280 music, philosophy, and modernity
new approach is that ‘A language I do not understand is not a lan-
guage’ (ibid.: 106 ), because there would be no sense in which it makes
the kind of connections on which meaning depends.
Music plays an extensive role in the exploratory texts of the early
1930 s that lead toPhilosophical Investigations. The straightforward rea-
son for this would appear to be that music provides a series of analogies
which allow Wittgenstein to expand his notion of language. However,
if we can no longer be happy with the notion of specifying the limits of
language, because in modernity it is no longer clear what ‘language’ is,
we cannot assume that we can clearly delineate what ‘music’ is either.
The point, as Wittgenstein comes to realise, is that we use all sorts of
means with which to communicate, and the elements of these means
rely on their relationships to other elements within particular prac-
tices. He talks, for example, of a colour sample as ‘only part of a lan-
guage’ (ibid.: 54 )–i.e. it is significant only in relation to other colour
samples – and says that ‘(Words are not essential to language)’ (ibid.:
4 , 186 ). Elsewhere he insists ‘Don’t forget here as well that verbal lan-
guage is onlyoneof very many languages and that there are transitions
from it into the others. Consider the map in terms of what in it cor-
responds to the expression of verbal language’ (ibid.: 194 ). No form
of language is self-contained, because some elements of each notional
‘language’ can play a role in more than one other language, and the
borders between these languages need not be fixed.^7
In the subsequent remarks he considers these transitions in terms
of music: ‘I must, if I speak to myself, already play on a given/existing
language-piano’; ‘If I use a word in language then it is either because
I wish to play [‘anschlagen’, the word for striking a key on the piano]
it as an already familiar note, or that I want to establish that I will use
the word in future in such a way’ (ibid.: 170 ). A vital role is played
here by gesture, which will be inseparable from ‘music’: ‘The child as
well onlylearnsone language by means of another one. It learns verbal
language via the language of gestures. But adults must presuppose or
wait for the understanding of the latter. Nobody thinks of teaching the
child the language of gestures’ (ibid.: 3 , 249 ). An ability to perceive
sense in gestures is therefore constitutive of what it is to be a being
that can understand. Otherwise we would be faced with the regress
7 This is why the attempts of Lyotard and others to use Wittgenstein to argue for radical
incommensurabilities between language-games are doomed to failure. In which language-
game does one locate assertions about the incommensurability between other language-
games? (See the Conclusion.)