The materialist's upper hand: a response to Uccisore

A particularly astute contributor to the board has recently argued that materialism faces similar problems to dualism in explaining the relationship between mental and physical events. Despite the considerable merits of his argument, in this response I argue that modern materialism is already well equipped to deal with the problems he raises. I start by outlining the notion of supervenience, and show how it solves the problem of causal under-determination that will be extracted from the essay. I then go on to sketch a wa that supervenience might be argued for by the materialist. Finally, I address a final consideration that can be found in Ryan Smith’s essay and suggest that it alone provides no convincing response to eliminative forms of materialism. The basic thrust of the response is to show that materialists can account for the relationship between physical and mental events that avoid the problem of under-determination that is proposed in Smith’s essay, and that therefore materialism does not suffer from the same difficulties that dualism does.

The first serious problem that Smith highlights for materialists seems to be that materialists have no way of accounting for how it can be that a certain physical state gives rise to a certain mental state:

For a start, the criticism seems to be waged at epiphenomenalists, who are often considered to be dualists. Perhaps this is part of the problem: the argument does not seem to be any response at all to forms of eliminative materialism that simply deny the existence of mental events altogether. However, let’s consider forms of non-eliminative materialism:

Non-eliminative materialism: a form of materialism that allows for the fact that mental events exist. *1

So, the problem here seems to be that the non-eliminative materialist wants to say that certain brain states entail certain mental states, and yet have no way of showing that a certain physical state guarantees a certain corresponding mental state. I’ll call this the problem of under-determination of mental states.

Let us now solve this problem using the notion of supervenience. A relationship of supervenience can defined as:

A supervenes on B when any object which possesses B must necessarily possess A.

A common example would be that the property of beauty (if such a thing exists) must supervene on physical properties. Simply put if two paintings, painting x and painting y, had the exact same physical properties and painting x is beautiful, then it seems necessarily the case that painting y must also be beautiful. With beauty this makes quite a lot of intuitive sense: if someone stood in front of two identical paintings and declared one to be beautiful and the other un-beautiful, we would almost certainly think he is lacking in understanding of the relevant concepts. The materialist can make a similar claim about mental states: they can say that a certain physical state does determine a certain mental state, that the correct explanation of the relationship between mental and physical states is that the former supervenes on the latter. Thus, a certain mental state guarantees the occurrence of a certain physical state. Take, for example, the fiercely nominalist philosophy of Donald Davidson:

At this point it may look remarkably like the response has quite spectacularly missed the mark. Perhaps Smith could argue that the key concern was that the materialist had no story to tell about why he believes the relationship of supervenience holds. The two are at a standpoint, the skeptic is saying “but what if the relationship doesn’t hold”, the materialist seems to be responding “it does”. However, there is still more to be said. The notion of supervenience has been introduced as a way of defining the relationship between the mental and the physical which avoids troublesome notions of causation and as the first part of the solution. What has been established so far is that:

  1. If a relationship of supervenience holds between the physical and the mental, then the under-determination problem would be solved.
    What is now needed is support for a second premise:
  2. A relationship of supervenience between the physical and the mental holds.
    Perhaps this is the real challenge that Smith is waging – that there is no reason why this should be seen to be the case. However, if this is indeed the accusation, then I believe it to be wildly misplaced. The materialist here has vast amounts of evidence from modern day psychology which he can call upon to support the 2). For example, he might (quite correctly) claim that research in to depression has shown that depressive symptoms are commonly identifiable by certain common physical symptoms, or that certain types of interference in the physical (such as frontal lobotomies) produce specific alterations in mental states. In essence the materialist is simply claiming that a relationship of supervenience simply does hold. There is no doubt that this assumes far more about psychology than we currently know, and that there is an interesting ongoing debate in the philosophy of psychology as to whether this is the case. Yet, this does not mean the materialists argument is not now a viable position. In essence, what is happening is that the materialist is calling upon empirical evidence to support his theory. And this seems perfectly in tune with a materialist account: all the materialist aims to do is show that mental states are completely explainable in physical terms.

The materialist’s position can now be formulated fully:

  1. A physical brain state occurs
  2. A mental state occurs that supervenes on the physical
  3. The physical brain state causes bodily movements etc.

