for the mental to be causally efficacious given physicalism is for it to be reducible to the physical (Kim 11). I find his argument compelling, and in this paper, I will examine one of the more interesting objections to Kim’s argument and solution to the PMC called the autonomy solution (sometimes called the dual explanandum strategy).

There are essentially two views about the mind that entail that the mind is not reducible to the physical, and thus distinct from the brain. One view is to postulate the mind as a separate entity or substance that causally interacts with the body. This view, called substance dualism, was originally proposed by Descartes. It has been largely abandoned today due to its lack of ability to explain how the mind and body interact, so I will not consider it here. The second view, called property dualism, 46 has become popular in contemporary literature to explain mental causation. Property dualism denies that the mind is a separate substance from the body and holds a physical ontology. It also asserts that when physical systems reach a certain complexity, they exhibit properties that are not physical. For example, property dualists claim that the physical complexity of the human brain has given rise to mental properties, like intentions or beliefs, which are understood as being purely mental and distinct from any physical property of the brain. Moreover, these emergent properties cannot be reduced down to the physical properties (Kim 9). Rather, the mental properties supervene on the physical properties. A is said to supervene on B, if and only if, there can be no change in A without a corresponding change in B. Thus, if a mental property is instantiated, then it has a physical property as its base that instantiates it (Kim 9). As an example, take the desire of mine to get coffee this morning. From supervenience, we can say that the desire I had must have also had a corresponding physical state in my brain that gave rise to or generated my desire. Without this physical state in my brain, the desire could not have been generated, but the desire cannot be explained and understood fully in terms of the physical state—in other words, the mental is not necessarily reducible to the physical even if it supervenes on it.

So, even though there is no separate substance that is the mind, the mind has properties that cannot be explained in terms of the physical such as desires, beliefs, intentions, etc. This seems to leave open the idea for the possibility of mental causation. There are good reasons for considering such a view in order to explain things like consciousness and the fact that mental states are multiply realizable by different brain states. For example, you and I can have the same belief about

Personally, I think the claims that substance dualism fails to be able to explain interaction between the mind and the brain are vastly 46 overstated. Moreover, I think substance dualism is able to make a more accurate account of reality than any of its competitors—by being able to explain consciousness and other subjective experiences; consequently, I think it, or something like it, is true. However, as I have mentioned above, in this paper I am focusing on physicalism’s inability to make sense of mental causation, so I will be ignoring substance dualism for the rest of this paper.

College Station (that it is located in Texas), but different parts of our brains correspond to these beliefs. We have the same mental state but different brain states. However, this is exactly the kind of view that Kim attacks in his paper. If Kim’s argument is valid, then any kind of property dualism entails that mental events are also causally inefficacious.

To understand Kim’s argument, we first must consider what assumptions the argument makes. The argument seeks to undermine any physicalism that is nonreductive about the mental, like property dualism. Physicalism is committed to the causal closure of the physical (CCP): Any physical event that has a cause at time t, had a physical cause at t (Kim 10). This is the first assumption that Kim makes, and it is one that is undeniable for any physicalist. The second assumption Kim makes is the principal of causal exclusion (PCE) which states: If an event, e, has sufficient cause, c, at time t, then no other cause brought about e (Kim 10). The last assumption, the principal of generative exclusion (PGE), is more general than the PCE, and it claims: If an event e, or an instantiation of a property P is generated by an event c, then the existence of e or of P, is not generated by anything other than c (Kim 11). Some of these assumptions may be controversial or debatable, but for the sake of the paper I will assume they are valid and reasonable.

Kim’s argument seeks to refute any kind of mental to mental or mental to physical causation that is proposed by property dualism. So, let us take the example mentioned at the beginning of the paper where my sensation of tiredness caused my desire or intention to get coffee and apply Kim’s argument to it. The argument starts off by claiming that my sensation of tiredness caused my intention to get coffee. Because the mental supervenes on the physical, my intention must have a supervenient physical base. Suppose this base is my b-fibers firing in my brain. But now, my intention has two supposed generative sources, my sensation and my bfibers firing. By the PGE, my sensation and my b-fibers firing cannot both be generative sources for my intention. The only way for my sensation and my b-fibers firing to be generative sources for my intention is to claim that my sensation caused my intention by causing my b-fibers to fire. Thus, my sensation caused my b-fibers to fire. However, by the CCP, my b-fibers firing must also have a physical cause. Suppose this cause is my a-fibers firing in my brain. Now there are two purported causes for my b-fibers firing: My sensation and my a-fibers; a physical and a mental contributor. But by the PCE, one of these causes must go. By invoking the CCP again, my sensation must go as a possible cause, and therefore the firing of my afibers caused my b-fibers to fire, which in turn caused my intention. This

contradicts what we assumed at the beginning of the argument—that my sensation caused my intention—and thus shows that the mental is causally impotent if property dualism is true.

