A STUDY OF LINGUISTICALLY "MIXED" FAMILIES IN CATALONIA, SPAIN:

Children of linguistically "mixed" marriages in Catalonia, Spain speak both Catalan and Spanish (Castilian), but often choose between the two languages many times in the course of a day. Numerous factors influence this choice, as this paper demonstrates. The Catalan language, which has a minority status within Spain, is spoken by less than 20% of the population. Catalan clearly occupies a numerically inferior position in comparison to Spanish (Castilian), the native language of over 65% of the country's citizens. Nevertheless, the historic prosperity of Catalonia (northeastern Spain, capital Barcelona), combined with the high status of Catalan relative to Spanish, make speaking Catalan a desirable skill for social acceptance and profession advance in Catalonia.

Until recently, history has worked against the Catalan language. Politically repressed for more than 250 years, but especially during the Franco dictatorship (1939-1975), Catalan has enjoyed a great resurgence since the re-establishment of democracy in Spain in the late 1970's.

In the face of these competing factors, the linguistic behavior of the "mixed" household (one parent speaks Catalan, the other Spanish) becomes an excellent gauge of the language conflict taking place in Catalonia. Problems of bilingual education and language planning are major national questions in countries like Canada, Spain and Belgium. Furthermore, linguistic divisions within a nation often parallel social and economic ones. The children of "mixed" marriages represent a particularly interesting linguistic sample, due to the unusual family and societal pressures upon them, as the following paragraphs will demonstrate.

My field projects (see O'Donnell 1986 and 1988) discussed the linguistically "mixed" household briefly; however, the importance of the linguistic loyalties of those who hear and/or speak both languages in the home became increasingly evident as my research proceeded. According to Strubell (1981: 151), the percentage of Catalans who marry Spanish-speakers in Barcelona rose from 25.4% in 1950 to 34.6% in 1973.^[1] Thus, children of "mixed" marriages are both numerous and significant to the future of the Catalan language and to language planning in general. In the course of my research, I have examined the following questions:

1. Domains of linguistic behavior. Does linguistic use correspond with a specific work, social, or family situation? Three empirical sources provide us with information about language use and linguistic domains within the bilingual family: (a) direct self-reporting about code switching in the household, and in non-domesting social situations (b) actual observation of the family's linguistic interaction by the researcher or his confederates, and (c) the informant's use of Castilianisms.^[2]

2. Linguistic opinions and prejudices. What are the informant's attitudes and biases about his/her languages? Current research (O'Donnell 1986, 1988; Woolard 1982, 1984) indicates several avenues of investigation into linguistic opinions; among these are: (a) unsolicited informant comments about the status of Spanish or Catalan, (b) formal "matched guise" tests which make informants rate Spanish and Catalan speech on such primary traits as "intelligent," "cultured," or "leaderlike," and (c) direct questions about which language is "more appropriate" or "more comfortable" for a given situation.

3. Linguistic loyalty versus linguistic defection. What choices do these individuals, who live on the linguistic border, have to make as they align themselves with the language and culture of one side of the family or the other? One of the best character studies of a "border-dweller" is that of Woolard (1986) in which she follows the linguistic, cultural, and personal developments of "Ana," thirty-year old daughter of Castilian parents who could speak both Catalan and Spanish with equal fluency. Despite her bilingualism and respect for both Castilian and Catalan traditions and customs, Ana still experienced considerable personal conflict about her ethnic and linguistic identity. Numerous "Anas," whose allegiances belong to two different groups, were interviewed for this study. The vast majority of linguistic "defectors" in this study were from Castilian dominant homes, but chose Catalan as their preferential language among friends or classmates.

4. The effects of recent political and administrative changes on language use. What is the current state of linguistic normalization in Catalonia? What progress (or stagnation) has Catalan experienced in the wake of great governmental and institutional changes of the early eighties?

I examine all of this against the backdrop of numerous studies about linguistic prestige in general and the prestige of Catalan in particular.

