Objections to Melnyk’s Case for Physicalism (2007)
Stewart Goetz and Charles Taliaferro
1. Philosophy of Mind
We are grateful for this opportunity to reply to Professor Melnyk’s systematic, clear paper entitled “A Case for Physicalism about the Human Mind.”
According to Melnyk (henceforth M), “An individual item (object, property-instance, or process) is physical in the broad sense if, and only if, it meets either of two conditions: (1) it’s an item of a kind that can in principle be defined in the distinctive vocabulary of fundamental physics; or (2) it’s a physically realized item of a functional kind.” As is evident from M’s explication of the concept of deciding to act in light of (2), physicalism implies that what we ordinarily take to be a causally undetermined mental action is both caused and determined.
Consider, now, M’s writing of his paper. If M is at all like us, when he was approached by Professor Draper about participating in this exchange, M mulled over the reasons for and against getting involved and freely chose (decided) to write his paper for a reason. In our paper, we explained that from the first-person point of view, which is the point of view of M as a conscious subject, a choice is an uncaused mental event and a reason is a purpose that provides an ultimate and irreducible teleological explanation of that choice. For the sake of discussion, we will assume that M chose to write his paper in order to make a clear and persuasive case for physicalism. Given M’s choice and its explanation, a mental event (e.g., the choice itself or a resulting intention) causally produced events in M’s brain that eventually causally led to the movements of his fingers and the production of his paper, all of which ultimately occurred for the purpose that explained M’s choice.
If physicalism is true, important elements of this first-person point of view are mistaken. If M made a choice to write his paper, then that choice was not ultimately and irreducibly explained teleologically by a purpose. Instead, M’s choice was ultimately and irreducibly explained causally and it or something else it caused in turn causally produced events in his brain, which in turn causally produced the movements of his fingers and the production of his paper, where no event in this sequence ultimately occurred for an irreducible purpose. Rather than there being two irreducible kinds of explanation (a dualism of explanations), teleological and causal, the former of which obtains prior to and accounts for the obtaining of the latter, there is only one kind of explanation (causal). And rather than there being two kinds of events (a metaphysical dualism of events), free (undetermined) and determined, the former of which occurs prior to and accounts for the obtaining of the latter, there is only one kind of event (determined).
Given these implications of physicalism, why think it is true? According to M, “as far as we can tell, all human phenomena are dependent on neural phenomena; we never catch the human mind at work without also catching the human brain at work…. So it looks like there’s no human behavior for the causal explanation of which it’s required to assume the existence of anything that’s neither physical nor physically realized.” (The emphases are M’s.) Contrary to M, if we take the first-person data seriously, not all human phenomena are dependent on neural phenomena in the sense that occurrences of the latter always explain occurrences of the former. In the case of choices, the occurrences of some neural phenomena are dependent on, because they are causally explained by, the occurrences of some mental phenomena, which are in turn ultimately explained teleologically by other mental phenomena (reasons or purposes).
As we noted in our paper, a reason for choosing involves conceiving of a way the world might be but presently is not. The content of our reason for acting now is to bring about the state affairs that you come to read our reply to M, arguing that there are no good objections to the view that we make uncaused free choices. This common sense understanding of action creates two difficulties for physicalism: the problem of accounting for the conceptual content of intentional activities (our conceiving states of affairs that we wish to bring about) and the problem of accounting for reasoning itself.
Consider first the problem of conceptual content: it seems apparent that we can indeed conceive of states of affairs that do not obtain as well as propositions about the world (Minneapolis is a city in Minnesota) and about mathematics (6 is divisible by 3 and that 3 plus 2 plus 1 equals 6). These entities or objects are not definable in the distinctive vocabulary of fundamental physics. While the content of a conceptual entity (a proposition or a state of affairs we conceive) might refer to or be about something that is physical and located in space (e.g., writing an essay or Minneapolis, Minnesota), it seems to be neither physical nor spatially located, nor comprised of parts that are physical and spatially related to one another. Thus, while the written sentences ‘The Minnesota Twins play baseball in the city of Minneapolis’ and ‘Minneapolis is the city where the Minnesota Twins play baseball’ are both themselves physical entities that are spatially located on this page and about a sports team that is spatially located in an American city, the content of the conceptual entity (the proposition) that is expressed by these two written sentences seems as if it is not physical and not located anywhere in space. It is because of the seemingly nonphysical nature of conceptual entities that Plato affirmed the existence of the nonspatial world of abstract objects called Forms. We believe that contemporary physicalists have yet to overcome the challenge of avoiding abstract, nonphysical objects.