If the mental is causally impotent, then mental events and properties are purely epiphenomenal. This has led some philosophers to adopt epiphenomenalism. This view holds that physical events cause mental events, but mental events have no causal relation or impact on any physical or mental events, and this appears to be the case for property dualism if Kim is right. However, many are reluctant to accept this view. It is no wonder why so many are hesitant to affirm the mental as purely epiphenomenal, since the whole point of property dualism is to make the mental something that is special and non-reducible to the physical. But on physicalism, as Kim’s argument shows, the only way to allow for mental causation is to allow for the mental to be reducible to the physical. Kim puts it this way, “Unless you bring the mind fully into the physical domain, there is no hope of explaining how the mind can participate in the physical causal processes” (emphasis added, Kim 15).

Furthermore, if the mental is epiphenomenal and can be reduced to the physical, or more specifically the neurobiology of the brain, then haven’t we lost the very thing that makes the mental important? The mental is not something over and above the physical; rather it is just a physical causal structure like everything else. Kim eloquently makes this point and writes, “If [mentality is reducible], we can save its causal powers but in the process we may lose just the thing we set out to save, namely minds" (Kim 16). If a property dualist wants to keep anything relevantly special about the mental and hold the view that the mental is causally efficacious, then they need to refute Kim’s argument.

Kim’s argument is cogent, so in order to refute Kim’s argument his opponents need to show that one of his three assumptions are wrong. The CCP is nonnegotiable for any physicalist, so many objections to the argument attack the causal principles. These objections take many forms, but I do not have the space to address all of them here. As mentioned earlier, the most intriguing objection is the autonomy solution, so this is where I will focus. According to the autonomy solution to the PMC, the idea that mental properties must do the work of their physical realizers or face exclusion is wrong. Rather, mental properties can have causal relevance in their own way by causing mental effects or, as some suggest, behavioral effects. For example, whenever we use psychological explanations about the mental, we rarely say the physical realizers caused the behavior or some mental

effect. The explanation of the causes can be made independently of the physical and can actually add to our understanding, which is why the autonomy solution is sometimes called the dual explanandum strategy. The mental need not be causally excluded from the effect, since the mental and the physical are causally relevant to different aspects of the effect (Heil and Robb 2018).

Thus, there are two ways to explain some event. We can explain it in terms of the mental, in an intentional/rational way, or we can explain the event physically/ mechanically. Those who support the autonomy solution suggest that when we explain some event, both explanations are required since the mental and physical are causally relevant to different effects. The philosopher Donald Davidson has given an example of where both explanations are necessary (Silcox 2020): Consider two climbers, one in front of the other, that are scaling a mountain along a rope. In this unfortunate scenario, one of the climbers accidentally causes the death of the other. This accident happened because the climber in front thought that he would be safer if he loosened the rope, and this thought simply unnerved him, causing him to unintentionally loosen the rope, thus killing his friend. Now, Davidson points out, consider the same situation where the climbers are scaling the mountain, but this time one of the climbers dies because his friend intentionally loosens the rope.

In both cases, the climber who killed his friend wanted to loosen the rope. Moreover, these examples look identical to any observer watching, but there is a clear difference in the latter case where the surviving climber intentionally loosened the rope to kill his friend rather than unintentionally. If we were to explain things in a purely physical or mechanical way, then these two events would be the same. But if we add the intentional explanation, then we can differentiate between the two events and see them as they really are. This is exactly what is expected from the autonomy solution. According to the autonomy solution, the climber who loosened the rope had a physical cause that caused him to loosen the rope, and this explains his physical behavior and effects. But there are also the mental causes that were relevant to whether or not the action was intentional or unintentional, and in both cases the climber who loosened the rope, wanted to but only in the second did he intend to.

