boundary 2
“Correlationism”: The Dogma that Never Was
David Golumbia
Few theoretical movements have developed as quickly or as publicly as the one known, among other names, as “Speculative Realism”
and/or “Object-Oriented Philosophy” (hereafter collectively referred to as
SR/OOO). The views of each of the main writers associated with these
movements—Graham Harman, Ian Bogost, Iain Hamilton Grant, Levi
Bryant, and Timothy Morton, among others—vary widely, and even the
views offered by these writers individually often vary. What they share in
general is a conviction that, as the editors of one of the movement’s primary
collections, The Speculative Turn, put it, “reality appears in [recent continental] philosophy only as the correlate of human thought,” and that what
is needed now in contrast is for theory to “speculat[e] once more about the
nature of reality independently of thought and of humanity more generally.”1
I appreciate helpful comments on earlier versions of this article from David Berry, Terence
Blake, Lauren Boasso, Paul Bové, Suzanne Daly, Alexander Galloway, Gayatri Chakravorty Spivak, the audience at the 2011 “Radical Philosophy” conference at Columbia University, and the anonymous reviewers for boundary 2.
1. Levi Bryant, Nick Srnicek, and Graham Harman, “Towards a Speculative Philosoboundary 2 43:2 (2016)
DOI 10.1215/01903659-3469889
© 2016 by Duke University Press
Published by Duke University Press
boundary 2
2
boundary 2 / May 2016
Almost exclusively in the SR/OOO literature, support for this view
is attributed to a single brief book by a French student of Alain Badiou,
Quentin Meillassoux, titled After Finitude: An Essay on the Necessity of
Contingency. The book appeared in France in 2006 and was published in
Ray Brassier’s English translation in 2008. In After Finitude, Meillassoux
introduces the terms correlation and correlationism to advance the argument that the past 250 years of Western philosophy and critical and literary
theory, or what he typically calls “post-Kantian philosophy,” is contaminated
by a deep, thoroughgoing, and almost entirely unrecognized error, an error
so profound that philosophy must essentially begin again. Meillassoux’s
book is so important to SR/OOO that Harman, one of SR/OOO’s most
active proponents, quickly issued a monograph solely devoted to Meillassoux, Quentin Meillassoux: Philosophy in the Making, that focused almost
entirely on After Finitude; at the time neither Meillassoux’s 1997 dissertation, “L’inexistence Divine” (supervised by Badiou), nor his numerological reading of Stéphane Mallarmé, The Number and the Siren, had been
published.2
In some (though not all) ways After Finitude reads like a distillation
of Badiou’s work, and it often takes as proven facts what are actually argumentative propositions in Badiou.3 This is only part of what makes After
Finitude look very strange to readers with an Anglo-American philosophical background: it does not follow typical procedures of philosophical argumentation, and its take on philosophical history, including recent history, is
beyond idiosyncratic—it is a picture of philosophy that no notable figure but
Badiou has ever previously put forth.4 This would be less troubling if After
phy,” in The Speculative Turn: Continental Materialism and Realism, ed. Levi Bryant,
Nick Srnicek, and Graham Harman (Melbourne: re.press, 2011), 3.
2. Graham Harman, Quentin Meillassoux: Philosophy in the Making (Edinburgh: Edinburgh University Press, 2011); the Mallarmé book has been published in English as The
Number and the Siren: A Decipherment of Mallarmé’s “Coup de dés” (Falmouth, UK:
Urbanomic, 2012).
3. Adrian Johnston, in “Quentin Meillassoux and Alain Badiou’s Anti-Kantian Transcendentalism,” Contemporary French Civilization 33, no. 1 (2009): 73–99, offers an especially
thorough analysis of the relationship between the two writers.
4. In the same year that After Finitude appeared in France, Tom Rockmore’s In Kant’s
Wake: Philosophy in the Twentieth Century (Malden, MA: Blackwell, 2006) was also
published. Rockmore’s carefully documented, knowledgeable, and clearly argued work
ranges over much of the same conceptual material as does After Finitude, but Rockmore’s conclusions are both far more nuanced than and often almost completely incompatible with those of Meillassoux.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
3
Finitude did not purport to be a systematic critique of all philosophy since
Immanuel Kant. Among its most troubling characteristics in this regard are
its almost total lack of reference to secondary and interpretive writings on
Kant, and to contemporary philosophy in general, because Meillassoux
repeatedly tries to show that Kant’s influence has made contemporary philosophy correlationist. Instead of demonstrating that flaw in contemporary
philosophy, though, Meillassoux generally (with very few specifics) argues
that Kant himself is correlationist and that this foundational taint itself contaminates what comes after. Meillassoux thus purports to discover something unseen in Kant, and then to claim that this unseen thing has been the
dominant influence over all philosophy since Kant, while making no effort
to show how correlationism can at one and the same time have been so
strongly determinative and also largely unnoticed.
In fact, the highly varied commentary on Kant, much of which focuses
on the same questions that interest Meillassoux, must make any reader
question the brunt of Meillassoux’s attack: does all this work, riven with
serious disagreements about the nature of Kant’s project, somehow uniformly participate in a subtle error that so many philosophers have failed to
appreciate? For that matter, given the enormous differences in philosophical outlook of such figures as Bertrand Russell, Gottlob Frege, A. J. Ayer,
Wilfrid Sellars, Rudolf Carnap, Ludwig Wittgenstein (early and late), W. V.
Quine, Donald Davidson, Hilary Putnam, Michael Dummett, Saul Kripke,
David Armstrong, David Lewis, Thomas Nagel, Jerry Fodor, Paul Guyer,
and Bas van Fraassen—just to name some of the most prominent analytic
philosophers whose work touches on issues raised by Meillassoux—how
can one take seriously the claim that they are all correlationist, especially
where the only attempt to demonstrate this error is made by analyzing not
their own works but positions (and not even, for the most part, actual writings) of Kant?
Despite the widespread impression that Meillassoux has conclusively deflated most recent philosophy, in recent work his own version of
the claim has narrowed, so that he now speaks of “Continental philosophy,”
rather than philosophy tout court, as correlationist, but still without explaining (as several of his commentators have urged) why the detailed analytical discussion of the questions he raises can or should be dismissed out
of hand.5 What is most worrisome about this from an institutional perspec5. In “Iteration, Reiteration, Repetition: A Speculative Analysis of the Meaningless Sign,”
a talk delivered at the Free University in Berlin, April 20, 2012 (trans. Robin Mackay,
oursecretblog.com/txt/QMpaperApr12.pdf), Meillassoux appears to silently withdraw
Published by Duke University Press
boundary 2
4
boundary 2 / May 2016
tive is that in his reception in the English-speaking world, Meillassoux’s
work has been taken to show that Anglo-American analytic philosophy is
also so tainted by correlationism as not to be worth taking seriously, in part
because Meillassoux explicitly makes this claim. What jars the philosophically informed reader is that the issues thereby raised turn out to be central
ones in disciplinary philosophical discourse, which Meillassoux is understood to have eliminated the need to address. What Meillassoux’s work
appears to do is therefore to license a kind of philosophy ex nihilo and without connection to existing practice, and what it thus engenders is a philosophical discourse that repeatedly fails to respect most of the methods of
that practice: to state clearly its contentions, to define its terms, to distinguish between philosophical issues (particularly epistemology and metaphysics), or to demonstrate textually its historical-philosophical assessments. As such, it creates a spectral philosophical orthodoxy that is largely
incoherent and that few if any practicing philosophers actually endorse; for
not only is analytic philosophy not correlationist insofar as we can make
sense of what Meillassoux says about it, but the issues Meillassoux claims
have been overlooked in post-Kantian philosophy are instead often the
main topics of philosophical discourse.