There is a second problem for physicalism that arises from our conception of states of affairs and the existence of abstract objects. The problem is that physicalism seems incapable of accounting for the mental causation that is involved in reasoning or inference to a belief that occurs because of the apprehension of states of affairs. When we reason to a belief, mental events seem ultimately and irreducibly to explain the occurrences of other mental events. Consider an example in which you are asked “What is the smallest perfect number (a number that is equal to the sum of its divisors, not including itself)?” If you seek to discover an answer to the question, you are acting for an ultimate and irreducible purpose, whose optative content is that you acquire knowledge of the smallest perfect number. Now, as you think about the question and apprehend (a mental event) that 6 is divisible by 3, 2, and 1, which when added together equal 6, and apprehend that there is no number less than 6 that is equal to the sum of its divisors, you are caused by these apprehensions to believe (a mental event) that 6 is the smallest perfect number. The fact that on this occasion you come to believe that 6 is the smallest perfect number is ultimately and irreducibly explained in terms of the contents of the propositions that you apprehend. In other words, acting for an ultimate and irreducible purpose (that you acquire knowledge of the smallest perfect number) ultimately and irreducibly explains the occurrence of ultimate and irreducible mental-to-mental causation (apprehensions of contents causing a belief that has content). If you go on to convey verbally your belief that 6 is the smallest perfect number to the questioner, your choice and/or intention to do so (or some other mental event) will cause your mouth to move in the appropriate ways for the purpose of telling the answer to the questioner. Here there is mental-to-physical causation that is ultimately and irreducibly explained by a purpose.
If M is right, however, and physicalism is true, there is neither an ultimate and irreducible teleological explanation of your seeking to discover the answer to the question nor ultimate and irreducible mental-to-mental and mental-to-physical causation. There is neither of these because (in M’s words) “all human phenomena are dependent on neural phenomena.” (The emphasis is ours.) What this seems to imply is that everything that occurs in our mental lives, including our beliefs, is ultimately explicable in terms of physical causation alone without any explanation of that which is irreducibly mental by something else which is ultimately and irreducibly mental. If physicalism is true, it seems that the ultimate explanation of M’s belief that physicalism is true will mention only neural phenomena and/or more fundamental physical constituents. There will be no mention in that explanation of anything that is ultimately and irreducibly mental. So, even if our first objection is answerable and there are irreducible contents of apprehensions and beliefs which are physically realized (which we doubt), physicalism entails that M believes that physicalism is true, not because his apprehensions of irreducible contents ultimately cause his belief with its irreducible contents but because physical (neural) realizers of those apprehensions causally produce physical realizers of that belief. In short, physicalism seems to entail that apprehensions of contents are explanatorily impotent and epiphenomenal. This is a hard pill to swallow. Given the truth of epiphenomenalism and the explanatory impotence of apprehensions of contents, why should any of us try to reason with others about anything with the hope of changing their beliefs by having them apprehend contents of conceptual entities?
At this point, M might interject that physicalism does not exclude the ultimate explanatory efficacy of the mental. That which is irreducibly mental and physically realized (e.g., apprehensions of content) is sometimes ultimately causally efficacious in producing both other irreducible mental events which are physically realized (e.g., beliefs) and other physical events (e.g., finger movements) which do not serve as realizers of anything that is mental. If M were to respond in this way, then he would have moved a long way toward dualism because he would have conceded the reality of both ultimate and irreducible physical-to-mental and ultimate and irreducible mental-to-physical explanation. And if he were to allow for such ultimate and irreducible mental-to-physical causation (as well as mental-to-mental causation), it would be interesting to know why he is a causal determinist who will not allow for ultimate and irreducible teleological explanation.
