the transformation of language situations: the habitat model[1]
Gerhard Leitner
1. Abstract
Modeling English has become an important area of research with implications on the structure of the discipline and on educational policies. It was the result of the challenge by Braj Kachru in the early 1970s that English was no longer a language with one underlying system. It gave rise to a controversy began between Kachru’s “Englishes” camp and those that continued to argued for English having a core, a standard form universally applicable, and a dynamic set of peripheries. That controversy and the revisions of either model have implications beyond descriptive linguistics in applied linguistics and politics (Leitner 2009). Alternatives have emerged in the past decade. The broadest one, the Habitat Model, is not too far removed from Alexandr Svejcer’s thinking, and promises deeper insights into how habitats and whole societies change.
2. Taxonomic models
The nature of English began to be addressed academically in the 1970s when Kachru argued for the legitimacy of “Englishes” like Indian English. As American English had been accepted for decades, there was a precedent. The variability of English had, of course, been noticed before, though it did not turn out to be a major issue in debates about English language teaching in the 1930s (see Leitner 1982; 1992; 1984). The creativity of the non-standard and slang, for instance, were collected by Eric Partridge (1937) and many amateurs and experts in Australia (Leitner 2004a), and elsewhere in the 19th century. Two Americans, James Greenough and George Kittredge, dismissed the fear that either would create barriers to comprehension at the turn to the 20th century. Pronunciation was a more difficult matter and comprehension problems between English and Australian accents had been studied and dismissed in radio surveys. The BBC’s pronunciation advisor, Lloyd James, pleaded in the 1930s in vain with American film and radio serials’ producers to avoid British and American English heading into different directions. He also foresaw a danger to the unity of English due to the expansion of English in the colonies. Alexander Mitchell, a leading Australian phonetician and disciple of Daniel Jones, had managed to persuade the Australian Broadcasting Corporation in the 1940s to adopt a model of pronunciation that was close to Received Pronunciation but was Australian where a case could be made for it; he thought of place names and a small number of word pronunciations. In the 1950s the BBC Yearbooks commented on the loss of control of Britain as regards pronunciation. None of these observations was able to cast doubt on the assumption that English was a single language. Kachru’s work did. It was best, he suggested, to set up a taxonomy based on the presence of (accepted) features in geo-political units and the societal function of English as a native, a second and a foreign language. The likelihood of developing local norms diminished from the (native) Inner Circle to the (second language) Outer Circle and to the (foreign language) Expanding Circle. A flood of research confirmed his view, adding a forceful body of data.
The Three Circle model was persuasive and had a welcome political orientation in the 1970s. But it was static and taxonomic. The political credo could not be maintained when the impact of colonialism faded and English spread as a global phenomenon (Leitner 2009; Bruthiaux 2003). What is more, the nexus between linguistic difference and sociolinguistic acceptance had been given up so that a plethora of “Englishes” emerged (Leitner 1992). If all that counted was linguistic difference, it was a short step to accepting the Expanding Circle English in China as a legitimate, norm-setting variety, thus demolishing the model itself. Its sterility became particularly visible when it failed to pick up Rodney Moag’s (1982) proposal to inject dynamism by arguing for a life-cycle. While Peter Trudgill’s (2004) claim of the inevitability of variety formation in restricted scenarios was ill-guided as he rejected social forces as forces and believed in what he called mechanistic processes based on the demographic make-up of a community, he at least proposed a series of linguistic changes and developmental stages.
