You are on page 1of 19

Trustees of Indiana University

Anthropological Linguistics
Applied Sociolinguistics: The Case of Arabic as a Second Language
Author(s): Richard W. Schmidt
Source: Anthropological Linguistics, Vol. 28, No. 1 (Spring, 1986), pp. 55-72
Published by: The Trustees of Indiana University on behalf of Anthropological Linguistics
Stable URL: http://www.jstor.org/stable/30027945
Accessed: 19-01-2016 16:05 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

Trustees of Indiana University and Anthropological Linguistics are collaborating with JSTOR to digitize, preserve and
extend access to Anthropological Linguistics.

http://www.jstor.org

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

APPLIED SOCIOLINGUISTICS: THE CASE OF ARABIC


AS A SECONDLANGUAGE
Richard W. Schmidt
of Hawaii at Manoa
University

Introduction.
Fishman (1970) has observed that an ap1.
is in order whenever new language varieplied sociolinguistics
ties must be developed or whenever language varieties,
old or
be
deals
with
the
second
This
of these
new, must
paper
taught.
and is further limited to the case of
issues,
language teaching,
in the present case Egyptian Arabic
teaching second languages,
(the Arabic of Cairo) as a second language to native speakers of
to the question of English as a
English, with some attention
I will further
second language for native speakers of Arabic.
limit myself to "second" rather than "foreign" language teaching.
describes
the conventions,
Since sociolinguistics
patterns and
constraints
which together comprise native speakers'
knowledge of
what constitutes
appropriate
speech behavior in the speech communities
of which they are members, the relevance of sociolinguisand comparisons is greatest
tic findings
when we consider
learners who are presently
situated
in the target speech commuwith native speakers,
and
nity, who need to interact
successfully
who must as a minimum understand the cultural
values which underlie speech if they are to interpret
what is said with any
In
second
most
addition,
accuracy.
language learners consciously
to
some
level
of
communicative
aspire
competence in the target
not
be
true
in
language (this may
foreign language learning,
where grammatical competence may be accepted by teacher and
students alike as the primary or sole goal of the teaching/
and readily recognize the importance of
learning enterprise)
about the rules for the appropriate
knowing as much as possible
conduct of speech in the new community (Wolfson and Judd 1983).
This paper has several themes.
The first
is that almost
that sociolinguists
can discover about the socioeverything
of Arabic will have some relevance for the
linguistic
patterning
Second is
teaching and learning of Arabic as a second language.l
the principle,
illustrated
here by three examples of sociolinguistic
that while rules for the appropriate
conduct
patterning,
of speech vary considerably
from one society
to another, there is
an underlying universality
to sociolinguistic
rules which makes
and accessible
them readily comprehensible
to language learners,
while cross-cultural
contributes
to learners'
convariability
tinued interest.
Third, there is some evidence that sociolinguistic
rules are difficult
for learners to acquire on their own,
that
some
facts need to be taught if
suggesting
sociolinguistic
second language learners are to achieve the desired goal of
communicative competence.
55
This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

56

Anthropological

Linguistics,

Vol. 28, No. 1

In every speech
2. Variable rules in English and Arabic.
are linguistically
attributes
significant
community, socially
but
marked, but such markers are usually not context independent,
and variable.
rather probabilistic
Examples abound, but Fisher
(1958) offered a clear and "typical" example in his classic
study
of the alternation
of -ing and -in as the present progressive
suffix in the speech of a group of children in New England.
Fisher found several social parameters that played a part in the
used the -ing variant
of the variant used: sex (girls
selection
vs.
more than boys), social class,
personality
(aggressive
the
of
vs.
mood (tense
relaxed),
formality
cooperative),
and topic of conversation.
Fisher also found that
situation,
certain "formal" verb stems appeared to require the formal suffix
while informal verb stems
(e.g.,
correcting,
interesting),
punchin,
(e.g.,
usually occurred with the informal -in suffix
chewin).
Variation such as that discussed
by Fisher for New England
and easily
to
learners
is
interesting
language
English
factors
involved - sex,
the
social
because
comprehended by them,
in socioare
factors
etc.
common
social class,
formality,
the
is
varies
What
variation
particular
everywhere.
linguistic
features which counts in a particular
mix of sociological
speech
with those
features which correlate
community and the linguistic
facts.
sociological
of English,
in urban varieties
The study of such variation
of
York
New
with
Labov's
landmark
(Labov
English
study
beginning
rule community, in
1966), has led to the model of the variable
which the speech community is seen as a group of speakers who
In the
share a set of norms rather than a set of behaviors.
class
social
Labovian
speech community, higher
prototypical
to the norms of the standard
groups approximate more closely
and within social
of
variables,
linguistic
language in their use
of
than
to the prestige
men
class groups women are more sensitive
is seen as the result of
the standard norm. Style shifting
to speech" and the power of the
of "attention
interaction
norms (Labov 1972).
prestige
The case of Arabic fits somewhat uncomfortably into the
of
of the sociolinguistics
Labovian paradigm, and investigation
to further the development of "post-Labovian
Arabic is likely
(Romaine 1982) in several ways. It is clear,
sociolinguistics"
class is not the only important
for example, that socioeconomic
in
variable
Arabic
social
speech communities (level of education,
and the urban:rural distinction
may be as imporreligiosity,
while it is not clear that we can really define a single
tant),
of even a city such
set of norms which unites all the inhabitants
social
as Cairo.
variables
Different
may have different
and markers of social
as group differentiators
functions
at
and there may be different
identities,
patterns of prestige
the
basic
most
different
of the community.
levels
Perhaps
rule model to Arabic,
in applying the variable
difficulty
the
between a speech community
however, stems from
relationship

