The great achievement of the Modistae with regard to syntactic theory was the development of a formal model of sentence structure. In the fully developed modistic model, the sentence was analyzed into a set of ‘constructions,’ or linkages connecting one word with another, and each construction was then described in terms of the two primitive grammatical relations, that of dependens to terminans and that of primurn to secundum. The emergence of this model marked the culmination of the move toward formally explicit, conceptually parsimonious syntactic descriptions that had been going on since the early days of regimen-theory.

In many ways the syntactic theory of the Modistae reflects the Aristotelian philosophical paradigm within which they worked; medieval philosophers regularly drew on the ontology and metaphysics of Aristotle for the kind of tools of thought that modern linguists draw from set theory and related disciplines. For example, the one-to-one relationship of dependens to terminans in each construction lent itself well to treatment in terms of Aristotelian dichotomies such as matter-form and actual-potential, and it may be that this fact helped promote the idea that all constructions should in fact be viewed as one-to-one relationships.

However, it would be a mistake to think of modistic grammar as a force-fit of linguistic structure onto Aristotelian metaphysics. The Modistae, being realists, held that the perceptible world had a definite structure and that the semantic function of language was to represent this structure; hence they used Aristotelian ontology as a theory of the things language could signify, and as such it figured prominently in their semantics.

Verbs

The most central members of the word-class verb have the following two properties:

(a) Inflection. They are tensed: they have one or other of the inflectional properties ‘past tense’ and ‘present tense’. Thus in He lived in Sydney and He lives in Sydney the words lived and lives are prototypical verbs. In the present tense the verb – again prototypically – agrees with the subject as 3rd person singular vs general (i.e. not 3rd person singular): He lives in Sydney vs I/you/we I they live in Sydney.

(b) Functional potential. They function as the ultimate head of the clause (with the EVP and VP as intervening categories), as explained in the last chapter.

It is tensed verbs that are most sharply distinct from words belonging to other parts of speech: this is why non-tensed verbs may be regarded as less central members of the class. As we have seen, the non-tensed forms (more specifically the non-finite forms of the traditional paradigm) are traditionally spoken of as verbal nouns or verbal adjectives, and although we are rejecting that kind of description there is no doubt that they have closer affinities with nouns and adjectives than do tensed verbs. We thus take tensed forms as the prototype and include other words within the class on the basis of their functional resemblance to the prototype. The status of a non-tensed word as a verb is then clearest when it is functioning as ultimate head of a tensed clause as in He was writing the let.

Types and distinctive properties of pronouns

Traditional grammar has a number of different subclasses of pronoun: we will be looking at them in turn later in the chapter, but it will be helpful to begin by identifying at least their central members:

1. (1) Personal pronouns: I, me; we, us; you; he, him; she, her; it; they, them

2. (2) Reflexive pronouns: myself, ourselves, yourself, yourselves, himself, herself, itself, themselves

3. (3) Possessive pronouns: my, mine; our, ours; your, yours; his; her, hers; its; their, theirs

4. (4) Demonstrative pronouns: this, these; that, those

5. (5) Interrogative and relative pronouns: who, whom, whose, what, which, together with compounds in -ever – whoever, etc.

6. (6) Indefinite pronouns: anybody, anyone, anything; somebody, someone, something; everybody, everyone, everything; nobody, no one, nothing; none

Pronouns, we have said, are better analysed as a subclass of nouns than as a separate part of speech. We analyse them as nouns because the phrases they head are like those headed by common or proper nouns in terms of their functional potential and, though to a lesser degree, their internal structure. Functionally, pronoun-headed phrases are like other NPs in that they occur as subject, object, complement of a preposition, and so on: Everybody left; John criticises everybody; The change will be of benefit to everybody. There are, it is true, some differences in functional potential. A handful of pronouns have contrasting case-forms (I vs me, etc.), such that nominative forms like I and he don't occur as object or complement of a preposition, and accusative forms like me and him do not occur as subject, at least in kernel clauses.

Complements and adjuncts

Some of the initial concepts that we need in talking about the structure of kernel clauses may be introduced with reference to an example like

(1) Unfortunately, my uncle was using an electric drill at that very moment

We are taking the verb using as ultimate head of the clause, and immediate head of the VP was using (see 3.3); intuitively we can see that my uncle and an electric drill are more closely related to the VP and hence more tightly integrated into the structure of the clause than are unfortunately and at that very moment. We can think of my uncle, was using and an electric drill as combining to form, as it were, the structural ‘nucleus’ of the clause, with unfortunately and at that very moment being ‘extra-nuclear’ elements – or adjuncts as we shall call them. Adjuncts are always omissible, so that (1), for example, can be reduced to

(2) My uncle was using an electric drill

Let us now temporarily confine our attention to kernel clauses like (2), containing only nuclear elements, no adjuncts. At the first level of structure, such clauses consist of a subject followed by a predicate. The predicate we are analysing as the head, with the subject a dependent of it. The predicate position is always filled by an extended verb phrase; the subject is usually an NP, but it can also be an embedded clause or, occasionally, a PP (see 2.2).

In Chs. 4, 6 and 8 we have set out the grammatical properties which characterise the three major parts of speech – and it is clear that in a sentence like Both men became very angry, for example, we can distinguish very sharply between the verb became, the noun men and the adjective angry. Non-central members, however, are often significantly less easily distinguishable: the frequent appearance in grammars of expressions like ‘verbal noun’, ‘noun used as an adjective’, and so on, attests to this. In the present chapter I will consider in turn each of the pairs verb/noun, verb/adjective and noun/adjective, partly in order to develop and sharpen the earlier account of the distinctive properties of the classes, partly in order to draw attention to certain problems that arise in drawing boundaries between them or between the associated phrase classes.

We noted in 2.4 that what we are calling -ing forms of verbs are traditionally analysed as either gerunds (as in Taking [exams was a waste of time]) or present participles (as in [Everyone] taking [the coach should report here at 6 a.m.]). We argued against making any inflectional distinction between them on the grounds that no verb in the language exhibits any morphological contrast here. The traditional concepts are worth taking up again at this point, however, because they also cover words that we do not regard as -ing forms of verbs.

The term ‘grammar’ is used in a number of different senses – the grammar of a language may be understood to be a full description of the form and meaning of the sentences of the language or else it may cover only certain, variously delimited, parts of such a description. Here we shall use it in one of these narrower senses, embracing syntax and morphology. Syntax is concerned with the way words combine to form sentences, while morphology is concerned with the form of words. We will launch without delay into a discussion of basic concepts in syntax and morphology, returning in §8 to the distinction between grammar in this sense and various other components of a full description and to the basis for dividing grammar into syntactic and morphological subcomponents. The only terms that we shall need to anticipate are ‘phonology’ and ‘semantics’: phonology deals with the sound system, with the pronunciation of words and sentences, semantics deals with meaning.

Words and lexemes

Syntax deals with combinations of words, we have said, morphology with the form of words. But again the term ‘word’ has been used in a variety of senses.