The following text has been automatically reproduced by an Optical Character Recognition (OCR) algorithm. It may not have been checked over by human eyes. For matters of precision please consult the original pdf.

Habermas on Heidegger and Foucault

Habermas on Heidegger
and Foucau It
Meaning and Validity in
The Philosophical Discourse of Modernity

Rudi Visker
Meaning, Habennas warns us, should not be allowed to consume
validity. 1 For once that happens the further exhaustion of the
project of modernity and the loss of its nonnative content are
inevitable. This is clear from the writings of those Habennas
calls ‘the theorists of the counter-enlightenment’ – a category
broad enough to include authors as diverse as Heidegger,
Nietzsche and Foucault. One way or another, these philosophers
are all undoing the intrinsic connection between meaning and
validity. More-over, the political mesalliance of some of them
serves to show how catastrophic the replacement of critical
theory’s commitment to ‘the philosophical discourse of modernity’ by some blend of post-modernism and post-structuralism is
bound to be. For, according to Habennas, only a theory that
respects the internal relation between meaning and validity,
without eliminating the difference between the two, can be
entrusted with the delicate task of defending the legacy of modernity, whilst retaining a critical perspective on the way it is
materialised in society. And Habennas is sufficiently confident
in the results of his Theory of Communicative Action to claim the
title for his own theory and to counterpose it to a host of other
attempts, which are successively shown to be mistaken or to
have missed the opportunity of taking ‘the alternative paths’

(PDM, 295) implicit in their own problematics.

Such was the programme and the underlying intention of the
famous series of lectures Habennas gave on ‘the philosophical
discourse of modernity’: a series of criticisms in the first ten
lectures, followed by an attempt to show that the ‘alternative
way out of the philosophy of the subject’ (title of Lecture XI) lies
in his own focus on a communicative reason which remains
faithful to ‘the nonnative content of modernity’ (title of Lecture
XII). Ironically, perhaps, the book succeeded in doing exactly
the reverse: far from demonstrating the superiority of his own
conceptual apparatus, and preparing for an equitable exchange
among responsible intellectuals, Habennas seems to have estranged himself from his audience, to the point where his opponents accused him not only of seriously misunderstanding them,
but of not having read them at all, as the following quotations
from Derrida attest:

It is always in the name of ethics, of an allegedly demo-

cratic ethic of discussion, it is always in the name of a
transparent communication and of ‘consensus’, that the
most violent infractions of the elementary rules of discussion are produced. It is always the moralistic discourse of consensus – or at least that discourse which
pretends sincerely to appeal to consensus – that in fact
produces the indecent transgression of the classical
Radical Philosophy 61, Summer 1992

nonns of reason and democracy. Not to mention of elementary philology …. The most prominent example … is
Habennas …..2
With a stupefying tranquillity, here is the philosopher
of consensus, of dialogue and of discussion, the philosopher who claims to distinguish between science and
literary fiction, between philosophy and literary criticism, daring not only to criticise without citing or giving
a reference for twenty-five pages, but, even worse, justifying his non-reading and his atmospheric or hemispheric choices by this incredible alibi: ‘Since Derrida
does not belong to those philosophers who like to argue
[argumentationsfreudigen Philosophen, my emphasis!

(J.D.)] it is expedient [ratsam] to take a closer look at his
disciples in literary criticism within the Anglo-Saxon
climate of argument in order to see whether this thesis
really can be held’.3
Far from opening up a dialogue, Habennas seems almost to have
lost – at least in the eyes of some – his right to be a partner in any
future exchange. Instead of taking sides here, we might do better
to analyse what went wrong. For that something went wrong
seems beyond doubt: not only Derrida, but almost every other
author discussed by Habennas, found himself forced into a kind
of philosophical Procrustean bed, which left him speechless
before the critical questions addressed to him.4 No one can
claim, of course, to have a definitive reading of an author – not
even the author could seriously claim this privilege – and it
would be pointless to criticise Habennas merely because his
readings of Benjamin or Nietzsche, of Derrida, Adorno or
Heidegger, are somehow flawed or not particularly interesting.

To supply a different reading of some of these authors merely for
the sake of contradicting Habennas seems an uninteresting enterprise. But to show, by way of such alternative readings, that
what is at stake is not only a misunderstanding, but a systematic
misunderstanding; and that the ‘system’ of this misunderstanding is directed by a theoretical position which is both crucial to
Habennas’s own theory and contentious; that does seem warranted. And that is what I propose to do here. The whole problem
is focused on the relation between a ‘validity’ and a ‘meaning’

which, according to a theory that itself seems to be ‘consuming’

its opponents, should not be allowed to ‘consume’ such validity.

Could it be, we may wish to ask, that what we witness in
Habennas’s reading of Heidegger and Foucault is the effect of a
theoretical position which, rather than sacrificing validity on the

altar of meaning, as Heidegger and Foucault allegedly do, undoes the internal connection between the two conversely – a
theory of validity that in its turn consumes meaning?

When Habermas tells us that meaning and validity are intrinsically or internally connected, he seems to be thinking of a
reciprocal causality between meaning and validity, of ‘a dialectical relationship between the world-view structures that make
intramundane practice possible by means of a prior understanding of meaning, on the one hand, and, on the other, learning
processes deposited in the transformation of world-view structures’ (PDM, p. 320). In other words, ‘the concrete a priori of
world-disclosing language systems is exposed … to an indirect
revision in the light of our dealings with the intramundane’

(PDM, p. 321). Less technically formulated, and more appropriate to our concerns here: even if we were to adopt some kind of
framework-relativism (and I will later try to explain in what
sense Foucault and Heidegger can be regarded as frameworkrelativists), our discovery of horizons of meaning (e.g.

epistemes, paradigms) is still not going to provide us with criteria for the validity of the statements, or of actions we undertake
on the basis of such frameworks. On the contrary, the problem
remains of how to discover whether specific validity-claims
(claims to truth, rightness, sincerity) can be redeemed; and in the
light of whatever such ‘valid’ experiences we may have, our
framework will have to change.

