Any theory of mind needs to explain mental causation. Kim’s (upward) exclusion argument concludes that non-reductive physicalism cannot meet this challenge. One classic reply is that mental properties capture the causally relevant level of generality, because they are insensitive to physical realization. However, this reply suggests downward exclusion (if mental properties are causally efficacious, their physical realizers are causally impotent), contrary to physicalism’s assumption of closure. This paper shows how non-reductive physicalists can solve this problem by introducing a contrastive account of causation with non-exhaustive contrasts. That view has independent justification, because it is also needed to solve other puzzles. On this theory, both a mental property and its physical realizer can cause the same physical effect without lapsing into any problematic overdetermination when they cause that effect in contrast with distinct foils. This contrastive solution has advantages over previous accounts of mental causation and is defended against objections.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
Another serious problem is how mental content can be causally efficacious if mental content depends on history, as in teleosemantic theories. That other problem will be addressed in a companion paper.
Kim sometimes refers to “sufficient cause” instead of just “cause.” I will discuss this qualification below. As formulated, this premise allows that some events or properties do not have any cause, as in some interpretations of quantum mechanics, but I will ignore that complication here. Notice also that physicalism does not strictly entail closure of the physical, since physicalists could be eliminativists about causation. Nonetheless, I will follow Kim in discussing only physicalists who also accept closure of the physical, since other versions of physicalism are less popular and plausible.
My main argument here is flexible enough to allow other causal relata, such as states of affairs, property instantiations, and so on, although it would take another paper to show how this all works.
Different notions of reduction can be distinguished by separate kinds of distinctness (Bernstein 2016, p. 21). The relevant kind here is numerical distinctness, so that we can count two properties (mental and physical) that themselves differ in some of their meta-properties.
Bernstein (2016, p. 20) adds that two causes overdetermine a result only if (iv) each is sufficient to cause the effect in the way it occurs. Imagine that one executioner shoots the prisoner in the stomach, which would cause the prisoner to die slowly, but another executioner simultaneously shoots the prisoner in the head, which causes him to die quickly. Only the executioner who shot the prisoner in the head causes the prisoner to die in the way he dies. The shot in the stomach is sufficient to cause death, but not in the way that it actually occurs. Hence, this case does not involve overdetermination of the relevant kind. This additional requirement does not affect my argument, so I will ignore it for simplicity.
Mental-to-physical impotence is only one part of epiphenomenalism, which also claims that physical properties can cause mental effects (as when hitting someone causes pain). Notice also that mental-to-physical impotence does not yet exclude mental-to-mental causation. Non-reductive physicalists also have trouble making room for mental-to-mental causation, but I will focus here on mental-to-physical causation, which is also known as downward causation.
Many philosophers deny that counterfactuals are sufficient for causation, but they still do not deny that these counterfactuals are necessary for causation.
You would have given me ten dimes if you had not given me four quarters on the assumptions that you had ten dimes (as well as four quarters) and that we are friends so you want to help me get coffee. On those assumptions, the closest possible worlds where you do not give me four quarters are worlds where you do give me ten dimes.
For additional support from studies of vision, see Aizawa and Gillett (2013).
Another example comes from electrode grids in the brain that enable paralyzed patients to control prosthetics. See the discussion of Anderson’s pioneering work in Woodward (2008). Many examples like these show both that mental states are multiply realizable and also that the causation of bodily movements by such mental states is not sensitive to which particular physical neural state realizes the mental state. Of course, it is also possible that other mental states are not multiply realizable, but my argument claims only that some mental properties have physical effects, not that all do.
What matters here is sensitivity to realizers instead of sensitivity to circumstances, as in Woodward (2006). This point about realizer-sensitivity is shared by Yablo (1992), Woodward (2008), Menzies (2008), and List and Menzies (2009), but these predecessors differ in detail. In particular, Yablo explains mental properties as determinables with physical realizations as their determinates. Menzies (2008) argues forcefully that the relation between determinables and their determinates is not an accurate model of the relation of mind to brain. These internal disputes do not affect my main points, so I will ignore them here.
For completeness, what if p* would occur if m were realized by p1 or p2 but not by p3? Then p* is not sensitive to whether m is realized by p1 or p2 but is sensitive to whether m is realized by p1 or p3 and also to whether m is realized by p2 or p3.
Except in cases of overdetermination and pre-emption, this counterfactual is necessary for causation even on generative, productive, or energy-transfer theories of causation. Those views require more than counterfactuals for causation, but they usually still admit that this counterfactual has to hold in causal relations.
