The Contingency of Universality’: Some
Thoughts on Discourse and Realism
LILIE CHOULIARAKI
Constructionism and realism are dominantly regarded as incompatible meta-theories. In
this article, I argue rather that a realist epistemology offers some premises that can usefully
ground discourse analysis in social scienti. c research. This has implications for the latter’s
modes of theorising and its potential for social criticism. The argument is in two moves.
First, I discuss how the concept of discourse . gures in three major critical traditions of social
theoretical, sociological and political theoretical thinking. I argue that, for different reasons
each, these traditions offer a less than satisfactory answer to the ontological question of how
discourse . gures in the social and/or to the epistemological question of how discourse is
operationalised in theory and research. Second, I turn to a discussion of critical realist
meta-theory. I critically discuss Bhaskar’s ontology of the real and suggest that realist
epistemology should be combined with a constructionist ontology, along the lines of feminist
theorising. This is useful in providing discourse theory and analysis with more effective
accounts on the nature of the social and on the modes of social inquiry. In so doing, I sketch
a view of discourse informed by critical realist elements and point to its conceptual and
analytical advantages.
Constructionism and realism are dominantly regarded as incompatible meta-theories.
Although neither can be reduced to a homogeneous . eld in itself, the two
positions are generally used as binary poles, which organise contemporary debates
on what the social world is and how we can study it. Such debates, more often than
not, exclude the possibility of a fruitful dialogue between the two positions. This
polarisation has implications concerning the theorising and research of social life in
critical social sciences. Roughly put, the world is either seen as an effect of discourse,
constructed through social practices of signi. cation, or as pre-existing social practice
in the form of some intransitive referent. Constructionists accuse realists of essentialism,
of insisting on the illusion of some pure existence, whereas realists accuse
constructionists of idealism, of the illusion that all existence is contingent on
language and signi. cation. For some characteristic arguments against realism see,
for example, Edwards et al. (1995), Laclau & Mouffe (1985) and Potter (1997), and
for some against constructionism see, for example, Bhaskar (1997/1998a) and
Archer (1998).
My own starting point is a constructionist perspective centred on the concept of
discourse. Discourse . gures here as constitutive of social practice. Speci. cally,
ISSN 1035-0330 print; 1470-1219 online/02/010083-32 Ó 2002 Taylor & Francis Ltd
DOI: 10.1080/10350330220130386
Downloaded By: [Fielding Graduate University] At: 17:21 21 May 2010
84 L.
p.92 Social Semiotics Commentary
First, the meaning–power relationship. Laclau & Mouffe combine the Foucauldian
view of discourse as . elds of signi. cation enunciated from positions of
power with Derrida’s view of semiosis as an in. nite play of signi. cation, which
precludes the . xity of meaning. Derrida (1978) speci. es the logic of the semiotic as
a logic of iterability. The semiotic is seen as a ‘differential structure’, a structure that,
precisely because it lacks a centre, cannot arrest meaning in its fullness. So, in any
practice, there is always a meaning surplus (‘supplement’) that exceeds its context
and that expands the possibilities for re-signi. cation without limit. The performative
force of the text (its capacity to constitute the social) resides not in the social logic
that produces it (a´ la Bourdieu), but in the semiotic logic itself, in the sign’s
necessity/capacity to break with prior contexts and assume new contexts, hence new,
unpredictable meanings: this ‘breaking force (‘force de rupture’) is not an accidental
predicate but the very structure of the (…) text’ (Derrida 1978, in Butler 1997a: 148;
emphasis added).
p.96
Discourse therefore is the materiality of the social, insofar as it both has
a speci. c material logic, the performative logic of the semiotic (cf. Derrida), and
insofar as it has material effects: it ‘materialises’ everything there is (Laclau &
Mouffe 1990; Butler 1993). This view comes closest to the Foucauldian view of
discourse already set out than either of the views already discussed. The materialities,
which phenomenologically appear to us as referents outside the semiotic, are
Downloaded By: [Fielding Graduate University] At: 17:21 21 May 2010
The Contingency of Universality 93
not real objects, but rather truth effects. They establish their positive identities
precisely at the moment when the meaning–power relationship that brings them into
being manages to conceal itself—and genealogy is that analytical practice that
exposes the status of such material positivities as contingent.
However, does the claim that all is discursively constituted necessarily imply that
the epistemological-analytical ‘language’ of discourse should be the ‘language’ of the
social overall? Laclau and Mouffe’s ‘language’ does exactly that: . eld of discursivity,
oating signi. ers, elements, discourse, articulation, moments, nodal points—all
provide a rich and sophisticated theorisation of the social as a terrain of discursive
articulations, which emphasise its openness and indeterminacy, its constitution and
change through constant ‘resigni. cations’.4 But if the task of critical, including
genealogical, analysis is to show how certain materialities are postulated as more
physical, hence as more ‘immediately real’, than others, then we need to study both
how this ‘immediate’ materiality is ‘given’ in a practice and how it is ‘open’ to new
classi. catory struggles and meanings. As well as employing Laclau and Mouffe’s
existing theoretical language, which accounts for the latter (the openness of discourse),
we also need a language that accounts for the possibility of relative closure.
This means accounting for physical materiality in its speci. city: as a modality of
materiality that is indeed discursively constituted but whose logic of constitution
(and thus its ‘degree of resistivity’ to closure) makes it function differently from
the materiality of the linguistic semiotic as temporally unfolding interaction in the
here-and-now of a practice.5 So, although Laclau and Mouffe acknowledge the
extra-semiotic, their epistemological ‘language’ binds the social and its ‘physical’
materialities too close to discourse. This is a consequence, I have argued, of their
turn to Derrida’s view of signi. cation and his emphasis on the unlimited, rather than
conditionally limited, reiterative function of the semiotic (see Butler 1997a: 150–51
for a critique on Derrida’s view of iterability as inherent in the differential structure
of language, ‘abstracted from its social operation’). As a result, Laclau and Mouffe’s
‘language’ of description of the social ignores what might be called the ‘multi-materiality’of
social practice and underestimates the extent to which other, non-semiotic
(but always semiotised) modalities of matter may bear a performative force on social
practice.
No comments:
Post a Comment