There are two reasons to be interested in the study of endangered languages. They can be termed 'contributions to scientific inquiry' and 'cultural preservation,' and are represented by the following two quotes, respectively:
At this point in the history of linguistics, at least, each language offering testimony for linguistic theory brings something important, and heretofore not known or not yet integrated into the theory. In many cases, data from a 'new' language forces changes in the developing theory, and in some cases, linguistic diversity sets an entirely new agenda (Hale 1998:194).Documentation of languages that are near extinction will insure that these languages can contribute to scientific inquiry and to the cultural knowledge of those who are losing their ancestral language… As language represents an important component of any culture, the loss of a language can result in the loss of cultural identity (Goebl et al. 1996:659). This interest has developed out of my research in language contact issues, as endangered languages are generally endangered because of contact with some other language. Such languages are particularly good targets of research for scholars interested in language change. In my experience, however, such scholars do not receive much background on training in many fieldwork issues, and this paper brings attention to some important issues.
Another possible way to determine whether a language is endangered is by applying findings from sociolinguistic research concerning which groups of speakers are generally trendsetters in language use. Direct assessment of a language's status might be aided by interpreting variation data in light of what we know about prestige/standard forms and who uses them. As a rough example, if the young women in a community are not using the language in question, or are using it to a lesser degree than their male peers, it may indicate a dim future for the language, regardless of the total number of speakers. Similar comparisons are possible between other pairs of groups within a community that are comparatively more and less likely to be leaders of change.
One complication to this is the following apparent paradox: women lead in change (Labov 1990), but, over and over, in many fieldwork situations, it seems that women are responsible for maintaining the heritage language more than men. While this conclusion is not based on a large quantitative study, I can provide several examples. Haarmann (1985, cited in Grenoble & Whaley 1998b:49) reports that as Russian spread in the Soviet Union, males retained the native language less frequently than women. Janet Fuller's research in Pennsylvania German communities found that women were more like to speak Pennsylvania German than men. Men had more frequently given up Pennsylvania German in favor of English (Janet Fuller, p.c.). In my fieldwork in a village in southern Italy where a Francoprovençal dialect is spoken alongside of Italian (discussed in Section 5), the overwhelming trend is that women who marry into the community learn Faetar, but men in the same position do not. In Wendy Ryback-Soucy’s (1998) research on the Franco-Americans of Manchester, NH, she was able to find many more women than men who were willing to help her study the dying French language of that community. Furthermore, of the people she did record, a higher proportion of the women had maintained French. In Michael Clyne’s research in Australia (cited in Boyd forthcoming:2) gender is found to be an important factor in language maintenance: 'men tended to shift to the majority language more often than women.' Boyd, in research on language maintenance in the Nordic region, found that there were more instances of mothers speaking a minority language than fathers, especially among the Finns she sampled (though this is connected to the overall rate of exogamous marriages of each type). In all of these cases (as Boyd points out specifically regarding Clyne’s work), there are, of course, socio-cultural pressures which are responsible for women bearing most of the load of language maintenance. It may not be gender per se, but rather responsibility for child care and the more restricted social domain that women in rural areas often have that causes them to maintain the heritage language longer than men. Might the spread of women’s liberation in this century and the next be an indirect cause of the increasing rate of language death?
[A] pervasive predictor of the continued use of a language is the prestige attached to it... a language typically grows in prestige if [although not necessarily because] it is associated with a rich literary tradition (Grenoble & Whaley 1998a:ix).
Successful efforts to halt language decline typically involve bilingual education and cultural awareness programs, governmental support, and most importantly, the community's desire to maintain their ancestral language (Goebl et al. 1996:663).The connection between literacy and language vitality is due, at least in part, to the increased awareness of language that literacy brings. Illustrating this tie, Ostiguy et al. (1998) describes a study contrasting elementary and secondary school students in Quebec. They examine the students' sensitivity to register-correlated variables and find that the older students, having achieved greater literacy, are more conscious of morphological and syntactic variables, which are represented in the orthography, than of phonological variation. They report no significant difference between 'orthographically visible' and other variants for the younger students. Thus, the development of literacy is shown to make speakers more aware of their language.
Given that linguists have the ability to affect the future of a language while they are in the field, it is important that we be aware of certain issues which are common knowledge in other fields, such as anthropology, psychology, and sociology, but not necessarily in linguistics. There are a number of considerations that need to be taken into account before rushing headlong into the preservation of a language, although on the surface, this would seem to serve both scientific and cultural heritage purposes.