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

57

in which speakers vary rather narrowly between standard and


nonstandard variants within what is clearly
one linguistic
system
urban North American case) on the one hand, and
(the typical
on the other hand, in which the ends of the
situations
diglossic
continuum are so different
that one
diglossic
linguistically
to posit underlying forms in common. Nevertheless,
hesitates
and language teachers alike must recognize that the way
linguists
in which native speakers of Arabic usually speak is in fact
neither "pure" colloquial
nor "pure" classical
Arabic, but a
between the two poles (Badawi 1973).
To
variety which fluctuates
the degree that this fluctuation
is not an essentially
haphazard
kind of discrete
alternation
but a structured
continuum of
rule-bound behavior,
the variable
rule is a useful formal
mechanism for representing
that fluctuation.
One variable feature in Egyptian Arabic is the distribution
of stop, sibilant
and interdental
fricative
of the
pronunciations
Classical
Arabic interdentals
[0], [d] and [B] (Schmidt 1974,
As the result of two diachronic
sound changes, one
1977).
to stops as soon as a
interdentals
changing the Classical
distinctive
was being formed and the
Egyptian Arabic dialect
other changing the interdentals
to sibilants
beginning some time
after the fourteenth
century (Birkeland 1952) and still
producare
there
in
tive,
contemporary Egyptian Arabic numerous lexical
with interdental
sibilant
and stop variants,
triplets
fricative,
third.
Other lexical
items e.g. [0a:liQ] " [sa:lis] J
[ta:lit]
those that have not been in the colloquial
for
a
vocabulary
long
time, including newly coined technical
terminology and older
words that are generally
acquired by native speakers only through
formal education,
their character as learned words retaining
cannot now be colloquialized
with stops, but any Arabic word with
substitution,
by sibilant
[0], [d] or [B] may be colloquialized
[mumagOil] w [mumassil] actor, but not *[mumattil];
e.g.,
[0awra]el
but not *[tawra].
both
[sawra] revolution,
Synchronically,
and native speakers tend to identify
such variants with
linguists
different
classical
and colloquial)
and
linguistic
systems (i.e.,
to assign the alternations
to code switching,
code mixing or free
variation.
The important fact to note, however, is that such
switching and mixing are orderly rather than random, and the
variation
is really not free.
These alternations
constitute
a
Labovian sociolinguistic
one
correlated
variable,
with facts of
the social context.
The present writer investigated
the stylistic
use of
interdental
sibilant
and stop pronunciations
fricative,
in
lexical
items with potential
interdentals
(the TH-variable)
among
16 university
students and 12 working class males in Cairo
(Schmidt 1974).
In each of four discriminable
styles informal
interview
relatively
speech, formal interview style,
- a distinctive
reading passages and word lists
distribution
pattern appeared.
Interdental
fricatives
accounted for slightly
more than half of all realizations
of the TH-variable when

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

58

Anthropological

Linguistics,

Vol.

28,

No. 1

but in no other style.


Sibilants
reading from word lists,
predominated when reading from texts.
Stops did not occur at all
in the reading styles,
in the relatively
but prevailed
casual
were
parts of the interview.
pronunciations
Stop and sibilant
about equal in the more formal parts of the interviews.
The same general pattern held for the two informant groups
vs. working class) with one major difference.
(elite
university
While all of the university
students produced at least some
instances
of interdental
seven of the twelve
pronunciations,
no
at all, and the
interdentals
working class informants produced
of the
working class informants who did produce some instances
as the
classical
did so less than half as frequently
interdentals
thus
mean for the university
TH-variable
Arabic
The
group.
marker in Cairo,
appears to be a highly developed sociolinguistic
the
an indicator
at
least
two
dimensions of
which co-varies
along
and
to speech)
socioeconomic
class (although
(attention
style
level of attained education might define the groups more
is also Labovian in the sense that it
The variation
accurately).
the level of conscious
the
most
for
below
part
operates
awareness.
While there is general awareness that the interdental
standard Araare "correct" in terms of classical
pronunciations
as
bic and native speakers are quick to label stop pronunciations
is
status
the
much
of
sibilant
the
pronunciations
"colloquial,"
less clear and individual
members of the speech community are not
aware of the ways in which they shift among variant pronunciawhen
Educated native speakers are typically
tions.
skeptical
told that literate
native speakers often substitute
[s] for [0]
when reading Arabic, though in fact all educated speakers in this
study did so.
in this study which
One aspect of Cairo Arabic Th-variation
model was the absence of
did not fit the western sociolinguistic
on the part of women to the
any apparent greater sensitivity
norm. The differences
between
of the classical/standard
prestige
not
men and women on the TH-variable were
the university
signifiwas the behavior of the unidifferent
cant.
What was strikingly
men and women as a group as compared to the working class
versity
included only male subjects).
group (which, unfortunately,
some of the
A second example of variation
which illustrates
of
the
distribution
is
Cairo
of
the
community
speech
complexities
which
in
Arabic
uvular [q] and glottal
Q-variable),
['] (the
of -ing/-in
the distribution
several ways rather neatly parallels
The regular diachronic
in English as described by Fisher (1958).
C3 was from [q] to ['] in Cairo and
development of classical
in a great many cognate pairs in the two
Lower Egypt, resulting
coffee;
codes ([qahwa]:['ahwa]
[qa:la]:['a:l]
to say), but it is
to speak of a simple
and misleading
somewhat over-simplified
and ['] in
synchronic contrast between [q] in classical/standard
no
structured
In
interviews,
speakers used
Egyptian colloquial.
a
of [q] even
a
but
there
was
residue
when
from
text,
reading
[']
and a noticeable
in the most casual parts of the interview,