In significant ways, especially as a discourse, the entire SR/OOO
enterprise is built on the single support of Meillassoux’s “critique of correlationism,” but despite this reliance, Meillassoux’s critique fails to establish
its basic premises, and fails and will have to fail to demonstrate the accuracy of its interpretation of post-Kantian philosophy, since it is manifestly
out of step with that philosophy. This is not, necessarily, to discount SR/
OOO as an enterprise, let alone some of its individual exemplars, since it is
not always clear in philosophical or theoretical terms how much SR/OOO
the condemnation of analytic philosophy in After Finitude with which this essay is concerned: “What we might call ‘the era of Correlation’ . . . continues, in my opinion, to
dominate continental philosophy today” (3; emphasis added). Since his complaint is
pitched against “the various destructions and deconstructions that all traditional metaphysics have undergone over the last century and a half” (1), his failure to engage with
the huge body of analytic metaphysics becomes all the more pointed. Commentators
on Meillassoux have noted his remarkable lack of engagement with philosophy of science; see Adrian Johnston, “Hume’s Revenge: A Dieu, Meillassoux?,” in Bryant, Srnicek,
and Harman, Speculative Turn, 92–113; Fabio Gironi, “Meillassoux’s Speculative Philosophy of Science: Contingency and Mathematics,” Pli: The Warwick Journal of Philosophy 22 (2011): 25–60; and Christian Thorne, “Outward Bound: On Meillassoux’s
After Finitude,” November 23, 2011, http://sites.williams.edu/cthorne/articles/outward
-bound-on-quentin-meillassouxs-after-finitude/.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
5
analyses require the “critique of correlationism” to ground themselves.
Whatever the merits of SR/OOO, it cannot rest on the “critique of correlationism,” because “correlationism” in the terms Meillassoux describes it
is in large part philosophically incoherent and, to the degree that it can
be made clear, results in a demonstrably inaccurate description of recent
philosophical practice—itself an ironic state of affairs, given Meillassoux’s
desire to get us closer to accurate descriptions of things in the world.
Correlationism
Meillassoux writes that “the central notion of modern philosophy
since Kant seems to be that of correlation” (AF, 5).6 By “correlation,” Meillassoux means “the idea according to which we only ever have access to
the correlation between thinking and being, and never to either term considered apart from the other.” Correlationism, then, “consists in disqualifying
the claim that it is possible to consider the realms of subjectivity and objectivity independently of each other. Not only does it become possible to insist
that we never grasp an object ‘in itself,’ in isolation from its relation to the
subject, but it also becomes necessary to maintain that we can never grasp
a subject that would not always-already be related to an object” (AF, 5). The
loose phrasing and lack of definition in these statements exemplify a characteristic problem for Meillassoux, wherein overtly metaphysical claims are
embedded in epistemological or cognitive statements, statements about
what “we only ever have access to” rather than about what there is, and
as if Meillassoux will offer a perspective that does consider them separately, which he is repeatedly unwilling or unable to do.7 Also characteristic
6. Quentin Meillassoux, After Finitude: An Essay on the Necessity of Contingency, trans.
Ray Brassier (New York: Continuum, 2008); hereafter cited parenthetically as AF. Meillassoux frequently emphasizes words and phrases in his text; unless indicated otherwise,
any emphases in quotations are those found in the original. Where useful for clarity, I have
occasionally cited the French original, Quentin Meillassoux, Après la finitude (Paris: Seuil,
2006); hereafter cited parenthetically as AlF.
7. In a fascinating exchange after Meillassoux presented some of the arguments in After
Finitude after its publication, Peter Hallward points out this characteristic ambiguity in
Meillassoux’s work: “It seems to me that you shuttle between an ontological argument
that you associate with metaphysics . . . and use that to demolish what are essentially
epistemological arguments that underpin the correlationist post-Kantian position.” Meillassoux admits to the criticism without acknowledging the problem it creates for his position: “Correlationism—you’re right—is not an ontology, strictly speaking. The correlationist—it’s true—doesn’t say that reality is the correlation. It’s the metaphysics of subjectivity
Published by Duke University Press
boundary 2
6
boundary 2 / May 2016
is the unsupported institutional claim that philosophers “never consider”
either thinking or being apart from each other, when the entire disciplines
of philosophy of mind and cognitive science, to take two obvious examples,
explicitly consider thinking separately from any consideration of being. The
converse project, to think being without considering thinking, is made logically and conceptually problematic by the same issue that dogs Meillassoux, namely, the unavoidable use of cognition in any statement of it: how
do we think being apart from thinking, without thinking as we do it? It will
turn out that, despite his reliance on such statements as basic definitions
of correlationism, Meillassoux never crafts propositions that avoid the very
problems he claims to have identified, or in other words, to make coherent
or legible the problems he takes to be foundational.
Such characteristic vagueness in the statement of what should be
central claims are reflected in the odd syntax Meillassoux uses in his definition of correlationism (also present in the original: “Le corrélationisme consiste à disqualifier toute prétention à considérer . . .” [AlF, 18–19]), in which
correlationism turns out not to be a claim or view or even the “notion” Meillassoux claims elsewhere but an attitude or disposition that “consists in disqualifying” another claim, in this case the claim that “we can consider the
realms of subjectivity and objectivity independently of each other,” which
this assertion notwithstanding is clearly something philosophers including
Kant frequently do, to the extent that we can make philosophical sense of
the phrase. It is typical of Meillassoux’s writing that at critical moments,
just where precise statement of terms is vital, precision is instead swept
aside, as in the case of the “it is possible/becomes possible to consider”
in this statement. Possible for whom? In what domain? This is loose talk of
a sort most philosophers work hard to avoid, talk that specifically yokes in
a knowing subject as it overtly critiques others for failing to be clear about
their epistemological (or metaphysical) commitments. What does it mean
for something “to become possible” in the sense Meillassoux is using the
expression here?
While Meillassoux paints correlationism as incompatible with realism
(at least with what he calls, without bibliographic reference, naive realism),
he repeatedly assigns to correlationism an explicit rejection of the existence of the material world (though he seems not to notice that this itself
that says that. He just says we cannot know anything apart from what we can perceive or
conceive, etc. That’s all. I refuse to say, on the contrary, that I can’t say anything about the
absolute.” From “Presentation by Quentin Meillassoux,” in Collapse III: Unknown Deleuze
(Falmouth, UK: Urbanomic, 2007), 444–45.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
7
logically entails some “consideration” of that material world even in the act
of rejecting its existence, something that he claims correlationism rules
out). Along with his invocations of Kant and George Berkeley (and “postmodernism”) as enemies, this makes the distance between correlationism
and idealism hard to determine. Thus, Meillassoux writes that under his
analysis, “every variety of correlationism is exposed as an extreme idealism, one that is incapable of admitting that what science tells us about
these occurrences of matter independent of humanity effectively occurred
as described by science” (AF, 18).8 In a talk delivered after the publication
of After Finitude, Meillassoux appears to admit that he introduced the term
“correlationism” for a practical and seemingly illegitimate scholarly reason:
to avoid having to engage with the voluminous anti-idealist philosophical
literature that would make his assessments less tractable: “I have given
the name ‘correlationism’ to the contemporary opponent of any realism. By
this term, I wanted to avoid the usual ‘parade’ of transcendental philosophy
and phenomenology against the accusation of idealism—I mean answers
such as: ‘Kantian criticism is not a subjective idealism since there is a refutation of idealism in the Critique of Pure Reason.’”9 It is hard to know how
readers should take such sentiments. At best, Meillassoux seems openly
to admit that “correlationism” is just a new word for idealism, a doctrine to
which few if any philosophers subscribe today or, with the exception of the
“German idealists,” have subscribed to since Kant, and one to which he
here acknowledges Kant offers direct counterarguments in the first Critique
itself. Further, it is impossible to reconcile the identification of “the contemporary opponent of any realism” with the detailed and unattributed presentation in After Finitude of correlationism’s epistemological commitments,
from which many if not most opponents of doctrinal realism explicitly demur.