There are, then, at least some elements of the first-person point of view that seem incompatible with the truth of physicalism. Is there something that we know from the third-person point of view, the point of view of an observer of the physical world that is used to support physicalism, which conflicts with what we know from the first-person point of view? M says that “if human physiologists had known that human behavior is caused by signals carried by motor neurons that run from the skull, and had then opened up some human skulls for the very first time only to find that they were quite empty, or merely filled with blood, then the positing of a nonphysically realized mind to explain the complexities of human behavior would surely have been unavoidable; there would simply have been no physical candidate of sufficient complexity to do the explanatory job.”
Unless human physiologists are already committed physicalists, M’s description of them is highly questionable. Given that human physiologists are individuals with first-person points of view, what they believe about their own choices before they ever open up human skulls implies that some of what happens in their physical bodies only happens because it is ultimately and irreducibly teleologically explained by what happens in their minds. M states that human physiologists would posit a nonphysical mind (and, one can reasonably assume, a teleological explanatory role for the mental), if they opened up human skulls and discovered that they were empty. What we know from the first-person point of view, however, is not posited or hypothesized. It is a fundamental datum that itself must be acknowledged, as opposed to explained away, by an adequate account of reality. And as we argued in our paper, there is nothing that a human physiologist or any other scientist discovers or is required to assume in his work as a scientist that is incompatible with or undermines this first-person datum.
M makes much of a representational account of human action. This stress on representation, however, eclipses an additional, vital, and evident feature of our mental lives: subjective experience. Consider, for example, an experience of pain. Unlike choosing, an experience of pain seems not to have any representational content. When we are in pain, we do not pain that such-and-such. Rather, pain is a nonrepresentational experience of what we can think of as hurtfulness or, to use an awkward expression, ouchiness.
Suppose, now, that prior to writing his essay, M had boiled some water for tea and accidentally touched the burner. Other things being equal, M would have experienced pain and withdrawn his hand from the burner, run it under cold water and iced it for a purpose, namely, that he alleviate the pain and minimize damage to his hand.[1] At least, that is the ordinary view of what would have been going on with M. What does physicalism say would have been going on with M? Well, something like the following functionalist account will be offered: given that M is a mental system, he had (i) a sensor that formed a representation about his environment (the tea kettle, the burner), (ii) a mental representation of damage to a subregion of his hand, and (iii) a representation of a goal state where the hand is withdrawn from the burner. Given the event process that involved the representations in (i), (ii), and (iii), and other things being equal, M’s hand would have been caused to move away from the burner in accordance with the relevant physical laws. Notice, however, that on the functionalist account of what happens when M experiences pain, the ouchiness of pain (its subjective feeling) has disappeared. All that we are left with is a system of causal inputs provided by or in the form of representations that produce certain causal outputs. In other words, on the functionalist understanding of the mind, an experience of pain is exhaustively characterized in extrinsic, relational terms (inputs and outputs). There is nothing it is like intrinsically to be in pain that is not reducible to a relational analysis.
It is precisely this inadequacy of the functionalist characterization of pain that we pointed out in the story about Mary in our essay. Contrary to what physicalism would have us believe, the fundamental datum provided by an experience of pain is dualistic. When people experience pain, they experience an event of a kind that is nonrepresentational in nature. They also come to have desires that they alleviate their pain, beliefs that this alleviation is accomplished by withdrawing their limbs, and form intentions that they withdraw their limbs, where each of these mental events is representational in nature. Moreover, their experiences of pain do not cause the withdrawal of the limbs. Rather, on the basis of the stated desires and beliefs, people are given purposes or reasons to act, which are that they alleviate their pain and that they minimize damage to their limbs. These purposes then teleologically and noncausally explain agents forming their intentions to withdraw the limbs, where these intentions (or some other mental events) causally produce the physical events that ultimately lead to the relevant movements of the limbs to accomplish the stated purposes. There are two kinds of mental events, representational and nonrepresentational, and two kinds of explanation, teleological and causal, neither of which is reducible to the other. Physicalism fails to provide an adequate account of this fundamental datum. Instead, it disregards it.