Accommodating as this theory may seem to some weak
version of framework-relativism – it allows for changes in the
horizon of meaning – it is important to point out that such may
not be its ultimate intention. In fact, a similar argument is already
present in the early Habermas’ s assessment of Nietzsche, where
he remarks that for all the Nietzschean pathos about our truth
being only an extra-moral lie, this ‘fictional’ status of truth could
still be sublated, as it were, in the context of a transcendentallogical pragmatism which points to the fact that some fictions are
more serviceable than others in helping the species survive.

There are ‘gattungsgeschichtlich “bewiihrte” Fiktionen’5 (and
these are the only ones we should care about and seek to pro-


mote. And if Nietzsche, some twenty years later, is still considered by Habermas to be the ‘turning point’ for the mistaken entry
into postmodernity (PDM, title of Lecture IV), it is because
postmodernity for Habermas consists in the mistaken belief that
the fictional character of truth could be used as an argument
against truth itself. Indeed, Habermas’ s disappointment with
both Heidegger and Foucault lies in the fact that, instead of
taking the alternative route of a ‘communicatively revised’ transcendentalism left open by Nietzsche in his writings, they wander into some kind of relativism which, conflating meaning and
validity, can therefore be criticised by means of the same transcendental-logical pragmatics. Heidegger, he tells us, ‘jumped to
conclusions in identifying the disclosure of meaning-horizons
with the truth [i.e., the validity] of meaningful utterances’ (PDM,
p. 320). Heidegger allegedly overlooked that this ‘changed understanding of meaning has to prove itself in experience and in
dealing with what can come up within its horizon’ (ibid.) – an
argument that recalls the young Habermas’ s pragmatist critique
of Nietzsche.

Similarly Foucault, according to Habermas, simply reversed
‘power’s truth-dependency into a power-dependency of truth’

(PDM, p. 274). Instead of finding in knowledge the guarantee for
power (Bacon), power for Foucault is said to become the true
face of knowledge. As with Heidegger, here too Habermas is
convinced that such monstrous statements derive from a conflation of meaning and validity. Repeatedly, he points out that
Foucault’s ideas about an internal link between power and
knowledge fall prey to a simple genetic fallacy, by seeking to
draw conclusions about the validity of knowledge (e.g. that of
the human sciences) from an investigation into the conditions
under which such knowledge arises (e.g. the link between criminology and the prison in Discipline and Punish) or is applied
(e.g. PDM, pp. 272-73, 415-16, n. 7 and 10). Hence what
Foucault considers to be an internal link between knowledge and
power is, according to Habermas, no such thing: his genealogical
findings have no relevance to the validity of the disciplines
whose history he attempts to unearth. As with Heidegger, what is
overlooked by Foucault is the fact that ‘it is only the conditions
for the validity of utterances that change with the horizon of
meaning’ (PDM, p. 320) and that such changes do not affect the
problem of validity.

The obvious question is: Is this
true? Were Heidegger and Foucault doing what Habermas says they are? Were
they simply brushing aside the difference between meaning and validity as
such, relying on the existence of frameworks that are nothing more than
‘protuberances of power’ ,6 or the effect
of some ‘ursprungs’ -philosophical
‘background-occurrence’ named Being
(e.g. PDM, p. 295 and Chapter VI, passim)? Or are they, contrary to what
Habermas believes, engaged not in undoing the internal connection between
meaning and validity, but in articulating it differently? And if they could be
shown to be engaged in the latter enterprise, does not the real question concern
at what cost Habermas himself succeeds in preserving an internal connection between meaning and validity?

And could it be, in the light of the cost,
that the connection he seems to be
pleading for turns out once again to be
an external one?

Radical Philosophy 61, Summer 1992



Habennas never tires of telling us that we can only understand
the meaning of any given speech act when we know the conditions under which it can be accepted as valid (e.g. PDM, pp. 31213,319-21).1 Validity, then, is a matter of knowing the conditions for validity. For example, truth is a matter of knowing the
conditions under which a proposition can be confinned as true.

And, in order to know such conditions, one has to refer to the
horizon of meaning dominant at the time. When this horizon
changes, the conditions for the validity of utterances change in
their turn. But this is all that happens, as Habennas stresses by
introducing an adverb whose significance should not be underestimated: ‘only the conditions for the validity of utterances’

change (PDM, p. 320 – my emphasis). This adverb is important
for two reasons. First, because it suggests that Habennas believes theorists such as Foucault and Heidegger do not see this.

Secondly, because one may wonder whether the ‘internal’ link
between meaning and validity, even granting that it holds for the
whole reservoir of meaning and the entire spectrum of validity
(not only for truth, but also for rightness and sincerity) (PDM, p.

321), is not still conceived of in tenns of knowing, of theoria.