Varieties of Correlationism
For readers of Kant and the extensive commentaries on him, what is
especially jarring about After Finitude’s anti-Kantian rhetoric is that Meillas8. “Matter independent of humanity” is part of Meillassoux’s “ancestrality” or “archefossil” thought experiment, flawed for many of the same reasons that his correlationism diagnosis is flawed. Johnston, in “Hume’s Revenge,” and Peter Hallward, in “Anything Is Possible: A Reading of Quentin Meillassoux’s After Finitude,” in Bryant, Srnicek,
and Harman, Speculative Turn, 130–41, address most of the problems with the thought
experiment.
9. Meillassoux, “Presentation,” 408.
Published by Duke University Press
boundary 2
8
boundary 2 / May 2016
soux’s claims are so often directly contradicted by Kant’s own plain statements in the first Critique (including, as Meillassoux admits in the 2007
“Presentation,” the “Refutation of Idealism” in version B of the first Critique)
and elsewhere.10 Kant is far more careful in his discussions of the epistemic capabilities of the human mind than Meillassoux wants his readers to
believe, and that care forces Kant’s readers to think in a more fine-grained
way than Meillassoux allows about just what the world of objects might be
to which humans may have various kinds of access. The scholarly literature
on Kant is replete with works that specifically reject the idea that Kant is an
idealist, and many works stress not just the compatibility of Kant’s work with
empirical science, but the very empirical basis of the critical philosophy in
scientific thinking.11 Students of Kant must be puzzled by the arguments in
the first chapter of After Finitude given Kant’s repeated rejection of idealism, in famous passages such as this one from the preface to the second
edition of the first Critique:
We cannot have cognition of any object as thing in itself, but can
have such cognition only insofar as the object is one of sensible intuition, i.e., an appearance. And from this it does indeed follow that
any possible speculative cognition of reason is restricted to mere
objects of experience. On the other hand, it must be noted carefully
that this is always subject to this reservation: that we must be able
at least to think, even if not cognize, the same objects also as things
in themselves. (Bxxvi–xxvii)12
10. Because Meillassoux so frequently invokes the incompatibility of science and correlationism, of particular relevance is Kant’s 1786 Metaphysical Foundations of Natural Science, a work that is widely understood in Kant scholarship to show just how compatible
are Kantian philosophy and science. See in particular Michela Massimi, ed., “Kant and
Philosophy of Science Today,” Royal Institute of Philosophy Supplement 63 (2008); and
Robert E. Butts, ed., Kant’s Philosophy of Physical Science: Metaphysische Anfangsgründe der Naturwissenschaft, 1786–1986 (Dordrecht, the Netherlands: D. Reidel, 1986).
11. Close readings of Kant that paint a very different picture from Meillassoux’s include
Paul Abela, Kant’s Empirical Realism (New York: Oxford University Press, 2002);
Kenneth R. Westphal, Kant’s Transcendental Proof of Realism (New York: Cambridge
University Press, 2004); and Michael Friedman, Kant and the Exact Sciences (Cambridge: Harvard University Press, 1992). In Kantian Humility: Our Ignorance of Things in
Themselves (New York: Oxford University Press, 1998), Rae Langton offers a carefully
argued, virtual point-by-point refutation of Meillassoux’s interpretation of Kant avant la
lettre, thick with textual detail.
12. Immanuel Kant, Critique of Pure Reason, trans. Werner S. Pluhar (Indianapolis:
Hackett, 1996). All passages from the Critique are taken from this dual-edition version and
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
9
Kant goes out of his way to account for and even to embrace human
thinking about the noumena even as he is careful to restrict knowledge to
that of which we humans have experience. If we mean to contradict this
idea we must make the outlandish assertion that we have knowledge of
what lies outside of experience (knowledge here meaning something like
certainty ); such knowledge is traditionally referred to as religious faith, and
it is beyond peculiar for Meillassoux to assert that this sort of relation is a
secularization or empiricization of Kant’s thought, because it is resolutely
clear how secular and empiricist Kant means to be.13 Meillassoux thus puts
himself in the unfortunate position of needing to insist that human beings
can know with certainty objects of experience (i.e., not a priori results) of
which they have no experience whatever (neither direct nor indirect, as via
scientific measuring instruments), and that this knowledge is somehow
more scientific than what current philosophy allows.
Throughout After Finitude, Meillassoux also fails to acknowledge or
discuss Kant’s terminology, and this is nowhere more notable than in the
failure to grant the distinction vital to Kant, as in the foregoing passage,
between thinking about something and knowing something or, in modern
terminology, between cognition and knowledge.14 Meillassoux thus often
directly contradicts Kant: “What we have just claimed is that thought is
capable of discriminating between those properties of the world which are
a function of our relation to it, and those properties of the world as it is ‘in
itself,’ subsisting indifferently of our relation to it. But we all know that such
a thesis has become indefensible, and this not only since Kant, but even
since Berkeley” (AF, 3–4). We have already seen that Kant goes out of his
way to assert that we can think about the thing-in-itself; what Kant denies
is that we can know it the same way we know the objects of our experience.
In fact the human propensity to think about what it cannot know has long
been understood as one of the wellsprings of Kant’s project in general, and
follow the traditional practice of citing the editions respectively as A (1781) and B (1787)
with corresponding page numbers.
13. It is hard to escape the feeling that Meillassoux very much wants to access a kind of
religious faith, as evidenced in his dissertation, “L’inexistence divine: Essai sur le dieu virtuel,” and in his frequent references to God and to a “virtual God” in After Finitude. On the
apparent religious dimension of Meillassoux’s thought, see Johnston, “Hume’s Revenge”;
and Martin Hägglund, “Radical Atheist Materialism: A Critique of Meillassoux,” in Bryant,
Srnicek, and Harman, Speculative Turn, 114–29.
14. See, for example, Rudolf A. Makkreel, “The Cognition-Knowledge Distinction in Kant
and Dilthey and the Implications for Psychology and Self-Understanding,” Studies in History and Philosophy of Science 34 (2003): 149–64.
Published by Duke University Press
boundary 2
10
boundary 2 / May 2016
the thought that we are not capable of discriminating between primary and
secondary qualities as a matter of thought is, overtly, anathema to Kant.
