Variation in parts of speech systems
The languages of the world show some variation in what parts of speech their grammars contain. In a typical sentence of any language, we can identify constituents which are translated by, and seem to function in a similar way to, nouns and verbs in English or other familiar languages. But whether we want to call these words ‘nouns’ and verbs’ depends on a number of considerations; some researchers deny that there is any meaningful sense in which those two parts of speech can be said to exist universally (Foley in print; see also Croft 2003: 183, and Evans and Osada 2005). We will say more about nouns and verbs shortly. Once we move away from these categories, however, there is greater variation. Adjectives and adverbs are frequently missing from the inventory of grammatical categories found in the languages of the world. In Warao (isolate; Venezuela), for example, the single form yakera can be used without any modification both referentially, as a name of the abstract thing ‘beauty’ (1a), and as a modifier of both a referential object (1b), and a predicate (1c):
These uses would call for three quite formally distinct categories in English and many other languages: noun (1a), adjective (1b) and adverb (1c). In Ancient Greek (Indo-European, Greece and Mediterranean; extinct), for example, the translations of yakera in (2a–c) are as follows:
In contrast, Warao shows no morphological distinction between the three uses.
Differences in part of speech systems are not limited to the presence or absence of adjectives/adverbs. We can see this by comparing some rough part-of-speech inventories from four languages: Chickasaw (Muskogean; Oklahoma, USA), Wolof (Niger-Kordofanian; Senegal and Gambia), San Lucas Quiavini Zapotec (Zapotec, Mexico), and Warlpiri (Pama-Nyungan, Australia) as laid out in Table 9.1 (Munro 2006; Nash 1986):
