Consciousness / mind · philosophy

The Combination Problem (?) for Physicalism

I’m intending to turn this into a more formal paper, assuming that the idea does not wither and die during the research process. This is basically a thought dump to get my brain moving and thinking about things in more detail, and if I am lucky I will learn enough to look back on this and want to delete it ūüėÄ

Back to Basics

I think the proper way to start off this entry is by revisiting the Hard Problem of Consciousness. It seems to be the case that we can have a complete scientific description of the human body yet coherently ask, “Why are all these goings-on accompanied by an experiential aspect? Why does my brain activity – and only certain types of brain activity – result in my subjective experience, the¬†what it is like¬†to be me?” For any brain activity we can, to use that dangerous phrase,¬†in principle¬†describe each physical aspect – for example the activities of neurons, glia, and a host of neurochemicals – without any mention of subjective experience. Imagine how we would describe the mechanical operations of a computer, how we could talk about the electrical activity, the function of various chips, and how it all relates together to produce the UI that shows up on our monitor. We can at least¬†treat,¬†for our example,¬†the human body as this sort of robot, with the brain as the CPU; all behavior can be described in similar terms, exactly as we would describe the activities of a machine.

Physicalism of the mind is, roughly, the position that there is nothing more to the mind than physical brain activity, and that our mental states in some sense supervene on the physical base (though, as Daniel Stoljar points out in his 2010 book, Physicalism, it is not entirely clear what we actually¬†mean¬†by ‘physical’). And, obviously, the existence of the Hard Problem presents a real challenge to this position – how on earth could we consider physicalism to be true if a whole physical description of the brain does not require any mention of conscious experience? An argument based on this issue is the conceivability argument against physicalism.¬†Chalmers formalizes this argument in (to use one of many examples) his paper¬†Panpsychism and Panprotopsychism:

(1) P&~Q is conceivable.

(2) If P&~Q is conceivable, P&~Q is metaphysically possible.

(3) If P&~Q is metaphysically possible, materialism is false.


(4) Materialism is false.  

In this argument he states that P is the totality of microphysical truths about the world and that Q is any phenomenal truth (he uses the example of a person stating they are conscious). The argument is that if we can conceive of a situation where we have the totality of microphysical facts and yet do not have the phenomenal truth then, because physicalism would require that all facts Рincluding phenomenal facts Рare necessitated by the physical facts. Yet, if we can conceive of this situation and conceivability actually does lead to metaphysical possibility (both of these are contested, though I think (2) is the more commonly attacked premise) then phenomenal facts would, in fact, not be necessitated by the physical facts; therefore physicalism would be false. One of the possible outs to this dilemma is panpsychism.


Panpsychism is more or less the thesis that some of the world’s fundamental entities are experiential in nature. If we look at the above case of physicalism we can see that physics describes the extrinsic properties of these entities, and describes the world in relations of extrinsic properties. But what is the intrinsic nature of these entities? The panpsychist answer is that, for at least some of them, phenomenal or protophenomenal properties make up the intrinsic nature. The advantage here is that we get the world described by the sciences while also recognizing experience, the aspect of the world we are most intimately familiar with, as real and fundamental; Strawson would even call panpsychism “real physicalism.”

There are other advantages to this position. One example is how it meshes with our evolutionary picture of the world, where life has become increasingly complex as it adapts over time Рand at some point in this process conscious experience enters into the picture. A reductionist Dennett would have no difficulties here, but for any physicalist that wants to acknowledge experience as a real thing the evolutionary story poses a real concern. Since the physicalist does not want to say, like the panpsychist, that experience has been there all along they are committed to saying that, in some sense or another, experience is emergent Рregardless of what sort of biological structure is prerequisite for this emergence, we have a case of radical emergence. An entirely new property arises from physical material that is wholly detached from that property; a frequent example of traditional emergence in nature is the property of liquidity Рa collection of molecules in a certain state will express this novel property. But note that the behavior of the liquid can still be entirely described at the level of molecule to molecule interactions with nothing lost; however, in the case of experiential emergence we not only have no initial reason to suspect that certain physical arrangements would generate experience (another framing of the Hard Problem), but even after knowing that we have experience it is impossible to describe it at the reductive, physical level without losing a big part of the picture.

Couple this with the observation that experience, under traditional physicalism, seems to be an all-or-nothing sort of phenomenon Рyou are either conscious or you are not. An earthworm might have a very different sort of consciousness compared to the consciousness of a human being, but it is nonetheless present. You either have the experience of red or you do not, there does not seem to be some transitional, inbetween type of red experience. So in the physicalist picture experience radically showed up from non-experience at some point in the evolutionary tree. And this also raises other unsavory issues Рat this point of emergence it seems possible that tiny physical alterations would result in consciousness flickering on and off, which seems very much at odds with the gradual evolutionary process. To the contrary, with panpsychism, the experience never has to pop into existence from nowhere: the experience is there all along.