Perhaps the most startling fact about After Finitude is that despite
the overt claims in its first chapter to have identified the correlationist error
in Kant and post-Kantian philosophy, in the second and third of the book’s
five chapters Meillassoux admits that Kant was not a correlationist, relying
on some of those same features of the first Critique he overlooks in After
Finitude’s first chapter. Meillassoux accomplishes this by introducing in
After Finitude’s second chapter a distinction between “strong” and “weak”
correlationism. Kant, he now argues, is a “weak” correlationist, because
“the Critical philosophy does not prohibit all relation between thought and
the absolute. It proscribes any knowledge of the thing-in-itself (any application of the categories to the supersensible), but maintains the thinkability of the in-itself. According to Kant, we know a priori that the thing in
itself is non-contradictory and that it actually exists” (AF, 35). What is especially striking about this admission is that it precisely uses the language
with which Meillassoux before and after this remark defines correlationism,
while explicitly stating that it does not apply to Kant. Again and again, both
in Meillassoux and in SR/OOO, it is the prohibition on the relation between
thought (not knowledge) and the absolute that constitutes correlationism;
yet as any cursory reading of Kant shows, Kant observes no such prohibition, and indeed the idea that Kant would write so often about the thing-initself while denying that he or his readers could think about it is plainly nonsensical. Meillassoux never explains how one flavor of correlationism can
reject what he elsewhere names as its core tenet. Yet Meillassoux’s attacks
from this point of After Finitude forward take “strong” correlationism as the
operant form of the doctrine, even when, as in the book’s fourth and fifth
chapters, Kant returns as their main critical target.
“Strong” correlationism is an even murkier doctrine than the “weak”
variety (or correlationism sans modifier), and Meillassoux is even more evasive about naming practicing strong correlationists than about naming correlationists in general, since he thinks that those who subscribe to correlationism rarely if ever admit to doing so: the “strong model . . . seems to
be dominant today, even if it is never explicitly thematized as such” (AF,
30). “Strong correlationism is not always thematized as such by those who
espouse it” (AF, 43). That he has in mind a very familiar and tired critique
of literary theory and in particular poststructuralism as it is practiced by the
likes of Jacques Derrida, Michel Foucault, Gilles Deleuze and Félix Guattari, and Jean-François Lyotard is suggested by the naming of “the parti-
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
11
sans of ‘radical finitude’ or of ‘postmodernity,’ who dismiss every variety of
universal as a mystificatory relic of the old metaphysics” (AF, 43), but no
direct engagement with these figures is offered in After Finitude, and there
is much in the works of each of these writers that comports very poorly with
the core tenets of correlationism.15 Meillassoux is willing to name outright
as strong correlationists only two philosophers, who make a very odd pair:
Ludwig Wittgenstein and Martin Heidegger. The only attempt to demonstrate empirically the existence of strong correlationism as a live doctrine
anywhere in philosophy or theory at all in the entirety of After Finitude is this:
The strong model in this characterization seems to us to be represented as much by Wittgenstein as by Heidegger, which is to say,
by the two emblematic representatives of the two principal currents
of twentieth-century philosophy: analytic philosophy and phenomenology. Thus, the Tractatus maintains that the logical form of the
world cannot be stated in the way in which facts in the world can
be; it can only be “shown,” that is to say, indicated in accordance
with a discursive register that cannot be bound by the categories
of science or logic. Consequently, it is the very fact that the world
is sayable (that is to say, liable to formulation according to a logical
syntax) that cannot be bound by logical discourse. Whence proposition 6.522: “There are indeed things that cannot be put into words.
They make themselves manifest. They are what is mystical.” But
the mystical does not consist in other-worldly knowledge—it is the
indication of science’s inability to think the fact that there is a world.
Hence proposition 6.44: “It is not how things are in the world that is
mystical, but that it exists.” Similarly, we have already seen how for
Heidegger it is the very fact that there are beings, and that there is
a givenness of beings, that points to the rift inherent in representation: “Of all beings, only the human being, called upon by the voice of
Being, experiences the wonder of all wonders: that beings are.”16 In
both cases, the fact that beings are, or the fact that there is a logical
15. Martin Hägglund makes a convincing case that “Derrida is one of the intended targets”
of Meillassoux’s attack and offers significant reason to question its accuracy, in “Radical Atheist Materialism,” 115. In “Towards a Speculative Philosophy,” their introduction
to Speculative Turn, Bryant, Srnicek, and Harman specifically name “phenomenology,
structuralism, post-structuralism, deconstruction, and postmodernism” as “perfect exemplars of the anti-realist trend in continental philosophy” (3) that SR/OOO claims to upend.
16. Martin Heidegger, “On the Question of Being,” in Pathmarks, ed. William McNeill (New
York: Cambridge University Press, 1998), 308 [Meillassoux’s note].
Published by Duke University Press
boundary 2
12
boundary 2 / May 2016
world, is precisely what cannot be encompassed by the sovereignty
of logic and metaphysical reason, and this because of the facticity of
the “there is.” (AF, 41–42)17
This passage would be remarkable even were it not the exclusive
presentation of strong correlationism in a treatise that is said to dismantle
that doctrine entirely. Contra Meillassoux’s suggestion that Wittgenstein
and Heidegger together represent all post-Kantian philosophy, it seems
clear enough historically that Heidegger (or, given the textual evidence
Meillassoux relies on, the Heidegger of Being and Time) represents a
single strand of philosophical thought that might accurately be called “Heideggerian” or even “phenomenological,” and that if Meillassoux’s accusation was that “Heideggerian philosophy is correlationist,” it would seem far
less controversial, far less dramatic, and frankly far less interesting, even
if its conclusion would remain controversial, than the claim about “postKantian philosophy.” More startling are his comments on Wittgenstein, and
their implication in the view that has gripped readers so strongly, that all of
modern philosophical thought can be characterized as correlationist. It is
this pole of his accusation that demands the closest scrutiny.
It is fair to take Wittgenstein’s Tractatus as one of several representative starting points for analytic philosophy, but within analytic circles the
Tractatus comes somewhat late to deserve that designation alone; much
more common within the discipline would be the works of Gottlob Frege
or of Bertrand Russell and Alfred North Whitehead, which predate Wittgenstein by decades, and none of which have ever been understood as
tending toward idealism.18 If the point is to show that analytic philosophy is
correlationist, given that work’s repeated engagement with issues of metaphysics, epistemology, and cognition, Meillassoux has chosen a remarkably unrepresentative site from which to do it. However we understand their
meaning, it would be difficult if not impossible to demonstrate that Wittgenstein’s closing remarks in the Tractatus serve the institutional function for
17. Ray Brassier’s translation of After Finitude silently incorporates the D. F. Pears and
B. F. McGuinness translation of Ludwig Wittgenstein, Tractatus Logico-Philosophicus
(New York: Routledge, 1974).
