Physicalism, of which materialism is a historical antecedent, is primarily an ontological doctrine concerning the nature of reality and, specifically, mental reality. It is the view that reality is ultimately constituted or determined by entities—objects, events, properties, and so on—that are physical. This thesis is often combined with a claim about the explanatory supremacy of physical theory (physics).
Any formulation of physicalism raises the question, What is meant by "physical"? It is difficult to formulate a conception of the physical that is neither too strong, making physicalism obviously false, nor too weak, making physicalism trivially true. For example, what is physical may be simply identified through the language of physics. However, a problem arises over the conception of physics appealed to. Current physics seems too narrow because future extensions of physics would not count as physical; but the idea of a completed physics is too indeterminate because there is no clear idea of what that physics might include. One could attempt to characterize the physical in more general terms such as having spatial location or being spatiotemporal. However, this threatens to make physicalism trivially true because mental phenomena seem clearly to have spatial location in virtue of having subjects—persons—who have bodies. It may be preferable to appeal to the idea of a completed physics. Although at any particular time people may not know exactly what is physical and what is not (because they may not know whether they have completed physics), nevertheless what is physical is all and only what a completed physics countenances.
There are two main types of physicalist theses. First, there is eliminative materialism, or physicalism. According to this there are not, and never have been, any mental entities, events, properties, and so forth. Strictly speaking, this is not a view about the nature of mental reality. Second, there is a group of doctrines that fall under the general heading of identity theories, some of which are stronger than others. These can be divided into two main categories. The stronger doctrines may be called type-type identity theories, or type physicalist theories (Armstrong 1968, Lewis 1966, Place 1956, Smart 1959), and the weaker doctrines may be called token identity theories, or token physicalist theories (Davidson 1970, Macdonald 1989, Macdonald and Macdonald 1995).
Physicalist theories need to account for at least two different kinds of mental phenomena. First, there are the sensations, such as color experiences, pains, tingles, itches, and the like, which are typically, and perhaps essentially, identified in terms of how they feel to their subjects. Then there are the intentional states or events, such as beliefs, hopes, desires, and thoughts, which are typically, and perhaps essentially, identified in terms of their intentional contents, or their "aboutness." For example, a person's belief that water is transparent has the intentional content, water is transparent ; a content that represents the world around that person in a certain way, irrespective of whether the world happens to be that way. One of the biggest difficulties for physicalism is accounting for both of these kinds of mental phenomena. In the late twentieth and early twenty-first centuries, philosophers have expressed skepticism as to whether a thoroughgoing physicalist position is possible, and have maintained that physicalism (either token, or type) is true of at most one of these two kinds of mental phenomena (Chalmers 1996, Kim 1998).
Consider any mental phenomenon, such as being in pain now, or thinking right now that water is transparent. It is possible to talk about this phenomenon as an individual occurrence of a certain kind in the mental life of a person and discuss its properties. It is also possible to talk about the kind of phenomenon—pain, or the thought that water is transparent—of which this event is an individual instance. Physical phenomena too can be discussed in both of these ways. Type physicalism is the view that the mental types, properties, or kinds under which mental phenomena fall are identical with physical types, properties, or kinds. For example, pain—that type of phenomenon, occurrences of which are individual pains—is identical with some single type of physical phenomenon such as C-fiber stimulation.
Type physicalism has its origins in the doctrines espoused by the logical positivists and central-state materialists (Place 1956, Smart 1959). It is a strong form of physicalism because it is reductionist. Many who endorse it believe that nothing short of it counts as a proper physicalism. They argue that even if it is in practice impossible for sentences containing mental terminology to be translated into or replaced by sentences containing physical and topic-neutral terminology, any view that holds that all mental phenomena are physical phenomena, but mental properties or kinds are not physical properties or kinds, is not worthy of the name "physicalism."
the first objection to type physicalism
Type physicalism suffers from two serious objections. The first, from phenomenal properties, specifically concerns sensations such as color experiences, pain, afterimages, and the like. It is that phenomena of these kinds or types have "felt" properties, such as being reddish, stabbing, or vivid, whereas phenomena of physical types do not. Given this, and given Leibniz's principle of the indiscernibility of identicals, it follows that sensation types are not identical with physical types because the phenomena that fall under them do not share all of the same properties. A variant of this objection focuses on the distinctive point of view a subject has on its own experiences: A subject knows what it is like to have experiences in a way that others do not, and this subjective mode of access reveals the phenomenal aspect of the experience, whereas an "other"-oriented point of view does not (Nagel 1974).
