Physicalism has never held any appeal for me. ‘It is false that the mind is non-physical’ seems to me plainly false. It seemed obvious to me, even as a student, that physicalism must be false, if it is taken to be the view that there is nothing outside the things described by physics. Obviously there is a difficulty for physicalism here. At which time in its history are we to take the authority of physics about what there is? Not so long ago physics told us that there was no first event, for example, and that there was such a thing as a steady state for the expanding universe. Now we know that both propositions are false.
I have always been more interested, however, not in such details but in the systematic way in which physicalism restricts our ontology, if it does, in the interests of unified explanation. The most obvious restriction is on the mind as such. The physicalist wants to replace the mind and mental states in our theories with a physical entity and physical states. The central state materialist, for example, tells us that the mind is the relevant part of the brain and that mental states are brain states, nothing more. (Of course, since identity is commutative, it also follows that all brain states are mental states, perhaps not quite what is wanted: idealism!) But now we are missing phenomenal properties or qualia. Why would anyone want to endorse a theory of this sort? Leibniz argued for harmony in his sense: a maximum complexity of phenomena consistent with a minimum underlying complexity of explanatory law. Why would anyone sacrifice so many phenomena – all of the phenomena, in fact, in the sense of everything that appears! – for a set of laws and principles which doesn’t even mention them? It seems perverse.
So when in the late 1960s or so central state materialism was replaced by functionalism, it should perhaps have been anticipated that functionalism too would have difficulties with phenomenal properties or qualia. The mathematical and computer science glitz of functionalism blinded people, I think, until fairly recently. In retrospect it seems obvious that if we were probabilistic Turing machines, then we would not see colours or anything else. But we do. So we are not probabilistic Turing machines. Of course a lot of work is devoted to machine vision, and it is important and interesting, for example machine reading of bar codes, or image analysis, including optical character recognition. But do these things amount to vision in the sense in which human beings are said to see colours and other things? It would seem not. A machine vision system will work every bit as well even if it doesn’t see anything. So why is there any need to suppose that it does see anything?
And there you have the problem with all sorts of physicalism. All of them are consistent with the non-existence of the things they say are physical. You could have neurons firing away, and no qualia, no apprehension of phenomenal properties, nothing! You could have just a Chalmers-style zombie. Of course this is hotly disputed by the enemies of the zombies. These unfriendly philosophers say that if you have the right physical substrate, then the zombie must also be conscious, and so zombies are impossible.
The most extreme version of physicalism is eliminative materialism, which claims that our concepts having to do with mentality are defective in a way that ensures that they have no application at all. These concepts represent a folk psychology, every bit as bad in its way as folk tales and folk medicine. But really eliminative materialism is just the cleanest and most logical version of physicalism. Forget everything non-physical, because nothing non-physical exists. There are no minds, no beliefs, no hopes, no fears, no psychological states at all. Neurophysiology will give us better concepts and better classifications – we hope.
If this seems uncommonsensical, it is. I am aware of my mind working slowly, or quickly, or well, or badly, I apprehend my beliefs, I am aware of many of my hopes and fears, and cognizant of other psychological states as well. I know what I am thinking, and I know it pretty directly. So goes the view which I would take as the beginning of wisdom in the philosophy of mind. The physicalist can’t just cry down the elements in the world of which he disapproves, from an explanatory view, and hope to get away with it.
The history of twentieth-century philosophy of mind has been the history of the failure of physicalism. Behaviorism came first, and it missed mental states and phenomenal properties or qualia. In the face of overwhelming logical difficulties, and following up very natural lines of thought in psychiatry and neurophysiology, we got central state materialism. The logical difficulties of this view are also insuperable, but it didn’t matter, because before they could really register in the collective philosophical psyche we got functionalism. But here too the phenomena of mental life were pushed aside, in favor of a sort of high-powered behaviorism – what else is the Turing Test? And finally when all this became clear, and the jig was up, we got eliminative materialism, the logical conclusion of physicalism or materialism. Mental life is as much of a fiction is the life of witches and goblins. But really truly, is it? I think that we are pretty much where we began at the end of the nineteenth century and the beginning of the twentieth century. If this is right, and history repeats itself, can can expect a welcome period of introspectionism to add to the massive progress in physiology and psychology over the last hundred years, made all the stronger by a frank recognition of the failures of physicalism?