The point that (C1) is not true in the case of insensitivity depends on a replacement interpretation of (C1), according to which (C1) is true if and only if p* would not have occurred if p1 had been replaced by p2 or p3. In contrast, some philosophers argue that (C1) is true on a deletion interpretation of (C1), on which (C1) is true if and only if p* would not have occurred if p1 had been deleted and not replaced by anything like p1: “you simply snip it away as though you had a metaphysical hole puncher” (Bennett 2003). However, metaphysical hole punching would violate physical laws and lack the continuity that we find in the actual world. Thus, the world where p1 is replaced differs from the actual world in fewer and less important ways than they both differ from the world where p1 is simply deleted. That distance means that we should not interpret (C1) in terms of deletion if (C1), like other counterfactuals, are assessed in terms of the closest possible worlds, as is standard.
An anonymous referee objected that lack of causal dependence does not prove lack of causation, according to Lewis (1973). However, Lewis still holds that c is a cause of e only if there exists a causal chain leading from c to e, such that, for some intermediary event d, d causally depends on c and e causally depends on d. No such intermediary or causal chain occurs between p1 and p* in the case under discussion, because the counterfactuals required for causal dependence will not hold for p* (though they might hold for its particular realization p*’, as I will show below).
Defenders of the sensitivity reply might propose the disjunctive event (p1 or p2 or p3) as a cause of p* at the same time as p*. However, it is not clear that disjunctive events are individual events, how to specify when they occur (since their disjuncts might occur at different times), or how to explain why these physical disjuncts instead of others are grouped together without referring to the mental property that they realize. These issues will be discussed further in Sect. 7.2 below.
On some accounts of counterfactuals, (S) will be trivially true whenever both X and Y actually obtain, even without any connection between X and Y. I see this as a serious problem for such accounts of counterfactuals.
Similarly, Raatikainen contrasts having a belief with not having that belief (2010, p. 356) and having a brain state with not having that brain state (2010, p. 357). Even when he introduces a non-exhaustive contrast between “goes to the refrigerator” and “goes to the grocery”, he adds, “Let us suppose, for simplicity, that these cases exhaust all possible cases” (2010, p. 356). My point is that this simplification obscures the power of the contrastive approach.
Schaffer (2005) adapts this example from predecessors.
It is a kind, because several distinct ball flights would go beyond the wall and count as p*. In this way, p* is a property of its instances, the property of going beyond the wall.
Notice that p*’ is a token rather than a type (though p*’ could be seen as a subtype instead of a particular token). Nonetheless, this token has distinct properties of being beyond the wall and being in the first tier of seats. The question here is which of these physical properties in contrast with which other properties is caused by m or by p1.
The set of acts of hitting a home run into the first tier of the stands (p*’) is properly included in the set of acts of hitting a home run (p*), and hitting a home run (p*) is a determinable of which hitting a home run into the first tier of the stands (p*’) is one determination. However, these relations of inclusion and determination are distinct from the relation that holds between mental events and their physical realizers, as Menzies (2008) argues against Yablo (1992). Thus, we should not think of p*’ as a realizer of p*.
If pressed (thanks, Alex), I would individuate events by their spatio-temporal locations. However, my point does not depend on this specific way of individuating events.
Philosophers sometimes refer to a “dual explanandum strategy.” My view differs from that approach insofar as I focus on causation rather than explanation. Still, I do refer to dual properties as well as dual contrast classes, though only one event as the effect. To that extent, my view may be seen as a partially dual effect strategy.
Contrastivism about causation might seem to be only a way to more precisely characterize what was called proportionality. That would be fine with me if it were true. However, a crucial difference is that the requirement that a cause be proportional to its effect suggests that a single physical effect cannot be both caused by a mental property and also caused by a physical property, since each of those causes will be proportional to a different effect. That was what implied downward exclusion and why the sensitivity reply was not enough to solve the problem of exclusion. Contrastivism about causation allows the possibility that a single effect has both mental and physical causes even though those different causes cause different properties of that single effect in contrast with different alternatives.
Like Woodward (2008, pp. 239–40), I admit that an “aggregate profile” of firing rates, if properly specified, could fit the requisite counterfactuals and, thus, could be seen as causal. My point is only that we need mental terms or properties in order to specify which firing rates lie within that profile, motivate including those firing rates but not others, extend the profile to new cases, or complete the list of firing rates inside that profile. That is the sense in which a purely physical description of the aggregate profile will remain arbitrary, unmotivated, and incomplete.
