lunes, 24 de octubre de 2011

Gabriel Catren and the Correlationist Circle

Speculative Physics, Speculative Realism?

       Gabriel Catren's project, in short, is to advance a 'speculative physics', where the latter is to be understood as the a priori determination of the laws of nature, and more specifically those described by physics in quantum mechanics. Following Schelling and Hegel, and perhaps in a broader sense the work of Iain Grant, Catren thus seeks to rehabilitate the idea of a philosophy of nature, where the empirical scope of physics is ontologically grounded upon what he describes as a non-transcendental, a priori foundation.

     It is within the scope of this project that Catren finds himself at odds with Meillassoux's own. Although he openly endorses Meillassoux's claim that the post-Enlightenment philosophical tradition initiated a Ptolemaic counter-revolution against its own pretences, he claims that Meillassoux's endorsement of the contingency of the laws of nature is in fact yet another iteration of the reaction motif which fails to remain true to the lessons of the Copernican revolution. In order to show this, Catren briefly takes issue with what he takes to be a gratuitous dismissal of the possibility of establishing that the laws of nature obtain by necessity. Thus, against Meillassoux, Catren holds that the principle of factiality, which claims that contingency is absolute and not correlational, runs on the unjustified assumption that the necessity of the laws of nature couldn't be established by speculative thought.

    First, Catren disputes Meillassoux's assimilation of scientific-natural knowledge to inferential knowledge. In doing so, Meillassoux ignores that 'inductive reasoning does not play any role in contemporary physics'. Although that remark is stated in passing, it should be said that Meillassoux's does not equate scientific knowledge to induction, but rather is targeting the epistemological constraint that the post-Humean correlationist (and idealist) philosophies claimed to condition scientific activity as such. Thus, while it may be perfectly true that physics today makes no explicit reference to induction or inference, this misses the point; since the question is whether physics can be epistemologically grounded without having its principles underdetermined by the pragmatic-formal constraints relative to the knower; the cognizant pole of the correlational circle which is articulated inferentially. The latter claims that the facticity proper to knowledge undermines any claim at necessity that science could strive to; thus Catren's target should be to dispute facticity itself as an epistemological constraint for knowledge as a whole, rather than to claim physics makes no appeals to inference. For Meillassoux's claim about the contingency of the laws of nature is supposed to follow from the general epistemological construal according to which metaphysical claims (and physical claims by extension) are rendered relative to the experience of the knowing agent, and thus to transcendental conditions for disclosure; this is precisely what the circle of correlation tells us. In this regard, Catren would need to explain why physics in particular is not susceptible to the circle, and this he does not do in the paper, albeit more is to be said about this issue below.

     On any account, Catren's weightier argument comes in later. This argument proceeds in different stages. I will address each of these in order, for clarity:

1) First, Catren claims that Meillassoux conflates the formal ontology of worlds with the question about the necessity or contingency of the laws of nature. That is, he reasons that because the idea of grounding the laws of nature with necessity would require either a metaphysical first mover or an infinite regress, their contingency must follow. Since the actual necessity of the physical world could not be said to follow without the dogma of a metaphysical unexplained explainer, it must follow that there is no way to get from the a priori concept to the existence of actual physical reality, no 'ontological argument' as he puts it. Thus even if it was shown that many possible worlds correspond to reason, this has no saying as to whether laws are necessary or not, or whether an enquiry into the necessity of the laws of physics would be futile by necessity. In short, it does not suffice to establish contingency. According to Catren then, even if this were true it doesn't follow that physics could not consolidate an ontological argument it could still be true that 'a physical theory provides a provisional solution to a particular problem. For instance, quantum mechanics is the best solution that we have at the moment for explaining the objective consistency of nature.' (463) Finally, this leads Catren to Claim that ' is difficult to understand why the supposed impossibility of providing a satisfactory rational global model for the 'topology' of absolute knowledge... would imply the futility of such a project.' (Ibid).

