The Study of Language Attitudes

THE STUDY OF LANGUAGE ATTITUDES 1
ROBERT L. COOPER AND JOSHUA A. F1SHMAN
Attitde has served s a variable in many sociolinguistic studies (for a

methodological and topical review, see Agheyisi and Fishman, 1970;
for a recent compendium, see Shuy and Fasold, 1973). This has been
true in part because attitude is a central concept in the social sciences
generally and in part because sociolinguistic phenomena are complex
enough to motivate a search for equally complex predictive hypothetical
constructs. Thus, for example, language attitude appears s a catalyst
for a sound change (Labov, 1963), a defining characteristic of a speech
Community (Labov, 1966), a predictor of second-language achievement
(Anisfeld and Lambert, 1961; Lambert, Gardner, Barik, and Tunstall,
1963; Lambert, Gardner, Olton, and Tunstall, 1968), a reflection of
interethnic attitudes (Herman, 1961; Lambert, Anisfeld, and YeniKomshian, 1965), a determinant of interlingual intelligibility (Wolff,
1959), and a determinant of teachers' perceptions of their pupils' ability
(Seligman, Lambert, and Tucker, 1972).
While it is not remarkable that language attitude appears s a variable
in sociolinguistic studies, it is surprising that the work of attitude theorists
on the one band and the work of sociolinguists on the other should
have been conducted in relative Isolation from one another. Sociolinguistic investigators' use of attitude s a variable has proceeded largely
without reference to issues in the theory and measurement of attitudes,
and the procedures and data developed by sociolinguists have been
largely ignored by attitude theorists. This mutual Isolation is unfortunate
because each research tradition can benefit from the other. Sociolinguistic behavior can serve s a rieh source of data against which attitude theories can be tested. Moreover, some of the attitude measure1
This article is a revised and expanded version of a paper prepared for the International Seminar in Language Testing, San Juan, Puerto Rico, May 1973. The original
paper will appear in L. Palmer and B. Spolsky (eds.), Papers on Language Testing
1967-74 (Washington, D. C., TESOL, in press).
Unauthenticated | 187.204.45.15
Download Date | 10/10/12 4:41 PM
ROBERT L. COOPER AND JOSHUA A. FISHMAN
ment techniques developed by sociolinguists, notably the use of aural

Stimuli to obtain intergroup evaluation, s in the matched-guise (Lambert, 1967) and mirror-image (Kimple, Cooper, and Fishman, 1969)
techniques, could profitably be used more widely. Sociolinguistic studies,
on the other band, could benefit from the attitude theorists' elaboration
of attitude s a hypothetical construct. If sociolinguists made explicit
the theoretical assumptions underlying their use of attitude s a research
variable, they should obtain sharper attitude scales and more readily
interpretable results.
It may be useful at this point to attempt to distinguish language
attitudes from other attitudes. How can language attitudes be so distinguished? One solution is to define language attitude in terms of its
referent. Thus Ferguson (1972) defines language attitudes s "elicitable
shoulds on who speaks what, when, and how". A second solution is to
define language attitudes in terms of their consequences, i.e., those
attitudes which influence language behavior and behavior toward
language. The problem with the former solution is that it excludes
attitudes of interest to sociolinguistics, e.g., attitudes towards organized
efforts involved in language planning, language maintenance, or language
shift. The problem with the latter solution is that it is too broad. If the
latter solution is adopted, it is not clear how language attitudes can be
distinguished from other attitudes, since almost any attitude, under the
right conditions, might affect language behavior or behavior toward
language. We have chosen to define language attitude in terms of its
referent. We have amplified the referent to include language, language
behavior, and referents of which language or language behavior is a
marker or symbol. Thus, attitudes toward a language (e.g., Hebrew)
or towards a feature of a language (e.g., a given phonological variant)
or towards language use (e.g., the use of Hebrew for secular purposes)
or towards language s a group marker (e.g., Hebrew s a language of
Jews) are all examples of language attitudes. Conversely, attitudes
towards Jews or attitudes towards secular domains are not language
attitudes, although they might be reflected by language attitudes.
This paper outlines several issues in the theory and measurement of
attitudes in general and of language attitudes in particular and concludes
with a description of ongoing research motivated by many of these
issues.
THE STUDY OF LANGUAGE ATTITUDES
EIGHTEEN ISSUES
We can convemently classify attitude and language-attitude issues into

