Power for the Mental as Such
David Robb
Davidson College
 
I. Introduction

The identity theory promises an elegant solution to the exclusion problem (Kim 1998, 2005; Robb 2012). Mental properties face no threat of exclusion from, or preemption by, their physical "base" properties, for every mental property is its physical base. Mental causation turns out to be just a kind of physical causation, perhaps empirically inscrutable, but philosophically no more problematic than any other sort of physical causation.

But in spite of its attractions, the identity theory faces a number of objections. These fall into two broad categories. In the first category are "Leibniz's Law" arguments alleging that mental properties have some feature that no physical property does (or could) have. Candidates include privacy, subjectivity, qualitative feel, normativity, original intentionality, externality, and multiple realizability. As important as such objections are, I mention them only to set them aside, for in this paper I'm concerned with an objection in the second category. Here there are arguments challenging, not so much the truth of the identity theory, but its ability to perform some crucial philosophical task. It may be said, for example, that identities are not appropriately explanatory, so that the identity theory will fail to explain psychophysical correlations, thus leaving the mind-body problem unresolved (Kim 2005, ch. 5). And there is the related objection--attributed to Max Black by Smart (1959)--that even if sensations are brain processes, qualitative feel will remain unreduced, for it will just reappear as a property of these brain processes. The objection I confront in this paper belongs in the same broad category as these, though here the task for the identity theory is resolving the problem of mental causation. The worry is that even if the identity theory is true, it makes no significant progress on the exclusion problem, but merely relocates it.

In what follows, I'll argue that with the backing of a powers ontology, the identity theory can in fact deliver on its promise to solve the exclusion problem. I'll start by sketching the central worry--the as-such objection--and arguing that a version of it also threatens the identity theory's most prominent contemporary rival, non-reductive physicalism. I'll then present the basics of a powers ontology and argue that within its constraints, the as-such worry for the identity theory dissolves. It will not be as clear, however, that non-reductive physicalists can help themselves to the same solution. I'll end by considering an objection to this powers-based solution.

II. The as-such objection

The identity theory, as I'll understand it, says that every mental property is physical. This entails that every mental cause is physical. It will be useful to begin with this weaker thesis, for it is here that in recent history, the as-such objection has been most prominent.

Davidson (1980) says mental causes--events, on his view--are physical. This permits mental causation to be a kind of physical causation, apparently blocking exclusionary problems. But some critics (e.g., Honderich 1982) objected that even if Davidson thereby secures the causal efficacy of mental events, he has not secured their efficacy qua mental. The mental is part of the causal order, but not, it seems, as such. There are at least two ways to make the point. First, by Davidson's own lights, psychophysical causal relations are grounded only in physical laws, for only such laws are strict. But then only the physical nature of a mental event, not its mental nature, is engaged in mental causation. Second, there are reasons to think that the physical nature of a mental event suffices for its behavioral effect, thereby preempting any would-be efficacy of its mental nature, whether or not there are psychophysical laws (strict or otherwise) in play.[1]

Formulating this objection requires an ontological ascent to natures, or properties, of causes: to ask whether the mental as such causes behavior is to ask about the causal efficacy of mental properties--or at least this is how Davidson's critics have usually framed the problem, and in this I will follow them. The worry, then, is that even if mental events are physical, only their physical properties are efficacious in producing behavior. That is, when a mental event causes behavior, it is not as mental that it does so: the mental as such does no causal work.

One response to this problem--not to be attributed to Davidson--is to say about mental properties what Davidson said about mental events: mental properties are physical. In particular, mental properties are just those physical properties that are, Davidson's critics allow, causally efficacious with respect to behavior. This is what I'm here calling the identity theory. But now the as-such worry seems to arise again. For we can ask the same question about mental properties that we did about mental causes: granting that mental properties are, on the identity theory, causally efficacious, are they efficacious qua mental? After all, the reasoning goes, any mental property that's also physical has a mental nature and a physical nature. And the same arguments used against Davidson apply here, for it looks as if only the physical nature of a mental property is engaged in mental causation. Psychophysical property identity seems not to help at all with the as-such objection, and indeed makes the ascent to properties look idle, a spinning of wheels.

Something has gone wrong, but what? I suspect that most philosophers writing in this area would say the mistake here is not in the ascent to properties, but in the thesis of psychophysical property identity. Often such philosophers, like Davidson, say that mental causes are physical. But they reject the identity theory in favor of non-reductive physicalism: mental properties cannot be reduced to physical properties, yet they nevertheless bear some close relation to them, a relation intimate enough to warrant the physicalist label.[2] The usual strategy is then to defend compatibilism, here the view that the causal efficacy of a mental property is compatible with (not preempted by) that of its distinct, physical base property. And while there are challenges for non-reductive physicalism and the compatibilist project, the as-such worry does not seem to be one of them. Since non-reductive physicalists reject the identity theory, mental properties on their view are only mental,[3] and so there is no dual nature in mental properties to give the as-such objection the traction it needs.