Do you think that physicalism is true about photosynthesis?
Apples to oranges.
Eric, I don’t think that physicalism is true about photosynthesis if photosynthesis involves light. If it involves electromagnetic radiation only, that’s different a different story. But my view is that there is such a thing as light. Have you seen Avrum Stroll’s work on the question whether water is H2O? Bees can see in colour, besides seeing in UV patterns on flowers.
I asked not because I’m interested in photosynthesis specifically, but to get a sense for how you would handle more straightforward biological processes. I could have easily asked about apoptosis or DNA replication. If you bite an antiphysicalist bullet about photosynthesis or apoptosis, I wonder how much common ground there will be on more controversial matters.
In your post, you went through the textbook 20th century story (behaviorism->identity theory-> (Turing machine) functionalism) but didn’t address more modern views that would say that consciousness is a brain process like respiration or photosynthesis or apoptosis in any detail. E.g., we have the work of Craver/Gillett/Bechtel about biological mechanisms and explanation: is there an argument that consciousness cannot be such a phenomenon? Do you think all biological phenomena just neatly fall into the 20th century philosophical taxonomy already given (i.e., identity theory etc)?
Sometimes the “biological” perspective ends up being cast as a species of identity theory. I’m not sure I like this, but let’s say this is right. I didn’t see a strong reason to reject identity theory in the above. I don’t say this to be snippy, but have always found the arguments against identity theory to be weak, and this is especially obvious when you try to extend such arguments to standard biological processes like photosynthesis (which is why I asked you about that).
(Ignoring teleofunctionalism for now… 🙂 )
Right, right, I understood that you were not meaning in instigate a discussion about physicalism in connection with photosynthesis specifically. My point was that even there you have to make a choice about physicalism. You can talk about colour and light and heat and all the rest, or you can relegate them, in their full reality, to the mind, as properties of sensations, much as Newton and Locke did. I think the biological perspective you describe is explicitly a species of the identity theory. You wrote, ‘Consciousness is a brain process’. But that just is the identity theory. In fact it is the title of the piece by U.T. Place that kicked off the identity theory in the 1950s!
What is wrong with the theory? I did mention the fact that it makes brain processes into consciousness, undesirable for a physicalist. But there is a huge literature on things like Kripke’s 1972 argument, which convinced a lot of people. Suppose”a” and “b” are rigid designators. They refer to what they do refer to in all possible worlds. From this it follows that ‘a=b’ is contingent. But that means that the identity of mind and brain is not necessary. In that case it is due to the meanings of “a” and “b”. But “mind” does not mean “brain”, even if they happen to refer to the same thing. Therefore the mind is not the brain.
Or again, a has the property of being necessarily identical with itself. It is necessarily identical with a, therefore. Then if a=b, b also has the property of being necessarily identical with a. But the mind could well have been something other than the brain, just as the gene could well have been something other than a bit of DNA. We can’t argue that the gene is DNA as a matter of what the word “gene” means, or what the concept is. The discovery that the gene is DNA is an empirical one and the proposition that it is is necessary.
But there are loads of other arguments, like Chalmers’ zombie argument, or Jackson’s “knowledge argument”.
A few things…
1. Physicalism about biological processes?
So are you saying physicalism is wrong about consciousness, but not wrong about apoptosis, or physicalism is wrong about both? Where are you wrt biological mechanisms? And if you are a physicalist wrt apoptosis, would you be an ‘identity theorist’, or something else in that taxonomy? Whether we talk about light hitting the plant that photosynthesizes or EM radiation hitting the plant, I think it shouldn’t matter, but let’s just say EM radiation (I work with infrared light all the time, which we cannot see, so to me it carries no implications about sensations).
