On Correlationism and the Philosophy of (Human) Access: Meillassoux and Harman

1 Correlationism and the Philosophy of (Human) Access

The loosely demarcated movement known as Speculative Realism (SR) got its title from a conference named Speculative Realism: A One-Day Workshop, held at Goldsmiths University in April 2007. ^[1] The speakers – and original members – were, Ray Brassier, Iain Hamilton Grant, Graham Harman and Quentin Meillassoux, even if the influence of SR has since spread well beyond the work of these respective philosophers. It would however be important to note from the outset that there are important and fundamental differences between the positions of the various thinkers that are often grouped under this umbrella term. ^[2] For this reason, many have since questioned the existence of such a movement, leading Graham Harman ^[3] to proclaim that he remains the only one amongst its four original members to still be fully committed to using the term. Nevertheless, the existence or status of such a group need not be a source of concern. As Harman rightly asserts, the realism/anti-realism debate has at least gained more prominence in current continental philosophy after a long period of being dismissed and criticised as a supposed “pseudo-problem.” ^[4] This fact, in turn, ought to attest to the impact and influence of this otherwise varied group of thinkers.

What is often said to almost exclusively unite all the original and current proponents of SR is their commitment to the critique of what Quentin Meillassoux terms ‘correlationism’ ^[5] or what Graham Harman calls the ‘philosophy of (human) access.’ ^[6] I shall have occasion to expand on both these expressions throughout the course of this paper. As a preliminary point, it may however be noted that both are to an extent similar in terms of what they critique, namely (what proponents of SR see as) the prevalent tendency within Kantian and post-Kantian thought to treat the relation between thought and world as the primary subject matter of philosophy. In making such a claim, they argue that philosophy since Kant lamentably negates the possibility of thinking or knowing what the world could be like ‘in itself’, that is, independently of our all-too-human relation to it. This prevalent similarity has in turn led many authors – both in recent literature ^[7] and even Meillassoux and Harman themselves – to treat the two terms as synonymous. Against this tendency, I maintain that the two expressions are in actual fact not congruent, and that the difference between them ought to be retained and also emphasised since it hints at important differences between the positions of Harman and Meillassoux. More precisely, the term ‘correlationism’ is directed against thinkers who prohibit the possibility of knowledge of a non-metaphysical absolute, while the ‘philosophy of (human) access’ targets the tendency within post-Kantian thought to posit a fundamental and exclusive chasm – or lack thereof – between human and world alone. In other words, and as I shall show below, the stakes of keeping the two terms separate thus boils down to the question of whether the future of SR should proceed epistemically by seeking a rational refutation of the ‘correlationist’ argument (Meillassoux), or ontologically through a speculative form of realism which refuses ‘access philosophy’ by generalising finitude beyond the human realm (Harman). In this paper I shall expand on these claims by analysing Harman and Meillassoux’s terms in order to evaluate their similarities and differences.

2 Quentin Meillassoux’s ‘Correlationism’

As has already been pointed out above, SR is united in its critique of what Meillassoux terms ‘correlationism.’ ^[8] In the opening pages of his now seminal work entitled After Finitude, he maintains that correlationism represents the “central notion” of all continental philosophy since Immanuel Kant’s famed ‘Copernican Revolution’, even if he has also suggested elsewhere that its roots can already be found in the works of Berkeley and Hume. ^[9] Meillassoux in turn defines ‘correlationism’ as:

[1] 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. [2] We will henceforth call correlationism any current of thought which maintains the unsurpassable character of the correlation so defined. [3] Consequently, it becomes possible to say that every philosophy which disavows naïve realism has become a variant of correlationism. ^[10]

It may then be possible to further explicate the definition above by breaking it down into three distinct yet interrelated claims. The first claim (numbered 1 in the quote above) indicates that correlationism refers to any philosophical position which (implicitly or explicitly) asserts that it would be impossible to attain knowledge of what being might be like independently of thought or vice-versa. As Paul J. Ennis points out, speculative realists understand “transcendentalism, phenomenology and postmodernism” to be correlationist insofar as all of them lay emphasis on questions of human access to being, rather than being itself. ^[11] Meillassoux ^[12] further asserts that, at its most basic level, all correlationism rests on an argument he calls the ‘correlationist circle’, which may in turn be formulated as follows: against the realist’s allegation that one can make positive claims about the nature of being as it exists independently of thought, the correlationist would claim that the realist is essentially guilty of circular reasoning, since there can be “no X without givenness of X, and no theory of X without a positing of X”, such that X “cannot then be separated from this special act of positing, of conception”. ^[13] The correlationist thus insists that one can only have access to the (cor)relation between thought and being, rather than being in itself. As Peter Hallward rightly points out, Meillassoux claims that all correlationist philosophies

