Reviewed by: Theories of case Helen de Hoop Theories of case. By Miriam Butt. (Cambridge textbooks in linguistics.) Cambridge: Cambridge University Press, 2006. Pp. xiii, 258. ISBN 9780521797313. $37.99. The first sentence of the introductory chapter states that the book is meant to serve as an introduction to various notions of case within modern theoretical linguistics. In fact, the book is much more than that. It not only provides an introduction to theories of case, but moreover puts together the various notions and insights coming from these theories in an illuminating way. The first chapter ('Introduction', 1–11) introduces the topic of the book: case. The important question addressed in this chapter is what the exact range of phenomena is that can be properly labeled 'case'. Notoriously, case is used to mark the semantic relationships between verbs and their arguments. But while some languages encode this relationship via morphological case, languages such as English encode it structurally, that is, in terms of designated positions. Thus, word order or structural position might be conceived of as an alternative strategy for marking the relationship between a verb and its arguments, and in that sense, it can be viewed as 'case' as well, which is of course a very common perspective in generative syntax. It is also clear, however, that word order and overt case marking are not truly alternative strategies for marking the relationship between a predicate and its argument since they are not complementary. Some languages have no case but yet a lot of word-order freedom, while other languages have both case and a strict word order. Similar problems as to what counts as 'case' are discussed with respect to head-marking versus dependent-marking languages, adverbial and nominal case, case stacking, and case and finiteness. Butt concludes that the notion of case employed in syntactic theories is an abstract notion used to characterize the interaction between verbal semantics, grammatical relations, and word order. The rest of the book is devoted to three guiding themes in order to allow for comparison across the theories: thematic roles, grammatical relations, and not surprisingly, overt case. Recurring topics throughout the other chapters are transitivity, unaccusativity, passivization, quirky case, and ergativity. Ch. 2 ('Foundational perspectives', 12–22) explains how modern theories of case developed from ancient traditions. B argues that due to the use of the Greek and Roman naming of cases, the fundamental assumption that case must be tied directly to semantics is still implicit in modern syntactic theorizing. This stands in stark contrast to the Indian tradition, where the case markers were simply numbered rather than associated directly with certain semantics. In the Arabic tradition, case is embedded within a larger theory of government in which the governor (a verb, preposition, or nominal) strictly determines the form of the governed element. This view led to a differentiation between two levels of representation at which government can take place, an underlying and a surface level of representation, reminiscent of certain versions of generative syntax. Ch. 3 ('Grammatical relations', 23–45) and Ch. 4 ('Structural case', 46–90) focus on the development of modern syntax (1950s onward). In early theories, such as transformational grammar and in particular relational grammar, grammatical relations were understood as abstract syntactic concepts that mediate between verbal semantics and case morphology. In Ch. 3 thematic roles in interaction with grammatical relations are introduced, while Ch. 4 explores the basic approach to case within the generative theories of government and binding and the minimalist program. In generative syntax, grammatical relations are associated with the assignment of structural (abstract) case. B argues that in contemporary generative syntax there is a trend toward incorporating ever more precise semantics into the structural representations. Ch. 5 ('Linking theories', 91–149) surveys a number of important linking theories, in particular Joan Bresnan's lexical functional grammar (LFG), Paul Kiparsky's linking theory, and Dieter Wunderlich's lexical decomposition grammar, theories of how predicate-argument structures are linked to syntactic structures. In these three approaches to argument linking, a number of features [End Page 620] play an important role. In LFG, the two relevant features are [+/restricted], which refers to the semantic restrictedness of a thematic...
Read full abstract