    The problem with this line of argument is that it leaves entirely undecided what this putative 'provisional' explanatory purchase endowed to physics is supposed to supervene on, once it is accepted that modally a plurality of possible worlds prevent an ontological streamlining of principled necessity to actual reality. The sense in which quantum physics or any other theory could be 'best' to describe the cohesion of the universe will necessarily be an attempt to describe reality from the purview of its present cohesion. But this does not seem to threaten Meillassoux, for his point is not that a description of locally operative principles could be produced by science or otherwise, but that their alleged necessity fails to obtain. In order to show that the contingency of the laws of nature fail, Catren would need to show not only that quantum mechanics provides the 'best' salient description for reality as we know it, but that the principles it describes can be established as necessary. For the gulf between the manifold possibilities or possible worlds rendered available to thought and presupposed a priori necessity of a set of laws only occurs in sight of the epistemic constraints set on the knower who theorizes. Claims to the 'partial' and 'fallible' nature of science seems on that account to subordinate Catren's putative claim for natural necessity to a kind of pragmatic instrumentalism: our most cohesive theory is taken to describe reality, but we know that it will eventually be displaced in favor of other theories given our epistemic limitations. But this seems to want to have your cake and eat it: either quantum mechanics is endowed with a priori necessity and thus its laws are not contingent to instrumental progress in science, but perhaps susceptible to specification, or they are subject to such variability in which case science must surrender its pretensions of a prioricity. Of course, Catren will claim that this merely states the inherent limitation of all epistemic traction before the real, and it doesn't solicit a wholesale negation of the necessity of the laws established by thought. But Catren surely thinks more: he thinks it is possible to show that the laws of nature are necessary. And yet since this is the goal of Catren's project rather than an accomplished result, it seems that it is he who is presupposing that such a necessity must exist, rather than Meillassoux assuming that they must not exist.

      For therein lies precisely Meillassoux's wager against Hume: against the skeptic who claims that even if we cannot establish the necessity of causal laws it doesn't follow that causality doesn't obtain, Meillassoux rather asks why should we assume it does obtain if reason indicates otherwise. Thus, Catren cannot simultaneously claim physical laws are endowed with a priori necessity and that at a loss for showing how the former the instrumental cohesiveness of physical description should suffice to justify belief in such a necessity. For the latter Catren would need to show, against Meillassoux, that something like the 'frequentialist' implication holds, such that the pragmatic success of physical description and the 'cohesiveness' it enjoys whilst framing nature in fact suggests that necessity must be rendered plausible. But at a loss for any such argument, Catren's tantalizing promise for a speculative physics seems to preemptively want to discount contingency as an unfounded assumption, while itself resting on dubious super-empirical virtue of physics, which are meant to indicate the 'provisional' state of the science.

      Now, Catren points out that the underlying motivation behind his disavowal of the necessity for such epistemological legitimating of the status of physics concerns the simultaneous endorsement of the principle of reason and the abandonment of the principle of ground. The latter is understood as the transcendental machinery of representation which enforces the circle of correlation in Meillassoux's case; and more generally any form of metaphysical 'grounding principle' which remains refractory to science's subtractive modus operandi. Thus, against Meillassoux's avowal of induction, Catren contends that contemporary physical science does not need to run the gaunlet of having to assume a metaphysical first-principle, or an infinite regress. Quite the contrary, to endorse the principle of sufficient reason adequate to science enforces the disavowal of any notion of ground: "Science does not progress by trying to found itself on a last self-posited metaphysical or transcendental reason, but by trying to absolve itself from any kind of presupposed background." (469).

