Language and Human Understanding

XII. Marrying Philosophy and Grammar in Distinguishing Types of Noun Expression

528 XII Marrying Philosophy and Grammar in Distinguishing Types of Noun Expression Section 1. The Varied Types of Meaning of Abstract Nouns (a) The roots of grammar lie in the semantics of speech: the awareness of langue involved in speech In chapter IX we saw that in any semantically structured expressive speech, the subordinate sentence constituents must be of different kinds. The distinction between referring or “naming” expressions and saying or “verbal” expressions, the latter “incomplete” or “unsaturated”, was fundamental, because of the overarching functional role of the latter. We also picked out what we called “application controllers” (typically quantifiers), and, within verbal expressions or predicates, we indicated the need to pick out certain key lexical elements as “content-introducing”. The capacity to make such logic-inspired distinctions is what enables us to penetrate the grammar of the language. However, much more is required for the development of grammatical theory. Some sentence-factors lie on the surface and are easy to identify by simple quotation, sometimes in the split-up form anticipated in chapter I, seen in the lexical predicate “look ... (optional) at” where “... (optional)” indicates the option of intruding some intervening expression, for instance, “intensely” in this case, or “the man” in “phone ... (optional) back”. This, crucially, involves understanding the “split-up” sentence-factors concerned as constituents, not mere components, as explained in chapter I. The same understanding of Marrying Philosophy and Grammar   529 sentence-factors as constituents rather than as components also allows us to count a lexical factor or lexeme such as “break”, qua lexical factor, as a constituent when we meet “broke” and “broken”. And constructions and other structural features, marked by the order of words and the morphological adaptation of lexemes, sometimes assisted by sentence-intonation, are other constituents. Here we see the importance of the shared practical knowledge of a language implicit in communal language-possession (“knowledge of langue”) and its accessibility to consciousness. We register the interconnections between “breaks”, “broke”, the participles “breaking” and “broken”, the nounform gerund “breaking”, the infinitive “to break”, and the unmarked present “break” in our practical use of this family of words in different utterances, including utterances exchanged between different times or speakers, different combinations with the copula and various auxiliaries, or embedded in varied clausal constructions. The language-user is aware of such lexical connections and of the semantic significance of these features, including intonation, without his or her needing to label or classify them—but their accessibility to reflective awareness provides the basis of the possibility of “grammar” as a part of linguistics. Developing grammars depends on appreciating the ways words and lexemes are used in speech. This appreciation also gives access to the roots of morphology, important for understanding live utterance as well as for morphological creativeness.1 (i) The variety of ways of forming nominalizations from predicates or clauses Thus the fact that langue as well as parole is present to consciousness so as to be the subject of reflection makes possible the generative capacity which appears most obviously in the way in which new topics of discourse, or new logical subjects, can arise. At least some such generative capacity is involved whenever we single something out—for instance, when we topicalize a presupposition of a “Yes/No” question, or pick it out by the appropriate kind of wh-question, and once having thus picked it out, thereby ipso facto having made it the subject to the verb “to be”. It is this which at an elemental level enables us to single out individual terms or expressions—for instance, proceeding from “He said that I was clumsy” to “Clumsy is what he said I was”. In speech and in writing systems without quotation marks, there is nothing marking the fact that here it is an expression of or for something, not the thing 1. I discussed the distinction between lexically significant morphemes and lexically nonsignificant morphemes, the latter important for perception and sometimes for associations derived from their etymology , in the appendix to chap. II, on morphology, in this volume. The distinction is fuzzy, but it is clear that “ante” from Latin and “anti” from Greek both retain productive lexical significance, vide “ante‑date” and “antioxidants ”. 530   Rewriting the philosophy of grammar referred to, which is being “topicalized”—so that the subject is being spoken of in what logicians now call the formal mode of speech.2 This capacity to single out...