Professional Documents
Culture Documents
Associate Editors:
Jacob L. Mey (Odense University)
Herman Parret (Belgian National Science Foundation, Universities of
Louvain and Antwerp)
Jef Verschueren (Belgian National Science Foundation, Univ. of Antwerp)
Editorial Address:
Justus Liebig University Giessen, English Department
Otto-Behaghel-Strasse 10, D-35394 Giessen, Germany
e-mail: andreas.jucker@anglistik.uni-giessen.de
Editorial Board:
Shoshana Blum-Kulka (Hebrew University of Jerusalem)
Chris Butler (University College of Ripon and York); Jean Caron (Université de Poitiers)
Robyn Carston (University College London); Bruce Fraser (Boston University)
Thorstein Fretheim (University of Trondheim);
John Heritage (University of California at Los Angeles)
Susan Herring (University of Texas at Arlington); Masako K. Hiraga (St. Paul’s (Rikkyo) University)
David Holdcroft (University of Leeds); Sachiko Ide (Japan Women’s University)
Catherine Kerbrat-Orecchioni (University of Lyon 2)
Claudia de Lemos (University of Campinas, Brazil); Marina Sbisà (University of Trieste)
Emanuel Schegloff (University of California at Los Angeles)
Deborah Schiffrin (Georgetown Univ.); Paul O. Takahara (Kobe City Univ. of Foreign Studies)
Sandra Thompson (University of California at Santa Barbara)
Teun A. Van Dijk (University of Amsterdam); Richard J. Watts (University of Berne)
85
Daniel E. Collins
Reanimated Voices
Speech reporting in a historical-pragmatic perspective
REANIMATED VOICES
SPEECH REPORTING IN A
HISTORICAL-PRAGMATIC PERSPECTIVE
DANIEL E. COLLINS
The Ohio State University
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Conventions for citing Cyrillic sources . . . . . . . . . . . . . . . . . . . . . . . . xix
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi
C 1
The pragmatics of reported speech . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Reported speech as intention and as perception . . . . . . . . . . . . . . . . 1
1.2 A “thick description” of reporting . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Some shortcomings of nonpragmatic reductionist approaches . . . . . . . 10
1.4 A method for historical-pragmatic analysis . . . . . . . . . . . . . . . . . . . 16
1.5 Previous applications of the “Method of Residual Forms” . . . . . . . . . 23
C 2
The text-kind: A pragmaphilological overview . . . . . . . . . . . . . . . . . . 27
C 3
Testimony . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.1 The standard reporting strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Strategic choices in the standard tag . . . . . . . . . . . . . . . . . . . . . . . . 42
3.3 Direct speech: Verbatimness or relevance? . . . . . . . . . . . . . . . . . . . 49
3.4 A convention of explicit typicality . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.5 From conduit to collaborator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
viii T C
C 4
Residual forms in testimony . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.1 Participial tags and foregrounding/backgrounding effects . . . . . . . . . . 75
4.2 Alternative tag verbs and different participant status . . . . . . . . . . . . . 85
4.3 Free direct speech and cohesion effects . . . . . . . . . . . . . . . . . . . . . . 92
4.4 Complementized reports: presupposition and grounding effects . . . . . 103
4.5 Fused reported speech and streamlining effects . . . . . . . . . . . . . . . . 116
4.6 Narrative reports of speech acts . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
4.7 Free indirect speech and the boundary with narrative . . . . . . . . . . . . 133
C 5
The question framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
5.1 The general function of the judges’ reported speech . . . . . . . . . . . . . 151
5.2 Dyadic questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
5.3 Reduction of the tag clause and rapid-fire exchanges . . . . . . . . . . . . 162
5.4 Verba-dicendi tags: turning the tables on the judge . . . . . . . . . . . . . . 165
5.5 Free direct speech and crossplay within a dominant encounter . . . . . . 167
5.6 Stage-managing: The dative with infinitive after speech-act verbs . . . 172
C 6
Reporting from judicial-referral hearings . . . . . . . . . . . . . . . . . . . . . . 177
6.1 The nature of judicial-referral hearings . . . . . . . . . . . . . . . . . . . . . . 177
6.2 Verifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
6.3 Falsifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
6.4 Preliminary reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
C 7
Layered reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
7.1 Preliminary remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
7.2 Additional layers of direct speech: narratives and dyads . . . . . . . . . . 205
7.3 Complementized indirect speech: rejoinders, hearsay, and hypothetical
reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
7.4 Additional layers of uncomplementized indirect speech . . . . . . . . . . . 228
7.5 Additional layers of fused reported speech . . . . . . . . . . . . . . . . . . . 233
7.6 Two functions of intercalated tags: Cohesion and disclaimers . . . . . . 237
7.7 Narrative reports of speech acts . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
T C ix
C 8
Reporting the verdict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
8.1 Preliminary remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
8.2 Reported speech in the adjudication . . . . . . . . . . . . . . . . . . . . . . . . 248
8.3 Reported speech in the judges’ ratio decidendi . . . . . . . . . . . . . . . . 252
8.3.1 Indirect and nondirect speech . . . . . . . . . . . . . . . . . . . . . . . 253
8.3.2 Fused reported speech . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
8.3.3 Direct speech . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.3.4 Intercalated verba dicendi and other quotation markers . . . . . 265
8.3.5 Narrative reports of speech acts . . . . . . . . . . . . . . . . . . . . . 270
8.3.6 Free reported speech in the ratio decidendi . . . . . . . . . . . . . 274
C 9
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
9.1 The purposiveness of reporting in trial transcripts . . . . . . . . . . . . . . 285
9.2 The main functions of the reporting strategies in the corpus . . . . . . . 288
9.3 Some methodological implications . . . . . . . . . . . . . . . . . . . . . . . . . 298
Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Appendix: Text-kind and date of the investigated trial transcripts . . . . 343
Name index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Subject index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
Acknowledgments
I would like to thank the many people who supported me in the writing of
this book. I owe special thanks to two department chairs — Stephen
Summerhill, who helped me to understand that a book of this kind was
possible; and Irene Masing-Delic, who showed constant solicitude about the
manuscript and provided me with travel support during the revision process.
I am also grateful to my friend Rafael A. Madan, who arranged for me to
use the Washington house of the Agrupación Católica Universitaria as a
research retreat, and to all the members of the Agrupación and residents of
the Washington house for their hospitality during my three visits. My thanks
also to the scholars who have commented on various portions of the book —
in particular, my teachers Henrik Birnbaum and Henning Andersen; my
colleague Brian Joseph; and Andreas H. Jucker, Richard J. Watts, Leslie K.
Arnovick, and other participants of the Historical Pragmatics panels at the
1998 International Pragmatics Association Conference and the 1999 Interna-
tional Conference on Historical Linguistics. I am especially grateful to
Andreas and Leslie for inviting me to present papers in these panels and to
the Ohio State University College of Humanities, Center for Medieval and
Renaissance Studies, and Department of Slavic and East European Languag-
es and Literatures, which provided financial support to enable me to partici-
pate. The anonymous referees of Pragmatics and Beyond gave me valuable
advice on improving the manuscript. Marika Whaley deserves special
recognition for her work in editing the manuscript and for the advice that she
provided during the revision process. I would like to thank Predrag Matejic
for allowing me to use the font OCS/PM, which he designed. Last but not
least, let me express my thanks and love to my family for bearing with me
during the process of writing and revising.
Preface
is not until someone forms a mental image of its meaning and chooses to
present it to interpreters in a specific context, for his own communicative
purposes. The goals of the reportee (attributed speaker), if any, are most
often irrelevant. The act of representation that mediates between reports and
their anterior utterances (real or projected) not only allows but even compels
reporters to impose their will upon both form and content through acts of
“responsive understanding” (Vološinov 1929/1986: 122–23), which can
include selection, choice of reporting strategy and contextualization, conden-
sation or amplification, and evaluation.
The pragmatic factors that are integral to the very definition of play
a central role in each individual act of reporting. Thus, in examining the
patterns of in a given language, it is not enough to treat them (as is done
in most of the previous studies of Old Russian) in a decontextualized manner
as an inventory of mere syntactic constructions, as if they were functionally
equivalent and randomly distributed (see 1.3). Rather, one must establish the
contextualizations of the various strategies and determine how they relate to
the intentions of the reporters. The goal should be to achieve an understanding
of that is as contextual and as close to the “emic” interpretive framework
of the participants as possible — that is, to create a “thick description” of
reporting, one that does justice to the “multiplicity of complex conceptual
structures, many of them superimposed upon or knotted into one another”,
which are involved in any social activity (Geertz 1973: 10).5
The fact that reporting is an intentional activity, an exercise of the will, does
not imply that every step in the process is necessarily conscious. Speakers
acquire an ability to reason from ends to means as part of their pragmatic
competence (cf. Brown and Levinson 1987: 58, 61, 64, 85); this ability is
automatized to such an extent that it is ordinarily not a subject of introspec-
tion.6 Thus intention can, but need not, imply deliberation or calculation; it
should be understood as a mental state oriented to a goal in the world — an
“operation-order” that motivates action (the means to the end), often on a
subconscious level (Austin 1966/1989: 274–77, 283–86).7 Intentional acts are
by nature purposive, directed at effecting a desired state of affairs or avoid-
ing an undesired one.
4 R V
‘continue’, and obratit’sja ‘turn to.’ Finally, there are verbs of mental
activity — predpoložit’ ‘surmise’ (2×) and zaključit’ ‘conclude’; manner-of-
speaking verbs — vskričat’ and voskliknut’ ‘exclaim’; and an attitudinal verb
— nadut’sja ‘pout, sulk.’
It is inadequate to explain diversity of this kind as mere “elegant varia-
tion” (to use Page’s term, 1988: 27). In using a wide variety of tags, authors
are trying to narrow the readers’ range of interpretive possibilities in order to
further their own communicative goals. Such use of nuanced vocabulary,
which is especially though not exclusively typical of modern literary
languages, is “a speaker-based strategy” (Lakoff 1984: 483–84), in which the
author gives the readers relatively little of the responsibility for sense-making
— in this case, evaluation of the represented speech events.
The repertory of tagging devices is not limited to verbs. In Old Russian
(), for example, could be indexed by adjuncts (2a), citation particles or
quotative markers (’s), textual conveyor nouns, or nominal labels (2b).10
(2) a. A po skaske [služylyx ljudej [tot
and by deposition- [of.service people]-. [that
Semejka ubit v [Pegoj orde
Semejka]- kill- in [Skewbald Horde]-
And, according to the deposition of the service people, the afore-
mentioned Semejka was killed in the Skewbald Horde (1648;
Tokarev (ed.) 1970: 896).
b. [V”spros” Inokentiev: “Gosudar’ Pafnotej!
[question Innokentij-]- [lord Pafnutij]-
Poveli… napisati zavěščanie o [monastyr’skom
bid- write- testament- about [of.monastery
stroenii…”
order]-
Innokentij’s question: “Lord Pafnutij! Bid [someone] write down
[your] testament about the monastic rule…” (ca. 1478; Dmitriev
and Lixačev (eds.) 1982: 496).
Another arena of choice is found in the composition of the tag clause apart
from the actual quotative device. In , reporters were faced, inter alia, with
decisions about tense and aspect, which could be varied for different effects,
T P R S 7
and about the order of elements within the tag, when it was not dictated by
rules of information structuring.
A further choice — one with ramifications for the salience of the report
in the discourse — involves the position of the tag vis-à-vis the report. For
example, in , tags can be preposed (3a), postposed (3b), or intercalated
(interposed/medial) (3c).
(3) a. I igumen mitropolitu tak rek”: Jaz…
and abbot- metropolitan- thus speak- I-
xožu po [staroi pošline…
go-.1 by [old custom]-
And the abbot spoke to the metropolitan in this way: “I… govern
according to the old custom…” (1391; ASÈI 3: 16, no. 5).
b. i vy buděte mně v” s(y)ny i
and you-. be-.2. me- in sons-. and
dščeri g(lago)let’ [g(ospod)’ vsedr”žitel’
daughters-. say-.3 [Lord Almighty]-
And ye shall be my sons and daughters, saith the Lord Almighty
(2 Corinthians 6: 18; Gennadij 1499/1992: 247).
c. i [t”t” pop” s [temi tvoimi gorodskymi
and [that priest]- with [those your- of.city
ljudmi… [dvorjan moix pere[bi]li: a
people]-. [servants my]-. beat-. and
bili, skazyvajut, na sm(e)rt’
beat-. say-.3. to death-
And that priest with those subjects of yours from the city… beat
up my servants; and beat [them], they say, [almost] to death (ca.
1451; ASÈI 3: 25, no. 9).
I treat intercalated reports as a separate strategy rather than a subtype of
postposition; this contrasts with the way the phenomenon is viewed by some
other scholars (e.g., Hermon 1979; Green 1980 with reference to English).
My reasons for this relate specifically to premodern Slavic languages, though
I suspect that they have a broader typological bearing.
First, conflation of intercalation and postposition obscures a striking
difference in distribution: as in English (see Partee 1973: 411; Hermon
8 R V
occur only within discourses — contexts that contain motivations for the
choice of one strategy over another.
The second alternative, syntactic change, can undoubtedly affect . For
example, long-term drift with continuous synchronic dynamism may be seen
in the gradual decrease in the use of bound reporting modes after ’s in the
history of Russian (see 1.2), or in German , where there is a tendency for
the indicative to replace the subjunctive and, where the latter is maintained,
for the present tense to replace the preterit (Hammer 1983: 265–66; Jespersen
1924/1992: 295–99). However, it must be emphasized that the extension of
one strategy at the expense of another generally has pragmatic causes; it is
a diachronic reflection of speakers’ synchronic choices, which are based on
socially-motivated preferences (cf. Keller 1985). Likewise, new strategies
develop as the result of individual acts that are exploratory and purposive:
“As they select novel and traditional expressions in accordance with their
individual hypotheses about their appropriateness… the speakers in effect
negotiate the norms that they look upon as their community norms” (Ander-
sen 1989: 25).20 Innovative strategies are acceptable to the community for
pragmatic reasons, e.g., because they are communicatively effective or
aesthetically pleasing. Thus it is not enough to offer exclusively historical-
syntactic explanations for the coexistence of reporting strategies (as is done
in much of the literature on ); one must also uncover the factors that
motivate the distribution in a given synchronic slice.
The third alternative, stylistic differences, can likewise play a role in the
reporter’s choices; however, they cannot account for every kind of variation.
It is almost always the case that various reporting strategies coexist within
the same styles, genres, and texts; the prevalence of one variety or other is
motivated by the needs of the discourse. Unless mere taxonomy is the goal,
one must propose some functional explanation to account for the overlapping
distribution of reporting strategies or indeed any other stylistic phenomenon.
possible to link the strategies with functions, and more than one function
must be operative; that is, the contextualizations of the strategies must have
known or inferable functions, which generally will reflect socially institution-
alized speech events or complexes of speech events.
Clearly, no methodology can make it possible to recover every possible
motivation for the selection of a strategy — e.g., the contextualization of a
reporting device — in a corpus of premodern writings; the real question is
whether it is valid and feasible to offer even partial matches between form
and function. The immediate answer is that one could not read and under-
stand premodern texts at all without inferring, however imperfectly, some
matches of this kind.21 Yet having enough knowledge of grammar and
lexicon to do accurate close readings still does not guarantee that one can
make form-to-function pairings that are “emic” — i.e., that even roughly
approximate the pragmatic competence of the original writers. However, as
when one is analyzing a modern foreign language, the risk of interference
diminishes in proportion as one becomes familiar with the cultural setting
and with a broad range of representative speech genres, which contributes to
intuitions about typical cooccurrence patterns and contextualizations. In other
words, the ability to make probable form-to-function matches depends, in
large part, on the “capacity for a truly empathetic reading” — the ability “to
transport oneself, as it were, into the very milieu in which [the texts] were
produced, reproduced, and read” (Birnbaum 1985: 171–72). While conclusive
verification may ultimately be impossible, the validity of an analysis can be
appraised, at the very least, for its plausibility on culture-specific and
typological grounds.
The first step in the analysis is pragmaphilological, in that it relates to
“the contextual aspects of historical texts” (Jacobs and Jucker 1995: 11–12).
Its aim is to establish the diplomatic structure, real-world function(s), and
reading conventions of the text-kind that furnishes the corpus. This informa-
tion is indispensable for “empathetic reading”, for acquiring a sense of the
knowledge and inferential patterns of the intended audience. The medieval
Russian trial transcripts used in this study furnish relatively rich internal
evidence for their situational context, and external evidence of somewhat
later provenience may also be found, e.g., in law codes and in travel writings
by foreigners; information of this kind is often not available in premodern
sources (ibid.: 6–7). Trial transcripts will be described in some detail in the
next chapter; for present purposes, it is enough to note that they typically
T P R S 19
in and of itself for following that convention (Lewis 1969). Hence a choice
that is intentional for some agents can be made automatically by others. At
first glance, this possibility seems to cast suspicion on the assumption that
any individual report reflects a context-sensitive use of discourse strategies
for specific effects — that is, effects other than the face-oriented one of
being seen to conform to convention.22 To a large extent, this problem is
resolved by locating the text-kind in its social context. Each section of trial
transcripts had an institutional function that the preferred reporting strategies
were proven to subserve efficiently. This rational motivation must have been
accessible to the pragmatic competence of the writers who formulated the
convention. Since, in order to persist, the convention must have been
validated continually by felicitous uses, its motivation must also have been
accessible to the enfranchised scribes who perpetuated it and who were
imitated by less competent writers unable to perceive its rational base. Thus
even scribes who were “merely” following convention were partaking, albeit
at one remove, in institutionalized intentions, and the result — the effect on
the intended interpreter — was the same. However, one may well doubt that
a scribe could be competent enough to follow conventions appropriately
while being utterly insensitive to the linguistic context.
The fourth step is form-to-function mapping; the goal is to identify
factors that could have motivated the correlations between reporting strate-
gies and contextualizations established in the third step. This procedure
entails a minute search for contextualization clues in the discourse, including,
in particular, the known or inferable function(s) of the given section (speech
genre) and the real-world speech events represented therein. Once I isolate
these potential communicative factors, I examine how the formal properties
of the strategies could have promoted the effective encoding of the message
— the represented speech act — in the given contextualization, from the
inferred standpoints of the writer and the intended interpreter. The pertinent
formal properties (see 1.3) include whether there is a tag clause and, if so,
how it is placed; what verbs of saying or quotative particles are used in it;
whether there is a complementizer; and what deictic pivot is chosen for the
report. Typological data on the communicative functions of analogous
strategies in other languages often prove relevant in this phase of the
investigation.
It turns out that the majority — sometimes the overwhelming majority —
of reports in each contextualization are presented by strategies that are not
22 R V
only conventional but also motivated. However, there is often a small residue
of deviant or problematic instances. It has been said that pragmatic analysis
does not need to account for 100% of the tokens in a corpus to be valid,
given the inconsistencies of human behavior (cf. Hopper 1979: 221, Note 5);
however, while this observation is valid on the whole, it does not absolve
one from trying to account for the exceptions in a fine-grained analysis. One
of the advantages of operating with prototypes is that they establish a
framework in which to explain many of the exceptions in a non-random
manner, so that the results in fact come closer to accounting for 100% of the
examples in the corpus.
The fifth step in the analysis is to account for the problematic tokens by
means of the “method of residual forms” mentioned above. I examine the
context and content of the exceptions (the residual forms) to isolate any
features that distinguished the represented speech events from the prototypes.
It is almost always possible to find atypical features in most of the deviant
cases — a fact that validates both the posited prototypes and the methodolo-
gy itself. Identifying additional contextualizations of this kind as potential
conditioning factors leads to a second phase of form-to-function mapping —
comparable to the fourth step, above — with the aim of discovering the
properties of the unconventional strategies that would made them appropriate
for such “highly entailing effects” (Silverstein 1985: 142–43). This accounts
for nearly all of the apparent exceptions. Many of the deviations fall into
classes that may be viewed as speech genres in their own right — e.g.,
choral reports (see 4.3) or boundary-setting descriptions with a mix of
gestural testimony and spoken commentary (see 4.1–4.3). Other deviations
are unique but still amenable to classification, like the case in which a
litigant calls one of the judges as a witness — a striking change in footing
(in the sense of Goffman 1981: 137). Indeed, many of the exceptions appear
to index changes in the participation status or configuration of the partici-
pants in the represented trials.
By now it will have become clear that the methodology outlined above
presupposes not only that the writers of the investigated documents were
rational in their choices but also that they were pragmatically competent in
the speech genre. The latter assumption may not be self-evident; given the
large percentage of conventional tokens, one might suspect that the low-
frequency residual forms were not purposive but rather inappropriate usages.
While this cannot be ruled out, there is no reason to ascribe residual forms
T P R S 23
take only one example — that , which is typically preferred, e.g., with
reports that are foregrounded in narrative (see 3.5), is actually simpler than
alternative strategies like , since it involves “decentering” or displacement
of indexical ground (Hanks 1990: 206). Assuming for the sake of argument
that it is, one can still find many contexts in which putatively more complex
reporting methods are “basic usages” (cf. D. Collins 1996; Ebert 1986).
(This, indeed, is a general problem for the concept of markedness in gram-
mar, epitomized by the preference for the marked perfective in narrative, the
prototypical past-tense contextualization.)
Finally, the results of my investigation do not bear out the notion
(Silverstein 1985: 142–43) that residual forms invariably have the foreground-
ing effect predicted for marked forms, whether in the Praguean sense of
“actualization” (Havránek 1932, 1964) or in any other sense. While they
undoubtedly have “highly entailing effects”, these do not necessarily equate
with foregrounding. The results of my analysis suggest that an element can
be contextually unexpected without being prominent; this is particularly the
case when the expected element is itself associated with a relatively high
degree of foregrounding.
A program of analysis that is similar to the method of residual forms has
been outlined by Brown and Levinson (1987: 282), who observe that
“nonoccurrence of expected forms, ‘exceptions’ and ‘breeches’, can be made
sense of by analyzing the rational sources for those occurrences in an
otherwise homogeneous style with other rational sources.” A further kindred
method has been proposed by Derjagin (1971: 151–52), whose approach
derives from the ancillary historical discipline of diplomatics. Derjagin’s
methodology is interesting in the present context primarily because he
applies it to medieval Russian legal/administrative texts. His first step is to
establish the syntactic and lexical patterns that are standard in a given text-
kind. Next he identifies features characteristic of individual writers by
comparative analysis of a large corpus of texts. The final step is to juxtapose
the works of a single scribe to isolate any peculiarities that may be ascribed
to the principal for whom the document was written (the person who
commissioned the document and who is explicitly represented in the first
person). In Derjagin’s view, the usage of individual authors becomes a factor
only in the last two phases of analysis. This approach has three major
shortcomings. First, it provides no explanation for residual forms other than
idiosyncracy or individual whim. Second, it fails to account for the fact that
26 R V
there can be variation even within individual usage. Third, it sets up a model
of the writer as a slave. In separating convention from deviation (“individual-
ity”), it does not do justice to the fact that conventional strategies can be,
and usually are, intentional — in other words, that they are an aspect of
individual usage.
C 2
The text-kind
A pragmaphilological overview
Medieval Russian trial transcripts are detailed protocols of trial hearings that
include extensive representations of dialogue — the testimony of litigants and
witnesses interspersed with the questions and statements of the judges — set
in a third-person narrative. While these documents adhere to the general norm
seen in narrative in that they recount sequentially ordered past actions, they
differ from it in two important respects. First, there is usually little apparent
causality in the chain of narrated actions, except in the representations of
adjacency pairs (dialogic exchanges); the question-and-answer dyad is the
central organizational principle of the text-kind. Second, the main focus of
interest or information load is the ; much of the narrative framework
consists of tags (attributions), while motions and nonverbal behavior are
often treated as incidentals or (if one can draw any conclusions from silence)
not reported at all. Moreover, even as “dialogue discourses” (Longacre
1983: 44), the transcripts are peculiar in that the does not so much tell a
story as deliver raw information; the writers leave their intended interpreters
the task of sifting through the data and determining their ultimate relevance.
Most of the extant trial transcripts are records of real-estate lawsuits; a
few are accounts of criminal proceedings involving property violations such
as theft, arson, or abetting runaway serfs.1 The text-kind is thought to have
arisen around the mid-thirteenth century (Leont’ev 1969: 45–46); the oldest
surviving examples, which are discussed in this study, date from the first
decades of the fifteenth century. While there are some earlier documents
relating to lawsuits, they are not trial transcripts in a strict sense; this is the
case, for example, with a verdict record of 1284, which contains no reports
28 R V
of testimony and is framed from the first person of the judge (Avanesov (ed.)
1963: 62). The oldest surviving document with protocols of dialogic testimo-
ny embedded in narrative is a “monastic-rule charter” of 1391 (ASÈI 3:
16–17, no. 5); this edict has a different structure than trial transcripts, and
part of the narrative is presented from the judge’s first-person viewpoint.
The corpus used in this study comprises 154 trial transcripts (some extant
in more than one copy) from the territories that consolidated to form the
Muscovite state.2 To my knowledge, this is an exhaustive corpus for the
period from 1400–1505 — that is, the period prior to the official review of all
charters in 1506, at the beginning of Vasilij III’s reign. While many of the
documents are originals or contemporary copies, some are later copies with
lost protographs; these evidently reflect the medieval legal custom of word-
for-word copying — verifiably so, in some cases where there are multiple
copies from a single protograph (see also 3.4). I make special note of instances
where there are traces of editing that are pertinent to reported speech.
The trial transcripts in the corpus are attested in three main subtypes —
trial records, judgment charters, and referred trial records.3 These subtypes
are not distinguished consistently in medieval sources. In addition, copyists
sometimes confused trial transcripts with similar text-kinds such as records
of amicable settlements (e.g., ASÈI 1: 235, no. 326; 288, no. 397; 320, no.
431).
Trial records (sudnye spiski — literally, “trial copies”) are transcripts of
judicial hearings held on disputed properties, which follow the order of the
real-world events. Their writers, who are usually anonymous but were
apparently professional scribes, took notes of some kind during the hearings
and then drew up polished versions in a conventional format prior to the
judges’ deliberations. The purpose of trial records was evidently to enable
the judges to review — indeed, reexperience (see 3.5) — the testimony when
this was necessitated by the complexity of the evidence or delays in the trial.
Some forms of delay were institutionalized. Real-estate trials were
generally held on the disputed property, and the on-site judges did not
always have the legal authority to hand down verdicts. In such cases they
referred the trials to courts of higher instance presided over by grand princes
or princesses, high-ranking boyars, or metropolitans. The higher judges then
held hearings in which the trial records were read aloud, and the litigants
were asked to verify that they were accurate (see Chapter 6).4 This procedure
was a safeguard against errors and falsifications; if one of the litigants
T T-K 29
disputed the record, it was submitted to the “men of court” (sudnye muži) —
official observers of the on-site hearings — for corroboration. After estab-
lishing that the records were accurate, the higher judges used them to review
the testimony prior to issuing their verdicts (see Chapter 8).
It must be stressed that the primary motivation of the verification process
was not to ensure accuracy in the written document, although that was
certainly a by-product. By custom, the higher judges and litigants listened to
the transcripts, which were read aloud by an officer of the court. In addition,
the trial judges sometimes made preliminary reports, again in oral form, at
the beginning of the referral hearings. Medieval Rus’ was a “mixed-orality”
culture, in which “the influence of writing remain[ed]… partial” (Fleischman
1990: 20). Trials, in particular, were an institution that was still largely oral
in its basis.5 Thus the written aspect of trial records was incidental; the docu-
ments were, in their essence, a means of reanimating oral situations (see 3.5).
Relatively few trial records have survived in separate form, probably
because they reflect an unresolved stage in the judicial process. The majority
of the extant transcripts include both a trial record and a narrative account of
the verdict. Like the judicial referral, the verdict seems to have been primari-
ly an oral institution. The judges apparently announced their decision aloud;
court scribes then drew up a conventionalized verdict record in third-person
narrative, which was appended to the trial record and the judicial-referral
record, if any, to make up the completed trial transcript. There are several
facts that provide evidence for this two-tiered process, including a verdict
record written in the first person from Pskov (Northwest Rus’; GVNP no.
340); occasional cases of slipping into the first person in verdicts from
Northeast Rus’ (see Chapter 8); and specific references to oral verdicts in
seventeenth-century trial dossiers (see the texts in Jakovlev 1943/1970).
Trial transcripts with verdict records fall into two subtypes, which seem
to have been equivalent in their force, though administratively distinct.6
Judgment charters — Kleimola’s (1975) translation of pravye gramoty —
were awarded to the victorious party as proof of ownership; they gave their
bearers the legal leverage to implement the judgment and to protect them-
selves against future infringements. Referred trial records (dokladnye
(sudnye) spiski) are documents issued by courts of higher instance, with
accounts of the judicial-referral hearing and verdict appended to the trial
record. These were usually given to the judges of the on-site trial, who
executed the judgments of the higher courts and issued judgment charters to
30 R V
the victors; however, they were apparently sometimes given directly to the
victorious litigants in lieu of a special judgment charter.
The trial transcripts investigated are relatively uniform in structure. They
typically consist of three or four sections, each with its own characteristic
formulae. They begin with an incipit, which provides orientation by identify-
ing the main participants and, in some cases, the location of the trial — often
nothing more specific than “having stood on the [disputed] land” (see
Kleimola 1972: 357–58). Many transcripts open with a brief formula indicat-
ing that some ruler has authorized the trial (cf. Kaštanov 1970: 37–38); this
generally takes the form po X- slovu “in accordance with X’s command”
(by X- word-).
While the authorization formula allows the documents to be dated to a
range of years (sometimes more precisely, with the help of other internal
evidence), the date itself is never included in the setting information. The
explanation for this absence — an omission, by modern expectations — is
that the documents themselves “did not yet fully constitute the legal act
which they set down”, as Kittay (1988: 211) notes of medieval charters from
Western Europe. Prior to the verdict, the written transcripts were used only
in the context of “personal attendance and verbal confirmation” (ibid.); this
can be seen in the procedures at judicial-referral hearings, where the validity
of the documents depended on the memory of participants — the trial judge,
litigants, and men of court — who carried knowledge of the date and
location within them. Only with the issuing of the verdict, when the tran-
scripts were cut loose from this context of active, verbalized (or verbalizable)
remembering, do their dates become potentially relevant as orientation for
readers/interpreters with no involvement in the trial process. Even so,
relatively few of the documents are internally dated, as compared with other
chancery text-kinds; of the 152 transcripts investigated, only one (AFZX no.
248) has the date within the verdict, while 34 others have it accompanying
the seals and/or signatures in the end protocols.7 Only one of the 13 tran-
scripts from before 1462 is dated (ASÈI 3: no. 32). Significantly, this
document comes from the chancery of Metropolitan Photios, a Greek, and is
dated in the ecclesiastical (originally Byzantine) style, by indiction; in other
words, it reflects, in part, legal institutions in which writing played a greater
role than in ordinary medieval Russian law. Of the remaining transcripts with
internal dates, only 4 (out of a possible 57) are from before the 1490s; 11
(out of a possible 53) come from the period 1491–1500, and 19 (out of a
T T-K 31
possible 29) are from 1501–5.8 These figures suggest that the date was
gradually assuming greater importance; however, during the period investi-
gated it remained optional, of uncertain relevance for any of the conceivable
interpreters.
In most incipits, the judge is introduced after the authorization formula
(if any) with the set phrase exemplified in (1a), from a judgment charter of
ca. 1448–52, which I chose as the main illustration in this chapter for the
sake of its brevity:
(1) a. †Si sud sudil [Mixailo Fedorovič(’).
[this trial]- judge- [Mixajlo Fedorovič]-
† Mixailo Fedorovič [Saburov] judged this trial
(ASÈI 3: 58, no. 35).
The word order in this prepatterned sentence is, as expected, “presentational”
— inverted, as compared with the neutral Subject-Verb order — because the
subject is asserted; the topic is the direct object “this trial”, which is presup-
posed from the very existence of the record. The formula exemplified in (1a)
can be elaborated with a participial (gerund) clause specifying the venue of
the trial — stav” na/v” X- “having appeared at/in X” (stand- on/in
X-). The next sentence is normally a set phrase, again in presentational
word order, that introduces the litigants, with plaintiff first (1b):
(1) b. Tjagalsja [arximandrit Feofan s Vašutoju, da…
sue- [abbot Feofan]- with Vašuta- and
Abbot Feofan sued Vašuta and [three other respondents are
named] (ibid.).
This formula has several possible elaborations, which need not be discussed here.
The incipit is followed by the trial record proper, the transcript of the
lower-court hearing (see Chapters 3–5). Typically this opens with the plain-
tiff’s charge against the respondent (2a); the judge is then depicted asking
the respondent for a rejoinder (2b):
(2) a. Tak rek [arximandrit Feofan: Žaloba
thus speak- [abbot Feofan]- complaint-
mi, g(ospodi)ne, na [tog(o) Vašutu, da…
me- lord- against [that Vašuta]- and
32 R V
Testimony
the frame in which direct reports of testimony were most frequently set.
In the transcripts investigated, one may identify a single standard report-
ing strategy, which accounts for the majority of the reports attributed to the
litigants and witnesses. This strategy is attested in virtually all the cases of
what may be termed prototypical testimony, the speech event that was
perceptually central in the trial institution. Prototypical testimony — a
concept that will recur throughout this study — is an uninterrupted statement
made by one of the principals (the plaintiff or respondent), which provides
information intended to establish the speaker’s rights to the disputed proper-
ty; it is elicited by the judge(s) in a face-to-face exchange, presumably in
some standard configuration determined by judicial custom. The prototypic-
ality of this speech event stems from the fact that, under ordinary circum-
stances, it was better suited than any other kind of utterance to supply
information that could be probative for the verdict — proofs of ownership
offered by the parties who were interested and legally liable in the lawsuits.
Further evidence that testimony in this strict sense was perceptually central
may be found in certain abridged transcripts (e.g., ASÈI 2: 504–5, no. 465),
where the only statements reported in are those that conform to the
prototype, along with the questions that elicited them. (The latter reflect the
prototyical speech event for the trial judges — requests for information about
the disputed property, addressed to and cooperatively answered by individual
participants, again in a face-to-face exchange.)
The standard formula for introducing testimony is given in two predictable
positional variants in (1a–b).2 This tag, which forms a link in the poly-
syndetic chain of sequenced narrative clauses in the authorial framework (see
3.2), always precedes the report, without any intervening complementizer.
(1) a. Initial report: tak(o) rek(l”)/rekli X- …
thus spoke-./ X-
X spoke in this way: []
b. Elsewhere: i X- tak(o) rek(l”)/rekli…
and X- thus spoke-./
And X spoke in this way: []
This reporting strategy is otherwise characteristic of only a few of the text-
kinds that are known to have existed in legal-administrative writing — in
T 37
Non-responses, which are rare, can be signalled in unusual ways (see 4.6).
As in narratives of dialogue in general, it was evidently expected that
questions would be answered by their addressees, in accordance with the
norms of cooperative dyadic conversation. The intended interpreters of trial
transcripts had good reason to infer that this convention was in effect, given
the relative power of the judge in the real-world situation (cf. Brown and
Levinson 1987).
The predictability of speakers in the question-and-answer structure
explains, in part, sporadic cases in which the tag clause is omitted (see also
4.3). Likewise, because the plaintiff was always the first speaker, the
attribution of the opening complaint could occasionally be omitted (3);
interpreters with the expected background knowledge of the text-kind could
infer it from the incipit, in which the plaintiff is conventionally the subject.
(3) Tjagalsja Savka Mandakov… s Ysačkom s
litigate- [Savko Mandakov]- with Isačko- with
[Vaskovym s(y)n(o)m” Obnorin(a)… To, g(ospodi)ne, [otvotčiki
[Vas’ko- son]- Obnorin- lord- [rentmen
kirilovskie… otveli u menja, u Savki,
of.Cyril]-. take.away-. at me- at Savko-
počinok k” [svoei pustoši… x Kočevinskoi…
clearing- to [. empty.land]- to Kočevinskaja
Savka Mandakov… sued Isačko, son of Vasko Obnorin… “The rent-
men of Cyril [Monastery]… took away from me, Savka, a clearing
[and annexed it] to their uncultivated property of Kočevinskaja…”
(ASÈI 2: 192, no. 285).
However, null tag clauses, like other forms of ellipsis, are not typical in trial
transcripts; the scribes aimed at maximal clarity and, in particular, tried to
make the relationship between the authorial context and the explicit to
preempt misinterpretations.
This effort to be explicit, at least with information potentially relevant
for the judgment, is reflected in the typical ways of referring to the repre-
sented speaker. As a rule, the reported speaker is identified briefly by a
unique name, nickname, or title — e.g., Saltyk da Vislo and Kirilo in (2) or
poselskoj (“estate manager”; Kaštanov 1970: 387, no. 27). When witnesses
are presented as a group, in many cases only one or two of them — possibly
T 39
the spokesmen in the real-world trial — are named individually; the others,
mentioned by name in the litigants’ testimony, are referred to by the noun
phrase i ego/ix” tovarišči “and his companions” (and his/their fellows-
.) or variant (e.g., ASÈI 2: 409, no. 401). Such generic labels are
hardly ever used to present multiple litigants, who were more individuated
in the scribe’s perception because of their prominence in the trial.
The preferred referential strategy in the tag clauses is thus to use noun
phrases with unique referents, even though the referents are usually already
thematized in the preceding narrative. Pronominal anaphora scarcely ever
occurs in the tag clauses, even though it is common within the ; indeed,
it is rarely used with the subjects of other narrative clauses. This avoidance
of anaphora is a convention specific to trial transcripts; it is not typical of
other kinds of narrative or legal-administrative writing. One of the only cases
in the corpus (4) appears in a passage that exhibits several other unusual
features as well (see also 4.2):
(4) I Mixal’ tako rek: Jaz, gospodine, zemli ne
and Mixal’- thus speak- I- lord- land-
znaju, mne sja zemlja dostala novo, a
know-.1 me- land- obtain- newly and
znaet, gospodine, zemlju [Fedor Kokoš, poselskoj
know-.3 lord- land- [Fedor Kokoš steward]-
[velikogo knjazja. I sud’ja vsprosil Kokoša, i
[grand prince]- and judge- ask- Kokoš- and
Kokoš molvit: Jaz, gospodine, mež ne
Kokoš- say-.3 I- lord- boundaries-.
znaju. I sud’ja tako rekl: Daj že
know-.1 and judge- thus speak- give-
ty mne starožilcev, xto znaet [zemle
you-. me- old.residents- know-.3 [land
Čekmakovskoj s [Putilovskoju zemleju meži.
Čekmakovskaja]- with [Putilovskaja land]- borders-.
I on postavil [Ontipu desjatskogo, da Frola…
and he- put.up- [Ontipa tenthman]- and Frol-
And Mixal’ spoke in this way: “I don’t know the land, lord, I just
obtained it, but Fedor Kokoš, lord, the grand prince’s steward, knows
40 R V
the land.” And the judge asked Kokoš, and Kokoš says, “I don’t know
the boundaries, lord.” And the judge spoke in this way: “Then give
me witnesses who know the boundaries of the Čekmakovskaja proper-
ty and the Putilovskaja property.” And he put up Ontip the Tenthman
and Frol… (ASÈI 1: 321, no. 431).
This passage provides a clear illustration of why the scribes avoided pronom-
inal subjects. The referent of the pronoun in the underlined clause is ambigu-
ous; it could be interpreted either as Mixal’, the respondent, or as Fedor
Kokoš, the last speaker before the judge’s command, whom Mixal’ names as
co-respondent. As the nearest possible referent other than the judge (who is
excluded because of his role in the trial), Kokoš is the most likely referent
of the pronoun, but Mixal’ is actually more probable in view of the partici-
pation framework of the speech event. The interpreter would have to fall
back on background knowledge for the information that it was the litigants
who were ordinarily responsible for producing witnesses. Even this does not
exclude Kokoš, since Mixal’ has tried to transfer responsibility to him.5 As
the estate manager, Kokoš would certainly be expected to know the lands
under his care or, at least, to be able to name knowledgeable locals as
witnesses. Such potential difficulties in interpretation were undesirable in a
text-kind whose purpose was, in large part, to provide enough information to
ensure a fair judgment. As a rule, the scribes strived to made their texts as
explicit and intelligible as possible and accordingly avoided anaphora and
other potentially ambiguous devices. (However, null reference is often
employed in one particular unambiguous context; see 4.3.)
There are three chief ways in which the standard tag formula can be
elaborated. First, the predicate can include an adjunct identifying the
attributed speaker as the spokesman for a group:
(5) a. tak rek Stepanko i v” [vsěx” kr(e)st’jan
thus speak- Stepanko- in [all peasants]-.
město…
place-
Stepanko spoke in this way on behalf of all the peasants… (ASÈI
1: 478, no. 589).
Second, the entire tag can be preceded (5b) or, less often, followed (5c) by
a clause consisting of a past gerund (deverbal adverb) plus one or more
T 41
’s. This kind of elaboration specifies the location of the trial or given
segment of the trial:
(5) b. Stav na mežě, tak rek Davydko…
stand- on boundary- thus speak- Davydko-
Having stood on the boundary, Davydko spoke in this way…
(ASÈI 1: 246, no. 340).
c. I Okiš, da i Larivon… tak rkli,
and Okiš- and Larivon-… thus speak-.
stav na sudě…
stand- at court-
And Okiš and Larion…, having stood [or appeared] in the place
of judgment, spoke in this way… (ibid.: 247).
As orientations, such gerund clauses often occur either with the tags for the
opening turns, as in (5b), or at episode boundaries, as in (5c), where witness-
es mentioned previously in — given but not fully activated in the
discourse — are introduced for the first time in the narrative. In a few cases,
new participant configurations are conveyed by other devices — for exam-
ple, the prepositional phrase pered” sud’eju “before the judge” (before
judge-),6 or the adverbial clause i vystupjasja “and, having come
forward” (and step.forth-).7
Third, the standard tag could be elaborated by a noun phrase identifying
the represented addressee. There is evidence from other text-kinds that the
reči could take an indirect object (or, in ecclesiastical writing, a preposi-
tional phrase) referring to the addressee; however, this is rare in trial
transcripts. Omission of this information could have little effect on the
interpretability of the text; the judge was the only ratified addressee in the
represented speech event, and trial procedure (if the internal evidence of the
written transcripts is any guide) required the participants to speak only when
so instructed by the judge. In most transcripts, economy was thus more
important than explicitness in this matter; there is only one transcript in
which a reference to the represented addressee (sud’i judge-) appears
repeatedly in the tag clause (14 out of 31 tokens). Significantly, in 12 of
these cases, the participants seem to have undergone a change in footing.
They are depicted speaking without prompting or in other contexts that
42 R V
1980: 592). However, the majority of preposed tags in trial transcripts are not
episode-initial. Hence the relation established by preposing would seem to
give the tag undue salience, especially in a “dialogue discourse” (Longacre
1983: 44), in which the most important information is conveyed in . After
all, the attribution was generally predictable, since answers were expected to
follow questions in the dyadic structure (see 3.1).
In fact, preposing the tag was well motivated for several reasons. First,
preposed tags served as a form of punctuation, demarcating adjacent reports
from one another. Hypothetically this could have been accomplished by a
string of postposed tags, but note that the trial record begins with a preposed
one; in any case, only preposed tags could serve to focus attention preemp-
tively on the reports. Back-to-back reports, like “colliding speeches”
(Brandsma 1996: 257) in general, would have posed processing difficulties
not only for the readers, given the graphic conventions of the transcripts, but
also for the hearers — a category to which at least some of the intended
interpreters belonged.
Second, the fact of the utterance, which was asserted by means of the
tag, was indeed relevant information for the purpose of the text-kind, not a
mere afterthought; the content did not have legal meaning unless it was clear
whose position it established. In the verbal duel of the trial, the individual
voices were not collaborating to tell a unified story.
Third, pre-indexing the reports ensured that acts of testimony would not
be interpreted without prior information about their source, which was
crucial for appreciating their legal significance. This partially phatic function
— much like the identification of the caller at the onset of a telephone
conservation — likewise served the needs of hearers, who were unable to
scan the transcripts as they were read aloud in the actual trials.
The repetition of the same prepatterned tag, with variations only in the
identity of the speaker, promoted interpretability in and of itself in that it
helped to establish a narrative framework — a plotline that provided orienta-
tion clues to the dialogue. Syntactically, the frame is a chain of tag clauses
begun by the adverb tak(o) in the first turn at speaking (see (1a),
Section 3.1), occasionally interspersed with other kinds of narrative clauses.
Each link in the chain is conjoined to its predecessor by the sequencing
conjunction i ‘and’ (see (1b), (2), and (4), Section 3.1).
It should be emphasized that the combination of and narrative based on
the sequentiality of real-world events was intentional rather than automatic,
T 45
since the scribes had other means of representing trials at their disposal.
Other principles of organization may be seen, e.g., in verdict records,
especially those appended without trial records to other kinds of documents,
as in (6), taken from an immunity charter of 1454.11
(6) potomu čto [tě pustoši izstariny za šest’desjat lět
because [those fields]-. of.old for sixty- years-.
potjagli x Karinskomu selu… a i [sam tot
belong-. to [Karinskoe village]- and [self that
Vasko skazal, čto [tě pustoši pašut
Vas’ko]- say- that [those fields]-. plow-.3.
xrest’jane [Karinskog(o) sela, kak sja dostalo
peasants-. [Karinskoe village]- since obtain-
[to Karinskoe za manastyr’; a Vasko iskal
[that Karinskoe] for monastery- and Vas’ko- seek-
ix za [kn(ja)žie zemli za ondrěevskie.
them- for [prince- lands]-. for Andrej-..
… because those fields of old, over sixty years, belonged to the vil-
lage of Karinskoe… and even Vas’ko himself [the losing litigant] said
that the peasants of the village Karinskoe have been farming those
fields since that Karinskoe became the monastery’s property; [1] but
Vas’ko was seeking them as Prince Andrej’s lands
(ASÈI 3: 83, no. 54a).
Here there is minimal attribution and no indication of the real-world ordering
of the represented speech events; the report marked [1], in particular,
conveys a claim that was probably voiced at the beginning of the trial.
Clearly this nonsequential, summarizing form of organization is highly
evaluative and so would not have served the institutional function of trial
transcripts — providing judges with unprejudiced raw material for their
deliberations.
Another component in the standard framework was the rek(li) (from
reči), which, as a perfective preterit verb, served to create a plotline of
foregrounded events, in accordance with the narrative norm. The narrative
norm in ecclesiastical writing (Church Slavonic) of the same period was the
aorist (Chvany 1985a). While this tense was obsolete in northeastern
chancery writing, it occurs instead of the preterit in the fifteenth-century
46 R V
events. It is doubtful that they included the vocative out of a desire for
verisimilitude, given that they evidently screened out other expressions of
interpersonal meaning and expressive elements that were presumably uttered
during the trials. (For example, the corpus does not contain a single insult,
despite the confrontational nature of the proceedings.) Clearly the vocative
had some function other than appellation.
It may be observed that g(o)s(podi)ne tends to occur in almost every
clause of the testimony in second position — after the the enclitics in
Wackernagel’s position or after the first major constituent. This is a slot
often occupied by macrostructural connectors and other cohesive devices,
and indeed the vocative may be interpreted as a marker — a unique and, in
the technical sense, typical feature of that could be exploited to promote
coherence by reinforcing other indicators and by disambiguating passages
in which such indicators were absent. As Fludernik (1993) cogently argues
in her theory of schematic language representation, typicality markers of this
kind play an important role in the construction of all modes of reporting.