Thus, neither the physical nor the mental need to be excluded as causes or explanations because they refer to different parts of the event. Applying this reasoning to the example of my desire to get coffee this morning, we can say that my sensation of tiredness caused my desire or intention to get coffee and also caused me to believe that I desired coffee, and my behavioral actions of actually going to get the coffee were caused by physical causes. Some who support the autonomy solution advocate a stronger version that claims that mental events can

cause certain behavioral effects, but these advocates still hold that the mental does not cause any physical effects as to not violate the CCP (Silcox 2020).

There are two reasons why the autonomy solution to the PMC is not in fact a solution: It makes an explanatory mistake and it cannot explain mental causation without denying one of the tenants of physicalism, namely supervenience. To consider the former, I must first explain what I mean by an explanatory mistake. The supporter of the autonomy solution argues that the mental and the physical independently cause different aspects of the same event. Thus, the fact that we can explain the event in two different ways, the intentional/rational or physical/ mechanical, means that there are two different sets of causes that work independently to produce the distinct aspects. You could explain my sensation of tiredness and my desire to get coffee as reasons or causes as to why I went and got coffee from Starbucks this morning. This would be explaining the event in a rational way. It uses a reference to an agent, me, and the reasons the agent did what he did. The other way to explain the event is to say I had some physiological cause that made me to get tired and had some chemical reaction in my brain due to this cause, and these combined prompted me to go to Starbucks. This is explaining the event in a physical manner. It does not reference any of the mental events or states and makes no reference to agents in order to explain the event.

Of course, using these two explanations adds to our understanding of the event. This is how we can differentiate between the climber example given by Davidson. This is to say, they add to our epistemological understanding of the event, but using two different explanations does not necessarily add to our metaphysical understanding. Using two different explanations gives us knowledge we may have not had before by allowing us to use two different concepts to describe the event, i.e. rational concepts and physical concepts. But the fact that we can understand or explain an event in two different ways, does not mean that each of the explanations are metaphysically relevant to the causes that brought about the event. This is a common response that many physicalists make to refute dualist arguments.

For example, in her paper, “In Defense of the Phenomenal Concept Strategy”, Katlin Balog uses this idea to explain why Mary, the character in Frank Jackson’s famous “Knowledge Argument”, learns something new, but this fact does not imply the existence of some new metaphysical thing. To understand the “Knowledge Argument”, consider a woman, named Mary, that learns everything there is to know

about the physical world all while locked in a black and white room. Consequently, Mary knows everything about colors and color vision, but she has not seen any colors. Upon leaving the room, Mary sees a ripe tomato that manifests the color red vividly. Many of us, myself included, have a strong intuition that Mary learned something new when leaving the black and white room, namely “what it is like” to see red. Jackson concludes that because Mary learns something new when leaving the room, even though she knew all the physical facts, some things that are not physical must exist (Jackson 291). However, if we use Balog’s reasoning, then Mary learned something new not because things that are not physical exist, but because she was able to use new concepts, namely phenomenal concepts, to understand red in a different way (Balog 2012).

Nothing about this necessitates that there must be something new metaphysically. Rather, it just necessitates something new epistemologically. Similarly, just because we can explain some event or effect in a rational way as opposed to a physical way, and gain new knowledge about it, does not entail that these rational or intentional ‘causes’ have any metaphysical significance. These ‘causes’ have epistemological significance as we learn why a person did what they did, but we merely have employed new concepts to understand the event in a different way. Thus, the fact that we can explain things two different ways does not entail that both explanations have causal significance. And given physicalism, it is very likely that the only metaphysically significant causes are physical.

Other than the explanatory mistake, the other reason the autonomy solution fails is that it cannot explain the efficacy of the mental without denying supervenience. Suppose that these explanations of the rational or intentional do actually imply the mental has some sort of causal relevance to either itself or some behavioral effects. Does this solve the problem of causal exclusion for the property dualist and explain how mental causation works? Not at all. For if the supporter of the autonomy solution accepts Kim’s assumption of the CCP and accepts the supervenience of the mental on the physical (both seem to be necessary for any physicalist theory of mind), then they must claim that the mental is causally inefficacious. To say it is causally relevant you would have to explain how it is causally relevant.