The only problem that now remains is the question of why it is the case that this relationship of supervenience holds. Perhaps there is a possible third claim from Smith here, after all that was his initial question, wasn’t it? That the materialist still needs to tell us why mental properties supervene on the physical (and not just give us his reasons for suposing that the mental does supervene on physical properties) But the charge now seems somewhat irreverent. The materialist need only argue that the relationship does hold in order to start explaining mental phenomena as physical phenomena. That they have no account of why it holds is no more troubling to the materialist than the fact that physicist has no account of why the laws of physics hold is a problem to physics. Such an explanation is simply unnecessary for the success of the materialist’s account. The dualist (as Smith noted) has a problem explaining how mind interacts with body – however the problem for the dualist is much more serious because without such an account there is simply no definable relationship between the body and the mind whatsoever. The materialist already has such an account (that of supervenience), and therefore has the explanatory upper hand. Also note, if the dualist did come up with such an account (as I believe is reasonably possible), he would then have exactly the same problem that the materialist now has – he also would have to explain why this relationship holds. So the materialist does not have the same problem that the dualist has here (insofar as the dualist’s problem is in giving an account of the relationship between mind and body). All explanation has to stop somewhere - I can see no reason why the materialist can’t simply stop his explanation here and leave the question of why the relationship of supervenience holds unexplained. On theother side of things, this option does not seem open to the dualist - who so far in our discussion has not shown how his thoery can explain the relationship at all, and so has a much more severe (and very well documented) problem. In short, I believe what the dualist is charged with lacking in the classic ‘causal explanation’ problem is precisely what we have just shown the materialist has readily to hand: an account of the relationship between the mental and the physical. Thus, the two do not share the same problem.

As noted, the problem of under determination is waged only at a non-eliminative type of materialism. An eliminative materialist would object to it simply by saying that there is no question whatsoever of mental states being different to brain states because there is no such thing as brain states other than what is physical:

Eliminative materialism: the theory that states that all mental states are reducible to physical properties in a way that completely eliminates mental states. That is to say, there is nothing more to mental life than a set of physical facts.

There is one more separate argument against materialism in Smith’s paper. I believe this time that this is directed solely against eliminative materialism, that is, the complete denial of mental events – or the complete reduction of the mental to the physical. I can’t think of a better way of expressing it than simply quoting it in bulk:

I believe this is simply a presumption against the materialist, as opposed to an actual argument. Consider the statement: “When we describe a mental event, explaining the neurological states involved would have nothing at all to do with the information we are trying to get at” – isn’t it the case that if I were an eliminative materialist, then I would be claiming that every time someone spoke of sadness and happiness, then what they were actually talking about would be a series of physical processes and nothing more? Isn’t it also explicitly their claim that a complete physical description of physical states would tell us all we need to know (indeed, everything there is to know) about mental life? I think the argument suffers a viscous circularity here – it assumes that when we talk about mental states that we are referring to non-physical states, which is something that I think the (eliminative) materialist is entitled (and presumably more than willing) to deny. But once this assumption is dejected, there is no argument here at all. *2

Finally, I would like to make a brief comment on the following passage:

I wager by now the problem here should be obvious. It seems that here the materialist has the same problems as the dualist because he has become a dualist. He has now worked mental events in to his explanation of the way the mind works, he has even suggested that it causally interacts with physical things: “This mental decision results in nervous impulses”. So, the materialist here is no materialist at all! No surprises, then, that he faces the same problems as the dualist! But as we have already seen, there is no solid reason for the materialist to resort to this final (decidedly dualist looking) position – he already seems to have met the obstacles that have been put in his path.

The conclusion here is this: that the dualist can formulate his position in a way that solves the under-determination problem. Furthermore, once the position has been formulated by an appeal to the notion of supervenience, it is possible for the materialist to provide evidence that the relationship of supervenience does hold, and thus maintain that a full and satisfactory explanation of mental events in physical terms is theoretically possible (even if the modern scientist lacks the capacity to give such an explanation for mental life). The question of whether the materialist could explain why a relationship of supervenience holds left was left untouched. This, however, has been argued to remain unproblematic for the materialist and incomparable to the problems that the dualist faces. *3

* Footnotes:

*1 A definition that draws in part from discussions in Philip Goff’s ‘A Non-Eliminative Understanding of Austere Nominalism’, which also suggests one very interesting method of formulating such an account and from which other parts of this paper are drawn.

*2 As it happens, I share Smith’s general concern that one should be deeply suspicious of eliminative accounts: they seem to deny the facts that are outstandingly obvious to us, that there is such a thing as mental experience. The point is more that this is merely a presumption, not an argument per se.

*3 Smith also sketches an independent reason for denying materialism. I have left this untouched because I believe it to be the real reason for the acceptance of some form of dualism. I think it might run close to Jackson’s arguments, specifically the Mary’s room example. The strikingly similar philosophical zombies argument might also provide a presumption against the supervenience claim – but this is not within the scope of this essay.

  • John Levington, Birmingham.


Davidson, D. 1970. “Mental Events,” reprinted in D. Davidson, ed., 1980, 207-225.