posit some sort of fundamental mediation between the subject and the object of thought, such that it is the clarity and integrity of this relation (whether it be clarified through logical judgement, phenomenological reduction, historical reflection, linguistic articulation, pragmatic experimentation or intersubjective communication) that serves as the only legitimate means of accessing reality. ^[14]

It would however be important to point out that, for Meillassoux, the correlation between the subject and object – or mind and world – represents but one form of correlation. He insists that correlationism can take various forms including those of the “subject-object, consciousness-given, noetico-noematic correlate, being-in-the-world, language-reference, etc.” ^[15] From these examples, it should then be ascertained that Meillassoux does not argue that correlationism relies on the subject/object dualism or on representationalism. Rather, he holds that many thinkers – such as Husserl and Heidegger, to name but two prominent examples – have notably criticised dualist or representationalist models, but only in order to then insist that the relation itself take precedence over the related terms. In some cases, it is even said that the relation itself comes to constitute the two poles.

Following from the first claim, Meillassoux’s second claim (numbered 2 in the quote above) outlined above emphasises that, for correlationists, the relation between ‘thinking and being’ is ‘unsurpassable’, such that it would be impossible to move beyond the strict boundary limits of the correlation in order to speak of things as they are independently of the way they are given. As a result, speculative realists maintain that post-Kantian thinkers such as Husserl, Heidegger, Derrida and Foucault classify as correlationists to the extent that they deride the realism/anti-realism debate as a “preoccupation of mediocre thinkers”, insisting instead that knowledge can only be limited to what goes on between thought and being. ^[16] For Meillassoux, the ultimate idea of correlationism is thus that “to be is to be a correlate”, ^[17] and its ultimate claims rest on the supposition that any elaboration of “exteriority” is always essentially “relative […] to consciousness, a language, a Dasein, etc.” ^[18] It would however be important to stress that Meillassoux is not here arguing against the rather evident idea that one must necessarily relate to something so as to conceive it; every philosophy – whether realist or correlationist – must essentially accept that it would be impossible to know something without relating to it. Rather, the correlationist claim differs from the realist one insofar as the former asserts that whatever can be known must be indexed back to a knower, such that it is impossible to know anything about objects as they are in themselves, independently of how they appear in their relation to a subject. It is for this reason that Meillassoux claims that the ‘co-’ in ‘correlationism’ constitutes the “grammatical particle that dominates modern philosophy”, ^[19] insofar as it is designed to fend off the possibility of stepping outside and beyond the limits of the correlation.

Finally, the third claim (3) implies that correlationism represents the implicitly dominant anti-realist dogma of Kant and his successors. Correlationism, as understood by SR, represents a prevalent, implicit, yet specific form of (transcendental rather than subjective) idealism which disavows realism by primarily negating its mind-independence component. To this, one may possibly object that many thinkers are not flat out anti-realists since they do not explicitly deny the existence of whatever may exist independently of thought. Derrida, for instance, is said to be the prime example of a correlationist par excellence. Yet he had often explicitly expressed his frustration with readers who misinterpret his work as a flat-out idealist denial of all referents. Derrida’s interview with Richard Kearney provides a representative example of such statements, when he proclaims that

I never cease to be surprised by critics who see my work as a declaration that there is nothing beyond language, that we are imprisoned in language […] Every week I receive critical commentaries and studies on deconstruction which operate on the assumption that what they call ‘post-structuralism’ amounts to saying that there is nothing beyond language, that we are submerged in words – and other stupidities of that sort. ^[20]

In response to such claims however, a speculative realist would insist that such attempts to fend off charges of idealist anti-realism do not quite cut it, for ultimately his ‘deconstructive’ project still leaves us perpetually ensnared within the unsurpassable correlation between language and whatever it is that may exist beyond it.

