 1'ork:
Longm<ln.
Stringer, E. T. (1996). Action research. ThoLls,lOd Oaks, CA: Sage.
497
>1--' \,X1 G (19CJ7)
Jerney, \ . . ~.
Lost in translation: Time and voice
.
in qualitative research. In
_ . R' r,. _. ]
\'V. G. Tierney & Y. S. Lincoln (Eds.), Repre5e1lfa~I0I1L~lld the text" e-, TaI~I.lfIg
the lwrralwe lIOlce pp. _73-'6)
> > ( .). Albant'o
J' State UOlvcrslty ofNev/iork Press.

>1--> I [" ~i[ (1997) Framer framed. Ne ..v York: Routledge.

.,1111 . 1 \ . - f' TIP f . Arts
~lurner,. V. (ICJ86)
_ . The mlthropology of performance. New lor (; er o[mmg
Journal Publications.
Ulmer. G. (1989). Teletheory. New York: Routledge.. .
'Vl C
nest, . -
(1993). foreword. In A. D. Smith, Fires ill the IJmTOr: CroW/1 HeIghts,
.. .. G deNY. Anchor
BraoHvlf, and other idmtities (pp. XVIl-XXll). at en I t y , . .
'X/,[ C (1994). Race matters. New York: Vantage.
"\' es,
"'/>11> >. (>'. (>1997) , .Par'ldise
1 ZtnSOIl, < revisited. In \XI. E. I'\.el1same
>[ . , Atlas of the new
(Ed)
West (pp. 15-(5). New York: \'\~ WI. Norton. A Method ofInquiry
Wolfe, T. (1973). The new journalism. In T. \\lolfe & E. \YJ. Johnson (Eds.), The new
journalislII:.4/1 anthology (pp. 1-52). New York: Harper & Row.

Laurel Richardson

The wrirer's object is~or should be-to hold the reader's

attention.... 1 want the reader to turn rhe page and keep
on turning to the end.
-Bd)did Tuc::hmJn, j\kw Yor~: Time" Fchudry 2, 1989

In the spirit of affectionate irreverence toward qualitative research, I

consider writing as a method of i"quiJy, a way of finding out about
yourself and your topic. Although we usually think about writing as a
mode of "telling" about the social world, writing is nor JUSt a mopping-
up activity at the end of a research project. \Xfriting is also a \-vay of
"knmving"-a method of discovery and analysis. By writing in different
ways, we discover new aspects of our topic and our relationship to it. Form
and Content are inseparable.

AUTHOR'S NOTE: I thank Erncst Lockridge for mau}' discussions about this chapter, and his
reading uf it llIultipk times. I also r!l;mk Arthur Boc],""'r. Norm",., Dc"";,,. ('"'Tory" rll;.,.
l"'lic!leJlr: hnc) l'atti Lmherj Y\,onna Lincoln, Meflghal1 MCJI'ri", and Jolm Vall }daanen [or
their generous and v:lluable critiques. And, finally, I am grateful to the many student's who
have told me they found rhe earlier \'ctsicn of rhis chapter useful; rhey have givcn me the
energy and the will to revise it.

498 499
 "., ... '" ,,~"" ' ..... ", ~" .~~.
\Xlriting as ;] method of inquiry departs from standard social science
practices. It offers an additional-or alternative-research practice. In
swndard social scientific discourse, methods for acquiring data arc distinct
from the writing of the research report, the latter presLlmed to be an un-
problematic 3ctivity, a transparent report aboUt the world studied. \X1hen
\ve view writing as a method, however, we experience "language-in-use,"
how we "word the world" into existence (Rose, 1992). And then we
"reword" the world, erase the computer screen, check the thesaurus, move
a paragraph, again and again. This "worded worlel" never accurately, pre-
cisely, completely captures the studied world, yet we persist in trying. \V'rit-
ing as a method of inquiry honors and encourages the trying, recognizing
it as embryonic to the full-fledged attention to the signi ficance of language.
\\7riting as a method of inquiry, then, provides a research practice
through \vhich we can investigate how \ve construct the \vorld, ourselves,
:ll1el others, and henv standard objectifying practices of social science
unnecessarily limit us and social science. \X1riting as method does not take
\vriting for gramed, bur offers multiple \vays to learn to do it, and to nur-
ture the writer.
I have composed this chapter into two equally important, but differ-
ently formatted, sections. I emphasize the equally because the first section,
an essay, has rhetorical advantages over its later-born sib. In the first sec-
tion, "Writing in Contexts," I position myself as a reader/writer of qualita-
tive research. Then, I discuss (a) the historical roots of social scientific
writing, including its dependence upon metaphor and prescribed writing
formats; (b) the postmodernist possibilities for qualitative writing, includ-
ing creative analytic practices and their ethnographic products; and (c) rhe
future of ethnography. In the second section, "\X!riting Practices," I offer a
compendium of wriring suggestions and exercises.
Necessarily, rhe chapter reflects my own process Jnd preferences. I
encourage researchers to explore their own processes and preferences
through vvriting. \X!riring from our Selves should strengthen the commu-
nity of qualirative researchers and the individual voices within it, because
\ve will be more fully present in our v./ork, more honest, more engaged.
<t> Writing in Conte}!!s
I have a confession to make. For 30 years, I had yawned my way through
numerous supposedly exemplary qualitative studies. Countless numbers of
500
..... , '':;1' , ",,'-'", ,'-'u '-', "'LI'""'''!
texts I abandoned half read, half scanned. I would order a new book with
great anticipation-the topic was one I was interested in, the author was
someone I wanted to read-only to find the text boring. It \vas not that the
\vriting was complex and difficult, bur that it suffered from acme and
chronic passivity: passive-voiced author, passive "subjects." "Coming out"
to colleagues and students about my secret displeasure with much of quali-
tative writing, I found a community of like-minded discontents. Under-
graduates, graduates, and colleagues alike say they have found much of
qualitative writing-yes-boring.
\V'e have a serious problem: Research topics are rivering and research
valuable, but qualitative books are underread. Unlike quantitative work,
which can be interpreted through its tables and summaries, qualitative
work carries its meaning in its entire text. Just as a piece of lirerature is not
equivalenr to its "plot summary," qualitative research is not contained in
its abstracts. Qualitative research has to be read, not scanned i its meaning
is in the reading.
Qualitative work could be reaching wide and diverse audiences, nor just
devotees of individual topics or authors. It seems foolish at best, and nar-
cissistic and wholly self-absorbed at worst, to spend months or years doing
research that ends up not being read and not making a difference to any-
thing but the author's career. Can something be done? That is the question
that drives this chaprer: How do we create texts that are vital? That afe
attended to? That make a difference? One way to create those texts is to
turn our attention to writing as a method of inquiry.
I write because I want ro find something out. I write in order to learn
something that I did not know before I wrote it. I was taught, however, as
perhaps you were, too, not to \vrite until I knew what I wanted to say until
my points were organized and outlined. No surprise, this static \;fiting
model coheres with mechanistic scientism and quantitative research, But, I
will argue, this static writing model is itself a sociohisrorical invention thar
reifies the static social world imagined by our 19th-century foreparenrs. The
model has serious problems: It ignores the role of writing as a dynamic,
creative proceSSi it undermines the confidence of beginning qualitative
researchers because their experience of research is inconsisrent with rhe
writing model; and it contributes to the flotilla of qualitative writing that is
0imoly not intere~tin[! to l''''''1d 6(,C::Lll~e ndl"c.rencc to tl... " mCH~c~l rcqu;re" ........,-;r_
en to gilence their own voices and to view themselves as contaminants.
Qualitative researchers commonly speak of the importance of the indi-
vidual researcher's skills and aptitudes. The researcher-rar!lcr than the
SOl
 Writing: A Nlethod of InqUiry

survey, the guestionnaire, or rhe census rape-is rhe "instrument." The

language \vould be one \vithotlt w d ' .
more honed the researcher, the bener the possibility of excellem research. ud Johnson's dictiowu:y. I .. orf" s'"onl,> unambIguous symbols. SaJ11-
Students are taught ro be open, to observe, listen, question, and partici- sougH to IX unlvo I .. '
< .
much like the univocal I ". . _ . . ca meanings 111 perpetuity,
pate. Yet they are taught ro conceptualize writing as "writing-up" the te- 1985, p. 4). nCdl1lngs of standard arithmetic terms" (Levine,
search, rather than as an open place, a method of discovery. Promulgating
Into this linguistic world tI,e '[ . d
"writing-up" validates a mechanistic model of writing, shutting down the LV arqUis e Condo t ' d
term social science. Heconte d'd I, "I_ rce Intra uced the
creativity and sensibilities of the individual writer/researcher. " n e t 1dt I..nowledge of th h" I
easy and error almost ilnp 'bl" .f e trut \VOl! d be
One reason, then, that some of our texts may be boring is that our sense , USSl e lone adopt:l . 1
moral and social issues (gu Jt d' L ' _ el precise al.l g 11 ag e about
of Self is diminished as we arc homogenized through professional social- . ( e 111 evme 198.) ~) B' I l
lIterature and science St. d , - ,p.b.) tie 19th century,
ization, rewards, and punishments. Homogenization occurs through the , 00 as two separate dom<' . I
ahgned with "art" an:l" 1_ , " ' . ams. ~Iterature was
suppression of individual voices and the acceptance of the omniscient . , l cu ~ure ; ltcomall1ed I, ,! -, f"
voice of science as if it were our own. How do we pur ourselves in our own res, erhics, humanity, and morality" (Cliff It lIe va ues a caste, aes.ther-
to metaphoric and a b' . 1 are, 9.86, p. 6) and the nglll"s
texts, and \vith what consequences? How do we nurture our own individ- In Iguuus angU'lge G' .
that its words were obje~ti\'e ' ," lv~n to SCience was the belief
<..-'
uality and at the same time lay claim to "knowing" something? These are
nonmet"aphoric. ' preCIse, unambiguous, noncontextual, and
both philosophically and practically difficult problems,
Bur because literary writing was takin a s ' . '.
ponance, status imp'1Cr and. .1 I g econd seM to SCience In im-
rnlt 1 va ue sam' r ,. .
<
Historical Contexts: Writing Conventions , -", "
to make literature a part f ' '. t:' Itelary WrIters attempted
Language is a constitutive force, creating a particular view of reality and
I .
eoml11ated both science d f"
0 sCience. By the lare 19ch cent
,. -
,"., I' "
Ul"), lea Ism'
an ICtJon Wtltlllg (CIa JI 199 1 )
Balzac spearhe<lded rI,e r' 1'. ug 1, - . Honore de
of the Self. Producing "things" always involves value-what to produce, , ea Ism movement' r,.. .
what to namc the productions, and what the relationship between the as an "historical organislll" "'1" ' l 10 .ltelature. He VIewed society
. \Hr 1 SOCIa spe' ", V " ,
producers and the named things will be. \Vriting "things" is no exception. cres." \X1riters deserving of '. 1 cles a dn to zoologICal spe-
praIse, lC comended . '."
l\To textual staging is ever innocent (including this one). Styles of writing SOIlS orcauses" of "soci lff~ -" 1 ". ,mustll1vestlgate the rea-
. a e, ecrs - t le first p' . I " , ,
are neither fixed nor neutral but reflect the historically shifting domina- ety IS based (Balz.1C 184 7 /1965 nnclp es Upon whIch SOCl-
, ,. - pp 747- 7 49) r B 1
cion of particular schools or paradigms. Social scientific writing, like all an "instrument of _. '1:" '. - - . "or a. zac, the nove! Was
- SCienn IC mqulry" (CrawEo d 19 '
ocher forms of writing, is a socia historical construction, and, therefore, Balzac's lead Emile 2 1 d _ ' r, 51, p, 7), Following
, a a argue tor "'natur l" ". I' . ,
mutable. famous essa}' "The N '.1 5 _" a Ism. III ltelature. In IllS
0\ e as aCIaI SCIence" h . d 1 ."
Since the l7th century, the world of writing has been divided into two nature, the naturalistic evoluri " ' ealglle t lar the' return to
separate kinds: literary and scientific. Literature, from the 17th century little all the m'1I1ifes[atl' j' Ion whl~h marks the century, drives little by
< un a lllman Il1t"'Il'"
. . I
onward, was associated with fiction, rhetoric, and subjectivity, whereas path." Literature is to be " . l: Ibel~ce I~~O r 1e same scientific
science was associated with fact, "plain language," and objectivity p,171), governed by SCIence (2ob, 1880/1965,
(Clifford, 1986, p. 5). Fiction was "false" because it invented reality, As [he 20th cemury unfold 1 1 .
unlike science, which was "true" because it purportedly "reported" "ob- entifie \vriting and lire .e~, t 1e relatlonships between social sci-
jeceive" realiey in an unambiguous voice. ,r . rary wrltIng grew in complexit TJ "
so Id demarcanons between "fact" d " f . . " - y. 1t:' presumed
During the 18th century, assaults upon literature intensified. John "imagined" were blurred Tl bl a.~ IctlOn and b.etween "true" and
. . le Urnnp- \-vas . 1 1 1I
Locke cautioned adults to forgo figurative language lest the "conduit" writinR ~{)." d,c p,,!,li-_ _. . . 0 mus[ lOl y ( e Bred ~llol.1nct
between "things" and "thought" be obstructed, David Hume depicted " , , u, JVULnatr'm, In wJJar Tom Wolfe dl bl dl
C

new JournalIsm," writers consciousl bi J 1e t 1e

poets as professional liars. Jeremy Bentham proposed that the ideal "fact" and "f" t " 1-' . y un'eel the boundaries bcrvvecn
_IC IOn ane ConSCIOl I d I
1S Y llla e t 1ernsclves [he center of the

502
503
 Writing: A Method of Inquiry

story. (For an excellent extended discussion of the new journalism, see Metaphor
Denzin, 1997, chap. 5.)
New journalists also encroached upon ethnography's province, bor-
rowjng its methods and reporting social and cultural life not as "report- . ~letap.h~r, a .literary de:rice, is the backbone of social science writing.
ers," bur as social analysts. Joining those trespassers were fiction writers LIke tI.le splO~, it bears welght, permits movement, is buried beneath the
such as Truman Capote, Joan Didion, and Norman lYfailer. Professors surface, and lInks parts togethet into a functional, coherent whole. As this
of literature <1\vakened and reawakened interest in novels by minority me_~apJhor about ~eta~llOr sllgge~ts, .the essence of metaphot is experi-
and postcolonial writers by positioning them as "ethnographic novels"- enung and understandmg one tlung m terms of another Th' ,
plished threw:rh con . (" . , . IS IS accom-
narratives that tell about cultures through characters (see Ba, 1987; g lpanson e.g., IvIy love IS lIke a green green toad") 0
Hurston, 1942/199-1). analogy (c.g. "the evening of life"). ' r
By the 1970s, "crossovers" between writing forms spawned the naming Social scientific writing uses metaphors at ever\' "le\'eJ "5 .] .
de d . . oCJa sCIence
of oxymoronic genres: "creative nonfiction," "faction," "ethnographic pen s upon a deep epistemic code regarding the \vay "that knO\vled '"
fjction," the "nonfiction novel," and "true fiction." By 1980, the novelist and under~t~~l,di.ng in genera.1 are figured" (Shapiro, 1985-1986, p. 198~~
E. L. DoctorQ\v \vould assert, "There is no longer any such things as fiction lv~eta~~~o.rs extel na!, to a parnclll~r piece of research prefigure the analysis
or nonfiction, there is only narrative" (quoted in Fishkin, 1985, p. 7). \~Ith "'" tllJth-value code belongmg to another domain Uameson 1981)
Despite the aerual blurring of genres, and despite our contemporary For exam p.le, the use of enlighten for knowledge is a light-based m;taphor'
understanding that all writing is narrative writing, I would contend that what Dernda (198 1 ) f .J ] . " ' ,
, ' . . ..:. rc ers to as ~ le lellOcentnc view of knowledge, the
there is still one major difference separating fiction from science writing. paSSlve recel,pt at rays. Immanent In such metaphors are philosophical and
The difference is not whether the text really is fiction or nonfiction, but ~allle
r
commitments so entrenched and familiar that thev can do tJ '.
I']' . len par-
the claim the author makes for the text. Claiming to write "fiction" is dif- Isan, wor (. ,l11 t le gUIse of neutrality, passing as literal.
ferent from claiming to write "science" in terms of the audience one seeks, T_leoretl~al schemata are always situated in complex, systematic rneta-
the impact one might have on different publics, and how one expects pl~ors. ConSIder the following statements about theory (ex~mples ins ired
"truth claims" to be evaluated. These differences should nor be over- b) Lakoff & Johnson, 1980, p, 46): p
looked or minimized.
\vhenever there are changes in writing styles and formats, we can ' What is the fOllndatioll of your theory?
expect inreJleccual interest in documenting and tracing those changes. ~ Your theory needs support.
Today, scholars in a host of disciplines are tracing the relationships be- ' Your position is shaky.
tween scientific and literary writing and are deconstructing the differences '\) Your argument is falling apart.
between them (see Agger, 1989; Brodkey, 1987; Brown, 1977; Clough, '} Let's construct an argument.
1992; Edmondson, 1984; l\llishler, 1989; Nelson, Megill, & McCloskey, 10 Tl,le form of your argument needs buttressing.
1987; Simons, 1990). Their deconstructive analyses concretely shovv how t Gwen your fmmeLUorl~, no wonder your argumem fell apart.
all disciplines have their own sets of literary devices-not necessarily fic-
tion writillg devices-and rhetorical appeals such as probability tables, Th ' j ' ,
e l~,a JClzed ~vords ,express our customary, unconscious use of the meta-
archival records, and first-person accounts. p,IlOr Theoty IS archltccure
Each social science writing convention could be discussed at length, but L . . "l-h,
.. e metap j lOr, moreover structures rhe ac-
tIons we ta 1(e in theorizing - d I] l ]' , '
. - an \\ lat we )E Ieve constitutes theory. We rry
I will address here only (a) metaphor and (b) writing format. I choose these
[[} 1~1I11d:1 theoret-;c31 ylii~h
slntei',r"'} .. we tILtH exptritnce as a struCfltre
convemions because they are omnipresent, and because I believe they are Wh.d, has a IIlorm and a f,Jl{ll
J

,
'r dat'.1011,
- wh'!e J1 we t Jlen experIence
. as an ed,-.
good sites for experimenting with writing as a method of inquiry (see the (Ice sometimes quite grail( . i_, sometimes
"
, J 10 need of shorillg flP, and 50111e-
section headed "\Yfriting Practices"). tunes in need of (i,'s n/(1tl.
. II'mg, or,
. more recently, decolfstrltC(mg..
504
505

Hisroricallv, theory constructors have deployed combative metaphors.
Spurt, gall/e, ~nd Wdr are COmmon ones. These metaph~~ic schemes do
not resonate with many women's interests, and, in addmon, they have
conrribuced to an academic intellectual culture of hostility, argumenta-
tiveness, and confrontation. In the 19705, feminist researchers intro~llced
and acted upon a different metaphor: "Theory is story." Nor only, lS the
personal the political, the personal is the grounding for tl~eory. \X11th the
new metaphor for their work, many feminists altered thelr resear~h. ~nd
writing practices; \vomen talking abom their experience, narratlVlzlllg
their lives, telling individual and collective stories became understood as
\".'omen theorizing their lives. The boundary betwccn " narrative ' " an-d
"analvsis" dissolved.
1vl~taphors arc everywhere. Consider f1mctionalism, role theory, gl-11~1C
theon' drall/aturgical analogy, OIganicisll/, social euoilltiollisll/, the socIal
systel;;', ecology, {abeling theory, equilibriull/, IJ1Lll/iln capi~L~!, r~sollrce
lIIobilization, ethnic ilfsurgency, developillg countries, stratIfication and
sif!,//ificance tests. Metaphors organize social scientific work and affect
ti~e inrcrpretations of the "facts"; indeed, facts are interpretable ("l~ake
sense") only in terms of their place within a metaphoric s.tructure: [he
"sense making" is always value constituting-making sense In a particular
\vay, privileging one ordering of the "facts" over others.
Writing Format
In addition to the metaphoric basis of social scientific writing, there are
orescribed writing formats: How we are expected to \vritc affects what \ve
~an write about. The referencing system in social science, for example, dis-
courages the usc of foamotes, a place for secondary arguments, novel con-
jectur~s, and related ideas. Knowledge is constitllted as "focused," "prob-
lem" (hypothesis) centered, "lincar," straightforward. Other thOllghts
are extraneous. Inductively accomplished research is ro be reported de-
ductively; the argument is to be abstracted in 150 words or les~; and
researchers are to identify explicitly with a theoretical label. Each ot these
conventions favors-creates and sustains-a particular vision 0. \vh:~
constitutes knowledge. The conventions hold tremendous matenal "a~1
symbolic po\ver over social scicntists. Using tl~em i~lC[ea~es the p,robab1h[~
of one's work being accepted into "core" SOCial SCIence Journals, but the)
3re not prima facie evidence of greatcr-or lesser-truth value or sigl1lfl-
cance than social science \vriting using other conventions.
506
Writing: A A~ethod of Inquiry
Additional social science writing conventions have governed ethno ~
raphies. Needful of distinguishing their work from travelers' and lTIissio~
ari~s' reports as \-~cJl as from imaginative writing, ethnographers adopted
an tmpersonal, dllrd-persO!1 voice to explain "observed phenomena" :md
to trumpet the authenticity of their representations (see Tedlock, Chapter
6, Volume 2).john Van Ivlaanen (1988) identifies four conventions used in
traditional ethnographies, or "realist tales": (a) experiential iluthor(itv),
where the author exists only in the preface to establish "I was there" and~"I
am a researcher" credentials; (b) documentary style, or a plethora of Con-
crete, particular details dut presume to represent the typical activity, pat-
tern, culture member; (c) the clilture member's {Joillt of view, putativelv
prescnted through quotations, explanations, syntax, cultural cliches and
so o~; and (d) ~lItelpretiv~ 01111lipotellce of the ethnographer. Niany o'f the
c1~SSIC books tn the SOCIal sciences are realist tales. Thesc include Kai
ErIkson's Everything in Its Path (1976), \X!illiam Foote \V'hvte's Street Cor-
ner Society (1943), Elliot Liebow's ]"'a/ly's Comer (1967), and Carol
Stack's Ail Gllr [(ill (1974).
Other genres of qualitative writing-such as texts based on life histories
or in-depth intervieWs-have theif own sers of traditional conventions
(see IVIishler, 1989; Richardson, 1990). In these qualitative texts, re-
searchers establish their credentials in the introductory or methods sec-
t~on; tl:ey write the body of the text as though the document and quota-
(Jon smppets arc naturally present, valid, reliable, and fully representative,
rather than selected, pruned, and spruced up by the author for their tex-
tual ap~ea"rance. A:in cultural ethnogtaphies, the assumption of scientific
authonty IS rhetoncally displayed in these other qualitative texts. Exam-
~l~s of conventional "life story" texts include Lillian Rubin's W/orlds of
lam (1976), Sharon Kaufman's The Ageless Self (1986), and my own The
New Other \'(Ioman (Richardson, 1985).
Postmodernist Context
\V'e are fortunate, now, to be working in a postmodernist climate (see
~gger, 1990; Clifford & MarclIs, 1986; Denzin, 1986, 1991, 1995;
utcbean, 1980; Lehman, 1991; Lyotarcl, 1984; Nicholson, 1990:
!Urner & Rt"unet", '199(,")", ~inH:: ....."111:;[1 a l11Llltitude of approaches to lmmv-
mg and telling exist side by side, The core of postmodernism is the doubt
that any method or theory, discourse or genre, tradition or novelty, h3S
a universal and general claim as the "right" or the privileged form of
507
 11'11 t::1'\.r'1'\.t:: IMIIVl'l, LVMLV"", IV'", "","'-' ,,,_, '~""-J""" ,,-,, ,'-',.
Wriiing: A Method or Inquiry
Jutboritacive knowledge. Postmodernism suspects alI truth claims of Similarly, when a man is . . I .
masking and serving particular interests in local, cui rural, and political abuse" 11 . .exposee ro thc dIScourse of "'childhood sexual
, e m3\, recategonze '1 d . I I' . "
experiences E. -1 ~. '- _ 'd< n remem Jcr liS own traumatic childhood
L

struggles. Bur it does nor automatically reject conventional methods of

. . xp:nence an memory arc thus - ~ _ _.
knowing and telling as false or archaic. Rather, it opens those standard Interpretations governed b ' . I' Up~'.l to contradictory
methods to inquiry and inrroduces new methods, which are also, then, individual is bo~h site and }"S~~l,J ln~eresrs an.d prc:'admg discourses. The
subject to critique. tity and for remakin:r mem~'~, Jeer o. tI~ese. ~lscurslvc struggles for iden-
The posrmodernist context of doubt) then, distrusts all methods and compering discagllrs _ . }. Because 1I1d1\'lciuals are subject to multiple
es 111 mJnv realm '1' t " .
equa!!y. No method has a privileged status. The superiority of "science" contradictory not stable f- _ j '... jL S, t lelr su JjectlVIE}' 15 shifting and
. ' , L\el, ngll
over "literature"-or, from another vantage point, "literature" over "sci-
ence"-is challenged. Bur a postmodernist position does allow us to know
Poststructuralism rhus oi i- .
social science- Eacll I'S 1-11 P nts to tIle COllll/lIlal cocrealioJl of Sell and
"something" without claiming to know everything. Having a partial,
.' '- Ovvn rllroUg 1'" -I r .
knowing about the sub' :. _. 1 r Ie or leI'. \.1l0WlIlg rhe self and
local, historical knowledge is still knowing. In some ways, "knowing" is knowledges, Postsrruc'tllr!lell~.1 areI l11tertwi.ued, panial, historical, local
, Sill, r len permlts-Il' "
< . ~ '.
easier, however, because postmodernism recognizes the siruationallimira- us to reEl . , .. d}, lllVltes-no lJlCltes-
ect lipan Our method and . I " ..'
rions of the knower. Qualitative writers are off the hook, so to speak. They S I II '. exp arc new ways ot knOWIng
peCllca y, postsrructuraJism Sli ~,,", _. . . .
don't have to try to play God, wriring as disembodied omniscient narra- tive v,rriters: First it d'. _ . ggests lwu Impoltanr thlllgs to qualita-
tors claiming universal, atemporal general knowledge; they can eschevv Sons writing frOi~ pa:~~c\~:al:S to .:Il~derstand Ol:r~eh:es reflexively as per-
the questionable meranarrative of scientific objectivity and still have frees us [rom rrying to \\"1" _ p~sltllons at .specIC.lc tImes; and second, it
. L e a Slllg e text In 1 'I - .
plenty to say as situated speakers, subjectivities engaged in knowing! once to everyone Nurturi -, \V lIe 1 We say everyrhlng at
telling about the world as they perceive it. of "science wririll'g" ng Our ~wn vellces releases the censoriolls hold
on Our conscIOUS 1ess ,. II I
A particular kind of postmodernist thinking that I have found especially ters in Our psvche: \V'ritin . 1_ j"'1 'd , as we as t Ie arrogance it fos~
helpful is poststmcturalism (for an overview, see Weedon, 1987; for appli- , g IS '\ a Il ate as a method of knowing.
cation of the perspective in a research setting, see Davies, 1994). Post- Creative Analytic Practices:
structuralism links language, subjectivity, social organization, and power. CAP Ethnography
The centerpiece is language. Language does not "reflect" social reality, bur
produces meaning, creates social reality. Different languages and different
In the wake of postmodernisr-includi ". ~ . ..'"
discourses within a given language divide up the world and give it meaning queer and critical I .. ng posrSLI uctllrallSt, [enl11llst
, race tIeorY-Crltlql f r' '. '
in ways that are not reducible to one another. Language is hm\' social orga- practices, qualitative \'Vorl- ". les~ tlaelt~onalqllaiJtallvewriting
nization and pmver are defined and contested and the place where our forms Scie .. '- no~\ appears 10 multIple venues in different
. nce-wrItIng prose IS not held' . .'
sense of selves, our subjectivity, is constructed, Understanding language as genre has been bl ' I I _ sacrosanct. [he ethnographic
< lin ee, en arged . 1 ,. -d . I
competing discourses, competing ways of giving meaning and of organizing versarions readers'the'lt :l' a tele to 11K ude poetry, drama, con-
the vvorld, makes language a site of exploration and struggle. other in th'at they arc < ~r, anl so on. These .cthnographies are like each
rO
Language is not the result of one's individuality; rather, language con- settled upon calling tr J~I~ed rthrough erea~fUe analytic practices. I have
liS c ass 0 ethnognph ' I
srructs the individual's subjectivity in ways that are historically and locally etlmograp!J1/ or C4P II I"' L Its erea/we {/lW ytic practice
"' "e. mograp IV 1 hlS label - I d
specific. \XThat something means to individuals is dependent on the dis~ work and old "I- _I -' can mc u e new work, future
, er V\ 01 '- \V lerever rl I J
courses available to them. For example, being hit by one's spouse is dif~ rional social scien,-f' ~' .. le aut lOr las moved outside COJ1ven-
I Ie Wfl[1I1g.
ferently experienced if it is thought of within the discourse of "normal
marriage," "husbands' rights," or "wife battering," If a woman sees male Ye/ :I,"e": 'l'O'r""I" c"'''hcep , oc ","oUf11l is prOblematic, 5ubJecr to cmil]ue
are t lOUg t about wi .
violence as "normal" or a "husband's right," then she is unlikely to sec it as nagra I - I . lat to name these genre-breaking erll-
plies, t le more I lIked r1 I
'",<viEe battering," an illegitimate use oE power that should not be tolerated. <1cr- CAP. TI le camp ex metaphoric resonances of the
'onym . 'e English
, woreIcap Comes fro III rhe Latin for head, caput.
508
509
 !I" 1 ... "r " ... '1""'\1 l V I ' , ... '1""'\1...'-'1""'\1 ' V I ' , 1""'\1 , ..... I~I...' "1...-'1...1' I,." ,,-,,.
vvnrmg: A Method ot Inquiry

Using "head" to signal e[hnographic breaching work can help break down Gammel, & Taylor, 1994), allegory (Lawton, 1997, pp. 193-214), conuersa-
the mind/body duality. The "bead" is borh mind and body and more, tion (see Ellis & Bochner, 1996b; Richardson & Lockridgl~, 1998), lavered
too. Producers of CAP ethnography are using their "heads." The prod-
accounts Uago, 1996; Ronai, 1992, 1995), w,itil1[J-<tOl'ie, l,ee r '''~'ton,
l-lc;;ts, "ltl'Ol.-lgb lTJ.cdlal<;J thruughout the body, cannot 11lanifest without
15:77; Richardson, 1995, 1997; Sr. Pierre, 199701, 1997b), and mixed genres
"headwork." (see Angrosino, 1998; Brown, 1991; Church, 1999: Davies, 1989; Dorst,
Cap-both noun (product) and verb (process)-has multiple common 1989; Fine, 1992; hooks, 1990; Jipson & Paley, 1997; Jones, 1998;
and idiomatic meanings and associations, some of which refract the play- Larher, 1991; Lather & Smithies, 1997; Lee, 1996; Lind'n, 1992; Pfohl,
fulness of [he genre: a rounded head covering; a special head covering 1992; Richardson, 1997; Rose, 1989; Stoller, 1989; Trinh, 1989; Ulmer,
indicating occupation or membership in a particular group; the top of a 1989; Visweswaran, 1994; Walkerdine, 1990; Williams, 1991; Wolf,
building, or fungus; a small explosive charge; any of several sizes of writ- 1992),
ing paper; purring the final touches on; lying on top of; surpassing, outdo~ For more than a decade, what r am calling CAP ethnography has been
ing. And then there are the orher associated words from the Latin root, labeled experimelltal or altematiue (see V:m lvIaanen,1 995). Uninten-
such as capillary and capital(ism), which humble and conrexrualize the tionally, however, those labels have reinscribeq tradition]l ethnographic
labor. practices as the standard, the known, accepted, preferred, tried-and-true
The practices that produce CAP ethnography are both creative and ana- mode of doing and representing qualitarive research. I believe that re-
lytic. Those holding the dinosaurian belief that "creative" and "analytic" inscription is now unnecessary, false, and deleterious. CAP ethnographies
are contradictory and incompatible modes are standing in the path of a are not alternative or experimental; they are in and of themselves valid
meteor. They are doomed for extinction. Witness the evolution, prolif- and desirable representations of the social. Into the foreseeable future
eration, and diversity of new ethnographic "species" during the past two these ethnographies may indeed be the most valid and desirable represen~
decades. tations, for they invite people in; they open spaces for thinking abollt the
Here is bur a sampling of the many "species" of CAP ethnography: social that elude us now. '
autoetlmogmphy (Behar, 1993, 1996; Bruner, 1996; Church, 1995; Ellis, CAP ethnography displays the writing process and the writing product
1993, 1995a, 1995b, 1998; Frank, 1995; Geertz, 1988; Geda, 1995; as deeply intertwined; both are privileged. The product cannot be sepa-
Goetting & Fenstermaker, 1995; Karp, 1996; Kondo, 1990; Krieger, rated from the producer or the mode of producrion or the method of
1991, 1996; Lawrence-Lightfoot, 1994; McMahon, 1996; Shostak, knowing. Because all research-traditional and CAP ethnography-is
1996; Slobin, 1995; Steedman, 1986; Yu, 1997; Zola, 1982), fiction- now produced within the broader postmodernist dimaee of "doubt,"
stories (Cherry, 1995; Diversi, 19983, 1998b; Frohock, 1992; Richardson readers (and reviewers) want and deserve to know how the researcher
& Lockridge, 1998; Rinehart, 1998; Shelton, 1995; Sparkes, 1997; Stew- claims to know. How docs the author position the Self as a knower and
art, 1989; Williams, 1991; Wilson, 1965; Wolf, 1992), poetr), (Baff, 1997; teller? These questions engage intert\vined problems of subjectivity,
Brady, 1991; Diamond, 1982; Glesne, 1997; Narum, in press; Patai, authority, authorship, reflexivity, and process on the one lnnd and repre-
1988; Prattis, 1985; Richardson, 1992a), drama (Ellis & Bochner, 1992; sentational form on the other.
Paget, 1990; Richardson, 1993, 1996a; Richardson & Lockridge, 1991), Postmodernism claims that writing is always partial, 10Gll, and situa~
performance texts (Denzin, 1997; McCall & Becker, 1990; tional, and that our Self is always present, no matter hO\v nmch we try ro
Mienczakowski, 1996; Richardson, 1998, 1999a, 1999b), po!)'uoca! texts suppress it-but only partially present, for in our writing we repress p~rts
(see Butler & Rosenblum, 1991; Daly & Dienhart, 1998; Krieger, 1983; of ourselves, too. Working from that premise frees us to wrire material in a
Pandolfo, 1997; Schneider, 1991), readers' theater (see Donmoyer & variety of ways: to tell and retell. There is no such thing as "gerring it
Yennie-Donmoyer, 1995), responsive readings (see Richardson, 1992b), right"-only "gerting it" differently contoured and nuanced. \X1hen using
aphorisms (Rose, 1992, 1993), comedy and satire (sec Barley, 1986, creative analytic practices, ethnographers learn about their topics and
1988), lJisna!/Jresentations (see Harper, 1987; Jacobs, 1984; McCall, about themselves that which was unknmvable and unimZlginab!e using
5]0 5] ]
 " " I 1,.1~r " l , . ' ' ' ' ' ' ' - ' ' ' , .... " , " - v r . , ' ' - ' ' ' , ' ".~ "~, ,~ __

vrTlung:.'4 lVlethod ot Inquiry

conventional analytic procedures, metaphors, and writing formats. Even

and interior monologue. Through these techniqucs, the writers Construct
if one chooses to write an article in a conventional form, trying on differ-
sequences ,~f ~vents, or "plots," holding back on interpretation, asking
ent modes ohvriting is a practical and powerful \\'a)' to expand one's inter-
pl-ctivc skiHs, rais\.: ont:'s consciousnr.:ss, and bring a fresh perspective to
readers to rehve" the events emotionally. with the write1"". TJl<.-'se lliAl'ri-t-
one's research. tjve~ seek to meet literary criteria of coherence, verisimilitude, and inter-
est. Some narratives of the Self are staged as imaginativ;:= renderings; oth-
It is beyond this chapter's scope for me to oudine or COOlment here on
ers arc staged as personal essays, striving for honesty, revelation the
the scores of new ethnographic practices and forms. And it is far beyond
"larger picture." In either case, auroethnographers are somewhat reli~ved
that scope for me to discuss practices that exceed the written page-
of the problem of speaking for the "Other," because they are the "Other"
performance pieces, readers' theater) museum displays) choreographed in their texts.
research findings) fine-art representations, hyperrexls, and so on-
although I welcome these additions to the qualitative repertoire. Instead, r Rel~ted to a:ltoethnography withom necessarily invoking the writing
strategIes mentioned above arc narratives abom the \'\'fiting process itself.
\vill address a class of genres that deploy literary devices to re-create lived
I call these writing-stories (Richardson, 1997). These are narratives about
experience and evoke emotion;:d responses. I call these cliocativc rep.resen-
COntexts in which the writing is produced. They situate the author's writ-
tations. I resist providing the reader with snippets from these t-orms,
ing in othcr pans of the amhor's life, such as disciplinary constraints
because snippets will not do them justice. 1will describe some texts, but r
academic debates, departmental politics, social movement~, communit,,:
have no desire to valorize a new canon. Again, process rather than product
is the purpose of this chapter. .
structur.e~, research interests) familial ties, and personal history. The~
offer CrItical reflexivity about the writing-self in different COntexts as a
Evocative forms display interpretive frameworks that demand analySIS
valuable creative analytic practice. They evoke new questions about the
of themselves as cultural products and as methods for rendering the social.
self an~ the su.bject; they remind us that Our work is grounded, contextual,
Evocative representations are a striking \vay of seeing through and beyond
and rhtzomanc. They can evoke deeper parts of the Self, heal wounds
socia! scientific naturalisms. Casting socia! science into evocative forms
enhance the sense of self-or even alter one's sense of identity. )
reveals the rhetoric and the underlying labor of the production, as well as
. In Fields of Play: Constructing all Academic Life (1997), I n;ake exten-
social science's potential as a human endeavor, because evocative writing
SIve use of .writing-stories to contextualize 10 years of my sociological
touches us where we live, in our bodies. Through it we can experience the
work, c.reatlng a text more congruent with POStstructliral understandings
self-reflexive and transformational process of self-creation. Trying out
of the situated nature of knowledge. Putting my papers and essavs in the
evocative forms, we relate differently to our material; we know it dif-
chronological order in \\'hich they were conceptualized) I sorted t1~em into
ferently. We find ourselves attending to feelings, ambiguities, temporal
tlvo ~ilcs-"kceper" and "reject." \X7hen I reread my first keeper-a presi-
sequences, blurred experiences) and so on; we struggle to find a textual
dentIal address to thc North Central Sociological Association-memories
place for ourselves and our doubts and uncertainties. . ..
of being patronized, marginalized, and punished by my department chair
One form of evocative writing is alltoetlmography. (ThiS toplC IS fully
and dean reemerged. I stayed with those me~ories and wrote a writing-
covered bv Ellis & Bochner in Chapter 6 of this volume; see also Fine
et aI., Vol~lme 1, Chapter 4.) Autoethnographies are highly personalized)
S~o? abom the disjunction between my departmental life and my dis-
revealing texts in which authors teU stories abom their own lived expe-
CIP.ll~ary reputation. \"X7riting the story was 110t emotionally easy; in the
Wntm I 1" h 'f' ,
riences) relating the personal to the cultural. The power of these nar~a ~ was re Ivmg orf] IC experIences, bur writing it rel'~ased the anger
aI:d pam. NIany academics who read that srory recognize:t as congrucnt
tives depends upon their rhetorical staging as "true stories," sto~ICS
WIth their experiences, their untold stories.
about events that really happencd to the writers. In telling thcse stones,
" , , teclmlques
the writers call upon such f!ctlon-vvntl11g ' I
as (raman 'c reol< l 1
~ :vorked chronologically
Wntl1 . .
through the keeper pile, rereading and rhen
_ '. .
ng t le Wfltlllg-Story evoked by the rereadlllg. Dlfferenr facets, dIffer-
strong imagery, fleshcd-ollt characters, unusual phrasings, puns, s~lbtexts)
ent COntexts. Some stories required checking my journals and files, bur
allusions, flashbacks and flashforwards, toile shifts) synecdoche, dIalogue,
InOSt did not. Some stories were painful and took an interrninable length
512
513

of time to write, but writing them loosened their shadow hold on me.
Other stories are joyful and remind me of the good fortunes I have in
friends, colleagues, family.
"\.\1rit.ing-stories sc.nsiri,w us to rhe potential consequences of all of o.ur
writing by bringing home-inside our homes and workplaces-the ethIcs
of representation. \"X7riting-stories are not about people and culrures "out
there"-ethnographic subjects (or objects)-they arc about ourselves, our
work spaces, disciplines, friends, and families. What can we say? With
what consequences? Writing-stories bring the danger and poignancy of
ethnographic represcntation up close and personal.
Each writing-story offers its writer an opportunity to mJke a situated
and pragmatic ethical decision about whethcr and where to publish the
story. For the most part, I have found no ethical problem in publishing
stories that reflect the abuse of power by administrators; J consider the
damage done by them far greater than any discomfort my stories might
cause them. In contrast, I feel constraint when \\lriting about my family
mcmbers. Anything I have published about them, I have checked out wit.h
them; in the case of more distant family members, I have changed their
names and identifying characteristics. Some of my recent writing I will not
publish for a while because it would be too costly to me and my familial
relations to do so.
Gtaduate students have found the idea of the writing-story llseful for
thinking through and writing about their research experiences. Some usc
the writing-story as an alternative or supplement to the traditional meth-
ods chapter and, as Judith Lawton (1997) has done, to link the narratives
of those they have researched.
Yet to be developed as a subgenre of writing-stories are \vhat we might
call microprocess writing-stories (see also Meloy, 1993). \"X!ho has not
looked at the computer screen, read a paragraph he or she has written, and
then chosen to alter it? Who has not had their subsequent writing affected
by what they have already written? How docs the process of writing pas-
sages and reading them back to yourself "open new questions and issues
that feed back and emanate from the earlier passages?" (A. l~ Bochner,
personal communication, May 10, 1998). HO\v is a changed Self evoked
through the hands-on/eyes-on feedback process?
Related to this 5ubgenre is computer technology and the textual page
layout: typefaces, font sizes, split pages, boxed inserts, running bottom
text, images, frames. How are choices made? \\1ith \vhat impact on the
producer and the reader? How does the ease of manipulating page formats
514
Writing: A Method of Inquiry
and typographical style contribute to-or distract from--the evocarive-
ness of the text? Authors' discoveries about their topics and themselves?
These are questions looking for \vriting-stOljes.
Unlike the two forms discllssed above, an evocative fotm about which
there is an extensive literaturc is ethnographic fiction (see :3anks & Banks,
1998). (For a more extended discussion of this and other narrative forms
see Tedlock, Volume 2, Chapter 6.) "Fiction writing," accc'rding to novel~
ist Ernest Lockridge (personal communication, 1998), "is IIsing thc imagi-
nation to discover and embody truth." Social science writers who claim
that their work is fiction privilege their imaginations, see.O\:ing to express
their visions of social scientific "truth." Usually they encase: their stories-
whether about themselves or a group or culture-in settings they have
studied ethnographically; they display culmral norms through their charac-
ters. In addition to the techniques used by self-narrators (sce above),
ethnographic fiction writers might draw upon devices such as alternarive
points of viev", deep characterization, third-person voice, and the omni-
scient narrator. (l do not think any ethnographic fiction writers, yet, write
from the point of viev\' of the unreliable narrator; see Lockridge, 1987.)
There are some advantages and some disadvantages to claiming one's
ethnographic writing is fiction. 5 raging qualitative research as fiction frees
the author from some constraints, protects the author from criminal or
other charges, and may protect the identities of those studied. But compet-
ing in the publishing world of "literary fiction" is very difficult. Few suc-
ceed..Moreover, if one's desire is to effecr social change through one's
research, fiction is a rhetorically poor writing strategy. Policy makers pre-
fer materials that claim to be not "nonfiction" even, but "true research."
Another evocative form is poetic representation. A poem, as Robert
Frost articulares it, is "the shortest emotional distance between tvvo
points"-the speaker and the reader. Wh'iting sociological interviews as
poetry, for example, displays the role of the prose trope if, constituting
knowledge. \"Vhen we read or hear poetry, we are continually nudged into
recognizing that the text has been constructed. But all texts 3re con-
structed-prose ones, too; therefore, poetry helps problem;1tize reliabil-
ity, validity, transparency, and "truth."
\X7riting "data" as poetic represenrations reveals the constraining belief
that the purpose of a social science text is to convey information as facts or
themes or notions existing independent of the contextS in whi:::h they were
found or produced-as if the story \ve have recorded, transcnbed, edited,
and written up in prose snippets is the one and only [rue one: a "science"
515
 Il'll t:1~t"I~t IAIIU1'l, CVALUAllvn, i'-\1'1U r.t:rrq:;.JCI'
story. Standard prose writing conceals the handprint of the sociologist
who produced the final \'.'ritten text.
When people talk, moreover, wherher as converSCllltS, storytellers,
informants, or interviewees, their speech is closer to poetry than it is to
sociological prose (Tedlock, 1983). Writing up interviews as poems, hon-
oring the speaker's pauses, repetitions, alliterations, narrative strategies,
rhythms, and so on, may actually better represent the speaker than the
practice of quoting in prose snippets. Further, poetic devices-rhythms,
silences, spaces, breath points, alliterations, meter, cadence, assonancc,
rhyme, and off-rhyme-engage the listener's body, even if the mind resists
and denies. "Poetry is above all a concentration of the power of language
\vhich is the power of our ultimate relationship to everything in the uni-
verse. It is as if forces we can lay claim to in no other way become present
to us in sensuous form" (DeShazer, 1986, p. 138). Sertling words together
in new configurations lets us hear, see, and feel the wodd in new dimen-
sions. Poetry is thus a practical and pO/uerful method for analyzing social
worlds.
"Louisa 1vlay's Story of Her Life" is an example of poetic construction
that challenges epistemological assumptions (Richardson, 1997). It is a 5-
page narrative poem I created from a 36-page transcript of my in-depth
interview with "Louisa May," an unwed mOther. In writing Louisa f./lay's
story, I drew upon both scientific and literary criteria. This was a greater
literary challenge than a sociological one because Louisa May used no
images or sensory words and very few idioms. The poem, therefore, had to
build upon other poetic devices, such as repetition, pauses, meter, rhyme,
and off-rhyme. \Vithout putting words in her mouth, which would vio-
late my sociological sensibilities, I used her voice, diction, hill-southern
rlwthms, and tone. I wrote her life-as she told it to me-as a historically
sit~tated exemphlr of sense making. Her life, as she speaks it, is a "normal
one." The political sub text, as I wrOte it, is "1vlother Courage in America."
Ethnographic drama is another evocative way of shaping an experience
without losing the experience. It can blend realist, fictional, and poetic
techniques; it can reconstruct the "sense" of an event from multiple "as-
lived" perspectives; it can allow all the conflicting "voices" to be heard,
relieving the researcher of having to be judge and arbirer (Davies et al.,
1997;]ohnston, 1997); and it can give voice to what is unspoken bur pres-
ent, for example, "cancer" as portrayed in Paget's (1990) cthnographic
drama or abortion as in Ellis and Bochner's (1992) drama. \Vhen the mate-
rial to be displayed is intractable, unruly, multisited, and emotionally
516
IMIIV'~
vvrmng: A Met/10d ot Inquiry
laden, drama is more likely to recapture the experience dun is standard
writing.
Constructing dr3m:J raises the postmodern debates :J.hout \'omP' and
"writt~n" texts. \X!hich Comes firsr? \"XIhich one should be (is) privileged,
and WIth what consequences? \X1hy the bifurcation between "oral" and
"written"? Originating in the lived experience, encoded ;IS field notes
transformed into an ethnographic play, performed, raped-recorded, and
then reedited for publication, the primed script might \vell be fancied the
~efinitive or "valid" version, particularly to those who privilege the pub-
lIshed over the "original," the performance, or even the lived experience.
\Vhat happens if we accept this validity claim? Dramatic con;;truction pro-
vides.multi pIe sites of invenrion and potential contestation for validi ty, the
blurflng of oral and "vritten texts, rhetorical moves, ethical dilemmas, and
authority/authorship. It doesn't just "talk about" these issues, it is these
issues (see Davies et aI., 1997; Johnston, 1997; Richardson, 1997).
A last evocative form ro consider is mixed genres. The scholar draws
freely in his ~r her productions from literary, artistic, and scic,lltific genres,
~ften breakmg the boundaries of each of those 3S well. In these produc-
tlOns, the scholar might have different "takes" on the same ropic, \vhat I
think of as a postmodernisr deconstruction of triangulation.
In traditionally staged research, \\'C valorize "triangulation." (For dis-
cussion of triangulation as method, see Denzin, 1978; Flick, 1998. For an
application, see Statham, Richardson, & Cook, 1991). In tri;-lIlgulation, a
researcher deploys "different methods"-such as interviews, census data,
and documents-to "validate" findings. These methods, however, carry
the same domain assumptions, including the assumption that: there is ~
"fixed point" or "object" that can be rriangulated. But in postmodernist
mixed~genre texts, we do not triangulate; we crystallize. \1;,'e recognize
that there are far more than "three sides" from which to approach the
world.
I propose that the central imaginary for "validity" for pcstmodernist
texts is not the triangle-a rigid, fixed, two-dimensional ob.iect. Rather,
tbe central imaginary is the crystal, which combines symme-:ry and sub-
stance with an infinite variety of shapes, substances, trarsl11urations,
rnultidimensionalities, and angles of approach. Crystals grow, change,
alter, but are not amorphous. Crystals are prisms that reflect externalities
and refract within themsclves, creating differcIlt colors, patterns, and
arrays~ casting off in different directions. \Vhat we sec depends upon our
angle of repose. Not triangulation, crysrallizatioIl. In po:itJl1odernist
5J7
 "" U-..r ,,, .... ,,....., IV", ..... ,.........V"', 'V", ,.." , ..... '''... ' " .......... " ".." , .... ',
ll1Lxed-genre texts, \ve have moved from plane geometry to light rheorYl
where light can be bot/; waves and particles.
Crystallization, without losing structure, deconstructs the traditional
~.l",_, u[ lI va l:d\ty\\ (wt: [",,,,1 1.u." then..: is no single truth, we see how texts
validate themselves), and crystallization provides us \vith a deepened,
complex, thoroughly partial, understanding of the topic. Paradoxically,
\ve know more and doubt what we know. Ingeniously, we knO\v there is
always more to know.
The construction and reception of the narrative poem mentioned
above, "Louisa May's Story of Her Life" (Richardson, 1997), is emblem-
atic of crystallization. That work generated alternate theories and per-
spectives for writing and for living, deconstructed traditional notions of
validity, glancingly touching some projects, lighting others. Ivly life has
been deeply altered through the research and writing of the poem, and
"Louisa 1\IIay" has touched wide and diverse :llldiences, even inspiring
some to change their research and writing practices.
In one section of Fields o( Play (1997), 1 tell t\vo interwoven srories of
"writing iJIegitimacy": Louisa !vIay's story and the research story-its pro-
duction, dissemination, reception l and consequences for me. There afe
multiple illegitimacies in the stories: a child out of \vediock; poetic repre-
sentation of research findings; a feminine voice in social sciences; erh-
nographic research on ethnographers and dramatic represemation of that
research; emotional presence of the writer; and work jouissance.
I had thought the research story was complete, not necessarily the only
story that could be told, but one that reflected fairly, honestly, and sin-
cerely what my research experiences have been. I still believe thar. Bur
missing from the research story, I came to realize, \vere the personal, bio-
graphical experiences that led me to author such a story.
The idea of "illegitimacy," I have come to acknowledge, has had a com-
pelling hold on me. In my research journal I \vrote l ".Ivly career in the
social sciences might be viewed as one long adventure into illegitimacies."
I asked myself, \'V'hy am I drawn to constructing "texts of illegitimacy,"
including the text of my academic life? \"X7hat is this struggle I have with the
academy-being in it and against it at the same rime? Ho\\' is my story like
and unlike the stories of others struggling to make sense of themselves, to
retrieve suppressed selves, to act ethically?
Refracting '''illegitimacy'' through allusions l glimpses, extended vie\vs,
I came to write 2. personal essay, "Vespers," the final essay in Fields o( Play.
518
vvrmng: A Method ot Inquiry
"Vespers" located my academic life in childhood experiences and memo-
ries; it deepened my knowledge of myself and has resonated with others'
experiences in academia. In rum, rhe writing or "Ve~pen:" J,:;;: refr=,c~<:d,
again, giving me desire, strength, and enough scJf-knO\vledge to narra-
rivize other memories and experiences-to give myself agencYl to con-
struct myself anew, for better or for Worse.
\'V'e also see this crystallization process in several recent mixed-genre
books. Margery Wolf, inA Thrice-Told Tale (1992), takes the same event
and tells it as fictional story, field notes, and a social scienriJic paper. John
Stewart, in Drill/(ers, Drummers and Decent Folk (1989)., writes poetry,
fiction, ethnographic accounts, and field notes about Village Trinidad.
In Schoolgirl FlctioJls (1990), Valerie Walkerdine develops/displays rhe
theme that "masculinity and ferninin'ity are fictions which take on the sta-
tus of fact" (p. xiii) by incorporating into the book journal entries, poems,
essays, photographs of herself, drawings, cartoons, and annotated tran-
scripts. Ruth Linden's j\;[aking Stories, l\1aking Selves: Feminist Reflections
017 the Holocaust (1992) intertwines autobiography, academic writing,
and survivors' stories in a Helen Hooven Sanrmver Prize in \'qomen's
Studies book, which was her dissertation. John Va~ !vIaanen's Tales (rom
the Field (198 8) presents his research on police as realist, cOllfessional, and
impressionist narratives. Patti Lather and Chris Smithies's 'Houbling the
AJlgels: Women LiviJlg \,Iith HN/AIDS (1997) displays high theory, re-
searchers' stories, women's support group transcripts, and historical and
medical information, using innovative text layouts. John Dorst's The
11?rittell Suburb (1989) presents a geographic site as site, image, idea, dis-
course, and an assemblage of texts. Stephen Pfohl's Death 11t the Parasite
Cafe (1992) employs collage strategies and synchronic juxtapositions,
blurring critical theory and militant art forms.
In some mixed-genre productions, the writer/artist roams freely
around topics, breaking our sense of the externality of wpics, developing
our sense of ho\'v topic and self are twin constructed. Sman Krieger's
Social Science alld the Self: Personal Essays 071 an Art Form (1991) is a
superb example. The book is "design oriented," reflecting Krieger's
attachment to Pueblo potters and Georgia O'Keeffc, and, as she S3YS, it
"looks more Wee a pot or a painting than a hypothesis" (p. 120). Trinh T.
lVlinh-ha's\\"lo mail Native Otber (1989) breaks down \vriting conventions
within each of the essays that constitute the book, mixing poetry, self-
reflection, feminist criticism, photographs, and quotations to help readers
519
 11'11 tr;;rru:; IM1I\..lI'1, I..VMI..Ur'I' ''-I'', """".' 'U.. , ,,'-..>'-,. ,r" ''-'"
vvmmg: A Method of Inquiry

, , , e postcoloniality, In I'ue [(/tOWl! Riuers: Liues of Loss (1l~d Libera~

expenenc . , . " _,. .. _,,1 q Ies and questions. I struggle with them in my reaching, writing, and collegial dis-
!' (1994) Sara Lawrence-Lighrtoot uses hcnon-\vntlllg lel. 101 l "
roll - - . . d f- "'lanai cussions. I have no definitive ans\vers, bur I do have some thoughts on the
self-reflexiviry to tcll stories of beins Afro-American an pro eS5 c,' issues,
,_>~" l,,_,,_.L p".,',.-!.,. "'I
~~;;:;;:~;;;';~'a:;'A~adeJJ1iCLife (1997) in its entirety tells the story of ~y
, .. \' . .., -[1 .... <." 01".. ,...", ''':0'- .... ,1 0'''.'''-<-''.'' }"-f." PL,)_
WriliIlg IS 3. process of discovery. 1\1r purpose is nor EO turn us into
poets, novelists, or dramatists-few of us will wrire well enough to suc-
intellectual and political struggles in academJa through p~rson~l, e~,sd~Sl
dra mas poems writing-stories, e-mail messages, and soclOlog) IdrttCE,llCls, ceed in those competitive fields. Most of us, like Poe, will be at best only
. <1 , I Caro yn IS almost poets. Rarher, my intention is to encourage individuals to accept
Anthologies also present mixed genres, Some examp es are. <: ._ < '.

d Arthur Bochner's Composillg Ethnography: Altematwe FO/,'tL. 01 and nurture their Own voices. The researcher's self-knovdedge and knov.;/-
an /',
alia ltatwe . 11 lIlg ('19960) 1 Ellis <and Ivlichael Flaherty's Investlgatmg
, \\I,'!' < I .. I edge of the topic develop through experimentation with point of vie\v,
SlIbjectiuit1 l ; Research all Lived Experiellce (199~), ~uth Be 1a1. _~~1C. tone, texture, sequencing, metaphor, and so on. Another skill, another
D b ' h Gordon's W/0mcIl \Ykiting ClIlture (199_)). I he book, se[]~s language-the studenr's own-is added to the student's repertoire. The
e ora in Symbolic Interaction, and .
Stlldies the'Journal Q /fa I'Itatn' Ie' I Ilqlt.
11')1 1111X
. science-\vriting enterprise is demystified. The deepened und ~rstandil1g of a
genres in their pages. Self deepens the text. Even the analysis paralysis that afflicts some readers
of postmodcrnism is attenuated when writers view their work as process
Whither and Whence? rather rhan as definitive represemation.
Students will not lose the language of science when they learn to write
The contcmporary postmodernist contcxt in which \VC wor,k as qllalit~ in other \vays, any more than students who learn a second language lose
. - esearchers is a propitious one, It provides an opportUnIty for us to their first (Y. S. Lincoln, personal communication, 1998). Rather, acquir-
tlve' r , . "t'qile and re-vision WrIting. . IIlOug I1 we are fxc"er -0
At l present
. _ ing a second language enriches students in two ways: It gains them emry
reView, Cll J , < . , ]' ff" .

r tcxts in a variety of forms to diverse audiences, we have (1 erc~t can into a new culture and literature, and it leads them to a deepened under-
ou 'n;s arising fro~,
self-consciousness about claims to authorsh,p, au: standing of their first language, not just grammatically, but as a language
:~~;it~, truth, validity, reliabiliry, Self-reflexivity brings to consclO,llsness that constructs how they view the world.
some ~f the complex politicaliideological agendas h'dden,,'; ~,ur ;' fltmg, \Xlriting in traditional ways does not prevent us from writing in other
Truth claims are less easily validated now; desires to speak lor or lers ~re ways for other audiences at other times (Denzin, 1994; Richardson,
t The greater freedom to experiment with textual for~., howe\ er~
~
1990). There is no single way-much less one "right" way--of staging a
dsuspec .
oes notguaran
('ee a better product. The opportunities for Wfltl,n g worthy
-., I.' I . _ 'ting text. Like wet clay, the material Can be shaped. Learning altl~rnative \vays
texts-boo (5 an d art!c
I ' I~eS d1at are "good reads' -are mu tip e, eXCI 'II . '. ', of writing increases our reperroires, increases the numbers and kinds of
ancI 'd em,andl'ng . But the work is harder. The guarantees are fewer. lere IS audiences we might reach.
a lot more for us to think abom. .. _ r. 1 for As I write this chapter, I imagine four friendly audiences: graduate stu-
I , for us to think about is \vhetherwrJnng CAP ethnograp
A ne t1lDg d , . 1y e dents, curious quantitative researchers, traditionally inclined qualitative
ublication is a luxury open only to those \'1ho have ac~ emlc sll1ecur .
~re the tenured doing a disservice to students hy introcluclOg to thes:, thel~, researchers, and creative analytic practitioners. I want to clarify and
teach-and, yes, proselytize.
" 'f er1' forms of writing? \\Till teaching students hcrcncisms desl~J1J
er l Id I I u "stlOns \X1ho is your audience? \X1hat arc your purposes? Understanding how to
,I,em' Alienate them from their discipline? (Wall v"',e as {t l~se. qed _
((lit
stage your writing rhetorically increases your chances of getting published
about' students' learning a second language? ) A re.1ate d'Issue 1s, If 'StU
. ClllS1
and reaching your intended audiences. Deconstructing traclitJonal writing
~re taught writing as inquiry, what criteria should be bro~l~hr to bear U~~J I
practices makes writers more conscious of writing conventions and, there-
;hcir work? These are heady ethical, pedagogical, aesthetic, and practluJ
fore, more competently able to make choices.

