- Nominal contact in Michif by Carrie Gillon, Nicole Rosen, Verna Demontigny
This book is the fifth volume in the series ‘Oxford studies of endangered languages’, whose mission, according to the general preface by series editor Stephen R. Anderson, is to ‘support the publication of theoretically informed work on endangered languages, while striking a balance’ between the goals of basic description and contributions to theoretical concerns (ix). Carrie Gillon and Nicole Rosen (henceforth G&R), with Verna Demontigny, provide by far the most comprehensive description to date of noun phrases in Michif, which is perhaps the most-discussed bilingual mixed language in the language contact literature. Demontigny, identified as G&R’s ‘primary Michif language consultant’ (xi), is well known and highly respected as one of the most proficient native speakers of Michif and is a long-time teacher of the language. G&R credit her with providing the native-speaker intuitions that they prioritized in their analysis (4); but since the analysis itself is apparently G&R’s, they are cited as sole authors in this review.
The main theoretical implication of G&R’s results, they argue, is that Michif is not a mixed language at all, but rather an Algonquian language with considerable French borrowing, primarily in the noun phrase. They then extend their conclusion to all other bilingual mixed languages,1 concluding that there is ‘little synchronic motivation for a new category of “mixed languages”’ (178), because all of the so-called mixed languages are ‘likely to behave mainly like one unified language, with some extra features from the other language, once the spotlight is on their structure’ (169). [End Page 806]
In this review I begin by outlining G&R’s structural analysis of Michif noun phrases—DPs (determiner phrases), in their terminology. The reason for the focus on noun phrases is that Michif has generally been characterized as a language with Cree (Algonquian) verb phrases and sentential syntax but with French noun phrases. That is, the traditional claim (in e.g. Thomason & Kaufman 1988:228–33, Bakker 1997, and Bakker & Papen 1997) is that the Michif noun phrase comprises French lexicon, phonology, and grammar, while the verb phrase and sentential syntax are Cree. It is well known that Cree parts of the structure leak into the French noun phrases, but (with the exception of a very few borrowed French verb stems) not vice versa (e.g. Rhodes 1986:289). As an illustration, a typical Michif sentence is G&R’s ex. 4b (p. 8), reproduced below, where French-origin words are in plain type, the only Cree-origin word is in italics, and vti stands for a transitive inanimate verb stem.
After outlining the contents of the book’s main chapters, I comment on theoretical and methodological issues that arise from the authors’ discussion.
G&R’s presentation of the data and analyses is clear and for the most part easy to follow, even for scholars unfamiliar with their generative theoretical framework, and most of the specific analyses are convincing. Brief arguments against the mixed-language status of Michif are scattered throughout the text, but discussion of the broader theoretical issue—namely, whether mixed languages exist—covers fewer than ten pages in all. As explained below, I find their theoretical claims about mixed languages problematic, both with respect to Michif and with respect to mixed languages in general, including pidgins and creoles as well as bilingual mixed languages. Moreover, the book lacks adequate information about how the Michif data was collected, a crucial component of any linguistic analysis but especially vital for an analysis of an endangered language.
After a guide to pronunciation and orthography and an introductory chapter sketching the history and status of Michif, ‘general issues of language contact’, and an overview of Michif structure, the authors address their main topic in the next five chapters: 2, ‘Mass/count’; 3, ‘Plurality’; 4, ‘Gender’; 5, ‘Articles’; and 6, ‘Demonstratives’ (with some further...