Shows that a phonological description that recognizes the functional variety of phonological rules is more illuminating than one in which data are handled merely as the output of a set of completely undifferentiated processes. Emphasizes the value of distinguishing motivated and unmotivated processes in phonology. (Author/RM)

In language use, redundant linguistic items are reduced in size, replaced with a proform or left out. This paper examines the nature of the processes involved and the conditions under which they operate. (CHK)

This paper concerns the theory of systemic grammar developed by Halliday, Hudson and others. It suggests modifications of Hudson's generative version, and the model presented resembles transformational grammar. (CHK)

Discusses noun phrases which involve typically predicative nouns, focusing on those anaphoric noun phrases that apparently have to be derived from predicates dominating an entire sentence. The ensuing modification of Bach's hypothesis provides evidence for a particular theory of relativization referred to as "the promotion analysis." (DS)

Zwicky's analysis of syntactic notions as possible candidates for "head," based on constituent-structure theory, is discussed in detail. This analysis is contrasted with the results gained form one provided in a dependency-based theory in which "head" is the name of a grammatical relation category. (Author/LMO)

All modern Romance languages except Rumanian have a class of causative + infinitive construction in which the infinitive subject surfaces as an agentive. This article investigates the question of how agentivization of the infinitive subject is to be handled in these languages. (DS)

The Spanish feminine article /el/, ordinarily the singular masculine definite article, has been used as evidence of the need for obligatory disagreement rules. Others explain the anamoly by means of referral rules. A third solution is suggested: an allomorphy rule which can be interpreted syntactically or phonologically. (LMO)

Discusses the types of arguments available for settling questions of phonological constituency. On the basis of these arguments the most frequently assumed syllable structure is selected in which the relationship between "Peak" and "Coda" is closer than that between "Peak" and "Onset." (Author/LMO)

Responds to criticism of an earlier (1984) paper in which it was argued that the complements of Welsh control and raising verbs should be analyzed as verb phrases (VPs) and not as clauses with empty subjects. The l984 position is defended against the traditional analysis in the three critiqued areas.

Refutes the theory that indirect requests are ambiguous. Arguments for it are examined and an attempt is made to expose the weaknesses in the kinds of tests generally used to detect "illocutionary" ambiguity. An alternative analysis in the framework of semantics and the pragmatics of directive speech acts is suggested. (AMH)

Reviews the comments of Goldsmith on Firthian Prosodic Analysis (FPA) and its relationship with Autosegmental Phonology (AP). It is argued that AP has maintained the insights of FP while providing additional analytic possibilities. The authors examine Goldsmith's misinterpretations and clarifies the characteristics of FPA. (94 references) (CK)