Monday 13 February

Investigating small-scale multilingual communities: an ethnographic perspective from the Vaupés (Northwest Amazonia)

Pattie Epps

While interaction among speakers of different languages is a given in multilingual societies around the world, the dynamics of this interaction and its linguistic outcomes vary considerably in the context of local practices and ideologies. This variability comes into particular focus in small-scale speech communities, where social and cultural contexts are sometimes very different from those that are more typical of contemporary urban and/or globalized communities. The Vaupés region of the northwest Amazon offers a fascinating illustration of the importance of an ethnographic perspective in understanding the dynamics of multilingual interaction and contact-driven change. In this region, many of our expectations about the outcomes of contact appear to be turned on their heads: high levels of multilingualism are accompanied by low rates of lexical borrowing (with more verbs borrowed than nouns), closely constrained code-switching, heavy structural diffusion, and the apparent absence and/or invisibility of shift among local languages. While explanations for these patterns have appealed principally to the local marriage practice of linguistic exogamy, a more fine-grained ethnographic perspective on the dynamics of multilingual interaction indicates that many other practices are relevant to the constellation of outcomes. This study explores two principal challenges in studying multilingually-induced change in small-scale communities like the Vaupés. First, how and to what degree of certainty can we establish the role of contact in shaping the contemporary languages, when large corpora, historical records of the languages, and even fine-grained studies of dialectal variation are for the most part entirely lacking? Second, how have particular cultural, social, and ideological practices and structures contributed to directing the contact-related change that has occurred – such as marriage patterns, specialized discourse forms (myths, incantations, etc.), and beliefs about the roles and responsibilities of individuals within a larger social universe?

Discussant:  Ruth Singer (UMelb/CoEDL)

Tuesday 14 February

1. Big languages in small communities: a variationist sociolinguistic approach to language contact and change

Shana Poplack

The survey methodology characteristic of large-scale sociolinguistic studies of urban settings has often (rightly!) been accused of downplaying if not ignoring many of the key explanatory factors available from fine-grained ethnographic observation. And while it is true that many elusive and unexpected linguistic facts opaque to “quantitative” sociolinguists have emerged from such methods, we are often in the dark about just how representative they are beyond, or even within, the community studied. On the other hand, what survey methodology offers in breadth tends to come at the expense of depth.

In this presentation, I report on the long-term efforts of the University of Ottawa Sociolinguistics Lab to marry ethnographic observation with survey methodology to elucidate key issues of language variation, contact and change. Canada, one of few officially bilingual countries, offers an ideal research venue to address these issues. Its two national languages enjoy grossly unequal status: here French is denigrated, despite its prestige worldwide. This is because its “quality” is widely considered (by francophones and anglophones alike) to be inferior, thanks to the many changes it has purportedly undergone on Canadian soil, due to contact with the majority language, English. The few anglophones remaining in Quebec, the sole Canadian province in which French is the official and majority language, are equally convinced that their English has deteriorated irreparably due to contact with French.

Our research has established that four key components were missing from these ideologies: 1) the distinction between stable linguistic variation and change, 2) systematic characterization of source and recipient linguistic varieties as they are actually spoken (rather than prescribed or intuited), 3) contextualization of these varieties in historical and comparative perspective, and 4) an understanding of the competing roles of community-based and normative forces in promoting or impeding linguistic change.

Over the last few decades we have been engaged in research aiming to identify, date and document the mechanisms and trajectories of change, and especially, to distinguish contact-induced change from internal evolution. These efforts have given rise to a number of methodological innovations in corpus construction, data elicitation and analysis, as well as a renewed realization of the crucial importance of systematic comparison with appropriate benchmark varieties. I detail these developments, explore what they have enabled us to discover and what we have yet to learn, and assess their applicability to other languages, communities and contact situations.

