Illusionism as the default theory of consciousness

As I’ve already noted, Tom asked me in a comment to look at this post by Richard Carrier on what it means to say that consciousness is an illusion.  I’m going to do that, but first I want to talk about this paper by Daniel C. Dennett that Carrier references that is arguing that the default theory of consciousness should be illusionism about it, and in fact the title says that it’s the obvious default theory.  As it happens, I had a long discussion on Jonathan MS Pearce’s blog with a commenter called im-skeptical where in the midst of that long discussion I also talked about the issues with calling consciousness an illusion, and so I’ll repeat some of those arguments here.

Anyway, Dennett’s main argument centers around an analogy with stage magicians.  With them, they do a lot of things that look strange and impressive, but ultimately they are all tricks and at least currently we automatically accept that they are performing illusions and are tricking us into thinking that things are not the way they really are.  So the woman, for example, is not really sawed in half and stuck back together.  There’s a trick that the magician forms to make us think that that’s the way things are, but it isn’t.  The same thing, then, not only can but should be applied to consciousness:  we should think that what phenomenality is telling us about what consciousness is all an illusion, and moreover that should be the default position that anyone takes wrt consciousness.  As Dennett says:

In short, when it comes to stage magic we assume, until positively shown otherwise, that the effects are achieved by some hard-to-imagine concoction of everyday physical causes and effects. Here is where anybody, philosopher or scientist or visionary, is apt to suffer a failure of imagination and mistake it for an insight into necessity. As the noted illusionist Jamy Ian Swiss has said, ‘No one would ever think that we would ever work this hard to fool you. That’s a secret, and a method of magic’ (2007, the e.g.conference, http://www.egconf.com/ videos/how-magic-works). This is not just an interesting observation. It draws attention to a fact that puts all philosophers on notice: nobody would, or should, take seriously a would-be explainer of stage magic who declared that it was just undeniably, intuitively obvious that no possible sequence of ordinary physical events could account for the feat just observed. We philosophical illusionists say that before you run off half cocked with theories about consciousness as one sort or another of ‘real magic’, you should try to explain it all as an illusion engendered by nature.

Now, the first problem here is that in order to adopt an illusionist approach, what you have to accept first is that our phenomenal view of consciousness — which, incidentally, is the only direct view we have to even think this “consciousness” thing actually exists — when taken at face value really, really seems to indicate that consciousness really is all those things that Dennett doesn’t think it actually is, which means that it’s separate from the brain and immaterial and so on and so forth.  But if our experience of consciousness really seems that way, then how can the default position be that it actually isn’t that way, but only appears to be that way?  I ran into the same sort of discussion with im-skeptical, and used the example of the branch bending in water to argue against that sort of presumption.  If I stick a branch in water and it looks like it bends, someone else observing that could not reasonably insist that because of what we know about sticks and what we know about water it has to be an illusion, and so the person who says that it is doing exactly what it looks like it is doing must prove that it’s actually doing that or else we assume that it’s an illusion with an explanation to be provided later.  At a minimum, this runs the risk of, well, doing what a lot of naturalists do and effectively making it impossible to ever prove that the branch bends in water, because any empirical evidence that we could muster towards showing that is dismissed on the basis that the branch simply can’t bend in water and so any observations must be considered illusory.  But even if the opponent does not go that far, it seems very odd to place the burden of proof on the person who is saying that things are really as they appear without providing sufficient evidence that it is indeed an illusion and so not what it looks like.  Surely before we consider something an illusion we would want to have an explanation for why it looks the way it does even though things aren’t that way, or at least have some sort of strong contradictory evidence for it being the way it appears.  For the branch, we can test it by creating a case where, say, if it bent it would touch something that it wouldn’t if it didn’t bend, and that would be sufficient to show that the branch does not actually bend in water.  And, of course, we have fill theories of optics to show that it isn’t the branch that bends, but the light that bends, which is why it looks like it bends even though it doesn’t.