     However, the ambiguity that we indicated above persists in Catren's account, since on the one hand this deposition of any notion of ground is supposed to account for the perpetual revisability of science, while Catren defends simultaneously the a priori necessity of its laws. Of course, the idea is that the lessons of physics force us to adapt to a post-humanist, de-anthropomorphized conception of the universe, where the valence of objective representation construed as a transcendental problematic is dismissed (on this account it is not surprising to see Catren's work overlap with the vocation of some contemporary post-Deleuzians/post-Landians, like Reza Negarestani, whose project seems at times very close to Catren's). But this seems to render the epistemological demand to explain how cognitive claims about reality, either advanced by science or philosophical speculative physics, is supposed to gain traction on an extra-experiential reality. Although Catren's observation that science's scope pushes us to "understand nature in non-correlational terms" is well taken, it is unclear what this prescribes for philosophy, and whether it suffices to circumvent the question of representational access and so of representation altogether. The putative authority endowed to science to not have to scale its descriptive methods through any transcendental machinery surely does not by itself provide an argument for philosophy about why the transcendental problematic is obsolete, any more than a marine biologist's focus on marine life is not a reason to think we have to stop asking questions about how we know the world. Thus it is not clear that 'the principle of ground' can be dispensed of in favor of a physics-friendly principle of reason without further argument.

     But in any case, Catren is right to insist in that the crucial problem assuaging Meillassoux is that his entire argument depends on hijacking the facticity that the correlationist claims inevitably places in the epistemological circle. This leads to the second point.

 2) Catren states that Meillassoux presupposes the impossibility of determining that the laws of nature obtain to argue for their contingency. This is inaccurate and conflates the generality of the argument, since Meillassoux hijacks in the correlationist argument the claim that it is impossible first to know that the conditions for knowledge are necessary, i.e. thought is factical. This is doubtlessly more general than claiming that the necessity of the laws of nature or physics is impossible to ground, since facticity is targeted at epistemological conditions for knowledge, which are at least not transparently tethered to physical laws. But this is a minor gripe, since the point surely also applies to physical principles, taken as epistemic conditions for objectivation. Still, even if we suppose in a naturalistic register that the conditions of knowledge are those specified by physics, the question concerns whether facticity obtains or not. However, Catren's argument focuses on what he deems to be a confusion of two dilemmas in Meillassoux's argument, which confuses a limitation for knowledge with an idealist determination by knowledge. The two dilemmas are:

 1) If we cannot determine the necessity of rational laws is this because of a limitation of our capacities, or because or their contingency?
 2) If physical laws are contingent, is this contingency correlative to thought, or absolute in itself?

      Accordingly, Catren claims that while Dilemma 1 is a perfectly good one, Meillassoux obscures it by focusing instead on the (fictitious) Dilemma 2. The difference in short is between epistemological criticism and ontological idealism, according to Catren. "To summarize, we can say that Meillassoux's proof... begins with an unquestioned (and probably) false) presupposition (namely, that physics cannot discover any rational necessity in physical laws), and proceeds by means of an illegitimate 'deductive' inference (namely, that of absolutizing a supposed limitation)."

      Again, we should remark that, strictly speaking, Meillassoux's construal of facticity cannot be circumscribed to 'physical laws', since the specific argument against the laws of nature follows rather after the principle of factiality has been argued for, as one of the three 'figures of factiality' (which follows from Meillassoux's argument against the frequentialist implication...) To ask whether facticity is relative to thought or not is not necessarily to ask whether physical laws are necessary or not, but more generally whether any knowable principle could be necessary. Thus the focus of the argument is the facticity of knowledge, rather than the contingency of natural laws; since the latter are in a sense subordinated to the former qua objects of knowledge. We have already seen why Catren's attempt to circumvent the transcendental problematic are insufficient for now, but let us leave this for the moment.

 As we know, Meillassoux's argument proceeds from absolutizing facticity and rejecting that it is a mere correlate of thought since, if it were, we would have accepted the idealist identification of all thinkable possibilities as necessary correlates of thought. In that regard, we should claim that Catren's 'real' epistemological criticism following from Dilemma 1 is part of what is included in Meillassoux's refutation against the correlationist 'possibility of ignorance'.  Recall that in the process of absolutizing facticity the correlationist, as hehijacks the speculative philosopher, claims that Meillassoux has illegitimately confused a mere limitation of our knowledge for an absolute limitation on things themselves. Thus while the speculative philosopher thinks that contingency is necessary lest we become idealists, the correlationist insists that for all we know this contingency is still as conceived for us, while the possibility of necessity obtaining outside our knowledge remains. Thus it would seem that Dilemma 1, as Catren presents it, is actually considered by Meillassoux as a possible objection which arises precisely in the course of absolutizing facticity; and so from within Dilemma 2.

