222 LANGUAGE, VOLUME 76, NUMBER 1 (2000) that it has finally seen the light ofday. [Marc Pierce, University ofMichigan.] Word order patterns in Early Modern English. By Bjorg Barren. Oslo: Novus Press, 1998. Pp. xviii, 461. Bsekken's book is a corpus-based study of word order in Early Modern English (EME) based on some 3000 pages of text dating between 1480 and 1730. In the history of English there was a major restructuring from Old English to Modern English. While the former had some kind of verb-second constraint, the cunent language has a dominant verb medial SVO order. B's largely descriptive study is a contribution towards answering the questions of when this shift took place and what the contributing factors were. The domain of investigation is thus non-subject-initial declarative main clauses, inverted XVS and noninverted XSV. B's main finding is that during the EME period the language stabilized, leading to dominant XSV order towards the end. B identifies 1680- 1730 as the period during which the most radical decrease in the use of inversion structures took place. The body of the book investigates particular factors which might account for the gradual shift. Ch 5 looks at the effect of text type and individual author on the rate of inversion. In general, the results here are somewhat enatic. For example, B observes that while some authors use inversion quite consistently, others show considerable variation even within texts of the same category. Chs. 6-8 discuss patterns with three particular initial elements: predicates, direct objects , and adverbials. B suggests that numerous factors including textual cohesion, emphasis, the category of the fronted element, the form of the subject , the relative weight of the subject and verb, and a tendency to reserve the postverbal position for focused and heavy elements are relevant for whether or not inversion is present. Ch. 9 documents the behavior of certain adverbs known to have triggered inversion in Old and Middle English (for example here, then, neither, never). B claims that individual adverbs developed differently. For negative adverbs, which require inversion in Modern English, this means that so-called negative preposing was not always a unified phenomenon. For nonnegative adverbs, inversion rates in most cases dropped dramatically over the EME period, converging on the modern situation. Ch. 10 investigates the discourse status, given vs. new information, of the initial elements and subjects. B finds that there is a strong tendency for the subject to be given information but that inversion increases the possibility that the subject will be new information . Initial elements, by contrast, may be either given or new information with roughly equal frequency. Finally, Ch. 1 1 looks at the types of verbs in inversion contexts, finding significantly higher rates of inversion with intransitive verbs, particularly verbs of appearance or existence, over transitive and linking verbs. The work contains numerous illustrative examples , statistical studies, and various cross-classifications of the data. Furthermore, each chapter ends with a useful summary of the observed patterns. Ch. 12 is a fine comprehensive overview of the work's findings and could profitably be read first. Not surprisingly , B concludes that word order in EME is ultimately determined by a complex interplay of syntactic , discourse, and pragmatic factors. [Eric Potsdam , Yale University.] Adverb placement: A case study in antisymmetric syntax. By Artemis Alexiadou . Amsterdam: John Benjamins, 1997. Pp. x, 256. Cloth $65.00. Adverbs have for a long time challenged linguists in many domains, and there has been a recent revival of interest in their analysis and implications for linguistic theory. In this book, Alexiadou argues for an original, largely syntactic approach to adverb placement . Ch. 1 introduces the vanous puzzles posed by adverbs . A sees the two pnmary facts for syntax to explain as (1) the limited number of adverb classes and (2) their rigid order in the clause. Ch. 2 summarizes the work's theoretical foundation: Noam Chomsky 's minimalist program and Richard Kayne's antisymmetric clause structure. In this framework, adjunction is disallowed, specifiers are licensed via feature checking with a head, and X' projections have strict specifier-head-complement ordenng. These assumptions entail the work's main proposal: individual adverb classes are...