Saturday, 17 August 2019

Imaginary Foundations, the Zombie Argument, and Modal Inertness

Recently published work by Wolfgang Schwarz, to an extent I'd thought impossible, offers an explanation of why there seem to be facts of phenomenal consciousness that we can know for sure. 'Red looks like this!', one says, as one inwardly points at the sensation, and one feels there is an indubitable fact here.

The basic idea, as I understand it, is that by having a cognitive architecture that allows us to keep track of sensation by means of belief-like things that we have full credence in, we can do things more efficiently than without such a mechanism. That's only a very rough explanation, but see:

- The main paper of Schwarz's on this, 'Imaginary Foundations'.
- A second paper outlining some of the ideas from 'Imaginary Foundations' in a more basic way. 

Schwarz's explanation of the seeming existence of phenomenal facts that we know with certainty leaves room for various metaphysical views about whether there are facts of phenomenal consciousness, and whether or not they're reducible to non-phenomenal facts. But Schwarz leans towards a sophisticated sort of eliminativism; it's not wrong to call phenomenal reports 'true', but really they don't state facts about the world. Nor do they seem to state a priori necessary facts, like mathematical sentences seem to - these apparent phenomenal facts have the flavour of a posteriori contingent matters, as Schwarz clarified for me in an exchange on his blog (link below). And so we might think there really aren't any facts here.

- See here for a blog post of Schwarz's, and the comments where he clarifies why it doesn't seem right to think of them as like non-external-worldly facts, the way we might think of mathematical facts. 

Now, one thing I'd like to do is explore the prospects of sticking with Schwarz's explanation of the seeming existence of phenomenal facts, but drawing a different metaphysical moral (or perhaps being critical of the very framework in which the apparent metaphysical options appear).

But another thing I'd like to do is run with the whole Schwarz package - the explanation of the seeming as well as the sophisticated eliminativism about phenomenal facts - and see what adopting this package might enable us to say about the Zombie Argument against materialism, the relationship between conceivability and metaphysical possibility, and the like.

Schwarz has already indicated briefly, in 'Imaginary Foundations' how his explanation of the seeming can explain why a philosophical zombie (p-zombie) - a being just like us but without any inner consciousness - might be conceivable. And similarly, how it might seem like Mary, the colour expert who has never seen red, gets a new bit of knowledge when she finally sees red; the knowledge that red looks like this

What I'd like to do, on this basis, is to put the sophisticated eliminativism in the picture and see what this lets us say about the Zombie Argument. And I have a hunch that there's an interesting and, as far as I know, novel position we could take here.

The idea is basically that we could regard these phenomenal consciousness reports as inert with respect to metaphysical possibility. If you have some big description which is true of some set of metaphysically possible worlds, or even just one, then adding or removing phenomenal consciousness propositions - the ones which we have in our minds for broadly computational reasons, although (by our sophisticated eliminativism) they don't really describe substantial facts - won't affect the metaphysical modal status of the description.

This seems to open up a new way of being a physicalist (or, for that matter, being a non-physicalist who believes in God or has other commitments which makes them a non-physicalist, but is suspicious about irreducible phenomenal consciousness). Physicalism (and other metaphysical views which do not posit irreducible consciousness) is often taken to entail that p-zombies are metaphysically impossible. If you need it to be the case that p-zombies are metaphysically impossible, then in the face of the Zombie Argument (see Chalmers (1999), (2009)), it can look like you really only have two broad options:

- Deny that p-zombies are conceivable in a strong sense (a sort of conceivability which is robust with respect to getting more non-modal facts and getting more rational). 
- Deny that (strong) conceivability entails metaphysical possibility.

But, if phenomenal consciousness reports are modally inert in the way indicated above, a third option presents itself. One can accept that p-zombies are as conceivable as one likes, and that descriptions involving phenomenal fact statements, and zombie-like descriptions where these are all negative, can both count as describing metaphysically possible worlds. By adding or removing phenomenal propositions, one just doesn't change which world or range of worlds one is talking about.

This enables one to both avoid setting implausibly high bars on what we should say is conceivable, and to avoid having to reject the idea that conceivability (of the right sort, and perhaps given the right information). One doesn't have to be a pre-Kripke style modal rationalist. One can accept that there are necessary a posteriori truths, but maintain that given certain relevant empirical truths, the modal situation becomes a priori. (I have a paper on this, and some ideas in my PhD thesis and a paper in progress.)

One issue here is that there might be more than one set of notions that are being called the metaphysical modal notions in contemporary philosophy. One set is friendly to moderate modal rationalism, but the more metaphysically loaded set may not be, or may not even be in good standing at all. I've found Rosen's 'The Limits of Contingency' very illuminating on this point. You might want to be a skeptic about the second, more metaphysical set of 'metaphysical modal notions' while being a moderate rationalist about the first set.

Nevertheless, it looks to me like Schwarz's ideas about the seeming existence of phenomenal facts give us a powerful way to be skeptical about irreducible phenomenal facts (whether this is because of physicalist or otherwise naturalistic predilections or just because of more specific suspicions about ideas about consciousness), while maintaining a (moderate, Kripke-proof) modal rationalism. Could there have been p-zombies? Sure, but that's not actually a different way for things to have been!


Chalmers, David J. (1999). Materialism and the metaphysics of modality. Philosophy and Phenomenological Research 59 (2):473-96.

Chalmers, David (2009). The Two-Dimensional Argument Against Materialism. In Brian P. McLaughlin & Sven Walter (eds.), Oxford Handbook to the Philosophy of Mind. Oxford University Press.

Haze, Tristan (2019). Linking Necessity to Apriority. Acta Analytica 34 (1):1-7.

Rosen, Gideon (2006). The limits of contingency. In Fraser MacBride (ed.), Identity and Modality. Oxford University Press. pp. 13--39.

Schwarz, Wolfgang (2018). Imaginary Foundations. Ergo: An Open Access Journal of Philosophy 5.

Schwarz, Wolfgang (forthcoming). From Sensor Variables to Phenomenal Facts. Journal of Consciousness Studies.