Explaining Language Structure Through System Interaction (review)

Lexicon, morphology, syntax, and phonology are the interacting systems that structure language by coding functional domains. This book, consisting of twelve chapters including the introduction (Ch. 1) and the conclusion (Ch. 12), tries to assess how the interaction of different linguistic levels works in several languages. The approach is typological. However, the language sample is not balanced as to genetic affiliation, but rather presents languages that are controlled to some extent by the authors. Functional domains, defined as classes ‘of mutually exclusive types of expressions constructed by a set of specific means of coding’ (26), are not the same across languages. Rather, specific functional domains are grammaticalized in each language to the exclusion of others. This assumption implies that language-specific functional domains must be discovered, rather than a priori assumed and then looked for in languages, as is commonly done. The difference between these two procedures permeates the whole book, but it is treated in detail especially in the last group of chapters (9–11).

Chs. 2–8 discuss the interaction of various means of coding. Ch. 2 is devoted to the lexicon and discusses patterns of lexicalization and morphological processes involving lexical categories, like reduplication of verbs or adjectives. Ch. 3, ‘Coding through linear order’, tackles, among other things, verb agreement for 3rd sg. in English, arguing that its function is to distinguish verbs from nouns (63–65). Ch. 4 is devoted to the function of case marking. It contains a comparison of Polish, a language with nominal inflection, with English. The authors argue that the tendency of English to use the whole VP in answers, while Polish only uses an NP, is due to the fact that Polish NPs are marked for case, while English ones are not (the discussion could have profited from a wider perspective: there are languages such as Italian without case marking in which answers are pretty much as they are in Polish). In Ch. 5, the authors describe the interaction of phonology with other coding means, devoting a long discussion to lenition in Welsh. Ch. 6 is devoted to agreement, under the assumption that agreement is not triggered by a controlling category, but rather must be viewed as an independent means of coding. For example, number coding on the verb is not triggered by the number of a given argument, but rather indicates whether that argument is singular or plural (154). The fact that agreement obtains where one expects it to is due to the fact that ‘although agreement does not exist, disagreement is disallowed’ (157). Ch. 7 is devoted to a related phenomenon, that is, nominal classification (gender), which is seen as a means for solving the problem of reference tracking, rather than as a system for categorizing entities (171). That the latter is not a function of nominal classification is shown by the fact that it does not exist in all languages (but this argument could also be used against the interpretation given by the authors). Ch. 8 discusses matrix clause coding (subject-to-object raising).

In Chs. 9–11, the perspective changes and the approach worked out in the first part of the book is applied to selected functional domains. In Ch. 9, the authors explore the functions of the dative case in a number of languages and argue that expressing ‘external’ possession is not among them. Ch. 10 is devoted to the coding of locative predication. Finally, Ch. 11 discusses the domain of reference.

The book presents an idiosyncratic approach, offering new solutions to old problems. On several issues more research is needed, especially based on thorough analysis of data from other languages.