This is true even of , which is not intrinsically mimetic, as often claimed,
but often features typification strategies, i.e., relies on adequation to type
rather than to token.17 Because the texts were written from a covert third-
person viewpoint, the appearance of a typical non-narrative feature such as
g(o)s(podi)ne was sufficient to identify a passage not only as but also as
testimony — indeed, as nonreported portions of testimony; the vocative
never occurs in the authorial narrative and tends not to appear in additional
layers of within the testimony.18
Table 1 summarizes the distribution of g(o)s(podi)ne by clause type in a
sample of 40 northeastern transcripts (nearly one quarter of the corpus). These
data show that the scribes tended to insert it in clauses that could be perceived
as potentially separate segments, as defined by a change of theme, topic, or
subject. The likelihood of its being repeated from one clause to another varies
inversely with the cohesiveness of the two clauses. These facts bear out the
interpretation that g(o)s(podi)ne served as a signal of continuity with the
preceding and, by its very repetition, made the into a cohesive chain.
T 49
In the following sections I shall examine the factors that made the
preferred strategy for reporting dialogic testimony in the narrative framework
of trial transcripts. This discussion will require a foray into debates about the
nature of the direct mode. Here my goal will be to rule out two prevalent
and related hypotheses — first, that is verbatim reporting; second, that
reflects the reporter’s commitment to faithful transmission of the wording
and structure of the anterior utterance.
At first glance, the fact that was the principal way of representing
testimony in trial transcripts seems not to require any explanation. After all,
is usually considered the most “simple”, “basic” mode of reporting not
only in but in all languages.20 This assumption is, I suspect, founded
sometimes on the ontogenetic priority and typological universality of but
more often on the notion that is in essence the reproduction of “original
utterances.” It is not, in fact, entirely clear that the assumption is justified or,
indeed, what it would mean from a pragmatic perspective, in which speech
reporting is treated as intentional activity. Whenever a choice is available,
there are text-kinds and discourse contexts in which modes other than are
not only statistically predominant but also perceptually normal (“basic”) for
pragmatically competent speakers (cf. Aaron 1992 on Obolo narrative). For
instance, in the earliest Slavic writings (roughly four to six hundred years
older than the trial transcripts examined here), there are several well-defined
situations in which is far less typical than , though reliance on context-
50 R V
blind overall frequency has misled some scholars into thinking that was
nonexistent or incipient (for discussion, see D. Collins 1996; cf. Ebert 1986
on in languages of Nepal). As examples of this kind show, the coexistence
of differently distributed reporting strategies imposes a methodological
imperative: one must seek out the factors that underlie the context- and text-
kind-based preferences for each strategy, including the one with overall
statistical predominance, before generalizing about the “basic” mode of
reporting in the language as a whole. I would argue that the notion of default
modes is at best an oversimplification and often an illusion; putatively
“basic” modes have pragmatic rationales.
In any case, the appearance of in trial records was by no means
inevitable, even in representations of dialogue; for example, Muscovite
interrogation records (rassprosnye reči), which consist mostly of reports, make
extensive use of and relatively little of .21 The scribes had other strategies
at their disposal and used them extensively in other sections of the docu-
ments (see Chapters 6–8), occasionally even in the trial record (see Chap-
ter 4); thus their use of reflects a continual exercise of their speech will.
If default is a pragmatically inadequate explanation, what did motivate
the choice of as the standard mode for reporting testimony? It cannot be
dismissed as “mere” convention, given that conventions have rational bases
and serve as solutions to coordination problems (Lewis 1969). One answer
that immediately springs to mind is that the scribes of trial transcripts chose
as the only feasible means of providing an exact (literal, word-for-word,
verbatim) rendition of the trial dialogue for the judges’ information, in the
manner of modern court reporting. In such a case, the documents, at least in
intention, would be transcripts in a strict sense — reproductions of the actual
proceedings. This interpretation would accord with the traditional and still
widespread conception of the nature of , which Coulmas (1985a: 41)
characterizes as “generally received opinion” — the verbatimness or “repro-
ductionist” (Sternberg 1982b) hypothesis. According to this model, is a
mode that features “word for word reproduction (indicating the quoted
speaker’s style, in the sense of his lexical choices)” (Banfield 1973: 9). This
explanation of is institutionalized in traditional terms like direct speech
and oratio recta, which presuppose the possibility of “straight” quotations,
without authorial tampering.22 Indeed, the term direct speech (or equivalent)
is occasionally used to denote not only a mode of reporting but also what
was actually said by the anterior speaker.23
T 51
For all its intuitive appeal, the verbatimness model of is not tenable,
as has been demonstrated trenchantly in a number of recent studies. It is, of
course, possible to reproduce anterior utterances by means of (Sternberg
1982b: 111), as can be typically seen in written citations from written
sources; indeed, as Haberland (1986: 225) points out, “if one wants to
commit oneself to the form of the model as well [as the content], then one
has to avail oneself of direct speech.” However, verbatim reporting is only
one subtype of ; given the existence of others (see below), it would be a
fallacy of composition to include the notion of verbatimness in the definition
of the category: “From the premise that … can reproduce the original
speaker’s words, it neither follows that it must perforce do so nor that it
ought to do so nor, of course, that it actually does so…” (Sternberg
1982a: 68). In fact, it has been shown empirically that is often used in a
non-reproductive manner — as “constructed dialogue” (Tannen 1989; see
below). In spoken language, verbatim reporting is ordinarily precluded by
memory constraints, except when the anterior utterance is very brief,
formulaic, or specially memorized (a situation that is far from prototypical).
Memory limitations come into play almost immediately, particularly when
there is intervening talk; according to Chafe (1994: 215), “distal speech
events” are stored in long-term memory in “verbally uncommitted” (non-
verbatim) form.24
Memory constraints do not, of course, exclude the possibility that
reporters, in using , are operating under an illusion or pretense of verba-
timness. Accordingly, some scholars have offered definitions of the category
that treat verbatimness as a purport, a social rather than empirical issue. Thus
Coulmas (1986b: 2) states that “evokes the original speech situation and
conveys, or claims to convey, the exact words of the original speaker”;
however, he qualifies this by observing that different cultures can have their
own criteria for “what counts as the same” (ibid.: 10–13). Another influential
view locates verbatimness in norms of conversational cooperation — in a
“faithfulness principle” akin to Grice’s Maxim of Quality (Short 1988): by
selecting , the reporter has obligated himself, to the extent that it is
possible, “to report faithfully (a) what was stated and (b) the exact form of
words which were used to utter that statement” (Leech and Short
1981: 320).25 In this reporter-commitment model, the fidelity of depends
on felicity conditions — e.g., sincerity and the effort to be accurate. The
verbatim that would result from fulfillment of these conditions would be
52 R V
be expected to say. The issue of actual wording is irrelevant; the reports are
constructed to convey what the reporter thinks the represented speaker’s
position is or ought to be. (Cf. also Morawski 1970: 700 on fictitious quotes
attributed to real authors.) There is no contextual clue that such “factional”
reports are fictive, nor is any implicature of nonverbatimness possible. This
category of direct reports thus poses problems for the view (expressed, e.g.,
by Roncador 1988) that the most important aspect of verbatimness is not
“factual adequacy” but rather the extent to which the reporters assume
responsibility for the wording by signalling their interference.
In general, approaches to that appeal to verbatimness, whether as an
empirical fact, as a purport, or as a conversational maxim, do not provide an
adequate picture of the nature and extent of the category. They ignore its
selective character and privilege messages (sentences or words), when in fact
speakers can employ to demonstrate any aspect of language use, including
code and manner of delivery (Clark and Gerrig 1990: 775–78). Moreover, the
message or proposition can be treated as secondary to the other aspects being
demonstrated (“supportive, annotative, or incidential”; ibid.: 780). Hence
reports with signals can include departicularizers (vague referents) like
this and that, which can substitute for part or all of the message (ibid.).28
Examples of this kind are troubling not for only verbatimness-based treat-
ments of but also for the more moderate alternative view that “the use of
only implies a commitment to the content of the model speech act, viz. to
its intention (illocutionary force) and its propositional content”, not to its
form (Haberland 1986: 225). Clearly “what counts as the same” is not always
wording or structure; often it is the gist or the basic ideological position.
It might be argued that, no matter how common they are in everyday
language, the “anti-mimetic” types of cited above stray too far from any
conceivable prototype to count as cogent arguments against the fundamental
verbatimness of the category. However, countering this objection is the “truly
hair-raising sloppiness and disregard for reliable representation” (Fludernik
1993: 17) of actual anterior utterances that is typical in speech and nontech-
nical transcripts of spoken language, even when a high degree of fidelity is
possible. Here again verbatimness is not flouted but simply irrelevant; there
is no perception that the report is “faithless”, and precisianist challenges to
wording will be viewed as uncooperative unless they bear on the interpreta-
tion in some way. As Fludernik shows, the effect of identity or similarity in
is achieved not just by imitation but also (and more often) by typification
54 R V
of written texts. This adverb derives from the prepositional phrase iz ust”
‘from the mouth’, used of documents composed on the basis of oral instruc-
tions (SRJa, s.v.). Thus in a contrast was made between written and
unwritten sources, without regard to verbatim transmission.
While the concept of verbatimness, as noted above, certainly existed in
medieval Russian culture, there is no evidence for any kind of across-the-
board commitment to fidelity in form and content. The authoritative religious
texts of Eastern Orthodoxy provided models for wording changes and para-
phrases in , as in “verified” reports repeated in different forms within a
single book of Scripture (Savran 1988: 29–36) or non-identical reports in
parallel Gospel passages (Ong 1981: 22–23, 1982: 64–65). On the other hand,
medieval churchmen were familiar with the fourth-century bishop Spiridon
of Trimithus’ objection to synonymic substitutions in Gospel citations even
when they preserved the content (see Arximandrit Panteleimon 1904, 12:
348–49). Uspenskij (1984: 382) asserts that, in ecclesiastical language
(Slavonic) of the late Muscovite era, “any deviation from the correct
designation may be associated with a change in content.” However, this
attitude, which is best documented for the time of the Old Believer Schism
(mid- to late seventeenth century), seems to have applied mainly to the
copying of scriptural and liturgical texts, i.e., reflect a genre-specific doctrine
of verbatim transmission of written sources. Outside of canonical genres, one
can see great deviations from fidelity in cases in which it is verifiable. For
example, Avvakum, the first leader of the schismatic Old Believers, had no
qualms about presenting his opponents’ version of a sacred text (which he
regarded as heretical) in that was patently unfaithful (7):
(7) Napečatano: [duxu lukavomu molimsja
print- [spirit evil]- pray-.1.
It is printed: “Let us pray to the Evil Spirit” (ibid.: 378).
Also telling is a case in which Avvakum attributes the words of Judas to his
persecutor (Uspenskij 1987: 56–57) — that is, uses without regard to
either form or content as a way of revealing his understanding of what his
reportee was “really saying.” These are not cooperative usages, but they do
provide evidence that direct reporting could flout verbatimness in medieval
Russian culture.
In legal-administrative text-kinds, the concept of verbatimness is likewise
restricted to reports of written sources. Documents were generally quoted in
T 57
whatever sketchy notes they were able to make in the course of the hearings.
(There is no reason to think that the testimony was ever held up for the
scribe’s benefit.) There is evidence of a slightly later period for the existence
of such notes. In a judgment charter of 1531, the plaintiff refused to verify
the trial record during a higher-court hearing. Thereupon the judge of higher
instance, who happened to be the grand prince, took the routine step of
calling on the “men of court” (official witnesses) for confirmation; their
statements prompted him to decide in favor of the respondents. However,
there was an unusual sequel: six months later, the plaintiff produced a
document that he claimed to be the genuine rough (“black”) draft (spisok”
černoj prjamoj). Both of the trial judges denounced this as a forgery, and the
grand prince refused to withdraw his original verdict (N. P. Lixačev (ed.)
1895: 188–92, no. 10). Whether or not it was authentic, this rough draft
proves that trial scribes produced some kind of preliminary version before
drawing up the fair copy of the transcript.
In representing the trial dialogue, the scribes, like modern court recorders
(Walker 1986), parliamentary reporters, and newswriters (Slembrouck 1986,
1992) filtered out many features of orality — overlaps, interruptions,
hesitations, disfluencies, auto-corrections, vagueness, and, to a large extent,
the redundancy that is characteristic of spontaneous dialogue but has little
informational value in writing. (Cf. the hemming and hawing in the Ameri-
can trial transcript — mechanically recorded in the first instance — quoted
in Philips 1985.) The scribes also edited out most features that had primarily
interpersonal meaning (phatic and emotive devices). Moreover, while one
cannot know what was actually said, it is a suspicious circumstance that so
much of the testimony adheres to the maxims of quality, quantity, relevance,
and manner, especially given that many of the trial participants could not
have been familiar with the conventions of the speech event. The scribes
must have chosen what to report with a strict view to relevance, as defined
by the institutional function(s) of the text-kind. (For what it is worth, none
of the cases in which the transcripts were challenged mentions any objection
to wording discrepancies.) Presumably what the scribes intended to record
was a paraphrase accurate enough to satisfy the litigants that their positions
were represented fairly — an adequation to type (Fludernik 1993), as defined
by the purpose of the text-kind. Any stricter adherence to verbatimness, even
if it were possible, would have exceeded the demands of relevance and
hence would have constituted uncooperative behavior.
T 59
or sice. The resulting construction, which Otin views as nascent , purport-
edly allowed nonverbatim reporting by introducing an element of comparison
(ibid.: 55–56).
Otin’s explanation is untenable even apart from the theoretical problems
with the verbatim model discussed above and the general unverifiability of
any claim about verbatimness in premodern texts. To begin with, the tak(o)/
sice… jako constructions cited by Otin (ibid.: 56) are not correlative. In a
correlative construction with adverbs of manner, two predicates would be
compared to one another. If jako is indeed a comparator in Otin’s examples,
the reports that it complementizes must be correlative not to the main clause
but to omitted reports signalled by tak(o) or sice.
A second problem is that Otin is not sifting the textual evidence proper-
ly. The tak(o)/sice… jako… construction, which by my observations is quite
rare, appears only in later texts. By contrast, jako plus (jako recitativum),
which Otin treats as a secondary development, is attested in the earliest
Slavic writings, as indeed is full-fledged (see D. Collins 1996). Moreover,
if the tak(o)/sice… jako… and jako recitativum constructions really did
represent inchoate , one would expect to find cases of reports with indirect
deictic orientation; in fact, examples of that kind are scantily attested, if at
all, and, again, appear after the earliest cases of clear-cut .
A final problem with Otin’s hypothesis is that it is based on faulty
reasoning. If the scribes felt a need to indicate verbatim by means of
tak(o)/sice, they must have perceived it as a special variety; concomitantly,
they must have viewed the form of without tak(o)/sice — the unmarked,
default variety — as at least potentially non-verbatim. Thus there would have
been no need to single out the nonverbatim type as a special category by
means of jako. In any case, it is unclear why the relative adverb of manner
should signal non-verbatimness when its demonstrative counterpart was
signalling the opposite.
The fact that the use of tak(o) by itself was sufficient to introduce an
element of explicit comparison (typification) is supported by cases of
“omitted speech” (Thompson 1996: 518) or “narrator’s report of voice”
(Semino, Short, and Culpeper 1997: 24–25) — clausal ellipsis in which the
adverb stands in anaphorically for a report (8):
(8) I Mamon s tovarišči tako rkli… I
and Mamon- with fellows-. thus speak-. and
T 61
pišet…
write-.3
And Ivan Vasil’evič looked at the document, and in the document
is written… (ASÈI 1: 319, no. 430).
The verb of writing appears in the present tense because the tag is a narra-
tion of perception (a mode similar to ), with the judge as subject of
focalization. This can be seen explicitly in cases in which the tag clause is
subordinated to a verbum percipiendi by means of the complementizer ože
(ož’) or aže (až’), both typically used for objects of perception.36
It is noteworthy that the tag in (9a–b) does not include an attribution, for
all that it features a report index (the tag verb) and a redundant reference to
the document as textual conveyor (the adjunct prepositional phrase). The
verb pišet(”) in this formula does not have a recoverable subject and is
functionally impersonal — a widespread usage in writings (Sreznevskij,
s.v. “pisati”), though personal uses are also common.37 The use of subjectless
or impersonal verbs to index reports was well motivated, given the general
purpose of attributions in the text-kind (see 3.2). While identification of the
speakers in the trial was important not only for coherence (tracking) but also
on legal grounds, the attribution of the authors (or principals, Goffman
1981: 144–45) of cited documents was superfluous. It was generally reported
within or inferable from the citation, and in any case had no evidentiary
value; there was no need to (re)construct the position of the author or
principal of documents produced outside of the trial context. (While informa-
tion in the tag was also predictable, the tag itself was functional, since it
promoted coherence by indexing the citation. Thus there are only a few
examples in which it is omitted — in some cases, due to increased predict-
ability in the context.)38
There is a strong preference for the direct mode of reporting in document
citations of this kind. While there are a few cases of paraphrase (“speech
summary”),39 verbatim transmission of written sources (not counting errors
or minor corrections) was clearly the norm. This is confirmed in several
cases in which the document survives in the original or in later copies. Even
when there is no means of corroboration, the verbatim nature of the citations
may be assumed on the basis of their diplomatic structure, which allows
them to be compared with extant documents of the same text-kind. Only
rarely do citations depart from the expected formulaic structure sufficiently
T 63
to suggest that they are abridgments or paraphrases rather than close copies
of their originals.40
While tak(o) is standard in the clauses that introduce , it is scantily
attested in citations of written documents (only three tokens). Significantly,
its presence in these cases can be ascribed to typification marking. Two of
the examples, which are citations of donation charters, appear back to back;
the first of these is quoted in (10):
(10) I [kn(ja)z’ Mixailo Andrěevič(’) v”zrěl” v” [Olešinu
and [prince Mixajlo Andreevič]- look- in [Oleša-
danuju gramotu; ož v gramotě pišot tak:
donation writ]- in writ- write-.3 thus
[Tě svoi zemli Oleša dal [živonačalnoi
[those . lands]-. Oleša- give- [life.giving
Troici… [c(e)rk(o)v’ s(vja)tuju Troicju, da [Savrasovskuju
Trinity]- [church Holy Trinity]- and [Savrasovskaja
derevnju, da [Gridinu derevnju i s lěsy, i
village]- and [Gridina village]- and with forests-. and
s požnjami. A poslus(i) v [danoj gramotě:
with fields-. and witnesses-. in [donation charter]-
[kn(ja)z’ Semen Fedorovič(’) Starodubskoj, da [Timofěi
[prince Semen Fedorovič Starodubskij]- and [Timofej
Dmitreev s(y)n” Korjakina… A [gramotu danuju
Dmitrij- son]- Korjakin- and [writ donation]-
pisal [Oleša sam svoeju rukoju.
write- [Oleša self]- [. hand]-
And Prince Mixajlo Andreevič looked at Oleša’s donation charter; lo,
in the charter [it] is written thus: “Oleša gave those lands of his to the
Life-Giving Trinity [Monastery]… [namely] the Church of the Holy
Trinity and the village of Savrasovskaja and the village of Gridina,
with both the forests and the fields. And the witnesses in the donation
charter [were] Prince Semen Fedorovič Starodubskij and Timofej, son
of Dmitrij Korjakin… And Oleša wrote the donation charter himself,
with his own hand” (ASÈI 1: 352–53, no. 467).
64 R V
If the citations in (10) and its sequel are compared with extant donation
charters, they are revealed to be highly abridged paraphrases rendered in
indirect speech rather than verbatim copies. All of the 113 donation charters
from the archives of Trinity-Sergius Monastery published in the first volume
of ASÈI, which range from 1392 to 1505, are written in the first person and
have a different, more complex diplomatic structure than the citation in (10).
Significantly, the scribe of the given transcript — an original rather than a
copy — did not use tak(o) in tagging a subsequent citation, a letter from an
absentee witness (ibid.: 353–54). While this is clearly an excerpt, given the
absence of the incipit, closing formula, and other inferable information, it is
in and almost certainly a verbatim transmission of the written words,
rather than a paraphrase like (10). Therefore, in a series of reports that could
be reproduced, the scribe indicated the non-verbatim renditions by tak(o),
which he omitted before the verbatim rendition expected for citations. (The
citation from the absentee witness’s letter shows that tak(o) was not simply
mechanically transposed from the standard tag clause. The motivation for the
was probably that the letter served as a written substitute for testimony;
indeed, it is followed without any intervening narrative by the statements of
witnesses who are present at the trial, which are likewise presented in .)
The remaining instance of a documentary citation tagged with tak(o)
(ASÈI 3: 55, no. 32) is in ; though selective, it is more or less verbatim, as
can be seen by comparing it with its surviving original, a trial transcript of
ca. 1416–17 (ibid.: 53–54, no. 31). This need not be interpreted as a counter-
example to the use of tak(o) to signal typification, as illustrated in (10), since
the citation is somewhat reworked and lacks the incipit and end protocol
found in the original transcript. Judging by other examples, the general
practice was to cite earlier trial transcripts in their entirety, regardless of their
length.41
goals, which generally appeal not to anterior speakers but to the addressees
with whom they wish to collaborate in sense-making. Typically, their choices
are guided by several factors, including the purpose of the text-kind or
speech in which they are engaged, their assessment of the speech situation,
and their individual and genre-based understanding of cooperative behavior
and relevance.42 In particular, the reporters must gauge what and how much
information they need to give their addressees in order to aid their task of
interpretation and steer them toward the intended understanding.43 In doing
this, they are influenced by knowledge founded on their own experiences as
hearers or readers. This applies not only to reporting but to the selection of
any discourse strategy; as Green and Morgan (1981: 168) observe, “Produc-
tion will be heavily influenced by the speaker or writer’s assessment of how
an utterance is likely to be interpreted by a real or hypothetical audience.”44
Accordingly, reporters choose the direct mode of reporting because, by
virtue of its formal properties, it allows them to provide their intended
audience with a different quality and, potentially, quantity of information
(which in the present sense is not limited to content or proposition). As I
shall show below, this approach to does not contradict the many function-
oriented studies that have correctly emphasized its capacity to create vivid-
ness, immediacy, or involvement, nor, indeed, does it rule out the fidelity-
based (evidential) hypothesis. However, it does view the mode from a
different vantage point in that it focuses on the work of the interpreter. A
case can be made that this interpreter-based approach actually accounts for
the other effects that have been attributed to .
The property that distinguishes from other reporting strategies is not,
as shown above, its treatment of the wording or message but rather its
perspective. The reporter embeds some aspect of language use from a
projected (represented) speech event in the ongoing speech event; the
embedded segment has the same deictic orientation as the projected speech
event and, optionally, other features not meant to be assigned to the authorial
or narratorial voice of the ongoing speech event. (There can be ambiguous
cases in which the interpreter has to infer the perspective, e.g., passages
without any shifters or other distinguishing features, which are nevertheless
most conveniently treated as direct by default because they are marked by
paralinguistic signals typically associated with that mode or because they are
contiguous to clear-cut stretches of and contain no indications of the
authorial or narratorial perspective.) This perspectival property, which makes
T 67
refraining from “interference” in the report (see Leech and Short 1981: 324).
By contrast, nondirect modes of reporting tend to give the impression that
the reporter “intervenes as an interpreter between the person he is talking to
and the words of the person he is reporting” and hence that the represented
speaker is not being allowed to speak for himself (ibid.: 320; Caldas-Coult-
hard 1994: 304). This concept of narratorial interference or control has been
applied not only to reports of actual anterior utterances but also — presum-
ably in a figurative sense — to fictive reports: “The character’s individual
manner… appears unmediated in his own ” (Uspenskij 1973: 43; see also
Pascal 1977: 2; Short 1988: 67).
The general notion of a “cline of interference” in reporting modes (Leech
and Short 1981: 324; Slembrouck 1986: 48–49) is undeniably useful; howev-
er, it would be more realistic without the Conduit Metaphor and the implicit
assumption of reporter commitment. In reality there is no report without
some degree of authorial interference. Whatever the desired impression may
be, in reality the reporter is not a go-between but the author of the report,
regardless of whether, or to what extent, he has appropriated heteroglossic
material; except in special cases, the represented speaker is neither a partici-
pant nor a consultant in the act of reporting.51 Every report is subordinated
to the communicative purposes of the reporter (Sternberg 1982b). In fact, the
impression of interference does not relate, except in a metaphorical sense, to
how much the reporter lets the “original reportee” (if any) speak. Rather, it
relates to the extent and explicitness of the reporter’s own work in imposing
meaning — that is, to the degree to which he leaves, or seems to leave, the
task of interpretation to his addressee. This “attempt to avoid interpretation
and bias (or to seem to avoid them)” (Longacre 1983: 131) is not at all the
same as letting the represented speaker have full “say”, i.e., control of the
report’s meaning, even if it creates that illusion. (Represented speakers are
not the authors, in any immediate sense, of the statements attributed to them.
In cases of actual anterior utterances, they may, if reporters are cooperative,
be the principals — “someone whose position is established by the words
that are spoken, someone whose beliefs have been told, someone who is
committed to what the words say” (Goffman 1981: 144).)
Accordingly, the reporter can use as a form of apparent, if not
necessarily actual, self-suppression; by treating his addressee as a witness, he
can cede, or seem to cede, responsibility for imposing meaning on the report.
Sometimes this strategy is hearer-based by default. For example, in some
T 71
possible from analysis and to suppress their own evaluations of what they
recorded. By presenting reports as , the scribes relegated as much of the
interpretive work as possible to those who were the ratified interpreters,
legally empowered to dispense justice.
Concomitantly, the scribes adopted other hearer-based strategies. For
example, in order to avoid evaluating the actions of the participants, they set
the reports in a minimal, prepatterned framework, consisting chiefly of tag
clauses with semantically general ’s rather than evaluative speech-act
verbs. These tags performed the important ancillary function of delimiting
turns at speaking and hence of signalling the shift of viewpoints (interests in
the litigation), but they provided no information about the represented
speaker’s illocutionary intent or about the manner in which the depicted
utterance was produced. The use of “a large and finely-differentiated
lexicon” is a speaker-based strategy; explicit lexical items encode the
speaker’s evaluation and leave little to the imagination of the interpreter
(Lakoff 1984: 484). Thus the low type-to-token ratio of the authorial narrative
is hearer-based; it contrasts not only with other text-kinds such as ecclesiasti-
cal writings but also with other contextualizations within the transcripts.
Deference to the hearers may also be seen in the preference for the
syntactically inexplicit forms of parataxis and polysyndeton. Hypotaxis is
almost never found in the narrative framework, though it appears in the
transcripts in other contextualizations and occurs extensively in other kinds
of writing. By making syntactic relations explicit, hypotactic structures
function as evaluations that the authors use to maintain greater influence
over the interpretation (Lakoff 1984: 484, 486; Labov 1972: 387–92; Schiffrin
1981: 48–49). By contrast, paratactic devices are hearer-based: while convey-
ing the same notional relations between clauses, they leave to the interpreters
the tasks of inferring the nature of the connections and unravelling other
ambiguities (Lakoff 1984: 486–87).
Another facet of the choice of was the “theatrical”, re-enactive
character of discussed above. It is doubtful that this was motivated by
esthetic concerns, as in modern literature; even though the transcripts exhibit
purposive, “artful” use of language, aesthesis in and of itself cannot be
plausibly regarded as a major goal of the text-kind. If it were, the scribes
could be reproached with artless overuse of vividness, given the low “report-
ability” (Labov 1981: 227–28) of many of the represented messages.
T 73
the trial primarily by ear — that is, to witness the trial as an oral proceeding,
again in the manner of the on-site participants.
In closing this chapter, I would like to propose, if only as a thesis for
further discussion, that this compulsion to “witness” (recreate) the represent-
ed speech event is not a by-product of mimesis but rather the very heart of
the direct mode of reporting. While much of the literature on from Plato
onwards has focused on its purportedly mimetic nature, many instances of
, as shown above, cannot be considered imitative in any strict sense;
moreover, as Sternberg (1982a, 1982b) emphasizes, all is to some extent
mimetic. What is unique about is the deictic adjustments that it imposes
on its audience. Thus, while the concept of mimesis is oriented towards
reporters’ fidelity to their “originals”, this may well be less central than the
effect that they intend to have on their interpreters. The quintessence of ,
I would suggest, is not mimesis but methexis (“participation”), to adapt a
term from Platonic and early Christian philosophy — an act that, in accor-
dance with culturally determined interpretive conventions, allows the
audience to participate both in the new event (the re-creation) and, vicarious-
ly, in the prior event, real or imagined, that is being represented.52
C 4
There are three optional elements in the formula given in (1) — the
subject noun phrase conveying the represented speaker, a prepositional
phrase conveying the addressee, and the cataphoric adverb tak(o). (The
corpus does not provide information about the relative order of the last two
elements.) The subject of the gerund, which is postposed to the verb when
it occurs in the tagging formula, is usually the same as that of the main
clause. While the typical pattern in such cases is zero reference, the
subject of the gerund can be made explicit under various circumstances, e.g.,
when it denotes a subset of the main-clause subject or when it is stranded
from the main-clause subject.
The adverb tak(o) is likewise postposed to the verb, not preposed as in
the standard tag formula; it appears immediately before the report if there is
an explicit subject. Since clause-final position is ordinarily correlated with
newsworthy information in , the postposition of the optional elements
suggests that the constatation of the speech act, as conveyed by the verb, is
highlighted less here than in the standard tag. This inference is borne out by
the use of a nonfinite .
Cross-linguistically, participial clauses tend to be used as backgrounding
devices.3 While they retain certain verbal categories and can perform the
verb-like function of predication, participles convey fewer of the categories
needed for full descriptions of events than do finite verb forms (cf. Hopper
and Thompson 1984). From a pragmatic perspective, they compel the
interpreter to look in adjacent clauses for much of the information that is
canonically “verbal” (tense, mood, and person). This inferencing process
(which, generally speaking, is more extensive with participial elements than
with adjacent finite verbs) creates a subordination-like association between
clauses; hence it makes participles and gerunds effective devices for
presenting events as being either dependent on other events or less prominent
than those denoted by main-clause finite verbs. This is especially the case
with forms like the gerund that signal non-sequentiality, since sequenced
events tend to be in the foreground (Givón 1984–90, 2: 840; Hopper and
Thompson 1980: 281).
As noted in 3.2, the institutional purpose of trial transcripts favored
foregrounding not only of the testimony but also of the attribution, which
comprised an identification of the speaker and a constatation of the speech
act. The attribution was important because it facilitated interpretation of the
drama-like texts and because every statement made by litigants or their
78 R V
witnesses was binding and potentially probative for the outcome of the trial.
Thus prima facie the gerund tag would seem to be less than optimal for
testimony (at least of the prototypical variety), since the combined effect of
the gerund, the conjunction a, and the placement of the tag verb in nonfinal
position (as seen in the postposing of tak(o) and/or the subject) was to
background the constatation of the utterance — to present it as relatively less
event-like than reports tagged with the standard formula.
One explanation for the use of the gerund tag is that not all of the
reports so introduced represent prototypical testimony occurring in a dyadic
framework. In many cases, the evident purpose of the is to explain the
sequenced, foregrounded nonverbal actions of topicalized participants.
Commentary of this kind generally belongs to the background of discourse
(Hopper and Thompson 1980: 280). However, it is by no means clear that
every instance of the gerund tag has a backgrounding function, and some of
them present information that is not presupposed and potentially newsworthy.
In addition, the peculiar properties of gerunds made them suitable for other
pragmatic effects; this fact may pose difficulties for a blanket interpretation
of participial clauses as backgrounding devices.
It must also be kept in mind that the tag and the report are independently
subject to foregrounding or backgrounding. The only reporting method
attested in the corpus with the gerund tag is , which implies that the
reports themselves are foregrounded — an inference supported by the
optional occurrence of the cataphoric adverb in final position in the introduc-
tory clause. Thus the use of the gerund to introduce would seem to
establish a kind of middle ground in the discourse. The gerund clause signals
that the represented utterance is ancillary, the that the content is newswor-
thy. The combined effect may be to suggest a lesser degree of newsworth-
iness than in prototypical testimony; in any case, the attribution information
is treated as less newsworthy than with the standard tag.
The gerund tag formula has three main contextualizations in trial
transcripts — in descriptions of boundary-setting; with commentary on
unexpected events; and in a special incipit formula. In the first contextuali-
zation, boundary-setting descriptions, longtime residents serving as partisan
witnesses are depicted conducting the judge(s) around the metes and bounds
of the disputed property and indicating where they think the landmarks are
(see Kleimola 1972: 361–62 and 1975: 37–40). The witnesses’ actions are
expressed in a narrative clause (often very extended) consisting of the
R F T 79
preterit of the verb povesti ‘lead’ (with the judge as understood patient) plus
strings of prepositional phrases — e.g., “And [they] led up from the stream
Medvežka along a ravine, and, having gone a little ways along a marshy
stream, to the left past an ivy bush (it stands on the marshy stream) and past
a little birch…”, etc. (ASÈI 1: 553, no. 639). Usually the descriptions end
with a report in which the witnesses sum up their actions — e.g., “Having
stopped, [they] spoke in this way: ‘On the left, lord, is the grand prince’s
land, and on the right is the land of Trinity Sergius Monastery’” (ibid.).
There can also be reports in the midst of the descriptions.
Diverse reporting strategies are found in this contextualization. While the
most common strategy is introduced by the standard tag formula, gerund
tags are also attested, as are alternative tag verbs, , , and fused speech.
The reason for this latitude is clear: the speech reported from boundary-
setting procedures is not the prototypical testimony associated with the
standard tag. To begin with, it does not occur in a dyadic framework. The
witnesses are on a more active footing during boundary-settings and tempo-
rarily assume responsibility for the conduct of the trial. The remarks that
they make are offered of themselves, not elicited by the judge (if internal
evidence can be trusted; note that elicitation is generally made explicit in
other contextualizations). Reported speech is not the sole or even primary
conduit of evidence in such passages; the main emphasis is on the witnesses’
movements, and the testimony is, for the most part, ostensive rather than
descriptive. Thus the reports that occur are not representations of “canonical
talk” whose primary context is a speech event; they depict “coordinated task
activity”, in which “nonlinguistic utterances have the floor” (Goffman 1981:
140–43). The reports serve chiefly as commentary on or interpretation of the
adjacent narrative of events; thus they can be marked as less important than
the prototypical testimony in the dyadic framework of the trial dialogue. Their
attributions are also predictable, since boundary-setting canonically involved
only a single set of speakers — hence the occasional use of (see 4.3).
The gerund tag formula is one of the chief ways of indicating that the
given report is ancillary to the surrounding description.4 At times it is used for
a more specific effect, as in (2) — to distinguish running commentary [1]
from the speech acts that close out the descriptions, which are more event-
like in the plotline and are generally introduced by the standard tag [2]:
80 R V
along that same ravine to the boundaries, to the lime tree, that they
had demarcated with Ivan Bitjagovskij. And, having taken their stand
at that lime tree, [2] Mantyrej and Misail’s witnesses El’ka S’janov
[and] his fellows spoke in this way to the judge: “Where we, lord,
have led, … to the right [is] the land and wood of Andrej Okljačeev’s
village Isxodskoe, and to the left is the land and wood and plowed
field and stubble fields and hamlets of the Theotokos Simonov
Monastery’s village Korobovskoe (ASÈI 2: 435–36, no. 411).
The report with the gerund tag formula [1] is followed by two similar cases,
which are omitted here for the sake of brevity. All three of these reports are
commentaries on the preceding narrative of actions. Judging from (2) and
two later boundary-setting descriptions in the same transcript (ibid.: 437–38,
438–39), the scribe used the standard tag formula for major stopping-points
and the gerund formula for minor pauses and commentary. The same
distributional strategy occurs in in a lengthy boundary-setting description in
another trial transcript, in which the running commentary of the respondent’s
witnesses is tagged eight times with a r’kuči and once, at the very end, with
tak(o) rek(li) (AFZX 225–26, no. 258). An earlier boundary-setting descrip-
tion in the same transcript follows this pattern, though less consistently: the
first four remarks during the boundary-setting are tagged with tak(o) rek(li),
the next five (which occur after a long stretch of description) with a r’kuči,
and the concluding remark with tak(o) rek(li) (ibid.: 224–25).
The second main contextualization for the gerund tag with is with
commenting on narrative events that contradict institutional expectations. As
in boundary-setting descriptions, the reports function not as events per se but
as amplifications of events; hence they are good candidates for background-
ing, like other collateral material. In one of the examples, the reported
utterance is an attempt to explain why a witness is not competent to testify;
in four others, including (3), it conveys litigants’ excuses for not producing
witnesses or evidence that they had promised prior to an adjournment.5
(3) I sud’ja dal srok Ondrejku [dve
and judge- give- term- Ondrejko- [two
nedeli samomu stati i [gosudarja svoego knjazja
weeks]-. self- stand- and [liege . prince
Borisa postaviti… I na [tot srok… Ondrejko
Boris]- put- and on [that term]- Ondrejko-
82 R V
litigants’ speech, undoubtedly the central part of the testimony, takes place
in a dialogue with the judge that usually runs for the entire length of the
trial, the witnesses typically make one individual or collective statement and
then cease to be participants; they have their say and then retire “backstage”.
Moreover, the witnesses’ testimony has a different illocutionary intent: the
litigants speak to establish their own claims, the witnesses to corroborate
those claims. It is significant here that the witnesses are the only participants
who are asked to testify on oath — i.e., to indicate the reliability of their
evidence. This may be seen from the formulae in (5a–c), which routinely
accompany questions addresssed to witnesses (173 tokens in the corpus).10
(5) a. S”kaži(te) (brat(’)e) (v”) (bož’ju) prav’du
tell-./ (brother(s)-) (in) [(God-) truth]-
Tell [us], brother(s), (in) God’s truth…
b. S”kaži(te) (brat(’)e) kak” pravo (pered” bogom’)
tell-./ (brother(s)-) as truly (before God-)
Tell [us], brother(s), as is true before God…
c. S”kaži(te) (brat(’)e) po velikogo knjazja kr’st’nomu
tell-./ (brother(s)-) by [grand prince]- [of.cross
cělovan’ju
kiss]-
Tell [us], brother(s), in accordance with [your] fealty oath to the
grand prince…
One may therefore posit that the witnesses’ peculiar footing and the discrete
character of their testimony may have influenced the occasional use of tag
verbs other than the standard reči. Note that the the latter never appears
in formulae like (5a–c).
There are six instances of witnesses’ testimony presented as tagged by
the preterit of s(”)kazati (usually written sk-), the most widespread of the
alternative verbs of saying (see also Chapters 6–8). Four of the cases tag
what may be termed multiplex testimony, in which a single group attribution
covers a series of separate reports, each of which is a response to the same
question (see 4.3). Hence the witnesses’ statements do not appear in a
straightforward dyadic context (6):
88 R V
be compared to that of the gerund instead of the preterit in some of the cases
discussed in Section 4.1.
The tendency to present nonprototypical speech events by nonstandard
strategies is illustrated clearly by a case of the verb ot”kazati ‘answer’,
which, given its meaning, could conceivably have been used for any state-
ment elicited by a question. In fact, there is only one example (7), which
conveys a speech event that is unusual for the text-kind:
(7) I [knjaz(’) Ivan Jur’evič(’) poslal k” mitropolitu
and [prince Ivan Jur’evič]- send- to metropolitan-
[s(y)na svoeg(o) kn(ja)zja Ivana, velěl ego sprositi,
[son . prince Ivan]- bid- him- ask-
on li posadil [těx mužikov, Okulika da
he- settle- [those peasants]-. Okulik- and
Olferka, na tom na [Šiškinskom selcě, i skol’
Olferko- on that- on [Šiškinskoe village]- and how.much
davno oni tut živut. I [mitropolit
long.ago they- there live-.3 and [metropolitan
Zosima otkazal [kn(ja)zju Ivanu Jur’evič(u) tak: [To
Zosima]- answer- [prince Ivan Jur’evič]- thus [that
seliščo Šiškinskoe — zemlja [Simanovskog(o) manastyrja…
village Šiškinskoe]- land- [of.Simonov monastery]-
And Prince Ivan Jur’evič sent his son Prince Ivan to the metropolitan
[and] bid [him] to ask him whether he had settled those peasants,
Okulik and Olferko, in that village Šiškinskoe, and how long they had
been living there. And Metropolitan Zosima answered Prince Ivan
Jur’evič as follows: “That village Šiškinskoe is land of Simonov Mon-
astery…”, etc. (ASÈI 2: 411, no. 402).
Here the judge, verifying an assertion made during the hearing, is communi-
cating with an witness through an intermediary, outside the context of the
trial hearing. As the mediated question is itself unusual for the text-kind, it
is not surprising that it is presented as , i.e., differently than prototypical
questions (see Chapter 5). The metropolitan’s relayed answer is presented as
evidence and given as , as if part of the hearing. The tag clause contains
no reference to mediation, and it is followed without seam by the continua-
tion of the trial narrative. Nevertheless, the heterogeneous nature of the
90 R V
5.3); pronominal anaphora in the text immediately following (8) (see (4) in
3.1); and the preterit of reči as a tag for the judge’s speech in the same
passage (see 5.4). Cross-linguistically, the historical present tends to be used
to increase vividness by substituting the perspective of the ongoing speech
situation for that of the narrative. In other words, it is an icon of foregroun-
ding, like other elements replicating “the ego-hic-nunc context of the speech
event” (Chvany 1990/1996: 288). Recent studies treat the historical present
as an involvement strategy that allows listeners to experience the narrated
event themselves and draw their own conclusions about its significance.12
As shown by trial transcripts and other text-kinds, the historical present
in legal language tends to co-occur with — in particular, reports
ascribed to adversaries. It is most common in emotionally charged contexts
such as petitions, where the writers or principals express personal viewpoints
and appeal to the sympathy of their addressees. This explains why it often
occurs within the in trial transcripts while being rare in the narrative,
which is written from the relatively impersonal perspective of an effaced
scribe, often unnamed in the text, who presumably had no stake in the
outcome of the trial. This general distribution supports the idea that the
historical present had an evaluative, involvement-creating function in , as
in other languages.
It is significant that one of the few cases in which the historical present
occurs in the authorial narrative involves the verb m”lviti, which was typical
of emotive contexts in northeastern documents. According to Mixajlovskaja
(1980: 47–50, 181–83), this verb was colloquial and had negative connota-
tions in . However, this nuance may not have been present in every
dialect; for example, in western and northwestern documents, m”lviti appears
to have been neutral, as it is in West Slavic (where it denotes unfocused
linguistic action). Also problematic is the notion that the verb was colloquial,
given that it can appear in trial transcripts and other kinds of chancery
writing, which are not colloquial in character any more than they are literary.
There is no evidence that its occurrence in (8) and elsewhere in the tran-
scripts is due to stylistic mistakes.
In fact, most of the examples of m”lviti found within the dialogue seem
to be used purposively, to convey a sense of futility or even exasperation
(e.g., the admission of defeat in Kaštanov 1970: 372, no. 16) — a nuance
furthered in some cases by the use of the historical present.13 The same
connotation of ineffective speech is found in (8), since the uncooperative
92 R V
As noted above, the standard reporting method in the trial narrative fore-
grounds both the report and its attribution (the constatation of the speech
act); the latter is prominent because any statement made by one of the
interested parties in the lawsuit is potentially probative. While tagged
with a preposed narrative tag is accordingly the most common method by far
for reporting testimony, there are also a number of cases of free ()
(“zero quotatives”, Mathis and Yule 1994; “null quotation formulae”,
Longacre 1994). This mode of reporting is typologically widespread. To a
large extent, the literature has emphasized its capacity to create “drama” —
vividness or involvement above and beyond the effects of tagged ,
particularly at discourse peaks. The use of this strategy in trial records shows
that there can be other effects than involvement; indeed, some of the cases
could only be infelicitious if involvement were a factor, given the low degree
of reportability.
It is generally agreed that felicitous use of depends typically on the
identity of the reportee being recoverable or inferable from the context, e.g.,
through “regular alternation of speakers” and/or “content appropriate to a
given speaker” (Longacre 1994: 130). (It can also be felicitous in the less
common cases in which the attribution is immaterial, as when “talking
heads” are depicted collaborating in the development of an argument or story
— for example, Eco’s three fantasizing editors in Foucault’s Pendulum.) In
3.1, I discussed two cases that illustrate this principle in that the was
used to present the opening statements of the trial hearings, where the
attribution was easily recoverable. Interpreters familiar with the text-kind
would be able to infer that the plaintiff, whose identity was established in
the preamble, would be the first represented speaker — a deduction rein-
R F T 93
forced by the itself, which included trial-initiating formulae and forms
dependent on the plaintiff’s viewpoint.
If predictability of attribution alone was sufficient to motivate , the
strategy should have been possible with most of the testimony presented in
a dyadic framework, since the represented speaker’s identity can generally be
inferred from the tag to the preceding question. The fact that is actually
rare in this context suggests that such predictability was not the primary
factor. Scribal carelessness — i.e., nonpurposive usage — is not a likely or
viable explanation, because the three cases attested in dyadic all appear in
representations of the same kind of speech act — a litigant naming and
presenting his witnesses — and involve the same formula (9).14
(9) I sud(’)ja vsprosil Stepanka i v” vsěx”
and judge- ask- Stepanko- in [all
kr(e)st’jan město: Komu ž to u vas
peasants]-. place- who- that- at you-.
vědomo, [dobrym ljudem, starožilcem? Vědomo
be.known- [good people old.residents]- be.known-
to, g(ospodi)ne, u nas” [Ostašu Paninu da Fetku…
that- lord- at us- [Ostaš Panin]- and Fedko-
And the judge asked Stepanko on behalf of all the peasants, “To what
good people, longtime residents, among you is this known?” “Among
us it is known to Ostaš Panin, and to Fedko…” [etc.]
(ASÈI 1: 478, no. 589).
Though found in a dyadic context, the underlined report is not prototypical
testimony whose purpose is to provide evidence about the disputed property;
rather, it is a preliminary to the gathering of such testimony. There are
several additional cases in the corpus in which the act of naming witnesses
is conveyed by an uncanonical method (see Section 4.6).15 This follows the
principle that deviations from the conventional reporting methods tend to
occur when the represented speech event is atypical in some respect — a
factor also present with the used to present the opening charge. In (9),
the atypical strategy may have been promoted not only by the predictability
and lesser importance of the attribution but also by the high degree of
cohesion within the question-and-answer dyad (cf. the repetition of to and
vědomo and the echo of u vas by u nas), which may have worked against the
94 R V
use of a tag clause (a delimiting device). Any doubt about the attribution
would have been dispelled by the use of the vocative g(o)s(podi)ne, a
testimony marker (see Section 3.2).
Another contextualization in which is found is in boundary-setting
descriptions. Three such cases, all from a single document, were discussed
briefly in 4.1 (ASÈI 2: 391–82, no. 388); it was noted that a seventeenth-
century copy of the same trial record, thought to reflect a fifteenth-century
version made shortly after the lawsuit, features the same reports presented as
with a gerund tag (ibid.: 394–95, no. 388a). While it is impossible to
determine when the gerunds were inserted, the fact of the substitution
provides independent evidence for the general preference for tagged ,
which I identified on the basis of statistical predominance (see Section 3.1).
Boundary-setting descriptions are, indeed, a favorable contextualization
for , as for other uncanonical reporting methods; ten cases are attested,
counting the three mentioned above. In each instance, the presence of is
signalled unambiguously by forms dependent on the reported speech event,
as in (10a).
(10) a. I Uvar”, i Gavša, i Ignat tak rkli:
and Uvar- and Gavša- and Ignat- thus speak-.