four categories: the nature of attitude, the determinants of attitude, the
effects of attitude, and the measurement of attitude. It should be noted
that issues raised with respect to attitudes generally apply with equal
force to language attitudes. Thus, for example, issues about the nature
of attitude are equally issues about the nature of language attitude.
The Nature of Attitude
(1) Does attitude have a characteristic structure? Most theorists view
attitude s an intervening variable, although a minority view it s only
a set of stimulus-response or antecedent-consequence relationships.
(For the latter view, see De Fleur and Westie, 1963.) Among the former,
however, there is disagreement s to whether the structure of the hypothetical construct is best thought of in terms of a cognitive-affectiveconative (thinking-feeling-acting) analysis or in terms of a means-end
analysis (McGuire, 1969). The means-end analysis defines a person's
attitude toward a referent s a composite of the perceived usefulness
of the referent with respect to bis goals weighted by the relative value
he places on each goal. Seyeral questions arise from the issue s to
whether or not attitude has a characteristic structure.
(a) What is the relative Utility of the cognitive-affective-conative
approach s compared to that of the means-end approach?
(b) What are the relationships between measures derived from each
approach?
(c) Are measures of the cognitive, affective, and conative components
of attitude so highly intercorrelated that they should be thought of s
measuring the same thing or are they relatively independent entities?
According to Triandis (1971), the evidence on this issue is mixed.
(d) If there is an affective component to attitude, what are the specific
feelings (e.g., fear, contempt, compassion) subsumed thereby? According
to Harding, Proshansky, Kutner, and Chein (1969), there has been little
work done with respect to the feelings involved in interethnic attitudes.
Specific emotions such s anger, admiration, and empathy cannot be
readily inferred from the typical Instruments used to tap the affective
component, e.g., positive evaluations on semantic differential scales.
Aside from the problem of determining WHICH scales constitute the
affective component (e.g., are evaluations in terms of BEAUTY more
affective than those of HEIGHT and are evaluations of both more affective
than those of INTELLIGENCE?), there is the problem of determining what
are the feelings which underlie such responses.
(2) To what extent can language attitudes be componentialized ? Two
types of factorization may be possible: in terms of the structure of
attitude (e.g., means-end or cognitive-affective-conative components)
and in terms of the object of the attitude. With respect to the latter, the
attitudes of a speech Community towards a language may be favorable
for use in some societal domains but not in others. For example, some
ultraorthodox Jews feel that Hebrew should be reserved for prayer and
for the study of religious texts. The use of Hebrew for secular purposes
is anathema to them. Thus, it would be misleading to assess attitude
towards a language globally or in an uncontextualized fashion or for
one domain alone. Similarly, attitudes of a diglossic speech Community
towards its functionally differentiated language varieties differ globally
but they also differ s a function of social context, with one variety being
favorably viewed for use in one set of contexts and the other variety
being favorably viewed for use in another (Ferguson, 1959a).
(3) Do the features attributed to languages have a characteristic
structure? Ferguson (1959b) has pointed out that most mother-tongue
groups positively value their first language but that the positive features
which they attribute to it may vary from group to group. Thus for example one mother-tongue group may consider its language particularly
well-suited for rhetoric whereas another may consider its mother tongue
particularly well-suited for singing. If languages were rated in terms of
many different characteristics, would the ratings form different factors
and would the same factors be found from language to language?
(4) Ferguson (1959a) also notes that occasionally a speech Community
negatively evaluates its language. This observation raises two related
questions:
(a) For languages which are negatively evaluated by their Speakers,
are the attributes the same (e.g., sloppy, careless, undignified, incorrect,
deviant, unfit for logical thought or serious intellectual expression)?
(b) If a language is negatively evaluated both by its Speakers and by
others, do both groups attribute the same characteristics to it?
(5) What features of a group give its language high prestige? Gumperz
(1958) points out that in India the language varieties spoken by the
ritually highest groups do not necessarily have the highest prestige,
and Nader (1970: 279) suggests that "the prestige factor which may
encourage admiration, borrowing, or emulation in language need not
be related to the affluent position of one group or another, or of one