2. Down the Kripkean hole?
I am not swayed by Kripke. To the extent I’d go in for these modal arguments, following Papineau I’m fine to take ‘a=b’ as necessarily true. Yes, because of my epistemic imperfections, as a child I may have thought that this identity was contingent, just like I had really bad pre-theoretic semantic intuitions about the speed of light. This is why I prefer science to conceptual analysis: conceptual analysis seems a horrible guide to the structure of reality.
As you know, none of these conceptual arguments against identity theory is exactly knock-down (same with zombie, Mary, etc) or uncontroversial. But that wasn’t really my point, I was more trying to get a sense of where you saw more recent biological approaches, and it seems you would lump them in pretty squarely with ‘identity theory’ (and regardless of the surface grammar, the expression ‘Consciousness is a brain state’ does not commit me to a facile identity theory, or even necessarily commit me to any identity theory, but let’s leave that for now).
3. What happened to Leibniz?
Ignoring all that, stepping back, I was a bit on board with you in your initial diagnosis that we haven’t really gotten much past Leibniz’s mill levels of intuition lobbing. The problem is that most of these highly technical arguments about rigid designators and how identity claims will work across possible worlds are fun and all, but seem more like exercises in cleverness, with well-worn counterarguments, than serious contenders to convince someone who is impartial.
That is, I was on board with the “simple” approach precisely because I find most of the philosophical arguments weak: zombies, modal stuff, etc.. It is the more root intuitive considerations that still hold sway, the root intuition that pulls people toward Mary, or whatever, even though the arguments are clearly unsound. People feel the pull of the Mill, or feel that there is an incommensurability between the neuronal and the phenomenal, and simply can’t shake it. It is this root that makes people feel the pull of the unsound arguments.
4. The vitalism cautionary tale
Despite the fact that many people feel the pull of such (ultimately unsound) arguments, we see enough examples from history (in particular the history of vitalism) to be wary of getting too metaphysically invested in the grip of their intuitions.
You might say that vitalists were dealing with “easy” problems like reproduction or whatever, but it is actually not that clear–some of them treated biological phenomena as irreducibly teleological, and this incommensurability between mechanistic and teleological events was one of the key motivating factors behind their vitalism. The history of real vitalism is actually very complicated, so I’ll just leave that there. I’m just saying that I know Chalmers has chalked it up to “Oh that was just easy problems” but that would be a glib assessment of the conceptual history.
Which brings us back to the simple approach. You started with the “Aw shucks I’m just here asking simple questions.” Then you end up plumbing these Kripkean arguments just to answer simple questions about respiration, as if they are really enough to settle anything. There seems to be a tension. Do you really think Kripke has improved on Leibniz? To me it seems like so much analytical flim flam…
1. I can imagine that there might be a psychological story to tell about the death of some cells, those in animal or human bodies! So maybe I wouldn’t be a physicalist about it, but unfortunately I know very little about the subject, cell death I mean. But with respect to photosynthesis, about which I only know a tiny bit more, if what you are imagining is a physical world without visual light but only ER then you do have a problem about light qualia. What happened to them, then? Are we to think that the visible world is an illusion or what? Or a property of all the brain processes? Here again, you can describe the physical world without qualia, in which case there is a problem about relating them to what is going on in the physical world, say by making them properties of sensations, and we have the mind-body problem, or you can describe the physical world with qualia, in which case there is a about the relationship between visible light and ER. If you say light is ER, you make light and optical effects invisible, which is odd, since if anything is visible it’s light and colours and edges and so on.
2. The only reason I started on about Kripke was because you posed the question, ‘Hasn’t something happened since the textbook history of the identity theory?’ and ‘Is there an argument that consciousness cannot be such a phenomenon?’ and my answer was, ‘Yes, something happened, for example starting with Kripke.’ Then you objected that this was so much flim-flam. OK, but along with the appearance of functionalism, it was the “flim-flam” arguments that killed the identity theory.
