Once More with Content

Greetings to everyone. My hit total passed 10,000 a few days ago, and I’d just like to thank everyone who has been reading this blog since I started it up in August last year. I’m still working my way through writers block, but this was going around in my head, so I’ve put it on paper (so to speak).

Graham recently posted a two part response (here and here) to my last post (here) in our ongoing discussion over the viability of his object-oriented philosophical position. There’s a lot there to respond to, and I suspect that he’s misunderstood some of what I said, and sidestepped some objections I don’t think he’s entitled to sidestep. However, its also clear that I’ve misunderstood him in a few places (and that I still don’t get other bits of what he’s doing, alas more reading required…), so I’ll try to be as even handed as I can.

Getting the most glaring correction out of the way, it seems that my claim that Graham derives the distinction between real objects and their qualities from the Husserlian distinction between sensual objects and their qualities was outright false. I don’t think I can get to the heart of the precise way that Graham does derive this distinction (through Xubiri and Leibniz) without doing additional reading that I’m at present unable to. On this note, although Graham has slightly cleared up my confusion with regard to precisely what he means by the way objects are in excess of (or at least distinct from) their qualities, it has created certain further confusions as to how exactly the distinction between essence and eidos works, but I suspect this is another thing that needs to be solved by more reading on my part.

An additional thing I will mention is that Graham took issue with my reading of Kant, and I think he’s wrong about Kant, at least as far as I can see. Kant has a far more intricate account of experience than most give him credit for, and the issue of the unity of an object in experience is a matter of particular importance for him (although, I would add, Kant certainly handles it in more logical terms than Graham and Husserl are inclined to). However, I’m not going to tackle this dispute here, because a debate about how to read Kant is tangential to the main points to be discussed, and could take up a whole post to itself.

Moving on to my counter-points, I’m going to tackle the problem of causality first, before going over Graham’s issues with my account of phenomenology and how it relates to his project. I had wanted to say something about the real qualities/sensual qualities distinction, but this post is long enough as it is, and my thoughts on the matter aren’t yet as clear as I’d like. Instead I’m going to close by tying what I’m saying into some comments on remarks Graham recently made about eliminativism (here).

1. The Problem of Causation

Now, in my last post I said that I couldn’t see a single problem unifying the occasionalist tradition that Graham has been presenting (specifically, the islamic tradition, Descartes, Malebranch, Leibniz, Hume, Kant, Whitehead and Latour), despite the fact that their positions do share similarities. Graham has taken issue with this, pointing out that they all share the problem of how entities can directly interact, albeit it that this problem is inflected differently among them, e.g., that for Descartes it is only a matter of how there can be mental-physical interaction, rather than interaction as such, and how for Kant and Hume this is more of an epistemological quandry. Now, I’m happy to accept this as obvious. The point I was making was that this problem of how entities can directly interact is not quite the monolithic issue that Graham presents it as. This is because there are different reasons for taking the idea of direct causal interaction between entities to be problematic in each case, some of which are theological (the islamic tradition, Malebranch), others of which are motivated by independent metaphysical (Leibniz and Whitehead) or epistemological concerns (Hume and Kant). The substantial point is that we need to have a good reason to think that there is a problem of causation of the kind Graham poses at all, and that the tradition isn’t unified by some such reason. This isn’t to say that we needn’t provide a metaphysical account of direct causal interaction, just that it is not obvious that this account must involve mediators of any kind.

Now, as I noted in my last post, Graham does have his own reason for holding there to be a genuine problem with direct causal interaction. My issue was that, if the phenomenological perspective Graham takes is to be justified in virtue of its ability to solve this problem, then the reasons given for thinking that it is a genuine problem must be articulable independently of that phenomenological perspective. Graham seems to take himself to meet this constraint:-

My account of the problem is not phenomenological: my account of the solution is phenomenological. Someone could reject the entire phenomenological aspect of my theories and still be persuaded that there is a problem concerning the direct relation between entities: for instance, none of the prior occasionalists were phenomenologists. The occasionalist problem is completely independent of phenomenology, both for me and in reality.