Znaem, g(o)s(podi)ne; poidite [sic], g(o)s(podi)ne,
know-.1. lord- go-.2. lord-
za nami; a my, g(o)s(podi)ne, tebja po
after us- and we- lord- you-. along
meže vedem. Iz polěs’ja poveli…
boundary- lead-.1. out.of wood- lead-.
na bereg po vetlu po vilovatuju,
onto bank- as.far.as willow- as.far.as winding-
po [samye rozsoxy. Po [ta města,
as.far.as [very forks]-. as.far.as [those places]-.
g(o)s(podi)ne, znaem: to, g(o)s(podi)ne,
lord- know-.1. lord-
meži [mitropoliče požne s [Sysoevoju
boundaries-. [metropolitan- field]- with [Sysoj-
požneju.
field]-
R F T 95
And Uvar and Gavša and Ignat spoke in this way: “We know,
lord; come, lord, after us, and we, lord, will lead you along the
boundary”. From the wood [they] led… [etc.] to the bank as far as
a winding willow, as far as the very forks [of the willow]. “This
far, lord, we know. There, lord, are the boundaries of the metro-
politan’s field with Sysoj’s field” (ASÈI 3: 462, no. 477; also in
AFZX 253, no. 308).
Direct deictic elements and the vocative g(o)s(podi)ne ‘lord’ serve as boundary
signals, reinforcing the interpreters’ expectations of what should come next,
based on their background knowledge of the text-kind and trial procedure.
As noted in 4.1, the statements made during boundary-setting depart
from the dialogic framework characteristic of prototypical testimony in the
trial hearing. The witnesses are depicted speaking without prompting, and the
primary form of testimony is gestural rather than verbal; much of the
emphasis is on what the witnesses show the judges rather than on what they
say. Reported speech tends to be concentrated at the end of the descriptions,
as a summarizing evaluation of the ostensive testimony; often, though by no
means always, it appears in a set form like (10b) (where A and B are
individual or institutional owners, and X and Y the names of properties):
(10) b. napravě — zemlja (NP-) NP-/, a nalevě —
to.right land- (X-) A-g/, and to.left
zemlja (NP-) NP-/
land- (Y-) B-/
To the right is A’s land (X), and to the left is B’s land (Y).
The occurrence of a report in some form was conventional in boundary-
setting descriptions; the fact that the speaker’s identity was easy to recover
permitted, though it did not motivate, the felicitous use of tagless strategies.
One factor that apparently promoted was the possibility of viewing the
witnesses’ statements and the nonverbal actions that they performed during
boundary-settings as parts of a single complex event — a single instance of
multimodal testimony. As the and demonstrations were perceived as more
closely related than two adjacent turns in ordinary testimony, they could be
treated as a cohesive unit by omission of the intrusive tag that would
otherwise delimit them — an icon of distance (cf. Haiman 1983: 67; Mathis
and Yule 1994: 67).
96 R V
[1] And Fegnast spoke in this way: “For that wood, lord, we have a
donation charter of Prince Mixail Andreevič… [etc.] and, lord, a copy
of that charter is before you.” [2] And the judge looked at the copy;
and in the copy was written: [text]. [3] “And that wood, lord…” [etc.]
(ASÈI 2: 329, no. 337).17
Experienced interpreters — especially those familiar with the cited text-kind
— would have had no trouble determining where the documents left off and
the testimony resumed, given the clues provided by vocatives and other
features dependent on the represented speech event. The attribution of the
was predictable because it served as commentary on the documents; it
did not need to be demarcated in a way that would have falsely implied the
onset of an entirely new turn at speaking (not to be equated with a new
utterance). The absence of a tag has the effect of clustering the report with
the document that it elucidates. A comparable case of creating multi-
modal testimony occurs with a report placed before a document citation to
explain why a copy is being offered instead of the original (ASÈI 2: 416, no.
404). As background to the documentary evidence, the report can be treated
as part of the same testimony.
Multimodal testimony is analogous to another common contextualization
for — multiplex testimony, in which a single question is followed by a
series of separate reports with a collective attribution (see 4.2). Multiplex
testimony permits a number of tagging techniques, all of which differ, to
various degrees, from the standard way of representing testimony in the
investigated transcripts, in accordance with the principle noted in 4.1 that the
further a contextualization departs from prototypical testimony, the more
hospitable it is to noncanonical reporting methods. In multiplex testimony,
the parallel responses to a single question are presented as equipollent parts
of a unified speech event with a single purpose — to support the claim of
one of the litigants. In other words, they are treated as a single instantiation
of testimony delivered over more than one utterance, just as reports and
gestures in boundary-setting descriptions or with document citations are
treated as a single instantiation of testimony delivered in more than one
medium. This recalls the use of when reporters wish to indicate the
“convergent behavior” or “similarity and shared knowledge” of two reportees
(“two speakers, one voice”; Mathis and Yule 1994: 64, 72).
The distinct nature of multiplex answers is manifest even when there is
98 R V
you stand?” And Maksim spoke in this way: “I, lord, remember for 60
years…” [etc.] And Manak spoke in this way: “And I, lord, remember
for 50 years…” [etc.] And Vas’ko Bužonina spoke in this way: “And
I, lord, remember for 40 years…” [etc.] And Jurka Melexov spoke in
this way: “And I, lord, remember for 45 years…” [etc.] And the judge
asked Olekseik… [etc.] (ASÈI 2: 519, no. 481).
The same pattern is found in a combination of multiplex and multimodal
testimony, when an absentee’s letter is treated in parallel with the of co-
witnesses (ASÈI 1: 353–54, no. 467), and in multiplex written evidence,
when two documents produced by the same party are cited back to back
(e.g., ASÈI 1: 538–39, no. 628). Rather than being linear like the reports
whose tags begin with i, the parallel responses in multiplex testimony fan out
from the plotline; their order can make no difference to the meaning of the
document. Indeed, it possible that the statements in (12) have been reordered,
since the witnesses are quoted in a different sequence than they are named
in the tag; the editors found the displacement of da Maksima odd enough to
warrant distinct punctuation. Ordinarily the of longtime residents goes
from oldest to youngest (Kleimola 1975: 35).
The use of a instead of i in multiplex testimony signals the scribes’
perception that only one speech event is taking place — a complex response
to a single question. Significantly, the standard pattern of coordination, with
i instead of a, is employed when the parallel answers to a single question
cannot be viewed as a unified event — e.g., when the speakers are opposed
parties (ASÈI 2: 523, no. 483), when one of the witnesses changes the topic
(ibid. 1: 573, no. 651), or when one of the witnesses’ testimony clashes with
that of the others (AFZX 98, no. 103; ASÈI 1: 466, no. 584; 468, no. 485).20
The perceived unity of multiplex testimony motivates the use of methods
like that minimize the interruption of going from one witness’s response
to another. In (13), the multiplex reply begins as tagged choral [1], then
fans out into separate untagged reports [2–4] before merging back, again
without a tag, into choral speech [5].21
(13) I [knjaz’ Vasilej sprosil Rodivonika s
and [prince Vasilij]- ask- Rodionik- with
tovarišči: Skažite v [bož’ju pravdu, č’ja to
fellows-. tell-. in [god- truth]- whose-
100 R V
Here has the effect of coalescing the individual acts of testimony into a
single continuous whole — one statement, as it were — whose parts are less
demarcated and more cohesive than are ordinary statements tagged by
intrusive narrative clauses. Accordingly, can be used for the testimony
of agreeing witnesses in contexts where the statements of uncooperative
witnesses appear as tagged (AJu 16, no. 8; ASÈI 2: 422, no. 406).
It is typical for the first response in multiplex testimony, which can be
choral, as in (13), or individual, to be tagged, and for some or all of the rest
to be .22 (The untagged reports are thus alternatives to tagged reports with
the non-sequencing conjunction a; indeed, they themselves often begin with
a.) Even when the opening tag refers to more than one speaker, the first
report can represent a single individual response. There is never any ambigu-
ity in the attribution of the untagged reports; first-person singular forms and
proper names signal shifts into individual or the transition from one
speaker to another, while first-person plural forms pinpoint the onset of
choral speech.
As (13) illustrates, tends be used in the context of choral speech —
reports presented as if uttered by several speakers simultaneously (cf. Tannen
1989: 113). In (13), the individual statements of how long the witnesses
remember are separable from, though contiguous with, the choral testimony
at the beginning and end, which identifies the owner of the property.23 In
other examples, the transition from individual to choral can occur within the
bounds of a single sentence; the joint information appears in a complement
clause after the final speaker’s individual statement.24 Such constructions, in
which one witness caps off the testimony of an entire group, plainly reveal
the constructed nature of choral speech. They also serve as a strong proof of
why must be approached as a discourse phenomenon rather than a
syntactic construction. To view the joint complement of the parallel instances
of ‘remember’ as mere clausal deletion would overlook the basic function of
the construction — to present the statements as equipollent parts of a whole,
as more similar to one another than to other statements in the hearing.
The chief function of choral speech is to reduce the diffuseness and
redundancy of multiple utterances with the same propositional content. It
may sometimes also be a way of framing statements by group spokesmen;
this can be inferred from cases in which the tag is collective but the itself
reflects a single perspective (e.g., ASÈI 1: 538, no. 628). Both of these
functions link choral speech with the pragmatic effects of discussed
102 R V
attested. Thus one must account separately for the complementizer and the
deictic orientation.
In indeterminate cases, it is best to speak neither of direct nor of indirect
but rather of nondirect speech, though I would hesitate to identify this as a
separate reporting strategy. Nondirect speech is not a true halfway point
between the two poles of the formal continuum of reporting strategies.
Because it lacks the deictic features that make salient in discourse, it
more closely resembles indirect speech in its pragmatic effects. From a
functional standpoint, would seem to be polarized into direct and less-
than-direct types.
Another point that needs to be stressed from the first is that was not
used as a means of avoiding verbatim reproduction — a traditional interpre-
tation, corollary to the reproductionist approach to (see 3.3). The state-
ments presented as were by and large as accessible to the scribes’ observa-
tion as those presented as . Some of them are actually adjacent to direct
reports; they can even represent the answers to questions in , as in (14),
below. Thus it seems likely that the contextually unexpected use of where
is ordinarily preferred was a way of achieving certain pragmatic effects.
In fact, we see in the cases of the same correlation between nonproto-
typical speech events and unconventional (or less conventional) strategies as
in other residual forms. For example, one of the tokens represents a state-
ment made in the interlude between the on-site trial and a higher-court
hearing (14) — a “low-keyed” transition between major episodes (cf. Diver
1969: 49, 57). During the trial, the plaintiff had used a copy of a charter to
prove that his monastery owned the disputed property; he stated that the
original was in the keeping of a chancery secretary, Ivan Kobjak Naumov
(ASÈI 1: 539, no. 628). Copies were not considered firm evidence in
trials; thus the respondents challenged the authenticity of the charter (ibid. 1:
540, no. 628). This prompted the judge, Prince Vasilij Ivanovič Golenin, to
visit Kobjak before submitting the trial record to the grand prince.
(14) I [knjaz’ Vasilej, priexav k Moskve, vprosil [dijaka
and [prince Vasilij]- come- to Moscow- ask- [clerk
Kobjaka: Prislal esi ko mne spisok z
Kobjak]- send- ..2 to me- copy- from
[danye gramoty [Vasil’ja Borisoviča, a gramotu mi
[donation writ]- [Vasilij Borisovič]- and charter- me-
R F T 105
reports in a less salient manner than is normal in the chief episode, the trial
narrative. The statements that are reported in interludes are never testimony
proper and often have a predetermined character. In some cases, they relate
to adjournment procedures and have no bearing on the evidence; in others,
they corroborate testimony made during the trial hearing, so that their
content is largely predictable. Corroboration is the key factor in another case
of complementized in this contextualization (15), which cannot be
identified as either direct or indirect due to the absence of deixis referring to
the represented speech situation.
(15) A k Elizaru sud’ja poslal [Ivana Turab’eva da
and to Elizar- judge- send- [Ivan Turab’ev]- and
Timošku sprašivat’ o Ikonniče zemle. I Elizar
Timoška- ask- about [Ikonniča land]- and Elizar-
skazal, čto [ta zemlja Ikonniče… po [Spasskuju dorogu
say- that [that land Ikonniča]- up.to [Spasskaja Road]-
protivu Gorbova [zemlja mitropoličja Kulikovskaja, a
opposite Gorbovo- [land of.metropolitan Kulikovskaja]- and
kupil [Ivan Fedorovič u Mixaila do moru,
buy- [Ivan Fedorovič]- at Mixajlo- before plague-
a v knigax pisana v dannyx [Danila
and in books-. write- in of.donations-. [Danilo
Ivanovičja bojarskaja zemlja, a ne [velikogo knjazja.
Ivanovič]- [boyar land]- and [grand prince]-
And the judge sent Ivan Turab’ev and Timoška to Elizar to ask about
the Ikonniča property. And Elizar said that that property Ikonniča… as
far as the Spasskaja Road opposite Gorbovo belongs to the metropol-
itan’s property of Kulikovskoe, and Ivan Fedorovič bought [it] from
Mixajlo before the Plague, and in the donation books of Danilo
Ivanovič it is recorded as boyar property and not [property] of the
grand prince (AFZX 228–29, no. 259; trial transcript quoted in another
transcript).
The represented speaker Elizar in (15) is a witness for the plaintiff who was
unable to testify in situ due to illness (ibid.: 228). As in (14), the message of
the (including the boundary-setting description, whose beginning is
omitted here) is presupposed from the prior text; it repeats and corroborates
R F T 107
And Zaxarij spoke in this way: [1] that they had [indeed] set the
boundaries, in accordance with the landmarks, [2] but, he says, he
does not recall whether the survey went up to this place or not
without [seeing] the document; and, he says, he does not recall
which peasants were with them on that survey. [3] And, said
Zaxarij, the boundary writ by which they surveyed and set the
boundaries is in the grand prince’s treasury. [4] And if Mixajlo
Šapkin and Golova, when departing, awarded those lands to the
monasteries and gave [them] their own writs, I, lord, was not with
them then; at that time the grand prince had sent me to Murom to
do scribal work (ASÈI 2: 270, no. 310).
While this report features the standard tag, it begins as clear-cut , indicated
by third-person verbs and pronouns coreferential with the subject of the tag
clause. The first sentence of the report [1], which conveys Zaxarij’s corrobo-
ration of previous testimony (cited in the previous question, which is in ),
features the highly presuppositional devices of nominalization and anaphora
(ix kladen’ja, literally “of their placing”). Nominalizations are low in
discourse saliency as compared with finite predicates (see Chvany 1990/
1996: 293). The second and third sentences [2] convey new information —
Zaxarij’s admission that he is unable to testify on other issues. Here the tag
is reinforced, as in (14), by intercalated present-tense ’s, which reiterate
the fact of reportedness without demarcating the clauses as separate utterances.
The non-initial tag skazal Zaxar’ja in the fourth sentence [3] may have the
same continuity-promoting function; however, it is more likely that the change
in tagging strategies, from present to preterit, marks, as it were, a new
paragraph — a transition from non-probative, negative remarks to positive
statements that could be of use for the judgment. The switch to in the
final sentences [4] marks a further change in the character of the speech
event and is iconic of increasing relevance; the report now conveys a state-
ment of personal experience bearing on one of the respondent’s claims. (The
fourth sentence in (16a) can bear a different interpretation; see 4.5.)
The second indirect report in the same transcript occurs in an ostensibly
non-dyadic framework (16b). Here the judge Mixajlo Gnevaš Stoginin is
represented testifying in response to the preceding statement; there is no
report of any intervening question. When the respondent turns the tables by
stating that Gnevaš had previously been appointed to judge a suit over the
property, Gnevaš has to step out of his judicial role to explain himself.
110 R V
The priests’ statements [1a–b] are coalesced because they are identical in
content. Their testimony may have been cited separately because it was made
under a peculiar kind of oath: laymen swore “before God” or “by a cross-
kissing oath” (see 4.2), clerics “by the priesthood” (po svjaščen’stvu ‘by
priesthood-’).
The underlined passages in (17a) are in unambiguous , with the
subjects of the tag clauses coreferential with the implicit subjects of third-
person verbs within the reports. The omission of the subject pronouns
indicates relatively cohesive relations between the report and the tag clause;
it may imply subordination, as does the complementizer (see Haiman and
Thompson 1984: 511–12). By contrast, multiplex testimony conveyed as
usually includes first-person subject pronouns.
Several motivations can be posited for the use of in (17a). First, it can
be seen as another means of minimizing redundancy in multiplex testimony,
where the reports tend to overlap in content. Indirect speech is a relatively
economical reporting mode (cf. Čumakov 1975: 19, 26) because it eliminates
certain elements referring to the represented speech situation that are
commonly found in testimony. The use of a complementizer in (17a)
obviates the disambiguating function ordinarily performed by these elements.
A second possible motivation for the use of complementized in (17a)
is the corroboratory and hence partially presupposed character of the witness-
es’ — a feature shared by the indirect reports in (14)–(16). The statements
of witnesses tend to have predictable content because of their circumscribed,
normatively partisan role in the trial; indeed, the litigants are often represent-
ed anticipating what their witnesses will say, as in the given transcript.
(Usually the witnesses are not mentioned at all until the judge asks the
litigants for corroboration; see Kleimola 1972: 360.) In (17a), the fact that the
witnesses support the respondent is contextually new information, but the
actual message conveyed by their statements is presupposed (or partially so,
as when they assert their longevity).
The backgrounding effect of the in (17a) becomes obvious when it is
compared with a passage from later in the transcript (17b) — the statement
of the plaintiff’s witness, which is tagged in the conventional way and
presented in a salient manner, as :
(17) b. Ivan velel Stepanka postaviti, i Stepanko
Ivan- bid- Stepanko- put- and Stepanko-
R F T 113
factor favors the choice of explicit subordination in the form of the comple-
mentizer č’to, which gives the attribution in the narrative framework greater
salience than the report — a striking effect in a “dialogue discourse”
(Longacre 1983: 44), where the most important information tends to be
within . Presupposition of the message may also favor , which avoids the
discourse-slowing deixis shifts of . Slowing the pace of interpretation, as
in , makes the report contextually salient — an effect that can be likened
to the modern Russian convention of putting spaces between each letter of
a word for emphasis. By contrast, does not place the report in special
prominence; it offers a quick account, much like a summary, of a message
that has, in effect, already been processed.
In Sections 4.1–4.3, it was noted that, as predicted by the method of
residual forms, unconventional strategies tend to index statements that depart
from the prototype for the represented speech event. This tendency also
applies to the cases of discussed above; it may be a factor conspiring with
presupposition to promote complementized . The differences between
ordinary testimony and the multiplex statements exemplified in (17a) were
discussed in 4.2–4.3. In (16a, b), the indirect reports reflect a drastic change
in footing; it was not a normal part of the judges’ role to bear witness to a
litigant’s claims or to account for their own conduct. The use of complemen-
tized in (14)–(15) may also be influenced by the unusual nature of the
represented speech events as compared with testimony during the trial
hearing. The given speech events have a different participation framework
than the trial hearing, and not only because of the change in venues (cf.
Goffman 1981: 128). The interested participants — litigants and witnesses
— are absent or inactive and thus unable to respond to the new evidence. In
(14), the secretary Kob’jak does not fall into one of the usual categories of
witnesses — longtime residents (znaxari, starožil’ci), or reputable spokesmen
of local opinion (dobrye muži).
If deviation from the prototypical speech event does play a role in the
choice of complementized , it raises two potential problems. First, why
should the statements of witnesses who fulfill social expectations in (17a) be
presented as when that of the uncooperative witness in (17b), whose
footing in the trial changes drastically, is presented in the conventional
manner? Second, why should , a strategy typically used in backgrounding,
be employed to index deviant speech events, which one one think would be
a foregrounding function? Recall Silverstein’s remark (1985: 142) that “there
R F T 115
In the cases of and examined above, the report clauses retain the
syntactic features of independent clauses, regardless of whether they are
explicitly complementized, since the grammatical relations within the reports
are not controlled by elements in the tag clauses. Indeed, the tags themselves
are omissible, as seen in . The deictic orientation in , while congruent
with that of the tag clause (narrative), is not grammatically dependent on it,
while the optional use of a complementizer is “a coding acknowledgment
that the two clauses are semantically still independent of each other, at least
to some extent” (Givón 1980: 371; see also 1989: 110; Thompson 1986: 520).
By contrast, in the strategy that I will term fused , the report clause
lacks some or all of the properties of independent clauses (see Givón 1980,
1989: 108–12; Haiman and Thompson 1984: 511; Li 1986: 36–37). In ,
fused may be divided into fused topic statements, or syntheses of content
(cf. Potebnja 1899/1958–65, 1–2: 296), and fused reports of messages. In the
latter type, which is the only one attested in the narrative framework, the
notional subject of the is treated as the direct object of the and
R F T 117
(20), and the others setting information ((18)–(19), (21)). Two ((18)–(19))
appear in “low-keyed” transitional episodes (cf. Diver 1969) and reflect
unsequenced, non-testimonial phases of the trial. The messages themselves
carry a relatively low information load; the boundary-setting reports are
particularly terse as compared with similar passages in other transcripts. In
(21), the message is prepatterned and thus partially presupposed.
While fused preserves the central propositional content, it renders
statements less diffusely than either and by omitting inferable informa-
tion about verbal categories. This terseness makes it suitable for presenting
brief reports, especially ones that carry a low information load and need not
be presented in a prominent, elaborated manner, with everything made
explicit for the interpreter. Fused is not suitable for presenting structurally
more elaborated statements with multiple clauses, as this would require
sustained inference due to the reduced cohesion of stranded constituents.
Fused can thus have a streamlining effect, which makes it suitable for
conveying presupposed or otherwise backgrounded reports. The small-clause
structure has the textual effect of eliminating the demarcation between report
and authorial context (in the present instance, narrative), though fused
does not go as far in this respect as (see 4.7), since it still marks repor-
tedness in an explicit manner. The lack of “gear-shifting” (Page 1988: 33)
between and narrative increases the pace of interpretation. This acceleran-
do or glossing-over effect tends, in general, to counteract discourse salience.
At this point a seeming contradiction emerges: if fused reports are
incorporated as constituents of the main clause, they must reflect the salience
characteristic of main clauses. This, in fact, is true not of the report as such
but of the separate constituents of the report, as is illustrated clearly in (18)
by the topicalization of the notional subject and the assertion of the notional
predicate. While the separate parts of the are thus amenable to salience
effects, the report as a whole presents neither a figure nor a ground that can
be treated separately from the narrative tag clause. Indeed, the capacity to
single out particular items within the reported information is clearly a special
property of the strategy.
It is noteworthy in this connection that fused constructions have been
treated specifically as salience devices. Nichols and Schallert (1983) view the
accusative with null ‘be’ as a subtype, derived by deletion of an underlying
participle, of the accusativus cum participio (an ecclesiastical-register con-
struction unattested in trial transcripts). They argue that writers chose these
R F T 123
While fused reports (see Section 4.5) are incorporated grammatically into the
narrative to a greater degree than and , they nevertheless retain, if not
their identity as independent clauses, at least their propositional structure and
perhaps also some elements of their original surface structure. A greater
degree of integration can be found in narrative reports of speech acts
(’s), to use Leech and Short’s term (1981: 323–24). In this strategy,
is, as it were, “submerged” (Page 1988: 34) into the narrative by means of
speech-act verbs (’s); verbal acts are recounted in much the same way as
R F T 125
wording — that is important for the outcome of the trial. The use of
would introduce elements of the speech situation with no relevance for the
verdict, give the reported information a degree of prominence incommensu-
rate with its functional load, and add diffuseness without a clear function in
the discourse. By contrast, the choice of for the preceding exchange [1–2]
is well motivated for two reasons. First, the two litigants’ statements occur
in a string of dyadic exchanges; second, they serve as testimony (or testimo-
ny-like information) in that they convey the litigants’ joint agreement to
settle the suit by oath. The juxtaposition of with has the effect of
contrasting Dmitrij’s words (testimony) [2] with his deeds (a speech act
viewed less as a message than as a form of narratable action) [3].
Two further cases of ’s conveying nonfeasance where verbal action
is expected appear in what is, to my knowledge, a unique ecclesiastical
lawsuit over the jurisdiction of a monastery (25).32 The judge is the metro-
politan; the litigants are proxies for the prince of Belozersk and the archbish-
op of Rostov.
(25) I [gospodin mitropolit vsprosil Feodora: Starym
and [lord metropolitan]- ask- Feodor- old-.
pak [knjazem rostovskim i belozerskim i
[princes of.Rostov]-. and of.Belozero-. and
[bojarom starym est’ li komu vedomo, [kotorye
[boyars old]-. ..3 who- be.known [which
prežnie arxiepiskopy v [Kirilov monastyr’
previous archbishops]-. to [Cyril- Monastery]-
[pristavov svoix slali, a igumena i
[bailiffs .]-. send-. and abbot- and
brat’ju sudili, i desjatilnicy ix v”ezžali
brethren- judge-. and tithemen-. their come-.
i [pošliny svoi imali? I [Feodor
and [taxes .]-. take-. and [Feodor
Poluxanov i na to pered gospodinom [dovoda
Poluxanov]- also to that- before lord- [argument
nikakova ne učinil že
not.any]- make-
130 R V
Even in their diversity, the narrative reports of the litigants’ speech acts
share certain pragmatic functions. Like the other nondirect reporting strate-
gies, ’s occur when the form of the message or the message itself has
a low informational value for the judge’s decision — that is, when it has
little weight for the primary function of the text-kind. The statements
presented as ’s are not testimony; they relate either to trial procedure or
to non-events (anti-testimony, as it were). Like fused , ’s can be
viewed in part as reductive devices that free non-evidentiary segments of
narrative from being encumbered by the superfluous details, perspective
changes, and diffuseness characteristic of the standard strategy, . In other
words, ’s are “useful for summarizing relatively unimportant stretches
of conversation” (Leech and Short 1981: 324).
At the same time, because of their descriptive, summarizing character,
’s function as a means of embedding evaluation into the narrative — a
way of commenting on events without breaking the flow of narrative clauses
(see Labov 1972: 371). This probably explains why ’s are so rare in the
authorial narrative of trial records. If the scribes had strayed from the (admit-
tedly partial) objectivity of their reporting methods by including their own
evaluations of the events and statements, they would have been acting as
judges themselves; this would have undermined the central purposes of the
text-kind — to refresh the memory of the trial judges and to provide informa-
tion to judges of higher instance, as well as judges in future trials to whom the
transcript could be submitted as evidence of the victorious litigant’s ownership
(see Chapter 2). The contextualizations in which ’s freely occur — within
the testimony and in the verdict — are precisely those in which participants
other than the scribes express evaluations. Thus ’s are a reporting
method in which the writer or speaker assumes most of the interpretive
responsibility — a speaker-based discourse strategy, in which “explicitness
and clarity are primary desiderata” (Lakoff 1984: 481). Such reports are rare
in the trial record because that contextualization, by reason of its function,
favors hearer-based discourse strategies (ibid.: 482) that give the intended
interpreter the primary responsibility for imposing meaning on the text.
R F T 133
Free indirect speech () in the European languages has frequently been
treated — and continues to be so, even in the face of abundant counter-
examples — as a primarily or exclusively literary phenomenon of post-
medieval origin. In Russistic scholarship, the emergence of is generally
ascribed to the late eighteenth or early nineteenth century, the period of the
rise of modern Russian literature. In the following discussion and in subse-
quent chapters, I will provide evidence that existed in texts of purely
utilitarian purpose, where it was motivated by factors other than those
typically cited for its equivalent in modern languages (e.g., distancing,
polyphonic complication, interior monologue).36
Many studies of have discussed the profusion of terms that have been
used to designate this method. I prefer the term free indirect speech because
it is parallel with direct and indirect speech and because it describes ostensi-
ble features of the strategy. (The most frequent English alternative, repre-
sented speech, is problematic, in my opinion, because every form of is a
representation.) Because is a heterogeneous category, I use a definition
based on central examples; some kinds of reports that have traditionally been
viewed as do not fit the two essential criteria of this definition. In
establishing the very existence of the reporting mode in or any other
premodern language, it is necessary to use a stringent definition to exclude
possible members of other categories.
The first criterion is that, as implied by the label free, central examples
of appear in the discourse without explicit signals of reportedness — tag
clauses, adjuncts, or citation markers — in the authorial context (“absence of
an indication of the speaker and of reported speech by means of a reporting
clause”, Slembrouck 1986: 60). This includes unbound reports that follow
’s or occur as fadeouts from tagged or .37 Reports signalled only by
particles are a grey area; I treat them as tagged . The criterion definitely
excludes any report with a contiguous tag, even when the latter is “parenthet-
ical” (intercalated or postposed) or an adverbial adjunct, and even when the
report itself contains syntactic constructions or lexical and stylistic elements
putatively forbidden in tagged (treated as by, e.g., Banfield
1973: 25–26, 1982; Leech and Short 1981: 329–31). These fall into my
category of tagged , even if they are not prototypical examples.38
The second criterion is that central examples of , like tagged , share
134 R V
for sources of the heteroglossia (Fónagy 1986). Therefore, is the most
hearer-oriented (to use Lakoff’s [1984] term) of the varieties of . Its
reported status is identified primarily or solely by the interpreter’s ability to
distinguish between information sources and to reconstruct the configuration
and viewpoints of participants in the represented speech event. This knowl-
edge can be promoted by various contextualization cues, including “instantial
norms” (Thompson 1996: 513) established within texts or genres, as well as
“assessments, interpretations, judgments, ‘intended meanings’ more plausibly
attributable to a character than to the narrator” (Prince 1987: 35).
The ambiguity between and narrative explains, to some extent, why
the strategy had been used for hundreds of years before it came to the
attention of linguists. Many of the early investigators believed, indeed, that
was a modern, purely literary development; for example, C. Bally
claimed that it only occurred “spasmodically” in French before the nine-
teenth century, while E. Lerch connected it with the rise of the European
novel (cited in Pascal 1977: 11, 15–16) — a view revived by Banfield
(1993: 354–57).42 These claims are contradicted by the occurrence of in
various premodern languages: French, as seen in the earliest specimen of Old
French literature, the Sequence of St. Eulalia (Lips 1926: 118–29; Ullmann
1957: 99); English, e.g., in a 1612 account of witchcraft trials (Leech and
Short 1981: 332); Danish (Haberland 1986: 232–33); and other languages.
Despite this body of evidence, Banfield (1993: 340) calls (“represented
speech and thought”) a “relatively recent literary form” and asserts that “the
general consensus of scholarship first finds the style in a fully developed
form in French in La Fontaine and in English in Jane Austen and Sir Walter
Scott” (ibid.: 354; cf. Lee 1993: 382 on French).
The validity of this dismissal depends not so much on empirical evidence
as on how the “fully developed form” is defined — by universal criteria or
by optional characteristics. Thus A. Neubert distinguishes full-fledged
from a “not sufficiently marked” inchoate form found in premodern English
fiction, which he terms “auctorial” (cited by Doležel 1973: 50, Note 42).
In fact, many perfectly good cases of in modern literature could equally
be dismissed as “insufficiently marked” if the required marking is lexicon
rather than deixis. The optional characteristics may define the prototype
(though that has yet to be established), but they do not exhaust the category.
In Russistic scholarship it has traditionally been asserted, quite errone-
ously, that is an exclusively literary construction that emerged in Russian
136 R V
are in , with first-person elements referring to the speaker and vocatives
referring to the hearers. Then comes another clause of [4], this time
without any marker of reportedness; the referent of the third-person predica-
tive possessive pronoun is Kuz’ma, the owner of the stolen bees, as shown
by Fedorec’s subsequent testimony (ibid.). The non-narrative present tense in
this segment depends on Kuz’ma’s perspective. The report concludes with a
further slip into , as defined by first-person elements and vocatives [5].
It is perhaps arguable whether both indirect reports in (27a) should be
treated as and the direct reports that follow them as , since by
convention the entire turn at speaking falls under the scope of the tag. The
writer evidently felt that the first shift to [2] threatened some loss of
cohesion, since he signalled the continuation of the report by means of a
quotative particle. (Intercalated quotation markers are well-suited to maintain
cohesion because their syntactic position does not disrupt continuity with the
preceding context; see 7.6). By contrast, the second indirect report [4]
matches the formal criteria for , even in juxtaposition with attributed to
the same speaker in the same turn at speaking. Slipping into is a common
phenomenon: “So dependent is on context that it appears more readily in
the vicinity of verbs of speech or thought, or next to instances of direct or
indirect discourse…” (Prince 1987: 35).
At first glance, it scarcely seems possible that the repeated fadeins and
fadeouts in (27a) could reflect a purposive use of reporting strategies. One is
tempted to explain the switches away as later corruptions, since the docu-
ment only survives in an eighteenth-century copy. This explanation is moot
in the absence of any textual history; however, it should be kept in mind that
copyists generally avoided changing the substance of the texts as much as
possible in order to maintain their legal utility. (Exceptions to this conven-
tion can be found in formulary redactions and some ecclesiastical writs with
an edifying purpose.) Alternatively, the slipping might be ascribed to the
scribe’s carelessness or difficulty in maintaining Kuz’ma’s first-person
perspective. This explanation is unconvincing, since is used consistently
throughout Fedorec’s subsequent testimony and in the questions of the
judges. Morever, it seems odd that the writer should have slipped out of a
direct perspective only in passages representing Kuz’ma’s claims about his
own actions and belongings while maintaining it when he discusses the
actions of others.
There is, in fact, a regularity in the switches that is based on distinctions
R F T 139
writer estimated the legal weight of different portions of the message. Unlike
real-estate lawsuits, criminal trials were most concerned with establishing
guilt; the restoration of the stolen property was a secondary issue.
Backgrounding and summarizing are apparently at work in a case of
in the witnesses’ statement (27b), which follows the defendant’s confes-
sion and belongs to the testimonial evidence. (It is noteworthy that the
judges’ question to the witnesses, which would disrupt the continuity of the
testimonial portion, is omitted.)
(27) b. I [te muži, prišed pered bojare, da [vse te
and [those men]- come- before boyars- and [all those
reči po tomu ž bojaram raskazali,
statements]- as.per that- boyars-. tell-.
kak Kuz’ma govoril. I [tot Fedorec pered
as Kuz’ma- say- and [that Fedorec]- before
nimi ne zapersja togo, čto on u Kuzmy
them- deny- that- that he- at Kuz’ma-
pčel ukral, i s [temi pčelami
bees-. steal- and with [those bees]-.
Kuzma ego jal, po tomu ego
Kuz’ma- him- take-. for that- him-
obmuževali i vinu položili
denounce-. and guilt- place-.
And those men, having come before the boyars, told the boyars all
the same things that Kuz’ma had said. And that Fedorec did not
deny before them that he had stolen bees from Kuz’ma, and
Kuz’ma took him with those bees [in his possession]; for that
reason [they] denounced and indicted him (Krotov and Smetanina
1987: 68, no. 5).
Initially the witnesses’ statement is presented as a involving the preterit
of raskazati ‘tell, relate’ with the noun rěči ‘statements’ as its object. As
discussed in 4.6, ’s are used as reductive devices to present relatively
minor information in an economical manner and thus to prevent the narrative
from being bogged down in more details than would be necessary for present
purposes. These motivations are clearly at work in (27b), since the witnesses
do nothing more than confirm Kuz’ma’s statement and Fedorec’s confession.
R F T 141
The status of the underlined passage following the is not immedi-
ately evident. If interpreted as new narrative rather than an elaboration of the
preceding, it could reflect an “eye to eye” confrontation (face to face with
the accuser) like those found in sixteenth-century interrogation records (e.g.,
AAÈ 1: 143–44, no. 172). The verbs zapersja ‘denied’, obmuževali ‘de-
nounced’, and vinu položili ‘indicted’ would then be part of the authorial
narrative. Three facts militate against this view in favor of the passage being
. First, it is unclear why such a confrontation would be necessary when
the culprit himself had admitted his crime both in and out of court. Second,
as shown by the opening complaint (27a), the question that follows it, and
Fedorec’s confession, the witnesses performed the speech acts denoted by
obmuževali and vinu položili when they caught Fedorec red-handed, prior to
the current speech event. This suggests that the verbs form part of a report
rather than the authorial narrative; they are “personal preterit” (the recollec-
tion of the represented speakers) rather than “narrative preterit” (cf. Padučeva
1996: 381). Third, the causal po tomu ‘for that reason’ is an evaluation of
motives rather than an objective account of observable actions; as such, its
most probable source is the witnesses rather than the writer, given the covert
style of narration. Spitzer (1928/1988: 74–75) shows that causal clauses that
are formally part of a third-person narrative often do not belong to the
viewpoint of the narrator but rather express the rationales of dramatis
personae, as filtered through the narrator’s perspective (“pseudo-objective
discourse”).45 In other words, causal clauses often provide clues to some
internal focalization.
As , the underlined statement in (27b) can be understood as a fade-in,
a selective amplification of the preceding dealing only with the
evidence that the witnesses can uniquely offer — their corroboration of
Fedorec’s confession and their own subsequent actions. When occurs
alongside ’s, it represents a relative “movement towards directness”
(Leech and Short 1981: 325, 336). Like other nondirect strategies, the in
(27b) serves to report information of lower contextual importance without
foregrounding it. This ranking function is similar to the backgrounding,
deferring use of in the opening statement. Since much of the witnesses’
testimony is presupposed from Kuz’ma’s statement and Fedorec’s confession,
there is no need to report it in the same actualizing manner as the preceding
dialogue; that would retard the pace of the narrative more than necessary for
perception of the content. The lack of a tag and subordinator in promotes
142 R V
The of the judges in the trial narrative, which is illustrated in some of the
excerpts in the last chapter, has a different character from the reports of the
litigants and witnesses. The content of their statements was evidently limited
both by their social role and by the conventions of the text-kind. The
transcripts themselves depict the judges acting primarily as moderators
during the on-site trial; their principal task (other than listening) was to elicit
as much evidence as possible: “judges tended to limit themselves to an
essentially supervisory role, ensuring each participant an opportunity to
counter evidence produced by the opposition with proof of his own” (Klei-
mola 1975: 74–75; see also 80–81). The judges assumed a leading role only
at the moment of the verdict. Indeed, they were not always legally empow-
ered to render verdicts; in many cases they were restricted to gathering
evidence, which they had to refer to a court of higher instance (see Chap-
ter 2) — one of the motivations for the very existence of trial transcripts as
a text-kind.
The trial judges’ peculiar role is reflected in the way that they are
identified in the transcripts. As was discussed in 3.1, the litigants and
witnesses are usually indicated by their names or nicknames in tag clauses
preposed to their reports; less frequently, titles or other lexical substitutes are
used, but in general potentially ambiguous forms of reference are avoided.
By contrast, it is rather common (ninety transcripts) for the judges to be
identified in tag clauses not by their names but by the generic terms sud’ja
‘judge’ or, in a few cases, pis’c’ ‘chancery secretary.’ More explicit forms of
reference (names and/or unique titles) are found in fifty-seven of the tran-
scripts, and a mixture of unique and generic terms occur in eight others; in
152 R V
many of these examples, the judges were prominent boyars or princes, whom
there was presumably some social pressure to identity. However, in several
cases the judges are left anonymous not only in the tags but also in the
incipit and end protocols, so that there is no clue whatsoever to their
identity. In addition, when there are several judges hearing a trial, named or
unnamed, they are always presented as a collective, never individuated; that
is, choral speech is the norm.1 In one transcript involving teams of judges,
the collective attribution sud’i ‘judges’ is used when the teams act together;
when they act separately, they are labelled not by their names but by their
liege lords — sud’i knjaži Borisovy Vasil’eviča ‘Prince Boris Vasil’evič’s
judges’ and velikogo knjazja sud’i ‘the grand prince’s judges’ (Antonov and
Baranov 1997: 287–88, no. 296). Even here, the judges are presented as a
chorus rather than as individual voices.
These data suggest that, in many cases, the scribes considered the judges’
identity irrelevant for the purpose of the documents. If the trial judges
referred the trials upwards, they would be at hand to supply the higher
judges with all the necessary information. In any event, there was no need
for precision in attribution, because what the judges said was not testimony
and had no direct bearing on the verdict.
What, then, was the function of the judges’ , if it did not supply the
evidentiary information central to the text-kind? It was not included for the
sake of verisimilitude, even though the question-and-answer format undoubt-
edly reflected actual trial procedures. One may note that the questions are
omitted altogether in certain other text-kinds that report speech from dialogic
exchanges — for example, inquest and interrogation records (doprosnye reči,
rassprosnye reči) and depositions (skazki) (e.g., AI 1: 174, no. 120; 192, no.
130). If these documents are compared to trial transcripts, one of the
differences that emerges is that each episode tends to focus on the testimony
of a single speaker or a group of associated speakers; thus that there is little
danger of misattributing partisan information. This contrast reveals the
essential conventional function of the judges’ in trial transcripts — to
enhance interpretability by promoting coherence in the discourse. (It is
noteworthy in this connection that the judge’s questions, along with the tags
of the interested participants’ , are omitted altogether in a trial précis used
for other purposes than ordinary trial transcripts [ASÈI 3: 348, no. 319].)
One of the ways in which the judges’ increased coherence was by sep-
arating the interested participants’ reports from one another (thus supporting
T Q F 153
the function of the preposed tags) and by creating dyads, the divisions that
are the fundamental organizational units in the transcripts. The sentences that
tag the questions — distinguished, as a rule, from those that introduce the
responses by the use of a verb of asking rather than a — partition the
discourse by indicating the beginning of a new dyad; the messages conveyed
by the questions act like titles in establishing the overall theme of the dyad.
The perceptual reality of this structure is borne out graphically in fourteen
out of the fifty-eight autographs and in one of the six contemporary copies
published in diplomatic editions, in which the questions as the beginnings of
the dyads are singled out by capital letters (indicated by boldface), while the
responses as continuations are not marked in any special way:
(1) a. I sud(’)i sprosili Onikeja da Semena… I
and judges-. ask-. Onikej- and Semen- and
Onikei i Senka tak rkli… I
Onikej- and Senka- thus speak-. and
sud(’)i sprosili Savki… I Savka tak
judges-. ask-. Savka- and Savka- thus
rek… I sud(’)i sprosili Makuty da
speak- and judges-. ask-. Makuta- and
Filiska… I Filisko i Makuta tak rkli…
Filisko- and Filisko- and Makuta- thus speak-.
And the judges asked Onikej and Semen… And Onikei and Senka
spoke in this way… And the judges asked Savka…. And Savka
spoke in this way… And the judges asked Makuta and Filisko…
And Filisko and Makuta spoke in this way…
(ASÈI 2: 193–94, no. 285).
This system of capitalization divides the trial dialogue into subsections
initiated by the questions rather than into individual turns at speaking.2 The
same question-dominant partitioning method operates in the nondyadic
context of multiplex testimony (1b) — a further proof that the answers in
this contextualization tend to be treated as a single instantiation of testimony
spread over more than one (see 4.2).
(1) b. I [knjaz(’) Ivan Jur’evič(’) sprosil” [manastyr’skix
and [prince Ivan Jur’evič]- ask- [of.monastery
154 R V
signal a shift to a new turn at speaking and perhaps even a new kind of
verbal encounter (see 4.2). (This function is also seen in the sole case of
g(o)s(podi)ne, which occurs in a transcript of a trial being judged by two
teams of judges, one appointed by the local prince, the other by the grand
prince. At a certain point, the local prince’s judges, who otherwise speak in
chorus with the grand prince’s judges, are depicted turning from a litigant to
consult their colleagues on a procedural matter. The vocative is used to mark
this transition from the dominant encounter to byplay.)3
Another way in which the judges’ promotes coherence is by providing
a context (frame) for the testimony. The continual shifts of perspective
caused by the alternation of represented speakers had the potential to disrupt
coherence — an effect that could only have been aggravated by the possibil-
ity of layering within the report. This was mitigated by the redundancy
generated by the judges’ questions, which provided an ample context for
each statement in the line of testimony. The wording of the questions in the
written transcript created expectations in the minds of the intended interpret-
ers, just as it did in the actual trials. The echoes of the judges’ inquiries in
the testimony, like other repeated elements (cf. Halliday and Hasan 1976: 13;
Brown and Yule 1983: 194; Tannen 1989: 50–51), promoted connexity and,
in particular, enhanced the internal cohesion of the dyads. Elements topical-
ized in the questions became quickly processed givens in the answers; this
focused attention on the less expected parts of the replies, the new informa-
tion (a typical function of frames, cf. Tannen 1989: 49–50). The very fact
that a question was asked highlighted the following response, since it is
ordinarily assumed that questions will be answered; cf. the cataphoric use of
the adverb tak(o) in the standard tagging formula. Highlighting is a general
function of questions, which, like other collateral information or comparators,
raise the possibility of alternative nonevents to heighten the significance of
the actual events (see Grimes 1975: 64; Labov 1972: 384–85, 387). (There
are, to be sure, isolated cases in which the judge’s report has been omitted,
if analogous contexts in other transcripts are any indication — e.g., when
witnesses just named by a litigant are depicted testifying without any
intervening question [ASÈI 1: 247–48, no. 340 (2×)]. This is a context with
high predictability and little risk of mistaking one position for another.)
The framing function of the judges’ questions became less important in
transcripts of the late sixteenth and seventeenth centuries. During this period
trials came to be conducted primarily through petitions and documentary
156 R V
evidence, and verbal confrontations became less central to the process. The
format of these lawsuit records resembles that of the earlier interrogation
transcripts mentioned above, to some extent. Testimony became increasingly
compartmentalized, and separate hearings were held to corroborate the
written evidence of a single party or speaker in the suit. With these limita-
tions on the possible speakers in a given hearing, it would have been
uneconomical to maintain the diffuse dyadic structure characteristic of
transcripts of the earlier period, and the framing function of the judges’
questions, in particular, became superfluous. Consequently, in later tran-
scripts (e.g, Jakovlev 1943/1970: 331, no. 4.1; 342, no. 5.7), the questions
were often concentrated at the beginning of the transcript or, in many cases,
omitted altogether as the least informative portion of the trial dialogue.
Features such as anaphora and the citation marker dě(i) now functioned as
the chief signals of partitioning in the discourse.
As noted above, the subject of this formula can involve more specific terms
(e.g., titles and/or names). The genitive object5 is a name or noun phrase
referring to the addressee; this identification is necessary because the ratified
addressee can change from dyad to dyad. The tag verb v”sprositi can appear
in the variant spellings v(”)s(”)prositi or vosprositi.6
A related verb of inquiring, s”prositi (with spelling variant sprositi), also
frequently appears. As the following passages suggest, the two verbs are
evidently in free variation, without any difference in meaning, dialect, style,
or function:
(3) a. I sud’ja vs”prosil Kazaka da i eg(o) tovariščov:
and judge- ask- Kazak- and his fellows-
Otvečaite!… I sud’ja vs”prosil [stroitelja Ieva:
answer-. and judge- ask- [abbot Iev]-
Počemu žo ty zoveš(’) [tě požni
why you-. call-.2 [those fields]-.
sp(a)sk(i)mi izstariny?… I sud’ja vs”prosil
of.Savior-. of.old and judge- ask-
Kazaka da i eg(o) tovariščov: A vy
Kazak- and his fellows- and you-.
zovete [tě požni [velikog(o) kn(ja)zja,
call-.2. [those fields]-. [grand prince]-
komu to u vas vědomo…
whom- that- at you-. be.known
[1] And the judge asked Kazak and his companions,
“Respond!”… And the judge asked Abbot Iev, “Why do you
claim those fields are Savior [Monastery’s] of old?… [2] And the
judge asked Kazak and his companions, “And you claim those
fields are the grand prince’s, whom do you have as witnesses to
that?” (ASÈI 2: 500, no. 463).
b. I sudia sprosil Vaska: Otvěčai!… I sudia
and judge- ask- Vasko- answer- and judge-
sprosil [starca Semiona: Počemu že ty [tu
ask- [monk Semion]- why you-. [that
požnju zoveš(’) monastyr’skoju?… I sudia
field]- call-.2 of.monastery- and judge-
158 R V
There are a few cases in which the object of the verb is omitted, so that
there is no indication of the addressee — e.g., (6) [2] (see Section 5.2).