When we chose to author our paper and provide a response to our interlocutor’s essay, we presupposed as a fundamental datum that we are substantial individuals with the potential or capacity to endure or persist across time into the future. While we have properties such as (i) the power to think and the power to choose, and exercise them as agents when we think and choose, and (ii) the capacity to desire, the capacity to believe, and the capacity to experience pain, which are actualized in us as patients when we believe, desire and experience pain, we are neither identical with our powers and capacities or their respective exercisings and actualizations nor systems that consist of them. We are substantial individuals that have powers and capacities.
Once again, physicalism gives us a completely different picture of ourselves. According to M, “Humans have minds, but it doesn’t follow that minds are genuine persisting objects.” Though minds are not genuine persisting objects, “physicalism about the human mind can easily treat minds as persisting objects—by treating an organism’s mind as its mental system, on a par with its digestive system or immune system.” In our view, this involves a profound category mistake; persons are not, and cannot be systems. While we think systematically when we author our essay, we are neither the same as systematic thinking nor a system of thought, just as a runner is not the very same thing as his running. M says “an organism has a human-type mental system just in case it houses some complex object or other that … generates representations of states of affairs far removed from the organism in time, space, or magnitude, representations of regularities in the world, and representations of its own representations.” Surely, however, we are something more substantial than some complex objects or other which are housed in organisms and generate representations. As we pointed out in our essay, what we seem to be are simple substantial individuals (souls) that have a multiplicity (complexity) of properties. And it is because we are such entities that we strictly persist self-identical (without the addition or loss of separable parts, of which we have none) through time.
Given everything counterintuitive that physicalism implies about what we are as human beings, it is imperative that we focus on the most general consideration that M claims supports physicalism. This consideration is that physicalism provides a more economical explanation than that provided by dualism in so far as it requires fewer ontological commitments, and this principle of economy is a good-making feature of an explanation. Well, it might be a good-making feature of an explanation given certain data. It is not a good-making feature of an explanation, however, when the ontological commitments implied by that explanation are simply used to eliminate the data that need to be explained. It is this eliminativist nature of physicalism that undermines M’s enumerative induction argument for physicalism. According to M, “[t]here is a strong empirical case for holding that concrete phenomena of many different kinds—chemical, biochemical, cell-biological, histological, physiological, geological, meteorological, and astronomical—are physical or physically realized…. Of course, that all nonmental phenomena are physical or physically realized doesn’t entail that mental phenomena are too. But it can still be some evidence that they are.” It can be some evidence that they are, if you have already eliminated all of the fundamental data of the mental that are not physical or physically realizable. We agree wholeheartedly that if there are no events such as uncaused choices that are ultimately explained teleologically and experiences of pain and pleasure that have intrinsic, nonrelational natures, then there might be some reason to think that what is mental is physical or physically realizable. This eliminativist position, however, will not convince anyone who is not already a committed physicalist. In the end, it seems to us that physicalists simply assert that the mental is physical and, because it is, anything about the mental that seems not to be physical must simply be eliminated.
It is important that we stress that we do not deny that there are neural-mental correlations of the kind that M highlights. What we insist upon is that these correlations are discovered and the fact that they are discovered presupposes that there are two metaphysically distinct kinds of events, physical and mental, whose instantiations are temporally correlated. In other words, correlation of this kind is not sufficient for identity. M is well aware that a dualist can acknowledge the existence of correlations and two-way causal interactions between physical and mental events. What often goes unmentioned by physicalists is that the seemingly substantive simplicity of the self (again, see our essay) is presupposed by, and sets the agenda for, an interesting issue in brain science known as the binding problem. The binding problem originates with first-person experience and the unity of our conscious lives.[2] Neurobiologists wonder about how this unity relates to the workings of the complex brain. For example, scientists are aware that the visual system has cells and regions of the brain that are especially responsive to stimuli originating from properties (e.g., color, lines, angels, shape, and movement) of physical objects. When we see a physical object, however, we have a first-person unified experience of a single object. The neurobiologist is interested in discovering where in the brain all of the effects of these diverse stimuli are bound together into a single, unified visual experience of an object. In other words, in light of the unified nature of our first-person visual experience, the neurobiologist searches for a corresponding single point in the brain which in virtue of its singularity captures the nature of our first-person experience.