18. Thus in his widely cited history Origins of Analytical Philosophy (Cambridge, MA:
Harvard University Press, 1996), the renowned analytic philosopher Michael Dummett
focuses almost exclusively on Gottlob Frege, Bertrand Russell, Alfred North Whitehead,
and Edmund Husserl, and discusses Wittgenstein only briefly, despite Dummett’s lifetime
commitment to Wittgenstein. Rockmore, in In Kant’s Wake, traces a similar genealogy
that has little if anything in common with the one developed by Meillassoux.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
13
analytic philosophy that Meillassoux claims for them. Further, this passage
reinforces the sense of After Finitude as partly a distillation of Badiou’s own
philosophical prejudices, as this interpretation of Wittgenstein is identical
(down to the specific passages quoted) with the one advanced by Badiou
in his very arch Wittgenstein’s Antiphilosophy.19
What makes all of this so jarring to the student of analytic philosophy
is that the standard interpretation of the Tractatus, the one that makes Wittgenstein’s work coincide to some extent with the perspectives of Russell
and Frege and, perhaps even more important, the one that has been influential in twentieth-century philosophical practice and that is taught in
Anglo-American schools, is that in the Tractatus Wittgenstein is a realist of
a fairly direct sort. Many statements in the work make this abundantly clear.
Thus in a recent essay, Hilary Putnam offers straightforward descriptions of
Wittgenstein that sound almost like the position Meillassoux recommends,
rather than the one he opposes:
The so-called “realism” of the Tractatus, the “realism” that consists
in taking at face value the totality of possibilities represented in the
language (which are also the propositions of science), is what Kant
called “empirical realism.” But Kant thought the possibility of empirical
realism could only be seen from the perspective of his “transcendental idealism.” Wittgenstein (in the Tractatus and pre-Tractarian writings) is saying that transcendental idealism is unintelligible nonsense.
I spoke a moment ago of taking the propositions of “the” language—
which is also “ordinary language,” according to the Tractatus, and
also the language of science—at “face value”; but taking them at face
value is the only way there is to take them! That’s the point.20
Meillassoux argues that “the Tractatus maintains that the logical form
of the world cannot be stated in the way in which facts in the world can be,”
but this is a characteristic, uncharitable, and highly unorthodox reading,
since “precisely the point that Wittgenstein was making in the Tractatus” is
that “there is no standpoint available to us outside the language” (emphasis
added) from which to make sense of Meillassoux’s observation about logical form—there simply are no statements inside our language that provide
objective validation for it.21 Asserting that we do have a standpoint available
19. Alain Badiou, Wittgenstein’s Antiphilosophy (New York: Verso, 2011).
20. Hilary Putnam, “Wittgenstein and Realism,” International Journal of Philosophy 16,
no. 1 (2008): 5–6.
21. Putnam, “Wittgenstein and Realism,” 8.
Published by Duke University Press
boundary 2
14
boundary 2 / May 2016
from which to make such a determination—in Putnam’s words, to distinguish between the propositions “chairs exist” and “chairs really exist”—is
to deny the entire thrust of Wittgenstein’s philosophy in order to accuse it of
an error that Wittgenstein himself goes out of his way to prevent his reader
from making, that few readers of Wittgenstein have thought he was making,
and that few if any followers of Wittgenstein attribute to him.
In his analysis, Meillassoux quotes two of the most gnomic propositions in the Tractatus, ones that specify certain phenomena Wittgenstein
calls “mystical.” Mystical is a term that is notoriously hard to define even
for Wittgensteinians, but it clearly points to consideration of some kind of
metaphysical issues. If the point is that the fact of the existence of the
external world is a “mystical” fact, then that would seem to put the fact of
its existence outside of any direct human relation to it, which is just what
Meillassoux and SR/OOO keep asserting; and if by “mystical” Wittgenstein is taken to mean that our contact with the external world in general
is in some way metaphysical and not, therefore, of a propositional or truthreferential form (the way his claims are commonly understood), that would
seem to constitute exactly a kind of contact between “thought and the
absolute” (AF, 128) that Meillassoux claims is the cardinal error of correlationist thought.22 Rather than being forgotten, though, such a belief would
seem to occupy a central place in the most influential body of post-Kantian
philosophy in the English-speaking world. It also tends to suggest that contrary to the claims of SR/OOO’s novelty, what it offers is very close to the
metaphysical realism of which Putnam himself has long offered withering
critiques.23 It is no less close to what a philosopher with a very different orientation, Bernard Williams, calls the “conception of the world as it is independently of all observers.”24
Meillassoux on Philosophical Practice
Such gestures read less like oversights and more like characteristic features of After Finitude from its opening pages. Its first sentences
22. See Brian McGuinness, “The Mysticism of the Tractatus,” Philosophical Review 75,
no. 3 (July 1966): 305–28, for a relevant (and unexceptional within analytic circles) explication of this part of Wittgenstein’s language that contradicts Meillassoux’s interpretation.
23. See, for example, Hilary Putnam, Reason, Truth, and History (New York: Cambridge
University Press, 1981) and The Many Faces of Realism (Peru, IL: Open Court, 1988).
24. Bernard Williams, Descartes: The Project of Pure Inquiry (New York: Penguin, 1978),
241.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
15
offer a strange and symptomatic allegation about the state of philosophy:
“The theory of primary and secondary qualities seems to belong to an irremediably obsolete philosophical past. It is time it was rehabilitated. For
the contemporary reader, such a distinction might appear to be a piece of
scholastic sophistry, devoid of any fundamental philosophical import. Yet as
we shall see, what is at stake in it is the nature of thought’s relation to the
absolute” (AF, 1). While Meillassoux might be fairly understood to be writing
primarily about philosophy in France, his authority and that of the critique
of correlationism, to say nothing of the literal meaning of his words and
the figures like Heidegger and Wittgenstein he invokes directly, are widely
understood to extend to the entirety of the Western philosophical tradition,
so these opening words of After Finitude must give any philosophically
informed reader serious pause, because they so completely misrepresent
the state of philosophical study. As an index of this strangely out-of-touch
quality of After Finitude, compare it with introductory comments by the US
philosopher Lawrence Nolan in his 2011 edited volume, Primary and Secondary Qualities: The Historical and Ongoing Debate:
There are few philosophical doctrines, if any, which are more familiar than the distinction between primary and secondary qualities.
Most philosophers cut their teeth on it as part of their undergraduate training in the discipline, and for good reason. Studying the doctrine of qualities introduces one to a wide swath of issues in metaphysics, epistemology, philosophy of mind, philosophy of perception,
and semantics. It also forces one to confront questions about the
relation between philosophy and science, and about the nature of
scientific explanation.25
The question of primary and secondary qualities is not part of an
irremediably obsolete philosophical past at all; it is not sophistry; it is of
fundamental philosophical import rather than being “devoid” of it; and so
on. Further, despite the lack of argumentative tone and content offered by
Nolan, it is his work, rather than Meillassoux’s, that provides evidence for
the institutional claims that undergird it: the book includes a bibliography of
over two hundred works from across the entire twentieth century and early
twenty-first that address various aspects of the question, not one of which
Meillassoux mentions.
25. Lawrence Nolan, introduction to Primary and Secondary Qualities: The Historical and
Ongoing Debate (New York: Oxford University Press, 2011), 3.
Published by Duke University Press
boundary 2
16
boundary 2 / May 2016
This is not just a bibliographic error; as Nolan suggests, questions
of the existence and nature of objects, the relation between perception and
metaphysics, the status of thought and knowledge, are not just of interest but are the actual substance of disciplinary philosophy in the Englishspeaking world, and beyond it as well. The very act of studying these discourses is predicated on learning to distinguish among epistemological,
cognitive, and metaphysical claims, to discriminate in one’s writing among
such claims, and to learn the ways of making philosophical claims linguistically precise so as to enable clear discussion of them.
