LINGUIST List 4.1021

Wed 01 Dec 1993

Disc: Word-classes

Editor for this issue: <>

Directory

1. RichardHudson50, Word-classes and cross-classification

Alexis Manaster-Ramer assumes that word-classes (aka parts of speech - a
dreadful term!) are cross-classified, but it's an interesting and important
question whether this is really true. In contrast with say phonemes, it's by
no means obvious that they are. He quotes the range of pro-forms that can
stand in for verbs, nouns an on, but why should we think that they
constitute a word-class (rather than, say, sharing the property of involving
identity-of-sense anaphora)? As for ad-X, where X ranges over various word-
classes, it's true that there are differences, but why should these involve
cross-classification, rather than a non-cross-classifying distinction
between adjective and adverb? I notice that he doesn't assume the much
more standard cross-classifying analysis of word classes in terms of +/-
N and V, which also strikes me as not obviously true.

My own conclusion, for what it's worth, is that word-classes are in fact
hierarchical (in an inheritance hierarchy), just as in traditional grammar,
in sharp contrast with morpho-syntactic features, which are truly cross-
classificatory.

Dick Hudson
Dept of Phonetics and Linguistics,
University College London,
Gower Street,
London WC1E 6BT
(071) 387 7050 ext 3152