It would also be important to note that, for Meillassoux, correlationism comes in varying intensities. In After Finitude, he essentially distinguishes between the weak model and the strong model of correlationism, and further subdivides the latter into two. ^[21] In Quentin Meillassoux: Philosophy in the Making, Harman accurately characterises Meillassoux’s treatment of correlationism in terms of a ‘spectrum.’ This idea of a ‘Correlationist Spectrum’ seems to be a suitable way of representing Meillassoux’s claims, since it clearly differentiates between the most common forms of correlationism, yet also illustrates that Meillassoux is not simply committed to limiting the forms of correlationism to just these two.

Against the so-called ‘naïve’ form of realism, weak correlationism makes the claim that it is impossible to achieve knowledge of things-in-themselves. Weak correlationism – like all its other forms – affirms what Meillassoux calls the ‘correlationist circle’, an argument which consists in the insistence that one may never claim to know the in-itself without falling into self-contradiction. For Meillassoux, the weak correlationist thus defeats the naïve realist’s faith in the apprehension of the things-in-themselves by claiming that it is impossible to separate what is posited from the act of positing it. ^[22] Meillassoux himself names Immanuel Kant as an exemplar of such a position, even if, as I shall show below, Harman does not quite fully agree with this assessment. ^[23] As is well known, in his Critique of Pure Reason Kant maintains that one can never know things-in-themselves (or ‘noumena’), since human knowledge is limited to what he called ‘phenomena.’ This in turn leads Kant to proclaim that philosophy ought to focus on things as they are knowable, rather than attempting to engage in futile speculation about the nature of things as they are independently of human knowledge. Nevertheless, for Meillassoux, Kant qualifies as a weak correlationist. ^[24] This is insofar as he maintains that it would nevertheless be possible for humans to think certain qualities of the thing-in-itself; for instance, Kant maintains that it is possible to think that the things-in-themselves exist, and that they are non-contradictory. In this context, it would be worth highlighting that Meillassoux is sceptical about the “miraculous operation” by which Kant moves from the world as it is for us to the properties of the in itself. ^[25] As shall be noted later, this suspicion on Meillasoux’s part illustrates that he is in actual fact convinced by the claims which strong correlationism levels against its weaker counterpart.

Meillassoux in turn defines ‘strong correlationism’ as the implicitly dominant model of all post-Kantian philosophy. Unlike its weaker cousin, strong correlationism relies on what he calls the ‘correlationist two-step.’ ^[26] Meillassoux himself names Wittgenstein and Heidegger as archetypes of strong correlationism, even if it may be argued that the charge equally applies to varieties of post-structuralism, post-modernism and phenomenology amongst many other possible examples. ^[27] Like the aforementioned form of (weak) correlationism, the argument of the ‘correlationist two-step’ begins with the affirmation of the aforementioned ‘correlationist circle’, which – as has already been shown – insists on the “inseparability of the act of thinking from its content.” ^[28] However, unlike its much weaker relative, the strong variant involves the further “belief in the primacy of the relation over the related terms.” ^[29] In other words, the strong correlationist sees thought and world as so tightly interwoven together that it would be completely impossible to imagine one without the other. In other words, and as Peter Gratton precisely points out, for the strong correlationist “reality and human beings go together like conjoined twins: where you find one, you find the other”, such that it “rules out of bounds any discussion [or thought] of ‘reality’ as it is” in itself. ^[30] For instance, and as has already been asserted, Kant deems it possible to think that things-in-themselves are non-contradictory. By contrast, the strong correlationist is said to differ from its weaker variant insofar as they maintain that “not only that it is illegitimate to claim that we can know the in-itself, but also that it is illegitimate to claim that we can at least think it”; things-in-themselves, in other words, can neither be known nor thought. ^[31]

Added to this, Meillassoux distinguishes two further kinds of strong correlationism. The first form of strong correlationism identifies itself as an heir to the Kantian project, and seeks to “uncover the universal conditions for our relation to the world.” ^[32] The second form of strong correlationism however dismiss the possibility of such universal conditions as an antiquated remnant of metaphysics, insisting instead that the relation is “itself finite, and hence modifiable by right.” ^[33] Meillassoux further claims that such a position is most prominently represented in the works of various post-modernists.

