Parts of speech is the traditional term for the major classes of words that are grammatically distinguished in a language. While all languages make parts-of-speech distinctions, there are rather striking differences between languages with regard to both the kind and the number of such distinctions that they make. A field worker investigating an unfamiliar language may therefore find it useful to know what generalizations can be made about parts-of-speech systems. What, for example, can be said about the ways in which, and the limits within which, parts-of-speech inventories may differ from one another? Which parts-of-speech distinctions are universal and which language-specific? What are the ways in which languages that lack a particular part of speech express the semantic equivalent? And what relations are there between the parts-of-speech system of a language and the language's other typological characteristics? It is the aim of this chapter to provide some answers to such questions.
By way of orientation, the present section sets forth some general assumptions that underlie the presentation in the rest of the chapter. First, then, it is assumed here that the primary criteria for parts-of-speech classification are grammatical, not semantic. As has been amply demonstrated in the linguistic literature (see, for example, Fries (1952)), the familiar notional parts-of-speech definitions, such as ‘a noun is the name of a person, place, or thing’, fail to provide an adequate basis for parts-of-speech classification, since there are many cases in which their applicability or inapplicability is unclear. Grammatical criteria, on the other hand, are not open to this objection.