Prestige

Scholars from Weinreich (1964) to Woolard (1982, 1984) and O'Donnell (1986, 1988) have defined, redefined, and attempted to measure the prestige of various languages. Weinreich (1964: 78) stressed the importance of using a language "as a means of social advance." Other linguists, from Fishman (1964, 1965) to Labov (1972) have demonstrated the inadequacy of the high-versus-low prestige classifications of languages and dialects. Fishman (1965: 68) maintained that what might be a low-prestige dialect or language in one context becomes a symbol of "group membership" or "group solidarity" in another. A goal of this project will be to "redefine [prestige] in terms of the people using it and the situation in which it is used," to use Labov's words (1972: 308).

Woolard (1982) developed a test to examine speakers' attitudes towards the Catalan and Spanish languages. In her test, administered to 240 high-school students in Catalonia, she elicited positive and negative reactions to speakers using Spanish and Catalan. The study showed clearly that Catalan, not Spanish, was rated more favorably in terms of formal prestige, based on primary traits like "cultured," "intelligent" and "leaderlike." My studies provide up-to-date information about linguistic preferences and usage. Like Woolard, I found that Catalan had become the "more refined" language in social situations. However, any conclusions about the advance (or decline) of Catalan will be highly dependent upon which area of Catalonia was studied.

Vallverdú (1985a: 90) defines the "normalization process" as "when there is both an establishment of language norms . . . and an extension of the language to all social levels." Vallverdú (1985) has established several levels of normalization, from stable bilingualism (two languages or dialect varieties which exist side by side without one dominating or replacing the other) to substitutional and "dialectalizing" bilingualism. A substitutional situation exists in Brittany (France) where the Breton language has lost ground to French in both its social and geographical extension. According to Vallverdú (1985a: 43), the Occitan language (Southern France) maintained its position geographically, but has been "fragmented into numerous local dialects or patois, through a process of dialectalizing bilingualism." Linguistic normalization can only occur in an area where stable bilingualism exists.

Unfortunately, in Catalonia, Spanish has long been the dominant language. Woolard (1982: 84) explains that "until 1975, Castilian [Spanish] was the only official language used in government, education, technology, the mass media, and much of the formal business administration." Ninyoles (1971: 165-170) describes the state of Catalonia as one where "total diglossia" exists. In a diglossic situation, two languages (like Catalan and Spanish) exist side by side in a given population, but one of the two is the standardized "high" variety, while the other represents the "low" variety. Despite its high formal prestige (relative to Spanish), Catalan has traditionally occupied the lower position in actual linguistic use. Proscribed from official and mass media use, Catalan became the informal, "street-and-home" language of native-born Catalans.

Strubell, both a scholar and a language planner in the Catalan semi-autonomous government, proposes (1981: 19) that "timid solutions [to the linguistic problem] are not solutions at all." Some feel that Catalan's having survived the Franco dictatorship (1939-1975) proves that a slow, careful normalization process will give positive results in Catalonia. Strubell expresses a view that is much less sanguine (1981: 21-22): "Catalan culture is dying, and the normalization process can make it die more slowly . . . only by attaining full dignity and normalcy of the Catalan language, once the institutional normalization process is completed, can Catalan recover from the injuries it has sustained over so many years." The consequences of failing to normalize Catalan can be extremely grave: Strubell (1981: 26) points to Occitan (southern France) and Gaelic (Ireland) as languages that are dying of neglect. Strubell (1981 and personal communication) indicates two important factors that influence whether immigrants will learn Catalan: (1) the percentage of immigrants as compared to native-born Catalans in an area, and (2) the length of time that an immigrant has lived in Catalonia. In areas where the immigrant population reaches a certain percentage (according to Strubell, about 34%), linguistic assimilation to Catalan drops off dramatically. This "critical percentage" theory is central to my analysis of any linguistic data I obtain, and to how I compare my research to others' works.

Another important factor in our cross-generational study of Catalan-Castilian linguistic hegemony is whether the recent Catalan language planning efforts, especially since 1983, have enhanced the status of Catalan and encouraged its use among non-Catalans and members of bilingual families in Catalunya. Woolard (1982) addressed the question of linguistic "backlash" by Spanish speakers who felt forced to speak Catalan, or rejected or ridiculed if they tried to do so.^[3] Preliminary results from three areas studied (Fraga [Aragon], Arenys de Mar and Mataró [Catalonia]) suggest positive results from the language and educational reforms of the late seventies and early eighties, at least in reported ability to speak Catalan and desire to do. Woolard's conclusions (1988) support this favorable assessment of the campaign for normalízació lingüística 'linguistic normalization.'