One response is to argue that the problem is purely conceptual and does not threaten physicalism, which is an ontological view about what sorts of things there are in the world, not a view about concepts (Levine 2001, Loar 1997, Tye 1999). Consider the type-type identity expressed by "Brain State B is the red-feeling sensation." To the objection that such identities are false because first-person access to experiences reveals them to have properties that physical states do not have, the response is that the apparent difference in properties arises from the distinctive nature of human experiential (or phenomenal) concepts alone. Certain concepts, such as the concept red-feeling sensation (or reddish sensation ), are ones that can only be possessed by being put into direct contact with experiences that fall under them, without the mediation of other information or concepts that one might have of those states. Because the phenomenal concept red-feeling sensation enables subjects to be put in direct contact with their own red-feeling experiences in a way in which no concept of Brain State B could do, it puts them in a position to recognize directly and in an immediate way their own phenomenal red-feeling experiences. Possession of the concept Brain State B could not put any subject in a position to recognize directly and in an immediate way its own red-feeling experiences. So, even having met the experiential requirement on the possession of the concept red-feeling experience, a subject might be under the illusion of thinking that the red-feeling sensation has a property that Brain State B lacks. Whether or not this response succeeds depends on whether, in acquiring a new concept, such as the concept red-feeling sensation, one learns a new fact about the world that one did not know before, despite being in possession of the concept Brain State B.
the second objection to type physicalism
The second objection to type physicalism is that from multiple realizability. This claims that mental kinds or properties may be realized in physically diverse types of ways, hence there is no single physical property with which a given mental property may be identified. The point is that even if each mental property were in fact to be realized by a single physical one, it is possible for it to be realized by physically diverse ones. The reason is that the introspective and behavioral basis upon which attributions of mental properties are typically made is silent on the potential internal physical realizers of them. Given the claim that identical things are necessarily identical, the mere possibility that a given mental property should be realized by a physical property other than that which in fact realizes it is sufficient to refute the claim that that mental property is identical with any physical property that may realize it. This objection is not independent of a modal argument that trades on the thesis that identical things are necessarily identical (Kripke 1980). This begins with the conceivability of a mental state type's existence in the absence of any physical type of state, and argues that, because what is conceivable is possible, it is possible that mental state types could exist in the absence of any physical state type. The argument concludes that, because it is possible that mental types should exist in the absence of any type of physical phenomenon, mental state types are not identical with any type of physical phenomenon. A version of this argument is held to be particularly decisive against type physicalism with respect to sensation states.
One response is to argue that mental types are identical with disjunctions of physical types. For example, pain may not be identical with C-fiber stimulation, but it may be identical with the disjunctive property, C-fiber stimulation, or A-fiber stimulation, or …, and so on (properties picked out by predicates formed by disjoining predicates that pick out all the possible physical realizers of mental properties). However, it is unclear whether these are bona fide properties. They do not have a unity of their own, viewed from a physical perspective; and it is arguable that a reason is needed, apart from the fact that they all realize a given mental property, to think that they are properties in their own right (Macdonald 1989).
Against this, it might be claimed that because any given mental predicate may correlate with an indefinite number of physical predicates, this may pose problems for formulating laws connecting mental with physical properties; but it does not follow that there is not a single physical property that is the extension of a given mental predicate. Mental properties are identical with the physical properties picked out by disjunctive physical predicates, but their autonomy is secured by their participation in real regularities, and so they do have a unity of their own, despite being identical with disjunctive physical properties (Antony 2003).
In a similar but more radical vein, it might be claimed that although there are mental and physical predicates, there really are only physical properties, so there are no type-type identities of any kind that might be problematic for physicalism (Kim 1998). This reductionist response avoids the problem of multiple realizability altogether, but only by taking an eliminativist stand on mental properties. Alternatively, it might be claimed that the only type-type identities licensed by physicalism are species-specific (as in, for example, that expressed by "pain in humans is identical with C-fiber stimulation"). None of these claims is unproblematic: the first, because it threatens to make mental properties non-nomic, which seems to undermine the commitments of type physicalism; the second, because it is eliminativist; and the third, because it leaves questions such as "What makes pain in humans and pain in dogs both pain?" unanswered.
Many consider one or the other of the above objections to be decisive against type physicalism and have opted instead for a weaker view: token physicalism. According to this, each individual mental event or phenomenon is identical with some physical event. One influential version of this is the view known as anomalous monism (Davidson 1970). Token physicalism is compatible with the multiple realizability of mental properties by physical ones because it is not committed to the view that each individual occurrence of a given mental kind is identical with an occurrence of the same type of physical phenomenon. It also appears to avoid the objection from phenomenal properties in its original form because it can concede that mental kinds have associated with them felt aspects with which no physical kinds are associated. To the objection that mental events are not identical with physical events because it is no part of the nature of any physical event that it have a felt aspect, the following reply can be made. If token physicalism is true, no physical event is essentially of a mental type; but given that it is of a given mental type, it has what is essential to being of that type. Thus, if this pain is identical with this C-fiber stimulation, then it is not essentially a pain. However, given that it is, as it happens, a pain, it has (though not essentially) what is essential to being of that type, namely being felt.