520
521
 " .. ~". " ... " " ........ ,
.. ~ .. ,~~''', ....... ,,'' ....... ,,~. " ...
The new ways of writing do, however, invoke conversation about crite-
ri8 for judging an ethnographic \vork-new or tT<lclitional. Traditional
ethnographers of gooclv./ill have legitimate concerns abour how their stu-
cl",,~s! ..v",,,t:. ........ \\\ l,,,, cva{l.:,.,t",d i( dle)- c:hOl);';C tD ,yriLt ct\i' cthn0'6raplly. I
have no definitive answers to ease their concerns, but I do have some ideas
and preferences.
1 sec the ethnographic project as humanly situated, always filtered
through human eyes and human perceptions, bearing both the limitations
and the strengths of hum8n feelings. Scientific superstructure is always
resting on the foundation of human activity, belief, understandings. I
emphasize ethnography as constructed through research practices. Re-
search pr8ctices are concerned with enlarged understanding. Science
offers some research practices; literature, creative arts, memory work
(Davies, 1994; Davies et aL, 1997), and introspection (Ellis, 1991) offer
still others. Researchers have many practices from which to choose, and
ought not be constrained by babits of other people's minds.
I believe in holding CAP etbnography to high and difficult stan-
dards; mere novelty does not suffice. Here are five of the criteria I use
when reviewing papers or monographs submirred for social scientific
publication.
1. Substal1tive contribution: Does this piece contribute to our understanding of
social life? Does the writer demonstrate;] deeply grounded (if embedded)
social scientific perspective? How has this perspective informed the con-
struction of the text? (See "\XTriting Practices," below, for some suggestions
all how to accomplish this.)
2. Aesthetic merit: Rather than reducing standJrds, CAP ethnogrJphy adds
another standard. Does this piece succeed aesthetically? Does the lIse of
creative analytic practices open up the text, invite interpretivc responses? Is
rhe text arrisrically shaped, satisfying, complex, and nor boring? (Creative
writing is a skill that can be developed through reading, courses, workshops,
and practice; see the suggestions listed in the "\XTriting Practices" section.)
3. Reflexivity: Is the author cognizant of the epistemology of posrmodernism?
How did the :lllthor come to write this text? How was the information gath~
cted? Are there ethical issues? How hJS the author's subjectivity been both a
producer and a product of this text? Is there adequate self-awareness and self-
exposure for the reader to make judgments about the point of view? Does the
author hold him- or herself accountable [0 the standatds of knowing and tdl-
ing of the people he or she has srudie.d?
522
~~ .. "" ...... ,' ,"'w'y. M IYIt:'UIUU ur InqUiry
4. Impact: Does this affect me? Emotionally? Tmc!JeCfllally? Does it generate
new questions? lvlove me to write? 1vlove me to try flew research pracrices?
Ivlove me to action?
{caffry: Does tills text embodY;l fleshed out', embodied sense
5. L-.;prr:ssiofl uf a
of Jived experience? Does it seem "true"-a credible aCcount of a cultural
social, individual, or communal sense of the "rea!"? '
These arc five of my criteria. Science is olle lens; creative ans another. \\le
see more deeply using two lenses. I want to look through both lenses to
see a "social science art [orm." '
I strongly disagree, then, with those who claim ethnography should be a
"science guild," a "craft" vvith "tacit rules," apprentices, trade "secrets,"
and "disciplined," "responsible" journeymen (i.e., professors) who enact
rules that check "artistic pretensions and excesses" (sec Schwalbe 1995.
see also ,Richardson, 1996b). This medieval vision limits erhno~raphi~
expl~ra~lOn, patrols the boundaries of intellectual thought, and aligns
quabtatlve research ideologically with those \vho would discipline and
punish post1110dern ideas within social science. Policing, however, is
always about bodies. It is always about real live people. Should the medi-
eval vision triumph, what real live people are likely to be excluded?
\Vhat I have learned from my teaching and conversations with col-
leagues is this: IVIinorities within academia, including ethn::c and racial,
postcolonial, gay and lesbian, physically challenged, and r,;wrning stu-
dents, find the turn to creative analytic practices as beckoning. These
researchers desire the opportunity to be "responsible" to the "guild" while
honoring their responsibilities to their traditions, their cultures, and their
sense of the meaningful life.
\X7elcoming these researchers creates an enriched, diversified, sociallv
engaged, nonhegemonic community of qualitative researchers. Evervon~
profits-the communities of origin and idemification and the qualjt~nive
research community. The implications of race and gendt'f would be
stressed not because it would be "politically correct," but beGl LIse race and
gender are axes through which symbolic and accual worlds have been con-
structed. Ivlembers of nondominanr worlds know that, and would insist
that this knowledge be honored (see Iviargolis & Romero, 1998). The
blurring of humanities and social sciences would be welcomed nor becmse
it is "trendy," bur because the blurring coheres more truly with rhe life
senses and learning styles of so many. This new qualitative community
could, through its theory, analytic practices, and diverse membership,
523

reach beyond academia, teaching all of us nbollt social injustice and meth-
ods for ~lleviating it. \Vhat qualitative researcher interested in social life
vmuld not feel enriched by membership in such a culturally diverse and
;nlJ;t;ne: C0n1m1.-H1;tY~
Furthermore, CAP Ethnography is now firmly established \;vlthjn the
social sciences. There are prestigious places for students and others to pub-
lish. Sociological Quarterly, Symbolic Interaction, American Anthropolo-
gist, Jour/wi of Contemporary Etlmography, jOllrnal of Aging Stlldies,
Qualitative Inquiry, Illtemational journal ofQualitative Research ill Edu-
cation, Qualitative Studies ill Psychology, Qttalitatiue Sociology, Wltlilwto
jOllmal of Education, and Text and Performance Qllarterly routinely pub-
lish CAP ethnography. The annuals Studies il1 Symbolic Interaction and
Cultural Studies showcase evocative writing. Publishers such as Rout-
ledge, University of Chicago Press, University of lvlichigan Press, Indiana
University Press, University of Pennsylvania Press, Rutgers University
Press, Temple University Press, and Sage Publications regularly publish
nc\v ethnography by both well-known and lesser-known authors. Alta-
lvlira Press (formerly a division of Sage) boasts the excellent Ethnographic
Alternatives book series, which is dedicated to qualitative research that
blurs the boundaries between the social sciences and humanities. Ne\".'
York University Press has launched the Qualitative Studies in Psychology
series, which is receptive to creative-analytic texts. Trade and univer
sity presses are increasingly resistant to publishing old-style monographs,
and traditional ethnographers arc writing more reflexively and self-
consciously (sec Thorne, 1993). Even those opposed to postmodernisfl1
legitimate it through dialogue (Whyte, 1992). Throughout the social
sciences, convention papers include transgressive presentations. Entire
conferences are devoted to experimentation, such as the "Redesigning
Ethnography" conference at the University of Colorado and the Year 2000
COllCh-Stone Symbolic Interaction Symposium.
At least three weU-respected interpretive programs-at the University
of Illinois (under Norman Denzin), the University of South Florida (under
Arthur Bochner and Carolyn Ellis), and the University of Nevada at Las
Vegas (with Andren Fontana and Kate Hausbreck)-teach creative ana-
lvtic practices. The Ohio State University's Folklore Studies (under Amy
Shun;an) and its Cultural Studies in Education Ph.D. program (under Patti
Lather) privilege postpositivism. Dissertations violating the traditional
five-chapter, social science vVTiting style format arc accepted in the United
Stntes, Canada, England, New Zealand, and Australia. Elliot Eisner
524
YYr/ung: A lVletHod ot InqUiry
(1996), art educator and past president of the American Educational Re-
search Association, has gone further. He proposes that novels should be
accepred as Ph.D. dissertations in education. All of these changes in aCa-
den,.i", pr~...;ti<;~:J urI,; <If.~nJ of parml(rrm cfJangl2s.
In the 19505, the sociology of science was a new, reflexively critical
area. Today, the sociology of science undergirds theory, merhods, and
interdisciplinary science studies. In the 19605, "gender" emerged as a
theoretical perspective. Today, gender studies is one of the largest (if not
the largest) subfield in the social sciences. In part, science studies and gen-
der studies thrived because they identified normative assumptions of
social science that falsely limited knowledge. They spoke "truly" to the
everyday experiences of social scientists. The new areas hie us \vhere we
lived-in our work and in Our bodies. They offered alternative perspec-
tives for understanding the experienced world.
Today, the postmodernist critique is having the same impact on the
social sciences that science studies and gender have had, and for similar
reasons. Postmodernism identifies unspecified assumptions chat hinder us
in our search for understanding "truly," and it offers different practices
that work. \Ve feel its "truth"-its moral, intellectual, aesthetic, emo-
tional, intuitive, embodied pull. Each researcher is likely to respond to
that pull differently, which should lead to \''fiting that is more diverse.
more author centered, less boring, and humbler. Thesc are propitiou~
times. Some even speak of their work as spiritual.
And Thence
The ethnographic life is not separable from the Self. WID we arc and
What we can be-what we can study, how we can write about that wbich
we study-is tied to how a knowledge system disciplines itself and its
members, irs methods for claiming authority over both the subject matter
and its members.
\'7e have inherited some ethnographic rules that arc arbitrary, narrow,
exclusionary, distorting, and alienating. Our task is to find concrete prac-
tices through which \ve can construct ourselves as ethical subjects engaged
in ethical ethnography-inspiring to read and to write. Some of these
practices involve working "\vithin theoretical schemata (sociology of
knowledge, feminism, critical race theory, constructionism, poststruc-
turalism) that challenge grounds of authority; writing on topics that mat-
ter, personally and collectively; jouissallce; experimenting \\lith different
525

writing formats and audiences simultaneously; locating ourselves in multi-
ple discourses and communities; developing critical literacy; finding ways
w \vrite/present/teach that are less hierarchal and univocal; revealing
institutional secrets; using positions of authority to increase diversity,
both in academic appointments and in journal publicacions; self-
reflexivity; giving in to synchronicity; asking for what we want, like cats;
110t flinching from where the writing takes us, emmionally or spiritually;
and honoring the embodiedness and spatiality of our labors.
\"X1har creative analytic practices in ethnography vvill eventually pro-
duce., I do not know. Slit I do know that the ground has been staked, the
foundation bid, the scaffolding erected., ;md diverse and advenmrous
settlers have moved on in.
. and Forever After
The Hafldboo/<. editors really do want 311 the contributors to predict the
furure of qualitative research. I thought I had. Oh, how r resist! Bur here
goes.
Forty years ago., I was an undergraduate who detested the yearlong
course '''History of \"Xlestern Civilization"-1,500 years., five continents,
700 countries, six trillion names., dares, wars, and places. I thought the
final would decimate me. But fortune smiled. In addition to the zillions of
"objective" questions, we were given a take-home essay: "\X!hat is the
future of history?" r said-in 10 pages or less-that the future of history
was both toward unity and toward diversity. I got <In A -I- on that ess<lY I
think I'll stick with it. That's the vvay I see the futllre of qualitative re~
search, too. We will be clearer <lbout its domain and more welcoming of
diverse representations.
The domain's metaphor will be the "text"-or some other equally out-
rageously encompassing image-bur the meaning and construction of text
"vill Far exceed the written page, the computer screen, and even the hyper-
text: two-dimensional, three-dimensions, refr<1nive, layered texts. Dis-
cl-lssions of the boundaries betvvcen literature and science will seem
quaim, as "writing"-in the future understood as any textual constrUC-
tion-will be routinely understood as a "method of inqniry." And, there-
fore, it wilJ have t:o be chal!engec1!
Oh, dear!
526
T,rIWly. t-I/Vlt1lfIUU ur InqUiry
~ Writing Practices
Writing l [he creative effort. should come first-at least for s)me
part of every day of your life. It is a wonderful blessing if yell
will use it. You will become happier, more enlightened, alive,
impassioned, light hearted and generous to everybody else. Even
your health will improve. Colds will disappear and all the other
ailments of discouragement and boredom.
-8rUlda Ucl:ord, II You
\.Y/~nl to Write, 1938/1987
In what follows, r suggest some ways of using writing as ;1 method of
knowing. r have chosen exercises that have been productive for students
because they demystify writing, nurture the researcher's voice, and serve
the process of discovery. J wish I could guarantee them to bring good
health as welL The practices are organized around topics discussed in the
text.
Metaphor
Using old, worn-out metaphors, although easy and comfortable, after a
while invites srodginess and stiffness. The stiffer you get., the less flexible
you are. Your ideas get ignored. If your writing is cliched., you'll not
"stretch your own imagination'" (Ouch! Hear the cliche of poiming OUi
the clichC!) and you'll bore people.
1. In traditional social scientific writing, the metaphor for theory is
that it is a "building" (structure, foundation, construction, deconstruc-
tion, framework, grand, and so on). Consider a different metaphor., such
as "theory as a tapestry" or "theory as an illness." \"Xlrite a paragraph about
"theory" using your metaphor. Do you "see" differently and "]:eel" differ-
ently about theorizing using an unusualmctaphor?
2. Consider alternative sensory metaphors for "knowledge" other
than the heliocentric one mentioned in the text. \X'hat happens when you
rethink/resense '"knowledge" as situated in voice? In touch?
3. Look at one of your papers and highlight your metaphors and im-
ages. \Vhat are you saying through metaphors that you did not realize you
were saying? \\That are you reinscribing? Do you want to? C;:m you find
different metaphors that change hmv you "see" ("feel") the material? your
527
relationship to it? Are your mixed metaphors pointing ro confusion in
yourself or to social science's glossing over of ideas?
4. Take a look ::If George Lakoff and Mark Johnson's i\tletaphors \\lc
Ul'2 By (19RO). It is;l. wonderful book, a compcllJiul11 of examples of met-
aphors in everyday life and how they affect our ways of perceiving, think-
ing, and acting. \Vhat everyday metaphors are shaping your knowing!
writing?
Writing Formats
1. Choose a journal artide that exemplifies rhe mainstream writing
conventions of your discipline. Hmv is the argument sClged? \Vho is
the presumed audience? rImv does the paper inscribe ideology? How
docs the amhor claim authority over the material? \X1here is the author?
\X!here are you in this paper? \Vho are the subjects and who arc rhe objects
of research?
1. Choose a journal article that exemplifies excellence in qualit8tive
rese8fch. How has the article built upon normative social science ,vriting?
How is authority daimed? \Vhere is the author? \Vhere arc you in rbe arti-
cle? \Vho arc the subjects and who are the objects of research?
3. Choose a paper you have written for a class or that yOll have pub-
lished that you think is pretty good. How did you follmv the norms of your
discipline? \Vere you conscious of doing so? How did yOLl stage your
paper? \Vhat parts did the professor/reviewer laud? Hmv did you depend
upon those norms to carry your argument? Did you elide some difficult
areas through vagueness, jargon, calls to authorities, or other rhetorical
devices? \\lhat voices did you exdude in your writing? \Vho is the audi-
ence? \\lhere arc the subjects in the paper? \,(1here arc you? How do you
reel abour the paper now? About your process of constructing it?
Creative Analytic Writing Practices
1. Join or start a writing group. This could be a \vriting support group,
a creative writing group,;] poetry group, a dissertation group, or another
kind of group. (On dissertation and article writing, sec Becker, 1986; Fox,
1985; Richardson, 1990; Wolcott, 1990.)
2. \\lode through a creative writing guidebook. Natalie Goldberg
(1986,1990), Rust Hills (1987), Brenda Ueland (1938/1987), and Deeua
\1{1e instein (1993) 3.1l provide excellent guides.
528
3. Enroll in a creative writing "vorkshop or d"lSS. These experiences
are valuable for both beginning and experienced researchers.
4. Use "writiug up" your field notes as an opportunity to expand your
writing vocabulary, habits of thought, and attentiveness to your senses,
~Illd as a bulwark against the censorious voice of science. \"Vhere better (0
develop your sense of self, your voice, than in the process or doing your
research? Apply creative writing skills to your field notes. "'rClU may need to
rethink what you've have been taught about objectivity, science, and the
ethnographic project. \\lhat works for me is to give different labels to dif-
ferent content. Building on the work of Glaser and Straus:; (1967), I use
four categories, which you may find of value:
ObseruatirJllllotcs (ON): These are as concrete ,wei demiled as I am able to
m3ke them. I want to think of them as fairly accurate renditions of what I
see, hear, feel, taste, and so on. I stay dose to the scene as I experience it
through my senses.
$ Methodological l10tes (MN): These are messages to myself regarding how ro
collect "data"-who to talk to, Wh,lt to wear, when to pho:lC, and so on. I
writea lot of these because I like methods, and I like to keep a process diary of
my work.
~ Theoretical /lotes (TN): These arc hunches, hypotheses, poststrucU1Lliist
connections, critiques of what r am doing/thinking/seeing. ! like writing these
because they open my field 110te texts to al ternative inrerpret3tions and <l crit-
ical epistemological stJl1ce. They provide a way of keeping me from being
hooked on one view of reality.
-- Personal notes (PN): These arc uncensored feeling statements about the re-
search, the people I am t,lIking to, my doubts, my anxieties, my pleasures. I
want all my feelings out on paper beclUse I know they are affecting what/how
r lay claim to know. I also Irnow they are a gre<lt source for hypotheses; if I am
feeling a certain way in a setting, it is likely that orhers might feel that way
too. Finally, writing personal notes is a way for me to know myself hetter, a
way of using writing as method of inquiry imo the selL
5. Keep a journal. In ir, write about your feelings about your work.
This not only frees up your writing, it becomes the "historiol record" for
the \vriting of a narrative of the Self or a writing-story abollt the wriling
process.
6. \X' rire a writing autobiography. This \voulcl be the story of how you
learned to write: the dicta of English classes (wpic sentences: outlines? the
five-paragraph essay?), the dicta of social science professors. your expef1~
529

cnees with teachers' comments on your papers, how and where you write
now, your idiosyncratic "writing needs," your feelings about \vriting and
abour the writing process. (This is an exercise that Arthur Bochner uses.)
7. Ir you '.v;sh to cxpcrinl.ent vvirh evocative writing, a good place 10
begin is by transforming your field notes inro drama. See what ethno-
graphic rules you are using (such as fidelity to the speech of the partici-
pants, fidelity in the order of the speakers and events) and what literary
ones you arc invoking (such as limits on how long a speaker speaks, keep-
ing the "plot" moving along, developing character through actions). \V'rit-
ing dramatic presentations accentuateS ethical considerations. If you
doubt that, contrast writing up an ethnographic event as a "typical" event
\vith writing it as a play, with you and your hosts cast in roles thar \",ill be
Dcr[ormed before others. \\Tho has ownership of spoken words? How is
~urhorship attributed? \Vhat if people do nor like how they arc character-
ized? Arc courtesy norms being violated? Experiment here with bOth oral
and \vfitten versions of your drama.
8. Experiment with transforming an in-depth interview into :J. poetic
representation, Try using only the words, rhythms, figures of speech,
breath points, pauses, syntax, and diction of the spc<Jker. \X'here are you in
the poem? \Xlhat do you kno"v about the interviewee and about yourself
that you did not know before you wrore the poem? \Vhat poetic devices
have you sacrificed in the name of science?
9, Experiment with \'\-'firing narratives of the self. Keep in mind
B<lrbara Tuchman's warning: "The writer's object is-or should be-to
hold the reader's attention.. , . I want the reader to turn the page and keep
on turning to the end. This is accomplished only \vhen the narrative moves
steadily ahead, not v,;hen it comes to a vveary standstill, overlaccd with
every item uncovered in the research" (in New Yor/.~ Times, February 2,
1989).
j O. Try \vriting a text using different typefaces, font sizes, and texwal
pbcement. Hov", have the traditional ways of using prim affected what
you know and how you know it?
. 11. \vJrite a "layered text" (see Lather & Smithies, 1997; Ronai, 1992).
The layered text is a strategy for putting yourself into your text and pur~
ring your text into the literatures and traditions of social science. Here is
one possibility. First, write a short narrative of the Self about some event
that is especially meaningful to you. Then step back and look at the nar-
D.rive from your disciplinary perspective and insert into the narrarive-
beginning, midsections, end, wherever-relevant analytic statementS or
530
"'''''''::1' 1'"I",CU'UU UI IIllfuuy
references, using a different typescript, alternative page placemcnt, split
pages, or other ways to mark the text. The laycring can be multiple, with
different ways of marking different theoretical levels, theories, speakers.
and so on. (Thb is an exercise that Carolyn Ellis uses,)
12. Try some other strategy for writing new erhno.sraphy for social
scientific publications. Try the "seamless" text, in which previous litera-
ture, theory, and methods arc placed in textually meaningful ways, rather
than in disjunctive secrions (for a excellent example, see Bochner, 1997);
try the "sandwich" text, in which traditional social scien:e themes are the
"white bread" around the "filling" (C. Ellis, personal communication,
April 27,1998); or try an "epilogue" explicating the dl(:oretical analvtic
\vork of the creative rext (see Eisner, 1996). .
13. Consider a fieldwork setting. Consider the various subject posi-
tions you have or have had within it. For example, in a store you might be a
salesclerk, customer, manager, feminist, capitalist, paren!:, child, and so
on. \\Trite about the setting (or an event in the setting) from several differ-
ent subject positions. \X!hat do you "know" from the difEcrent positions?
Next, let the different points of vie\\' dialogue vvirh ead other. \'X!har do
you discover through these dialogues?
14, Consider a paper you have written (or your field n')te5). \\fhat have
you left our? \Vho is not present in this text? Who has been repressed or
marginalized? Rewrire the text from that point of view,
15. \\Trite your "data" in three different ways-for example, as a narra-
tive account, as a poetic representation, and as readers' theater. \\That do
you know in each rendition that you did nor know in the other renditions?
How do the different renditions enrich each other?
16. \Vrite a narrati\re of the Self from your point of view (such as some-
thing that happened in your family or in a seminar). Then interview an-
other participant (such as family or seminar member) and have that person
tell you his or her story of the event. Sec yourself as part of the other per-
son's story in the same \vay he or she is part of your story. How do you re-
write your story from the other person's point of view? (This is an exercise
Carolyn Ellis uses.)
17, Collaborative writing is a way to see beyond one's own naturalisms
of style and attitude. This is an exercise that I have used in my teaching, but
it \'{ould be appropriate for a \vriting group as well. Each member writes a
Srory of his or her life. It could be a feminist story, a success story, quest
Story, cultural story, professional socialization story, reaLst tale, confes-
sional tale, or another kind of story. All persons' srorics are photocopied
531
 YVIHlIlY f-\ IVlemoo or mqUiry

for the group_ The group is then broken into subgroups (I prefer groups of
three), and c;lCh subgroup collaborates on writing a new story, the collec-
tive story of its members. The collaboration can take any form: elrama,
po,.n,-y. (;e~;",n. na,.",\,;~,<,. ,-,f ,-1-" ."...J.vcs., ,ec.. . l;"""
chooses. The collaboration is shared with the emire group. AI! members
then write abom their feelings aboLlt the collaboration and \vhat happened
to their stories, their lives, in the process.
13. I\/lemory work (see Davies, 1994; Davies et al., 1997) is another
collaborative research and writing strategy. Stories sh3fed in the group are
discllssed and then rewritten, "vith arrcntion paid to the discourses tholt arc
shaping the stories in each of their teHings. As more people tell their
stories. individuals remember more details of their own stories, or de-
velop lln\' stories. Participants discover what their stories have in COIll-
mon, perhaps even writing what Bronwyn Davies (1994) calls a "collective
biography...
19. Consider;] parr of your life outside of or before aGldemia with
\vhich you have deeply resonated. Use that resonance as a '\vorking meta-
pbor" for understanding and reporting your research. Students have crc-
ared excellent repons by using unexpectecllenses, such as choreography,
principles of flower arrangement, art composition, and sporrscasting, to
vicw their lives and the lives of others. \'\!riting from that which resonates
with your life nurtures a more integrated life.
20. Different forms of \vriting are appropriate for different audiences
and different: occasions. Try writing the same piece of research for 3n
academic audience, a trade audience, the popular press, policy makers,
research hosts, ~111d so on (see Richardson, 1990). This is an especially
pmverful exercise for dissertation students who may want to share their
frightfully sterile and dry because you are so quick, snappy, and
efficient about doing one thing after another that you have no time
for your o\vn ideas to come in and develop and g~nrly shinc.
~ ..l,,,,<~ver: ,1,,-, ;;ul,sroup ~D[LlIdl1 UcbnrJ, Ii You
W!cm! to \%!'':, 1938/1987
'$> References
Agger, B. 09S 9). Reading science: A literary, IJOIitical and sociological alli/lvsis. Dix
Hills, I\fY: Gencr<ll Hall. ~
Agger, B. (1990). Jhe decliltc o( discourse: Reading, ruriling and ,Tsistallce ill 1)051-
modem capitalism. Bristol, PA: Falmer.
Angrosino, IVI. V (1998). Opportullity I-Iollse: Ethnographic stori':!s o(ll/cntal retar-
datioll. \\Tal nut Creek, CA: AltalvIira.
133, IVr. (1987). So lung a letter (lvI. Bode-Thomas, Trans.). l'onsmourh, NI-I:
Heinemann.
Baff, S.J. (1997). Realism and n<ltllralism 'Inc! dead dudes; Talking about litcrature
in IJ rh grade English. Qualitative Inqlfiry, 3, 468.490. ~
Balzac, 11 de (1965). Preface to The human coll/edy, from At the Sign of the Cilt and
racket (c. BeJJ, Trans., 1897). In R. Ellman & C. Feidelse'll,]r. (Eds.), The
modem traditiOI/o' Backgl'OlIIlds of modem litem/me (pp. 246-254). New
York: Oxford University Press. (Original work published :842)
Banks,1\., & Banks, S. P. (Eels.). (1998). Fietioll and social rescmc/;: By ice or fire.
Walnut Creek, CA: Altaivlira.
Barlcy, N. (1986). Ceremo11y: An anthropologist's misadl'clItllr;!S i1l the A(ricall
bush. New York: Henry Holt.

results in a "user-friendly" way with those they studied. Barley, N. (1988). Not {/ pleasant sport. New'York: Henry I-lolt.
21. \'V'ri te writing-swries (sec Richardson, 1997), or reflexive accounts Becker, H. S. (1986). \Y/riting (or social scientists: How to (inish )'0111' thesis, book. or
article. Chicago; University of Chicago Press. -
of how you happened w write pieces you have \vritten. Your \\Titing-
swries can be about disciplinary politics, departmental events, friendship Behar, R. (1993). Ti'l7IIslated woman: Crossing the border with Esperm1za's slor\,.
Boston: Be3coll. .
networks, collegial ties, family, anel personal biographical experiences.
Behr, R. (1996). The vulnerable observer: Anthropology that bi'e<:I/~s YOllr hear!.
\\7riting-stories situate your work in contexts, tying VI/hat can be a lonely
Boston: Beacon.
and seemingly separ3tive task to the ebbs and flows of your life, your self.
Behar, R., & Gordon, D. A. (Eds.). (1995). \\'Iomell writillg cultme. Berkeley: Uni-
\V"riting these srories reminds us of the continu8l cocreation of the self and versity of California Press. .
social science.
Bochner, A. (1997). It's abom time: Narrative and the divided self. QlIalitatiue
Inqllil'Y, 3, 418-438.
Willing is doing something yOLl know already-there is no new
Brady,1. (Ed.). (l991). Anthropological poetics. Savage, IvID: ROWllUll &: Linle
im3ginative understclnding in it. And prescntly your soul gets fielcl.
532 533
 " . ~." "~",, '''''', ~., ,~~,-" '''''', ..... ..., "U "~J~" ,... , ,,,,.,
Brodkey, L. (1987). /\cademic writing as social pmctice. Philadclphi<l: Temple Univer-
sity Press.
Brown, K.1\/1. (! 99-] ).l'daJlw Lola: A Vodoll/nics/css ill13rooN)'II. Berkeley; Univer-
sin' DE Cl1i{ornl;1
Pre<.;f;.
Brown, R. 1-1. (1977). A /JOetie (or sociology. Cambridge: Cambridge Universiry
Press.
Bruner, E. j'\'1. (1996). My life in an aShnllll. Qlfalitative Inquiry, 2, 300-.319.
Buder, S., & Rosenblum, B. (! 99 J). Concer ill! lUO voices. San Francisco: Spinster.
Cberry, K. (1995). The best ye,lrS of their lives: t\ portrait of a residential home for
people with AIDS. Symbolic Illtemclioll, 18, 463-4g6.
Church, K. (l9(5). Forbidden lwrmtiues: Critical autobiography as SOCiill sciellcc.
Newark, NJ: Gordon & Breach.
Church, K. (l999). Fabrications: Stitchillg ourselves toge/her fOnline]. Ottawa:
C3nadian Ivluseum of Civiliz;ltioll. Avail<lble Inrernet: hnp:llwww/grannyg.
bc.GI/Fahricari ons/i nclex. html
Clifford, J. (1986). Introduction: Partial trurhs. In J. Clifford & G. E. I'vlarcus
(Eds.), \'/ri/ilfg CIIltltre: The poetics and pofitics of ethnography (pp. 1-26).
Berkeley: University of California Press.
Clifford,j., L~ Ivlarcus, G. E. (Eds.). (1986). \'{,Irit ing wltwe: The poetics m/({/,ofitics
of eflmograjJhy. Berkeley: University of California Press.
Clol1gh, P. T. (1992). The elld(s) of ethnography: From realislll to social criticism.
Newbury P~l["k, CA: SJge.
Crawford, !v1. A_ (l951). Illtrodltc!iollto Old Coriot. New )'orl(; Penguin.
Daly, K., & Dienh:-m, A. (1998). Navigating dle family domain: Qualit<lt:ive field
dilemmas. In S. Grills (Ed.), Doing ethnographic research: Fieldworlc. sct/ilrgs
(pp. 97-120). Thousand Oaks, CA: Sage.
Davies, B. (1989). Frogs and snails and femillist tales: Preschool children. Sr. Leo-
nJrds, Austnlia: Allen &.: Unwin.
Davies, B. (1994). Poststmctumlis{ theor), and classroom practice. Gee!ong, Vic-
[ari3, Australia: Deakin University Press.
Davies, B., Dormer. S., Honan, E., IVlcAllister, N., O'Reilly, R., Rocco, S., &
\\I;llker, A. (1997). Ruptures in the skin of silence: A collective biography.
Hecate: A Womlm's hrterdisci/,/illary JOflmal, 23(1), 62-79.
Denzin, N. K. (1978). The research act: A theme/ieal introdllction to sociological
methods (2nd ed.). New York: l\/1cGraw-Hill.
Dc-nzin, N. K. (1986). A postmodern social theory. Sociological Theory, 4, 194-
204.
Denzin, N. K. (1991). Images of postl11odern societ)'. Newbury Park, CA: S3ge.
Denzin, N. K. (1994). EvaluJting gualitative research in the poststtuctmJI
moment: The lessonsJ3mesJoyce teaches lIs.llltematiol1a!IOltrlwl ofQlIali-
taliue Studies ill Edllc'1tioll, 7, 295-308.
534
""""y. J> IVICl!'VV VI If'yUlly
Denz.in, N. K. (1995). The cillematic society: The vo)'ellr's gaze. Tholls~lI1d Oaks,
CA: S'lge.
Denzin, N. K. (1 997).illterjJretiue ethnography: Etlmogrophic practices for the 21st
","'f'u-y. Th""_"",nd Onk"l C,\; S'lsC:,
Derrida,./. (1982). MargillS alphilosophy (A. Bass, Trans.). ChiGlgo: University of
Chicago Press.
DeShazer, M. K. (1986). Illspiring womell: Reilllagillillg the llltise. New York:
PergJtl1on.
Diamond. S. (1982). Totellls. Barrytown, N'{: Open Book/Station I-lill.
Diversi,1'v1. (19983). Glimpses of street life: Representing lived e"\ perience through
short srories. Qllali/ative IlIqlliry, 4, 131-147.
Diversi, i\-t. (1998b). Lare for school. Wailwfo jOllmal of Educatioll, 4, 78-86.
Donmoyer, R., & Yennie-Donmoyer,.J. (1995). D~lta as drama: I<eflecrions on the
use of readers theater as a mode of qualitarive data display. QlialitL1tir!e
Inqltiry, 1,402-428.
Dorst, .J. D. (1989). The rurittell suhurb: All Amcric'1II site, all etlmc"gnrphic
dilemma. Philadelphia: University of Pennsylvania Press.
Edmondson, R. (1984). Rhetoric ill sociology. London: IvlacmilLtn.
Eisner, E, (l996). Should a novel COUIlt as a disserrarion in education? Research ill
the Tcaching of Ellglish, 30, 403-427.
Ellis, C. (l99J). Sociological introspection and emotional experience. SYlllbohc
InteractioJl, 14, 23-50.
Ellis, C (1993). Telling rhe srory ofsllddcn death. Sociological Qrldrfcrly, 34, 711-
730.
Ellis, C (1995a). Fillal negotiations: 11 stOfY of lot-'e, loss, and chrollic illness. Phila-
dclphi<:l: Temple Univcrsity Press.
Ellis, C (1995b). The other side of the fencc: Seeing black and white in <:I sma!!
southern town. QlliIlitatiue inquiry, 1, 147-168.
Ellis, C (1998). "I hate my voice": Coming to terms wirh minor bodily stigrn3s.
Sociological Quarferl}', 39, 517-537.
Ellis, C, & Bochner, A. P. (1992). Telling and performing personal srories: The con-
straints of choice in abortion. In C. Ellis & 1v1. G. Flaherty (Ecls,), lnuesti-
gating sllhjediuit}': T{esearch on liued experience (pp. 79 -1 01). Newbury
Park, Ct\: Sage.
Ellis, c.,
& Bochner, A. I~ (Eds.). (1996a)_ Com/Josing c/lmography: iIl/enld/iue
forllls of qualihail'/:.' writing. \X alnu[ Creek, CA: AltaMira.
'
Ellis, C, & Bochner, A. I~ (1996b). Introduction: Talking over ethnography. In
C. Ellis & A. P. Bochner (Eels.), Call/posing et/mograIJhy: Alknrafiue (orllls o(
qualitatiue writing (pp. 13-48). \Valnur Creek, CA: Alw!Vlic;l.
Ellis, c., & Flaheny, M. G. (Eds.). (1992). lJIuestigating subjectivity: Rcsc.trch Oil
lived experience. Newhury Parle, CA: Sage.
535
 ",' ~ .... ,'"- 'n, '~", .... ' n ... ~'-" ''-''', ,_." ''-' " .... , ,,,_ .... ,..,. 'r'" ''.- VVlIlIUy; 1-\ IVle(rlUU or mqUiry

Erikson, K. T. (1976). Everything ill its path: Dcstmclioll olthe colllllllllrit)' ill the
BII/~J!o Creel:. flood. New York: Simon & Schuster.
Fine, At (1992). Dismptiue uoiccs: The possibility of femillist rescl7fch. Ann Arbor:
Un;vns;n' o( /d;ch;g::m Press.
Fishkin, S. F. (1985). From ftlct to (iction: }ollrJ/alislIl and imaginative writing ill
AlI/erica. Baltimore: Johns Hopkins University Press.
flick, U. (1998). An introduction to qllalitatizJe research: Theory, method and {[!Jpli~
catiolls. London: Sage.
Fox, 1\/1. F. (Ed.). (1985). Scholarly writing t1ltd publishing: Isslles, problems, and
solutions. Boulder, CO: \X!cstview.
Frank. A, (1995). The IIJOltllded storytellcr: Body, illness, alld ethics. Chic:!go: Uni-
versity of Chicago Press.
Frohock, E (1992). Healing powers. Chiclgo: University of Chicago Press.
Geertz, C. (1988). Worhs alld liues: The t7lllhropologist 115 t1lfthOl: Sranford, CA:
SUlilford University Press.
Ceria, J. E (! 995). An uncommon friendship: Ethnographic fiction :lround finance
equity in Tex'ls. Qualitatiue Inquiry, 1,168- J SS.
Claser. B. C., & Str.:mss, A. L. (1967). The disC(wery ofgrouuded t{}cory: Strategies
lor qlh1lilatiuc research. Chicago: Alcline.
Glesnc, C. E. (1997). That r;]r( feeling: Re-presenting resezlrch through poetic tran-
scription. Qualitative TlIquiry. 3, 202-221.
Corning, A., & Fenstermaker, S. (1995).lndiuiduallmiccs, col/ectit'e l'isiOl/s: Fifty
years of IUOlllel1 ill sociology_ Philadelphia: Temple University Press.
Goldberg, N. (1986). Wkitillg dOllm the bones: Freeillg the II/riter withill. Boston:
Shambal:!.
Goldberg, N. (1990). Wild milfd: Liuilfg the writer's life. New York: BantJlll.
I-''1arper, D. (1987). \'(1orl~illg Imowledge: Sbl! and COllI/lilli/it)' ill a sml1l/ sho!J. Chi-
C3g0: University of Chicago Press.
I-lills, R. (1987). W+iting in gelleml alld the short story ill partiwlm: Boston:
Houghton l\'lifflin.
hooks, b. (1990). Yct/wing: Race, gel/del; alld cultTlral tlOlities, Boston: South End.
Hurswn, Z. N. (1991). Dust trachs all a road. Ne\vYork: J-larperCollins. (Origin;]]
work published 1942)
Hutcheon, L. (1988). A poetics (or postl1lodemislll: History, theory, fictiol!. New
York: Routledge.
jacobs, J. (1984). Themall: All attempted escape (rom euerydilY lire. Prospect
Heights, TL \,\laveland.
Jago, B. J. (1996). Postcards, ghosts, and bthers: Revising family stories. Quoli!i1-
fiue Inqlliry, 2, 495-516.
JeuTIeson, F. (1981). The political lIIlCOllSciof/s: Narratiuc as a socially symbolic aet.
Irll,lca, NY: Cornell University Press.
Jipson, j., E.-[. Palcy, N. (Eels.). (1997). Daredevil research: Re-aeRtillg allalytic prac-
tice, New York: Petcr Lang.
Johnston) Iv1. (with Educators for Collaborative Change). (199 7 ). COlltradictiems
ill cothl!Jortifioll: NeIU thillkillg 011 SC{;Ool/ll11iuersily partmTshi/Js. New lork:
Teachers College Press.
Jones, S. H. (1998). }(aleidosco!)c 1/otes: \Vritil1g women's Itlltsic 'il1d organizational
Cltlture. WZt!lllH Creek, CA: A!taMira.
Karp, D. (1996). Speahillg o( sadness, New York: Oxford University Press.
Kaufman, S. (1986). The ageless self' Sources afmcallillg in later I/(e. .l\-,tJdison: Uni-
vcrsity of \X1isconsin Press.
Kondo, D. 1\. (1990). Crafting selues: Power, gelldel; and discollr:;cs o(idelltitl' il! a
Japal/ese ulorl~!)!ace. Chicago: University of Chicago Press. .
Krieger, S. (1983). The mirror dance: Identity ill a lllolllell's com/lilli/it)'. Phib.del~
phia: Tcmple University Press.
Krieger, S. (1991). Social sciellce and the self: PersoJlal essa),s Oil ,m art (01'111. New
BfIlflswick, l\IJ: Hmgers University Press.
Krieger, S. (1996). The fall1iI-y silucr: Essays on rcll1tiolfshitls among women. Berke-
ley: University of California Press.
Lakoff, G., & Johnson, Iv!. (1980). AJetaphors we liue by. Chicago: University of
Chic3go Press.
Lather, P. (1991). Gcltillg smart: felllinist research alld pedagogy with/ill the post-
modem. New 'York: Routledge.
Lather, l~, & Smithies, C. (1997). HOlfbling the allgels: W/ollleJl living luith HJl.lj
"UDS. Boulder, CO: \X1estview.
Lawrence-Lightfoot, S. (1994). I'vc ImowlI riuC1"s: Liucs o( loss alld liberatioll.
Boston: Addison-Wesley_
Lawton, J. E. (1997). Recollcept/lalizing {l horizontal career line: A study of seuen
e.c-pericl1ccd w[h11l English teachers approaching career e;rd. Unpublished
doctoral dissertation, Ohio State University.
Lee, V (l99(~). Gralln)' midwives alld black n'omen luritas. New r'nrk: Routledge.
Lehman, D. (1991). Signs of the timcs; Decollst.mctioll and the fall oj Pmzl de !vImr.
New York: Poseidon.
Levine, D. N. (1985). The flight (rolll ambiguity: Essays ill social ,md wIturi11 the-
or)'. Chicago: University of Chicago Press.
Lieho\\; E. (1967). ]~J!!)"s comer: A stltdy o(Negro street comer me.'l. Boston: Llule,
Brown.
Linden, R. R. (1992). }vlaking stories, waking selues: Feminist reflectiolls 011 the
Holocaust. Columbus: Ohio State University Press.
Lockridge, E. (1987). F. SCOtt Fitzgerald's Trompe l'oeil and Tho: great Gatsby's
buried ploL}oflrnal ofNarrafil'e Technique, 17, 163-183.

536 537

Lyurard, j.-E (191:\4). The postnlOdcnI conditioll: A report (Jlt Imowledge
(G. Bennington & B. j\'Llssumi, Trans.). lvlinne:>polis: Universiry of IvIinnc-
sCJta Press.
l\i'brgnl;<:. r:.,
&: R.mlH'l"(). fvl. (100gj. Tht: df'\l:1ttmcnt it: VP1'Y n1:1!e, v~ry white, very
old, and very conservative; The functioning of the hidden curriculutJl in
gradu:ue sociology departments. Hanmrd f:dllcatiolwl Rel'ii'll', 68,!~32.
I\kC:>l!, M. lvI., & Becker, H. S. (1990). Performance science. Socia! ProlJle1lls, 37,
11(;-132.
lvteCall, 1\'1. .i\'1., Gammel, L., &. T:l}'lor, S. (1994). The aile abo lit the t;mllcr's
d<lllghter: StereotyjJes alld self jlortr<lits. IVIinneapolis: Country Characters.
tvldvbhon, J\t (1996). Significant absences. Qualitative illquiry. 2, 320-336.
i\leluy, J. M. (1993). Problems of writing and represemation in qualitative inquiry.
l/1tcrlliltiOlwl.lolll"llal of Qlwlitatiue Studies itl Educatiol/, 6, 315-330.
lVlicnczako\vski, j. (1996). An ethnographic act: The construction of consensu,11
theater. III C. Ellis & A. P. Bochner (Eds.), CO/llposing et!mography: Altenw-
til'e it:mlls o(q/wlitatil'C writillg (pp. 244-2(6). \\la lnut Creek, Ci\: AIt:ll\-lira.
0Aishkr, E. C. (I ()89). l?esearch il/teruiewillg: CClI1te.YI and narratiue. C1lllhridgc,
i\IA: Harvard University Press.
Nelson, J. 5., rVlegilL 1\., & IVIcCloskey, D. N. (Eds.). (1987). The rhetoric of the
hllllldll scienccs: L.anguage and argull/ent itl scholarship alld fnll/tall af/~lirs.
rVladison: University of \X1iscol1sin Press.
Nicholson, L .J. (Ed.). (] 990). Femhlisl!I/postlllodemislll. New York: Routledge.
Norurn, K. E. (i 11 press). School p:urerns: A sextet. Il/tenwtional.lollrJIal o{Qrwlita-
tille Stl!dies ill Edllcatiol/.
Paget,Ivt (1990). Performing the texr.jourl/al o(Colltempomry Ethllogri1phy, 19,
136-155.
Pandolfo, S. (1997).lmjJi1sse of the ill/gelS: Scenes fro1l1 ([ MOroCCdll S{HICC of melJ/-
ory. Chic3go: Ulliver~ity of Cbicago Press.
Parai, D. (1988). Constructing a self: A Br:1Zili3n life story. Femillist Studics, 14,
142-163.
Pfohl, 5.]. (l992). Deat/; at the n11"c1site Cafe: Socia! sci2llCi!. (fictiolls) and the post-
modem. New York: St.l'vlanin's.
Prattis, 1. (Ed.). (1985). Re(lectiolls: The allthrojJ%gical tIluse. \X!:lshington, DC
American Anthropological Association.
Richardson, L. (1985). The l/ew other woman: COlltcl1ljJorary single women ill
a(f7irs with lJlarried mel1. New York: Free Press.
Riclnrdson, L. (1990). Writillg strategies: Reaching diuerse audiellces. Thousand
Oaks, CA: Sage.
Richardson, L (J 992a). The consequences of poetic representation: \'i/riting the
other, rewriting the self. In C Ellis & M. G. Flaherry (Eds.), ltwestigl1tillg
subjectivity: Research 01/ livC/,.-1 experiCllce. Newbury Park, CA: Sage.
538
"''''''=,. ""'V'",."'UUUllffl/UIlY
Richardson, L. (1992h). Resisting resistance narratives: A repres:nration for COlll-
lllunicarion. In N. K. Dcnzin (Ed.), Studies iI/symbolic il/teraction: II rest'dl"Ch
dlllllWI (Vol. 13, pp. 77-83). Greenwich, CT: JA!.
r-t.;ch:1rdson. L (199.1). The C'-151.:: of l!Jc~li:ip]Jcd line: Funic;;, dralllancs and
Iransgressive validity. Sociological QUi/rterly, 3-1, 695-710.
RIchardson, L (1995). Writing-stories: Co-authoring "The se:l .110nSter," ~l writ-
ing-story. Qualitatiue Inquiry, 1, 189-203.
Richardson, L (1996a). Educational birds. jOlll"lwl of COl/telll/)Or,lry Et!mogmp!}y,
25,6-15.
Richardson, L (1 996b). 1\ sociology of responsibility. QUilJitati/ic Rescilrch, 19.
519-52.:1.
Richardson, L. (1997). Fields of pllly: COllstructillg illll/cadclI/ic li!;,. Nt'w Bruns-
wick, N.J: Rutgers University Prcss.
Rich:mJson, L. (] ~)98). The politics of location: \,\lhere am I now? QlIl1litiltft'C'
Iltquiry, -I, 41-.:1R.
Rlch~rdson, L. (1999a). De:ld again in Berkeley. QI,,7litdtil'c IIlquiry, 5, 141.J+1.
Richardson, L. (1999b). Paradigms lost [Disrillguished Lectllre]. Symbolic illieritc-
tirm, 11. 79-91.
Richardson, L, L\: Loclcridge, E. (1991). The sea monster: An ethnographic drann.
Symbolic IJlter,zctio/l, 14,335-340.
Richardson, L, & Lockridgc, E. (1998). Fiction and ethnography: A conversation.
Q/wlitdtit'c II/quiry, 4, 328-336.
Rinehan, R. (1998). Sk8ing. W/ailwto./Oltl"lwl olEdl/cation, 4, 87-100.
Ronai, C. R. (1992). The reflexive self through narrative: A night in the life of ;111
erotic dancerlresearcher. In C. Ellis & jVI. G. Flaherty (Eeb.), 11I1.'cstig'ltillg
sllbjectiuity: Research oilliued experience (pp. J 02-124). Newbury Park, CA':
Sage.
Ronai, C. R. (1995). JVlultiple reflections of child sexual abuse: An argument for;l
layered aCCOUllL./OIt/"lla! olColltcmporilrY Ethllogmphy, 23. 395-426.
Rose, D. (1989). Pi/Items ol American clI!tllre: EthnograjJhy dnd estl"<7ngelllcllt.
Philadelphia: Universiry of Pennsylvania Press.
Rose, E. (J 992). The //lewld. Boulder, CO: \\l;liting Room.
Rosc, E. (1993). The u..-'o/"/!/de. Boulder, CO: \\;Jaiting Room.
Rubin, L. B. (1976). Worlds ol/will: Life in the wor!::.hlg-c1ass t~-"Jlily. New York:
Basic Books.
Schneider, J. (1991). Troubles with textual ::lUrhnrity in sociology. Symbolic Interl/C-
tion. 1-1,295-320.
Schwalbe,1v1. (1995). The responsibilities of sociological poers. QIf..7!ita!iue Sociol-
ogy, 18, 393-4ll.
Shapiro, M. (1985-1986). IVlet3phor in the philosophy of the sociJ sciences. elll-
tllm! Critique, 2, 191-214.
539
She Iron, A. (1995). The man at the end of the machine. SYlllbolic I!Juraction, IS, Weedon, C. (19S7).1~1/1inisf practice alld !Joststmcfumlist theory. New)(Jrk: nasil
50S-5! 8. BIJchvell.
Shost';]k, A. (Ed.). (19%). Priuilte sociology: Ulls/Jari/rg r'-'flectiolls, 1IliCOllllllon \'\lcinstcin l D. (19')J). m'iling (01' )'011)' life.- A gil iii,! am] rXllllpLl/liol1 to fhe iII/fa
iY"" . Di", I--lill,;, hT)', General l-lall. worlds. New \'orb j-brpcrCollills.
Simons, j-f. W~ (1990). Rhetoric ill (he humall sciences. London: Sage. \'('hyre, \\( F. (1943). Street comer society: The social stmctlfre o{ dll {tohm sIll/}[.
Sinhin, K. (1995). Fieldwork and subjectivity: On rhe rit'LIa!iz~Hion of seeing a Chicago: University of Chic<lgo Press.
burned child. Symbolic Interaction, 18,487-504. Whyte, \X~ F. (1992). 1n defense of Street comer society. JOIIl"lwf 0/ COllfclIIj)ol"l1rl'
Sparkes, A_ C. CJ 997). Ethnographic ficrion and representing, the 3hsent mher. Etlmogrdphy, 21, 52-68. .
Sport, Ed[f({/tioll, and Society, 2, 25-40. \\:Jilli3IllS, P..1. (1991)_ The alchemy ofmce alld rights: Di'1r)' of a lall' professf.JI: Cam-
St:lck, C. B. (1974). All OIlr kill: Strategies (or survival in a blach WlIl/Ill/llity. New bridge, i\1.A: Harv:lrd University Press.
'kJrk: Harper & Row. \Xli/son, C (1965), Crazy February: Death and li(e ill thc AJdy,ill hi,;hl,mds of A'Ie\"-
Statham, A., Richardson, 1,., & Cook, J. A. (1991). Gender and !tniversity teaching: ico. Berkeley: University of California Press.
A Hugotidtcd differellce. Albany: St<lte University of New "y'ork Press. \\7olcort, H. F. (1990). \\?rilillg Ifp qualitative reseilrch. Newbury [1,lrll, CA: Sage.
Stcedman, K. (1986). Landscape for a good WOJlli1Il: A story of two liucs. New Wolf, J'vi. A. (1992). A thriaHold talc: Fcminism, postmodcJ"lIislll, alld etlmogrdphic
responsibility. St,mford, CA: Stanford University Press.
Brunswick, N.J: Rutgers University Press.
I'll, 1>L (1997). [-fullgry lightni/lg: Notes o{a WOIlUl/l ,mthropologi5t in \~;I!e-::lle!il.
SlcW;lrt, J. (989). Drinkers, drlm/mas and decCltt (o!l~: Ethnographic Iwrrativcs of
Albuquerque: University of New J'dc:,ico Press.
Village Trinidad. Albany: Stare University of Nc\v 'fork Press.
Zob, E. (1965). The novel as soci<:J! science. In R. Ellm;ln & C. Fc:icklsotl, Jr. (Eels.),
Stoller, I: (1989). The taste ofetlmographic (hings: ]'1JC senses ill allthropolo6'Y' Phil~
The modem traditioll: Backgrollnds ofJr[odcnr literatllre (pp. 2 7 0-239). New
adelphia: University of Pennsylvania Press.
York: Oxford University Press. (Original work published I S SO)
Sf. Pierre, E. A. (1997a). Circling the rext: Nomadic writing practice~. Qlwlitatiue
Zola, 1. K. (1982). Missing pieces: A chronicle of lil/ing with 11 dis{rbility. Phil;ldel-
Inqlliry, 3, 40.3~417.
phia: Temple University Press.
St. Pierre, E. A. (1997b). Nomadic inquiry in the slllo()[h spaces of the field: A pref-
acc. III!CnratirJl1al jorll"!url o(Qlfalitatiue Studies ill Educatioll, 10, 175-189.
Tedlock, D. (l9S3). The spohen word and the lUork o( interpretation. Philadelphi<l:
University of Pennsylv:mi<l Press.
Thorne, B. (1993). Gender /Jlay_ New Brunswick, N.J: Rutgers University Pre~s.
Trinh T. tvi. (19B9). \'{loman, natiue, other: Writillg postcoloniality and (eminism.
Bloomington: Indi<lna University Press.
Turner, v., & Bruner, E.1VL (Eds.). (1986). The anthropology ofexperience. Urbanu:
University of Illinois Press.
Ueland, B. (1987). [fyoll wallt to Ivrit e: /I book aboltt art, indepelldcJlce and spirit.
Saint Paul, MN: Graywolf. (Original work published 1938)
Ulmer, G. (1989). Te!etheory: Gm/lllllatology ill the age of uideo. New York:
Routledge.
V<ln lVlaanen, J. (1988). Talcs o( the (ield: 0" writing etlmography. Chic3go: Uni-
versity of Chicago Press.
Van Ma;:men,]. (EeL). (1995). Re/JTesentation in ethnography. Thousand Oaks, CA:
Sage.
Vis\veswarnn, K. (1994). Fictions o(feminist ethnography. IVIinneapoiis: University
of IVlinnesota Press.
\X!;:llkerdine, V (1990). Sclwolgir! fictions. London: Verso.