These examples can be regarded either as random deviations or as idiosyn-
cratic but purposive usages. Although the evidence is slight, it suggests that
T Q F 163
the departure from the convention may have been promoted by certain
features of the context.
(7) I sud(’)i sprosili Kur’ana: Komu to
and judges-. ask-. Kur’jan- whom- that-
vědomo… I Kur’an” tak rek: Vědomo, g(ospodi)ne,
be.known and Kur’jan- thus speak- be.known lord-
[ljudem dobrym starožilcem [Ivašku Braginu, da [Ivašku
[people good old.residents]-. [Ivaško Bragin]- and [Ivaško
Šestakovu… I sud(’)i v”sprosili: Skažite v pravdu,
Šestakov] and judges-. ask-. tell-. in truth-
č’i to požni, na koix stoim?
whose-. fields-. on -. stand-.1.
[Ivaško Šestakov tak rek…
[Ivaško Šestakov]- thus speak-
And the judges asked Kur’jan, “To whom is that known…” And
Kur’jan spoke in this way: “It is known, lord, to good people, the
longtime residents Ivaško Bragin and Ivaško Šestakov…” And the
judges asked, “State in truth, whose fields are those on which we are
standing?” [And] Ivaško Šestakov spoke in this way…
(ASÈI 2: 321, no. 334).
This passage exemplifies the rhetorical use of to introduce new partici-
pants (cf. Larson 1978: 60–64), which is typical in trial transcripts. The
fact that the respondent has just named his witnesses creates a presupposition
that one of the named parties will be the next addressee, in accordance with
ordinary procedure. This presupposition is supported by the content of the
question, since the formula S”kaži(te) v” prav’du ‘state in truth’ is reserved
for interrogating witnesses. Thus the context obviates the need to identify the
addressee, especially as the witnesses’ testimony follows the question
without interruption. (This is reminiscent of the use of names alone as tags
in dramatic texts, where the reader expects a change of speakers with every
paragraph. Cf. also the use of with statements in boundary-marking
descriptions, where the identity of the speakers is predictable and there is a
quick alternation between narrative of actions and .)
In (6) [2] (see Section 5.2), the omission of the object also seems to be
conditioned by context-generated expectations, although here the addressee
164 R V
is given in the narrative rather than in the preceding . The ellipsis occurs
in an exchange in which the judge has just been interrogating the same
addressee, whose identity is thus well established in the context. This is also
true of a further example from the same transcript:
(8) I sud’ja vsprosil Korovaja: a ty pak
and judge- ask- Korovaj- and you-.
zoveš’ Ikonniče zemlja [velikogo knjazja
call-.2 Ikonniče- land- [grand prince]-
počemu? I Korovaj tak rek: vedomo, gospodine,
why and Korovaj- thus speak- be.known lord-
[ljudem dobrym, čto [to Ikonniče zemlja [velikogo
[people good]-. that [that Ikonniče]- land- [grand
knjazja. I sud’ja vsprosil: komu vedomo, čto to
prince]- and judge- ask- whom- be.known that
zemlja [velikogo knjazja? Vedomo, gospodine, to [Ofone
land- [grand prince]- be.known lord- that- [Ofona
Stafurovu…
Stafurov]-
And the judge asked Korovaj, “And you, why do you claim Ikonniče
is the grand prince’s land?” And Korovaj spoke in this way: “It is
known, lord, to good people that that Ikonniče is the grand prince’s
land.” And the judge asked, “To whom is it known that it is the grand
prince’s land?” “It is known, lord, to Ofona Stafurov…” (AFZX
227–28, no. 259).
Judging from internal evidence, most litigants named their witnesses right
after they asserted that they could produce them; Korovaj’s hesitation thus
gives the appearance of hedging, as, indeed, it does in the other example
from this transcript. The question with the abridged tag in (8) is a rejoinder,
with predetermined content, to Korovaj’s tantalizing statement. Clearly the
reduction of the tag is motivated by and, as it were, iconic of the decrease in
new information and the increase in the pace of the dialogue that presumably
results from it. The same cumulative effect probably accounts for why
is used for Korovaj’s reply at the end of (8), and also why the is omitted
in the otherwise standard tag clause following the abridged question in (6)
[2] (Section 5.2). In general, ellipsis of parts of the tag is characteristic of
T Q F 165
In addition to the formula cited in (2) (see Section 5.2), there are thirteen
instances (in eight transcripts) in which of the judges is tagged with
tak(o) rek(li) ‘spoke in this way.’ Most of these tag clauses include datives
referring to the judge’s addressees, parallel to the genitive object in the
conventional formula — a clarifying device motivated by the judge’s
customary role as the initiator of exchanges. The use of this strategy with the
judges’ is not surprising, given it is the standard formula for tagging
testimony, the basic speech event in the text-kind. Moreover, none of the
thirteen examples involves an interrogative .17 Ten are directives, some
with declarative preambles, while the other three involve refusals, in two
cases accompanied by promises. Thus the violation of convention goes in the
direction of semantic appropriateness. Although the reči is semantically
broad enough to tag any kind of reported sentence, it is associated proto-
typically with noninterrogative ’s, unlike verbs of inquiring, although, as
mentioned above, directives that elicit information are relatively close to the
’s that verbs of inquiring convey. In one transcript, indeed, the choice of
verbs clearly reflects the division between interrogative and stage-managing
’s: questions and commands that elicit verbal testimony are tagged by the
preterit of v”sprositi, while the four statements about trial procedure and
directives relating to the participants’ movements are tagged by tak(o) rek(li)
plus the dative of addressee (ASÈI 1: 556–60, no. 642).
In trial transcripts, the reči, as the core of the standard tag, is
prototypically associated with responses in dyadic exchanges. This dialogic
function accounts for the use of tak(o) rek(li) in refusals, where the judge is
put in the non-dominant position — unusual for the text-kind — of having
to respond to interested participants (9):
166 R V
the other cases of refusal, the judges are likewise depicted reacting to situa-
tions initiated by interested participants that they perceive as outside their
legal competence (ASÈI 1: 235, no. 326; ASÈI 1: 559–60, no. 642). Thus the
reports tagged by tak(o) rek(li) reflect sharp changes in footing; the judges
cease to be dominant moderators, and other speakers assume, for a time, the
leading role in driving on the dialogue. (The reversal is particularly striking in
the second case, in which stonewalling litigants threaten to go over the judge’s
head if he does not decide the suit in their favor.) The verb reči, associated
with responses, is more motivated in these contexts than verbs of inquiring
would be, since the intent of the judge’s statement is not to elicit information.
It may be possible to treat some of the other reports of the judges tagged
by tak(o) rek(li) as reactive and discourse-continuing rather than dominant
and discourse-initiating — e.g., when the judge is answering a request for an
adjournment (ASÈI 1: 558, no. 642). This may account for the unusual
instance in which the plaintiff produces a witness preemptively in his
opening complaint; in violation of the customary procedure, the judge reacts
to this not by asking the respondent to counter but by ordering the plaintiff
and his witness to show him the boundary — a tagged by tak(o) rek(li)
(ASÈI 2: 364, no. 370). When the judge does eventually turn to the respon-
dent, he asks him not for a rebuttal but for a demonstration of his version of
the boundary — another procedural anomaly tagged by tak(o) rek(li) (ibid.).
Another of the examples depicts the judge reacting to a bipartisan agreement,
initiated by the respondents, to accept whatever version of the boundary a
specific group of peasants will offer (AFZX 139, no. 157; cf. also ASÈI 2:
351, no. 358). These cases are further illustrations of the tendency, seen
frequently in the transcripts, for unusual speech events to be depicted by
means of uncanonical reporting strategies.
Free direct speech is not a typical strategy with the judges’ reports; this is
hardly surprising, since the omission of a tag clause would tend to counteract
the framing function discussed in 5.1. However, is employed in a
strikingly purposive manner in one idiosyncratic trial transcript, where it
occurs five times over a series of ten questions. While the judge is depicted
interrogating five allied participants in this sequence, the tag clauses are only
168 R V
omitted before certain questions to the plaintiff, Ondron. This is clearly not
the result of scribal error; on the contrary, the writer of this extraordinary
transcript utilized selectively to create a tightly structured discourse.
From the scribe’s viewpoint, there are only two real participants in the
series of questions, the judge Mikita and the plaintiff Ondron; and there are
only two issues being discussed in this dominant encounter — which
properties Ondron intends to claim and what basis he has for those claims.
Each issue forms the subject of an episode or paragraph consisting of several
dyads. Within each paragraph, only the first of Mikita’s questions is present-
ed by means of the conventional tag clause; the following questions are
consistently given in , while Ondron’s responses are all presented as
introduced in the standard way. This full ellipsis of the tag clause may, to
some extent, be a device for conveying rapid-fire exchanges, as with the
cases of partial ellipsis discussed in 5.3.
This structure is complicated by the fact that, in three cases, Ondron
asserts that someone else owns the properties in question. Each time this
occurs, the judge’s question to the putative owner is tagged, even though his
following question to Ondron is presented as . In effect, the scribe is
treating Mikita’s brief exchanges with Ondron’s allies Torop, Uvarko, and
Frolec as crossplay — “communication between ratified participants and
bystanders across the boundaries of the dominant encounter” (Goffman
1981: 133). The omission of the tag in questions to Ondron is iconic of
continuation — a return to the main speech event in the sequence.
The sequence of questions begins with two dyads dedicated to clarifying
the extent of Ondron’s claims (10a). Mikita’s first question is tagged, with
Ondron identified as the addressee [1]; his second is presented as , with
an appellative (a nominative instead of the conventional vocative) at the end
to show that Ondron is still the addressee [2].
(10) a. I Mikita vsprosil Ondrona: Tvoja li to
and Mikita- ask- Ondron- your-.
zemlja Merinovo? I Ondron tak rek:
land- Merinovo- and Ondron- thus speak-
To, gospodine, Merinovo — zemlja [šurina
that- lord- Merinovo- land- [brother-in-law
moego Gridkina… A Podymskaja [zemlja
my]- Gridka-. and Podymskaja- [land
T Q F 169
While the judges’ directives are mostly reported as tagged , there are
certain classes of exceptions that tend to be presented in a semantically
reduced, grammatically dependent form. As the judges’ chief role in the trial
hearing was to gather evidence, most of the commands reported in the
transcripts serve as requests for information and require verbal responses.
The framing function that the commands peform in adjacency pairs motivates
the use of , with its diffuseness and deviations from the narrative orienta-
tion. However, the judges also had a secondary role as “directors” or “stage
managers” responsible for organizing the movements and judicial actions of
the interested participants. Directives stemming from this role are typically
conveyed in a way that reduces the intrusiveness of the in the narrative —
as infinitival complements after the velěti ‘bid, command’ (see also 4.6).
This kind of speech event is not a prototypical dyadic interchange, as
Longacre (1983: 57–58) points out: “Such structures contain initiating
utterances but probably should not be considered to contain resolving
utterances since the resolution is non-verbal.”
Most of the directives reported as complements of velěti reflect three
kinds of situations: the judges are setting a time for reconvenement,18 having
documents read before the court,19 or ordering the litigants to produce
previously named witnesses.20 Such commands are concerned with the parti-
cipants’ movements or other behavior, and only indirectly with testimony;
they expect no response besides compliance, and, as a rule, their reports are
followed not by but by narrative. Thus in one case the judge is depicted
sending an intermediary for information (ASÈI 2: 411, no. 402; see (7) in
4.2); this is stage-managing, not eliciting testimony by direct communication.
The only recurrent situation in which the infinitive after velěti overlaps
with is in orders initiating boundary-setting procedures. Here the illocuti-
onary intent may involve both eliciting testimony and directing the partici-
pants’ movements; however, cooperative responses are not primarily verbal
in nature. The prevalent strategy is ; ’s are only attested five times.21
It is suggestive that, in a transcript where both methods are used, a command
followed by narrative of action is given as a , while one that unexpect-
edly occasions dialogue is presented as tagged (11):
T Q F 173
after velěti ‘command’ or, when appropriate, after the phrasal verb dati/
učiniti s“rok” ‘set a date (for reconvening)’ (13):
(13) I o sem Suxoj reksja doložiti [gosudarja
and about this- Suxoj- promise- ask- [sovereign
velikogo knjazja a [oboim istcem i
grand prince]- and [both litigants]-. and
znaxorem učinil srok stati pered [velikim
witnesses-. make- term- appear- before [grand
knjazem o [tem dele… a čto na [tom selišče
prince]- about [that matter]- and in [that village]-
seno v stozex, i Suxoj [togo sena ne
hay- in stacks-. and Suxoj- [that hay]-
velel svoziti s [togo selišča s Pavlecova
bid- take.away- from [that village]- from Pavlecovo-
do doklada
before referral.hearing-
And Suxoj promised to ask the sovereign grand prince about this, and
he set a term for the litigants and witnesses of both sides to appear
before the grand prince in that matter… And, as regards the hay in the
haystacks in that village, Suxoj commanded that it not be taken away
from that village of Pavlecovo before the higher-court hearing (AFZX
234, no. 261).
These reported commands recall the way that the judges’ directives are
typically conveyed in another non-dialogic context, the verdict (see Chap-
ter 8).24
C 6
As previously noted, the judges who heard the testimony in the trials were
not always able or empowered to hand down verdicts. There are many
transcripts that illustrate a further stage in the legal process, in which the
trial judges referred the suits to judges of higher instance, who could be
grand princes or princesses, metropolitans, or boyars with special judicial
authority (see Kleimola 1975: 68–74). The general term for hearings before
higher judges (which were not restricted to lawsuits) is doklad ‘judicial
referral.’ Judicial referral was one of the motivations for trial transcripts as
a text-kind; the higher judges used the records of the on-site trial hearings as
the principal source of information for their verdicts. Written records of the
higher-court hearings and verdicts were then appended to the recopied trial
records as proof of the judgment in the event of future litigation. The
resulting documents are known as “referred-judgment trial records”
(dokladnye sudnye spiski) or, if certified by court secretaries, “signed
referred-judgment trial records” (podpisnye (dokladnye) sudnye spiski).
However, many medieval copies are labelled more loosely as “judgment
charters” (pravye gramoty), the term for any trial transcript that includes a
verdict. The scribes who composed the appendices are often anonymous;
however, given the nature of the judicial system, there is good reason to
presume that they were not the scribes who transcribed the trial proceedings
in situ. Thus referred trial copies can potentially reflect the interference of a
second writer.
Most judicial-referral records are brief and staged according to the set
pattern illustrated in (1), which evidentally reflects a ritualized procedure.1
178 R V
to falsify the trial transcript [3] (see 6.2). This verification usually completes
the judicial-referral record; however, in some cases it is followed by reports
of additional testimony — documents the litigants were unable or unwilling
to submit during the on-site trial; statements elicited by the judge from
government officials; or corroborations of trial record by the “men of the
court” (sudnye muži), the observers of the on-site trial (see 6.2–6.3).
The verification process recounted in (1) was one of the principal
institutions in which trial records were utilized in medieval Russian law. It
was motivated, at least in part, by a recognition that the reporting methods
were fallible (see 3.3). This also accounts for the use of typicality markers
like takov sud ‘such a trial’ in reference to the reported dialogue, in contrast
to the token marker used in reference to the written document in ses(’) spisok
‘this [trial] record’ (see 3.4). As noted in Chapter 2, the document did not
“fully constitute the legal act” (Kittay 1988: 211) until it was ratified as the
official reading with the verdict; prior to that, it was a memorandum
ancillary to the recollections of the litigants and men of court. Indeed, the
primarily oral, reconstitutive nature of the judicial-referral hearing is striking.
The higher judges, though presumably literate, are conventionally said to
have “heard” the trial, with its voices reanimated by the spoken performance
of the trial record and, as suggested in some transcripts, by the spoken report
of the trial judge (2).
(2) I [knjagini velikaja, vyslušav sud, po spisku i
and [princess grand]- hear- trial- through copy- and
po [sud’inu slovu po Vasil’evu, velela…
through [judge- word]- through Vasilij-. bid-
And the grand princess, having heard the trial, according to the [trial]
record and the [trial] judge Vasilij’s statement, ordered…
(AFZX 259, no. 308).
Judicial-referral hearings led to significant shifts in the trial participants’
footing; these are reflected in the composition of the text. As a rule, once the
trial judges have made their reports, presented their trial records, and
produced the litigants, they retire into the background; they reappear in the
verdict, first as addressees and then as executors of the higher judge’s
directives. The litigants mostly play a limited corroboratory role similar to
that of the witnesses in the on-site trial. This reduction in their prominence
180 R V
6.2 Verifications
And the plaintiff Karpik, also representing his brother Fedko, and
Mitja Galaseev, representing the monk Aleksandr, said that they
had had a trial such as [is/was] written in this copy
(ASÈI 3: 222, no. 209).
b. I [oba istca skazali: Sud im
and [both litigants]-. say-. trial- them-
takov byl, kak v spisku pisano.
such- be- as in copy- write-
And both litigants said they had a trial such as [is] written in the
copy (ASÈI 2: 419, no. 405; see also ibid.: 416–17, no. 404).
The presence of can be seen from the third-person pronouns referring to
the represented speakers; thus there is a shift from to in the transition
from the judge’s question to the litigants’ choral response. Given the
absolute identity of the contextualizations, the absence of the complementizer
in (4b) should perhaps be ascribed to scribal idiosyncrasy. It is likely,
indeed, that both of the transcripts with this strategy are the work of a single
scribe; they both record lawsuits of 1490–98 involving the Simonov Monas-
tery and have the same trial judge, the same judge of higher instance, and
the same chancery secretary as signatory (ibid. 2: 417, 419). Moreover, their
verification formulae share another peculiarity — the omission of the
demonstrative sem’ in the second part of the correlative construction.
In a further case (ASÈI 1: 401, no. 523), uncomplementized tagged by
s(”)kazati appears in the second of two conjoined reports, the first of which
is complementized . Here the first report conveys the verification of the
trial record, while the second, uncomplementized one recounts the corrobora-
tion of an earlier judicial-referral record. (The transcript contains two such
records, one from the court of a boyar judge, the other from the court of the
son of the grand prince, who had been given the privilege of deciding all
real-estate litigation.) Judging from other contextualizations, there appears to
be a tendency to omit the complementizer in the second report of a tauto-
subjective sequence (see 7.4, 8.3.1). In any case, the scribe of the second
judicial-referral record shows a proclivity for reduced structures; he also
omits the judge’s and does not individuate the litigants in the tag clause,
despite the fact that one of them is a proxy. These abridgments are evidently
motivated by the fact that the information in the second judicial-referral
record is repetitive.
184 R V
change not only in the character of the reported information but also in
footing of the represented speaker in the hearing. The respondent ceases to
be a passive corroborator and takes the initiative; it is his reported statement
[2] that determines the further direction of the speech event.
It is noteworthy that the additional testimony in (5a) [2] is conjoined to
the preceding context by a instead of the sequencing conjunction i. As noted
in Sections 4.2–4.3, the non-sequencing a is used contexts such as multiplex
testimony to imply that the reports thus conjoined are viewed as distinct
parts of a single statement. If this is true in (5a), the portion of the respon-
dent’s statement that overlaps with that of the plaintiff is presented as choral
[1], while the rest is presented individually, of necessity, but still not as a
fully separate turn at speaking [2]. This bracketing is reinforced graphically
in (5a) and three other autographs (ASÈI 1: no. 583, 587, and 589) where
the initial a is not capitalized, unlike the i of the following turn at speaking
(boldfaced).
As (5a) shows, any new documents that the litigants adduce are summa-
rized and backgrounded through the use of a complementizer and without
any features of “direct” (non-narrative) orientation. This usage shares the
motivation for — more precisely, the disincentive for — that is under
discussion. The judgment charter mentioned in (5a) [3] is only cited as a
confirmation of the respondent’s previous assertion [2], which it echoes; in
other words, the summary of its pertinent information is largely presupposed.
In general, tends not to be an optimal strategy for summarizing because
of its diffuseness and the perspectival information that it conveys, which is
superfluous in the given function.
By contrast, extracts from land-cadastre books, which were housed in
state chanceries, are cited in (5b). In such cases, the judge sends a
secretary — an official of some dignity — to obtain the information from
the chancery secretary outside the hearing;
(5) b. I [knjaz(’) Vasilei Ivanovič(’) velěl [d’jaku svoemu
and [prince Vasilij Ivanovič]- bid- [secretary .
Oleške Bezobrazovu u d’jaka u Mikifora is
Oleška Bezobrazov]- at secretary- at Mikifor- from
[kostromskix knig [manastyr’skie zemli
[of.Kostroma books]-. [of.monastery lands
R J-R H 187
Here again the report is treated like the verification procedure immediately
preceding it — as complementized tagged by s”kazati; note the emphatic
third-person pronoun referring to the represented speaker. (Where there is no
emphasis, subject pronouns are often omitted in tautosubjective complement
clauses.) The fact that statements are corroboratory and partially presupposed
does not imply that they are trivial. Thus the indirect report in (6) is quoted
in the verdict (ibid.) as being of pivotal significance, since the plaintiff’s
claim to the land had revolved around its not being a different property.
There is considerable evidence that complementized was generally
preferred for reporting simple, presupposed, corroboratory messages in
chancery language of the fifteenth and early sixteenth centuries. This can be
seen in other text-kinds that record official verifications. For example, many
of the surviving fifteenth-century wills were submitted to high churchmen for
probate; during these hearings, the scribes and witnesses were called on to
corroborate the testaments. The records of these proceedings, illustrated by
an example from 1433 (7), were appended to the proven wills.
(7) A [sija duxovnaja gramota [Ione vladyce javlena, a [pop
and [this spiritual writ]- [Iona bishop]- show- and [priest
Genadej i vo [vsex posluxov mesto tuto ž
Genadij]- in [all witnesses]-. place- there
stojal pered vladykoju i skazal, čto [sja duxovnaja
stand- before bishop- and say- that [this spiritual
gramota pered nimi pisana. A [Ivan Vasil’evič’
writ]- before them- write- and [Ivan Vasil’evič]-
skazal, čto [sju gramotu on pisal.
say- that [this writ]- he- write-
And this will [was] shown to Bishop Iona, and the priest Genadij,
representing all the witnesses, stood there before the bishop and said
that this will [had been] written before them. And Ivan Vasil’evič said
that he had written this document (ASÈI 1: 87, no. 108).
Complementized after s”kazati, as in the underlined segments, is the sole or
main way of presenting in 14 out of the 16 cases of this text-kind examined
(including all of the cases in ASÈI, AJu, and AJuB).5 In one example, an
indirect report is continued by an even more integrated strategy, fused
(ASÈI 1: 100, no. 6); in another, the first report is complementized and the
190 R V
The use of here is indicated by the first-person pronoun and the vocative
referring to the participants of the represented speech event. This example is
particularly unusual in that it depicts the witnesses as well as the litigants
participating in the verification routine.
Even though is not optimal for representing presupposed corrobora-
tions, there are at least two reasons why it is not surprising to find it in this
contextualization. First, is associated with testimony, the prototypical
speech event for the text-kind. It would require little stretch of the imagina-
tion to view the litigants’ statements and behavior during the judicial-referral
hearing as additional evidence (cf. Kleimola 1975: 85–86). Analogous
verifications of documents, including earlier trial transcripts, can be recount-
ed within the record of the on-site trial on a par with ordinary testimony —
as they are, in a way, since the history of the cited documents is not an
observable for the judge or scribe (see AFZX 229, no. 259; ASÈI 2: 419, no.
405; 461, no. 422). In such cases, the answers not only verify but also
amplify the information contained in the old trial transcript; even negative
responses add to the body of evidence by casting doubt on the claims of the
opposing parties or, ultimately, on the veracity of the speakers. Second,
is associated with dyadic reports in narrative, and the verification procedure
is dyadic (or quasi-dyadic, given the choral speech) and set in a minimal
narrative. Indeed, in some transcripts, the trial narrative flows continuously
into the judicial-referral record, without the closure provided by the final
, as if the statements in the higher court continue those of the lower
without any temporal break (see, for example, ASÈI 1: 434, no. 557; and
541, no. 628).
While the use of in the judicial-referral verification can be interpreted
as an extension or continuation of the pattern ordinarily employed for
testimony in the trial record, there is another possible explanation. The
chronological distribution of the examples suggests a drift from direct to
indirect style. Especially noteworthy in this regard is the preponderance of
complementized in the early-sixteenth-century transcripts (13 out of 15
cases). However, the earlier transcripts investigated do not provide fully
secure evidence, given that the time gap between the earliest attested
examples of and is some forty years at the outside. Preliminary
examination of later trial transcripts shows that complementized after the
preterit of s”kazati likewise predominates in judicial-referral records from the
reigns of Vasilij III Ivanovič (1505–33) and Ivan IV Vasil’evič (1533–84).8
192 R V
Only a single case of tagged by tak(o) rek(li) was found (AJu 31, no. 16).
It is possible that the two instances of uncomplementized tagged by
s”kazati, which occur in separate transcripts of the same lawsuit of 1498/99
(ASÈI 2: 453, no. 418; 455–56, no. 419), represent an intermediate stage in
the proposed drift or a compromise between the two strategies. Perhaps a
further stage can be seen in an early sixteenth-century transcript from
Muscovite-annexed Novgorod, where the verification report is tagged by
s”kazati and complementized by č’to (Koreckij 1969: 290, no. 2). In any
event, there is no evidence for any unusual contextual factors to distinguish
these examples from the better-represented cases presented as .
6.3 Falsifications
The remarks of the higher judges are generally treated like those of the on-
site judges in the trial record. When they elicit testimony, they are presented
as after a verb of asking; when they relate to trial procedure (“stage
managing”), they are presented as grammatically integrated speech after
velěti, that is, as part of the third-person narrative (cf. the nondialogic
commands to clerks and other court personnel in ASÈI 1: 493, no. 595; 509,
no. 607; 514, no. 607a; 3: 186, no. 172; 269, no. 250). Procedural statements
by other participants likewise tend to be integrated into the narrative; thus,
when the litigants agree to call on the men of court for corroboration, their
statements are given as ’s, as in (9).11
The same distinction between narrated procedure and reported testimony
appears in two examples in which bailiffs presenting the men of court make
preliminary reports before the judges. In both these cases, the litigants’
separate responses in the verification procedure, which are presented as ,
are followed by narratives of action, including ’s (ASÈI 1: 493, no. 595;
3: 186, no. 172). In the first instance (10), the bailiff’s report begins as a
continuation of the narrative:
(10) I pered [knjazem Vasil’em Ivanovičem [nedelščik Vaska
and before [prince Vasilij Ivanovič]- [bailiff Vaska
Karačev [sudnyx mužej… postavil; a [sud’ja Ivan
Karačev]- [of.court men]-. place- and [judge Ivan
Kuzmin da [starec Efrem stali že. A
Kuz’min]- and [elder Efrem]- stand-. and
[otvetčika Stepanka Dorogu, — skazal [nedelščik Vaska
[respondent Stepanko Doroga]- say- [bailiff Vaska
Karačov, — dal na poruku, a vyručil ego,
Karačev]- give- to bail- and bail- him-
gospodine, u menja [Ofremko Bulgak Jakušov syn, da
lord- at me- [Ofremko Bulgak Jakuš- son]- and
[Oleško Dmitrov syn Bobrovnikov; i tot, gospodine,
[Oleško Dmitr- son Bobrovnikov]- and that- lord-
Stepanko zbežal [s] [svoimi poručniki.
Stepanko- run- [with [. bondsmen]-
[1] And the bailiff Vaska Karačev placed the men of court… before
Prince Vasilij Ivanovič; and the [trial] judge Ivan Kuz’min and the
R J-R H 195
monk Efrem [the plaintiff] also appeared. [2] But, said the bailiff
Vaska Karačev, [he] set bail for the respondent Stepanko Doroga,
“and, lord, Ofremko Bulgak, Jakuš’s son, promised me to be his sure-
ty, as did Oleško Dmitr’s son Bobrovnikov; and that Stepanko, lord,
ran away with his sureties” (ASÈI 1: 493, no. 595).
In accordance with normal expectations, the passage marked [2] should be a
continuation of the narrative [1] affirming that the respondent — the party
who rejected the trial transcript — is present; note the fronting (topicaliz-
ation) of the corresponding noun phrase otvetčika Stepanka Dorogu. Instead,
[2] initiates the bailiff’s report of how the respondent has absconded. As the
events described are not observables in the ongoing speech event, the scribe
has to disclaim responsibility for this information; however, he demarcates
it in a way that minimizes the transition from the narrative, by intercalating
the preterit of the s”kazati — the usual tag for the of officials other
than the judges — right before the point at which he identifies the subject,
who is also the information source.
As an interruption, intercalation is a strategy that delays the complete
reception of heteroglossic information for various pragmatic ends (see
Shapiro 1984: 73–76; Leech and Short 1981: 333, 351 n. 11; Page 1988: 27).
In (10), it allows the reported information to be perceived as a continuation
of the narrative by postponing the attribution. This effect of seamless
transition is furthered by the absence of elements exclusively dependent on
the reporter’s perspective in the underlined clause. While the strategy is
ambiguous here, the subsequent clauses are clear-cut , with a first-person
pronoun and vocatives oriented on the bailiff’s viewpoint. This information
is pertinent to the verdict, since flight constituted an admission of guilt (see
Kleimola 1975: 89–90), and the sureties were financially culpable for their
actions. Hence serves as a foregrounding device, and the report itself
counts as testimony. By contrast, the underlined quasi-narrative report is to
some extent presupposed, since it describes a standard procedure; as back-
ground for the more immediately relevant information that follows, it need
not be conveyed in a salient manner.
A second bailiff’s report (ASÈI 3: 186, no. 172) follows much the same
pattern.12 Presented as a continuation of the narrative, it is attributed by an
intercalated , just as in (10). There is no fade-in to . Presumably was
not optimal here, since the report is limited to background information
196 R V
without any bearing on the verdict (accounting for the absence of one of the
men of court, who has died in the interval between hearings).
The of the men of court can be presented like standard testimony
(ASÈI 1: 494, no. 595) or like typical corroborations, with s”kazali plus
(ibid. 3: 186, no. 172). In the case with (in the passage following (10)),
the statement goes beyond the restricted content of the verification formula
— a reference to the absconding litigant. While this information has already
been established in the context, it zeroes in on the truly crucial point in the
judgment and thus merits foregrounding. By contrast, the statement in is
restricted to the verification formula. The same pattern may be observed in
a case in which a litigant returns to recant his falsification (11):
(11) Po [semu spisku, stav pered [Mikitoju Vasil(’)evičem
by [this copy]- stand- before [Mikita Vasil’evič
Beklemišovym, … Andrěiko skazal, čto sud byl
Beklemišov]- Andrejko- say- that trial- be-
ne takov, kak pisano v [sem spisku. Da sšed s
such- as write- in [this copy]- and leave- from
suda, opjat(’) stav pered Mikitoju, skazal, čto
court- again stand- before Mikita- say- that
sud byl takov, kak pisano v [sem spisku
trial- be- such- as write- in [this copy]-
In accordance with this copy, having appeared before Mikita
Vasil’evič Beklemišov, Andrejko said that he did not have a trial such
as [is] written in this copy. And, after leaving the court, he appeared
again before Mikita [and] said that the trial was such as [is] written in
this copy (ASÈI 1: 400, no. 523).
The underlined report is simply a positive restatement of the verification
formula. It differs from the other recantations (ibid.: 509, no. 607; 514, no.
607a; Kaštanov 1970: 359, no. 9) and from the unrepentant retraction in (9),
which are in clear-cut , in that it does not include any new information
such as expressions of guilt or resignation. The largely presupposed nature
of the report may explain why it does not contains any of the features of
that clash with the narrative perspective — that is, any of the elements that
could actualize and foreground information perceived as probative and
prominent. In the absence of personal pronouns, there is no way to determine
R J-R H 197
In several transcripts, the trial judges are depicted making a statement at the
beginning of the higher-court hearing, before the verification procedure. Here
the choice of strategies is influenced by various factors, including the given-
ness and diffuseness of the reported information. In some cases, the report
is said to be an account of the trial; it is then not depicted but merely referred
to by means of a (12).13 Presumably the message did not have to be
recounted because it was more or less equivalent to the written transcript.
198 R V
Layered reports
given reports took place, it would be risky, at best, to attribute their form,
including the choice of reporting strategy, to the represented reporter, whose
statements, after all, are only known through the filter of the writer’s
language. In this sense, despite their distinct epistemological status, the
additional layers of in trial transcripts present methodological problems
similar to the ones posed by the in fiction, where “there is no direct
‘original’ prior to or behind an instance of [] or []” and where “the
supposedly ‘derived’ utterances are not versions of anything, but themselves
the ‘originals’ in that they give as much as the reader will ever learn of
‘what was really said’” (McHale 1978: 256; see also Chatman 1975: 221–22).
The layered reports found within the trial dialogue are attested in many
of the same strategies as the questions and answers attributed in the authorial
narrative. However, the distribution of the strategies is quite different —
indeed, almost in an inverse proportion. This undoubtedly reflects the
contextual diversity and distinct function of the additional layers of speech,
as compared with the quotations in the narrative.
Because speech within speech can convey a spectrum of verbal interac-
tions limited only by the inventory of speech genres in medieval Russian
society, it represents a much wider range of types than the testimony that
contains it. Even though the scribes provided minimal background and
avoided evaluations in their exposition of the trial hearing, they depicted the
participants speaking “for themselves”, without any constraint on what they
could say — presumably as a reflection of reality. Since representations of
the participants’ statements can contain not only complicating action clauses
but also extensive orientation, reports of current states of affairs, and
evaluations of events (Labov 1972: 363–65; Grimes 1975: 51–64), it follows
that third-person narrative is not the sole or even principal context in which
additional layers of can be found. First-person narrative, projections in the
future, commands, and other frames are also attested.
The functional differences between the record of the testimony and
additional layers of are manifest. As already observed, the trial dialogue
is intended to provide the judges with ample and unbiased information, in
order to facilitate their decisions; thus it is generally presented in as neutral
and comprehensive a form as possible. By contrast, the chief function of the
within is rhetorical; it is often intended to persuade — to advance or
discredit one side of an argument. Unlike the authorial narrative, it always
takes a definite, personal stance towards the events that it describes. This
L R 205
Although is by far the most common strategy for reporting the partici-
pants’ statements in the trial dialogue, it is rarely used with additional layers
of within those statements. Only twelve clear examples are attested (two
in variant transcripts from the same trial), as against more than sixty cases of
indirect or nondirect speech. Whatever other factors were present, it can be
assumed that the need for clarity militated against additional layers of ,
which in scriptio continua could be misinterpreted as the onsets of new turns
at speaking in the trial dialogue. This was especially true when the layered
reports were tagged with tak(o) rek(li) and contained the vocative
g(o)s(podi)ne. Another disincentive was the difficulty in keeping track of
attributions and deictic pivots when multiple perspectives are present; this
undoubtedly accounts for the total absence of additional layers of , a
strategy largely dependent on predictability of attribution.
The layered direct reports share several traits with the direct reports in
the trial dialogue. For example, is privileged with reports in narrative
sequences, especially when the represented is “hyperforegrounded” as a
narrative peak or a crucial point in a line of argumentation. Thus is
preferred for the responses in dyadic or quasi-dyadic exchanges, and is
sometimes also used for the initiating ’s in such dyads. Dyadic responses
function, of course, as the peaks of minimal narratives.
All these factors can be seen in the two cases in which layered reports
are presented in accordance with the conventions for testimony. Both
examples occur within narrative accounts of legal or quasi-legal proceedings;
they are presented as responses in dyadic exchanges with third-person
speakers. In general, the configuration of the participants in the represented
speech events resembles that of the participants in a trial hearing. In the first
case, the layered report is a rebuff to a warning that the represented report-
ers officially witness (1). The onset of the dyad is presented as tagged by
the izvěčati ‘warn, notify (of misconduct)’ (on this , see Zaliznjak
1986a: 177):
206 R V
been committed when “the community either failed to identify the murderer
or refused to yield him once discovered”; Kaiser 1980: 68.) This report,
which is tagged like testimony, has evidentiary value, because it shows that
potential witnesses for the respondents once made an official deposition that
the contested land was the property of the plaintiff.
In both examples of layered tagged by tak(o) rek(li), the reporters are
bystanders rather than active participants in the exchanges that they depict.
In (1), the witnesses were only overhearers, not the addressees of the retort
addressed to their leader Oksenko, while the reporters in the second case make
no mention of their own role in the interrogation. Thus the reporters have a
low degree of involvement; their status in the represented exchanges resem-
bles the recording scribe’s detached, observational role in the trial hearing.
The difference between such detached, disinterested reports and more
involved accounts emerges with clarity when the testimony of Oksenko’s
witnesses in (1) is compared with Oksenko’s own version of the same
exchange (2), with a mixture of first- and third-person reports. As depicted
in the trial transcript, Oksenko is the transmitter of the warning (and hence,
from his own viewpoint, the representative of the law); thus he is highly
involved both in the events recounted in his testimony and, as plaintiff, in
the trial itself. Unlike his witnesses, Oksenko was not in a position to
observe impartially his own actions or those of his opponents; as his goal is
to persuade the judge, he has no reason to report statements with any degree
of objectivity.
(2) I jaz im, gospodine, izvečjal ot poselskogo,
and I- them- lord- warn- from steward-
čtob iz [manastyrskogo lesu iz Dubrovinskogo
that- from [of.monastery forest]- from of.Dubrovino-
[sečenyx drov ne vozili, a bole togo by ne
[cut- wood]-. cart-. and more that-
sekli; ini, gospodine, skazyvajut to
cut-. others-. lord- say-.3.
les ne troickoj ne monastyrskoj, a
forest- of.Trinity- of.monastery- and
zovut [tot les monastyrskoj Dubrovinskoj
call-.3. [that forest of.monastery of.Dubrovino]-
208 R V
double duty, not only actualizing crucial evidence but also disassociating the
reporter from responsibility (Clark and Gerrig 1990: 792; Sternberg 1982b).
A third evaluative device in (2) is the use of the m”lviti (molviti) [4],
which also tags seven other cases of layered in the transcripts. Like reči,
this verb tends to appear in dyadic or at least reactive contexts, but it seems
to have a special emotive force. Often, as in (2) [4], statements tagged by
m”lviti do not resolve the dyad in a way expected by or acceptable to the
reporters (cf. Longacre 1983: 73); they are uninformative, otherwise uncoop-
erative, or unsuccessful. This connotation emerges clearly in some of the
other examples of layered . In one case, the represented reporters had
petitioned the steward and received his promise of a deed for the property
under contention — a promise that he did not live to fulfill. While the
petition is given as tagged by a , the steward’s promise is given as
introduced by m”lviti (ASÈI 3: 187, no. 173). In three other cases, litigants
are depicted defending themselves against the charge of having “kept silent”
— being negligent in prosecuting their claim — by telling how the grand
prince responded to repeated petitions with promises that have come to
nothing. The grand prince’s statements, conveyed as tagged by m”lviti,
have failed to have their expected perlocutionary effect (3).1
(3) Bivali esmja, gospodine, čelom [velikomu
hit-. ..1. lord- forehead- [grand
knjazju ne odinova, i [knjaz’ velikij tak molvil:
prince]- once and [prince grand]- thus say-
Kak poedet [moj pisec Pereslavlja pisati, i
when come-.3 [my clerk]- Pereslavl’- write- and
vy emu ukažite [moi zemli, i
you-. him- show-.2. [my lands]- and
on mne skažet, — i pisec v Pereslavle
he- me- tell-.3 and clerk- in Pereslavl’-
davno ne byval.
long be-
[1] “We petitioned, lord, the grand prince more than once, [2] and the
grand prince spoke in this way: ‘When my clerk comes to write up
Pereslavl’, you can show him my lands, and he will tell me’; and the
clerk hasn’t been in Pereslavl’ in a long time” (ASÈI 1: 452, no. 571).
210 R V
for , which puts the interpreter in control as a witness able to draw
personal conclusions about the statement thus depicted. While this “direct-
experience” effect (Clark and Gerrig 1990: 793) could potentially create
empathy, this risk could be offset by the use of evaluative introducers like
m”lviti and by a careful selection of what to report from direct perspective.
The zeroing-in, foregrounding effect of could conspire with presupposed
negative evaluations to create distancing rather than empathy, holding, as it
were, the given under microscopic scrutiny.
The distribution and foregrounding effect of additional layers of can
be readily compared with the usage in the modern American trial transcript
studied by Philips (1985). According to Philips (ibid.: 154), (“quoting”)
signals a high degree of relevance — “information being presented as evi-
dence directly related to proof of the elements of a criminal charge” (see also
162, 168). By contrast, background information and other parts of the
testimony that are perceived as less relevant (“central”) to the outcome are
presented transcript as ’s (“topic naming”) or (“reports of substance”;
ibid.: 160–62).
The most common way of conveying reports within the trial participants’
statements is by the use of complementized indirect or nondirect speech
tagged by forms of the s”kazati or its imperfective partner, s”kazyvati or,
in a few instances, by other verbs. There are 92 examples of this strategy in
the trial transcripts investigated (or 93, given an ambiguous example). In
most cases, the complementizer is the explicative conjunction č’to; the
synonymous conjunction koe is also attested in a single transcript, where it
is the only complementizer used (ASÈI 1: 419, no. 540).
The contextualizations in which complementized is preferred fall into
several speech genres. Twelve examples occur as preambles to judges’
questions and function as requests for amplification of the immediately
preceding remarks or previous testimony. In such cases, the queries are
intended to clarify the content of the , to resolve apparent contradictions,
or to pursue lines of inquiry stemming from the reports. Six of the examples
are tagged with s”kazyvati in the present tense (5).
L R 213
is attested in five examples, resembles requests for amplification like (5) and
had the same basic motivations in the actual trials.5
(6) [Vaši starožilcy skazyvajut, čto [te
[your- old.residents]-. say-.3. that [those
zemli paxali na monastyr’ let s
lands]-. plow-. for monastery- years-. about
šestdesjat’, i vy o čem monastyrju za
sixty- and you-. for what- monastery- for
tolko let molčali, i [velikomu
so.many- years-. be.silent-. and [grand
knjazju este o [tex zemljax
prince]- ..2. about [those lands]-.
bivali li čelom?
hit-. forehead-
“Your witnesses say that they farmed those properties for the monas-
tery for some sixty years. Why did you go so many years without
complaining to the monastery? And have you petitioned the grand
prince about those properties?” (ASÈI 1: 540, no. 628).
If reports of this kind are compared with their antecedents in the trial
record, they turn out to be far from mechanical transmissions. The messages
are greatly condensed, with omission of perspectival information like the
vocative g(o)s(podi)ne; in addition, they show the inevitable changes of
meaning that result from recontextualization. This is epitomized by (6), in
which the judge, as depicted in the transcript, subverts the illocutionary
intent of the testimony; the witnesses had evidently wanted to inculpate the
opposing side’s supporters but unwittingly unmasked the nonfeasance of
their own principals.
Subversion, indeed, is the function of fourteen other cases that challenge
the statements of the opposing party. In these examples, the reports are
adduced exclusively as material to be refuted. The verbs in the tag clauses
are mostly in the present tense, as with the requests for amplification and for
reaction discussed above.6
(7) A čto, g(o)s(podi)ne, skazyvaet [černec’ Kirilo, čto
and that lord- say-.3 [monk Kirilo]- that
L R 215
for judgment [4]. The content of the false testimony, while relevant to the
charge, is presupposed and presented as a matter of lesser concern through
the use of devices that integrate it into the narrative — complementizers and
. Isaja has shown his enemies to be liars and has no need to dispute their
words as such.
In other cases, the layered reports tagged by the perfective preterit, which
convey statements made in earlier legal procedures or confrontations, are
situated in narratives (fourteen instances in four transcripts).8 These indirect
reports differ from direct ones tagged by tak(o) rek(li) or m”lviti (see
Section 7.2) in that they are nondyadic, and there is no special focus on the
reported information per se. The narrative in which the ’s occur is (unlike
the authorial framework) not primarily a sequence of ’s. The layered
reports serve as background; with the surrounding narrative, they set the
stage for that part of the testimony directly relevant to the case at hand. For
example, in one of the cases, a witness for the respondents is recounting the
results of an obysk, an investigation conducted by collecting depositions from
reputable members of the community (9); however, the underlined report is
not the ultimate point of his statement but rather an explanatory preamble
whose purpose is to bring out the full significance of the more recent
conflicts that prompted the lawsuit.9
(9) vylgal sebě u Timofěja u Mixailovičja
get.falsely- . at Timofej- at Mixajlovič-
[gramotu lgotnuju na [tu rozsěč(’), a skazal,
[writ of.privilege]- for [that clearing]- and say-
g(ospodi)ne, to lěs [velikog(o) kn(ja)zja… I my,
lord- forest- [grand prince]- and we-
g(ospodi)ne, bili čelom [Timofěju Mixailovičju;
lord- hit-. forehead- [Timofej Mixajlovič]-
i Timofěi… obyskal [ljudmi dobrymi [vseju
and Timofej- seek.out- [people good]-. [all
volostiju, čto, g(ospodi)ne, Onikei da Senka skazali
region]- that lord- Onikej- and Senka- say-
Timofěju, čto [tot Savka popaxal [našu zemlju
Timofej- that [that Savka]- plow- [our land
monastyrskuju kočevinskuju; i Timofěi, g(ospodi)ne,
of.monastery Kočevinskaja]- and Timofej- lord-
218 R V
would retard the pace of the narration by singling out the statement per se
for special attention. Such a ritardando effect would only be well motivated
if the reported information itself was a special focus of the discourse.11
The preterit perfective is also used to tag a further case of layered
complementized conveyed with situational specifications (10). This report,
which appears in an early-sixteenth-century larceny trial, forms the preamble
to a question following up on testimony that the defendant is a known
criminal. In Muscovite law, the fact of a prior conviction created a presuppo-
sition of guilt (see Kaiser 1980: 91, 139). Like the testimony in (8), the
judge’s inquiry, as recorded in the transcript, has the illocutionary intent of
exposing a contradiction and does not expect an exonerating reply:
(10) Pered [knjazem Ondreem Vasil’evičem v [tom dele
before [prince Andrej Vasil’evič]- in [that matter]-
sud tebe byl li i bit li
trial- you-. be- and beat-
esi knutom v [toj tat’be so [knjaž-
..2 whip- in [that theft]- from [prince-
Ondreeva suda? I peredo mnoju esi nyneča
Andrej- trial]- and before me- ..2 now
skazal, čto [tem senom tebja podkinuli, a koli
say- that [that hay]- you-. frame-. and when
u tobja [to seno vynjali, i ty ego
at you- [that hay]- take.away-. and you-. it-
skazyval svoim, a skazyval esi, čto ego
say- .- and say- ..2 that it-
kupil, i u kogo že ty paki [to seno
buy- and at whom- you-. [that hay]-
kupil?
buy-
“[1] Were you tried in that matter before Prince Andrej Vasil’evič?
And were you flogged for that theft after Prince Andrej’s judgment? [2]
And you have now said before me that you were unjustly convicted in
[the theft of] that hay; [3] but when they took that hay away from you,
[3a] you called it your own and [3b] said that [you] bought it. [4] From
whom was it that you bought that hay?” (Kaštanov 1970: 413, no. 40).