Conclusions

In studies of linguistic attitudes and performance, the researcher often expects that he or she will not replicate the results of similar surveys. Linguistic preferences are inherently unstable, but Catalonia's linguistic conflict exacerbates this instability. Thus my findings are sometimes quite different from those of Strubell (1978) or Woolard (1982, 1984), and other researchers. Linguistic research involves plotting points on a graph, and indicating tendencies, rather than presenting stable character traits of a population

The kinds of information I obtain relate to linguistic normalization, language prestige, and the role of the "mixed" marriage in the evolution of Catalan. I continue to find, as I did in O'Donnell (1988), that Catalan is gaining ground in the fight for full normalization. Furthermore, I anticipate that Catalan will continue to be highly rated as a prestige language, as it was by Woolard's (1982, 1984) informants. Certain language planning strategies remain impossible or impracticable in a "nation" like Catalunya, although demographic evidence would suggest their efficacy.^[4] Thus, Strubell's (1978) evidence that a "critical mass" of native Catalan (as compared to the Castilian-speaking immigrant) population must be present in a given municipality for linguistic normalization to take place might argue for demographic "quotas" limiting numbers of immigrants in given areas. Such a suggestion would be unfeasible at best, and could easily be construed as racist at worst. In addition to the dangerous human implications of this population "distribution" theory, the suggestion makes even worse linguistic sense, as demonstrated by Woolard (1982, 1986, 1988). As Woolard writes (1988: 12), Catalan use has spread because "there seems to be a loosening of the bond between Catalan language and native Catalan ethnolinguistic identity. It no longer matters so much to young Catalans who speaks Catalan, but rather simply that it is spoken." Only by reducing the barriers between the two communities, and by finding points de rencontre that motivate the Castilian to learn Catalan and the Catalan to encourage learning it, will the ethnolinguistic boundaries become more permeable.

Despite careful application of survey research techniques to this study and despite the obvious importance of "mixed" families in the Catalan linguistic conflict, certain problems remain in completing such research. The major drawback of any study of "mixed" marriages is that little information about the demographic profile of the "mixed" population exists. Linguistic censuses and local population information do not always separate "mixed" linguistic families from monolingual families. I model my linguistic samples after the general population of the target city, while concentrating on the more numerous younger mixed households. The difficulty in finding "mixed" family informants also makes using several interviewers difficulty, and this may have biased the study.

The children of the "mixed" marriage are the first line of defense, or of expansion, of the Catalan language; especially since the number of "mixed" marriages (using Strubell's calculations) surpassed 35% in Catalonia's most populous area. The number of "mixed" households is increasing in Catalonia, while the Spanish-speaking immigrant population has a higher birthrate than does the native Catalan population. Thus, the bilingual family continually gains importance. Follow-up studies to confirm the tendencies discovered in this survey are a logical outcome of my research. In the three areas I studied (Fraga, [Aragon], Arenys de Mar and Mataró [Catalonia]), the Catalan language had extremely different levels of prestige. In Fraga, part of the "Western Strip" which is linguistically Catalan but politically part of Aragon, children of "mixed marriages" tended to identify their mother's language (whether Catalan or Spanish) as the one they spoke better. Meanwhile, children in the Maresme region of Catalonia (where both Arenys and Mataró are located) tended to identify CATALAN as the language they spoke "with greater ease," whether their mother or father was the Catalan speaker. Some speakers from "mixed" households denied having a Castilian parent, saying, "my father came here very young and is very Catalan." I base my conclusions about the linguistic attitudes of "mixed" marriage children on questions eliciting lexeme choice, statements about linguistic prestige, and the informants' actual use of Catalan (or Spanish) in the interview, and indirect comments about language prestige like the denials of "mixed" parentage mentioned above.