However, given what I’ve just said above, I think Graham is slightly confused here. Yes, one can have non-phenomenological reasons for holding that there is a problem of direct causal interaction, and the occasionalist tradition has a number to choose from, but I (and I suspect most people) don’t accept them, and we’d need some extra persuading to take up any of them. Graham’s own reason for thinking that there is a problem is provided by his account of how objects withdraw from one another (taken from Heidegger, although I’m not entirely convinced of this, again, more reading required…), and although this is not just an account of experience (as it is about real rather than just sensuous objects), it is nonetheless motivated from within his phenomenological perspective. Graham appropriates Heideggerian phenomenology to demonstrate that objects withdraw from our experience of them, and to generalise this to encounters between objects as such. In essence, he has an account of experience which purports to show that objects withdraw from experience, and thus, if we reject the fundamentals of this account of experience, then we also reject his reason for taking their to be a problem of causation (if not all potential reasons).

I think this point is demonstrated even better by taking a look at another bit of Graham’s post:-

Moreover, I do not accept that the burden of proof is still on me to show that direct relations between entities are impossible. I have published on this at length, and given detailed reasons for why I think this to be the case. And people who claim that direct contact is possible are going to have to claim that one thing can adequately grasp another through contact. But this is impossible: no amount of knowledge about a tree, even if it were God’s own knowledge, is itself a tree.

Again, I have to reiterate my disclaimer that I haven’t read everything Graham has published on this subject, but I think the last two lines of the above paragraph highlight the dependence of Graham’s account of the problem on his phenomenological approach. Here Graham claims that those who deny that there is a problem of direct causal interaction must show how one thing can grasp another adequately, or know another adequately. However, this is only the case if the debate is already understood in the generalised phenomenological terms that Graham advocates, i.e., if we already agree that we must understand interactions between entities in intentional terms. However, that we must do so is far from obvious. If we don’t accept that a direct causal interaction between two entities need be understood in terms of one grasping another at all (nevermind to what degree), then claims about the supposed structure of such intentional grasp (i.e., withdrawal) have no hold on our account of causation.

Now, in the first part of his response, Graham does suggest an alternative reason for the problematic status of direct causal interactions between entities, one which is indeed independent of his phenomenological standpoint. He frames the problem as an extension of the classical problem of the discrete and the continuous. His idea is that if one genuinely pursues the Aristotelian idea that there are genuine discrete substances then one is lead to the problem:-

But if things are discrete, then to some extent they are cut off from one another. And so in a sense the occasionalist tradition is merely being one level more candid than Aristotle, and pushing the discreteness of things to its maximum. If things are discrete (as Aristotle, the occasionalists, and I all hold) rather than emerging from some sort of pre-individual or quasi-articulate flux, then you do have to wonder how they can make contact at all. It’s funny that it took Islamic theology to alert us to this problem, but their motives are essentially irrelevant: what they provided was a model of a universe made up of countless little islands, all linked by God. In some ways, it’s Aristotle to the nth power.

Now, I can sort of see where Graham is going here, but the case isn’t yet adequately articulated. Aristotle himself had no problem with direct efficient causation between discrete substances, and so we need some special, additional insight about discreteness that shows that it must involve the kind of radical isolation Graham (and some of the occasionalists) advocate. Moreover, this needs to be demonstrated in a way that does not appeal to the notion of withdrawal, or any other phenomenological notions for that matter, if this is to be a genuinely independent reason for accepting that there is a problem of direct causation.

There is an additional hint at where Graham is coming from in his response to my claim that positing direct interactions between real entities is no less problematic than positing direct interactions between real and sensuous entities. Graham’s response is to appeal to an ‘existence proof’:-

the contact between real and phenomenal objects occurs constantly. I am a real object, and I am in contact with all the phenomenal/intentional objects that populate my experience. I’m not sure what sense there would be in proving this, because– there it is. The burden of proof would lie on the other side. They would have to show either that I am not a real object, or that there are no sensual objects in experience, and I don’t see either of those points as provable.

I’m afraid that I’m unimpressed by this. This whole point is dependent upon the ontological interpretation of the structure of experience that Graham provides, specifically on the distinction between real and sensuous objects. Its only in the context of this ontological posit that contact between real and sensuous objects is obvious and ubiquitous. An alternative ontology that denied the existence (or reality) of anything such as sensuous objects could treat such supposedly obvious facts as that I am interacting with Popeye when watching a cartoon in very different terms. It’s very difficult to deny that the phenomenal character of our experience is object-directed, even when directed at so-called fictional objects, but its not difficult to deny that this constitutes a special ontological category of sensuous objects opposed to real ones. All one needs to do is deny that an immanent description of the phenomenal character of our experience places any constraints on ontology, i.e., to reject phenomenology as a starting point.

