question of no longer regarding them as stable or static things or facts, but
as dynamic processes: a ‘word’ covers a series of semantic, morphological
and phonetic variations each of which has its own history; an ‘utterance’ is
not the inscription of an ideal proposition, but the site of a polyphony whose
result is multi-accentuality; and a language – English, for example – is not
an ideal type realised in individual speech acts, but (as we have seen) a fluid,
dynamic set of dialects, registers and styles. Alternatively put, language is
no longer conceived as a stock of words (like languein Saussure), or of rules
(like Chomskyan competence), but as a system of variations – a formulation
which (as we have seen) is not as paradoxical as it might seem. Here is what
Vygotsky, the Soviet psychologist, has to say about the meaning of words:
The discovery that words’ meanings evolve leads the study of thought and
speech out of a blind alley. Word meanings are dynamic rather than static
formulations. They change as the child develops; they change also with the
various ways in which thought functions.^3And Vygotsky adds: ‘The relation of thought to word is not a thing but a
process, a continual movement back and forth from thought to word and
from word to thought.’^4
The first quotation seems to be a matter of plain good sense: we know very
well that the meaning of words changes all the time and we are used to
resuscitating these superseded meanings and telling these stories, in all their
nuances, for the benefit of younger generations, who pay no attention to
them. The second quotation indicates the enemy by negation: a positivist
conception of language that wants to isolate and fix facts so that they are
more readily manipulable by science. To replace these ‘facts’ by processes is
to affirm that the object of the ‘science of language’ is not to discover the laws
of nature, but to describe pragmatic maxims, of the kind ‘do not say x, say
y’ – or indeed ‘if you say x, then you must also say y’ – where it is clear that
power relations obtain and that these ‘rules’ are made to be challenged,
exploited, changed. This means that what are called ‘exceptions’ (leaf/leaves,
mouse/mice, etc.) are not so much defects of grammatical construction, which
a refinement of the rules will soon suppress, as indices of a semantic and
Propositions (1) • 141(^3) Vygotsky 1962, p. 124.
(^4) Vygotsky 1962, p. 125.