There are two kinds of anti-reductionisms about the mind. One is methodological in which one is skeptical that a reductive theory of the mind can be produced about the mind given the limits of science. The other is metaphysical anti-reductionism which commits to mental kinds not being type-identical to the canonical language of physics. One need not commit to metaphysical anti-reductionism to be a methodological anti-reductionist, and I think we should be only methodological anti-reductionists.
Metaphysical anti-reductionism is not a tenable theory because the arguments for it are not sound enough to sufficiently demonstrate that we ought to believe that the mind is not physical. Methodological anti-reductionism is more tenable thesis given the limits of using testimony to construct either computational, causal, or neurological identities. To represent arguments for methodological anti-reductionism will be Thomas Nagel's “What Is It Like To Be A Bat?” and to represent arguments for metaphysical anti-reductionism is Frank Jackson's “Epiphenomenal Qualia.” While the arguments covered here are not nearly all of them, the arguments in both works have played a fundamental role in constructing the frame of the debate between reductionists and anti-reductionists.
In “What Is It Like To Be A Bat?”, Nagel argues the essential part of the mind-body problem is consciousness (322). He claims that 'without consciousness the mind-body problem would be much less interesting [and] with the mind-body it seems hopeless” (322). By hopeless, Nagel is referring to the reason that reductionist theories “do not even try to explain [consciousness]” (322). He rightly claims that a complete theory of mind could not “exclude the phenomenological” and “if physicalism is to be defended, the phenomenological features must themselves be given a physical account” (323). Brian Cooney explains that, by phenomenological, Nagel means “the way that an object appears, is experienced or perceived; the way something is for a conscious subject” (323).
Early in Nagel's paper, Nagel concludes from the modal argument and the what-it-is-like-to-be argument that the subjective character of experience is not captured in any of the familiar reductive theories (323). In other words, he concludes that the subjective character cannot be analyzed in terms of functional (or intentional) states or causal roles because the physical theories are compatible with the absence of the mental (323). He also articulates the intuition, which drives this anti-reductionism, that to talk about mind objectively seems to eliminate the subjective, which was the essential feature any complete theory of mind is attempting to explain.
Later in his paper, Nagel admits to the possibility of a reduction and reveals himself to not be a metaphysical reductionist. First, he claims that the external world is the essence of the internal world, not merely a point of view of it (327). This implies that Nagel thinks that the mind depends on a real and existing world to exist. While this does not mean he is making type-identities with specific physical things, it commits him to at least a weak supervenience thesis. Second, he claims that the current concepts do not provide an account of how a reduction of the subjective to the objective may be done (327). This implies that Nagel leaves open the route that it may indeed be possible to create a physicalist theory of the subjective. This seems to contradict his earlier statement about the nature of objective explanations necessarily failing to explain the subjective, but because Nagel uses the word 'seems,' he was not absolute in that stance. Third, Nagel claims that the inadequacy of physicalist hypotheses do not justify the conclusion that physicalism is necessarily false (328). Because Nagel allows for the possibility of metaphysical reductionism but provides skeptical arguments to at least the current reductive programs, he is an example of a methodological anti-reductionism.
While I think that methodological anti-reductionism is reasonable currently, there is significant arguments against it. I can hardly cover all of them here, but I will cover the three most significant. The first is the explanatory argument. It is the strongest in my opinion and is the reason I am not a metaphysical reductionist. The explanatory argument is that there are certain questions about the nature of the mind that are sufficiently explained by an identity or functional thesis. The reason that a relationship holds between two mental states is explained by the same reason that a relationship holds between two physical states. The reason I have a mental state rather than another person is explained by the fact that the mental state has something to do with physical location of my parts which other people do not have. The reason that mental events happen simultaneously with physical states rather than after or before physical states is because they are identical with at least something about physical states.
The problem with the explanatory argument for reductionism is demonstrated by Nagel's modal argument. Basically, if we can conceive of a person with identical physical facts as we but lacking a mind, we have not sufficiently explained mental events with physical events. As long as the physical theories of minds are compatible with its absence, it is not really a theory of mind but merely a theory of brains or computers. The modal argument does not mean that reductionism is false, but it challenges that these theories have discovered type-identities for mental kinds in the physical. The type-identity is essential part of many reductions because type-identity is a discovery of the essence of something. If one discovers the essence of something, that holds in all possible worlds. Nagel is right that we have not reach the point in which we can say that we have discovered the nature of the phenomenological, but he leaves it open that we might develop concepts that will make a reduction possible.
The second argument for reductionism is an induction from the history of successful reductions. Scientists reduced heat to mean kinetic energy, life to biological processes, fire to combustion, etc.... The argument is that there is no reason to believe that the mind cannot be reduced in the same way. Those who conservatively hold to the old dualistic intuitions are foolishly holding to a kind of thought that has been wrong about so many things like the nature of mental illness, sleep, and learning (Churchland 132). We should therefore feel confident in our our physicalism about mental states.
The weakness with this inductive argument is that this kind of reduction is not identical to past reductions. Past reductions have always been objective-to-objective, not subjective-to-objective. While there is a phenomenological quality to heat, what we are trying to explain when reducing heat to mean kinetic energy is its objective character. This does not mean there is not an objective story that separates our brains processing of sense data from the external referrant like light frequency, but rather that the subjective character is incompatible with current physicalist reductions, as Nagel suggests.
The third argument is that qualia are not essential to mental states. Paul and Pat Churchland argue this in their response to the inverted spectrum argument in “Functionalism, Qualia, and Intentionality.” They argue that when we are talking about the mental state red, how we experience it is not essential in our description. They use a tiger for an example. While the tiger is initially discriminated by its black stripes and yellow fur, there are also albino tigers (Churchland 351). If the phenomenological character were different, the functional role would remain the same for tigers. In other words, “sameness of functional role dominates over differences in qualitative character” (Churchland 352).