Goff, P. ‘A Non-Eliminative Understanding of Austere Nominalism’. Findable online at … nalism.pdf .

Smith, Ryan. ‘DesCartes, Materialism, and the problem of Mental Causation’, viewtopic.php?f=9&t=167822.

Lovely… you managed to capture pretty much every objection which I thought to level at the arguments presented by the opposing party.


Wish you hadn’t included that bit… It does not seem necessary.

We need not deny mental experience, only equate them with physícal events in the brain. trying to keep in mind that we’re all really working with mental models of the world, it strikes me that the true barrier some of us face here is based on language and conception.

I’d gladly defend this perticular point of view if I’m asked…

Otherwise… I thought this was an excelent counter, and I can’t wait to see Uccisore’s reply.

Good job :smiley:

First of all, thanks for the thoughtful response to my paper- it’s good to know they’re being read, and it’s always nice to have chance to polish up the work through criticism.

 Let me kill two birds with one stone off the bat, Dualism itself and eliminative materialism.   Though I don't mention him as much as I should, this is a paper in response to Descartes' Meditations, and the Passions of the Soul to a lesser degree. As far as I've ever seen Descartes argues for a conceptual difference between mind and matter (i.e., thoughts don't appear to be extended, we need to think of ourselves as mental beings but not necessarily as physical beings), without ever sealing the deal with an argument for real, substantial dualism, though others have tried.  I think the conceptual difference, though, is enough to deal with the eliminativists, or at least put the burden of proof on them. First, I don't think an eliminativist can doubt the conceptual difference- thinking about the mind is not the same as thinking about the brain, just as thinking about Bat Man is not the same as thinking about Bruce Wayne.  They are two different concepts, with different content, that happen to be wrapped up by the exact same object in reality (or so the materialist would content). If it needs to be proven that the conceptual difference exists, I'd be happy to do so, but I consider it too obvious to spend time with, unless someone really needs this work shown. 
  So, One can talk about Bruce Wayne or Batman, without referencing the other. One can even know they are the same person, while maintaining the independent concepts - for example, Alfred may see Bat-manning as something Bruce does with his evenings, and Wonder Woman may see Bruce Wayne as Bat-Man's cover story.  Even if they are in possession of the same facts. Also, compare our conception of the Sun as a life-and-warmth giving yellow circle in the sky, to the conception of the star these 8 planets orbit that an alien would have upon visiting our solar system. Once these aliens visit the earth, and learn something about our culture, they could in turn understand why we conceptualize the sun the way we do. 
In both these cases, though, one of the conception pairs [i]still points to[/i] the other.  When we point up at the Sun, we're still pointing in the direction of the star, even if the aliens wouldn't call that direction 'up'.  When Bat-Man is inexplicably absent, we can explain that absence by something going on in Bruce Wayne's life. Also, in both cases, the Sun, star, Bruce, and Bat-Man are all the same kinds of objects as each other's pair member- persons, celestial objects, etc. Even if one doesn't believe it, one can [i]just see[/i] that Bat Man is the type of thing that could be Bruce Wayne. The premise gets off the ground in a way that "Bat Man is the number 14" does not. Bat Man and 14 are not the kinds of conceptions that could be of the same object- we know that from the conceptions, with no knowledge needed of a hypothetical object. 
It seems to me, the things that compose the conception of the mind (being without extension, grasping propositional content, bring composed of ideas and experiences), are so completely different than the things that compose the conception of the brain, that the burden of proof must be on the eliminativist who wants to contend something so counter-intuitive. Note that any argument about brain damage, drugs, and so on linking brain events to mental events are arguments not for eliminativism, but for supervenience. which I'll talk about next. 

I guess the first thing I’d say is that I don’t see how you think the materialist has explained anything here. You say that the materialist has the concept of supervenience, and that this puts them in a position superior to the dualist, even if the precise interaction between the mental and the physical is yet to be described. But from what I can tell, supervenience isn’t an explanatory concept of anything, it’s just a word for the fact that when things are going on in minds, things are also going on in brains. I haven’t met a dualist that would deny this is the case (though it certainly may have exceptions, such as God), so I don’t see what’s been contributed here.
Supervenience isn’t an ‘account of the relationship between mental and physical events’, it’s just a word that labels it, whatever it may happen to be. As far as I can tell, all I need is a term of my own- ‘substance defenestration’, say-, and to allege that this term means the sort of interactions mental and physical substances have (without saying a word about what that interaction is like), and I’m in the same position as the supervenience materialist.
Even if, however, supervenience explained something, I don’t think it explains the thing that’s problematic here. Even the epiphenominalist could easily contend that every time something happens in the brain, something also happens in the mind, and that mental effects are supervened on physical brain activity in the same way that color is supervened on vision. And for all of that, they’d still fall prey to skepticism, because of the issues of causation I bring up. The causative problem for the materialist is not ‘how does the brain cause thinking’, but rather, ‘how does thinking cause physical action’. It’s not enough that the brain doing stuff is supervened by the mind doing stuff. The specific content of that mental stuff (“Boston is south of here”), has to be directly linked to the physical end results- my body getting in the car and driving it to Boston. This obligates the materialist to nominalism at the very least, and probably to eliminativism, which I think we both see is a non-starter. My contention is that seeing how the brain moves muscles around is easy, and seeing how the mind can understand Boston is easy (or at least intuitive), but seeing how the mind moves muscles is hard, and seeing how the body responds to the understanding is also hard.


