The “interaction problem” is traditionally regarded as the main objection to Descartes’ brand of dualism. I’ve discussed it many times here at the blog, and of course it is addressed in my book Philosophy of Mind. The problem concerns how a res cogitans or “thinking substance” and a res extensa or “extended substance” can possibly have any causal influence on one another given the way Descartes characterizes them.
There are various ways the problem might be spelled out. Sometimes it is framed in terms of the idea that it is mysterious how something with no length, width, depth, surface, spatial location, or any other physical attributes (all of which a res cogitans lacks) can get in contact with something that does have such attributes (as a res extensadoes). Sometimes it is framed in terms of the idea that such interaction would violate the law of conservation of energy, insofar as for a res extensato influence a res cogitans would seem to require the physical world as a whole to lose energy, and for causal influence to go in the other direction would seem to require the physical world as a whole to gain energy. As I suggested in a couple of relatively recent posts (here and here), the best way to understand the interaction problem is in terms of the idea that the Cartesian conception of mind and body makes their causal relationship comparable to demonic possession, and cannot account for the unity of the human person.
But here’s yet another way to think about it. The Cartesian conception of mind and body, at least when worked out consistently, arguably makes of res cogitans and res extensa two worlds that are so self-contained and complete that there is simply nothing for either one to do vis-à-vis the other. Each is like a fifth wheel relative to the other.
Hence, on the one hand, we have the neo-Cartesian notion of a “zombie” (in the philosophical sense of the term) as the natural consequence of the mechanical conception of matter that Descartes, along with other early modern philosophers and scientists, made central to the modern understanding of nature. I say “neo-Cartesian” because the notion of a zombie is not explicit in Descartes. But it is implicit, and appeal to it has become a key move in the argumentation of contemporary Cartesians. Descartes notoriously thinks of non-human animals, which lack res cogitans, as automata, behaving as if they experience pain, pleasure, and other sensations, but in fact devoid of consciousness. If you add to Descartes’ conception of an animal as an automaton the idea of something which behaves as if it were uttering meaningful speech and as if it were manifesting intelligent behavior but is really just mimicking these things, you essentially have the idea of a zombie – of a creature which is physically and behaviorally identical to a human being but is devoid of any mental properties.
Part of the reason the mechanical conception of matter entails the possibility of zombies is that it takes matter to be devoid of anything like color, sound, taste, odor, heat, cold and the like, as common sense conceives of these qualities. On the mechanical conception, if you redefine redness (for example) as a tendency to absorb certain wavelengths of light and reflect others, then you can say that redness is a real feature of the physical world. But if by “redness” you mean what common sense understands by it – the way red looks in conscious experience – then, according to the mechanical conception, nothing like that really exists in matter. And something similar holds of other sensory qualities. The implication is that matter is devoid of any of the features that make it the case that there is “something it is like” to have a conscious experience, and thus is devoid of consciousness itself.
Another aspect of the mechanical conception of nature that entails the possibility of zombies is the thesis of the causal closure of the physical. The idea here is that everything that happens in the physical world at any one moment of time can be entirely explained in terms of the state of the physical world at an earlier moment of time together with purely physical laws. (Notice that I said explainedby rather than determined by. You don’t need to be a determinist to hold to the causal closure of the physical.) Every physical event, on this view, has a physical cause sufficient to account for it. If this is the case, there is nothing for distinctively mental properties to do, which is why contemporary Cartesians and materialists alike often take an epiphenomenalist position vis-à-vis the mind, according to which mental properties are simply “along for the ride,” as it were, and don’t actually have any efficacy relative to bodily behavior.
The causal closure thesis is arguably the more crucial thesis vis-à-vis zombies, because one could in principle take the view that zombies are possible vis-à-vis phenomenal consciousness but not vis-à-vis rationality. If one did hold this, then one might in principle argue that while there could be a creature physically and behaviorally identical to us that was not conscious, there could not be one that is physically and behaviorally identical to us that was not rational. But if we factor in the causal closure thesis, this becomes untenable. If physical causes suffice to account for everything that happens in the physical world, then that would include apparentlyrational speech activity and apparently rational behavior. Actual rational thought processes qua rational (as opposed to qua physical) would be unnecessary. Hence a zombie devoid even of rationality would be possible.