540 541

Anthropological Poetics
Ivan Brady
Poetics is a topic llsually associ8ted with the systematic study of Jiter~
3rure, but, especially in its modern concern with texts as "cultural
anifacts," it extends to anthropology in several vvays.l One obviolls con-
nection is that, like other academic disciplines, anthropology is "literary"
in that it conveys its information primarily through writing. This textual
base lets anthropology share with 1110re conventional studies of poetics an
interest in text construction, the authority of the text, semioric beh,lVior
and the production of me,ming in discourse, and, in general, aU the philo-
sophical and critical problems associated with mimesis-the representa~
[ion and successful communication of experience in any form, especially
as problematized in texts. It also brings to the fore something that anthro-
pology is predisposed to engage because of its 0\'.'0 diverse history: debate
over the place of art and science in the social sciences and the humanities.
Anthropology has intellectual camps at both extremes, that is, strong
science orientations that conscientiously attempt to exclude more artistic
AUTHOR'S NOTE; This chapter is ;] sllbst:J.nti~Jly revised and exp;]ll(le.d version of Brady
(1996). It h;JS benefitled from generous readings by James A. Boon, Norman K. Demin,
gobcn Boroisky, j);Jn Rose, f\liJes Richardson, Yvonna S. Lincoln, James \V Fcrn:llldcz, and
Barbara TcdJock, ,md r wish to rh,mlc them. Any remaining wrongheadedness is enrirely my
own. "Sh3m:m's Song" is r<:primed wirh permission of Station I-Jill Press (Barrytown, New
York), from Smnle)' Di:lmoncl's Totems (1982, p. 94).
542
MIIU IfOpO/OglCOI J-'oetlcs
or humanistic methods and interpretations, and vice vcrs", (see rujimarJ,
1998). Overall there is a compromise of identity: Anthropology sees itself
as an "artful science" (Brady, 19903, 1993), That leaves roOill for en!J~lUil1g
n variety of posrfJloclern challenges from other disciplines, including much
that has misen under the labels literary and pocl;c (Brady, 1991 b).
Cultural anthropology in particular encompasses both individualized
studies and systematic comparisons of cross~ClJltural experience, so its
potential poetic sources are broadly based. That hJ5 allownl amhropolo~
gists to feed into and draw reciprocally on several disciplin::s \vhile devel-
oping their own specialized studies of the forms and comeiE of poe ric pro~
duetion in their own societies, in other cultures, as wdl as in their
communications about other cultu res ..'. Playi ng in such ficlc!:; begs;] variety
of critical issues, from cuI rural relativism to competing philosophies of
representation that have long histories of contemplation .11 other disci-
plines and less attention in anthropology. Theorists from ot:ler disciplines
have reached into anthropology on similar issues (sec Clifford, J988;
Clifford & .iVlarcus, 1986; Krupclt, ] 992). Bur the cross-poaching th;u
underwrites anthropology's "literarizarion" and the "anrhropologizing"
of literary studies is hard to meJsure and subject t'O varicus inrerpreta-
tions. J It has also resulted in some confusion over the degree to which
anthropology ought to rely on other disciplines instead of crcating its 0\'1'11
adaptive solutions to the challenges posed, for example, t<::garcJing "the
effort to situate discourses sociologically, to show how discourses func~
tion, compete, and clash within sociopolitical arenas, and to trace hmv
discourscs are transformed historically" (Fischer, 1988, p. 8; compare
Krupat, 1992, pp. 51-52; Taylor, 1996).
One famous connection between anthropology and other disciplines
that study poetics is structuralism. Although now less popular (having
been relegated mistakenly by some to the graveyard of things buried by
"postmodern" growth-see Brady, 1993), the structuralism developed in
this century by linguists Ferdinand de Saussure and Noam Chomsky, psy-
chologist Jean Piaget, and especially anthropologist Claude Levi-Strauss
had a profound impact on literary studies in the 1960s and 1970s:1 The
concept of poetics <md what might constitute a legitimate studv of it has
been permanently changed as a result, and the po;sibilities f'Jr demarcat-
ing something clearly as a17throfJological poetics have been complicated in
the process. s That doesn't disappoint most of the modern critics who
operate happily bCf\veen the cracks of conventional divisions. Poetics is
now more than ever an interdisciplinary topic that rightfull) resists such
543
 11'1 I l;:J\rrU:;.IMl lVI", o:.VMLVM' 'V", MI"'-' "Lr !"L-"_" 'rI' ,OJ, 1
rednctions. Nevertheless, anthropology's footprints tend to be distinctive,
so marking its path through this terrain is not impossible. Even though
anthropological inguiry into poetics lacks coherence as a specialized field)
clone v'lith it and by marking the intersections of their \vork with like-
minded activities from elsewhere.
\V'ithout departing entirely from the common literary grounds of writ-
ing as a form of expression and what constitutes dllthor-ity in the first
place, anthropological studies of poetics differ from those of most other
disciplines by rcaching into performance issues in discourse, includ-
ing studies of rimed and vvorldview, from tribal societies to modern
theater, and their relationships to langl18ge ;md culture. li Ethnopoets in
particular-cultural anthropologists, linguists, and poets who study other
cultures from a poetic perspective-never lose sight of the language that
structures and facilitates discourse, in oral or written form.! These are
some of the reasons the subject of anthropological poetics gravitates natu-
rally to sociolinguistics and linguistics in general (which has its own gra-
dations of science-minded versus more artful practitioners), residually t'O
the empirically rigorous studies of semantics and cullllre popularized in
anthropology in the 1960s and 1970s as "cognitive anthropology" (see
1yler, 1969), and directly to discourse-centered production in any form,
including writing. They are tied together by the instrumentClI role lan-
guage plays in each instance.
The emphasis on the centrality of language in culture was also one of
the original attractions of structuralism and poetics to linguistics and at
the same time a brace for some of the earliest discontents with logical posi-
tivism in ethnogrJphy, which hist'Orically has tried to keep its rhetorical
bases and authorship "invisible" as a pretext of clinical dist3llce. An
increased understanding of the collaborative nature of fieldwork and the
need for more "reflexive" perspectives on it relative to constructivist (e.g.,
the reader reception theories of scholars such as semiologist Umberto Eco
and literary critic Roland Barthes) and multi vocal interpretations of litera-
ture (following especially the seminal work of Russian critic 1vlikhail
Bakhtin) have also spurred the development of a critical dialogics for lin-
guistics and ClTIthropology as a whole.1! The challenge for a dialogic poetics
is that it "must first of all be able to identify and arrange relations between
points of view: It must be adequate to the complex architectonics that
shape the viewpoint of the author toward his characters, the characters
toward the author, and all of these toward each other" (Holquist, 1990,
544
p.. 1(1)~ Ethno.grapl~ers have been attracted to this kind of argument pre~
Clsely because It defllles both the relaTionships and the COlJ:;tralnts 011 self-
conscious cross-cultural research ;-mel writing. Bur (he sllb~,equent rurn 10
linsui,;rics ~l-S scicncl: und IIIlJdcl for CUlrUG11 and rcxtu:d studies and to bn-
gU3ge in some strict sense as the key [0 ethnographic in'lestlgation has
proved to be inadequate-or at least incolllp!ere-and is the -source of
many of the indetcrminacies in philosophy and method rlUt characterize
curr-cnt.debate on ethnographic representation and interprc[;]tlon.')
~. he lIltc.rdisciplin3? stretch of what call be marked ;]s the amhropo-
Joglc~l vcr~lOn of poetlcs thus covers a lot of ground. It r8flf;es from a self-
CO~~SCIOUS mtercs~ in {JOel]")'., conccivcd brightly' by poet Rita Dove (1994)
as the art of makmg the interior life of one individual ;Jvail;Jble to others"
(p.1S; see also Prattis, 1986, 1997), on the one hand, to milch marc inclu-
sive analytic interests that Clre perllClps best contained bv Fre!lCh critic Paul
V~lcry's (1964) concept of poetics as "everything rha~ belrs on thc cre-
auon or composition of works h<lving language at once as their SUbst<.lflCC
~nd their ~nstrumellt" (p. 86; sec also Brady, 1991a), 011 thc tither. Splicing
IIUO th8t fUZZy frame\vork frol11l11any directions, amhropological poetics
nevertheless settles mainly into three subcategories of inqu.ry, which are
nor mutually e~clusive: (a) erhnopoetics, (b) literary anthropology, and
(c) anthropological poetry_
<> Ethnopoetics
Erhn.opoetic~ may be .the mosr conspicuously anthropological activiry in a
poetic domalll. Denl11S Tedlock (1991), a pioneer in the field defines it as
the "study of the verbal arts in a worldwide range of langu;~ges and cul~
tures" (p. 81). It focuses primarily on "the vocal-auditory ch;lnllel of com-
munication.in which speaking, chanting, or singing voices ;ive shape to
proverbs, nddles, curses, laments, praises, prayers, ptoph,~cies, public
announcements, and narratives" (p. 81). The goal in such sudies i~ "not
0:1]y to analyze and interprer oral performances bur also to make them
dlr~ctly a~~essibJe through rranscriptions and translations that display
their qualitIes as works of an" (p. 81)..Much has been done eO reach this
goal since erhnopoetics was invented as a special genre of inquiry in the
United States in the [ate 1960s.
. Poet-ethnographer Jerome Rothenberg coined the term dlmofJoetics
In1968 and is properly considered to be "tbe father of All1eric.111 e[l1no-
545

poetics" (Tarn, 1991, p. 75). His most instructive thoughts on this topic
arc collected in edited volumes that also iHustrmc the diverse intcliectl18J
inspif<1tiol1 this movement owes to various social scientists) ethnogra-
pl..,er:;, i'tnd puet:; \illduding especially Jienry D:Hlid Thore:lU, Gertrude
Stein, Ezra Pounu, Anhur Rimbaud, and \\!illiam BlakeL and to other
influential social thinkers (such as Giambattista Vico and surrealist Tristan
Tzara-see Rothenberg & Rothenberg, 1983). Some of Rothenberg's
other work (e.g., 1981, 1985) digs deeper into the roots of "oral poetry"
(see Finnegan, 1992a, 1992b) and related traditions as old as the late Pleis-
tocene (see Bmdy, 1990b; Tarn, 1991, p. 15) and (]s momentoUS as the
birth of theater and poetry in shamanism (Rothenberg, 1981). A common
theme throughout is an attempt ro close the distance modern thinking
tends to put bet'vveen "us" and "them," both historically and as these artifi-
cial boundaries are llsed to separate us from the performative traditions of
"ex-primitives" around the globe today (see Bauman, 1977, 1992;
N[acCannell, 1992; Schechner, 1995). The author of numerous books or
poetry, Rothenberg \vas also a cofounder (with Dennis Tedlock) and co-
editor of the radical magazine AlclJerillga/Etlmopoetics, which featured
"transcripts, translations, and tear-out disc recordings of performances by
indigenous verbal anists from Africa, Asia, Oceania, and the Americas"
(D. Tedlock, 1992, pp. 81-82). It was keenly focused on developing
ethnopoetics, on freeing poetries of all kinds from the "monolithic great
tradition" of \\Testern literature, and on exploring new techniques of
translating the poetry of tribal societies. Although Alcheringa is no longer
in print, its experimentalism continues to characterize the field of
ethnopoetics roday (see especially Tedlock, 1983, 1990, 1992).
The innovative work of linguist-anthropologists Dell Hymes, Robert
Duncan, and George Quasha, anthropologist-poet Stanley Diamond, and
other poets who, like Rothenberg, have had some training in anthropol-
ogy or linguistics, including David Antin and Gary Snyder, must also be
considered in calculating the intellectual history of ethnopoetics. Their
collective scholarship has influenced all the others who have contributed
to the development and preservation of this genre since the early 1970s.
One such person was anthropologist Victor Turner. \\Triting more than a
decade and half ago aboUt the connection of ethnopoetics to his own work
on ritual, Turner (1983) suggested that ethnopoetics offers a way of re-
newing recognition of "the deep bonds between body and mentality,
unconscious Jnd conscious thinking, species and self" that "have been
treated without respect, as though they were irrelevant for analytical
546
" , nl" U/JU1UyILUI rUl;'ilC.'>
purposes" in much of the intellectual discourse from anth,opology's colo-
nial period (p. 338). Hc also nored poignamly that the resurgence of
erhnopoetics "come~ Jt;1 time when knowledge is beJng Increasecl of other
cultures.. other worldvinvo;, other life styles, when \\lesterners, endeavor-
ing to tfap non-\'{les tern philosophies and poerries in the corrals of their
own cognitive constructions, find that rhey have caught sublime monsters,
eastern dragons, lords of fructile chaos, whose wisdom makes our knO\vl-
edge look somehow shrunkcn and inadequatc" (p. 3J8). Appropriate to
bridging ethnopoetics and anthropology's struggle for;:] postcolonial
idemity, such concerns also lead directly into the rea!Jr of humanistic
anthfopology.lo
Anthropologist-linguist Keith Basso (1988) points to <1norher major
problem still on the ethnopoetic horizon that is centered in the larger issue
of hegemonic discourse-a cultural bias or override in th::: linguistic and
intellectual forms we use to appropriate and represell t cros~i~cllltural expe-
rience. There is, Basso says, "a growing com'inion among linguistic
anthropologists that the oral literarures of No.tive Americll1 people have
been inaccurately characterized, wrongly represented, and improperly
translated." For the better part of a century, "the 5poken producrions of
Native American storytellcrs have been presented as pieces :If prose whose
formal divisions are marked by paragraphs and senrenccs." Recent re-
search indicates that this is a fundamental distortion of the record. It
appears that "Native American storytellers often spoke--and in some
Indian communities continue to speak today-in form~ of measured
verse" (p. 809).11 The ethnopoetic task is to decide on th,~ kinds of evi-
delKe that "attest to the existence of these poetic forms" and to ans\ver a
variety of rclated questions: "Givcn a properly recotded text, together
with a knowledge of the language in which it was made, hO\\ should analy-
sis proceed? \Vhat kinds of theoretical constructs are called for along the
way, and how should these constructs be modified and refined?" (Basso,
1988, p. 809).12 Answering these qucstions sweeps through many of the
postmodern challenges that have surfaced in ethnopoetics and anthropol-
ogy in general in the past 20 years. The concerns cluster around cultural
and historical "situatedness"-our inability ro devise a culture-free or
purely objective view of anything. It is hard to avoid the ?rgumenr that
scholarly work is culturally constructed, rhetorically conditioned, tropo-
logical, empowered with a poim of vie\'V, and addled with imperfections,
distortion, and incompleteness. Everything a scholar produces is, in effect,
"textual" in every sense of the term (sec also Rofty, 1979, [981; \'{Illire,
547
 "" ,-,.' ,,,_or,, ''-''', ,-,,..,,-vn, ''-''', r;" ..... ,,'-, "'--''-'' ,r;, ''-'"
and history per se (l\1anganaro, 199Gb). But, despite its needs for objectiv-
ity and the defiance of ethnocentrism, ethnographic writing can "no lon-
ger be seen as a natural or organic extension of content"; the anrhropolog-
leal \vrltt:r CirtltT plugs into preexisting plot gtl'Ucturcs or creates a new
structure out of the amalgam of old forms" (.Manganaro, 199Gb, p. 15). By
allowing no culturally or linguistically unconstructed access to reality-no
story structures unconditioned by history-and by virtue of its roots in
the "fuzzy" cross-cultural margins of anthropological fieldwork, ethno-
graphic representation is thus conducive to developing the kinds of com-
peting interpretations that go with the dedar3.tion of "fuzzy" boundaries
on any topic. In that capacity it easily lends itself to further analogizing of
the situation as ambiguous "text." 21
The attempt to reconcile the role of the author, the place of fiction, and
related literary enmeshments in ethnography's realist writing tradition
Jeclds to two underexarnined developments in the field, one old and one
ncw, and it begs a third. First, it shows the need for a critical reexamination
or \vriting constructions and realist assumptions in anthropology's only
established poetic genre, the ethnographic novel. 22 Second, there is an
active search on in many quarters of anthropology to adopt more obvi-
ously literary forms that can be used to enhance communication of the
ethnographic experience in the realist tradition, including those with
greater "writer consciousness" (e.g., where the author appears in first per-
son as narrator and actor in the ethnographic account, contrary to the
positivist tradition).13 Although there are now many variations on this
theme,24 Levi-Strauss's 7Jistes ll-olJiques (1961) is by far the best known
of all such works and was foundational in "its resistance to the duty of
directly rendering the anthropological subject. "25 It "represents the an-
thropological text that perhaps most influenced current speculation on
the nature of anthropological representation" (Manganaro, 199Gb,
p. 17). One enduring consequence is that anthropologists arc much more
alert to their roles as writers and critics today in ethnographic and histori~
cal research,26 and to the role of writing and criticism more generally in
anthropological method and theory.27
Third, the ne\v emphasis on creative construction and prolonged atten-
tion to anthropology's inevitable textual involvement has created a kind
of epistemological havoc. It not only pits positivism against all of the great
deconstructive "undoings" of late, it also means the loss of invisibility in
writing as part of the realist tradition. That is a direct challenge to science's
550
- - - - - - - - - - - - _ . - - - - .. _ . _ - - -
mythical perpetuation of the unobtrusive sign (comp~lre Barrhes, 1972).
By accentuating the form of the message rather than tbe contents of its
speech acts, the literary or poetic takes for its primary objeer wh:1r science
does nor. In the: {orrnalislsensc, we can cull OUl a measure of literariness in
any text and call it poeticity: the degree [Q which a \vork flags rhe linguistic
nature of its own being; the degree to \vhich it emphasizes materiality ver-
sus transparency through self-referencing linguistic forms (see Jakobson,
1987). A text high in poeticity, \vherher prose or verse, in this sense signals
itself in place instead of disappearing; it ce!ebr8tes the signifier over the
signified without abandoning the basic communication funcrion it shares
as discourse with more scientific or hiswrically ;lL1thentic Gllculations,
albeit through very different channels (compare Levine, 1987, 1994).
\X1riting in a way that docs not call attention ro itself in oreler to ;l11mv the
objective observer to focus directly on the "reality" of the subject is a delu-
sion-a special kind of fiction and desire that helps ro turn "culture" inro
"nature" in the common view. It hides precepts and politics alike behind
the power of the sign and makes them look invisible (Brady, 1991a,
p. 216). The narrative discourse of science, as in art, is cre::ned OtH of the
interaction of such culrur8l conventions and situations, the way the author
deploys them through pMticular language codes, and the reader's process
of reception (creative construction) that releases meaning from the texr.
Clarity of meaning itself takes on new meaning in this context. It lacks
absolutes. It becomes a manipulable cultural code, reckoned \vith familiar-
ity and history-still" fuzzy" around the edges. It cannot be a culture-free
peek at the universe enabled by a mythical objectivity (Brady, 1991b,
p. 19; see also Bruner, 1986, 1990; compare Barthes, 1968).
Science, of course, needs language to function, but unlike literature, it
neither aSserts nor sees itself as situated luithiJ1 language. The conspicu-
ously poetic author willingly appears in his or her text as an artisan whose
constant display is the craft of language (the language and form of the
poem is claimed by the poet and must be read through that claim-a pro-
prietary function) or as a person who visibly leads the prose narrative from
within. The scientific author tries to avoid both, aspiring to invisibility in
every place except the opening credits of his or her cover page (compare
Geertz, 1988, pp. 7-8; Manganaro, 1990b, pp. 15-16). The resulting gap
between author and text is a pit for various detached composition strate-
gies, including anthropology's great timeless fiction of writing in the
ethnographic present. Aurhorial invisibility enhances the prospect for
551

smuggling distancing time frames (we are "present" and full of timci
Others are "past" and "timdess")2H into the argument, not to mention
excessively clinicJI assumptions of philosophy, including the possibility of
';mirror of nmure minds" in observation and communication and rebted
forms of objectivity (Rorty, 1979). Calling "into question the very lan-
guage by which modern science knows irs language," thereby bringing its
invisible signs back into conscious orbit, Barthes (1986, p. 5) concludes,
can thus only be done by writing, itself a condition of language.
Philosophical justification in the criticism of interpretive anthropology
and its textual rum have waned since Dreyfus and RabinO\v's (1983) chal-
lenging evaluation of Michel Foucault's thinking and its relevance to cul-
tuml anthropology.L9 Any serious reengagement of interpretive anthro-
pology \\'ith modern science (and, to some degree, with archaeology and
physical anthropology-compare Hodder, 1986; Tilley, 1990) will have
to resurrect the historical points at which epistemological concerns began
to recede in allthropology and then build from there. Literary anthropol-
ogy, as parr of the emerging critical tradition in textual studies, rests on a
similat fate if it is ever to close the loop of its critical departure from mod-
ern science concerns with explanation per se and much that has been
rejectcd arbitrarily in the zeal to construct powerful arguments about rep*
resentation, historical situatedness, and author-ity in general. As Fabian
(1990) suggests, ';Dialogical and poetic conceptions of ethnographic
knowledge touch the heart of questions about othering. But they have a
chance to change the shape of ethnography only if they lead to literary
processes that arc hermeneutic-dialectical, or 'practical,' rather than rep-
resentational" (p. 766). On the other hand, many of the participants in
anthropology's literary turn have no interest whatsoever in dosing that
larger loop with any method, theory, or language that resembles the status
guo ante in their field (Brady, 1991b, pp. 12-1.3). Theirs is a more creative
turn to be considered criticaJiy before any larger inclusions are
attempted.l{) Following ideas that have "already been claimed by various
forms of radical anthropology," which see "the discipline as a handmaiden
to an era of European and American imperialism," Nathaniel Tarn (1991)
suggests that intellectual studies-perhaps properly conceived as \\7ittgen~
steinian language games-should keep their identities as such and not be
promoted to some Archimedian level of essential Truth or "used for
oppressive purposes or any pretence at superior knowledge" (p. 57).31
Anthropological poetry has surfaced in a similar humanistically grounded
context as a distinctly literary activity crossed over to social science.
552
,....." 'U JI Vp'V'U,::!".. VI I Ut::"LI'-.o
* Anthropological Poetry
If ethnopoetics is the most conspicllous anthropological activity in a
poetics domain, anthropological poetry is easily qualified in the converse
as the most conspicuously poetic activity in an anthropological domain.]2
Such work tends to focus on cross-cultural themes, esoteric cultural infor-
mation, the experiences of doing fieldwork in other cultures, and so on. Irs
distinctive-characteristic is the presentation of that information in marked
poetic form. Set in an anthropological mode, poetry is a conspicuously lin-
guistic message loaded with critical commentary on the nature of the world
and our place in it (BradYl 1991a, p. 216). Through poetic portrayal,
although perhaps fictional at the level of precise rime and sequence of
events, anthropological poets often attempt to convey the cross-cultural
circumstances and events of their fieldwork in an authentic and penetrat-
ing way (see Flores, 1982, 1999; Prattis, 1986). The aim in this poetry is
not to exist only for its own sake or self, or merely to entenain. bm to flag
its language without losing its historical or ethnographic referentialiry and
authenticity-thereby constituting a paradox of the first ordeL]] But it is
the conspicuously linguistic forms (such as the self-halting process of poetic
line phrasing) that give it poeticity, not its commentary-that is its anthro-
pological parr. This raises the issue of writer consciousness once again, the
degree to which authors should or must appear as such in their discourse,
and therefore crosses the boundary that traditionally has divided scientific
writing (and observation) from other forms.
Poetics leads to the aesthetic, which, as part of a general concern for the
making of meaning in what we do and study (see Brady, 1991b; Flores,
1985), can be engaged in radical forms, including but not limited to
poetry, without completely abandoning seemingly contrary interests such
as scientific observation. Although poetry is not science and does not
aspire to be science (Diamond, 1986c, p. 132), some of the work of at least
the kind of science produced by ethnographers is intruded upon by the
anthropological poet. By varying their forms of expression to include
poetry, anthropologists attempt to say things that might not be said as
effectively or at all any other way. This is consistent \vith the need to djs-
cover and examine critically all of the ways a subject (including social and
cultural relationships) can be represented. In that diversity the anthropo-
logical poet finds a measure of truth. But the Cartesian critic sees another
version: By reporting fieldwork experiences through poetry, the author
invokes a form of subjectivity to do the work of a form of objectivity, the
553
 ",.~." "~""'_", h"'~~''''~'',
conventional ethnography (see Tun, 1991, p. 246). In the process, infor-
marion may be conveyed more as what "might" or "could be" than "what
is" or "what was" as concrete historical fact, and that is problematic
(coulllerintuirive) to modern science and to ethnography thm emulates its
methods.
The pecking order of arguments in this instance has not always been the
same. Poerry has occupied more respectablc positions in some of the Truth
camps of the past. It has noc always been so marginalized as intellectual
activity.Jl Aristotle recognized 2,400 years ago that because poctry exacts
universal judgments from anion, from history, it call be considered (with
more than a little irony for the present argument) at one levelll10re scien-
lific (or philosophical) dmn history. Iris a medium in which "the lessons of
history do not become any more intelligible and they remain undemon-
strared and therefore merely probable, bur they become more compendi-
ous and therefore more useful" (Collingwood, 1956, p, 25). Nevertheless,
rejected as method by the received wisdom of science since the Enlighten-
ment, in pan for its declaration of "ringing true" rather than "being true"
in some particular empirical and historical accountability or "what actu-
ally happened," in part for working in a kind of "fifth dimension," inde-
pendent of time and therefore "of a sort that scientists cannot recognize"
(Graves, 1971, p. 35; see also Bruner, 1986 l p. 52), poetry is reintroduced
only with difficulty to any context of specific empirical accoumability,
such as concerns anthropology and the social sciences.]5
The challenge of poetry for anthropology thus has several aspects. In
addition to studying the uses to \vhich poetry can be pur, the contexts in
which it appears as discourse, and the variability of its forms in specific
cultures (its ethnopoetic dimensions), the challenge includes the develop-
ment of a genre of writing and reponing that systematically tries to incor-
porate satisfying and edifying poetic quality (foregrounding for clarity as
\velJ as aesthetic functions and the practical usc of metaphor as a tool of
discovery) without sacrificing the essence of ethnographic accountability.
In writing anthropological poetry, an author attempts to evoke a compara~
ble experience or set of experiences through the reader's experience \vith
rhe text on the t\vin assumptions that all humans are tied together through
certain substantive universals of being and that the beings we encounter
are sufficiently like ourselves to be open to empathic construction, discov-
ery, and reporting (Quine's principal of charity-see Shweder, 1996).
Perhaps one can also "extend to all discourse what has been said of
poetic discourse alone, because it manifests to the highest degree, when it
554
..... , ' " HUjJUIUY'LUI rut:tIL~
is successful, the effeer \vhich consists in awakening experiences which
vary from one individual to another.... The paradox of communication is
rhar it presupposes a common mediuJl1, but one which works-as is clearly
seen in the limiting case in which, as often in poetry, the aim is to transmit
emotioJ15-only by eliciting and reviving singular, and therefore socially
marked, experiences" (Bourdieu, 1991, p. 39). In this \vaYl ,vherher
through poetry, a great novel, or a play, the anthropological writer invites
us to "live through" other experiences vicariously and, "through the
power of metaphor, to come away with a deeper understanding of ... the
hum~1ll condition" (Coward & Royce, 1981, p. 131). Anthropological
humanists (some of whom are poets) look at this as the challenge of writ-
ing "from within" rarher than "witham" and that of "good writing" in
general, which, as mentioned earlierl is an important part of anthropol-
ogy's emerging poetics,}6
Clifford Geertz is well-lcnown for the CJuality of his writing l 3S are a few
other anthropologists. J7 But theirs is not typical writing in the discipline.
In bctl as noted previously, and contrary to the kind of disciplinary pro-
tection history claims generally for the public accessibility of its writings
against the obtuse jargonizing of the social sciences, not all anthropolo-
gists strive for more interesting ;md edifying ways of communicating
anthropological experience. Appearing "roo literary" (and certainly \vrit-
ing poetry begs the issue) is generally believed ro undermine scientific
authority.JU It can put the writer in the "wrong" intellectual camp, and that
can have severe career consequences for research funding, promotions,
and other political concerns. Some of the resistance is simply against pas-
sion in discourse (see Fabian, 1994, p. 100; Taussig, 1987). For holistic
anthropology (v.,rith its interest in preserving the four-field character of the
discipline: cultural anthropology, archaeology, physical anthropologYl
and linguistics), the real danger in exaggerating the focus on "author-
centrism" is slipping into a purely rhetorical or aesthetic legitimntion of
ethnography "by ontologizing representation, writing, and litetary form"
(Fabian, 1994, p. 91). That may elide or preempt altogether the question
of objectivity, which-as one of the few gates of conversation open to
reconciling conventional scientific concerns with advancing poetics and
the need for epistemological study in ethnography-nevertheless is a
neglected issue in postmodern anthropology (Fabian, 1994, p. 91; see also
Megill, 1994; Tiles, 1984). Tn rhe middle of ir all is an ongoing confusion
between powerful explanation and good vocabubry (Rony, 1981, p. IS8;
see also Abrahams, 1986; Brady, 1991b).
555
 '" , ... ", " ... 'r->, ,..... , " ... 'r-> .....,,..,,, ,...... ", ,..", , .... " ... , " ...... ,-" ,,..,, ' ..... ' ,
Lamenting the generally sad state of anthropological discourse today-
tbat it is often "painFul co read," enormously self-serving, corrupted with
esoteric and bland jargon, incestuous in its preoccupations when it might
l:><..-. bette,- "",rv"'J. "v;,J~ a 111<::',,", uunvarJ yi~ion of ir~ purpOSl:s and
nication strategies-Tarn (1991) suggests that "if the latest generation
writes well it may be due to mass alienation from the academic side of the
discipline (with all its dangers) and to such phenomena as 'ethnopoetics' "
(p. 56)-with its emphasis on comparative expression. A properly modern
anthropology would find a way to reintegrate its writing of science and its
humanisms. ''"It would pay the greatest attention to the way in which ...
ethnography was written, striving to go beyond belles lettres tovvard a lan-
guage with scientific and literary properties both, but governed primarily
by literature, so that its results could be available to all culturally liter-
ate readers" (Tarn, 1991, p. 57). Addressing in particular what authors
have attempted in what he calls "auro-anthropologies" ("personal ethnog-
raphy," "reflexive ethnography," and so on; see Crapanzano, 1980;
Dumont, 1978; Rabinow, 1977), Tarn (1991) says that such efforts run
parallel to his "argument that the genre so long looked for which would
assure a complete union of the poetic and anthropological enterprises
(without reducing either completely to the other, witham sacrificing the
option of lopsided emphasis, of an anti-union, or of for all practical pur-
poses complete independence in the conscious formulation of anthro~
po logical discourse] (should such be desirable) lies not in the keeping of
the anthropologist who cannot, for all his/her efforts, get beyond belles
!ettres, but with the poet who, in theory, still can. This is the question of a
language which, without turning away from scientific veracity, abdicates
not one jot of irs literary potential. Undoubtedly utopian, the search is at
home in poetry, incurably utopian, and probably nowhere else" (p. 256;
see also Richardson, 1994; Rose, 1991b).
As poets, anthropologists still have to get some purchase on their audi-
ences. \\Till it be poetry addressed to the existing elite, to anthropologists
as such, perhaps as a form of extending the ethnographic tradition, or
addressed to "the people at large, or that section of it which has not been
consumerized out of existence as a reading and listening public?" (Tarn,
1991, p. 64). Market matters, and some poetry published in the social sci-
ences is of dubious quality by literary standards and therefore less market-
able. J9 But what makes the distinction? \\That is defensibly good or agree-
ably bad poetry? There is a whole industry of writing about this in general,
of course, and there has been since the first critics forayed as experts into a
556
MflLnfapaloglCO! I'-oencs
world not of their own making. I will not attempt to sort that pile our here,
especially at the level of craft (although such discussion is relevant), except
to echo some carly-in-the-cemury thinking by l. A. Richards (1929): "Ir i,
COHlnlU- less imporr3m to like "good' poetry and dislike 'bad,' than to be able to use
them both as a means of ordering our minds"; Jncl that what matters is the
quality of the reading we give to poems, "not the correctness with which
we classify them" as good or bad, right or wrong (p. 327). The ease with
which a poem might slip into the mind to good effect is as much CI function
of the reader's susceptibility by topic and variable inclination to appropri~
ate such things as it is a function of form. There may be wide agreement on
the value of a particular poem or body of work, but there cue no absolutes
of poetic stimulation and effectiveness. 1V11.1cl1 depends on rhe situation at
hand. As with any text, the same poem can travel with very differem
effects across personal and cultural boundaries_
Nonetheless, after all is said and done, perhaps we can agree that lhe
most effective poetry stirs something up in you-an emotion or passion
that reaches beyond the shallow, that gravit8tcs to deeper experiences and
the sublime. The best of it is powerfully orienting, inspir8tional, if nor
more directly mantic or prophetic:lO In ethnographic or cross-cultural
poetry, the effect is likely to be a significanr realization of identity with
what otherwise could not possibly be claimed as Own. Through words anel
their specialized forms, a kinship is evoked that is grounded in empathy
and interpretation. It huddles with the universality of being human and
sends a large message: Perhaps we arc all Cheyenne) Arapaho, Tuvaluan,
and Ik. Perhaps we arc also all ethnographers now, as Rose (1991 b, 1993)
says. The best poets can make these things work as ideas and as platforms
for social action (compare Rothenberg) 1994).
The late anrhropologist-poct-social-activist Stanley Diamond (1987),
reflecting on the ugly, the beautiful, and the sublime in poetry md quar-
reling with Keats's famous line that '"'a thing of beamy is a joy for ever"),
wrote:
The mere deepening of gratification from the "joyful" experience of beauty
as rrurh does not achieve the sublime. The experience of the sublime is both
transcendental and quintessentially culrural at the same time. Language it-
self is the transcendence of the biological, it is the mediulll of culture, ;:wd
culrure is a rope-bridge thrown across a biological chasm. As rhe objecrive
realizations of the human essence, that is, as the existenrializing of our hu-
man possibilities, culture(s) is the arena for the construction of meanings: it
557
 '" , .... ", " ... ,r" ''-''', ... ,,...,... V,...,, ''-''', ,..." ,L.F r.Lr r.L.JL'' ,M' '\.J'~
represents a struggle then- is constant and renewed in each generation, and
evident in the livcs of individuals as they strive to become cultured human
beings.... There arc no certainties herc, only struggle and conrillBency, pain
'om,l ro:"U1.'o"nioll. G,,-nificat"lufl/ sarisf:;ction , Qr happifl{'::i.1 ilft: nor J[ issue.
13U\', we encounter joy. This is the joy t113t one finds in Lear, as he hurls his
words into the l'crrible gulf that engulfs him. The joy is in the words, in his
matured sensibility, in his challenge to nature and human defeat. The joy is
in the challenge, and in l'he formulation of his meanings. Or observe the
final shuffling off of guilt by Oedipus at Colonnus, as Sophocles etherealizes
him in <l be<lIIl of light. Or l'he conclusion of the \X!innebago medicine rite,
when the inil'iate finally achieves his emancipation from society, after be3r~
ing aU the abuse tl13t society may heap upon him. Or, for that matter, the or-
dinary riruals of maruring and variegated experience known in every
primitive sociel'y, whereby growth is attended by pain, where a new name
may be earned, and where the past is arduously incorporated into the pres-
ent, preparing the individual for the next ritual round as he moves higher in
the spiritual hier:lrchy of his society. That is where the joy is. And finally, it
is this joy, not Keats's beam}' or truth, which defines the sublime, beyond
the confines of the merely aesthetic, breaking all the formal rules of aesthet~
ics, beyond the range of the romantic imagination. For we arc not' talking
of im<lgination here, but of experience and its meanings, whether in the
culture of dreams, the culture of the hunt, or in the ceremonies of rebirth.
And finally, [am talking of the sacred space, the sacred silence tbat lies be-
yond language, bur remains grounded in language. (pp.170-171; compare
Burke, 1958)
All but the most serious and sensitive writers would be hard-pressed to
capture these existential dimensions in generally accessible language.
Ethnopoets strive for such experience and communication, even as they
eschew as unrealistic any expectations of ever reducing the sacred or the
sublime to strictly clinical or analytic forms. A battle is fit on this very
quest: If the anist sees it, she believes she has some prospect of saying it
poetically, of conveying with less prospect of empirical distortion the
nature of the experience as panhuman emotion; if tbe scientist sees it or
grips its "felt meaning," she can only hasten its transformation into some~
thing else (or some things else) by attempting to appropriate and express it
through clinical forms. Either way, there is a problem. Coleridge's "The
Rime of the Ancient lvlariner" refuses reduction to the subject of sailors
and seawater or to critical summaries of its formal properties and inspira-
ttons, just as the statistical expression of cross-cultural trait distributions
558
AnrnropOloglcOl t'oencs
in the ancient A/fiddle East cannot yield its exactness and rhetoricJI imeg-
rit}' to interpretive statements high in poeticity or uncommon metaphor:l'
Poetizing such things may reveal new dimensions of the problem, ro he
surc, and that can be a profound contribution to humanistic concerns, but
it is not necessarily a proper substitute for the original statement or form.
\Vhilc the upshot of such things can be shared widely in language, culture,
and experience, the confining exactitudes of competing poetic and scien-
tific linguistic forms restrict movements and thus hook again into th,lt
entangled truth of disciplinary and intellectual discourse: Not all subjects
travel equally wel1. 4l The movement is not hopeless. Translation is possi-
ble. But it is by definition a changing frame.
Beyond that we need to ask who does the poet speak (or? There is
debate over this in anthropology, not only for its poets but for all of its
writers, as there is in every discipline that has entered the crisis of repre-
sentation. Sticking to anrhropological poets for the moment, for they may
have stretched the forms of representation in the social sciences the most,
consider Diamond's 0982, p. 94) poem "Shaman's Song":
I talk to flowers
IVly fingertips withstand
The glance of roses
\\That do you know of the Bear
His body, my spirit
Rises everywhere
Seeing what the leaf sees
And the cloud
Ambiguous as a "'loman
Drifting through stones
I have lain with the otter
Under white water
On beds softer than birds
\\:rhat do you know of the Fox
Bearing the message of death?
Diamond was not a shaman. He was not a Native American, although the
spirit of indigenous cultures of the Americas resonates through this work
and through the multiple (Native American, Anglo, and anthropomorph-
ized animal) voices that structure his epic poem Going Wiest (1986a).~J In
these works, as in all others he produced, especially in the waning years of
559

his life, Diamond aimed for the sublimc by clrawing on the Everyman he
met time and again through cross-culmral expericnce and by raking on
rhar persona. Here was Arisrotle's universal-and Diamond's mine, Pcne-
rraring ;md can~picllaus experiences were laid out bdore him like so much
crackable glass and Diamond felt it under his feet with evcry cross-cultural
step. Treacllightly, he said, and his poetry \vas the gateway sign: "Things
can break here." In that he spoke for everyone, as brother, as kin, as alien-
ated Jew, as Everyman who has ever suffered the great smotherings of life
ancl iclentity that flow from cross-cllimral oppression. Translation and
multivoice reporting of human commonalities in purpose, interests, pasts,
and futures were not only possible, they were, to Diamond's eternal
credit, imperative (see Rose, 1983/1991a). Everyman was for him simul-
taneously author, subject, and audience, a social unit whose variable
parts were collapsed into reciprocal self-a\vareness through the rituals of
poetry.
Two related themes found throughout Diamond's work, including his
provocative "How to Die in America" (1986b), are that individuals and
cuIt-mes transcend that which conSUllles them, no matter how painful the
circumstances, or perhaps because of the pain, and that vast unintended
consequences come from colonial cannibalizing, v.,rith its miles and miles
of cross-cultural casualties on the beachheads of the world, in Biafr,],
India, New .Mexico, the Dakotas, New York, and California-not to ll1en~
tion the great erasures of tribal peoples in Russia, Spain, Iraq, Australia,
China, ;-mel the other Americas. Diamond's poetry seems to come from
everywhere because his quest for internalizing and expressing the sublime
had no firm cultural boundaries. Transcultural mysteries and truths were
in his view everywhere subject to appropriation through the carefully
opened eye and ~ar, once saturated with experience. Diamond believed
that no culture could hide its fundamental life from the prying eyes of
the mystic or the excavations of rhe traveling poet, especially when a
mature social scientist carried them barh on his shoulders. His discourse
was full of crossovers and cbanging centers. He knew that: anthropology
struggled for such knowledge and he believed that none of it would come
without intellectual passion in discovery and representation. By shuck-
ing preconceptions to the best of his ability and immersing himself in
the lives of others, he devoted the last years of his life to saying these things
through poetry, having been, as he said, "first a poet, then an anthro-
pologist, then a poet again" (personal communication, lvlarch 1983; see
also Diamond, 1982; Rose, 1991b). His writings make it obvious that he
560
Anrnropologlcol I-'oetlcs
never really changed professions, only from time to time his manner of
professing.
I~ the truth of Diamond's poetic work mixed up ''''ith beam}' Jnd yet
anchored empirically in valid ethnography (compare ~'IcAIlister, 1998)2 Is
it best judged as poetry, as skill in form, plusscd by the enjOYment or illu-
minating and edifying introspection it produces? Or is it best judged like
all fiction, not exclusively or necessarily in terms of its historical accuracy,
bur on a sliding scale of erotics and believability in the realm of being
human? Does it stir something valuable up in us, perhaps rare or orhenvise
unobtainable, even as it rings truc on life as \ve know it, or argue it, or
both? These arc bur a few problematic slices of Geenz's (1973) c!Jim that
all ethnography is in important vlays fictional-something constructed,
something fashioned, and never from whole cloth. The crisis of repreSCrl-
tation in anthropology followed most directly from that pioneering work.
The poets have never held center stage in the ensuing melee, bur we can
assert here that the fundamental issues of ethnographic representation do
not change for conspicuously poetic versus more clinical texts. They are
textual through and through, no matter what the purported form of repre-
sentation. 44
By the same token, as convention would have it in society generally, and
specifically in thc curious mix of subtle inferences, methodological man-
dates, and editorial imperialism in the \Vestern world concerning appro-
priate language for scientific reponing, the form of address often depends
heavily on the nature of the problem to be addressed. As it happens, l\!Iilcs
Richardson (1998a) says, "social scientists have long rurned ro poetry" as
a reportorial mode, but the content generally has diverged from the aca-
demic centers of their fieldwork and therefore from a collision course with
more conventional forms of ethnographic reporting: "Their poems ...
rarely addressed the rich ethnographic record they compiled nor the
anguish they felt about the free individual encountering coercive culture"
(p. 461). Some of the most notable poetry written by anthropologists in
previous generations-by Edward Sapir, Loren Eiseley, and Ruth Bene-
dict, who make up most of a short list-was similarly decentered as eth-
nography per se bur competitive enough to be published in nonanrhro-
pological media oriented toward such concerns. But no anthropology
journals catered to such interests, and the mainstream senriment was gen-
erally then what it is now; Poetry is an aside, an amusement, that belongs
elsewhere. Other than what might be called a "poetic mentality" in the
humanism of the times, and for some an affinity with the enticingly poetic
561
 11'11 cr;n..;.c lAIIV1'l, CYALUAIIVI'l, J.\!'lU Ktf-'Kt;:Jt:I'lIJ.\! JUN
work of Sir James Frazer, there was no ideology afoot that might sponsor
or enfranchise a more specifically poetic genre of anthropological writing,
including pOetfY.
Tbat dlIT1<ltc has changed perceptibly with the growth of more explic-
itly interpretive \vorks in the field, that is, with the advent of post-
modernism, the growth of "tcxtualism" because of its focus on text con-
struction and consumption, and the linguistic emphases of influential
structuralists such as Jakobson (1987) and Levi-Strauss (1962). lain
Prattis's Reflections: TIJe Anthropological Muse (1986), a collection of
poetry and artwork specifically addressed to fieldwork experiences by cu!-
tural anthropologists and linguists, established a precedent. So did reviews
of Stanley Diamond's and Paul Friedrich's poetry in anthropology's flng-
ship journal, the American Anthropologist, in the early 1980s. 45 At a time
\vhen much of this was at least a source of heated debate in the field, then
editor George rdarcus published Diamond's "How to Die in America"
(1986b) poem in the first volume of Cultural Anthropology. As the editor
of Anthropology Gnd I--Iumallism Quarterly, Miles Richardson initiated a
poetry and fiction competition that featured the winners in the journal.
The competition has been continued by the present editor, Edith Turner,
and by her poetry editor, Dell Hymes, and the topics continue to widen.
Friedrich's Bastard Moons (1979a), Diamond's Going West (1986a), anel
Dennis Tedlock's Turner Prize-winning Days From a Dremn Almanac
(1990) explore issues of language, ethnographic truth, and shamanic
dreams" (Richardson, 1998a, p. 461). Recent poetry in the Americil}]
Anthropologist, under the editorship of Barbara and Dennis Tedlock,
includes contrihutions by Friedrich (1995), Hymes (1995), anel
Richardson (1998b) on a variety of themes central to ethnographic ex-
perience. A new journal devoted to poetry, poetics, ethnography, and
cultural and ethnic studies, called appropriately Cross-Cllltural Poetics
[XcpJ and edited by Ivlark Nowak, published its first issue in 1997. Now
the journal Qllalitatiue Inquiry, under the editorship of Yvonna Lincoln
and Norman Denzin, is following suit (see Brady, 1998a; Richardson,
1998a). It will carry poetry and fiction as regular features, not because it is
trendy to do so, say the editors, but because "social scientists have for roo
long ignored these important forms of ethnographic representation and
interpretation" (N. K. Denzin, personal communication, November 8,
1998).
Looking in another direction, if the anthropologist-poets have been
few in number to date (albeit increasing in the postmodern afterglow),
562
,.."III II U/JUIUYIl.UI r-Ut.'lIL::.
amhropology as a whole has inspired numerous poets from other disci-
plines. "Poets precisely because of their 'otherness' in their home societies
arc attuned ro Otherness wherever it may be found, Hence they find an
affinity, dml to their own delight a socia! affinity, with the fruits of other
cultures, particularly those of the <despised and rejected' " (Turner, 1983,
p. 339). To which Tarn (1991) adels: "The inspiration that anthropology
has afforded poets-to go no further back than Pound or Eliot, or St:. John
Perse and Segalen, or Neruda, Vallejo, and Paz, in the age of Frazer,
Harrison, or the amhor of From R;tllal to Romance-to say nothing of
Ivlarx and Engles, Freud or Jung, IvIauss, Durkheim or Lcvi-Strauss-can
scarcely be said to have abated when we now have a virtual school of
'ethnopoetics' devoted both to the accessing of 'primitive and archaic
pac tries' inco our culture through the best available techniques of twenti-
eth century translation and to the mutual effecr upon each other of such
poe tries and our own-granted that Native poets very much continue to
produce poetry all over this worlel" (p. 63).
Anticipating mutual effects, it is reasonable to ask where one goes with
a penchant to write poetry in anthropology. Just adopting poetry as a form
of writing guarantees nothing. But poetry can be informative and useful,
even exhilarating when it reaches the sublime. Properly cast, it can be both
fire bellows for and bubbling lid on the pot of intellecruallife, a source of
transformation that is admirably unquiet in any position: It "turns every-
thing into life. It is that form of life that turns everything into language"
(Ivleschonnic, 1988, p. 90). It is in fact an an "far larger than any descrip-
tion of its powers" (Vendler, 1988, p. 6). For these reasons and others
already detailed, some competition between poetry and science as genres
may in fact be "healthy and entertaining" (FabiJn, 1990, p. 766), even
functional (Brady, 1993). It follows that having a hall full of pocts (as
nightmarish as that may seem to some scientists) can give academies of arts
and sciences a better overall pulse, if not a better future of discourse and
discovery. \X!riting poetry has also "helped individual anthropologists to
overcome alienation from experience" (Fabian, 1990, p. 766) and that, as
much as anything, has serious implications for engaging posnnodern con-
cerns about authority in ethnography. The penchant of the anthropologist
to wax poetic, if honed and pointed in these ways, can push into the heart
of ethnographic contemplation, epistemology, and theory. Hanging alit in
that zone for a while (as nightmarish as that D13Y seem to some poets)
might give \vay to progress in the conversation over Wh3t divides us so
readily in the academy today.
563

There are Jn3.ny other paths to that end, of courst. Poetry is not neces-
sarily the right tool for every job. As the old ad3ge goes; if one has only a
hammer ro work with, after a while everything stJrts to look like a nail. In
this InStanCe, "To Bcck the solution for a problem reg3rding the produc-
tion of knowledge in different or better representations of knowledge is to
reaffirm, nor to overcome, the represenrationist stance," \vith all of its
compulsions to order in conceptions of language and culture and related
problems (Fabian, 1990, p. 766; see also Friedrich, 1979b). Remaining
attentive to "the transformative, creative aspects of ethnographic knowl-
edge" (Fabian, 1990, p. 766) is not entailed automatically by writing
poetry (or reproducing dialogues) as an alternative to strict representa-
tiona! ethnography. "To preserve the dialogue with our interlocutors, to
assure the Other's presence against the distancing devices of anthropo-
logical discourse, is to continue conversing with the Other all all levels
of writing, not just to reproduce dialogues" (Fabian, 1990, p. 766) or to
slip unwittingly from being "mltural historians into itinerant bards,
clowns, or preachers" in a humanistic or poetic commitment to "being
with others" (Fabian, 1990, pp. 766-767; see also Tarn, 199 J, p. 75). We
can do better-and more.
One more thing to do in this \vriter's arena is to counter sole reliance on
the centrifugal or "distancing" devices of anthropological discourse-the
"Others are never Us" ideas, substituting third-person writing for first-
person experiences, writing in the ethnographic present, and so on.
Developing alternatives depends on paying special attention [Q the pro-
cesses through which ethnographic knowledge is produced in the first
place (Fabian, 1994)."l(, That does not require a lapse into the old static
models of structuralism, although much of what a poetics yields as the sub-
lime might be seen as the recognition of common deep structures (langne)
in a variable field of surface particulars (parole). Certainly there is room
for some edifying resusciration here. But the movement to more centripe-
(31 or "closing" discourse in anthropology should at minimum include an
ontological and ethnopoetic emphasis on the performative aspects of cul-
ture, those that can be "acted out," as opposed to focusing exclusively on
the more limited category of "what members of a culture know [that] is
'informative' in the sense that it can be elicited and produced as discursive
information" (Fabian, 1994, p. 97}.47
Such concerns necessarily draw edll10graphers and poets into the larger
philosophical fields of process, structure, and agency-including the
dYl181nics of knov"lcdge :md ritual, theater and history (sec Brady, 1999;
564
M' HII/ U/-,UIUYH_Uf r"Ut'Ul.""
Richardson, 1994)-and thereby increase the prospects of producing
larger-level understandings of the experiential fields we draw from 35
observers and comribute to :lS actorS ~nd \vriters. Bl1t it should also be
remembered that the original point of creative mystery in every study IS
also where the .iVluses caJ[ on the writer, and that is precisely "where
humanistic anthropology must both stall the process of knowing and open
It up imellectually-where we must catch ourselves in the act of rushing
headlong into conventional formats, of jumping to conclusions about
where the experience must take us and how to communicate it to others,
and begin to build a successful poetics into the fr3mework" (Brady, 1991b,
p. 20). By attempting to reconcile the spread in this connection between
analytic problems and the substance of fieldwork, anthropological poetics
GlIl do more than capture (or create) and convey the poe tries and litera-
tures of various cultures. It can require greater philosophical justification
for its ethnographic endeavors and ,It the same time help to crase some
unnecessary distortions of detachment from its objects of inquiry.
<i> Coda
Here, then, is the poetic turf more anthropologists than ever arc traversing
nowadays. It is neither defined nor covered exclusively by anthropologists
or traditional anthropological interests. As observers of common problems
(in and out of their OVi'n societies), anthropologists differ in research meth-
ods and strategies and sometimes produce incomparable results. Adding
other disciplines from the wings only seems to complicate matters. But
common ground for all is not beyond the pale. One source of that for
anthropologists and other social scientists is that they all seek \vays of
"speaking in the name or the real" about the people and behaviors they
srudy-of representing fairly and accurately what cannot easily be known
or demonstrated in the foggy partialities of cross-cllirural experience, the
imperfections of culturally situated observations, and the many voices
insinuated in mutually constructed truths (compare Brady, 1983, 1991c).
Another, as poet Tarn (1991) says, lies in the knowledge that "nothing can
bide from liS for long the fact that we all face the same problems in the end;
that the poetic fate, like the human, is universal" (p. 14).
Discovering and assessing tbe common denominators of human exis-
tence bas always had some priority in anrhropology,Pl and the need to
bridge [he inherently hermeneutic studies of el"hnography and larger-level
565
 " , ' ... , , , " ... ' ' " ' ' ' ' - ' ' ' , .... '"' ...V""' , ' ' - ' ' ' , ,-;, , u " ... , " ... .., ... , , ' ' ' ' ''-'" MfJUJropOloglcol t'oetlcs

statements about the whole organism (universals of language, culture, you travel depends on where you start. The destination is in the longest of
cognition, and behavior) is getting renewed attention-in full view of past long runs the same: the audiovisual roOI11, rhe bookshelf, the electronic
limitations, such as incommensurate theory construction,lallsuase biases] library, and, perhaps above all else, the mirror in the cemer of the house of
pc>tLL'coc,1 and praB,nc-..L;<; ,-",I"~iun.,,, bcP""C<O'l e~hno5n:1F'hcr5 i."lnd dH:: tJLVUP:;; "\Yhv-nn;~,Yc~-'" Inquiring Illll1ds \V3nr to know. \YJe have :1lways wanted to
they study, the impossibility of absolute objectivity, and the need for suc- know. Who has the key? Never mind. There seems to be more than one
cessful measures of validity in ethnographic assertions:19 But the poets door and some of them aren't locked. See you inside.
h8ve traditionally approached human commonalities differently, hanging
more on meaning than behavior-more on humanistic interpretations [Refrain: Is this jllst more anthropological borrowing (rom the l1ltlltiple-
than clinically valid8table truths. From the perspective of anthropological reality world ofphysics? No. An argument for 1111compromising relativism,
poetics, resurrecting a systematic scientific examination of "vhat is univer- 01' just for toleral1ce of diversity of mgl/mcllt and approach in the absence
sal in life and culture-"thinking big" again (see Simpson, 1994)-is more of an ouennchil1g paradigm for cultural anthropology? Neither. Do eth-
than a big task. It requires jumping the gap of irony and paradox from a nographic methods houe to be reduced to "deep hanging ant" (Geertz,
great (Geertzian) descent into dec-1i1s of local knowledge, "thick descrip- 1998)? No" Do we haue to abandoll the search (or commensurate theories
tions," and related critical intellectual developments over the past 20 and lfJ1iuersals of language, clllture, and behavior? No. Should we force
years, back into the whole package of logical positivist methods and rlOslalgia-driuen models ofthe !Jast on present ethnographic research for the
assumptions that have been criticized so extensively by the great descend- next millennium? No need to ask. Wlon't work.]
ers. Besides, as Aristotle discovered, the poets' generality is exposed in the
rich particulars of their work. The poets believe that the commonalities of
life can be plucked from there by large-minded observers of \vhatever per-
Notes
suasion. \X1hether or not these apparent sharings can be defended as unas-
sailable or universal truths by other standards is another matter. The paths
I. Hut compare Bachelard (1964) on the "poetics of Sp;JC(;," Bachdard (1971) on the
for collecting them arc different, so are the logbooks, and therefore the
"poetics of reverie," Brown (1977) on;l "poetic for sociology," and HaJlyn (1990) on the
form of the arguments down the line. Scientists normally create conglom- "poetic structure of the world."
erates from heavily laundered particulars, drawn from \vide fields of clini- 1. For ethnographers, the poetic focus on communications about othetcultures is as
cally cut cloth, to reach their generalities. much a concern about how the story of fieldwork is told as anything else. See Bruner (19il4),
So which to choose? No matter which extreme is harnessed-soft art, I'ranis (1986), Van Ivlaanen (1988), I1k,mganaro (1990a), Brady (1991a), and O'Neli
(1994).
hard science-let me reiterate what can be called by now (via my pressing
redundancies) the maxim that some things can't be said exactly or as effec- 3. See Beaujour (1987, p. 470) and Fischer (1988, p. S); compare Burke (1989,
p. 188), Krupat (1991, pp. 51-52), Robbins (1987), Tarn {1991, p. (;3), Eing (l994), and
tively any other way. Extremes 3re validated and perpetuated by that James, Hockey, and Dawsun (1997).
defensible fact, even as the proponents themselves exaggerate it in their 4. See Levi-Strauss (1962, 1967, 1969), Riffaterre (I 970}, Scholes (1974), Culler
bookend fights over turf and exclusivity, But what about the stretch in be- (1975,1977), Todarov (1977,1981), and Harland (1987).
tween? Rifts are made to be mended; the challenge in gulfs is to bridge 5. Sec Boon (1972), Todorov (1981), Eagleton (1983), Brady (1991<1), and Brooks
them. The boldest of anthropology's new interpreters and poets have (1994). The burgeoning domain of "cuhural studies" also begs the question of disciplinarl'
gone halhvay around the horns of this problem by trying to raise the illu- division and proprietary inrerests in these subjects (cflmpare lvlarclls, 1998). Although there
is much more in book form and articles elsewhere using this label,;l compatison of the r~mge
mination of our human commonalities to an art form without losing the
of interests covered in the literary journals Cultllml Critique and Poetics Today relative to
ability to inform more traditional social science concerns at the same time, those of the American Anthropologist, Cultum! AntIJro!Jology. and Allthropology and
in part on the premise that the collective work of scientists and pacts is in Hl/llulI/ism is instructive.
the broadest view complementary. In the pool of common knowledge, the 6. See Turner (1974,19112.1, 1982b), Bauman (1977, 1992), Schcchner (1985,1995),
pattern is ancient and very human: many in the one, one in the many. How Graham (1995), Dening (19%), and Brady (1999).