220 R V
Here, as in the layered report in (8), the underlined tag clause includes a
topicalized reference to the judge’s role in the hearing (“before me”). This
implicitly calls attention to the solemn, binding nature of the act of testify-
ing, which the defendant Gridja has violated. The juxtaposition of Gridja’s
past offense [1] with the present charge against him [2] (cf. nyneča ‘now’)
reflects a presupposition of his guilt. This is made explicit in the statements
following the underlined segment [3], in which hearsay ascribed to Gridja by
hostile witnesses and denied by Gridja himself is taken as given, as back-
ground for a question. The question itself [4] contains the emphatic particles
že and paky, which, by signalling repetition, call attention to the fact that a
satisfactory answer has yet to be given.
The two hearsay reports in (10) — the first conveyed as a [3a] and
the second as [3b] — actually represent a third layer of set within ,
since they are themselves heteroglossic. It is noteworthy than the hearsay is
tagged with yet another form — the preterit of the imperfective
s”kazyvati, which indexes a different kind of speech event than the present
or preterit perfective. This way of indicating hearsay is found in another
instance in which a judge is asking his addressees to confirm that they did
say what preceding speakers had alleged of them (AFZX 116, no. 125). The
statements tagged by the imperfective preterit are repeated from the testimo-
ny; as presupposed reports, it is not surprising to find them in complemen-
tized . The use of the preterit also has a straightforward explanation. The
putative original statements are supposed to have been made before the trial
hearing and hence outside of the judge’s (and scribe’s) observation; the
messages that they convey have not yet been acknowledged by the speakers
to which they have been ascribed, so that the attribution has not become a
given, activated as “present” testimony like the statements repeated from the
trial hearing in (5)–(7). As for the choice of aspect, it is typical in the
transcripts for imperfective preterits to be employed in the judges’ questions
with untagged heteroglossic information that is being offered for verification
(11). The confirmations themselves can be perfective if that is contextually
appropriate.
(11) I sud(’)i sprosili Danila: Davyval li
and judges-. ask-. Danilo- give-
ty… pustoš(’)… I Danilo tak rek: Dal
you- field]- and Danilo- thus speak- give-
L R 221
esmi… pustoš(’)…
..1 field-
And the judges asked Danilo, “Did you donate []… the
field…?” And Danilo spoke in this way: “I gave []… the field…”
(ASÈI 2: 316, no. 333).
This quasi-evidential use of aspect can be explained by the fact that the
actions denoted by the perfective are situated in a definite network of
contextual relations, either explicitly, as in (8)–(10), or in the implicit view
of the reporter. By contrast, those denoted by the imperfective are treated in
a decontextualized manner, and the primary issue under discussion is the
occurrence of the actions. (Cf. the case in which the judge adduces an
invalidated claim of the plaintiffs as the chief reason for the forthcoming
negative judgment; ASÈI 2: 389–90, no. 387.)
Another category of hearsay involves statements that litigants or, more
frequently, witnesses ascribe to their fathers or grandfathers. These are most
often presented (14 out of 16 examples) as complementized indirect or
nondirect speech tagged, like other hearsay, by the preterit of s”kazyvati or,
in two cases in a single transcript, by slyxati ‘hear’ (AFZX 214, no. 249
(2×)). (Verbs of hearing can act as the converses or relational opposites (see
Palmer 1981: 97–100) of ’s, with their focus on the addressees rather than
the speakers.) The choice of the imperfective for ancestral hearsay is
determined by the fact that the statements are said to have occurred in the
unsituated past; the imperfective thus functions as constatation of the ,
without specifying whether the action was iterative or punctiliar. The focus,
indeed, is on the source of the witnesses’ knowledge — i.e., on the repre-
sented speaker rather than on the .12 The tag clauses involving s”kazyvati
all include first-person forms — either dative pronouns or possessive
adjectives — referring to the reporters, who depict themselves as the
addressees of the layered report.13
(12) a. My, g(o)s(podi)ne, slyxali u [otcev”
we- lord- hear-. at [fathers
svoix; [otci naši skazyvali, čto to
.]-. [fathers our]-. say-. that
zemlja [velikogo kn(ja)zja…
land- [grand prince]-
222 R V
“We, lord, heard from our fathers; our fathers said that it [was]
the property of the grand prince” (ASÈI 2: 425, no. 407).
The witnesses in medieval Russian trials generally testified on what they
“remembered” or “knew” (cf. the term znaxar’ ‘one who knows’); the
authority of their statements was thought to increase in proportion with their
longevity (cf. the term starožil’c’ ‘longtime resident’; see also Kaiser
1980: 128). This was undoubtedly the social motivation for the ancestral
hearsay exemplified in (12a), which added, in effect, several decades to the
witnesses’ memory. By reporting a speech event in which they were partici-
pants (addressees), the witnesses were actually asserting something about
their own knowledge (cf. Townsend and Bever 1977: 5), as can be seen from
the dual attribution in (12b):
(12) b. a skazyvali nam, gospodine, [otci naši da
and say-. us- lord- [fathers our]-. and
i my zapomnim, čto izstariny [to
we- recall-.1. that of.old [that
Popkovo tjanet k Biserovu…
Popkovo]- belong-.3 to Biserovo-
“And our fathers, lord, told us, and also we remember that of old
that Popkovo belongs to Biserovo” (AFZX 107, no. 114).
The primary purpose of the witnesses’ testimony, whether it invoked only
their own recollections or ancestral hearsay, was to state their own ability to
corroborate the claims of the litigants who called them. In keeping with this,
the chief element foregrounded in their testimony is not the content of their
memory nor (in ancestral hearsay) the actual message of the layered speech,
which tends to be presupposed; it is the attribution — their assertion that
their knowledge, including what their forbearers told them, bears out their
principal. The use of complementized in such contexts is well motivated:
the report is presented as ground rather than a more salient (event-like)
figure that could vie with the more pertinent information, the attribution, for
the attention of the interpreters (see Talmy 1978: 640; Wallace
1979: 214–15). The emphasis on the attribution also explains the inverted
word order of the tag clauses in (12b) and most of the other examples of
ancestral hearsay: the represented source of the message is asserted in the
L R 223
normal position for new information. (In (12a) the represented source has
already been topicalized.)
This analysis is confirmed if the cases of ancestral hearsay in comple-
mentized after the preterit of s”kazyvati are compared with the sole
instance of in this context, which is tagged by the historical present of
m”lviti (12c).
(12) c. a [otec moj, gospodine, togdy žil za
and [father my]- lord- then live- under
mitropolitom v sele v Kulikovskom, i jaz,
metropolitan- in village- in Kulikovskoe- and I-
gospodine, s [otcem svoim i kosil na
lord- with [father .]- mow- on
[tex luzex, na kotoryx stoim, i
[those meadows]-. on -. stand-.1. and
[otec moj… mne molvit tak — nynečja my
[father my]- me- say-.3 thus now we-
[te luzi kosim za
[those meadows]- mow-.1. as
mitropoliči, potomu čto ešče ne sel nixto
of.metropolitan-. because yet settle- no one-
na Parašine, a kak na Parašine sjadut
on Parašino-, and when on Parašino- settle-.3.
ljudi žiti, i oni to u nas u
people- live- and they- at us- at
mitropoličix [te luzi
of.metropolitan-. [those meadows]-
ot”imut, zanež to zemlja [velikogo
take.away-.3. because land- [grand
knjazja Parašinskaja [te luzi…
prince]- of.Parašino- [those meadows]-.
“… And my father, lord, then lived under the metropolitan in the
village of Kulikovskoe, and I, lord, with my father mowed [hay] on
those meadows on which we stand, and my father… spoke to me
in this way: ‘Now we mow these meadows as the metropolitan’s,
224 R V
because no one has yet settled in Parašino, but when people come
to live in Parašino, they will take those meadows away from us,
from the metropolitan’s [people], because those meadows are the
property of the grand prince, of Parašino’”
(AFZX 230–31, no. 259).
There are at least three factors that may have influenced the unusual report-
ing strategy in this example. First, like the other cases of layered and
unlike the other cases of ancestral hearsay, the report is embedded in a
narrative. Second, the message of the layered report is unexpected. For one
thing, it provides far more information than is usual in ancestral hearsay; for
another, it is not a straightforward, presupposed corroboration of the princi-
pal’s position. The represented reporter is testifying on behalf of a servant of
the grand prince in a lawsuit against the metropolitan’s steward, yet he is
depicted testifying that he and his father were servants of the metropolitan at
the time of the narrated speech event. Therefore, the underlined report, which
ascribes the disputed property to the grand prince, is in effect the admission
of a party opponent, unique in the investigated transcripts. This fact, which
takes on a devastating significance in the verdict, distinguishes the report
from the other instances of ancestral hearsay. The unexpected nature of the
additional layer of makes it contextually salient and promotes foreground-
ing, a function usually performed by . This foregrounding effect, which is
a form of evaluation, may also have motivated the use of the historical
present in the tag and the choice of the emotively charged m”lviti (see
Section 7.2), which may, indeed, have its “frustrative” nuance here.
The third probable factor in the choice of in (12c) is the diffuseness
and complexity of the layered report. The cases of ancestral hearsay in
complementized (12a–b) involve relatively simple clauses with a non-
coreferential third-person subject. By contrast, the layered report in (12c) is
relatively lengthy and more hypotactic than is typical for the language of
trial transcripts. The use of a complementizer would lead to multiple embed-
dings, while a switch to the narrator’s indirect orientation would have caused
referential complications; for example, the phrase i oni to u nas u mitropoličix
te luzi ot”imut, which includes a coreferential inclusive first-person plural
pronoun, would have two competing third-person plural referents.
The final category of complementized report that should be noted is hypo-
thetical , which falls into two subtypes. The first is projected statements;
L R 225
these include first-person utterances with the ’s cělovati kr’st” ‘swear [lit.
kiss the cross]’ and s”latisja ‘cite, refer to’ (13). The latter denotes ’s in
which litigants simultaneously identify witnesses and predict how they will
testify or in which they make reference to documentary evidence to which
they have no immediate access.14 The present tense is used because the ’s
have performative force; they are claims that are pertinent to the ongoing
trial, like rejoinders.
(13) Vedomo to, gospodine, našie že volosti
be.known- that- lord- our- region-
xristianom [Jakušu Tokarevu… da [Bornjaku Ontuševu,
peasants-. [Jakuš Tokarev]- and [Bornjak Ontušev]-
na nix sja, gospodine, i šlem, čto to
on them- lord- cite-.1. that
zemlja [velikogo knjazja.
land- [grand prince]-
“That is known, lord, to peasants of our region, Jakuš Tokarev… and
Bornjak Ontušev; I cite them, lord, that that is the grand prince’s
land” (AFZX 223, no. 258).
The imperative of the verb s”prositi ‘ask’ introduces a hypothetical report
complementized by č’to in a similar situation, in which the represented
speaker asks the judge to ask witnesses about his assertion (ASÈI 1: 492, no.
595). This is essentially the same illocution as in (13). The verb of asking is
used factively, as it were: “Ask the hundredman and peasants, (they will say
that) that village Kozlovo is the monastery’s property of old”.
The second subcategory of hypothetical occurs only in questions and
is tagged with preterit verbs. Here the judges are trying to obtain further
information; however, unlike in requests for amplification, they are referring
not to statements made in the course of the trial but to ’s that the address-
ees could be expected to have performed before the trial (hence the preterit)
in accordance with normative procedure (14a).
(14) a. Bivali li este čelom o [toi
beat-. ..2. forehead- about [that
zemlě [g(o)s(u)d(a)rju velikomu kn(ja)zju, čto u
land]- [sovereign grand prince]- that at
226 R V
While it is far more common for the additional layers of indirect or nondirect
speech to be explicitly subordinated, there are 17 cases in the corpus in
which such reports follow their introductory clauses without any intervening
complementizer. The tag verbs represented are the ’s s”kazati/s”kazyvati
and kazati, the z”vati ‘call, claim (as/that)’, and the verb slyxati ‘hear’,
which is the semantic converse of s”kazyvati. The uncomplementized reports
occur in some of the same contextualizations as complementized ones. Nine
of the examples appear in the judges’ questions as requests for amplification
or reaction.16 Five others come in testimony as subversive requotations of
statements made (or allegedly made) by the represented reporters’ oppo-
nents.17 Like their complementized counterparts, the uncomplementized
reports in these contextualizations are tagged with the present imperfective,
which indicates that the content of the requotation is topical — a claim that
is still being made by one of the parties in the litigation. This also accounts
for the present imperfective tag in an example in which a litigant ascribes his
L R 229
own position to the witnesses that he has just named (AFZX 227, no. 259).
There is also a case of ancestral hearsay tagged by the hearer-oriented verb
slyxati (ibid.: 106, no. 114) in the preterit imperfective, as is expected in this
contextualization, since it denotes a speech event in the unsituated past. The
remaining case is tagged by the preterit of the perfective s”kazati; like
other additional layers of tagged in that way, the report appears in a
narrative context and depicts a statement situated in an earlier legal proce-
dure (ASÈI 2: 194, no. 285).
These overlaps in contextualization suggest that the absence of the
complementizer does not index any distinctions in meaning (at least, as far
as can be determined). Presumably one of the discourse factors that promot-
ed it was pacing, since its general effect is to smooth transitions by eliminat-
ing a barrier between the narrative and that shares the narrative reference
points. It is noteworthy that four of the examples follow other reports
attributed to the same speakers — complementized (15a), or ’s (15b);
AFZX 222, no. 258; 260, no. 308). In all of these cases, there are null
subjects in the clauses introducing the uncomplementized reports, since their
referents have already been thematized; in addition, all of the tags are linked
to the preceding context by the non-sequencing conjunction a, which is
typical in multiplex testimony and other situations with subdivided speech
events (see 4.1, 4.3).
(15) a. A čto, g(o)s(podi)ne, skazyvaet [černec’ Kirilo, čto
and that lord- say-.3 [monk Kirilo]- that
[tě pustoši dal [Danilo Blin, a
[those fields]- give- [Danilo Blin]- and
skazyvaet sem(’) lět pustoši kosili,
say-.3 seven- years-. fields- mow-.
to, g(o)s(podi)ne, lžet Kirilo…
lord- lie-.3 Kirilo-
“… And, lord, as for the fact that the monk Kirilo says that
Danilo Blin gave those fallow fields and says [they] have used
those fields for seven years, Kirilo, lord, is lying…” (ASÈI 2:
315–16, no. 333; see also (7) in 7.3).
b. vylgal sebě u Timofěja u Mixailovičja
get.falsely- . at Timofej- at Mixajlovič-
230 R V
reports are followed by narratives of action with the same subjects (see 4.3).
By contrast, if the underlined passage in (23) is interpreted as speech within
speech, the subject of skazyvajut must be indefinite.)
As the examples show, fused was not a homogeneous phenomenon in
. While the notional subject of the report is consistently treated as direct
object, there are a number of different ways of realizing the underlying
predicate, up to and including its omission in syntheses of content, where the
topic is reported without any message (24).24
(24) Dvorskoj, gospodine, na [tom lese byl da ukazal,
steward- lord- in [that forest]- be- and show-
gospodine, nam mežu ot [Kijasovoj pustoši prjamo
lord- us- boundary- from [Kijas- field]- straight
lesom, a ne tudy, kudy Nikon skazyvaet.
forest- and there where Nikon- say-.3
“The steward, lord, was in that forest and showed us, lord, the bound-
ary — from Kijas’ fallow field straight through the forest, but not
where Nikon says” (ASÈI 1: 459, no. 581).
While the reduction of the report is not always as extreme as in (24) and
similar cases, fused is characteristic of reductive contextualizations where
there is no particular focus on the report as a whole (although individual
portions of the report can be put into emphatic positions). Like the use of
omission of the complementizer, the choice of fused is due, in part, to
pacing; such integration eliminates the transition out of and into the sur-
rounding non-reported discourse. In addition, certain common types of
sentences are evidently amenable to fused reporting when in isolation; for
example, there are five cases, including (22), indicating that certain parties
possess (or do not possess) charters giving them rights to the disputed proper-
ties.25 The use of fused in this contextualization is a general preference in
the text-kind; similar cases may be found in verdicts (see Chapter 8).
“I, lord, bid them to build farmhouses in those villages, and the Ivašov
Ugol peasants, lord, told me… that those villages Altynovo and Du-
brovka are lands of the grand prince, and belong, they say, to Ivašov
Ugol” (AFZX 116, no. 125).
(For a similar example, see ibid.: 181, no. 204.) In such cases, the intercalat-
ed tags, without implying the occurrence of a new , specify the informa-
tion source of an otherwise ambiguous segment of polysyndetic discourse
that is stranded from the tag in the outer context. This resembles the typical
use of the grammaticalized citation particle dě(i) in other text types.26 Such
explicitness is, of course, particularly necessary when reporters (actual or
represented) are citing claims they do not wish to endorse.
The inobtrusiveness of intercalated tags makes them suitable for convey-
ing heteroglossic information from previous testimony that is assumed for the
sake of argument, as is typical in requests for amplification and other
rejoinders (26). This function is particularly common (fourteen cases) when
the use of a preposed tag clause would lead to unusual syntactic complexity,
e.g., when the report is embedded in subordinated relative (26a), topicalizing,
or complement clauses (26b).
(26) a. Komu to vědomo, čto [tě derevni…
whom- that- be.known that [those villages]-.
[vaši zemli monastyr’skie [Vasil(’)evskie
[your- lands of.monastery]-. [Vasil’evskaja
slobodki, čto vam, skazyvaete, dal
slobodka]- you-. say-.2. give-
[Roman Ivanovič(’)?
[Roman Ivanovič]-
“To whom is it known that those villages… are your (the monas-
tery’s) properties pertaining to Vasil’evskaja Slobodka, which you,
you say, were given by Roman Ivanovič?” (ASÈI 2: 202, no. 288).
b. A vy celuete li krest na tom,
and you-. kiss-.2. cross- on that-
čto kažete [tot lug [velikogo knjazja
that say-.2. [that meadow]- [grand prince]-
[Dorofeevskoj zemli?
[Dorofeevskaja land]-
240 R V
“And do you swear that (you say) that meadow belongs to the
grand prince’s Dorofeevskaja property?” (AFZX 260, no. 308).
Example (26b) illustrates particularly well the difference between intercalat-
ed (parenthetical) and preposed (embedding) tags. Its placement is actually
ambiguous, since the complementizing conjunction č’to serves as the first
phonological word.27 However, the addressees are not being asked to swear
to the observable fact that they say the given message (“And do you swear
that you say that meadow belongs…”); rather, they are being asked to
uphold the content of the message, which is taken as a given for present
purposes but marked as reported lest it be mistaken for the position of the
represented speaker. Rephrasing examples like (26a–b) in a way that would
allow preposed tags would lead to further structural complexity or a degree
of diffuseness that would be problematic with presupposed information.28 It
is noteworthy that, in two cases, intercalated tags appear in noninitial clauses
in multiclausal complements (second clause, ibid. 212, no. 248; third clause,
ibid: 212–13, no. 248), where they also serve to promote cohesion under
conditions of syntactic complexity, like the example in (25).
When reports are presented in this way, the intercalated tags can have a
distancing effect similar to that produced by hedges. The reporters are using
the heteroglossic information for their own purposes without dwelling on it
per se or endorsing it as the truth; in some cases, they are even preparing to
undermine it in what follows. The noncommittal character of the strategy
explains its prevalence in the judge’s requests for amplification, where eight
cases are found, including those in (26). However, this quasi-evidential
function is contextual rather than inherent. Every report marker is a form of
disclaimer, given that reporters have the option of accepting heteroglossic
information as their own truth. Greater degrees of distancing can be ex-
pressed by using report markers “pleonastically”, where the fact of hetero-
glossia is evident, e.g., after the negative-presuppositional l”gati ‘lie’:
(27) tot, gospodine, [Ivaško Vnuk lžet, čto ot
that- lord- [Ivaško Vnuk]- lie-.3 that from
nego, skazyvaet davyval najmu pjat’ altyn…
him- say-.3 give- rent- five altyns-.
“That Ivaško Vnuk, lord, says falsely that for [that property], he says,
he paid five altyns in rent…” (AFZX 127, no. 140).
L R 241
The norms of reporting in the final major section, the verdict (found in the
judgment charter and referred trial transcript subtypes) differs strongly from
the patterns found elsewhere in the documents. In all of the verdicts exam-
ined, the dyadic question-and-answer structure found in the trial record is
abandoned for a concise third-person, past-tense narrative devoted exclusive-
ly to the actions — more precisely, the ’s — of the judges, who are the
topic of the discourse and the subject of most or all of the main clauses. The
verbs conveying the judges’ actions appear in the tense that is the narrative
norm — the preterit in Northeast Russian texts, the aorist in Pskovian and
Novgorodian ones. The third-person style of the verdict precludes the use of
appellative elements, even though the judges’ ’s had addressees in the
actual context that is represented.
As a rule, the verdict section begins with the conjunction i, which is used
elsewhere to link consecutive events in the polysyndetic chain of the narrative
framework. Thus the verdict is presented as the final event — the last word,
as it were — in the trial hearing. In cases of judicial referral, the verdict serves
as the conclusion of the judicial-referral record, which is usually joined to
the preceding context asyndetically, so that it forms a separate narrative from
the trial dialogue. In most of the trial transcripts that survive as autographs,
where there is sparing use of capital letters, the initial letter of the verdict
section is capitalized (boldfaced in (1)).1 This usage suggests that writers
viewed the verdict section as a discrete paragraph in the discourse:
(1) I [Kostantin Grigorievič(’) v [těx zemljax
and [Konstantin Grigor’evič]- in [those lands]-.
246 R V
with the old custom… except in spiritual cases, but the abbot judges
his brethren, the monks, himself… And the archbishop is not to send
his bailiffs to Cyril Monastery… (ASÈI 2: 281, no. 315).
In (4), taken from a lawsuit over ecclesiastical jurisdiction, the adjudication
of the verdict begins in an integrated, attributed form as the complement of
the prisuditi. It then slips into a sequence of unattributed clauses (only
part of which are cited here), with nonpast indicative verbs or unbound
infinitives conveying imperative modality (prescribed future behavior). The
meaning of the passage would be essentially unchanged if the first directive
were expressed as nondirect speech or if the later commands were presented,
like the first, as infinitive complements. However, the latter would create
multiple embeddings, which could be hard to process, especially given the
graphic format. Moreover, integration with past-tense ’s indicating
temporally limited events would be less effective than unbound present-tense
discourse for conveying directives meant to have ongoing validity.3
Directives of limited validity are, in fact, generally conveyed by ’s
involving the volitional verb velěti ‘bid’, which takes an infinitival comple-
ment with a dative for the recipient of the command (if explicit). (In prohibi-
tions, the is negated, even though the scope of the negation is the action
conveyed in the complement clause.) Thus the commands are presented in
the same way as procedural directives in the trial narrative (5).
(5) I [knjaz(’) velikii Vasilei Dmitreevič(’)… [těx zeml’
and [prince grand Vasilij Dmitrievič]- [those lands
c(e)rk(o)vnyx… ne velěl prodavat[i] ni kupiti nikomu,
of.church]-. bid- sell- nor buy- noone-
a kto imet prodavati ili kupiti ili arximandrita
and who- will-.3 sell- or buy- or abbot-
ne slušati, tě[x] velěl [veliki [knjaz’] [arximandritu
obey- those- bid- [grand prince]- [abbot
mixailov’skomu vo[n] metati.
of.Michael]- thence drive-
And Grand Prince Vasilij Dmitrievič… commanded that no one sell or
buy those church lands; and if anyone buys or sells or does not obey
the abbot, the grand prince commanded the abbot of St. Michael to
drive them out (ASÈI 3: 54, no. 31).
252 R V
The strategy of velěti with the dative plus infinitive maintains the narrator’s
perspective and allows the reported command to be condensed into its barest
essentials, while contextually trivial deictic information such as person, tense,
and mood is eliminated. However, in elaborated commands such as (5), in
which the complements of velěti have dependent clauses of their own, the
verbs in the subordinate clauses remain finite and retain the grammatical
marking of independent verbs. Examples of this kind demonstrate how bound
can slip into less integrated forms — in the given instance, nondirect
speech — in cases in which compact presentation can lead to syntactic
complexity and processing difficulties. (Slipping into is not the only
possible interpretation here. Semino, Short, and Culpeper (1997: 30) view
English constructions involving elaborated ’s as a special subcategory
— “narrator’s representation of speech act with topic” — that has “a greater
summarising effect than is normally associated with .”)
congruous with those used for within in the trial record. There was a
strong tendency to reduce such requotations; this was probably promoted by
the givenness (recoverability) of the message, which favored compact
presentation, and by the transition from a dyadic framework to the continu-
ous narrative format of the verdict. Much of the detail — in particular,
deictic elements — needed to convey the give-and-take of conversational
exchanges would be superfluous outside the dyadic framework, in speech
reported at second hand — indeed, at third hand, since the judges’ statements
themselves are converted and reduced from their original first-person
orientation.
In accordance with this reductive tendency, there is a strong preference
for nondirect and grammatically integrated forms of in the ratio decidendi.
The best-attested strategy is s”kazati ‘say ()’ followed by complemen-
tized indirect or nondirect speech (6). The only complementizer attested is
the all-purpose conjunction č’to ‘that’. This strategy is attested, sometimes
alongside other methods, in 34 examples in 25 documents.7
(6) I [knjaz(’) velikii Ivan Vasil(’)evič(’)… [Vaska Danilova
and [prince grand Ivan Vasil’evič]- [Vasko Danilov]-
velěl obiniti… potomu čto skazal Vasko, čto [sud
bid- find.wrong- because say- Vasko- that [trial
takov byl, da i to sam že skazal, čto
such]- be- and that- self- say- that
za sorok lět o [toi požně molčal, da
for forty- years-. about [that field]- be.silent- and
n(y)něča [tu požnju pokosil.
now [that field]- mow-
And Grand Prince Ivan Vasil’evič… bid [the trial judge] to decide
against Vasko Danilov… because [1] Vasko said that there had been
such a trial [2] and also said himself that he had been silent about that
field for forty years and had now mowed that field
(ASÈI 2: 384, no. 383).
While complementized is rarely used for testimony, it is the most common
strategy with within the testimony, where statements are filtered through
dominant subjective viewpoints.8 In addition, it occurs frequently with
verificational statements in judicial-referral reports — that is, with given,
R V 255
And Vasko spoke in this way: “… That field was Paša’s meadow,
and after Paša his son Filja mowed that meadow during the
plague; and that, lord, was more than 40 years ago… And after
them from that time… no one mowed that meadow (ibid.).
The statement that Vas’ko kept silent about his claim to the property for
more than forty years does not correspond directly to any of his remarks; the
author has read between the lines of the testimony in (7b). Nothing in the
record suggests that Vas’ko admitted such nonfeasance, though a statement
to that effect is tagged as his .
The requotation in (6) features omission of both the coreferential subject
pronoun and the vocative g(o)s(podi)ne, which is characteristic of the switch
from direct to nondirect speech. Since the vocative relates to the original
speech situation, it offers no information pertinent for interpreting the judge’s
reported decision; consequently, its omission — attested in all the cases of
complementized speech in the verdict — reduces the semantic complexity
caused by the shift in viewpoint, which does not carry any functional (infor-
mative or disambiguating) load in this context. The same is true of the
deletion of coreferential subject pronouns, which is common though not
obligatory in requotations in the ratio decidendi. Like other redundant
elements, such pronouns would facilitate interpretation of the testimony in
the trial record (as in the real-world dialogue that it reflects); there was no
need for this function in the verdict, inasmuch as the critical evaluation was
a fait accompli. The greater brevity and simplicity that resulted from these
deletions probably added to interpretability of the causal clause.
In addition to complementized reports, there are fourteen tokens of
uncomplementized indirect or nondirect speech, in twelve transcripts. There
is one case of the kazati, which, as usual, appears in the present tense, as
does the verb of knowing in the second clause of the rationale (ASÈI 3: 88,
no. 56). The remaining examples are all tagged by the preterit of s”kazati,
like the complementized reports discussed above. In most of the examples,
the report is the first or only item in the rationale; hence the tag clause
comes immediately after the causal conjunction.9 This suggests that distance
between the causal conjunction and the reported-speech clause plays some
role in the pragmatic choice between inclusion or omission of the comple-
mentizer; in other words, one of the motivations may have been to avoid
explicit multiple embeddings in close proximity.
R V 257
elements (to varying degrees) and to the use of grammatically bound forms
of or lexical evaluations of illocutions in place of clauses.
Syntactic factors and discourse cohesion also seem to exert some
influence. The occurrence of complementizing conjunctions is evidently
linked with the need to avoid ambiguous structures that could potentially be
misinterpreted in the continuous linear organization of the text. For example,
when there is some interrupting element such as a prepositional phrase in the
matrix clause right before the subordinate clause (e.g., po kr’st’nomu
cělovaniju ‘in accordance with a cross-kissing oath’; e.g., ASÈI 3: 291, no.
276), explicit complementizers seem to be preferred in order to promote the
proper segmentation of the text. Bolinger (1972: 38) notes a typologically
parallel use of that in English when the complement clause “is separated
from the main verb by intervening complements or interpolations… to
preserve the identity of the clause.”
Conversely, complementizers can be omitted when the predicate comes
first in the speech clause rather than the subject, because there is less
possibility of misassigning reported information — in particular, noun
phrases and prepositional phrases — to the matrix clause when the word
order is marked/salient than when it is neutral. Likewise, it would appear that
the complementizing conjunction is omissible when the itself begins with
a complementizer such as kak” that clearly demarcates it from the tag clause
(e.g., ASÈI 2: 198, no. 286; 203, no. 288); this possibility may also be
promoted by the production and interpretation constraints on multiple
embeddings. Syntactic parallelism also has some disambiguating power in
this respect, as is shown by the first two reports in (8).
The tendency to reduce in the ratio decidendi is carried further when the
report loses its clausal identity and fuses with the tag. In seven examples (in
six documents), the subject of the is encoded as the object of s”kazati
‘say ()’ in the accusative case, or genitive when the verb is negated by
raising (cf. Noonan 1985: 90) (10).
(10) I… [Vasilei Grigor’evič(’) Naumov… [Okiša Oljunova da
and [Vasilij Grigor’evič Naumov]- [Okiš Oljunov]- and
260 R V
Given the compact and partially given character of reports in the ratio
decidendi, the diffuse form of prototypically associated with new
messages (testimony) is dispreferred. There is one clear instance, which is
tagged with the standard formula tak(o) rek(li) (13):
(13) I [knjaz’ Ivan Jur’evič’ velel… [Andreja Okljačeeva
and [prince Ivan Jur’evič]- bid- [Andrej Okljačeev]-
obviniti… potomu čto Ondrej i ego znaxori
find.wrong- because Andrej- and his witnesses-.
zabludilisja… vodili po [ugrešskoj zemle
go.astray-. lead-. through [Ugrešskij land]-
[Kostjantinovskogo sela; i [ugrešskoj krylošanin [Ofonasej
[of.Konstantinovo village]- and [Ugrešskij cleric [Ofonasij
262 R V
while the preceding, unattributed clauses serve only to set the stage for it.
Thus the use of dě(i) in (17), like that of the ’s in (15)–(16), may be a
highlighting technique.
Another factor may have been the need for cohesion. The three clauses
that open the ratio decidendi cohere strongly because they share a single
theme; the second and third elaborate the content of the in the first. By
contrast, the fourth clause is less tightly connected to the preceding text
because of the shift of focus from the steward’s obligations to his actual
behavior. This shift is indicated by the retopicalization of the given referent,
as signalled by the pronominal rather than zero subject in the clause tagged
by dě(i).
false (spisok sudnoj obolživil) in the higher court (ASÈI 1: 494, no. 595).
Here the verb obol”živiti (obl”živiti) ‘declare false’ is not a straightforward
equivalent of the litigant’s statement from the verification procedure (ibid.:
493) but an analysis of its illocutionary intent. The original report, which is
in , could have been transplanted to the verdict without violating clarity or
stylistic conventions; the further stage of reduction is used — as often in the
verdict — to indicate the significance of the statement as filtered through the
judge’s interpretation. This same principle of labelling is found in a verdict
that emphasizes the respondent’s act of naming witnesses as background for
the crucial information that the witnesses did not bear him out (AFZX
98–99, no. 103). The v”imenovati ‘name’ functions as a narrative
substitute for the formulaic expression used to call witnesses (a list of names
without any ; ibid.: 98). Since it follows a string of three independently
tagged clauses, the represents a slip into a more integrated, evaluative
form in a contextualization that favors compactness.
Highly evaluative perlocutionary-based ’s are found in several
verdicts. Typically, these feature negations of ’s that refer to the illocuti-
onary intent desired by the principal. In some cases, the ’s refer to the
testimony of uncooperative witnesses and involve ’s such as
posluš’stvovati ‘bear witness’ (ASÈI 1: 548, no. 635); muževati ‘bear witness’
(AFZX 98–99, no. 103); and m”lviti v” ist’cevy rěči ‘say what litigant says’
(Gorčakov 1871, appendix [separate pagination]: 65, no. 4.5). Another
perlocutionary-based centers on očistiti ‘prove ownership of’ (literally,
‘clear’) in (20):
(20) I [kn(ja)z’ veliki… Kirila velěl obviniti,
and [prince grand]- Kirilo- bid- find.wrong-
potomu čto [Danilo Blin [těx zemel’… po gramotam
because [Danilo Blin]- [those lands]-. by writs-.
ne očistil…
clear-
And the grand prince… bid [the trial judge] to decide against Kirilo,
because Danilo Blin did not manage to establish his ownership of
those lands… with [his] documents (ASÈI 2: 318–19, no. 333).
The plaintiff Kirilo had based his case on the fact that Danilo Blin had
donated the disputed property to the monastery. In the higher court, Danilo
274 R V
produced documents to prove that the property was originally boyar land
(ibid.: 318); however, this new evidence was not accepted for technical
reasons. The use of the očistiti in (20) probably reflects the judge’s
evaluation of the perlocutionary success of Danilo’s statement. Note that the
negated perfective in (20) implies not inaction but an action essayed without
the desired achievement. Inaction — failure to perform a legally normative
— can also be reported as grounds for an adverse judgment. The typical
verb used is m”lčati ‘be silent’ (e.g., AFZX 107, no. 114); however, more
specific negated ’s like bivati čel”m’ ‘petition’ can also be found (ibid.).
In such cases, silence is viewed as intentional withholding of judicially
relevant information. (On verbs of reticence and silence, see Verschueren
1985: 73–121.)
The voice of the court is also heard in verdicts in which the rationale is
that a litigant had lost an earlier suit over the same property (ASÈI 2: 462,
no. 422); had neglected to petition for redress in a timely manner (AFZX
107, no. 114; ASÈI 1: 541, no. 628); had failed to name new witnesses when
his first were discredited (ibid.: 535, no. 492); or was unable to name
witnesses altogether, as in (23a):
(23) a. I [Kostantin Grigorievič(’)… iščei…
and [Konstantin Grigor’evič]- plaintiffs-.
obvinil, potomu čto… [sami oni
find.wrong- because [selves they]-
iščei i starožilci…
plaintiffs-. and old.residents-.
And Konstantin Grigor’evič… decided against the plaintiffs…
because… they themselves [were] plaintiffs and longtime resi-
dents (ASÈI 2: 445–46, no. 414).
b. Kak im, g(o)s(podi)ne, u nas starožilcom na
how them- lord- at us- old.residents-. for
[tu zemlju byti? starožilci, g(o)s(podi)ne, na
[that land]- be- old.residents-. lord- for
[tu zemlju my…
[that land]- we-
“How, lord, could we have them, longtime residents, for that land?
The longtime residents, lord, for that land are us” (ibid.: 443).
The juxtaposition of (23a) with its anterior report in the testimony (23b)
reveals it to be a clear case of , with the first person replaced by the third.
The crux of the verdict was evidently the interpretation of the term “longtime
resident”, which in legal usage denoted a special kind of witness (Kleimola
1975: 35–37; Kaiser 1980: 137–38). The judge evidently found an anomaly
in the peasants’ claim to be at once litigants and “witnesses”; this crossing
of purposes motivated his decision to reject their testimony.
Unattributed reports of testimony — more precisely, — occur in
some cases in which the crucial evidence against the losing litigants was
their witnesses’ claim not to know or remember information about the
disputed property (24)–(25).18 Though the reported status of cognitive
R V 277
villages have been built amidst those lands”). This is an untagged, condensed
renarration of the respondents’ testimony (cf. AJu: 15, no. 8). While the
ultimate source of the first part [3], Mixajlo Volynskij’s cadastre book, is
attributed by an adjunct prepositional phrase, the second part [4] (about the
“new villages”) is an unattributed synthesis of information from two other
cadastre books, which are explicitly named in the respondents’ testimony.
Indeed, since it is ascribed neither to the ultimate sources nor to the speakers
who introduced it into evidence, it is doubly unattributed (or triply, given
that the judge is the immediate source).
The other case of untagged in (25) is at the onset of the rationale [1],
in the statement “they had sought those lands after forty years” — i.e.,
longer than the three- to six-year statute of limitations legislated by the
Lawcode of 1497 (Grekov (ed.) 1952: 28; see also Kleimola 1975: 30–31,
87–88). The plaintiffs did not, in fact, admit that their opponents had been
in de facto possession of the land for forty years; this information was
furnished, probably unwittingly, by their witnesses (AJu: 15, no. 8). In the
filter of the judge’s interpretation — i.e., “taken out of context” — the
witnesses’ statement assumed a different significance, which subverted the
intention of the original testimony (cf. (23a)).
This evaluative use of was neither an accident or an abuse; interpreta-
tion is the essence of the requotations in the verdict. Indeed, changes in
meaning — the addition of a new perspective, if nothing else — are an
inevitable effect of any recontextualization of because of the inseparabili-
ty of meaning and context (Bakhtin 1975/1981: 340; Sternberg 1982a;
1982b: 108–9; Tannen 1989: 100–101, 105–110). This is true not only of the
ratio decidendi, where evaluation is institutionalized, but also of the other
sections, including even the testimonial portion with its objective style of
narration. As Sternberg remarks (1982b: 108), no matter how objective a
reporter tries to be, “the difference… is ultimately one of degree: to quote
is to mediate and to mediate is to interfere.”
Within this difference of degree, it is evident that the typical manner of
reporting in the testimonial portion differs from the one generally employed
in the verdict in the demands that it makes on the interpreter. Whereas the
standard strategy in the trial record gave the intended interpreters most of the
responsibility for sense-making, those used in the verdict often gave them
hardly any at all. This was appropriate, given the function of the completed
document and of the verdict in particular: it was now essential to leave the
280 R V
And Semen Nakvasa brought the [trial] record to me…, [1] and in
it was written that [2] [he] asked my witnesses… about the border
of the Glumovo property with my property… [3] And they stated
before him that they did not know the border between those prop-
erties. [4] But he asked the Glumovo witnesses, [5] and they said
that the border of the Dobromer’e property with the Glumovo
property [was] the … [lacuna — DEC] road; and they also led
[him] along the road, and from the road… [boundary-setting de-
scription with lacunae — DEC]. And my witnesses and villagers
did not follow them along that border. And Nakvasa placed the
witnesses of both parties before my court; [6] and before me they
said the same thing as is written in Nakvasa’s [trial record]. And
I… for that reason found in favor of Vasilij and against the
Dobromer’e men (ibid.: 507).
While the narrative in (26b) resembles trial records in content, it resembles
the ratio decidendi (apart from the use of the first person) in its preference
for less-than-direct speech — complementized indirect or nondirect speech
([1], [3]); fused speech ([5]); and ’s or syntheses of content ([2], [4],
[6]). While the account of the hearing includes representations of both the
questions and the answers, it cannot be considered dyadic in format, since it
relays only the theme of the questions, not the message ([2], [4]); moreover,
the represented answers appear to be conflations of several statements ([3],
[5]). All of these reports are filtered through the perspective of the grand
prince; in addition, they are all backgrounded, since they are recorded not to
supply information for a verdict but as an ex post facto record of a favor
granted to a petitioner.
In many ways, the style of reporting in (26a–b) and in the ratio decidendi
anticipates the patterns seen in trial records of more than a century later. In
the late sixteenth and seventeenth centuries, the judges’ interpretation of
documents in camera came to assume greater importance than testimony
offered in face-to-face confrontations; hence more and more of the space in
trial records was given over to the judges’ reasoning. Accordingly, oral
testimony was relegated to a primarily corroborative role in the judicial
process and was increasingly presented not in the diffuse, undigested form
of but in the less-than-direct reporting modes characteristic of compact,
writer-based contextualizations.
C 9
Conclusions
The scribes who wrote the investigated trial transcripts drew from a rich
inventory of reporting strategies. After deciding initially to report rather than
aver, the scribes could choose from among four modes — , , ’s, and
fused , a small-clause construction not found in — and from a variety
of tag formulae, which could be preposed to or intercalated in the reports. In
addition, with and they could choose between complementized and
uncomplementized and between tagged (attributed) and untagged (free)
varieties. Though has been treated in previous studies as a postmedieval
innovation, it is, in fact, well attested in medieval trial transcripts, and it can
even be the dominant reporting strategy in other premodern text-kinds, e.g.,
third-person memoranda summarizing orders (nakazy) to the tsar’s officers.5
The distribution of the various kinds of suggests that they are func-
tionally polarized into two broad types, in accordance with their effect in
creating prominence in the discourse. Reporting strategies may be classified
as relatively compact in proportion as they blend in with the surrounding
nonreported discourse, and as relatively diffuse in proportion as they stand
out from it. This salience-related distinction is not equivalent to the tradition-
al dichotomy of and , since it is not defined by deixis alone; nondeictic
factors also can have the effect of integrating reports into the authorial
discourse or, conversely, singling them out.6 Another reason why the
C 289
reporting is more hearer-based — that is, it requires more effort on the part
of the interpreter, who then has a greater role in sense-making; by contrast,
compact discourse is more speaker-based.
The institutional purpose of the trial record — the “dialogue discourse”
that recounted the on-site trial hearing — was to provide judges with
detailed, impartial accounts of the on-site hearings as raw material for their
verdicts. Accordingly, in this contextualization the conventions of reporting
were directed at maximizing the interpretive contribution of the judges while
minimizing that of the scribes, who generally avoided evaluations and
effaced themselves to the point of anonymity. This hearer-based orientation
is seen first of all in the choice of a diffuse strategy — signalled by a
preposed narrative tag clause — as the standard way of reporting testimony.
By virtue of its formal properties, typically functions not as an analysis
but as a quasi-dramatic reenactment of events, even though under
ordinary circumstances it cannot be considered a true facsimile (see the
discussion of verbatimness, above). Consequently, the occurrence of in
the trial record compelled the interpreters to adopt the perspective of the
participants in the represented speech event in order to impose coherence on
the . In other words, the interpreters were obliged to hear the trial hearing
as it were from inside, without any apparent interference from the reporters;
they became, in effect, witnesses of the hearings, like the judges who
observed the actual trials, with no other vade mecum than their own recon-
structive abilities. Thus the standard reporting strategy matched the organiza-
tional principles of the actual trial hearings, in that it allowed “every man…
to tell his own tale and plead for himself so well as he can”, as Dr. Giles
Fletcher, an early English ambassador to Muscovy, wrote in Of the Rus
Commonwealth (1591/1966: 71).
The hearer-based orientation that motivated insets is further reflected
in the extended frame. The use of standardized preposed tags, though often
redundant, ensured that the interpreters could keep track of the positions of
the principals and established a predictable structure that facilitated quick
recognition of represented utterance boundaries. (It is noteworthy that the
reči, which appears in the standard tag clause, is infrequent in additional
layers of within the testimony; presumably this is due to the fact that it
could have been misinterpreted as a signal of a new turn at speaking in the
authorial narrative.) Likewise, the diffuse dyadic format, with the judges’
questions in , was not chosen for the sake of verisimilitude, though it
C 291
mirrored the structure of the actual hearing; rather, it permitted the scribes to
supply an ample frame for testimony independent of the narrative. This, in
turn, compelled the intepreters to situate the testimony in a context consisting
primarily of represented speech events and thus to perceive it much as the
participants in the hearing did, though in a different medium. In addition,
the questions topicalized elements that the interpreters could process quickly
while focusing on the new information in the responses. Moreover, since
statements were conventionally followed by questions and vice versa, the
dyadic format provided many additional clues such as modality or the
occurrence of specific vocatives to supplement the delimitive function of the
tag clauses.
The hearer-based orientation also motivated other features of trial records
— in particular, the spare, covert style, reflected in the preference for
minimal lexical elaboration and paratactic organization (cf. Lakoff 1984).
Paradoxically, the existing treatments of syntax typically treat parataxis
as if it were “simpler” and less “sophisticated” than hypotaxis, even though
the latter actually makes fewer demands on the interpreter. Likewise, the use
of “dialogue discourse” and minimal narrative limited the possibility of
authorial intrusions. The scribes concentrated important information within
the and conveyed events without indicating how they unfolded and
without analyzing their relative weight by departures from real-world
sequencing. This gave the interpreters evaluative responsibility for making
connections between the represented events (cf. Rader 1982: 192) and so
fulfilled the social function and, in particular, the cooperative norms of the
text-kind. The use of a third-person narrative style likewise assisted the
interpreters by ensuring that any first- and second-person elements could be
taken as signals of .
As mentioned in 9.1, violations of the standard pattern in a given
contextualization occur chiefly with reports that depart from the prototypical
speech event in that contextualization. The prototypical speech event in the
trial record was testimony, which in this context must be understood in a
narrow sense: an uninterrupted statement by one of the principals, which
provided information intended to establish the speaker’s rights to the
disputed property and was elicited by the judge(s) in a face-to-face exchange,
presumably in a standard configuration determined by judicial custom. There
can be little doubt that this speech event was perceptually central in the trial
institution; no other kind of utterance was as likely to supply information
292 R V
that could be probative for the verdict — proofs of ownership offered by the
parties who were interested and legally liable in the lawsuits. (There was, of
course, a different prototyical speech event for the judges — requests for
information about the disputed property, addressed to and cooperatively
answered by individual participants, again in a face-to-face exchange.)
The violations of the standard in the trial record include cases in which
the litigants’ remarks were not directly concerned with providing evidence,
or when they were presupposed, limited in content, or transparent in attribu-
tion. Other examples suggest that unconventional reporting strategies could
indicate that the represented speaker was on an unexpected footing in the
trial, as when two judges were asked to give evidence, or when witnesses
testified against their party’s interests or made statements outside the hearing.
Another type of nonprototypical speech event is the opening complaint,
which ordinarily deals with the respondents’ malfeasances rather than with
the plaintiffs’ own proofs of ownership. Since this nondyadic contextuali-
zation is relatively close to the prototype, the standard strategy predominates.
By contrast, in certain other nondyadic situations, there is a distinct prefer-
ence for less-than-direct strategies — and fused . This is the case in the
incipit and in narrative accounts of reconvenements, where reports generally
convey backgrounded orientation information. Similarly, ’s are common
in reports of nondyadic stage-managing utterances by both the litigants and
the judges. Since reports of this kind usually had no evidentiary value for
the verdicts, it would have been a violation of relevance to present them in
the detailed form of , which would have brought the narrative to a
standstill. The judges’ procedural commands are generally followed not by
dialogic responses but by citations of documents or narrative passages; this
obviated the framing function ordinarily performed by the judges’ . When
they do serve, in accordance with the prototype, as frames for verbal action
such as testimony or refusals, they tend to be presented in the conventional
way, as .
