Attitudinal contrasts between minority and majority languages in contact

Wolfgang Wölck (University at Buffalo and Universidad Nacional Mayor de San Marcos, Lima)

Abstract

Formal and informal attitude studies carried out in language contact situations over the past forty years have shown a recurrent pattern of distribution of the social and linguistic evaluations by and of speakers and their linguistic varieties: the majority language/variety and its speakers tend to evoke positive reactions along the instrumental-institutional dimension, being labeled as, e.g., 'rich, educated, successful', while the minority variety and its speakers elicit positive responses in the personal-affective semantic category (e.g., 'smart, honest, responsible'). Negative personal characteristics (e.g., 'dumb, unreliable, deceitful') correspond to the majority, negative institutional ones (e.g., 'poor, uneducated') to the minority variety. This pattern appeared most obviously throughout the several phases of our longitudinal survey of Quechua-Spanish bilingualism in Peru, but was also evident in the diglossic contexts in Scotland and Northern Germany and among African Americans, as well as among French-English bilinguals in Canada. Brief methodological remarks are included in the presentation.

 

1. The relevance of language attitude studies

It was one of the major innovations and a great accomplishment of the newly developing inter-discipline of sociolinguistics in the 60's to have given respondents' reactions and attitudes to language their proper place as important subjective data side by side with traditional "objective" data as a more complete basis for our analysis and understanding of the use and development of human language (cf. Wölck, 1977:744). Psychologists had long given the study of attitudes a prominent place in their study of human behavior (cf. Allport 1935). William Labov (1966) was one of the first linguists to complement his objective analysis of linguistic variation with speakers' subjective reaction measurements. The real introduction of language attitude studies to linguistics, more appropriately to sociolinguistics and particularly to contact linguistics, was made in the work of social psychologists, with Wallace Lambert and his Canadian colleagues playing a leading role (1960 and 1966). Their major goal was the comparative evaluation of two languages in contact and of their speakers.

Much of the methodology used by Lambert and by other prominent students of language attitudes like Giles and Powesland (1975) has a long and well-documented history which is, however, unfamiliar to most modern socio-linguists. According to Oppenheim (1966:105): "of all the tasks or purposes of interviews and questionnaires, that of attitude measurement has undergone the greatest amount of technical development." Some of the best-known techniques of psychological measurement were developed for attitude studies, with Thurstone's milestone contribution going back to 1929, Likert's major works being published in the 1930's, Gutman's in the 1940's. For a recent review of language attitude research in multilingual settings, see Vandermeeren (1996).

2. Methodological principles

2.1. Evaluative terms

Most formal language attitude studies follow the method developed by Osgood, Suci and Tannenbaum (1957) for the 'measurement of meaning', known as the 'semantic differential'. They developed a set of fifty pairs of attributes, many of them polar opposites or antonyms, with supposedly universal applicability to the semantic description or differentiation of popular concepts. These terms were put on the ends of a seven-point scale on which respondents were asked to check their responses to a particular concept or stimulus, e.g., 'snow, sin', or 'foreigner' (p. 49). The attributes fell into three categories or 'factors': evaluation, potency and activity. Typical evaluative pairs were 'good:bad, beautiful:ugly, honest:dishonest'; potency contrasts: 'strong;weak, brave:cowardly, rugged:delicate; activity pairs: 'fast:slow, active:passive, hot:cold' (pp. 36-38). The universality of these terms was so well accepted that they were translated into several other languages and tried and tested with other nationalities. One would, however, only have to look at the title of Michael Jackson's first hit album "BAD" to question the cultural and ethnic universality of some of these terms.