566 567
 7. Drawing on the oral behavior of the speech community, for eXilmple, ethnopoet
David t\min (19S3), par;]phrJ.~il1g,Dell Hymes, writes: "Allover the world in a great variet>'
of languages people ;l1HlO11nCe, greet, rake lea\'c, invoke, introduce, inquire, reqL1es~,
demand) Corl1l1hHld, ~lJil.\ll,;nrfti1[! t:nCOlJfilgc, beg, ,lnsurer, f1Jlllc\ repofr, de~cfjbe, narl'ate,
'''1<'''I''~C, ."'"'''''''' ;'b" "'-", "dY'~"l <..Itol""', Icfll~l:, ,lp\lID,l,;iIX! n:pruilcll, IDIi:t:, [,Wilt, inSUlt,
praise, discuss, gossip. Among this grab-bag of human Iangu;Jgc activities ate a number of
more or les~ well-defined llniverS<11 discourse genres, whose cxpectation strllcwres arc the
source of all poetic ilctivity. If there is any place that we should look for an ETHNO-
POETICS ir is here, among these universal genres, whcre alilingllisric invention begins.... 1
rake the 'poetics' part of ETHNOPOETICS to be ... the structure of those linguistic acts of
invention and discovery through which the mind explores the transformational power o[
language ,md discovers and inVentslhe world itsel[" (p. 45 I). Albeit less dynamic in orienta-
tion :11lc! less concerned with cultural process rhan product, there is <lIsa connective tisslle in
that perception for strucrural studies of the ways what h<ls been discovered and invented is
pm into place in particular societies, th;H is, the logical Jnd symbolic arrangements rhat
characterize the cognitive COntcnt <lnd social relations of whole cultures (including anthro-
pology itsdf~see Boon, I l.J82, 1904,1989; Hrady, 1993).
R. for more nn dialogics, see Tcdlock (J9B3, 1987b), Hill (1986), Fdd (I9H7),
f-Iolqllist (1990), \'\leiss (1990), Br<ldy (1991 h), Bruner and Corbin (1991), Duranti (1993),
;md Emerson (1997).
9. See H<lrland (1987), Clifford and Ivlarc\lS (1986), Clifford (1988), Fahi<ln (1994),
iVlarcus {1998}, and Gccrrz {1998).
10. Sec, for example, Tarn (1991), Wille (l9:Il), SCOtt (1992), and I1r:luy and Turner
(1994).
1 I. See also Rothenberg (19R 1, j 985), Swann (1983), Tedlock (1983), Tedlock Jnd
Ted lock (1985), Krnskrir)' (19B5), and Sherzer and \Voodbury (1987). On the anthropology
of folk narratives, see Jackson (1982), \Yjilbert and Simone<lu (1982), Swann (1983),
Sherzer and Woodbury (1987), Narayan (1989), Abu-Lughod (1993), B:lsSO (1995),
Graham (1995), CamIre and Echeverri (l996), and ReichclDolmaroff (l996).
12. Compare D. Tedlock (I 972, 1985, 1987a, 1991, 1993), Culler (1977, p. 8),
Rifbterte (1984), Sherzer (1987), and Graham (1995).
13. If 1101 literature per se~see, for example, Spradley and IvIcDollough (1973),
Langness and Frank (1971), Dennis and Aycock (1989), Handler (l9S3, 1985, 1990), Han-
dier and Segal (19[;7), Richardson (1990), <lnd Benson (1993).
1-1. See also lvbrclls and f'ischer (1986), I'-!:Indlcr (1983,1985,1990), Handler and
Segal (1987), fischer (1991), and ~vlarcllS (1998).
15. Of course, the whole enterprise is "interpret'lve." Furthermore, interpretive anlhro-
pology~as pan of the poststrucruralist, deconstructionist movement in contemporary phi-
losoph)-' and social scicnce~is nor necessarily (a) Jntiempiric<ll (How can any discipline
operate withollt an empirical ground?), (b) antiobjeetive (see Brady, 1991b; Rorty, 1979,
pp. 361-3il]; Spiegelberg, 1975, pp. 72-73), or (c) <lntisciencc (compare Barrett, 1996;
Holton, 1993; jennings, 19B3; Kna.uft, 1996; Lett, 1997; Milxwe!l, 1984; O'IvleJra, J 989;
Sangren, 1991; Sll'<lllkman, 1984). But responsible socia.l science of this kind docs reject (<IS
necessarily incomplete. attlong other problems) dogmatic empiric-ism that forecloses on the
swdl' of meaning in favor of an exclusive focus on beha.vior (see Brady, 1993, p. 277, 11. 28;
Fernandez, 1974; Pobnyi & Prosch, 1975; Rabinow & Sullivan, 1987).
568
16. Following especially G~ertz's (1973) classic observJtions on a Balinese cockfight.
17. See Marcus and fischer (1986, p. 72), lvbrclls (1980, 1998), IVlarcusand Cushman
{1982J, Boon (1972,19B2, 1989), Clifford and j'vlarctls (1986), Fernandez (1974,1985,
1986, 1988, 1989), Jnd much that has been published in the IDUfini Cult UTili A"thrn!lDloQl'.
rIle diSCUSSion of amllfOfmloml :1.Ild Im:h [lJ~cour!:e in T~ylor (19%), :lnd note 11, a.bove, on
folk narratives. See also Fischer (1988, 1989, 1991), Fabian (1990, p. 767: "If writing is parr
of a sysrcm of inreJ1ecrll:l1 <lnd polirical oppression of rhe Other, how can we avoid contrib
uting to that oppression if we go on writing?"), lvbnganaro (1990a), Tarn (1991), Karp and
Levine (1991), Said (199"1), Krupat (1992), IvtarcllS's (l993) review o[J(rupat, and Masci;l-
Lces, Sharpe, and Cohen (1993) on some important aspects of the politicizing that follows
from cthnographic encapsulations of other people's texts and customs.
18. See, [or example, the best-selling novels by Silko (1977) and IVlomaday (1989), and
SLiter's (1982) ethnogr;lphic exegesis of Brazilian literalltnl de corder.
19. See, (or example, Dennis and Aycock (1989), Rosaldo (1989), Manganaro (1990a),
B. Tedlock (1991), Hmdy (1991 b), Benson (1993 ),james et al. (l(97), and J\,farcus (1998).
20. See, for example, Feld (J 982), Jackson (1986), Turner ("1987), Dennis (l9S9), Fox
(1989), Stewart (1989), Richardson (1990), B. Tedlock (l992), Gorrlieb and Graham
(1993), Lim6n (1994), Behar (1993, 1996), Srewan {I 996), and Dening (l99ila, 1998b).
21. For related readings on tex[]]JI authority and ethnographic tepresentation, see
Geertz (1973,1983,1988, [995.1998), Lanser (1981), Bruner (1984), Fernandez (1985,
1986), Webster (lSlS6), Fabian (1990), Weinstcin (1990a, 1990b), Simms (1991), Tsing
(l994), Dening (1996, 1998a, 1995b), Rapport (1997), <lnd Banks and Banks (1998); com-
pare '"Eltlssig (1993) and J\:lonafi-Haller (1998). for J recent debate on who speaks [or whom
in historical ethnography, see Obeyesekere (1992), Sahlins (1996), and Borofsky (1997);
compare Brady (1985).
22. On that aCCOlmr, see the diverse works of La Farge (1929), Bohannan (writing <IS
Elenore Smith Bowcn, 1954), Turnbull (1 %2), Stewart {I 9(2), IvLmhiessen 0963, 1975),
Kurten (1980), Thomas (1987), Handy (1973), Thompson (1983), jackson (1986), and
Kn;tb (1995); compare the narrative form in Turner's (1987) empathic memoir.
23. See Turnbull (1 %2,1972), Rabinow (1977), Dumont (1978), Crapanzano (l980),
Turner (1987), Denni~ (1989), Dennis and Aycock (1989), and B. Tedlock (992).
24. Sec especially the innovative combination of Gottlieb Jnd Graham (1993), Fischer
(1983) on Ivlichel Leiris's eHort "to combine sympathetic ohservation of the Other, the
unavoidability of literary self-inscription, and the imperative of culmral critique" (p. 8);
compare Beaujoir (1987), Tarn (1991), ;ll1d Wolf (1992).
25. Tarn (1991) echoes a sentiment shared by many abour Levi-Strauss: He might have
been a creative writer with equal success, "had he chosen that path of expression" (p. 56).
MJnga.naro (1990b) adds, "The writerly sense of 7/istes TrolJiqlles, importantly, arises nor
only our of a literary style adopted by the anthropologist-author, but from rhe very perspec-
tive that the wtiter takes to his subject" (p. 16). It is also noteworthy that Levi-Strauss has
been roundly criticized for the [ormillisrn of his other works (see, e.g., Eagleton, 1983;
Geerrz, 1973, pp. 345359; Harland, 1987; Prattis, 1986; compare Boon, 1972, 1982;
Brady, 1993). The large-minded LeviStrauss seems to have tollched all the bases at one tittle
or another.
26. A theme plainly visible in Dening (1980, 1998a), Geertz (J998), and Botofsky
(1000).
569
 27. See Fox (1991), [Joggie, DeWalt, ;llld Dresder (! 992), Dorofslcy (I99'-1), [brrctt (1996),
Jessor, Colby, and Shweder (1996), Knauff (19'){i), Denzin (ll)97), Lett (l997), J;llnes er <11.
(]()97), Layton (1997), R3ppon (1997), ,md !vlarcus (J990); wmpare O'ivleara (1997).
2S, fllf imprmnnr 5InrCIllrl1LI Ofl rim prohl81ll, co, \1IAlf (Ilig}), ""b;,n (IligJ),
T\c.",i~j,y \1;H:'7 1 woo), TllOlllil~ (LliD', I'n I), and Dening (1 ~)95, I ~)Yh).
19. See R3binow (1983, 1984, J 986), Jarvie (1983), \X!llthnow, Humer, Bergesen, and
KllrzwciJ (1984), Ulin (1984), Rabinow and Sullivan (987), 1300n (1981), Swearingen
(1986), Loriggio (1990), Brady (199Ia, 1993), Duranti (1993), and Rappon (1997);
compare Scholte (1966) for some earlier thinking, and, morc recently, cominuing the
epistemological effon in anthropology with smart and pr,lCtical arguments, D.n'ies (1999).
30. See, for example, Reck's (1978) Ivk,,-icalll:thnography, Rose (1993) on the de,lth (If
"tvIalinowski-style" ethnography and the binh of'1 Illore poetic and nmltigenre manifesto .IS
its replacement, Boon (1902, pp. 9-12) 011 some of the confusions of rhis style ;md it sup-
posed diHerentidtion as "ethnography" from rhe "armch:lir" speculations of Sir James
George Frazer, Burke (1989, pp. 183ff.) on the need fot anthropologists ro "recognize the
factor of rhetoric in their own field" (lvbrctls, 1980). and Boon (I n 2) on "standards of
'convincingness' in various cross-cullllr,!! sryles and gemes, just as there are Clnons of veri-
similitude in redlist ethnogmphy" (p. 21).
31. Fur more all [he relativity and eligibility of such material ,1S languagt: games or simi-
lar focus frames, see Gueui (1984), de Zengoriw (1989), Del1zin (1997), and Gellner
(199S).
32. This is less tfnt.' on both COlllHS in linguistics-see, for example, Napoli and R;lnclo
(1979) and Bright (1983, 19(5).
33. Dilefl1ma is the other side of paradox. As Paul de l'vlan once observed: "It is in rhe
essence of language 10 be capable of origination, bUI of never achieving the absolute identity
with itself that exiSlS in the n.ltural object. Poetic language can do nothing but originate
anew over and over ag'lin; it is always constitutive, able to posit regardless of presenCe but,
by the same roken, unable to give a foundation to what it posits except as'lll intent of con-
sciousness" (quoted in Donoghue, 1989, p. 37).
3'-1. See Vendler (1985, 19SB) on the variable marginality of poetry. Not all cultures
share Americ<1's questionable v'1Juation of poetry. It has conspicuously more status as an
activity elsewhere (Br'lzil, Spain, Engbnd-~to name a few). Sec Bishop and Brasil (1971),
LordI.' 0984, p. 87) on poetry being something other than.1 luxury, Tarn (1991, p. 15) on
poetry's "survival V;JIlle" for humankind, Lavie (1990) and Dehar's (1993) sensitive and rich
texts on the tribulations of letting their poetry and poetic mel1l'alities out in oppressive con-
tex!s, and Richardson (1994) on poetry being a "/JIHClica! and fJo/uerfl/l method for analyz-
ing social worlds" (p. 511).
35. Tarn (1991, p. 254) sees this as a kind of competition of "vocations" thm can be
most discouraging in its petsonal, political, and philosophiClI entanglements. But "Ir should
be remembered rh<ll' these seemingly insurmountable problems do not prevent great artists
from ;lIsa being great scientists (and vice versa-e.g., Goethe; see Gould, 19(1), that "con-
tempt for one is no qualification for citizenship in the other" (grann, 199], p. 77.5), and thar
the opposition that pits imagin~ltinn against reason, when overdrawn on the b.utlegrounds
of art <1nd science, may he <1 kind "synecdochic blbcy" anyWaY-3 c;Jregory mistake in its
opposition (see Brady, 1991;1; Burke, 1989, p. tl7).
570
36. For more on 3nthropological humanism ~md writing, see \'\.lill.: (1991) and Brady
,mel "furner (1994); compare Harris (1997_ p. 2 l ) ] ) Jnd Lett (1997).
37. See, for example, Levi-Strauss (1961), Turnhull (1962,1972), Van Ll\vick-GoodalJ
(1li71), H,\tl'I~ (1977, In7l, Denlng (I 9ROLS.1hl;n, (1911 L 1911" 1991;U"bm II QQn,
Tllom:Js (l9X7), RO~:lldo (]ll~lJ), NJ.rJ.y'l11 (1(j~l), D. T",dlocL:: (l9(0), L,denlla!l (1991),
B. l~dlock (1992), Rose (1993), and Behar (1993,1996). Fabian (1990) \lotes wryly that
"Geenz prnb8bly deserves credit for initiating the new literary awar~ness in anthropology
not 50 much because he fraternized with literary critics bur because he dared ro write well
and got away with it" (p. 761). Fernandez (personal cOlllmunication, February 27,1998),
playing nn SOllle of his own influential work (e.g., 197A,19f.l6), says that Geenz is above all
else a "master metaphorist.'
38. However, separating these genres and doing bOlh can generate a laudable efkcr.
Scientists who also write novels, pOetry, or \'ery "Iileruy" memoirs only ,eem to enhance
their reputations (see, e.g., Levi, 1984; Lighul1an, 1993; S.lgan, 1980, 1(85). On a related
mJtter, when asked if his exercises in poetry improved his science, Nobel Prize-winning
chemist and estimable poet Roald Hoffmann said no, at least not directly. Writing poetry
for Hoffmann makes him feel better about himself as a person, as a human being, and thai
helps his science considerably (personal C0I11Il111llicatinn, February 26, 1997; see also
Chandrasekhar, 1987; I-Ioffm~l1l11, 19117 ,1990a, 1990h, 1995). Evell with separation in
practice or redefinitioll in humanistic terlllS, the functional linkage of ;lI"t-inscience cannot
be denied. It m.1Y be Jess direct, less obvious, less e"plored, and less reponed than it could
be. 15m it is always present.
39. There are important exceptions. Sec, for example, Friedrich (1979;1), Diamond
0982,1986'1, I 986b), some of l'ranis (19S6), various contrilmtions over the years to
Anthropology ill/d Hl/miIl/iS/Il (Quarterly), Fox {1989}, Stewart (1989), Tedlock (1990),
Richardson {1998a, 1995b}, and Flores (1999); compare Hal! (l98R) on phases in becom-
ing a poet.
'-10. Leavitt (1997) is pioneering, work on poetry and prophecy, words and power,
mantic prose ~lnd ecstatic experiences-"stirrings" thaI rarlge (rom the aesthetic and emo-
tional to fundamenrally physical responses and their diverse cultural expressions in ritual,
politics, healing, ;llld messianic movements, to 1l.1Ine;1 few. See also Ahu-Lughod (1986),
Trawick (19B8,1997), Dobin (1()90j, ;jnd CsorrLls (1997).
41. See Fernandez (1974) for all impon'lIlt argument (wirh discussion) on sensitivity to
and the mastery of metaphors (tropes) in ethnographic narration .llld related highly orga-
nized expressive activities, L.akoff and Johnson (1980) and Lakoff and Turner (989) for J
clear enllnciation of common versus llllcommon metaphors in everydd)' life, Van Den
Abbeele (1992) on tr;jvel as a common metaphor and much that is smuggled into its use, and
Fernandez (1986) and Fernandez and Herzfeld (1998) on the place of sociJl and cultural
poetics in the study of meaning in performance.
42. Playing all! these constraints is in pan how logical positivism gor to be known less as
the backbone of universally ilpplicable methods and more as a list of things that could not (or
were not allowed to) be studied in depth, or at all (e.g., meaning versus behavior, aesthetics
and emotions, and sO on).
43. It is nOt rhe same thing as letting mulriple authentic voices surface in a single text or
performance (see Trawick, 1997), of course, bur Diamond believed in~<lnd displayed (hy
assuming the roles of orner spenkers) in sOllle of his poetry-the prospect thar some people
571
could speak for othets to advantage for all on som~ Occl3ions; thilt certain knowledge
bound<lries could be crossed with insight and pmver, withom imperialism or blind culrural
exploitation, withom a franchise of superiority or oppressive ethnocemrism. Early in his
GOh'S W,st {1986,,), foc examplc, the anthropomorphized Offef speaks for rhe Mohicans
''''L' _".\oL""l u'"'' I"'; cep"C"e'H'''iYc~ v[ ",,11 u.ll\:r N'l1:i YC l\IIlt:IH':illl:l) tJirUUj;11 c.'jl){t:riC
IU10wl-
edge of their irnpendingcollision with history: "Because we knew the future! And we under-
stood their [Ivlohicln, Algonquin] I~gends! Bener th<ln they did" (p. 11).1\ l<Jrger poim of
this boundary crossing thar is not Ollways obvious, especially where ethnic pride is ~H stake, is
d10lt indigenous identiry doesn't necessarily translate to experr on ,Ill that contextll;llizes it.
j\'luch depends on tilt: form and cultural origins of the quesrions asked and the culmral and
intellectual range of ,1llswers sought (see Brady, 1985).
44. There are no sepOlrate languages for science and poetry, only speciJli7.cO vOGlnubr-
ies and variable cotlte"t5~culturJI "pte-texrs" and "subtexts" rhar reassure, inform, Jnd
might even misinform rhrough m~l1lipubrions designed to present the truth we /1.'''/11 to find,
as opposed to something that might be calculated rhrough other measures as more <1ccur~!te,
anJlytical1y satisfying, or less ptejudiciOlI. Ivloreover, there is plenty of slipP~lge in rhe divi-
sion of labor between scientific and poetic teXIS. Asserting bi<1s conlTol, for example, includ-
-mg unmasking ethnocenrrism, is seen as obligatory in scientific observation. rXlsc!osing
obsetvation<11 bias directly by emulating ("flagging") it in performJJ1ce ilnd leXt is ,1 poetic
oprion that may accomplish much the same rhing, albeir through a r:Idic;l1ly differcnl chan-
nel and probably t'or differellt ends. Here, as elsewhere, comext is pfJeticJ!!y everything t'or
detetmining meaning, and on occasion it can itself be all but slave to authorial illlcmiom
and sociolinguistic form.
45, As book review editor of the American AI/lbmpa/agist ar the rime, I commissioned
rhese teviews (see Rose, 1983IJ991<1; ~yler, 1984). One earlier and interesting example 01
poerry in the AA was published by then editor Sol Tax. Referred to a5 "Puzzled Ph.D. C1l1-
didare," and published anonymollsly (Anonymous, 1954), the poem was sent to Tax by
lvlclville Herskovirs, who rook it' off the bull~tin bOJrd <1r the Dep;1frlllcnt at' Anthropology,
Northwestern Universiry. It was actually "\'\1<15n't It a Thought Titanic" by then graduate SlH-
dent (and now ;mrhropologist and poetics scholar at the University of Chicago) J;J.lT1es \\~
fernandez.
46, For special takes on the producriou of ethnographic knowledge, see Borofsky
(1987, 2000), Geenz {l973, 1983, 1988, 1995, 1998}, Dening (19XO, 1996, 19')8a,
1998b), Brady (1991b), and Behar (1993, 1996).
47. See also Tedlock (1983, 1(90), Schechner (InS, 1995), Fern;mdez (19S8),
Graham (1995), Dening (19%) :lnd l:1.ylor (1996), Poetry, of course, fits into the per-
formative traditions of its makers di rectly, <lnd it C<1n be an im porram somce for understand-
ing indigenous histories. As Charlot (1985) says of I-:Ia\v<lii:ln po~rry, "The fact rhat poeuy
has been used frequently for imporrant occasions and purposes suggests that it' has a utility
thus far overlooked by historians" (p. 29), See also Vendler 0995, p. 6) on Ihe historical
meaning of rhythms, stanza forms, personae, and genre; and Fernandez and Herzfeld
(1998) for a strong argument that "poetic principles guide all effective and nffcctilic sociJI
interaction" (p. 94) and <1 discussion of how that articulates with the problem of
performativity in human action generally.
48. See Brown (1991) and Uvi-Strauss's whole corpus of writings; see also Whitten
(1988) and Kilbride (1993),
572
49. Sec, for eXJtnple, .lessor et al. (1996), Shore (1996), Bloch (199R), Jild Brady
(l998b); compare \Vicrzbicka (]990).
50. Some poets live there, along with a group of narcissists, clever historians, and a
tinhorn totalitarian or two. They are visitecl regularly by social and natural scientists. The
projecr IS ro see what !llalu:~s
tile re~idents tick. The prohlem i~ rh . n everyone in the room is
ticking. Everyom: is looking back ar evetyone who's looking at. Nobody knows for sure
who's studying whom, Of, for that maner, whose conversation should dominate. Somebody
said there was order in the ch<1os. Two poets and the ghost of Aristotle said, "We raId yOll
so." Two scientists wanted ro bOltle it hut couldn'r gel a grip on ~ither irs substance or its
me:ming. New schools of methods grew up in the exact Spot of rhis discovery. A plaque
marks it to this very day: HIE FOUNTAIN OF IvlUSES STARTED HERE. LANGUAGE Or-TRUTH SOLD
NEAR HEI1-!:. PLEi\SE DO NOT LOITER.
-J) References
Abrahams, R, D. (1986). Ordinary :lfld extraordin;uy experience. In V Turner &
E,lV1. Bruner (Eels.), The anthropology of experiellcc (pp, 45-72). Urbana:
University of Illinois Press,
Abu-Lughod, L (1986). Veiled sentiments: HOllor a!/d poel1j' iii a Bedouin society,
Berkeley: University of California Press.
Abu-Lughod, L. (1993). Writing women's luorlds: Bedouin stories. Berkeley: Uni-
versity of Cnlifornia Press,
Anonymous (f. \YJ. Fernandez). (1954). Puzzled Ph.D. candidate, In S. Tax, This
issue and others. American Anthropologist, 56, 742.
Antin, D. (1983). Talking to discover. In ,1, Rothenberg & D. Rothenberg (Eds,),
Symposium of the whole: A rallge of discoursc toward all etlmopoetics
(pp. 450A61), Berkeley: University of California Press,
Bachelard, G. (1964). Thc poetics of space, Bosron: Beacon.
Bachelard, G. ('1971). On poetic imaginatiol! and reuerie (C G.:mdin, Trnl1s.). India-
napolis: Bobbs-MerrilL
Banks, A., [I,[ Banks, S. P. (Eds,). (1998). Fiction and social research: By icc or (ire.
\Valnut Creek, CA: AltalVlira,
Barrett, S. R, (1996). AnthroIJOlog)': 11 student's guide to theory and method,
Toronto: University of Toronto Press,
Barthcs, R, (19G8). W+itilrg degree zero (A. Lavers & C. Smith, Ttans.). New York:
Noonday.
Barthes, R. (1972), Mythologies (A, Lavers, Trans.), New York: Hill & Wang,
Barthes, R, (1986). From science to lirerature, In R, Barthes, 1he rustle oflanglloge
(R. Hownrd, Trans,; pp. 3-10). Berkeley: University of California Press.
Basso, K, (1988). A review of Native American discourse: Poetics alld rhetoric
U. Sherzer & A. C. Woodbury, Eds,). American Ethl!ologist, 15, 805 -810.
573
 Basso, E. (1995). The last Cillllfiuals: A Soulh American oml histon'. Austin: Univer-
sity of Tex:ls Press. .
B,luman, R. (1977). Verbal (lrt (IS per(orl1ulIlce. Prospect Heights, 1'-: \Xfavebnd.
IJaUITUn, R. (Ed.). (1991). Fnlbhll'~\ wlllll'dl Ilcr!omli711ccs. aI/(./ pOfJII1,n elltertaill_
ments. New York: Oxford University Press.
Beaujoir, IvL (1987). iVlichel Leiris: Ethnogr.::tphy or self-portrayal? Review essay of
Sulfur 15, featuring new trzmslations of lVlichei Leiris's work. Cllitural
Anthropology, 2, 470-480.
Behar, R. (1993). "Hallslaled tuOJlli7JI: Crossing the horder /llith Es/)eril//za's story.
Boston: Be.::tcon.
Behar, R. (1996). The I/lilncmble obsert'cr: AJtthro/)()logy thatlm'i1!;:s YOllr heart.
Boston: Be'Kon.
Benson, r~ (Ed.). (1993). Anthropology <11/[1 litemtllre. Urhana: Univcrsitv of illinois
Press. .
Bishop, E., & Brasil, E. (Eds.). (j 972). All anthology of tlucnticth-celltury Bra::JIi<1l1
jJOetfy. iVliddlelOwn, CT: \\iesleyan University Press.
Hloch, i'vI. E. F. (1998). How lUe think they thi!l/~: Allthropologicdl approaches to
cognition, memory, alld literac)'. Boulder, CO: \'{festview.
Bonn,.f. 1\. (1972). From s)'lI1bolism to structuralism: Leui~Str{/l!ss in a literar" tm-
ditioll. New York: Harper & Row. -
Boon, J. A. (1982). Other tribes, other scribes: Symbolic '1I1thropo{ogy ill the com-
jJamtiue stlldy of C1fItIll'CS, histories, religious, and texts. Clillbridge: Cam-
bridge University Press.
Boon,J. A. (1984). Folly, Bali, and anthropology, or sCltire ClCroSS cultures. In E. Ivi.
Bruner (Ed.), Text, play, and story: The construction and reconstruction o(
sel( and society (pp. 156-177). \,\lashingwn, DC: American Ethnological
Sociery. >.
Boon, J. A. (1989). A(finities and extremes.: Criss-crossing the bittersweet ethnology
of East Indies history, Hindu-Balinese cuItllre, alld ll1do-European allure.
ChiGlgo: University of Chicago Press.
Borofsky, R. (1987). Making history: Pukapllkan and anthrolJO/ogical constructiolls
of kuoluledge. New York: CClmbridge University Press.
Borofsky, R. (1994). Assessing cultural anthropology> New York: IV1cGraw-Hill.
Borofsky, R. (1997). Cook, Lono, Obeyesekere, and Sahlins [with commentary].
Current Allthropology, 38, 255-282.
Borofsky, R. (Ed.). (2000). Rell1emlm1Jlce of Pacific pdstS: All illl/itation to remake
history. Honolulu: Univetsiry of Hawaii Press.
Bourdieu, P. (1991). Lallgllage al1d symbolic pOluer. Cambridge, MA: I-In.rvard Uni-
versity Press.
Bmven, E. S. (pseudonym of L. Bohannan). (1954). Retunt to 1t.7llghler: All 11l1t/JrO-
polngical mwel. New York: Harper & Row.
574
Brady, 1. (Ed.). (1983). Speaking in the name of the real: Freeman imd IVleacl on
Samoa [Special section]. American Anthropologist, 85, 908-947.
Brady, L (1985). Review of Tlwalu: A history, by S. FClniu, V. Idemia, T. Isako, et a!.
Tn P.7cific History RilJliogn11,lJ)' an,} Com"","l. 10r:.r; (rr- _r;25,1>. C"n!n"Tc"
Journal of Pacific Hisrory.
Brady, L (1990'1). Comment on "Is anthropology art or science?" by IVl. Carrithers.
CUTTellt Allthropology, 31, 273-274.
Brady, 1. (1990b). Review of The uiolCllt imagination, by R. Fox. Americall Ant/no-
pologist, 92, 1078-1079.
Brady, L (Ed.). (1991<1). AntlJrolJo/ogical poetics. Sav3ge, !vfD: Rowman &
Lirtlefield.
Brady,1. (1991b). Harmony and argument: Bringing forth the artful science. In
L Brady (Ed.), Anthropological poetics (pp. 3-30). S:lVage, IvID: Rowman &
Littlefield.
Brady, I. (1991c). The Samoa reader: Last word or lost horizon? CUITent Ant/Jro-
polog)', 32, 497-500.
Brady, 1. (1993). Tribal fire and scribal ice. In E Benson (Ed.), /\ntlnopo/ogy {/nd lit-
eratllre (pp. 248-278). UrbanJ: University of Illinois Press.
Brady, l. (1996). Poetics. In D. Levinson & M. Ember (Eels.), The encyc/o/Jedia of"
wltllml aJlthrolJOlog)' (Vol. 3, pp. 951-959). New York: Holt.
Brady, L (1998a). A gift of the journey. Qlwlitatiue Inqlliry, 4, 463.
Brady, r. (l998b). Two thousand and what? Anthropological moments and meth-
ods for the next century. Americall Anthropologist, 100, 510-516.
Brady, L (1999). Ritual as cognitive process, performance as history. Currc//t
Allt/Jropolog'y, 40, 243-248.
Brady, L, & Turner, E. (1994, June). Imrodu([ion. In 1. Br;Jdy & E. Turner (Eels.),
Humanism in anthropology, 1994 [Special issue]. Anthropology and HUlllall-
iSIl1, 19,3-11.
Brann, E. T. H. (1991). T1JC world of the imaginatioll: SHill and substance. Savage,
IvlD: Rowman & Littlefield.
Bright, \XZ (Ed.). (1983). Discovered tongues: Poems Iry linguists. San Francisco:
Carvine.
Bright, \"Xl: (1985). \Y/ord formations: Poems by lingllists. San Francisco: Corvine.
Brooks, P. (1994). Aesthetics and ideology: \"\1hat happened to poerics? Critical
IlIquiry, 20, 509-523.
Brown, D. E. (1991). HIlIlh1lll1niuersals. New York: McGr;Jw-J-lill.
Brown, R. H. (1977). A poetic (or sociology: Toward a logic of discovery for the
hUll1all sciences. Cambridge: Cambridge University Press.
Bruner, E. M. (Eel.). (1984). Text, play, and story: The constmction and reconstruc-
tioll of self alld society. \\lashingron, DC: American Ethnological Sociery.
575
Bruner, E. IV1., & Gorfain, P. (1991). Dialogic n,lfration and the paradoxes of
1"bsad3. Tn r. Brady (Ed.), AnthrolJOlogicill poetics (pp. 177-203). Savage,
i\'1D: Rmvl1lan & Littlefield.
BruHer,]. (1986). Aclualmiuds, {Jossible lUorids. Cambridge, JvlA: Harvard Univer-
sity Press.
Bruner, J. (1990). Acts olmeaning. Cambridge, MA: Harvard University Press.
Burke, E. (1958). A philosophical cnqtlil)' into tbc origin of ideas of thc sublime and
bealltiful. Norre Dame, IN: University of Notre Dame Press.
Burke, K. (1989).011 symbols and society U. R. Gusfield, Eel.). Chicago: Universiry
of Chicago Press.
ClIlelre, H., & Echeverri,.J. A. (1996). Cool tnblCco, sweet coca: Teachillgs o( all
Indian sage (rom the Cololllbian A1IIa:::.on. Devon: Themis.
Chanelr3sekhar, S. (1987). IilIth and heauty: Aesthetics ,md nlOtiuatiollS III science.
Chicago: University of Chicago Press.
Charlot, j. (1985). The Hawaii,lII poetry o(reh/;ioJl alld politics. Honolulu: Insti-
nHe for Polynesian Studies.
Clifford, J. (1988). The In"e/licamel1t o( wltlfre: Truent iefh-cmtlfry etlJllogmphy, lit-
emture, and art. Cambridge, ./'vIA: Harvard University Press.
Clifford, J., & Ivbrcus, G. E. (Eels.). (19S 6). Writingwlturc: The poel ics alld polit ics
of ethnography. Berkeley: University of California Press.
Collingwood, R. G. (1956). The idcd o( history. New Y()tk: Oxford University
Press.
Coward, H. G., & Royce,.f. R. (1981). "[f)ward an episremologicll basis for human-
istic psychology. In]. R. Royce & L. P Ivlos (Eds.), [-Jllllldllistic psychology:
Concepts al1d criticisllls (pp. 109-134). New York: Plenuill.
Crapanzano, V. (1980). Tilhami: Portrait o( a Moroccall. Chicago: Universiry of
Chicago Press.
Csnrdas, T. ]. (1997). Prophecy and rhe perfornunce of metaphor. American
Amhropologist, 99, 321-33l.
Culler, J. (1975). Structuralist poetics: Stmcturalislll, linguistics, alld the study of
litemtnre. Ithaca, NI': Cornell University Press.
Culler,]. (1977). Foreword. In T. Tarodov, The poetics of prose (pp. 7-13). TthaGl,
NY: Cornell University Press.
Davies, C. A. (1999). Reflexive ethnography: A guide to researching selues mrd oth-
ers. New York: Routledge.
Dening, G. (1980). Islands and beaches: Discolfrse on a silent land: Marc]lIeSi1S
1774-1880. Honolulu: University of Hawaii Press.
Dening, G. (1995). The death o(W/i//iam Gooch:.A histol}"s anthropology. Hono-
lulu: University of Hawaii Press.
Dcning, G. (1996). Performances. Chicago: Universiry of Chicago Press.
Dening, G. (1998a). Readings/writings. Victoria: Ivlelbourne University Press.
576
Dening, G. (1998b). \"X/riring, rewriring the beclCh: An essay. Rethinking Hi~tory,
1(1},143-172.
Dennis, P. A. (1989). Oliver LaFarge, wrirer and <lnrhropologist. In P. A. Dennis LX.
WZ Aycock (Eels.), Literature <1111 t1!rthropo!o.'?}' (pr. 209-219). Luhhnd:: TeX:1S
Tech University Press.
Dennis, P. .1\., & Aycock, \'V. (Eds.). (1989). Litemtllre and allthrotJulogy. Lubbock:
Texas Tech Universiry Press.
Dcnzin, N. K. (1997). Interpretive etlJllography: EtlmogratJhic pmctices (or the 21 st
eentllry. Thousand Oaks, CA: S"ge.
de Zengmita, T. (1989). On Wingenstein '5 remarles on Frazer's Golden bough. ClIl-
tllml AnlhrotJOlogy, 4, 390-398.
Dialllond, S. (1982). Tolems. Barryrown, NY: Open Book/Srarion Hill.
Diamond, S. (1986a). Going Il-'est. Norrhhampron, I\/IA: Hennes I-louse.
Diamond, S. (1986b). How to die in America. ClIltural Alltlnopolo!:.,ry, 1,447-448.
Diamond, S. (l986c). Preface. Dialectical Anthropology, 11, HI-132.
Di"mond, s. (1987). The beauriful and the ugly are one rhing;, rhe sublime ;lJlorhc.r:
A reflection on culture. Cultllral Anthropology, 2, 268-271.
Dobin, H. (1990). kIer!in's disciples: Prophecy, poetry, and pOlUcr ill Relwiss,mcc
England. Stanford, C}\: Sranford Universin' Press.
Donoghue, D. (l989,.June 29). The strange case of P~llli de Man. New Y()r1~ Rel'iclIl
ofBoohs, pp. 32, 37.
Dove, R. (1994, January-February). Whar does poetry do for us? Vilgillia (Univer-
sity of Virginia <llumni newslener), pr. 22-27.
Dreyfus, H. L., & Rabinow, P. (1983). tdiche! FOT/calllt: Beyond structllralislll (lnd
herllleTlClltics (2nd cd.). Chic.:lgo: Universiry of Chicago Press.
Dumont, J.-E (1978). The headmal! lInd I: Ambiguity and ambimlc/lCi! ill the
fie/dworkiJlg experiel1ce. Austin: Universiry of Texas Press.
Duranti, A. (1993). Truth and inrentionaliry: An ethnographic cririque. Cultllral
AllthmfJOlog}', S, 214-245.
Eagleton, T. (1983). Literary theory: All introductiON. ./'vIinneapolls: Universiry of
I\11nnesota Press.
Eco, U. (1979). The role of the reader: Explorations in tl}(! semiotics of texts.
Bloomington: Indiana University Press.
Ellis, c., & Bochner, A. E (Eds.). (1996). COlllposing ethnography: illtcrlWlilJe
(orl1ls o( qualitatiue writing. \V'alnur Creek, CA: AltaMira.
Emerson, C. (1997). The (irst hwrdl"ed years of Mikhail Bal:.hti!l. Princeton, Nj:
Princeton University Press.
Fabian, J. (1983). Time and the other: How anthropology I/[al~es its object. New
"York: Columbia University Press.
Fabian, .J. (1990). Presence and representarion: The orher and 3nthropological
writing. Critical Inquiry, 16,753-772.
577

Fahian, J. (1994). Ethnographic objectivity revisited: Ftom rigor to vigor. Tn
A. IVlegili (Ed.), Rethinking objectil/ity (pr. fi1-lOS). Durham, NC: Duke
University Press.
felrl, 5. (1')82). SOl/ill! alld ~elltillzelll, Birds. /fIeenl,,". nodi,,. a",~ W"." i" JZ"I"j;
8xpregioll. PhiL:tdelphia: University uf Pennsylvania Pr~ss.
Feld, S. (1987). Dialogic editing: Interpreting how Kaluli tead sound nnd senti-
memo Cllltllmi Anthropology, 2, J 90-21 O.
Fernandez,]. \X( (]974). The mission of metaphor in expressive culrure [with com-
memaryJ. Clfrreltf Anthropology, 15, 119-145.
Fernandez, J. \'\( (] 985). Exploded worlds: Texts as il metaphor for ethnography
(and vice-versa). In S. Diamond (Ed.), Amhropology a(ter '81: St:ne o( the
art, state of society, parr II [Special issueJ_ Dialectical Anthropology, 10(1-2).
15-26.
Fernandez,). \Y/. (1986). Persuasions and IJerlorll1a1rccs: The pIa)' of tropes ;'1 cll/-
tllre. Bloomington: Indiana University Press.
Fernandez,]. \X~ (1988). Andalusia on our minel:;: Two conrrasting places in Sp;lill
as seen in a vernacular poetic duel o( the bte 19th cenfllry. Cultllml Antlnu-
Ilo/oR)" 3, 21-35.
Fernandez, J. \"Y/. (1989). Comment on Keesing's exotic readings of cllltur;:1! texts.
Current Anthropology, 30, 470-471.
Fernandez,]. \\Z, & Herzfeld, Iv1. (1998). In search of me,ll1ingful methods. In
H. R. Bernard (Ed.), A halldbook of1/lcthod in cllituml i71ltlnopology (pp. 89-
129). Thollsand Oaks, CA: Sage.
Finnegan, R. (1992a). Oral poetry. In R. Bauman (Ed.), Folklore, ClIltllr{J! perfor-
mances, and IJOplllarelltertainments (pp. 119-117). New York: Oxford Univer-
sity Press.
Finnegan, R_ (l992b). Oml poetry: Its nature, significance, alld socia! cOl1text.
Bloomington: Indiana University Press.
Fischer, Iv1. .IvI. J. (1988). Scientific dialogue and critical hermenemics. Cultural
Anthropology, 3, 3-15.
Fischer, ?v!. tv!.]. (1989).l\'1useums and festivals: Poetics and polirics of Represenra-
tions Conference, the Smithsonian Tnstitmion, September 26-28_ -1988,
1. Karp and S. Levine, organizers. Cllltllml Allthropology, 4, 104-221.
Fischer,l'v1. [vLl (1991). Anthropology as cultural critique: Inserts for the J990s:
Cultural studies of science, visual vinual-realities, and post-rr,lllma politics.
Cllituml Anthropology, 6, S15-537.
Flores, T. (1981). Field poetry. Anthropology and Humanism Quarterly, 7(1), ] 6-
22.
Flores, T. (1985)_ The anthropology of acsthetics.In S. Diamond (Ed.), Anthropol-
ogy after '84: State of thc art, state of society, part II [Special issue]. Dialecti-
cal AlIthro!lology, ]0(1-2),27-41.
578
, "'L'"~I-''''''V~''-''''' , U'-<o'-"
Flores, T. (l999). 111 Illace: Poems by Toni Flores. Geneva, N\': Hobart & William
Smith Colleges Press.
Fox, R. (1989). Thc uiolellt imagination. New Brunswick, NJ: Rutgers University
Ple~s.
Pox, R. G. (Ed.). (199J). Rccaptllrillg alrthrop%f:,ry: \"?urkillg ill the presellt. Scinta
Fe, NIVI: Schoo! of American Rcsenrch Press.
Friedrich, P. (1979a). Bastard 1I1001lS. Chicago: Benjamin & j\'bnha \''laitc.
Fricdrich, P. (1979b). Linguisric relativity and the ordeHo-chaos continuum. In
J. IVlaquer (Ed.), Onlinglfistic i1mhropology: Essays ill hOJlorof Harry Hoijcr
(pp. 89-139). Malibu, CA, Unclena.
Friedrich, P. (1995). The world-listener & Citics (two poems). Amcricall AntiJro~
pologis!, 97, 658~659.
Fujimara,.J. H. (1998). Authorizing knowledge in science and anthropology. Amen'-
can Anthropologist, 100,347-360.
Geenz, C. (1973). The intcrpretation of wltllrcs: Selected essays. New York: Basic
Books.
Gecrtz, C. (1983). Local blOwledge: Further essays ill interpretivc .mthropology.
Ncw York: Basic Books.
Geenz, C. (1988). l'\lor/(s and liues: Tile (J/rthropologist as i1ltthOl: Stanford, CA:
Stanford University Press.
Geenz, C. (1995). Disciplines. Rmitrlll, 14(3),65-101.
Geenz, C. (1998, October 22). Deep hanging out. New York Rcuiew of Books,
pp. 69-7IJ.
Gellner, E. (1998). Lang/lage altd solifllde: W/ittgenstein, Malinowski, and the [-Iabs-
blllg dilemma. New York: Cambridge University Press.
Gonlicb, A., & Graham, P. (1993). Parallellvorlds: An lIltIJrollologist and a writer
ellcorlJlter Africa. New York: Crown.
Gould,S.). (1991). Ivlore light on leaves: Can a great artist also be a great scientisr?
Natllril! History, 2, 16-23.
Graham, L. (1995). Performing dreams: Discourse of iII/mortality amollg the
XC/vallte of celrtral Brazil. Austin: University of Texas Press.
Gnn;es, R. (197], December 2). Science, technology, and poetry. New Scientist,
PI'. 34-35.
Guetti, J. (1984). \X1irtgenstein and literary theory. Raritan, 4(2), 67-84.
Hall, D. (1988). Poetry alld ambitio}/: Essays 1982-88. Ann Arbor: University of
Michigan Press.
Hallyn, F. (1990). The poetic stmctllre of the [uorld: Copenzicl/s and Kepfel: New
York: Urzonc.
Handler, R. (1983). The dainty and the hungry man: Litcrature and anthropology
in the work of Edward Sapir. In G. \'{~ Stocking,]r. (Ed.), Observers observed:
Essays all ethnographic research (pp. 208-231). Madison: University of \'\lis-
cOllsin Press.
579

Handler. R. (1985). On dialogue and destructive analysis: Problems in narr:1ting
n:ltionalism and cthniciry.jolmul! ofJll/thro/Jological Res(!l1rch, 41, 171 ~ 18 I.
I-Lllldler, R. (1990). Ruth Benedict and the modernist sensibility. In lvL }.1<ll1g'll1aro
([eL), AA,,,l,,,,,,;,,, ,,,,tlm:>l-,o/ob,)'; P1'01Il (icUworl.:. to text (pp. 1 (;3 -ISO). Prince-
ton, NJ: Princeton Universiry Press.
Handler, R., & Segal, D. ("1987). Narr;lting multiple realities: Some lessons from
.lane Austen for ethnographers. Allthropology alld HUII/anisll/ Quarterly,
12(4), 15-2l.
Handy, \X~ C. (J 973). 71Jl1lUler fr01l1 the sea. Honolulu: University of Haw:lii Press.
Harland, R. (1987). Superslrlfcflfmlism: The philosophy of structuralism d/l(l post-
stTlletflralism. New York: Methuen.
Harris, ]. (1997). Giono's invisible sheep: Lacanian mirroring :lnd modeling in
Walcon's "Another life." SOllth Atltll1tic Quarterly, 96, 293-309.
Harris. Iv!. (1977). Ca!lnibals and hiltgs. New York: H.ando111 House.
H;lrris, lvI. (1987). The sacred cow alld the abominable p(fJ. Riddles of jood altd wl-
ture. New '{ark: Simon & Schuster.
Hill, J. H. (1986). The rcfiguration of the anthropology of language. Cf/ltmo/
Allthmpolob')l, 1, 89-102-
Hodder,1. (1986). Reading the past: Current a/Jproaches to il1terpretatioll in archae
ology. New York: Cambridge Universiry Press.
Hoffmann, R. (J 987). Thc melamict state: Poellls by Roald [-[O((l/WIlII. Gainesville:
University Presses of Florida.
Hoffmann, R. (1990a). Gaps and verges: Poems by Roald I-1o(fmalm. Gainesville:
University Presses of Florida.
Hoffmann, R. (1990b). lvlolecular beaut}'. ]ollmal ofAesthetics a/ld Art Criticism,
4S,191-204.
Hoffmann, R. (1995). The same and not the sallie. New York: Columbia Uni\lcrsity
Press.
Holquist, M. (1990). Dialogism. Bakhtin alld his world. New York: Routledge.
Holton, G. (1993). Science and anti-scicnce. Cambridge, I'vlA: Harvard University
Press.
Jackson, 1\'1. (1982). Allegories of the wildc/"Iless: Ethics and amb(r;/Iity ill KIf/"i_1I!ko
llarratiues. Bloomington: IndianJ University Press.
Jackson, 1"v1. (1986). Bar(/lua and the wa)'s birds fly i/1 the sky. W:::Ishingron, DC:
Smithsonian Instirution Press.
Jakobson, R. (1987). Language ill literature (K. Pomorska & S. Rudy, Eds.). Cam-
bridge, Ivli\: Harv<ud University Press.
James, A., Hockey, j., & Dawson, A. (Eds.). (1997). After writing wllme: Episte-
mology and praxis ill contemporary anthropology. New York: Routledge.
.l3.fvie, I. C. (1983). The problem of the ethnogrClphic real. ellrmlt Anthropolog-)'.
14,313-325.
580
f-\nrnropologlCQI f-'oeflCS
Jennings, B. (1983). Interpretive social science and policy analysis. In D. Calbhan
& B. Jennings (Eds.), Ethics, the social sciences, <1!u{ policy analysis (pp, 3-
35). New '{ork: Plenum.
jes:;or, R., Colby, 1\., &= Shweder, R. A. (Eds.). (1996). EthlIogmphy L1lld Imll/{/n
developmcnt: COlltext and meanil/g ill social inquiry. Chicago: University of
Chicago Press.
Karp, I., & Levine, S. D. (Eds.). (1991). Exhibiting wItwes: The poelics and politics
of 1III15eU1l1 disphly. Washington, DC: Smithsonian Instirution Press.
Kilbride, P. L. (1993). Ami-ami-universalism: Rethinking cultural psychology and
ami-anti-relativism. RCI/iews in Anthropology, 22(1), 1-11.
Knab, T. J. (1995). A W(1/' of witches: A jOl1mey ilzto the underworld ol the {(Jlltelll-
porary Aztecs. New York: HarpcrCollins.
Knauff, B. M. ('199(;). Genealogies jor the In'esclIl in C1I!lllrd anthropology. New
York: Routledge.
Kroskrity, P. V (1985). Growing with stories: Line, verse, :md genre in an Arizon::l
Tewa texr. .foltr1wl ofA.l1thropological Research, 41,183-199.
Krupat, A. (1992). Etlmo-criticism. Ethnography, history, literatllre. Berkeley: Uni-
versity of C<llifornia Press.
Kurten, B. (1980). DilIIce of the lZ?:CI: New York: Berkeley.
LadermaJl, C. (1991). Tailling the wind ofdesire: Psychology, lI1edicine, and aesthet-
ics i1/ klalay slwJJ1allistic performance. Berkeley: University of Cllifornia
Press.
La Farge, O. (1929). Laughing boy. Boston: Houghton lvlifflin.
Lakoff, G., & Johnson, M. (1980). Metaphors I[le lille by. Chicago: University of
Chicago Press.
Lakoff, G., & Turner, 1v1. (1989). More thall cool reason.' A fic!dguide to {metic met
afJhOl: Chicago: University of Chicago Press,
Langness, L. L., & Frank, G. (1978). Fact, fiction, and the ethnographic novel.
Anthropology and HlfIIll1llism Quarterly, 3(1-2), 18-22.
Lanser, S. S. (1981). The l/ormtit/e act. Princeton, N]: Prince ron University Press.
Lavie, S. (1990). The poetics of military oCClipation: 1I'1zeiJla allegories ol Bedof/ill
identity 11I1der israeli and Eg}'/Jtiau mle. Berkeley: University of California
Press.
Layron, R. (1997). All il/troduction to theory in anthropology. Cambridge: C::I111-
bridge University Press.
Leavitt, .I. (Ed.). (1997). Poetry mId 1./1"Ophecy: The anthropology ofil1spiration. Ann
Arbor: University of Michigan Press_
Lett, ]. (1997). Sciellce, reaSOII, and anthropology: The (Jrillciples of ratiO/wI
inquiry. Lanham, MD: Roman & Littlefield.
Levi, P. (1914). The periodic table. N!:w York: Schocken.
Levine, G. (1987). Literary science-scientific literature. Rariti711, 6(3), 24-41.
581
 .. '~", -" .~~." ,~",.".~ ,,~, "'--''-''''
Levine, G. (1994). \X/hy science isn't lirerature: The imponance of differences. In
A.l\Jcgill (Ed.). [~ethin!;,ing objectiuity (pr. 65-79). Durham, NC: Duke Uni~
vetsifY Press.
fJ:vj.Srr.llJu:, C. CJ96'I1. 1ht;flJJ! fWI/Ji,1!1J!:, /Ill (wllJ1'npnlog;,c,,1 5'i,,,}y ,,f ,,,.,,,,,,,!,,,.,.
societies ilT Bm::.i! (J. Russell, Trans.). New York: Atheneum.
Levi-Strauss, C (1962). "Les chars" de Charles B;:wdclaire. I.:HoTllme, 1( 1),5-11.
Levi-Strauss, C. (1967). Structural allfhropo!oS:.'}' (C Jacobson & B. G_ Schoepf,
Trans.). Garden City, NY: Doubleday.
Levi-Strauss, c.
(J 969). The raw and the coo!ced (/. \X1cightman & D. \X eightm;ln,
'
Trans.). New York: Hatper & Row.
Liglnrnan, A. (J 993). EilTstein's dreams. New York: Time W!arner.
Lim6n,1- (1994). Dancing Il'ith the dellil: Society and wltuwl poetics in M exic<1n-
Americilll South Texas. Ivladison: University of \Visconsin Press.
Lurde, A. (1984). Sister olltsidel: Trumansburg, NY: Crossing.
Loriggio, F. (1990). Anthropology, literJry theory, and the tr,lditiolls of modern-
iSI11. IIl!\'1. I'vlangal1<lro (Ed.), Modemist unthro!Jology: From fieldwork to text
(pp.115-141). Princeton, N]: Princeton University Press.
iv1acCanncll, D. (1992). Empty lIteeting grounds: The tomist papers. New York:
Routledge.
IVLmganaro, LvI. (Ed.). CI 990a). Model'llist illltlnopology: From fie!dwor!~ to text.
Princeton, NJ: Princeton University Press.
ivLlIlg,anato. ;V1. (1990b). Textual play, power, and culrur'll critique: An orientation
to modernist anthropOlogy. In M. lYbnganaro (Ed.), Modernist alrthrotJol-
og)': From fieldwork to text (pp. 3A7). Princeton, NJ: Princeton Universiry
Press.
j\ilucus, G. E. (1980). Rhetoric and the ethnographic genre in anthropological re-
se3rch. Cllrrelll Anthropology, 1/, 507-510.
IV1::lrcus, G_ (1938). Parody and the parodic in Polynesian culrural history. Cllltliral
Anthropology, 3, 68-76.
Marclls, G. E. (J 993).1\ review ofEtlmo-criticism: Ethnography, history, literatllre,
by A. Krupat. Americall Anthropologist, 395, 766.
TvIarcus, G. E. (J 998). Ethnography through thick and thin: A nell' research imagi~
nary for anthropology's changing professional culture. Princeton, NJ: Prince-
ron University Press.
Iviarcus, G. E., & Cushman, D. (1982). Ethnographies as texts. Amlfwl Reuie/u of
Anthropology, 11, 25-69.
IVlarclls, G. E., &. Fischer, 1\'1. M. J. (1986). Anthropology as CIIltmal critique: All
experimental moment ill the hl1mall sciences. Chicago: University of Chicago
Press.
Ivbscia-Lees, F. L, Sharpe, P., & Cohen, C. B. (1993). The postmodernist [Urn in
anrhropology: Camions from a feminist perspective. In P. Benson (EeL),
582
'""I ,tt IIU/JUIUYIl.UI 'Uelll.:>
i\lIthrofJOlog)' ,md literatme (pp. 225-148). Urbana: University of IIJinois
Press.
!\'lanhiessen, r. (1963). At play ill the fields of the Lord. New )'orlc Vintase.
l\.'L"'!,;c",,,.on. f1. (1975")_ r;..,,- 7<"'-("5'" N<;:w Yorl~; ViULi.lSI:,
IVlaxweJl, N. (1904). From Iwowledge to wisdom: A rwoluticJII ill the aims ,md
lIlethods of science. New York: Blackwell.
tVlcAl!isrer, j. \X( (1998). Is beauty a sign of trmh in scientific theories? American
Scicntist, 86, 178-183.
Megill, A. (Ed.). (J994). Rethinking objectivity. Durh'lnl, NC: Duke Universiry
Press.
Iv1cschonnic, I~r. (1988). Rhyme and life. Critical Inquiry, J 5, 90-107.
I\'liner, E. (1990). Comparative poetics: All interCf/ltuTilI essay 011 thenri!'s of litcrd-
tllrc. Princeton, N]: Princeton University Press.
j\'10IlBcby, N. S. (1989). The ancient chi/d. New )'ork: HarperColiins.
I\lorz~lfi-Haller,I~ (J 998). Beyond rextual an3lysis: Practice, inreracting discourses,
and the experience of distinction in Bors\Van~l. enltwal Anthropology, 13,
522~547.
Napoli, D. ]., &. Rando, E. N. (Eds.). (1979). Li/lguistic /III/se. Edmonton: Linguis-
tic Research.
Narayan, K. (1989). Storytellers, sail/ts, and scoundrels: FoIl: n,nratiuc iI/ Hindu
religious tcaching. Philadelphia: University of Pennsylvania Press.
Obeyesekerc, G. (1992). The a/JOthcosis o(Captain Cooh: Europe'1J1lJ1ytlmwking
ilt the P.:/ci(ie- Princeton, NJ: Princeton University Press.
O'Meara,]. T. (1989). Anthropology as empirical science. American AI1t1Jropolo-
gist, 91, 354~369.
O'Me<lra, ]. T. (1997). Causation and the struggle for a science of culrure [wirh
commentary]. Current Allthropology, 38, 399-41rl.
O'Nell, T. D. (1994). Te!lingabour whites, ralking<1bour Indi<1lls: Oppression, resis-
rallce, 3nd contemporary Inelian identity. Cultllml AlIthrofJOlogy, 9, 94- J 26.
Poggie,]. ]., Jr-, De\XTalr, B. R., & Dressler, \Y,Z WI. (Eels.). (1991).AlltlJropological re-
search: Process and aptJlicatiolt. Albany: State University of New York Press.
Polanyi, J'v1., & Prosch, H. (197 S). M.eaning. Chicago: University of ChicJgo Press.
Prattis, J. I. (Ed.). (1986). Reflections: The anthropological IlIltse. Washington, DC:
American Anthropological Association.
Prarris,.J. 1. CJ 997). Anthropology at the edge: Essays 011 mlture, s)'lI1bol, alld COJl~
sciousncss. Lanham, MD: University Press of America.
Rabinow, P. (1977). Reflectio1lS 071 fieldwork ill ldol'OCco. Berkeley: University of
California Press.
Rabinow, P. (1983). Humanism as nihilism. In N. Haan, R. N. Bellah, P. Rabinow, &
\\( Iv!. Sullivan (Eds.), Social science as moral inql/ir-y (pp. 52-75). New York:
Columbia University Press.
Rabinow, P. (Eel.). (J 984). The FoucaT/lt reaJel: New York: Pantheon.
583
 WI I t::KI-I{t:: IAIIVI'<, t::VALUAIIVl'l, AI'lLJ l'u:.nU::.;ll:'I'l,,,"",1 ''-''-'
Rabinow, P. (1986). Representations arc social bets: IVlodernity ,llld post~
mudcrnity in anthropology. In J. Clifford & G. E. IVIarclls (Eels.). Whti/lg CIt/-
turc: The poC/ics and politics of ethnography (pp. 234-261). Berkeley: Uni-
Vt:r.~ilY of California Press.
Rabinow, P., & Sullivan, \"X~ lVL (Eels.). (1987). Il1terpretive sucial scie/lce: A second
looh. Berkeley; University of California Press.
jZappon, N. (1997). Transcelldent i/ldiuid1fal: Towards a literary alld liberal I1nthro-
polog)'. New York: Routledge.
Reck, G. G. ("1978). 1/1 the shadow of Tlaloe: Life in a Mcxin7lluillage. Prospect
Heights, IL: Waveland.
Reichel-Dolm3roff, G. (l996). YlIrIl!Jar{; Studies of dll Aml1Z0lliall foundation
lJIyth. Cambridge, IVIA: Harvard University Press.
Richards, I. A. (1929). Practical criticism. New 'York: Harcourt Brace Jovanuvich.
Richardson, L. (1994). Writing: A method of inquiry. In N. 1(. Denzin & '{ S.
Lincoln (Eds.), Halldbooh o( qualitatiue research (pp. 516-529). Thollsand
Oaks, CA: Sage.
Richardson, Iv1. (1990). Cry IOIlCSOIIIC and other accollnts of the aJfthropologist's
project. Albany: Stare University of New York Press.
Richardson,Ivl. (19983). Poetics in the field and on the page. Qlwlitat il/e IlIqllir)', 4,
451-462.
Richardson, [..,1. (l998b). The poetics of a resurrection: Re-seeing 30 years of
change in :1 Colombian community and in the amhropologic81 cnrerprise.
American Anthropologist, 100, 11-21.
Riffaterre, Iv1. (1970). Describing poetic structures: Two appro:lChes to Baude-
laire's "Les chats." In J. Ehrmann (Ed.), Structllralism (pp. 189-23 0). Garden
City, NY: Doubleday.
R.iffaterre, !vL (1.984). Semiotics ofpoetry. Bloomington: Indiilna Universiry Press.
Robbins, B. (1987). Poaching offthc disciplines. Raritan, 6(4), 81-96.
Rony, R. (1979). Philosophy and the mirror of natl/re. Princeton, NJ: Princeton
University Press.
Rony, R. (1981). Nineteenth-century idealism and twentieth-century rextualism.
Monist, 64, 155-174.
Rosaldo, R. (1989). Clflillre and trllth: The remahing of social allalysis. Boston:
Beacon.
Rose, D. (1991a). In search of experience: The anthropological poetics of Stanley
Diamond. In 1. Brady (Ed.), Anthropological poetics (pp. 219-233). SaYage,
MD: Rowman & Littlefield. (Reprinted from Americall Alllhropologist,
1983,85)
Rose, D. (1991b). Reversal. In 1. Brady (Ed.), Anthropological poelics (pp. 283-
301). SelVage, l"vlD: Rowman & Littlefield.
Rose, D. (1993). Ethnography as a form of life. In E Benson (Ed.), Allthropology
and literature (pp. 192-224). Urbana: University of Hlinois Press.
584
MIIU IIUfJUIOg!COl r-oencs
Rothenberg, J. (l98 1). Pre-faces alld other writings. New Y(Jr!c: New Directions.
Rotlli:nbe_rg, .J. (Ed.). (1985). Technicialls of the sacred: A Ti7llge 0/ !Joelries from
Afne11, Amcrica, Asia, Emope alld Oeeal11rr- Berkeley: University of Califor-
J
ni~ Prf'-<-.'" '
Rothenberg, J. (1994). "Je est LIn :mtre": Erhnopoetics and the poet as other. Ameri-
call Anthropologist, 96, 523-524_
Rot:henl~e~g, j., & Rothenberg, D. (Eds.). ("1983). SymposiulI1 ofthe whole: A range
0/ discourse tOWi1rd all etlmopoetics. Berkeley: University of California Press.
Sagan, C. (980)- Cosmos. New York: Ballantine.
Sagan, C. (1985). COli tact. New York: Pocket Books.
S<lhlins,1V1. (198]). Historical lI1etalJhors alld /IIJ,thiciil realities: Stl'llctllre il1 the
early history of the Salldwich Islands hillgdom. Ann Arbor: Universin' of
l'vfichigan Press. .
Sahlins, .M. (1985). Islands of history. Chic<lgo: University of Chicago Press.
Sahlins, 1"'1"1. (1996). How "natilies" thillk: AboUI Captain Cook, (or cJ,:alllple. Chi-
cago: Universiry of Chicago Press.
Said, E. (1991). The politics of knowledge. Raritan, 110), 17-31.
Sangren, l~ S. (1991). Rhetoric <lnd the authority of ethnography. In S. Silverman
(Ed.), Inquiry and debate ill the human scielfces (pp.17?-3(7). Chicago: Uni-
versity of Chicago Press.
Schechner, R. (1985). Between theater alld allthropology. Philadelphia: University
of Pennsylvania Press. .
Schechner, R. ("l995). The flltllre of ritual: \X/rilings all wit lire alld performallce.
New York: Routledge.
Scholes, R. (1974). Structuralism ill literatl/re: An introdllction. New Haven. CT:
Yale University Press. '
Scholte, B. (1966). Epistemic paradigms: Some problems in cross-cultural research
on social anthropological history and theory. Alnerican Anthropologist, 68.
1192-1201.
Scott, D. (1991). Anthropology and colonial discourse: Aspects of the
demonological construction of Sinhala cultural practice. ClIltural Anthropol-
ogy, 7,301-326.
Shankman, P. (1934). On semiotics and science: Reply to Renner and Scholte. Cur-
relll Anthropology, 25, 691-692.
Sherzer,]. (1987). A discourse-centered approach to language and culture. I\lIIeri-
CilJlAnthropologist, 89, 295-309.
Sherzer,.J., & Woodbury, A. C. (Eds.). (1987). Native American discollrse: Poetics
and rhetoric. New York: Cambridge UniversitY Press.
Shore, B. (1996). Cognitioll, wltflre alld the problem of meaning. New York:
J
Oxford University Press.
Shweder, R. A. (1996). True ethnography: The lore, the law, and the lure. In
R. lessor, A. Colby, & R. A. Shweder (Eds.), EtlmograpbJI and hll1lWIl devel-
585
 opll1ent: COif text alfd lIIea/rillg ill social inquiry (pr. 4~52). Chic:lgo: Univer-
siry of Chicago Press.
Silko, L. Iv1. (1977). CeremoJlY. New 'lark: Penguin.
Sinll11g, N. T. (1991 I. Ihill/s ofcOJItilet: A stud), of t1.w ;nter/Jla)' amI ;lItersectio/1 0/
traditionol and lIoll~traditio!lal literatI/res, c/!ltllres, and me/ltalities. New
York: Pace University Press.
Simpson, D. (1994). Literary criticism, localism, and local knowledge. Raritan,
]4(1),70-88.
Slater, C. (1982). Stories all a stringe The Bmziliflll Iitemtllm de cordc/o Berkeley:
University of California Press.
Spiegelberg, H. (1975). Doing phenomellology: Essays DII.md ill phenolllenology.
The Hague: iVlaninus NijllOff.
Spradley,.J. E., & IVfcDonollgh, G. P. (Eels.). (1973). Anthropology throllgh litcra-
tl/re. Boston: Linle, Brown.
Stcwarr, J. (19(~2). ClIrl/iug road: Stories by Jolm Steu'art. Urbana: University of Illi-
nois Press.
Stewart, J. (1989). Drill/~crs, drulllmers, dlrd dael1t folh. Albany: Slate University of
New York Press.
Stew~lrt, K. (1996). A space all the side of the road: Cllltllml poetics in an "other"
Amcrica. Prlnceron, NJ: Princeton University Press.
Swann, D. (Ed.). (1983). Smoothillg the ground: ESSd)'S ill Naliue American ol"dllit-
eratme. Berkeley: University of California Press.
Swearingen,.J. (1986). Oral hermeneutics during the transition to llreracy: The
contemporary debate. Cultural Al1throf)()logy, 1, 138-156.
Tarn, N. (1991). Views from the wCiluillg tIlollllfaill: Selected essays in poetics and
anthropology. AJbuquerque: University of New MeXICO.
T:111ssig, M. (1987). ShamanisllI, co/o/lialis1l1, alld the wild lIIalT. Chicago: Univcr-
siry of Chicago Press.
Taussig, .roil. (1993). i'vIilllesis (lnd afterit)': A partiwlar history of the senses. New
)'ork: Routledge.
'Elylor, L. J. ("1996). "There are two things thar people don't like to hear aboUt
themselves": The anthropology of Ireland and the Irish view of anthropol-
ogy. South Atlantic Quarterly, 95, 113-226.
Tedlock, B. (1991). From participant observ,ltion to the observation of participa-
tion: The emergence of narrative ethnography. jOllntfll of Anthmpo/ogical
Research, 47, 69-94.
"I'edlock, B. (1992). The beautiful and the dallgemlls: Dialogues with the ZUlli Illc/i-
<11rs. New York: Penguin.
lecllock, D., & Tedlock, D. (1985). Text and textile: Language and technology in
rhe arts of the Quich Maya.]olll'/1al of Anthmpological Research, 4-1, 121-
l46.
586
Tcdlock, D. (Trans.). (J 972). Fillding the ccllter: Nmmt iFC poetry of the ZUlli huh
tillS. New York: Dial.
Tedlock, D. (1983). The s{Johcllword dl/d t/;e wor/~ of illterpretl1/ioll. PhibdelphiJ:
Univcl<;ilT of Pennsylv'1ni", Pr~,ss.
Tedlock, D. (Trans. & Commenraror). (1985). POllOI V/I/;: The iVlayan boo/~ of the
dl.l/UII of life. New York: Simon & Schuster.
Tcdloek, D. (1987a). Hearing a voice in an ancienr rcxt: Quich IVlaya poeries in
performance. In.J. Sherzcr & A. C. Woodbury (Ecls.), Niltiuc America/l dis-
COltfse: Poetics alld rhetoric (pp. 140-175). Cambridge: Clmbridgc Univer-
sity Press.
Tedlock, D. (l987b). Quesrions concerning dialogical anthropology (with a
response from Stephen Tyler). j01lma{ o{Anthropologicill Research, 43, 325-
344.
Tedlock, D. (1990). Days from [1 dremn almillli-IC. Urb,lIla: University of lliinois
Press.
Tedloek, D. (1991). The speaker of rales has more rhan one string to pL1Y on. In
I. Drady (Ed.),Al1thropo{ogical poetics (pp. 309-340). S;wage, J\'ID: Rowm::1n
& Littlefield.
Tedlock, D. (1991). Ethnopoetics. In R. B;ultllan (Ed.), FolJdore, WI/11hz! per-
(omlr7l1ces, dlld poplllarelltertaillll1ellts (pp. 8185). NewYor!.::: Oxford Uni-
versity Press.
Tedlock, D. (1993). Breath 011 the mirror: Mythic uoiccs '-/lid uisiolls of the lh'ing
Mdya. Albuquerque: Univcrsity of Ncw i\'lexico Press.
Thomas, E. M. (1987), ReilldeerlllOon. New York: Pocket Books.
Thom<:ls, N. (1989). Ollt of tiltle: History and euollltirm ill anthropological dis-
course. New York: C<:lmbridge University Press.
Thol11as, N. (1991). Against ethnography. Clllwml Anthropology, 6, 306-311.
Thompson, \X~ L (1983). BIlle jade from the ltIol"l1illg star: All essay alld a cycle of
poems all QlIetzalcoatl. WIest' Stockbridge, ~ILA: Linclisfarne.
Tiles, h.'1. (1984). Bachclard: Science Wid objectiuit)l. New York: Cambridge Univcr-
siry Press.
Tilley, C. (Ed.). (1990). Reading lJIilterial culture. Cambridge, Ivl.A: Blackwell.
Todorov, T (1977). The poetics of prose (R. How<:lrd, Trans.). Irhaca, N''{: Cornell
University Press.
Todorov, T. (1981). Illtroductioll to poetics (R. Howard, Trans.). Minneapolis: Uni-
versity of lvlinnesora Press.
Trawick,1'v1. (1988)_ Spirits and voices in Tamil song. Amcrican Ethnologist, ./5,
193-215.
Trawick,Iv1. (1997). Time and mother: Conversations With a possessing spirir. Tn
J. Lec1Vitt (Ed.), Poetry and prophecy: The anthropology of illspiration
(pp. 61-75). Ann Arbor: University of IVlichigan Press.
Tsing, A. L. (1994). From rhe margins. ClIltllral Allthropology, 9, 279-297.
587
 MI HI IIUfJUlUYILUI rU!o'lll-~

Turnbull, C. (1962). The (orest people. G~mlen City, N"!': Doubleday. \X/hirten, N. E.,.I1". (1988). Toward a critical anthropology. Alllericml Ethnologist,
Turnbull, C. (1972). The 1llOIIIltain people. New York: Simon & Schuster. 15,732-742.
Turner, E. (987). The s/Iirit alld the drum: A memoir ofA(rica. Tucson: University \X1ierzbicka, A. (1996). Sel/wlltics: Primes and IIl1il'ersals. Oxford: Oxford Univer-
01 Arizona Press. .
-Imner, V (1974). DmnUls, fields, and 1I1cfaplJOrs: Symbohc action ill/Jllnwn soci- \Xfilbert, J., & Simoneau, 1(. (Eels.). (J 982). Folk literature 0/ the Mafoca Illdialls.
ety. Irhaca, NY: Cornel! University Press. Los Angeles: UCLA Larin AmeriGl Center Publicarions.
Ti.lrner, V (Ed.). (198201). Celebration: Studies ill festivity alfd ritf/al. Washington, \\'ilk,S. {l991).HlIIJ/onistic il11thropology. Knm..v ille: University of Tennessee Press.
DC: Smithsonian Institurion Press. \Villi3ms, B. (1998, November 19). The end of explanation? Nell' Yor!::. Reuiclu o(
Turner, V (l9;\2b). From ritual to theatre: 7'llC /JIlllUlIl seriousness o( play. New Boob, pp. 40-44.
YCJrlc: Performing Arts Journal Publications. Wolf, E. (1982). Ellro{Je and the people luithou! history. Berkeley: University of
Turner, V (1983). A revic\v of "Ethnopoetics." In.J. Rothenberg ~l.[ D. Rmhenberg California Press.
(Eels.), Symposium of the whole: A range ofdiscourse toward an etlJlloJJOetics Wolf, IV1. A. (1992).!\ thrice-told tale: Feminism, pastil/adem ism, alrd etl}//ogra/Jhic
(pp. 337-342). Berkeley: University of California Press. respo/lSibility. Stanford, CA: Stanford UniversiLY Press.
Tyler, S. A. (Ed.). (1969). Cognitiue allthropology. New York: Holt, Rineharr & \Vurhnow, R., Hunter,.J. D., Bergesen, A", & KurzweiL E. (Eels.). (1984). Cultural
\'{/insron. analysis: The worh o( Peter L. Bergel; Mary DouglLIs, fdiche! Fouct/ult, and
Tyler, S. A. (1984). The poetic rurn in postmodern anthrDpology: The poetry of jiil'geJl Hauenllas. Boston: Routledge & Keg~lJl Paul.
Paul Friedrich. American Allthropologist, 86, 328-336.
Ulin, R. C. (ln4). Understanding wltures: Perspectiucs in olfthropology alld social
theory. Ausrin: University of Texas Press.
V:J!ery, P. (1964). Aesthetics (R. Mennheim, Trans.). New york: Bollingen Founda-
tion.
Van Den Abbee!e, G. (1992). Ihwel as metaphor: From MOlltaiglle to Rousseau.
klinneapolis: University of Ivlinnesota Press.
Van Lawick-Goodall, J. (1971).111 the shadow of man. New York: Delta.
Van i'vIaanen, J. (1988). Tales of the field: On writing ethnogmphy. Chicago: Uni-
versity of Chicago Ptess.
Vendler, H. (1985, November 7). Looking for poetry in America. NC/u York Revieu'
o( Books, pp. 53-60.
Vendler, H. (1988). The music of what happells: Poems, poets, critics. Cambridge,
[vIA: Harvard University Press.
Vendler, H, (:1995). SOf/1 sa)'s: 011 recent {JOetl)'. Cambridge, MA: Harvard Univer-
sity Press.
Y\lebstcr, S. (1986). Realism and reification ill the ethnographic genre. Critique of
AnthrolJOlogy, 6, 39-62.
\,\/einstein, F. (199001). History and theory after the fa!!: All essay all inteI1}retatioJl.
Chicago: University of Chicago Press,
"Weinstein, F. (1990b), Who should write history? SUNY Research, 10(3),20-11.
Weiss, \XZ A. (1990). Challenge to authority: Bakhtin and ethnographic description.
Cit/turd Allthropology, 5, 414-430.
White, H. (1973). Metahistory: The historical imaginatioll ill nineteenth-century
Europe. Baltimore: Johns Hopkins Univetsity Press.