To repeat the point, proving that there are no sensual objects is not equivalent to demonstrating that the phenomenal character of experience is not object-directed. The battle is to be fought over the more basic question of how ontology is to be constrained by experience, not over what the structure of our experience, as it seems to us, is.

2. Phenomenology and ‘Experience’

This leads us nicely into Graham’s problems with my account of phenomenology. The overarching theme of Graham’s response to my criticisms is that they attack a position (or castle) that he doesn’t occupy. To explain this in brief, I attempted in my previous post to provide reasons for rejecting phenomenology, by showing that the kind of content that makes up the phenomenological sphere (what I there called pre-predicative phenomenal content) is redundant. Graham tries to sidestep this criticism by denying that phenomenology posits anything like the kind of pre-predicative phenomenal content I was talking about. I think he’s perhaps misunderstood what I mean here, and so I’m going to try and clarify the objection.

A lot of the difficulty here stems from the term ‘pre-predicative’, which I did hesitate to use in the last post, but went ahead with because it is a term that both Husserl and Heidegger themselves use. Graham seems to have misinterpreted my attack on pre-predicative content as a rejection of the idea that there can be experience that is not theory laden, as if the pre-predicative content of experience is what we retrospectively apply our theories to. This is totally wrong. I accept that, for Husserl, we can encounter things as things in terms of the roles they play in our theories. For instance, I can look up into the night sky and see the moon as a giant spheroid, pock-marked with meteor craters, partially illuminated by the sun, and orbiting the earth at astronomical speed, an experience very foreign to aboriginal tribesmen and the otherwise scientifically uneducated. Graham seems to think that I think that the problem with phenomenology is that it posits some kind of pre-theoretical realm of experience. This would be a problem if it were the case, in the same sense that the Kantian positing of a manifold of raw sensuous intuition is, but it isn’t the problem I’m tackling.

Graham is right to point out that the ‘pre’ in ‘pre-predicative’ really means prior to language or explicit articulation. Given this, the term is highly problematic, because it actually isn’t meant to exclude what we ordinarily think of as predication from the content of experience. For Husserl and Heidegger, I encounter the tree as a tree, and as green, large, healthy, etc. We’d ordinarily understand these different determinations simply as predicates, so the choice of the term ‘pre-predicative’ can be very misleading. Nonetheless, Husserl and Heidegger do try to draw a line between the content of experiences and the content of explicit, linguistically articulated judgments, or assertions. I must admit that I know more about how Heidegger handles this than Husserl, but I can sketch the rough outline of what they propose. If we take my encounter with the tree as an example, all of the various determinations of the tree that I encounter are not discrete parts of the experience, but are (or at least can be) effectively simultaneous. The experience of the tree is initially a unitary phenomenon. Moreover, these various determinations aren’t encountered as pure generalities, but are bound up with the details of their instantiation. For instance, seeing that the tree is healthy is not just a matter of seeing it as an instance of some genus, interchangeable with any other, but is tied to the specific signs of health that this tree displays.

What proper predicative judgment does is thus twofold. On the one hand, it pulls the various determinations of the thing apart so that they may be addressed individually. On the other, it abstracts from the specific details of the thing’s various predicative determinations in a way that facilitates communication. In both cases, it makes explicit what is implicit within the experience. As such, it doesn’t add anything that isn’t already bound up in the pre-predicative content of experience (which isn’t to say that we can’t make assertions about what is beyond our current experience, only that such statements would refer back to some possible experience). It is in this sense that phenomenology takes the conceptual content proper to such judgment/assertion to be derivative upon the pre-predicative content of experience. Anyone can string a set of words together to form a predicative statement, but only those who have the ability to refer the terms back to some possible experience can count as grasping the content expressed by it. We can thus see why the worry that the words we use can become disconnected from experience is a common theme in phenomenology (one most famously expressed in Heidegger’s idiosyncratic account of idle talk).