At this point, it should be noted that, in After Finitude, Meillassoux speaks of correlationism somewhat ambiguously. He sometimes seems to attribute it exclusively to thinkers who adopt a sceptical stance towards things-in-themselves; at other times he however also seemingly includes various idealist positions which, for Meillassoux, are said to “hypostatise the correlation” between thinking and being. ^[34] In an essay entitled “Iteration, Reiteration, Repetition”, he makes however the distinction between these two positions more explicit and distinguishes a ‘strict sense’ of correlationism form the broader ‘Era of Correlation.’ Meillassoux subsumes the aforementioned weak and strong forms of correlationism under the rubric of ‘correlationism in the strict sense.’ The broader ‘Era of Correlation,’ on the other hand, is defined as consisting of two opposite movements, namely ‘correlationism in the strict sense’ and ‘subjectalism.’ As has already been outlined above, correlationism, in its strict sense, denies ‘thought all access to the absolute’, and comes in varied ‘sceptical’ forms, such as those of transcendentalism, phenomenology and postmodernism. ^[35] Meillassoux, in turn, defines subjectalism as a non-materialist form of absolutism which absolutizes “thought itself,” or “certain remarkable traits of thought.” ^[36] Meillassoux further subdivides subjectalism into two poles, namely the vitalist pole – which, for Meillassoux includes thinkers such as Nietzsche and Harman (see below) – and the idealist pole – which includes thinkers such as Hegel and Berkeley. In spite of their differences, Meillassoux however claims that both vitalists and idealists are similar insofar as their “refusal of anthropocentrism” leads to a form of “anthropomorphism” which imbues reality with “subjective traits whose origin is in truth all too human.” ^[37] In sum, Meillassoux thus defines the ‘Era of Correlation’ as an

antimaterialist […] era that has shut us up inside the correlation, either through an antispeculative gesture – which alone I call correlationism – or through a speculative gesture – which I call subjectalism – that absolutizes the correlation of thought and the world through the choice of various traits, all of which are present in human activity. ^[38]

A further detailed analysis of Meillassoux’s own path beyond correlationism is unnecessary given the scope and aims of the present paper, and shall therefore be omitted. On the basis of the outline provided, it would nevertheless be possible to notice a common thread running through all forms of correlationism as described by Meillassoux. In spite of their varying assertions and commitments, all of them are committed to a stringent emphasis on finite human knowledge, and thereby deny any possible acquaintance with the ‘great outdoors.’ Meillassoux’s major misgiving with all forms of correlationism thereby lies in their inability to arrive at some form of direct and absolute knowledge of the ‘in-itself.’ Against this correlationist commitment, Meillassoux’s thought seeks a speculative (against correlationism) non-metaphysical (against subjectalism) form of materialism which is able to bypass the limits of human finitude and accede to a thinkable and knowable absolute. ^[39]

Relative to this, it may be noted that Harman sees Meillassoux’s anti-correlationist stance just described as essentially inadequate insofar as it still rests on the all-too-human commitment to the fundamental difference separating humans from the rest of the world, and to the claim that the former’s superior rational capacities are able to adequately model the latter. Stated differently, Harman claims that Meillassoux perpetuates another form of ‘philosophy of human access’ which the former dubs ‘epistemism’, insofar as he sees the latter’s form of speculative philosophy as still too faithful to the idea that “reason ought to be able to attain the direct presence of the thing.” ^[40] Thus, Harman finds a problem with Meillassoux’s anthropocentric tendency to reduce what is real to the tiny sliver available to human knowledge. In this way, he sees the latter’s anti-correlationist stance as a ‘knowledge-centred brand of realism’ which ultimately rests on a passage beyond human limits, while remaining within the sphere of human knowledge. ^[41] This is insofar as Meillassoux “seems to see no problem with fully translating a thing into knowledge of that thing, identifying its primary qualities with the mathematizable ones.” ^[42] As I shall show, Harman’s anti-access position stands in contrast to Meillassoux’s position insofar as it postulates an ontological form of realism which rejects any epistemological commitment to absolute knowledge, whilst simultaneously opposing the stale anthropocentric “dictatorship of human beings in philosophy.” ^[43]

