French through Corpora: Ecological and Data-Driven Perspectives in French Language Studies ed. by Henry Tyne et al. (review)

This edited volume presenting the state of the field of corpus studies in French will be of use to francophones and non-francophones alike. The works are united, of course, by their use of corpora. However, the division of the book into parts by area of linguistic enquiry (‘Diachrony’, ‘Syntax’, ‘Sociolinguistics’, and ‘Learning and Teaching of French’) highlights the different sets of concerns around corpus methods in each field. It also shows the diversity of the uses of such methods, all the while evoking the challenges imposed by a relative lack of developed corpora in French. In their Introduction, the editors explain the relationship between ‘ecological’ and ‘data-driven’, the two themes uniting the volume: ‘language is not just a neutral collection of data, hence the inclusion of the concept “ecology” in the title’ (p. x). The articles in Parts 1 and 2 exemplify the ‘data-driven’ approach, understood by the editors as one that ‘build[s] conclusions on the sole basis of the findings’, not one that cherry-picks tokens to suit the researcher’s hypothesis (p. x). In Part 1 (‘Diachrony’), the authors address methodological and theoretical concerns, including the unavoidable lacunae in any historical language corpus. The articles in Part 2 (‘Syntax’) show a high level of engagement with advanced corpus analytical tools. In contrast to the first two parts, the last two are more concerned with an ‘ecological’ approach. In her Foreword, Carol Sanders states that the ideal ecological corpora provide ‘“natural, undoctored” language respectful of the context in which the data is gathered’ (p. viii). In Part 3 (‘Sociolinguistics’) such corpora are relatively scarce: there is frequently a mismatch between big data and traditional sociolinguistic methods. In Part 4 (‘Learning and Teaching French’), a preoccupation with ‘authenticity’ is present as possible uses of corpora for both students and teachers are outlined. For the historical linguist and the syntactician, suitable data seems easier to come by and to analyse, even if the organization of this data is incomplete at present. However, the holistic, authentic, and ecological corpora prized by sociolinguists and linguists working in language acquisition still seem rare. For this reason, perhaps, the articles in Parts 3 and 4 present less convincing arguments for the use of corpora, especially large ones. As the scope of corpus linguistics is being delimited, the book highlights the importance of evaluating a variety of methods, theories, and findings. It also calls attention to gaps in current data and research. The editors ask whether corpus linguistics should be considered ‘“merely” a methodology’ or as ‘something rather more, given its tremendous impact on all that we know about language use’ (p. xi). Since the authors featured deploy corpus-based tools effectively in a great variety of fields without being constrained by a uniform methodology, the reader is led to believe that it is ‘something rather more’.