Levi's Object Oriented Ontology - Virtual Structures and Local Manifestations

- Levi's Object Oriented Ontology -
Virtual Structures and Local Manifestations
_______________________________________________

   Levi has posted here a really good answer to my post from Saturday, articulating how his own version of OOO might address some of the core issues raised by my inquiry. I think it is a good beginning for the purposes clarifying some of the specifics around some tightly knit questions about individuation and the like, in their metaphysics. I think Levi does very well in beginning by distinguishing between metaphysical and epistemological realism.

More fundamentally, I think it is imperative to distinguish three determinations about real objects: that they are (existential), how they are (qualitative), and how many they are (quantitative). Following Levi's presentation, Bhaskar's argues first that real objects qua partitions of the world must exist: scientific practice happens on the basis of manifestations, and objects are precisely the conditions for these local encounters. Bhaskar also allows us to know that objects should have an endogenous structure in order to condition the possibility scientific practice. As Levi acknowledges, this leaves as epistemologically inaccessible from certainty the unified qualitative identity of particular objects qua virtual wholes (how they are), and their numerical status (how many they are).

    Nothing here, it should be noted, yet explains that the world must be external to the field of the subject, however, but only that the world must be objectified, differentiated and structured in order for there to be scientific practice. This is not at all unlike Kant, who claims that causal necessity and the possibility of scientific practice can only actually obtain on condition of the categorical determination that differentiates the field of sensibility of space-time for the subject of representation. For Bhaskar the isolation of entities and can only occur because the world is externally so divided, insofar as there is an asymmetry   between an object and its powers, i.e. the object can be 'out of phase'. This will turn out to be problematic for the externalist thesis, as I shall indicate below. But let us move on.

Let us begin with the last of these points, addressing how the withdrawal thesis, which anchors a local manifestation to the real object, is to be justified through condition (1), i.e. that objects can be out of phase with their qualities. It is important to notice that this doesn't determine that by necessity every object will actually be out of phase with its qualities, and therefore that there must always be incongruence between the real object and the sensual counterpart, or between the potential virtual structure and its actual local manifestation. It rather determines that it can never be necessary that an object is fully actualized, i.e. that it locally manifests the entirety of its powers. The object may thus be out of phase, and this contingency explains both the possibility and interest of scientific investigation, since it an object were fully transparent at every time, no such inquiry would be required. Of course, there's a further unstated premise in Levi's account here, which states that an object may never manifest itself completely at any point in time: there are always aspects of it which remain withdrawn. It should be underlined that this specific point doesn't emerge in Bhaskar's argument, as presented by Levi in his post.

In passing, I wonder how much of Bhaskar’s account as presented in Levi's summary is endorsed by Graham. In any case, it is a very clear formulation to start the discussion! With respect to the connection between their two accounts, I would for example inquire about the following formulation:

“Within my framework, all local manifestations are acts or activities of objects or generative mechanisms. Here what we ordinarily refer to as properties or qualities are doings on the part of objects. The red of a billiard ball, for example, isn’t a quality that that billiard ball has or is.” 

   I preemptively anticipate here that an asymmetry is sketched between Graham and Levi’s accounts, insofar as for the former it would be strange to say the real parts/qualities of the object are something the object does but is not, insofar as for Graham there is a sensual object and sensual parts of the object. Of course, these do not belong to the real object which is its host, even if they do belong to the unified intentional real object which emerges from their relation. Alternatively, if Levi means that the acts of objects are what in Graham’s account correspond to the sensual objects/parts caused by their real counterparts, both accounts might be reconciled. But Levi would still need to say that the produced result, the actualized local manifestation, is an object in its own right, like Graham's sensual objects, which I think is not really his own position. But in this I digress. 

   My next observation resides particularly with respect to Levi's account on the apparent epistemological consequences that follow from these metaphysical theses. Particularly, the consequences that follow given the epistemic inaccessibility of how real objects / virtual structures must correspond to their real powers, as inferred through their local manifestations (quality); and how many real objects correspond to local manifestations(quantity). We are thereby to inquire about how local manifestations are suitable indexes to infer integral, real objects as their causal, virtual anchors. For example, the redness in the billiard ball is by implication the redness of a virtual power in a real, virtual object. This follows from the thesis that objects are structure and the world is differentiated, and the possible gap between an object's powers and its actual manifestations. But this doesn't yet help us in determining the qualitative identity of this virtual object as a whole, or indeed about whether the local manifestation corresponds to a virtual power in one virtual object or many.

 Levi seems perfectly at ease with accepting this result. He claims that indeed we have no resources to attain certainty about the qualitative identity of virtual structures, or about the integrity in the real counterparts to the local manifestations we perceive. So, against such a model, construing knowledge around certainty appears an incommensurable demand for thought.

   Yet I think my worries are here not so much about certainty as a condition for knowledge, but about the possibility of degrees of adequation. If there are no epistemic criteria to distinguish which descriptions/perceptions/actions might be better suited to coin their real counterparts than any others, then it seems we are delivered back into a form of correlationist agnosticism about the real, where the latter is thinkable, perhaps even as the necessary anchor for our experience, but never known in its ontic specificity. Therefore, the virtual powers we take to be manifested in actuality might correspond to anything our verbal stock of terms and descriptions might want to stipulate is in the 'great outdoors'. There might be a theory about how experience or experimentation  isolates objects in a way that restricts the scope of individuation of virtual structures on the basis of the powers manifested and thus coined as potencies in their real hosts, but this is a tricky issue for OOO given the irreductionist thesis. I'm not sure Bhaskar's scientific realism here can help Levi anymore than Kripke's naturalism can assuage Graham. 

I would consequently take issue with the following claim:

“As we saw in the previous section, experiment involves situating entities in controlled and isolated settings. If this is to be possible, it follows that the world must come in chunks. The entities of the world must be differentiated or independent. Were this not the case, then it would not be possible to isolate entities so as to conduct controlled experiments on them.”

      Bhaskar’s argument is probably here more sophisticated than what Levi has schematically presented, so I give him the benefit of doubt. However, while the world is said to come in chunks, this is only insofar as the local manifestations we witness in experience must entail virtual counterparts as their hosts, and since experience is salient with respect to a multitude of such manifestations, by implication the world must be likewise exhibiting structural complexity.

Nothing yet tells us about how local manifestations become anchored on real objects qua unified virtual structures, however, so that it could be said that there are many chunks qua virtual objects, rather than a singular, stratified, and partitioned virtual object which produces the entirety of local manifestations as locally actualized. This is a real problem, insofar as the ontology of objects presumes to distinguish itself from the processualist thesis by claiming that multiple individuals subsist, and not follow from the single apeiron,  the Deleuzean virtual of multiplicities in perplication, or whatever else. The problem is that without epistemic criteria to determine how local manifestations can be tethered to real objects, either in quality or in quantity, it becomes impossible to justify the claim that the world is composed of many real objects. For all we know, local manifestations might all follow from the singular substantial field, which would reduce it to something akin the Deleuzean differential intensive spatium.This would be in perfect coincidence to the thesis of out-of-phasing, differentiation and structure. The appeal to a hierarchical structuration of the world in emergence does not assuage this worry therefore, insofar as it is perfectly possible to imagine the multitude of local manifestations as emerging from a singular substance, or a problematic field of singular differential points in the Idea, prior to intensive individuation.

 The claims about the necessary transitivity of real objects are likewise unhelpful in this regard, insofar as nothing precludes the idea that what is independent of mind/society/language/experience might be one thing rather than many, or even a non-unified purely differential field of pure potentialities ala Deleuze’s spatium.