The of the witnesses is conveyed more diversely than that of the
litigants; this is explained by the fact that the statements of these secondary
participants are not prototypical testimony in the sense defined above. The
witnesses were on a different footing in the trial than the litigants; they
spoke on oath, often acted as a group, and were only responsible for evi-
dence contained in their own memories. The witnesses’ evidence generally
falls into two categories: boundary-setting descriptions, which differ from
C 293
This case study of reporting in medieval Russian trial transcripts has at least
two larger implications for historical-linguistic methodology. First, it demon-
strates a rigorous, non-arbitrary method of recovering the crucial factors of
intention and purpose in premodern writings. The method approaches the
data from two directions — from the observable formal properties of
discourse strategies, which are seen to be conducive to specific textual
functions; and from the needs of speech genres and text-kinds, which
motivate one or a subset of those functions in a given context. These
institutional needs or aims provide a social locus for intentionality that lies
not in the unrecoverable whims of individual writers but in the norms of
cooperative behavior in institutional situations.
The second ramification follows from the generally purposive nature of
reporting as one kind of behavior involving large-scale textual units. It was
shown that the choice of reporting strategies could, in virtually all cases, be
linked with textual functions that matched the needs of the given contextu-
alization. The method of residual forms proved that this was true even in
cases when the strategies in question were infrequent — perhaps even
statistically insignificant — in the corpus. If this is any indication of what
may be expected in other cases of alternative discourse strategies, it reveals
C 299
a possible pitfall for approaches that pay insufficient attention to often fine-
grained details of context and genre.
To put it briefly, it is premature to interpret the nonoccurrence or
scarcity of a unit of syntax, discourse, or lexicon as an evolutionary indicator
before one has determined whether there are likely contextualizations for the
unit in the attested texts. The crux is, of course, how to define what the
likely contextualizations are; typological data cannot be more than suggestive
here, since the functions of higher-level units can vary from one language to
another. In fact, one is compelled to examine an exhaustive or, where that is
not feasible, at least a representative range of contextualizations; yet in order
to decide what is representative one must have a good working knowledge
of the repertory and reading conventions of the text-kinds produced in the
given culture, and hence of the culture itself.
It is not difficult to find object lessons to justify this cautionary note;
indeed, some may be found in the history of the study of . The existence
of in premodern Russian was overlooked — or, perhaps, not looked for
— not only because of false a priori assumptions about its modern, Western
European origin but also because of the decontextualized nature of most
previous approaches to reporting (as to discourse in general) in medieval
texts. In order to recognize , one must be familiar with interpretive frames
and with the reading conventions of specific text-kinds; however, most of the
previous investigations drew their conclusions about from isolated
sentences — a dubious unit in medieval discourse — culled from miscella-
neous sources.
While studies that cull examples from indiscriminately chosen texts can
accidentally cover a large number of contextualizations, they are unlikely to
give any weight to the functional, generic, and contextual factors that can
motivate variation. In any case, assembling a corpus that represents a
sufficiently broad spectrum of functions — when it is possible at all —
requires extensive background research, which is not exclusively linguistic
and does not always lead to the immediate location of the sought-for tokens.
This preliminary work is often omitted; consequently, conclusions have
sometimes been built on the shifting sand of unrepresentative data.
It has been said with some justification that the study of discourse “is of
necessity the study of particularity” (Becker 1984: 435) — an observation
that, without rejecting the validity of generalizations per se, warns against
overgeneralizing neglect of contextual detail.7 The results of my case study
300 R V
1986: 65) that “there is not a single new phenomenon… that can enter the
system of language without having traversed the long and complicated path
of generic-stylistic testing and modification.” Before positing sweeping long-
term tendencies or independent developments in the linguistic code, one
should first take into account “the teleology of the authorial context” (to use
Bakhtin/Vološinov’s term), as reflected in such factors as the individual
characteristics of primary sources, the function and evolution of text-kinds
and the speech genres that they contain, and the changing needs of the
interpreters and social institutions that the text-kinds serve. This highly
contextualized level of analysis has as yet been little practiced — a fact that
gives particular urgency to the further development of the field of historical
pragmatics and to pragmalinguistic studies of text-kinds and speech genres.
Notes
Preface
1. See, for example, Arnovick 1999; Brown and Gilman 1989; Shippey 1993, van der Walle 1993,
and the articles in Jucker, Fritz, and Lebsanft (eds.) 1999 and Jucker (ed.) 1995 (in particular,
the discussion in Jacobs and Jucker’s introduction).
2. See, inter alia, Baynham 1996; Fludernik 1993; Thompson 1996, and the articles in Coulmas
(ed.) 1986; Dirven et al. 1982; Janssen and van der Wurff (eds.) 1996; Lucy (ed.) 1993;
Verschueren (ed.) 1987. For further references, see van der Wurff 1997.
3. See D. Collins 1996; J. Collins 1987; Johnstone 1987; Mayes 1990; Philips 1985; Romaine and
Lange 1991; Silverstein 1985, 1993; Tannen 1986, 1989: 98–133; Yule 1993.
4. Bakhtin acknowledged his authorship on various occasions; Vološinov’s widow supported his
claim (Clark and Holquist 1984: 146–48, 166; Ivanov 1973: 44; V. L. Maxlin in Vološinov
1929/1993: 177).
Chapter 1
1. On the universality of (in particular, direct speech), see Coulmas 1986b: 2; Cram 1978: 42;
Li 1986: 39; Lucy 1993: 9; Silverstein 1976: 50; Wierzbicka 1974: 271.
2. See Jakobson 1957/1984: 41; Longacre 1983: 44; Vološinov 1929/1986: 85–87.
3. On prior texts, see also Bakhtin 1975/1981: 278–80; Becker 1984, 1988: 24–26; Chafe
1994: 212; Coulmas 1986a: 1–2; Fónagy 1986: 283; Hopper 1988: 117–23; Prince 1987: 46;
Tannen 1989: 99–100.
4. On metapragmatics, see also Lucy 1993: 17; Silverstein 1976: 50–52, 1977: 146–47, 1985,
1993: 36–45; Verschueren 1987: 125–26, and other works in Verschueren (ed.) 1987.
5. On “thick” descriptions, see Ryle 1966–67/1971, 1968/1971; on “emic” understanding, see
Becker 1984.
6. On the automatization of intentional strategies, see Enkvist 1987: 24; Keller 1985: 213.
7. Note that intentional acts can subsume acts whose intentionality is irrelevant (Blackburn
1994: 196). Uttering a relative clause is intentional; the accompanying movements of the speech
organs are usually not (Keller 1985: 213).
304 R V
8. The sequence in which I discuss these decisions is not intended as a claim about the order in
which they are actually made in the reporting process.
9. Among the many alternative terms are verbs of saying/utterance/communication, linguistic action
verb(ial)s (Goosens 1987; Verschueren 1987), and metapragmatic descriptors (Silverstein 1985).
10. For the citation conventions, see “Conventions for Citing Cyrillic Sources” in the front matter.
11. See, e.g., Munro 1982; Coulmas 1985a, 1985b; Roncador 1988; Frajzingier 1991; the studies
in Coulmas (ed.) 1986 and Janssen and van der Wurff (eds.) 1995. For bibliography, see van
der Wurff 1997.
12. On ’s with infinitives and participles, see D. Collins 1994: 202–44, and the references cited
there.
13. On the nonconvertibility of and , see Banfield 1973; McHale 1978: 250–57, Partee 1973.
14. Banfield’s attempts (1973, 1978, 1982, 1993) to work English and French into a generative
syntax proscribe many of the features that are typical of the mode. For discussion, see McHale
1978: 252–57.
15. Even as object clauses, reports are syntactically problematic. See Munro 1982 for discussion.
16. See Bulaxovskij 1958: 409; Ivanov 1965: 80–81; Machek 1968, s.v. “dít”; Šaxmatov 1941: 268;
Sreznevskij s.vv. “děti, děju” and “dějati”; Vasmer 1953–58, s.v. “de.”
17. See Hickmann 1993: 66–67; Leech and Short 1981: 324; McHale 1978: 258–60; Page 1988: 35;
Semino, Short, and Culpeper 1997; Tannen 1986: 323.
18. On slipping in , see Bulaxovskij 1958: 416; Lopatina 1979: 445–46; Sundberg 1982: 178. On
, see Borkovskij 1981: 222–30; Koduxov 1955: 136–40; Lopatina 1979: 446–47; Peškovskij
1956: 485; Vinogradov and Istrina (eds.) 1960: 413–14. On other languages, see Fónagy
1986: 276–77; Haberland 1986: 230–31; Holt 1996: 243, Note 1; Schuelke 1958; Tannen
1986: 314, 1989: 117–18; Yule 1993.
19. See, for example, Givón 1980; Wierzbicka 1987, 1988: 23–168.
20. Cf. Keller 1985: 236; see also Lewis (1969) on the emergence of conventions in language.
21. Cf. Ehlich (1981: 160–63) on the parallel activities of “philologists” and “linguist native speakers.”
22. The possibility that writers may have had private agendas poses similar difficulties. However,
unless there is internal evidence for such agendas, there is no reason to suspect them in the
kinds of documents used in this study, which had disinterested writers and clear-cut institutional
functions.
23. Inability to make a match of this kind does not, of course, exclude the possibility of a rational
motivation that cannot be reconstructed due to inadequate information.
Chapter 2
1. For descriptions of typical lawsuits, see Dewey and Kleimola 1973: 42–45; Kleimola 1972,
1975. On general characteristics of the text kind, see also Čerepnin 1948–51, 2: 230, 1952: 649;
Dewey and Kleimola 1973: 41–48; Kaštanov 1998: 160; Kleimola 1972, 1975; Leont’ev
1969: 45–46; Pokrovskij 1973: 20–22.
2. Not included in this count are a few documents that are cited within other transcripts but are
not independently attested.
N 305
3. On the terminology, see Dewey and Kleimola 1973: 47–48; Kaštanov 1998: 160; Kleimola
1972: 356, 1975: 10; Pokrovskij 1973: 21.
4. On the referral process, see Kleimola 1972: 367, 370; 1975: 68–74; Leont’ev 1969: 40–41.
5. Cf. the reliance on communal memory, as evidenced by the testimony of “longtime residents”
(starožil’ci).
6. The Muscovite Law Code (Sudebnik) of 1497 legislates different processing fees (see Grekov
(ed.) 1952: 21–22, §§15, 16, 22, 24; see also the commentary, 64–65, 68). On these subtypes,
see also Kaštanov 1988: 151, 1998: 160; Kleimola 1975: 6; Leont’ev 1969: 45.
7. One later summary of a trial transcript, of uncertain function, has the date at the beginning,
much as in a chronicle entry (ASÈI 3: no. 319); however, this paraphrase evidently had a
nonjudicial purpose.
8. See ASÈI 2: no. 229, 388a, 464 (no year), GVNP no. 340 (1463–90); AFZX no. 117, 157, 204,
248, 259, ASÈI 2: no. 416, 418, 419, 422, 3: no. 105, 276 (1491–1500); AFZX no. 306, ASÈI
1: no. 651, 658, 2: no. 309, 310, 336, 338, 428, 495, 3: no. 48, 172, 173, 221, 223, 224, 250,
251, 478, Kaštanov 1970, no. 40 (1500–1505). In tallying up the total number of transcripts for
a specific period, I have assigned the documents without internal dates to the earliest possible
year given by the editors.
Chapter 3
1. On perspective and the typology of , see, inter alia, D. Collins 1996; Doležel 1973: 26, 28–29;
Ebert 1986: 157; Goffman 1981: 147–52; Koduxov 1955: 118–19; Li 1986: 29–34; Palmer
1986: 163–65; Romaine and Lange 1991: 228–29; Roncador 1988: 3–4. Alternative definitions
of are discussed in 3.3.
2. There was a different convention for introducing the judges’ ; see Chapter 4.
3. For settlement records presented as dialogue discourse, see ASÈI 1: no. 420, 421, 422, 428,
432; 2: no. 210, 289, 290, 335, 409, 429; 3: no. 213, 214; Kaštanov 1970: no. 70. For a
default-judgment charter, see ASÈI 2: no. 167.
4. The following forms occur: masculine rek(”) or rekl(”), rekli or r(’)kli. The letters
transcribed ” and ’ were essentially diacritic and often omitted in writings of the Muscovite
period.
5. Such buck-passing may be seen in other transcripts, e.g., ASÈI 1: 324, no. 336; AFZX 97, no. 103.
6. See AFZX: 116, no. 125; ASÈI 2: 251, no. 296; Kaštanov 1970: 357, no. 9.
7. This is attested in two related transcripts — ASÈI 2: 415–16, no. 404 (5×); 417, no. 405.
8. See ASÈI 2: 435 (2×), 436–37 (5×), 438, 439 (3×), 440, no. 411.
9. I am not claiming that the decision-making process followed the order presented here.
10. Given that preposed tags are less salient than (“eclipsed by”) the following direct reports, it seems
somewhat problematic to characterize this relation as “separate: equal” (Thompson 1996: 520).
11. For a similar example, see ASÈI. 1: 315, no. 427.
12. On punctuation in , particularly with reference to , see also Čerepnin 1956: 159, 374–76;
Osipov 1992; Preobraženskaja 1983: 102–3; Rinberg 1985: 49; Starovojtova 1988: 70. The same
306 R V
basic observation could be made for premodern writings in other European languages, e.g.,
Hungarian (Fónagy 1986: 257) and English (Lennard 1995: 67–68). On the diverse uses of
quotation marks in premodern English texts, see also Page 1988: 30–31).
13. On the usual editorial practices, see ASÈI 1: 8; Čerepnin 1956: 564–69; Kotkov and Popova
1986: 3.
14. On this word, customarily abbreviated as (>n or (c>n, see Čerepnin 1956: 246; Zaliznjak
1990: 9–24.
15. Cf. Brown and Levinson 1987; Dewey and Kleimola 1973: 41–42; Kleimola 1972: 364.
16. See Volkov 1974: 115–16; Zaliznjak 1986b: 272, 289, 1990: 12–14.
17. On typicality markers in , see also Bauman 1986: 67; Chafe 1994: 217; Dubois 1989.
18. On vocatives as a marker, see Roncador 1988: 4; Wierzbicka 1970: 644.
19. In heterosubjective conjuncts, many of the cases without g(o)s(podi)ne show topic continuity
involving elements other than the subject.
20. For representative expressions of this view, see Otin 1969: 54–55; Wierzbicka 1974: 267, 272;
Xaburgaev 1974: 425–26; Čumakov 1975: 16; Longacre 1983: 132; Li 1986: 39–40.
21. On the use of in this text kind, see also Larin 1961: 30. For examples, see AI 1: 340–41, no.
179.2; ibid. 2: 62–63, no. 53; ibid.: 123, no. 92; ibid.: 249–50, no. 212; ibid.: 314–15, no. 262;
ibid.: 341–42, no. 282; ibid.: 351–52, no. 290; ibid.: 354, no. 295; ibid.: 357–58, no. 301; ibid.:
358–59, no. 303; ibid.: 364–65, no. 307; AJuB 3: col. 271–76, no. 330; Kotkov, ed. 1984: 161,
no. 126; Kotkov, ed. 1990: 81–82, no. 67; ibid.: 108, no. 89; Mordovina and Stanislavskij
1979: 101; Stanislavskij 1981: 288, no. 2; ibid.: 293–94 no. 8; ibid.: 298–99, no. 18; ibid.: 300,
no. 21; ibid.: 303–4, no. 28; ibid.: 305, no. 32; and Kotkov et al., eds. 1993: 119–47.
22. The reproductionist model is common in Russistics — e.g., Čumakov 1975: 16–17; Gvozdev
1958: 274; Koduxov 1955: 114–19; Peškovskij 1956: 484; Rozental’ 1988: 247; Vinogradov and
Istrina (eds.) 1960: 402. For reproductionist treatments of in , see Molotkov 1958: 28;
Sprinčak 1960–64, 2: 119; Otin 1969; Lopatina 1979: 442; Rinberg 1985: 53; Schmücker-
Breloer 1987: 248–49. For other languages, see, e.g., Xaburgaev 1974: 425–26; Li 1986: 40;
Massamba 1986: 99–100.
23. has also been treated as an “opaque” context (cf. Quine 1960: 151) in which referential terms
are necessarily de dicto rather than de re, i.e., expressions of the “original speaker” (see
Coulmas 1986b: 3–4; Kiparsky and Kiparsky 1970: 157–58, Note 7; Li 1986: 29–30; Partee
1972; Zwicky 1974: 198–99). There are several reasons to doubt this approach, apart from its
reproductionist character. First, it only pertains to reports of actual utterances. Second, it is
logical-philosophical contraband to associate verbatimness in natural language with truth value.
The question “Is this a valid argument?” is not equivalent to “Is this an exact rendition of what
was said?” Third, actual observation proves that reporters allow substitutions in unless they
have some special reason to avoid them. (For further criticism, see Cram 1978: 42–44.)
24. On and memory constraints, see Chafe 1994: 215–17; Clark and Gerrig 1990: 796–97;
Coulmas 1986b: 25, Note 8; Mayes 1990: 332; Tannen 1986: 313–14.
25. See also Comrie 1986: 266; Coulmas 1986b: 2, 6, 11–13, 25; Leech 1974: 353; Li 1986: 40;
Romaine and Lange 1991: 231–32, 243, 263; Short 1988: 66–67, 70.
26. On non-verbatim uses of , see also Baynham 1996: 66–68; Chafe 1994: 216; Clark and Gerrig
1990; Dubois 1989; Fludernik 1993: 408–14; Fónagy 1986: 255, 278–82; Haberland
1986: 224–25; Hanks 1990: 206; Mayes 1990; Page 1988: 6–7, 25–26, 29–30; Romaine and
Lange 1991: 229–31; Sternberg 1982a, 1982b; Tannen 1986: 313–14, 1989: 98–119.
N 307
27. On co-construction of (“chiming in”, Couper-Kuhlen 1999: 18, 21), see also Mathis and Yule
1994; Yule 1995: 189.
28. On departicularizers in , see also Chafe 1994: 216; Haberland 1986: 223–24; Fludernik
1993: 410; Mayes 1990: 334; Sternberg 1982a: 96, 99–104.
29. On strengthening hedges, see Brown and Levinson 1987: 145.
30. Cf. Romaine and Lange’s observation (1991: 231–32, 243, 263) that different tags are used
with to indicate differing degrees of commitment to fidelity.
31. See Coulmas 1986: 10–13; Goody 1977: 118; Lyons 1977: 17; Ong 1981: 21–22, 1982: 57–68.
32. For example, ASÈI 3: 270, no. 251; Anpilogov 1977: 79, 208, 212.
33. This word was usually written without the o, with a superscript k over the a. Thus there is
usually no way to distinguish between tako, which in the later medieval period occurred mostly
in ecclesiastical texts, and the innovative form tak, first attested in a text of 1356 (Vasmer
1953–58, s.v.).
34. On cataphora in discourse, see also Halliday and Hasan 1976: 17–19, 69, 75.
35. Cf. Romaine and Lange (1991: 244–51, 262) on the strategy be like in recent American English.
36. See AFZX 97, no. 103; ASÈI 1: 335, no. 447; 352–54, no. 467 (3×); 2: 119, no. 188; 150, no.
229 (2×); 389, no. 387 (2×); 416, no. 404; 418, no. 405; 3: 55, no. 32.
37. The true impersonals pisano ‘write-.’ and napisano ‘write-.’ are also found
in seven examples. See AFZX 97, no. 103; ASÈI 1: 491, no. 595; 2: 335, no. 338; 451, no.
418; 544, no. 496; 3: 55, no. 32; Kaštanov 1970: 371, no. 16.
38. Two cases directly follow perception reports similar to (9b) (ASÈI 2: 320–21, no. 334; 378, no.
381). Others occur when the citation is the second in a series (ibid.: 265, no. 307; 3: 182–83,
no. 172 (2×)).
39. See ASÈI 2: 422, no. 406; 3: 463–64, no. 478.
40. For possible examples, see ASÈI: 201, no. 288; 437, no. 411; 523, no. 483.
41. Cf. AFZX 227–29, no. 259; 258–59, no. 308; ASÈI 1: 472–73, no. 587; 2: 324–27, no. 336;
332–33, no. 338; 418–19, no. 405; 459–61, no. 422. For exceptions, see ibid.: 333–34, no. 338;
422, no. 406; 436–37, no. 411 (said by the editors to be an abridgment).
42. See Bakhtin 1952–53/1986: 76–81, 86; Fludernik 1993: 20, 22; Grice 1975/1989; Lakoff
1995: 196; Slembrouck 1992: 109–10; Sperber and Wilson 1986; Walker 1986: 214, 220.
43. In a sense this is true even of mouthpieces or, at least, of the ambassadors mentioned above,
since the rulers they addressed would require as unmediated a representation of the grand
prince’s position as possible.
44. On assessments of audience, see also Bakhtin 1952–53/1986: 97–99, 1975/1981: 279–81; Lakoff
1984: 481; Sinclair 1988: 15; Walker 1986: 216–17, 220.
45. Clark and Gerrig (1990: 787) treat both and as demonstrations that differ in perspective:
“Free indirect quotations, like direct quotations, are demonstrations that are components of
language use. It is just that free indirect quotations take the vantage point of the current instead
of the source speaker.”
46. See also Fónagy 1986: 255; Goffman 1981: 149–52; Mayes 1990: 346; Tannen 1983: 365,
1989: 125.
308 R V
47. Zipf (1949: 20–21), approaching speech behavior from a different standpoint, makes a similar
distinction between speaker’s economy, where the hearer must exert more effort, and auditor’s
economy.
48. It would be more precise to say addressee-based, since such strategies are oriented to the
knowledge and sense-making ability of ratified addressees rather than to hearers of other kinds
(cf. Clark 1992: 205–6, 248–74; Goffman 1981: 260). Moreover, they are not restricted to
spoken language but also appear in writing.
49. See, e.g., Baynham 1996: 78; Chafe 1982: 48, 1994: 217, 223; Čumakov 1975: 19; Jespersen
1924/1992: 290; Leech and Short 1981: 319–20; Li 1986: 41; Mayes 1990: 326; Page
1988: 32–33; Schiffrin 1981: 58–60; Schuelke 1958: 97; Tannen 1983: 365, 1986: 311,
1989: 25–26; Thompson 1996: 512.
50. See, for example, Fludernik 1993: 435; Glock 1986: 46, 48; Larson 1978: 60–68, 183; Li
1986: 40; Philips 1985: 154–55, 169; Tannen 1983: 364.
51. This is not to be confused with the moral or ethical question of the extent to which one bears
responsibility for the use or misuse of one’s statements by others — an issue that need not be
discussed here.
52. Cf. Huizinga (1944/1955: 14–15) on “the mystic repetition or re-presentation” involved in
sacred rites.
Chapter 4
1. See Barnet 1965: 62–65; Borkovskij and Kuznecov 1965: 290, 318; Kuz’mina and Nemčenko
1980: 152–54; Stola 1958.
2. One complication is the aspect of the gerund. Since the verb reči was perfective by the fifteenth
century, r’kuči might be expected to be perfective as well and hence to convey sequentiality;
however, its use in discourse suggests that it was imperfective (cf. Barnet 1965: 104–5;
Larin 1975: 217).
3. See Chvany 1985b: 15; 1990/1996: 292–93; Fox 1983; Givón 1984–90, 2: 839–40; Hopper and
Thompson 1984: 740–41.
4. For examples, see ASÈI 1: 556, 557, no. 642 (2×); ibid. 2: 394, 395, no. 388a (3×); 436, no.
411 (1×); AFZX 231, no. 259 (2×); 259–60, no. 308 (2×).
5. For the other cases, see ASÈI 2: 431–32, no. 410 (3×); 536, no. 493. It was not, of course,
obligatory to use a gerund tag in such contexts; the standard formula could be used, as in ASÈI
2: 539, no. 493.
6. See also ASÈI 2: 119, no. 188; 2: 377, no. 381; AFZX 181, no. 204; 213, no. 249; and in
transcripts from Pskov and Dvina (GVNP 148, no. 92; 188, no. 132; 326, no. 340). The same
incipit is found in a 1391 monastic statutory charter (ASÈI 3: 16, no. 5) and in default
judgment charters (ASÈI 1: 238–39, no. 329; 362, no. 479; 2: 102, no. 167; AFZX 221, no.
257; 229, no. 259).
7. See N. P. Lixačev (ed.) 1895, pt. 2: 198, no. 10 (1543); AJuB 1: no. 52.5 (1547).
8. See, for example, AJu no. 24 (1584), 52.7 (1561), 52.8 (1567); DAI no. 157 (1609); Ivina (ed.)
1983: no. 154 (1566); Kotkov and Filippova (eds.) 1978: no. 26 (1555), 83 (1576–77); Liberzon
N 309
(ed.) 1988–90, pt. 2: 656 (1578); N. P. Lixačev (ed.) 1894: no. 2 (1583), 3 (1589); N. P.
Lixačev (ed.) 1895, 2: no. 11 (1551), no. 15.4 (1584). For conservative texts that largely
maintain tak(o) rek(li), see AJu no. 22 (1547), 23 (1571); AJuB 1: no. 52.6 (1555); N. P.
Lixačev (ed.) 1895, pt. 2: no. 12 (1552).
9. Brat’e is the vocative of the collective (functionally plural) noun brat’ja. Typically the two
vocative forms were spelled identically in scribal usage — as brate, with a superscript t.
10. E.g. Antonov and Baranov 1997: 117, no. 146. For the other cases, see D. Collins 1994: 138,
Note 192.
11. For the remaining cases, see ASÈI 1: 477, no. 588; 2: 452, no. 418; 455, no. 419. Another
instance of this appears in nonmultiplex testimony (ibid. 1: 547, no. 635).
12. On the historical present in , see also Nikiforov 1952: 174. For other languages, see Chafe
1994: 207–11; Johnstone 1987; Page 1988: 48; Schiffrin 1981; Schuelke 1958: 97; Tannen
1983: 365, 1989: 214, n. 13; Uspenskij 1973: 71; Vinogradov and Istrina (eds.) 1960: 482.
13. Cf. the strong association of the historical present and emotive ’s such as go, be like, be all
in American English (Schiffrin 1981: 58; Romaine and Lange 1991: 243).
14. For the other cases, see ASÈI 3: 221, no. 209; AFZX 228, no. 259.
15. The distinct status of this speech act is borne out in seventeenth-century trial dossiers, where it
is typically performed by means of a special text-kind, ssyločnye pamjati (“referral memoran-
da”; see 4.4).
16. For other examples in boundary-setting descriptions, see AFZX 109–10, no. 117 (3×); 260, no.
308; AJu 16, no. 8; ASÈI 1: 321, no. 431 (integrated ); 492, no. 595; 2: 197, no. 286; 202,
no. 288; 435, no. 411 (2×); 438; 444, no. 414; 452, no. 418; 455, no. 419 ( and integrated
); 540, no. 493; 3: 76, no. 50 (2×); 220, no. 209. The same usage is found after citations of
documents (AFZX 97, no. 103; ASÈI 1: 235, no. 326).
17. For an additional example, see ASÈI 1: 538–39, no. 628.
18. This pattern is violated in a few cases where some of the parallel tags are conjoined with i
(AFZX 212, no. 248; ASÈI 1: 460, no. 581; 2: 415, no. 404). Other factors may be at work in
these examples.
19. For similar examples, see AFZX 106, no. 114; 116–17, no. 125 (2×); 214, no. 249; ASÈI 1:
247–248, no. 340 (with a choral tag); 288, no. 397; 400, no. 523; 464, no. 583 (2×); 466–67,
no. 584; 469, no. 585; 470–71, no. 586 (2×); 474, no. 587 (2×); 476–77, no. 588 (2×); 478–79,
no. 589 (2×); 481, no. 590 (3×); 483, no. 591 (2×); 485, no. 592 (2×); 487, no. 593 (2×);
488–89, no. 594 (2×); 573–74, no. 651 (3×); 2: 151, no. 229; 321, no. 334; 411, no. 402; 439,
no. 411; 519–20, no. 481; 3: 188, no. 173; 218–19, no. 208; 271, no. 276 (2×); Kaštanov
1970: 411–13 (2×). For cases of multiplex testimony by co-respondents, see ASÈI 1: 398, no.
522; 501, no. 604; 556–57, no. 642.
20. For an exception in which conflicting responses are presented as part of the same speech event,
see ASÈI 2: 407, no. 400. Clearly this is a case of nonevaluation rather than a true counter-
example.
21. For similar cases, see AFZX 110, no. 117; AJu 15, no. 8; ASÈI 1: 248, no. 340; 450, no. 571;
506–7, no. 607; 512, no. 607a; 552, no. 639; 2: 407, no. 400; 452, no. 418; 455, no. 419; 3:
182, no. 172; Gorčakov 1871, appendix: 63–64, no. 4.5; Koreckij 1969: 289, no. 2. For amid
tagged , see ASÈI 1: 248, no. 340; 3: 69, no. 48.
310 R V
22. The standard tag is used in all but two cases, which feature the s(”)kazati (ASÈI 2: 452, no.
418; 455, no. 419; see 4.2).
23. For similar cases, see AFZX 223, no. 258; AJu 16, no. 8; Antonov and Baranov 1997: 118, no.
146; ASÈI 1: 540, no. 628; 2: 263–64, no. 307; 434–35, no. 411; 3: 69, no. 48; 219, no. 208;
235, no. 218; Gorčakov 1871, appendix: 63–64, no. 4.5.
24. See AFZX 109, no. 117; 127, no. 140; 230, no. 259; 259–60, no. 308; ASÈI 1: 528, no. 615;
557, no. 642; 540, no. 628; 2: 313–14, no. 332; 423, no. 406; 469, no. 428; 540, no. 493; 3:
143, no. 105; 221–22, no. 209; 321, no. 334.
25. See Kleimola 1975: 9–10, especially note 46. For examples of trial dossiers, see AJu no. 25, 26,
29; AJuB 1: no. 52.9, 104; D’jakonov (ed.) 1897: no. 57; Kotkov (ed.) 1990: no. 98; N. P.
Lixačev (ed.) 1895: no. 5; Zabelin 1848; and Jakovlev 1943/1970, appendix.
26. For other instances of this procedure, see AFZX 234, no. 261; ASÈI 2: 496–97, no. 458.
27. For the other example, see ASÈI 1: 321, no. 431. Excerpt (20) is is unusual in that the
respondent conducts the boundary-setting rather than his witnesses.
28. See Borkovskij and Kuznecov 1965: 307–8; Nikiforov 1952: 115, 157, 165; Uspenskij
1987: 152–53.
29. For treatments of this kind of formula, see earlier in the same transcript or ASÈI 1: 248, no.
340.
30. Note that the constructions can also occur in non-narrative contexts, where there is no plot.
31. On the definition of speech-act verbs, see Verschueren 1980: 4–5, 1985: 5; Wierzbicka 1987.
32. The heresy trials known from the first half of the sixteenth century have a different character
(Kazakova 1960: 285–318; Pokrovskij (ed.) 1971).
33. Cf. the use of dovoda… ne dovel ‘did not produce an argument’ as a within the judges’
in Antonov and Baranov 1997: 287, no. 296.
34. See GVNP 326, no. 340 (2×); ASÈI 2: 434, no. 411 (in a relative clause).
35. For the other examples, see ASÈI 3: 234, no. 218; GVNP 326, no. 340.
36. For a survey of the vast bibliography on , see Fludernik 1993.
37. Cf. Vološinov 1929/1986: 134; Leech and Short 1981: 325–32; Clark and Gerrig 1990: 788;
Thompson 1996: 514–15.
38. See also Slembrouck 1986: 47–48, 59–60, Page 1988: 33, 35, 37.
39. See Vinogradov and Istrina (eds.) 1960: 428 on ; Doležel 1973: 27 on Czech; Fónagy
1986: 284 on Hungarian; and Bamgbos® e 1986: 78–80, 95 on Yoruba. On in , see also
Vološinov 1929/1986: 138–40, 155–59; Koduxov 1955: 136–38, 140–44; Uspenskij
1973: 33–36; Čumakov 1975: 38–48.
40. See Banfield 1973: 21–22; Doležel 1973: 32–39; McHale 1978: 265; Spitzer 1928/1988.
41. This is not the place to discuss the arguments for a “narratorless” , advocated by Banfield
(1973, 1978, 1982, 1993). See the counterarguments in Fludernik 1993; McHale 1978: 253–57,
1983; McKay 1978: 14–20; Padučeva 1996: 348–49; Toolan 1992: 129–37; Wierzbicka
1974: 294–97.
42. Among the scholars claiming that is exclusively literary are C. Bally and E. Lorck (cited by
Pascal 1977: 13, 18–19), Banfield (1973, 1978, 1982, 1993), and Padučeva (1996: 337).
Vološinov (1929/1986: 156) sees it as primarily a written phenomenon dependent on the
“’silencing’ of prose.” Other investigators have pointed to examples in spoken narratives (Pascal
N 311
1977: 18–19; Ullmann 1957: 98; Haberland 1986: 232–33; Clark and Gerrig 1990: 787;
Hickmann 1993: 67) and in non-literary text kinds such as parliamentary reporting and
journalism (McHale 1978: 282–83; Slembrouck 1986; Short 1988: 72).
43. See, e.g., Kozlovskij 1890: 4 (the first Russian study of ); Vološinov 1929/1986: 139;
Švedova 1952: 113; Bulaxovskij 1954: 442–44; Koduxov 1955: 138, 145, 168; Molotkov
1958: 27.
44. Some of the reports examined here for convenience’ sake are technically not but free
nondirect speech, since they do not contain any forms with person marking. Given that
reporting strategies form a continuum, free nondirect speech does not represent a distinct
category and has the same non-actualizing discourse effects as central .
45. See also Bakhtin 1975/1981: 305–6, 318; Doležel 1973: 35–39, 50; Thompson 1996: 513.
46. For other cases in boundary-setting procedures, see ASÈI 2: 534–35, no. 492; 540, no. 493.
47. For additional counterarguments to the literariness claim, see Fludernik 1993: 4, Polanyi 1982.
Chapter 5
transcripts (GVNP no. 92, 340). It is the predominant verb in AJuB 1: no. 103.1; ASÈI 1: no.
326, 587, 589, 592, 595, 607, 607a, 615, 628; 2: no. 229, 336, 338, 416, 418, 492; 3: no. 288,
364; Kaštanov 1970: no. 9, 27.
9. It occurs to the exclusion of other verbs in AJuB 1: no. 103.2; ASÈI 1: no. 340, 521, 539, 557,
582, 584, 585, 588, 591, 594, 604; 2: no. 285, 286, 306, 307, 332, 333, 337, 374, 383, 401,
402, 404, 406, 407, 414, 458, 465, 493; 3: no. 50, 105, 209, 221, 223, 224, 251, 276, 478;
Kaštanov 1970: no. 6, 16, 40; Koreckij 1969: no. 2; Krotov and Smetanina 1987: no. 5; GVNP
no. 132. It is the predominant verb in ASÈI 1: no. 447, 522, 524, 538, 571, 581; 651; 2: no.
287, 288, 296, 310, 334, 375, 405, 410, 419, 421, 422, 483; 3: no. 208.
10. See AJu no. 12; ASÈI 1: no. 538, 587, 607, 607a; 2: no. 287, 288, 336, 405, 422, 483, 492; 3:
172; Kaštanov 1970: no. 9.
11. Eleven are originals or contemporary copies (ASÈI 1: no. 524, 538, 587, 592, 607, 607a; 2: no.
287, 288, 334, 422; 3: no. 208), and six are later copies (AJuB 1: no. 103.1; ASÈI. 1: no. 595;
2: no. 229, 375, 418, 483).
12. ASÈI 1: no. 447, 571, 581, 615, 628; 2: no. 405, 410, 492; Kaštanov 1970: no. 9, 27.
13. ASÈI 1: no. 522, 2: no. 416, 3: no. 288.
14. ASÈI 1: 235, no. 326; 478, no. 589; 541, no. 628; 3: 386, no. 364.
15. ASÈI 2: 252, no. 296; 455, no. 419. V”prositi is the sole verb of inquiring in the remaining
example (Gorčakov 1871: 51, no. 4.2).
16. For examples, see AFZX 228, no. 259; ASÈI 1: 459, no. 581; 509, no. 607; 514, no. 607a (a
variant of the preceding document); 540, 541, no. 628; 2: 196, no. 286; 202, no. 288; 313, no.
332; 369, no. 374; 497, no. 458; 523, no. 483; 539, no. 493; 3: 291, no. 276; Gorčakov 1871,
appendix (separate pagination): 50, no. 4.1; Kaštanov 1970: 359, no. 9.
17. See AFZX 139, no. 157; 229, no. 259; ASÈI 1: 235, no. 326; 321, no. 431; 450–51, no. 571
(2×); 557, 558, 559–60 (2×), no. 642; 2: 351, no. 358; 2: 364, no. 370 (2×).
18. See ASÈI 2: 431–32, no. 410 (5×); 352 no. 467 (2×); 536, 539–40 no. 493 (3×); 3: 84–85, no.
55; 302, no. 288.
19. See AFZX no. 129 (2×); 229, no. 259; 257, no. 308; Antonov and Baranov 1997: 117, no. 146;
ASÈI 1: 539, no. 628; 2: 420–21, no. 406 (3×); 459, no. 422; 537–38, no. 493 (2×); 3: 141, no.
105.
20. See AFZX 106, no. 114; 109, no. 117; 228, no. 259; ASÈI 1: 481, no. 590; 502, no. 604; 2:
252, no. 296; 372, no. 375; 411–12, no. 402 (2×); 425, no. 407; 434, no. 411; 448, no. 416;
533–34, no. 492 (2×); 3: 240, no. 221; 290, no. 276; and, in a northwestern transcript, GVNP
327, no. 340.
21. For the other cases, see AFZX 231–32, no. 259 (2×); 260, no. 308; ASÈI 1: 434, no. 557.
22. The occurrence of the nonreflexive rek in one case of this formula (ASÈI 1: 474, no. 587) is
probably a mistake for rek”sja, and it is so treated by the editors (ibid.: 475, Note 5). For
further examples of rečisja, see AFZX 117, no. 125; 228, no. 259; ASÈI 1: 248, no. 340; 290,
no. 397; 1: 403–4, no. 525; 414, no. 537; 416, no. 538; 417, no. 539; 465, no. 583; 467, no.
584; 471, no. 586; 477, no. 588; 478, no. 589; 481, no. 590; 483, no. 591; 485, no. 592; 487,
no. 593; 489, no. 594; 493, no. 595; 509, no. 607; 514, no. 607a; 548, no. 635; 2: 197, no. 286;
202, no. 288; 254, no. 296; 314, no. 332; 317, no. 333; 322, no. 334; 392, no. 388 (with jatisja
in its variant, 395, no. 388a); 416, no. 404; 419, no. 405; 455, no. 419 (with jatisja in its
variant, 453, no. 418); 461, no. 422; 501, no. 463; 520, no. 481; 523, no. 483; 535, no. 492;
N 313
541, no. 493; 3: 88, no. 56; 219, no. 208; 222, no. 209; 292, no. 276; 462, no. 477. For further
examples of jatisja, see ibid. 1: 236, no. 326; 400, no. 523; 402, no. 524; 452, no. 571; 469, no.
585 (with rečisja in a variant copy); 502, no. 605; 2: 365, no. 370; 370, no. 374; 373, no. 375.
23. For similar examples, see ASÈI 2: 535, no. 492; 3: 88, no. 56.
24. For further examples with velěti, see AFZX 110, no. 117; 228–29, no. 259 (cited within another
transcript); 232, no. 259; 261, no. 308; ASÈI 2: 194, no. 285. For further examples with učiniti
or dati s“rok”, see ASÈI 1: 560, no. 642; 2: 263, no. 307.
Chapter 6
1. In a few documents, there is no record of the hearing; the verdict of the higher court comes
immediately after the trial copy (AFZX: 98, no. 103; ASÈI 1: 501, no. 463; 2: 392 no. 388;
395, no. 388a. In others, the hearing is recounted without (ibid. 1: 322, no. 431; 2: 370, no.
374; 373, no. 375; 390, no. 450).
2. For the other cases, see AFZX: 232, no. 259; 213, no. 248; ASÈI 1: 401, no. 523; 420, no. 540;
465, no. 583; 467, no. 584; 469, no. 585; 471, no. 586; 475, no. 587; 477, no. 588; 479, no.
589; 481–82, no. 590; 483, no. 591; 486, no. 592; ibid: 489, no. 594; 493, no. 595 (2×); 503,
no. 604; 509, no. 607 (2×); 513–14, no. 607a (2×); 560, no. 642; 575, no. 651; 2: 194, no. 285;
198, no. 286; 202–3, no. 288; 330, no. 337; 314, no. 332; 317, no. 333; 322, no. 334; 440, no.
411; 462, no. 422; 470, no. 428; 520, no. 481; 535, no. 492; 541, no. 492; 3: 70, no. 48; 186,
no. 172 (2×); 189, no. 173; 219, no. 208; 235, no. 218; 269, no. 250; 272, no. 251 (2×); 292,
no. 276; Gorčakov 1871, appendix (separate pagination): 56–57, no. 4.3.
3. For the other document summaries, see ASÈI 1: 465, no. 583; 467, no. 584; 471, no. 586; 477,
no. 588; 479, no. 589; 486, no. 592; 2: 317–18, no. 333; 3: 70–71, no. 48 (tagged with skazal).
For the references to cadastre books, see ibid.: 469, no. 585; 475, no. 587; 481, no. 590; and 483,
no. 591. The transcripts from ASÈI 1 are all records of lawsuits of 1495–99 between Kostroma
peasants and the Trinity-Sergius Monastery’s rent collector Ofonasij; they have the same trial
judge and higher judge. Thus they may reflect the practice of a single scribe or (more probably,
given the orthographic differences) a group of scribes working under the same supervision.
4. See ASÈI 2: 317–18, no. 333 (tagged with tak(o) rek(li)); 3: 70–71, no. 48 (tagged with skazal).
5. See ASÈI 1: 182, no. 253; 287, no. 394; 338, no. 450; 343–44, no. 456; 345, no. 457; 378, no.
499; 442, no. 562; 524, no. 612; 3: 102, no. 67a; 3: 103, no. 68; 138, no. 100.
6. See ASÈI 1: 297, no. 406; 334, no. 446; 2: 404–5, no. 399; 524, no. 483a; 3: 36, no. 19. In two
other referred purchase deeds (Kotkov and Filippova (eds.) 1978: 9, no. 1; GVNP: 243, no.
220), reports in the higher-court record are presented as tagged by the preterit of v”sprositi
and reči. However, the appendices contain no mention of the deeds to which they are attached,
so that they are not really verification records.
7. For the other examples, see ASÈI 1: 402, no. 524; 404, no. 525; 435, no. 557; 3: 85, no. 55; 3:
88, no. 56; 269, no. 250; Kaštanov 1970: 358, no. 9. The first two cases may belong to the pen
of a single scribe; both are copies of originals from 1485–90, with the same respondent, trial
judge, and higher judge.
314 R V
8. See, for example, AJu 28, no. 14; 45, no. 20; 52, no. 22; Ivina (ed.) 1983: 11, no. 3; 51, no. 46;
74, no. 63; N. P. Lixačev (ed.) 1895: 170, no. 7; 231, no. 12; Novosel’skij et al. (eds.) 1975: 45,
no. 40; 48, no. 41; 83, no. 77; 197, no. 194; 234, no. 230; and 261, no. 255.
9. For the falsifications, see ASÈI 1: 400, no. 523; 493–94, no. 595; 509, no. 607 and the variant
from the same trial, 514, no. 607a; 3: 186, no. 172; 269, no. 250; 271, no. 251; Kaštanov
1970: 359, no. 9.
10. See ASÈI 1: 509, no. 607; 514, no. 607a; Kaštanov 1970: 359, no. 9.
11. See also ASÈI 1: 493–94, no. 595; 3: 186, no. 172. In a further instance, the echoes a
performative statement given in (Kaštanov 1970: 359, no. 9).
12. This passage is misinterpreted by the editors of ASÈI. See D. Collins 1994: 360–61 for
discussion.
13. For similar examples, see AFZX 117, no. 125; ASÈI 1: 434–35, no. 557.
14. See AFZX 214, no. 249; ASÈI 1: 320, no. 430; 541, no. 628; 2: 152, no. 229; 318, no. 333;
365, no. 370; 384, no. 383; 408, no. 400; 3: 77, no. 50.
15. Cf. a fully presupposed report of a litigant’s nonappearance made by a surety, which is
presented as complementized (ASÈI 1: 236, no. 326).
Chapter 7
1. For the other cases, see ASÈI 1: 507, no. 607, and the variant transcript from the same trial,
513, no. 607a.
2. A further case of after m”lviti (AFZX 230–31, no. 259) is discussed in 7.3.
3. For s”kazyvati, see AFZX 214, no. 249; ASÈI 2: 316, no. 333; 523, no. 483; 534, no. 492; 537,
no. 493. For the other verbs, see AFZX 222, no. 258 (kazati); ASÈI 1: 397, no. 521 (govoriti);
AFZX 234, no. 261; ASÈI 2: 150, no. 229 (z”vati); ASÈI 1: 459, no. 581; 2: 468–69, no. 428
(nazyvati).
4. The same transcript includes another unique case in which govoriti appears as a phatic device
in the judge’s speech (ASÈI 1: 396, no. 521) — a context in which s”kazati is abundantly
attested.
5. For the other examples, see ASÈI 1: 514, no. 607a and the variant transcript from the same
trial, 509, no. 607; 2: 269–70, no. 310 (2×).
6. For subversive requotations treated in a similar way, see AFZX 231, no. 259 (2×); ASÈI 1: 416,
no. 538; 2: 268–69, no. 310; 316, no. 333; 452, no. 418 and in a variant transcript from the
same trial, 455, no. 419; 503, no. 464; 3: 269, no. 250; Gorčakov 1871, appendix (separate
pagination): 51, no. 4.2. One fragmentary example may involve a statement made prior to the
trial (ASÈI 3: 245, no. 224).
7. The prepositional phrase peredo mnoju co-occurs in a tag clause with the perfective preterit in
another example (ASÈI 1: 557, no. 642), where its precise motivation is unclear.
8. Imperfective preterit tags can be found in narrative contexts when the given report serves
primarily as a frame or background, e.g., when it initates dyads or explains subsequent
nonevents (ASÈI 1: 538, no. 628; 2: 411, no. 402; 3: 173, no. 187; 3: 222, no. 209; Kaštanov
1970: 355, no. 9).
N 315
9. Cf. the layered report (in complementized after s”kazati) that is set in a mini-narrative in
AFZX 116, no. 125. Here the reporter’s purpose is evidently to give background on the
witnesses whom he is about to introduce; that is, the emphasis on the source of the information
rather than the content.
10. See the other example in ASÈI 2: 193, no. 285; ibid. 1: 460, no. 581; Kaštanov 1970: 411–13,
no. 40 (10×). There are also two or possibly three cases of the negated perfective preterit of the
zaperetisja ‘deny’, functioning as collateral information rather than plotline (Krotov and
Smetanina 1987: 68, no. 5).
11. The gerund of the verb rosprositi ‘find out by means of interrogation’ also introduces a
complementized report in a narrative context (ASÈI 2: 369, no. 374). The backgrounding in this
example is to be expected, since it conveys the reason for a subsequent nonevent (a non-
judgment).
12. Cf. the use of the preterit of the imperfective javljati ‘declare’ to tag hearsay of a different kind
(ASÈI 1: 480, no. 590).
13. For the other cases after s”kazyvati, see AFZX 106–7, no. 114; Antonov and Baranov
1997: 117, no. 146; ASÈI 1: 485, no. 592; 2: 197, no. 286 (2×); 199, no. 287; 414–15, no. 404;
3: 141, no. 105 (2×); and, with a left-shifted topic clause intervening between tag and
complementized report, 1: 528, no. 615.