Then, on the side of res cogitans, we have the Cartesian idea that your mental life could be exactly as it is now even if the entire material world, including your body and brain, were illusory, parts of a hallucinatory deception foisted upon you by a malicious spirit (Descartes’ “evil genius”). This hypothesis reflects what contemporary philosophers of mind would call a radically internalisttheory of mental content, on which the contents of thought and consciousness are determined entirely by factors internal to the mind rather than its relations to anything outside it.
If this radical internalism is correct, then it seems that there is really nothing for the external material world to do vis-à-vis influencing what you think and experience. We have, in effect, the flip side of the zombie hypothesis. On the zombie hypothesis, the world is physicallyidentical to our world, but devoid of phenomenology. On the “evil genius” hypothesis, the world is phenomenologically identical to our world, but devoid of anything physical.
You can see why the Cartesian thesis that some ideas are innate would naturally develop into the thesis that all ideas are innate, as it essentially did in Leibniz (for whom every monad has all information packed into it upon its creation, and thus needs no “window” onto external reality). For if the mind could be in any phenomenological state whatsoever even in the complete absence of matter (and thus the complete absence of any mind-independent world or any sense organs with which to make contact with a mind-independent world) then what is needed in order to determine the content of any phenomenological state (and thus any thought or experience) must already be built into the mind.
So, again, the Cartesian picture of reality leaves us with two self-contained worlds. There is physical reality, which could be exactly as it is in the complete absence of any mental phenomena whatsoever, as in the zombie scenario. And there is mental reality, which could be exactly as it is in the complete absence of any physical phenomena whatsoever, as in the evil genius scenario. So what exactly does matter do vis-à-vis mind, and what exactly does mind do vis-à-vis matter? Interaction becomes problematic because it seems unnecessary.
This also helps us also to understand more deeply why, as I have often pointed out, there is no interaction problem on an Aristotelian-Thomistic (A-T) conception of human nature, even though it too regards the intellect as immaterial. For one thing, the A-T position rejects the thesis of the causal closure of the physical. And the main reason it does so is because the causal closure thesis conceives of causation entirely in terms of what A-T thinkers would call efficient causality. But efficient causality is only one of four fundamental modes of explanation – the others being, of course, formal causality, material causality, and final causality. The causal closure thesis also presupposes a reductionist conception of physical objects (never mind the immateriality of the intellect) that the A-T position rejects. It just gets the basic metaphysics even of the material worldbadly wrong. And of course, from an A-T point of view, the Cartesian also gets the metaphysics of human nature badly wrong, starting with the idea that a human being is a mashup of two distinct substances. There simply aren’t two things in the case of a human being in the first place, on the A-T view, and thus there is no question of how two things “interact.”
Nor does the A-T position have any truck with the internalism and innatism of the Cartesian position, committed as A-T is to the thesis that “there is nothing in the intellect that was not first in the senses.” Angels have innate ideas, but we don’t. For example, we need to abstract our ideas of universals like dog, tree, man, etc. from actual concrete particular material things which instantiate those universals – in which there must be concrete particular material things in order for us to have any ideas about them. Hence there is no sense to be made of our having exactly the phenomenology we have in the absence of any material world.
(Does that mean that angels could be subject to an “evil genius” type hallucination? No, because Descartes’ scenario involves a kind of sensory experience, and angels don’t have that. In fact it is not clear that, given A-T premises, we can really make coherent sense of the evil genius scenario. For the scenario involves sensory experience in the complete absence of anything corporeal. But if there is no corporeality, there can be no sensory experience; and if there is sensory experience, then there must be corporeality. The scenario seems possible only if we don’t think about the underlying metaphysics very carefully.)