   Unlike Graham’s brand of OOO, however, we must notice that Levi’s account entails that the local manifestations of objects can be qualitatively mapped onto their virtual potentialities, while for the former there is a severe ontological gap separating real objects/parts which remain intractable as such from their sensual doubles. In Levi’s case, the local manifestations of real objects can be perfectly said to correspond to a real ‘virtual power’, even if it is impossible still to specify what exactly the real object(s) which is manifested is.  Thus, Levi can claim that:

“The foregoing makes no claims as to whether it is atoms, subatomic particles, organisms, stars, baseballs, etc., that exist. These are questions for actual inquiry and cannot be answered a priori.” 

     Unlike Graham, there is no qualitative abyss which renders the local manifestation by necessity distinct from its host reality, part or object, i.e. they could be balls, baseballs, and Roger Rabbit. This is not to say that the local manifestation is identical to the virtual power it actualizes, but only that it can be said to be congruent with a qualitative determination, as described by our singular terms. Thus unlike Harman, for whom our singular terms and comportments cannot but fail to adequately target the real counterparts or counterobjects behind sensual comportments, Levi’s local manifestations correspond to real powers, without being able to specify the virtual unities to which they correspond. This is the case even though real powers, as Levi claims, are “nothing like their powers”; they are still the manifestation of these specific powers: the distinction is here thus strictly speaking modal, i.e. there is nothing qualitative differentiating virtual powers from actualized local manifestations. It is important to notice the import of this reading.

    So when Kant said that being is not a real predicate, like Heidegger explains, he does not mean that a real chair is identical to an actual chair, insofar as identity is irreducible to quality. He meant that as far as their qualities were concerned they were indistinguishable, only their positioning in a world rendered them distinct. Similarly, for Levi, local manifestations don't bear less or more properties than the virtual powers they actualize, they simply manifest locally these same powers as triggered by a perturbation in the world.

I should specify at this point why the rest of Bhaskar's list of conditions for objects is still problematic under these lights, however. As I see it, neither differentiation, nor structure, nor stratification clarifies numerically why we should advocate a plurality of objects rather than one, nor qualitatively how they must be so.

Differentiation only tells us we can isolate realities to locally manifest themselves, but never what or how they are nor how many they must be. Thus we see redness and roundness in an object we call "billiard ball", but nothing tells us how we exactly we individuate this object. We isolate ‘things’, but nothing is as of yet said about how criteria for differentiation in scientific practice allows for real individuation, to assuage these concerns. I'm sure for Bhaskar scientific experimentation plays a clearer role here: we toy with what is before us, and with time we can map the features or powers an object may exhibit as a whole. Thus the proper name we endow the object with is simply to signify the set of those powers observed by and through experimentation. This, however, seems to solicit the thesis that it science's experimental activity which approximately clarifies and gains putative authority in isolating the powers of objects, and determining their identities, which would make the determination of entities relative to constructions in scientific practice. But of course for Levi this needn't be so restricted: a sculptor may know about the powers of clay more than a molecular biologist. Yet nothing still tells us how we determine that a certain set of powers suffices to constitute a virtual totality in the form of an object. As we will see in the end, Levi's argument here will be to argue for the peculiar powers of particular objects, which remain irreducible to their parts or partitions.

It is also strange at this juncture that Levi mentions that we isolate ‘things’ rather than local manifestations: what makes the differentiation of chunks in the world pertain to objects rather than local manifestations qua subjective appearances, say, or sensual objects and their parts, as in Graham’s brand of OOO? It is also not specified why exactly the differentiation must entail the reality of external objects rather than being individuated by the subjective psyche within a sensible field which might be fully ocnstituted by the mind (as in dreams), or whatever else. We don’t quite know yet why or how isolation entails externalism, i.e. how even those entities which appear to subsist outside their relations can be said to be necessary as such outside their manifestation to the perceiver/thinker. The answer to this, apparently, is that the implication of a virtual structure foreign to the manifestation is what implies the externality of any entity as locally manifested.  But this foreignness to manifestation doesn't entail externality to thought, but merely externality to the individuation apparent in the manifestation.

    It is important to note that as far as potentialities in the virtual being of the object may be ‘out of phase’, there is no argument for the causal correspondence of what is individuated as a local manifestation to a single virtual potentiality, and not just virtual object qua whole. Thus, just like in Harman’s case a sensual object/part was undetermined in its numerical correspondence to its anchoring real object(s)/part(s), for Levi a local manifestation cannot be straightforwardly said to be the manifestation of a single power, unless we specify how our descriptions of manifestations are anchored on virtual powers in a one-to-one correspondence. This would entail that we know that what gives itself as a single local manifestation is qualitatively and quantitatively the same as its virtual power, unlike Graham's account for which this if left open. But this stipulated congruence is as of yet to be explained or accounted for. We can let this go for now.

 Structure likewise only tells us that there must be endogenous powers in the world of scientific practice for experimentation, but not how these structures are unified into concrete realities. It must also be said it is not explained why objects must have structures qua stable duration of a collection of powers, rather than a perpetually changing, totally differenc/tiated field of pure potentialities as static Ideal multiplicities specified by singular points and organized relations (without objective individuality), and which become enveloped by intensive contractions which then give objects, like Deleuze says. I don’t think here Levi has forced the ‘goo-theorists’ into having to provide a justification, since they explain the emergence of stable objects through intensive individuation. At least those of a Deleuzean orientation! But I should stay out of that debate anyway, since it’s not really my point of contention.  Hierarchy and intransitivity I mentioned above, and should need to further discussion for now.

Levi is too smart and knows too well these insufficiencies, so he delivers his own argument for the epistemic necessity of integral objects:

      “It could be that the computer is merely a collection of objects and not an object in its own right. I don’t claim certainty. Many philosophers, influenced– probably unconsciously –by a Cartesian tradition seem to treat certainty as a model of knowledge… When I make the claim that, for example, the Coca-Cola Corporation is an object, I am making both a very strange claim and one that ought to be defended with reasons. Here my argument is that the Coca-Cola Corporation both possesses powers and is interacted with by other entities in ways that are irreducible to any of its parts. The Coca-Cola Corporation is capable of doing things that its parts are not capable of doing. Likewise in the case of the computer. I believe the computer is itself an object, that it can’t be reduced to a mere aggregate, because it has powers and is capable of doing things that its parts are not capable of doing.”

    This is promising and interesting indeed! I’d like to know, nevertheless, how we are to exactly construe powers or potentialities. Given Levi’s account of hierarchy in objects, it must follow that entities within others have distinct powers from the higher end objective ‘genuses’:

“Higher scale entities have an autonomy, independence, or existence of their own characterized by their own generative powers. They cannot exist without the lower scale entities, nor do these higher order powers contradict the powers of the lower scale entities, but they are autonomous entities in their own right, irreducible to these lower scale entities”

     Of course, one can almost trivially say that my keyboard “S” key has powers that my computer as a whole does not have: i.e. its paint can be erased, it can be pressed to target a virtual ‘S’ on my screen, etc. But half of my S key surely also has powers that my S key doesn’t have! It can fit on smaller gaps, it has increased capacity to exert pressure against other surfaces, it can serve as an outline shape for a cubic drawing half the size of the square rendered by its double, etc. Needless to say, the same seems to apply to most partitions I can think of, and using standard like Quine examples we can get all sorts of funny scenarios. Not to mention Frege! Superman could surely do things Clark Kent couldn’t! Maybe this had to do with the psychological effect dressing up as a superhero has on the real bodily host, but this all remains rather fuzzy, since potentialities are never specified as distinctively physical, or indeed as pertaining to any descriptive domain for their individuation. So surely Levi cannot straightforwardly say that both Kent and Superman are local manifestations of the same real object, insofar as the latter is the same ‘embodied’ structure, in the sense of a physically extended body. Maybe Bhaskar could, but i'm not sure about Levi. And surely we don’t yet have a suitable index to distinguish even approximately (not with certainty) how certain qualities may resemble real counterparts or not.