3. About Carl Carver: Why isn’t he just a functionalist? He makes a sharp distinction between mechanisms (which are explanatory) and realizations. The higher-level realization is reduced to the lower mechanical structures. About Gillett: He regards consciousness as a sort of multi-dimensional kind of information-processing that integrates the activity of perceptual systems. This raises the same question is Tononi’s integrated information theory. Is the integrated information the same thing as consciousness? If so, we have the identity theory. Is it the function of the densely packed cells in processing information that is consciousness? If so, we have functionalism. About Bechtel: his emphasis on “interfield theories” just says that cognitive science is complicated with many cross-connections. It’s like the Wild West. But I don’t get a clear unified theory of mind out of this, or even a philosophical view, beyond a graded, cross-disciplinary functionalism. That’s fine, but what happens to the anti-functionalist arguments, especially those concerning phenomenal properties?
4. As I wrote earlier, I think we are still where Leibniz left us, and I mentioned Kripke, Chalmers and Jackson because their work goes some way towards supporting that view. Of course there has been some argument about it, but the arguments are still there for study.
OK thanks for that I have a better understanding of how you see how the technical views fit with the initial picture you started with.
I pushed on identity theory because what moved the physicalists toward functionalism was mainly the multiple realizability arguments. That’s what I was referring to as the weak arguments against identity theory. [The Kripke stuff is very different, and specific to consciousness. His argument doesn’t really apply to biological processes the way that the multiple realizability considerations do.]
So I was just trying to get a sense for whether you thought the recent work that has been much more sensitive to actual biology has work pushed the multiple realizability stuff in an interesting direction. I was actually surprised you went to Kripke, given that you gave a kind of textbook behaviorism -> identity -> machine functionalism trajectory. I should have been more clear about what I was thinking. But then I took the Kripke bait anyway and forget my point..so my apologies.
I do think that the biological view has room for many views, whether they be identity theories, some kind of (teleo)functionalism (which you didn’t include yet, which is why I brought it up), or some hybrid. So I agree that the new mechanistic views can hold a lot of different flavors, and it isn’t as simple as ‘brain theories are identity theories.’ It is an extremely rich and complex world, the world of real biological explanation, and I find it frustrating when philosophers try to shoehorn people (especially based on surface grammar, especially given that we scientists are pretty sloppy with language so for goodness’ sake be charitable) prematurely into 1950s categories like ‘central state materialism’ or whatever. These old categories that invoke all sorts of baggage and preconceptions constructed by people staring at their concepts.
Given that it is clear that biological reality and explanation is much richer than these categories, I was mainly trying to get your take on how the 20th century taxonomy fit onto the newer work. And now I see you probably agree it is complicated, without a simple one-size-fits all mapping. This seems right. Indeed, I think biological reality itself is not univocal about this.
I am poking at this post because I am worried you seem to be overstating your case against materialism. But for the record, I do find the materialist view problematic. For instance, the most promising materialist attempt to block the root Leibnizian intuition, the phenomenal concepts strategy, I find ad hoc, almost a case of special pleading.
That is, with materialism you end up with your own weird highly technical bits that are logically possible but not super compelling at an intuitive level. That is, even if technically they can save materialism, maybe they are just overly-clever analytic flim flam: that’s how the phenomenal concepts strategy still reads to me.
So I usually end up back at something like Leibniz’s mill (or whatever your favorite intuition pump is): not really satisfied with any view, but not willing to go down the dualist road because of the vitalist catastrophy.
That is, I’m agnostic. And my attitude is, it’s 2016, why do we need to be super confident? I was excited because I thought you mind end up doing something like that. But instead, it seems you are trying to go down much more confident routes. And that’s when I tend to jump ship.
I think I’ll stop pressing you on how you see photosynthesis: it seems you keep wanting to pull consciousness in with photosynthesis in a way I don’t quite get: I viscerally recoil from realism/idealism debates, so maybe let’s just leave that bomb there ticking for now.