Nevertheless, it seems to me that non-reductive physicalism does in fact confront a version of the as-such problem. Non-reductive physicalists say that mental properties, while not physical, bear some intimate relation to their physical base properties; call it realization just to have a label. Even granting that the compatibilist project can secure the causal efficacy of mental properties, why think such properties are causally efficacious qua mental? Why not think instead that they enter the causal order only qua physically realized?

To make the point more concrete, consider a prominent version of non-reductive physicalism, the "subset" account defended in Shoemaker 2001, 2007 and Wilson 2009, 201.[4] On this account, the (forward-looking) causal powers associated with a mental property on an occasion are included among the causal powers associated with its physical base property. (Since the inclusion is proper, the mental property is distinct from its base property, so there is no reduction, as desired.) Here then is a quick route to compatibilism: When a causal power of a mental property is manifested in behavior, this power will be among those associated with its physical base property. In that case, the causal efficacy of the physical property won't exclude the mental property's efficacy, but will rather include it, since the two properties share the (token) manifested power. For the physical property to be efficacious in this case just is for the mental property to be efficacious.

Suppose that the subset account successfully secures the causal efficacy of mental properties. Still, has it secured the efficacy of mental properties as such? For all that's been said so far, it could turn out that the mental property is efficacious, not in virtue of being mental, but merely in virtue of sharing the manifested power with its physical base property. This is, if not the same problem I've raised for the identity theory, a very similar one, and equally worrisome.

Before I present a solution--one based on a powers ontology--a few remarks are in order about how I conceive of as-such questions. First, there is a reading of these questions on which they take on an explanatory-epistemic character. To ask whether a cause, or a causally efficacious property, is efficacious qua F is to ask whether conceiving of or describing it as F is appropriate or useful relative to, say, some explanatory scheme (Burge 1993), context of causal inquiry (Horgan 2001), or contrast class (Sinnott-Armstrong ms). I do not deny the legitimacy and importance of these epistemic as-such questions. But here I'm looking at the metaphysics of mental causation, and so I am (temporarily) setting aside epistemic concerns. When I ask whether the mental as such is causally efficacious, I’m wondering how the world is independently of any particular epistemic aims. In the memorable phrase of British emergentist C. Lloyd Morgan, there is something that is the "go of the world", that drives it forward (Seager ms). I want to understand how mental properties as such can provide the go of the world, and in particular, how they can drive human bodies forward. How this is related to our explanatory practices and the like is an important question, but it is not my focus here, which is on the metaphysics alone.[5]

Second, it may be thought that as-such questions, while appropriately asked of causes, are not appropriate, or receive trivial answers, when asked of properties. On one tradition, properties are individuated as finely as the predicates (or concepts) that pick them out. To ask, then, whether a mental property is this or that qua mental is odd, even nonsensical, as it suggests a mental property is anything other than what is revealed by its individuating mental predicate. But it will be clear below--if it is not already--that this is not how I am thinking of properties. Indeed, the identity theory would be a non-starter if properties were so finely individuated, for there is no hope of a semantic reduction of mental predicates to physical predicates. Since the identity theory is a substantive, live thesis, properties are not merely the "shadows cast by predicates".[6]

Third, in other work I have consistently rejected the legitimacy of (metaphysical) as-such questions when they are asked of properties (Robb 1997, 2001, 2012; Heil and Robb 2003). We can ask if an object or event causes something qua this or that, and an affirmative answer will assert the causal efficacy of some property. But I've held that we cannot then ask whether the property itself is efficacious qua this or that. I've rejected such a question as presupposing a mistaken ontology of properties, in particular, an ontology on which properties themselves have (second-order) properties. As will become clear below, I still think this ontology is mistaken. Indeed, that point will be central in what follows. And so I still think that if as-such questions, when applied to properties, presuppose second-order properties, these questions are illegitimate. Nevertheless, the present paper is motivated by the thought that these questions can get some traction even within (what I take to be) the correct ontology of properties--get some traction and, I hope to show, receive some affirmative answers.[7]
 
III. Sketch of a powers ontology
 
I turn now to a powers ontology, for it can, I believe, deliver what the identity theorist needs in response to the as-such objection.

I take a powers ontology to be a version of a property ontology, according to which properties, and only properties, are ontologically basic, the "elements of being" (Williams 1966), the ultimate truthmakers. For example, bundled properties are truthmakers for truths about objects, so a property ontology can dispense with substance (Simons 1998; Robb 2005). Resembling properties are truthmakers for truths about types and kinds, so a property ontology can dispense with universals. Properties, then, I take to be particulars, also known as tropes or modes (Campbell 1990; Heil 2003). And internal relations among properties are truthmakers for relational predications, so a property ontology can dispense with irreducible relations (Mulligan 1998). A powers ontology is a property ontology plus the thesis that every property is a power. It's what a property ontology becomes when conjoined with the Eleatic Principle (Oddie 1982) that power is the mark of being.

On this ontology, what becomes of qualities, those properties that "color" our world, that "fill space"? Some philosophers have argued that a world without qualities is empty.[8] I find these arguments persuasive, and thus look for an ontology to accommodate, not eliminate, qualities. Accommodation in this case means identification: every quality is a power. I’ll return to qualities--in particular, mental qualities--in a later section. For the moment I set them to one side.

