concept of consciousness thus defined, if not a complete philosophical novelty,
then at least highly original: (i) consciousness is not interiority but the
internalisation of an exteriority; (ii) consciousness is not irreducible individuality
but an in-between, an effect of the sociality of interlocution; (iii) psychic
individuation cannot therefore simply be the result of an interpellation
(the Althusserian term is anachronistic here, but the idea that conscious-
ness derives from a social role, just as, in Mauss, the person is derived from
a mask – persona– is already unquestionably present in Voloshinov).^7
From these four founding concepts, Voloshinov draws three methodological
rules for studying language. The first enjoins us to never sever ideology from
the material reality of signs. Ideology is not abstract and ethereal; there is no
transcendent realm of ideas: it is always embodied – in gestures, intonations,
expressions. The second recommends that we never sever the sign from the
concrete forms of social communication – hence both the central importance
of a form of pragmatics and the rejection of any form of methodological
individualism. The speaker is always-already collective. The third encourages
us not to sever social communication from its base in the material infrastructure.
External linguistics takes priority; and language is to be analysed as social
and historical practice. The issue is not who speaks, and if she clearly expresses
what she means to say; it is to determine the place in the overall social
structure from which a voice – thisvoice – is raised.
This methodological starting-point has at least three significant consequences,
expressed by three original concepts, to which I have already alluded in
passing. The first is the concept of the multi-accentuationof the sign. We re-
call the test that Stanislavsky imposed on trainee actors: pronounce a simple
word – e.g. the exclamation ‘Good!’ – in sixty different ways, in order to give
the utterance thus produced sixty different meanings. Voloshinov makes this
exercise not a curiosity or an exception, but the starting-point of analysis.
Every sign is multi-accentuated not only because it contains a multiplicity of
possible meanings, realised by the use of the sign in concrete interlocution,
but in that it is history-laden, in that it sediments the meanings which these
realisations have imparted to it. To emit a linguistic sign is to take one’s place
in a chain of voices, which constrain – albeit never completely – the meaning
Continuations • 109(^7) See Mauss 1990.