When we prepared the first formal instrument for the study of the attitudes towards Spanish and Quechua in bilingual Peru (Wölck 1972 and 1973), we made a special effort to find community-specific terms for our scales. Even such an apparently unambiguous term as 'simpático' turned out to be unusable in Andean Peru. First, because of the different meaning attribution by the different genders, then because 'simpática', when applied to a woman always meant physically attractive there. In a genderized language the masculine adjective often has a totally different meaning from the feminine; as in (Buffalo) Puerto Rican Spanish, where 'guapo' implies strength and belligerence, 'guapa' easy availablility; and 'honrado' is honest, 'honrada' faithful (Schefter, 1972). An Osgoodian pair, 'boss:worker' which we translated into Spanish as 'jefe:empleado' turned out to have inverted polarity: 'empleado' had obviously more status as someone with a steady job, while 'jefe' was used as a jovial address to any male from children to adults, somewhat like 'gov' in England (cf. the attitude profiles in the Appendix).

In the development of a suitable community-sensitive attitude instrument, the selection of appropriate evaluative terms is the most difficult and most important task. For the description of some successful processes, see Wölck 1973 and 1986.

2.2. Stimuli and juries

The choice of the linguistic stimuli to be evaluated is the next most important task. Since the main goal of a language attitude study is to determine the subjective differences between the two languages in contact, in all our cases between a minority language and the majority language, as many external factors as possible would have to be controlled and held invariable. And since a real bilingual community consists mainly of bilingual speakers, the obvious and natural choice for the presenters of the stimuli would be bilingual individuals fluent in both languages. This natural choice was first made by Wallace Lambert (1966) in his bilingual attitude studies, when he had the same individual present stimuli in both languages, unbeknownst to the jury. Since then, this technique has been known as 'the matched guise'.

More difficult and second in rank of importance is the selection of the stimuli themselves. Again obviously, we want continuous natural discourse; because we know that prosodics and diction weigh as much or more heavily as social indicators than phonology, which is all you get from word lists or reading passages. Fishman's (1972) criticism of the situational (non-)typicality of some stimuli is another serious concern.

The composition of the jury members who react to the stimuli is the last and by no means the least important aspect of the design of the instrument. Like any population sample for a sociolinguistic survey instrument, the jury must be representative of the people studied. The best method appears to be the ethnographic procedure of 'community profiling,' most recently summarized by Labrie and Vandermeeren (1996).

My earliest awareness of the importance of subjective reactions to language goes back to my first fieldwork experience in diglossic Northeastern Scotland (Wölck 1965), where the status attribution to majority 'Sassenach' English differed notably from that of the local 'braid Scots', i.e., the Buchan dialect. Although there was no doubt about the higher status of Standard English (of the Scottish variety), the Standard speakers were often ridiculed as 'daft' and 'ablech' , meaning 'puny'.

3. The data

3.1. Peru

The first large formal language attitude study with representative data and results still easily available to us was part of our longitudinal survey of Quechua-Spanish bilingualism, carried out in four phases between 1968 and 1996 (see Wölck and von Gleich 1973, 1975, 1982, 1983, 1986, 1994 and 2000). Two composite attitude profiles, one from 1969 and the other from 1978, are attached in the Appendix. In each case the speaker is a working class ambilingual, the raters high school seniors and university under-graduates. Note that between the two dates there had been two major political changes affecting the status of Quechua: an education reform in 1972 had provided regulations for widespread Quechua-Spanish bilingual education, and in 1975 Quechua was constitutionally recognized as co-official with Spanish.

The 1969 results already show an interesting tendency of differentiation between the evaluation of the Quechua and Spanish speech samples: Spanish is consistently ranked higher on more objective institutional 'status' attributes, as e.g., urban (ciudad), higher class (clase alta), industrious (trabajador), and educated. Quechua wins on some of the more subjective affective personality traits, as prettier (bonito), stronger (fuerte), responsible and smarter (habil, sabido). - Note the significant crossover between smart and educated in the 1969 profile. And remember our mistake and its effect on the empleado/jefe scale mentioned earlier.Some of this distribution changed ten years later after the rise in the status of Quechua, as apparent in the 1978 profile, where the gap in both categories is smaller, except for the obvious institutional-instrumental differences, as in the rurality, class, and ethnic (cholo=Indian/criollo=Mestizo) scales, which are marked even stronger in1978.

