When language breaks down: Analysing discourse in clinical contexts (review)

Theoretically minded linguists—and particularly, those interested in the cognitive foundations of linguistic competence—sometimes tend to consider (clinical) discourse analysis with disdain. They should not, and Asp and de Villiers’s book magnificently shows why. When language breaks down is a blueprint for a research program and a toolkit for achieving it. But this is not a restriction in scope: the authors’ apparent modesty reflects their awareness of enormous challenges that still have to be met in order to describe various types of language impairment. The idea defended throughout the book is that detailed analyses of the discourse of individuals with neurodevelopmental or neurodegenerative deficiencies can result in more precise diagnoses and yield a better understanding of the neural architecture that underpins language interpretation and use. The case studies that illustrate each step of the book are drawn from A&V’s respective areas of expertise: neurodegenerative disorders (such as Alzheimer’s disease or vascular cognitive impairment) and autism spectrum disorders (ASD). The main clinical characteristics of these conditions are introduced in the second chapter, which also contains a (rather terse) summary of the communicative profile of each group.

Now, speaking of the analysis of a string of discourse is premature as long as one does not have a precise description of what is going on. Of course, the degree of precision that analysts need to meet and the level of coding they choose depend on the object of their inquiry. For instance, if what one wishes to assess is the degree of morphosyntactic complexity, detailed coding of conversational structure (such as topic continuation and shifts) might be unnecessary. This, however, presupposes that researchers already know what they are looking for; that is, that they have at their disposal diagnostic criteria that include precisely circumscribed linguistic problems. But, of course, many linguistic and communicative properties of neurodegenerative and neurodevelopmental deficiencies remain unknown, and more often than not nosological uncertainty looms large.

A&V provide the reader with tools to achieve the most exhaustive description possible. They explicitly adopt an eclectic approach to discourse analysis ‘drawing from such theoretical sources as conversation analysis, generative linguistics, functional linguistics and pragmatics’ (19). In Chs. 3 and 4, the authors explain how to break down and code a sequence of discourse at different analytical layers: conversational units (e.g. turns, false starts, repairs), intonation contours, speech act (or illocutionary) functions (e.g. statements, questions, commands) and the way these are syntactically realized (e.g. tagged or not, polar vs. wh-questions), markers of attitude and modality, argument roles, and the internal articulation of discourse (e.g. cohesion, focus).

The intended audience of this book is not only researchers in linguistics but also—and perhaps chiefly—professionals, ‘[n]eurologists, psychiatrists and neuro(psychologists) [who] are involved in diagnosis, assessment and treatment of people with neurological and/or neuropsychiatric dysfunction’ (24). Virtually no previous knowledge of the terms of the art is presupposed, and while the linguist might find definitions of ‘clause’ or ‘verb argument’ tedious, these are certainly useful for readers outside our discipline.

I am less enthusiastic about the way nonlinguists are supposed to be introduced to the theoretical background from which A&V draw their analytical tools. The fourth section of Ch. 2 attempts to provide concise overviews of conversational analysis, generative linguistics, functional linguistics, and pragmatics—all in less than five pages. No doubt it is important to explain where the analytical tools and concepts that are used come from, and to clarify their theoretical presuppositions. But such an introduction deserves much more space than A&V allow for, and has to be more detailed and less dense in order to be adequately understood by nonlinguists. As for the insiders, they will probably find this part of the book oversimplified and uninformative.

In Chs. 6 to 9 the authors show how to use discourse analysis to examine various linguistic, affective, and cognitive properties, and how to integrate such...