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

59

inhibition
of the glottal
when moving to more
stop pronunciation
with
variation
The parallel
formal topics.
is
English -ing/in
rather striking:
Fisher found that the choice of variants
changed
use of -ing in formal situations
from an almost exclusive
to a
use of -in in informal interviews.
predominance but not exclusive
A second parallel
and the
alternation
between the English -ing/in
is the relationship
between this variCairo [q]:[']
alternation
lexical
items.
In addition to two lexical
able rule and specific
items which are never colloquialized
with [I], [qur'a:n] and
there are numerous others which usually appear with
[qa:hira],
in
contexts
for colloquial,
so
[q]
normally construed as calling
that most Cairenes distinguish
between such pairs as [qawi]
applaud
strong and ['awi] yery, [saqqaf] to educate and [sa"'af]
even in informal speech.
In the case of the Arabic Q-variable
there is differentiation by sex.
In the study reported here, both the elite
univerand
the
males
class
used
males
the
standard
uvular
working
sity
than the university
women did.
In contrast
[q] more frequently
to the western sociolinguistic
that women are more
hypothesis
sensitive
than men to the prestige
of prescriptive
forms, in this
case Arab women seem to be deliberately
particular
choosing to
a
standard
variant.
particular
Or, we
downplay
phonological
a
there
is
non-classical
might speculate,
competing
prestige
variety for Cairo with ['], which educated women recognize though
educated men do not.
resolution
of this
Royal [1985] has suggested an interesting
of
puzzle for Cairo Arabic.
Royal concludes that the preference
women for ['] over [q], together with the tendency of Cairene
women to palatalize
[D] and [T] before high front vowels and the
of
women
to
consotendency
produce much weaker pharyngealized
nants are all part of a prestigious
for
social
system
signalling
gender through a fronting and raising versus backing and lowering
with fronted speech more typically
pronunciation
convention,
associated
with femininity
and backed speech with masculinity.
with the variable of phonoRoyal's research dealt specifically
(let us call this the Arabic Phlogical
pharyngealization
for which native speakers claim there to be
-variable),
masculine-feminine
and class differences.
Based on spectroof second formant transitions
in recordings of
graphic analysis
male and female speakers from two age groups in two Cairo
(one a folk quarter in the older part of the city,
neighborhoods
the other a westernized
affluent
suburb), Royal found that males
indeed produced markedly stronger pharyngealization
than females
in the affluent
In the folk neighborhood,
on the
neighborhood.
other hand, older subjects
did not observe this sex distinction
(pharyngealization
by both men and women was heavier than in the
affluent
suburb group) but younger subjects were seen to be
In this case, what is especially
acquiring the distinction.
is that both upper class men and upper class women
interesting
seem to be responding to a prestige
norm which distinguishes
between classes
but which is not in the direction
of classical

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

60

Anthropological

Vol. 28,

Linguistics,

No. 1

Arabic requires strong diffeArabic, since classical/standard


and nonpharyngealized
consorentiation
between pharyngealized
nants .
In sum, each of the three variable features we have
an interesting
deviation
considered for Cairo Arabic represents
from the pattern we would expect based on a variable
rule model
The
derived from North American sociolinguistic
patterns.
the
case
is
for
in
direcNorth
American
style shifting
typical
tion of a prestige
social
stratiwith
norm,
(usually national)
to the norm exhibited
on the
and greater sensitivity
fication
a
see
we
pattern of
part of women. In the Arabic TH-variable,
classical
towards the prescriptive,
norm, with
style shifting
influence
but
without
stratification
social
any
apparent
strong
on the other hand, shows
The Arabic Q-variable,
of speaker sex.
but
norm by all informants,
towards the classical
style shifting
to
on
of
women
the
with a countervailing
produce
part
tendency
variants than would be predicted on the basis of
fewer classical
in the
and educational
level alone.
their socioeconomic
Finally,
and
we see social stratification
case of the Arabic PH-variable,
both operating in favor of a less classisexual differentiation
in Cairene Arabic
cal pronunciation
that is now more prestigious
than
the classical
the
least
for
westernized
classes)
(at
upper
norm of strong pharyngealization.
.

Personal

address

(1985) have pointed out,


subject par excellence:

systems.

personal

As Philipsen

address

and Huspek

is a sociolinguistic

In every language and society,


every time one person speaks
to another, there is created a host of options centering
named and
around whether and how persons will be addressed,
described.
The choices speakers make in such situations,
and their meaning to those who interpret
them, are systein
Such
not
random.
matic,
language behavior,
systematicity
is universal,
whether of use or interpretation,
although
what elements comprise the personal address system and what
rules govern its deployment, vary across contexts
(p. 94).
Ervin-Tripp (1971) has provided an elegant formal analysis
of an American address system, in the form of a computer flow
and addressee
for setting
chart with a series of binary selectors
the
choices
For
Western
American
output
major
English,
identity.
are: Title + Last Name (Dr. Jones), Mr./Mrs./Miss + Last Name
Kin Title + First Name (Uncle George), First Name
(Mrs. Roberts),
and 0 or no-naming, which is the choice whenever an
(Bill),
addressee's
name is not known, or when a speaker is unsure of
which
The binary selectors
what address form is appropriate.
include: whether the
determine which address form is appropriate
addressee is an adult or a child, since children are nearly
first-named
by adults; whether speaker and hearer are
invariably
a courtroom, which
a
status
marked
in
e.g.,
setting,
co-present