Discussant:  Catherine Travis (ANU/CoEDL)

Wednesday 15 February

Investigating small-scale multilingual communities: a comparative view

Pattie Epps

An ethnographically grounded perspective offers significant insights into multilingual interaction and language contact in particular communities, as we have seen for the Vaupés region of the northwest Amazon. In so doing, it also provides us with powerful comparative tools for understanding the processes and motivations for language contact more generally, and – on the other side of the coin – the dynamics of language maintenance and diversification/spread over time. This study takes up this comparative exploration of patterns of multilingual interaction in the Amazon basin and beyond, and considers how an ethnographically grounded approach can provide new insights into many familiar questions: How do the outcomes of contact vary across small-scale communities in different parts of the world – in grammar and lexicon, and involving shift or maintenance? How do multilingual regions vary in the composition of their languages – are these closely related or do they belong to distinct families? are the family units small and relatively localized, or large and dispersed? are spreads contiguous or patchy? How do speakers make use of multiple languages – actively (as speakers) and/or passively (as hearers)? in everyday and/or in ritual contexts? in response to what social obligations or concerns? To what degree might patterns of multilingual practice be localized or widespread, and how might these distributions inform our understanding of global linguistic diversity? 

Discussant: Bethwyn Evans (ANU / CoEDL)

Thursday 16 February

Contact-induced change in the wild: Using linguistic variation to detect, analyze and distinguish it from internal evolution

Shana Poplack

The ability of language contact to trigger change has long been a contentious issue, particularly as regards the grammar. Examination of the recent literature suggests that this is because much of the relevant evidence stems from the (largely unacknowledged) inherent variability particularly salient in contact languages, and the presumed prevalence in them of nonstandard variants. Is variability coterminous with change? How can we tell when change has occurred, and how can we determine whether it is the product of contact or internal evolution?

Applying the variationist framework and the comparative method of historical linguistics to massive corpora of spontaneous bilingual speech, this presentation outlines a comprehensive empirical method for detecting change in contact languages, which we suggest is as applicable to small as big languages. We illustrate with a salient and highly stigmatized feature of North American French, preposition “stranding” (1a), widely believed to be a change induced by long-term contact with English, a language which strands prepositions nigh-categorically.

(1)   a.  J’avais pas personne à parler avec. (OH.013.1964)

            ‘I had no one to talk to.

        b.  Les anglaises avec qui je parlais, ils le croyaient pas. (OH.082.1695)

            ‘The Anglophones to whom I was talking, they didn’t believe it.

        c.  Pour faire quelque chose de different, du nouveau monde à parler [  ].(OH.002.040)

             [Why did you change schools?] ‘To do something different, new people to talk [  ].

(2)        Oui mais, il veut pas payer pour. (OH.013.260                 

    ‘Yeah, but he doesn’t want to pay for.

The English and French versions of (1a) are similar on the surface, but what of their underlying structure? Our working hypothesis is that this can be discerned by examining the distribution and conditioning of competing variants in discourse.  We show how to marshal this variable grammatical structure first to situate stranding with respect to the native French options for preposition placement with which it coexists (pied-piping [1b] and “absorption” [1c]), and then to ascertain its possible sources, by comparing  the constraints on variant selection operating in donor and recipient languages, contact and pre-contact stages of the recipient language, mainstream and “bilingual” varieties of the donor language, and in the speech of copious and sparse code-switchers. Detailed comparison with a superficially similar pre-existing native French construction, “orphan” prepositions (2), enables us to assess the extent to which the other logical catalyst for the emergence of bare variants – language-internal analogy – is operative.

Systematic quantitative analyses turn up several lines of evidence militating against the interpretations that a) bare prepositions are in fact stranded and b) their source is external.  Most compelling are the findings that the conditions giving rise to the contentious variants are the same as those operating to produce the native strategy of orphaning, while none of them are operative in English, the presumed source. Explicit comparison of copious vs. sparse code-switchers reveals no difference between them, refuting claims that the former are necessarily agents of convergence. Results highlight how surface similarities may mask deep underlying differences, a crucial problem in the identification of contact-induced change in multilingual contexts. 

Discussant:  Marie-France Duhamel (ANU/CoEDL)

  • Australian Government
  • The University of Queensland
  • Australian National University
  • The University of Melbourne
  • Western Sydney University