As argued above, the structure attributed to the entity does little help assuage these worries, and that the numerical/qualitative withdrawal of real/virtual objects remain lingering in OOO as a dangerous resemblance to the correlationist thesis in which the in-itself is thinkable, but unknowable. Indeed, in Levi’s brand of OOO, the existence of withdrawing realities is entailed by the admittance of differentiation for practice, as well as the other features Levi mentions. But I think there are still questions to be raised, many of which I’m sure Levi’s justifiably anticipated Democracy of Objects will deliver! Thanks to Levi for the time, courtesy, and attention.

Some Problems With Object Oriented Ontology: Reality, Relation, Knowledge

 

Some Problems With Object Oriented Ontology:

- Reality, Relation, Knowledge -

_______________________________________________________

I – Withdrawal and Relation, Reality and Sensation

I was just thinking about some of my lingering preoccupations and questions about Object Oriented Ontology, which have in some ways remained even after my series of exchanges and conversations with Graham and Levi, through e-mails, at last year’s OOO conference, and in the blogs. I think in the past I might have been a bit too obscure when formulating  these, so I want to reformulate some of my concerns, this time aided by a few diagrams which help illustrate where the problem lies as I see it.

Concretely, my questions arise with respect to the thesis of withdrawal, i.e. the thesis that real objects recede when entering relations to other objects. The latter is supplemented with the thesis that two objects only enter in relation inside a third object, in which the first two become mediated. This third object is endowed with a unity and provides a reductive buffer zone where the two initial objects meet. The real objects underlying the relation, for their part, remain ontologically subsistent even if the relation and thus the mediating object under which it enters with another object is destroyed.

 For clarity’s sake, I will use in what follows the example proposed by Joseph Goodson about myself and my computer. First, we run the hypothetical thesis that there exist two real objects with a subsistent, endogenous structure, myself and my computer. Let us say at one point I am in class, away from my computer, and no seeming relation between the two occurs then. We simply have two isolated substances of the following form:

  That is coarse enough for a diagram! It must be noted that Graham's The Quadruple Object delivers a much more rich and contrived set of diagrams that the embarrassing ones presented here. Hopefully, these modest attempts will suffice for the points I seek to advance provisionally. In any case, things get more interesting in the next step.

  Provided I have understood Graham correctly, the next stage would describe how these two real objects enter into relation. As described above, this involves the appearance of a third object which buffers the two. Expanding Heidegger’s thesis that theory distorts or reduces its intentional object in relation, Graham argues that all relations between objects must perform this occlusion/reduction. Thus, the third object which constitutes the intentional relation between the two hosts is an emergent new object also called real, insofar as it is an entity ontologically irreducible to its parts, and which only obtains when two or more real objects enter in relation. The real objects corresponding to this relation would nevertheless remain thereby withdrawn from what would be communicated/transmitted and given inside the unified buffer of this third and emergent real object.

  One might thereby question whether we should include within this third real object the 'real objects' that constitute their hosts. This is a fairly interesting point in its own right, but has no bearing for my argument here. In what follows I capture the withdrawal thesis by excluding the real objects from the new real object qua emergent intentional relation, insofar as the aspects buffered therein do not comprise the totality of the entities which enter into the relation. So, to run with our basic example, let us imagine that I return home after class, and sit down with to my computer and start typing a blog post in it. The next diagram expresses the relation which obtains thus:

APPENDIX - January 28th / 2011
I know the arrows for 'translation' indicated here are debatable for the reasons which follow from those outlined above, but I don't think their placing affects the argument I shall offer below in any case. Also, this diagram seems to contradict Graham's claim that relations are always asymmetrical, i.e. a real object only ever interacts with a sensual one, and vice versa. No two real objects ever touch, just as no sensual objects touch. This, however, is problematic as I see it, and warrants a different reconstruction. The reason is that at the very least a considerable number of relations produce a bilateral distortion of their objects: when I type in my computer the latter distorts me and apprehends only a sensual correlate relative to its capabilities, reducing me to some relation specific construction (the fingers pressing it...). But by the same token, my fingers likewise reduce the computer qua whole to a series of partial sensual counterparts, i.e. the keys which I press, the feel they produce, the screen I watch. This being said, this creates an issue if we want to say that the intentional relation is a single real object, paired with the thesis of asymmetry. This is because if the real me is inside the relation paired to its sensual related term, we would surely also need, at the side, the complementary couple of the real computer with its sensual me. But this seems awkward, since, what we get in that case is rather TWO relations whose unity is far from obvious. This would seem to imply that such cases imply two different emergent real relations obtain, which I do not think is what Graham advocates. If one constructs this bilateral distortion in a single emergent relation, then we have two unrelated halves within the same emergent object, in which case there seems nothing to justify their intrinsic unity.

For these reasons, I believe the thesis of asymmetry fails to capture the essential problem of withdrawal and relation: all an object ever encounters is an aspect of another, and by the same token, no object as a whole ever encounters another fully. Objects distort each other bilaterally.  When I type in my computer I surely am distorted by the computer and only part of me becomes invested in this relation; just as only part of the computer ever interacts with this part. The real totalities presumed as causing these sensual counterparts which encounter each other are never given to any of the terms. Likewise, unless all objects or some objects (cognitively endowed systems, i.e. humans) can epistemically access their total being in every occasion of relation by some form of privileged access, beings withdraw from themselves in every case, and not just from others. In any case, the crucial point is that in every relation, what is given to each of the terms is never the real object in its totality, but only a translation/distortion. This raises interesting questions about how self-relation works in OOO, but i'll leave that aside for now. With this in place, we can proceed to review the main qualms and quarrels with Graham's account.

II - The Qualms and Quarrels 
The crucial aspect about the second moment is that in my relating to my computer there is a) a third real object constituted as the real Daniel-PC unity, and b) that the real Daniel and the real PC withdraw from what is given to each term within this relation qua unified object. What this means is that what is given to a term in the relation is never strictly speaking another real object as such, but some aspect(s) of it which gets translated, and which becomes relative to the particular network proper to the objects as emergent, unified reality. I know that the jargon of ‘translation’ is really Levi’s, but it is useful here to indicate that what the Daniel-PC unity unifies is not the real objects of their hosts as such, but the sensual doubles to which they become reduced upon encounter.

   Thus when I type in my computer the keyboard relates to a reduced aspect of my whole (the fingers touching it) while most of me remains withdrawn from what this relation relates. By the same token, the keys themselves as registered by my fingers whilst typing only constitute a small part of the whole contrived circuitry, processes and parts which surely make up the computer as a whole. No matter how apparently exhaustive the relation may be between two objects, in their relating their substantive realities withdraw, and only sensual images appear to each of the terms. Distortions are given to distortions; aspects are given to aspects. Or to use Graham's phrase: real objects never touch each other directly. The function of the third real object is therefore to create a 'linked complex space' wherein specific aspects of each real object get translated by its counterpart, given as a new (also emergent) sensual object. It should also be noted that in Graham's dualism of sensual/real entities one may also distinguish between substances/parts. Therefore just as much as the real-PC withdraws from my mediation in relation to the sensual keys parts and sensed PC whole as I distort it, there (may be) real key-parts and a real PC whole that withdraws from my distortion. The same obviously applies to the computer's/keys distortion of my own substantive unity and parts. The withdrawal of entities thus applies in the case human relations as much in practice as it does in theorizing. Extending the Lacanian thesis of the non-sexual rapport, or the Heideggerean thesis of ontological 'forgetfulness', OOO advances a thesis about the non-rapport between any real objects.