There appears to have been a re-emergence of two of the original attitude dimensions, previously labeled as 'cognitive' and 'affective' (cf. Lambert and Lambert 1964), with the latter still quite appropriate and the former perhaps better re-named as the institutional-instrumental or 'status' dimension.

3.2. Other cases

This division between objective-institutional and subjective-emotive or affective values appears in the description of the attribution to the two varieties in many cases of contact between majority and minority languages. I already mentioned the instance of Scots-English diglossia above. In my work on High-Low German diglossia, I found a similar emotional attachment to Low German (Wölck, 1986). Low German 'Platt' was considered more 'gemütlich', its speakers friendlier. The Institut für Niederdeutsche Sprache (Institute for the Low German Language) in Bremen was then making regular news broadcasts in Low German. Listeners' comments praised them because 'even bad news sounds less terrible in Low German' ("selbst schlechte Nachrichten hören sich auf Platt nicht so schlimm an").

The selection of attributes for Black English and their application to Standard vs. Black English diglossia showed a comparable distribution (Grant and Williams, 1973). Speakers of Black English, dudes or mamas were fly, mellow, heavy, together and cool. Standard speakers, hoochies and joeys, were weak, dull, jive, tack and square. The latter are (negative) opposites of the former, in order. Note the date of the study cited, and expect current attributes to differ. Around the same time, Taylor published a study of teachers' attitudes to the two varieties and also noted that Black English was considered as 'cool' (1973: 392).

In her study of French-English bilingualism among Canadian children in a town across the border, Schneiderman (1975) observed a similar evaluative split. Her test instrument was a puppet show, where the puppets were similar in every other respect except that some spoke French, others English. The audiences considered the English puppets as meaner, cheating, lying, starting fights, more stupid, and they would not invite them to their birthday party; the French puppet was nicer, trustworthy, agreeable, smarter.

In her study of the triglossia or trilingualism among the Amish of Western New York, Lodahl (1997: 58ff.) noted the strongest affective loyalty to the unwritten bilingual dialect of Pennsylvania German, which many prefer to call 'Amish', as their most 'expressive' means of communication, especially when they 'get really excited' and which they use with greater 'ease' than English and definitely (biblical) Standard German. Ease of expression, incidentally, was one of the highest ranked attributes of a real bilingual in a study by Gallardo (1978).

4. Conclusion

There is no doubt about the affective attachment of members of minority communities to their language. It is, therefore, hardly surprising that the positive attitudes towards their home variety will be along the affective dimension, while the institutional status prestige remains with the language of the dominant majority language. Even though we might have predicted this, the research and observations cited above give some reliable evidence of this tendency. The recurrence of this dichotic division has motivated me to include this phenomenon in my list of contact linguistic universals:

'Minority languages evoke more positive personal affective reactions, majority languages more instrumental institutional values' (Wölck 2003: 31 and 36).

What still needs to be determined, however, is the dividing line between the two sets or dimensions. Note, e.g., that in both Peruvian profiles the majority language scores higher towards humility on the arrogance (prepotente) vs. humility scale than the Quechua minority language, on friendliness as well as on the responsibility scale; and after the officialization of Quechua Spanish has overtaken it on the sincerity scale.

Furthermore, this attitude division apparently only applies to natural bilingual and clearly diglossic communities, and not to just any existing difference between higher and lower status varieties of (the same) language. In their study of attitudes towards different varieties or, rather, styles of French used in Canada, d'Anglejan and Tucker (1983: 348) showed that Standard European French was consistently ranked higher than Canadian French on all scales, including the likeability and intelligence scales.

Besides confirming the attitudes and value attributions associated with majority and minority languages in contact, which we might have already suspected if not predicted from our knowledge of the affective loyalty to vernacular or 'native' languages, the formal proof of the attitudinal associations with majority and minority language varieties could help in gauging the chances and prospects of 'revitalizing' ailing minority languages (cf. Fishman, 1996) including the preparation of bilingual education programs as shown in Delgado-Okonkwo (1974), Taylor (1972) and Wölck and Nelde (2000).