Changed conditions of meaning, different concrete a prioris of
world-disclosing language systems, different epistemes, different frameworks, for Habennas these only necessitate a different
grasp of the conditions under which any given speech act can be
accepted as valid. Man’s relation to truth, for example, would
basically involve knowing the conditions that render a constative
speech act acceptable and then presumably living up to them (but
not to complicate matters further,8 I will leave this aside). And
the same would hold for man’s relation to other dimensions of
validity (sincerity, rightness). Any change of horizons of meaning, the very fact that there can be, as Habennas himself seems
willing to admit, a change in such horizons, is only deemed to be
important insofar as it changes conditions of validity. To be sure,
such changed conditions of validity will give rise to learning
processes which will in turn transfonn world-view structures,
and thus at first sight the ‘reciprocal causality’ between meaning
and validity seems to hold in both directions (though clearly
under the aegis of validity itself). But does the role Habennas
accords ‘validity’ here – the fact that, in conceiving of horizons
of meaning in tenns of conditions of validity, the stress is on
validity and not on conditions – does this theoreticist bias in
approaching the problem of meaning, this link between validity
and the knowledge of its conditions, not endanger Habennas’ s
claim to exclusive possession of a theory which intrinsically
connects both tenns without eliminating the difference between
them? Is there not something missing, something Habennas
might have picked up from such authors as Heidegger and
Foucault, had he been less convinced of the need to conduct a
rearguard action on behalf of the legacy of Enlightenment? Are
not Heidegger and Foucault, in their different ways, questioning
a move Habennas seems to take for granted: the assumption that
man’s relation to horizons of meaning, and to a validity which
draws its conditions from such horizons, can be exclusively seen
in tenns of knowing, of theoria? For what, after all, does it mean
for man to be in the truth, or to speak the truth? Could it not be
that Habennas, Heidegger and Foucault have been giving different answers to the same question?

Let me first try to show how the need for a non-theoretical
relation to truth arises just where Habennas would least expect
it: in the fact that ‘the conditions for the validity of utterances
change with the horizon of meaning’ – a fact that, pace
Habennas, neither Foucault nor Heidegger overlooked, but
which was at the centre of the framework-relativism they were

Discourse, limitation and exclusion: such were the tenns with
which Foucault initially tried to account for the fact that truth (or
falsity) can only occur within an order of truth. Certain conditions have to be fulfilled, certain models deployed, certain metaphors or concepts given preference over others, before a statement can be considered in its validity. Or better still: before a
statement can be judged either true or false, before it can be
considered as a candidate for truth or falsity, it has to fulfil
certain conditions which are more complex than those governing
truth or falsity in the strict sense. In order to be true or false, a
statement first of all has to be taken seriously by the court that is
to judge its merits, its validity. A statement has to be ‘in the true’;
it has to comply with the rules laid down by a ‘regime of truth’ ,
in order to be either true or false. Not everything can be said in
any given discourse; not all statements will be listened to in any
given age; and therefore the realm of truth cannot but be finite. If
there is to be truth (or falsity), it there are to be claims to validity,
there must be first of all a ‘regime of truth’ that prescribes a
principle – or set of principles – of relevance. And that is why
Foucault should not be taken as simply attacking or denying
truth, not even when he insists that there is a link between truth
and power, and that it is senseless to oppose power in the name of
truth, because truth itself is already a kind of power.9
Naturally, if one follows Habennas in taking such assertions
at face value, 10 one will be unable to see in this ‘politics of truth’

anything other than an attempt ‘scornfully’ to reduce ‘relationships of validity to the powers that triumph behind their back’

(PDM, p. 324). Rather than acknowledging power’s truth-dependency, Foucault’s insistence on the link between truth and
power would once more conduct us into the cul-de-sac where
Nietzsche allegedly left us: it would leave us with an analysis
that ‘strips the history of discourse-constitutive rules of any
authority based on validity and treats the transfonnative of
transcendentally powerful discourse fonnations .just as conventional historiography treats the ups and downs of political regimes’ (PDM, p. 255). Truth would no longer be truth were it the
mere expression of ‘power strategies [that] intersect one another,
succeed one another, [that] are distinguished according to the
type of their discourse fonnation and the degree of their intensity; but [that] cannot be judged under the aspect of their validity’ (PDM, p. 127 – Habennas’ s emphasis). All we would be left
with would be ‘a concept of power [that] does not free the
genealogist from [the] contradictory self-thematizations’ (PDM,
p. 295) familiar from all attempts to claim validity for a radical
critique of validity.

But the question is, of course, whether Habennas is not
merely reading this aporetic self-refutation of a radical critique
of reason that still wants to be reasonable (e.g. PDM, pp. 126-27,
341) into the works of authors such as Foucault, who, far from
wanting to abandon reason or truth, may simply be trying to shift
the meaning of such concepts in a direction which is not selfproclaimed irrationality, but a reason or a truth seeking to come
to tenns with its own conditions of possibility. For, as I already
pointed out, Foucault is not interested in denying truth. Like
Heidegger before him, the question he raises is what makes the
truth true. If what provides for the possibility of truth or falsity is
a certain regime involving selections and exclusions – a limited
principle of communication – then what follows is not that there
is ultimately neither truth nor falsity – or no communication at all
– but that a certain ‘self-evident’, ‘natural’ way of thinking about
these concepts has to be revised. And we not only need a revision
of these concepts but a theory which could explain why these
concepts have been conceived in their traditional fonn.

Such was Heidegger’s programme: to understand how it

Radical Philosophy 61, Summer 1992



came about that aletheia had only been understood as correspondence and adaequatio; and why it was that, in speaking the
truth, man forgot its essance (where the ‘a’ stresses the verbal
sense of Wesen) and its conditions of possibility. In raising such
questions, Heidegger was also pointing to the fact that truth for
man is not only a matter of taking validity claims seriously.

Truth, to put it another way, is not only a matter of truth or
falsity, but goes to the heart of man’s Being. Or again: Man, in
having to speak the truth, does not only have to comport himself
as a responsible claimant to validity. What the truth demands of
man is that, apart from operating within a realm of the true and
the false, he also relates himself to the fact that he is related to
such a realm, that ‘there is’ such a realm, and that his dependence
on this ‘there is’ (es gibt) says something about his own Being. In
other words, Heidegger realised that what is at stake in the
history of truth is not only truth itself, but man’s own essance
(Wesen). And this is why his attempt to think that history in
terms of a forgetting of aletheia, of the essance of truth, or of
what makes the truth true, was always linked with the attempt to
restore man to his own Being by preparing a ‘revolution
[Umwal:ung] of human Being’.”