Here is where I find that some very rudimentary questions can be raised, in spite of Levi's recent proclamations about how OOO has been circumspect in providing support for their claims. The first obvious observation concerns the status of real objects. Since every time I think about, type in, or generally relate to my computer, either in practice or theory, the real object in relation to me withdraws, how do I know that it is, in fact, one real PC that is withdrawing and not a multitude of PC-Parts, or of qualitatively different real objects altogether? More specifically, since every time I think/act towards my PC this will be towards a sensual distortion of the object, how can I ever know anything about the structure of real objects as such?

Graham’s answer to this crude objection is simply that we cannot know anything with certainty about the structure of particular real objects. In our first correspondence, as elsewhere, he claims that we nevertheless could bear standards 'for better or worse' at a loss for such certainty, and given the fallibility of knowledge. But this is dodging the issue; since given the irreductionist thesis advanced by OOO, no set of descriptions and no ontic register is said to gain privileged traction before the real.   Just like science keeps revising its stock of phenomena and forces over history, we must accept the overwhelming possibility that those entities and forces to which we endow unity in our relatings might turn out to be in reality totally different from the way they appear to us right now. It is important to notice, however, that this limitation in fact follows in principle from Graham’s thesis of withdrawal, since the real object can never coincide or be exhausted with its sensual double under relation. Thus I never know if there is in fact a ‘real PC’ underlying the sensual-PC given to me within the Daniel-PC unity, or if there rather is a variety of subsistent/independent "PC-parts" as proper ontological wholes, much like Quine’s famous butchered rabbits. No field of discourse or individuating description in fact escapes withdrawal; no activity or approach from one object to another can reduce the abyss that separates them. The 'real qualities' which determine the real object are forever precluded from knowledge, and they remain qualitatively foreclosed from whatever our grasp of sensual qualities may bring, i.e. which means that they couldn't ever coincide in the form of representational adequation. Here is where the theory of 'allure' through metaphor is supposed to do some work for Graham, insofar as it tilts the tension between objects given their irreducibility. But I won't address this here (**there's a comment below which very rapidly runs through this issue).

   The inevitable consequence of this position of agnosticism about the real seems to be that I have to admit that withdrawing from my distortion of the PC given to me as a sensual double there might be no real PC after all. Instead, there might be  anything akin one can  conjure or individuate in the imagination or outside of it. Thus what I take to be a PC might turn out to be a semblance produced by the Cartesian Evil Genius, Roger Rabbit smoking a blunt, a used napkin, Nixon, an Eastern Airlines ticket reservation booth from 1982 still awaiting passengers, a Chinese dumpling filled with a mixture of gun powder and paprika, eleven trillion bottles of expired baldness lotion, etc. Even if the 'real objects' and qualities remain withdrawn due to a qualitative difference from their sensual counterparts, nothing guarantees that they should be given one way rather than another to support the peculiar brand of sensual double. And by the same token, metaphysical description does no better in speaking of trees and beetles when attempting to allude to the real trees and beetles underlying our sensual distortions, than if it chose to speak about Roger Rabbit instead.

This is because even if there are merely sensual objects, as those entities conjured in our imaginings, with respect to the real no term/action/thought is ontologically closer to the real than any other, since the gap is qualitative and global rather than a matter of degree or local. Harman can thereby distinguish between purely sensual and real objects on the basis of ostentation, insofar as the latter anchors our acts of reference on some real correlate(s), but he cannot specify which of our singular terms target correspondent realities behind their sensual appearance, since they all remain qualitatively different from the unspecified real object(s) which withdraw, and which never touch.  Just like Graham states that he would need to 'be God' to know what is the underlying real object withdrawing from my construction of the sun as an astral phenomenon and that of the Incas as a light-weaving God, all the phenomena withdrawing from by our multiple descriptions and comportments remain spectral hosts, anonymous noumena lurking behind the veil of appearances. Without any epistemological criteria to gauge the adequacy of relational terms to resemble or adequately represent their real counterparts, all sensual objects stand in the same  epistemic footing. Of course, Graham knows very well that there are purely sensual objects in my imaginings, fictions and the like; but this doesn't need to occupy us now, albeit it raises its own stock of questions. To forecast a Fregean example: would we say that there is a 'real man' underlying Clark Kent and Superman qua identities attributed to them within our distorting and reductive culture? Or would we say that there are two disjoint realities there, a true schizophrenic split of the real rather than a mere sensual split? If we say that there is one reality behind these two sensual objects, then we must ask what provides us with the knowledge about this unity. And if Graham's answer is that we simply do not know what or how many realities underly each identity, we must accept that the world in-itself remains shrouded mystery, and that we lurk among, after all, mere phenomena, like Husserl anticipated.

Although this position is perfectly consistent with Graham's rejection of 'certainty' as something that can be reasonably obtained, it must be said that it displays a striking resemblance to the correlationist hypothesis according to which the real is thinkable but is unknowable. This is what I have called elsewhere the problem of virulent noumena, i.e. the potential proliferation of real entities as subsistent outside relation, given the lack of epistemological criteria to measure degrees of adequacy between thinking and being, concept and object. It is important to note that this "cognitive barrier" does not merely apply to human comportments or relations, but that any entity will be destined to perform its own proper brand of reduction/distortion, and thus ontological occlusion. Except, of course, the hypothetical God alluded to by Graham in his initial response.

But if real objects withdraw in such a radical way, then one must ask what solicits the thesis of their existence at all? That is, given that all I know are sensual objects, how do I know in fact that there are such things as withdrawing real substances, gaining  access to their general structures, and not the infamous Deleuzean flux of morphogenetic production in actualization, the formless apeiron from Anaximander, the Heideggerean 'being of beings' which is not a being, and such. Everytime I think about 'my PC' all I get is my sensual reduction of whatever realities withdraw from it; but the latter remain utterly intractable to thought. Graham’s answer here, as formulated in the OOO conference here at UCLA, was that confronted with the choice between accepting that there is a single, formless apeiron, or a multiplicity of objects, he advocates the latter. As construed, this seems to indicate that Graham thinks that it is a matter of ‘axiomatic decision’ or perhaps phenomenological clairvoyance that there exist many things rather than one. Thus, my impression is that he would resist a characterization of OOO as dogmatic on the grounds that he simply has chosen to affirm the subsistence of objects; not any more than Badiou’s endorsement of the axiom of the void set is dogmatic to found and launch set-theoretical ontology. Incidentally, I know Graham believes Badiou’s construal of consistency as an effect of the count already constitutes an anthropocentric reduction of objects to some form of bundle of qualities ala Hume, in Badiou’s ontology, and I think Graham critically misreads the latter by conflating consistency with human/subjective-constitution (the former is rather native to structure, which is perfectly objective). But let us obviate this for the time being.

However, the problem of virulent noumena outlined above becomes a further problem for Graham’s theory of relation qua an extension of phenomenological intentionality, outside the human eidetic-cohort. This is because it is not just real objects that remain veiled in mystery from my comportments with the world, but the relatings between these objects themselves. Just as I cannot know what real object(s) underlies the Daniel-PC unity, I also do not know what sensual doubles obtain when two real objects foreign to me meet. This is because every time I consider or interact with an object, this will be through the sensual construal of my own activity, and thus relative to my knowledge and the sensual correlates within the unities obtained by my relatings. How do we even know thus that other objects relate, and that they also individuate their objects by mediating them through sensual buffers? If all I ever know are the sensual objects of my  own comportments, and my own ways of translating objects, then it we are delivered right back to Husserl for whom the intentional object of experience was finally that endowed to with eidetic unity by consciousness. What epistemic warrant would thereby ever allow us to identify how objects relate outside this consciousness in a way that suddenly corresponds to the entities-processes we postulate through our singular terms?