© Wolfgang Wölck (University at Buffalo and Universidad Nacional Mayor de San Marcos, Lima)

REFERENCES

Allport, Gordon W. (1935): "Attitudes". In: Murchison, C. (ed.): Handbook of social psycholog. New York: Russell and Russell. pp. 798-844.

d'Anglejan, Alison and G.Richard Tucker (1983) "Sociolinguistic correlates of speech style in Quebec". In: Fasold, R.W. (ed.) (1983): Variation in the Form and Use of Language. Washington: Georgetown University Press. pp. 327-353.

Delgado-Okonkwo, Luz: (1974) Correlations between language attitudes and Achievement in bilingual school children. MA Thesis, SUNY/Buffalo.

Fishman, Joshua A. (1972): "Domains and the relationship between micro-and macro-sociolinguistics". In: Gumperz, J.J. and Hymes, D.H. (ed.) (1972): Directions in sociolinguistics: the ethnography of communication. New York: Holt, Rinehart and Winston. pp. 435-453.

Fishman, Joshua A. (1996): "Language revitalization". In: Goebl, H./Nelde, P.H. / Wölck, W. (ed.) (1996): Contact Linguistics: an international handbook of contemporary research. Vol. I. Berlin-New York: de Gruyter. pp. 902-906.

Gallardo, Andrés (1978): "What's a bilingual?". Buffalo Papers in Linguistics 1.1: pp. 17-42.

Gallup, George (1947): Qualitative measurement of public opinion: the quintamensional plan of question design. Princeton, NJ: American Institute of Public Opinion.

Giles, Howard and P.F. Powesland (1975) Speech style and social evaluation. New York: Academic Press.

Grant, Virginia and Cynthia Williams (1973): "Value terms in the Black vernacular" Unpubl. Research Report, SUNY/ Buffalo.

Guttman, Louis (1950): "The basis for scalogram analysis". In: Stouffer,S.A. (ed.): Measurement and prediction. Princeton, NJ: Princeton University Press.

Labov, William (1966): "Hypercorrection by the lower middle class as a factor in linguistic change". In: W. Bright, ed. Sociolinguistics. The Hague: Mouton. pp. 84-113.

Labrie, Normand and Vandermeeren, Sonja (1996): "L'analyse du profil de la communauté". In: Goebl, H./Nelde, P.H./Wölck, W. (ed.) (1996): Contact Linguistics: an international handbook of contemporary research. Vol. I. Berlin-New York: de Gruyter.

Lambert, Wallace E., Gardner, R.C. and S. Fillenbaum (1960): "Evaluational reactions to spoken language." In: Journal of Abnormal and Social Psychology 60. pp. 44-51.

Lambert, Wallace E., H. Frankel and R.A. Tucker (1966) "Judging personality through speech: a French-Canadian example". In: Journal of Communication 16. pp. 304-321.

Likert, Rensis (1932): A technique for the measurement of attitudes. Archives of Psychology No. 140.

Lodahl, Sonja (1997) Why they speak the way they do: Language use and attitudes among the Amish of Conewango Valley, New York. MA Thesis, University at Buffalo.

Nelde, Peter H., S. Vandermeeren and W. Wölck (1991) Interkulturelle Mehrsprachigkeit: Eine kontaktlinguistische Umfrage in Fünfkirchen. Bonn: Dümmler and Munich: Südostdeutsches Kulturwerk.

Oppenheim. Arthur N. (1966) Questionnaire design and attitude Measurement. New York: Basic Books.

Osgood, Charles E., G.J. Suci and P.H. Tannenbaum (1957): The measurement of meaning. Urbana, IL: University of Illinois Press. (repr. 1965)

Schefter, Maria (1972): The Puerto-Ricans of Buffalo, NY. Unpubl. MA Project, SUNY/Buffalo.

Schneiderman, Eta (1975a): Attitudinal determinants of the linguistic behaviour of French-English bilinguals in Welland, Ontario. Doctoral Diss. SUNY/Buffalo.