     Against the correlationist rebuttal, as we indicated above, Meillassoux claims that the contingency under which reality appears before thought cannot itself be rendered relative to thought, lest we proceed to identify all thinkable reality as correlated to our knowledge, i.e. it wouldn't have occurred to us not to be idealists if we did not accept of the thinkability of an in-itself non-relative to thought. For the putative 'strong correlationist' stand before the idealist is to insist precisely in that it is possible to think of the correlation not being just as it could be, ; facticity threatens the correlation with a loss for reasons for itself to be, and so opens a possible gulf between the thinkable and that which is relative to thought. In this regard, Catren's argument fails again to realize the true pivotal point in Meillassoux's argument is to be found in his putative hijacking of facticity, rather than in the secondary attack on the laws of nature.

      For how does Meillassoux proceed? First, we are told by the hypothesized correlationist that one can find no reasons to ground the necessity for the conditions for thought which organize the range of knowable phenomena, i.e. thought is factical. Meillassoux then goes to say that this shows that facticity must be an absolute principle of things themselves, unless we accept the infinity of thought inherent in the idealist. But at this point Meillassoux's argument takes a strange emphasis: the argument against the idealist, and the rejoinder to the correlationist 'possibility of ignorance', consists in insisting on the facticity of the agent of the correlation, that is, our knowledge that the very agent who embodies the conditions of thought necessary for the correlation could not be, and that this possibility cannot itself be rendered relative to thought lest we think it immortal. My capacity-to-be-other must be thought as absolute lest we claim, like the idealist, that even the thought of death is relative to thought. I quote Meillassoux in full here:

      "She does so by maintaining that we can think ourselves as no longer being; in other words, by maintaining that our mortality, our annihilation, and our becoming-wholly-other in God, are all effectively thinkable. But how are these states conceivable as possibilities? On account of the fact that we are able to think - by dint of the absence of any reason for our being - a capacity-to-be-other capable of abolishing us, or of radically transforming us. But if so, then this capacity-to-be-other cannot be conceived as a correlate of our thinking, precisely because it harbors the possibility of our own non-being. In order to think myself as mortal, as the atheist does - and hence as capable of not being - I must think my capacity-not-to-be as an absolute possibility, for if I think this possibility as a correlate of my thinking, if I maintain that the possibility of my not-being only exists as a correlate of my act of thinking the possibility of my not-being, then / can no longer conceive the possibility of my not-being, which is precisely the thesis defended by the idealist.

       For I think myself as mortal only if I think that my death has no need of my thought of death in order to be actual. If my ceasing to be depended upon my continuing to be so that I could keep thinking myself as not being, then I would continue to agonize indefinitely, without ever actually passing away. In other words, in order to refute subjective idealism, I must grant that my possible annihilation is thinkable as something that is not just the correlate of my thought of this annihilation. Thus, the correlationist's refutation of idealism proceeds by way of an absolutization (which is to say, a decorrelation) of the capacity-to-be-other presupposed in the thought of facticity - this latter is the absolute whose reality is thinkable as that of the in-itself as such in its indifference to thought; an indifference which confers upon it the power to destroy me."