This fairly straightforward summary of the thrust of Heidegger’ s
writings on truth, and the way they link up with Foucault’ s
programme,12 aims to show that what is at stake here might not
be an attempt to let meaning ‘consume’ validity, but to think
their interrelation in a different way. In fact, as Habermas’ s
criticism of both Foucault and Heidegger makes clear, his contention is not merely that meaning and validity are internally
related, but that this interconnection also involves a symmetrical
or ‘dialectical’ relation. If Habermas’ s criticism of Foucault was
simply that Foucault has severed the internal connection between meaning and validity by reducing truth to power – a
criticism which, as we have seen, is certainly part of the argument – then it would suffice to repeat that Foucault does allow
for truth, that he even seeks to defend truth by investigating
discourse as its condition of possibility, and that his use of such
terms as ‘power’, ‘regime’ or ‘politics’ vis-a-vis truth cannot be
taken to suggest what Habermas reads into it. As soon as one
realises that there is no real opposition in Foucault between, on
the one hand, the idea that truth is linked to power in the sense of
a limitation, and, on the other hand, an emphasis upon the
productive character of power; as soon as one realises that
power, in order to be productive, has to be selective and exclusive, one has already seen through what Habermas calls
Foucault’s ‘systematically ambiguous’ use of the concept of
power – a concept Foucault is said to have forged by amalgamating ‘the idealist idea of transcendental synthesis with the presupposition of an empiricist ontology’ (PDM, pp. 274 and 270).

In Habermas’s reading of Foucault, the fact that ‘genealogical historiography is supposed to be both at once functionalist
social science and at the same time historical research into
constitutive conditions’ (PDM, p. 274) points to the ‘irritating
double role’ (PDM, p. 273) Foucault preserved for the category
of power: ‘on the one hand, it retains the innocence of a concept
used descriptively and serves the empirical analysis of power
technologies … [whereas] on the other hand, [it] preserves from
its covert historical sources the meaning of a basic concept
within a theory of constitution as well’ (PDM, p. 270). The same
genealogy that descriptively lays bare power relationships as
conditions for the rise of scientific knowledge and as its social
effects is, according to Habermas, simultaneously forced to play
‘the transcendental role of an analysis of technologies of power


that are meant to explain how scientific discourse about man is
possible at all’ (PDM, p. 274). Whereas Habermas may be right
in detecting a certain ambivalence in the way concepts such as
power/knowledge come to function in Foucault’s texts,13 he may
be too quick to conclude that only this ambivalence lends ‘the
empirical analysis of technologies of power their significance as
a critique of reason and secures for genealogical historiography
its unmasking effect’ (PDM, p. 270). For it is only because
Habermas already seems to know what power is about that he is
able to dismiss the relevance of such concepts as power/knowledge. Because Habermas is convinced that Foucault is attempting the impossible in ascribing transcendental capacities to the
kind of empirical power strategies we all are already familiar
with, he cannot see that Foucault is (or should be taken as)
questioning this very familiarity by investigating the empirical
bearings of an unfamiliar ‘power’: a ‘power’ which possesses a
constitutive function. What ‘Foucault’ really discovers behindor, rather, in – the empirical power-strategies his genealogy is
trying to unearth, is a necessary limitation, a ‘power’ that does
not merely have a negative function: a ‘power’ that produces
because it limits. 14 For example, without some limits on the true,
there would be no truth at all: if anything can be said, there is no
longer an order of truth, but a chaos. Far from making validity
impossible, ‘power’ in the sense of a necessary, constitutive
limitation, is what, according to ‘Foucault’, allows for the possibility of validity as such. Meaning – the realm of a limited
discourse, the regime of truth – and validity are internally connected: only within such a realm can there be statements that can
be taken seriously as validity claims. Only because the number
of candidates for truth or falsity is restricted in terms of conditions of ‘well-formedness’ can there exist truth or falsity.

Discourse, then, is a set of rules that imposes a basic homogeneity upon candidates for truth and falsity. It can only allow
for the true and the false by first imposing on them the realm of
the ‘true’. Clearly, this version ofJramework-relativism, which
does not let the truth of statements depend on a framework, but
only their truth orfalsity, their possible truth, cannot be taken to
exclude objectivity. For there must be some objectivity if one is
to decide within the framework between those statements that
are true and those that are false. As D. C. Hoy puts it: ‘the way
the world is may determine what is true or false, but that will still
not explain what is actually said, or comes up for counting as true
or false’. 15 And here we have probably reached the heart of
Habermas’s objection. For even were he willing to grant that
there is an internal connection between meaning and validity in
Foucault, he would still be unwilling to accept its terms. For,
according to Habermas, Foucault refuses to let the meaning of
his frameworks be judged by the ‘innerworldly success of the
practice [they] make possible’ (PDM, p. 154). In other words,
what Habermas really opposes is the idea that the conditions for
truth and falsity cannot themselves be judged in terms of truth
and falsity (e.g. PDM, p. 255). Only conceptions of a meaninghorizon allowing for practices which will ultimately be able to
determine the validity of such a horizon can do justice to the
internal and symmetrical relationship Habermas posits between
meaning and validity. And, since neither Heidegger nor Foucault
seem to be willing to promote such a conception, Habermas
tends to suggest that their theories not only deny the symmetrical
relation meaning/validity, but any connection whatsoever. Validity is consumed by meaning as soon as one neglects the fact
that ‘whether the validity conditions are in fact satisfied to such
an extent that the sentences can also function is not a matter of
the world-disclosing power of language, but of the innerworldly
success of the practice it makes possible’ (PDM, p. 154 – my