When we apply that example to Dennett’s argument, it can make the situation look better for Dennett than it really is.  Someone could argue that we know the properties of branches and we know the properties of water and so today if someone sees a branch bend in water we could argue that such a thing is impossible based on the detailed scientific information we have for those things and so the most reasonable explanation is that it indeed really is an illusion.  While I’d disagree that we could assume that — since science has to be open to revision by empirical evidence and so really would have to take this new evidence seriously, more seriously than it could by Dennett’s argument — this isn’t an unreasonable argument.  But this is not the case for consciousness.  We don’t have fully fleshed out theories about the brain and how it produces consciousness to appeal to so that we can consider holding to those theories the default and treat violations of those theories as dubious events that we probably could explain away.  What we have are loose correlations between brain states and phenomenal states, and an argument that so far all the things that we know have phenomenal states and so even rudimentary consciousness have brains.  That’s nowhere near enough to make the brain state and naturalist/materialist model the default, given that a) all immaterialists who think that consciousness is not epiphenomenal will accept those brain correlations and b) to be an illusionist Dennett already has to accept that the primary — if arguably not the best, as Dennett would counter — evidence we have for consciousness really indicates that the naturalist/materialist theory doesn’t work.

The reason we can accept, by default, that a stage magician is tricking us instead of performing a “real” illusion is because we have proven that there is indeed a trick to all of those and that those tricks obviously get performed in those contexts.  We therefore are not assuming that stage magician tricks are illusions by default, but are instead making a reasoned conclusion that they are based on the context we have and all the investigations we’ve done we have pretty solid evidence that that is what they actually are.  The same thing applies to branches bending in water:  we did the work and the illusionists manage to indeed meet the burden of proof and show that that was indeed an illusion.  Illusionists about consciousness are nowhere near being able to do that yet.

So when Dennett says this:

Illusionism, I am saying, should not be seen as a lame attempt to deny the obvious, but as the leading contender, the default view that should be assumed true until proven otherwise. (I grant that my whimsical title, ‘Quining Qualia’, lent unintended support to the perception that illusionism is a desperate and incredible dodge, and for that little joke I now repent.)

It actually really does seem to be an attempt to dodge the issue by putting the burden of proof on the simple position that says that things really are as they appear.  It may indeed be wrong and those who deny Dennett’s illusionism may end up looking like someone who insists that the woman really was sawed in half, but Dennett hasn’t done enough to show that that’s the case and so no one should be overly concerned that they didn’t adopt his mere asserted case before the evidence had come in supporting it.  How Dennett talks about the failings of the non-illusionist position only highlights that issue:

I put ‘theories’ in scare quotes because most philosophical theories are just definitions defended, with no aspiration to make novel predictions but rather just to assign the phenomena covered by the ‘theory’ to some category or other. They at best clarify and articulate the implications of the everyday concepts involved. A weakness of such ‘theories’ is that, since they are largely driven by shared folk intuitions, they are always playing catch-up, seeing if they can accommodate newly discovered but unanticipated scientific discoveries, instead of pioneering perspectives from which new empirical questions can be asked and answered.

But can Dennett’s illusionism do any better?  It doesn’t seem like we can get any great empirical predictions from it either, or else he could use those empirical predictions as test experiments to prove his theory correct over all of these other “philosophical” theories.  The only advantage Dennett’s position seems to have, as far as I can tell, is that by insisting that it’s all done by the brain any new thing we discover in the brain technically fits into this theory, but any interactionist theory is going to be able to accommodate pretty much all of that.  If he hasn’t accrued enough evidence that the brain has to be doing these things and that there can’t be some kind of separate immaterial entity doing it by now, what makes him think that his theory will even tell us where to look for that, let alone find it?  And that’s on top of the fact that he has flat-out admitted that he doesn’t have the evidence and so wants to push the burden of proof on his opponents.  He’s not someone who has demonstrated that branches don’t bend in water and so in exasperation is telling those who still insist that they do that they need to prove that in light of all the existing evidence, but is instead someone who is coming over and insisting that branches don’t really bend in water because he strongly believes that they don’t (and, arguably, doesn’t want them to).