   So when I speak of the beetle crawling up the tree, how can I ever presume to describe this relation without surreptitiously anthropomorphizing it in terms of the sensual objects given to me? This occluding violence of non-human relations follows since I cannot but help myself to the vocabulary correspondent to sensual reductions within my restricted phenomenological sphere. But then it becomes impossible to specify what relates to what outside of me or indeed how it does so; which real objects and relations are adequate to my singular terms and which remain merely sensual fictions without a positive real objective anchor.  Let us here shift in example and imagine me walking back from class and seeing a lizard climb up a tree. This third completed diagram displays coarsely what I take to happen in this third moment of our construing non-human relations:

    Of course, strictly speaking, I couldn't even say I know there to be a 'real me' underlying my own sensual relations, let there be a story of privileged access here which I have missed. But the point is simply here that whatever I construe as a relation between the beetle and the tree can only take the form of a sensual correlate given for me, within a relation to myself, while their real counterparts and their substantive quality identity or numeric quantitative extension,  must remain occluded. Even though the diagram posits two anonymous terms (x,y), there could be in theory an infinity of such realities, of which we know nothing, within our relatings.

    The presumed de-anthropomorphizing role OOO would play in speaking about non-human relations between objects seems thereby vitiated, given the inevitable gap between real objects and qualities on the one hand, and their sensual counterparts on the other. The entirety of the world as specified by our singular terms becomes swallowed by the realm of the sensual, since the real object which withdraws is necessarily qualitatively distinct from what gets transmitted in the relation. It is thus not just that we could be wrong (i.e. that knowledge is fallible), but that we cannot but be wrong given the qualitative gap that obtains due to withdrawal. Knowledge thus seems confined to the realm of the sensual, much like for Kant it remained a category of the phenomenal. The point, in a nutshell, is that even when speaking for relations for other entities, these can only be specified as sensual entities in relation to us. Thus the agnosticism about the real objects can be extended to an agnosticism about non-human relations more generally.

As a result, OOO seems forced to oscillate between correlationist agnosticism, insofar as it affirms the withdrawal of the real in every occasion of relation and thus of knowledge, and a descriptive metaphysics in which the problematic of access to the world as it is in-itself becomes obviated as we deploy our terms and descriptions to match general features of reality. The in-itself thus becomes thinkable but unknowable in its particularity, even if we know it to be there and how it is there generally. There seems to be thus something of a regression implicit here to the pre-critical endorsement of an in-itself separate from the for-us, without grounds to base how in spite of their independence we have access to its general features through metaphysical description. Only metaphysics can tell us anything about real objects, their general ontological features, while the reality of particulars seems proscribed from thought.

    Indeed something akin to this seems to underlie Graham’s Whiteheadean pragmatic deflation of Meillassoux’s circle of correlation, when he simply considers human-world relation as one kind among many. Graham has (or had) a story to tell here about the role ostentation plays in anchoring our sensings to real objects, in fixing the reference of the relating into a real entity. But the problem pertaining to OOO is ultimately unsalvageable through this glossing appeal to ostentation and Kripke. For the latter, it is physical knowledge of spacetime in the natural sciences which determines the endogenous structure of the objects of reference, and saves him from the ontological indeterminacy of Quine; while for OOO no such privileged locus or principle of individuation exists to bridge our knowledge to the real, concept and object. The latter inevitably withdraws from us, as from all descriptive registers, except in their general features known to us through object-oriented metaphysics. How we ever gain access to this general knowledge follows from the axiomatic assumption about the existence of real objects, along with the theory of vicarious causation and withdrawal.

The problem here is finally that in denying that access to the in-itself can ever obtain, it becomes increasingly difficult to distinguish in what way Graham’s OOO constitutes a realism differently than the correlationist who claims reality is out there but remains unknowable (as a limit concept, as it is for Kant; as the 'being of beings' as it is for Heidegger, or whatever else). That Graham is able to describe how real objects are endowed with a general structure and claim that they actually ‘are out there’ seems justified by way of an appeal to the real outside our mediation, in unexplained congruence with the terms of his metaphysical theory. But the question remains: how can we ever know that real things exist and relate, if all I ever grasp is the sensual counterpart of some unavoidably withdrawn reality? In any case, I hope this very crude presentation of my position makes some headway into clarifying why I remain skeptical about the status of OOO’s putative realism.

Meillassoux's Answer to Hume's Problem

MEILLASSOUX'S ANSWER TO HUME'S PROBLEM:
- CAUSALITY AND NECESSITY - 

_____________________________________________________

 Introduction

       In this post I present Meillassoux's alternative answer to Hume’s challenge against causal necessity. I first present in schematic fashion the classical version of the ‘problem of causality’ as elaborated by Hume (2000, 2007), as well as the well known answer famously proposed by Kant’s (1999) ‘Copernican turn’ in the first Critique. I then proceed to diagnose some problems associated with the latter’s reply, and show how Meillassoux (2008) proposes a different answer altogether. Accepting Hume’s disavowal of causal necessity, I explain how Meillassoux lays the grounds for a new brand of what has been termed ‘speculative realism’; where the latter dislodges transcendental philosophy and its methodological restriction to investigating subjective conditions of access to the world. Following this line of argument, we will briefly review how Meillassoux attempts to overcome the correlationism proper to post-Kantian philosophy, i.e. the doctrine which asserts that we can have no access to things-in-themselves, but only to a correlation between us and the world[1].

  I - Hume’s Problem – Causality, Necessity, Transcendence

          Hume’s ‘problem of causation’ asks the following question: how can we establish the causal necessity of those beings or events described by science, given that our experience provides no guarantee whatsoever for it? Meillassoux’s own formulation is excellent: “Can one establish that in identical circumstances, future successions of phenomena will always be identical to previous successions?” (Meillassoux: 2008, Pg. 85) As is well known, Hume’s answer to this problem is to simply deny that causal necessity can be established deductively or be known with certainty, reducing our construal of it to being a function of habit obtained by extending inductively a chain of actual experiences:

        “It appears that, in single instances of the operation of bodies, we never can, by our utmost scrutiny, discover any thing but one event following another, without being able to comprehend any force or power, by which the cause operates, or any connection between it and its supposed effect… One event follows another; but we never can observe any tie between them…  It is allowed on all hands that there is no known connection between the sensible qualities and the secret powers; and consequently, that the mind is not led to form such a conclusion concerning their constant and regular conjunction, by anything which it knows of their nature.”  (David Hume, 1737)

         It is important to note that Hume does not flatly deny that the laws of nature actually obtain; nor does he deny that these rules are real in order to assert their merely ideal character. The challenge to causality does not amount to an idealist sublation of natural occasion, by affirming the latter's eventual subjective constitution. Rather, it states that we can never obtain knowledge of whether these rules obtain or not by necessity. That is, Hume’s position is not presented as being essentially ontological, but rather epistemological. Experience can tell us about recurrent events which occurred in the past, and about the actual occurrences in the present. By the same token, the deliberative logical principle of non-contradiction can tell us that an event p cannot obtain simultaneously to an event non-p, in a given situation. However, experience tells us nothing certain about the future; there is nothing seemingly contradictory in anticipating that things could turn out differently than they normally do. Thus, whatever we extrapolate from the present into the future cannot obtain from necessity, but merely by a function of habit. Let us for the moment run along with this abbreviated, coarse presentation of the Humean reply, and proceed into the thicker issues at stake.