3 Graham Harman and the ‘Philosophies of Human Access’

Meillassoux’s analysis of correlationism outlined above may in turn be fruitfully compared to what Graham Harman names the ‘philosophies of human access.’ In The Quadruple Object, Harman describes the latter as the “tacit or explicit credo of a now lengthy tradition of philosophy” which begins with Kant and German Idealism, but is said to persist to this very day. ^[44] Harman further defines it as a form of anti-realism which rests on a simple argument of the following form: against all proclamations of philosophical realism, the philosopher of human access (implicitly or explicitly) asserts that in order to think anything as unthought means to think it. ^[45] Therefore, the claim that one can think the unthought clearly constitutes a contradictory claim, for it is impossible for one to think the unthought without actually thinking it. ^[46] Like the correlationist, the philosopher of human access therefore privileges human access to the world by claiming that “human experience includes the totality of legitimate philosophical content”. ^[47] As a result, Harman asserts that the ‘philosophy of (human) access’ is a form of anti-realism which restricts “philosophy to operate only as a reflective meta-critique of the conditions of knowledge”, and thereby prevents philosophy from being able to speculate about the nature of the world independently of the human access to it. ^[48] Harman further claims that the basic line of reasoning of the philosopher of human access outlined above yields two possible conclusions, and names these the weaker and stronger inferences. ^[49]

Like all philosophies of human access, the ‘strong access’ version begins with the premise that one cannot think what is unthought, and goes on to conclude from this that “there really is nothing outside the human-world coupling.” ^[50] Harman regards the philosophy of George Berkeley as the historical exemplar of such a position. He however also claims that the ‘strong access’ version may be easily refuted with the following argument: While one may concede that the claim that “the thought of X cannot exist without the thinking of X” constitutes a tautological claim, it would nonetheless be illegitimate to conclude from this – as the strong access philosopher does – that X itself does not exist. ^[51] Harman thus objects to the strong access philosopher with the claim that such thinkers illegitimately move from a tautological claim that “there is no thinking without thinking” to the non-tautological conclusion that “there is no being without thinking.” Harman argues that the evident weakness of the strong access version has led many to refute it, opting instead to turn to what can be referred to as the ‘weak access’ position. ^[52]

While the strong access version involves the unqualified assertion that nothing exists outside of thought, the weak access version opts for a more philosophically guarded cynicism which maintains that something might very well exist beyond access but, even if this were the case, one would never be able to conclusively know it. For Harman, the ‘weak access’ argument may be sketched out in the following fashion: like the strong access version, the weak access argument starts by making a tautological claim that there cannot be any thought of some X lying outside of thought. The weak access philosopher then uses this tautology to derive the following inference: since there is no thought of X outside of thought, it follows that to attempt to think X as unthought immediately turns it into a thought. As a result, the ‘weak access’ philosopher treats a statement such as ‘unthought object outside of thought’ as literally devoid of all meaning, since they treat the statement ‘X outside of thought’ as synonymous to the ‘thought of X outside of thought.’ ^[53]

In light of this, and in spite of the differences between the strong and weak access versions, it may be noted that Harman finds both positions to be thoroughly problematic to the extent that they intrinsically remain tethered to what he has recently dubbed ‘onto-taxonomy.’ ^[54] He defines the latter as a characteristic of all philosophical positions which posit an a-priori chasm between “human beings on one side and everything else on the other.” ^[55] For Harman, such thinkers are guilty of propagating what he calls the ‘taxonomic fallacy’, namely they uphold the unwarranted anthropocentric assumption that some distinctive trait of human beings (whether it be language, reason, and so forth) automatically makes them “ontologically different in kind” from everything else that may possibly exist beyond thought. ^[56] Harman rejects such an assumption, and instead argues that the difference in degree – but not in kind – between the multitude of entities populating the world “must be earned rather than smuggled in from the seventeenth century as purported self-evident truths.” ^[57] This fact is crucial insofar as it illustrates one important aspect of Harman’s specific form of (speculative) realism: for Harman, both ‘access philosophies’ and self-proclaimed realist or materialist philosophies which claim privileged access to the way the world is in-itself remain ‘onto-taxonomic’ in their persuasion, and this is for two reasons; first, such philosophies start off by assuming that the world is different in kind from humans. Second, they assume that human reason is so special that it is able to reduce the ‘world’ to the totality of what can be thought about it.