3. Dynamism
Dynamism was on the agenda during the past decade. What is surprising is the longevity of the Circles Model as alternatives emerged that promised greater insights. As just mentioned, Moag (1982) had developed a taxonomy of features for all varieties. Change in linguistic texture, social functions and status, he argued, could be mapped out in his grid and be used to assign some variety to one of the three circles. Membership in a circle could, and did, change as some East African countries moved into the Expanding Circle. His life cycle borrowed a lot from creole studies, which had for long shown that developmental stages, stops and restarts linked linguistic development with extra-linguistic factors (Mühlhäusler 1996). Clines ranging from basilect to acrolect via the middle step, the mesolect, were extensions and were introduced into the study of Singaporean and Malaysian English by Platt et al. (1982). Clines are referred to, e.g., by Low/Azirah (2012) in South-East Asian English. As it is normally implied that higher lect speakers have expressions of lower lects at their disposal, varieties need not be defined as having reasonably firm systems, but as the result of choices. An extreme version of that view was Richard Hudson’s (2003) Word Grammar, which defined language as clusters of linguistic expressions that are available as speakers’ resources. There were precursors such as A.J. Aitken’s (1984; see also Johnston 2007) studies of Scots, Standard Scottish English and Standard (English) English. He believed that lexical items could be arranged in tabular form that reflected a greater or lesser degree of proximity to Scots, the one pole, or to Standard English (the other pole). Ian Malcolm’s studies of Aboriginal English and Aboriginal creoles, too, were based on the exploitation of repertoires to create meaning (see Leitner 2004b; Leitner/Malcolm 2007). Similar assumptions can be found in studies of language attrition and shift (Clyne 2003; Dixon 1997). Dynamism resides in the way speakers (can) exploit common repertoires to project loyalties.
It took long for dynamic models to appear in studies of varieties of English. Moag (1982) and Platt et al. (1984) were precursors. It was Schneider (2003) who developed a fully-fledged alternative to Kachru’s Three Circles. His model is based on contact, the impact it has on linguistic texture and social stratification. He maintained that developments could be clustered into stages, and were associated with two (or more) strands of communities, who ultimately followed a consensual path. Here is a partial summary:
It is the core thesis of this book that, despite all obvious dissimilarities, a fundamentally uniform developmental process, shaped by consistent sociolinguistic and language-contact conditions, has operated in the individual instances of relocating and re-rooting the English language in another territory, and therefore it is possible to present the individual histories of PCEs [= Post-Colonial Englishes, GL] as instantiations of the same underlying process. More specifically, it is posited that evolving new varieties of English go through a cyclic series of characteristic phases, determined by extralinguistic conditions. (2007: 5)
He posited five stages: (i) foundation, (ii) exonormative stabilization, (iii) nativization, (iv) endonormative stabilization, and (v) differentiation, which define a path from the transplantation of a range of dialects into novel environments by the first strand, the founders, to internal differentiation. The early input varieties of English would interact and develop (independently) by new communities, the two strands. They would mix as a result of interaction of dialects and contact with indigenous and/or migrant languages. The two (or more) strands of speakers are assumed to arrive consensually at a shared variety.
The model has well-known weaknesses. When Schneider talks of it being cyclic, he does not mean what one might assume, namely the possibility of regressions and new starts. It’s a straight forward path where some variety may stop and take off again. In creole studies, regressions are known, as is relexification when the ecology of a contact language changes. Many pidgins like Tok Pisin underwent relexification (Mühlhäusler 1996). It is known that varieties may lose their independence and become integrated into a cline of variation. Sometimes they take off again. Scots, which developed at the same time as Old English, lost its independence early in the seventeenth century. It was integrated into a dialect spectrum that ranged from Scots to near Standard English. After the devolution in the 1990s, it is re-gaining ground and some level of independence – although in a different shape. While Schneider is aware of such developments, he fails to propose an appropriate mechanism in his model. Furthermore, it is not clear why the two strands are to follow a consensual, and not a conflict-laden path and why the linguistic developments mentioned define these stages and not others. As linguistic levels develop at different speeds (Hudson 1996), it is unclear why and how such stages can be defined in the first place.
The model follows its predecessors in being entirely Anglo-centric, ignoring the other side of contact. The impact of English on the lexico-grammar of languages it is in contact with, of the effects of the space it carves out for itself on that of other languages, of language shift to English or alternatives, and language policy issues escape its vision. A comprehensive alternative, the Habitat Model, is clearly necessary.