There are – it should not even need to be said – plenty of counterarguments from physicalists, but the point here is that there are some motivating reasons to accept the panpsychist ontology over the traditional physicalist ontology. But this brings us to what is seen as the biggest hurdle for panpsychism to overcome: the combination problem.

The Combination Problem

When we think about the nature of our conscious experience there is one universal observation: it, to be conservative,¬†seems¬†to be unified. I cannot separate the noise of my keyboard from the tactile sensation of my fingers fluttering over the keys – any separation is conceptual, a useful heuristic, rather than a reflection of how our conscious experience actually appears to us. I can report that I am hearing specific sounds, having certain feelings, in a certain mood, but they are all tied together into a seemingly inseparable whole. While in his paper¬†The Combination Problem for Panpsychism¬†Chalmers notes three general varities of the combination problem, and points out that there has not been a solution that adequately addresses each variety, I think a very general understanding of the problem is all we will need for our purposes. I’m going to borrow the William James quote from his paper rather than digging it out of my own copy of¬†The Principles of Psychology:¬†

Where the elemental units are supposed to be feelings, the case is in no wise altered. Take a hundred of them, shuffle them and pack them as close together as you can (whatever that may mean); still each remains the same feeling it always was, shut in its own skin, windowless, ignorant of what the other feelings are and mean. There would be a hundred-and-first feeling there, if, when a group or series of such feelings¬†were set up, a consciousness belonging to the group as such should emerge. And this 101st feeling would be a totally new fact; the 100 original feelings might, by a curious physical law, be a signal for its creation, when they came together; but they would have no substantial identity with it, nor it with them, and one could never deduce the one from the others, or (in any intelligible sense) say that they evolved it. Take a sentence of a dozen words, and take twelve men and tell to each one word. Then stand the men in a row or jam them in a bunch, and let each think of his word as intently as he will; nowhere will there be a consciousness of the whole sentence. We talk of the ‚Äėspirit of the age,‚Äô and the ‚Äėsentiment of the people,‚Äô and in various ways we hypostatize ‚Äėpublic opinion.‚Äô But we know this to be symbolic speech, and never dream that the spirit, opinion, sentiment, etc., constitute a consciousness other than, and additional to, that of the several individuals whom the words ‚Äėage,‚Äô ‚Äėpeople,‚Äô or ‚Äėpublic‚Äô denote. The private minds do not agglomerate into a higher compound mind. (James 1895)

What James is saying, essentially, is that there is nothing about a collection of experiences that necessitates a higher level of experience that subsumes the rest – there is no reason that this collection should result in anything different than the collection of individual experiences. And, unfortunately for panpsychism, in order to account for our macroexperience, the unified field of consciousness that all of us possess, the fundamental, microexperiential pieces would have to combine in such a manner. This is an incredibly difficult problem for the position, and it might ultimately turn out to be fatal. If we cannot get to the reality of our conscious experience – the entire motivation for panpsychism – by adopting the position then there is simply no reason to make that ontological commitment. So if we treat panpsychism as the thesis that macroexperiential truths are necessitated by microexperiential truths this combination problem would make panpsychism false. Outside of the current cultural mindset that sees panpsychism as an absurd position – a judgement that waxes and wanes throughout Western thought – the Combination Problem is the most frequently cited reason to reject it.

A Combination Problem for Physicalism? 

If the Combination Problem is sufficient reason to reject panpsychism, despite the appealing aspects the position has over physicalism, then finding physicalism to suffer from a comparable sort of combination problem would remove the motivation to reject panpsychism¬†for that particular reason –¬†so long as there are no solutions available to the physicalist that are not available to the panpsychist. I think there is at least reason to¬†suspect¬†that such a problem might exist.

A highly recognized issue in neuroscience (and related fields) is known as the Binding Problem. We have a growing understanding of which areas of the brain respond to, process, and accomplish specific stimuli/tasks, but the problematic part is that all of this activity is spread out across different brain regions and yet is somehow bound into a concrete experience. And more than simply binding the various sensory stimuli into a unified phenomenal field there are issues with how various pieces come together that identify, for example, objects as such. There are certain features that ‘belong’ to the water bottle on my desk which allow me to distinguish it from the rest of the environment as a water bottle. All these disparate neural networks somehow come together as a complete whole – this should be tickling your intuitions but I’m inclined to think of this particular issue as a cognitive science problem shared by both panpsychism and physicalism, not a physicalist equivalent of the Combination Problem. Rather the exercise is intended to show a common conception of our conscious experience: that it is a¬†unified whole.