588 589

Understanding
-~-~----
Social Programs
Through Evaluation
Jennifer C. Greene
Social program evaluation is ;] field of applied social inquiry
uniquely distinguished by the expJ icit value dimensi 011S of irs knowl-
~dg~ claims, by the overt political character of its contexlS, ~lOcI by the
InevItable pluralism and polyvocality of its actors. J Social progrJl11 evalua-
tors use the methods and tools of the social sciences to address not 8bstracr
theoretical questions of interest to some scholarly colleagues, bm rather
the priority policy and practice questions of diverse social actors~
decision makers, program administrators, direct service st3ff, program
participants, and others. Social program evaluators aim to inform <lIlei
improve the services, programs, policies, and public conversations 3t
hand, in contexts lilce these:
Annette Barlow directs a COUnty United \\/ay III rural Georgia. Like the
boards of directors of United \V'ars elsewhere, Annette's boatd h"s increas-
ingly asked for information about the outcomes of the programs they fund.
Are babies in the COUnty being born any healthier as a result of the pre-
natal program? Are children who participate in the peer tutoring program
reading any better? Are seniors in the volunteer grandparent program less
590
'-"'U""'""'CU"U"'!:;I-'VL.IVI r ' v y ' v l " ; ; ' IIIIVUy" LVVIUUtlVl1
isolated, and, If so, docs this meaningfully affect their physical health, their
qu,:dity of life, and thcir economic self-sufficiency?
Marrin Segura is a newly elected U.S. senator from Arizon1 who finrk
hlm~e!f the junIOr minority member on the Scn~He Foreigll Relations
Committee. As he learns more abour American rehuionships around the
world, hc wonders about the influence on these rel,uionships of inrerna-
tional monetary policies, such as structural adjustment, tariffs, and rrade
agreements.
Evelyn Reynolds has 20 years of service as a soci31 worker in her county
social service department, \vorking primarily with public assistance pro-
grams and clients. As the programs have changed over the years-from a
proliferation of categorical benefits ro more-integrated benefit and ser-
vice packages to a phasing out of welfare altogether-so has Evelyn's job.
She finds herself with increasingly less <1uthority yet more responsibility
and with more and more desperate demands [or fewer and fewer re-
sources. "There must be some way to make this system work better!" she
asserted at a recent staff meeting.
Thomas Gteenbaum is the director of social services in Evelyn
Reynolds's stare. He is also experiencing tensions and contradictions in
his job. However, the pressures Thom;:ls experiences-anel thus his pri-
ority information needs-revolve around minimizing program COStS,
meeting federal regulations, and responding to the conflicting demands
of political action lobbyists.
Michael Grey \XTolf has been active in the Native American sovereignty
movement for more than a dec.lde. His recent work h"s focused on com-
paring Native and non-Native public elementary and secondary schools,
toward the goal of developing charter school legislation for Native
schools in his home swte.:'
In this chapter, I endeavor to use these and other specific contexts fa
redirect attention to the sociopolitical role of evaluation. I do so by high-
lighting the sociopolitical dimensions of evaluation approaches that in~
corporate qualitative methods and the changing face of this genre as it re-
sponds to challenges from within and without. These highlights 8re
embedded within the following organization for the chapter. The next
two sections locate evaluation within the social policy arena and qucll-
itative approaches within the overall landscape of social program evalua-
tion, respectively. The subsequent two sections portr<ly the philosophy
and practice of evaluation approaches that rely on qualitative methods.
591
 Phi/oso!Jhy refers to the general epistemological logic of justification for
t~1ese ap~roaches, and practice describes key facets of application. The
fmal section then locates approaches in this genre historically, connects
rhem to ~i!ll1ifiol1t theorPt;c,1 ~evelopmc",c. ;" ev"I"",;"", ~v"d 5"kb
t!1em thl'Ough comcmporary challenges toward a promising future jden~
tit}' and role.
The Contexts of Program Evaluation
The scenario~ just above, presented as characteristic contexts of social pro-
gram evaluatlon, clearly evince its politiGJ! inherence (Cronbach & Associ-
ates, 1980; Parton, 1987; \X7eiss, 1987). They directly engage the pluralistic
values of contemporary American democracy. These are Contexts abour
contested social policies and programs, about how and by whom resources
arc ~llocated, and about competing civic values, barb in the global arena
and m the 10c:11 community. The core coment of social program evaluation
is thus intertwined with political power and decision making about societal
priorities and directions. Evaluators working in these politicized contexts
must negotiate identities for themselves that maximize their credibility and
potential effectiveness. Different evaluators choose different grounds for
negotiation, grounds that lead to different evaluation purposes and roles,
yet at! are concerned with power and all are value based (Greene, 1997).
l\'1oreover, the work of social program evaluators is steered by the inter~
esrs of selected members of the setting being evaluated. (Interests here arc
value-based claims on resources.) In all evaluation contexts there are l1lul-
~ipl:,. often competing, potential audiences for evaluation-groups and
IndIVIduals who have vested interests in the programs being evaluated,
called stalwholdel's in evaluation jargon. These range from the powerful to
the povverless, from policy makers and funders like Annette Barlow's
United \Vay board and the U.S. Senate to program administrators like
Thomas Greenbaum and program staff members like Evelyn Reynolds, to
advocates like lVIichael Grey \Volf, and to the citizenry at large. Thus eval-
uators must also negotiate whose questions will be addressed and whose
interests will be served by their work, negotiations that again are inher-
ently value based.
Evaluation results then enter the political arena of social program and
policy decision making not as decomextualized, abstract, or tl~eoretical
knowledge clajms, but rather as interested knowledge claims, as empiri-
592
cally justified value judgments about the ll1erit or \vorrh of the programs
evaluated. Evaluators do nor JUSt claim to know about something, rhey
claim to know how good it is from selected vantage points. Ar root, EvaluJ.-
'~n" ~<, "bOHr- Yc~lLl;''5 (S'-Tiyc:n 1 :1.967, and judt'iinB (s~.lk<':l ] ~;'(j7). VdIUl;::J
permeate the evaluation landscape, frol11 values that are constitmive of rhe
programs being evaluated to values as professional ethics, to values as the
ethical aim of the practicc of evaluation (Sclnvandt, 1997b). I-renee, in
addition to negotiating evaluation purpose and audience, eV<lluators must
negotiate the criteria and standards upon which judgmEnts of quality will
be made. These criteria and the resulting judgments are inescapably power
and value based and thereby matters of debate and conflict in all contcm-
porary democracies.
1) Locating OualitativeProgram Evaluation
Because social program evaluation is inextricably intertwined \vith politics
and values, and because ev"l!uarors must navigate carefully amid competing
political and value agendas, it is essential that evaluators have a diverse set
of approaches to help guide practice. Currently, the diverse approaches
available to evaluators offer not only choices of methods, bur also alter-
native epistemological assumptions (about knO\vlcdge, the social world,
human nature) and distinct ideological stances (about rhe desired ends of
social programs and of social inquiry), These alternative philosophies and
methods (concerned with what and howwe know) and alternative ideolog-
ical stances (concerning the meaning of social and community life) are con-
stitutive of their respective approaches.
Table 16.1 offers a desc";!Jtiuc categorization of four contempor3ry
genres of evaluation approaches. Like all such categorizations, the bound~
aries of these genres are clear only in the presentation of them. In histori-
cal evolution, conceptual argument, and actual practice, genre boundaries
are considerably more fluid. The first genre, \vhich represents the histori-
cally dominant tradition in program evaluation, is oriented around the
interests of policy m8ker; and funders-characteristically, It involves
causal questions about the degree to which a program has attained desired
outcomes while retaining cost-efficiency compared to its critical com-
petitors. Recurrent demands for accountability in social expenditures
are well addressed by this genre. These include Annette Barlmv's United
\Vay board's desire for information on program oLJ(comes and state
593
 c,
:g I
c "ill 0
administrator Thomas Greenbaum's need for inform3tion on his social
"cc "' 'ti t "'c
>
'"" Ce- ~ \> ~
~

'0" ill ffi ill

'" '" '" " 0

'0
ill "'
'0 E
'D."
0.
,S
's welfare program's costS versus benefits. Also in this genre, science, in the

'" '0 ''"""

~ ill
CD x '0 0 ill :7 .0 e 0 guise of program evaluation, is espcci;;dly valued for its presumed objectiv-
0' '~" C
""'" '0
'" '" C E
" '"
Q
~ Q
'" E
0

'""'
~ Q
~ ~

"e ".!il ~
c 0
~ ~ E
2 ;; '" 8
u ~
'0 U .0 0
I "f ,x
cL
0
m W
J: J;
ity und lrurb cbims, still held as regulative ideals.
1;
,"" The second genre of evaluation approaches 3fOse largely in response to
0, b. :3
2 C
0
E C
2 uc C '" 0, !: ,~

.g
0 0
D. '0
0 0

"ti "
"
~
x
et "' "'
0 c0
0 0, ~ cf 0 > 0,
>
~
c ~ 1" the failure of cbssic experimental science to provide tIllst\vorthy, timely,
'" '"2
u ill
0

~lI1d lIseful information for program decision making. As early trials clearly
>
e- c E
ill 0 0 0
0
e- '" 0 '~ u c
~
> 0,

'" '" i;;" 'co Q, !:

"
~
C "iii 0,
0
.~ "' " e- "'
'~ ~
0. ~
D > e 2' C ~ u
,~ 'c"
'0 ~
0
E .0
.~ ~" e-0. 0. 0 .~ E
m
u 0
i '0 E
'" denlO11strated (one classic is the Head Start evaluation; Cicirelli & Associ-
" '"
~ 0
~ '" c 1" u
," "o ,S
"" '" e-
0
t
D. a 3: ~ '" E 0. 0 c 0

'" "
E
~ I
'" '" ~ ~ Q
arcs, 1969), the logic of experimentalism simply does not transfer \vell to
real-\vorld social contexts. This second genre then refocllsed attention on
C a> c the needs of decision makers, especially the pracrical needs of on-site deci-
Qf C'

,.
,S m' 0
.~'
.'" :;
Q 0
~
ill

-.s0
0
ill U
E 0E 2-
"
"' 6 <5 "-
'is
~ c u :; .~
0
;;
:g
i'
sion makers for program information useful for management decisions.
'U;0." '0 0
C
0
i 1: c0
~
>
:; c
'f- ,2 '0
10 '0 oi e-
0
"' Evelyn Reynolds's plea for information useful for program improvement
~ ':7" 1i c -ii u '~"
~
c 1: ~ .~ 0
'~"
.;::: 0

'~"
0
0 t c '" is well answered by this genre. With its ueilitarian value b3se, this genre of
"'E" 'is
u

'"t
'0
~
'"
~~ '<Ii u0
'-';l
0
0
0

>
f,
c
'"
0

.!::!' "' "'

0
0
0

~
0
'3'
ill
""
0
u
0
U
"
0

~
,<
;g
w uu
'c"
ill ~
1: C
~ ill
c '-5
m
U
C ,E .~ c
2
0 -ii c
Q
~ c'
,Q 0 0 ~
'" 0 "'c .'2
~ u
evaluation embraces eclectic methods choices in the service of practical
problem solving. IvIichael Parron's (1997) utilization-oriented c.v"1luation
~ -s:8
0 10 10 10 't:: a0 C
0 '0 u 0
.~

Q'"
g
0'
u
0
0
0
m U
-"u
~
0
~ ~ S
0 ~
.~ "' c
0
0. 0' '"
0.
0
u
C
0 u
0
u
0 '0"
'is
t
0.
0
C
.~ ,S>
u
0
- a 0
ill
~
u
U
c
0
is::1 quintessential example of this genre. His approach to qualitative eval-
uation also falls within this genre, given its utilitarian value base (Parton,

",,' J990),
x'
But it is in the third cluster that qualitative 3pproaches to evaluation
u
c
1"
'm ~ UJ
'0 ~
ill

'co
0
~
,~ e'
a0 '
'" E c 0
,,,'
" '9
m
ill '0
0
0
that arc rooted in long-standing philosophical and disciplinary traditions,
a0. 13'" ~ E 2
"'c 10 ~ ~ such as interpretivism and ethnography, respectively, have found their
a. ~
~

2 E
0 c 0 0 ~ u ~ 0
u home. In both theory and practice, qualitative approaches to evaluation
'"e' '" 'is
0
Oi
> u e0 1': iic ~
0
1! ~ "
E E 'w span multiple "moments" in qualitative inquiry (sec Lincoln & Denzin,
-,0 0 u :s e- '" E ~
'"
0 ill 0 ~
-, '0 uc : c0 C ~ '>
" '"
~
0 u Chapter 14, Volume 1), from the blurred genres and representation/legiti-
'R ~ "' '5
0: c0 0
ill
I
"' .2 C "' 0
mation crises of the 1970s and 1980s to the experimental posrmodcrnism
of today. Yet most of these approaches ill evaluatioll share a hermeneutic
m oi
~ '0 vie\\' of social scientific knowledge claims, a value orientation that pro-
c c6 c ~'
c0 motes pluralism, and a preference for qualitative methods. Also influ-
D. '"
.J;
0
E e
0

E ~u
'co
cc
E . u e 'Q" 0
enced by the responsive tradition within program evaluation (Abma,
0
:g
:5e- a~
2
"'c'o" ,"c'
'0
c C c' ~ ~
'E" ~
a E 19973; Stake, 1975), these approaches characteristically seek to address
8 c '0
E' <5 'u" 0
~'
0 0
~
c0 the interests and honor the experiences of stakeholders closest to the pro-
2 Q
'"
~
,"e i'2x c ,S>
u E '" o'
0U
~ i5- 0 '0 c 0; ~
c W D C grams evaluated-namely, program staff and beneficiaries-by giving
~
C
W ::;
D
E a:
0 C
D
U
Q
x
W ~
E
" -5 "5
0
Q
D ~

voice to their conrextualized program understanding. In-depth, con-

texruaJizecl understanding may also bc of interest to other stakeholders,
H
0. c'
0 for example, Arizona Senator lvlartin Seguro's interest in the connections
~
,","E '>n~ ;::
~
0, ~ E 0
to ;r; 0
H
S! '> 00
;':
Q
c d: Do '0 r:: bctween international monetary policies and U.S. relations abroad. The
....,
"-I 0
c
~
';;; .g i' ~
~
00
D
o~
0
u~'
.g E
'0 r::
~ Ji ou u - early work of Robert Stake (1975) and of Egon Cuba and Yvonna Lincoln
i:i'"
.12 ~ 0 E-
~ Do ~
e& &~
0
~ E ;': .E
1" 0c E
u "'-
''''
(3 .~ of'" (1981) significantly shaped the contours of this genre. As discussed later in

595
594
 this chapter (and as illustrated throughout this volume), contemporary
work in this genre is reshaping it in significant ways.
Finally, the fOllrth genre represents evaluation approaches that are
"lij\~Hlv ;de()l()['".~<::'1I.1) t-l~",-t cxpl;~~;tly ",ctv01.nc~. " pan:cuLl1- v"lu,", ",-sellcb,
for example, of Rawlsian social justice (House, 1990), cl11powcrmenr
(Fetterman, 1994), critical race consciousness (Ladson- Billings, 1998), or
social change (Ryan, Greene, Lincoln, lVlathison, & rVlenens, 1998; \\lfhit~
more, 1994)..Michael Grey Wolf's explicit desire to advocatc for Native
American sovereignty in his evaluative work well illustrates this genre. As
norcd, all evalu8tion approaches advance certain ideals and values. "They
privilege the stakeholder audiences \"1ho share those ideals and they SlllTl-
man the methodologies that enable their realization" (Greene, 1997,
p. 27). Distinctively in the fOLlrth genre, the essential ratiomtles for evalua-
tion are, first, the advocacy of ideals and values and, second, the answering
of certain program questions. For most other evaluators, answering pro~
gram questions is the sC8ted first priority. \Vithin this fourth genre of con~
remporary evaluation approaches are a diverse array of philosophical
arguments, including critical soci,-ll science, feminisms, andneo-i\IIarxislll.
\Vithin evaluation, British ,md Australian democratic evaluation (IVlac-
Donald, 1976) and action-oriented evaluation (Carr & Kemmis, 1986)
are early exemplars. GLIba and Lincoln's (1989) fOllrth-generation evalua-
tion also \varrants examination as it advances an activist ideology with
grounding in a constructivist philosophy.
Even as I endeavor in Table 38.1 to present the four genres of evalua-
tion approaches as significantly differentiated by ideology, in addition to
epistemology and technique, the methodological domination of evalua-
tive thinking still reigns (Greene, 1992; Schwandt, 1998). This is nowhere
more evident than in the naming of different approaches to evaluation by
tbeir primary methods, as in qualitative euallfatiall. This naming obscures
important differences among evaluative approaches that rely on qualita-
rive methods, such as those promoted by Stake, Guba and Lincoln, Patton,
Eisner, Schwandt, ancllvlacDonald (which I discuss in my contribmion to
the first edition of this Handbao!?; Greene, 1994). It also misdirects ,-ltten-
rion toward holU an evaluation is done (qualitatively) instead of why (ro
be contextually responsive, to help improve the program, to promote
democratizing conversations, to ad va care for pluralism). Further, the
methodological domination of evaluative theory and practice falsely sim-
plifies the political complexities of the contexts in which evaluators work.
596
+ The Logic of Justification for
Evaluations Conducted Qualitatively
Although v;,ri<1tiollS exist in the philosophical underpinnings of different
qualitative approaches to evaluation, most share a core set of assumptions
and stances that are aptly labeled COl1stmctivist or illfel1Jrctiuisf. As
Schwandt (1994) notes:
Proponents of these persuasions share the goal of understanding the com-
plex world of lived expetience from the poim of view of those who live ir.
This gaul is variously spoken of as an abiding concern for the life world, for
the ernie point of view, for underst,mding mC<lning, for grasping the acror's
definition of a situation, for Verstchcll. The world of lived reality and situa-
tion-specific meanings that constitute the general object of investigation is
thought to be constructed by social acwrs. That is, particular ncwrs, in
particular places, at particular times, fashion meaning out of events and
phenomena through prolonged, complex processes of social inrer<1ctiOll in-
volving history, language, and action. (p. 118)
He goes on:
Constructivists are deeply committed w the ... view that whar we take to be
objective knowledge and truth is the result of perspective. Knowledge and
truth are created, not discovered by mind. They emphasize the pluralistic
and plastic character of reality-pluralistic in the sense that reality is ex-
pressible in a variety of symbol and language systems; plastic in the sense
that reality is stretched and shaped to fit purposeful acts of intentional hu-
l11<ln agents. (p. 125)
Constructivism
Constructivist inquirers seek to Iwdcrstand cOlltextlwlizcd meaning, to
understand the meaningfulness of human actions and interactions-as
experienced and construed by the actors-in a given contexr. This aim is
based on the aSsllmption that the social \vorld, as distinct frol11 the physical
world, does not exist independently, "out there," waiting to be discovered
by smart and technically expert social inquirers. Rather, the emotional,
linguistic, symbolic, interactive, political dimensions of the social worlcl-
and their meaningfulness, or lack thereof-are all constructed by agentic
human actors. These constructions are influenced by specific historical,
597
 -- --_._._ .. _... '" -~-'~ ... -"'. ... - .... --"' ..
~ _. _. __ .. _..
geopolitical, and cultural practices and discourses, and bv the inrcnriolls-
these consrruc~
quality and meaningfulness of understanding. Yet, unlike some of their
noble and orbenvise-of those doing the constructing. So more radical peers who disclaim the existcncc of any privileged proce-
rions <lfe multiple and plural, contingent and contextual.
And so rhe ba'ic t '1 [, '1' " I' I dures that wil! enhance the acceptability of an inquirer's interpretations
., ", (0 Socia mqull'V IS not to Olsenve, owinl pCOp""-
(c'5" n<~"onc, 199.2 j Srnith J 1989 J p. lGO), lIlV~~ ,",vn:;;tnl\;tivj:;;t cvahh1l0r"
Lie::; of Lilt: c.\ternal "vor/d, ro answer population representation questions,
favor some procedural structure for their work. This is partly because the
or to extract and conn en observed effects with causes. Rather, the first
task is to understand people's constructions of meanings in the COntext contexts of social program evaluation conrinuc to demancl assurances of
methodological quality and data integrity. Evaluations conducted qllalita~
being studied, because it is these constructions that constitute social reali-
tivcly can make little contribution to socia! policies and programs if they
ties clOd underlie aIJ human action. This task itself is an interpretive one,
r,equiring the inquirer to "elucidate the process of meaning consrruc- arc not perceived as crediblc-defensible, enlightcning, and useful-by <1t
least some evaluation users.
[Jon ... to construct a reading of these meanings; it is to offer the inquirer's
construction of the constructions of the actors one studies" (Schwandt, In this respect, evaluators arc particularly concern cd about criteria and
1994 11) 'fl ' h' '" mcthods [or warranting their evaluative knowledge claims as empirically
-, ,p. _0. lat IS, t e lIlterprenvc mqUircr cannot know the meanings
of another's life experience, bur only the inquirer's own inscriptions or based representations of progr8111 experiences ;md not: as biased inquirer
representations of said meanings (Lather, 1991). In these \vays, inrcr- opinions, Philosophically, this poscs a contradiction, that being "the
prerivist, constructivist inquiry is unapologetically subjectivist-the in~ attempt to provide <1 methodological foundation for knov,.-ledge based on
llonfoundational assumptions" (Smith, 1989, p. 159). The pwcrice sec-
quirer's worldview becomes part of the construction and represenration
tion that follows outlines some of the ways evalu;:ltors do this, for exam-
of meaning in any particular COntext. Inquirer bias, experience, expertise,
and insight arc all part of the meanings constructed and inscribed. ple, using time-honorcd procedures of triangulation. These responses to
the criteria problem generally follow Smith's (1989, 1990) lead in recast-
Values are therefore also intertwined with IOlO\ving, as knowing is
ing the concerns from ones of foundational methods to ones of heuristic
intertwined \vith being and acting. In constructivist, interpretive work,
procedures, appealing, for example, to procedural criteria as grounds for
therc are "no facts without values, and different values can actually lead [Q
judging the goodness of interpretations (Schwandt, 1994, p. 130). FOI~
diffcrenr facts" (Smith, 1989, p. 111). That is, different knowers'holding
many qualitative evaluators, nonetheless, the contradictory demands at
different ideals and values can construct different meanings, even in the
philosophy and practice, especially around criteria and procedures for
same situation. So constructivist, interpretive inquiry honors the value
warranting quality, remain strong.
dimensions of lived experience and human meaning, bur does not pre-
scribe or advance any particular set of values. Rather, the values of a par-
Beyond Constructivism
ticular constructivist inquiry become those held dear by the constructors
of meaning in that inquiry-the members of the setting srudied, the
Constructivism characterizes the philosophical logic justifying many of
inquirer, and the larger society. Constructivism is thereby value pluralistic.
the evaluations conducted qualitatively today. But constructivism itself is
-Methodologically, constructivism is most consonant \vith natural set-
not unchallenged, and some evaluators have engaged these challenges.
tings, with the human inquirer as the primary gatherer and interpreter of
Two brief examples here will illustrate these developments at the level of
meaning, with qualitative methods, with emergent inquiry designs, and
philosophy.J I \vilJ return to the topic of contemporary challenges and
\vith contextual, holistic understanding, in contrast to interventionist Dre-
future directions in the concluding section of the chapter.
diction and control, as the overall goal of inquiry (Guba & Lincoln, 1989;
Lincoln, 1990; Patton, 1990), First, for more than a decade, Thomas Schwandt (1989, 1996, 1997a,
1997c, 1998) has offered practical philosophy 3S an alternative W3Y of
Perhaps marc significant, constructivism suppOrts the decentering of
conceptualizing the practice and discourse of social science, including
inquiry/evaluation discourse from questions of method to questions of
evaluation, "Practical philosophy is concerned with the mode of activ-
purpose and role. The quality of technique becomes secondary to the
ity called the practical (praxis). Its subject matter is how an individual
598
599

conducts her O[ his life and affairs as a member of society" (Schwandt,
1998, p. 9). To practice evaluation practically means to shift: radically from
a methodological ro a political-ethical frame, and to be less concerned
(. I '
fJ IJout !,~errcct:l1lg ;lflL ,\varr;lnnng our L::no'\.vlcdge cbi,-ns (Schwandt, 1996;
and more concerned about helping practitioners to deliberate well and to
develop their own \vise practice.
Second, posrmodcm challenges exist to all contemporary forms of
evaluation practice (lVlabry, 1997). In much postmodern thought, for
example, claims to know are not just partial and contingent, but also fun-
damentally indeterminate. Further, not only is the author of a postmodern
represenwtion of kllO\ving also a major constructor of what is claimed to
be known, bur so is the reader. One response to these challenges is to aban-
don modernist principles and preoccupations in favor of actually practic-
ing evaluation postmodernly. Tincke Abma (1997a, 1997b, 1998) offers
inspirational examples of postmodern evaluation practice that accept
postmodern incredulity and doubt without giving way to nihilism and dis-
engagement. For example, Abm;:J (1997a) promotes in her work the ide;}
and experience of playfulness: "A playful person is nor too attached to his
or her personal persuasions and appreciates the power of rcdescribing, the
power of bnguage to make new andedifferent things possible and impor-
tant" (p. 44). Abma also invokes the "self-reflexive, polyvocal, ,md multi-
interpretable" (1998, p. 434) texts of postmodern writers in endeavoring
to craft her evaluation reports as "open, ambiguous, and unpredictable ...
without summary, conclusions and recommendations" (1997b, p. 106)
and thereby as invitations to dialogue (1998).
" The Practice of Evaluations Conducted Qualitatively
So br, I have described the unique contexts and character of social pro-
gr2.l11 evaluation, presented a typology of contemporary genres of eval-
uation theory and practice, and explicated the constructivist philosophy
underlying most "qualitative" evaluation approaches. In turning now to
practice, I need to acknowledge that in the field, evaluators rarely practice
a "pure" form of their craft, either philosophically or methodologically.
The complex, pluralistic demands of evaluation field contexts evoke in-
stead multiple, diverse frames for guiding practice and invite dialogue
among them. In fact, evaluation expcrtise today is marked by its dialectical,
dialogical temperament, its openness to multiple forms and layers of
600
..... " U .... ' ... L'-U' .... "'=' .......'.... 'u,' , ..... ~'u"'" ""'-'.... ~" ........ ' ....... "u.,
understanding, and its responsiveness to contextual needs for understand-
ing, rarher than its adherence to any singular philosophy or approach.
I offer here an example of evaluation practiced qualitatively that illus-
trales commun features of thb genre. After lligl1ligiHing these features, I
will discuss the continuing chaHenges of this approach to evaluation.
An Example
Envision an evaluation of a state-level welfare reform initiative, a
reform that aims to shift three-quarters of the people currently on the
state's welfare rolls to sustainable jobs over the next 5 years. This reform is
based on the premise that long-term economic self-sufficiency for many of
the people currently dependent on welfare requires an approach thar firsr
concentrates on "developing human capital" and then shifts to v.;ork
placements. The devclopmem phase of this program includes intensive,
individualized education, personal development, job training, and job
search activities.
To address evaluative questions abollt the quality and effectiveness of
this welfare reform for both policy and local program audiences, the quali-
tativelv minded evaluator would seek primarily to understand how the
progr;m is experienced by individual participants-welfare clients, pro-
gram staff, educational and economic development leaders-in particular
contexts, for it is in these contextualized experiences that the meanings of
program quality and effectiveness are shaped and molded. The qualitative
evaluator of this welfare reform would likely purposefully sample for
intensive dara gathering a small number of program communities and par-
ticipants to obtain heterogeneous samples on relevant characteristics (for
example, demographics, economic and employmcnt histories, cultural
norms) in order to capture a broad range of diversity of experiences and
meanings expected. For this sample, then, the evaluator would likely con-
duct extended on-site observations in order to develop a descriptive por-
trayal of what happens in the new welfare program and how this portrayal
va;ies across contexts. How is the concept of "human capital" envisioned
and implemented in these varied program sites? \Vhat: is a typical program
experience like for different types of welfare cliems? How do community
characteristics shape the definitions of long-term employlllent variously
adopted by the program sites?
The evaluator would likely sample 10-15 welfare clients in each site for
individual interviews and reviews of their program records in order to
601
 understand Wh~H prognm ~
P'l't' -'
,,! JClpants tmc me
-. I . rf I .
a~llI1g U ,or not, In the pro-
L

gram and how this meaning' . ,11.- d . rion would access and analyze macro-level information-most probably in
IS Ie cete In th-Arr 11 '''' .
evaluator would also lil--,j" I t : : .1ogra,n,l Journeys. The quantitative form-on relevant welfare dependence, economic, <111d job
. "C} saJnp e some prJ ff
some ecoIlomi I -1 ~ (gram Sfa and managers market indicators and \vould assess program costs against outcomes.
, , c (eye opmcl1llcaders ~nd Ml I" _ , '
'H o,;LJ,ch :;,He fur i!1cliviC]U31 0 _. .' . - - l1E' Ley CUrrlTTItll'lry ",,,,,,,h e r ,, 'Th""c cvalu,-'l;ve dar_a ,vould rl,en l-,c ~voven ;n,O;"'ln oVi.r",ll portr'-ly",1 of
" r gl Oup lntcrVICV,/S re -. r J -I -' .
o f and experiences with the, If gMc lI1g t lell' perceptions program quality and effectiveness as grounded in insider experiences and
. - lew we 'are progr \Vl . I
Views regarding the human' 1 I . am. Jar IS t 1e range of meanings and as bounded by the opportunities and constraints of the
a!1C t 1C eCOIlOlTIJc dey 1 I
t I1C new progr;:l1n ~ \X1hat . e armcnr emp lases of broader marketplace. This portrayal would be offered as a holistic repre-
L agreemems and dls'{ _
< '. .
communitv about the hnds I ~ . < grcemems eXist Wlthm the sentation of those studied, with possible but not assured extensions and
. \. 0 economic self-suff' ',I' ,
the new progxram) Altern,' I I . ICI,e,n q 1elng fostered in relevance to those not studied. This portrayal would be filled with snap-
. . a Ive )', t lese pro . "f[ I '
ceptlOlls might be mol' ""fE I . gram sta ane commUlllt)' per- shots of program activities ;:md vignettes about accomplishl11enr and dis-
c c ectlve v ;:md repre .' I I
structured quantl't'lt,',I,'" 1 1 ' " 1 ' 1 ' . sentatlve y gar lcred with appointment, fulfillment and disengagement, This portrayal would thus
'< ... surveys.
<l

The evaluator would then iIl['e' I ' offer a contextualized, complex, dynamic, value-laden, bmnecessarily
grate t lese multlpl ~. -I' 'd I '
a 1Jour program eng"" d' e 111llVI lla srones partial statement about \velfare reform. Although holistic, hermeneutic
< bement an experience i '
These community narrative' I' 1 m,o communIty narratives. understanding is the goal of evaluation practice qualitatively, human phe-
' . L S \Vou ( conve'll t1 . I I, '
nons of the conrextualized ,.' . I . ~e ,e\ a lIdtor s ,reprcsenra- nomena in their sociopolitical contexts remain extraordinarily complex.
., , ,expenenna meal1l r '1 -II '
Inltlatlve in each localit'll studied 0 . ng or t l~ we are rdorm All representations of interpretive understanding are thus necessarily
might highlight the dI'Sj:uIlcrilre b' . ne COnI1l11~1I11tY.I,larra[]Ve,for examplc, partiaL
. .. enveen t lC" . t .' I
rhetoric of the w lr ' I pIO eSSlOlla advancement"
e -Ire re orm program and the . . _ ' I' ,
lo\v~skjlleclsetof d ' If', plograms 1111ltcd,mostly Evaluation as the Telling of Stories
e UcatlOna 0 fenngs A ~I . I I
plemenrarity of the program's' . . . " I no~ ler ITIlg 11:, lighlight the com-
.... . actiVity sc ledule w'tl tl d 'I I
practices of a s' 'f' .' I 1 le ar y r lythms and As highlighted by this example, evaluation practiced qualitatively is a
'. < Iglll ICant proPOrtIOn, a!thou- h i ' I] . _
t!cIpanrs in that setting Yet'lnorl ' h g lOt a ,of the program par- narrative craft: that involves the telling of stories-stories about individu-
ler mIg t centralJ r f
<
tlClpants' experiences of their ~ } eature program par- als and groups of people in their own complex and dynamic communities,
program as oppressiv dd I ' , stories that enable understanding of what these communities share with
and Core program staff \\'110 are I' c , e an e llimanrzmg,
JItter i V resewf I b I'
~ u a out t lelr recent trans-
... ' < -
fer to this "the \V . fl' . others and what is unique to them, stories with an explicit authorial signa-
, orst 0 t lC neIghborhood welf' fe ",
narrative5, that is would IlolI'srI' II d 'b, are o. Ices. Community ture (Barone, 1992), and stories with the aim of understanding, and often
. ' . ca y escn e '1:l J. _ .
meamngful connecrions 'md d' . : III ana yze Ol explarn the action, as the improvement of practice or the reframing of the policy con-
< lSJuncrures of thIS w If f ....
\v!thl11 each of its multi I - e dre re orm 111ltlatlve versation, toward the appreciation of pluralism and complexity.
. . p e COntexts. Beyond d '. .
polmcaJ policy and action im. . . ' _, escnptlon anel analySIS, As a narrative craft:, this approach to evaluation heavily favors qualita-
1994) would in I, 'I ' I erpretatlOns of these narratives (\'V'olcorr tive methods, methods that require direct engagement with members of
vo \.e t le va ues and l r ~f f' d' , '
stakeholders. 1e It: s 0 In IVldllal evaluators and the settings being studied and that gather information about their experi-
Furthermore, these COlTImunit .. ences in their O\"/n words. Such methods uniquely enable understanding of
level events and inreraetions bllt~ n arr at:1\rcs arc shaped not just by miero- insider experiences and perspectives. On~site observations and personal
' d lso b y macro-level' . I' ,
an d especially job marl-et fa -r I' economIC, po mcal, inrervie\vs are thus the mainstays of evaluation practiced qualitatively,
\. . C ors, orces 'l11d trend TI '
gain more meaning if placed 'I' I' ~ ,. s.. 1e narratiVes would supported by revie\"ls of relevant documents and records. However, most
Wlt: l1n r lelr large' d'
nomic and political contexr FLlrrl . r state an reglOnal ceo- evaluations also require information about a large, representative sample
, , " ler acosr-ben f' l' f I '
Imtiative is in all probal'l' ' - e It ana YSIS 0 t lIS reform of program participants or information about macro social-politieal-
. ' < )1 Hy, an expected < fl l ' .
addition to the intensive study of I ..,eJ parr 0 t.le eva uanOn. So, ll1 economic factors and trends or cost-benefit information, as illustrated by
. se ecte: communIty cases, this evalua- the welfare reform example above. For these kinds of purposes, qIlilntita-
602
603
 hue methods are clearly acceptable within qualitative evaluation. In facr,
in all genres of evaluation practice, evaluators are likely to attain more
~~Jmprehensiveand in-depth program understanding by explicitly inviting
olalogue around data from different methods and aDorn,che, {C,oeno &
Ccu-,-,,,,,,,lli, ~'7/1"1,
Good storytellers carefully select and richly portray their settings and
make the actions of their protagonists explicable within these settings. In
like fashion, qualitative evaluations are framed by the careful selection of
one or more cases to study and by the rich, multilayered descriptions of the
contexts of those cases. The \\lelfare reform example is framed bv the
selection of several communities as instrumental cases (Stake, 1995)-
cases selected not for their intrinsic interest but for their potenti31 to pro-
vide insight into the overall reform initiative. Understanding the interper-
sonal, civic, socioeconomic, cultural rhythms of these communities is
essential, 3S constructivist mcaning is greatly embedded in COntext. \1(1hat
community characteristics, for example, might influence a program par-
ticipant's sense of hopefulness or despair, connectedness or isobtion,
and then hoy\' are these feelings related to her or his program experi-
ence? Locating these instrumental cases within the larger state economic
context-through more quantitative methods and indicators-is also a
critical part of me cll1ing making and understanding the overall reform
endeavor.
As a narrative craft, evaluation practiced qualitatively is highly depen-
dent on the narrator-evaluator. Different evaluators will tell different
stories, and more experienced, knowledgeable, sensitive, and insightful
evaluators can usually tell more meaningful stories. An evaluator's Dum
positioll in his or her work reflects both personal biography and political
partisanship, both who the evaluator is, as an individual and an evaluator,
and \vhose interests he or she chooses to advance in the study (Greene,
] 996, p. 285). \Xlith a constructivist worldviev.'!, most evaluators in this
genre are partial to the interests of those closest to the program being stud-
ied, namely, program staff and participants. These are the voices and per-
spectives most prominently featured in the welfare reform example above.
Bur, with a constructivist \vorldview, most evaluators in this genre are 3lso
committed to pluralism and thus to representing the full array of perspec-
tives and meanings in their stories-for a comprehensive story, all are also
needed.
Finally, evaluators must define and pOSition not just themselves in their
work, but their work in the vi/orld (Greene, 1996, 1998b). Evaluators must
604
be accountable for the differences their stories make. \"\7ith their con-
structivist worldview, evaluators in this genre most comfortably position
their stories as guides for the improvement of specific contextual practices
\Pc,LL""'l ] 99717 Cl:> '-'f'f'0nuuiLk" [V1- prVbri.llll !c:~nlin!5 dud in~_q:,I1L .LJy
diverse interested stakeholders (Stake, "1997), or as vehicles for retratlllng
the larger policy conversation (\X!eiss, 1998). Perhaps less comfor.tably,
constructivist and imerpretivist evaluators can position their stones as
insistent demands to attend to the vital complexities, the legitimate plural-
ism, and the politics of power that constitute tbe fabric of contempo~~ary
social life in the United States. To have a voice in the fmure direction 01 the
social practices and policies they study, evaluators in this genre must offer
not onlv convincing stories, but challenging ones, and they need nor only
to tel! ;hcir stoties bur also actively and assertively to claim Sp;KC on the
social agenda for them.
The Challenges of Evaluation as Storytelling
But how do those who read and listen to the evaluator's story about the
meaningfulncss of this state-level welfare reform decide if it is a good
storv? How do they know the evaluator is not just advocating for her own
vie\~poinr about welfare reform in the telling of this story? J-J.0:v does the
evaluator know where the boundaries of her O\vn advocacy he:' And how
does she as a "qualitative" evaluator fairly and justly fulfill her responsibil-
ities-to those who have commissioned the evaluation, to the members of
the settings studied, and to the larger citizenry, the collective good: \Xlhos,e
interests are advanced, for example, when the evaluator judges thlS state 5
approach to welfare reform to be an effective one, bur welfme reform
itself to be a misguided policy direction?
These questions reveal the essential value dimensions of all social pr?-
gram evaluation, across all genres. They conrest the criteria to be used.ll1
judging the quality of the inquiry (How good is this story?) and the quailty
of the program evaluated (How good is this program?). They ch:l11er:.ge
the presence of the evaluator in the story (Are you just ad~ocat1l1g ,tor
vour own viewpoint?) and they query the very pllfpose at cvalual:1on
(Whose interests are being advanced?). As noted, these questiol~s arc ger-
mane to all forms of evaluation. They are more visible in evaluatIons prac-
ticed qualitatively because this evaluation genre explicitly acknowledg~s
the value strands of knowledge claims. And they are more problematlc
because this evaluation genre is philosophically nonfoulldationaI and
605
 V"U"""'.UP'U"'':J -'''''-IU', 1"''d'U'''''' ""'--'~'':;;f'' ' - . v , ...... "'"',,

pr3ctically v3lue pluralistic. The constructivist juStification fur this genre

provides no ready answers to these challenges, which instead must be .
,In 0
utreached hand. Bur even such concerted , ~ efforts \vill fan shorr, and
addressed by each evaluator in each COntext. I . listic inclusiveness is rarely fully attamCt . ".
In the welfare refQrm evaluarion eX~Il1I11~ "60ve, <I", evoI"""H mc~,.. el,
p ura .
Moreover evaluators 1Il. r1'lIS genre I-now
\. that thell' claIms
t1l'-l.t oo..::t
, , to .un-
P1T\--ilo..::So..::d
d
~Hl
J1iat
lJlc inquiry crjrerl~, for example, some of rhe rfUsrvvorthiness
1d. , , _ _
~t-C e'.Hl:>tinltivc vi the pet:>pcetn_ _ .c:> do I
un '-<::<,;Q5
dc;l-~jlan ... d d . I Tl e welfarc reform eva nator,
criteria of Lincoln and Gliba (1985). These criteria (credibility, applicabil- thereby unavoidably intcrcstc an p~lrtJa. 1 .1"' ntex. m'il mean-
ity, dependability, and confirmability) are constructcd to parallel COll\'en- ' . Id"
with her constnlct!vlst war Vle\V, IS 1,-
II-ely to empf13SIZC co qual'lt)l _
In like
<

-. . ' J ' I ents a program '- _

tiona I inquiry criteria (internal validity, external validity, reliability, and
ingfl\Incs~ over ~OS~;;~C1~~~~;:t:~:Z:~:I~:deals p rivilege the interests of
neutrality, respectively) and thus can be recognized as familiar and under- manner, I atton s ( ,
' 1" makers Cuba an
d L'
mco n s
I ' (1989)
"-
founh~general1on
-"rood as legitimate by critical evaluation audiences, site-level program on-sIte c eClSlOn , .. 1'1 a hy with an empowerment
staff, and state-level program decision makers alike:l Procedural guides evaluation integrates a constructiVIst p 11 os P , -f I. d St;::J1zc's
for fulfilling these criteria 3rc also familiar, as they rely heavily on method- . j., I g' ~o champion t I1e 1l1tereSts
. 0 f the . least
. powel u, an I. ,
IUCO a } l. .... . '_ rinci les cngender sympat lIes
ological argumenrs and techniques-sampling for diversity, triangulating (1997) constructivist beliefs a~d ref~r~~,;t p d Peaningfullv vvith reform.
for agreement, and monitoring bias_ lvLm}' importanr evaluation clients for those who are also struggllllg mIg 1tI yan ill "b" tors of this I.
need familiar or recognizable evaluation constructs like these in order to In shan, while soundly
. 1-
dis~labiImiIng ~dvocoafCYaIlascv~~:1t~~~~'1~ld
dema c canmgs , '-,
embrace

;I~:~: l~~~rg:~~~f~~:a~~e with their own pI;iIO:~~~,tt~(;ll~~O~;~~;;~n~~~

accept un familiar ways of knOWing and thinking. Although not as respect-
ful of the nature of constructivist thought as Some other proposed lists of
own leanings underscore the Importance co . . ~
inquiry criteria-for example, Harry \X/olcott's (1990) understanding,
Guba ilnd Lincoln's (989) authenticity, and action researchers' war- critical markers of social meaning and actIOn.
mllted actioll (Greenwood & Levin, 1998)-these trustworthiness Con-
structs arc pragmatic choices for evaluators concerned about the accept-
ability and usefulness of their stories, especially for "modernist cliems" i> The Journey of
(Raben Stake, personal communication, lvlay 6, 1999). Evaluations Conducted Qualitatively
Bur still, usefulness for what purposes, for vl-'hose aims? \X!hercas quali-
tative inquirers in other domains can uninhibitedly experiment with vari- Beginnings
ous political and action agendas for their work, social program evaluators
.. . -'1 .s to rogram evaluation emerged
are more constrained, both by tradition and by context. Advocacy is a Constructivist, quahtatlve apPl~ac:e P970 (S "1- 1967 1975), in
daunting specter in many evaluation circles, challenging as it does the tra- . S 1 . commul1lty III thc 1 s ta '-c, ,
withm the U.. eva uatlon . . , t Ilectual thought and societal
ditional Stance of the disinrerested, distanced, neutral evaluator whose "1 "nificantevol11tlOnsm In e . .
tandem Wll.l [\Ova Slg I' 1- 1991) 5 The first (r)evolutlon
only job is [() find out the "truth" about the effectiveness of the social pro- ' f (C ] 1985'Greene&McCmroc" .
bebe I-s d 00h (, . not , . - .
experlmenta I '
sCicnce a s the paradigm~
for social pro-
,
gram at hand (Greene, 1997). Bur advocacy as the promotion of some was t 1e -et ron1l1
b
I. d . I . !lces) a dethroning resultlng
interests over others is unavoidable in contemporary social program eval- I . ( d othcr app Ie socIa SCle , I
gram eva llatlon an . . . " " 't hilosophy of science that hac
uation. There are simply too many stakeholders with too many varied from major fractll~e.s In th~ PO~ltl\ IS : Enli' htenmem era of the 18th
interests for any single evaluation to address all of their concerns fairly and 'd d most scientifiC practice Since th g d d'
gUI C y 6 , . J delptes and discussions that ensue regar mg
justly. In this regard, evaluarion practiced qualitativelY-characteristically
more than other genres-strives [Q reach pluralistic ideals ...vhile clearly
cenrur :. It ...vas ~nll~ t lC alis~ th~t
interpretivist philosophies and quali-
alternal.lVes to experiment. - . d' Initially contested on both
acknowledging its partialities. Pluralistic ideals imply inclusiveness of per- I mered evaluatIve Iscourse. I
tative approac
d 1 . les I ed pract!ca. I groUll d s (as" expressed in evaluation's qua i-
0-~
spective and voice. That is, the views of both the usually vocal and the USll- .
metho ooglca an < I, Reichardt, 1979; Reichardt &
ally silent stakeholders are sought, the latter often \vith an extra stretch of iative-quantitalive debate, e.g., Coo \. . . . t 01- strong root in
Rallis, 1994), qualitative approaches to evaluatlon 0 '-
606
607
 -'---"-"':J --_._. , ... _-:J .. _._._--,_ .
'-:J' - ' . ' -

the rhinking of many evaluation rheorists and methodologists. Largely

Evaluating Program Theories
through their sensible and appealing practice, such approaches have now
become an accepted alternative throughout the evaluation community.
.
Early IJrogram evaluanons , t Ile G reat S0 CIe t)/ era of the _United States
10 _
The second major ,cr of changes that CI1Cou raged and en" hlerl ,Ik eme c_
2,""c.<'o and I:5l~VWUl oj ljUaliLHive nppJoaches to evaluiltiun involved
"vere styled c-.['-e~ .
e'""l-'<=r>n,cn.ta
1 t
c~"
-ov of -~ ot
f"051""'''' CI~co'_'cJ1
,.'d
1'0 ram out-
the h)'f-.othc-

sized causal links between program inputs and expecte P f g <

changes In the belief systems of the larger U.S. society during the 1960s
comes Ev~llIations of that era aspired to measure outcon~esor
Plrogram
and 19705. Thomas Cook (1985) describes these changes as follows:
P artici~:.:m:s and to compare these to measured outcomes. or slim ~r peo-
L The decline in aurhority accorded standard social science theory following IJle who did not participate ,In t Ile programs.. ' Little attention was . givend S[Q
the perceived failures of the Great Society (see also HOllse, 199.3); g athering data on the actua I programs, as lill ' plemented
. or d"
experience
dOd I . ~ 0_
he rocesses bv which observe d program effects_ happene or " I not lap
The decline in <!urhority accorded political figures following the Vietnam t en Pwere unknown-r . 'IlUS t,Ile 1ale
1 I 0 f "black box" evaluations f Il for thest:
'1
d
\\lar, rhe \'Vatetgate scandal, ,md other government debacles; and,
~arlY studies, signaling the disfavor into which they gene;"IIY Cd ~:\sr:~~~
.3. An increase in the cultural, value, and polirical pluralism or socicty, as mani- alone studies. As Lee Cronbach and ASSOCiates (198.0) olserv e ' ~ g the
fest, for cxample, in the civil rights and women's movements. e\/alu3tive question ' . " a dOff
lllvltes I, e renriated answer rnstead 0 f leavlllg
f I' ...
program plan, the delive.r ,of the program,.an.~,le
< ,
y , d 1 esponse 0 C Jents as r.5 ' .
unexamined elements wlthm a closed black box (~.)' " . ( . bl
There \vas considerable consonance berween lhcse shifts in societal think- S . . tl early' 1980s a number of evaluation theOrists not.1 y
ing and rhe emergence of alternative paradigms, including intcrpretivism rartmg III le " ~ '983 and C-arol \X!eiss
L n' -I' an 1987; Huey Chen & Peter ROSSI, 1
l -, ,
;, , ,,:
and constructivism, for understanding and doing social science.
19en-;')lCl,e\.ill
o :1n ,
to.reargue
-
for the .Importance 0 f" rIlC'ory-driven evaluation ,. " I
and -evaluation
. be
directed at undcrsrandlllg .
an d asscsslllg
, . a pr ognlln ~
s t le-

Present-Day Connections to Dry of change," From Carol Weiss (1998):