     We should first note Hume’s answer to the skeptical problem challenges in the same stroke the Leibnizean principle of sufficient reason and all instances of rationalist clairvoyance into the real. For the last two, the stability of the laws of nature and the epistemological warrant on causal necessity is guaranteed by the deduction of the existence of a supreme being or principle which, once established as obtaining/existing, guarantees these laws[2]. More radically atheistic than his contemporaries, Hume’s empiricism harbors no supreme principle for being, and thereby holds the epistemological problematic open given the logical possibility of an abrupt change in natural recurrence, which threatens the continuity of scientific practice and its experimental repeatability. The classical example, from the Enquiry Concerning Human Understanding, argues over the trajectory of billiard balls set in motion before an observer:

       “May I not conceive that a hundred different events might as well follow from that cause? May not both these balls remain at absolute rest? May not the first ball return in a straight line, or leap off from the second in any line or direction? All these suppositions are consistent and conceivable. Why then should we give the preference to one, which is no more consistent or conceivable than the rest? All our reasonings a priori will never be able to show us any foundation for this preference. (Hume 2000: Pg, 84)

         This way, in one swift blow, Hume sought to challenge not only the naïve metaphysical realism which presumed to describe the structure of the world in-itself, but also the epistemological naivety of rationalist philosophies claiming we possess privileged access to ‘clear and distinct ideas’, accessible through speculative reflection[3]. Thus Hume’s reduction of causality to habitude paved the way for a skeptically bolstered instrumentalism about science, where the objects and events postulated through the latter’s experimentation become reduced to heuristic fictions, grounded by nothing besides the conjunction of successive experiences, and subordinated to human ends. As far as metaphysics is concerned, the challenge paved the way for the view according to which being-in-itself is not just unknowable, but also unthinkable, and so that the hypothesis of a consistent reality which exists independent of our thought amounts to little more than a ‘limit concept’ at best (Kant), or a speculative fiction at worst (Heidegger)[4]. At its most radical, being becomes entirely identified with and sacrificed to thought; such as in Hegelian Absolute Idealism. In the latter's dialectical movement, the in-itself merely constitutes the point of passage of Nature as externalized objectivity to its sublation in the subjective unfolding of the history of Spirit (ad the in-and-for-itself). Although Hume’s position does not yet endorse the properly ontological denial of the existence of a world indifferent to our conception of it, it already accepts the epistemological constraint according to which experience does not allow us to know anything about it with certitude.

II - Kant vs. Meillassoux – Transcendental Idealism vs. Speculative Realism

     It was Kant who provided the most decisive reply to Hume. As is known, he famously accepted Hume’s delimitation of knowledge to lived experience, while he nevertheless attempted to salvage causal necessity from the deflationary agnosticism about the in-itself. In order to do this, Kant stipulated that since causal necessity cannot possibly find its ground on lived experience, its source must be different altogether; and he locates this ground in the constituting activity of the agent of experience itself, i.e. what he calls ‘the transcendental subject’. In other words, Kant inflects the traditional metaphysical problematic about the structure of the world into the subject, reorienting thought to the question about how must the content given to experience conform to the constraints structurally delimited by the faculties of thought. This is the root of what has been called Kant’s transcendental turn, which amounts to nothing less than a ‘Copernican Turn’ in philosophy:

   “Up to now it has been assumed that all our cognition must conform to the objects; but all attempts to find out something about them a priori through concepts that would extend our cognition have, on this pre-supposition, come to nothing. Hence let us once try whether we do not get further with the problem of metaphysics by assuming that the objects must conform to our cognition, which would agree better with the requested possibility of an a priori cognition of them, which is to establish something about objects before they are given to us.” (Preface to Critique of Pure Reason B/XVI)

       Kant thereby rejects Hume's agnosticism about causality by claiming that empirical knowledge is constituted by a priori subjective faculties. As such, experience of particular beings or events, and the causal relations between them, becomes not productive to knowledge about things-in-themselves (noumena), but to how things appear to us through these faculties (phenomena). The class of the noumenal, however, must be distinguished from the realm of objectivity: the latter obtains as functions of synthesis given for the transcendental subject, while the former constitutes the limiting concept of being, separable from the way it is given to us through sensibility (CPR, 270-274). Given the externality of thought to being-in-itself, the noumenal becomes a category of action rather than a given as concept for reflection, and is therefore deemed non-objective in that sense[5]. Thus the principles of pure reason Kant also calls the 'conditions of possibility' of all knowledge; it is a matter of "transcendental truth, which precedes all empirical truth and makes it possible." (Ibid; a 146, B185).

      The laws of nature, by implication, are salvaged only as phenomenal occurrences, relative to subjective constitution rather than externally representing the world ‘as such’. This is why the Critique must take enquiry into the synthetic a priori at its core; the synthesis of categorical concepts and formal intuition under which that which is given to experience provides knowledge. But it thus makes no sense to speak of what is given to experience, irrespective of how it is given as experienced. This is what Meillassoux finally calls ‘weak correlationism’, i.e. the thesis that the in-itself is thinkable but utterly unknowable. In Kant’s own version, knowledge of the content of the in-itself is proscribed, although one can know a priori that it must exist outside the strictures of experience. (Meillassoux: 2008, Pg 35)[6].

       Meillassoux intervenes precisely at this juncture, and highlights how Kant salvages causality by making representation in consciousness depend on the stability of phenomena. It is not that causality must of necessity obtain irrespective of the subject, but that if consciousness exists then this can only be because there is causality (in addition to the other categories of the understanding) at the level of phenomena. In other words, since consciousness is consciousness-of-something, then the latter entails that the conditions requires for them must obtain necessarily. However, Meillassoux draws attention to the fact that what the three major approaches to the problem of causality share (the metaphysical realist, the empiricist skeptic, the transcendental idealist) is that they do not call into question the truth of causality as such, but merely question whether thought can furnish a reason for knowledge about its necessity. Even Hume never really doubts the existence of causal laws in nature, but merely denies that reason can ground their necessity, since he “…believes blindly in the world that metaphysicians thought they could prove” (Ibid, Pg. 91). In fact, his argument  claims that we might extrapolate general laws/causes for natural behavior, but that what we cannot unearth is the first principle(s) which makes these laws necessary: “[A]s to the causes of these general causes, we should in vain attempt their discovery… These ultimate springs and principles are totally shut up from human curiosity and enquiry”. (Hume: 1957, Pg. 45)

     However, Meillassoux highlights, the underlying motivation for burdening thought with the task to ground the truth of causal necessity rather than disputing this truth tout court resides not in reason’s venerable teaching, which tells us that the billiard ball can logically take ‘a hundred different outcomes’. Rather, this temptation arises, just as with Kant, from the stability in nature, as made apparent through perception. The perception of such stability/regularity suggests that these principles must be necessary, even if thought is unable to provide the reason(s) for this necessity.

     Meillassoux’s strategy is thereby to turn the traditional retort to Hume’s problem upside down, and begin by assuming the falsity of the principle of causality, rather than its truth, castigating the tradition’s unwillingness to follow reason against the ploys of sensation:

      “How could reason, for which the obvious falsity of causal necessity is blindingly evident, work against itself by demonstrating the truth of such a necessity? It is our senses that impose this belief in causality upon us, not thought…In any case, it is astonishing to note how in this matter, philosophers, who are generally the partisans of thought rather than of the senses, have opted overwhelmingly to trust their habitual perceptions, rather than the luminous clarity of intellection.” (Meillassoux: 2008, Pg. 91)

      It might seem at this stage that this strategy obviates rather than confronts the transcendental philosopher’s answer to the problem. Yet Meillassoux’s point is that there is no putative reason why we should accept the correlationist burden of deliberating over the possibility of grounding causal necessity.  The latter can be safely assumed to be false and thus truly without reason, in continuity with the teachings of intellection. Thus for the 'speculative realist', it is not simply us who are at a loss for reasons for knowing whether causality obtains necessarily or not, but it is thought which openly provides no suitable indication to even assume causality must obtain at all. In fact, thought itself furnishes us with the positive knowledge about the possible fluctuation in the regularities correlationists deem as necessary, and not just an absence of reasons for their proper grounding.