The binding problem also encompasses different modes of perception. For example, because one of us is currently hearing voices in the distance, feeling the keys of his computer, smelling the odor of brewing coffee, and seeing words on the screen, the neurobiologist is interested in locating the singular spot in this person’s brain wherein all of the effects of corresponding stimuli are bound together. So far, scientists have failed to find what they are looking for. Our point, however, is not that the failure to find a point of binding in the brain is evidence for the existence of the soul as the substance in which the binding is located. Rather, it is that the mere fact that the binding problem exists is confirmation of the reality of the apparent substantive simplicity of the self. The existence of the binding problem is confirmation that practicing scientists themselves, unlike philosophical physicalists, take seriously our first-person experience of ourselves as unified, simple subjects. And it is because we take seriously our experiences of ourselves as simple substantial individuals that we remain convinced that the dualist view of the self is true.
Though the binding problem presents a problem for M’s physicalism of acknowledging and accounting for a subject’s unity at any particular moment of time, it leads to a corresponding problem of acknowledging and accounting for the identity of that subject over, across, or through time. For example, while it is true that you are the numerically same person over seven years, it is not the case that (strictly speaking) your body is the numerically same body over seven years, as most of its parts are acquired and lost during that time. The bare fact of personal identity over time haunts the physicalist elimination of the soul.
2. Mind and Theism
Though we are at odds with M’s physicalism, we find some of his comments about the evidential relationship between dualism and theism very informative. For example, we agree with M’s statement that “given theism’s commitment to a nonphysical and nonphysically realized mind, its prior probability will be raised if we already know of examples of nonphysical … minds.” Of course, where we differ with M is that we believe, while he does not, that there are examples of nonphysical minds. Moreover, we agree with M that in order for theism to “avoid being disconfirmed by the existence of moral evils, theism requires that human agents freely choose to act in one way rather than another.” Of course, where we differ with M is that we believe, while he does not, that human agents indeterministically freely choose to act in one way rather than another for ultimate and irreducible purposes. In short, we agree with M’s more general point that if one believes that this world is a certain way, then one will be much more inclined to believe that theism is true.
What about God’s creative activity, given the truth of theism? M compares an account of God creating the world whereby, in one case, physicalism is true and in the other dualism is true. He suggests that in the latter case, God would need to do more work. We understand creation to be, in one sense, complex (God wills there to be and to be sustained a complex, contingent good world) and singular and unified (God wills that the world be itself unified causally). So, on a dualist view, there is no reason to think that God would sequentially be involved in many discrete acts of creating this or that law, that moon or that bush, and so on. The whole of creation can reasonably be understood as the outcome of a singular but determinate divine will.
In conclusion, a theistic framework successfully accounts for both the reliability of fundamental physics and the origination of conscious life. Unlike physicalism, it recognizes the reality of our mental lives and exhibits the extraordinary simplicity of providing a single explanation for them and our bodily lives. In terms of economy of explanations, theism has a comprehensive, explanatory power that locates the existence and continuation of the cosmos with all its laws in a single, divine, good reality.
Continue the Debate
Notes
[1] Even if the heat of the burner caused an initial small movement of M’s hand away from it before he experienced pain, this movement was quickly followed by movements whose source was M, not the heat of the burner, and which were teleologically explained in the way that we describe in light of M’s experience of pain.
[2] For a nice explication of the binding problem, see John Searle’s “The Mystery of Consciousness: Part I,” The New York Review of Books (November 1995): 60-66.
Copyright ©2007 Stewart Goetz and Charles Taliaferro. The electronic version is copyright ©2007 by Internet Infidels, Inc. with the written permission of Stewart Goetz and Charles Taliaferro. All rights reserved.