Relative to this, it would then be important to note that, for Harman, the problem with the ‘philosophy of human access’ is not finitude – as Meillassoux claims of ‘correlationism’ – for in fact he believes this to be an inevitable condition of all forms of relation, whether human or otherwise. Harman’s path beyond ‘access philosophies’ thus proceeds by essentially generalising (rather than rejecting) finitude as a feature of all inter-objective relations. More specifically, for Harman finitude is not to be understood as “the product of human sensation, but of the perspectival stance of any entity whatever ^[58]”, to the effect that “the real [ontological] distinction” is not to be found in the ‘onto-taxonomical’ rift “between thought and nature, but between objects in themselves and objects as caricatured by others in their relations.” ^[59] By way of an example often used by Harman, his claim is that when humans relate to cotton, they necessarily do so by translating it in a finite human manner; for instance, they relate to it as a form of equipment used for cleaning or as part of a theory about botany. Nevertheless, he also insists that this fact also holds true for the relation between fire and cotton for instance, such that when fire interacts with cotton or causes it to burn, the former only relates to the latter by translating it in limited terms which might be relevant to it but not to humans, insects, or rocks. ^[60] Thus, Harman states that fire only relates to the flammable qualities of cotton, not the cotton in its being. This generalisation of the thesis of finitude thus attests to Harman’s refusal to take the ‘access philosopher’ position seriously; in effect, he believes that the argument of the ‘weak access’ philosopher is not at all convincing, and thereby not a true problem which one ought to show sympathy towards. Rather, he views it as a fallacious argument and a “sad degeneration from a robustly realist attitude ^[61]” which needs to be thrown out. He claims that all ‘philosophies of human access’ seem like an “indefensibly narrow […] claustrophobic honey trap” which are “both inadequate and false.” ^[62] From such claims, it is therefore clear that Harman’s position progresses through the refusal of the ‘access’ problem rather than its refutation.

Against Harman, Meillassoux claims that the above described Harmanian solution to the ‘philosophies of human access’ is indefensible as it ultimately amounts to nothing more than a ‘Rhetoric of the Rich Elsewhere.’ ^[63] This is insofar as Meillassoux argues that Harman’s path beyond access entails the unwarranted generalisation of certain “experienceable traits of our always human existence”, thereby illegitimately “hypostasizing between things the relation [of finitude] that humans maintain with them.” ^[64] In other words, he claims that Harman’s anti-access philosophy remains faithful to the ‘Era of Correlation’ in the form of a ‘vitalist’ variant of ‘subjectalism’ (see above). The importance of the difference between Harman and Meillassoux becomes more telling if one considers the way they each deal with figures from the history of philosophy. Meillassoux is unequivocally dismissive of the ‘catastrophic’ positions adopted by thinkers such as Kant and Heidegger, insisting that each belongs to differing forms of anti-realist correlationism. ^[65] Harman’s attitude towards Kant and Heidegger is however somewhat more complicated. In Harman’s reading of Heidegger and Kant, he argues that both of them are to some extent realists for the following reason: Kant insists on the existence of things-in-themselves, while, for Harman, Heidegger holds that “being [can never be] fully exhausted in its manifestations to Dasein.” ^[66] Nevertheless, he also contends that Kant and Heidegger remain ‘access philosophers’ due to the fact that they only conceive of this mind-independent reality as some locus of resistance. Thus, he is critical of their inability – or, rather, unwillingness – to speculate on the nature of being independently of Dasein or synthesis – for in fact, Kant ignores the problem of what things are independently of minds, whereas Heidegger dismisses the question altogether.

4 ‘Correlationism’ and ‘Access Philosophies’ as Forms of Idealism

As has been noted, Meillassoux claims that the commitments of the correlationist exposes all of its variations to be a form of ‘extreme idealism’, insofar as it rests on the assumption that one cannot represent something without the act of representation, thereby transforming the “in itself” to “the in itself for us.” ^[67] Furthermore, this same claim may also be observed in Harman’s treatment of the ‘philosophies of human access.’ Harman asserts that “the skeptic immediately flips into an absolute idealist, since the phrase ‘things in themselves’ is emptied of all possible meaning and becomes just another way of saying ‘things for us’.” ^[68]

The claim that the authors associate ‘correlationists’ and ‘access philosophies’ with forms of ‘extreme idealism’ or ‘absolute idealism’ has however been met with scepticism. In an interview with Graham Harman, Meillassoux himself also admits that the term ‘idealism’ is “loaded with ambiguity”, and that “there are numerous correlationists who refuse to be recognized as idealists.” ^[69] This somewhat ambiguous move from ‘correlation’ or ‘access’ to outright idealism has in turn earnt Meillassoux and Harman a great deal of criticism by authors such as David Golumbia ^[70], Peter Gratton ^[71], Peter Hallward ^[72] and Dan Zahavi ^[73] amongst others. It may then be useful to briefly consider this move, as well as a representative sample of its criticisms. In an essay which is otherwise highly critical of speculative realists, for instance, Zahavi concedes that “the speculative realists are certainly right in their assessment of how widespread correlationism is.” ^[74] Nevertheless, he also maintains that it is both “controversial” and “historically incorrect” to maintain that thinkers such as Heidegger and Husserl are flat out idealists. ^[75] This claim is, in turn, also reiterated by Gratton. Echoing Hallward’s critical assertion that correlationism wrongfully conflates epistemological conditions with ontological claims, Gratton points to the difficultly of justifying how the statements of correlationism can be said to lead to a “crude idealism.” He argues that it is most certainly possible to maintain that “such and such are the epistemological or linguistic conditions for knowledge” without having to commit oneself to the ontological claim that “things in the world ‘depend’ on thinking for existence.” ^[76]