Radical Philosophy 61, Summer 1992

According to Habennas, this is precisely what is forgotten by
Foucault and Heidegger when they reserve ‘the title of truth for
[something] which no longer has anything to do with a validity
claim transcending space and time’ (PDM – p. 154 – my emphasis; compare p. 155). Instead of seeing that language has ‘to
prove itself’ through praxis (PDM, p. 335, and compare p. 154),
these authors are said to let ‘the “truth” of semantic worlddisclosure found the propositional truth of statements [and]
prejudice the validity of linguistic utterances’ (PDM, p. 331 my emphasis). They ‘hypostatise’ the ‘luminous force of worlddisclosing language’ and no longer feel the need to let it ‘prove
itself by its capacity to throw light on beings in the world’ (PDM,
p. 154). Heidegger, for example, allegedly ‘supposes that beings
can be opened up in their Being with equal ease by any given
approach’ (PDM, p. 154) and hence finds himself defending a
,super-foundationalism of a history of Being abstracted from all
concrete history’ (PDM, p. 104 – my emphasis). In fact, for
Habennas, this ‘abstraction from all concrete history’ – extravagant as this claim may be regarding a philosopher who devoted
all his efforts to think history – once more seems to concern the
alleged denial in Heidegger (or Foucault) that the ‘accumulation
of knowledge could affect the previous interpretation of the
world and burst a given totality of meaning’ (PDM, p. 331), i.e.,
with the ‘dialectical’ feed-back between meaning and validity:

‘As a result, intramundane praxis cannot get learning processes
going’ (PDM, p. 331).

Since it is ultimately the possibility of such learning processes, and the way to conceive them, that are in question, one
would expect Habennas to do more than implicitly refer to the
reader to the incorporation in his earlier work of a notion borrowed, inter alia, from Piaget’s ‘reconstructive science’. Suffice
it to point out that Habennas seems to think that learning processes, since they also concern the truth of frameworks, not only
operate within frameworks, but also allow one to proceed from
one framework to another in such a way that the gains of learning
compensate the dis-Iearning process involved: the truth of the
next framework somehow has to encompass, or be on a higher
‘developmental level’ than, the truth of the foregoing one (e.g.

PDM, p. 84), for otherwise one would once more be surrendering to what he calls ‘the imperative force of an illumination
compelling one to one’s knees’ (PDM, p. 255). Since, as far as I
can see, one cannot find a compelling reason in Habennas’ s
published writings to adopt this Piagetian theory and to preserve
it in the face of its critics, including those sympathetic with the

Radical Philosophy 61, Summer 1992

project,16 and since, on the other hand, a
discussion of the role and the position
of the ‘reconstructive sciences’ in
Habennas’s work far exceeds the scope
of this article, and would ultimately involve a full deconstruction of the whole
Habennasian project, I will have to restrict myself to a few remarks that return us to the relation between meaning
and validity.

The central question in the debate
over meaning and validity seems to be
whether what is claimed here (and elsewhere in Habennas’ s work), in the
name of a symmetrical relationship between meaning and validity, is not doing the converse: letting validity consume meaning by making meaning ultimately depend on validity. It should be
clear by now that one cannot even begin to investigate this problem, if one
lets oneself be ‘blackmailed’ by the implicit presuppositions in
Habennas’s criticism of Heidegger and Foucault. Given that
Habennas’s own analysis of the ‘symmetrical’ relation between
meaning and validity does not prevent him from placing it under
the aegis of validity, the fact that Heidegger and Foucault seem
to be defending a ‘non-symmetrical’ relation under the aegis of
meaning cannot constitute grounds for simply dismissing their
attempt. Nor can it be taken as a reason to conclude in their
favour. As long as the discussion is couched in the tenns
Habennas imposed on his adversaries, the only thing which
would pennit us to choose would be proof that, at all times and as
a matter of principle, crucial experiments for the assessment of
regimes of truth are possible; i.e., that it is always possible to
assess the conditions for candidacy to truth and falsity laid down
by such regimes in such a way that what are considered to be
‘relevant’ statements in a given regime could somehow be
shown to be inferior/superior to what is the case in another
But why should the discussion be these tenns?

What is really at stake when Habennas decides that the transition
between such regimes cannot merely be a matter of ‘conversion’

(Kuhn) to another regime or to another ‘style’ of scientific
reasoning (Hacking); that it has to involve knowing, or being
able to show, ex post factum, that what one learns in such a
conversion compensates for what one dis-Iearns in abandoning
another ‘style’ or ‘regime’? Why can’t we simply say that what
is deemed relevant has changed: Foucault’s happy positivism?

And why can’t we add, with Heidegger, that history is precisely
the name for the experience that we cannot account for everything in tenns of beings or of a highest being, but are ultimately
exposed to the sending of a destiny (Geschick) which is given to
us and never fully in our control?18 In other words: what hidden
anxiety has been setting the stage for this debate?

As we have seen, the crux of Habennas’ s debate with Heidegger
and Foucault – indeed, his motivation in taking it up – resides in
his fear that these philosophers will deprive us of ‘the transcendent moment of universal validity [that] bursts every provinciality
[including the provinciality of a framework, R.V.] asunder’

(PDM, p. 322). This seems to suggest that, although truth for
Habennas may be dependent on cultural practices, it can never
be immanent in such practices. Although a truth-claim will


always be made from within a provinciality, it can never satisfy
itself solely with immanence. Of necessity, truth has a transcendent moment: ‘though never divorced from social practices of
justification, from the rules and warrants of this or that culture
[discourse etc.], truth cannot be reduced to any particular set
thereof’ . 19
But the question is, of course, whether the transcendent
character one intuitively associates with truth – embedded as it
seems to be in our practices of telling the truth – allows us to
infer that comparisons between ‘frameworks’ along Habermas’ s
transcendental-pragmatist lines must eo ipso be possible. Might
it not be that truth’s transcendent character is something one only
experiences within a regime of truth, within a provinciality?