Now, Graham takes it that my arguments for the redundancy of pre-predicative phenomenal content, and thus for the non-existence of the phenomenological sphere, are dependent upon taking such content to be a form of pre-theoretical, ‘raw’, or ‘primal’ experiential content. He takes it that arguments against the ‘myth of the given’, only have traction against such full blooded conceptions of the given. He claims that the conception of experience that phenomenology in general, and his own project specifically, depends upon, is much more innocuous, and thus not susceptible to my critique. To quote him at length:-

When Deontologistics and his allies say that “the phenomenological realm does not exist,” what they seem to be saying is that “the pre-inferential given does not exist.”

Now, I can understand why they want to say that, and even share that wish. But when I speak of “the phenomenal/sensual realm” I mean something much more harmless than that (and I also take most phenomenologists to be saying much more harmless). All I mean is this: experience exists. Simple. Why is such a banal statement so important to me? Because my sense of reality is of a realm so withdrawn from all access and all contact, that if real objects were the whole story then there would be no experience at all. And yet there is experience. We do not hover in a void of darkness, encountering nothing, as all entities slip away into vacuous concealment. There is in fact an intentional realm. Within this realm we find rocks and mountains, and we also find Popeye and HAL…

I think we all agree that there is such a thing as experience. And I’m not talking about pre-predicative, “given” experience, I’m talking about any experience at all, no matter how scientifically mediated or how idiotically stupefied. And as I see it, the achievement of phenomenology is not to say that there is some privileged layer within that experience that is the “given” part of it, and that all scientific inference must bow down before the given. No, the achievement of phenomenology is to discover that this experience is broken up into objects that vary in their presentations to us. And that is definitely true to an equal degree of mountains, the Higgs boson, and Popeye. All can be considered from different aspects at different times while retaining a certain identity.

Now, I don’t think that what Graham is claiming when he says ‘experience exists’ is as innocuous as he is making out. However, there are a bunch of points to make about the above passages.

First, one shouldn’t confuse the ‘pre-inferential given’ with the ‘pre-theoretical given’, as these aren’t the same thing. The former is specifically about the source of the content (i.e., that it is not arrived at through inference), whereas the latter is about the character of the content itself (i.e., that it not contain reference to theoretical notions). The kind of content that phenomenology deals with isn’t pre-theoretical, but it still is pre-inferential. When I see the tree, I don’t infer that it is a tree, or that it is large, or anything else for that matter, it is simply presented that way. Now, as my previous post noted, we need some account of non-inferential inputs into discourse, if we are to take seriously the contribution that perception makes to the process of justification. The relevant question was whether we needed to posit something standing in between the perceptual stimulus and the observational statement it engenders, i.e., whether we need to posit some ‘given’ content which is presented to the observer, or whether we can focus directly on the observer’s representation of what is perceived in making the statement. The argument given there was that we needn’t do so. The argument didn’t depend on whether what is given is spartanly pre-theoretical or richly woven with references to theoretical notions. It was an argument against givenness as such.

Secondly, it’s a little problematic that Graham unpacks the significance of his claim that ‘experience exists’ in relation to his peculiar notion of reality. The reason for this is that, as I’ve noted above, Graham’s idiosyncratic conception of reality as consisting in radically isolated objects that are fully withdrawn from one another is not independent of, but motivated by his phenomenological approach. I have to emphasise here that I’m not saying that Graham’s conception of reality can’t genuinely be a conception of reality because it is motivated by an account of experience. Such terrible arguments just return us to the stale debates over correlationism. I’m simply saying that Graham needs to secure the foundation of his approach – the phenomenological standpoint – without recourse to the more complicated ontological edifice he builds upon it. The fact that we don’t hover in a void of darkness does not necessarily lead to the conclusion that there is an intentional realm in the sense that Graham intends, populated as it is with ontologically distinct intentional objects. This just reiterates the point I made at the end of the last section.

Finally, there is indeed some loose sense in which we can all agree that there is experience. There might also be some good sense in which we can also agree upon the rough facts about what experience is like, or how it seems (e.g., that it is object-directed, and aspectual). However, such agreement does not automatically translate into agreement about what experience is. The move that some make is to infer that we can at least be certain that experiential seemings exist, but this is really very flimsy. I see no good reason to hold that the character of our experience forces us to give any particular ontological status to its elements. This doesn’t prove that Graham’s realm of intentional objects doesn’t exist, all it does is deny that their existence is guaranteed by the object-directed character of our experience. Again, it seems entirely consistent to accept that the character of my experience is object-directed and nonetheless to deny that there are anything like intentional objects. At the very least, we need a good argument showing why it isn’t consistent.

