This paper discusses the usage of some constructions which display non-canonical word order, despite having grammatical counterparts with canonical order—S V (O) (ADV)—whose truth conditions are identical: (1a) Outside stood a little angel vs. (1b) A little angel stood outside. It is claimed here that non-canonical constructions like 1a allow language users to exploit alternative word orders for multiple purposes, ranging from the facilitation of encoding to the creation of many specific rhetorical effects. A large number of syntactic constructions may be transformed into inverted structures of two target types, and these may be exploited to serve a variety of pragmatic functions.

It is commonly believed that agents constitute a single primitive category, distinct from other categories such as patient. This paper challenges such an assumption, arriving at three conclusions: (a) A variety of rules classify agents into affected vs. non-affected types. (b) These rules group affected agents with other affected roles such as patient, dative, and experiencer. (c) The affected/non-affected distinction is more basic than role distinctions such as agent and patient. Among the correlates claimed here for the affected/non-affected contrast is ‘causee’ case-marking, an area that has attracted recent controversy in Comrie 1976 and Cole 1976.

Some conservative Italian dialects possess simplified color systems, sometimes having fewer color terms than Classical Latin. Thus a hypothesis formulated by Berlin & Kay 1969 is confirmed: evolution of color systems is not always linear; regressions and subsequent re-elaboration may be observed. Italian dialects show some of the modalities of diachronic change of color systems.

The longitudinal emergence of verb inflections (-ing, -s, and -ed/IRREG) was observed in the spontaneous speech of four American English-speaking children. The occurrence of the inflections, their linguistic/non-linguistic contexts, and the conditional USE of the inflections (occurrence in optional or obligatory contexts) are described. The major results of the study are that the inflections -ing, -s, and IRREG emerged in the children's speech at the same time, but were distributed selectively with different populations of verbs. The results are discussed in terms of two major factors in the emergence of the verb inflections: the sentence relations between verbs and other constituents, and the semantics of verb aspect. The results are consistent with the general principle of ASPECT BEFORE TENSE in linguistic theory and in language development, and provide indirect evidence about the rules that children learn for verb inflectional morphology.

A distinctive feature of Finnish morphology is the possessive suffixes, which are found not only on possessed nouns, but also on adjectives, postpositions, and untensed verbs. Traditional grammars have taken the view that these suffixes arise through a rule of agreement with a genitive specifier, which may be subject to subsequent deletion rules. This paper shows that they are not agreement markers, but rather clitic allomorphs of the reflexive pronoun. X-bar theory is used to formulate rules governing their distribution.

Some Polynesian languages have accusative case systems, while others have ergative systems. Chung 1978 has argued that the Proto-Polynesian case system was accusative, and that the case systems of the ergative languages are the result of a PASSIVE-TO-ERGATIVE re-analysis, which re-interpreted passive structures as superficially active and transitive. We argue here that the passive-to-ergative re-analysis accounts historically for the unexpected properties of two syntactic rules found in ergative Polynesian languages: (a) the Raising rule of Niuean, and (b) the Relativization rule of Tongan and many Samoic-Outlier languages. Such an account is desirable since these rules are typologically rather unusual. Niuean Raising counter-exemplifies Postal's (1974) suggestion that Raising is universally restricted to complement subjects, while Relativization in Tongan and many Samoic-Outlier languages violates Keenan & Comrie's (1977) Accessibility Hierarchy. We conclude by arguing that certain differences in the histories of Raising and Relativization in Polynesian follow from the different characterizations of the two rules in universal grammar.