Without an explanation of how mental types relate to physical ones, token physicalism threatens to succumb to the charge that it is dualist because it countenances the existence of nonphysical properties or types. A common strategy is to advance a supervenience doctrine concerning the relation between mental and physical properties, according to which physical properties, although distinct from mental ones, in some sense determine them (Hellman and Thompson 1975). There are many varieties of supervenience theses. One difficulty is in finding a thesis strong enough to do justice to the claim that physical properties determine mental ones without being so strong as to entail identities between mental and physical properties or types, and with these, reducibility. Another, related problem, is explaining how it could be that mental types or properties supervene on physical ones in a way that dispels the worry that mental properties have no causal powers of their own.
See also Causal Closure of the Physical Domain; Dualism in the Philosophy of Mind; Functionalism; Mind-Body Problem; Nonreductive Physicalism; Philosophy of Mind; Reduction; Reductionism in the Philosophy of Mind.
Antony, L. "Who's Afraid of Disjunctive Properties?" Philosophical Issues 13 (2003): 1–21. An argument for the identity of mental properties with disjunctive physical ones.
Armstrong, D. M. A Materialist Theory of Mind. London: Routledge & Kegan Paul, 1968. Defense of a type-type identity theory of the mental and physical.
Block, N., ed. Readings in the Philosophy of Psychology. Vol. 1. Cambridge, MA: Harvard University Press, 1980. Articles on type-type identity theories, token identity theories, reductionism, and functionalism.
Davidson, D. "Mental Events." In Experience and Theory, edited by Lawrence Foster and J. W. Swanson. Amherst, MA: University of Massachusetts Press, 1970: 79–101. Highly influential argument for a token identity of the mental and physical.
Hellman, G., and F. W. Thompson. "Physicalism: Ontology, Determination, Reduction." Journal of Philosophy 72 (1975): 551–564. Defense of a supervenience doctrine.
Jackson, F., R. Pargetter, and E. Prior. "Functionalism and Type-Type Identity Theories." Philosophical Studies 42 (1982): 209–225. Discussion of the relation between functionalism and type-type identity theories.
Kim, J. "Supervenience as a Philosophical Concept." Metaphilosophy 12 (1990): 1–27. Discusses supervenience as a covariance relation and its relation to reduction.
Kim, J. Mind in a Physical World. Cambridge, MA: MIT Press, 1998. A reductionist physicalist position regarding intentional states only that suggests handling the multiple realizability objection either by denying that mental properties are genuine or by maintaining that mental properties are identical with disjunctive physical ones.
Kripke, S. Naming and Necessity. Oxford: Basil Blackwell, 1980. A modal argument against type-type and token identity theories.
Levine, J. Purple Haze. Oxford: Oxford University Press, 2001. A defense of the view that phenomenal property problems with type physicalism are purely conceptual.
Lewis, D. "An Argument for the Identity Theory." Journal of Philosophy 63 (1966): 17–25. An argument for a type-type identity theory.
Loar, B. "Phenomenal States." In The Nature of Consciousness, edited by N. Block, O. Flanagan, and G. Guzeldere, 597–616. Cambridge, MA: MIT Press, 1997. A defense of the view that phenomenal property problems with type physicalism are conceptual.
Macdonald, C. Mind-Body Identity Theories. London: Routledge, 1989. A survey of type-type and token identity theories, and a defense of a token identity theory.
Macdonald, C., and G. Macdonald. "How to Be Psychologically Relevant." In Philosophy of Psychology: Debates on Psychological Explanation, edited by C. Macdonald and G. Macdonald. Oxford Basil Blackwell, 1995. Advances an argument for a token identity theory that attempts to explain how mental properties could supervene on physical ones in a way that dispels the worry that mental properties have no causal powers of their own.
McGinn, C. "Can We Solve the Mind-Body Problem?" Mind 98 (1989): 349–366. An argument for the view that the mind-body problem cannot be solved because of the conceptual disparities between the mental and the physical.
Nagel, T. "What Is It Like to Be a Bat?" Philosophical Review 83 (1974): 435–450. Argues against identity theories of the mental and the physical on the basis of a disparity between the mental and physical in points of view.
Place, U. T. "Is Consciousness a Brain Process?" British Journal of Psychology 47 (1956): 44–50. A defense of central-state materialism for sensations.
Smart, J. J. C. "Sensations and Brain Processes." Philosophical Review 68 (1959): 141–156. A defense of central state materialism for sensations.
Tye, M. "Phenomenal Consciousness: The Explanatory Gap as a Cognitive Illusion." Mind 108 (1999): 705–725. A defense of the view that phenomenal property objections to type physicalism are due to a confusion about phenomenal concepts.
Cynthia Macdonald (1996, 2005)