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

61

requires that personal ties be masked; whether the addressee is a


and if so whether older or younger generation,
relative,
since
Uncle George is appropriate while Nephew Billy is not; whether or
not the addressee is a friend or colleague
and, if a colleague,
of higher rank; whether the addressee has given dispensation
to
the speaker to use a less status marked address form (please call
me George); whether or not the addressee has by virtue of office
some special
claim to a specific
title
etc.
(Mr. President),
While American culture is generally
viewed as very informal and
while native speakers consider the system to be straightforward,
there are additional
Whenever full titles
cannot
complications.
be realized,
as when the last name is not known, some titles
can
be abbreviated
Dr. without a last name in the case of an
(e.g.,
M.D. but not a Ph.D.), but most cannot (Excuse me, Mr. is not a
abbreviation
for Excuse me, Mr. Smith if one cannot
possible
remember the gentleman's name). There are no widely agreed upon
norms for how to address in-laws,
and conflicts
between personal
in
such
cases
force a resort to no-naming.
preferences
frequently
There is also a complicated system of address forms used in
Not surprisingly,
choices of address forms are
Egyptian Arabic.
based on familiar
social categories:
to
sex, age, age relative
the speaker, setting,
etc.
But there are differences
as well,
some of which derive from different
social determinations
of
common selectors.
For example, at the American University
in
Cairo, large meetings of the entire faculty are status marked
situations
for American and Egyptian colleagues
more
alike,
i.e.,
formal address forms will be used, but departmental meetings are
so marked only for some Egyptian professors
and virtually
no American professors.
The selector
"friend or colleague" may also be
defined differently.
Ervin-Tripp observes that for an American
assistant
to call a new colleague
of the same age and
professor
rank by Professor + Last Name would be considered very strange on
the U.S. West Coast, but not so in Egypt.
There are many differences between the American and Egyptian systems in the area of
kinship terms, both in the forms used and in the persons to whom
One is the interesting
use in Arabic of
they may be applied.
what Ayoub (1964) has called "bi-polar" kin terms, ego addressing
alter with the term which in its literal
sense would be appropriate for alter addressing ego.
A father may call his son father
and be called father in return.
As Ayoub points out, this phenomenon of the senior borrowing from the kinship vocabulary of the
social context,
when the senior wishes
junior occurs in a special
the junior to do something, but chooses a conciliatory
request
form rather than an abrupt command. American English and Egypin the ways in which kinship terms can be
tian Arabic also differ
extended to non-kin in solidary
fashion.
is
['abla] elder sister
used by students to address any female teacher.
While such extensions happen sometimes in American English,
so that many children address close family friends who are not in fact related as
Aunt X or Uncle Y, in Egypt one may use a kin term appropriate
in
to address strangers.
If I hail a cab in
age and generation

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

62

Anthropological

Linguistics,

Vol. 28,

No. 1

Cairo, I shout [ya: taks] but once I am in the cab, if I light a


and offer one to the driver (it would be rude not to) I
cigarette
if he is approximately my
often address him as [ya: 'ax] brother,
age, or [ya: 9ammi] uncle if he is, say, 15 years my senior.
there are major differences
In terms of output categories,
between the American English and Egyptian Arabic systems.
In the
more
to
choose
has
Arabic
terms,
options
simplest
many
among.
name is known, the problem of address is not
When the addressee's
A student whose name appears on a class
solved.
automatically
list as Mohamed Moustafa Abdel Hamid may be Mr. Abdel Hamid in
English, but Mr. Mohamed in Arabic, or perhaps Mr. Moustafa,
names such as Mohamed, Ahmed, Ibrahim are
since very common first
often dropped in favor of second names. This never happens with
comwomen's names, but with married women there are additional
when
names
not
their
in
do
Women
legal
change
plications.
Egypt
names
in
use
their
husbands'
nor
do
professional
they
they marry,
but upper class women may use their husband's first
situations,
name is Omneya
So a woman whose official
or last names socially.
+
or
LN)
Omneya (FN)
Fayek Kassabgy may be Mrs. Kassabgy (Title
in English, but in Arabic she might be Omneya (FN) or Karima (an
additional
unofficial
personal name known only to close friends
and family).
At the school where she teaches she would be Miss
to professional
Omneya (Title + FN) or simply Miss or ['abla];
be
Madame
she
would
Omneya (Title + FN); and
probably
colleagues,
as
Madame Gazala (Title +
she might be addressed
socially
husband's LN) or Madame Wissam (Title + husband's FN) or Madame
+ husband's FN). For a
Doctor Wissam (Title + husband's title
the more common address terms
woman lower on the social scale,
+ FN or husband's FN, or ['um] + FN of eldest son
would be [sitt]
lower in the soStill
or, if she has no sons, eldest daughter.
cial structure,
mostly with recent immigrants from the countryside, one finds conscious avoidance of any personal address forms
for women; one can hear a husband address his wife as [ya:
gama:9a], literally
group.
In the case of address forms used to strangers whose name is
not known, the range of output forms is also far greater in EgypWhile Americans have only
tian Arabic than in American English.
in use (so
limited options such as Sir, and Ma'am, restricted
the system of titles
that no-naming is generally
preferred),
Without any
which may be used alone in Arabic is extremely rich.
some forms heard during
here are listed
attempt to be exhaustive,
a single 20 minute metro ride, arranged roughly from most formal
to least formal, although in fact several dimensions of choice
are involved.
The speaker is a ticket taker, and the frame is
which does not permit no-naming.
[taza:kir
ya: ___] tickets,
__,
to a
term of great respect,
taza:kir
ya: afandim
well dressed man or (more
common) woman