Important Evaluation Theories
The [program] theory gives guidance to t Ile eva Iua tor ,bout
, where to look
and what to find out. (p. 65)
In its contemporary form, social program evaluation is a young field,
JUSt beginning ro develop and refine its own theories (Shadish, Cook, & . of a [progr<lm ] t Ilcory can .expose nai've
Thc mere construcuon ", lilTand simplis-
before the
Leviton, 1991). Among the important theoretical developments likely to . expectatlons.
[lC : . . . The . . . theory. can be a Icmnmg roo OJ g
continue to shape evaluative thinking and practice over the coming decade evaluation begins. (po 66)
are two sets of ideas: (a) the reclamation of theories of progra11l action and
change as significant conceptual frameworks for evaluation, and (b) the \' ,iI the' theory
Once the data are in hand, [the ,:H1<llysis assesses ] Il~Wf\l'. f nhe de-
di-
redefinition of evaluation's primary role as an opportunity for aetilJe, gell- II I I
scribes what ncrua y lappenec ... - This, is important
. I morl11<luon 0
I , can rethink the
eratr'lJe, and inc!usilJe learning, a redefinition that emphasizes the process rectors and staff of the program u~d.c~ study, s~ t lar.t ~~rk our as ;lntici-
of evaluation over its products. \\fhat these developments in evaluation unders[',mdings and replan the ;lctJvltles that dId not \
pared, (p, 66)
theory share is a recenrering of evaluation around its sociopolitical role,
instead of i[$ methods and techniques, and a valuing of the educarive and
, [a Iso] p
Theory.based evaluatlon ovides
r" cxp!<lnations, storicsbl"of means
(_ 6R)and
action porenrial of evaluation. They also derive strength from evaluation
ends, thM communicate readily to policymakcfs and the pu Ie. p. l
practiced qualitatively, through significant but not exclusive reliance on
interpretive methods, and through SUPpOrt for, if not complete allegiance
t
Another key argument for theory-lase d eva IUeation
. is. 1
its. potential to
_ t' ns
to, a constructivist \Yorldvie\v. I describe each development briefly below.
contribute to 'generalizable knowledge about how sOCIa lnterven 10
608
609

work and the condirions and facrors rhar enable and obstruct their success
(Cronbach, ] 982; \1{!eiss, 1998). Evaluation practiced qualitatively has
much to offer this goal of social problem solving. Holistic program por-
nay,]I, elll iDm'"rc pro~LUTI conHrucr, inro contextlJ~lized lived experi-
ences vvithin diverse community values. Evaluators' program portrayals
can complement regression coefficients, say, for net earnings increases of
job training program participants, with stories of human pride and shame,
accomplishment and failure. Program portrayals can also challenge
the generation of effect sizes, which homogenize and centralize program
parricipation, with the dissonance and complexity of human diversity.
Indeed, this increasingly urgent dialogue about social problems and
how best to solve them requires the dialectical participation of all evalua-
tion perspectives, and a critical if not leadership role for the insistence
on compIcxity, comextuality, and human agency offered by qualitative
approaches to evaluation.
Evaluating With Stakeholder Engagement
Offering some counterpoint to the reclamation of program theory as an
important evaluation agenda is the expanding development and applica-
tion of participatory and collaborative approaches to evaluation (Ryan et
al., 1998; \X!hitmore, 1998). These approaches emphasize the active
engagement of stakeholders in the evaluation process for purposes of
enhancing ownership and thus usefulness of the evaluation results
(Cousins & Earl, 1995; Patton, 1997) or for purposes of promoting some
form of democratizing social change, such as social justice (I-louse, 1990)
or empowerment (Fetterman, 1994).7 In these (admittedly diverse) ways,
participatory evaluators frame evaluation primarily as an opportunity for
engagement, learning, and action in that context. And the process of con-
ducting the evaluation-the ways in which stakeholders are involved,
vvbich particular stakeholders participate, how less-powerful voices can
be fairly heard, who speaks for and with whom-becomes of central
importance, not the issue of \vhat methods are used or even what substan-
tive results arc obtained.
In ongoing developments, utilization-oricnted participatory evaluators
are advancing the significant role evaluation can play in orgallizational
leaming, forging connections to contemporary emphases on strategic
planning and quality management within organizational developmcnt cir-
cles (Cousins & Earl, 1995; Patton, 1994; Preskill & Torres, 1998). And
610
understanding ::'ociol programs J hrough tvoluation
social action-oriented participatory evaluators are advancirg the signifi-
cance of evaluation's potenrial for broadening and deepening our deliber-
ations and dialogues about important social iggucs (Greene, '1 997; H ou~e
&: I-Io...vc 1999; Ryan et ell., 199R). In nil dimensions of tbese ongoing
advances'in parrici patory evaluation, qualitative evaluation is playing and
will continue to playa central, although again not exclLlsiv,~, role. Inclu-
sion, not cxcl usivity, is the defini ng character of contemporary dialectical,
dialogical evaluation.
<> Future Directions
As vividly illustrated throughout this volume, ways of rhinking about and
doing "qualitative research" continue to evolve and change. Social pro-
gram evaluators, like other applied social scientists, arc cl~a,lcnged by ,the
ongoing evolutions in the philosophy and ideology of SCience. Here IS a
sampling.!!
Challenges to the Very Nature of Our Qualitative Data
The [qualitative] interview is fundamentally indeterminate. The com~lex
play of conscious and unconscious thoughts, feelings, fears, powers, deSires,
and needs on the part of both the interviewer and interviev'F~e cannot be
captured and categorized. . . . \Xlhen we think we "imerpr'~t" what ~he
meaning or me<lIlings of nn interview are, through various (bra reductIon
techniques, we are overlaying indeterminacy with the determinacies of our
meaning-making, replacing ambiguities with [our] findings or construc-
tions. \X1hen we proceed as if we have "found" or "construcced" the best, or
the key, or the most important interprewtion, we are misportraying what
has occurred.... [Instead in the analysis] the researcher fills ... [the inter-
view's] indeterminate openness with her or his interpretive baggage; im-
poses names, categories, constructions, conceptual schemes, theories upon
the unknowable; and believes that the indeterminate is nOW located, con-
structed, known. Order h;JS been created. The restless, approJriativc spirit
of the researcher is (tcmporarily) at peace. (Scheurich, 1995, p. 249)
Challenges to Our Representations as Meanings
Postmodernism opens space for new forms of representing social science in-
quiry by challenging the assumptions of what arc seen as accepted forms of
611
 ' _ " , _., ' .... ...,n' ' ' - ' ' ' , n ' H ..'
prcscming the findings of inquiry. PJget ... poims out, "there is something
odd ,thout privileging an analysis of discourse in its least robust form, a writ-
ten text, exploring it' in great de!:"il while ignoring the speakers' miens and
intention,,_ n.
'" 1n t'he creat'ion of new representations of inquiry, we need ro struggle
to represent the complexities and indetermin<lcies of particip<lllts' experi-
ences ... [and] to acknowledge our role in the construCt'ion of the represen-
t',ltion, our voice in dle presenrmion. [Further] as, in postmodern terms,
knowledge is p<lrtial, conditional ~llld cont'exrual, so arc represclH8.tions.
(Goodyear, 1997, pp. 64-65, 69)
Challenges Regarding the Political Ethics of Our Work
How do we Iwndle "hot" inform:uion, especially in times when poor and
working-class women ,md men are being demonized by the Right and by
Congress? .. For instance, what: do we do with information abour the ways
in which women on welfare virtually have ro become we!fJre cheats to sur-
vive? ("S1Ire he comes alice II mOllth i-lIld gives l1Ie sOllle 11I01/cy. [nul}' haul' to
ti/kea beatiJlg, b1lt the kids /feed the 1I101/CY.") A few [of those we study] use
more drugs than we wish to know ... some underartend to their children
well beyond neglect. ... To ignore these data is t'o deny tbe effects [of hard
economic times], To report the d~1t;] is to risk their likely misinterpretation.
1n a moment in histOry when there are few audiences willing to reflect on
the complex social roots of communit'y and domestic violence and t'he im-
possibility of sale reliance on '.ve!fare, or even to appreciate the complexity,
love,hnpe, and p<lin rh<lt fills the poor and working class, how do we display
the voyeuristic dirty laundry that liners our (brabase? At the same rime,
how can we risk romanticizing or denying the devastating impact of the cur-
rent assault all pnor and working-class families launched by the Stare, rhe
economy, neighbors, and sometimes kill? (Fine & \"X!eis, 1996, pp.158-159)
These challenges to the very core of what it is \ve do when we practice
the craft of social program evaluation with a constructivist, qualitative
vvorldview arc dauming but not defeating. They continue the erosion, par~
tially initiated by constructivism, of evaluation's preoccupation with
methodological rigor and technical expertise. They affirm the construc-
tivist evaluator's emphasis on the meaningfulness of Jived experience,
even as they unravel our confidence in our ability to know, understand,
and interpret such experience. They also substantiate constructivist eval-
uators' self-consciousness about values, about our own position in our
\vork, and aboLlt tbe location of our work in the world.
612
'U.rI'.LJL"lll-'.' 'VI'I
utIUt:r.sLUrlUiny .:10001 f"'rograms I nrougn eVOIuonon
As socia! program evaluators, we have responsibilities to multiple audi-
ences, including the powerful policy makers, the al1-but-pO\veriess poor
people who are often tbe intended beneficiaries of tbe programs Vo/e eval-
u;lte, anel tile citizenry ::It lJrgc. Our work, therefore, must respectfully
balance social scientific theories of knowledge construction, interpreta-
tion, and representation with the political realities of social policy mak-
ing. Such balance is attained not through partisanship, but through
explicit commitment to inclusiveness, to pluralism, to ensuring thar ail
stakeholder voices are part of the conversation (Datta, 1999; House &
Howe, 1998). Inclusiveness respects both constructivist holism and demo-
cratic dialogue. \'(/e must acknowledge tbat making such a commitment
explicit is often viewed as advocacy or partisanship, and so our challenge
becomes one of creating impartial, nonconfronrational expressions of this
commitmem (Datt;], 1999).
The need is not: to find the 01lC correct way to do program evaluation,
but r3ther to conduct our craft so that it: responds to multiple audiences,
includes multiple perspectivcs, constructs multiple and diverse under-
standings and equally diverse rcprescmations of those und,~rstandings,
and encouragcs dialogue and conversations. I believe that the construc-
tivist, qualitative genre of eV8lu8tion-with its valuing of responsive-
ness and pluralism-is extremely well positioned to be an active player in
this dialogical evaluation evolut:ion. It is now up to us, each of us as
constructivist evaluators, to claim a seat at the rable.
<I> Notes
1. Social program evaluators work at all levels of government and t11ll~ social pro-
gramming, fro III positions both imide <lnd external to governmenr 3genci,;s. Their work
is frequently m31ldared by program fllnders~l1amely, governmcnt ;1I1d fOllndations~
tr3dition;1lly far knowledge, developmcnt, or accountability purposes (Chdirnsky, 1997),
Practicing evaluators have been trained in applied social science graduate progr;rms or in
policy ;rnalysis and evalu;nian gr;:ldllare programs. In addition to social programs, evalua-
tiolls arc conducted on objects (proclllCl evaluation), on people (personnel e'13Iu;lrion), 3 m l
on ptogr,lllls in the private sector such as executive development ;rnd human resource trrrin-
ing. Although Scriven (I9\J5) and others argue that the logic of evalurrtioll is the same :rcrass
these eli fferem objeers of evaluation, I be lieve tbey constinr te essentially difft'rent tasks, nod
so require qualitJtively rrnd politicrrlly di fferem responses, In this chapter, I f(,cus on qun! ita-
[iv;: evalu3tion nf social programs and the policies they embody, predomina;-nly in the U.S.
public domain. iVly ;Jrglllllcnts here may nor transfer well t'O mher evaluation contests; this
is a juclgmcnr for [he reader,
613
I
.1.
 ,._ ~. H_"
" L . .. ,ro., ,....",
.., I am indel)[l'd to Alyce Sporred Be:!r for the idea of this cvalll:!rion comen.
~l. Thl'se two examples :lre drawn from Greene (1998,1).
"I. From Lincoln and Cub:l (J 985); Evall1:ltion findings wi!! 1](' (a) credible when conso-
nant with contextll<llized lived experiCllcr, (b) applicable 1'0 orlJer simihr COIlII!.\'fg wht"n' ~-o
)""i;;"". b l ' '\,u~,-, cluinB tl,,;- '-'ppl}ing, (l) ile/lCllItt/IJle wIlen rile merhods decisions l1l:Jde 3re
ddell5ible :!nd reasonable for th:Jt contexr, ;:Ind (d) confirmable when inferences can be
traced bJck through analyses to d,1t3 actually coUecred.
5. See l1ly chapter in lhe firsr edition of this Handbool:. for an elaborated discussion of
rhe history of qualitative program eVilluatioll (Greene, 1994).
(J. Amol: g lhe numerous aCCOlHHS of this revolution in the philosophy of science, rhe
works of philosopher Richard Bernstein (1983, 1992) arc among the mosr ;Iccessihle.
7., !h(:se [\.1'0 clusters of purposes for parriciparory evaluarion rdlccr irs dual origins in
(al U.t1IIZiHlOll (Panon, !997) and stakeholder-based (Cold, 1983) evaluation, and (b) from
ulHslde rhe evalu:1t!on field, liberatory lraditions of participarory and p;lrrlciparorv action
rest';'Irch (Frl'lre, 19 7 2; Greenwood L(,,: Levin, 1998). See Whirmore (199B) for cla'borared
(!lscllsslOns of participarory e\'aluarion's histories, rheories. and practices.
I). These examples are drawn from Greene (1998a).
\j>- References
Abnla, 1. A. (J 997<l). Playing wirh/in plurality. Evaluation, 3, 1.5-48.
Abma, T
A. (1997b). Sharing pmver, facing ambiguity. In L.lvbbry (Ed.), Emll/I/-
tlOlI .uhf thc postmodem dilemma (pp. 105-119). Greenwich, CT: .fAJ.
Abma, T. A. (1998). Text in an evaluative cOn(ext: Writing for dialogue. Eva!lwlion,
4, 434-454.
Barone,1. E. (1992). On the demise of 5ubjecrivity in educational inquiry. Clfl"rieu-
IUI/1Ini]ltiry, 12, 25-38.
Bernsrein, R. J. (1983). Bcyond objectiuism and relatiuism. Philadelphia: University
of Pennsylvania Press. . Greene,
Bernsrein, R. j. (1991). The new constellation: The ethical-llOlitical horizons of
lJIodemitylpos{l/1odemity. Cambridge: lvlIT Press.
Dickman, L. (Ed.). (1987). Using IJrogram theory ill eualuatioll. San Francisco:
Jassey-Bass.
Carr, \\~ L, & Kemmis, S. (1986). Becoming critical: Education, /wOluledgc, a/ld
action l"csearch. London: Falmer.
Chelimsky, E. (1997). The coming transform<ltions in eV<l!uation. In E. Chelimsky
& \\~ R. SI13dish (Eels.), EualutItio/! for the 21st century (pp. 1-1.(--;). Thotls3nd
Oaks, CA: Sage.
Chen, I-I.-T., & Rossi, P. H, (1983). Evaluating \vith sense: The theory-driven
approach. Eualltatioll RCl'iew, 7, 183-301. . Greene,]. C. (1998b). Qualitative, interpretive evaluation. In A. j. Reynolds &
Cicirelli, \~ G., & Associates. (1969). Tile i11lpact ofl-.Jcad Start: All euallltllioll of
the effects of Head Start 011 children's cognitive alld affective del'e1ojJ11Icllt
614
unOl;'r.storlwng .::l000l r-rograms 1nraugn cvwuunurJ
(Repon ro the Office of Economic Opportunity). Arhens: Ohio University/
\'\!cstinghol1se Learning Corporation.
Cook, T. D. (1985). PU5rpositivisr critic:1lml1ltiplism. In L Shorbnd &= IV1. Ivl.lvlark
(F,L." Sod,,{ 5d,,'''''~ ,,,,,,I :iOCied polic)' (pp. 2162). ll",vo.;dy I lillo, Ci\; SalS""
Cook, T. D., & Reichardt, C. S. (Eds.). (1979). Qualitativc tlnd quantitativc JIIeth~
ods ill eualuatioll research. Beverly Hills, CA: Sage.
Cousins,]. B., & Earl, L (1995). Participatory euahwtioll ill edt/cation: Studies in
e/.'aillatioll lise and OIgtlnizatiO/Ialleaming. New York: Falmer.
Cronbach, L]. (1981). Des(r;ning eua/ltations of educational and soda! progral/1s.
S~ln Frnncisco: Jossey-Bass.
Cronb:lCh, L J., & Associares. (1980). Toward reform of program cvaluation. S<lll
Francisco: Josscy-Bass.
Daml, L.-E. (1999). The crhics of evaluation neutr<llity and advocacy. [n J. I...
Firzpatrick & Ivl. J"lorris (Eds.), Cllrrcnt alld emerging ethical challenges in
evaluation (pr. 77-88). San Francisco: Jossey-Bass.
Fay, B. (1987). Critical social science. Ithaca, NY: Cornell Univcrsit;; Press.
Fetterman, D. i'vi. (1994). Empmvermenr evaluation. Evaluation Practice, 15, 1-16 .
Fine, M., & \'{1eis, L. (1996). Writing the "wrongs" of fieldwork: Confronring our
own research/writing dilemmns in urban erlmographies. Qlwli:atiue Illquiry,
2, 251 ~174.
Freire, P. (1972). Pedagogy of the oppressed. New )'ork: Herder & Herder.
Gold, N. (1983). Sl'akeholders and program evaluation: Characterizations and
reflections. In A. S. Bryk (Ed.), S!lheholder~basedevaluation (pp. 63-71). San
Francisco: .lossey-Bass.
Goodyear, L K. (1997). ':4 circle that it's time to opell": Using performal1ce as a
representatiol1 of i/ participatory eual1latiol1. Unpublished m<lst'cr's thesis,
Cornell Universiry.
J. C. (1991). The practitioner's perspective. ClIrriCllllllllInqlliry, 22, 39-
45.
Greene, J. C. (1994). Qualic:Hive program evaluation: Practice and promise. In
N. K. Denzin & Y. S. Lincoln (Eds.), Halldbooh of qllalitatiue rcseardJ
(pr. 530-544). Thousand Oaks, CA: Sage.
Greene, J. C. (1996). Qualitative evaluation and scientific citizenship: Reflections
<lnd refrJcrions. Eualui-1tiol1, 1, 177-289.
Greene, J. C. (1997). EvaluJrion as advocacy. EuaIliatiOlI Practice, 1S. 15-35.
Greene,J. C. (1998a). Balancing philosophy and practicality in qualitative evalua-
tion. In R. DJvis (Ed.), Proccedings of the Stahe Symposium 011 Edllcational
Evall/alion (pp. 35-49). Champaign-Urbana: University of Illinois Press.
H.]. \'i/alberg (Eds,), Evaluation research for ed,lcatiollol prodllctivity
(pp. 135-154). Greenwich. CT: JAI.
615
 .... - .,~. ",--.,.,,--,, ,,....,, ,'-'"
Creene,]. c., & Caracelli, \~ J. (Eels.). (1997). Aduallces i7lIl1i.yed-method eUallli7-
tioll: The chaIlenges alld belle/its of illtegratillg diuerse Iwmdigms. San Frall-
cisco: jossey-Bass.
Cherne, J. C/ &- l\ilcClintodi! C. (J 99'1 J. The fJvolmio!'l ot c""luar;cm IIH:t!lodol ogv.
Jheo)'y Into Praclice, 30, ]3-21.
Greenwood, D.)., & Levin, ]v1. (J 998). llllroductioll to actioll research. Thousand
Oaks, Ct\: Sage.
Guba, E. G., & Lincoln, )~ S. (J 981). Effective eualuatioll: Improving the usefullless
of eu{IIl/atioll results through reslJO!1siuc alld ]wtflmlistic approaches. San
Francisco: Jossey-BJss.
Cuba, E. G., & Lincoln, ): S. (1989). Fourth generation wallfatiolJ. Ncwbur\' Park,
C1\: Sage. . In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qUlllitatiue research
I-!ouse, E. R. (I990). Methodology and jusrice. In K. A. Siromik (Ed.), ElIalliatiOJ/
and social justice (pp. 23-36). S:m Francisco: ]osscy-Bass.
House, E. R. (1993). Professional eu(/l1/ation: Socia! iml;act find political collse-
q!lelzces. Newbury Park, CA: SJge.
House, E. R., & I-lowe, K. (1998). De/iberatiue eualuatiolf. l\lper presented Jt the
<lnnU<llllleering of the American Evaluation Association, ChicJl:;o.
Ladson~8iJlings, G. (1998). JUSt whar is criticJI race theory Jnd what is'it doing in a
"nice" field like education? Intematiollal Journal of QlIalitatiue Studies ill
Education, 11, 7-24.
Lather, p. (1991). Getting SII/art: Feminist research alld t1edagogy with/in the post-
lIJodem. New York: Routledge.
Lincoln, )'. S. (1990). The making of a constructivist: A remembrance of rransfor-
!11Jtions past. In E. G. Gub<:l (EeL), The Immdiglll dialog (pr. 67-87). Newbury
Park, CA: SJge.
Lincoln, Y: 5., & Guba, E. G. (] 985). Nalumfistic inquiry. Beverly Hills, CA: Sage.
Mabry, L. CJ 997). A postlllodern test on postlllodernism? In L Mabry (EeL), ualu-
atioll alld the pastil/adem dilemllla (pp. 1-19). Greenwich, CT: Till.
l'vlacDonJld, B. (1976). Evaluation and the comrol of education. In D.A. T:'1wney
(EeL), CurricululI/ evaluatioll loda)'; "Hem!s alld im!JlicatioJls. London:
FaJrner.
Patton,1v1. Q. (1987). Evaluation's political inhcrency: Practical implications for
design Jnd lise. In D.]. Palumbo (Ed.), The politics of program ellaillalioll
(pr J 00-145). Newbury Park, CA: Sage.
Panon, M. Q. (1990). Qtw!italive evaluatiolJ alld research mcthods (2nd cd.).
Newbury Park, C1\: SJge.
Patton, M. Q. (1994). Dcvelopmemevaluarion. Evaluatioll Practice, 15,311-310.
Patton, Iv1. Q. (1997). Utilizatioll-(ocused cvaillation: New ceJttur)' editioll. Thou-
sand Oaks, CA: Sage.
PreskiJl, H. S., & Torres, R. (1998). Eua/zwtiue learning ill orgallizatiolts. Thousand
OJks, CA: 5:1gc.
616
UlllIt>l:>lLlIILlH Iy JUl.-lUI ,ILlyILlIII:> I III uuy" '-- VUIUUW.... ' I
Reichardt, C. S., [\: Rallis, S. E (Eds.). (1994). The qlw!itatiue-qIllilltitative debate:
New perspectiues. San Francisco: Jossey-Bass.
Ryan, K. E.} Greene, J. c., Lincoln, r~ S., Mathison,S., & Mertens, D. (1 ~)98).
.~ Adv:lnt"gc5 and c!,;llleng<'s or usins indusive ev,dU<:ltiol1 approachc~ in evalu-
ation practice. American JOUr/wi of Evaluatioll, 19, 101 ~ 122.
Scheurich, J. J. (1995). A posnnodern review of research interviewing. IJltema-
tiollal JOlfmal of Qualitative Studies ill Education, 8, 139-252.
Schwandt, T. A. (1989). RecJpturing moral discourse in evaluat:ol1. Educational
Researchcl; 18(8), J 1-16,34.
Schwandt, T A. (1994). Constructivist, inrerpretivist approaches 0 human inquiry.
(pp. 118-137). Thousand Oaks, CA: Sage.
SchwJndt, T A. (l996). FJrewelJ to criteriology. Qllalitatiue Illqlliry, 2, 58-71.
Schwandt, T A. (1997a). Evaluarion as practical hermeneutics. E:mhll1tioll, 3, 69~
83.
Schwandt, T A. (l997h). Reading the "prohlem of eVJluation" :n social inquiry.
Qllalitatiue I1Iql/ir)', 3, 4-25.
Schwandt, T. A. (1997c). Whose interests are being served? Progra:11 evaluation as a
conceptual practice of power. In L Mabry (Ed.), EuallfatiJll alld the pOS[-
modem dilemllla (pp. 89-104). Greenwich, CT: JAI.
Schwandt, T. A. (1998, April). Recaptltring moml discourse il1 evalltatioll-
reuisited. Keynote ;lddress delivered <:It the annual Kelly Conference, Ottawa,
ON.
Scriven,Ivl. (1967). The methodology of evaluation. AERA f\ilol.'ografJ/J Series ill
Curriculltm Eualrlltioll, 1,39-83.
Scriven, ?v1. (1995). The logic of evaluation and evaluation practice. In D. Fournier
(cd.), Reasoning ill cl/{.J!/fatioll: ilzferential lillks alld leaps (pp. 49-70). San
Francisco: Jossey-Bass.
Shad ish, \Y,J. R., Cook, T D., & Leviton, L. C. (1991). FOl/ndations o/program eVL1I-
uation. Newbury Park, CA: Sage.
Smith,]. K. (1989). The nature a/social and educational illquir)': Empiricislll versus
interjJretatioll. Norwood, Nl: Ablex.
Smith, J. K. (1990). Alternative research paradigms <:Ind the probkm of criteria. In
E. G. Guba (Ed.), The {Jaradigm dialog (pp. 167-187). Newbury Park, CA:
Sage.
Stake, R. E. (1967). The countenance of educational evaluation. Teachers Colfege
Eecord, 68, 523-540.
Stake, R. E. (1975). Ev{.J!uatillg the arts ill education: A respc'llsiue approach.
Columbus. OH: Iv1errill.
Stake, R. E. (1995). The art of case stud}' research. Thousand Oa:(s, CA: SJge.
Stake, R. E. (1997). The fleeting discernment of quality. In L lvIabry (Ed.), Evalua-
tioll and the postll/odem dilelllllla (pp. 41-59). Greenwich, CT: ]AI.
617
 \:lC~SS, c. H. (j 972). E/failialiol/. Englewood Cliffs, NJ: Prentice H:1IJ.
Wms. C. H (19R7) WI .. , I . .
_ 1Clt po [tlcs <lnd cvaJuntlon rcscClrch mcet.ln D. J Palumbo
C" / < '

(Ed.), 1he polzllcs of IJrogmll! c/lahw/ion (pp .17_70) TI . 10 I C

S"age. . " ' . lOU~,Hl( .1 ($
' A.
\.~; ;.,>., C.
~l y.;J1:5J .Eua{lwliO/l (lml teL). Upper Saddle River, NJ; Prentice Hall
11.

\X Jlltmor~, ~. (J ~94). To tell .rhe [.nlth: Working with oppressed groups in partj'ci-
. approaches to. mqll1fY. In P.. Reason (Ed)
p;](ury "1,I,c,p a IlOll 111
. IJiI1lWI1
.
mqlllty (pp. 82-98). Thousand Oaks, CA: S:lge
\Xfhirmore F (Ed) (199R) U d d.
~ . , j,
C
Influencing the
-. ' .,. . .. ' .. II erslalf 1l1g and /nl1cticing [Ji1l'liC//Jaj(J1')' eua{lll1-
t/01l. San FranCISCO: .tossey-Bass.

\\loJc1tf, H. ~'. (~:)90). On seeking-and rejecting-validity in qualitative research.

Policy Process \X7ith
n E. \\1. Elsner & A Pesh!-" (E i,) (-; /" . _ .
~1tl _t S. , ~1ta Ila/we fIltJlllry ill edllcatioll: The
, COllIl!llllJIg
Wolcon H F (JCI9-') 1'-
" "

;:."
'.

/"
'.
debate (pp. 121~152). New 'rode Teachers College Press
'- ..
Qualitative Research
" ' _ . . ." - , . hlllSlorlJllJIg qua Ilalive data: Descriptioll, ilIw/l'sis, and
I!Ilt!rpretatlOlI. TholJs~md Oaks, CA: Sage. -

Roy C. Rist

1\'lorc than 20 years ago, James Coleman wrote, "There is no body of

methods; no comprehensive methodology for the study of the
impact of public policy as an aid to future policy." This 1l0\v-famous quote
still rings true. Indeed, one can argue that in the intervening decades, the
tendency in policy research and analysis has become ever m'Jre centrifu-
gal, spinning off more methodologies and variations on methodologies,
more conceptual frameworks, and more disarray among rh'Jse who caB
themselves policy analysts or see themselves working in the area of policy
studies. A number of critics of the current scene of policy studies and the
attendant applications of so many different methodologies have argued
that any improvements in the techniques of policy research h3ve not led to
greater cbrity about what to think or what to do.1Vlore charitably, it could
be said that the multiplicity of approaches to policy research should be

AUTHOR'S NOTE; The I'iews expressed here arc thosc of thc amhor, anc1l1o endorscment by
the \\/orlct B<lnk is intended or should be inferred.

618
619
I
i
J.
 "'''Ut;:", ''-'' II) lilt" rUIILy rIULt':>:> VVIlII \.dUUIILUUVl;:' 1"\.t':>t'UfU I

\velcomed, as they bring different skills and strcngths to what are admit~
ted!y difficult and complex issues. which this is clone needs to be reformulated. \YJe are well pas.t ~he time
\vhen it is possible to argue that good research will, ~ccam.e It IS good,
Regardless of whether onc suPpOrts or challenges the contention that
influence the policy process. That kind of lin~J[ relatlon 0:: r~sear,c~1 to
po/icy reseQrd1 has lwei CLJ1triflli!Q/ ill1noct 0n ti,e J~Mwl.,dse I""',e 'ele
a
nua ~o policy making, the b;ttoJll line remains much the same: \\lhat pol-
- - ~--,. III pi)" V
i.lCUVll ovt il viable \\'i.I)' inb
which to tlnnk
_ about how kno\\
d ledge.
can inform decision making. The relation is bo~h l11~re su tie an more
icy researchers tcnd to consider as improvements in their craft have not
tenuous. Still, there is a relation. It is my intent III thIS chapta co add~ess
significantly enhanced the role of research in policy making. Instead, the
IlOW some at' 1I,e - - jlol"lgCS
'-~ of knowledge and action are, formed,
. _ partlcu-
I 1
proliferation of persons, institutes, and cenrers conducting policy-related
larly for the kinds of knowledge generated through qualttatIve researc 1.
work has led to more variation in the manner by which problems are
defined, more divergence in the ways in which studies arc designed
and conducted, and more disagreemenr and COntroversy over the 'Nays
in which data are analyzed and findings reported. The policy maker <!> The Nature of Policy Decision Making
now confronts a veritable glut of differing (if not conflicting) research
information. ..
Policy making is multlchmcnslOna . I an d mu lf ted Research
tJ ace. _ is but
, one
A sobering but provocative counterintuitive logic is at work here: fn- . I
(and often 1ll111or at that) among t le numlJcr 0 ,
L f frequently ~_
contradIctory
_.

creased personnel, greater allocation of rcsourccs, and growing sophisti- and competing sourccs that seek to influence what.ls an on.go111g.and L-o.n-
cation of methods havc not had the anricipated or demonstrated effect of stanrIy evolving process. The emphasis here on polIcy ma!(]n,~' bel~~ a PIO~
greater clarity and understanding of the policy issues before the country. cess is deliberate. it is a process that evolves through q:cles, \\ lth each ~~cl
Rather, current efforts have led to J marc complex, complicated, and par- more or less bounded, Illorc or less constrained by tIme, funds,' poh~.lCal
tial vie\\' of the issues and their solutions. Further, as Smith (1991) would suppOrt and other events. lt is also a process that circles back on Itself, lter-
argue, this tendency to greater complexity has left both the policy makers .
ates the , same decision issue tllne . an d 0 ft en cloes not come to
an d agam,
and the citizens less able to understand the issues and to see how their closure. Choosing not to decide is a frequent outcome. . .
anions might affect the present condition. . process sugges t s the need for a modlfl-
Such a dcscription of the polley . f
\Vhereas one may gram that early analyses, for example, in the areas of I . f i
cation if not a fundamenta re fTam1l1g, 0 t le rat dicional understandmg
_, '-: 0

education or social welfare, were frequently simplistic and not especially policy'making. in this latter, more traditional approach, dec:~slOn maJb(l~g
sophisticated in either the dcsign or application of policy methods, the in the policy arena .IS uneIerstOQ(Jas' a dl'screte event , ' und,_naken
_ ; ' d . ,:'" a
inverse does not, in and of itself, work to the advantage of the policy . . ,- j ' "
d -fined set of actors workll1g 111 "rea tIme an d mOVll1g to Clelr
' . eClSlon
maker. Stated differently, to receive a report resplendent with "state-of- c, . . I . m S
on the basis of an analYSIS of theIr a ternatlves. welS " ._ (198)) has TIlcely
the~Jrr" methodologies and complex analyses that tease out: every nuance summarized this notion of "decision making as an event :
and shade of meaning on an issue may provide just as little guidance for
effective decision making as did the former circumstances. The present . I'Jrerature picture decision making
Both tbe popular and tbe aC;lC1COliC '. .as an
fixation on the technical adequacy of policy research witham a commen- event a group of authOrize, d cIeC1SIOn
.. ma Ieers, assemble ,at parn:ular tlmes
surate concern for its utilization is to relegatc that work to quick obscurity ,
and places, review a problem (or opportumt) y ~ ~o nsider a"numrer
_ of d'a
alter-_
(Chelimsky, 1982).
native courses of action with more or less e~pllClt calclilan.o.n of.,.tb.e a ~L ~
If this adminedly brief description of the Current state of policy re- maes and disadvalHagcs of each option, \",clgh the altcrnatl\'cs a'.'.l am.st td,felr
search approximates the reality, then a fundamental question arises: Is the o
oals or preferences, and then selecr an a1ternatlve
. tl1a t seemS' we sUlteor
presumption correct that research cannot be conducted that is relevant to
gachieving
. their purposes. Tbe resulr .IS a cIeCISlOn.
' (p._J
1")

rhe policy process? It is my view that the presumption is nor correct. Re-
search can contribute to informed decision making, bur the manner in
She also nicely demolishes this view when she writes:
620
i 621

J
 IIfHut:'IIUII~ Lfle r-OIlCY r-rocess vVlrn \.juatltorrve Keseorcn

Civcn the frar;melll':nion of 'wthor" '. .

and Jegislarl\:e committees' "111el '[1 !t,~ _;)I'c~().ss !1l~11 ([pie bUrC<lllS, deparrmen lS,
-, c It: llSJOllHCll stJgcs! ' 'I' I
. . this view would contend, and the policy ;:lctiol1s can be deliberate, di-
alesce imo decisions II _', -I" 1 . - ,)} \\ lIe 1 actIOns co-
. . " 1t: [1,ltJtlona modelo/-I,-'," /_" . rected, and sllccessful.
Stylized rendition of rC"l/it, 1-1- 'f" l eCISJOn 018 'ltlg IS a hIghly
). lent! Icall011 of all\' clear' .- -fl' . These comments should not be taken as a diatribe against, research or an
makers Gill be difficulr. ('omp,;",,,,,., ",;.rJk,"'Yd ~ cur g!'.Jllp 0 [eelSIOn ;;u5urut:nr then l-;:nQwkdbc CUUlH51ur lli:LuJ;;IIL Quite tht: contrary. System-
heyaCf/on :1lrhou"h I > _ _1_
uu;aUt:ldL lIas [;l](en the
"- to> H: O[ ::; It:: mill' be tl/l;l\l:1' I I' 1
ing to be-or W'ls-d ~ ." ) TI - , :' Ie r 1ar liS or ler action was go- atic knowledge generated by research is an important and necessary com-
, cCIS1VC.
except in rerms of "taking ' . f"
1e goals at polic I ' - r _ .
. } arc 0 len equally diffuse,
<-
ponent in the decision-making process. Further, it is fair to note that there
. C<lleo somcundesrr_bl" .. is seldom enough research-based information available in the policy
lOllS arc considered "md! _ f . :1 C sltu:mon. \'\1/1Ic h Opl11-
W lat set o' advantages or disadv<lllt ,. . ,
, , . ' <
sessed, mal' be JlllpOSsible t II' 1 ' ' - arena. \Xlilliam Ruckelshaus once nored th8t although he W:1S the adminis-
. ' .- 0 te 1n [le JlHer,Jetive !.' '. ages <lIe
. as-
process of (ormu!;ltinr.: p I'., TI ' _ . ' ~lltl !IpartlClp<lnt, dIffuse trator of the Environmental Protection Agency, he made rnany decisions
. , u 0 IC}. 1t: complexltV ot ' " ~. I. '.
lluklng otten defi-,s ne t .. " - gO\crnmt:JHa decIsIon when there was less than 1Oc;---r) of the necessary research infurmation :nraiI-
, t:. <l co m panrnellt<lllz3tion. (p. 26)
able t:o him and his staff. The rclevJnce and usefulness of policy research
wiH not become apparent, however, unless there is a reconsideration of
Of particular relevance here is tha.t r1 ' f ' .. what is understood by decision making in the policy process. A redefini-
ongoing set of Jclj'ustmcnr "d '- 1e ocus ~n decIsIon making as an tion is needed of the context in \\'hich to look for a linkage b,~tween knowl-
r S, or 1111 coursc correct!) _/". 1
orhavingtopinl1ointd' ~ , , ' I ' " . (nS,C[llllJ1atestlebind edge and action. Unpacking the nature of the policy cycle is the strategy
.' . 1t evell[-t lat is the c ~ . ~ I,
ner-Ill which resc3rch 1 I, > fl ' '. X<1ct time, p acc, and l11an- employed here to address this redefinition of policy decision making.
las Jecn [n 'UCntl;]! 0 r, r p .
becausc the specifics C"l 'jd 1_ . n po Ie} .. arenrhetlcally,
en se om 1(;' supplIcd ,I ' I
should havc an imp'lCt 0 d~"'.~ I ' . ' tle notIOn (lat research
. '- n eelSlon rna (]J1g seen t 1 b
and more an 8rticJe of f 'rl "[1 '- - 15 0 lave ecome 1110re 01> The Policy Cycle and Qualitative Research
al 1. lat researchers J." .'
understood decision making d '" J ldve so persIstently mis-
< _'

fJu-~n' '" '. ,an )t't lave constanrly sought to be of in-

1 t: ce, IS d SltU3Uon deserving of ~ ,.' I ~ 1 There arc tVI-'O levels of decision making in the policy arena. The first
receives. So Ion _ ' . I '- COllSIC era Jiy more analysis than it involves the establishment of the broad parameters of gove:~nll1el1t action,
brought to bear ~p'~nr::~I~r~l~rs p~esul.lle that research findings mUst be
such as providing national health insurance, establishing a national energy
will be miss,'n 1 .' g e event, a d,screte act of decision making they
, g t lOse CIrCUmstances' d . ' . policy, restructuring the national immigration laws, or reexamining the cri-
can be usefuL However th . ' ,-l~ processes where, III fact, research teria for determining the safety and soundness of tbe country's financial
ing" and to "process de'cis;o~e~:~~:;~~,~O~~~W~?from "ev~nt decision mak- institutions. At this level and in these instances, policy re~;earch input is
serving an "enli I
. "
" ,?
essltates lookmg at research as
g Henment fUllctlOn . 111 Contrast to an "el' . likely to be quite small, jf not nil. The setting of these national priorities is a
tJOll (see Janowitz, 1971; Patton 1988. ' V I ' 19meenng[Ul1c_ political event, a coming together of a critical mass of politicians, special
V' > WeISS, 1988)
' ,

lew1l1g policy research as servin an " j' Jl' . . interest groups, and persons in the media who are able :OlffiOr-g them to gen-
that policy resc3I"chers work witl ~.'-, el~ l~ Ht:nmcnt ~unctlon suggests erate the attention and focus necessary for the items to reach the national
to create a en t . j 1 po IC} fl1dkers and theIr staffs over time
< 11 extlla understa d' l . agenda.
wiJl exist over time '1[1(} str,>, < n mg ajJout an Issue, build linkages that Iron triangles" built by the informal linking of supporters in each of
, "re constant y to ed .. I
ments and rese'lrch find,', . 1 Ucate a Jout new develop- these three arenas are not created by the presence or absence of policy re-
. c. ,lgslntlcarea.This's' j'
mg perspective, where it i s , '" I: l.n COntrast to t le engmeer- search. One or another research study might be quoted in support of the
beJr to detenlline the dire'cPtresumcl~ that ~uffJclent data can be brought to contention that the issue deserves national attention, but iteS incidental to
. JOn anc JntensIty of th J d l' " .
nve, much as one can de r 1 I ' . : . e Inrenoe po ICy IDltla- the more basic task of first \-varking to place the issue 0::1 the national
~Jr~dge. If the policy direc:i:I~~s ~~~ff~~i:~~I~a:l~l1~ ,for. the building of a agenda. If one wishes to influence any of the players dllting this phase
1l1rormation relev:mr to tl . d . 1 } xpllclt, then the necessary of the policy process, it is much more likely to be done through personal
'- le eve opment of the policy can be collected , so
contact, by organizations taking positions, or through the creation of
622
623
 sufficien t static in the policy system (for example, lining up special inrerest
Smallwood offer a relevant departure point with their de:;cription of the
groups in opposirion to a proposal, even as there arc groups in favor). This
actors involved in policy formulation:
works to the benefit of the opposition in that media coverage will have to
be seen ro he "balanced" iJnd COVtrJ,ge of [h~ ()J)pnqit;()l1 c"n Cr<~"t"e ,.I-,e
:'"p,-<c,-,,,ic,,, tb,-,~ tlKrc jj rlUl rllE Strong unified suppOrt for a position that I n genenl the prinCIpal :1ctors In. polK)',~l rnulation
or, are the ((legitimare'! or1
otherwise would seem EO be the case_ . . . makers: people who occupy POSt'(lO~S 1.1.1 the ',:rovernment3
form,ll policy ''.. . ,_

Once the issue is on the agenda of key actors or organizations within the . . I y ~S.SI'gn priorities
arella that emitle them to aut I10ntatn'c , ~ .Jcommit
. 3nd I i Th~level
rc
Irces These people include elected offJcl<lls, leglslatOls, d.ll 1 g
policy establishmem, there are possibilities for the introduction and utili-
SOl ..
adminisrratlve _ appOllltees,
. . If.,1
<cac 11 must
1 a \\ 1 0 1 . follow prescnhed . paths
'to
zation of policy research.1t is here at this second level of policy making_
1-- I"cv Since these forma I po Itcy Illa I~er s r epresent
. d.v<crse
'conStlt-
.
the level where there arc Concerns abour translating policy intenrions inro rna ~~ po I ....
llcllCles-eleet'ofal, _ .. '.. "
admllllstrrlf1\e, .
and bureaucratic-rhe rolicy making 1 1
policy and programmatic realities-that I will focus in this chapter.
Process- offers many points of access rI1roug I1 \V I'11~ I11~. uerest .':;'ffronps
... alll Ot 1-
1.-
The frame\',.'Ork in which the comributions of policy research in gen-
ers ftom arell,lS outSide . gO\.ren.1Il.1Ctlt can
. exel -. 'ose .. lllflllCtlCe.fThus. 1 po 1.-IC)
eral and qualitative research in particular can best be understood is that of m'lhllP usuall\' involves a dtverse . . set , 0 fall tl1 ontatlve 'or:Jnnd,
I. po
_ IC)f
the policy cycle, a concepr that has been addressed for more than a decade " " 1-'
'- D I 0 .. peratc within the governmenra 1arena, Pi, I IS '1 l Iverse :,ct 0
(see, e.g., Chelirnsky, 1985; Guba, 1984; Nakamura & Smallwood, 1980; rna ,-t:rs, W 1. 0 . [ 1 Outside arenas, who
special interest and other constnuency groups r~lll --, '1 .
Rist, 1989, J 990,1993). I will develop my discussion of the policy cycle press their demands on these formal leaders. (pp . .) _-.L)
here according to its three phases-policy formulation, policy imple-
mentation, and policy aCCountability. Each of these three phases has
As the fonnuhuion process begin.s,.there are a numbe~l~f prefssin g l~lles~
its own order and logic, its own information requirements, and its own . .. . . I ~ ompI 1l1g ()f W
dons. Answering each questlon necessitates t It: c llat:ever
dd'.' .1
policy actors. Further, there is only Some degree of overlap among
I fle
' I I e pus
information is currently avaJia) I develop
.- . mew."1]-'
0 a ItlOna TI
the three phases, suggesting that rhey do merit individual analysis and
undersranding. information when the gaps are too great d I . 'y b ~no\l,;n.
'. 1'n what IS curren: 'oad sets of le
information needs can generally be clustered aroun t l;.c:e f I alation- <
The opporrunities for qualitative research \vithin the policy cycle are
thus defined and differentiated by the information requirements at each
q uestions. Each of these c1usrers is highly relevant to po orm "I' ,~ I?
.. f or t l'
in each there are important opporrul1lt:les le presentatIOn
_ an d Ut:lIZa
phase. The questions asked at each phase are distinct, and the information
tion of l]ualitativc research. d d f
generated in response to these same questions is Llsed to different ends. It is
to a detailed examination of these three phases of the policy cycle and the Thc first ser 01- information
. necds revo 1ves ara und ,.'an un" erstan mg 0
,hc rob!em
the policy issue at hand. \'7hat arc the contours of thiS Issue: Is], p
manner in which qualitative research can inform each phase that I now
t:Urn. -
or condition one thar is larger now t1lan be fore ., abou
< 1 t t .Ie I..
same, 1.or
' lIer) Is anything knovm about W 1lctIlcr t'jle nature of t le
. cone
. mon ldS .
sma. - .
C-I,angeeP Do the same target populations, areas,
or instlrutlOl1S
' d..expen-
b
<I> Policy Formulation ence this condition now as earher? How we JJ c~~n
< '-- ' tht. L-on mon '..
et
defined;' How well can rhe condition be measured? \X1hat.are the differden
. f h' ~b'litv of policy
. . ' b I condition lts causes an ItS
Nakamura and Smalhvood (1980) define a policy as follov1ls: ;~A policy can interpretations and understandm.gs a. out t:e
effects;' The issue here, stated elifferenrly, 15 one 0 tea I '1- _ 1-
be thought of as a set of instructions from policy makers to policy imple-
makers .. to define dearly and understan d t1le pIO. blem or cone mon t lat
menrers that spell our both goals and the means for achieving those goals"
tIle)1 are facing and for which they are expected to develep a response. ]
(p. 31). How is it that these instructions are crafted, by whom, and with
Cl1arles, Lindblom (1968) has I1lce
. ]y capture dsom e a f. the f'conceprua
'. , f
what relevant policy information and analysis? The answers can provide
'
complexity . makers as t I
facing poltcy , to cope with the delnlt!on
ley tf} 0 a
important insights into the process of policy formulation. Nakamura and
policy problem or condition:
624
625
 "''' ... ~,.~'''~ ".~, ~ .. ~,.

r:~Jicy makers are nor faced \'lith a given problem. Instead thev have' _ '-1
tlty and formulate ['heir) -II. R' _" . ' . IU Ilen- the changing nature of social conditions are all germane to the questions
c,,. 'XTI .. I 1- n~) t:1l1. IOllng breaks om III dozens of Amcricm
ItS. \ 1"tISflC!1roblCtll" i'd' ' . 1 posed here.
. .",. - . J' <llnt<lltllng aw ilnd order? Racial d' .. .
[J~Jn~ InCipient revolution? Bl:l c k power? Low income l L 1/ __ ISCflll1l11;J- There is an additional contribution that qualitative worl~ can make at
L-,nG,," .,{ con .-"1,~",,,,;,,,. ,_ I - __ \ _ _ J\v essness nt tile ~hi,; ,;L-'-'6'" or d,,,, p",li",1'" 1~':<:-''''''''>''] and ir i" "hat: of >;ll[driu Q lobe inrcndcd ~Llld
organization? A!jcn3ri~ln; ('~:~ ;.-l) FC<I"clul n:l()fIT! movement? LJrh;ll1 dis-
unintended consequences of the various policy instrument~; or tools that
might be selected as the means to implement the policy (Salamon, 1989).
The second duster of question f _ , There is a present need within the policy community to ascertain what
nusly in res _ 1. I"' 5 oeuses on what has taken place previ- tools work best in which circumstances and for which targer populations.
. - ,ponsc to r 115 cone ![IOn or problem \\!Ilar .
I ,t ",' I' . "programs or project's Very little systemaric work has been done in this area-which frequently
ld\C prevIOUS y been 1I1iti"11ed' .r '10'"- Iong dd the}' J ~ LI - ,. .~

were thev) \'(11 Ht ]n ,-! - f.c r- < I
_ ast. I ow successful leaves policy makers essentially to guess as to the trade-off, between the
'were req-~ire -pc T_I e 0 Jlllle: lIlg \vas required? How many staff members
l . 1 ow rcccpn\'c were the po ulation .... choice of onc rool and another.
these iniri"'I'"c''
"' _ . request h "I - dOd p,
5. DIeI t 1ll":\' ] . . 5 or .inStitutions to Information of the kind provided by qualitative research can be of sig-
Did t1.. . ," _ .. _ . _ . e p or 1 t ley reSIst the Jl1tervenrions) nificant help in making decisions, for example, about whether to provide
1C ~rc\ lUllS cHarts address the same condition 0 . II". .
rentl}' eXI . 'f- r PIO) em as Cur- direct services in health, housing, and education or provide vouchers to
'. -. SfS, or \vas It oj o1:erent? If it \vas different h ) If.
S"ll1lC \Vh t _ .'_ ] I' .' - , ow so. !t was the recipients, whether to provide direct cash subsidies or tax credits to
., , ) arc yet acc !tlonal efforts necessary) Arc I _ .'
groups involved? \\7hat may ex l' . r 1e S8mc Interest
I ' .. employers who will hire unemployed youth, and whether to increase
. I"
'-g roup CO;] mOil:'
~ . - p am an) changes 111 the present imcrcsr ... funding for information campaigns or to increase taxes as strategies ro dis-