On the weaker version of the autonomy solution, you cannot do this without leading to absurdities. Consider what is being claimed by the weaker version of the autonomy solution, that some mental event causes another mental event or effect without causing any behavioral or physical effects. As some philosophers have pointed out, such as Daniel Dennett, this would mean that someone’s desire to get coffee causes them to believe they are about to get coffee, but is not causally

relevant to their action of getting the coffee (Moore 2020). This seems incredible. How does my desire cause my belief that I am about to go to Starbucks, but have nothing to do with my actual action of going to Starbucks? If you think the mental has any causal powers, then you ought to claim that the mental is causally relevant to its related action or behavior if its powers are to do anything meaningful. But if someone were to endorse such a view of the autonomy solution, then they would have to deny this.

Not only are the conclusions from this reasoning implausible, but concluding something such as mental to mental causation without reference to the physical is to deny that the mental supervenes on the physical. As Karen Bennett writes regarding this strategy, “Why does the dual explanandum strategy require denying that the mental supervenes on the physical? Because the supervenience claim extends the exclusion argument to problematize mental to mental causation as well” (Bennett 326). According to supervenience, any mental property or effect must have a physical supervenient base that generates it. This physical supervenient base must also have a physical cause that generated it based on the CCP. This excludes the mental from having any causal potency.

There is the stronger version of the autonomy solution that claims that not only can the mental cause mental effects or events but also behavioral effects, so it avoids the absurd conclusions that the weaker version must make. But this does nothing to help the supporters of the autonomy solution’s case with regards to supervenience. For the same argument above can be applied to the behavioral effects, i.e. they must have some physical cause and the mental event that ‘caused’ the behavioral effect must have also had a physical cause. The only way to get around this is denying supervenience and run afoul of the CCP. Again, these are not options for the physicalist. Therefore, if someone supports the autonomy solution, then it seems they have to affirm epiphenomenalism since the mental is causally impotent.

Consequently, the autonomy solution does not solve the PMC for the property dualist. The supporter of the autonomy solution’s attempt to claim that mental properties are distinct from the physical, and are therefore not reducible to the physical, leads to the conclusion that the mental is causally inefficacious. Moreover, they claim that the mental and the physical enjoy causal relevance independent of each other and explain different aspects of an event. However, just because an event

can be explained in two different ways by referencing the different independent ‘causes’, does not necessitate that two different causes exist. Instead, we are able to do this because of our ability to utilize different concepts. Further, given physicalism, I noted that the attempt to explain mental to mental or mental to behavioral/physical causation with the autonomy solution fails because in order to keep the causal efficacy it must deny supervenience. Hence, all the causal work is done by the physical, and explaining the rational ‘causes’ of an event simply gives us new knowledge and a different understanding of the event.

I find that many of the other objections to Kim’s argument have similar difficulties that the autonomy solution has, and thus it seems that in order for there to be mental causation, on physicalism, the mental must be reducible to the physical. But as we saw earlier, for a physicalist to affirm reductionism is to give up the very idea of what we mean by the mental. As Kim notes, “If we save minds’ causal efficacy only by giving up minds, that seems like no way to save it” (Kim 13). The problem of mental causation remains unsolved, but at least we can say the autonomy solution, or anything like it, fails as an adequate response.

Balog, Katalin. "In Defense of the Phenomenal Concept Strategy." Philosophy and Phenomenological Research

(2012): 1-21.

Bennett, Karen. "Mental Causation." Philosophy Compass (2007): 316-337.

Heil, John and David Robb. "Mental Causation." The Stanford Encyclopedia of Philosophy (2018) https://

plato.stanford.edu/entries/mental-causation/#ProIIIExc.

Jackson, Frank. "What Mary Didn't Know." The Journal of Philosophy (1986): 291-295.

Kim, Jaegwon. "How Can My Mind Move My Limbs? Mental Causation from Descartes to Contemporary

Physicalism." Philosophic Exchange (2000): 1-13.

Moore, Dwayne. "Mind and the Causal Exclusion Problem." Internet Encyclopedia of Philosophy (2020)

https://www.iep.utm.edu/causal-e/.

Silcox, Mark. "Donald Davidson: Anomalous Monism." Internet Encyclopedia of Philosophy (2020) https://

www.iep.utm.edu/anom-mon/.