Analysts, whether professional or amateur, of the Catalan linguistic situation adopt several positions on the evolution (or deterioration) of Catalan as a "normalized" language. Some strong advocates of Catalan normalization see favorable tendencies in language use among Catalans, immigrants, and visitors, but warn against any "lowering of our defenses." Others point to the latest newspaper poll showing increases in a given linguistic indicator of Catalan's health (e.g., a slight increase in 25-35 year olds who say they understand Catalan) as a clear indication that the local language is "winning out." Other Catalans, in the course of candid interviews, call their tongue a "llengua cremada" ('burnt-out language') which has the fate it deserves: that of a second-class status.^[5]

While these analyses may reflect their proponents' political strategies more than their actual evaluation of the linguistic state of affairs, there is no proof that the "scare tactics" approach is any more effective than the "bandwagon" method (Catalan is winning, so get on the bandwagon). The linguistic fieldworker in Catalonia, especially one who does not have member status in either ethnolinguistic group, should feel no obligation to support one thesis or another in the linguistic conflict. Few observers, or even casual visitors to Catalonia, can long remain personally neutral in the linguistic and ethnocultural dichotomies of Catalonia, but it is precisely this "taking sides" that exacerbates the conflict and reinforces the barriers between the groups. Nowhere are these personal and group conflicts more acute than in a linguistically "mixed" household, where the bilingual child may speak one language to the mother, while switching to another in the father's presence. Perhaps the personal pressures are greater in a bilingual family than in a bilingual society like Catalonia, but the microcosmic nature of this social unit makes it an excellent sociolinguistic and ethno-cultural barometer of the Catalan conflict.

My three field studies strongly suggested that, in both Catalonia proper and in the Catalan-speaking Fraga area of Aragon, official attempts to endow Catalan with institutional prestige (teaching it in the schools, using it on all street signs and official documents, for example) have furthered both the formal and informal^[6] prestige of Catalan. Among my own informants, this appears in both subtle and obvious ways. Still, the children of Spanish-speaking immigrants or "mixed" marriages who bear Catalan first names and become fluent in the language^[7] of the adopted nació in catalanoparlante schools remain the best proof of Catalan's high prestige and expanded domain. This corresponds with the preliminary results of my Mataró study: almost 80% of the "mixed" households chose CATALAN, not Spanish as their dominant home language.^[8]

Works Cited

Alba, Victor. 1975. Catalonia: A Profile. New York: Praeger Publishers.

Badia, Margarit, Antoni M. 1969. La llengua dels barcelonins. Vol. 1. Barcelona: Edicions 62.

Bastardas i Boada, Albert. 1986. Llengua i immigració. Barcelona: La Magrana.

Blom, Jan-Petter and John Gumperz. 1972. Social Meaning in Linguistic Structure: Code Switching in Norway. In Gumperz and Hymes, eds. 407-434.

Domingo, Xavier. 1987. De la resistencia a la normalización. Cambio-16. March 2, 1987. 63-67.

Ferguson, Charles. 1959. Diglossia. Word 15. 325-340.

Fishman, Joshua A. 1964. Language Maintenance and Language Shift as a Field of Inquiry. Linguistics. 9.32-70

___. 1965. Who Speaks What Language to Whom and When? La Linguistique. 2.67-88.

Giles, Howard and Robert N. St. Clair. 1979. Language and Social Psychology. Baltimore: University Park Press.

Gumperz, John and Dell Hymes, eds. 1972. Directions in Sociolinguistics. New York: Holt, Rinehart and Winston.

Kahn, R. L., and D. F. Cannell. 1957. The Dynamics of Interviewing. New York: John Wiley and Sons.

Labov, William. 1972. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press.

Ninyoles, Rafael-Lluis. 1971. Conflicte Linguistic Valenciá. Substitució linguística i ideologies diglossiques. Palma de Mallorca: Tres i quatre.

O'Donnell, Paul E. 1986. The Castilian-Catalan Border in the Baix Cinca (Aragon): A Lexical Study. Ann Arbor Dissertation. University of Michigan.

___. 1987a. Interviewer and Third-Party Effects in Linguist and Anthropological Interviewing. University of Michigan-Flint.

___. 1987b. Black English Revisited: Style and Grammar in a Social Dialect. Field Study. University of Michigan-Flint.

___. 1988. Catalan and Castilian as Prestige Languages: A Tale of Two Cities. Language Problems and Language Planning 12(3).

Payne, Stanley. 1951. The Art of Asking Questions. Princeton: Princeton University Press.

