In lieu of an abstract, here is a brief excerpt of the content:

Chapter 1 Introduction 1.1. Preliminaries The study of complementation and complementizers has been an area of great interest for syntactic theory in recent years. In this book I present a description and analysis of complementation in Bulgarian with particular emphasis on questions and relative clauses (the so-called WH constructions) and their interaction with the set of clauseintroducing words known as complementizers. WH constructions and clauses containing complementizers have been studied intensively in English and a number of other, mostly Germanic or Romance languages, and a variety of hypotheses concerning the structure and derivation of these constructions and particularly the role of the complementizer position (“COMP”) have been put forward in the literature. An examination of these constructions in Bulgarian, a language which had not previously been studied in any depth in a generative framework,1 raises problems for some of these hypotheses, provides additional evidence in favor of others, and brings to light new types of data which a theory of WH constructions must account for. The goals of this book are both descriptive—to provide an analysis of a part of the grammar of Bulgarian with detailed data that will be of interest to Slavists and Balkanists as well as general linguists—and theoretical—to use these data to shed light on broader crosslinguistic questions of the nature of COMP, complementizers, and WH constructions. These two goals are not at all incompatible; in fact they complement and strengthen each other. A sound theoretical framework is essential to the analysis of linguistic data, and detailed, carefully analyzed data in turn are necessary for the framing and testing of linguistic theories. It is particularly important for evaluation of competing hypotheses to look at the facts of a variety of languages. A theory which is able to account, in an insightful way, for the facts of a language other than the ones it was originally based on obviously derives some support from this. Although an account which does not work for the “new” language is not thereby disproved or rendered untenable for other languages, it does lose any claim to universality, and may lose much of its explanatory power if phenomena which it is meant to explain are also found in the “new” language. 1 There have been a few generative studies of some aspects of Bulgarian grammar, especially clitics (Ewen 1979, Hauge 1976, Moneva-Dolapčieva 1976). Penčev (1984) deals with WH constructions in a classical transformational framework, but not in great depth or detail. 2 ASPECTS OF BULGARIAN SYNTAX If data are crucial in evaluating theoretical issues, then it is equally true that theoretical considerations play a very useful role in organizing and clarifying data. In the present case, the concept of COMP makes possible a very different (and I believe better) classification of clause types than that used in traditional grammars of Bulgarian , and at several points theoretical questions prompt detailed investigation of constructions which are only mentioned in passing, if at all, by the standard handbooks. Some of the most critical data in the analysis of Bulgarian WH constructions are of a type not considered interesting by the writers of traditional grammars, either because they are colloquial, and thus not included in prescriptively oriented works, or because they are ungrammatical. The idea that what cannot be said (and why) is as important as what can be said is an important feature of generative theory. The ultimate goal of modern generative linguistics is to understand and represent the knowledge that a native speaker has of his language, including the ability to distinguish possible from impossible structures. A generative grammar is a set of rules which are capable of producing any grammatical sentence in the language, but no ungrammatical ones; in order to formulate the rules and the conditions on their application precisely it is often necessary to look in detail at sentences which speakers reject as impossible. Following standard practice, I mark all ungrammatical sentences with an asterisk (*). “Bulgarian” for the purposes of this book means “Modern Standard Bulgarian”, that is, the language used by urban, educated speakers in present-day Bulgaria. This includes a range of styles or registers, from very colloquial to very formal, but excludes regional dialects and older forms of the language, although I will occasionally cite such forms for comparative purposes. The bulk of the data on which this book is based was gathered during eight months of research in Bulgaria during 1980–822 and several shorter visits between 1972 and...

Share