Despite Meillassoux’s admission that Kant is not a strong correlationist and does not adhere to the central doctrine that is supposed to
characterize correlationism and make it so objectionable, in the fourth and
fifth chapters of After Finitude Kant returns as the signal villain of modern
thought, with little if anything being done to explain why, not subscribing to
the offending doctrine at which After Finitude is pointed, he remains a target of critique. In the fifth and final chapter of After Finitude in particular,
Meillassoux makes much more explicit that correlationism’s most critical
sin is the supposed lack of fit between correlationist philosophies and scientific practice.
Now Meillassoux writes that his particular concern is for the “capacity whereby mathematized science is able to deploy a world that is separable from man . . . that rendered possible the essential alliance between
the Galilean and Copernican revolutions” (AF, 115). “In speaking of ‘the
Copernican revolution,’” Meillassoux writes, again bucking most received
notions of the history of science and philosophy, “what we have in mind is
not so much the astronomical discovery of the decentering of the terrestrial
observer within the solar system, but rather the much more fundamental
decentering which presided over the mathematization of nature, viz., the
decentering of thought relative to the world within the process of knowledge” (AF, 115).
Meillassoux only fully states the central argument of After Finitude in
its final pages. That argument rests on a unique and, at best, idiosyncratic
interpretation of many foundational precepts of both modern philosophy
and science:
Even as science, by virtue of its power of decentering, revealed to
thought the latter’s own speculative power, philosophy, at the very
moment when it was ratifying this takeover, did so by abjuring all
speculation, which is to say, by renouncing any possibility of thinking the nature of this revolution. Something akin to a “catastro-
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
17
phe” occurred in this changeover from metaphysics to science as
a guarantor of knowledge—Copernican science provided the impetus for philosophy’s abandonment of speculative metaphysics, but
this abandonment was reflected back onto Copernican science as
philosophy’s Ptolemaic interpretation of the latter. Thus, philosophy’s message to science was: “it is you (and not speculative metaphysics) that holds the reins of knowledge, but the underlying nature
of this knowledge is the very opposite of what it seems to you.” In
other words, in providing the impetus for philosophy’s destruction of
speculative metaphysics, science also destroyed any possibility of a
philosophical understanding of its own essence. (AF, 120)
Like much in After Finitude, this purportedly new reading of Kant
is a familiar one from the philosophical literature with which Meillassoux
engages at only a surface level. That Kant’s metaphorical use of Copernicus admits of more than one reading, for example, is a point covered
by most analytic discussions of the first Critique, including ones as widely
available as Paul Guyer’s 1987 Kant and the Claims of Knowledge.26
Guyer’s discussion shows that there is no case for Meillassoux’s accusation that post-Kantian philosophy denies the existence of the external
world in an unreflective or dogmatic fashion, even among those readers of
Kant who believe he offers such an idealist perspective, let alone among
philosophers in general. Further, the focus on the Copernican metaphor
obscures the far more important engagement of Kant with Newton: “Kant
envisages his Critique on the order of Newton’s Principia, and following
the method detailed in his Opticks,” writes the analytic philosopher Robert
Hahn.27 “Kant envisages Hume’s attack on the validity of causal inference
as an attack on the possibility of all empirical knowledge,” he goes on, so
that “Kant’s argument against Hume thus becomes ex hypothesi an argument for the validity of Newtonian physics.”28
Accepting his own rereading of the Copernican revolution in Kant’s
philosophy as inarguably valid, Meillassoux then asks:
Why did philosophy not take the course exactly opposite to the one
followed by transcendental or phenomenological idealism, viz., the
26. Paul Guyer, Kant and the Claims of Knowledge (New York: Cambridge University
Press, 1987).
27. Robert Hahn, Kant’s Newtonian Revolution in Philosophy (Carbondale: Southern Illinois University Press, 1988), 2.
28. Hahn, Kant’s Newtonian Revolution, 12.
Published by Duke University Press
boundary 2
18
boundary 2 / May 2016
course of a thought capable of accounting for the non-correlational
scope of mathematics, which is to say, for the very existence of science, the latter being properly understood as the power to decenter thought? Why did philosophy, in attempting to think science, err
towards transcendental idealism instead of resolutely orienting itself,
as it should have, towards a speculative materialism? (AF, 121)
Soon after this he concludes: “The deception that presides over
this catastrophe is revealed to be what we earlier referred to as the
‘de-absolutizing implication,’ that is to say, the idea that there is an irrefutable inference from the end of metaphysics to the end of absolutes. Since
science has convinced us that all metaphysics is illusory, and since every
absolute is metaphysical, then it follows that, in order to think science, we
must renounce every form of absolute” (AF, 125). “Yet this,” Meillassoux
argues, “was not what modern science actually required” (AF, 125). His
alternative version of “philosophy’s task consists in re-absolutizing the
scope of mathematics—thereby remaining, contrary to correlationism,
faithful to thought’s Copernican de-centering—but without lapsing back
into any sort of metaphysical necessity, which has indeed become obsolete” (AF, 126). Few passages in After Finitude lay their cards so clearly on
the table, because in the lexicons of most philosophical practice, including
that of Kant, the terms metaphysical and absolute (neither of them terribly
precise, especially when used in such a general manner) point at just the
same things, and Meillassoux admits this in one sentence while denying it
in the next. “Since every absolute is metaphysical,” he writes (“et puisque
tout absolu est de type métaphysique” [AlF, 174]), with the only available
justification being that the terms are identical, followed closely by the task
of “re-absolutizing the scope of mathematics . . . without lapsing into any
sort of metaphysical necessity.” Perhaps such a distinction is plausible, but
one searches After Finitude in vain for any detailed explication of what it is
supposed to mean, including his unconvincing and symptomatically decontextualized excursus on the Cantorian transfinite.
Despite the widespread adoption of the critique of correlationism
in SR/OOO, among many of whose prominent writers a wide variety of
objects are offered as primary concerns, the attentive reader must note
how specific are those objects to which Meillassoux believes correlationism has denied proper status: they are almost without exception mathematical. “Our speculative reformulation of what we must now call ‘Kant’s
problem’”—despite his earlier clear admission that Kant is no correlationist and his failure to identify Kantians who are—“can be stated as follows:
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
19
how is a mathematized science of nature possible?” (AF, 126). Meillassoux appears not to notice that this problem strongly resembles the one
around which the first Critique itself is overtly structured but which is not
mentioned once in After Finitude—“How is synthetic a priori knowledge
possible,” where mathematical propositions are the canonical examples of
such knowledge)—but by removing it from its historical context and deferring it for so long in his own inquiry (only two pages of text remain in After
Finitude after he poses this question), Meillassoux deprives his reader of
the chance to reflect on the familiarity of the question and his answers to it.