14. For other cases of s”latisja with a report, see the other instance in AFZX 223, no. 258; also
ASÈI 2: 533, no. 492; 3: 141, 143, no. 105 (2×); Koreckij 1969: 289, no. 2. For cases of
cělovati kr’st” with a report, see AFZX 260, no. 308; ASÈI 3: 386, no. 364; Koreckij
1969: 290, no. 2.
15. Likewise, Hanks (1990: 206) treats as “decentered” (indexically displaced from the reporter’s
frame).
16. Six are tagged by s”kazyvati (ASÈI 1: 247, no. 340; 2: 194, no. 285; 3: 182, no. 172; 268–69,
no. 250 (2×); Kaštanov 1970: 359, no. 9), two by z”vati (AFZX 227, 230 no. 259 (2×)), and
one by kazati (AFZX 260, no. 308).
17. Four are tagged by s”kazyvati (AFZX 222, no. 258; ASÈI 2: 315–16, 317, no. 333 (2×);
Kaštanov 1970: 355, no. 9), and one by kazati (ASÈI 2: 438, no. 411).
18. For the other examples, see AFZX 227, 230, no. 259; Kaštanov 1970: 355, no. 9.
19. For the other examples, see ASÈI 2: 194, no. 285; 315–16, no. 333.
20. For the other examples, see AFZX 222, no. 258; 260, no. 308; ASÈI 3: 182, no. 172.
21. The prepositional phrases can be left implicit (ASÈI 1: 235–36, no. 326; 3: 54, no. 31).
22. For other requests for amplification, see ASÈI 1: 235–36, no. 326; 461, no. 582; 541, no. 628; 2:
539, no. 493; 3: 54, no. 31; Kaštanov 1970: 413. For other subversive requotations, see ASÈI 1:
541, no. 628 (with the nazyvati ‘call, claim’); 2: 407, no. 400; 414, no. 404; 3: 54, no. 31.
23. For another excuse for nonfeasance similar to (22), see ASÈI 2: 264, no. 307. For another excuse
for a no-show witness (as well for absence of promised documents), see ibid. 2: 432, no. 410.
24. For other syntheses of content, see ASÈI 1: 193, no. 285 (after zaslyšeti ‘hear-’); 396, no.
521 (after slyxati ‘hear-’); 450, no. 571; 512, no. 607a and in a variant transcript from
the same trial, 507, no. 607; 2: 539, no. 493; 3: 85, no. 55. For examples with interrogatives or
relatives, see ASÈI 1: 502, no. 604; 2: 193, no. 285; Kaštanov 1970: 359, no. 9; 372, no. 16;
Krotov and Smetanina 1987: 68, no. 5 (2×)).
25. See also ASÈI 1: 541, no. 628; 2: 264, no. 307; 432, no. 410; 539, no. 493.
316 R V
26. For sporadic instances of dě(i) in the transcripts investigated, see 4.7, 8.3.3. The particle occurs
once in an additional layer of within the testimony of a poorly preserved trial transcript of
the early sixteenth century (ASÈI 3: 245, no. 224), where it performs a disassociating or even
dubitative function.
27. There can also be ambiguity in main clauses when the tag occurs after a noun phrase that can be
interpreted either as its subject or as the subject of the report clause (e.g., ASÈI 2: 532, no. 492).
28. For further examples in complement clauses, see AFZX 127, no. 140; 212–13, no. 248 (2×);
ASÈI 1: 501, no. 604; 507, no. 607 and in the variant transcript from the same trial, 512, no.
607a; 2: 268–69, no. 310; 3: 291, no. 276. For other cases in relative clauses, see AFZX 230,
no. 259; 260, no. 308; ASÈI 2: 194, no. 285. For an instance in a topicalizing clause, see ASÈI
2: 438–39, no. 411. For examples in main clauses, see ibid.: 369, no. 374; 532, no. 492.
29. Cf. the intercalated tags in complements after iskati ‘seek; sue’ (ASÈI 2: 268–69, no. 310) and
m”lčati ‘be silent; not report’ (ASÈI 1: 507, no. 607; 512, no. 607a).
30. For further examples, see AFZX 120, no. 129; 227, no. 259 (2×); ASÈI 1: 320, no. 430; 398,
no. 522 (2×); 507, no. 607; 513, no. 607a; 2: 151, no. 229 (2×); 369, no. 374 (2×); 371, no.
375; 415, no. 404 (2×); 445, no. 414; 449, no. 416 (2×). The perfective zam”lčati is attested in
ibid.: 1: 247, no. 340.
31. For further examples of biti/bivati čel”m’, see AFZX 227–28, no. 259 (4×); ASÈI 1: 289, no.
397; 396–97, no. 521; 452, no. 571; 460, no. 581; 501, no. 604 (3×); 507, no. 607; 513, no.
607a; 540–41, no. 628 (2×); 2: 193, no. 285; 369, no. 374 (4×); 432, no. 410; 445, no. 414
(2×); 490, no. 450; 500, no. 463; 503, no. 464; 490, no. 450; 537, no. 493. On the use of this
phrasal verb, see also Tarabasova 1963.
32. For some examples with velěti, see AFZX 97, no. 103; 116, no. 125 (2×); 229, no. 259; ASÈI
1: 460, no. 581; 501, no. 604; 507, no. 607 (2×); 512, no. 607a (2×); 539, 541, no. 628 (2×);
2: 280, no. 315 (3×); 439, no. 411; 445, no. 414; 490, no. 450 (2×); 496, no. 458; 500–501, no.
463 (4×); 503, no. 464; 536–37, no. 493 (2×); 3: 54, no. 32; 58, no. 35; 84, no. 55. For ukazati,
see ibid. 3: 461–62, no. 477 (2×). For prikazati ‘command, entrust’, see AFZX 228, no. 259;
Morozov 1988: 304, no. 2.
33. For other cases, see AFZX 97, no. 103; 227–28, no. 259 (2×); ASÈI 1: 193, no. 285; 2: 534,
no. 492 (s”kazati); 1: 193, no. 285; 459, no. 581; 3: 141, no. 105 (s”kazyvati); 1: 418, no. 539;
502, no. 604; 2: 415, no. 404; 541, no. 495 (govoriti/govarivati); Krotov and Smetanina
1987: 68, no. 5 (izgovoriti).
Chapter 8
1. For similar cases, see ASÈI 1: 465, no. 583; 471, no. 586; 475, no. 587; 509, no. 607; 515, no.
607a; 2: 194, no. 285; 198, no. 286; 203, no. 288; and 412, no. 402. No capital letter is
indicated at the beginning of verdicts in three other autographs (ibid. 1: 479, no. 589; 2: 426,
no. 407; 3: 462, no. 477).
2. See, for example, Jakovlev 1943/1970: 321–22, no. 1.5–1.6.
3. Similar cases can be found in northwestern trial transcripts (GVNP 149, no. 92; 327–28, no. 340).
4. For a legal-historical examination of these passages, see Kleimola 1975: 82–83, 86–91.
N 317
5. The ratio decidendi can also be given in a prepositional phrase when judgments are “in
accordance with [po]” documents presented during the trial (e.g., ASÈI 2: 203, no. 288; 314,
no. 332; 3: 143, no. 105). Such prepositional phrases can be treated in parallel with causal
clauses (ibid. 1: 541, no. 628).
6. In a few early transcripts, the rationales incorporate reports of higher-court verifications. Since
none of the attested cases involve denials, the reports were inferrable, if not given, since
verification was a routine. In two other cases, the reports evidently comprise information
presented during the higher hearing but not recorded in the transcript itself (ASÈI 2: 98, no.
286; 541, no. 493).
7. For other examples, see AFZX 98–99, no. 103; 229, no. 259 (quoted in another transcript); 232,
no. 259; ASÈI 1: 248–49, no. 340 (2×); 489–90, no. 594; 503, no. 604; 528–29, no. 615; 554,
no. 640; 2: 194–95, no. 285; 200, no. 287 (2×); 2: 271, no. 310; 370, no. 374; 423, no. 406;
426, no. 407; 458, no. 421; 470, no. 428; 541, no. 493 (2×); 3: 186, no. 172 (2×); 219, no. 208
(4×); 243, no. 223; 245, no. 224; 272, no. 251; 3: 292, no. 276; and 464, no. 478.
8. The report in (6) may not be prototypical , if the deictic adverb nyněča ‘now’ is oriented to
the original speaker’s viewpoint, as opposed to being a generalized present including the time
of the verdict.
9. In addition to (8)–(9), see AFZX 98–99, no. 103; 140, no. 157; Antonov and Baranov
1997: 118, no. 146; ASÈI 1: 460, no. 581; 2: 198, no. 286; 203, no. 288; 3: 77, no. 50; 88, no.
56; 186, no. 172; 240, no. 221; 245, no. 224; 292, no. 276.
10. For the other cases, see AFZX 140, no. 157; Antonov and Baranov 1997: 118, no. 146; ASÈI
3: 71, no. 48. On null anaphora and cointerpretation, see Halliday and Hasan 1976: 13, 142–44;
Brown and Yule 1983: 185, 193; Nichols 1985: 173–74.
11. For the other cases, see ASÈI 3: 77, no. 50; 292, no. 276. Postposition of the subject is also a
factor in one of the coreferential examples — Antonov and Baranov 1997: 118, no. 146.
12. For the other examples, see Antonov and Baranov 1997: 118, no. 146; ASÈI 2: 501, no. 463;
541, no. 493; 3: 71, no. 48; 272, no. 251.
13. On repetitions as cohesive devices, see Halliday and Hasan 1976; Brown and Yule 1983: 194;
Tannen 1989: 50–51; and (with special reference to citations markers) Kieckers 1920: 202.
14. For a similar example involving the preterit of s”kazati ‘say ()’, see ASÈI 3: 71, no. 48.
15. Similarly, in the formula kladusja/kladem”sja na X (lay-..1./ on X- ‘I/we
agree to accept what X says’), the imperfective is replaced by perfective položitisja in the
renarration (e.g., AFZX 140, no. 157).
16. For similar cases, see AFZX 107, no. 114; ASÈI 2: 445, no. 414; 3: 243, no. 223; Gorčakov
1871, appendix (separate pagination): 65, no. 4.5.
17. Other untagged reports of no-shows are found in AFZX 111, no. 117; ASÈI 1: 354, no. 467;
494, no. 595; 2: 318–19, no. 333 (directed and executed verdicts); 3: 462, no. 477.
18. For similar examples, see ASÈI 2: 410, no. 401; 3: 88, no. 56.
318 R V
Chapter 9
Aaron, Uche E.
1992 “Reported speech in Obolo narrative discourse”. In Language in Context:
Essays for Robert E. Longacre [Summer Institute of Linguistics and The
University of Texas at Arlington Publications in Linguistics 107], Shin Ja
J. Hwang and William R. Merrifield (eds.), 227–40. Arlington, Texas:
Summer Institute of Linguistics.
Allingham, Margery
1955/1990 The Estate of the Beckoning Lady. New York: Avon Books.
Andersen, Henning
1989 “Understanding linguistic innovations”. In Language Change: Contributions
to the Study of its Causes, Leiv Egil Breivik and Ernst Håkon Jahr (eds.),
5–27. Berlin: Mouton de Gruyter.
Anpilogov, G. I.
1967 Novye dokumenty o Rossii konca XVI-načala XVII v. Moscow: Izdatel’stvo
Moskovskogo Universiteta.
1977 Nižegorodskie dokumenty XVI veka. Moscow: Izdatel’stvo Moskovskogo
Universiteta.
Antonov, A. V., and K. V. Baranov (eds.)
1997 Akty služilyx zemlevladel’cev XV — načala XVII veka: Sbornik dokumentov.
Vol. 1. Moscow: Arxeografičeskij Centr.
Arnovick, Leslie K.
1999 Diachronic Pragmatics: Seven Case Studies in English Illocutionary Develop-
ment [Pragmatics and Beyond, New Series 68]. Amsterdam: John Benjamins.
Arxeografičeskaja Kommissija
1836 Akty, sobrannye v bibliotekax i arxivax Rossijskoj Imperii Arxeografičeskoju
Èkspedicieju Imperatorskoj Akademii Nauk. 1836. Vol. 1. St. Petersburg:
Arxeografičeskaja Kommissija.
1838 Akty juridičeskie, ili sobranie form starinnogo deloproizvodstva. St. Peters-
burg: Arxeografičeskaja Kommissija.
1841 Akty istoričeskie, sobrannye i izdannye Arxeografičeskoju Kommissieju.
Vol. 1. St. Petersburg: Arxeografičeskaja Kommissija.
320 R V
Battistella, Edwin L.
1990 Markedness: The Evaluative Superstructure of Language. Albany, New
York: State University of New York.
Bauman, Richard
1986 Story, Performance, and Event: Contextual Studies of Oral Narrative.
Cambridge: Cambridge University.
Baynham, Mike
1996 “Direct speech: What’s it doing in non-narrative discourse?” Journal of
Pragmatics 25: 61–81.
Becker, A. P.
1984 “The linguistics of particularity: Interpreting superordination in a Javanese
text”. In Proceedings of the Tenth Annual Meeting of the Berkeley Linguistics
Society, Claudia Brugman and Monica Macaulay (eds.), 425–36. Berkeley:
Berkeley Linguistics Society.
1988 “Language in particular: A lecture”. In Linguistics in Context: Connecting
Observation and Understanding [Advances in Discourse Processes 29],
Deborah Tannen (ed.), 17–35. Norwood, New Jersey: Ablex.
Birnbaum, Henrik
1985 “Orality, literacy, and literature in Old Rus’”. Die Welt der Slaven 30:
161–96.
Blackburn, Simon
1994 The Oxford Dictionary of Philosophy. Oxford: Oxford University.
Bloomfield, Leonard
1933/1984 Language. Chicago: University of Chicago.
Bolinger, Dwight
1972 That’s That [Janua Linguarum, Series Minor 155]. The Hague: Mouton.
Bondar’, I. P.
1967 “Obstojatel’stvennaja svjaz’ nekotoryx glagolov reči v staroslavjanskom i v
drevnerusskom jazykax”. In Očerki po russkomu jazyku i stilistike, 192–201.
Saratov: Saratovskij Gosudarstvennyj Pedagogičeskij Institut.
Borkovskij, V. I.
1981 Sintaksis skazok: Russko-belorusskie paralleli. Moscow: Nauka.
Borkovskij, V. I., and P. S. Kuznecov
1965 Istoričeskaja grammatika russkogo jazyka. Moscow: Nauka.
Brandsma, Frank
1996 “The presentation of direct discourse in Arthurian romance: Changing
modes of performance and reception?” In The Medieval “Opus”: Imitation,
Rewriting, and Transmission in the French Tradition, Douglas Kelly (ed.),
245–60. Amsterdam: Rodopi.
Brecht, Richard D.
1974 “Deixis in embedded structures”. Foundations of Language 11: 489–518.
Brown, Gillian, and George Yule
1983 Discourse Analysis. Cambridge: Cambridge University.
322 R V
Chatman, Seymour
1975 “The structure of narrative transmission”. In Style and Structure in Litera-
ture: Essays in the New Stylistics, Roger Fowler (ed.), 213–57. Ithaca, N. Y.:
Cornell University.
Chvany, Catherine V.
1973/1996 “On ‘root’ and ‘structure-preserving’ — disposable blades for Occam’s
Razor”. In Selected Essays of Catherine V. Chvany, Olga T. Yokoyama
and Emily Klenin (eds.), 104–27.
1985a “Backgrounded perfectives and plot line imperfectives: Toward a theory of
grounding in text”. In The Scope of Slavic Aspect [UCLA Slavic Studies 12],
Michael S. Flier and Alan Timberlake (eds.), 247–73. Columbus, Ohio:
Slavica.
1985b “Foregrounding, ‘transitivity’, saliency (in sequential and non-sequential
prose)”. Essays in Poetics 10: 1–27.
1990/1996 “Verbal aspect, discourse saliency, and the so-called ‘perfect of result’
in Modern Russian”. In Selected Essays of Catherine V. Chvany, Olga
T. Yokoyama and Emily Klenin (eds.), 286–99.
Clark, Herbert H.
1987 “Four dimensions of language use”. In The Pragmatic Perspective: Selected
Papers from the 1985 International Pragmatics Conference [Pragmatics and
Beyond Companion Series 5], Jef Verschueren and Marcella Bertuccelli-
Papi (eds.), 9–25. Amsterdam: John Benjamins.
1992 Arenas of Language Use. Chicago: University of Chicago.
Clark, Herbert H., and Richard J. Gerrig
1990 “Quotations as demonstrations”. Language 66: 764–805.
Clark, Katerina, and Michael Holquist
1984 Mikhail Bakhtin. Cambridge, Massachusetts: Belknap.
Collins, Daniel E.
1994 The Pragmatics of Reported Speech in Medieval Russian Trial Transcripts:
The Contextualization and Interpretation of Metapragmatic Framing Strate-
gies in Old Russian Juridical Language, 1400–1505. Unpublished Ph. D.
Dissertation, UCLA.
1996 “The pragmatics of indirect speech in Old Church Slavonic and other early
Slavic writings”. Studies in Slavic and General Linguistics 23: 21–86.
1997 “The function and social context of proščal’nye gramoty (Toward a thicker
description of a medieval Russian text kind)”. Zeitschrift für Slawistik 42:
192–231.
Collins, James
1987 “Reported speech in Navajo myth-narratives”. In Linguistic Action: Some
Empirical-Conceptual Studies [Advances in Discourse Processes 23], Jef
Verschueren (ed.), 69–84. Norwood, New Jersey: Ablex.
324 R V
Comrie, Bernard
1976 Aspect: An Introduction to the Study of Verbal Aspect and Related Problems.
Cambridge: Cambridge University.
1986 “Tense in indirect speech”. Folia Linguistica 20: 265–96.
Coulmas, Florian
1985a “Direct and indirect speech: General problems and problems of Japanese”.
Journal of Pragmatics 9: 41–63.
1985b “Nobody dies in Shangri-La: Direct and indirect speech across languages”.
Georgetown University Round Table on Languages and Linguistics 1985,
140–151.
1986a “Introduction: Conversational routine”. In Conversational Routine: Explora-
tions in Standardized Communication Situations and Prepatterned Speech,
Florian Coulmas (ed.), 1–17. The Hague: Mouton.
1986b “Reported speech: Some general issues”. In Direct and Indirect Speech
[Trends in Linguistics 31], Florian Coulmas (ed.), 1–28. Berlin: Mouton de
Gruyter.
Coulmas, Florian (ed.)
1986 Direct and Indirect Speech [Trends in Linguistics 31]. Berlin: Mouton de
Gruyter.
Couper-Kuhlen, Elizabeth
1999 “Coherent voicing: On prosody in conversational reported speech”. In
Coherence in Spoken and Written Discourse: How to Create it and How to
Describe it [Pragmatics and Beyond, New Series 63], Wolf Bublitz et al.
(eds.), 11–33. Amsterdam: John Benjamins.
Cram, D. F.
1978 “The syntax of direct quotation”. Cahiers de lexicologie 33: 41–52.
Čumakov, G. M.
1975 Sintaksis konstrukcij s čužoj reč’ju. Kiev: Vyšča Škola.
Derjagin, V. Ja.
1971 “K voprosu ob individual’nom i tradicionnom v delovoj pis’mennosti (na
materiale važskix porjadnyx XVII v.)”. In Russkij jazyk: Istočniki dlja ego
izučenija, S. I. Kotkov (ed.), 151–88. Moscow: Nauka.
1980 “Ob istoriko-stiličeskom issledovanii aktovyx tekstov”. Voprosy
jazykoznanija, 1980, no. 4: 97–107.
Dewey, H. W., and A. M. Kleimola (eds.)
1973 Russian Private Law in the XIV — XVII Centuries: An Anthology of Docu-
ments [Michigan Slavic Materials 9]. Ann Arbor, Michigan: University of
Michigan Department of Slavic Languages and Literatures.
Dirven, Rene, Louis Goosens, Yvan Putseys, and Emma Vorlat
1982 The Scene of Linguistic Action and its Perspectivization by Speak, Talk, Say
and Tell [Pragmatics and Beyond 3 (6)]. Amsterdam: John Benjamins.
R 325
Diver, William
1969 “The system of relevance of the Homeric verb”. Acta Linguistica Hafniensia
12: 45–68.
D’jakonov, M. A.
1897 Akty, otnosjaščiesja k istorii tjaglogo naselenija v Moskovskom gosudarstve.
Part 2: Gramoty i zapisi. Jur’ev: Tip. K. Matisena.
Dmitriev, L. A., and D. S. Lixačev (eds.)
1982 Pamjatniki literatury Drevnej Rusi: Vtoraja polovina XV veka. Moscow:
Xudožestvennaja Literatura.
Doležel, Lubomír
1973 Narrative Modes in Czech Literature. Toronto: University of Toronto.
Dubois, Betty Lou
1989 “Pseudoquotation in current English communication: ‘Hey, she didn’t really
say it’”. Language in Society 18: 343–59.
Du Bois, John W.
1986 “Self-evidence and ritual speech”. In Evidentiality: The Linguistic Coding of
Epistemology [Advances in Discourse Processes 22], Wallace Chafe and
Johanna Nichols (eds.), 313–36. Norwood, New Jersey: Ablex.
Ebert, Karen
1986 “Reported speech in some languages of Nepal”. In Direct and Indirect
Speech [Trends in Linguistics 31], Florian Coulmas (ed.), 145–59. Berlin:
Mouton de Gruyter.
Ehlich, Konrad
1981 “Native Speaker’s heritage: On philology of ‘dead’ languages”. In A
Festschrift for Native Speaker [Janua Linguarum, Series Maior 97], Florian
Coulmas (ed.), 153–65. The Hague: Mouton.
Enkvist, Nils Erik
1987 “A note towards the definition of text strategy”. Zeitschrift für Phonetik,
Sprachwissenschaft und Kommunikationsforschung 40: 19–27.
Fleischman, Suzanne
1990 “Philology, linguistics, and the discourse of the medieval text”. Speculum
65: 19–37.
Fletcher, Giles
1966 Of the Rus Commonwealth. Albert J. Schmidt (ed.). Ithaca, N. Y.: Cornell
University.
Fludernik, Monika
1993 The Fictions of Language and the Languages of Fiction. London: Routledge.
Fónagy, Ivan
1986 “Reported speech in French and Hungarian”. In Direct and Indirect Speech
[Trends in Linguistics 31], Florian Coulmas (ed.), 255–309. Berlin: Mouton
de Gruyter.
326 R V
Forsyth, J.
1970 A Grammar of Aspect: Usage and Meaning in the Russian Verb. Cambridge:
Cambridge University.
Fox, Barbara
1983 “The discourse function of the participle in Ancient Greek”. In Discourse
Perspectives on Syntax, Flora Klein-Andreu (ed.), 23–41. New York:
Academic Press.
Frajzingier, Zygmunt
1991 “The de dicto domain in language”. In Approaches to Grammaticalization
[Typological Studies in Language 19], Elizabeth Closs Traugott and Bernd
Heine (eds.), 1: 219–51. Amsterdam: John Benjamins.
Geertz, Clifford
1973 “Thick description: Toward an interpretive theory of culture”. In Clifford
Geertz, The Interpretation of Cultures, 3–30. New York: Basic Books.
Gennadij
1499/1992 Biblija. Vol. 8. Reprint. Moscow: Moskovskij Patriarxat.
Givón, Talmy
1980 “The binding hierarchy and the typology of complements”. Studies in
Language 4: 333–377.
1984–90 Syntax: A Functional-Typological Introduction. 2 vv. Amsterdam: John
Benjamins.
1989 Mind, Code and Context: Essays in Pragmatics. Hillsdale, New Jersey:
Laurence Erlbaum.
Glock, Naomi
1986 “The use of reported speech in Saramaccan discourse”. In Pragmatics in
Non-Western Perspective [Summer Institute of Linguistics Publications in
Linguistics 73], George Huttar and Kenneth Gregerson (eds.), 35–61.
Arlington, Texas: Summer Institute of Linguistics.
Goffman, Erving
1981 Forms of Talk. Philadelphia: University of Pennsylvania.
Goody, Jack
1977 The Domestication of the Savage Mind. Cambridge: Cambridge University.
Goosens, Louis
1987 “Dealing with linguistic action verbs in depth”. In Linguistic Action: Some
Empirical-Conceptual Studies [Advances in Discourse Processes 23], Jef
Verschueren (ed.), 95–116. Norwood, New Jersey: Ablex.
Gorčakov, M.
1871 O zemel’nyx vladenijax vserossijskix mitropolitov, patriarxov i sv. Sinoda
(988–1738): Iz opytov izsledovanija v istorii russkogo prava. St. Petersburg:
Tip. A. Transhelia.
Green, Georgia M.
1980 “Some wherefores of English inversions”. Language 56: 582–601.
R 327
Havránek, Bohuslav
1932 “Úkoly spisovného jazyka a jeho kultura”. In Spisovná čestina a jazyková
kultura, Bohuslav Havránek and Miloš Weingart (eds.), 32–84. Prague:
Melantrich.
1964 “The functional differentiation of the standard language”. In A Prague
School Reader on Esthetics, Literary Structure, and Style, Paul L. Garvin
(trans.), 3–16. Washington, D. C.: Georgetown University.
Hermon, Gabriella
1979 On the Discourse Structure of Direct Quotation [Center for the Study of
Reading, Technical Report 143]. Cambridge, Massachusetts: Bolt Beranek
and Newman, Inc.
Hickmann, Maya
1993 “The boundaries of reported speech in narrative discourse: Some develop-
mental aspects”. In Reflexive Language: Reported Speech and Metaprag-
matics, John A. Lucy (ed.), 63–90. Cambridge: Cambridge University.
Holt, Elizabeth
1996 “Reporting on talk: The use of direct reported speech in conversation”.
Research on Language and Social Interaction 29: 219–45.
Hopper, Paul J.
1979 “Aspect and foregrounding in discourse”. Syntax and Semantics 12: 213–41.
1988 “Emergent grammar and the a priori grammar postulate”. In Linguistics in
Context: Connecting Observation and Understanding [Advances in Dis-
course Processes 29], Deborah Tannen (ed.), 117–34. Norwood, New
Jersey: Ablex.
Hopper, Paul J., and Sandra A. Thompson
1980 “Transitivity in grammar and discourse”. Language 56: 251–99.
1984 “The Discourse Basis for Lexical Categories in Universal Grammar”.
Language 60: 703–52.
Huizinga, J.
1944/1955 Homo Ludens: A Study of the Play-Element in Culture. Boston: Beacon
Press.
Ickler, Nancy
1977 “Topicalization and relativization in Old Russian”. In Proceedings of the 3rd
Annual Meeting of the Berkeley Linguistics Society, Kenneth Whistler et al.
(eds.), 656–69. Berkeley: Berkeley Linguistics Society.
Issatschenko, Alexander V.
1980 “Russian”. In The Slavic Literary Languages: Formation and Development
[Yale Russian and East European Publications 1], Alexander M. Schenker
and Edward Stankiewicz (eds.), 119–42. New Haven, Connecticut: Yale
Concilium on International and Area Studies.
Ivanov, Vjač. Vs.
1965 Obščeindoevropejskaja, praslavjanskaja i anatolijskaja jazykovye sistemy:
Sravnitel’no-tipologičeskie očerki. Moscow: Nauka.
R 329
Koduxov, V. I.
1955 “Sposoby peredači čužoj reči v russkom jazyke”. Učenye Zapiski Lenin-
gradskogo Gosudarstvennogo Pedagogičeskogo Instituta 104: 107–72.
Koreckij, V. I.
1969 “Vnov’ otkrytye novogorodskie i pskovskie gramoty XIV-XV vv.”. Arxeo-
grafičeskij ežegodnik za 1967 god, 275–90.
Kotkov, S. I. (ed.)
1990 Pamjatniki južnovelikorusskogo narečija: Konec XVI-načalo XVII v. Mos-
cow: Nauka.
Kotkov, S. I., and I. S. Filippova (eds.)
1978 Pamjatniki russkoj pis’mennosti XV — XVI vv.: Rjazanskij kraj. Moscow:
Nauka.
Kotkov, S. I., and Z. D. Popova
1986 Očerki po sintaksisu južnovelikorusskoj pis’mennosti XVII veka. Moscow:
Nauka.
Kozlovskij, Pavel
1890 “O sočetanii predloženij prjamoj i kosvennoj reči v russkom jazyke”.
Filologičeskie zapiski 4–5: 1–7.
Krotov, M. G. and S. I. Smetanina (eds.)
1987 “Istočniki po istorii feodal’nogo zemlevladenija Velikogo knjažestva
Rjazanskogo v XV v.”. Sovetskie arxivy, 1987, no. 1: 66–69.
Kuno, Susumu
1987 Functional Syntax: Anaphora, Discourse and Empathy. Chicago: University
of Chicago.
Kurihara, Takehiko
1985 “The syntax and semantics of quotation reconsidered”. Hokkaidȧo Daigaku
7: 67–89.
Kuz’mina, I. B., and E. V. Nemčenko
1980 “K istorii kratkix form dejstvitel’nyx pričastij v russkom jazyke”. In
Obščeslavjanskij lingvističeskij atlas: Materialy i issledovanija 1978, R. I.
Avanesov (ed.), 151–200. Moscow: Nauka.
Labov, William
1972 “The transformation of experience in narrative syntax”. In William Labov,
Language in the Inner City: Studies in the Black English Vernacular, 354–96.
Philadelphia: University of Pennsylvania.
1981 “Speech actions and reactions in personal narrative”. Georgetown University
Round Table on Languages and Linguistics 1981: 219–47.
Lakoff, Robin Tolmach
1984 “The pragmatics of subordination”. In Proceedings of the Tenth Annual
Meeting of the Berkeley Linguistics Society, Claudia Brugman and Monica
Macaulay (ed.), 481–92. Berkeley.
332 R V
Lixačev, D. S.
1946/1986 “Russkij posol’skij obyčaj XI — XIII vv.”. In Issledovanija po
drevnerusskoj literature, 140–53. Leningrad: Nauka.
1952 Vozniknovenie russkoj literatury. Moscow: Akademija Nauk SSSR.
Lixačev, N. P. (ed.)
1894 Mestničeskie dela 1563–1605 gg. St. Petersburg: Sinodal’naja Tipografija.
1895 Sbornik aktov sobrannyx v arxivax i bibliotekax. St. Petersburg: V. S.
Balašev.
Longacre, Robert E.
1983 The Grammar of Discourse. New York: Plenum.
1985 “Discourse peak as zone of turbulence”. In Beyond the Sentence: Discourse
and Sentential Form, Jessica R. Wirth (ed.), 79–98. Ann Arbor, Michigan:
Karoma.
1994 “The dynamics of reported dialogue in narrative”. Word 45: 125–43.
Lopatina, L. E.
1979 “Sposoby peredači čužoj reči v ramkax složnogo predloženija”. In
Istoričeskaja grammatika russkogo jazyka: Sintaksis: Složnoe predloženie,
V. I. Borkovskij (ed.), 422–47. Moscow: Nauka.
1985 “Sredstva organizacii teksta pri peredače čužoj reči: Na materiale sudebnyx
dokumentov XVII v.”. In Vostočnye slavjane: Jazyki, istorija, kul’tura, J. N.
Karaulov (ed.), 57–64. Moscow: Nauka.
1987 “Zakonomernosti sintaksičeskoj struktury teksta v pamjatnikax narodno-
razgovornoj raznovidnosti russkogo literaturnogo jazyka XVI — XVII vv.”.
In Drevnerusskij literaturnyj jazyk v ego otnošenii k staroslavjanskomu, L. P.
Žukovskaja (ed.), 237–45. Moscow: Nauka.
Lucy, John A.
1993 “Reflexive language and the human disciplines”. In Reflexive Language:
Reported Speech and Metapragmatics, John A. Lucy (ed.), 9–32. Cam-
bridge: Cambridge University.
Lucy, John A. (ed.)
1993 Reflexive Language: Reported Speech and Metapragmatics. Cambridge:
Cambridge University.
Ludolf, Heinrich Wilhelm
1696/1959 Grammatica Russica Oxonii A. D. MDCXCVI. Reprint. B. O. Unbegaun
(ed.). Oxford: Clarendon.
Lyons, John
1968 Introduction to Theoretical Linguistics. Cambridge: Cambridge University.
1977 Semantics. 2 vv. Cambridge: Cambridge University.
Machek, Václav
1968 Etymologický slovník jazyka českého. 2d ed. Prague: Academia.
Martysevič, I. D.
1951 Pskovskaja sudnaja gramota: Istoriko-juridičeskoe issledovanie. Moscow:
Izdatel’stvo Moskovskogo Universiteta.
334 R V
Massamba, David P. B.
1986 “Reported speech in Swahili”. In Direct and Indirect Speech [Trends in
Linguistics 31], Florian Coulmas (ed.), 99–119. Berlin: Mouton de Gruyter.
Mathis, Terrie, and George Yule
1994 Zero quotatives. Discourse Processes 18: 63–76.
Mayes, Patricia
1990 “Quotation in spoken English”. Studies in Language 14: 325–63.
McHale, Brian
1978 “Free indirect discourse: A survey of recent accounts”. PTL: A Journal for
Descriptive Poetics and Theory of Literature 3: 249–87.
McKay, Janet Holmgren
1978 “Some problems in the analysis of point of view in reported discourse”.
Centrum 6: 5–26.
Miller, George A., and Philip N. Johnson-Laird
1976 Language and perception. Cambridge, Massachusetts: Belknap.
Mithun, Marianne
1986 “Evidential diachrony in Northern Iroquoian”. In Evidentiality: The Linguis-
tic Coding of Epistemology [Advances in Discourse Processes 22], Wallace
Chafe and Johanna Nichols (eds.), 89–112. Norwood, New Jersey: Ablex.
Molotkov, A. I.
1958 “K istorii sintaksičeskix konstrukcij dlja peredači čužoj reči v russkom
jazyke: Konstrukcija prjamoj reči”. Učenye Zapiski Leningradskogo
Gosudarstvennogo Universiteta 235: 23–49.
Morawski, Stefan
1970 “The basic functions of quotation”. In Sign, Language, Culture [Janua
Linguarum, Series Maior 1], A. J. Greimas et al. (eds.), 690–705. The
Hague: Mouton.
Morozov, B. N.
1988 “Gramoty XIV — XVI vv. iz kopijnoj knigi Rjazanskogo arxierejskogo
doma”. Arxeografičeskij ežegodnik za 1987 god, 298–309.
Munro, Pamela
1982 “On the transitivity of ‘say’ verbs”. Syntax and Semantics 15: 301–17.
Nichols, Johanna
1980 “The syntax of Old Russian mıæněti (sja)”. In American Indian and Indo-
european Studies: Papers in Honor of Madison S. Beeler [Trends in Linguis-
tics 16], Kathryn Klar, Margaret Langdon, and Shirley Silver (eds.),
421–39. The Hague: Mouton.
1985 “The grammatical marking of theme in literary Russian”. In Issues in
Russian Morphosyntax, Michael S. Flier and Richard D. Brecht (eds.),
170–86. Columbus, Ohio: Slavica.
R 335
P’ecux, Vjačeslav
1990 Novaja moskovskaja filosofija: Roman. In Vjačeslav P’ecux, Ja i pročee,
218–333. Moscow: Xudožestvennaja Literatura.
Peškovskij, A. M.
1956 Russkij sintaksis v naučnom osveščenii. 7th ed. Moscow: Gosudarstvennoe
Učebno-pedagogičeskoe Izdatel’stvo.
Philips, Susan U.
1985 “Reported speech as evidence in an American trial”. In Georgetown
University Round Table on Languages and Linguistics 1985, 154–70.
Pokrovskij, N. N. (ed.)
1971 Sudnye spiski Maksima Greka i Isaka Sobaki. Moscow: Glavnoe Arxivnoe
Upravlenie.
1973 Aktovye istočniki po istorii černosošnogo zemlevladenija v Rossii XIV —
načala XVI v. Novosibirsk: Nauka.
Polanyi, Livia
1982 “Literary complexity in everyday storytelling”. In Spoken and Written
Language: Exploring Orality and Literacy [Advances in Discourse Processes
9], Deborah Tannen (ed.), 155–69. Norwood, New Jersey: Ablex.
Potebnja, A. A.
1899/1958–85 Iz zapisok po russkoj grammatike. 4 vv. in 3 pts. Reprint. V. I.
Borkovskij (ed.). Moscow: Učpedgiz.
Preobraženskaja, M. N.
1983 “Specifika složnyx sintaksičeskix konstrukcij v uslovijax drevnerusskogo
teksta”. In Struktura predloženija v istorii vostočnoslavjanskix jazykov, V. I.
Borkovskij (ed.), 102–8. Moscow: Nauka.
Prince, Gerald
1987 A Dictionary of Narratology. Lincoln, Nebraska: University of Nebraska.
Quine, Willard Van Orman
1960 Word and Object. Cambridge, Massachusetts: M. I. T.
Rader, Margaret
1982 “Context in written language: The case of imaginative fiction”. In Spoken
and Written Language: Exploring Orality and Literacy, Deborah Tannen
(ed.), 185–98. Advances in Discourse Processes, 9. Norwood, New Jersey.
Reddy, Michael J.
1979 “The Conduit Metaphor — a case of frame conflict in our language about
language”. In Metaphor and Thought, Anthony Ortony (ed.), 284–324.
Cambridge: Cambridge University.
Remnick, David
1994 “The exile returns”. The New Yorker, February 14, 64–83.
Rimmon-Kenan, Shlomith
1983 Narrative Fiction: Contemporary Poetics. London: Methuen.
R 337
Rinberg, V. L.
1985 “K probleme konstrukcii prjamoj reči i svjaznogo teksta v istoričeskom
sintaksise”. In Vostočnye slavjane: Jazyki, istorija, kul’tura, J. N. Karaulov
(ed.), 49–57. Moscow: Nauka.
Romaine, Suzanne, and Deborah Lange
1991 The use of like as a marker of reported speech and thought: A case of
grammaticalization in progress. American Speech 66: 227–79.
Roncador, Manfred von
1988 Zwischen direkter und indirekter Rede: Nichtwörtliche direkte Rede, erlebte
Rede, logophorische Konstruktionen und Verwandtes [Linguistiche Arbeiten
192]. Tübingen: M. Niemeyer.
Rosen, Harold
1980 “The dramatic mode”. In Coming to Know, Phillida Salmon (ed.), 152–69.
London: Routledge and Kegan Paul.
Rozental’, D. È.
1988 Russkij jazyk. 2d ed. Moscow: Vysšaja Škola.
Ryle, Gilbert
1966–67/1971 “Thinking and reflecting”. In Gilbert Ryle, Collected Papers, 2:
465–79. New York: Barnes and Noble.
1968/1971 “The thinking of thoughts: What is ‘Le Penseur’ doing?” In Gilbert
Ryle, Collected Papers, 2: 480–96. New York: Barnes and Noble.
Savran, George W.
1988 Telling and Retelling: Quotation in Biblical Narrative. Bloomington: Indiana
University.
Šaxmatov, A. A.
1941 Sintaksis russkogo jazyka. E. S. Istrina (ed.). Leningrad: Učpedgiz.
Sayers, Dorothy L.
1927/1987 Unnatural Death. New York: Harper and Row.
Schiffrin, Deborah
1981 “Tense variation in narrative”. Language 57: 45–62.
Schmücker-Breloer, M.
1987 Peredača prjamoj reči v delovyx tekstax XVII v.: Po materialam arxiva
Požarskix. Istorija russkogo jazyka i lingvističeskoe istočnikovedenie, V. V.
Ivanov and A. I. Sumkina (ed.). Moscow: Nauka.
Schuelke, Gertrude L.
1958 “’Slipping’ in indirect discourse”. American Speech 33: 90–98.
Semino, Elena, Mick Short, and Jonathan Culpeper
1997 “Using a corpus to test a model of speech and thought presentation”.
Poetics 25: 17–43.
Shapiro, Marianne
1984 “How narrators report speech”. Language and Style 17: 67–78.
338 R V
Shippey, T. A.
1993 “Principles of conversation in Beowulfian speech”. In Techniques of
Description: Spoken and Written Discourse, John M. Sinclair, Michael
Hoey, and Gwyneth Fox (eds.), 109–26. London: Routledge.
Short, Michael
1988 “Speech presentation, the novel and the press”. In The Taming of the Text:
Explorations in Language, Literature and Culture, Willie Van Peer (ed.),
61–81. London: Routledge.
Silverstein, Michael
1976 “Shifters, linguistic categories, and cultural description”. In Meaning in
Anthropology, Keith H. Basso and Henry A. Selby (eds.), 11–55. Albuquer-
que, New Mexico: University of New Mexico.
1977 “Cultural prerequisites to grammatical analysis”. Georgetown University
Round Table on Languages and Linguistics 1977, 139–51.
1985 “The culture of language in Chinookan narrative texts: or, on saying that…
in Chinook”. In Grammar Inside and Outside the Clause: Some Approaches
to Theory from the Field, Johanna Nichols and Anthony C. Woodbury (eds.),
132–71. Cambridge: Cambridge University.
1993 “Metapragmatic discourse and metapragmatic function”. In Reflexive
Language: Reported Speech and Metapragmatics, John A. Lucy (ed.),
33–58. Cambridge: Cambridge University.
Sinclair, John McH.
1988 “Mirror for a text”. Journal of English and Foreign Languages 1: 15–44.
Slembrouck, Stef
1986 “Towards a description of speech presentation and speech reportage in
newspaper language”. Studia Germanica Gandensia 9: 44–115.
1992 “The parliamentary Hansard ‘verbatim’ report: The written construction of
spoken discourse”. Language and Literature 1: 101–19.
Sperber, Dan, and Deirdre Wilson
1986 Relevance: Communication and Cognition. Oxford: Blackwell.
Spitzer, Leo
1928/1988 “Pseudo-objective motivation in Charles-Louis Philippe”. In Leo
Spitzer, Representative Essays, Alban K. Forcione, Herbert Linden-
berger, and Madeline Sutherland (eds.), 41–81. Stanford, California:
Stanford University.
Sprinčak, Ja. A.
1960–64 Očerki russkogo istoričeskogo sintaksisa. 2 vv. Kiev: Radjanska Škola.
Sreznevskij, I. I.
1893–1912/1989 Slovar’ drevnerusskogo jazyka. Reprint. 6 vv. Moscow: Kniga.
Starovojtova, O. A.
1988 “O vydelenii granic složnogo sintaktičeskogo celogo v drevnerusskom
tekste: Na primere konstrukcii s prjamoj reč’ju”. Vestnik Leningradskogo
Gosudarstvennogo Universiteta, 2d ser., no. 3: 70–74.
R 339
Sternberg, Meir
1982a “Point of view and the indirections of direct speech. Language and Style 15:
67–117.
1982b “Proteus in quotation-land: Mimesis and the forms of reported discourse”.
Poetics Today 3: 107–56.
Stola, R.
1958 “Zur Frage der Entwicklung des deepričastie im Altrussischen”.
Slavjanskaja filologija: IV Meždunarodnyj s”ezd slavistov, 189–94. Moscow:
Akademija Nauk SSSR.
Sundberg, Hagar
1982 The Novgorod Kabala Books of 1614–1616: Text and Commentary [Stock-
holm Slavic Studies 14]. Stockholm: Almqvist and Wiksell International.
Švedova, N. Ju
1952 “K voprosu ob obščenarodnom i individual’nom v jazyke pisatelja”.
Voprosy jazykoznanija 1952, no. 2: 104–25.
Talmy, Leonard
1978 “Figure and ground in complex sentences”. In Universals of Human Lan-
guage, Joseph H. Greenberg (ed.), vol. 4: Syntax, 627–49. Stanford,
California: Stanford University.
Tannen, Deborah
1983 “’I take out the rock — dok!’: How Greek women tell about being molest-
ed (and create involvement)”. Anthropological Linguistics 25: 359–74.
1986 “Introducing constructed dialogue in Greek and American conversational
and literary narrative”. In Direct and Indirect Speech [Trends in Linguistics
31], Florian Coulmas (ed.), 311–32. Berlin: Mouton de Gruyter.
1989 Talking Voices: Repetition, Dialogue, and Imagery in Conversational
Discourse [Studies in Interactional Sociolinguistics 6]. Cambridge: Cam-
bridge University.
Tarabasova, N. I.
1963 “Ob odnom frazeologizme v častnoj perepiske XVII veka”. In Issledovanija
po lingvističeskomu istočnikovedeniju, S. I. Kotkov and O. A. Knjazevskaja
(eds.), 144–55. Moscow: Nauka.
1964 “O nekotoryx osobennostjax jazyka delovoj pis’mennosti”. In
Istočnikovedenie i istorija russkogo jazyka, S. I. Kotkov and V. F. Dubrovina
(ed.), 157–72. Moscow: Nauka.
Thompson, Geoff
1996 “Voices in the text: Discourse perspectives on language reports”. Applied
Linguistics 17: 501–30.
Timberlake, Alan
1975 “Hierarchies in the genitive of negation”. Slavic and East European Journal
19: 123–38.
Tixomirov, M. N., and A. V. Murav’ev
1982 Russkaja paleografija. 2d ed. Moscow: Vysšaja Škola.
340 R V
Tokarev, S. A. (ed.)
1970 Materialy po istorii Jakutii XVII veka (Dokumenty jasačnogo sbora). Vol. 3.
Moscow: Nauka.
Tolstoj, L. N.
1978 Vojna i mir. Vol. 1–2. Moscow: Xudožestvennaja Literatura.
Toolan, Michael J.
1990 The Stylistics of Fiction: A Literary-Linguistic Approach. London: Routledge.
Townsend, David J., and Thomas G. Bever
1977 Main and subordinate clauses: A study in figure and ground. Bloomington:
Indiana University Linguistics Club.
Ullmann, Stephen
1957 Style in the French Novel. Cambridge: Cambridge University.
Uspenskij [Uspensky], B. A.
1973 A Poetics of Composition: The Structure of the Artistic Text and Typology of
a Compositional Form. Valentina Zavarin and Susan Wittig (trans.).
Berkeley: University of California.
1984 “The language situation and linguistic consciousness in Muscovite Rus’:
The perception of Church Slavic and Russian” [California Slavic Studies
12]. In Medieval Russian Culture, Henrik Birnbaum and Michael Flier
(eds.), 365–85. Los Angeles: University of California.
1987 Istorija russkogo literaturnogo jazyka (XI — XVII vv.) [Sagners Slavistische
Sammlung 12]. Munich: Otto Sagner.
Valk, S. N. (ed.)
1949 Gramoty Velikogo Novgoroda i Pskova. Moscow: Akademija Nauk SSSR.
van der Walle, Lieve
1993 Pragmatics and Classical Sanskrit: A Pilot Study in Linguistic Politeness
[Pragmatics and Beyond, New Series 28]. Amsterdam: John Benjamins.
van der Wurff, Wim
1997 “Reported discourse: A bibliography”. Unpublished.
Vasmer, Max
1953–58 Russisches etymologisches Wörterbuch. 3 vv. Heidelberg: Carl Winter.
Verschueren, Jef
1980 On Speech Act Verbs [Pragmatics and Beyond 4]. Amsterdam: John
Benjamins.
1985 What People Say They Do with Words: Prolegomena to an Empirical Con-
ceptual Approach to Linguistic Action [Advances in Discourse Processes 14].
Norwood, New Jersey: Ablex.
1987 “Metapragmatics and universals of linguistic action”. In Linguistic Action:
Some Empirical-Conceptual Studies [Advances in Discourse Processes 23],
Jef Verschueren (ed.), 125–40. Norwood, New Jersey: Ablex.
Verschueren, Jef (ed.)
1987 Linguistic Action: Some Empirical-Conceptual Studies [Advances in Dis-
course Processes 23]. Norwood, New Jersey: Ablex.
