of desire) and a discursive, linguistic aspect (collective assemblages of
enunciation). Their principal characteristic is the ontological mix that occurs
in them. Thus, the feudal assemblage – Deleuze and Guattari’s canonical
example – is made up of human bodies (the knight, his lady and his vassals),
animal bodies (the knight’s horse), buildings and objects (the manor and the
sword), but also of utterances and texts (courtly love poems, decrees of justice),
and, hence, of institutions (courts of law) and rituals (marriage). This concept
is important in a variety of ways. It makes it possible to go beyond the
separation between material infrastructure and ideal superstructure, by
demonstrating the imbrication of the material and the ideal: language is
precisely the site of this imbrication. It therefore underscores the material
character of linguistic phenomena. But it also stresses their social character:
the source of utterances is not the individual speaker, but the assemblage of
enunciation, which is collective. The individual speaker speaks in that she is
spoken by the collective assemblage. This in no way prevents her discourse
from exhibiting the idiosyncratic originality of a style. The court poet produces
creative work in that he is spoken by a collective assemblage: in terms of
literary criticism, this is called a genre, as long as we conceive a genre as
an institution (a set of constraints that are both material and ideal), and not
only as an abstract set of rules. Finally, the canonical example clearly shows
that linguistic phenomena are historical phenomena: each assemblage has
its conjuncture.
Class struggle
That language is involved in the class struggle is self-evident: I have tried to
show that language is a political phenomenon. Politics finds its natural medium
in language: programmes, speeches, debates, and so on. I have said enough
about the specific importance of slogans, a theory of which represents Lenin’s
contribution to the Marxist philosophy of language, for there to be no need
to return to it. But we must go further. If language is the archetypal form of
historical human praxis, and if the history of humanity is the history of class
struggles, there must be a closer link between language and class struggle,
over and above the obvious linguistic divisions (dialects, jargons, styles,
speech genres) analysed by socio-linguistics. In the work of Bourdieu (whose
relationship with Marxism is complicated), we find a set of concepts (cultural
capital, authority-authorisation, intimidation, etc.) that make it possible to
200 • Conclusion