Whether or not one believes in absolute truth, the status of mathematics is both special and problematic for almost all philosophers and
deserves and receives special consideration. Few if any examples can be
found to prove the converse of Meillassoux’s suggestion: that there exists
a host of correlationist philosophers who deny that mathematical statements present particular metaphysical and epistemological challenges
precisely because they appear to embody principles that are discoverable
by human reason and yet seem true regardless of any discovery or knowledge of them. More disturbing from an institutional perspective is that a
philosophical and mathematical doctrine has existed since at least the
early 1900s that is explicitly committed to the existence of mathematical
objects as somehow special and different with regard to most other objects
in the world. This doctrine is known as mathematical Platonism. On its most
charitable interpretation, After Finitude seems to be advocating for such a
view, without ever acknowledging the existence of such a doctrine. (While
Meillassoux’s teacher, Alain Badiou, gives short shrift to mathematical Platonism, he does at least discuss it briefly in his works; Meillassoux never
mentions it.)29
Platonism in mathematics is taken seriously across a range of analytic circles, where it is understood as the “metaphysical view that there are
abstract mathematical objects whose existence is independent of us and
our language, thought, and practices.”30 Figures as diverse and central to
29. See Alain Badiou, “Platonism and Mathematical Ontology,” in Theoretical Writings
(New York: Continuum, 2004), 49–58. In a manner somewhat consistent with Meillassoux’s odd exegetical procedures, Badiou disavows the doctrine primarily by arguing that
Plato did not subscribe to it, appearing not to recognize that the term is not used in
contemporary philosophical discourse as a description of Plato’s views, while endorsing many aspects of what is today called Platonism. Further, Badiou applauds central
aspects of the archetypal Platonist Kurt Gödel’s positions, calling him “(with Cohen) the
greatest of Cantor’s heirs” (55).
30. Øystein Linnebo, “Platonism in the Philosophy of Mathematics,” Stanford Encyclo-
Published by Duke University Press
boundary 2
20
boundary 2 / May 2016
analytic philosophy as Gottlob Frege, Bertrand Russell, and Kurt Gödel, as
well as more recent writers including Crispin Wright and Bob Hale, Stewart
Shapiro, and Mark Balaguer, all defend the doctrine in one form or another.
Central works by W. V. Quine, Hilary Putnam, and Michael Dummett consider the position seriously and even accept some parts of the Platonist
doctrine. Much as with his statements on primary and secondary qualities,
it is hard to understand even in the most literal and bibliographical sense
what evidence Meillassoux is relying on to support his assertion that modern thought has “renounced” “the absolute scope of mathematics” (AF,
125). Yet he discusses no contemporary philosophy of mathematics other
than Badiou, and there only to endorse Badiou’s controversial and unorthodox interpretation of Cantorian set theory.
Meillassoux’s Metaphysics
One of the many ironies of the suggestion that correlationism represents the unacknowledged dogma underlying all post-Kantian philosophy is that the anticorrelationist position developed by Meillassoux closely
resembles one of the most prominent positions in twentieth-century analytic philosophy, sometimes even associated with the Tractatus with which
Meillassoux so quickly dispenses. Putnam famously calls this doctrine
metaphysical realism and considers it central to analytic philosophy from its
inception: “On this perspective, the world consists of some fixed totality of
mind-independent objects. There is exactly one true and complete description of ‘the way the world is.’ Truth involves some sort of correspondence
relation between words or thought-signs and external things and sets of
things. I shall call this perspective the externalist perspective, because its
favorite point of view is a God’s Eye point of view.”31 “The world consists of
some totality of mind-independent objects” could be taken as a fair summary of Meillassoux’s doctrinal commitments. Of course, Putnam opposes
this view, not because of doubts about the reality of mind-independent
objects (which Putnam, like and in accordance with Kant, endorses) but
precisely because of the language we humans must use to characterize
the totality of mind-independent objects and our knowledge of that totality.
Putnam describes the opposing perspective this way:
pedia of Philosophy (Fall 2011), plato.stanford.edu/archives/fall2011/entries/platonism
-mathematics/.
31. Putnam, Reason, Truth, and History, 49.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
21
I shall refer to it as the internalist perspective. It is characteristic of
this view to hold that what objects does the world consist of? is a
question it only makes sense to ask within a theory or description.
Most “internalist” philosophers, though not all, hold further that there
is more than one “true” theory or description of the world. . . . There
is no God’s Eye point of view that we can know or usefully imagine;
there are only the various points of view of actual persons reflecting
various interests and purposes that their descriptions and theories
subserve.32
As in so much analytic philosophy, note that the status of objects
and our relation to them plays a central role in Putnam’s writing, here from
a classic 1981 text that is one of the most-cited monographs in the analytical canon.
The problem for Meillassoux is that he continually and unreflectively
assumes not the existence of a noumenal realm—as we have seen, since
Kant such a metaphysical postulate has been philosophically acceptable, to
at least a substantial group of practicing philosophers—but our true, direct,
and absolute access to it. These are different questions; one is metaphysical, and it is this question that Meillassoux so often claims to be pursuing,
while denying the plain assertions of so many philosophers to endorse the
doctrine (the existence of the external world regardless of human access to
it); but the other is epistemological, and it is on these lines that Meillassoux
fails so often to deliver on his promises.
Thus Meillassoux typically offers purportedly anticorrelationist formulas like “The mathematizable properties of the object . . . are effectively
in the object in the way in which I conceive them, whether I am in relation with this object or not” (AF, 3; emphasis added); writes of “that outside which was not relative to us, and which was given as indifferent to its
own givenness to be what it is, existing in itself regardless of whether we
are thinking of it or not; that outside which thought could explore with the
legitimate feeling of being on foreign territory” (AF, 7; emphasis added);
and asks, “What is it that permits mathematical discourse to bring to light
experiments whose material informs us about a world anterior to experience?” (AF, 26; emphasis added). In each case, and in many others like
them, it is only by cursory reading that we can ignore the explicit invocations of human cognition that do not simply color but ground the nomi32. Putnam, Reason, Truth, and History, 49–50.
Published by Duke University Press
boundary 2
22
boundary 2 / May 2016
nally metaphysical claims being made. Further, the philosophical sloppiness of such statements points exactly to what any careful critic, especially
readers of Putnam and other philosophers who have written about the
subject, must suspect: that despite his claim to decenter the human subject in favor of objectivity (ironic, given its overt disavowal of Heidegger
and post-Heideggerian continental philosophy), Meillassoux is committed
exactly to the establishment of a very powerful, very centered, and very
absolute human subject, one typified exactly by the philosophical position
from which Meillassoux writes, which is precisely that of the God’s Eye view
that Putnam works so hard to locate within realist doctrines, and to dispel.
Meillassoux’s anthropocentric practice is not merely a subtle philosophical fault: it is a profound mistake at the core of his work, one that—
because, unlike most proponents of realism in Anglo-American philosophy, Meillassoux refuses even to acknowledge the possibility that human
perception might not provide perfect access to objective reality—ends up
informing nearly all of the argumentative and conceptual work Meillassoux,
and SR/OOO following him, claims to do. As David Berry has pointed out,
at a very practical level, SR/OOO’s claim to equalize the human with all
other objects results in a “performative contradiction,” since while “it would
be perfectly legitimate to outline a formalist theory or methodological position that, for the sake of the approach, limits the requirement to treat human
actors as particular or special in relation to others, . . . it is quite another to
then extend this claim into a philosophical system which is part of a special order of discourse particular to human beings, that is, philosophy.”33 If
we could make sense of the idea of nonhumans doing philosophy, and of
that philosophy being of a sort both to be recognizable to humans as philosophy and yet not to be the same thing we call philosophy, the claim to
have escaped the human orbit might deserve greater consideration. Meillassoux makes it sound as if it is prejudice to note that, so far as we know,
only human beings engage in the specific disciplinary practice we call philosophy, and that human beings are inherently constrained by the conditions of our own existence (in ways we both do and do not understand) in
our assessments of external reality—of which following Kant we can neither
deny the existence nor claim to know its exact nature.