'prior ideological clarification'... calls for an open, honest assessment of the state of the language and how people really feel about using and preserving it, replacing wishful thinking and denial of reality with an honest evaluation leading to realistic recommendations. Personal and community attitude are as important as–if not more important than–the technical aspects that are less emotional... We often find that those who vote 'Yes' to 'save the language and culture' expect someone else to 'save' it for others, with no personal effort, commitment, or involvement of the voter.Why is this? Although it might seem that preserving a language can only be beneficial, both to science and to the community that speaks a language, there are a number of reasons that a community, or individual speakers, may not be enthusiastic or even supportive of language preservation efforts. I discuss some briefly here; they are more thoroughly discussed in Marks & Dauenhauer (1998:64-68).
Perhaps the toughest to tackle is 'lack of approval' from God, the belief that if God wanted the language to survive, it would not be dying, and that no artificial means of preservation should be implemented. Such views are supported by the spread of dominant religions which do not make mention of indigeneous languages and cultural practices. Similarly, but without invoking deities, there are those who believe that languages run a natural course and there is no need to attempt to alter it. In such situations, interventions by field workers may not be welcome and the linguist needs to find other ways to reward research participants.
Second, preservation efforts that depend on the development of literacy in the community may cause a threat to the speakers' image of cultural competency. If a language has traditionally been oral-only, then adding a written component drastically changes the very nature of the language, particularly when this addition is implemented by an outsider. In some way, this suggests that the native (illiterate) speakers are not performing adequately, in their own language.A related issue is resistance to the need for training to develop literacy because of the perception of simplicity of one's native language. Not all people are aware of the need for extensive time and effort to learn a language, written or spoken. For example, some members of the Faetar community were amazed at how long it took me to learn Faetar, since all of them had easily accomplished it as toddlers. Similarly, the mother of a Peace Corps worker visited her son in Niger. The son had learned enough Hausa to communicate, but the mother knew none and was unable to communicate with any of the locals. Several woman asked the son why he didn't just 'give [his] mother Hausa,' so she could talk to them (Jill Nagy, p.c.).
The resistance to learning a new mode of the language (such as reading or writing) can be increased if community members see the lack of government or administrative support for literacy efforts as indicating the lack of importance of the language. This would be especially true if there is debate as to whether the language should even be offered in the school system.
Finally, as in many anthropological research contexts, there is the issue of 'ownership' of the language, especially of stories and histories told by speakers to the linguist. This is exacerbated when texts are transcribed and translated, as that inevitably changes them.
How much should these factors guide where linguists do fieldwork? Perhaps they should not influence which languages we choose to study, but they should determine how we interact with the speakers of the endangered language and what we try to give back, that is, how we share our findings with them.
The language contact community consists of two neighboring mountaintop villages, Faeto and Celle, with a combined population of about 600, located in southern Italy. Faetar, an endangered language, and Italian are spoken by virtually all the speakers. Most of the inhabitants are retired or unemployed, many are farmers, and a few work in the service industry, both in Faeto and in larger towns in the region. Earlier this century, the population was ten times as large as it is now (Valente 1973:39), but there has been an economically driven mass exodus from rural regions in Italy (as in much of Europe). This trend suggests that Faetar is in danger of disappearing within a few generations. This Francoprovençal dialect exists in southern Italy due to migration from southeastern France around the 14th century. Faetar is, therefore, a doubly endangered language, as Francoprovençal has been virtually exterminated in France by aggressive majority language legislation (Brenzinger 1998:274). Meanwhile, it is undergoing change due to frequent contact with Italian (For more detail, see Nagy 1996).
Also contributing to the rapid decrease in the number of speakers of Faetar is the fact that the language is unwritten, or at least, unread. There are a few speakers who occasionally write down texts in Faetar, each using their own orthography and always providing an Italian translation. Even these writers cannot readily read their texts, and I have never seen anyone attempt to read them except for the sake of exhibiting their novelty.
Being the only linguist to study a particular language, as is often the case for endangered language field workers, with no other scholars to set you back on track if you go astray, brings with it a unique set of problems. One feels a need to be all things to all people. I have had to wear a number of different hats for this research project. In the rest of this section, I describe some of what is under each hat.