individual or another".
The Determinants of Attitde
(6) To what extent do responses to aural Stimuli, such s matched-guise
or mirror-image responses, reflect attitudes toward a language s the
symbol of a group and to what extent do they reflect attitudes toward
the language itself? Responses have typically been interpreted s representing reactions to groups (and thus to language s symbol). However, there may be present in these responses a component of attitude
toward the language itself.
(7) What is the influence of context on matched-guise and mirrorimage responses? Agheyisi and Fishman (1970) suggest that differential
responses to two language guises may reflect in part the interaction
between context and language. Thus, for example, if context is not
varied and if one language is considered inappropriate for use in that
context by members of a speech Community, the Speakers employing
that language guise may be downgraded.
(8) What are the bases of language-attitude responses? To what
extent do they reflect a stereotype in the sense of a failure of rationality ?
To what extent are they based on direct experience ? To what extent are
they generalizations from attitudes toward related constructs (e.g.,
attitudes towards the functions allocated to the language or attitudes
towards Speakers of the language) ?
(9) What are the circumstances under which a language comes to be
negatively evaluated by its Speakers? Conversely, what are the circumstances under which Speakers who formerly devalued their language
come to revalue it positively? The current revaluation of Black English
by many (but by no means all) Black Americans is a case in point.
(10) What has the greater effect on liking, demographic or ideological
similarity? In the United States, for example, does a Black Separatist
prefer a Black integrationist or a White segregationist? This issue, a
perennial one in attitude research (see McGuire, 1969), is one for which
language attitude researchers can be particularly helpful. The question
can be studied by a verbal-guise procedure in which language and
message can be systematically varied, keeping the Speakers constant.
Here, language could serve s a marker of group membership and the
message could serve s a marker of ideological affiliation.
10
The Effects of Attitde

(11) What are the behavioral consequences of language attitude? Typically, only modest relationships are found between attitude measures
and the overt behaviors which such scores are designed to predict. Can
a general Statement be made about the relationships between language
attitudes, language behavior, and behavior toward language?
(12) What is the effect of language on persuasion ? There is an extensive
literature on the effects on persuasion of manipulating various channel
variables (see McGuire, 1969) but little if anything on the effect on persuasion of manipulating language. Would a bilingual be more persuaded by
a message in one of bis languages than in another even if the attributed
source remained the same?
(13) What are the relative contributions of channel and source
variables to persuasiveness ? This question, another issue in general
attitude research (see McGuire, 1969), could be tackled by language
attitude researchers, because language can be used both s a channel and
s a source variable. For example, bilingual respondents could be asked
to react to messages in each of two languages spoken by a source identified s the SAME bilingual authority. Here, language would serve s a
channel variable. Respondents could also be asked to react to messages
in each of two languages where the messages are identified s being
delivered by two DIFFERENT authorities. Here, language would serve
s a source variable. The relative influence on persuasiveness exerted
by language s a channel and s a source variable could then be
compared.
(14) What is the relationship between the prestige of two language
varieties and the type and extent of borrowing one from the other?
Emulation is not always unidirectional, with Speakers of the less prestigeful variety borrowing from the more prestigeful. The reverse direction
is sometimes found s well. Emulation, furthermore, is selective. Some
features but not others are borrowed. In Israel, for example, Sephardi
and Ashkenazi Hebrew are not generally accorded the same prestige,
yet Ashkenazim borrow some features from Sephardi Hebrew and
Sephardim borrow some features from Ashkenazi Hebrew. Can generalizations be formulated and tested with respect to the influence of
relative prestige on the issue of who borrows what features from
whom?
11
The Measurement of Attitde

(15) What relationships exist among attitude scores obtained from
diflferent classes of attitudinal measurements (physiological or psychological reaction, situational behavior, verbal report)?
(16) What relationships exist among measurements of attitudes s
obtained toward an object, toward a Situation, and toward the object
in that Situation? According to Rokeach (1968: 119), the study of attitudes towards situations has been neglected. "The Splitting off of attitudetoward-situation from attitude-toward-object has severely retarded the
growth of attitude theory. It has resulted in unsophisticated attempts
to predict behavior accurately on the basis of a single attitude-towardobject, ignoring the equally relevant attitude-toward-situation." If
attitude toward Situation were included either by means of multipleregression analysis or by means of contextualizing the attitude object
(obtaining measures of attitude towards the object in specific contexts),
improved correlations might be obtained between attitude scores and
overt, non-attitudinal behavior. With respect to language attitudes,
we can ask what relationship exists between context-free and contextladen measures. We can also ask whether one type of measure has higher
validity than the other.
(17) Do indirect measures of attitude (i.e., measures whose purpose
is not apparent to the respondent) have higher validities than direct
measures? Triandis (1971) states that little research has been done on
their respective validities. With respect to language attitude, we can ask
whether the measure of interethnic attitudes based on reactions to aural
Stimuli are more valid than measures based on responses obtained from
conventional intergroup attitude scales. Lambert (1967: 94) Claims that
the matched-guise technique "appears to reveal judges' more private
reactions to the contrasting group than direct attitude questionnaires
do, but much more research is needed to adequately assess its power
in this regard". On the other band, Webster and Kramer (1968) found
an inverted U-shaped relationship between degree of prejudice towards
French Canadians and evaluational reactions to a French Canadian
guise. The issue of the relative validity of direct and indirect languageattitude measures raises the allied question of the relationship which
exists between the two types of measure. To what extent do they assess
the same attribute?
(18) How reliable are indirect measures of language attitude? Indirect attitude measures are typically less reliable than direct ones. Is
12
this also true for indirect language-attitude measures? If so, it could