Robinson, D. and S. Rohde. 1946. Two Experiments with an Anti-Semitism Poll. Journal of Abnormal Psychology. 41.136-144.

Ryan, Ellen Bouchard. 1979. Why Do Low-Prestige Languages Exist? In Giles and St. Clair, eds. 145-157.

Strubell, Miguel. 1978. La llengua dels Matrimonis Mixtos a Barcelona. ICE Barcelona. Unpublished article.

___. 1981. Llengua i pobació a Catalunya. Barcelona: Edicions de la Magrana.

Vallverdú, Francesc. 1981. El conflicto lingüístico en Cataluna: historia y presente. Barcelona: Ediciones Peninsula.

___. 1985a. El fet lingüística com a fet social. 6th ed. Barcelona: Edicions 62.

___. 1985b. Dues llengues: dues funcions? 6th ed. Barcelona: Edicions 62.

Weinreich, Uriel. 1964. Languages in Contact. Findings and Problems. The Hague: Mouton and Co.

Woolard, Kathryn A. 1982. The Problem of Linguistic Prestige: Evidence from Catalonia.

___. 1984. A Formal Measure of Language Attitudes in Barcelona: A Note from Work in Progress. International Journal of the Sociology of Language 47.63-71.

___. 1986. The 'Crisis in Concept of Identity' in Contemporary Catalonia, 1976-1982. McDonogh, ed. Conflict in Catalonia. 54-71.

___. 1988. The Consequences of Political Change and Language Planning for Language Evaluations in Autonomous Catalonia. Under review: Language in Study.

Zang Mier, Jeanne. Catalan: An Atypical Minority Language. Michigan Romance Studies IV. 180-196.

^[1]The percentage of persons from Catalan-speaking families who marry children of castellanoparlantes is certainly lower than Strubel's 34.6%. He found this high percentage by counting Catalan-born informants who married people born outside Catalonia. My study (l989) used a purely linguistic criterion for choosing informants: One parent must be from a Castilian-dominant household, while the other comes from a family where Catalan is the majority language.

^[2]In O'Donnell (1986), I found that my informants were more likely to use Castilianisms for outside-home words like acera 'sidewalk' and buzón 'mailbox;' than for domestic terms like different fruit and vegetable names.

^[3]As Woolard (1982: 87) notes, "if they [Catalan policy makers] do not find politically adequate means of acknowledging the implications of the informal social prestige of [Castilian], they risk backlash from a large and threatened minority."

^[4]The preferred terms in Catalan are nació 'nation' to refer to Catalonia proper (as opposed to els paísos catalans: 'the Catalan lands), and estat 'state,' which refers to Spain as a whole (with all its "autonomous entities").

^[5]One Catalan professional who severely criticized the "co-officiality" of Castilian and Catalan may actually have been reacting to all the inconvenience the legal equality involved. Where documents in his office once appeared in Spanish only, he now had to provide legal and contractual terminology in both languages. Many linguists and anthropologists observing the linguistic conflict in this nació may overlook one obvious factor in language shift and language maintenance: convenience. As Catalan-Spanish co-officiality becomes an accepted fact of life for office workers, lawyers, and public servants, linguistic assimilation may be the most convenient solution for both Catalan- and Spanish-speakers.

^[6]Formal prestige relates to a language's value for social advance (see Weinreich 1964: 78), while informal prestige involves what Fishman (1965: 70) calls its value as a symbol of "group membership," or solidarity.

^[7]Woolard (1988: 12-13) notes that young Catalans no longer care so much "who speaks Catalan, but rather simply that it is spoken. . . . Castilian-speaking listeners no longer penalize second-language speakers of Catalan with reduced solidarity." Furthermore, young Catalans may not even be sure of who addresses them in their native language: "It may well be, in fact, that young urban Catalans of 1987 cannot distinguish native and non-native speech styles as well as their counterparts of 1979." All of this bodes well for the future of full bilingualism in Catalonia.

^[8] I used the concept of a "majority language within the home" (also used in O'Donnell l986 and l988). Some families switch back and forth from Catalan and Castilian, but in all cases I studied, one language dominates in the household, in that it is used by the majority of the family members in their verbal interactions with other household members.