Those who want to leave metaphysics and turn to the distinct linguistic issue of when it is proper (or true?) to say that c causes e will also need to specify rules about which (if any) contrasts are relevant in which conversational contexts. I do not address such linguistic issues here, because I want to keep metaphysics and linguistics separate, and also because I would adopt a Pyrrhonian approach here parallel to my Pyrrhonism in other areas (Sinnott-Armstrong 2006).
Aizawa, K., & Gillett, C. (2013). Levels, individual variation, and massive multiple realization in neurobiology. In J. Bickle (Ed.), The Oxford handbook of philosophy and neuroscience (pp. 539–581). New York: Oxford University Press.
Bennett, K. (2003). Why the exclusion problem seems intractable, and how, just maybe, to tract it. Noûs,37(3), 471–497.
Bernstein, S. (2016). Overdetermination underdetermined. Erkenntnis,81, 17–40.
Davidson, D. (2001a). Causal relations. In Essays on actions and events. New York: Oxford University Press
Davidson, D. (2001b). Mental events. In Essays on actions and events. New York: Oxford University Press
Fitelson, B. (2013). Contrastive bayesianism. In M. Blaauw (Ed.), Contrastivism in philosophy (pp. 64–87). New York: Routledge.
Fried, I., Mukamel, R., & Kreiman, G. (2011). Internally generated preactivation of single neurons in human medial frontal cortex predicts volition. Neuron,69, 548–562.
Hall, N. (2004). Two concepts of causation. In J. Collins, N. Hall, & L. Paul (Eds.), Causation and counterfactuals (pp. 225–276). Cambridge: MIT Press.
Kim, J. (2005). Physicalism, or something near enough. Princeton: Princeton University Press.
Kim, J. (2007). Causation and mental causation. In B. McLaughlin & J. Cohen (Eds.), Contemporary debates in philosophy of mind (pp. 227–242). Oxford: Blackwell.
Lewis, D. (1973). Causation. Journal of Philosophy,70, 556–567.
List, C., & Menzies, P. (2009). Non-reductive physicalism and the limits of the exclusion principle. Journal of Philosophy,106(9), 475–502.
Loewer, B. (2007). Mental causation, or something near enough. In B. McLaughlin & J. Cohen (Eds.), Contemporary debates in philosophy of mind (pp. 243–264). Oxford: Blackwell.
Maslen, C., Horgan, T., & Daly, H. (2009). Mental causation. In H. Beebee, C. Hitchcock, & P. Menzies (Eds.), The Oxford handbook of causation (pp. 523–553). New York: Oxford University Press.
Menzies, P. (2008). The exclusion problem, the determination relation, and contrastive causation. In J. Hohwy & J. Kallestrup (Eds.), Being reduced: New essays on reduction, explanation, and causation (pp. 196–217). New York: Oxford University Press.
Raatikainen, P. (2010). Causation, exclusion, and the special sciences. Erkenntnis,73, 349–363.
Schaffer, J. (2005). Contrastive causation. Philosophical Review,114(3), 297–328.
Schaffer, J. (2010). Contrastive causation in the law. Legal Theory,16, 259–297.
Shapiro, L. (2000). Multiple realizations. Journal of Philosophy,97(12), 635–654.
Shenoy, K. V., Sahani, M., & Churchland, M. M. (2013). Cortical control of arm movements: A dynamical systems perspective. Annual Review of Neuroscience,36, 337–359.
Sinnott-Armstrong, W. (2006). Moral skepticisms. New York: Oxford University Press.
Woodward, J. (2006). Sensitive and insensitive causation. Philosophical Review,115, 1–50.
Woodward, J. (2008). Mental causation and neural mechanisms. In J. Hohwy & J. Kallestrup (Eds.), Being reduced: New essays on reduction, explanation, and causation (pp. 218–262). New York: Oxford University Press.
Yablo, S. (1992). Mental causation. Philosophical Review,101(2), 245–280.
Thanks to Sara Bernstein, Trey Boone, Johannes Himmelreich, Kevin Hoover, Jennifer Jhun, Christian List, Jeff McMahan, David Robb, Alex Rosenberg, Justin Snedegar, anonymous referees, and audiences at Australian National University, Duke University, Hebrew University in Jerusalem, and the Universities of California at Santa Barbara, Oxford, Reading, St. Andrews, and Sigtuna for helpful comments on drafts and critical discussion of these issues. I am also grateful to the John Templeton Foundation for financial support while writing this paper, but, of course, the opinions expressed in this publication are those of the author and do not necessarily reflect the views of the John Templeton Foundation.
Funding was provided by John Templeton Foundation (Grant No. 48365)
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Sinnott-Armstrong, W. Contrastive mental causation. Synthese (2019). https://doi.org/10.1007/s11229-019-02506-0
- Multiple realizability