   This positive knowledge is what Meillassoux develops into an absolute, called the principle of factiality, i.e. the principle according to which everything could be other than what it is; or which affirms the lack of reason for anything to be the way it is. Although it lies outside the scope of this essay to explain how Meillassoux laboriously arrives at this principle, we should note that it is drawn as a result of a direct criticism against the insufficiency of correlationist philosophy. For this, he emphasizes how the latter is ultimately forced into denying the autonomy of the ancestral phenomena described by science (such as the fossil-record, which speaks of phenomena before the conditions of the correlation were possible), and thus all knowledge of the absolute or the ‘in-itself’. Correlationism does this by way of subordinating natural spacetime to logical conditions of manifestation for transcendental access, in some form of other (Ibid, Chapter I). Thus, for example, Kant logically subordinates physical space-time to being forms of intuition for the transcendental subject; just like Heidegger logically subordinates ontic space-time to Dasein’s ekstatic temporality. In addition to the agnosticism about the in-itself, the correlationist asserts that thought cannot furnish an ultimate reason for things to even appear the way they do, i.e. we can know those principles which regulate experience, but we can find no reason for their necessity. This is what since Kant onwards has been labelled the 'facticity' of thought: its incapacity to furnish a reason for the principles which structure it. Thus against the idealist, who expressly identifies the in-itself with thought, the correlationist insists on the facticity of the correlation on the basis of the contingency of the agent of the correlation. I can find no reason for me to exist as I do, and this opens the possibility that things might be different than they appear to me.

     Meillassoux’s coup consists first in showing how in the process of denying the autonomy of the phenomena described by science, the correlationist must already have accepted the absoluteness of the facticity of every being, if it seeks to escape the idealist identification of being and thought. For this, it is important to remember that the correlationist, unlike the idealist, purports to advance a position of ignorance, i.e. we don’t know if the in-itself is different from thought, but we know that this is at least possible. For Meillassoux, however, the principle of factiality reveals itself as an absolute, drawn directly from the implicit consequences of the correlationist argument, which cannot be once again relativized to us lest we become absolute idealists. Ray Brassier’s (2007) reconstruction is here profoundly illuminating:

            “Thus [the correlationist] finds itself confronted with the following dilemma: it cannot de-absolutize facticity without absolutizing the correlation; yet it cannot de-absolutize the correlation without absolutizing facticity. But to absolutize facticity is to assert the unconditional necessity of its contingency, and hence to assert that it is possible to think something that exists independently of thought’s relation to it: contingency as such. In absolutizing facticity, correlationism subverts the empirical– transcendental divide separating knowable contingency from unknowable facticity even as it strives to maintain it; but it is thereby forced to acknowledge that what it took to be a negative characteristic of our relation to things – viz., that we cannot know whether the principles of cognition are necessary or contingent – is in fact a positive characteristic of things-in-themselves…” (Brassier: 2007, Pg. 67)

       In other words, Meillassoux’s explicit challenge to the alleged correlationist agnosticism about the in-itself, is to insist that in order to refute the idealist’s claim that the thinkable is what actually is for-us (that there is no thinkable outside of the correlation), one must accept as thinkable the possible eventuality of every occurrence. This includes the eventuality that the correlation might not obtain and that thus the in-itself could be different than the for-us, i.e. its contingency. This eventuality must in turn not be merely relative to the correlation, since if it was it wouldn’t have occurred to us not to be idealists. That everything is possible is not merely a possibility relative to us, but expresses the lack of reason for things themselves to be as they are; the principle of sufficient reason fails.

      These consequences follow, Meillassoux reiterates, from the correlationist acceptance of Kantian facticity: I can find no ultimate reason for my own finite being, just like I cannot know why there are twelve rather than thirteen categories, and so on. This is not to assert, like Hume, that we can find no necessary laws, even within our subjective sphere, contra-Kant. Rather, it simply means that the structure of thought remains a factical given, and that it is not regulated by an ultimate superior ‘first principle’ which endows it with its necessity. The principle of sufficient reason does not hold for being in itself anymore than it does for phenomenal appearances. The falsity of the principle of sufficient reason is thereby extended and positivized by Meillassoux into an absolute, as expressed by the principle of factiality, i.e. the absolute necessity for the contingency of every being:

       “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... The correlationist does the opposite of what she says - she says that we can think that a metaphysical thesis, which narrows the realm of possibility, might be true, rather than the speculative thesis, which leaves this realm entirely open; but she can only say this by thinking an open possibility, wherein no eventuality has any more reason to be realized than any other. This open possibility, this 'everything is equally possible', is an absolute that cannot be de-absolutized without being thought as absolute once more.” (Ibid, Pg. 57)

III – Chance Outside Probability

    These schematic considerations aside, we can say that Meillassoux’s avowal of the principle of factiality is a radicalization of Kantian facticity, while his disavowal of the principle of sufficient reason (and therefore of causal necessity) is a radicalization of Hume’s reduction of the latter to successive conjunction. Taken together, the speculative realist view leads to a picture of the cosmos as a ‘chaosmos’, i.e. a universe where causal necessity is indeed merely a fiction, in which things could change suddenly and without reason. Whereas the correlationist claims that one does not know whether the world itself conforms to our appearances, the speculative philosopher hijacks his argument and positivizes this knowledge into a property of things themselves. Reason thereby dissolves causal necessity, the better to expound a vision of nature as subjected to nothing but the absolute contingency of every being. Every cause might produce any effect whatsoever, assuming it obtains as non-contradictory[7]. This is what Meillassoux calls a ‘hyper-chaos’, the veritable non-metaphysical absolute which sets speculation apart from every theism:

    “We see an omnipotence equal to that of the Cartesian God, and capable of anything, even the inconceivable; but an omnipotence that has become autonomous, without norms, blind, devoid of the other divine perfections, a power with neither goodness nor wisdom, ill disposed to reassure thought about the veracity of its distinct ideas. We see something akin to Time, but a Time that is inconceivable for physics, since it is capable of destroying, without cause or reason, every physical law, just as it is inconceivable for metaphysics, since it is capable of destroying every determinate entity, even a god, even God. This is not a Heraclitean time, since it is not the eternal law of becoming, but rather the eternal and lawless possible becoming of every law. It is a Time capable of destroying even becoming itself by bringing forth, perhaps forever, fixity, stasis, and death.” (Ibid, Pg. 63)

       Meillassoux thus shifts from the traditional question about a) "How we can establish the necessity of the laws of nature?" to the question about b) "How nature manifests stability given the contingency of natural laws and of logical causality". How can laws be contingent and yet the universe appear to be stable? This way, the principle of factiality thereby deflates Kant’s claim that consciousness implies the necessary stability of natural laws, downgrading it to the claim that science and consciousness require stability. From the latter, one cannot infer that causal necessity obtains from stability. Therefore, Meillassoux adamantly rejects the necessitarian argument intrinsic to the Kantian position, while he accepts that scientific practice does depend on the stability of natural laws.