I choose this research paradigm, rather than collecting all my data via elicitation sessions with one or two informants, in order to capture the way the language is actually used as a mode of communication. This would not, of course, be possible in severely endangered language situations where the language is no longer used for communication, but exists only as a device for indicating cultural membership on particular occasions. However, where possible, language in its natural domain should be the target of research. As pointed out by Sankoff (1980:48):
[T]he distribution of linguistic features cannot be understood solely in terms of their internal relationships within grammar, but must be seen as part of the broader sociocultural context in which they occur.There is question in the field of endangered language research as to whether severely endangered languages (sometimes called moribund) behave differently than healthy languages (cf., Dorian 1994, Schilling-Estes & Wolfram forthcoming), particularly in terms of forming a homogeneous speech community. Because this is an unsettled issue, it is important to record as many speakers as possible, due to the possibility of wide variation in speech patterns. Enough data to analyse quantitatively must be collected in order to see if, in fact, normal patterns of social and stylistic stratification apply. Here is one example from my fieldwork of the benefit of large-scale fieldwork. One scholar preceded me in sociolinguistic research in Faeto: Dieter Kattenbusch, who conducted fieldwork in Faeto in the 1970's. Based on his observations, Kattenbusch (1979:145) predicted that there would be no speakers of Faetar by the year 2000, as the only inhabitants of Faeto who would be alive then would be those who had been in contact with Italian, a more economically and socially useful language, since youth. The majority of his Faetar survey respondents also thought that the language would die within twenty years (ibid. 143).
Twenty years later, during my fieldwork in Faeto, I hear the same predictions (that Faetar will be dead twenty years from now) and see the same pattern. That is, the speakers who were Italian-speaking adolescents during Kattenbusch's fieldwork have become Faetar-speaking adults; and there is a new batch of Italian speaking adolescents. During adolescence, the Faetani speak (and are reported to speak) Italian even at home, perhaps because they attend high school in a neighboring non-Faetar speaking town. Later, if they stay in Faeto, they resume speaking Faetar, especially if they become parents. (Presumably, many of those who are very anti-Faetar leave when they are old enough to, and therefore, are not present in the research sample.)
The ability to conclude that the lack of adolescent speakers is, at least partially, an age-graded phenomenon, rather than purely a sign of language death, exists only because a large body of data from a wide variety of speakers was collected both by Kattenbusch and by me. If the number of Faetar speakers had been given only as a fraction of the whole population, rather than discretely for various age groups, such a conclusion would not be possible.
To further complicate matters, scholars of language death who attempt to correlate variable linguistic usage with social factors often rely on ‘objective’ factors such as speakers’ demographic characteristics (i.e. age, sex, social status) rather than ‘subjective’ factors such as speakers’ feelings regarding the endangerment of their language and culture (Schilling-Estes & Wolfram, forthcoming).
There is a need, at this stage, to document every possible social and linguistic variable because it is not yet clear which ones will be relevant. Until language death is more clearly understood, we cannot filter out the 'unimportant' or 'irrelevant' details, because we do not know which they are (Grinevald 1998:265).paucity of such research is partially due to the lack of speakers to record in moribund language contexts. Yet, it is exactly this dearth of speakers that makes it important to collect data from a wide variety of speakers in order to understand the behavior of endangered and moribund languages. When there are not enough fluent speakers available to provide an adequate sample, the linguist must consider contributions from semi-speakers, emigrants (who used to be part of the speech community), and non-native speakers (who became members of the community). Such speakers do share much, but not all, of the language and language use norms with core members of the community, forcing us to reconsider what we mean by the concept of 'speech community.' The contributions of linguistic data from such speakers must be carefully considered and not weighted equally with that of native, local speakers.
For this purpose, I have made notes of many aspects of the language and compared the phonemic inventory, allophonic variation, morphological and syntactic structures to descriptions of nearby Italian dialects, standard Italian, and Francoprovençal, in order to develop an accurate structural description of Faetar. Such knowledge is necessary to develop a theoretically-based account of the variation in different linguistic contexts.
Before jumping into it, I conducted a 'prior ideological clarification' check and found a great deal of support within the community for the preservation of Faetar. In addition to the many requests for me to write some kind of book about Faetar, there already exist several preservation-oriented institutions such as the annual tour of the town's church choir performing songs in Faetar throughout the region, a village museum with cultural objects labeled in Faetar, publication of a local magazine and several books with sections written in (variously transcribed) Faetar, and several festival activities such as treasure hunts with the list posted in Faetar.
On a positive note, producing a grammar of Faetar should help slow the demise of the language since large numbers of speakers and government support have been identified as safeguards against the demise of a language (Krauss 1992, Grenoble & Whaley 1998b:31). The latter is possible in this situation if a grammar is created, and it can work toward creating the former.