Schneiderman, Eta (1975b): "An examination of the ethnic and linguistic attitudes of bilingual children" International Review of Applied Linguistics 13:

Taylor, Orlando L. (1973) "Teachers' attitudes toward Black and non-Standard English as measured by the language attitude scale". In: Shuy, R.W. (ed.) (1973): Language Attitudes: Current Trends and Prospects. Washington, D.C.

Thurstone, L. L. and E.J. Chave (1929): The measurement of attitudes. Chicago: University of Chicago Press.

Vandermeeren, Sonja (1996): "Sprachattitüde". In: Goebl, H./Nelde, P.H./ Wölck, W. (ed.) (1996): Contact Linguistics: an international handbook of contemporary research. Vol. I. Berlin-New York: de Gruyter. pp. 692-702.

von Gleich, Utta (1982): Die soziale und kommunikative Bedeutung des Quechua und Spanischen bei Zweisprachigen in Peru. Doctoral Diss., Universität Hamburg.

von Gleich, Utta and W. Wölck (1994): "Changes in language use and attitudes of Quechua-Spanish bilinguals in Peru". In: Peter Cole et al., (eds.) (1994): Language in the Andes. Newark, DE: LAS. pp. 27-50.

Wölck, Wolfgang (1965): Phonematische Analyse der Sprache von Buchan. Frankfurter Arbeiten aus dem Gebiet der Anglistik und Amerikastudien, Nr. 10. Heidelberg: Carl Winter.

Wölck, Wolfgang (1973): "Attitudes towards Spanish and Quechua in bilingual Peru." In: Shuy, R.W. (ed.) (1973): Language Attitudes: Current Trends and Prospects. Washington, D.C.; pp. 129-147. reprinted in Fasold, R.W. (ed.) (1983): Variation in the Form and Use of Language. Washington: Georgetown University Press. pp. 370-388.

Wölck, Wolfgang (1975): "Metodología de una encuesta sociolingüística del bilingüismo quechua-castellano." In: Matos, R.A. and Ravines, R. (eds.) (1975): Lingüística e Indigenismo Moderno en América. Lima: Instituto de Estudios Peruanos. pp. 337-59.

Wölck, Wolfgang (1976): "Community Profiles: An alternative to linguistic informant selection." In: International Journal of the Sociology of Language 9. pp. 43-57. (reprinted in Linguistics 17.7)

Wölck, Wolfgang (1986a): "Language attitude studies: Problems and suggestions." In: Hartig, M. (ed.) (1986): Perspektiven der angewandten Soziolinguistik. Tübingen: G. Narr. pp. 35-48.

Wölck, Wolfgang (1986b): "Language use and attitudes among teenagers in diglossic Northern Germany." In: Ureland, P.S. and Nelde, P.H. (eds.) (1986): Language Contact in Europe. Tübingen: Niemeyer. pp. 97-109.

Wölck, Wolfgang (2000): "Ketschua: die grösste einheimische Kleinsprache Amerikas." In: Zimmer, St. and Kelz, H.P. (eds.): Die europäischen Kleinsprachen an der Schwelle des 21. Jahrhunderts. Bonn: Nomos.

Wölck, Wolfgang (2003): "Are there universals in contact and conflict linguistics?" In: Bochmann, Klaus et al., (eds.): Methodology of conflict linguistics. St. Augustin: Ansgar. pp. 29-39.

Wölck, Wolfgang and Peter H. Nelde (2000): "Sociolinguistic prerequisites for a German-Hungarian bilingual-bicultural education ptogram." In: Peyton, J. et al. (eds.) (2000): Language in action. Creskill, NJ: Hampton Press. pp. 94-109.

6.1. Standardvariationen und Sprachauffassungen in verschiedenen Sprachkulturen | Standard Variations and Conceptions of Language in Various Language Cultures

Sektionsgruppen | Section Groups | Groupes de sections

       Inhalt | Table of Contents | Contenu  15 Nr.

Webmeister: Peter R. Horn     last change: 16.6.2004