Now we have to flesh out a satisfactory account of what constitutes something being¬†physical¬†such that this potential combination problem presents itself. Luckily, for our purposes, this is not particularly complex, and the only thesis about the physical we need to put forth is that physical things are composite – they are divisible, if not infinitely, down to incredibly small levels. We can divide a human being into cells, divide cells into their various parts, divide those molecules into their atoms, and those atoms down into neutrons, electrons, and protons. And we could keep going, perhaps all the way down to the planck scale. So with these two things in mind, let’s put forth a version of what Kant called the Achilles Argument – an argument with a very long history, originally coming about as an argument for the immortality of the soul:

(1) Consciousness is a simple unity.

(2) Physical things are composite.


(3) Consciousness is not physical.

When we say that consciousness is simple we do not mean it in the sense that there is not a rich complexity to our conscious experience, but we instead use it to distinguish the sort of unity consciousness possesses to other sorts of unities. For example, I can build a castle out of Legos and refer to it as a unified structure, but it can still be clearly divided down into individual legos without any difficulties. The sort of unity consciousness possesses, on the other hand, is not divisible in such a way Рthis is what we mean by consciousness being simple. So in order to attack (1) we have to deny that, even though consciousness is unified, it is not a simple unity; however, this goes against our entire phenomenology, and in making this attack it seems that we would have to say that in some sense we are mistaken, that our experience is an illusion of sorts. A hard pill to swallow. There is often a confused response to this premise as well, as an individual might point out that conscious experience could be, and probably is, generated by neurons firing in synchrony Рthat neurons behave as a unity. But this is no different than the example with the Lego castle as each neuron is its own entity, relationially bound to but distinct from the neurons around it, and even the neuron itself is composed of similar structures Рall the way down.

Denying (2) also seems to be a difficult bullet to bite. The idea that physical things are divisible is not just what we see when interacting with the world, but it is also the presupposition that all of the sciences rest on. This mechanical picture, where various natural entities are interacting with each other to carry out chemical reactions and cohere into larger structures,¬†requires¬†that physical things be divisible into their constituent parts. What would we mean by saying that protons and electrons exist if we did not see the atom as an entity that can be broken down? One line of response to this is that while physical things are divisible, they are not¬†infinitely divisible.¬†So that at some point physical things as well bottom out into a truly simple entity. The issue with this can be seen if we read up to the William James passage, particularly the section about words and sentences. Let’s pretend that there is an ultimate, fundamental physical substance that is truly unified – it cannot be broken down any further. If we only put¬†part¬†of consciousness onto each of these entities then we have the same situation with trying to force together a complete sentence out of a collection of words – it simply does not work, and is in fact directly analogous to the Combination Problem, where you have fundamental entities that must somehow cohere into a whole sentence (or thought!). And it does not seem particularly appealing, in fact it might even be ad hoc, to say that each of these entities contains the whole thought – but that is what is necessary in order to resolve this issue of divisibility. One possible avenue that I do not know enough to comment on is that our quantum picture of the world might point to the physical world¬†actually¬†being unified in the necessary way and that our treatment of physical things as composite is nothing more than a very successful heuristic – much like Newtonian physics does not reflect the actual physical reality but is a very effective, and much simpler, approximation over a certain scale. But I think issues around Scientific Realism happily pop their heads into the ring at that point.

Is The Achilles Comparable? 

In order to determine whether or not the Achilles argument against physicalism is comparable, in the sufficient sense, to the Combination Problem that panpsychism faces we need to look at the physicalist responses and see if any of them are both promising and/or unavailable to the panpsychist. At least of the possible responses I have though of I do not see such an option. To attack the first premise is to, in some sense, break down the conscious experience into something that is not actually unified in the simple sense. But it seems that doing so is also, and in actuality is (in breaking down the subject; interpretation of split brain patients is an interesting path to pursue here), an avenue that the panpsychist takes against the Combination Problem. And, as we have seen above, attacking the second premise puts you in a place that very much resembles the problem in that we end up with various words trying to assemble themselves into a verse. I think the quantum mechanical suggestion is no better here. The very fact that panpsychism takes on the exact same structural and relational scientific picture that physicalism does makes rejoinders against the second premise also available to the panpsychist.

On the same note, however, this means that the Achilles problem is as much able to be directed at the panpsychist as the traditional physicalist, but I’m not sure that’s a particular concern – it does not seem to be something that can be tacked on to the Combination Problem to put panpsychism’s issues over and above physicalism’s. And so we are back where we began: panpsychism has an advantage over physicalism in that no¬†radical emergence¬†is required for the existence of experience, it being there all along, and panpsychism does not suffer from problems that are not shared by physicalism. It’s possible that there is a significant difference in character in assembling things in the structural manner the physicalist shares and assembling the intrinsic, phenomenal attributes that are unique to panpsychism. I’m undecided on a lot of these questions.

Conclusion, Sort Of?

This was basically a bunch of musing and thought dumping on my part, there are a lot of suggestions I’ve made here that I suspect are wrong, both in obvious and subtle ways, but this sort of exercise is always great.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s