It would however be possible to reply to such critiques by focussing on the modifications which ‘correlationists’ and ‘access philosophies’ bring to idealism. Such positions may in actual fact be said to refine the main idealist thesis. Whereas an orthodox idealist would reduce everything to a single pole construed in terms of ‘ideas’ or ‘Geist’, the ‘correlationist’ or ‘access philosopher’ reduces everything to a double-sided relation, irrespective of whether such relation is construed in terms of the subject-object, noetico-noematic, language-reference, Dasein-Being correlation and so forth. Thus, this refinement of idealism in correlationism is designed to fend off metaphysical realism, but it does little to alter the basic idealist thesis, insofar as correlationism reduces “every real being to being dependent on the relation to an originary ground, which is itself […] reduced to an anthropological determination (whether experience or language).” ^[77]

Thus, as Harman rightly points out, the ‘correlationist’ – and, one may argue, the ‘access philosopher’ – essentially insists that there are only two real beings, namely humans and world, which nevertheless inseparably exist in a permanent (cor)relation with one another. ^[78]

5 Conclusion

On the basis of what has been said, a certain parallel between the critique of ‘correlationism’ and that of the ‘philosophies of (human) access’ may be noted. Both positions essentially name implicit forms of idealism which hold that “whatever structure there is in the world has to be transcendentally imposed or generated or guaranteed.” ^[79] Such similarity has in turn led many to assume – incorrectly, in my view – that the two terms are ultimately congruent and hence interchangeable, and it is for this reason that many have even characterized correlationism as the only fundamental point of convergence between the various forms of SR. Harman is however somewhat unclear about the relation between the two terms; at times, he seems to regard them as “not identical but similar enough.” ^[80] At other times however, he proclaims that ‘correlationism’ could “simply replace [his] own ^[81]” term, and this is for two main reasons; first, he claims that Meillassoux’s term is more “crisp, snappy and memorable” than his own. Second, Harman claims that it “leaves its target no escape” since “it fully grants that the correlationist is not an idealist in the strict sense, but is obsessed instead with a correlation that includes a world-pole no less than a mind-pole.” ^[82]

In spite of Harman’s assertions, I nevertheless claim that the difference between the two terms should be retained and even emphasized. My reasons for holding this are twofold: First, it may be argued that Harman kept using his own term independently of Meillassoux’s and vice-versa, in order to emphasise their own specific forms of realism/materialism, as well as their respective differences. Furthermore, I am of the view that Meillassoux’s ‘correlationism’ and Harman’s ‘philosophies of human access’ seem to be more illustrative of their own specific and ultimately divergent misgivings with many of the major figures of continental philosophy. On the basis of what I have outlined throughout this paper, it is clear that Meillassoux’s term ‘correlationism’ names the various philosophies who proclaim a fundamental inability to pass beyond the limitations of finite human knowledge. Conversely, Harman’s major qualm with the ‘philosophies of human access’ relates to their inability – or better, unwillingness – to create a speculative ontology which moves beyond the narrow confines of what is given to our all-too-human modes of understanding. The stakes of keeping the two terms separate thus boils down to the question of what kind of future is in store for SR. In the last instance, thinkers who still choose to work within this nascent way of doing philosophy would need to decide whether the basis of this speculative non-idealist gesture should be ontological or epistemological in nature. In other words, contemporary and future speculative realists would need to decide whether to follow Meillassoux in proceeding epistemically by seeking a path towards a non-correlated knowledge of the absolute, or whether to support Harman’s quest for a speculative ontological form of realism which generalises the thesis of finitude in such a way that it refuses to take the argument of the ‘access philosopher’ seriously. ^[83]