The only reason to accept the illusionist theory as the default is if you buy that consciousness just is what the brain does … and that’s precisely what the phenomenal ideas of consciousness and how they appear is challenging.  Dennett cannot assume his position as the default in order to “refute” his opponents and put the burden of proof on them.

This is especially the case when he tosses out his idea of how things could go under his model:

How might this go? When you seem to see a red horizontal stripe (as a complementary-colour after-image of a black, green, and yellow American flag), there is no red stripe in the world, no red stripe on your retina or in your brain. There is no red stripe anywhere. There is a ‘representation’ of a red stripe in your cortex and this cortical state is the source, the cause, of your heartfelt conviction that you are in the presence of a red stripe. You have no privileged access to how this causation works. We have a good theory of how colour perception works, with its opponent processes and refractory periods, so you can probably explain the early or distal links in the causal chain from eyeball to conviction, but you simply don’t know what the proximal or immediate causes are that put you into a state of subjective conviction and the attendant further sequelae (‘and then what happens?’). (And this is true of your access to normal, not illusory, vision as well, of course.) The red stripe you seem to see is not the cause or source of your convictions but the intentional object of your convictions. In normal perception and belief, the intentional objects of our beliefs are none other than the distal causes of them. I believe I am holding a blue coffee mug, and am caused to believe in the existence of that mug by the mug itself. The whole point of perception and belief fixation is to accomplish this tight coalescence of causes and intentional objects.

Now, this isn’t all that clear (and I’ll say more on Dennett’s clarity when I look at Carrier’s post, since Carrier talks a lot about it) but the issue here is that it founds like Dennett is saying that the experience is not the cause of our beliefs, but that our beliefs cause the experiences.  So I don’t come to believe in the mug because I see or feel the mug, but the mug itself somehow makes me come to believe in it which then seems to produce that experience of the mug.  This … is a very, very odd theory.  I’d have to delve deeper into Dennett to really determine if it’s at all credible but the only thing it has in its favour is the idea that, yes, for it to be true all of our ideas about our phenomenal experiences would have to be illusions, and lots of things would go away at that point.  In fact, it would seem like he’d demonstrate that Chalmers’ phenomenal zombies are possible and could be walking around even now, which is probably not what he’d want.  Given how greatly it violates our intuitions and the philosophical problems it creates, I feel perfectly reasonable in saying that he had better have some really strong evidence in favour of this idea before I’ll consider it a contender, let alone the default or most reasonable theory.  And the entire rest of the essay is built around insisting that he doesn’t need to provide evidence.

Let me end with Dennett with this:

We illusionists advise would-be consciousness theorists not to be so confident that they couldn’t be caused to have the beliefs they find arising in them by mere neural representations lacking all ‘phenomenal’ properties. Of course they could; just ask stage magicians — illusionists in the everyday sense — who specialize in provoking false but passionately held beliefs in things that they seemed to see but didn’t see.

I actually in fact hold that we could indeed be caused to have our beliefs by mere neural representations lacking phenomenal properties.  So does Chalmers.  I just don’t think that our beliefs are in general produced by neural representations that don’t have or produce phenomenal experiences, and so think that when I come to believe that the mug exists that it’s because I see it or touch it and that experience causes the representations which then cause the beliefs.  I actually argued this for a graduate course in Cognitive Science once, arguing that the normal procedure is that we have a phenomenal experience which produces a representation that doesn’t have (or have to have) phenomenal content which then produces or can be broken down into specific beliefs that we can act on, but we can get the representations and beliefs in other ways, through reasoning and the like.  What I would argue, then, is that the phenomenal experience portion is consciousness and the representation part isn’t.  But if Dennett wants to insist that that is consciousness, I can oblige.  And if Dennett wants to go along with that move, then any sort of illusionism goes away because we would separate what Dennett considers to be “consciousness” into one bucket and the phenomenal experiences into another, and so everything can indeed be what it appears to be without stepping on each other’s toes.  But he needs to be clear about what he thinks of as consciousness, and all of the people he’s arguing with are not going to accept tossing aside phenomenal experience in order to explain consciousness.