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

63

taza:kir

ya: be:h

to a man who
highly respectful,
appears to have high status

taza:kir

ya: usta:z

literally
professor;
respectful
but not as deferential
as the
above forms

taza:kir

ya: siyadtak
ya: HaDritak

roughly sir or ma'am


[HaDritik].
Sex
[siyadtik],
of speaker is important: women
say [HaDritak] more often than
men and almost never use
or [be:h]
[siyadtak]

taza:kir

ya: Hagg(a)

literally

one who has made the

pilgrimage.
Respectful
term of address to
religious
older but not high status
Not used to a friend
person.
or acquaintance
unless it is
known that addressee has in
fact made the pilgrimage
taza:kir

ya: sayyid
ya: sitt
ya: 'a:nisa

to
Mr., Mrs., Miss; generally
an equal, though note that
servants usually address the
woman of the house as [sitt];
[sayyid] used most often to a
young man

taza:kir

ya: misyu
ya: mada:m
ya: mazmaze:l

to an
again, Mr., Mrs.,,Mi;
or
one
equal
perceived as
Social
higher in rank; polite.
class of speaker determines
whether these or forms closer
to French will be used

taza:kir

ya: m9allim

but used by
literally
teacher,
extension
to a foreman, plant
floor supervisor,
pimp, hashish
etc.
When addressing a
dealer,
stranger the form is most often
used with rough types

taza:kir

ya: 9ari:s

literally
bridgegroom;
to a young boy

taza:kir

ya: 9amm
ya: 9amm ide:x

literally
side,

wise

uncle
uncle.

jocular,

on father's

term to an elderly
passenger

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Respectful

lower class

64

Anthropological
taza:kir

ya: ba:sa

Linguistics,

Vol. 28,

No. 1

pasha; while under the system

in Egypt, the
of Turkish titles
outranked
the
pasha
bey, the
term [be:h] remains an address
form of respect,
while [ba:*a]
is used only sarcastically.
Can be used with children,
friends,
equals and subordinates, but an addressee who
himself as higher in
perceives
status would take offense

taza:kir

ya: afandi

also sarcastic
(cf.
above, a respectful

taza:kir

ya: kapten

can be used to a young


captain,
man who looks like an army
private

taza:kir

ya: usTa

to anyone who appears to be a


or unskilled
skilled
worker

taza:kir

ya: rayyis

boss, usual
waiters

taza:kir

ya: 'ibni
ya: binti

my.b.y,

taza:kir

ya: walad
ya: bint

and
unfriendly
boy, sirl;
often followed
threatening,
an order to leave the train
move to a second class car.

[afandim]
term)

term of address

son, my girl;

to

friendly

by
or

The sociolinguistic
es.
patter4. Discourse sequencing a
and
to
not
is
limited
of
course
of
phonological
language
ning
but extends to the organization
of discourse
lexical
variation,
in speech events such as lectures,
sermons, interdiscussions,
as
events
such
transactional
views, meetings,
ordering a meal in
and the like, all of which
a restaurant,
casual conversations
Violations
middles and ends.
have rules for proper beginnings,
and
commented
are
noticed
native
such
of
rules by
upon
speakers
and since these rules vary so much from
by other native speakers,
on the part of
and misperceptions
violations
culture to culture,
cultural
non-native
assumpspeakers who operate under different
to result in severe crosscultural
are likely
tions and attitudes
misunderstandings.
Only a brief example will be given here, dealing with the
This is of
sequencing of openings to telephone conversations.
an
ethnomethodolohas
interest
since Schegloff
(1968)
provided
of some aspects of conversational
openings which
gical analysis
has
are apparently universal,
while Godard (1977)
given an exam-

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

65

variation.
ple of crosscultural
Schegloff's
analysis
begins with
two rules for American telephone conversations:
(1) the answerer
and (2) the caller provides the first
A
speaks first,
topic.
to provide a more abstract
single deviant case leads Schegloff
tied to the more general goal of accounting for cooranalysis,
dinated entry into two-party conversations
in general,
in which
the first
utterance
of the conversation
is not the answerer's
"hello" but the ring of the telephone,
which Schegloff
characas the first
a
terizes
of
Summons-Answer
The
part
sequence.
first
this
the conditional
part of
adjacency pair establishes
relevance of the second part.
Given the first
element, the
second is expected,
so that if there is no answer the caller will
infer that no one is home and if the answerer lifts
the telephone
receiver he speaks first.
A further characteristic
of the category Summons is that it is a summons for some reason, so the
caller's
to raise the first
substantive
right and obligation
on
is
the
established.
other
Godard,
hand, has contrasted
topic
some aspects of the expected behavior of the caller and answerer
at the beginning of telephone conversations
at private residences
in the United States and France.
which
Among the differences
Godard notes are: in France, but not in the United States,
the
caller checks the number, and once assured that he has reached
the right house will give his identity
before he asks for his
intended addressee;
in the United States,
but not in France, the
caller
often questions
the answerer's
Is this
(e.g.,
identity
Robert?) or may ask to speak to someone other than the answerer
this
(e.g.,
Hello, may I speak to Jane, please?).
Interpreting
behavior in terms of French norms, Godard found such behavior
rude, since in France the caller must name himself,
incredibly
converse with the answerer before asking for
and, if acquainted,
What seems normal to an American strikes
a French
anyone else.
as
a
treatment
of
the
answerer
as
no more
speaker
dehumanizing
than a mechanical extension
of the telephone.
Telephone conversational
openings in Cairo are different
from both those in France and in the United States.
The
taken from a corpus of 215 telephone
following
conversation,
recorded in Egypt, is in many respects
beginnings
typical:
Answerer:
Caller:
Answerer:
Caller:
Answerer:
Caller:
Answerer:
Caller:

alo:
alo:
alo:
mi:n byitkallim?
Who's speaking?
'inta mi:n?
Who are myou?
'abul magd mawgu:d?
Is Abu elMagd there?
'ana 'abul magd..
mHammad?
I'm Abu el-Magd. Mohammed?
'aywa.. 'izayyak ya: be:h?
Yes.. how are yogu, Bey?

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

66

Anthropological

Linguistics,

Vol.

Answerer:

28,

No. 1

'ilhamdulila:h,
wizayyak 'inta?
Fine, and you?
Caller:
'ilhamdulila:h
'ahlan wasahlan
Answerer:
literally
you're welcome
Caller:
walla:hi
ya: 'ax, 9ayz as'alak Ha:ga
want to ask you something.
Really, brother,
_
This call may strike some western readers as bizaare (simply
norms for conversabecause they are familiar with very different
but there is nothing unusual about such a call
tional openings),
in Cairo.
The first
thing to be noticed is that at the beginning
with no selfthere is a series of hello's
of the conversation
In his second turn, the caller
identification
by either speaker.
from the answerer, but this is refused by
requests identification
of the
the answerer, who counters by demanding the identity
in the next
caller.
Instead of providing self-identification
of the answerer.
instead guesses the identity
turn, the caller
then
This is confirmed by the answerer, who
attempts to guess the
are
of the caller.
After both guesses
confirmed, talk
identity
and responses until
proceeds through a series of greetings
or at
after twelve turns, the message is introduced,
finally,
that he does have a topic to introindicates
least the caller
A common reaction of foreigners
duce.
living in Egypt is that
at all in Cairo,
there are no rules for telephone conversations
of behavior and norms of
but there are indeed rules, regularities
not only for the series of ritualized
greetings
interpretation,
at the end of the sample call presented here but also for the
initial
exchanges.
apparently disorderly
is
The initial
problem for entry into any conversation
for talk.
suggests that simply lifting
Schegloff
availability
while the answerer's
the telephone receiver establishes
presence,
However,
it
seems for telephone
establishes
availability.
hello
as opposed to some other types of Summons-Answer
conversations,
is not
face to face encounters)
availability
sequences (e.g.,
are
identities
until
and
cannot
talk
established
proceed
securely
norms for establishcultures provide different
known. Different
identiA German telephone answerer typically
ing such identity.
when answering, without knowing the
fies himself automatically
Most Americans do this only when answerof the caller.
identity
seldom when answering a telephone in a
ing a business telephone,
and Egyptians apparently never provide such
private residence;
turn.
of self during the answerer's first
identification
the usual case, an
When an answerer does not self-identify,
of the
American caller typically
attempts to confirm the identity
this
has
and
(Is
George?)
apparently
person he intended to reach
by voice, will
or, if he has been able to make the identification
while an Egypti(Hi, George, this is Bill),
proceed to greetings
to respond hello to the
is likely
an caller
in the same situation
in my sample
answerer's hello (one third of the conversations
most
This strikes
first
have hello as the caller's
utterance).

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

67

Americans as over-insistent,
since although it is possible
to
find examples of repeated hello's
in American telephone conversations these are usually in cases of an exceptionally
poor
connection or on those occasions when an answerer picks up the
in which case
telephone and does not immediately say anything,
the caller's
hello is analyzed by Schegloff
as a repetition
of
the summons. The clearest
indication
that the Egyptian rules are
different
is the fact that Egyptian informants report that they
do not view such repeated hello's
as over-insistent,
and are not
offended by them.
What Americans find most offensive
about Egyptian telephone
second turn in the sample call,
calls,
however, is the caller's
his [mi:n byitkallim?],
the demand to know the identity
of the
answerer when voice identification
has not succeeded.
American
callers
do sometimes request the identity
of answerers, of
course.
out
that
a
when
caller formulates a
Schegloff
points
location
to
relation
some
as X's home or X's
such
by
person,
if the voice of the answerer is not recognizable
as that
office,
of X, then the result may be a request for identification.
Something like that seems to operate in the Egyptian speech community
that a high percentage
also, and may be related to an expectation
of telephone calls will result in wrong numbers.
The resulting
rule is slightly
different
for the two speech communities: in the
United States a caller requests identification
only if there is
evidence that the party reached is not the party intended; in
unless there is positive
Cairo, one requests identification
evidence that the party reached is the one wanted.
Note, however,
that the Egyptian answerer does not satisfy
the caller's
demand
for identification,
and here the rules for the two cultures
coincide.
In the United States,
a called person responds to a
demand for identification
without discomfort or anger only if he
that he is a person who would not normally be expected
recognizes
to be in that place.
Most (but not all) Egyptians similarly
reMore importantly,
port that they find such demands offensive.
in
none of the calls recorded did Egyptians answer such demands.
Egyptian and American telephone conversational
beginnings
are alike,
then, in several respects.
They all have the same
of sequenced interaction:
categories
summons, answer, greeting,
introduction
of message.
They are alike in the fact that an
answer may self-identify,
but usually does not, and in the availof identification
But there is an
ability
by voice quality.
in the fact that there is a strong
equally strong difference
reluctance
on the part of both callers
and answerers in Egypt to
give any self-identification
before ascertaining
the identity
of
the other, the reasons for which must be sought in the general
culture.
are an attempt to ascertain
Repeated hello's
identities
voice.
If
this is unsuccessful,
through
demand
Egyptian callers
identification
from the answerer far more frequently
than do American callers,
although this does not result in called persons
giving up their rights to prior self-identification
by callers.