The real problem here is that what Graham wants to establish when he says ‘experience exists’ is something self-sufficient, in the sense that it can be described entirely on its own basis. Now, this isn’t meant to imply any kind of idealism or solipsism. Obviously, Graham wants to demonstrate that in virtue of the structure of our experience, there must be real objects beyond our experience. The important point is simply that those structures of our experience which imply that there must be reality beyond experience can be described in terms that do not appeal to such realities. I touched on this in my last post when I discussed phenomenology’s traditional indifference to causation. What I meant by this is that phenomenology needs to be able to describe universal structures of consciousness which are independent of any particular causal structure that they might be instantiated in (Merleau-Ponty complicates this slightly, but we needn’t worry about him here). Graham radicalises this indifference, as he extends the intentional structure of consciousness to all objects. For his account to work, the description of the universal structure of the intentional relation must be completely independent of anything like biological mechanisms, lest we be unable to ascribe it to marbles rolling on kitchen tables. What Graham is appealing to is thus not the innocuous fact that all of well-formed homo sapiens have some form of conscious awareness, and that it has a certain phenomenal character, but to the more loaded idea that this awareness consists in a self-sufficient and ontologically privileged site from which philosophy should begin.

Now, at this point some will say that experience can indeed be described in such an immanent way, as if it were such a self-sufficient and ontologically privileged site. This is just what phenomenology does. However, my argument is not that phenomenology cannot do this, but rather that this fact alone does not establish the truth of the ontological privilege it claims for experience. If it were the case that this were the only way to give an account of experience, both in terms of the way it seems to us, and in terms of the role it plays within the process of justification, then there might be good reason to accept phenomenology’s truth. However, the point of my previous post was to show that there are other ways of accounting for these features which are not only adequate but potentially superior (this was the point made by my argument that phenomenology’s account of conceptual content is not sufficient). On the one hand, we can account for the role experience plays in justification in terms of the properly universal structures of discourse, which are constituted by norms that are independent of the causal mechanisms by which they are instantiated, and, on the other, we can account for the way experience seems to us, or its phenomenal character, in terms of the particular causal mechanisms in which both our perceptual capacities, and our capacities for following the aforementioned discursive norms, are instantiated. If this approach (or something similar) works, then although we might be able to provide nice self-sufficient phenomenological analyses of the structure of our experience, they are entirely redundant.

3. Conclusion

This brings me to a related exchange that Graham had (here and here) with John Effay of Rebarbazon (here) about eliminativism. Now, I’ve made my own eliminativist sympathies known before (here), although I come at it from a somewhat unusual angle. I’m not quite as gung-ho as some people about the Churchland’s proclamations about how neuroscience can/will change the way we think and live (which isn’t to say I’ve made up my mind entirely about its potential influence). As such, I’m fine with criticisms of the more radical gestures eliminativists sometimes make about this. However, Graham goes farther than this in his response to John, and implies that eliminativism holds that a completed neuroscience has specifically metaphysical implications. Now, there might be someone out there who does think this, but I haven’t come across them. Graham rightly notes that one would have to first give some kind of metaphysical privilege to the human in order for this to be the case, something which all eliminativists are quite opposed to.

Neuroscience can have absolutely no positive metaphysical implications whatsoever. However, this doesn’t mean that it can’t have some negative implications. From a metaphysical standpoint, eliminativism is a deflationary position. What it does is to suggest that the purportedly self-sufficient realm of phenomenal experience which forms the starting point of much philosophy (and metaphysics) can actually be accounted for (and indeed, can be understood in ways which reveal novel aspects of it) by a scientific study of the causal mechanisms which make us up. In short, it tries to undermine the basis for certain approaches to metaphysics, rather than provide us with new roads to travel down.

To conclude, the overall point I am trying to make is that it is perfectly possible to do philosophy in a way that doesn’t treat experience as sacrosanct. If we are to be coaxed into approaches that do give a form of ontological privilege to experience, then they need to give us good reasons for it, and as of yet, I can’t see any.