of ~;~cctyl~l~dt~lstcr of qLlehsri(~ns relevant to the policy formulation stJge courage sl11oking. These arc but three examples where different policy
tools are available and where choices will have to be made among them.
e OCllses on W .at IS known of the revi ' -_ '-.
impacrs thar \vould help one c h - , p. ous etfon~ and theIr Key among the activities in the policy formulation stage is the selection
'd . _ . _ . OOse among present-day optlons. Con- of the most appropriate policy strategy to achieve the desired objective.
Sl erIng trade-offs among vanous levels of f f ' _.
ferem level f ~ . I e art m companson to dif- Central to rhe design of this strategy is the selecrion of one or more rools
s 0 cost 15 Jut one among several kinds of I I
co . I I_ . . e ,Ua re evant to available to the government as the means to carry out its intentions. Quali~
nSle ermg t lC policy Options TI I
<. lere may' a so be d'lt I
frames neCessar bef " ' . . ' a on t 1(': tIme tarive studies of how different tools are understood and re:;ponded to by
yore one could hope to see Impact T d I'f h
twee I J I [I s. ifa e~o s C* target populations is of immense importance at this stage of the policy
._ .n r le el1gr lOt 1e developmental stage of the program and rhe evel1-
;ualjlmpact<,> Jre rej~vant, particularly if there arc considerable pressures
process.
Unfortunately, although the demand for analysis of thif~ type is grear,
~~~~~ w.r-,t~~erJ1l ~Iolllt;ons.Th~ rendency to go to "weak thrusr, wea-k effect" rhe supply is extremely limited. The qualitative study of policy tools is
. .CgIl::S IS \\~e une e~swodll1 these circumstances. Alternativelv if revi-
ous efforts dId neceSSItate a considerable period of ,. f ., P an area that is yet to be even modestly explored within the research
lille or measurable
Outcomes to appear how did th
hold I 1"1 ' -
r
<- -
.
e po ICy makers ll1 those circumstances
community.
Although qualitative research can be relevant at this stage, it is also the
f . t]on.to rile PH) ic suppOrt and keep the coalitions intact long enougth
or 1e 1 esu ts to emerge? case that its applications are problematic. The basic reason is that seldom
Q~alitative research is highly relevant to the informati ,,_ I I.
is there enough time to both commission and complete ne\v qualitative
Stl-e ~
'"- 111e po IICY evc Ic. Studies
e < b on tI -" ~. ] . onc I1t;CC sat t lIS research within the existing window of opportunity during policy for-
.ff"
r]1C eIIcnng llltcrpr' .
, ,. f .]leSOua constructlOnorprobJems
L

- -., on mulation. Thus the applications have to rely on existing qualitarive

.. ~ . . - e a IOns 0 SOCIa conditions , on t],e b 11l.] CImg an d sus- research~and that mayor may not exist. Here is one key means by which
rammg 01 coalmons for change on previous p . .. . c-
i _ . ' - rogram IIlltlatives and their good, well-crafted qualitative work on topical social issues can find its way
~::ac.ts, ?!1 COl1lm~l~lty and organizational receptIvity to pro rams all
o~DaJ1lzarlOnaIstabJiIty and cohesion during ,I f I.' g ,. into the policy arena. As policy makers start on the formulation effort,
. .1eormu atlon stage, and on their need to draw quickly on existing work puts a premium on those
626
627

research studies that have worked through maUers of problem definition,
the social construction of problems, community srudies, rerrospectlve
assessments of prior initiatives, and so on.
The problellldric nature of the appliGHions of Gll:llit:J.tive ro!';e,'1rch ;:It
t!lis smge is further reinforced by the fact that seldom are researdl funds
available for srudies rbat address the kinds of questions noted above in the
three dusters. If the problem or condition is nor seen to be above the hori-
zon and thus on the policy screen, there is linle incentive for a policy
mJker or program manager to use scarce funds for what would appear to
be nonpragm.1tic, "theoretical" studies. And by rhe rime the condition has
su fficien tly changed or become high Iy visible JS a soci;ll issue for rhe policy
community, qualitative work is hard-pressed to be sufficiently time sensi-
tive and responsive. The windO\v for policy formulation is frequently very
small and open only a short time. The information that can be passed
through has to be ready and in a form thar enhances quick understanding.
The above cOr!Snail1ls on the use of qL13litJtive research at this stage of
the policy cycle shonld not be taken as negative judgments on either the
utility or the relevance of such information. Rather, it is only realistic to
acknowledge that IlJving the relevant qualitative research available when
it is needed for policy formulation is not always possible. As noted earlier,
this is an area VI/here there are potentially significant uses for qualitative
studies. But the uses are likely to corne because of scholars and researchers
who have taken on an area of study for their own interest and to inform
basic understandings in the research community, rather than presuming
before they begin that they would influence the formulation process. ft is
only the infrequent instance where there is sufficient time during the for-
mulation stage for new qualitative work ro be conducted.
It should be stressed here that the restrictions on the use of qualitative
work during the fonnulation phase of the policy cycle come much more
from the llarure of the policy process than from the nature of qualitative
worle. The realities of the legislative calendar, the short lives of most senior
political appointees in anyone position, the mad scramble among compet-
ing special interest groups for their proposals to be addressed and acted
upon, and the lack of concentration by the media on any issue for very
long all inhibit rhe developmem of research agendas that address the
underlying issues. This is ironic because it is clear that the country will face
well into the foreseeable future the issues of health care allocation and
quality, immigration comrols and border security, educational retraining
of dislocated workers, and youth unemployment, to name but fOUf areas
628
IIII/LII::IIUIIY Lilt' rUllcy rrULl;;'::;S VVIHI \.dUUlItUUVt' (\t'jt'UICIl
that have heretofore persistently swyed ncar or at the top of the nation81
policy agenda. Basic, in-depth qualitative worle in these and other key
areas could inform the policy formulation process for years to come. Bur
tile prc::;::;urc::; and CitrucLUwl inccntive; in tllc policy ::i)'::itcrn aU go in the
other direction. To wit: Develop short-term proposals with quick impacts
to show responsiveness and accommodate all the vested interests in the
iron triangle.
In sum, with respect to this first phase of the policy cycle, qualitative re-
search can be highly influential. This is particularly so with res?ect to
problem definition, understanding of prior initiatives, commulllty and
organizmional receptivity to particular programmatic approaches, ~nd
the kinds of impacts (both anticipated and unanticipated) that mIght
emerge from different intervention strategies. This information would be
invaluable to policy makers. But, as noted, the use of the material can be
hindered bv such factors as whether or not the information exists, is
known to tl1c policy community, ;md is available in a form that makes it
quickly accessible. Overcoming these obstacles does not gr:arantee the use
of qualitative research in the formulation process, but one can be strongly
assured thar if these obstacles are present, the likelihood of the use of qual-
itative mJterial drastically diminishes.
<3> Policy Implementation
The second pl1J.se of rhe policy cycle is that of policy implement~tion. is I:
in this stage that the policy initiatives and goals established durmg poltcy
formulation are to be transformed into programs, procedures, and regula-
tions. The knowledge base that policy makers need to be effective in this
phase necessitates the collection and analysis of differen~info~mat.i~nf~om
that found in policy formulation. \vith the transformatlOn ot- pohc~es mto
programs, the concern moves to the operational activities of the POlICY tool
and the allocation of resources. The concern becomes one of how to use the
available resources in the most efficient and effective manner in order
to have the mosr robust impact on the program or condition at hand. As
Pressman and \'V'ildavsky (1984) have written in this regard:
Policies imply theories. "'\Vhether stated explicitly or not, policies point to_a
chain of causation between initial conditions and future consEql!ences. If X,
then Y Policies become programs when, by authoritative acton, the initi;ll
629

conditions ~rc cre~ted. X now exists. Progr~ms Il1Jke the theories opera-
tional by forging thc first link in the L.WS'll chain connecting actions to ob-
jectives. Civcn X, we act to obt:Jin y. Implememarion, then, is the ability to
forge ,ub,rquclH links ill [he causal chain so JS to ohtJ;n the desired I'e"dt,.
{p. XXii)
The research literature all policy and program implementation in-
dicates that that is a particularly difficult task to accomplish (see, e.g.,
Hargrove, 1985; Pressman & \X1ildavsky, 1984; '{in, 1985). Again, qUOt-
ing Pressman and \X1ildavsky:
Our norlll<11 expectations should be that new progr<1lllS will fail to get off the
ground and thar, <1t best, they will tJkc consider~ble tilllc to get started. The
cards in this world arc st<1cked Jgainst things happening, as so much effort is
required to make them work. The remarkable thing is that new programs
work at all. (p. 109)
It is in this conrext of struggling to find ways of making programs work
that the data and analyses from qualitative research can come into play.
The information needs from qualitative research at this stage of the policy
cycle cluster into several areas. First, there is a pressing need for infor-
mation on the implementation process per se. Qualitative researchers,
through case studies, program monitoring, and process evaluations, can
inform program managers responsible for the implementation of the pol-
icy initiative.
Qualitative work can focus on such questions as the degree to which the
program is reaching the intended urget audience, the similarities and con-
trasts in implementation strategies across sites, the aspects of the program
that are or are not operational, whether the services slated to be delivered
are in fact the ones delivered, and the operational burdens placed on the
institution or organization responsible for implementation (i.e., Is there
the institutional capacity to respond effectively to the new policy initia-
tive?). The focus is on the day-to-day realities of bringing a new program
or policy into existence. This "ground-level" view of implementation is
best done through qualitative research. The study of the rollout of an
implementation effort is an area where qualitative work is at a clear advan-
tage over other data collection strategies.
A second duster of research questions amenable to qualitative work
in the implementation arena focuses on the problem or condition that
630
""'~~"~"''::J .,'~, u"~r' 'u,-c".> ,.,"" ",::"U,-"'CL""'" ""''''=''.''U,
prompted the policy or progr31l1 response in the first place. No problem or
condition stands still simply beclllse the policy community has decided to
rake action on what was known at the time the decision was made. Prob-
I"",,, n"d co"ditio",. ",I""'ljc-bot], DeCore ,,,,d a{t",r '-'- t~v[;'-'r ,-c:;pon:;c i:;
decided upon. Thus the challenge for qualitative researchers is to continue
to track the condition, even as the implementation effort swings into
action. Qualitative work can providc ongoing monitoring of the situa-
tion-whether the condition has improved, worsened, remained static;
whethcr the same target population is involvcd 3S earlier; whether the
condition has spread or contracted; and whether the aims of the program
still match the assllmptions and previous understandings of the condition.
Qualitative work can provide an important reality check for program
managers as to whether the program is or is not appropriate to the current
condition. Qualitative work that monitors the condition .:n real time can
playa key role in the continuous efforts of program managers to match
their services or interventions to the present circumstanc,::s.
The third cluster of necessary policy questions during this implemen-
tation phase of the policy cycle focuses on the efforts mad:: by the organi-
zation or institution to respond to the initiative. Here, for example, quali-
tative dara would bc relevant for learning how the organiz8tional response
to tbe condition or problem 113s been conceptualized. Are the socia! con-
structions of the problem that were accepted ae the policy formulation
stage by federal policy makers accepted during implcmentation by the
program managers and staff months later and perhaps thousands of miles
away? What has been the transformation of the undersrardings that have
taken place when the policy or program is actually being implemented?
Do the policy makers and the program implementation folks accept the
same understandings as to the intent of the policy-let alone the same
understandings of the problem that the policy is suppose to address?
Another aspect of this need for qualitative data concerns the organiza-
tional response. Here questions would be asked that address the expertise
and qualificltions of those responsible for the implement;ltion effort, the
interest shown by management and staff, the controls in place regarding
the allocation of resources, the organizatiollCll structure and whether it
adequately reflects the demands on the organization to respond to this
initiative, what means exist in the organization for deciding among com-
peting demands, the strategies the organization uses to clarify misunder-
standings or ambiguities in how it defines its role in implementation, and,
finally, what kinds of interactive information or feed bade loops are in
631

place to assist m::magers in their ongoing efforts to move the program
toward the stated objectives of the policy. It is information of precisely this
type on the implementation process that Robert Behn (1988) notes is so
organizational goals.
'" Policy Accountability
The third stage in the policy cycle comes when the policy or program is suf-
ficiently mature that one can address questions of accountability, impacts,
or outcomes. Here Jgain, the informJtion needs arc differenr from those in
the nvo previous stages of the policy cycle. The contributions of qualitative
research can be pivotal in assessing the consequences of the policy and pro-
gram initiative. Just as the questions change from one part of the policy
cycle to another, so too docs the focus of the qualitative research necessary
ro answer these S;lme qucstions.
First there is thc matter of what the program or policy did or did not
accomplish: \Vere the objectives for the program met? Qualitative re-
search can specifically help in this regard by addressing, for example,
whcther the community and police were 3ctively working together in a
neighborhood "crime watch" program, whether the appropriatc target
audience of homeless persons in another program received tbe health ser-
vices they \vere promised, and whether in a third program youth were
given the type and quamity of on-the-job training that resulted in Sllccess-
ful placements in permanent positions.
\vhen a program reaches the stage that it is appropriate to discuss and
assess impacts, qualitative research provides a window on the program
that is simply not available in any other way. Qualitative research allows
for the study of both anticipated and unanticipated outcomes, changes in
understandings and perceptions as a result of the efforts of the program or
policy, the direction and intensity of any social change that results from
the program, and the strengths and weaknesses of the administrative!
organizational structure that \vas used to operationalize the progr8m. Pol-
icy makers have no equally grounded means of learning about program
impacts and outcomes as they do with qualitative research findings.
These grounded means of knowing also carryover inro what one might
traditionally think of as quantitative assessments of policy. Qualitmive
work can provide to program managers and policy makers information on
632
"",u .... ' ''''''''':1 ",c::, <.."' .... 1' ,'-'...."''''" ,,,,-,, .... ,'.... -''-'-', .... ,,
~'-''-'''''-'",
how confident they can or should be in the measures being used to deter-
mine program influence. Although the intent may be that of a highly reli-
able and replicable instrumenr that allmvs for sophisticated quantifica-
The issues of reliability and validity are well known in the research liter-
ature and need not be reviewed here. Suffice it to say thar policy maleers
and program managers have been misled more than once by investing a
great deal of time and effort on their instrumentation without equal
emphasis on answering the question of whether their mea:;ures were the
appropriate ones to the problem or condition at hane!. Swdies of school
desegregation and busing or health care in nursing homes are but two
areas vI,here a heavy emphasis on quantifying outcomes and processes
have left key aspects of the condition undocumented and thus unattended
to by those who should have been paying attention.
There is an additional 3spect of this first cluster of information needs
that merits special 8etention vis-i-vis qualitative research. This has to do
with whether the original objectives and goals of the policy stayed in place
through implementation. Gne message has come back to policy makers
time anel ag8in: Do nor take for granted that \vhat was intended to be
cstablished or pu tin place through a policy initiative wi II be \vhat one finds
after the implementation process is complete. Programs and policies make
countless midcourse corrections, tacking constantly, malcng changes in
funding levels, staff stability, target population movement~, political sup-
parr, community acceptance, and the like.
It is through the longitudinal perspective of qualitative ',vork that such
issues can be directly addressed. Blitzkrieg assessments of programs are
simply unable to pick up the backstage issues and conflicts that will inevi-
tably be present and that may directly influence the direction and success
of the program (Rist, 1980). To ignore staff turnover in a program that is
highly staff-intensive in thc provision of services, for instance, is to miss
what may be the key ingredient in any study of implementation. Bur recog-
nizing that it may be an issue in the first place is one of the ways in which
qualitative work distinguishes itself from other research srrategies.
The second cluster of information needs that emerge wh,:;n a program is
being assessed for impacts and outcomes is that of addressing whether and
what changes may have occurred in the problem or condition. Central to
any study of outcomes is the determination of whether in fact the condi-
tion itself has changed or not ancl what relevance the program or policy
did or did nor have to the present circumstances.
633
 Although it is rudimentary to say so, it is worth stating explicitly that
problems Ciln c11ange or not, rotally independently of any policy or pro-
gram initiative. Conceptually what \ve have is a situation in which impacts
t!6uld I'll' '~(\111d not !1C1VC nec",-reel, ",n,-~ 1:1.,,, ,'on""'-1ucnc<;o ,sou!ell.,e c1.ntl '-'
o
or no change in a program or condition.
For example, a positive outcome of a policy could be 110 worsening of a
condition, that is, no change in the original status that first prompted the
policy response. Developing local intervention programs that stalled any
growth in the number of child abuse cases could be considered a positive
outcome. The key question is, of course, whether the evidence of no
growth can be anribured to the intervcntion program or somc other factor
that was affeering the community independent of the intervention pro-
gram itself, such as broad media coverage of a particularly savage beating
of a child and, in the aftermadl, considerable additioncd media coverage of
how patents can cope with their urges [() injure their children. - QU;:llirative researchers who come to know an organizati')ll thoroughly
Qualitative \vork in this instance could foclls 011 such impacts as the
outreach efforts or the program to arrr:lct pare!1ts who had previously
abused their children; effons to reach parents who arc seeking help [0
build better skills in working \vith their children; patterns and trends in
child abuse as discussed by school teachers, day care providers, and others
who have ongoing and consistent COlHacr with children; and whether and
hm\' parents arc now coping with the stresses that might cause them to
abuse their children.
The above discussion also generates an additional area in \vhich qualita-
tive work can assist at this stage of the policy cycle. It is the close-in and
intensive familiarity \vith the problem or condition that comes from con-
ducting qualitative work that would allow the researcher to make judg-
ments on whether the situation is of a magnitude and nature that further
action is necessary. If the study indicates that the problem or condition is
diminishing in severity and prevalence, then further funding of a program-
mmic response may not: be necessary. As a contrary example, the data from
qualitative work may suggest that the condition has changed directions-
that is, moved to a new target popubtion-and a refocusing of the pro-
gram is necessary if it is to be responsive.
Social conJitions do not remain static, and the realization that the char-
acteristics of a condition can change necessitates periodic reexamination
of the original policy intent (policy formulation). Qualitative researchers
can position themselves so that they can closel y monitor the ongoing char-
acteristics of a condition. \"X1ith this firsthand and close-in information ,
634
thev are well suited to suggest any necessary changes w both the policy
for;l1ulation and implementation strategies for subsequeIl! intervention
effons.
The t],ird ;nfOrl11Cltion need [It this srnge of the policy cyd~ \,,-,here quali-
tative work can be of direct use comes with the focus on accouIHability.
Here qualitative work can address concerns of management supervision,
leadership of the organization with clear goals in mind, the attention to
processes and proccdures that would strengthen the capaci.-y of the orga-
nization to implement the policy initiative effectively, tht' use of data-
based decision making, and the degree of alignment or congruence be-
tween the leadership and the staff. All of these issues speak ;:Iircerly to the
capacity of ,111 organization to mobilize itsel f to provide effecti ve service to
its customers. If the organization is not positioned to do so, then there are
clear issues of accountability that rest v.;ith the leadership.
and from the inside \vill he in a unique position from which to address the
treatment and training of staff, reasons for attrition and low morale, the
service-oriented philosophy (or lack of it:) among the staff andlcadership,
the beliefs of the staff in the viability and worthiness of the program to
address the problem, the quality ,-md quantity of information used \vithin
the program for decision making, and the like. These are true qualitative
dimensions of organizational life. It is essential that these be swdicd if
judgments are to be made on the efficiency and effectiveness of any partic-
ular programmatic strategy. These judgments become cemral to subse-
quent decisions on the potential selection of a policy tcol that would
require a similar program intervention.
There arc clear concerns of management accountability that must be
discussed and assessed whenever programs are to be funded anew or redi-
reered. Some of these concerns deal directly with impacts on the problem
or condition, whereas others focus on the internal order and logic of the
organization itself. Stated differently, it is important during the account-
ability phase to detcrmine the degree to which any chan~es in the con-
dition or problem can be directly attributed to the program and whether
the program optimized or suboptimized the impact it had. Likewise, it is
important to ascertain \-vbether the presence (or absence) of Jny doc-
umented impacts is the result of the coherence of rhe policy formulation
or the nature of program implemcntation. Finding that instance where
coherent and robust policy initiatives are operationalized \vithin a welJ-
managed organization necessitates the complex asses~mcnt of what
635
 impacts can be Jrtribured [() the policy and what to its successful im-
plemcm3tion. QuaJitacive research h<ls a perspective on how to under-
rake rhis kind of assessment rhar other research approaches do nor and
for lvhich dlt mher :1[lpfOilc!leq \vould have to rely heavily on proxy
measures.
'" Policy Tools
The analysis thus far has focllsed on the naturc of the policy cycle and how
each phase of the cycle has different information requircments for policy
makers and program managers. The effort has been to document how qual-
itative research call play al1 active and positive role in answering the infor-
mation nceds at each of these phases and for both the policy makers and the
program managers. In this section, the attention shifts to a focus on what
arc termed policy 100/5.
Such 3n emphasis is important because a deeper understanding of the
tools avaibble w government and how each can be more or Jess effectively
llsed to achieve policy objectives can clearly inform all three stages of the
policy cycle. Key to the eFforts in policy formulation is the selection of an
appropriate tool-be it a grant, a subsidy, a tax credit, a loan, a new regula-
tion, the creation of a government-sponsored enterprise, or the provision
of direct services, to name but 7 of the more than 30 tools currently used
by government.
The selection of one rool rather than another is a policy choice for
which few guiding data are available. Further, research to help policy mak-
ers in this regard is extremely sparse. Policy makers decide either based on
past experience with a tool ("\X1e used tax credits before, let's use tax cred-
its again") or because they have a clear proclivity for or against a particular
tool (conservatives would resist direct govcrnment services and seek
instead a tool that locaces the activity in the private sector, c.g., grants for
the construction of public housing or the privatization of all concessions in
national parks). Ie is safe to assert that neither qualitative nor quantitative
researchers have shown much interest in this area. Beyond the \vorles of
Linder (1988), Linder and Peters (1984, 1989), May (1981), and Salamon
(1981,1989), there is not much research, either theoretical or empirical,
to be cited.
\Vhat folJO\vs is an effort to identify four areas where qualitative work
could be highly valuable to discussions regarding policy rools. For each of
636
these areas, there is at present a nearly complete research void. It should be
stressed that the shan discussion to follow is not meant to he a definitive
statement on how qualitative work can address the information needs of
policy makers as tbey choose among (Ools, nor is it the ddir:.itive research
agenda on the strengths and weaknesses of differenr tools.
It needs to be restated that few researchers of any persuasion have
moved into this difficult bm highly policy-relevant area. The reasons for
this hesitancy are oLltside the bounds of this discussion, but it is clear that
the policy analysis and research communities have, with few exceptions,
steered wide of this POft of inquiry. Building primarily on the works of
Linder, Peters, and Salamon, what follows is offered as a modest agenda
for those qualitative researchers who are interested in exploring new and
untested ways of involving qualitative work within the policy arena. A
more elaborate and detailed research agenda in this area is still well over
the horizon.
As noted, four areas amenable to qualitative study will be briefly dis-
cussed. These are resource intensiveness, rargeting, insLtLltional con-
straints, and political risks. The tentativeness of this proposal has to be
stressed yet again. There may well be multiple other ways in which to
frame the qualitative srudy of policy tools. \X hat follmvs here is predicated
'
on the previous discussion regarding the policy cycle. The framework for
the qualitative study of policy tools is essentially a matrix analysis,
whereby each of these four areas can be studied in each of the three phases
of the p~licy cycle. All 12 combinations \vill nor be individually addressed
here; rather, the focus v.liB be on the four broad areas that can help to clar-
ify the trade-offs among tools.
Resource illtellsiuelless refers to the constellation of concerns involving
the complexity of the operations, the relative costliness of different
options, and the degree of administrative burden that different tools place
on organizations. Tools vary widely in their complexity, their demands on
organiz.ations for technical expertise to administer and manage, their
direct and indirect costs by sector, and the degree to which they are direct
or indirect in their intent. And just to complicate matters m,Jre, the mix of
these concerns for any given rool will shift as one moves from one phase of
the policy cycle to another. Keeping the financial costs Ie-wand federal
involvemem to a minimum, for example, may be high priorities in Wash-
ington during the policy formulation stage, bur these wiE also have the
consequences during the policy implementation stage of serving few of the
eligible target population, adding complexity through mandated state
637

administration, and reducing direct impaets.1'vlanaging roxic waste clean-
ups is but one example that is somewh~H parallel to this brief scenario.
For qualitative researchers, the challenges here arc multiple, not the
least because they would necessitate more direct attention to org:ln;7,~1
ti(~I1~1 <:lIlalpis. Bur [here IS Jlso the de"lf opportunity to ask questions
wlthll1 organizations and to assess organizational capacity in ways that
have not traditionally been done. Administrative burden has not been a
topic of much (if any) qualitative research, but it is a vcry real consider-
ation in the policy arena. Learning more of how to conceptualize this con-
ce~n, how it is understood at various levels of government and within the
prIvate sector, and how different tools vary in this regard would be of con-
s~derab!e interest to policy makers in departments as well as those respon-
SIble for regulator and administrative oversight in organizations such as
the Office of lVIanagemcnt and Budget in the \X7hite House.
At present, ::I concept such as administrative burden is ill defined and
subject to widely varying inrcrpretations. In the absence of any systematic
research, one person's definition and experience with "administrative
burden" is as good as any other person's-Jnd maybe berter if he or she
has more institutional or organizational influence. Additional cxamples
concerning such concepts as "operational complexity" and "instiwtional
capacity" arc readily apparenr.
Target;,zg refers to the capacity of the policy tool to be aimed at particu-
~ar populations, problems, or institutions for whom the rool is primarily
Intended. A tool that, for example, sceks to help homeless persons who arc
mentally ill and also veterans would be highly targeted. Such;} tool would
be differentiated from a tool th8t is either diffuse or low in target specific-
ity, for example, a tax credit for the interest earned in individual retire~
Il1ent accounts.
There are several key aspects of the targeting issue for a policy rool that
gualitative researchers could address. First, there is the matter of the preci-
sion of the targeting. Qualitative researchers, in reference to the example
just given, could help policy makers work through the strategies and defi-
nitional problems inherent ifI determining Vl/110 is or is not homeless, who
has or has not been diagnosed as mentally ill, and how to screen homeless
veterans for service when documentation, service records, and so on are
all likely to be lost or when persons simply cannot remember their own
names.
A second aspect of targeting in seJecting a policy tool is that of the
amenability of the tool to adjustment and fine tuning. If the characteristics
638
"",,-,C' , ..... u,~ <' 'c , u,,,.. y "I u'--'='':>':> flU' ~UU'f<U<IVC "c-,><""",--"
of the target population start to change, can the rool be adjusted to
respond to this change? Flexibility in some instances would be highly
desirable, whereas in others it may be irrelevant. For example, it \vould be
btudkiul tV ChU05C il pulicy instfumtIll tllat responds to lluctliJ(]ons Jnd
variations in the refugee populations coming inro rhe Un ired Stares,
whereas it would be unnecessary in the instance of an entitlement program
for which age is the only criterion for access to services.
Qualitative studies of different populations targeted by tools and the
need (or Jack thereof) of specificity in the targeting would be highly useful
in policy formulation. There is also the opportunity in thi~ area to explore
whether those who have been targeted by a program believe this to be the
case. Establishing community mental health centers could have some in
the target population coming because of the "community health" empha-
sis, others coming for the "menral health" emphasis, and still others not
showing up at all because they are not cerwin whose community is being
referred or because they would never want anyone in their own neighbor-
hood to know they ha\;e memal health problems. Linkin:s services to tar-
get populations in the absence of sllch qualitative informa;:ion suggests i~1
mediately the vulnerability and precariousness of presuming to establish
service c'enters witham the dewiled knowledge of the populations for
whom the effort is intended.
The example of community mental health centers leac.s to a third con-
sideration in the targeting area-that of adaptability across uses. Can com-
munity mental health centers also serve other needs of the designated pop-
ulatio~, for example, nutrition and education, as well as serve as centers
for entitely other target populations \vho are in the same residential vicin-
ity? Can they serve as centers for the elderly, for latchkey chi.ld.r~n, for
infant nutrition programs, and so on? The issue is one (If flexlbI1Jty and
acceptance as well as neutrality in the perceptions of ;:he other target
groups. There may be groups \vho would not want to .come to a mental
health center, but who would be guite pleased to meet In a church or at a
school. Gaining insight on these matters is clearly impocant as decisions
are made on the location and mix of community services to be offered at
anyone location. Qualitative studies on these issues can inform policy
m~kers and program managers in ways that will clearly affect the success
or failure of different strategies.
Institutional capacit)' refers to the ability of the institution to deliver on
the tasks that have been delegated to it. \'xlhen a policy option clearly relies
on a single institution to achieve certain objectives-for example, using
639

tb~ public schools as the vehicle to teach English to non-English-speaking
chJ1drcn-there has to be some degree of certainty that the institution has
t!le capacity to do so. Countless experlcnces with different policy initia-
t1t!e~ h!1ve !:h~I.Vl1 t;me ::'l11.d "f':":n d-l:ll "'Orn_", :ns\;nn:c>ns s;,nply did nvt
have, at the nme, the capacity to do what Was expected of them.
Further, there can be constraints placed on the institution that make it
difficult if not impossible for the objective to be achieved. In addition to
~!1e I~~re.r.e~dily anticipate~1~onstraints of funding, staff availability, qual-
Ily ollacllmes, and low polmcal support, there are also constrai!1ts associ-
ated with the degree of intrusiveness the institution can exercise as well as
the level of coerciveness allowed. The hesitancy of policy makers to allow
intrusive efforts by the Internal Revenue Service to coliect unpaid taxes
has a dear impact on the ability of the organization to do so. The same Can
be said with respect to the IRS on the matter of coerciveness. Policy makers
have simply decided to keep some organizations more constrained than
ochers in carrying out their functions, for fcar of abuse. Policy tools that
have to rely on voluntary compliance or are framed to have an indirect
effect fa,ce ~onsrraints different from those where these do not apply.
QualItatlve research into the domain of institutional constraints and
how it is t~at these constraints play Out in the relation of the organization
to the fulfIllment of its mission is not, to my knowledge, now being done.
It may be argued that it is not necessary, as the constraint dimension for
any policy tool is too removed from research influence. That is. any con-
straints on an organization are more philosophical and ideolo~icai than
operationaL let the issue of institutional capacity and what docs or does
~ot hinder the ability of the organization to achieve its stated objectives is
Important to understand explicitly. 1 policy makers establish the parame-
~crs around an organization to the degree that it can never clearly achieve
Its goal (e.g., the IRS and unpaid back taxes), then there is a built*in level of
failure that ought not be ignored and for which the institution should not
be held accountable.
P~lit~cal risk is the fourth dimension of the study of policy tools where
qualItative research can directly contribute. Here the issues cluster around
c.ancerns of unanticipated risk, chances of failure, and timing. The selec-
non of a policy tool is made with some outcome in mind-either direct or
indirccr. Yet there is always the possibility of unanticipatcd Outcomcs-
again either direct or indirect. The selection of a tool necessarily has to
ta~(e into accounr the risk of unknown outcomes and how thes'e might
<lfreer the success of the policy.
640
-- ------ .~-- -'=' -
Quaiitative research, by the n:ltme of its being longitudinal, done in
naturalistic settings, and focused on the constructions of meaning devel-
oped by participants, is in a unique position from which to assess the possi-
bi1;rr of tools ]'''V;tl6 tI,C ;n.l!"[\CI-S ;nrcnded by policy m::tkerr;_ Low fisk of
unknown outcomes-for eX:1l11ple, in increasing the securi-:y at U.S. fed-
eral courthouses-eliminates some level of uncertainty from the decision
that does not happen when the risk of unknown outcomes is quite high,
such as moving to year-round school schedules or as was learned when the
movement to cleinstitutionalize the mentally ill resulted in tens of thou-
sands of mentally ill persons being left on their own with no means of sup-
port or treatment.
One other aspect of the political risk faCtor that qualitative research can
adclress is the sustainability of the policy initiative. Close-in studies of the
operational life of a policy initiative can gain a perspective en the commit-
ment of those involved, their belief in the worthiness of ,he effort, the
amount of political support they are or are not engendning, and the
receptivity of the urget population to the effon. If all these indicators arc
deCidedly negative, then the sustainabiliry of the initiative is surely low,
It is difficult to achieve success in policy efforts in the best of circum-
stances; it is that much harder when all the indicators point: J1 the opposite
direction. Qualitative research should have a distinCi window from which
to judge matters of political risk. Understanding of the participants, will-
ingness to assume the causal linkage posited in the policy itself, and the
degree of risk of unknown outcomes all influence the likelihood that any
policy tool \vill achieve its intended results.
<> Concluding Observations
In reviewing this assessment of the contributions of qualitalive \'i'ork to the
policy process, it is apparent that the contributions arc more in the realm of
the potential than the actual. There is no broad-based and :;llstained tradi-
tion within contemporary social science of focusing qllaliotive work spe-
cifically on policy issues, especially given the real time con:;traints that the
policy process necessitates. 'ret it is also cle3f that the opportunities are
multiple for such contributions to be nude. The issue is chiefly one of how
to link those in the rese3fch and. 3cademic communities who are knO\vl-
edgcable in conducting qualitative research studies to those in the policy
arena \\'ho C3n commission such work and who will make use of the find-
ings. The analysis of different strategies for building these linkages would
641
 ""''-'''''''''''''':::1 tI,<=", v""'y' 'V\.-<="-O>-" . " " , \,:''-'U'''UUYC 'U;;-o>C'..I'l.."

require a separate chapter; suffice it to say here that much hard thinking
Jnd numerous exploratory effons will be required for the potential to
become the actual. The issues of institutional cultures, academic re\vard
.l:y!.:[em[.;, puhliC:.1I:;on n:9u;rcmcnts, funJing sources, and methodological
limiwtions are but five among many that will have to be addressed if the
linkages are to be built. And even beyond the resolution of (or at least the
careful thinking about) these issues is the fundamental question of whether
there is the will to bring qualitative work directly [11to the policy arena.
IvIuch of what has been written here will remain speculative unless and
until there is some consensus among the practitioners of qualitative re-
search that making this transition is \vorrhwhile. The policy community is,
I believe, ready for and would be receptive to anything those in the qualita-
ti ve research community could offer, should they choose to make the effort
to do so.
'" Note
1. I W;lI1t to stress eddy on rll;][ in this chapter I will flor seek to develop distinctions
among various cOl"H'cmionally used terms for Cju:lfitative research. Thus, in the pages that
follow, rerms such as ijllil!ifatiuc //Iork, qUil/ita/iuc research, and ql/alitative methods will all
be llSed to denote the same frame of reference. I most frequently use dlC term that appears in
the title of this handbuok, ijllafitaliue research. [leave it 10 other authors in this volume to
develop those distinctions as appropriate. I would also notc, in defense of not trying t'O spec-
ify in lllllch detail JUSt exactly wh:l[ the meaning is hehind the lISC of anyone of these terms,
r!J,n early reviewers of this ch:Jpcer suggested <H IC<lst fOUT other terms I might use in lieu of
those I have. These terms included lIatllralis/ic, constructionist, interpretive, and et/;-
llogti7!J!Jies. I am sure th:n the delineation of distinctions has 3n important p13ce in this book;
it is just not my intent to do so here.
J 31so want to note early on that I am not going ro try to differentiate among various qual-
itative C!;Hil collecrion strategies, or means of analysis, as to their particular spheres of poten-
tial influence. Thus in this chapter I will nor try to indicate what policy relevance or influ-
ence one might expect from case Studies (and there ate multiple variations in this single area
alone) in contrast, [or eX3mple, to Illultimethod studies. IVly intent is to place qualitative
work broadly withill the policy arena, not to develop a prescriptive set o[ categories abOUt
which methods or modes of analysis are likely to le3d to what types of influence.
,;- References
Behn, R. D. (1988). IVlanaging by groping along. Joumal of Polic)' Analysis and
1l'1mwgemellt, 7(4).
CheJimsky, E. (1982). f\'Iaking evaluations rclev<:Jnt to congressional needs. GAO
Rel!iel/.~ 17Ct).
Chelimsky, E. (198.1). Old p,merns and new directions in progr~lIn evaluation. In
E. Che!imsky (Eel.), Progra!J/ elw{,wtioll: Pdtlems alrd directions. Washing-
ton, DC: American Sociel"j' for Public Administration.
Guba, E. G. (1984). The effect o( d..,(initions o[ poli<.:y un the naturc dilL! outcomcs
of policy analysis. Educational Leadership, 42(2).
H8rgrove, E. (1985). The missing link: The stud)' of the ill1plel/l?lftatioll of social
policy. Washington, DC: Urban Institure Press.
JanO\vitz,IVI. (1971). Sociological me/hods alld SOcidl policy. New York: General
Learning Press.
Lindblom, C. E. (J 968). The policy lIlahing process. Englewood Cliffs, N]: Prentice
Hall.
Lincler, S. H. (1988)..T\'bnaging support for social research and development: Re-
search goals, risk, and policy instruments. }emmal of Policy Allolysis o/fd
t'/Imragemelll, 7(4).
Linder, S. H., & Peters, B. G. (1984). From social theory to policy design.Iortrlwl of
PI/olic }Jo/icy, 4(3).
Linder, S. H., & Peters, B. G. (1989). Instruments of government: Perceptions and
contexts. Journal of Public Policy, 9(1).
IVlay, P. ]. (1981). Hims for crafting alternative policies. Policy Analysis, 7(2).
Nakamura, R. T., & Smallwood, F. (1980). The politics of policy illlplementation.
New York: St. r\'laflin's.
Patron, I.vI. Q. (1988). Qualitative emll/ation and research methods (2nd ed.).
Newbury Park, CA: Sage.
Pressman,.1. L., & \X'iJdavsky, A. (1984). Implementation Ord ea.). Berkeley: Uni-
versity of California Press.
Rist, R. C. (1980). Blitzkrieg cehnogr<.lphy: On the rransformatior. of a method into
a movement. Educatiollt11 Researche/.; 9(2).
Rist, R. C. (1989). Managemem accounwbiliry: The signals senr by auditing and
evaluaeion. JOII171<11 of Public Policy, 9(3).
Rist, R. C. (Ed.). (1990). Progm1l1 cualrwtiolll1l1d the IlUllfagemeJlt ofgOl/emment:
Paltems alld prospects across eight natiolls. New Brunswick, N]: Transaction
Books.
Rist, R. C. (1993). Program evalt.l<.ltion in the United States General Accounting
Office: Reflections on question formulation and utilizacion. In R. Conner et
al. (Eels.), Adut1/1cIng public fJOlicy wall/alion: Learning flOm intematiollal
experiences. A.. msterdam: Elsevier.
Salamon, L Ivt (1981). Rethinking public management: Third-parry government
and the changing forms of government anion. PI/b1ic Poli,:y, 29(3).
Salamon, L. IV1. (1989). Beyond priuatization: The tools of go~'emmcllt actioll.
\Xiashingron, DC: Urban Institute Press.
Smieh, J. A. (1991). '17JC idea brohers: Tbinh tal1hs and the rise oftho: new policy elite.
New York: Free Press.

642 643
Weiss, C. H. (19E:2). Policy research in the context of diffuse decisioll making. In
R. C. Rist (Ed.), Po/ic}' stl/dies rel/iew aWl/wI. Beverly Hills, CA: S;]ge.
\"\?eiss, C. H. (1988). Evaluations for decisions: Is ,1Ilybody there? Docs anybody
carel ElIll/lIlliinll 1'1'Ilrlia. 9IIL
Yin, R. K. (19HS). Studying the implement;<tion of public progr"lms. In \'v. WilHams
(Ed.), Stl/dying implementation. Chatham, NJ: Chatham House.

'4- Suggested Further Readings

Bcmelmans-Vielec,1v1. L., Rlst, R. c., & Veelung, E. (Eds.). (1998). Carrots, sticks,
ond Ser1l1011S: Polic)' instmmcllls and fheir evalllatioll. New 8nlllswick, NJ:
Transaction Books.
Elmore, R. (1987). Instrumems and strategy in public policy. Policy Studies Rel-'icw,
7(1),6378.
Gray, A., Jenkins, 8., & Segswonh, B. (Eels.). (j 992). BlIdget lng, <1l1ditillg alld eurd-
[fatioll: htllCliolls aud illtegratioll ill SCI/ell gOl'cmJllellts. New Brunswick,
N.f: Transaction.
Hooel, C. C. (1986). The fools ofgouetlllllellt. Chatham, NJ: Chatham House.
Leellw, F. L., Rist, R. C., & Sonnichsen, R. (Eels.). (j 994). Call gOl'erlll11el1tS leam?
Compara/iue perspectives all evaluation alld OIgmlhatiollalleaming. New
Brunswick, N]: Transaction.
Mayne,J., & Zapico-Goni, E. (Eels.). (1997). iVfOllitorillgperfomulIlce ill the public
sector: FlIture directiolls (i'om illternatiollal experience. New Brunswick, NJ:
Transaction Books.
Rist, R. C. (1990). Mamgemcnt accountability: The signals scnt by aueliting and
evaluation. Jourllal of Public Policy, 9, 355-369.
Toulc1l1ollele, .f., & Rieper, O. (Eels.). (1997). Politics and practices of intergovern-
mental evaluatioll. New Brunswick, N.J: Trans;:Jction Books.
Vedung, E. (1997). Pltblic polic)' alId program emlltatiou. New Brunswick, N.J:
Transaction Books.
.. Chapter 2
Gubrilllll, J. F., & Holstein, J. 1\. (Eels.) (2002). Handbook of iltteruiew research:
COli text alld methods. ThouS,lllcl Oaks: Sage.
.. Chapter 3
Babbie, E. (1998). Obseruillg ourselves: Essa)'s ill socit11 research. Prospect Heights,
IL: \"\/avelanel.
Evans,]., & Hall, S. (Eds.) (1999). The visual Cltltllre reader. Ln~dl)~l: Sage..
Gold, R. L. (1997). The ethnographic method in sociology. Qualtta!we Illqfllry, 3,
388-402.
SchensuL S. L.. Schensul,j. R., & LeCompte, M. D. (1999). Essenti:ll ethnographic
m~thod~: Observations, interviews, and questionnaires. Inl Scheusel & J"l.
D. LeCompte (Ecls.). Ethnographer's Toolbt, Vol. 2. \"I/;]lnut Creek, CA:
AltaMira.
Wolcott, H. F. (1995). The art offieldworh. \X'alnllt Creek, CA; AI:alvIira.

<Co Chapter 5

Emmison, M., & Smith, P. (2000). Researching the uisllal. London: Sage.

<Co Chapter 6

Bochner A. P. (2001). Narrative's virtues. QI/alitative Inquiry, 7, "13j-157.

Boe Ilncr,'A ., L& E11'[5 ,C. (700
-
7) Etlmographicallv_ spcahing: i\lItoetfmograph)', liter-
_.'

atllre, alld aeslhetics. Walnut Creek, CA: Alt'aIvIira.

644 645
 ",_ "",_' ,~,,,"" '- n,""'" .... ,,, ..........-'

Goocbll, H, L. (2000). \'(lritil1g the new ethnography. Wal nm Creek, CA: Alta!vlira. $ Chapter 10
Rosel1walJ, G., & Ochberg, R. (1992). Storied liues: The wltllral politics of self-
Greenbaulll, T. L (1998). The halulbooh (or (OCIIS group research, 2nd edition.
1/nderstilnding. New Haven: 1::lle University Press.
Til/m3nn-l-id y, L (2nnll. Rd"'''''' g"" ,w,l,k,;gl,l, [I",J.',,,L,,,,/;,,g !,i..... l-J,ip
Thousand Oaks: Sage,
<1cross sexltal orienta/ioll. Walnut Creek, CA: i\ltalvlira.
$ Chapter 11

. Chapter 7
Bernard, H. R. (2002). Research methods in anthropology, 31'd edition. Thousand
Oaks, CA: Sage.
Dey,1. (1993). Qoalitatiue data analysis: A lIscr (rimdly guide (or social scientists.
London: Routledge.
Krippendorf, K. (1930). Content analysis: Au i/ltroduction to its methodology.
Beverly Hills, CA: Sage.
Miles, Iv1. E., & Huberman, A. M. (1994). QlIalitatiue data mtalysis: An expanded
sourcebook, 2/1d edition. Thousand OOlks, CA: Sage.
SmlU5s, A., & Corbin, J. (1990). Basics o( q/wlitative research: Grounded theory
procedures alld techniques. Newbury Park, CA: Sage.
Weller, S. \XZ, & Romney, A. K. (1988). Systematic data collection. Thousand Oaks,
CA: Sage.
$ Chapter 8
Flick, U. (2002). All introduction to qualitatiue research, 2nd edirion. (See espe-
cially Part 6.) London: Sage.
~ Chapter 9
Agar, Iv!. H.(1996). The pro(essiollal stranger: All ill(orlllal illlrodllCtioli to etlmog-
mphy. Orlando, FL: Academic Press.
Bern;lrd, H. R. (2001). Research methods ill ll11thropology: Qlhllitatile arid qUailti-
tatiue appro,1ches. \Valnut Creek, CA: Alt.liViir:l.
Fettctman, D. IVI. (1998). EIIJllOgraphy: Stet, by step. Thollsand O:llcs, CA: Sage.
LeCompte, M. D., & Schcnsul,J . .1. (1999). Desigllillgallil COlldllctillg etlmogmphic
research. \'Valnm Creek, CA: Altal'vlir,l.
Wolcon, H. F. (1999). EtlJ/logmphy: A luay o( seeing. \Valnu! Creek, CA: AlraIVlir3.
. Chapter 1 2
G3damer, H-G (1995). 'Ji-Ilth alld IIlethod (2nd rev. cd.; J. \Xi e'lI1she,mer & D. G.
Ivlarshall, Trans.). New York: Crossroad.
Hazelrigg, L (1989). Cla;1II5 of lmowledge. Tallahassee, FL: Florida Swre Univer-
sin' Press.
jones, R. (1992). PhysiCS (or /he rest olliS. Lincolnwood, II-: Conrel~rorary Books.
Rorr)', R. (1985). Solidariry or objectivity? In]. Raichman & C. \X1e:it' (Eels.), Post-
analytic philosophy (pp. 3~19). New York: ColumbiCl University Press.
Sheman, N. (1993). E/lgellderings: COllstmctions of kllOlulcdge, {/lIl!J.Jrity ilI/d priv-
ilege. New York: Routledge.
Smir.h, j. (1993). After the demise of empiricislII: Thc problem of judging social and
cducational iI/qui/y. Norwood, N J: Ablex.

Sacks, H. (1992). Lectures all colll'ersatioll (G. Jeffcrson, Ed.). Oxford, UK: . Chapter 1 3
Blackwell.
Bochner, A. P., & Ellis, C. (Eds.) (2002). Ethnogmphically sp:;ahing: AlIto-
Silverman, D. (Ed.). (1997). Qualitative research: Theory, method, and practice.
e!hnogmphy, litemtlll'c, and aesthetics. \'V01lnur Cteek, CA: AltJ..1vlira.
London: Sage.
Silverman, D. (1998). HaTlley Sachs: Social science and conversation allalysis. New
York: Oxford University Press. ~ Chapter 14
Silverman, D. (2000). Doing qllalitatilJe research: A rm1ctical halldbook.. London:
Sage. Beh:lr, R. (1993). Ii'amldled womall: Crossing the border with EspEranzo's stor)'.
Silverman, D. (2001). Illtelprcti1lg qllalitatiue data: ldethods (or analysing tal!:.., tcd Boston: Beacon.
am! interaction, 2nd edition, London: Sage. Bochner, A., & Ellis, C. (Eds.). (2002). Etlmogmphically s!le,1kil/g. Thollsand Oaks,
ten Have, l~ (1998). Doing conversation analysis: 11 !Jrt1cticalguide. London: Sage. CA: Sage,

646 647
 Schwandt, T. A. (forthconling). El-'aluation p1'iletice reconsidered. Balrimore, IvID:
Dcnzin, N. (J997). htterpretiue ethnograph)': Eth/lographic practices for the 21st
cell/ury. Thollsand 0;lk5, C1\: Sage. Peter Lang.
Ell is, c., & Bochner, A. (Eds). (1996). Composillg i!tlJlZogmpby: Allcmatil!e forms
O(Qlffllitl1l;I!(lwritiHg. \l{l"lnut Creek, Ct\: AlraMirJ. i> Chapter 17
Jones, S. H. (1998). Kaleidoscope notes: \Y/ritillg women '5 music flirt! org<1llizalional
w!tllrc. \\'?all1lJt Creek, CA: AltaMira. ' 000) Voice-~ 0/- the Iloor: Call anyone hcar liS? Oxford, UK:
N arayan, 0 . (Ed) (
~ < ., -'-

Luber, l~, & Smithies, C. (1997). Troubling the a/lgels: \\'!OIllCIl!ivil1g with HIV/ Oxford University Prcss.
AiDS. BOLllder, CO: \\1estvie'lv. Risr R. (1999). Linking cvalu;j[ion utilizarion and govern'c1llcc: Fundamental ch~l_-
, Icnges for coumnes , bw'Id'Il1g evCl I ' "ulpa..
u;.H!on CI'[\, In R Bo)de & D. LemaIre
. _ _ _ .
R..ichardson, L (1990). Writing strategies: Reaching diverse dudiences. Thousand
Oaks, CA: Sage. (Eds.) Building effectiFe ellalaation capacity: Lessons Irol1l fmlet/Ct'. Ncw
Richardson, L. (1997). Fields of !lld)': COllstmeting an academic life. New Bruns- Brunswick, NJ: Transacrion Books. .
wick, NJ: Rutgers University Press. Roche, C. (1999). Impact assessments for deuc!opll1ellt agencies: Lel1tll1l1g to /lallte
Trinh 1'. M. (1989). \Y!01l1dlf, IUI/h'e, other: \\/rithlg postcolonialily and femiitislll. chmrge. Oxford, UK: Oxfalll Publishers.
Bloomington, IN: Indian<l Universiry Press.
Van ~v[<l:anen, J. (1988). Tilles o( the ficld: On wriling ethnography. Chic<lgo, IL:
Universiry of Chicago Press.

,. Chapter 1 5

Br<ldy, /. (2002). Thc tillte at Darwin's reef: Poetic cxplorations ill anthropology alld
histo!"y. \X!;dnur Creek, CA: Alralvfira.
Denzin, N. K., & Lincoln, "{ S, (Eds.). (1002). The qualitalive inquiry reader. Thou-
s,Jnd Oaks, C1\: S:Jge.
Hirsch, E. (1999). HOll! to read a IlOellI and (all ill laue with poetry. New York:
Harcourt Brace.
Koch, K. (1998). Mahillg )'0111' O/Ult days: The pleasures of reading and turfting
fnlet,.y. New 'rode: Simon <lnd Schuster.
Nowak,1V1. (2000). Reuenal1ts. IVlinneapoJis: Coffee House Press.

i> Chapter 1 6

House, E. R., LV. I-lowe, K. R. (1999). Val1/cs ill cual"atiof1 and social research.
Thousand Oaks, CA: Sage.
Kushner, S. (lOaD). Personalizing eua/uation. London: Sage.
IVlerrens, D. M. (1999). Indusive evaluarion: Implicarions of tmnsforrnative theory
for c\'a[udtion. American j01l/7uz! of Evalltation, 10, 1-14.
Patron, 1,,1. Q. (2002). QIl(/!iteltivc research and euaI1latio!l1l1etlJods, third edition.
Thous<wd O<lb, CA: Sage.
RY<ln, K. R., & DeSrefano, L. (Eds.). (2000). Evaluatioll as a dcmocratic process:
Promoting inc/usioN, dialogue, ilnd deliberatioll. Nnv Directions for Evalua-
tion no. 85. San Francisco: Jossey~Bass.
649
648
 Author Index

Abelson, R. P., IHO Austin,.J. L., 354

AbmJ., T. A., 595, GOO Azgnoli. L. J., 116
Abrah'lltlS, R. D., 555
Abramson, J~ R., 380
Abu-Lughod, L, 121, '-13,210 Lb, lvl., 5(H
Acker, J., 374 Babhie, E., 64, 82, 85
Adams, R. N., 76 B:lff, S. ]., 510
Adler, p" J07, 108, 114, 143,109, 113 ['>:lb.;!", c., 95
Adler, P. A., 107, J 08, 114, 143,109,213 Baker, H., Jr., 460
Agar, M., 55, 128, 275, 279, 2HJ, 292, Bakhtin, Ivl. Ivl., 202
293,398,399,410 B:lles, R. F., 348
Agger, B., 504, 507 B:l!z:lc, H. de, 503
Akeroyd, A. v., 55 Banks, A., 232, 515, 549
Albanese, J. P., IH4 B:lnks, S. P., 232, 515, 549
Alben, S. 1\iL, 162 Bar;lka, A., 460, 479
Alexic, S., 459 Barik, G., 127
Allen, J. T., 169 Batkin, S., 27S
Althcidc, D. L, 108 Barlcy, N., 510
Altork, K., 122 Barnett, G. A., 270, 27)
Anderson, B. A., 380 Barnett, J-1. G., 126,394
Anderson, N., 66 B:ltone, T. E., 599, ii03
Angrosino,lv1. V, 1] 6, 119, 124, 128, t'>3rretr, S. R., 143
134,138,231,511 B:ltry, K., 374
Anzaldua, G., 202, 479 Barth, .1-, '145
Araujo, L., 277 Banhes, R., 202, 5S1, 552
Argyris, D., 404 Bartlett, P. E, 397
Arnowitz, S., 2. 7 B35S0, K., 54 7
Artieri, G., 18] Bateson, G., 33,194
Ashton-Jones, E., 130 Bauman, R., 546
Ashton-Voyoucalos, S., 282 Bechtel, R. ]., 285
Atkinson, P., 54, 56, 62, 63, 93,219,222, Bechtel, W, 163
313,332,336,3<11,347,348,349, Becker, H. 5., 6,14,17,23,66,77,19"1,
356 273,279,286,510,528

651
 " ~'" . , .... " , ' ..... \e::V,.... ...., 'r-\" V '- 1Y1r-\1 '-"'r-\'-J

Behar, R., 2, 28, 84,116,117,120,126, Brodkey, L, 504

134,202,212,470,510,510 Chippindale, c., 165 Crol1hach, L. f., 592, 609, 610
Brokensha, D., 397
Belln, R. I)., 632 Chishollll, H.., 442 Cruikshank, .I., 470
Brouwer, lvI., 183
Be!!, .I., 110 ChO\v, E. N., 369 Curran, C. E., 139
Brown, K. lVI. , 511
nf.lldlJr. It. 10 Chriti:ln 1 B. T. 1 4(-)0
tlenmayor, R., 96, 369, 374, 375", 383 \jl1fl~[Jilllb, \j, G" "ttiJ, "!60, "II\]
Broyard, 11.., 22S
Hennert,]. \\~, 39'1 Church, K., 210, 510, 511 Dale, T., 172
Bruner, E.IV1., 26, 212, 470, 474, 475, Cicirclli, V G., 595
Benson, P., 218 Daly, K., 510
507, SlO
Berc!son, E., 277, 348 Cicourel, A. \~, 22, 62, 67, 71, 78, 92, 94, D':\nadradc, R., 28, 263. 264, 272, 280,
Bruner, J" 214, 218,551,554
Berg, B. L, J 72 95,348 281
BllclchoJdt, D., 348
Berger, D., 215 Cbndinin, D. ]., 21 (J Danielson, W: A., 269
Buford, B., 218
Bernard, H. R., -108, 261, 263, 270, 277, Clark, C. c., 283 Danowski, J. :1,., 270
Hulmer, M., 275, 278, 286
279,282,284,286,289,293,401, Clark,]. A., 311 D;lnon, 1'\'1., 269
Burbee, L., 263
406, '~10 Clifford, ]., 2, 26, 112, 502, 503, 507, Dash, L.. 459, 462
Burke, L 558
Bernstein, R., n 1 543,548 D:wies, B., 511, 5 Hi, 517, 522, 532
Burkhart, G., 123
Bern.:man, G. D., 411 Clough, P. T., 21, 26, 27, 8'1, S8, 202, Davis, A. \'., 459, '173, '17l). 4S0
Burton, D. Iv1., 269
Bessinger,.f. B., 269 211,466, A70, 503, 504 D;lVis, M., i23
Bunon, Ivi. L., 284
Best,.J. A., 275 Cochrane, G., 408 0;1\\'5011, P, '175
Butler, j., 467
Bettini, L., 275 Coffey,1\., 54, 56, 3B, 332, 3411, ,1'19 Deck, t\., 209, 2! I, 21'1
[-Iutler, S., 510
Bhaskar, R., 432 Cohen, A., 210, 212 D('esc, .1.,277
Bickm!ln, L., 609 Cohen,]., 283 Degh, L., 459, 4li5
Hilden, S. 1(, 279, 311 Cohen-Cruz, j., 459, 469 Demhn, R., lIS
Cahill, I... 5., 139
Birringer, ]., 467 Colby, B. N., 285 Derner::Irh, Ivl. J., 71
Callaway, 1-1., 218 Cole, D., lOS
Blackwood, E., 126 Denison,.J., 45~
Campbell, D. T., 22, 157,28'1 Cole:;,./. lVI., 169
Bloch, M., 164 DClllin, N. K., 12,22,71;, HI, 82, 86, %,
Cannel, C. F., 70, 85
Bloor, 1V1., 178,286 Collingwood, R. G., I (~9, 554 99, 121, 129, 132, 134, 166, 168,
Cannon, L. \',377
Bochner, A. I~, 3, 28, 210, 211, 213, 216, Collins, P. H., 85, 369, :no, 376, 382, 170,195,209, 2JO, 214, 218, 234,
C;lrace1Ji, V.I., 604 46B, 473, 479
217,218,224,470,511,514,516, 2(\6,349,368,373, 45i, 452, 459,
Carey, J. \',12,277
520,53],549 Compte, IvI. D., 311 -161,467,470, 4S0, 50'1, 507, 510,
Carley, K., 271, 272, 273
Boelen, \XI. A. IV1., 89 Conklin, H., 263 517,521,562
Carmines, E. G., 284
Bogardus, E. S., 71 Connelly, r-. lVI., 210 Derrida, j., 88, 156,202,222,505
C;lrr, D., 120
Bogdan, R. C, 22, 279, 311 ConCluergood, D., 467, 469, 470, '171 DeSh:lZcr, Iv!. K.. 516
Carr, \XI. L., 5%
Bolton, R., 123, 130 Converse, j. ]\11., 65, 67, 69, 70 Desvousges, \\,~ H., 7!
Cuspecken, P. E, 471
Borgmti, S. P., 261, 262, 271, 287, 289 Cook, D. A., 5, 6, 7 DeVault, Ivl., 210
C;lrver, R., 487
Boster, ]., 262, 264 Cook,]. A., 374, 517 DeWalt, B. R., 3%, 402, 4 JO
Cassell, j., 135
Boster, J. S., 402 Cook, T. D.. 284, 607, 608 Deyhle, D., 460, 479
Casson, R., 280
Bourdiell, P., ]64,555 Cooke, J., 46S Dey, I., 276, 278, 279
Ceglowski, D., 459
Bowen, E. S., 212
Cerne;J, M. M., 408
Copeland, A., 192 Di~mond, 5., 214, 510, 553, 557, 559,
Bowler, L, 93 Corbin,.J., 23, 2713, 279, 280, 32B 560,562
Chakraborty, G., 284 Coriega, CA., 479
Bradburn, N. M., 69, 70, 85 Dienhan, A., 510
Chalfen, R., 192
Bradley, C, 284 COllser, G. T., 226 Dill, B. T., 370
Chambers, E., 390, 408, '109
Brady, L, 210, 218, '159, 468, 470, 510, Cousins,]. B., 610 Dillman, D. A., 97, 'JH
Ch;lmbers, R., '108
543,545,546,551,552,553,555, Cowan, G., 283 Dilthcy, W L., 30
Chapple, E. D., 410
562,563,564,565 Coward, H. G., 555 Dingwall, R., 61, 91, 93
Chariry, A., 459, 464, 465
Braithwaite, lvI., 159 Crabtree, B. F., 276 Divetsi, Ivl., 7, 459, 5 J 0
Charrnaz, K., 279
Br~maman, A., 469 Craig, R. T., 283 Doerfel, J'vl. L., 271
Chavez, L R., 128
Brandes, S., 210 Ctapanz,Hlo, V, 68, gB, 214, 556 Doig, I., 478
Chelimsky, E., 620, 62'1 Crawford, L., 210
Brewer, D. D., 128 Donley, L., 159
Cllen, H. -T., 609
Briggs, C L, 399 Crawford, l'vl. A., 503 Donmoyer, H.., '171,510
Cherry, K., 510
Brin, T., 27,1 Cressey, D. R., 27R, 286 Dormer, S., 516, 517, 522, 532
Chertyholmes, c., 431
Crites, S., 220 Dorst,]. D.,511,519
652
653
Doucet, 1.., 291, 292 Fidding, N. G., 53, 5,1. 55, 5(" 284, 313, Gatewood, ]. B., 287 Griffin, c., 371
Dougbs, J. D., 67, hi, 80, 84, 86,89,91 318,334,336 Ge3ring, .I., 131 Grossberg, L., 5, 6,10,11
Dove, R., "IS 1,545 Filsread, W J., 22 Geer, B., 14,1.3,278 Cuba, E. G., H, 33,156, 15S, 170, 27S,
Drass, K. 1\., 311 finan. T L.. 40S Geerrz, c., 20, 24, 26, 96,144,217 , '106, 279 1 21l4! 106\ 595, S9K, 60fi, 6U7,
Drc5:,lt:r, \'[ \\(, .'PJfi, 401, '110 Finch, ].,363,568 510, S1~, 5,19, HI, so::: 1, 567
Dreyfm, I-I. L., 552 Pinc, M" B2, 370, 371, 372, :)73, 475, Gdberg, L., 275 Gllbrium, J. E, 62, 63, 64, 91, 92, 94,
Dubisch, J., -J12 511,612 George, A. L., 184 210,3'12,343,345,346,348,349,
Du Bois, \Y: E. n., 65, 475 Finnegan, R., 546 Gerbner, G., 283, 284 352,371,373
Dumom,./. .r.. 212. 556 Fischer,1"1. D., 294 Gergen, K. .J., 122 Gnla, R. 1'''1.,139,142,143
Dunb;lr, c., }r., 459 Fischer, 11,1. IV!.]., 25, 68, 80, 84, 210, Ceria,]. P., 510 Gupta, 1\.,1, 19, 28, lOS, 120, 121
Dunphy, D. c., 277, 2S5 543,549 Cera,]., J 71
Fishkin, S. F, 50'1 Gibson, \\~, lSI
Fiske, IV!., 72 Giddens, A., 164,431 H;lcking, L, 220
E;lrl, L., 610 Fje!!Jmn, S. JVl., 282 Gilbert,.f. R., 275 Hagaman, D., 194
Eddman.1vL. 124 Flahcny, Ivl. G., 218,470, S20 Gilkes, C. T, 370 1-L111e, IV!., 263
Edmondson, R., 5(H fleisher, Ivi. S., 262 Giroux, H. A., 124 Hammersley, M., 284, 406, 421, 433,
Edwards, E., 192 Fleiss,]. L., 123 Gladwin, c., 278, 287 '134,435,436,437,454
Eisner, L, 531 Flick, LJ., 8,15, 99, 517 Gbdwin, H., 282, 287 Hammond, J. L., 12X
Ellis, c.,J, 28,80, LlJ,20'J,210,2Jl, Flores, E. T., 128 Glaser, B. G" 22, 23, 67, J94, 278, 279, Hannerz, U., 12J
213,215,217, lIS, 470, 5 JO, 51 I, no res, T., 553 328,529 Hanson, N., 428, -130, '-136
516,520,522,53J,549 FlucllrLobbal1, c., 135,137,411 Gbssie, H., 161 Harding, S., 202, 374
Ember, C. R., 21lA Fonow, l'vI. Iv1., 374 Glassman, Iv\. ro., 70 l-Jareven, T., 121
Emher, 1"1., 28'-1 FonrOlna, A., 71, 75, 76, 77, 8,1, 90, 36'-1, Glassner, B., lIS, 3,13, 3'15, 3'16 Hargrovc, E., (;30
Erickson, C. L., 166 366,367,368,373 Glenn, E. N., 375 Harman, R. c., 262
Erickson, r., 26,1, 166 Ford, N. A., 488 Glesne, C. E., 510 H:lrpt't, D., 186, 510
Erikson, K. T., 89, 507 roster, G. Iv1., 394 Gluck, S. B., 68, 79, %, 99 Harrington, J. A., 262
Escobar, A., 411 Foucault, Ivl., 202, 342 Goetting, A., 215, 5 I0 Harrington, \\7., 21B, 459, 461, 462, 463,
Espiritu, Y. L., 3{~9, 370 Fox, K. \Z, 124 Goetz, J. P., 311 464
Essevcld, J., 374 Fox,1v1. E, 528 GoHman, E., 469 H<lrris, K J., 408
Ettet~Ltwis, G., 37'-1 Frake, C. 0., 263 Gold, R. L., 112, 113 Harris, \'i/. J., 460, 479
Ewald, \\~, 193 Frank, A., 215, 232,510 Goldberg, N., 528 Hanigan,]. A., 266
Frankel, iVL, 70 Golde. P., 404 Hardey,./. f\., 401
Freeman, D., 77 Goldm,lll, A., 2 I 0 Hartsock, N. C Iv1.. 202
Fabian,]., 549, 552, 555, 563, 56'! Fregoso, R. L., 480, 483 Gonzales, 1\;1. c., 210 Harvey. L, 66
Fackler, r. Ivl., 4(~5, 466, 470 French, D., 211 Gonzalez, F. E., 459, 479 I-latch, E., 169
Fan, D. P., 2115 Frey, J. H., 69, 71, 78, 364, 366, 367, Goodenough, WE., 263 H~l\vkins, A. H., 210
Farringdon,.J. M., 269 368,373 Goodman, L, 123 Hay:lno, D. 11,'1., 209, 2 J3
F;lrringdon, j\1. G., 269 Friedman, N., 210 Goodm<ln, N., 431 Hazelrigg, L, 427, 437, 440, 441, 444,
Fawcett, s. n., 408 Friedrich, P.., 562, 564 Goodye;lr, I... K., 612 "150
Feld, S., 548 Frohock, F., 510 Gooskens,1.,195 Healey, D.,.136
Feldm<ln, H., 128 Furbee, L., 185 Gorden, R. L, 70, 85, 87 Heath, C. c., 355
Fenste.rmaker, S., 215, 510 Gordon, D. A., 2, 212, 520 Heath, S. n., 40J.
Ferguson, F., 75 GotrScl13lk, L. A., 285 Heidegger, JV1., 222
Ferguson,,)., 1, 19,28, :lO8, 120, 111, Gadamer, H. ~G., 222, 431, 439.447 Gould, S. 1-. 145 Henderson, B., 470
126 Gallagher, D., 418, 446 Gravel, A., 211 Henley, N. IVI., 262
Ferguson, L., 159 Gammel, L., 511 Gr3vcs, R., 554 Herirage, J. c., 340, 341, 356, 357
Ferre, J. I~, 465, 466, 470 Garcia, A. 1"1., 369 Green, J. \'{(, 404 Herndon, c., 211
Fetterman, D. M., 409, 596, 610 G;lrfinkel, H., 94, 342, 350, 352 Greene,]. c., 409, 592, 596, 604, 606, HcrtZ, R., 75, 113, 97, 218
Fiedler, L., 477 Garro, L, 289 607,610,611 Hesse, Iv!.. 427, oi31
Fielding,]. L., 284 Gates, H. L., Jr., "ISO Greenwood, D. ]., 404, 606 Heyde, D. L D., 375, 378

654 655
 ~_~~~~''''~' H'~ .'" ~", '''... '''''-' '--0:"""''''' 'r", V ... j'""-,, .... '_.h'-.J f.\urnor moe.\

Higginborham, L, 377 lralic, G., 269 [(ennedy, E. L., 123 Ltc, R. i'd., 5.3, 54, 55, 56, 3 U, 318, 334,
Higgins, P., 210 Kerlinger, E, 429 .3JG
HiJJ, C. E., 287, 289 Kiesinger, c., 2.10 Lee, v., 5" I 1
Hill, P. L, 4S0 Killick, A. P., '121
J"ck~un, 1\" 2111 Lehman, D., 507
Hill. P... T. G., '170. 475 -,.lebun, 13 .. 192 Kink<lid,.J., 213 LeightOn, A. H., 394
Hills, R., 52B Jackson, L., 128 Kirchler, E., 266, 167 Leinhardt, G., 280
Hindess, B., 437 J<lckson, 1\'1., 26, 2X, 210,212,217,224, Kirk,]., 284, 406 Lejeune, !:, 210
Hinsley, c., 129 459,460,469, '170, 473 Kittredge, \V., 473, 478 Lemert, c., 469
Hirschman, E, c., 284 Jacobs, J., 510 Kitzingher,]., 372 Lemonnicr, P., 160, 162
Hirsh, E., 120, 130 .Iago, B. .1-, 511 Klein, D., 382 Leone, M., 161
Hobbs,.J., 275, 281 JJkobson, R., II;}, 55l, 5(iI Klein, R. E., 189 Leroi~Golirhan, A., 162
Hodder, 1.,158,159,166,169,172, Jamc~on, E, 505 Klich, IC, 186 Leung, i\-L L. A., 377
552 J<lllg, H. Y., no Kluckhohn, c., 392, 410 Levin, M., 'IDA, 606
J-[o(sradtcr, D., 440 Jankowi~lk, \1/. R., 11 () Koester, S, K., 128 Levinc, D, N., 503
Hollrrook, B" 313, 332 Janowitz, 1\'1., 622 Kolbe, R. H., 284 Levine, G., 551
Holland, D., Iii I Janssens, L., 141 Kolker, A., 124 Levi-SrrJuss, c., 5, 550, 562
Hullway, w., 210 Jarrett, R. L., 370, 377, JSO Kondo, D., 210, 510 Lewin, E" 123, 212
Holmberg, A, R., 39'1 Jehn, K. A" 2S5, 291,292 Komrba, J. A., 84, '[70 Lewin, K., 403
Holquist, lv1., 544 Jenkins, .I. E., ,170 Kr<lClUCr, 5., 284 Lieblich, A., 56
Holsrein, J. A., 62, 63, 64, 91, n, 94, Jerome, N. \xZ, 'lOS Kr:llner, M., 459, 461, 462 Lieho\\', E., 121, 507
210,342,343,346,349,371,373 Jipson,].,511 Krieger, S., 81, 209, 212, 213, '171, 510, Liefting, lvI., 195
Holsti, O. R" 277, 28'1, 185
Honan, E., 516, 517, 522, 532
Johannsen, ./. I'd., 11
Johnson, 1\., 284
519
Krinitzky, N., 379
Light, R. .J., 283
Limerick, I~ N., 474
hooks, b., 2, 5, 369, 459, 460, 479, 480, Johnson, J., 210,282. Krippendorf, K., 275, 277, 283, 284 Lincoln, Y. S., 9, 29, J3, 96,156, 158,
484,486,511 Johnson,]' Iv!., 77, SA, 117, 108 Krizek, R. L., 21 () 170,218,278,279,28'1,406,409,
Hoose, B" 139 Johnson, Nt, 505, 528 Krueger, R. A., 365, 367, 384 451, '159, '170, 471, 521, 595, 596,
Horney, Iv1. A" 336 Johnson, S. c., 264 Krup<lt, A., 543 598,606,607,610,611
House, E, R., 432, 596, 608, 610, 61 I, Johnston, I'v1., 516, 517 Kruskal, J. B., 266 Lindblom, C. E., 625
613 Joncs, L., 479 Kuhlmann, A., 111 Linden, R. R., 212, 511, 519
Howe, K., 611, 613 Jones, S. I-I., 7, 212, 232.,459,473, Kuhn, T. S., 202, 418, 430, 436 Linder, S. H" 6.36
Hsiung, P. -c., 120 511 Kuklick, H., 109 Lindesmirh, A. R., 278, 286
Hubbard, A. S., 119 Jordan,]., 459, 479, 4l)(i Kulick, D., 212 Lindsay, E. A" 275
Huber,]., 12, 15 Joyce,./. A., 460, 479 Kur<lsaki, K. S., 276 Linz, D., 2H4
Huberman, A. lVI., 108,275,276,177, Jules-Rosene, B., 213 Kvale, 5" 62 Lionnet, E, 211, 214, 215
178,280,182,311,314 Loasos<l, D. L, 169
Hughes, E, c., 14,23,278 Lockridge, E., 5 10, 51 I, 515
Hughes, L., 458, 475 Kahn, R., 70, 85 Ladson-Billings, G., 459, 479, 596 Lofland,]., 22, (;7, 7'1, 76, 79, 86, 279,
Hughes, I~, 128 Kaid, L. L, 284 Laff<lI,].,285 311
Humphreys, L, 89 Kaplan, C. D., 128 Lakoff, G., 505, 528 Lofl-and, L H" 22, 24, 86, 279, 3 J 1
Hum,]. G., 279 [Cup, D., 510 Lang, S., 123, 117 Lonkila, M., 279, 313, 332
Hurston, Z. N., 114, 504 Kassis, H., 269 Larcom,J., 110 Lop"lta, H. Z., H7
Hutcheon, L., 507 Kaufer, D. 5., 272 Lather, P., 7,131,219,447,448,449, Lopate, P., 226
Hutchinson, S. A., 279 Kaufman, S" 507 450,470,511,519,530,598 LopnGarza, IvL, 128
Hyman, l-I. I-I., 70 Kay, K., 276 Latour, B., 163 Lurde, A., 225, A80, 481
Hymes, D., 562 Kearney, M. H., 278, 279, 280 Lawtence-Lightfoot, S., 510, 520 Loughlin,.J., 345
Kelle, V., 277, 294, 313, 332 Lawton, ]. E., 511, 514 Lubiano, \\'Z, '158
Kclly, E. E, 285 Leap, W. L., 123, 212 Luff, I-~, 355
lmrich, D.]., 184 Kemmis, 5" 596 Leary, D., 432, '134 Lynd, H. M., 65
lrurita, V. r., 279 Kempton, \'Z, 281, 402 LeCompte, Iv1. D., 36 Lynd, R. S., 65
Irwin, K., 279, 280 Kcndall, P. L, 72 Lee, G., 473 Lyot<lrd, J, -F., 202, 507

656 657
 "~'" ., ... , "''-' ,,:,v"''''' '''','VI.. "'M'L,,'MI....J

J\bbry, L., 600 l\'lcKinnon, A., 269 lvlukaia, T., 218 Pandolfo, S., 510
l\'lacCannell, D., 54(; McKinnon, L l\L, 284 Mullin, c., 284 Parini,]., 218
Iv1Jccoby, E, E., 65 IV1cLcllan, E.. 276 JVluruock, G. P., 275 Parker, L, 459, 460, 179
IVlaccoby, N., 65 lvlur p by. C., 1S8 Park Fuller) L T 2.11
McMahon, M., 510
t{~"D~'n"l,J, fI" 5')6 Ivlurphy, S., 2/S, 279, 280 r"r~on5, 'c, 352
McTavish, D. G" 285
lvbcIntyn::, A, 220 !\Itild, G. H., 352 Murray, S. P., 131 Pawi, D., fl8, 79, 510
}AacKenzie, D., 431 Medu, Ivl.,194 Patterson, B. R., 275
i'vIadurc, M., 443 Megill, A., 504, 555 Parwn, M. Q., 274, 592, 595, 598, 605,
IVbcQuecn, K. 1\-1., 276 l\1ehan, H, 213 Nagel, T., 430 607,610,622
lvladdox, R., 118 IVleloy,]. lvI., 514 Nakamura, R_. T., flI'-I Payne, D., 210
lvladriz, E., 369, 370, 371, 377, 380 lVlerriman, N., 169 Narapn, lZ., 110 Payne, S. L., S I
lvlahmood, C. K., J28 Merrens, D., -.j09, 596, (j I 0, 61 1 Neal, L, 479 Pelegrin, J., 165
IVlairs, N., 225 !vlenon, R., 210,366 Nebeker, K. c., 460,479 Pelto, G. I-L, 108, 109, Ill, 101
Malinowski, D., 10, 74, 7R, 88, 212., :192, Nelson, c., 5, 6, la, 11 Pelto,I~1-, 108,109,112,401,402
i'vlenon, R. K., 72
411 rvIeschollnie, H., 563 Nelson, J. 5., 504 Perakylii, A., 359
i\lalkki, L H., 121, 130 /'vlctzger, D., 263 Nerlove, S., 263 Perchonock, N., 264
J\bmber, S.,I 79 Nero, C. L, 479 Perez-Stable, E. J., 378
Ivlieczkowski, T., 128
Mangabcira, W, 313 Neumann, M., 209, 213, 21B Pesquera, B, jvL, 379
Mienczakowski, J., 124, '.J71, 5JO
Ivlanganaro, IV1., 548, 550, 551 lvIi!es,1'1. B., 108,275,276,277,278, Newron, .,121 Peters, B. G" 636
IVlanicas, 1:,431,431,436 2BO,282,287,294,311,313,314, Ngeva, J., 119 Petcrs,1- D., 121
iVlanning, P. 1(.,286,398 315,319,322,323,330,335 Nicholson, L L 507 Pfohl, S. J., 511, 519
l\L>.rcheni, C., 284 I'dil!cr, D., 160,161,162 Noddings, N., 214 Pirro, E, E., 285
lvIarcus, G. E., 2, 15, 16, (is, 80, 84,110, Miller, j., 343, 345, 346 Nolan, j. lvI., 262, 270 Pizarro, lVI., 459,480
110,210,507,543,548,5'-19 1\1111cr, L, 21 J Noriega, C. A., 479, 480, 4S3 Plath, D, \',37
Margolis, E., 194,513 ivlilJer, M., 282 Narum, K. E., 510 Poggie, J. j" Jr., 396, 402, 410
l'darin, G" 375, 378, 379 Miller, M. L, 284, 406 Nussbaum,]' F., 275 Polan}'i, M., ,146
1vlarkhal1l, A.IVI., 97, 232 Miller, \II. L, 276 Polkinghorne, D. E., 131, 132
Marrindale, c., 269 Pool, L dc S., 91, 259, 269, 284
Mills, C. \'7., '164, 465, 170
ivlanfnez, H., 287, 288, 289 IV1ilstein, B., 276 Oakley, A., 48, 81, 82, B3, 90, 99, 221, Popper, Ie., '-128, '-136
[vlarx, Ie, 459 l\'liner, E., 548 365,368,375 Postman, N., 145
Mason, J., 357 Mingers, J., 289 Obba, c., 84 Porter,]., 349, 353
1Vlarhews, H. E, 287, 289 Mink, L, 219 O'Brien, JV1., 283 Pam, R., IH4
Marhison, S., "109, 596, 610, 611 Mishler, E. G., 64, 504, 507 Ogilvie, D. lVI., 277, 285 Powell, L C, 475
Iv!atsumOlo, V" 111 IVlircheJJ, S. K., 28"1 Ohnuki-Tierney, E., 210 Pratt, M. L, 211, 214
IVlaxwclJ, J., 432, '135, 437, 454 Okcly,j., 212, 218 Pt;1nis, I., 5-10
Mohler, I~, 285
May, P.J., 636 Monaco, ]., 5, (j Olesen, \Z, 369 Prattis, J. L, 545, 553, 562
l\1;Jynard, M., 367 !Vlomgomery, E., 394 Olkes, c., 26 Preiss,). J., 76
Mbebc, lvI., 119 1v100rc, C. c., 284 Olson, G. A., 120,121,129,130 Preissle, J., 36
McAllister, J. W, 561 O'Reilly. R., 516, 517, 522, 532 Prenderg;lSt, G. L, 2{i9
./Idoore, H., 158, J 69
1vIcAlJister, N., 516, 517, 522, 532 Moore,]. \'>7., 375 Orme, n., 169 Preskil!, H. 5" 610
IvIcCal!. G. J., 119 I'I/loreno, E., 117 Ortner, S. B., 28 Presser,S., 70
Jv!cCaJ1, M. M., 510 Osgood, c., 270 Pressman, J. L, 629, 630
I'I/!organ, 0, L, 71, 86, 99, 363, 364, 365,
IvlcCleary, R., 286 366,367,371,3S4 ' Oxman, T. E., 285 Price, L, 281
McClintock, c., 607 Morgan, 1\'1., 277 Oxtoby, i'vl. J., 277 Pticc-\Vil!iams, D. R., 2,\;4
McCloskey, D. N., 504 Morrill, c., 17 Priot, L, 349
IVlcElroy, A., 409 Propp, V. L, 348
lvlorse,.f. lvI., 274, 276
IvIcGhee, R., 161 Paget, iVL A., 471, 5J6 Punch, 1'\'1., 90, 428
.!v1onan, I-I" 13'-1
lv1cCormick, R. A., 139, 1'10 Paisley, \'7. 1-, 28'-1 Purdy,]., 22
rVlosteller, F., 269
McKay, N. Y., 480 Ivlotzafi-Haller, E, 113 Paley,N.,511 Purvis, J., 367
Palmquist, P., 271, 272 Putnam, H., 430, 43J, 439
McKemie, D., 269 l'vlowen, J. c., 2i:l4

658 659
Quinn, N., 263, 273, 280, 2\37 Rose, D., :U4, 511, 556, 557, 560 Schwandt, T. A., 12,53,222,438,445, Smirh,1v1. S., 277, 185
Quinney, R., 129, 191 Rose, E., 500, 510 471,593,596,59 7 ,591\,599,600 Smith, R., 75
Rosen, J., 465 Schwanz, g., 80 Smith, WE., 186
Rosenau, l~ lvl.) 4+1 Schw:nrl, R. c., J 57 Smirhies, c., 7, 447, 449, A70, 51 J, 519,
P,D.b;mnv, P., 77, 2.12, 552, 556 Rosenbaum, [VI., 278, 279, 280 Schweizer, T., 286 530
Ragin, C. c., 286, 321 Rusenberg, S. D., 285 Scrimshaw, S. C. 1'vl., 402 Snow, D., 17
Rallis, S. E, IDS, 607 Rosenblum, E., 510 Scriven, 1v!., 593 Snyder, R. L., 178
Ramsey, P., ]]9 Rossi, P. H., 609 Se;:Jrle, J. R., 354 Sohier, R., 279
Randall, tvL, J7S Rossman, G. B., 108 Sechrest, L., 157 Sorensen, i'vl. ~L., 168
Rapport, N., 549 Rorer, D., 284 Secord, P., 434, 436 Sp~lflish, Ivl. T., 71
Rasmussen, P., 76, 78, 36 Rothenberg, D., 546 Seidel,j. V., 55, 277, 311 Sparkes, A., 428, 446, 4"!7, 510
fLahje, \\~, lSI) Rothenberg, j., 546, 557 Seidman, L E., 62, 85, 92 Spears, N. E., 284
Reason, 1'.,471 Rowlilnds, M., 161 Sev<lreid. fl.. 405 Specror, j., 172
Redpath, H., 269 Royce, J. R., 555 Sewell, \\~ f-1., ] 09 Spence,]., 191
H.eed Danahay, D., 209, 210, 2 J 1, 213 Rubin, L. B., 507 Shaffer, C. L, 285 Spicer, E. H., 394, 411
Reichardt, C. S., 607 Ruebush, T. K." II, 2[l9 Shamdasalli, P. N., 8(';, 366, 367 Spiggle, 5., 284
Reinharz, 5., 83, 210, 212, 374 Runyan, D., lS4 Shange, N., 479, 4\W, 4\32 Spindler, G., 15
Rhett, K., 226 Rushforrh, S., 263 Shanks, 1\'1.,161,166 Spindler, L., 15
Rich, A., 22 1 Ryan, G. W, 263, 2il6, 269, 270, 275, Shapiro, G., 235 Spradley,]. P., 67, 75, 76, 77, 86, 263.
Richards, I. A., 557 282,236,287,288,289,293 Shapiro. [v1., 505 275, 279, 398, 400
Richards, L, 176 Ryan. K. E., 409, 5%,610,611 Shelley, G. A., 275 Stacey, j., 368
Richards, T. j., 276 Rylko-B<lller, B., 409 Shelly, A., 3 II Stack, C 13.,379,507
Richardson, L., 3, 84, 89, 209, 211, 214, Shelton, A., 214, 510 Stake, R. E., 593, 595, 604, 605, 607
219,345,346,470,472,507,510, Shimkin, D. n., 404 Stanley, J. c., 22, 284
511,513,516,517,518,520,521, Sacks, H., 341, 348, 349, 350, 351, 352, Shostak, A., 210, 214, 510 Swrr, L., 79
523,528,532,556,565 354,355,358 Shoner, j., 219 Starham, A., 517
Richardson, IvL, 459, 561, 561 Salamon, L. !d., 627, 636 Shweder, n.. A. 554 Steedman, K., 510
Ricoeur, 1\ 157 SandeJowski, 1'1., 27'1, 279 Siberr, E., 311 Stcin, H. E, 404
Rieger, J., 191 Sanjek, R., 37, 157 Silvcrlllill1, D., 17, 57, 62, 6J, 76,92,93, Stepick, A., 128
Riemer,]. W, 109, 213 Sarup, 1'1., 62, 92 94,95, lOS, 259, 341, 341, 346, Stcpick, C. D., 128
Rigby, 1\ 405 Salta, L E., 613 347,349,351,357,358 Stern, C 5., 470
Riley, !vI. \\Z, 215 Saussure, F. de, 341 Simons, H. W/., 504 Stewart', D. \\1., 86, 366,367
Rinehart, R., 459, 510 Schaefer, D. R., 97, 98 Simpson, D., 5 (;(, Stcwart, ./., 510
Rist, R. C, 624, 633 Schank, R. c., 280 Sims, N., 459, 461 Stimson, G., 340
Robbins, [vI. c., 261 Schedmcr, R., 546 Singer, E., 70 Stocking, G. \'V.,Jr.,I, 19, 108, 109, 130
Roberts, C. W, 184 Schegloff, E. A., 341, 358 Singer,]., 275 Stoller, P., 26, 511
Robinson, \'\Z 5., 286 Schein, E. H., 405 Sleath, n., 28'1 Stone, I~ j., 277, 284,285
Rocco,S., 516, 5"17, 522, 532 Schensul, J. .f., 58, 403 Slikkerveer, L../., 397 St. Pierre, E. A., 511
Rodriguez, R., 113 Scheurich, J. J., 6],76,448,449,479, Slobin, K., 510 Strilthern, ['vI" 210
Roffman, J~, 22 611 Sluka, J. A., 128 Srrauss, A. L., 14,22,23,67, 194,278,
Romero, 1vL, 523 Schlanger, N., 162 Smallwood, F., 624 279, 280, 32li, 529
Romme, A. G. L., 286 Schnegg, M., 270 Smirh, A., 186 Strauss, C, 281
Romney, A. K, 201, 263, 266 Schneider, J., 510 Smith, A. D., 459, 46B, 469, 479, 483, Stringer, E. T., 470, 471
Ronai, C. R., 7, 12:-1, 218, 232,459,311, Schncidman, E. 5., 285 484 Srronach, I., 443
530 Schnurr, P. P., 285 Smith, D., 202, 112 Sudarlzasa, N., 83
RODS, G., 261 SdlOcpfle, G. 1'1., 107, 112, 114, 282 Smith, D. E., 83, 35 Sudman, S., 69, 85
Ropo, A., 279 Schrag, c., 452, 453 Smith, J., 430, 431, 441 Svarsrad, n., 274
Rony, R., 202, 222, 439, 441, 454, 547, Schuman, H., 69, 70 Smith,]. A., 620
532,355 Schurz, A., 93 Smith, j. .f., 262
Rosaldo, R., 1, 19, 10, 21, 26, 395 Schwalbe, M., 523 Smith, J. K., 598, 599 TJgg, J., 192

660 661
Tan:l!::J, G., 131 vall Dijle, T. A., 382 Wetherell, I'vl., 353 Wolfe, T., 461
Tarn, N., 5'-16, 552, 554, 556, 563, 564, \'an Gelder, r. J., 128 White, D. R., 2S-I Wong, L lVI., 475
565 Van IVb,men, .I., 37, 87, 2 J 11, 212, 28I, White, H., SA? Wood, H., 213
'r;ll15sig, tv1., 555 S07,511,519 \,\/hitlllore, E., 596, 610 Wright, K. D., 279
Ta:>, S., J~)J, 40) van lvlierlo, i'vl., 195 Whyre, \Vi. F., 77, R9, 121) 366, .3514, ,{OJ,
7-:1\;]or, c., '131, 445 VJnO~~ lVbrin, B., 375) 378, J7'J 507,52.-1
TGylor, D. W.) 275 ,..tn \X'iIliF.i31l, .1., -tOIi, 40S' Wilc1avsli:y, 1\., 629, 630 Yatcs, T., 164
Taylor, J., 211 Vaz, K. M., 84 Wilkinson, S., 364, 365, 367, 36,s, 370, Yennie-Donmoyct, J., 471, 510
1;1)do" L. J., 543 Vendler, H., 563 371,372,374,381,474 Yin, R. K., 630
f.1ylor, S., 511 Villenas, S., 460, 479 \Villiarns, B., 549 Yoc1cr, S., 2H2
1,1}'lor, S. J, 22, 279 Viney, L. L., 21:15 \X1iJllams, G., 263
Young,j.,410
Tedesco,}. c., 284 Visweswaran, K., 511 Williams, I~ 1., 510, 51 1
loung,]. c., 263, 287, 189
Tedlock, nO, 210, 212, 213, 516, 545, VOj-'scy, J\L, 343, 3"15 \Vi/liams, \Ii. L., 123
Young, r., 66
5'i6,562 \ViJlsm, D. G., 27.1
Yu, P. -L., 510
ten H,l\'e, P., 353, 3S6 WiJlson, ['vI., 121, 1/2
Yule, G. U., 269
Tesch, R., 259, .313 W;lirzkin, H., 271 Wilson, c., 5] 0
Thom;Js., D. K., lJO \'?<llker, A., 516, 517,522, .)32 \Vilson, D. Ivl. c., 275
Thompson, B., ] 58 \\'alkcr, n. L., 336 \X'jj"on, H. S., 279
Thompson, H., fll, n, R9 Wilver, B. D., JSO Zeller, R. A., 284
Walkerdine, V, 511,519 Zerger, A., 18,j
Thompson, ].,71 Wallace, A. F. c., 263 \\'isb, Iv'l., 26(i
Thorne, n., 524 \\lingensu..:in, L., 222, 34 I, 358 Zilbcr, T., 56
Wallace, D. L., 169
Thr:lshcr, r. lVI., 66 W'obst, lvl., 160 Zinn,IvL S., 378
Walter,]., 139, 142
Tierney, \'(~ G., 13], 218, 470 Wodalr., R., 280 Znaniccki, F., 285
W,llters, D. M., 116
Tiles, IVI., 555 Wolcott, H. E, 5, 22, lOS, 124, 131.390, Zola, E., 503
Warren, C. A. B., 82, 83 ZoJa, r. /(.,2]0, 510
Tilley, c., 157, 158,160, 161, 166,172, 408,410,42 1,428,528,606
Warren, D. M., 397
552 Wolf, D. L. 110,111, Il7, 127, IH Zuckerman, H., 71
Warwick, D. P., 89
Tillman-Healy, L. tvL, 124,210,217 \X'olf, Iv1. A., Ii, lID, Ill, 117, 120, lJO, Zueff, c., 285
\\las hburn, D., 16-1
Tompkins,.J,,221 132,133,510,511,5]9 Zussm;ltl, R., 215
Wasscrbll, R., 96
Torres, R., 610 Warson, R., 349
TOtlJmin, 5., 202 \V,lUgh, J. n., 124
Treichler, P. A., 5, 6, lO, 1 J Wax, 11,'1. L., 135,412
Trend, Ivl. G.. 402 \'l;"lax, R., 76, 77
Trinh, 1: Iv1., 202, 213, 479, 511, .519 \\le aver, A., 336
Trost,). E., 274 Webh, E. J., ]57
Trorter, Ivl., 111 WIeber, 11,'1., 3511
Trotter, R. T., 58 \Veber, R., 284
Tucker, G., 285 \Veber, R. r~, 28S
Turner, V, 26, 212, 469, 470, 474, 507, Weedon, c., 50S
546,563 Weinstein, D., 5, 528
TuvaJ-Ivl<lshiach, R., 56 Weillst'cin, M. A., S
Tyan, K. E., 424- Weis, L, 475, 612
Tyler, S. A., 210, .544 Weisner, T., 269
\Veiss, C. H., 592, 60S, 609,610,621,
622
Ueland, n., 527, 528, 533 WcitZtlliln, E. A., 287, 29-'1, 3] 1, 313, 315,
Ulmer, G., 8], 471,511 319,322,323,330,333,335
\'?elJet, S. c., 261, 266, 189
Werner, 0., 107, ] 12, 114,26'1,282, 2115
ValerY,P.,545 West, c., 23, 373, 468
V;ln Jer Dos, S. E., 195 Weston, K., 85, 124

662 663
 Subject Index
Action rcsc:lrch, 403-404, 606-607
t\meriGln Institute of Public Opinion, 65
Analytic bracketing, 94
Analytic methodology, 259-261, 260
(figure),290-291
analytic induction, Boolean approach,
285-187
case study, intcr/intracultural conflict,
291-292
case study, young adult drug userS,
292-193
classical content analysis, 282-2ll4
codebook developmenr, 276-277
cognitive m<lp an;llysis, 271-172, 273
(figure)
componential analysis, 263-264, 264
(table)
conceptual models, development of,
275-289
content dictionaries, 284-285
cull'ural domain dara, 263-267
crhnogrrrpbic decision models, 287-
289,288 (figure)
grounded theory and, 278-280
hierarchical duster an<llysis, 26'1
key-words-in-cnmcn lisIs, 269, 275
mental maps, 266~267, 267-
268{figures)
multi method analysis, 291-293
negative case analysis, 27S
overlapping c1ustcr analysis, 266
:;ampJing, 274-275
schema all<llysis, 280-281
665
strucmral/senumic network analysis,
270-271,271 (figure)
taxonomies, 264-2J;6, 265 (figure)
text marking proce;s, 277
thematic construct J.l1alysis, 275-276
visual display metlDds, 282
whole~text analpis, 274-289
word allJlysis, 269-273
word coums, 269-1.70
See also Data managememj Text/talk
an<llysis
Amhropological poetics, 423, 542-545
alternative discourse, dc\,c1opmenr ai,
564-565
anrhropological pn~try, 553-565
conventional writir_g and, 561-562
discourse, quality of, 556-559
cthnopoetics, 545-548
literary amhropolo::l"Y, 548-552
meaning, subjectiVE vs. objective
ethnography, ;i53-555, 566-567
pracrice of, 562-564
reprcsenration in, 559~56], 565-566
scientific authority and, 555
Applied ethnography, :i8-59, 389-391
action approach, 403-404
client needs, methodology <lnd, 408-
409
clinical approach, -j 04-405
cognitive approach:s, 398-400
computer-based programs for, 406
criticisms of, 405-407, 413-414
cultural process vs. stasis, 394-398
 ethics/morals issues in, 411-412
future of, 414-415
history of, 391-394
micro/macro analytic appro3ches, 400-
40.1
rese;lfcl1er l)i::lSIn, 4fJ7
rcsc;lrch mcthudologieslt~chniqu~s,
413
science parJdigms in, 41 ()
Strengths of, 407A08, 412-413
utility, criteria of, 409-410
Artifacts. See Marcri:JI culrure
Amhority claim, 27, 2H,1 09, 157
Alltocrhnography, 5 I-52, lOO-lOl, 512-
513
blurred genn;s in, 214-215
cumplete-member researchers, 211,
213
consequences of, 233-235
definition of. 209-210
empiricism vs. interpretive
perspectives, 101-203
ethic;)l issues in, 240-2"12
fiction, ethnography and, 231
future of, 1"13-145
generalizability and, 217, 224,229
interviews, inter::Jetive conversation,
233-239
literary auroetllllographies, 211
memory/emotional recall and, 230-231
methodological strategies for, 210-
211,236-243
native edlllogr;lphy and, 213
persom.! narrative in, 201, 204-206,
211, 213-214,21G-227
practice considerations in, 227-236
tadical empiricism and, 212
refle:.:ive crhnographies, 211-213
reliability issues in, 229
storytelling and, 219, 220-222, 232-
233
textual form, 239-240
Sec also Wtiting inquiry
Blurred genres, 3, 4, 24-25, 214-215
Borderbnd cultur<ll :lre<lS, nO-I21, 202
Bricoleurs, 4, 5
cryswllization process and, 8-9
definition of, 5-6
666
interpretive pr:lctices of, (1
rnoIHage method, 6-7
task characreristics of, 9
Bureau of Applied Social Research, 67
CAQDAS. SCI.' Compmcr-as:;isted :malysis;
Software tools
Call sal narrativc, 23
Chicago school, J, 21-12, 65-66
Cinematic practices, 483-486
Civic journalism, xi, 461
civic transformations and, 4/;-l-465
intimate journalistic practice, 462-463
undcrstandings in, 461
writing norms, cornmunir;]J"ian ethics
and, 465-466
See also Political interpretive pr;lCt'icc~
Coding. Sec Software tools; Whole-lext
analysis
Cognitive Illap an:llysis, 271-271, 273
(figute)
Collaborative inqlliry, 49,1] 1
Colonial ethnography, 18- r 9
Complcte-member researchers, 211, 213
Componemial analysis, 26.3264, 264
(t:lble)
Computer-assisted analysis, 14, 24, 53-56
applied ethnography and, 406
content dictionaries, 284-285
electronic interviewing, 96-98
historyof,311-313
Internet and, 55-56
interpretive process and, 54-55
software familics in, 54, 319-322
See also Software tools
Conceptual model-building, 277-278
analytic induction, Boole311 approach,
285-287
classical content analysis, 282-284
content dictionaties and, 284-285
ethnographic decision models, 287-
289,288 (figure)
grounded theory and, 278-280
schema analysis, 280-281
vistlal display methods, 282
Confessional tales, 212, 226
Constructivism, .34, 35, 37
material culture and, 50
meaning, COntexrllalizarion of, 133-
135
SCI.' also Rd:uivism
Cuntextual issues. Sec Inreractional
COllver~<ltil'll analysj~ (CA), 57, .353, 357-
358,357-358 (tables)
Cre:ltive analytic practice (CAP)
ethnography, 509-51 I
aurocthnography, 512-513
ethnographic drama, 516-517
ethnographic fiction, 515
evocative form~ in, 511-512
institurional acceptance of, 524-525
micrnprocess writing-srories, 514-5 15
miwd gentes, 517-520
poetic representation, 515-516
process/product relationship, 511
process suggestions, 528-533
quality work, critetia for, 520-524
triangul3rion vs. crystallization
process, 517-519
writing-stoties, 513-5 J 4
Creative interviews, 80, 91
Crisis of tepresematinn, 3, 4, 25-27, 28,
212,449
Criteria. See Relativism
Critical r<lce theory, 58, 478-479
cinematic practices, aesthetics and,
423-486
standpoint epistemology, en:Jctmenr of,
479-483
Crystallization process, H-9, 517-519
Cultural studies,ll, 27, 34,33<36
bordetland areas, 120-12"1
communities of memory, hybrid
cultures, 111
culrure, definition of, 390-391, 399
Sec ,-zlso !V!ateri,J! culture
Data management, 37, 52-53
frame substitution, 263
free lists, 262
paired comparison test, 262
pile sort, 262
srstematic elicitation techniques, 261-
263
triad test, 262
llnslTucwrecl interviews and, 79
667
Sec also Analytic methodology; Text/
talk analysis
Democratic ptactice, ;;i, 419-420, 422,
459
Discuurse analysis (011), 57, :H8-34~), 353
Documents. Sce !\/lare~ial culture
Dramarurgic;ll culture, 467-468
E.lectronic interviewing, 96-98
Emotionality:
interview methods, 70
observational practice and, 117-118
Ethics, 31
applied etbnograp1.Y and, 41 1-412
going native, 117
harm concept, intrusive research and,
137-138
institutional revie\v boards and, 135-
136
interviewing and, 38-90
privacy protection:;, fictionalized
accounts aillt 138-139
proportionare reason, 49,139-143
social progrJ.Jn evdllarion, political
ethics and, 5n, 612-613
Ethnic studies, 18-19, 34, 35
Ethnographic decision models (EDIvls),
287-289, 2S8 (figure)
Ethnography. Sec ApJ=lied ethnography;
AmoethnogrJ.plw; Observation
Evaluation, 3.1, 37-38
criteria for, 420-4:22
Sce also Social ptcgram evaluation
Evocative narratives, 217-219
Experimental inquiry. Sec Anthropological
poetics; Postlnodernism; \X/riting
inquiry
Feminist communitaf'anism, 465-466
Feminist research, 34. 35
amoethnographic form and, 211
experimental text~; and, 84-85
interviewing, feminist-based practice,
48,96
observational reseJrch and, 49
oral histories, 79
power, perceptions of, 120
 '-.-'-''--......'-.-",,''-' ,-..", . . . ,', , ... ", ", ... ",,>.;1 """UMI..,,M"VI.. "ilM' '--I'.'M'---'

types of, 369 See also Focus groups; Imerviews electronic interviewing, 96-98 Literary forms, xi, 3, 12
Sec also Focus groups; Gender; emotional dimension and, 70 aut'Oetllllographies, 211, 214, 231
Writing inquiry ethical considerations, 88-90 civic journalism, 4(;5-466
Ficrion. See Literary forms; Metaphorical Harm wncept l 137, 139- HO cthnomcthodologically informed experimental wrlung, 18-2~, 4~
prinKy
fjl,;tiUllulit;cu ilccoum:;,
Focus groups, 57-58, 71, 363-364 blurred genres, 1'125 feminist Crllic in, 48 protection and, 138-139
collective testimony, source of, 374- crisis of rcpresentation, 25~27, 28 framing interviews, 85-87 naturalism in, 22, ::8, 35
375 legitimation crisis, 28 fUlure of, 90-91, 94-96 See also Amhropobgical poetics;
collectivistic vs. individualistic research modernist phase, 22-24 gendered intcrviews, 81-85 Political interpretive practices;
3.nd, 364-365, 368, 352-383 praxis crisis, 28-29 group imetviews, 70-74, 73 (tahle) Writing inquil y
culturallsocial constraints and, 381 tradirional period, 19-21 hiswrical development of, 64-68 Lived experience,l2
empowerment, ethics of, 375, 382-383 informant/guide in, 77-78 constraints of, 16
feminist postmodernism and, 368-370 imer;tcrivc conversation, 138-239 intimate jOlirnalistT and, 462-463
future of, 384 Idemity roles. See Interprcrive practices; interpretation of, 87~88 matetialist-realist ontology, 35
groups, composition of, 379-380 Observation interv"lewer effects, 69-70, 91, 367, performance texts ;lIld, 17
history of, 366-367 Individual viewpoint, 16 380 reflexive/multi vocal rexts, 35
inrerviewer influence and, 367, 380 Institutional review boards (lRBs), 49, interviewer-respondent relationship, ~ocial context and, 31
interview guide in, 380 135-136 75-76,82,92-93 writing process ane', 27
market research and, 71, 365 Interactional contexr,1 14-115 imen'iewcr self-presentation in, 77 Scc also Amoethnography; lvIarerial
minority women and, 364, 367, 369- ethnographer, individual experience interview society, 63-64 culture
370,374,376-382 of,131-133 language in, 86 Lone Ethnographer conccpt, 20-21
recruitment straregies, 377-379 meaning, comextuali:r.ation of, 1.33- negotiated accomplishment of, 91 93
self-other distancing and, 371-372, 135 non-verbal responscs in, 87, 97
374,376-377 power plays, perception of, 118.-11.'1 opinion polls, 65 I'darket tesearch, 71,365
settings for, 373-374 roles/idcntities, negotiation process oral histories_ 79-80 Marxislll, 34, 35
social interactions and, 372-373 and. 124-128 postttlodcrn interviewing, 80-81 l"vIateri;ll culture, 50,172-173
rransform3.tjve process of, 375 situarional identity, adoption of, 115- rapport/understanding in, 78 artifact analysis, interpretive process
voice issues in, 368-369, 373, 374-375 118 respondent language/culture and, 76- and, 158~159
validation criteria and, 12\:l~133 77 coherent interpretation of, 170-171
See also Observation social psychology and, 65-66 communicativc/representational
Gang studies, 66 Internet, 55-56 struCt'llred intcrviewing, 68-70 objects, 160
Garage sale metaphor, 11 electronic imervicwing, %-98 survey research, 65, 66-67, 91 confirmation, meaning hypotheses
Gendcr: visual sociology ~lIld, 178, 180 trusr issues in, 78 and, 170-172
domestic decorative arts and, 159 Sce also Computer-assisted analysis unstructurcd interviewing, 74-79 correspondence, theory/dar,l and, 171
ethnographic practice and, 121-123 Imwpretive practices: See also Focus gronps; Text/talk grammar/dictionary of, 161-161
gendered interviews, 81-85 ani fact analysis and, 158 159 ;malysis interpretive methodology and, 166-
identity negotiation and, 126-127 art/politics and, 37-38 169
material culture and, 50 interviews, 87~S8, 91 meaning, reading/writing acts and,
photOgraphy, social construction of, paradigms in, 33-36, 34 (table) Judgment process. Sct! Relativism 156-157
191-193 personal identities and, 123-124 memory traces and. 162
General Inquirer, 285 postmodern interviewing and, 80-81 practical/cultural knowledge and, 162-
Generalization, 217, 124, 229 qualitntive rcsearch and, 9-11, 421 Knowledge: 164
Golden age. See Modernist/golden age See also Bricolellrs; Dar,l management; production of, 17 reimerpretations and,157-158
Grounded theory, 67-68, 278~280 IVlaterial culture; Political See illso Relativism; \X;'riting inquiry symbols, idcologic81 components/social
Group interviews, 70-71 interpn:rive practices K\\7IC (key-words-in-comen) analysis, convention and, 161-161
advantages/disadvantages of, 73-74 Interviews, 47-48, 61-61, 98-99 269,175 temporal dimension. meaning flux
groupthink and, 73 access issues, 76 and, 164-166
skills for, 71-73 confessional slyle, H7-88 rheory of imerpret;\tion for, 159-164
typcs/dimensions of, 71, 73 (table) creative interviewing, 80, 91 Legitimation crisis, 18 written documents/records, 156-158
mility of, 71-71 data collection, 79 Life history approach, 21 Materia1ist~realist ontelogy, 35

668 669
I\leaning construction, 133-135
bnguage usc and, 3'11
text readings, 156
Sec a/so Interprerive prilctices;
ilA"N!,-i"d ~l1ltl... e
i\lelting pot hypothesis, 19
l'vlembcrship categoriwtion analysis
(iV1CA), 57, 95, 349-352
lVlemoirs, 26, 212, 214-215
Ivlentalmapping, 266-267, 2<17-268
(figures)
l\'1elaphoricaJ texts, 7-9, 505-506
iVlinorities. Scc Critical race theory; Focus
groups; Political interpretive
pr;lCtlCeS
iVlodernist/golden age, 3, -I, 22~2-1
lv'lomems. Scc Historic moments
?vlomage, 6~7
;\'!tne evidence. Sec Ivlmerial culture
Narrative ethnographY', 2J2
National Opinion Research Center, 67
Native ethnography, 213
N~1turalism, 21, 2B, 35
Neorealism. See Relativism
Objeerivity, 31, 4B, 91,109
Observational research, 48-49, ]07-108
bias in, 111
classic tradition of, 11] -I J 4
collaborative methods and, 110-111
ethical dimension of, 135-143
ficrionalized accounts, privacy
protecrions, 138-139
fmure prospects for, ]43-145
lwrm concept, intrusive reseilrch and,
137-138
institlltional review bOilrds ilnd, 135-
136
interactive context of, 1 IA-D5
meaning, contex[tJillization of, 133-
135
naturalistic observation, lOB
objectivity and, 109
power plays, perception of, J 1 S~124
proportionate reilson and, 139-143
role typology, 113-J 14
sirtl~m:~dness and, ]09-110
670
situatiunal idcntity, aduprion of, 115-
liB
situational identiry, inn.:rilcrivc
dynamics of, 124-12S
",,]icl,,';nr, ",,,I, 1 OR_lOa, -I )R_l1;
Sec also Applied ethnogr3 p hy ;
Participant observation
Oral histories, 79-80, 111
Participant obsetvation, 13, 67, 74
conrradiction in, lUi
observarion of participant and, 26-27,
211-213
role typology, 1 13-11 '1
Performance te_xts, 7-8, 467
co-participation, actionlref1ection and,
'170-473
critiCill rilce consciousness and. 478-
-IB6
cultural meaning, daily enactment of,
4(i9-470
dramaturgical culmre and, 467--168
forms of, 469
moral dialogical texts, '169
systems of discoursc in, '16S-']{;9
Sec also Political inrerprerin, practices
Personal narrative:
autoerl1l10graphy and, 101, 204-206,
211,213-214
confessional tales, 211, 226
consequences of, 121-122
empirical science and, 11 ()
evocative narratives, 217-2J9
goals of, 224-226
postrnodernist narrative inquiry ;md,
217,211-224
storyrelling vs. story anillysis, 2J9,
220-112
rruth and, 219~220
Poetics. Sec Anthropological poetics
Policy. Sec Public policy
Political interpretive practices, 458-461
civic journalism, 461-467
civic rransformations and, 46+465
critical race consciousness,
performance-based aesthetic and,
4n-486
future of, 486-489
performilnce tex[!; and, 467-471
Red Lodge, MOlHana, performance/
expetience of, '173-478
Politics:
historical research prilClicc and, I S-19
program evaluation, 38
qualitative research, resistance to, 12-
13
See also Public policy
Positivism, ll-13, lA-IS, 3<1
Postmodern interviews, 80-81, Btl
Postmodernism, 3-4, R, 15-16,19,2829
Posrpositivism, 4,12,14-15,22,34
Power clements, 118-124,443-444
Praxis crisis, 19-19
I'rogt;m1 evaluation. Sec Soc!ill progr;lm
cvaluation
Proportionate reilson, 49, 139
critcriil of, 139-1'10
knowing/judging abollt, 140-[41
social reseatch and, 142-1'13
Public policy, 38, 424, 61')621
accountability/outcomes assessment,
631636
conceptual complexity of, 625-626
decision-milking, lliltllte of, 621-613
formulation of, 624-619
implementation of, 629-632
initiative-effect evallwrions, 626
organizationJ.lIinstitlltional respOnse,
631-632,639-640
policy cycle, qualitative research and,
623~624
policy tool selection, 636-641
political risk and, 640-641
ptior initiatives and, (;16
qualitative research, role of, 616-619,
641-642
research influence on, 622-623
resource intensivenessh-elarive costS
and,637-63g
targeting and, 630, 638-639
tracking process, 630-631
Qualitative research, 1~1
data collection/analysis, 37
definirional issues in, 3-5
evaluation, 37-38
cver>,ela>, social world and, J 6
671
genlTarive natllre of, Xl
historic moments in, 3 4. 19 29
history of, 17-19
individual viewpoint in, 16
34 (table), 37..38, '17
multimethod/ll1ultig'~1l1"(,
tocus of, 8,
11
other, research subject, 30-3 I
positivism vs. posrpflSltivism, 14-15
postmodcrn sensibilities and, 15 16
process of, 19-38, 3~~ (table)
quanritative rescarcll and. 13-17
researcher in, 3 I
resistances to. 1213
straregies of inquiry, ]f; 37
tensions within, 17
thick desctiptions in,16-17
triangulation \'s. cry;tallization
process, H-9
Sec 'lfso Bricoleurs
QU3ntitative research, 13-17
Quasi-fonndationalism, '12S, 431'132
descriptive v~llidit'y ;:nd, 435-436
interpretive v~llidit)' and, 436
knowledge, social/historical product',
434-435
neorealism, COl1sttlll:rivislll Jlld, 432
observation, theory te~ting and, 436-
437
5cienrific discourse, norms of, 4.14
truth-as-col"[espond~nct', -132-'134
Queer theory, 3,1, 35-3,;
Quilt maker. See Bricol:urs
Rilce. Sec Crit'iclJ r;1Ce 1heory
Reality, 11, 33, 35
Red Lodge, MO!Hilna, .:.73-478
Reflexivity, 4, 26, 51,125-126
amoerhnography, 211-2I3
tadical empiticism ~ndt 212
l{elativism, 33, 35, 420, 427-429
constructionism, moral responsibility
and, 441-442, 453
historical developm::m of, 42.9-431
judgment process ami, 440-'1-11, 443-
451
method, neutrality of, 419-430, 431
 ~~~~~~ ,~.".~ "., ~'"
nonfollndationa!ist perspective, 438,
'139
pluralism/mulripliciry and, 452-453
qUJsi-[oL!mbtion:disll1 and, 428, 43]-
4JH
representation, problem of, 449-,150
self-refuting nature of, 438-440
standpoim episremologies, 446, 447,
451-453
rheof}'-free knowledge, 430-'1.,1
[{e"e<Hcher as subject. See
Auroethnography
Schema analysis, 280-281
Self-reflection, 27
Siruatlofl~ll identity. Sec ObservJtional
research
Social policy. Sec Public policy
SociJI program ev;l!uarion, 424, 590-591
~lctivisr ideology ~llld, 596, 608
advocJcy sunce ill, 606607
case example ot', 60]-603
challenges to, 611-613
constructivist inquiry, 597-599
contexts of, 591-593
educHive/generative potenti'll of, 608
heuristic procedurcs Ill, 598-599
hisLOtical development of, 607-611
meollling/representations in, 611-611
methods/philosophies in, 593-596, 594
(table)
multi-interpretable texts, 600
narrative storytdrlllg as, 603-607
organiz.ationJI learning and, 610-611
policy-makcrs/fundcrs Jnd, 593, 595
political ethics of, 591, 612-613
pr;lctical problem~solving, 595, 599-
600
qualitative evaluarion, justificmion for,
597-600
Cjualit;ltive practice of, 600-607, 611
responsive conlextllalized
understanding in, 595-596
stakeholder engagement in, 6] 0-61]
theory-driven evaluation, 609~61 0
trustworthiness constructs and, 606
vJlue dimensions oi, 592-593, 598,
605-607
Social psychology, 65-67
~ ,.~ '"<:'-'" ... , 'n" .... ,.,,~, ... " .. ~, ....... .JuoJecr maex
Sofrware 10015, 294~295, 310-3]] discourse analysis, interprered reality gender ;lnd, 193
an:llysis, style/:lpproach of, 327-330 and, 348-349 ill1Jge, social cons-Tllction of, 192~193
Jpplications considerations, 324 empathic understanding, commonsense narrJtive swdies, IS6-1gS
bias widlin 1 313 and, 342 phenomenological perspecrive ~tnd,
choice G"idelines, J22-.331 intcIVje\Vs, J-fJ-J'-f7 191
cJoscncss-to-data issue, 331, 332 bllguage game and, 341-341 photo el"icit<1rion t:.chnique, lRB-l~)O,
code-and~retrieveprograms, 320 membership categorization analysis 194-195
code-based rheory builders, 320-321 and, 349-.152 recorded perception, history of, ] 77-
computer use, history of, 31] -313 narrative perspective and, 343-31.J5 181
conceprual nenvork builders, 32] sociJI comext and, 341 reseatch photogr3iJhy, visual social
consistency issues, 316 rape analysis, 356~35S science and, 181-1S5
consolidation tJsks and, 317 text-bJsed analysis, 347-353 IheoretiGl1 petspective, sociological
dangers of, 317-318 trJnscripts in, 353-358 consciousnes:, and, 194-196
databJse, nature/demands of, 324-327 Textual forms, xi, 3 Voice:
financial constraints and, 330-331 crystallization process and, 8-9 :H1dicnces and, IJI)
hardw;lre considerations, 323-324 performance texts, 7-8 feminist research nethods Jnd, 368-
m,lllu;llmethods debate, 331.J reflexivity in, 4 369
methodology choice and, 332-333 researcher-as-aurhor in, 12 focus groups and, 373,37-1-375
multimediJ clpabilities and, 321 ~322 Scc also Anthropological poetics; \\-Titing process an-=!, 131-132
Autoethnography; Literary forms;
qU;llity-of-research issues and, 335
rVlaterial wlnlte; Political
represcntational aids in, 317
imerpretive prJcticcs
research ,md development on, 335-337 Whole-text anillysis, 174
Thick descriptions, 16-17, 25, 98, 169
software f:Jmilies, 3] 9-322 codebook devclop:l1em, 276-277
TrJditional moment, 3, 4, 19-22
speed of, 316-3] 7 sampling and, 274-275
Triangulation, 8, 99, 5J7-518
textbase managers, 319-320 rext marking process, 277
text retrieval, 319 thematic construct analysis, 275-276
theorj'-building process and, 315 -316 See also Analytic mcthoJology; \X10rJ
Unstructured interviews, 7"~75
utility of, 314-318 analysis
access issues, 76
See also Computer-assisted anJlysis Word analysis, 269
creative interviewing, 80
Speech-iKt theory, 354 cognitive map analysis, 171 271,273
data collection Jnd, 79
Standpoint epistemologies, 446, 447, 45]- (figure)
informant/guide in, 77-78
453 interviewer-respondent telationship in, key-words-in-cont,~xt lists, 2(}9
Storytelling, 3 75-76 structurallsemJntic network analysis,
auroethnography and, 119, 220-222, interviewer self-presentation, 77 270-271,271 (figure)
232-233 oral histories, 79~SO word coulltS, 269-270
social program evaluation -and, 603- postmodern interviewing, 80-81 See also Analytic mcthodology; \XillOle-
607 rapport/understanding in, 78 text analysis
story anJlysis Jnd, 119, 220-222 respondents, language/culture of, 76-77 Writing inquiry, 499-500
See also Interviews trust issues in, 78 creative analytic practicc ethnography,
Survey research, 65, 66-67, 91 509-520,518-533
Survey Research Center, 67 crossover forms, 3 )3-504
Validntion, 108~109 doubt, conrext of, 50S
ethnographer, individual experience formats/genres in, 506-507, 528
Text/talk analysis, 56-57, 340-342, 359 of, J31~133 fmure of, 516
audio recordings, use of, 354-356 imerJctive context and, 119-] 31 historic writing conventions, 502-504
coding schemes and, 348 internal/external criteria in, 128-129 institutionJI acceplallce of, 524-525
content analysis, 348 Value-free science, 9, 11,31 language, sociill constructs and, 508~
conversation analysis, 357~358, 357- Verification issues, 11, 14 509
358 (tables) Visual mcthods,50-51, 176 metaphor" 50S-50Ii, 527-528
672 673
 '- ,<~"~."""~ . ",,~,,,,,~~

