[Some] theorists do not adequately respect the complexity, integrity and
uncomfortable implications of theoretical discourses - they are willing to
raid them, dismantle them and borrow from them to give authority and flesh
to what they already believe or to generate syncretic syntheses.
... there are many critics who wish to render
by tidying him up or by eliminating distortions due to the intrusion of
value bias, so that his work fits the procrustean bed of a positivistic
... there are those who reject the relevance of science for the analysis of
human behaviour and advocate instead one or other form of idealism.
Much recent writing has demonstrated that these are false alternatives not
least because the positivist model does not even fit the natural
sciences...; furthermore Durkheim's 'scientific rationalism' is not a
simple variant of positivism. ... whether correcting or dismissing it, both
positions share a dogmatic approach to Durkheim's work -judging it by some
already constituted theoretical discourse.
The characteristics and limitations of such an approach to theory were
in his discussion of the two different
ways in which
read classical political economy. In a dogmatic reading
the earlier text is reduced to the status of an anticipation (or
ramification) of the later discourse and its study will therefore yield
little of interest. In contrast to this he advocates a 'symptomatic
reading' by which one can identify 'the combined existence of sightings and
oversights which in an author "poses a problem, the problem of their
combination".' This helps one to search for examples of 'the correct answer
to a question that was never posed'
Balibar, 1970, pp. 19-22) and to recognize the presence of more
than one discourse in a text, to
separate them out and to discover if they complement, suppress or displace
each other thereby truncating each other's development. The symptomatic
reading will be effective
in so far as it divulges the undivulged event in the text it reads, and in
the same moment relates it to a different text, present as a necessary
absence in the first . . . the second text is articulated with lapses in
the first. (Althusser and
Balibar 1970, p. 28)
In practice Althusser distinguishes between two different kinds of texts,
one of which can indeed be assessed dogmatically, as having nothing to say
- although it may be internally consistent to the point that it only ever
poses questions that it has already answered _ whereas the other is not so
much ideological as pre-scientific. The latter are read symptomatically
from a position grounded in an already constituted scientific discourse.
This helps to identify its useful parts and to discover symptoms (of
incoherence) and to assess whether these merely indicate inadequacies or a
point at which discourses intersect.
Hirst's Durkheim, Bernard and Epistemology
is such a reading
of Durkheim. He stresses the power of Durkheim's critique of social
contract theories, applauds his consistent anti-humanist stance, his anti-
subjectivism and his assertion of the irreduceable nature of the social,
but criticizes him for misrepresenting Bernard and for succumbing to a form
of metaphysical essentialism.
Thus Durkheim is congratulated for anticipating structuralism and for
implicitly answering questions not put in his texts. His work is useful in
so far as it agrees with or anticipates structuralism. The rest is confined
to silence, or better is repressed, perhaps with the danger that it might
re-emerge unbidden in the critic's own discourse.
Fundamental to these arguments of Althusser and Hirst is the belief that
there exists a realm of scientific discourses that can be used to judge
others, demanding compatibility in so far as they overlap. The mode of
reading may be more sophisticated but it is still undoubtedly dogmatic.
Hindess has argued that we must abandon any dogmatic distinction
science and ideology. If theoretical discourses are to be criticised it can
no longer be because they are alleged to be derived from some empiricist
process or from epistemology but only because of an inadequacy at the level
of their concepts and the relations between these concepts.
1977a), pp. 223-7
Such analyses help us to specify what is worth retaining within a discourse
and to locate where it has become incoherent and thus how to rework it to
generate more valid arguments. Hindess here acknowledges the potential
incompatibility of internally coherent discourses but implies that the
analysis of discourses is an activity of discovery - of stumbling upon, or
unveiling what is already latent in the texts, thereby underplaying the
intertextual nature of reading.
Reading is intertextual in that discourses are often identified only
through the examination of more than one of an author's texts and a serious
reading will involve examining a wide range of these. We must ask whether
everything an author 'wrote, said or left behind is part of his work?'
M. 22.2.1969, p. 103 in
Foucault, M. 1986). In Durkheim's case these might include his
'marginal' writings on law, for example his somewhat Hobbesian 'La Science
positive de la morale en Allemagne' or his wartime articles on Germany,
where he pointed out that legal formalism could contribute to an unjust,
aggressive social order
1972). The texts that constitute an
oeuvre are inevitably a selection.
In this book an attempt is made to identify the discourses that traverse
Durkheim's texts, to disarticulate and evaluate them and to show how the
development of some of the more fruitful is truncated by the presence of
others that are less coherent.
The main focus is on The Division of Labour in Society,
Suicide, The Elementary Forms of the Religious Life, Moral
Education, Professional Ethics and Civic Morals, The Rules of
Sociological Method and Socialism.
Texts are also intertextual in that those who read them are always
themselves already positioned by other texts and by the relations between
discourses, discursive practices and the forms of subjectivity thereby
constituted. Social practices work in part through the
subjects and this is as true of reading subjects as speaking/writing
subjects or acting/feeling subjects.
"A meaning effect does not pre-exist the discursive formation
in which it is constituted. The production of meaning is an integral part
interpellation of the individual as a subject; in so far as,
other determinatons the subject is 'produced as cause of himself in the
subject form of discourse, under the influence of interdiscourse."
1982 p. 187)
The practice of reading, then, is always one produced by an interdiscourse.
Thus, for example, whilst Nisbet
is correct that the conservative
elements of Durkheim's work form 'one of the coherent systems of his
thought', he seems blind to the fact that, as Hunt
have argued, it is only one such system in a complexus of political
discourses - socialism, radicalism, conservativism - that traverse his
All three commentators, however, can be criticized for not attempting to
specify from 'where' they themselves engage in such readings - to read a
text is to be positioned by it, by the other texts with which it is
articulated, and by the other discursive formations within which one is