This argument is sufficient to explain our mental reference but it is not enough to rid the subjective character's existence from the equation. What makes the mind interesting is not this functioning but the subjective character. Even if we could provide a complete story of the mind in terms of functioning or identities, we would leave the question of 'what are qualia.' They seem to be more than functional because, so conceived, they are only contingently connected with functional roles, hence the non-essential role they play in mental states.
What demonstrates the importance of qualia to our understanding of consciousness is contained in the What-it-is-like-to-be argument. Nagel argues that no amount of physical information about bats can enable us to imagine what it is like to be a bat. When we imagine echo-location, we are just humans imagining we are bats, not actually being bats. While this argument relies on pointing out physical differences (e.g. between bats and humans), it clearly demonstrates that there is something worth explaining (e.g. what-it-is-like-to-be) that cannot be explained physically as of yet.
Given that we are methodologically prevented from making reductions, at least currently, are we justified in being meta-physical anti-reductionists about qualia? Frank Jackson in “Epiphenomenal Qualia” thinks so. He covers many of the same arguments as Nagel, so I will not repeat those here. I will focus on his strongest and most famous argument for metaphysical anti-reductionism, the Mary argument, and the explanatory argument for epiphenomenalism, which is not explicitly in his paper.
The Mary argument hinges on Mary learning a new fact when she experiences red after knowing the complete reductionist theory. Mary is a neuroscientist in a black and white room with a black and white television (Jackson 344). She has never seen the color read. She knows all the physical facts about what happens when a person testifies experiencing red in his or her brain (Jackson 344). When she leaves the room, she experiences the color red and learns what it is like to experience red like those who testified about the experience in her books. In other words, she learned a fact about everyone who experienced red. If she knew all the physical facts but still learned a new fact, that logically entails that she learned a non-physical fact. If there are non-physical facts, physicalism is false.
The best counter-argument is that Mary is not learning a new fact but is learning her relation to others experiences. John Perry in “Time, Consciousness and the Knowledge Argument” explains with a metaphor to a map. The idea that Perry is drawing upon comes from D. H. Mellor's “McTaggart, Fixity and Coming True.” Mellor argued that since the question “What time is it now?” is a relational statement between when the statement is uttered and the event of a certain time designation like May 1984 (70). Perry takes this and with an in-between metaphor about location. He nicely demonstrates that Mary's knowledge about the relation between her experience and everyone’s experience is not a discovery of something knew about the world.
The in-between metaphor uses location. Larry has a map. He knows that Salt Lake City is west of Little America. When he sees that he is at Little America, he can use the demonstrative 'here' and say Salt Lake City is west of here” (Perry 78). Larry did not learn a new fact about geography. There is nothing new about 'here' because 'here' in this context is Little America.
The same can be said of Mary. Mary like Larry only learned that she that her brain was having a neurological process. Nothing changed about the factual nature of seeing red, but rather that Mary was now in the seeing relation to red. Since Mary is not learning a new fact, we cannot say there must be non-physical facts, so the argument fails. This does not mean necessarily there is no non-physical facts just that we cannot conclude from this argument that there are.
As for Jackson's other arguments, I have covered them already. Despite what Jackson says, the Fred argument is identical to the what-it-is-like-to-be-a-bat argument from Nagel; the only difference is that Fred sees two distinct forms of red and bats have echolocation. Just as we cannot imagine what Fred's experience is like we cannot imagine bats and our physical explorations do not give us access to it. The Modal argument also does not provide definitive answer as to the nature of the qualia. If qualia end up being type-identical with something physical, the argument is moot.
Without the Mary argument, the only thing left is to attempt to explain why it is impossible to explain qualia. The explanation an epiphenomenologist would use is not that there is a deficiency of our scientific methodology but rather that qualia are in fact not physical. This, however, falls into the trap of an argument from ignorance. 'I cannot understand through reduction; therefore, a non-reductive explanation is true.' This is a common move for arguments for other supernatural entities. Just as it is wrong to use a god-of-the-gap argument it is wrong to posit an epiphenomenal realm to explain our ignorance of the mind. What we should be is agnostic about the nature of the mind, and only methodological anti-reductionists until we find a way of knowing this part of the mind.
Churchland, Pat. Churchland Paul. “Functionalism, Qualia, and Intentionality.”The Place of the Mind. ed. By Brian Cooney. Australia; Wadsworth Cengage Learning: 2000.
Churchland, Paul. “Eliminative Materialism.” The Place of the Mind. ed. By Brian Cooney. Australia; Wadsworth Cengage Learning: 2000.
Jackson, Frank. “Epiphenomenal Qualia.” The Place of the Mind. ed. By Brian Cooney. Australia; Wadsworth Cengage Learning: 2000.
Mellor, D. H. “McTaggart, Fixity, and Coming True.” Metaphysics: Classic And Contemporary Readings. ed. by Ronald C. Hoy and L. Nathan Oaklander. 2nd Edition. Australia; Thomson Wadswoth: 2005.
Nagel. Thomas. “What Is It Like To Be A Bat?” The Place of the Mind. ed. By Brian Cooney. Australia; Wadsworth Cengage Learning: 2000.
Perry, John. “Time, Consciousness and the Knowledge Argument.” Metaphysics: Classic And Contemporary Readings. ed. by Ronald C. Hoy and L. Nathan Oaklander. 2nd Edition. Australia; Thomson Wadswoth: 2005.