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

68

Anthropological

Linguistics,

Vol.

28,

No. 1

for second language learning and teaching.


Implications
5.
here - variable
rules in
The sociolinguistic
patterns discussed
and
and
discourse
Arabic, personal address forms
English
sequencing rules - have been chosen from a very broad range of possible
sociolinguistic
topics primarily because there are some data
on which to base a comparative sociolinguistics,
available
but
also because in each case there is some connection to problems of
second language learning and teaching.
rules of phonology,
In the case of variable sociostylistic
to
there is some evidence (Schmidt 1977) that these may transfer
the target language, so that the persistent
probpronunciation
lems of Egyptian Arabic speakers in mastering the English interto the fact that in Egyptian
dental fricatives
may be attributed
is
and sibilants
between interdentals
Arabic the alternation
while in English these sounds are
determined,
sociolinguistically
assigned to separate phonemes which are not in sociolinguistic
A pedagogical
is that a rather common
variation.
implication
device used by teachers of English in the Arab world, stressing
of English
the identity
c>
[0] and [d] with Arabic orthographic
for although there are minimal pairs
and 3 , seems misguided,
showing that Arabic [0, d] and [s,z] are phonemically contrastive
in careful speech, the fact remains that any printed r"\ may be
Identificamuch notice.
read aloud as [s] without occassioning
what
tion of English [0] with Arabic c4: is therefore
precisely
must be avoided.
of a sociolinguistic
variable
Notice, however, that transfer
error in the
in this case does not result in a sociolinguistic
some English sociolinguistic
sense of violating
norm. This is
because the only type of TH-variation which has a sociolinguistic
in English when produced by native speakers is the
interpretation
dese and dose for these and
substitution
with stops (e.g.,
sibilant
(zese and zose) are clearly
those);
pronunciations
of such
foreign and native speakers tend to be very tolerant
variables
when looking for sociolinguistic
errors.
Therefore,
to be relevant for Arabic as a second language,
which are likely
on those which trigger
it would be best to concentrate
initially
and
which non-native
for
native
reactions
speakers
by
strong
to
variants
which
are
carry normative
produce
expected
speakers
best candidate
The
the
within
speech community.
interpretations
for such research is probably the voice quality of pharyngealization (or Royal's more general series of contrasts
along the
This is a case
vs. backing/lowering
dimension).
fronting/raising
in which native speakers attach strong judgments to the sociois considered polite
weak pharyngealization
variable:
linguistic
and cultured,
but upper class men who exhibit weak pharyngealization are criticized
by working class men for speaking like
is viewed as an aspect of "the
girls;
heavy pharyngealization
but
with frankness,
Arabic language" and associated
original
consonants too strongly are
women who pronounce pharyngealized
This is also a case in which non-native
perceived as unfeminine.
speaker behavior may vary along the same dimension which triggers

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

69

Kahn (1975) compared


such value judgments among native speakers.
in the speech of male and female
the degree of pharyngealization
Arabic speakers and male and female American students trained by
a male native speaker instructor.
The comparison revealed that
between American males and females was less than
the difference
between the sexes for Arabs, and that the
half the difference
overall formant values for Arab women were higher than for the
the American women exhibited
American women, i.e.,
stronger
than
the Arab women. Before concluding that it
pharyngealization
is important for non-native
learners of Arabic to be taught by
native speakers of the same sex, however, we need to know how
in non-native Arabic is judged by native
heavy pharyngealization
A
speakers.
working hypothesis
might be that there is a threshhold level in phonology, below which non-native
speakers are
and above which they are vulnerable
to
simply labeled [+foreign]
the same kinds of judgments which native speakers make of other
native speakers.
Some of the contrasts
between the American English and Cairo
Arabic personal address systems clearly need to be taught to nonnative speakers.
Leaving aside the more exotic aspects of these
address systems (as seen by outsiders),
even the simple contrast
between Title + Last Name (the American pattern) and Title +
First Name (the basic Egyptian pattern)
is poorly understood by
non-native
and few Americans resident
in Egypt have any
speakers,
clear notion of how the Egyptian address system works.
In some
are simply amusing, as when an Amecases, the misunderstandings
rican secretary
in an Egyptian firm puzzled over why she was
listed
in the company telephone book as Ms. + FN (an anomalous
form in both cultures)
or when two librarians
at the American
in
Cairo
were
out
to find that one of
University
put
slightly
them was listed
by FN and one by LN. On the grounds that most
Egyptian faculty members have studied or lived abroad and would
a decision
was made to list them all
prefer the western pattern,
by last name, while "staff" were listed
according to the Egyptian
In a few cases, conflicts
between address systems have
pattern.
more serious consequences,
and I know of one case in which an
American supervisor
considered dismissing
an American English
teacher because she did not have her students'
The only
respect.
evidence for that was the fact that the students addressed the
teacher as Miss Mary, which the American supervisor
did not recognize as a respect pattern in spite of long residence
in the
Arab world.
As evidenced by the telephone conversation
example, it may
be the area of discourse
in which the risk of severe crosscultural misunderstandings
most frequently
arises and where second
should be most concerned.
In addition,
language instruction
there is some evidence that the sociolinguistic
aspects of discourse are extremely difficult
for non-native
speakers to acquire
on their own (Scarcella
The problem here for an
1979, 1983).
for language teaching is where to begin
applied sociolinguistics
within an area of sociolinguistic
that has so many
patterning

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

70

Anthropological

Linguistics,

Vol.