On the other hand, this activity highlights a significant hurdle to overcome in conducting this sort of fieldwork: the lack of training in applied linguistics. Developing pedagogical skills for teaching people about their language is generally not a part of the training received in a graduate program in linguistics, nor an activity that is rewarded in the academy. This has been succinctly pointed out by Grinevald (1998:155-6) in a discussion of the potential dissonance between the demands of the field and the demands of an academic career:
The point to realize is that there is no division of labor in the field, that the linguists, with their formal education are the main–supposedly expert–resource for whatever project is wanted, from literacy programs to bilingual education programs, to revitalization programs, to translation of legal texts.In addition to handicapping the researcher in her ability to complete the (non-theoretical linguistic) tasks which are expected of her, there can be professional repercussions. While such work is seen as important by scholars involved in endangered language preservation, purely 'descriptive' works are generally not given equal stature with theoretical treatises in the academy–they do not contribute as much to a scholar's professional reputation; and pedagogical tools, such as a grammar written for the (non-linguist) speakers of a language, even less so. This can make it difficult for junior scholars to sufficiently accomplish this aspect of the research, when the demand to produce peer-reviewed publications is high.
Each version of the grammar will include illustrative segments of recorded speech, accompanied by transcription and translation. This database will be stored digitally (on CD-ROM) and made available on the WWW, to which readers of the grammar will be directed. Texts will include samples from interviews, narratives, joke-telling sessions, and descriptions of pictures from a children's storybook. Items will be selected in order to represent the speech of men and women of different generations and of different occupational categories.
Clearly, this indicates the need to don yet another hat–that of the technician. A fieldworker must be
not only willing but able to creatively employ and modify existing software and hardware. This may be in response to practical [and] logistical challenges arising in the field (intermittent access to power supplies; unfavourable recording conditions), and it may be in response to questions and analytical problems that emerge in the course of transcription and statistical analysis (Miriam Meyerhoff, p.c.).
One means of overcoming several of the problems mentioned above is to include native speakers in the fieldwork, analysis, and documentation. This should lead to more productive working relationships and better understanding of the cultural context in which the language is embedded. Perhaps most important for cultural preservation is the greater degree of productivity this provides, in the spirit of 'Give someone a fish and they eat for a day; teach them to fish and they eat for life.' In addition to providing the linguist with greater insight and access to the community, this helps overcome the problem of cultural hegemony, the 'theft' of the language by the researchers. My awareness of this issue can be attributed to Grinevald (1998:150), who points out:
The reality is that there are very strong feelings [in] South American academic and linguistic circles about the necessity to develop linguistics done by Latin Americans, and to establish the institutional base that is needed for it. There is a widespread sentiment that foreign linguists who help these efforts, in a genuine and effective way, are very welcome, while others who continue to function on an individual(istic) basis, mostly concerned with furthering their own career in foreign universities, are not so.I believe that such attitudes are appropriate in all domains of fieldwork, and one of the best contributions a linguist can make to a community is to provide better access to academic instruction for all interested parties.
The major advance is, of course, the development of recording equipment, which has gotten steadily more portable, making high quality audio (and video) field recordings relatively easy and inexpensive to attain. This can allow for a division of labor between a fieldworker with good interview and elicitation skills and others who work in a lab or office on the transcription and analysis of the data. It also allows for more rapid recording of data and greater quantities of it.
Computers, especially portable ones, have also changed the way research is conducted and make the compilation of dictionaries, grammars and teaching texts easier. Desktop publishing, photocopying, and the Internet make such texts easier to distribute, both to other scholars and to members of the community under investigation. These technological advances also allow for long-distance collaborations with linguists and native speakers. Having set up a web-site describing my research (http://english-1.unh.edu/nagy/research.html), I receive e-mail messages from many Faetar speaking Americans that I don't know, and they have contributed to my research. Just this year, I received an e-mail message written in Faetar from Faeto. This is quite an advance– when I first set out to do fieldwork in Faeto in 1992, it was not even possible to arrange for accommodation ahead of time by telephone.Articulatory and auditory phonetics field experiments can be done with a great deal less heavy equipment with the advent of acoustic analysis software. There are, still, however, problems to be overcome in this domain. The lack of a reliable electric source in rural domains where many endangered languages survive is one, and the alternative of carrying a large supply of batteries into the field is not attractive.
Finally, in the near future, automated transcription of unanalyzed languages may become possible, causing a great decrease in the amount of time necessary for such work. Voice recognition transcription systems already exist for much-studied languages like English, but as the technology is generalized, this should provide a useful tool for other languages as well, though of course the transcription cannot be perfected until the language has been analyzed.
