An Overview of Lexical Semantics

This article reviews some linguistic and philosophical work in lexical semantics. In Section 1, the general methods of lexical semantics are explored, with particular attention to how semantic features of verbs are associated with grammatical patterns. In Section 2, philosophical consequences and issues arising from this sort of research is reviewed. There is a longstanding philosophical tradition, going back at least to Parmenides and Socrates, of paying much attention to the meaning of words. In the twentieth century, Wittgenstein, Quine, and others centered a great deal of their philosophical research around a number of general questions regarding the nature of meanings, how words might come to have meanings, the relationship between words and their worldly denotations, etc. Although it is a subfield of linguistics, lexical semantics can be viewed as a refinement of this philosophical tradition. Moreover, just as the philosophical study of word-meaning has been realized in various different ways, so too there are many distinct linguistic projects that fall under the rubric of ‘lexical semantics’. The present article focuses on one central aspect of lexical semantics. In particular, I will consider those aspects of word-meaning that appear to be intimately connected to aspects of the ‘syntax’ or structural design of human languages. While this type of research is extremely common in linguistics, it often plays a less central role in philosophical discussions. There are exceptions to this, however; e.g., James Higginbotham has postulated that What is crucial to the design of language is not some distinction, however drawn, between properly semantic information, on the one hand, and empirical or collateral information, on the other, but rather the distinction between information that has systematic grammatical effects and information that does not. (470) At this point, the question immediately arises as to what it might be for semantic structure to have ‘systematic grammatical effects’. How could semantics affect syntax, and what kind of evidence for such a view could there plausibly be? I address these questions in Section 1 with a somewhat detailed example that illustrates the kinds of methods linguists use to 120 An Overview of Lexical Semantics © 2007 The Author Philosophy Compass 3/1 (2008): 119–134, 10.1111/j.1747-9991.2007.00101.x Journal Compilation © 2007 Blackwell Publishing Ltd adduce evidence for syntactically relevant semantic structure in words. In Section 2, I explore two places where lexical semantics is philosophically relevant: in issues regarding analyticity and the identification/individuation of concepts. I conclude in Section 3. 1. Extracting Structure from Patterns: An Example of Lexical Semantics Research Let’s begin by considering a miniature example of how linguists use grammatical distributions of words as evidence for the presence of complex semantic structure in the words. (Parts of the following example come from Levin, English Verb Classes.) I stress that this example is only meant to illustrate the types of methods and evidence linguists use. In actual research, considerably more evidence, often involving multiple languages, is typically used. Consider the verbs break, cut, hit, and touch, which all appear as transitive verbs: (1) a. Margaret cut the bread; b. Janet broke the vase; c. Terry touched the cat; d. Carla hit the door. However, these verbs are not equally acceptable in the so-called conative construction: (2) a. Margaret cut at the bread; b. * Janet broke at the vase; c. * Terry touched at the cat; d. Carla hit at the door. Another difference in patterning appears when we examine the part-whole alternation: (3) a. (i) Margaret cut Bill’s arm; (ii) Margaret cut Bill on the arm; b. (i) Janet broke Bill’s finger; (ii) * Janet broke Bill on the finger; c. (i) Terry touched Bill’s shoulder; (ii) Terry touched Bill on the shoulder; d. (i) Carla hit Bill’s back; (ii) Carla hit Bill on the back. A third source of difference in acceptability comes from the middle construction: (4) a. The bread cuts easily; b. Crystal vases break easily; c. * Cats touch easily; d. * The door frame hits easily.