       However, we should note that at this stage of the argument Meillassoux seems to be claiming that the capacity of things-in-themselves to be other is established bythe factical knowledge of one's own capacity to be other. In other words, whereas Meillassoux originally claims that his argument does not depend on knowledge of anything actual, but merely considering that which is possible, in order to refute the idealist, facticity appears to be tethered to knowing that one could-not-be; that one cannot find any reasons for thought to exist. However, this by itself does not seem to solicit the thesis that every being will necessarily be contingent, but only that the agent of the correlation, that is one's thought itself, is contingent. This is not to repeat the argument in favor of the possibility of ignorance, since what we are claiming now is rather that while Meillassoux might have claims for the necessity of contingency on the basis of the knowledge of 'ourselves to be mortal' this is at best to know that we are necessarily contingent; that is, that the correlation is necessarily contingent, and perhaps that which appears within it as the realm of 'appearances'. But from this it seems totally illegitimate to conclude that being-in-itself construed as a mind independent reality is necessarily contingent, since there is no logical necessity from my capacity to acknowledge that I am a finite mortal and therefore contingent in being, as is my every thought and phenomenal correlate, to the necessity that everything that is thinkable as being independent of my thought and phenomenal correlates is contingent. The former is to assert that it must follow that facticity is absolute for the correlational agent; the latter is to say that facticity is a property of all things-in-themselves, and not just for the agent of the correlation. Meillassoux slides from one conclusion to the other through by attributing factical knowledge to the agent of a 'capacity-to-be-other', while within the scope of the circle of correlation the most facticity could show is the absolute contingency of appearances along with its conditions for objectivation. The difficulty creeps in the moment Meillassoux must conflate our knowledge of facticity to a knowledge of finitude, in his way to rebutting the idealist.

      But if this is the case then it bears radical consequences for Meillassoux's subsequent 'figures of factiality', contingency of the laws of nature included. If the principle of factiality is to be restricted to the realm of the phenomenal and to the reality of the correlational agent qua condition for phenomenal objectivation, then the three figures of the factial appear likewise circumscribed to the agent and its phenomena. And it is here that Catren's argument might be given some speculative weight: the idea that Meillassoux conflates epistemological criticism with ontological idealism. For now it seems that the contingency of the laws of nature no longer applies to being-in-itself simpliter, but only to the peculiar being under which phenomena are given within the circle. And since now positive knowledge seems circumscribed to the circle of correlation, at a loss for a positive knowledge of the actual, this seems to reactivate the Kantian idea according to which the laws of Nature, as we know them, are relative to appearances, and the domain of phenomena. But unlike Kant's static categorical framework, Meillassoux's absolute seems to solicit the possibility of an absolutely contingent modification in the laws that articulate appearances, while nevertheless remaining silent about the necessity/contingency of being outside the correlation.

      Indeed, it remains entirely open whether such beings even exist or not, and so the decision between realism and idealism seems suspended in favor of a kind of ontological agnosticism, paired with knowledge about the absolute certain of the contingency of the phenomenal. But this seems like transcendental idealism with a vengeance: we cannot know if being is outside thought, or if it would conform to any necessary laws, let alone whether these resemble our knowledge and appearances. At the same time, we are also unable to establish, a priori, that a given categorical framework for the stability of appearances obtains within the circle itself. Thus the laws of nature, taken in its restricted sense as the laws of appearance, are contingent in form as well as in their existence, i.e. that there is no reason for things to appear as they do entails that appearances could very well not appear, or that they could suddenly appear under radically different modalities than we have known thus far.

      The startling conclusion is that Meillassoux's absolutization of subjective facticity rehabilitates the possibility of the necessity of the laws of nature, understood as the laws of a mind independent reality, like Catren stipulates, but for this very reason forecloses the possibility of there being any knowledge of such laws, let alone a priori knowledge, or of their necessity. The only thing about the in-itself that I'm allowed to solicit is a nightmarish hybrid of Descartes and Heidegger: the transparency of my thinking, and the certainty of my death. The status of speculative physics would thereby seem, on that account, merely constricted to a regional study for the transient stability of phenomena presented under given conditions, and to the extent that a metaphysically fragile cohesion allows to discern. Every knowledge of the relation between the in-itself and the for-us is foreclosed in principle, excepting the knowledge that subordinates my being to the possibility of death as the one entity who I know must exist of itself. Thus the conditions for objectivation which remain the sole 'realist' ground for thought remain irreducible to the phenomena yielded within the correlational circle, but for this reason envelops thought as the dream of an opaque shadowy subjectivity which, for all we know, might lay suspended in utter solitude, shrouded by nothing but void.