account for the low correlations often reported between indirect languageattitude measures and other variables (see, for example, Lambert,
Hodgson, Gardner, and Fillenbaum, 1960).
WORK IN PROGRESS
The issues raised above have motivated a series of stuaies by faculty

and students at the Language Behavior Section of the Hebrew University
School of Education.2 All but one of the researches have been undertaken with the same group of respondents who have been intensively
studied. These respondents are 65 high-school students in Jerusalem,
drawn primarily from two government schools, one following a more
secularly-oriented curriculum, the other following a more religiouslyoriented curriculum. About equal numbers of each sex participated. All
of the respondents, who were drawn from the eleventh and twelfth-grade
classes, attended, over a two-week period, six sessions which lasted from
one to two hours each. The respondents attended these sessions during
their Passover vacation and were compensated for their participation.
Their responses to all items were anonymous. In addition to demographic
Information, the following broad classes of Information were obtained
from each respondent: attitude (both language attitudes and other
attitudes), language usage, and language proficiency.
Attitude Measures
Both direct and indirect attitude measures were employed. The matchedguise procedure developed by Lambert and his co-workers was used
with respect to two ethnic groups: American immigrants to Israel and
Israelis born in Israel (Sabras). The Americans were represented by
American-accented English and Hebrew. The Sabras were represented
by Israeli-accented English and Hebrew. An innovation in the present
study is the use of both context-free and context-laden Stimuli. The
context-free Stimuli consisted of Speakers reciting a list of calendar
dates. The contextualized Stimuli consisted of two sets of conversations.
One set was between a father and son at home. The son asked permission
2
Among the students involved in this research are Bryna Bogoch, Elizabeth Nadel,
Phyllis Rosenbaum, Yehudit Rosenbaum, Barbara Schaier, and Jean Vermel.
13
to go on a trip with some friends during bis vacation, and the father,
after raising several objections, finally agreed. The other set of conversations was between a teacher and Student (both male) at school. The
Student complained to the teacher about having failed an examination
in which bis answers (but not bis procedures) were correct. The teacher,
after giving bim a lecture on the importance of method, finally agreed
to allow the Student to take a substitute examination. The scripts for
the dialogues were based on role-playing protocols obtained from
Israeli high-school students. Respondents listened to sixteen voices
reciting dates and to eight versions of each of the two dialogues. They
listened to these over a three- day period, with one set of Stimuli (dialogue
or dates) per day. On the fourth day, they listened again to the same set
of Stimuli to which they had reacted on the first day. Inasmuch s the
order of administration was counterbalanced for contextualized and
non-contextualized Stimuli, test-retest reliabilities can be compared for
the two types of measure. The respondents listened to the indirect
attitude Stimuli in individual language laboratory booths, permitting
us to counterbalance the order of presentation of individual voices (for
the dates) and pairs of voices (for the dialogues).
It should be noted that the procedure described makes it possible
to derive separate scores for attitude toward group and attitude toward
language. Thus, for example, attitude towards Americans can be inferred from responses to American-accented guises (in both English and
Hebrew), and attitudes toward English can be inferred from responses
to the English guises (both Americans and Sabras). Similarly, attitude
scores can be derived for Sabras and for Hebrew. Both the context-free
and contextualized measures will yield scores for attitudes toward
English, Hebrew, Americans, and Sabras. Thus, context-free and contextualized scores can be compared for the same referents. In addition,
the contextualized measures will yield scores for attitudes toward fathers,
sons, teachers, and students.
Three types of responses were elicited to the aural Stimuli. Respondents
were asked to rate each voice with respect to each of six attributes,
previously obtained from Israeli high-school students s distinguishing
American immigrants from Sabras, e.g., high Standard of living, religiously observant, etc. Respondents were asked to indicate the extent to which
each voice aroused each of six feelings, e.g., liking, distrust. Finally,
respondents were asked to rate each voice with respect to the probability
that they would act in a given way should they meet the Speaker in each
of six situations previously described. One Situation, for example, is s
14
follows: "This person stops you on the street and asks you to help him
find bis way to a place which is nearby but not in the direction you are
going. It is hard to explain how to go there without taking him there
yourself. You are on your way to school and are exactly on time. You
know that if you take him to where he wants to go you may be late.
What is the probability that you will take him to where he wants to go?"
These three types of item were intended to elicit cognitive, affective,
and conative responses respectively. The 'affective' items are notable
for their direct probes of emotional content. Whether these a priori
clusters will emerge by factor analysis is of course one of the issues we
are investigating.
In addition to attempting to obtain cognitive, affective, and conative
measures, we have also obtained means-end attitude scores. However,
the latter have been obtained only with respect to English and Hebrew.
Respondents were asked to rate the personal importance of each of a
number of values or goals. They were then asked to rate the degree to
which each of the following personally facilitate or hinder attainment
of each goal: fluency in spoken Hebrew, fluency in spoken English,
being able to read and write Hebrew fluently, and being able to read
and write English fluently.3
Parallel to the cognitive, affective, and conative scores obtained by
indirect methods are cognitive, affective, and conative scores obtained
by means of conventional direct questionnaires. Respondents were asked
to rate, on the same scales that were employed for the matched-guise
procedure, Sabras and each of the following Immigrant groups: Americans, Frenchmen, Iraqis, and Russians. Respondents were asked to
rate each of these groups separately for males of the same age s the
respondent and for males of the same age s the respondent's parents.
In addition to the attitude scales described thus far, respondents were
given questionnaires designed to elicit directly attitudes towards education, the family, the typical high-school teacher, the respondent's
current English teacher, Sabras, American immigrants (these last two
scales were given in addition to the direct scales on Americans and Sabras
mentioned above with the latter, respondents were asked to rate the
degree of their agreement with sentences describing each group), and
each of six languages: Arabic, English, French, Hebrew, Russian, and
Yiddish. The six-language questionnaire included questions testing
knowledge about each language (e.g., number of Speakers, size of
3
The means-end study is being conducted by Jean Vermel.