I understand that having to defend yourself against two potentially very different perspective might take more time than you’d be willing to invest, and so feel free to take your time with my following remarks.

I would argue that recent advances in neuroscience and our understanding of the human brain has given us more than enough cause to consider the two (mental experience and events in the brain) one and the same. The one to one correlation of events seems to indicate equality, and with no exception, it indicates that concieving of two seperate events no longer adresses anything, neither mental nor physical. To concieve of brain events as seperate but somehow related to mental events fails to be a meaningful distinction at this point.

Seeing a shiny metalic fork in my hand and feeling a cold and hard sensation on the skin of my hand, NEED not be caused by the same object, but we are more than justified in assuming it is… to assume that some invisible thing that can only be sensed through touch happens to accompany shiny objects that can only be experienced through sight seems a meaningless distinction given the relationship we’d be forced to claim they have to eachother… might as well call it “steel” and atribute both types of sensations to the same thing.

But this is an unreasonable way to adress the issue, it seems to me… That’s like saying the software on my computer isn’t physical because the way we concieve of the software in form of coding is nothing like the way in which we concieve of the hardware in terms of electronics… and it would reduce the programming to utter gibberish to to think of the software as tiny indentations on a disc or some such… clearly that only speaks to which concept is the most economical for us given our needs and knowledge and says nothing about which more accurately describes reality.

Likewise the assumption is that IF we could understand the workings of a brain we could “read” it’s content, which would be identical to the contents of “the mind”, would effectively equate the two. And so far all evidence indicates that this is the case… I don’t see what it would mean to say that they are different things… we might as well call all visual experiences “visions” and set them apart as something non-physical… so why don’t we?

At the core of all of this, it strikes me as a discussion of the borders of what we consider our own identity… those things being what we consider the “ego” and not part of the “other” which is the physical and outside world… why it would strike us as counter-intuitive to accept a physical object (that is to say an object we experience through our five senses) as EQUAL to ourselves, and not just be “part of” us is not very surprising from an evolutionary standpoint…

Our concept of the brain comes from our five senses (we’re not born with knowledge of the brain)… and as such we cannot naturally get the feeling that it IS us… like other parts of our bodies we can accept it as PART of us, but to think of it as EQUAL to us… that’s an ability we’ve never needed to evolve… and I see no reason that we ought to expect that it come natural to us, such that a failure be evidence of it’s falsity.

I wouldn’t contest any of this. But what the materialists say is - dualism is incoherent ergo materialism is the proper conception. The materialist says - you can’t explain anything in non-physical terms in a satisfactory manner. At any rate I am not offering a total defence of materialism - I am just showing how it can meet your objections. Although Mad man P seems to have some pretty good responses to it!

I think this is just unfair. It isn’t just a word at all - I clearly defined the concept. And it wasn’t defined in a circular or vague manner either, unlike " the sort of interactions mental and physical substances have" (which is both). Supervenience was defined:

A supervenes on B when any object which possesses B must necessarily possess A.

This seems to me like a complete account of a relationship. I fail to see how any account of any relationship whatsoever could be formulated in a way that is clearer than this one. And it is such a clear, non circular account of the relationship that the dualist apparently lacks. More importantly, though - it makes no appeal to the notion of causation whatseover.

But this is the dualist’s problem, not the materialist’s. The point is that the Cartesian dualist wants to say that mind causes physical events - and this is what you’re intuitions are objecting to. How can spirit - a seemingly acasual concept, cause physical events? How does mind interact with body - if the two are distinct?

The materialist, though, rejects the claim that there is a thing called mind which is distinct from physical, thus there is no causal relationship between thought and the body. The materialist would claim that the “cause” of your going to boston would be explainable only in terms of other physical events (some neurons firing, some electrical signals etc).

It doesn’t. Well - not if you’re a materialist anyway. There is no ‘causative problem’ for the materialist because, unlike the dualist, they do not claim that a relationship of causation between mind and body exists.