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

Chapter 2 Word Order and Basic Sentence Structure Before undertaking the analysis of WH Movement and other COMP-related phenomena , which is the main object of this work, it will be useful to give a brief description of the basic facts of Bulgarian sentence structure and word order, so as to have a framework into which to fit the later discussion. This is particularly true since there is very little available in the way of useful description of Bulgarian syntax, especially word order. 2.1. Bulgarian Word Order Is Relatively Free The order of constituents in a Bulgarian sentence is quite free, and traditional grammarians have not been concerned with analyzing the limits on this freedom. The state of studies of word order in Bulgarian until very recently is pretty well summed up by the following statement from Popov, Stojanov, and Janakiev (1980: 321). Word order expresses the logical process of thoughts, which is the same in all people, therefore it occurs with identical features in many languages. […] The order of words is determined by logical, psychological, and grammatical factors . It depends on the way in which the speaker’s thoughts flow, the ideological and psychological content of the sentence, and the speaker’s relation to this content, the goal of the utterance, and the traditional habits and schemas of speech and grammar. (My translation, CR) The section on word order in traditional grammars typically consists of a statement similar to the above, followed perhaps by examples of several of the possible word orders of Bulgarian, classified as more or less normal or neutral. While the statement that word order depends on “logical, psychological, and grammatical factors ” is undoubtedly true, it is not particularly useful as a basis for syntactic analysis, especially since none of these factors is ever spelled out with any precision. Szober (1979) put forward the general principle that old information precedes new information in the sentence in Bulgarian. This generalization is certainly a step in the right direction, and seems to have some validity at least as a tendency in many languages (for examples, see Li and Thompson 1976 and references cited there). However, such a vague idea is too simplistic to handle constructions of any complexity. 16 ASPECTS OF BULGARIAN SYNTAX There has been some effort to correct this state of affairs, and some much more useful and enlightening work on word order by Bulgarian linguists has appeared in the last decade or so: Georgieva 1974 lists a great many possible orders (in simple, one-clause sentences only, unfortunately) and makes an attempt at sorting out the discourse functions of the different word orders; Penčev 1980 produces some very interesting results by working with the interrelationship of intonation with word order and other syntactic factors; and Penčev 1984 makes some interesting suggestions on word order as a reflection of theme-rheme structure. Popov 1983: 266–87 also contains a fairly extensive discussion of several ways of looking at word order, including “logical-grammatical” and “actual” (= theme-rheme structure). But these few studies barely scratch the surface; a great deal more work remains to be done. Bulgarian word order is relatively free, especially in colloquial speech, but of course not completely so. Word order is considerably less free than in so-called “W*” (non-configurational) languages (Hale 1981), since major constituents such as NP have a consistent internal structure and cannot be broken up. If anything, Bulgarian could be described as an XP* rather than W* language, that is, with phrasal constituents in random order rather than totally non-configurational; however, even phrase order is subject to some constraints, both syntactic and discourse-related.1 In what follows I suggest that, while the most normal surface constituent order is Subject-Verb-Object (SVO), underlying order is probably V-initial, and any material which precedes the verb in surface structure has been fronted either by WH Movement or by a rule which adjoins a phrase to the left of S′ or S. This adjunction rule creates two positions at or near the beginning of the clause: TOPIC position to the left of S′ (preceding the S′-initial COMP) and FOCUS position to the left of S (and therefore immediately following COMP). As the names of these two positions indicate, material moved to them tends to be interpreted as having particular discourse functions . In addition to TOPIC and FOCUS phrases, a sentence may have associated with it a “Left Dislocated” or “Right Dislocated” Noun Phrase, that...

Share