15
lexicon) and a set of semantic differential items for each language.4

Responses to the semantic differential items should enable us to test the
consistency of the factor structure of language attributes across languages.
Finally, social attitude questionnaires developed by Hofman (1974)
were administered. These yield scores for modernism, nationalism,
conservatism, internationalism; scores with respect to the centrality and
valence of being a Jew, an Israeli, and a Student; scores with respect to
public, private, instrumental, and sentimental orientations toward
language; and scores with respect to traditional versus progressive
orientation towards education.
All of the attitude measures described thus far were obtained by means
of verbal report, although the indirect attitude measures may perhaps
be thought of s situational s well. We also administered an item which
required an overt, non-verbal response. This was part of the six-language
questionnaire. Respondents were asked which languages, if any, they
would like to study during their summer vacation, if second-language
instruction were offered free at the appropriate level by their high school.
They were also asked to indicate their choices (if any) on a stamped,
addressed postal card which was provided and to mail it so that the
addressee (a person at a research Institute ostensibly not connected with
the university) might know the extent of interest in such courses. Two
overt responses can be noted in this connection: whether or not the
respondent mails the card and, if he mails it, whether or not he puts
bis return address on it. (The latter was not requested.)
Language U sage Measures
A language-background questionnaire was administered in which
respondents were asked about the frequency of their usage of Hebrew
and English for specific contexts and functions.
Language Proficiency Measures
Respondents were administered a cloze procedure in English and in
Hebrew. They were asked to restore words to written passages from
which every sixth word had been deleted. For native Speakers, the cloze
can be treated s a measure of reading comprehension whereas, for nonnative Speakers, it can be viewed s a global measure of second-language
4
The six-language questionnaire study is being conducted by Bryna Bogoch.