R 341
Zabelin, Ivan
1848 “Sysknoe delo o ssore meževyx sudej, stol’nika Knjazja Vasil’ja Bol’šogo
Romodanovskogo i dvorjanina Lariona Sumina”. Čtenija v Imperatorskom
Obščestve Istorii i Drevnostej Rossijskix pri Moskovskom Universitete 3 (7):
85–122.
Zaliznjak, A. A.
1986a “Novgorodskie berestjanye gramoty s lingvističeskoj točki zrenija”. In
Novgorodskie gramoty na bereste: Iz raskopok 1977–1983 gg., V. L. Janin
and A. A. Zaliznjak (eds.), 89–219. Moscow: Nauka.
1986b “Slovoukazatel’ k berestjanym gramotam”. In Novgorodskie gramoty na
bereste: Iz raskopok 1977–1983 gg., V. L. Janin and A. A. Zaliznjak (eds.),
260–306. Moscow: Nauka.
1990 “*Ogospodin””. In Voprosy kibernetiki: Jazyk logiki i logika jazyka, Vjač.
Vs. Ivanov (ed.), 6–25. Moscow: Akademija Nauk SSSR.
1995 Drevnenovgorodskij dialekt. Moscow: Jazyki Russkoj Kul’tury.
Zimin, A. A. (ed.)
1953 Pamjatniki prava feodal’no-razdroblennoj Rusi: XII — XV vv. [Pamjatniki
Russkogo Prava 2]. Moscow: Gosudarstvennoe Izdatel’stvo Juridičeskoj
Literatury.
Zipf, George Kingsley
1949 Human Behavior and the Principle of Least Effort. Cambridge, Massachu-
setts: Addison-Wesley.
Zwicky, Arnold M.
1974 “Bibliography IV: Direct and indirect discourse”. Ohio State University
Working Papers in Linguistics 17: 198–205.
Appendix: Text-kind and date of
the investigated trial transcripts
In the list below, the trial transcripts used in the study are listed by source (see the
References). The classifications of the text-kind and the dates are taken from the
editions; I have indicated “with referral” or “with verdict” in certain cases whose
classification does not correspond to the text-kind descriptions given in Chapter 2.
Where the documents are not originals, the dates of the extant copies, when given in
the editions, are provided in square brackets after the posited dates of the protographs.
AFZX 1
103 Pravaja gramota (with referral, 1462–64 [1525–50]
114 Pravaja gramota, end of the 15th century [1525–50]
117 Pravaja gramota (with referral, 1498 [1525–50]
125 Pravaja gramota (with referral, 1462–1505 [1525–50]
129 Pravaja gramota, 1495–99 [1525–50]
140 Pravaja gramota, end of the 15th century [1525–50]
157 Pravaja gramota, 1499 [1525–50]
204 Pravaja gramota, 1492 [1525–50]
248 Pravaja gramota (with referral, 1493–94
249 Pravaja gramota (with referral, 1473–89 [1525–50]
254 Pravaja gramota, ca. 1501–2 [1525–50]
258 Pravaja gramota, ca. 1501–2 [1525–50]
259 Pravaja gramota (with referral, 1498 [1525–50]
261 Pravaja gramota (with referral, 1473–89 [1525–50]
308 Sudnyj spisok, 1498–99 [1525–50]
Antonov and Baranov 1997
145 Pravaja gramota, 1500–1501 [1628]
296 Pravaja gramota, 1490–94 [1665]
ASÈI 1
326 Sudnyj spisok, 1462–73 [mid-16th century]
340 Dokladnoj sudnyj spisok, ca. 1464–78
344 R V
A Bloomfield, Leonard 23
Aaron, Uche E. 14, 49 Bolinger, Dwight 259
Allingham, Margery 55 Bondar’, I. P. 76, 301
Andersen, Henning 16 Borkovskij, V. I. 304, 308, 310
Anpilogov, G. I. 307, 318 Brandsma, Frank 44
Antonov, A. V. 152, 184, 213, 309, 310, Brecht, Richard D. 182
311, 312, 315, 317 Brown, Gillian 102, 155, 253, 264, 317
Aristotle 65 Brown, Penelope 3, 20, 25, 38, 69, 86,
Arnovick, Leslie K. 303 287, 306, 307
Austen, Jane 135 Brown, Roger 303
Austin, J. L. 3, 249, 277 Bulaxovskij, L. A. 136, 304, 311
Avanesov, R. I. 28
Avvakum, Protopope 56 C
Caldas-Coulthard, Carmen Rosa 52, 70
B Cassirer, Ernst 65
Bakhtin, M. M. xiv, xv, 1, 2, 4, 10, 11, Čerepnin, L. V. 46, 304, 305, 306
19, 20, 279, 301, 302, 303, 307, Chafe, Wallace 51, 57, 59, 67, 68, 69,
311; see also Vološinov, V. V. 71, 208, 303, 306, 307, 308, 309
Bally, C. 135, 310 Chatman, Seymour 68, 204
Bamgbose, » Ayo 310 Chvany, Catherine V. 8, 43, 45, 91,
Banfield, Ann 50, 133, 135, 304, 310 107, 109, 123, 211, 308, 318
Baranov, K. V. 152, 184, 213, 309, 310, Clark, Herbert H. 13, 52, 53, 55, 67,
311, 312, 315, 317 69, 71, 102, 199, 203, 209, 212,
Barnet, Vladimír 76, 123, 308 306, 307, 308, 310, 311
Battistella, Edwin L. 24 Clark, Katerina 303
Bauman, Richard 306 Collins, Daniel E. xv, 25, 50, 60, 103,
Baynham, Mike 52, 54, 69, 303, 306, 301, 303, 304, 305, 309, 314
308 Collins, James 1, 10, 23, 303
Becker, A. P. 299, 303, 318 Comrie, Bernard 160, 306
Bever, Thomas G. 107, 222, 227 Coulmas, Florian 20, 50, 51, 52, 145,
Birnbaum, Henrik 18, 203 203, 303, 304, 306, 307
Blackburn, Simon 303 Couper-Kuhlen, Elizabeth 12, 169, 307
350 R V
Cram, D. F. 303, 306 Givón, Talmy 19, 77, 116, 304, 308
Culpeper, Jonathan 13, 60, 252, 304 Glock, Naomi 308
Čumakov, G. M. 112, 306, 308, 310 Goffman, Erving 5, 22, 42, 62, 70, 71,
79, 114, 168, 203, 211, 227, 305,
D 307, 308, 311, 318
Derjagin, V. Ja. 25, 318 Goody, Jack 203, 307
Dewey, H. W. 304, 305, 306 Goosens, Louis 10, 304
Dirven, Rene 10, 65, 247, 303 Gorčakov, M. 273, 309, 310, 311, 312,
Diver, William 15, 104, 122, 218 313, 314, 317
D’jakonov, M. A. 310 Green, Georgia M. 7, 43, 66
Dmitriev, L. A. 6 Grekov, B. D. 113, 139, 279, 305
Doležel, Lubomír 135, 142, 305, 310, 311 Grice, Paul 51, 226, 307
Dömötör, A. 12 Grimes, Joseph E. 59, 130, 155, 204,
Du Bois, John W. 71 264
Dubois, Betty Lou 306 Gvozdanovićå, Jadranka 46
Gvozdev, A. N. 306
E
Ebert, Karen 25, 50, 227, 243, 305 H
Eco, Umberto 92 Haberland, Harmut 51, 53, 67, 135,
Ehlich, Konrad 304 145, 304, 306, 307, 311
Enkvist, Nils Erik 303 Hagenaar, Elly 11
Haiman, John 83, 95, 112, 116
F Halliday, M. A. K. 155, 307, 317
Filippova, I. S. 46, 308, 313 Hammer, A. E. 16
Fleischman, Suzanne xvi, 29 Hanks, William F. 25, 306, 315
Fletcher, Giles 290 Hasan, Ruqaiya 155, 307, 317
Fludernik, Monika 2, 5, 35, 48, 52, 53, Havránek, Bohuslav 25
54, 55, 58, 59, 69, 303, 306, 307, Hermon, Gabriella 7, 43
308, 310, 311 Hickmann, Maya 5, 304, 311
Fónagy, Ivan 5, 12, 67, 135, 145, 237, Holquist, Michael 303
303, 304, 306 , 307, 310 Holt, Elizabeth 71, 304
Forsyth, J. 160 Hopper, Paul J. 22, 77, 78, 303, 308
Fox, Barbara 308 Huizinga, J. 308
Frajzingier, Zygmunt 304
Fritz, Gerd 303 I
Ickler, Nancy 37
G Issatschenko, A. V. 287
Geertz, Clifford 3 Istrina, E. S. 15, 304, 306, 309, 310
Gennadij, Archbishop of Novgorod 7 Ivan Ivanovič, Grand Prince 197
Gerrig, Richard 13, 52, 53, 55, 67, 69, Ivan III Vasil’evič, Grand Prince 197
71, 102, 199, 209, 212, 306, 307, Ivan IV Vasil’evič, Tsar’ 191
310, 311 Ivanov, Vjač. Vs. 303, 304
Gilman, Albert 303 Ivina, L. I. 308, 314
N I 351
J L
Jacobs, Andreas 17, 18, 303 La Fontaine, Jean de 135
Jakobson, Roman 76, 303, 318 Labov, William 72, 130, 132, 155, 204,
Jakovlev, A. 29, 156, 310, 316 267
Janssen, Theo A. J. M. 303, 304 Lakoff, Robin Tolmach 6, 54, 68, 72,
Jespersen, Otto 16, 143, 149, 308 132, 135, 211, 243, 289, 291, 307
Johnson-Laird, Philip N. 68, 71 Lange, Deborah 10, 67, 69, 303, 305,
Johnstone, Barbara 303, 309, 318 306, 307, 309
Jucker, Andreas H. 17, 18, 303 Larin, B. A. 306, 308, 318
Juškov, A. I. 318 Larson, Mildred L. 163, 308
Lebsanft, Franz 303
K Lee, Benjamin 135
Kaiser, Daniel H. 113, 139, 192, 207, Leech, Geoffrey N. 4, 51, 70, 102, 103,
219, 222, 271, 276 124, 132, 133, 135, 141, 195, 265,
Karpov, G. O. 65 304, 306, 308, 310
Kaštanov, S. M. 30, 38, 91, 196, 206, Lennard, John 306
208, 211, 213, 219, 232, 243, 304, Leont’ev, A. K. 27, 304, 305
305, 307, 309, 311, 312, 313, 314, Lerch, E. 135
315 Levinson, Stephen C. 3, 20, 25, 38, 69,
Kay, Paul 294 86, 287, 306, 307
Kazakova, N. A. 310 Lewis, David 20, 21, 42, 50, 287, 304
Keller, Rudi 16, 303, 304 Li, Charles 116, 303, 305, 306, 308
Kieckers, E. 265, 317 Liberzon, I. Z. 308
Kiparsky, Paul and Carol 306 Lips, Marguerite 135
Kittay, Jeffrey 30, 179 Lixačev, D. S. 6, 65
Kleimola, Ann M. 29, 30, 78, 99, 112, Lixačev, N. P. 58, 246, 308, 309, 310,
113, 151, 166, 174, 177, 191, 192, 314
195, 198, 210, 271, 276, 279, 304, Longacre, Robert E. 27, 42, 44, 70, 92,
305, 306, 310, 316 102, 103, 114, 172, 173, 209, 211,
Klenin, Emily 311 303, 306
Klewitz, Gabriele 12, 169 Lopatina, L. E. 76, 85, 266, 304, 306,
Koduxov,V. I. 46, 304, 305, 306, 310, 318
311 Lorck, E. 310
Koreckij, V. I. 46, 192, 234, 30, 312, Lucy, John A. 303, 318
315 Ludolf, Heinrich Wilhelm 85
Kotkov, S. I. 46, 306, 308, 310, 313 Lyons, John 17, 307
Kozlovskij, Pavel 311
Krotov, M. G. 137, 139, 140, 312, 315, M
316 Machek, Václav 304
Kuno, Susumu 260 Maksim Grek 85
Kurihara, Takehiko 43 Martysevič, I. D. 113
Kuz’mina, I. B. 308 Massamba, David P. B. 306
Kuznecov, P. S. 308, 310 Mathis, Terrie 92, 95, 97, 307
352 R V
P S
Padučeva, E. V. 11, 134, 136, 141, 142, Savran, George W. 56
149, 310 Šaxmatov, A. A. 304
Page, Norman 6, 35, 46, 67, 102, 107, Sayers, Dorothy L. 2
122, 124, 134, 142, 195, 201, 228, Schallert, Joe 122–23, 124
249, 265, 267, 304, 306, 308, 309, Schiffrin, Deborah 72, 208, 211, 308,
310 309
Palmer, F. R. 221, 305 Schmücker-Breloer, M. 306
Panteleimon, Arximandrit 56 Schuelke, Gertrude L. 13, 304, 308, 309
Partee, Barbara Hall 7, 14, 304, 306 Scott, Sir Walter 135
Pascal, Roy 70, 134, 135, 310 Semino, Elena 13, 60, 252, 304
Pearce-Higgins, Lucinda 14 Shapiro, Marianne 195, 265, 267
P’ecux, Vjačeslav 5–6 Shippey, T. A. 303
Peškovskij, A. M. 304, 306 Short, Michael H. 4, 13, 51, 52, 54, 60,
Philips, Susan 10, 58, 203, 212, 303, 67, 70, 102, 103, 124, 132, 133,
308
N I 353
Asking, verbs of see Inquiring, verbs of subordination 186, 227, 295, 315
Aspect, choice of 6, 117, 206, 215, summaries 140, 186
220–21, 226; see also Imperfective; Backshifting 11, 103, 131, 134
Perfective Bailiffs, reported speech of 194–96
Asyndeton 178 Bessudnye pravye gramoty see Default-
Attitudinal verbs 6 judgment charters
Attribution see Tag formula Binary oppositions 24
Author-based styles see Speaker-based Birchbark letters 13–14
strategies biti/bivati čel”m’ 82–83, 226, 242, 243,
Authorization formula 30, 31 274, 281, 301, 316
Automatized uses 3, 5, 21, 24, 75, 303 Bloodwite 206–7
Autonomous language 294, 295 Boundary-setting descriptions 22, 97,
Averrals 4, 5, 42, 43, 71, 247, 288, 293 235
aže (až’) 62 with addressee indicated 42
compact strategies in 292–93
B commands initiating 167, 172–73
Background information see direct speech in 79, 81, 144, 146, 309
Commentary; Orientations free direct speech in 79, 84, 94–96,
Backgrounding 263, 283 103, 144, 163, 293
amplifications 81 free indirect/nondirect speech in
of attribution 77 143–46, 149, 277, 293, 311
commentary 78, 97, 119, 236, 293 fused reported speech in 79, 117,
corroborations 107, 112, 116, 293 119–20, 122, 293, 310
free direct speech 293 gerund tags in 78–81, 84
free indirect speech 139–40, 141, nondirect speech in 106
143, 148, 149–50, 287, 311 s(”)kazati in 88
fused reported speech 119, 120, 122, brate 86–87, 154–55, 309
123, 234, 235, 236, 293, 296 brat’e 86–87, 154–55, 309
gerunds 78, 82–83, 84, 293, 315 Byplay 155, 311
imperfectives 117, 228, 314
indirect/nondirect speech 107, 112, C
113, 114, 212, 227, 228, 287, 296, Cadastre books 184, 186, 192, 278–79,
315 313
narrative reports of speech acts 126, Capital letters 46, 153–54, 178, 186,
212 245, 311, 316
nominalizations 109 Cataphora 37, 59, 77, 78, 155, 210,
null subjects 8 231, 307
participial clauses 77–78, 84, 123 Causal constructions 12, 141, 252–53,
preposed elements 43, 77–78 256, 263, 274, 295, 317
presupposed elements 43, 113, 116, čelobitnye gramoty see Petitions
122, 195, 226, 234, 296 cělovati kr’st” 225, 315; see also Cross-
reports of less immediate interest kissing oaths
138–40, 141, 287
S I 357
Challenges 42, 58, 104, 188, 193, promoted by repetition 47, 155, 267,
214–15, 216 297, 317
Chancery writing see Legal- and second position 48
administrative writing and tak(o) 59
Choral speech 22, 52, 99, 101–102, in dyadic structure 93–94, 98, 155
152, 155, 180, 182, 183, 186, 191, and thematic continuity 112, 231,
218, 293, 309, 311 233, 246, 257–58, 295
Chronicles xv, 37, 305 and complementizers 257–59, 295
Church Slavonic see Ecclesiastical Collateral information, reported speech
register as 81, 83, 120, 155, 218, 235–36,
Citation particles see Particles, quotative 315
Citations of documents see Quotations of Colons, marking reported speech 13, 46
written records Commands
Classroom discourse 54 framing function of 172
Closing formula 173–74, 191; see also in layered reported speech 204, 241,
End protocols 242–43
Cognition, verbs of 49, 117, 276–77 of judges 33, 154, 159, 160–61, 165,
Coherence 172, 179, 199, 247, 250–52, 316
deictic shifts 46, 107, 114, 150, 155, of litigants 40, 125
205, 227, 263, 264, 296 procedural 125, 127, 149, 165,
direct speech 47, 48, 69, 263–65, 172–75, 194, 198, 251, 292
290, 296 Commas, double 46
free direct speech 103 Commentary, reported speech as 22, 43,
promoted by frames 44, 46, 115, 78, 79, 81, 88, 97, 103, 119, 121,
152–55 277, 293
promoted by judges’ reports 152–55 Commissive speech acts 174; see also
promoted by repetitions 44, 48, 59, Promises
256 Compact strategies 288–90, 297, 318
role of cataphora 59 in boundary-setting descriptions
role of intercalated elements 47–48, 292–93
264–65 in corroborations 293–294
and slipping 264–65 counteracting foregrounding 288, 293,
role of tag formula 15, 44, 62, 68, 318
290 free indirect speech as 150, 289, 295
role of vocatives 47–48 fused reported speech as 288–89, 293
Cohesion and givenness 131, 197, 240, 253,
and gear-shifting 46, 138, 263, 264 254, 258, 293–94, 295
free direct speech 93–94, 95, 98, in judicial-referral records 201,
101, 102 293–94
promoted by intercalated elements in layered reports 243
105, 138, 238–40, 263, 264–66, Narrative reports of speech acts as
267, 270, 297 132, 280, 283, 289, 294
and pacing 201, 293
358 R V
and coherence 47, 48, 69, 263–65, compared with nondirect speech
290, 296 104
as collaboration 69–70 and gear-shifting 96, 107, 119, 124,
in commentary 81 201, 227
complementized 11, 67, 103–4, 136, and givenness 186, 191, 200–201,
199–200, 264, 289 226, 261
in Contemporary Standard Russian 5, with habitual statements 52
14–15 as hearer-based strategy 67–72, 212,
and continuity of theme 96 290
creative character of 2, 57, 65 and historical present 91, 208
and direct experience 69, 102, 212 with hypothetical speech 52
and disassociation of responsibility illocutionary force 53, 68, 72
71, 195, 208–9 and immediacy 66, 297
decentered nature of 25, 107, 227, contrasted with indirect/nondirect
315 speech 11, 103–4, 107, 112, 115,
deixis in 14–15, 35, 66–67, 68, 74, 208, 226, 227, 299–300, 304
95, 107, 114, 132, 134, 150, 196, introduced by alternative verba
211, 218, 227, 262, 263, 288, 295, dicendi 85–92, 208–11, 314
305 introduced by gerunds 76–85, 94,
as demonstration 53, 67, 69, 307 281, 293
diffuse character of 122, 124, 129, introduced by speech-act verbs 5,
131, 132, 150, 172, 186, 201, 208, 83, 85, 86, 161, 205, 301
211, 218, 224, 261, 262, 280, 281, and involvement 66, 68–69
283, 288–90, 294, 295 in judges’ questions 156, 165, 172,
and distancing 71, 212 180–81, 183, 184, 194
in dyads 191, 205–7, 290, 296 in judge’s report 198–201, 294
emotive elements in 35, 67, 134 in layered reports 205–12, 223–24,
and empathy 212 296
and evaluation 68, 69, 71, 212, 224 and main-clause syntax 67
in falsifications 192, 294 in narrative frames 205, 211, 224,
with fictitious statements 52–53, 67, 263, 296
70 at narrative peaks 68, 92, 102, 205,
and foregrounding 25, 43, 68, 76, 78, 211
84, 112, 113, 129, 150, 156, 195, contrasted with narrative reports of
196, 205, 208, 210, 212, 224, 227, speech acts 126–29, 131, 172–73,
228, 265, 296; compared with free 194, 208, 210, 212, 280, 292, 294
indirect speech 138–41, 149–50; nonverbatim uses of 2, 51, 52–58,
compared with fused speech 121, 306, 307
122, 123, 234, 235; compared with and objectivity 71–72
indirect speech 107, 114, 115, 184, in Old Russian 9, 11, 35, 49, 161,
185–86, 211, 218–19, 222, 226, 306
293; compared with narrative ontogenetic priority of 49
reports of speech acts 126; opacity of 306
S I 361
with convergent statements 97, 101 evaluations in 134, 136, 141, 145,
in crossplay 168, 171 277–78, 279, 293
and direct experience 102 and foregrounding, compared with
and gear-shifting 96, 102–103, 163, direct speech 139–40, 141,
293 149–50; compared with narrative
and involvement 92, 102 reports of speech acts 141
with judges’ reports 167–71 as hearer-based strategy 134–35
in multimodal testimony 95, 96–97, in incipits 146–48, 149
99, 102, 103 and indirect speech 134
in multiplex testimony 97–102, 115 in layered reported speech 220, 274,
in naming witnesses 93–94, 120–21, 295–96
130 “literariness” of 133, 135–36,
in Old Russian 301 148–49, 310–11
in opening complaints 38, 92–93, as “modern” development 133,
146 135–36, 149, 288, 299
and pacing 99, 102–3, 163, 164–65, and narrative discourse 134–35
168, 293 with narrative reports of speech acts
at narrative peaks 92, 102 131, 133, 140–41, 149
reading conventions for 4 and narratives of perception 62
and redundancy 101 “narratorless” 310
and tagged direct speech 14–15, 115, in orientations 143, 148, 149
134 and pacing 149–50
with trivial attributions 38, 62, 79, personal preterit in 134, 141, 143
92–93, 95, 97, 102, 163, 205 and perspectival ambiguity 148–49
universality of 9 and polyphony 148–49
and vividness 92, 102 pronouns in 137–38
Free indirect speech (₎ 2, 11, 122, reading conventions for 135, 136,
133–50, 204, 250, 274–79, 288, 145, 293, 299
304, 307, 310, 311 and slipping 138, 139, 141, 146, 250
and backgrounding 139–40, 141, 143, and staging 136–40
148, 149–50, 287, 311 in summaries 140, 148, 150
in boundary-setting descriptions in syntheses 279
143–46, 149, 293, 311 terminology for 133, 136, 145
characteristics of 133–35 tense in 131, 134, 141–42, 143, 148
with cognition predicates 276–78 in testimony 133–50
as compact strategy 150, 289, 295 in transitions 150
in Contemporary Standard Russian with trivial attributions 141, 142–48,
133, 135–36, 310 149, 274, 276–77, 293, 295
in continuations 144, 277 in verdicts 250, 274–79, 295
deixis in 133–34, 135, 139, 143, 144, French 135, 142, 304
149–50 Old xv, 135
as demonstration 307 Fronting 84, 195, 232; see also
emotive elements in 134, 149 Topicalization
S I 365
Interference, authorial/narratorial 53, background 38, 40, 69, 95, 197, 294
69–70, 279, 290; see also Control experiential 4, 20, 66, 97, 145, 171,
of interpretation 286; see also Responsive
Interior monologue 133, 136 understanding
Interludes, accounts of 104–7, 118–19, koe 212
122, 124, 142–43, 150, 173–74,
188, 201 L
Interpersonal meaning 10, 48, 54, 58 Law Codes
Interposed tags see Intercalated tags Pskovian 113
Interpretability see Coherence Muscovite 18, 113, 139, 210, 279,
Interrogation records (rassprosnye reči) 305
50, 141, 152, 156 Layered reports 203–43, 253
Interrogatives 154, 160, 165, 315 and coherence 155
Intervention, effect of 70, 297 compact strategies in 243
Intonational cues to reported speech 4, in direct speech 205–12, 223–24, 296
12, 66 as evaluation 241, 243, 270, 280, 295
Intonational units 46 function of 204–5
Inversions see Word order, inverted in free indirect speech 220, 274,
Invitations 154, 235, 272 295–96
Involvement in fused reported speech 117, 208,
and direct speech 66, 68–69, 210–11 233–37, 243, 296
and free direct speech 92, 102 in indirect speech 209, 212–33, 254,
and historical present 91, 208, 210–11 296, 315
and m”lviti 210–11 intercalated tags in 237–41, 296–97
iskati 317 in judges’ reported speech 161–62
Iteratives 120, 226 as narrative reports of speech acts
iz ust” 56 208, 210, 220, 229, 241–43, 270,
izgovoriti 316 295
izvěčati 205 orientations in 204, 210, 212, 217,
226, 228, 230, 296, 315
J quotative particles in 296–97, 316
jako 59–60 situated 217, 221, 228, 229, 234, 255
recitativum 60 speaker-based strategies in 243, 296
jatisja 174, 190, 243, 312, 313 as universal 203
javiti/javljati 226, 315 unsituated 221, 228, 229
Judgment charters (pravye gramoty) 28, with standard tag 207, 290, 295
29–30, 31, 33, 58, 177, 186, 245, and verdict 248, 253–54, 280, 295
282 vocatives in 48
Least effort, principle of 54
K Legal-administrative writing 8, 25, 28,
kazati (kažet”) 8, 117, 119, 213, 228, 30, 36–37, 39, 45–46, 56, 57, 59,
233, 238, 256, 266, 314, 315 85, 91, 130, 136, 137, 189, 203,
Knowledge
S I 369
as continuum 12–13, 104, 134, 311 Requesting, verbs of 159, 160–61, 272
creative character of 2, 11, 20, 57, 65 Requests
deictically ambiguous see Nondirect for amplification 212–13, 214, 225,
speech 228, 234, 239, 240, 315
grammatically integrated 9, 37, 118, for reaction 213–14, 228
123, 173–74, 189, 194, 201, 208, Requotations see Layered reports
233, 254, 289, 295, 309; see also Residual forms see Prototypical speech
Fused reported speech events, departures from
inexplicitly marked 1, 2, 5, 122, Responsive understanding 3, 20
134–35, 142, 276–78 Retractions see Recantations
as inset in frame 1–2 Ritardando see Pacing
less-than-direct 37, 104, 283, 289, Rjadnaja gramota 46
292, 294, 295, 296, 297; see also Rjazan’ 136, 139
Compact strategies rosprositi 315
mediated by mental images 2–3, 4, rule 249
73–74 Rus’, Northeastern 29, 34, 45, 46, 48,
as object clause 11, 304 91, 245; see also Muscovy
open-endedness of 13, 15 Rus’, Northwestern 13, 29, 45–46, 113,
perceptual autonomy of 1–2; see also 192, 245, 250, 308, 311–312, 316;
Heteroglossia see also Dvina, Novgorod, Pskov
pervasiveness of 1 Russian, Contemporary Standard
as phenomenon of discourse 11, 16 deictically ambiguous reports in 12
as semantic motif 285, 318 direct speech in 5
submerged 124, 174, 248, 249–50, discourse themes in 96, 266
252 free direct speech in 14–15
terminology for xiii, 304 free indirect speech in 133, 135–36,
as universal 1, 5, 9, 49, 203, 303 310
within reported speech see Layered imperfective in 160
reports indirect speech in 5, 11
Reported thought xiii, 4–5, 10, 52, 135, intercalated tags in 8
138, 143 introducers in 5–6, 301
Reportedness 4–5, 12, 109, 122, 133, postposed tags in 8
135, 137–38, 143, 144, 145, 247, reflexives in 118
263–64, 276–77; see also Averrals report-clause types in 5, 9–10, 11,
Reporter-commitment model of direct 233
speech 49, 51–54, 55, 56, 70, 71, Russian, Old
307 direct speech in 9, 11, 35, 49, 161,
Reports, preliminary 29, 178, 180, 194, 306
197–201, 294 discourse theme in 96
Reproaches 162 evolution of reported speech in 288,
Reproductionist approach to direct 301
speech 49, 50–55, 64–65, 104, free direct speech in 301
107, 306
376 R V
fused reported speech in 116–17, 191, 203, 204, 207, 220, 243, 245,
122–23, 237 247, 249, 252–53, 274, 285, 286,
historical present in 91, 309 288, 290, 291, 293, 294, 297, 298,
hypotaxis in 72, 291 304
imperfective in 120 Scriptio continua 46, 76, 205
indirect speech in 9, 11, 103–4 Scripture 56
intercalated elements in 7–8, 9, 76, Self-quotations 10, 210, 218, 227
200, 238, 265–66 sei/sii/ses’ 181, 182, 183, 197; see also
multiple attributions in 9 Demonstratives
narrative norm in 15, 27, 45–46 Semi-uncial (poluustav) 57
narrative reports of speech acts in 9, ses’ see sei/sii/ses’
131 Setting see Orientation
negated commands in 128 Settlements, amicable 28, 37, 145–46,
null subjects in 77, 96 190, 293, 305
parataxis in 291 Settlement charter (rjadnaja gramota) 46
postposed tags in 7–8, 43 sii see sei/sii/ses’
preposed tags in 7, 43 Silence, reports of 209, 226, 242, 256,
punctuation in 46, 305 274, 277, 316; see also
quotative particles in 11, 76 m”lčati/zam”lčati
reflexives in 118 “Silencing of prose” 310
report-clause types in 3, 6–7, 9–10, Simonov Monastery 183
15 skazat’ 5, 85, 300
slipping in 9, 13–14, 200, 264, 304 s(”)kazati/s(”)kazyvati
small clauses in 9–10, 116–17, as default verbum dicendi 5, 88, 300
122–23, 237 displacing reči 85, 115, 300
types of tag in 6, 11, 76, 85, 161, with fused reported speech 117, 120,
301 233, 259
verba dicendi in 85–86, 91 as intercalated tag 268, 317
word order in 37, 43, 77 118 in judicial-referral records 182–84,
187–92, 195, 196, 198–200, 208
S with layered reported speech 212,
Saints’ lives 37 213, 215, 218, 220, 221, 223, 228,
Salience see Foregrounding 229, 233, 238, 314, 315
Salience, plot 123 with “omitted speech” 243, 316
Saliency Hierarchy 318 in other text-kinds 146
say 5, 10, 43 in ratio decidendi 254–56, 259, 266,
Saying, verbs of see Verba dicendi with testimony 87–88, 103, 110, 310
Schematic Language Representation, s”kazatisja 261
Theory of 48 Skazki see Depositions
Scribes 20, 21, 22, 27, 28, 29, 38, 39, s”latisja/pos”latisja 130–31, 225, 270,
40, 42–43, 47, 50, 57–58, 61, 315
71–72, 86, 91, 92, 104, 115, 132, Slavic languages 7, 49, 60, 91, 300–301
136, 146, 158, 177, 181, 182, 189,
S I 377
Slipping 9, 13–14, 67, 264, 304; see in layered reports 209, 213, 215, 225,
also Quotations, incorporated 226, 228, 233, 240–43, 296, 315
and coherence 264–65 in verdicts 34, 248–49, 270–74
of direct speech into (free) indirect Speech acts, indirect 272
137–39 Spokesmen 38–39, 40, 101, 114
in extended reported speech 264 s”prašivati 206
of indirect/nondirect speech into direct s”prošati 160
13–14, 108, 109, 195, 199–200, s”prositi 156–60, 181, 206, 225, 311
263–65; into narrative reports 273 s”rok” 175, 199, 200, 313
and foregrounding 265 Ssyločnye pamjati see Referral
of fused reported speech into direct memoranda
121 Stage-managing, reports of see
of narrative reports into Commands, procedural
indirect/nondirect speech 141, 146, Staging 136, 138–39
251–52, 294–95; into direct speech Starožil’ci see Longtime residents
29, 250, 294 Statejnye spiski see Diplomatic dossiers
and free indirect speech 138 Stopping-points see Episode boundaries
intentionality of 14, 138–39 Streamlining effects 122
slovo v” slovo 55, 57 Style indirect libre see Free indirect
slyxati 221, 228, 229, 233, 315 speech
Small clauses 9, 117, 122, 123, 213, Stylistic factors in reporting 15, 16, 35,
233, 234, 237, 259–60, 288; see 50, 54, 55, 91, 133, 145, 149–50,
also Fused reported speech 157, 165, 289, 302; see also Drift,
so to speak 59 stylistic; Emotive elements;
Socialization, linguistic 1, 5 Narration, covert style of
Song of Prince Igor’s Campaign xv Subjectless constructions 62
Sound-laws 23 Subjects, null 8, 112, 189, 236–37, 246,
Speaker-based strategies 6, 68, 72, 132, 256, 258, 317
211, 243, 280, 283, 290, 294, Subversion
296–97 in layered reports 214, 228, 234, 314,
Speech genres 18, 19, 21, 22, 128, 204, 315
212, 213, 227, 235, 298, 302 in ratio decidendi 247 255, 265, 267,
Speech will 4, 42, 50, 65, 288 279
Speech-act verbs (’s) see also sud” 198, 199
Narrative reports of speech acts Sudebnik see Law Codes, Muscovite
definition 310 Sudnye spiski see Trial records
bound forms after 9–10 Summaries 42, 43, 45, 95, 114, 148,
tagging direct speech 5, 83, 85, 86, 186, 197, 218, 258; see also
161, 205, 301 Syntheses
in judges’ reports 172–75 of documents 116, 186, 288, 313
lexical-reductionist approaches to 15 and free indirect speech 140, 148,
as speaker-based strategy 43, 72, 125 150
and indirect speech 114, 186, 218
378 R V
W
Wackernagel’s position 8, 48, 200, 238
Warnings 205, 207, 210
Wills 189–90
In the PRAGMATICS AND BEYOND NEW SERIES the following titles have been pub-
lished thus far or are scheduled for publication:
1. WALTER, Bettyruth: The Jury Summation as Speech Genre: An Ethnographic Study of
What it Means to Those who Use it. Amsterdam/Philadelphia, 1988.
2. BARTON, Ellen: Nonsentential Constituents: A Theory of Grammatical Structure and
Pragmatic Interpretation. Amsterdam/Philadelphia, 1990.
3. OLEKSY, Wieslaw (ed.): Contrastive Pragmatics. Amsterdam/Philadelphia, 1989.
4. RAFFLER-ENGEL, Walburga von (ed.): Doctor-Patient Interaction. Amsterdam/Phila-
delphia, 1989.
5. THELIN, Nils B. (ed.): Verbal Aspect in Discourse. Amsterdam/Philadelphia, 1990.
6. VERSCHUEREN, Jef (ed.): Selected Papers from the 1987 International Pragmatics Con-
ference. Vol. I: Pragmatics at Issue. Vol. II: Levels of Linguistic Adaptation. Vol. III: The
Pragmatics of Intercultural and International Communication (ed. with Jan Blommaert).
Amsterdam/Philadelphia, 1991.
7. LINDENFELD, Jacqueline: Speech and Sociability at French Urban Market Places. Am-
sterdam/Philadelphia, 1990.
8. YOUNG, Lynne: Language as Behaviour, Language as Code: A Study of Academic English.
Amsterdam/Philadelphia, 1990.
9. LUKE, Kang-Kwong: Utterance Particles in Cantonese Conversation. Amsterdam/Phila-
delphia, 1990.
10. MURRAY, Denise E.: Conversation for Action. The computer terminal as medium of
communication. Amsterdam/Philadelphia, 1991.
11. LUONG, Hy V.: Discursive Practices and Linguistic Meanings. The Vietnamese system of
person reference. Amsterdam/Philadelphia, 1990.
12. ABRAHAM, Werner (ed.): Discourse Particles. Descriptive and theoretical investigations
on the logical, syntactic and pragmatic properties of discourse particles in German. Amster-
dam/Philadelphia, 1991.
13. NUYTS, Jan, A. Machtelt BOLKESTEIN and Co VET (eds): Layers and Levels of Repre-
sentation in Language Theory: a functional view. Amsterdam/Philadelphia, 1990.
14. SCHWARTZ, Ursula: Young Children’s Dyadic Pretend Play. Amsterdam/Philadelphia,
1991.
15. KOMTER, Martha: Conflict and Cooperation in Job Interviews. Amsterdam/Philadel-
phia, 1991.
16. MANN, William C. and Sandra A. THOMPSON (eds): Discourse Description: Diverse
Linguistic Analyses of a Fund-Raising Text. Amsterdam/Philadelphia, 1992.
17. PIÉRAUT-LE BONNIEC, Gilberte and Marlene DOLITSKY (eds): Language Bases ...
Discourse Bases. Amsterdam/Philadelphia, 1991.
18. JOHNSTONE, Barbara: Repetition in Arabic Discourse. Paradigms, syntagms and the
ecology of language. Amsterdam/Philadelphia, 1991.
19. BAKER, Carolyn D. and Allan LUKE (eds): Towards a Critical Sociology of Reading
Pedagogy. Papers of the XII World Congress on Reading. Amsterdam/Philadelphia, 1991.
20. NUYTS, Jan: Aspects of a Cognitive-Pragmatic Theory of Language. On cognition, func-
tionalism, and grammar. Amsterdam/Philadelphia, 1992.
21. SEARLE, John R. et al.: (On) Searle on Conversation. Compiled and introduced by
Herman Parret and Jef Verschueren. Amsterdam/Philadelphia, 1992.
22. AUER, Peter and Aldo Di LUZIO (eds): The Contextualization of Language. Amster-
dam/Philadelphia, 1992.
23. FORTESCUE, Michael, Peter HARDER and Lars KRISTOFFERSEN (eds): Layered
Structure and Reference in a Functional Perspective. Papers from the Functional Grammar
Conference, Copenhagen, 1990. Amsterdam/Philadelphia, 1992.
24. MAYNARD, Senko K.: Discourse Modality: Subjectivity, Emotion and Voice in the Japa-
nese Language. Amsterdam/Philadelphia, 1993.
25. COUPER-KUHLEN, Elizabeth: English Speech Rhythm. Form and function in everyday
verbal interaction. Amsterdam/Philadelphia, 1993.
26. STYGALL, Gail: Trial Language. A study in differential discourse processing. Amsterdam/
Philadelphia, 1994.
27. SUTER, Hans Jürg: The Wedding Report: A Prototypical Approach to the Study of Tradi-
tional Text Types. Amsterdam/Philadelphia, 1993.
28. VAN DE WALLE, Lieve: Pragmatics and Classical Sanskrit. Amsterdam/Philadelphia,
1993.
29. BARSKY, Robert F.: Constructing a Productive Other: Discourse theory and the convention
refugee hearing. Amsterdam/Philadelphia, 1994.
30. WORTHAM, Stanton E.F.: Acting Out Participant Examples in the Classroom. Amster-
dam/Philadelphia, 1994.
31. WILDGEN, Wolfgang: Process, Image and Meaning. A realistic model of the meanings of
sentences and narrative texts. Amsterdam/Philadelphia, 1994.
32. SHIBATANI, Masayoshi and Sandra A. THOMPSON (eds): Essays in Semantics and
Pragmatics. Amsterdam/Philadelphia, 1995.
33. GOOSSENS, Louis, Paul PAUWELS, Brygida RUDZKA-OSTYN, Anne-Marie SIMON-
VANDENBERGEN and Johan VANPARYS: By Word of Mouth. Metaphor, metonymy and
linguistic action in a cognitive perspective. Amsterdam/Philadelphia, 1995.
34. BARBE, Katharina: Irony in Context. Amsterdam/Philadelphia, 1995.
35. JUCKER, Andreas H. (ed.): Historical Pragmatics. Pragmatic developments in the history
of English. Amsterdam/Philadelphia, 1995.
36. CHILTON, Paul, Mikhail V. ILYIN and Jacob MEY: Political Discourse in Transition in
Eastern and Western Europe (1989-1991). Amsterdam/Philadelphia, 1998.
37. CARSTON, Robyn and Seiji UCHIDA (eds): Relevance Theory. Applications and impli-
cations. Amsterdam/Philadelphia, 1998.
38. FRETHEIM, Thorstein and Jeanette K. GUNDEL (eds): Reference and Referent Accessi-
bility. Amsterdam/Philadelphia, 1996.
39. HERRING, Susan (ed.): Computer-Mediated Communication. Linguistic, social, and
cross-cultural perspectives. Amsterdam/Philadelphia, 1996.
40. DIAMOND, Julie: Status and Power in Verbal Interaction. A study of discourse in a close-
knit social network. Amsterdam/Philadelphia, 1996.
41. VENTOLA, Eija and Anna MAURANEN, (eds): Academic Writing. Intercultural and
textual issues. Amsterdam/Philadelphia, 1996.
42. WODAK, Ruth and Helga KOTTHOFF (eds): Communicating Gender in Context. Am-
sterdam/Philadelphia, 1997.
43. JANSSEN, Theo A.J.M. and Wim van der WURFF (eds): Reported Speech. Forms and
functions of the verb. Amsterdam/Philadelphia, 1996.
44. BARGIELA-CHIAPPINI, Francesca and Sandra J. HARRIS: Managing Language. The
discourse of corporate meetings. Amsterdam/Philadelphia, 1997.
45. PALTRIDGE, Brian: Genre, Frames and Writing in Research Settings. Amsterdam/Phila-
delphia, 1997.
46. GEORGAKOPOULOU, Alexandra: Narrative Performances. A study of Modern Greek
storytelling. Amsterdam/Philadelphia, 1997.
47. CHESTERMAN, Andrew: Contrastive Functional Analysis. Amsterdam/Philadelphia,
1998.
48. KAMIO, Akio: Territory of Information. Amsterdam/Philadelphia, 1997.
49. KURZON, Dennis: Discourse of Silence. Amsterdam/Philadelphia, 1998.
50. GRENOBLE, Lenore: Deixis and Information Packaging in Russian Discourse. Amster-
dam/Philadelphia, 1998.
51. BOULIMA, Jamila: Negotiated Interaction in Target Language Classroom Discourse. Am-
sterdam/Philadelphia, 1999.
52. GILLIS, Steven and Annick DE HOUWER (eds): The Acquisition of Dutch. Amsterdam/
Philadelphia, 1998.
53. MOSEGAARD HANSEN, Maj-Britt: The Function of Discourse Particles. A study with
special reference to spoken standard French. Amsterdam/Philadelphia, 1998.
54. HYLAND, Ken: Hedging in Scientific Research Articles. Amsterdam/Philadelphia, 1998.
55. ALLWOOD, Jens and Peter Gärdenfors (eds): Cognitive Semantics. Meaning and cogni-
tion. Amsterdam/Philadelphia, 1999.
56. TANAKA, Hiroko: Language, Culture and Social Interaction. Turn-taking in Japanese
and Anglo-American English. Amsterdam/Philadelphia, 1999.
57 JUCKER, Andreas H. and Yael ZIV (eds): Discourse Markers. Descriptions and theory.
Amsterdam/Philadelphia, 1998.
58. ROUCHOTA, Villy and Andreas H. JUCKER (eds): Current Issues in Relevance Theory.
Amsterdam/Philadelphia, 1998.
59. KAMIO, Akio and Ken-ichi TAKAMI (eds): Function and Structure. In honor of Susumu
Kuno. 1999.
60. JACOBS, Geert: Preformulating the News. An analysis of the metapragmatics of press
releases. 1999.
61. MILLS, Margaret H. (ed.): Slavic Gender Linguistics. 1999.
62. TZANNE, Angeliki: Talking at Cross-Purposes. The dynamics of miscommunication.
2000.
63. BUBLITZ, Wolfram, Uta LENK and Eija VENTOLA (eds.): Coherence in Spoken and
Written Discourse. How to create it and how to describe it.Selected papers from the Interna-
tional Workshop on Coherence, Augsburg, 24-27 April 1997. 1999.
64. SVENNEVIG, Jan: Getting Acquainted in Conversation. A study of initial interactions.
1999.
65. COOREN, François: The Organizing Dimension of Communication. 2000.
66. JUCKER, Andreas H., Gerd FRITZ and Franz LEBSANFT (eds.): Historical Dialogue
Analysis. 1999.
67. TAAVITSAINEN, Irma, Gunnel MELCHERS and Päivi PAHTA (eds.): Dimensions of
Writing in Nonstandard English. 1999.
68. ARNOVICK, Leslie: Diachronic Pragmatics. Seven case studies in English illocutionary
development. 1999.
69. NOH, Eun-Ju: The Semantics and Pragmatics of Metarepresentation in English. A rel-
evance-theoretic account. 2000.
70. SORJONEN, Marja-Leena: Recipient Activities Particles nii(n) and joo as Responses in
Finnish Conversation. n.y.p.
71. GÓMEZ-GONZÁLEZ, María Ángeles: The Theme-Topic Interface. Evidence from Eng-
lish. 2001.
72. MARMARIDOU, Sophia S.A.: Pragmatic Meaning and Cognition. 2000.
73. HESTER, Stephen and David FRANCIS (eds.): Local Educational Order. Ethno-
methodological studies of knowledge in action. 2000.
74. TROSBORG, Anna (ed.): Analysing Professional Genres. 2000.
75. PILKINGTON, Adrian: Poetic Effects. A relevance theory perspective. 2000.
76. MATSUI, Tomoko: Bridging and Relevance. 2000.
77. VANDERVEKEN, Daniel and Susumu KUBO (eds.): Essays in Speech Act Theory. n.y.p.
78. SELL, Roger D. : Literature as Communication. The foundations of mediating criticism.
2000.
79. ANDERSEN, Gisle and Thorstein FRETHEIM (eds.): Pragmatic Markers and Propo-
sitional Attitude. 2000.
80. UNGERER, Friedrich (ed.): English Media Texts – Past and Present. Language and textual
structure. 2000.
81. DI LUZIO, Aldo, Susanne GÜNTHNER and Franca ORLETTI (eds.): Culture in Commu-
nication. Analyses of intercultural situations. n.y.p.
82. KHALIL, Esam N.: Grounding in English and Arabic News Discourse. 2000.
83. MÁRQUEZ REITER, Rosina: Linguistic Politeness in Britain and Uruguay. A contrastive
study of requests and apologies. 2000.
84. ANDERSEN, Gisle: Pragmatic Markers and Sociolinguistic Variation. A relevance-theoretic
approach to the language of adolescents. 2001.
85. COLLINS, Daniel E.: Reanimated Voices. Speech reporting in a historical-pragmatic per-
spective. 2001.
86. IFANTIDOU, Elly: Evidentials and Relevance. n.y.p.
87. MUSHIN, Ilana: Evidentiality and Epistemological Stance. Narrative Retelling. n.y.p.
88. BAYRAKTAROGLU, Arin and Maria SIFIANOU (eds.): Linguistic Politeness Across
Boundaries. Linguistic Politeness Across Boundaries. n.y.p.
89. ITAKURA, Hiroko: Conversational Dominance and Gender. A study of Japanese speakers
in first and second language contexts. n.y.p.
90. KENESEI, István and Robert M. HARNISH (eds.): Perspectives on Semantics, Pragmatics,
and Discourse. A Festschrift for Ferenc Kiefer. 2001.
91. GROSS, Joan: Speaking in Other Voices. An ethnography of Walloon puppet theaters. n.y.p.
92. GARDNER, Rod: When Listeners Talk. n.y.p.
93. BARON, Bettina and Helga KOTTHOFF (eds.): Gender in Interaction. Perspectives on
feminity and masculinity in ethnography and discourse. n.y.p.