The mind-body problem is the problem of explaining how the mind and body are related. Put that way, the problem seems singular even if not simple. Is the mind simply the body or some part of it? If it is not, then what is it and how is it related to the body? Most important, if the mind is not the body, how do they interact?
The most obvious strategy for solving the problem is to insist that minds are bodies or some proper part of the body, perhaps the brain or the central nervous system. This position, the identity theory, claims that minds and brains are identical. René Descartes famously argued that the identity theory must be false. According to Descartes, it is conceivable that minds exist without bodies, and so it is possible that minds exist without bodies, and so minds are not bodies. Now the relationship between what is conceivable and what is possible is tricky and has attracted much attention. It is dubious that Descartes is licensed to move from conceivability to possibility. But what of the move from the notion that it is possible that minds are not bodies to the conclusion that minds are not bodies? That move might seem a non-starter. Much that is possibly the case is not actually the case. But what Descartes assumes, at least implicitly, is that if minds and brains are identical, then they must share all of their properties in common. Descartes assumes that whatever is identical must be indiscernible: For any two objects, x and y, if x is identical to y, then for any property x has y has, and vice versa. And so if minds are brains, then whatever property the one has the other also has. But it is not possible for brains to exist without brains. And so, if minds are brains, then minds could not exist without brains. Thus, if it is possible for minds to exist without brains, then it follows that minds are not brains.
Descartes’ response to the mind-body problem is that minds and bodies are radically different kinds of things; the latter material, the former immaterial. But Descartes’ solution, dualism, introduces two new problems. The first challenge is to make intelligible what an immaterial thing might be. The second problem, related to the first, is the problem of mental causation: How do minds causally interact with bodies? This problem for Descartes is especially acute: How do immaterial substances interact with material substances?
The contemporary version of the mind-body problem is not cast in terms of substances and is often cast, instead, in terms of laws, states, properties, or events. In terms of properties, for instance, the mind-body problem is this: How are mental properties related to physical properties? That is, how are mental properties related to the properties of interest to the natural sciences? And the problem of mental causation, in its modern guise, concerns the causal closure of the physical. It is commonly presumed that the laws of physics are causally closed; there are no physical events that are not caused by other physical events. Were this not the case, then there would be, at the explanatory level of physics, miracles: physical events with no physical explanations. But if the instantiations of mental properties are causally efficacious with respect to the physical, if having some mental property is the cause of some behavior, for instance, then we are left with only three possibilities. First, there are physical events that are not physically caused, and so there are physical events without physical causes. Second, whatever the mental causes the physical also causes, and so behavior is causally overdetermined. Third, mental properties are identical to physical properties. The first option is the modern version of Descartes’ solution. The third is the modern version of materialism, what is commonly called “physicalism.” The second? Well, about that there is much contemporary debate, to which we will return.
A more radical way to respond to the problem of mental causation, however, is simply to deny that the mental is causally efficacious. One way to do this is to claim that the mental, although it might be caused by the nonmental (e.g., the feeling of pain might be caused by a broken bone) and although it might cause what is mental (e.g., someone's experiencing pain might cause that person to remember being in pain), does not cause anything nonmental (and so the feeling of pain would not cause someone to avoid the cause of that pain in the future). This position, epiphenomenalism, has few contemporary adherents. The problem for the position is that, first, it seems obvious to many that the mental is causally efficacious, and second, if the mental does not cause the physical (if, e.g., it does not cause behavior), it is hard to see why we should believe it to exist.
Early in the 20th century, the more popular response to dualism's troubles was to deny the mental entirely, or at least to deny it any explanatory role. According to psychological behaviorism, what is explanatorily relevant is what is observable, and what is observable with respect to explaining human behavior are environmental stimuli and behavioral outputs. Psychological behaviorism attempts to “solve” the mind-body problem by denying the mental entirely. Another form of behaviorism, logical behaviorism, agrees with psychological behaviorism that what matters is what is observable, but insists that the mental is causally efficacious, or at least that appeals to the mental play an explanatory role. Logical behaviorists attempted to define mental terms in terms of behaviors or dispositions to behave. By comparison, we might think of something's being fragile as its being disposed to shatter if struck. If it shatters when struck, we might say that it shattered partly due to its fragility. But its fragility, it might be thought, just is its being such that it would shatter. There is an explanation, logical behaviorism contends, even though the explanatory circle is small. Logical behaviorism does not deny the mental, per se. Ordinary psychological explanations have a purchase. But, according to logical behaviorism, the mental is not separate from the body. Indeed, logical behaviorism denies that the mental is at all private.