16
proficiency (Oller, Bowen, Dien, and Mason, 1972). In addition, spoken

measures were obtained for each respondent. Respondents were asked
to read aloud the lists of English and Hebrew dates which they had
previously heard during the matched-guise procedures. (The dates had
been selected to reveal maximally American accent in Hebrew and
Israeli accent in English.) They were also asked to summarize each of
the two dialogues they heard: one in English and the other in Hebrew.
Finally, they were asked to read a list of 16 unrelated English sentences,
each representing an English pronunciation problem for native Speakers
of Hebrew. The spoken responses were tape-recorded in individual
language laboratory booths. The language proficiency and usage scores,
along with the demographic variables, will serve s criterion scores for
the attitude measures.
Demographic Variables
Among the Information sought by a personal background questionnaire
was age, sex, grade, school, degree of family's religious observance,
place of birth, recency of arrival if not born in Israel, school subjects
liked most, school subjects liked least, grade-point average, career plans,
and occupation of head of household.
The Effect of Language on Persuasion
All of the procedures described thus far were administered to the same
respondents. Another study was carried out with a group of ArabicHebrew bilingual Muslim adults in Jericho, a city in the Administered
Territories. The purpose of the study was to investigate the effect of
language on persuasion.5
Four one-minute passages were recorded by a fluent Speaker of Arabic
and Hebrew. One passage decried the evils of tobacco and cited scientific
evidence in support of its argument. The other decried the eifects of
liquor and cited traditional arguments in support of its position. The
different types of argument were employed in order to maximize for each
message the difference between responses to the Arabic and Hebrew
versions. It was thought that in Hebrew a scientific argument might be
more persuasive than a traditional one and that in Arabic a traditional
5
The study of the effect of language on persuasion was conducted by Barbara

Schaier. A detailed description of the study can be found in Schaier, Cooper, and
Fishman, 1974.
17
Islamic argument might bemore persuasive than a scientific one. Respondents were individually approached. Fach heard one passage in one
language and the other passage in the other. It was made clear to each
respondent that the Speaker of each passage was a Muslim from Jerusalem and that he had recorded both passages. Thus, any difference in the
average response to passages in each language can be attributed to
language and not to the ethnic identity of the Speaker.
Respondents were asked two types of question in order to assess the
two languages' relative effectiveness: direct questions such s whether
or not the respondents agreed with the message and to what extent they
thought the message logical and well-presented, and one indirect question.
The indirect question asked whether or not the respondents thought
that an increased tax should be placed on the commodity in order to
discourage consumption. No differences were found in response to the
direct questions: there was almost unanimous agreement with the
messages' content no matter which language was used. However, the
indirect question elicited a dramatic difference. Twice s many respondents endorsed an increased tax on tobacco when the message was heard
in Hebrew s when it was heard in Arabic, and twice s many respondents
endorsed an increased tax on alcohol when the message was heard in
Arabic s when it was heard in Hebrew. In other words, Hebrew was
more effective than Arabic for an argument based on scientific considerations and Arabic was more effective than Hebrew for an argument
based on traditional considerations, when effectiveness was assessed
indirectly. Work is now in progress to determine the generalizability
of these findings.
SUMMARY
Researchers in attitude theory and in sociolinguistics have largely ignored