Einar Haugen (1972) initiated a broader, ecological concept in his studies of linguistic minorities, language maintenance and shift and the retention of cultural identities. The conceptof language ecology implied dynamism and was borrowed from biology. It referred to the interaction of a language with its environment, which, paraphrasing Mühlhäusler (1996: 3), is more than the sociology of language or the pragmatics of speech situations. A brief description of the history of research comes from Stevenson (2000: 113):
Ethnolinguistic minorities understood as ‘communities of memory’ therefore need to be situated in an environment that is less tidy than the purely politically defined state. For this reason, Einar Haugen thirty years ago proposed an ecological approach to the study of multilingual societies, which would not only take into account all essential elements of the structure and development of a linguistic community but would acknowledge their interdependency (Haugen 1972). More recently, the creolist Peter Mühlhäusler has taken up Haugen’s ecological approach in relation to language change, language shift and linguistic imperialism in the Pacific region (Mühlhäusler 1996). Here again the emphasis is on the necessity of a holistic conception of the object of study, including all relevant features of social and physical environment, and conventional social constructs such as community, society and network are replace as sites of communicative action by the concept of ‘habitats’.[2]
Mühlhäusler (1996: 3f) lists ten questions to be discussed for each language, e.g., its demographic base, its domains of use, internal varieties, traditions in writing, standardization, but also the concurrent languages used. I will add to this list the degrees of openness to outside impact and its impact to the outside. While he emphasizes the metaphorical use of the concept, he takes it further and develops it in biological terms throughout the study of the Pacific and it is in this context that he introduces the term of habitat (1996: 19) to refer to the relationship of language to the environment. His extension of the term ecology to cover the intimate relationship of language with environment explains, e.g., that societally relevant knowledge in, say, medicine, sustainability, is lost when indigenous concepts in fauna fall into disuse.
A link between linguistic and society was known in European functional linguistics at the beginning of the 20th century. it was introduced into sociolinguistics by scholars like Alexandr Svejcer (1986; 1976). The (formerly) East German scholar Klaus Hansen (2008: 20) has reviewed that line of the research history and introduces a broad understanding of Svejcer’s language situation to the study of varieties of English. Clearly, there is a considerable overlap between Haugen’s and Mühlhäusler’s language ecology (that later also became ecolinguistics) and Svejcer’s and Hansen’s language situation, though it seems to me that Mühlhäusler goes much further in embedding languages in their social and physical environment. The concept of language habitat (or its plurals languages habitats and language habitats) is quite similar but borrows the term from a social understanding of habitat.
4. A Habitat Model – a case study of Australia
Though there is an impressive research history behind the Habitat Model, it was developed independently in Leitner (2004a/b) and applied to the language situation in Australia. One of the central objectives of the study of Australia’s language habitats was to develop an integrative approach or, as ecologists put it, ‘holistic approach’. I borrowed freely from Michael Clyne’s vast body of research and started by asking whether one can find a pervasive Australianness across all of Australia’s languages that reflected complex identities. The study showed the correlative processes of a decreasing impact of the historic descendent, British English, and the increasing impact of Australian English across the entire habitat. Contact is seen as multiplex, involving many partners in multi-dialectal and multilingual contexts, leading to change in the linguistic texture of all languages involved, in their respective statuses, and their functions as tools of communication or markers of identities. Contact recasts entire language habitats.
As was the case with English and other transplanted languages that have become dominant – such as French, Spanish, Portuguese or Mandarin etc. – they ultimately reduce the level of multilingualism and typically introduce hierarchies instead of parity. The goals of that model can described as
an attempt to show how all communities fit into a common framework. One must avoid sectionalizing and ignoring common features, while giving due weight to differences. But Australia's Many Voices [i.e. the title of the book, GL] is more than an attempt to cast a language net onto Australia, it provides a model for English elsewhere, indeed for other habitats, as in NZ, South Africa, the USA or Canada (2004a: 6).
The following diagrams illustrate the main processes in the transformation of a habitat due to the transplantation and implantation of English: