Work Papers of the Summer Institute of Linguistics, University of North Dakota Session




From the introduction: "[...] Clitics which double objects in Yagua are called Set II clitics. All other clitics are referred to as Set I clitics.

"In this paper, I argue that the relationship between Set I clitics and Set II clitics, and the effect of these clitics on word order, present some intriguing problems for theories of syntax, especially as seen in the phenomenon of clitic doubling, the co-occurrence of tautophrasal, coreferent clitic NP pairs associated with a single semantic or thematic (θ)-role. I argue that a careful study of Yagua clitic doubling makes at least the following contributions to linguistic theory:

"1. The crosslinguistic parametrization of nominal clitics proposed in Everett 1986, 1987 receives significant support. Simply put, clitics may vary crosslinguistically in whether they require a θ-role or morphosyntsctic case (henceforth Case).

"2. By establishing that pragmatically unmarked word order (VSO) and syntactically underlying word order (SVO) in Yagua are coexistent, mutually compatible components of Yagua grammar, the concept of basic word order, assumed implicitly by the majority of typological studies, is shown to be of little use, unless defined more precisely in terms of the notions just mentioned. [...]

"3. Yagua Set I clitics are sensitive to otherwise word internal morphophonological processes, such as Vowel Harmony, which do not affect Set II clitics. However, as we will see, there are strong arguments for attaching all clitics in the syntax, i.e., postlexically. This represents a serious problem for models which account for word internal phonology or inflection in the lexicon, e.g. Lexical Phonology (LP. Mohanon 1986, Kiparsky 1985, Pulleyblank 1986) and Lexical-Functional Grammar (LFG. Bresnan 1982). It supports instead the attachment of at least some morphemes in the syntax.

"The discussion is organized as follows. First, we survey the basic facts of Yagua clitic doubling and its interaction with word order. Section two presents additional facts about Yagua clitics which must be accounted for. In section three, it is shown that Yagua clitics are nonarguments. In section four, I propose an analysis of the facts in terms of Government and Binding Theory (GB). Section five adduces independent evidence for this analysis from some otherwise puzzling facts of Yagua reflexivization. The implication of this analysis for syntactic typology, and theories such as LP and LFG are discussed in the final sections. Results and major features of this study are summarized in the conclusion."

Included in

Linguistics Commons