the theoretical concerns and empirical findings of the other. This is
unfortunate because each research tradition can enrich the other. In
this paper some issues in attitude and language- attitude theory and
measurement were outlined along with some research procedures that
they have motivated. It is hoped that our data will be useful to workers
in attitude theory and sociolinguistics and that our results will encourage
cross-fertilization between the two spheres of inquiry.
The Hebrew University of Jerusalem,
Yeshiva University, New York
18
REFERENCES
Agheyisi, Rebecca, and Joshua A. Fishman
1970 "Language Attitde Studies: A Brief Survey of Methodological Approaches",
Anthropological Linguistics 12:5, 137-57.
Anisfeld, Moshe, and Wallace E. Lambert
1961 "Social and Psychological Variables in Learning Hebrew", Journal of
Abnormal and Social Psychology 63, 524-29.
De Fleur, Melvin L., and Frank R. Westie
1963 "Attitde s a Scientific Concept", Social Forces 42, 17-31.
Ferguson, Charles A.
1959a "Diglossia", Word 15, 325-40.
1959b "Myths about Arabic", Languages and Linguistics Monograph Series 12
(Georgetown Univc rsity), 75-82.
1972 "Soundings: Some Topics in the Study of Language Attitudes in Multilingual
Areas", Paper presented to the Tri-University Meeting on Language Attitudes, Yeshiva University, January.
Gumperz, John J.
1958 "Dialect Differences and Social Stratification in a North Indian Village",
American Anthropologist 60, 668-81.
Harding, John, Harold Proshansky, Bernard Kutner, and Isidor Chein
1969 "Prejudice and Ethnic Relations", in G. Lindzey and E. Aronson (eds.),
The Handbook of Social Psychology 5, 2nd ed. (Reading, Mass., AddisonWesley), 1-76.
Herman, Simon R.
1961 "Explorations in the Social Psychology of Language Choice", Human
Relations 14,149-64.
Hofman, John
1974 "The Prediction of Success in Language Planning: The Case of Chemists
in Israel", International Journal of the Sociology of Language l, 39-65.
Kimple, James Jr., Robert L. Cooper, and Joshua A. Fishman
1969 "Language Switching and the Interpretation of Conversations", Lingua
23,127-34.
Labov, William
1963 "The Social Motivation of a Sound Change", Word 19, 273-309.
1966 The Social Stratification of English in New York City Washington, D. C,
Center for Applied Linguistics).
Lambert, Wallace E.
1967 "A Social Psychology of Bilingualism", Journal of Social Issues 23:2, 91-109.
Lambert, Wallace E., M. Anisfeld, and G. Yeni-Komshian
1965 "Evaluational Reactions of Jewish and Arab Adolescents to Dialect and
Language Variations", Journal of Personality and Social Psychology 2,84-90.
Lambert, Wallace E., R. C. Gardner, R. Olton, and K. Tunstall
1968 "A Study of the Roles of Attitudes and Motivation in Second-Language
Learning", in: J. A. Fishman (ed.), Readings in the Sociology of Language
(The Hague, Mouton), 473-91.
McGuire, William J.
1969 "The Nature of Attitudes and Attitde Change", in: G. Lindzey and E.
Aronson (eds.), The Handbook of Social Psychology 3, 2nd ed. (Reading,
Mass., Addison-Wesley), 136-314.
Nader, Laura
1968 "A Note on Attitudes and the Use of Language", in: J. A. Fishman (ed.),
Readings in the Sociology of Language (The Hague, Mouton), 276-81.
19
Oller, John W. Jr., J. Donald Bowen, Ton That Dien, and Victor W. Mason
1972 "Cloze Tests in English, Thai, and Vietnamese: Native and Non-native
Performance", Language Learning, 22,1-15.
Rokeach, Milton
1968 Belief s, Attitudes, and Valucs: A Theory of Organisation and Change (San
Francisco, Jossey-Bass).
Schaier, Barbara, Robert L. Cooper, and Joshua A. Fishman
1974 Language, Technology, and Persuation: a Middle-Eastern Example. Paper
prepared for the Vlllth World Congress of Sociology, Toronto, August.
Seligman, C. R., W., E. Lambert, and G. R. Tucker
1972 "The Effects of Speech Style and Other Attributes on Teachers' Attitudes
towards Pupils", in: W.E. Lambert, Language, Psychology, and Culture
(Stanford, Stanford University Press), 338-50.
Shuy, Roger, and Ralph W. Fasold (eds.)
1973 Language Attitudes: Current Trends and Prospects (Washington, D. C.,
Georgetown University Press).
Triandis, Harry C.
1971 Attitde and Attitde Change (New York, Wiley).
Webster, William G., and Eraest Kramer
1968 "Attitudes and Evaluational Reactions to Accented English Speech",
Journal of Social Psychology 75, 231-40.
Wolff, Hans
1959 "Intelligibility and Inter-ethnic Attitudes", Anthropological Linguistics l,
34-41.

The Study of Language Attitudes

Uploaded by

Copyright:

Available Formats

The Study of Language Attitudes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Study of Language Attitudes

Uploaded by

Copyright:

Available Formats

How do the authors define language attitude?

How do the authors define language attitude?

What are some ways that language attitude has been used as a variable in sociolinguistic studies?

What are some ways that language attitude has been used as a variable in sociolinguistic studies?

THE STUDY OF LANGUAGE ATTITUDES 1

ROBERT L. COOPER AND JOSHUA A. F1SHMAN

Attitde has served s a variable in many sociolinguistic studies (for a

ROBERT L. COOPER AND JOSHUA A. FISHMAN

ment techniques developed by sociolinguists, notably the use of aural