28, No. 1

semantic formulas
turn-taking,
aspects:
openings and closings,
as
for speech acts such
criticizing,
praising,
apologizing,
verbal routines
(see Gregory and Wehba, this volume), and so on.
For the
The list of phenomena to be considered is a long one.
to second language pedagof sociolinguistic
findings
application
of sociolinguistic
in any of
ogy, the identification
patterns
these areas will be relevant.
But for an applied sociolinguistics which seeks to identify
likely problem spots for second
and
those areas, rather than simply
research
learners
language
available
whatever
is
(as I have done here), probably
applying
the best approach is to begin with a framework which is broad
phenomena. One such
enough to encompass a range of linguistic
framework which could guide research is the concept of politeas developed by Brown and Levinson (1978) who
ness, especially
interaction
relaas a principle
of face-to-face
view politeness
ted to both concern for speakers'
and addressees'
good image
claims of both, a univer(face) and respect for the territorial
sal framework within which there is great cultural
variability.

WORKSCITED
in Arabic kinship terms.
R. 1964.
Ayoub, Millicent
Bi-polarity
In Horace G. Lunt, ed., Proceedings of the Ninth
The Hague: Mouton. Pp.
International
Congress of Linguists.
1100-06.
Badawi, Said Mohammed. 1973.
Mustawaya:tu l'arabiyyati
fi: miSr (Levels of Contemporary Arabic
Imu'a:Sarati
Cairo: Dar al Ma'a:rif.
Egypt).
H. 1952.
Growth and Structure of the Egyptian
Birkeland,
Oslo: Avhandlinger utg. av Det Norske
Dialect.
Videnskape-Acad.

in
Arabic

in
Universals
1978.
Brown, Penelope and Stephen Levinson.
Esther
N.
Goody,
phenomena. In
politeness
language usage:
in Social
ed., Questions and Politeness:
Strategies
Press.
Interaction.
Cambridge: Cambridge University
1971.
Susan.
Sociolinguistics.
Ervin-Tripp,
Fishman, ed., Advances in the Sociology
Hague: Mouton. Pp. 15-91.
Fisher, John L. 1958.
variant.
linguistic
Fishman, Joshua A.
Introduction.
Publishers.

Social influences
Word 14:47-56.

1970.
Sociolinguistics:
Rowley, Massachusetts:

In Joshua A.
The
of Language.

on the choice

of a

A Brief
Newbury House

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

Arabic as a Second Language

71

norms: Phone
different
1977.
Same setting,
Godard, Daniele.
in France and the United States.
call beginnings
Language
in Society 6:209-19.
Arabic emphatics:
the evidence for
1975.
Kahn, Margaret.
of phonetic sex-typing.
determinants
Phonetica
cultural
32:38-50.
of English
The Social Stratification
1966.
Labov, William.
D.C.:
Center
for
New York City.
Washington,
Applied
Linguistics.
. 1972.
Sociolinguistic
of Pennsylvania
University

---

Philipsen,
Gerry and Michael
studies
sociolinguistic
Romaine, Suzanne,
Communities.

Patterns.
Press.

in

Philadelphia:

A bibliography
of
Huspek. 1985.
of personal address. AL 27:94-101.

ed. 1982.
Sociolinguistic
London: Edward Arnold.

Variation

in Speech

Anne Marie.
1985.
Patterns
Male/female Pharyngealization
in Cairo Arabic:
A Sociolinguistic
Study of Two
Texas Linguistic
Forum 27.
Neighborhoods.
Department of
Center for Cognitive Studies,
The University
of
Linguistics,
Texas at Austin.

Royal,

1979.
"Watch up!":
routines
Scarcella,
Robin.
prefabricated
adult second language performance.
Working Papers on
19:79-88.
Toronto:
OISE.
Bilingualism
---

in

of
. 1983.
Developmental trends in the acquisition
conversational
competence by adult second language learners.
In Wolfson and Judd. Pp. 175-83.

Emanual A. 1968.
Schegloff,
AA 70:1075-95.
openings.

Sequencing

in conversational

Variation
in Spoken
Schmidt, Richard W. 1974.
Sociostylistic
Brown
Ph.D.
Egyptian Arabic. Unpublished
University
dissertation.
---

. 1977.
variation
and language transfer
Sociolinguistic
in phonology. Working Papers on Bilingualism
12:79-93.
Toronto: OISE.

Wolfson, Nessa and Elliot


Language Acquisition.
Publishers.

and
1983.
Judd, eds.
Sociolinguistics
Massachusetts:
House
Rowley,
Newbury

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions

72

Anthropological

Linguistics,

Vol.

28,

No. 1

NOTE
1. Readers should be warned that the data on which this
paper is based were gathered more than ten years ago and thereA preliminary
fore may be dated in some instances.
conceptualization
of the material here was presented at the Symposium on
Annual Meeting of the
and Applied Anthropology,
Sociolinguistics
1975.
for
Amsterdam,
Society
Applied Anthropology,

This content downloaded from 95.186.253.178 on Tue, 19 Jan 2016 16:05:27 UTC
All use subject to JSTOR Terms and Conditions