         Yet the principal challenge against this view is what Meillassoux calls the frequentialist implication: the view that if the laws of nature were really contingent then the immeasurable excess of logical possibilities over actual regularities would entail the overwhelming probability of a frequent change in the natural continuity of the world. But, so the argument goes, since this excess obviously runs against the regularities perceived in actual experience, then nature must be endowed with a necessity which guarantees its stability. In other words, the objector claims that if laws could change without reason, they would, and since they don’t, then this must prove their necessity. But although Meillassoux does not contend the fact that there is a stability perceived in nature, which furthermore makes science possible, he contests the claim that laws ought to change frequently if they are not necessary. By the same token, if the die would show the same face upon every cast, the skeptic is lead to the belief that there must be a lead ball in the dice, just like the observer thinks the necessity of laws from the recurrence of particular occurrences. One thinks: “Consequently, there must be a necessitating reason, albeit hidden - just as there must be a lead ball imbedded in the dice - that explains the invariance in the result.” (Ibid; Pg 98)

       To answer the frequentialist implication, Meillassoux clarifies that the probabilistic reasoning it follows must presuppose the totalization of possibilities available to thought. Given a quantifiable number of logically existing possibilities, one draws the radical asymmetry between a priori possible occurrences from the finite reiterations within our own physical universe. The radical asymmetry between the a priori and the empirical is then said to suggest-imply a regime of physical necessity. This derived principle accounts for the antisymmetry  between empirical actualities  and the overabundant field of logical possibilities. Physical necessity furnishes then the apparent stability proper against the probabilistically  homogenous field of  logical necessity (in which every occurrence is equally possible to any other).

        Notice that the frequentialist argument remains unthreatened by a space of infinite possibilities in calculating probability, given that since the advent of Cantor’s set-theory it has become perfectly possible to operate over quantified infinities of variable cardinal order. Against the frequentialist implication, Meillassoux appropriates a further feature of the Zermelo-Fraenkel set-theoretical axiomatics, and denies that probabilistic reasoning is at all applicable to the universe as a whole[8]. In this he follows Cantor’s Theorem, denying that what is thinkable must be by implication totalizable; the detotalization of number in the transfinite is exemplified in the series of alephs. Meillassoux’s gloss here is useful, and can be quoted in full:

          “This succession [of infinities] is known as the series of alephs, or the series of transfinite cardinals. But this series itself cannot be totalized, in other words, it cannot be collected together into some 'ultimate' quantity. For it is clear that were such a quantitative totalization to exist, then it would also have to allow itself to be surpassed in accordance with the procedure of the grouping of parts. Thus, the set T (for Totality) of all quantities cannot 'contain' the quantity obtained by the set of the parts of T... For this totality of the thinkable is itself logically inconceivable, since it gives rise to a contradiction. We will retain the following translation of Cantor's transfinite: the (quantifiable) totality of the thinkable is unthinkable.” (Ibid; Pg 103)

      As a result, Meillassoux denies that a probabilistic calculus of the universe is possible to infer the absurd improbability of the stability of the world obtaining, if physical causal necessity did not obtain at the same time. Since the axiomatics of ZF set-theory opens a space for this detotalization of nature as such, the putative transitivity from stability to necessity is thereby called into question. In doing so, Meillassoux preemptively counters the argument that he has merely chosen a suitable axiomatics to prove his point, since the idea is rather that the open eventuality proper to ZF severs the causal link between stability and necessity, given that at least one such axiomatics allows us to think of its contingency.  He thus sets the stage for a reelaboration of the concept of chance apart from any kind of probabilistic reasoning, delivered instead to an absolute contingency regulated only by the unreason for anything to be, i.e. the principle of factiality.

      Yet Meillassoux accepts nevertheless that there is a limitation to what he has accomplished through his speculative hypothesis: he has not yet shown that the purported de-totalization alluded to through Cantor’s theorem’s actually obtains for nature, but merely draws the consequences that follow from its assumption. In order to establish the truth of non-totalization, one would require the absolutizing of the transfinite, as has been done already with contingency through the principle of factiality, deriving the former from the latter[9]. This would imply discovering an in-itself continuous with the transfinite core of mathematicity as such, something which Meillassoux does not shy away from setting himself as a goal, again in certain continuity with Badiou’s project:

          “We would have to be able to rediscover an in-itself that is Cartesian, and no longer just Kantian - in other words, we would have to be able to legitimate the absolute bearing of the mathematical - rather than merely logical - restitution of a reality that is construed as independent of the existence of thought. It would be a question of establishing that the possibilities of which chaos - which is the only in-itself - is actually capable cannot be measured by any number, whether finite or infinite, and that it is precisely this super-immensity of the chaotic virtual that allows the impeccable stability of the visible world.” (Ibid; Pg 109)

Conclusion

    Our brief recount of the details and problems surrounding Meillassoux’s attempted answer to Hume’s problem is merely tantalizing in its brevity. It outlines a series of speculative issues opened up in After Finitude, and which promise a more, full-blown elaboration in his forthcoming L’ Inexistence Divine. Provisionally, these indications should suffice to open up a space to return to Hume’s challenge, and attempt a retort which does not take the transcendental route, in which contingency frees chance from probability, and in which the cosmos is shorn from all vestige of natural necessity.

    In fidelity to Badiou’s avowal of the power of mathematicity to think of being resolutely, away from conditions of disclosure and the correlationist constriction, Meillassoux’s foray promises the recuperation of the world itself; albeit perhaps a very different world from the one commonly described by philosophical accounts still wedded to the principle of sufficient reason and the belief in causal necessity[10]. I cite here Meillassoux’s concluding lines from After Finitude, in which he sets thought to rejoin the passion for the absolute, long forgotten by the post-Humean obsession with the human:

      “Our only aim 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. If Hume's problem woke Kant from his dogmatic slumber, we can only hope that the problem of ancestrality succeeds in waking us from our correlationist slumber, by enjoining us to reconcile thought and absolute.” (Ibid; Pg. 121)

---

[1] As Meillassoux (2008) and Brassier (2007) show, the determination between self and world obtains through different iterations: the pure ego against the intentional object of consciousness (Husserl), Dasein’s conditions of ontological disclosure (Heidegger), community of speakers against the social consensus (Habermas), class struggle as mediating history in its material becoming (Marx). The crucial point remains, for the correlationist, the denial that anything like the ‘world in itself’ is knowable.

[2] Here the classical examples would be both Descartes and Leibniz. For the latter, the deduction of the existence of God guarantees that the best world possible must exist, and that therefore it is ours such as it is perceived. For the former, it is the existence of God which guarantees that there could be no deception with regards to what I take to be real through experience. Hume’s atheist short-circuiting of the principle of sufficient reason denies thereby that natural occasion can be safeguarded by transcendental skyhooks.

[3] The exemplary case in this regard is again, of course, Descartes.

[4] For this see Heidegger’s famous analysis of tool-being, where he reduces scientific phenomena to present-at-hand- actuality for humans. See: HEIDEGGER, Martin. Being and Time. Translated by John Macquirre and Edward Robinson, Harper and Row, 1962; Pgs. 64-155.

[5] Ibid; B25;

[6] As Kant reiterates, knowing existence, which is a modal attribution, does not concern a ‘real predicate’, since it merely positions the being in question without adding anything to it, i.e. an actual chair is no different as far as its quiddity (or ‘essence’, whatness, Wasein) is concerned than a possible chair. For a poignant discussion of this aspect in Kant’s first Critique and its relevance for his ontology, see Heidegger, Martin; “The Basic Problems of Phenomenology”, Indiana University Press, 1988.

[7] The impossibility of a contradictory being is the first of what Meillassoux calls ‘figures of factiality’, which are direct consequences drawn from the principle of factiality. For his argument in favor of this principle, see Meillassoux (2008); Pgs 69-71.

[8] Badiou (2006), Being and Event, Meditation 26.

[9] Nathan Brown told me in conversation that Meillassoux has apparently already completed this derivation; and is currently working on a theory of signification.

[10] The expression is Graham Harman’s; see his Guerrilla Metaphysics: Metaphysics and the Carpentry of Things, Open Court, 2005.