Perhaps the best example of the conflict between endangered language preservation and linguistic research is this cautionary tale. Before setting out for my second stint of fieldwork in Faeto, I had spent several days carefully constructing a series of recordings of Faetar words with the length of one consonant gradiently altered. The goal was to conduct a perception test to investigate contrastive consonant length in the language. I had high hopes for solving a perplexing problem based on speakers' reactions to these words and their ability (or lack thereof) to make lexical distinctions based on the phonetic gradation. However, each time I convinced someone to sit down and listen to the recordings on my laptop, the reaction was the same:
'A computer can't speak our language!'
Agheyisi, R.N. 1984. Minor languages in the Nigerian context: prospects and problems. Word 35:235-53.
Boyd, S. (forthcoming). A minority language as mother tongue or father tongue. Does it make a difference? Proceedings of the 3rd International Conference on Maintenance and Loss of Minority Languages. Tilburg: Tilburg University Press.
Brenzinger, M. 1998. Language contact and language displacement. In F. Coulmas (ed.) The Handbook of Sociolinguistics. Malden, MA: Blackwell. 273-284.
Coulmas, F. (ed.) 1998. The Handbook of Sociolinguistics. Blackwell.
Dorian, N. 1994. Varieties of variation in a very small place: Social homogeneity, prestige norms, and linguistic variation. Language 70.4:631-698.
Edwards, J. 1992. Sociopolitical aspects of language maintenance and loss: towards a typology of minority language situations. In W. Fase, K. Jaspaert, and S. Kroon (eds.), Maintenance and loss of minority languages. Amsterdam and Philadelphia: John Benjamins 37-54.
Gal, S. 1978. Peasant men can't get wives. Language in Society 7:1-16.
Goebl, H., P. Nelde, Z. Stary' & W. Wölck. 1996. Contact Linguistics: An International Handbook of Contemporary Research. New York: de Gruyter.
Grenoble, L., And L. Whaley. (eds.) 1998a. Endangered languages: Current issues and future prospects. Cambridge/New York: Cambridge University Press.
---- 1998b. Toward a typology of language endangerment. In Grenoble & Whaley 1998a, 22-54.Grinevald Craig, C. 1998. Language contact and language degeneration. In Coulmas 1998, 257-270.
Haarmann, H. 1985. The impact of group bilingualism in the SU. In I. Kreindler (ed.). Sociolinguistic perspectives on Soviet national languges: their, past, present and future. Berlin: Mouton de Gruyter. 313-344.
Hale, K. On endangered languages and the importance of linguistic diversity. In Grenoble and Whaley 1998a, 192-216.
Kattenbusch, D. 1979. Faeto und Celle: Frankoprovenzalische Sprachkolonie in Süditalien. Ein Beitrag zur soziolinguistik. In Schwarze (ed.) Italienische Sprachwissenschaft. Tübingen: Saarbrücken. 135-146.
Krauss, M. 1992. Endangered languages. Language 68:4-10.
Labov, W. 1990. The interaction of sex and social class in the course of linguistic change. Language Variation and Change 2:205-254.
Marks, M. & R. Dauenhauer. 1998. Technical, emotional, and ideological issues in reversing language shift: examples from Southeast Alaska. In Grenoble and Whaley 1998a, 57-98.
Melillo, A. 1990. La Tutela delle minoranze linguistiche. Bari: Università degli studi di Bari M.A. Thesis.
Nagy, N. 1996. Language contact and language change in the Faetar speech community. University of Pennsylvania Ph.D. dissertation. Philadelphia: Institute for Research in Cognitive Science.
Ostiguy, L., G. Gagné, & L. Laurencelle. 1998. Étude de la sensibilité aux variantes orales soutenues chez des écoliers québécois francophones. In C. Paradis, D. Vincent, D. Deshaies, & M. Laforest, eds. Papers in Sociolinguistics: NWAVE-26 à l'Université Laval. Quebec: Nota bene.
Ryback-Soucy, W. 1998. The French Influenced Dialect of Manchester, NH. Paper presented at the American Dialect Society, New York.
Sankoff, G. 1980. The Social life of language. Philadelphia: University of Pennsylvania Press.
Schilling-Estes, N. & W. Wolfram. (forthcoming). Alternative Models of Dialect Death: Dissipation vs. Concentration. Language.
Schmidt, A. 1990. The loss of Australia's Aboriginal Language Heritage. Canberra: Aboriginal Studies Press.
Valente, V. 1973. Bilinguismo nei dialettofoni delle isole franco-provenzali di Faeto e Celle in Capitanata. In Centro di studio per la dialettologia italiana (ed.).Bilinguismo e diglossia in Italia, Pisa: Pacini. pp. 39-47.