33. David M. Berry, “The Uses of Object-Oriented Ontology,” Stunlaw: A Critical Review
of Politics, Arts and Technology, May 25, 2012, stunlaw.blogspot.com/2012/05/uses-of
-object-oriented-ontology.html.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
23
The Return of the Human
Meillassoux ends After Finitude with a series of brief and obscure
formulations that appear to indicate what would be at stake for a philosophical practice stripped of correlationist blinders. First, “we must establish the following thesis[:] . . . what is mathematically conceivable is absolutely possible” (AF, 126); second, he means to demonstrate “that the laws
of nature derive their factual stability from a property of temporality that is
itself absolute, which is to say, from a property of time that is indifferent to
our existence, viz., that of the non-totalizability of its possibilities,” a matter “of deriving the absolute and now unconditionally necessary scope of
a particular theorem, viz., the theorem that allows us to maintain the nontotalizability of the transfinite” (AF, 127)—in both cases, characteristically
and explicitly invoking human existence and cognition even as he claims
to have put them aside. “Our only aim,” he writes, “has been to try to convince the reader not only that it is possible to rediscover thought’s absolutizing scope, but that it is urgent that we do so, given the extent to which
the divorce between science’s Copernicanism and philosophy’s Ptolemaism has become abyssal, regardless of all those denials that serve only
to perpetuate this schism”—“thought’s absolutizing scope” being again an
epistemological rather than metaphysical concept, about human thought
rather than escaping consideration of it. Meillassoux hopes nothing less
than to “[wake] us from our correlationist slumber, by enjoining us to reconcile thought and absolute” (AF, 128)—despite the apparent cardinal sin of
correlationism being its inability to sever the relation between thought and
being, Meillassoux’s goal is revealed explicitly as yoking them together.
When the “critique of correlationism” is invoked in SR/OOO writings,
it is typically understood as a demonstration that almost all prior thought
has been interested only in human beings, that this is some kind of not
just conceptual but ethical-cum-political flaw, and that a new form of practice can be developed in which humans themselves are, or at least human
access is, demoted in favor of a more egalitarian approach to the stuff of
the world.34 Yet Meillassoux’s profound lack of attention to the actual work
of philosophers, literary theorists, and even scientists—itself significantly
34. I am thus very sympathetic to the critique offered by Alexander Galloway, “A Response
to Graham Harman’s ‘Marginalia on Radical Thinking,’” An und für sich, June 3, 2012,
itself.wordpress.com/2012/06/03/a-response-to-graham-harmans-marginalia-on-radical
-thinking/, and his “The Poverty of Philosophy: Realism and Post-Fordism,” Critical Inquiry
39, no. 2 (2013): 347–66.
Published by Duke University Press
boundary 2
24
boundary 2 / May 2016
disturbing for a philosophy that claims to be getting in closer touch with
objects than have previous schools but that is unable even to perform this
operation on the material it should know the most about—means that he
continually fails to make good on these promises and fails to avoid the
conceptual and practical traps philosophers have uncovered in the very
issues he claims to be championing. By failing to read carefully in metaphysics or epistemology and to respect their procedures, SR/OOO confines its litanies of objects to those perceptible and knowable by human
beings, despite the clear consequences even of Kant’s philosophy that the
universe must in fact be full of objects about which human beings know little
if anything (exactly the position that in some modes SR/OOO appears to
recommend) and that the ones about which we do know come in a variety
of kinds. In claiming to decenter the human, but in “enjoining us to reconcile
thought and absolute,” Meillassoux achieves the opposite effect, putting the
human knower at the center of a universe to which he or she has perfect
access, a Ptolemaic Earth around whom everything revolves.35
At least in some of its instances, SR/OOO offers itself as a program
of not just metaphysical but ethical import, and some of its main figures
at times write as if one of our world’s current major problems is our failure
to take seriously enough the concerns of nonhuman (and even nonliving)
objects. Here, too, the failure to engage with the many varieties of materialist political and literary theory that deal with the stuff of the world is of serious concern, for the varieties of “thing theory,” material culture, and ecological approaches to criticism appear to have been engaged in just such
practices and must be written out for SR/OOO’s global critique to take hold.
Like Meillassoux’s paradoxical human de- and recentering, such institutional practices end up installing the SR/OOO writers as having some kind
35. In Anglo-American philosophy the discussion of objects has been widespread and
persistent since the early 1900s, and SR/OOO’s almost complete inattention to this
material is one of the most concerning and symptomatic features of its practice. Almost
unmentioned in SR/OOO are influential general works wherein objects figure prominently, such as W. V. Quine’s Word and Object (Cambridge, MA: MIT Press, 1960); and
Brian Cantwell-Smith’s On the Origin of Objects (Cambridge, MA: MIT Press, 1998).
There is also no engagement with analytic work on mereology, the study of parts, wholes,
and the metaphysics of objects in general. Among prominent contemporary studies in this
field, see Kathrin Koslicki, The Structure of Objects (New York: Oxford University Press,
2008); David Lewis, Parts of Classes (Oxford: Blackwell, 1991); and Peter Simons, Parts:
A Study in Ontology (New York: Oxford University Press, 1987). To my knowledge there is
also no sustained discussion in SR/OOO of analytic metaphysics of the sort associated
with David Armstrong and David Lewis.
Published by Duke University Press
boundary 2
Golumbia / “Correlationism”
25
of privileged or special access denied to other thinkers for reasons none of
the SR/OOO writers can articulate. Meillassoux condemns thinkers from
across the spectrum of human activities for their egocentrism, yet its actual
practice suggests that this is exactly how Meillassoux wants his audiences
to see his own work, while he repeatedly fails to demonstrate that egocentrism in his targets. Among the most concerning effects of his writing is to
discourage students from reading in the philosophical tradition works that
directly address the questions in which they seem interested.
The project SR/OOO and Meillassoux take as their starting point,
Kant’s project, was never intended solely as metaphysics or epistemology.
For Kant it connected directly to questions of how human beings ought to
live. Kant wrote the first Critique in part to enable to completion of the second, the Critique of Practical Reason, but the second Critique goes largely
unmentioned in Meillassoux and in SR/OOO. Kant used the impact of scientific inquiry on rational conceptual practice to argue against the relevance
to humans of an absolute realm not of mathematics but of a Supreme Being
whose existence determines the right course for human action: “Without
a God and a world invisible to us now but hoped for, the glorious ideas of
morality are indeed objects of approval and admiration, but not springs of
purpose and action” (A813/B841). Human beings—or more generally, rational actors—are “special” precisely because we are free to act in a way that
other (nonfree) actors are not. This may be construed as conferring on us
privileges, but to Kant it confers something much more like responsibility.
This responsibility is one we can only conceive of in the human context or
in that of other rational beings. To deny this fundamental thrust of Kantian
philosophy is to suggest that we bear no such responsibility, that we are
free to pick and choose which parts of our experience we consider worthy
of the ethical imperative Kant calls categorical, that we might decide not
merely that wooden posts, marbles, and iPhones are worthy of the same
ethical care due to other humans, but that other humans might well not
be. If there is one feature of a Kantian and post-Kantian philosophy that is
widely shared and that we consider abandoning only at our own peril, it is
this one.
Published by Duke University Press