methodological expansion and, 525- experimentation in, 28-29

526 interpretive practice and, 37, 422-423
pustmodernist comext and, 507-509 Sec also AuwcthnographYi Political
quality \Vork, crireri'l for, 520-524 interpretive practices;
,,,,,,,.,,,! ~.:t."'!:t).,

\'hiting process, 26-27

.::(){l-.::n1
Wriw:n records. Sec tvlaterial Culture
About the Authors
civic journalism, 465-466

Michael V. Angrosino is Professor of Anthropology at the University of

South Florida, where he specializes in mental health policy analysis, the
influence of organized religion on contemporary social policy, and the
methodology of oral history. He has served as editor of Human Organiza-
tion, the journal of the Society for Applied Anthropology, and is currently
general editor of the Sal/them Anthropological Society Pl'Oceedings Series
for the University of Georgia Press. His most recent book, Opportunity
HOllse: Ethnographic Stories of ldental Retardation (1998), is an experi-
ment in alternative ethnographic writing.

H. Russell Bernard (Ph.D., Illinois, 1968) is Professor of Anthropology at

the University of Florida. He has taught at Washington State University,
\V'est Virginia University, and the University of Florida. He has also taught
or done research at the University of Athens, the University of Cologne,
the National 1vluseum of Ethnology (Osaka), and Scripps InsritlHion of
Oceanography. He works with indigenous people to dC'lelop publishing
outlets for works in previously nonwritren languages. He also does re-
search in social network analysis, particularly on the :Jroblem of esti-
mating the size of uncountable populations. His publications include
Native Etlmogrdphy: An Gtoml Indian Describes His ClIlture (with Jesus
Salinas Pedraza; 1989), Technology and Social Change (edited with Pertti
Pelto; second edition, 1983), and Research Methods ill Anthropology (sec-
ond edition, 1994). I-Ie has served as editor of HllInlJl Organization
(1976-1981), [he American Anthropologist (1981-1989), and en!tnm!

674 675
 ..... '-' ................ ,,, ...... " , ..... "" ... ,,, " .... ,,, " .. ""'-'r~ ...' ,'""", .... n,,",,, ... ',,'"".......,

Anthropology Methods jo/mud (1989-1998), and is currently ediror of into working fuJly within the qualitative paradigm. In her current \\Titing
Field lvIethods. she conrinues to grapple with the meaning of quality whilc striving to uti-
lize qualitative inquiry in \vriting the self, in dcpicring th~ lives of womcn,

Institute for Interpretive Human Studies at the University of South

Florida. He is the coauthor of UnderstalIding ['(1111ily Commllll;cation and Norman K. Denzin is Distinguished Professor of COI11r:1unications, Col-
coeditor of Composing Etlmor;mphy: Alternatiue Porms of Qualitative lege of Communications Scholar, and Research Professor of Communica-
VI/riting (1996) as well as the AltaIvlira Press book series Etlmographic tions, Sociology and Humanities at the University of Illinois, Urbana-
Allernatiues. He has published more than 50 articles and monographs all Champaign. He is the amhor of numerous books, including Il1tel1Jretil'e
close relationships, communication theory, and narrative inquiry. His Etlmogri.1phy: Ethnographic Practices (or the 21st Century, The Cinematic
cu~tent research focuses all geriatric care managers as ethnographers of Societ~y: The Voyeur's Gaze, Images of Postmodem Soci.?ty, The Research
3gmg. Act: A Theol'etio.d Introduction to Sociological Methods, Il1terfJretive
IllteractiollisJ1l, Hollytuood Shot by Shot, TIle Recoverii1g Alcoholic, and
Ivan Brady is Distinguishcd Teaching Professor of Anthropology at the The Alcoholic Self, which won the Charles Cooley Award from the Society
State University of Ne\v York at Oswego and a SUNY Faculty Exchange for the Study of Symbolic Interaction in 1988. In 1997 he was awarded the
Scholar. He is a former President of the Society for Humanistic Anthropol- George Herbert A\vard from the Study of Symbolic Interaction. He is the
ogy and book review editor for the American Anthropologist. His primary editor of the Sociological Quarterly, coeditor of Qualit;Hive Inquiry, and
research has been in the Pacific Islands. His poerry has appeared in various editor of the book series Cultural Studies: A Research Annual and Stlldics
books and journals, including Reflections: The Anthropological Muse in Symbolic Interaction.
(edited by 1. Prattis; 1985), the Neuromlthropology Networl~ Newsletter,
and Anthropology {lnd Humanism (Quarterly). Carolyn Ellis is Professor of Communication ane Sociology and
Codirector of the Institute for Interpretive I-Iuman Smeies at the Univer-
Erve Chambers is Professor of Anthropology at tbe University of l\!lary- sity of South Florida. She is the author of Final Negotiiltiolls: A Story of
land, College Parle. He is founding editor of the publication Practicing Louc, Loss, and Chronic Illness (1995) and Fisher rolh: Tum Comnwnities
Anthropology and a past President of the Society for Applied Anthropol- all Chesapeahc Bay (1986). She is coeditor of Composing Ethnography:
ogy. His current research interests include ethnographic approaches to Altematiue Forms of Qualitatiuc Writing (1996), lllucstigatillg Subjectiu-
decision making in the areas of tourism development, natural resources it)': Research on Lived Experience (1992), Social Perspectiues 011 EmotioJl
management, and urban and regional planning. He is the author of (yolume 3), and the AltaIvlira book series Ethnographic Allernatiucs. Her
Applied Anthropolog)': A Practical Gnide (1985) and Native TOllrs: The current research focuses on illness narratives, autoethncgraphy, and emo-
Anthrotm/ogy o('Hauel and TouristJl (1999). His edited volumes include tional sociology.
HOllsing, Culture, and Design: A Comparative Perspcctiuc (1989) and
Tourism and Culture: All Applied Perspective (1997). Andrea Fontana is Professor of Sociology at the Univers:cry of Nevacla, Las
Vegas. He received his Ph.D. from the University of Calil:ornia, San Diego,
Deborah K. Deemer is Assistant Professor of Education at the University ill 1976. He has published articles on aging, leisure, theory, and post-
of Northern Iowa. Her interest in criteria for qualitative inquiry began modernism. He is the author of The Last Frontier: The Social Meaning of
when she was working in program evaluation as a Research Associate at Getting Old, coauthor of Social Problems and Sociologies of Everyday
Aherno College's Office of Research and Evaluation. For more than a dec- Life, and coeditor of The Existential Self in Society aldl Postmodemis111
ade she has struggled to disentolI1gle her subjectivity from realist preten- and Social hlqlfiry. He is former President of the Society for the Study
sions. Her coauthored contribution to this volume reflects her movement of Symbolic Interaction and a former editor of the journal Symbolic Illter-

676 677

action. His t'Wo most recently published essays are, respectively, a decon-
struction of the work of Hieronymus Bosch and a performance/play about
Ewinelli, the castrato.
James H. Frey is Dean of the College of Liberal Arts, Professor of Sociol-
ogy and founder :md former Director of the Center for Survey Research at
the University of Nevada, Las Vegas. He is author of Survey Research by
Telephone and Gouemlllcnt and SI701't: Public Policy Issues. He has pub-
lished papers on survey research, group interviewing, sporr sociology~
deviance, and \-vork in the leisure industry. He recencly edited an issue of
the AlInals ofthe American Academy 0/ Political and Social Science on the
social and economic impacts of gambling.
Jennifer C. Greene received her Ph.D. in educational psychology from
Stanford University in 1976. Since then, she has been engaged in the field
or sociJI and cd LlcationaJ program evaluation. \'(1orking first at the Univer-
sity of Rhode Island, then at Cornell University, and currently at the Uni-
versity of Illinois, she has concentrated on making her work useful and
socially responsible, both in theory and in practice. Her work has em-
phasized the development and refinement of various approaches to eval-
uation, primarily using qualitative methodologies, participatory ap-
proaches, ~Hlcl "mixed-method" perspectives and value stances. She has
evaluated a wide range of programs, including public policy education,
natural resource leadership training, remedial education, and youth em-
ployment. Her work generally focuses 011 educational programs and pro-
grams for families and children. In her publications she endeavors to share
the lessons she has learned about evaluation practice across multiple
conrexts.
Douglas H;;trper is Professor and Chair of the Department of Sociology
and Codirector of the Graduate Center for Social and Public Policy at
Duquesne University. He has held faculty appointments at the University
of Amsterdam, the University of Bologna, the State University of New
York at Potsdam, and the University of South Florida. He is the founding
editor of Visual Sociolof:,ry, the journal of the International Visual Soci-
ology Association. His books include Good Company (the sociology of
the tramp), \X/orl,,-ing Knowledge: Sl::.i11 and Conmumity in a Small Shop
(the microsociology of the shop), and Changing Yllor/;;.s (on cows and their
keepers). He has coedited books on visual sociology published by aca-
678
MUUUl [II!:' MUUIUI.:>
demic presses in Iraly and Holland. His first book, Good Company, has
been translated into Italian and French; his papers hav.; appeared in trans-
lation in French, German, and Italian. He is also codirector of the film
Ernie's Sawrni{{. l--:lis current sociological interests center on the sociology
of jazz. His 1998 ASA panel on that topic featured the first musical sociol-
ogy in ASA history, as performed by H. S. Becker ane Robert Faulkner.
Ian Hodder obtained his B.A. degree in archaeology from the University
of London in "1971 and his Ph.D. from Cambridge University in 1975.
From 1974 to 1977 he taught in the Department of Archaeology at Leeds
University, and since then he has taught at Cambridge, ending up as Profes-
sor of Archaeology. Since September 1999 he has be~n Professor in the
Cultural and Social Anthropology Department at Swnford University. He
is also a Fellow of the British Academy and has taught as a Visiting Profes-
sor at the University of Amsterdam, Paris 1/Sorbonne, :~he State University
of New York at Binghamton, the University of California at Berkeley, and
the University of Vienna. His books include Spatial Analysis ill Archaeol-
ogy (with Clive Orton, 1976),Reading the Pas! (1986), The Domestication
of Ellrol,e (1990), and The Archaeological Process (1999).
Yvonna S. Lincoln is Professor of Higher EdLlGItion~ Texas A&i'vl Univer-
sity, and coeditor of this volume and the first edition of the Handbook of
Qualitative Research (1994). She is also coeditor of the journal Qualita-
tiue Inquiry, with Norman K. Denzin. She is, with her husband Egon G.
Guba, coauthor of Effectiue Euallta/ioll (1981), Naturalistic Inquiry
(1985)~ and Fourth Generation Evaluation (1989); she is also the editor of
Olgmzizational Theory and Inquiry (1985) and coeditor of RepresentatioN
and tIle Text (1997). She has been the recipient of numerous awards for
research, induding the AERA-Division .J Research Achievement Award,
the AIR Sidney Suslow Award for Research Contributions to Institutional
Research, and the American Evaluation Association's Paul Lazarsfeld
Award for Contributions to Evaluation Theory. Shl: is the author of
numerous journal artides, chapters, and conference presentations on
constructivist and interpretive inguiry, and also on higher education.
Esther Madriz is Associate Professor of Sociology and Associate Director
of the Center for Latino Studies in the Americas at th.= University of San
Francisco. She has a master's degree in criminal justice administration
fro111 California State University in Sacramento and a Ph.D. in sociology
679
from Vanderbilt University. She previously taught at Hunter College in
New York City. She te8ches classes in criminology, violence against
women, and juvenile delinquency. Her field of expertise is fear or crime.
~I~p ;<; t11.C Cl.urJ.,o, ot n~Clny Clrt:cl...,,, on tl"" ,opic OlnJ !.,CI.<> c-,.l"o rlJbli~J.,ed
papers on the use of focus groups as a feminist methodology. Her recent
book Nothing Bad Ha!JpCllS to Good Girls: J<earofCri11le in WfOmcll's Lives
(1997) was nominated for the C. \Xlright Ivlills Award in ] 998. She is a
member of the board of directors of the (nstiruto Familiar La Raza in San
Francisco and sits on the editorial boards of Social justice and Peace
Review.
Kimberly A. 1Vlays de Perez is a nontraditional student holding a B.A. in
anthropology from the University of South Florida. She is especially inter-
ested in the role of culture in the practice of medicine. She currently works
as a medical translator at a Tampa, Florida, dinic that serves a variety of
Hispanic populations. She plans to use her background in anthropology in
her chosen career as a medical doctor.
Laurel Richardson is Professor Emerita of Sociology, Professor of Cultural
Studies in the College of Education, and Graduate Professor of \X-'omen's
Studies at the Ohio State University. She has wrinen extensively on quali-
tative research methods, ethics, 8nd issues of representation. She is the
author of seven books, including Fields ofPlay: COllstmcting all Academic
Life (1997), which was honored with the C. H. Cooley Award for the 1998
Best Book in Symbolic Interaction. Currently, she is interested in the rela-
tionships between conceptual and personal constructions of "timeplaces,"
narratives of the self, and knowledge practices.
Ray C. Rist is the Evaluation Adviser for the World Bank 1nsritute (WBI).
He is also the Head of the Evaluation and Scholarship Unit within the
\\TBI. Prior to his coming t:) the \'{1orld Bank in 1996, his career had
included 15 years in the U.S. government, with appointments in both the
executive and legislative branches. He also served for 12 years as a univer-
sity professor, \vith positions at the Johns Hopkins University, Cornell
University, and George \X!ashington University. He was the Scnior
Fulbright FelIO\v at the Max Pbnck Institute in Berlin, Germany, in 1976
and 1977. He has authored or edited 23 books, written more than 125
articles, and lectured in more than 40 countries. For the past 15 years, he
has chaired an international working group with represem<lrivc from 17
680
countries who are collabor:uing on research related to evaluation and gov-
ernance. He also has a 4-ye~l[-old granddaughter named 1\11011y.
fvlissouri-Columbia. He has scrved as coeditor of Cultural Alllhro!Jology
lvlethods]oumal (1993-1998) and has written and lectured on qualitative
data collection and analysis tcchniques, ethnographic decision modeling,
and response biases in the field. For 2 years, he was Associate Director of
the Fieldv,,rork and Qualitative Data Laboratory at the UCLA IVledical
School, \vhere he consulted with and trained researchers in text analysis.
His substamive interests in medical anthropology focus on how lay-
persons seleer among treatment alternatives across illnesses and cultures.
He has conducted fielth-vork in A'lexico and Cameroon and has published
in Social SciCltcc and iVIedicinc, Human Organizatioll, and Archives of
]I.;ledical Research.
David Silverman's interests arc in nonromanric qu~lIitative methodolo-
gies, professional-cliem communication, and conversation analysis. He is
the author of 14 books, the most recent of which are hzterpretiJlg Qualita-
tiue Data (1993), Discourses of COllnselling ( 1997), Haruey Sachs: Social
Science alld COl1uersatioJ1 Analysis (1998), and Doing Qualitatiuc Re-
search: A Practical Handbook (1999). Sillce ] 999, he has been Professor
Emeritus at Goldsmiths College, London Universily. He continues to
argue for a rigormls, theoretically based social science that maimai ns a dia-
logue with the wider community.
John K. Smith is Professor of Education at the University of Northern
Iowa. For the past 20 years, his interests have centered on the philosophy
of social and educational inquiry, with a special empbasis on the issue of
criteria. His work has appeared in such journals as :he Educational Re-
searcher,jollrnal of Educational AdministratioH, and Educational Eualua-
tiolt and Policy Analysis. He also has published two books: The Nature of
Social alld Educational Inquiry and After the Demise of Empiricism.
Eben A. \Veitzman received his Ph.D. in social and organizational psychol-
ogy from Columbia University. He is currently Assistant Professor, Gradu-
ate Programs in Dispute Resolution, University of M8ssachllsetts, Boston,
and Research Associate at the International Center for Cooperation and
Conflict Rcsolmioll. He is also the book and software reviews editor for
681
the journal Field Methods. During] 993-1995, he was a Visiting Professor
of Psychology in the New York University Psychology Department's pro-
gram in industrial/organization31 psychology. His interests are in organi-
zational development, cross-cultural conflict, conflier resolution, :1nd
intergroup relations. His current research focuses on intragroup conflict
in mediation, cultural differences in Jttiwdes toward conflict, organiza-
tional conflict, and the effects of cooperation Joel competition on small
group processes. He is the recipient of recent grams ro study "intragroup
conflict in mediation" and "cross-cul rural conflict on campus." His recent
publications on the use of software in qualitative research include articles
such as "Analyzing Qualitative Data \\lith Computer Sohware" and, with
the late IVlatthew B. IvIiles, "The State of Qualitative Analysis Software:
\\lhar Do \X1e Need?" and "Choosing Software for Qualitative Data Analy-
sis: An Overview," and the book Computer Programs for Qualitative Data
Allalysis (199.5), vi/hich he is currentlr revising in collaboration \virh Nigel
Fielding and R:l)' Lee.

682