In philosophy, the popularity of both versions of behaviorism went hand in hand with a theory of meaning that was popular during the early part of the 20th century, the verificationist theory of meaning, according to which the meanings of theoretical terms were supposedly the verification conditions for their correct applications. Verificationism and the behaviorist theories following in its train now have few adherents. With better theories of meaning and reference, and especially with better theories with respect to theoretical terms, both philosophy and the behavioral sciences once again began appealing to hidden objects, states, and processes without embarrassment.
Not everyone agreed that these included the mental, however. The latter part of the 20th century introduced new versions of eliminativism. According to defenders of this view, attributions of propositional attitudes (e.g., beliefs and desires) belong to a folk theory for explaining behavior. As a theory, this folk theory is subject to revision and even elimination as better theories are developed. If the best theories for explaining human behavior appeal to neuroscience, and if our folk theories are not reducible to the theories of neuroscience, then, according to these eliminativists, we should conclude that our folk theories of the mind are not simply incomplete, but false.
Another strategy that surfaced during the second half of the 20th century, and that has attracted far more attention, is functionalism. According to this view, someone is in pain, for instance, whenever she is in a state that plays a particular functional role, which is to say whenever she is in that state commonly caused by pain stimuli (e.g., pin pricks) and that commonly produces pain behavior (e.g., moving one's arm quickly away from the pin). Functionalism shares with behaviorism the idea that the meanings of mental predicates are tied with behavior, but it breaks with behaviorism in that, according to functionalism, mental predicates pick out properties and states that not only cause behavior but also cause each other. Pains produce not only behavior but also beliefs and memories. And part of the commonsense explanation for this appeals to the fact that we reflect on our memories, beliefs, and desires and that this reflection results in additional memories, beliefs, and desires.
One version of functionalism, realizer functionalism, is generally thought of as a version of the identity theory. According to realizer functionalism, because mental terms pick out whatever it is that plays a particular functional role, and because it is presumably the job of neuroscience to tell us what that is, mental state terms (at least when used to ascribe mental states to us) pick out neurophysiological states (of us). Thus understood, realizer functionalism solves the problem of mental causation by insisting that mental properties just are certain physical properties, and so there is no threat to the causal closure of the physical and there is no worry about causal overdetermination.
Another version of functionalism, role functionalism, holds that mental terms pick out properties and states at a higher level of abstraction, perhaps those that humans might share with machines having no neurophysiology. As commonly understood, role functionalism holds that mental properties are second-order properties—properties had by properties. Someone is in pain, according to this view, in virtue of having some first-order property (presumably some physical property) the instantiation of which plays the appropriate functional-causal role. So, for instance, if having some neurophysiological property plays the pain role for us, if having that neurophysiological property is typically caused by pin pricks, for example, and typically causes pain behavior, then we are in pain whenever that neurophysiological property is instantiated in us. But that neurophysiological property is not identical to pain. Rather, pain is the having of some property or other (in our case that particular neurophysiological property) that plays the relevant functional-causal role.
Role functionalism is theoretically compatible with both dualism and physicalism about the mind. It is compatible with dualism since it allows that the instantiation of some nonphysical property might play the appropriate functional role. It is compatible with physicalism since, presumably, nothing nonphysical actually plays that role. According to role functionalism, the property of being in pain is not, strictly speaking, a physical property or any particular physical property, and so creatures physically very different than us, and perhaps even machines of the right type, might share the property of being in pain. But assuming that pain is always realized physically, assuming that everything that is in pain has some physical property the instantiation of which plays the functional role of pain for that creature or machine, there is an important sense in which dualism is false.
Exactly how and whether dualism is false and physicalism is true by the lights of role functionalism has itself been an area of intense debate. According to one particularly influential view, token-token physicalism, although mental types (e.g., pain) are not identical to physical or neurophysiological types, particular instances or tokens of the mental (e.g., someone's particular pain at some particular moment) are identical to particular instances or tokens of the physical. By analogy, no one thinks that a particular shade of red, say crimson, is identical to redness, because something can be red but not be crimson. Nonetheless, it might be held, something's being crimson is just its being red.
Role functionalism was once thought to be a kind of philosophical panacea, a halfway house between a dualism that seems shrouded in mystery and a reductive physicalism that seems counter to common sense, and it continues to be the favored view in the philosophy of mind. But role functionalism faces several objections. According to one objection, being functionally organized in a particular way is insufficient for intentionality. We can imagine, it is thought, a creature or a robot with the right kind of functional organization, but without what we would think of as thoughts or understanding.
It is also challenging to see how functional organization is sufficient for consciousness, but here the problem is seemingly a problem for any theory of the mind, and especially for any physicalist theory. It seems possible, for instance, that someone functionally like me, someone who typically sees blue objects as blue, who typically comes to believe that such objects are blue after seeing them, who typically asserts that such objects are blue when asked, and so forth, might have experiences of blue objects that are qualitatively different than the experiences that I have when I see blue objects. We are functionally the same, she and I, but we are not mentally the same. The problem, the qualia problem, is to explain how functionalism in particular, and physicalism more generally, can explain the qualitative nature of many of our mental states.