Something that points to the paradoxical intertwinement of truth
as an infinite task and ‘truth’ or discourse or aletheia as a finite
realm within the necessity of such a task will be experienced
always anew, but which cannot itself be judged in terms of true
or false? Can we not take truth as an infinite task seriously,
without knowing the truth of that which allows for the true is
itself true? Are Heidegger and Foucault indeed undermining the
very possibility of truth by claiming that the realm within which
truth/falsity becomes possible involves limitations which are
neither true nor false since they are what allows for truth and
falsity? Can truth still be the task it is, once we admit that the
situation-transcending import of truth-claims does not by itself
point to the possibility (or, indeed, the necessity) of transcending, in the name of truth, that which allows for the truth? Do we
need the transcendental-logical pragmatist reassurance that
some orders of discourse, some regimes of truth, some ‘truths’

are more true than others, in order to take truth seriously? Can’t
we speak the truth, while accepting that ultimately ‘truth’ is
finite, and that we will never have a more than finite relation to
the truth?

If ‘truth’, aletheia, as Heidegger contends, always involves
some ‘untruth’; if a lethe is ineradicably present at the heart of
aletheia; or if, as Foucault suggests, in order to speak the truth,
man has to come to terms with the fact that there are necessary
limitations on the realm he will enter; in short, if truth is itself
finite and yet at the same time an infinite task, then taking this
infinite task seriously only seems possible whenwe take our own
finitude seriously: what we have is an infinite task, but never a
non-finite truth. Indeed, what makes the task a task is precisely
the fact that, in order to have truth, there will always and of
necessity be something which escapes it: e.g., what Foucault
points to when he differentiates between the ‘truth’ (as the
regime of the true/false) and the truth.

Viewed from this angle, Habermas’ s suggestion that we
must at least be able to decide in retrospect about the truth of a
regime of truth, about the validity of a given horizon of meaning,
that we must at least have the possibility of a non-finite truth,
may well be another way of seeking to deny man’s finitude, yet
another way of thinking that finitude is something we can easily
come to terms with, if only by the assurance that we are making
progress toward the non-finite, gradually throwing light on the
lethe that obscures aletheia’s heart. For what is revealed by this
Habermasian desire for the truth to be fully true, so true that it
even encompasses the truth of its own conditions of possibility?

What is revealed by this desire informing Habermas’ s debate
with post-structuralism, if not that truth cannot simply be related
to some uncomplicated desire for the truth? What complicates
this desire for the truth is not, as Habermas seems to fear, that the
link between truth and desire would in and of itself threaten the
truth. Rather, if there seems to be something threatening about
the truth itself, if there seems to be something about the truth
which man finds hard to bear, this is because the truth, qua
intertwinement between a finite ‘truth’ and an infinite task,

confronts man with his own finitude and with his desire to escape
this finitude. That is why for Heidegger the fact that, like truth,
man is finite, turns finitude itself into a task demanding a certain
comportment of man. And that is why, as suggested earlier, what
is at stake in this debate between Habermas and Heidegger must
ultimately be related to the question of what it means for man to
speak the truth. This question is all the more important since
Habermas’s insistence on the symmetrical relation between
meaning and validity seems to obstruct his access to the kind of
ethics he really needs: an ethics that not only explains the
possibility of dialogue, but which can also account for the possibility of its absence. 2o Paradoxically, it seems to be the
Heideggerian concern with an ethics of finitude which is simultaneously an ethics of truth, that could help us out here.

Let me briefly try to clarify this by taking issue with the
Habermasian claim that it is only the context-transcendent character of our notion of truth ‘that keeps us from being locked into
what we happen to agree on at any particular time and place, that
opens us up to the alternative possibilities lodged in otherness
and difference that have been so effectively invoked by poststructuralist thinkers’ (McCarthy, p. 370 and PDM, pp. 322ff.).

That is, the claim that the truth of ‘post-structuralism’ presupposes the truth of universal pragmatics – a claim which, as the
title of Habermas’s penultimate chapter indicates (PDM, pp.

294ff.: ‘An Alternative Way out of the Philosophy of the Subject: Communicative versus Subject-Centered Reason’), explicitly organises the argumentative architecture of The Philosophical Discourse of Modernity.

For Habermas, Foucault’s and Heidegger’s refusal to turn the
relation between meaning and validity into a symmetrical one,
where meaning would ultimately be subjected to a proof of its
validity, not only constitutes a threat to the meaningfulness of
the project of truth, but also to the very possibility of dialogue
and critique. Dialogue can only be what it is, can only find its
motive and potential, in the transcendent character of the truthclaims we make. As McCarthy observes, ‘without that idealizing
moment, there would be no foothold in our accepted beliefs and
practices for the critical shocks to consensus that force us to
expand our horizons and learn to see things in different ways’

(McCarthy, p. 370).

Is this so? Does the possibility of dialogue depend on the link .

