Recent times have seen a marked growth of interest in historical linguistics, as well as an increased role for diachronic factors within synchronic theory. This suggests that a fundamental re-assessment of the relationship between synchrony and diachrony in linguistic theory is in order, particularly with regard to its role in linguistic explanation. A number of ways are indicated in which diachronic factors enter at various levels as integral and at least equal partners in the over-all explanatory structure of linguistic science.

Mixture of Spanish and English, whether in isolated loan words or in code-switching of clauses and sentences, while socially motivated, is subject to clear linguistic constraints. Quantitative analysis of mixing in conversations of Mexican-Americans suggests specific functional constraints to express tense/aspect/mood and subject/object relationships, as well as structural constraints which permit only surface structures which are grammatical in both languages. Resolution of structural conflict plays a key role, so that lexical cores trigger longer phrasal switches if they govern rules which create non-shared surface structures. The relative frequency of mixes without structural conflict is constrained by discourse function.

Dependency network representations, which eliminate the need for transformations, are proposed as a means for showing the essential semantic relationships among the components of sentences, especially those that contain oblique complements (including by-phrases). Evidence from Quileute, a Northwest American Indian language, provides strong support for the position, advanced by Langacker & Munro 1975, that passive sentences are basically agentless, and for their suggestion that all oblique complements have external sources.

The material presented here is intended to illustrate the arguments presented in our article on Noun Phrase Accessibility (Keenan & Comrie [1972] 1977, hereafter NPA); it was originally intended that these data would be published as an appendix to that article. Exemplification is given for each strategy used to form relative clauses (RC's) in each language cited in Table 1 of NPA (pp. 76–9), except where adequate data appear in the body of the text. All information is based on work with native speakers, except where explicitly stated otherwise. Acknowledgments made in NPA (p. 63, fn.) also apply to the present publication.

It has been shown by Keenan & Comrie 1977 that restrictive relative clauses (RC's) in many or perhaps all languages are subject to several constraints which make reference to the Noun Phrase Accessibility Hierarchy and to a classification of RC's based on their surface-structure properties. It is argued here that our understanding of RC-formation in universal grammar is further enhanced if a modified version of the classification of RC's, proposed informally by Givón 1975, is used instead of the classification of Keenan & Comrie.