Where the qualia problem raises the worry that functionalism cannot avoid all of the problems of physicalism, another problem suggests that role functionalism cannot avoid a perennial problem for dualism. The problem, the problem of mental causation, is that if mental properties are not identical to physical properties, then it would seem that their instantiations would have nothing to do. One version of the problem maintains that any causal explanation of an event excludes all other explanations. The problem, if such there is, would seemingly apply to the properties appealed to by most of the natural sciences. Indeed, it would seem to raise a problem for all of the sciences except for physics since, if the argument were successful, it follows that the lower level causal explanations exclude all other higher level causal explanations. But role functionalism seems to face special challenges in explaining how the mental could, by its lights, be causally efficacious. If a mental property is a second-order property, if it is the having of some property or other that plays a particular functional role, then it would seem to be the first-order property that does the work by definition. And if instances of mental properties are identical to instances of physical properties, if the role functionalist endorses token-token identity, then those instances must play exactly the same causal role. And if they play the same causal role, if instantiations of mental properties never cause anything not caused by instantiations of physical properties, then that would seem a good reason to think either that mental properties do not exist or that they just are those physical properties.
One particularly interesting response to this problem is to think of mental property tokens not as identical to physical tokens, but instead to think of them as proper parts of physical tokens. The position agrees with realizer functionalism that mental properties are not second-order. And it agrees with dualism, but here a dualism about properties and not substances, that instances of mental properties are not identical to instances of physical properties. It also agrees with role functionalism that the mental is physical in virtue of being multiply realizable by the physical. To understand the suggestion, imagine an overly simplified view, a popular philosophical fiction. Imagine that pain in humans is correlated with C-fibers firing and that pain in Martians is correlated with A-Fibers firing. According to traditional role functionalism, if John the human is in pain, then his being in pain is not his having C-fibers fire, although it is true that he wouldn't be in pain if his C-fibers were not firing. His being in pain is his being functionally organized in the appropriate way and, thus, his having something or other that is playing the pain role. Perhaps we might say that his being in pain at this moment is identical to his having C-fibers firing at this moment, but being in pain is not having C-fibers fire. But then it is unclear what causal role there is for pain to play for John since any instance of pain in John just is an instance of C-fibers firing in John.
According to traditional realizer functionalism, if C-fibers firing is the realizer of pain for humans, then it just is pain for humans. And because A-fibers firing just is pain for Martians, it just is pain for them. But then it appears that humans and Martians, although it is true to say of both that they experience pain, actually share nothing in common when they are both in pain.
If we think that pain is multiply realizable, if we agree with the Role Functionalist that humans and Martians have something in common when in pain, then we must also think that the causal powers of C-fibers firing and the causal powers of A-fibers firing are relevantly similar. No doubt they do not share all of their causal features, for otherwise they would not be neurophysiologically distinct, but their causal features must overlap with respect to those features relevant for pain, for otherwise we would not treat them as alike psychologically. Pain, on this view, contributes the set of causal powers that is a proper subset of those powers contributed by both C-fibers firing and A-fibers firing, and so an instance of pain might be thought of as a proper part of any instance of C-fibers (or A-fibers) firing. Whether this strategy will prove successful is as yet a matter of considerable controversy.
- Anomalous Monism
- Consciousness and Embodiment
- Eliminative Materialism
- Explanatory Gap
- Mental Causation
- Reductive Physicalism
- The conscious mind. Oxford, UK: Oxford University Press. (1996).
- Scientific realism and the plasticity of mind. New York, NY: Cambridge University Press. (1979).
- Consciousness explained. Boston, MA: Little, Brown. (1991).
- Special sciences and the disunity of science as a working hypothesis. Synthese, 28, 77-115. (1974).
- Epiphenomenal qualia. Philosophical Quarterly, 32, 127-136. (1982).
- Physicalism, or something near enough. Princeton, NJ: Princeton University Press. (2005).
- Naming and necessity. Cambridge, MA: Harvard University Press. (1972).
- An argument for the identity theory. Journal of Philosophy, 63, 17-25. (1966).
- Mind, language, and reality. Cambridge, UK: Cambridge University Press. (1975).
- Physical realization. Oxford, UK: Oxford University Press. (2007).
- Mental causation. Philosophical Review, 101, 245-280. (1992).
Related Credo Articles
For centuries, scholars have struggled to define the nature of the human being. One of the key questions in this struggle deals with mind,...
Glossary Consciousness (phenomenal) What it is like to have or undergo an experience; the ways things seem to feel to you when you...
Roughly speaking, “the mind-body problem” has to do with the relation between minds and bodies or between psychological events and physical...