McCarthy/Habermas are making here between their views on
truth-transcendence, on meaning and validity, and on universality? Do we have to look for a non-finite truth that encompasses
all finite orders of truth in order to save a concept of universality
and, with it, the possibility of dialogue? What if we followed
Merleau-Ponty in substituting for such a ‘universality from
above’ a universality that would have to be conceived in a
‘lateral’ way?21 What if it were this kind of universality that
makes the true dialogue possible – not the transcendence of truth
that breaks all provinciality asunder and opens me up for the
other by also breaking his provinciality asunder, but the experience that, for all its transcendence, my truth-claim seems to be as
inextricably bound up with my provinciality (order or regime of
truth; aletheia), as the truth-claim of the other is with his? In
other words, what opens us up to ‘the alternative possibilities
lodged in otherness and difference’ mentioned by McCarthy (p.

370), might not be the situation-transcending character of the
truth-claims as such, but the fact that this transcendence is tied to
an immanence which at first sight closes us off from, estranges
us from the other. In the light of a certain ethics of truth – the
kind of ethics at the centre of Heidegger’s concern with the truth
– it is not the experience of a possible common identity (univerRadical Philosophy 61, Summer 1992



sality from above), but the experience of my own identity which
I cannot share with the other, that can throw a bridge between us
(lateral universality). In other words, what we may be in need of
is not, as Habermas thinks, ‘an alternative way out ofthe philosophy of the subject’ (PDM, title of Chapter XI and p. 301), but an
alternative way into it. For, after all, would we need a dialogue,
or an ethics of dialogue, if we were simply decentred, always
already – de jure, if not d..e facto – intersubjective?

Ironically, perhaps, the philosophy that feels obliged to protest against those who are sacrificing validity on the altar of
meaning may well find itself sacrificing subjectivity on the altar
of intersubjectivity, oblivious of the fact that it thereby hollows
out both concepts. Refusing to let meaning be consumed by
validity, Heidegger, on the other hand, seems to be in a far better
position to take Habermas’s problem seriously and to ground the
possibility of dialogue in an ethics of ‘truth’. For Heidegger, the
subject of truth is not decentred because it has no centre but the
one it shares with others; it is decentred because it has a centre
that is not that of the other. That is why finitude confronts us with
the difficult task of discovering that at the centre of one’s own
relation to truth there lies a non-universalisable core, a moment
of non-universality, from which we cannot and should not seek
to escape, and which makes both truth and dialogue the infinite
task we know they are. Having a centre which it may neither
abandon, nor comfortably nestle itself into, the subject of truth
which makes this finitude his own bursts open towards the other,
by discovering the wound with which its ‘truth’ had always
already afflicted it. At that moment tolerance is born, as we all
are: in a cry of pain. No longer a marginal virtue, but ‘the point at
which recognition of our human condition can begin’,22 it will
have to carry the weight of an ontological function.


Michel Foucault, ‘Truth and Power’, in idem, Power/Knowledge, Selected Interviews and Other Writings (1972-77), New
York: Pantheon, 1980.


I cannot repeat here the attempt to demonstrate the extent to
which the traditional reception of Foucault has been misled by
the ‘face-value’ of his texts (see my Michel F oucault. Genealogie
als Kritik, Miinchen: Fink, 1991).


Martin Heidegger, Vorn Wesen der Wahrheit. Zu Platons
Hohlengleichnis und Theiitet (Winter tenn 1931-32), GA 34, p.


For a more detailed analysis see my ‘From Foucaultto Heidegger.

A One Way Ticket?’, forthcoming in Research in Phenomenology.

I have tried to argue this in a number of places: ‘Can Genealogy
Be Critical? A Somewhat Unromantic Look at Nietzsche and
Foucault’ ,inManandWorld,23: 4, 1990, pp. 441-52; ‘Foucaults
Anfiihrungszeichen’, in B. Waldenfels and F. Ewald (eds),
Spiele der Wahrheit. Michel F oucaults Denken, Frankfurt a.M.:

Suhrkamp, 1991, pp. 298-319.

Since the tenn ‘power’ assumes a specific meaning here, the
ultimate mistake would be to oppose power-strategies because
they involve ‘power’. The problem is to look for a model of
critique that does not mix up the constitutive and the contingent
(as a certain ‘Foucault’ – this much should be admitted – found
himself doing) (see notes 10, 12 and 13).

D. C. Hoy, ‘Taking History Seriously: Foucault, Gadamer,
Habennas’, in Union Seminary Quarterly Review, 34: 2, 1979,

Thomas McCarthy, ‘Rationality and Relativism: Habennas’s
“Overcoming” of Henneneutics’, in J. B. Thompson and D.

Held (eds), Habermas. Critical Debates, London/Basingstoke:

MacMillan, 1982, pp. 57-78; B. Waldenfels, In Den Netzen der
Lebenswelt, Frankfurt a.M.: Suhrkamp, 1985, pp. 110-17; K.

Meyer-Drawe, ‘Zahmung eines wilden Denkens? Piaget und
Merleau-Ponty zur Entwicklung von Rationalitat’ ,in A. Metraux
and B. Waldenfels (eds), Leibhaftige Vernunft. Spuren von
Merleau-Ponty’s Denken, Miinchen: Fink, 1986, pp. 258-75.

