The aim of this paper is to argue that the process of code-mixing is constrained by the government relation that holds between the constituents of a sentence. The government constraint replaces a number of specific constraints that have been proposed in the literature to account for apparently ‘impossible’, ‘ungrammatical’ or ‘non-occurring’ types of intra-sentential switches. Code-mixing is a form of linguistic behaviour which produces utterances consisting of elements taken from the lexicons of different languages. Some examples are given in (1).

The terms definite and indefinite, when used in relation to NPs, can be handled in two different ways. One can treat definiteness as a formal (syntactico-morphological) category and make use of the syntactic features [±definite], as in Chomsky (1965). One can also go into the question of how definite and indefinite NPs are used, i.e. the question of when and why a (formally) definite NP is preferred to an indefinite one, or vice versa. If we take the latter approach, we are investigating definiteness as a semantico-pragmatic category.

The phenomenon that I describe in this paper is a specific class of indirectly conveyed meanings. The meanings involved are definitional, or partially definitional. Though conveyed by the utterance of sentences, they are not part of the meaning of those sentences. To begin with, I shall use the term ‘non-descriptive definition’ to refer to the act of conveying such a meaning by the utterance of a sentence that does not have that meaning. In due course, though, I shall refine the reference of that term.

A number of languages have what can be called prepositional complementizers, items which look like prepositions and have some of the properties of prepositions but appear in positions in which complementizers appear. In this paper I shall be concerned with Welsh prepositional complementizers and, in particular, with their implications for the analysis of subjectless infinitives. I shall consider the possibility of providing an analysis of the Welsh facts within the government-binding (GB) framework of Chomsky (1981, 1982, 1984), in which subjectless infinitives are clauses. I will argue that GB assumptions preclude a satisfactory account of the data. I will also show that certain facts relating to agreement pose further problems for a GB approach. I will then show that the facts are quite straightforward if one assumes that subjectless infinitives are bare VPs, as argued for Welsh in Borsley (1984a), and that VPs and Ss are members of the same basic category, as proposed in Borsley (1983, 1984b). It looks, then, as if we have some interesting evidence here for these assumptions.

This paper is an argument for the constraint on grammars known as the ‘true generalization condition’ (Hooper, 1976:13): all rules express generalizations true for all surface forms. I make this condition fully explicit by interpreting it to mean the prohibition of the three transformational rule-types: deletion, movement and feature-changing. The argument takes the form of a comparison of a recent autosegmental analysis of the intricate facts of Arabic root and pattern verb stem morphology with an alternative which observes the condition. I hope to show how the latter analysis in every empirical aspect is equivalent to the former in its claims about Arabic, and significantly differs, as the result of observing the true generalization condition, in its lack of numerous un-empirical claims made in the autosegmental analysis. In so far as both have descriptive adequacy, the analysis governed by the true generalization condition, termed ‘non-transformational’, has also explanatory adequacy in the sense of Chomsky, 1964: 28–9, since it is closely determined, or selected, by the true generalization condition.

As is well known, English has two genitive or possessive constructions, the ‘proposed’ and the ‘postposed’, exemplified in (1).

In each case we have an NP, with a head N (book, office, dog, house, plants) modified by a possessive expression (John's, a man's, mine, etc.). This expression is itself an NP in the genitive Case, and I shall refer to it as the ‘genitive phrase’. By contrast with other familiar languages more highly inflected than English, genitive Case is hot marked by an inflection on the head of a genitive phrase, but by the clitic ’s, which is attached right at the end of the phrase. The exception is where the genitive phrase is not a full NP but a personal pronoun, in which case we get an inflected form (irregular in pattern) as in these other languages: I - my/mine, he - his, etc. These possessive forms of pronouns have almost identical distribution to that of full NPs in the genitive (there are some differences which I shall point to below), and so it seems clear that they are genitives, despite the morphological difference; personal pronouns are highly irregular morphologically anyway, and not only in English. This is assumed in all recent work I know of, and I shall take it to be uncontroversial.

A long recognized problem for linguistic theory has been to explain why certain sounds, sound oppositions, and sound sequences are statistically preferred over others among languages of the world. The formal theory of markedness, developed by Trubetzkoy and Jakobson in the early 1930's, and extended by Chomsky and Halle (1968), represents an attempt to deal with this problem. It is at least implicit in that theory that sounds are rare when (and because) they are marked, and common when (and because) they are not. Whether sounds are marked or unmarked depends – in the latter version of the theory, particularly – upon the ‘intrinsic content’ of acoustic and articulatory features which define them. There was, however, no substantive attempt among early proponents of the theory to show what it was about the content of particular features and feature combinations that caused them to be marked, and others not.