No weaker version of a decisive test could do here, since, as
suggested above, authors like Heidegger and Foucault are framework-relativists in the sense that they think that what changes
between frameworks is not simply the stock of attainable truths
(in which case comparisons of scope would suffice), but also
what it means to attain truth (Heidegger) or to be ‘in the true’

(Foucault). The strange thing about Paracelsus, for example, is
not simply that he was serious about a lot of things we cannot
possibly be serious about, but that what it meant for him to be
serious about ‘truth’ seems to be something different from what
it means to us. This is why a counter-argument to the kind of
incommensurability at stake here (Hacking’s second type, discussed as ‘dissociation’ in his Representing and Intervening,
Cambridge: CUP, 1990, pp. 69-72) cannot rest with the attempt
to show that all or part of a framework’s empirical propositions
have some bearing on its ‘core propositions’, and that there is
therefore room for a ‘dialectical feedback’ between the two
(cognitive pressure feedbacking in such a way that the framework itself would change). Rather, since a framework is not
primarily defined by a set of core propositions, but by a ‘style of
reasoning’ (Hacking), including a certain view of what it means
to know, to argue seriously, to be relevant, to be ‘in the true’, the
attempt to refute this position should include a way of showing
that from one ‘framework’ to another, there is not simply
progress in knowing, but also in what it means to know, to be ‘in
the true’ , etc. Habennas, of course, is well aware that he needs
to come up with some kind of solution to this Hegelian problem
and that is why he has been trying to use (and extend) the results
of a Kohlberg-Piaget type of research to build in a diachronic
moment into the framework of his universal pragmatics.

E.g., Martin Heidegger, ‘The Question Concerning Technology’, in idem, Basic Writings, New York: Harper & Row, 1976,
p. 299: ‘Man can, indeed, conceive, fashion, and carry through







Jiirgen Habennas, The Philosophical Discourse of Modernity.

Twelve Lectures, Cambridge: Polity, 1987, p. 320 (translation
altered); henceforth: PDM.


Jacques Derrida, Memoires pour Paul de Man, Paris: Galilee,
1988, p. 225 (my translation).


Jacques Derrida, Limited Inc., Evanston: Northwestern UniversityPress, 1988,p.157 (Derridais quoting fromPDMp. 193; the
emphasis is his own).


See, inter alia, the excellent commentary byB. Flynn in the fifth
chapter of his forthcoming Political Philosophy at the Closure
of Metaphysics and J. M. Bernstein’s insightful discussion of
Habennas’s failure’ to recognize the philosophical discourses of
modernity [Foucault, Adorno, Derrida] as the philosophical
expression of artistic modernism’ (‘Frankfurter and French
Fries: Between Modernity and Modernism’ (review ofPDM) in
Art History, 11: 4, 1988, pp. 586-90 (quotation at p. 588).


Jiirgen Habermas, ‘Nachwort’, in Friedrich Nietzsche,
Erkenntnistheoretische Schriften, Frankfurt a.M.: Suhrkamp,
1968, p. 257.

The expression was omitted in the English translation (PDM, pp.

275-76 – see the Gennan original, p. 326).


For a systematic exposition of ‘Meaning and Validity’, see
Jiirgen Habennas, The Theory ofCommunicative Action, vol. 1,
Boston: Beacon Press, 1984, pp. 295ff.


See, e.g., PDM, pp. 346-47: ‘as a participant in discourses, the
individual, with his irreplaceable yes or no, is only fully on his
own under the presupposition that he remains bound to a universal community by way of a cooperative quest for truth.’ I have
emphasised the tenns I will analyse on a different occasion when
focusing on the role the alternative ‘yes or no’ is forced to play
in Habennas’ s universal pragmatics.

Radical Philosophy 61, Summer 1992





this or that in one way or another. But man does not have control
over unconcealment itself, in which at any given time the real
shows itself or withdraws. ‘




Thomas McCarthy, ‘Private Irony and Public Decency: Richard
Rorty’s New Pragmatism’, in Critical Inquiry, 16, 1990, pp.

369-70, and compare PDM, p. 323.

Whereas Habermas tries to ground the possibility of dialogue in
an ethics oflanguage, he accounts for the absence of dialogue not
in ethical but in political terms (systematically distorted communication). I suspect that ultimately this dissymmetry in his
analysis will prevent him from taking such distortions seriously.

Merleau-Ponty introduces the counter-concept of a ‘lateral
universality’ (as against an ‘overarching universality’) in the
course of his essay ‘From Mauss to Claude Levi-Strauss’, the
problem being precisely ‘how [to] understand the other without


sacrificing him to our logic or it to him’ – see Signs, Evanston:

Northwestern University Press, 1964, p. 115 (quote) and 120 Ca
second way to the universal ‘). I will develop my own interpretation of these terms below.

Ernesto Laclau, New Reflections on the Revolution ofOur Time,
London/New York: Verso, 1990, p. xiv.

This is the text of a lecture given in May 1991 to the Philosophy
Department at the University ofWarwick and to the Centre for Theoretical Studies at the University of Essex, where I enjoyed the hospitality of
the Philosophy Department as a Visiting Fellow during the spring and
summer terms of 1991. I am grateful to Jay Bernsteinfor his comments
on an earlier draft, though my alterations will doubtless fail to satisfy all
his criticisms.




St Catherine’s College, Oxford, 24-26 July 1992

Friday pm:

Roy Bhaskar on critical realism and dialectic


Alex Callinicos on the revolutions in Eastern Europe
Mike Haynes on the market and Eatern Europe
Gregory Elliott on ‘The End of Social Democracy?’

John Lovering on ‘A Post-Modern Arms Industry’

Sue Clegg on explaining child abuse
Tony Lawson on the critique of contemporary economic theory
Anthony Arblaster on politics and opera


Kate Soper on ‘Realism and rhetoric in the discourse on nature’

William Outhwaite on ‘Habermas, modernity and realism’

Jim Kinkaid on ‘Marx and Logic’

Rick Marsden and Phil Sharpe on realism and classical Meaxism
Francis Roberts on closure and the human sciences
Margaret FitzSimmons and Ulker Semen on realism, space and social theory

Some five other meetings are still being arranged including meetings on chaos theory and on
explanations in contemporary physics.

For further information contact: Rob Hoveman
45 Castlemaine Avenue, South Croyden, Surrey CR2 7HW


Radical Philosophy 61, Summer 1992

Download the PDFBuy the latest issue