Since every property is a power, every property will have associated with it a causal profile specifying what manifestations result from partnering the property with other properties, themselves powers.[9] Consider, for example, the shape of my house key. This property is the power to cut open cardboard boxes, to open locks of a certain sort, to cause a certain kind of impression when the key is pressed into soft clay, to cause a distinctive visual experience in suitably placed perceivers, and so on. Here I’ve left out most of the power partners for each manifestation, but this at least starts to give a sense of the profile’s "multi-track" complexity (Heil 2003, pp. 198-9; Williams 2010). But in virtue of what does a property satisfy its profile? One could characterize the relation between property and profile as that of role-filler to role, but this merely postpones the metaphysical question, for what is it about the property that makes it a role-filler for this role?

It seems to me there are two answers a powers ontology cannot permit. One says the property satisfies its causal profile in virtue of the laws of nature, here construed as external relations among the property and its various power partners. This is the Humean answer, for example, though some non-Humeans, such as Armstrong (1983), can say this as well. But whatever the attractions of this picture, it entails that properties in themselves are inert, not powers. Moreover, it means that properties alone are no longer the ultimate truthmakers, for these external relations are needed as well. A powers ontologist can't accept either of these consequences. A second answer says a property satisfies its profile in virtue of having multiple, intrinsic, second-order properties,[10] each of which corresponds to a unique part of the profile. So, for example, my key's shape is the power to cut cardboard boxes, and this is in virtue of the shape's having a second-order property: call it C. Now if a property has second-order properties, the powers ontology requires that these too are powers, so C is a power. This immediately raises a red flag, for a regress threatens if a power's having a causal profile depends on its having a property with a causal profile. And there's another problem as well: what power is C exactly? Presumably it's the power to cut cardboard boxes, and just this power. But I don't think it's intelligible that C is the power to cut cardboard boxes but not the power to, say, make a certain sort of impression in soft clay. How could there be a property, second-order or otherwise, that's the power to cut cardboard boxes (when partnered with rigidity, etc.) but not the power to make an impression of a certain shape in soft clay (with those same partners)? It starts to look as if C really must have the causal profile of the shape of my key. But this is to collapse C, which was supposed to be a second-order property, back into the first-order property that is its bearer.

A powers ontologist should not find this result surprising. If powers are at the ground level, it looks like a mistake to explain their powerful natures in terms of further powers. On the ontology, a property doesn’t have powers to do various things, it is the power to do those things (Marmodoro 2010). Back to the question, then: What is it about a property that makes it fill the role specified by its causal profile? Nothing about the property, but just the property itself. The property, that is, is itself the truthmaker for its causal profile. The causal profile is complex, but its truthmaker is simple.
 
IV. Power for the mental as such
 
Return then to the as-such problem confronting the identity theorist, now in the context of the powers ontology just sketched.

I assume for the moment, along broadly functionalist lines (Levin 2010), that what makes a mental property mental is its causal profile. Now if the identity theory is true, every mental property is physical, so the mental property will have a physical causal profile as well, most of which will include far more than what's required for the property to be mental. If these mental and physical components of the profile correspond to distinct natures--that is, distinct, intrinsic, second-order properties--of a mental property, the as-such worry is still with us. But the powers ontology blocks this: the mental and physical elements of the property's causal profile find their ontological correlate, or truthmaker, in just the property itself. Put another way, a mental property is its mental nature and its physical nature. But then for a mental property's physical nature to be causally engaged just is for its mental nature to be engaged, delivering a quick solution to the as-such problem.

The solution no doubt seems too quick, but I don't claim that it's easy, for it depends on the controversial details of the powers ontology sketched above, and there is at least one serious objection to the view, which I will consider in the last section. But setting these worries aside for the moment, I believe I can now pinpoint where the as-such objection goes wrong. After the ascent to properties, the critic assumes that if mental properties are physical, they aren't wholly mental, but rather partly mental, partly physical, thus raising as-such worries. Wanting a mental property that's wholly mental, the critic suggests non-reductive physicalism, on which mental properties are only (and so wholly) mental. As I argued earlier, however, it's not clear that moving to non-reductive physicalism helps with the as-such problem: even if mental properties are only mental, they may still not be efficacious qua mental. The lack of progress here suggests the critic was too quick in rejecting the identity theory. The mistake, in particular, was in thinking that a property that's both mental and physical must not be wholly mental. But a powers ontology permits it to be wholly mental and wholly physical. It's possible, that is, to be wholly mental without being only mental.

Before looking at some complications, it's worth pausing for a moment to consider whether non-reductive physicalists can help themselves to a similar solution. It seems to me they cannot, for they face a dilemma: either the causal powers associated with a mental property are included among those associated with its physical base property, or they are not. Suppose first that they are not. In that case, I don’t see how non-reductive physicalists can avoid causal competition between the two properties, thereby blocking a compatibilist solution to the exclusion problem (Wilson 2011). Their view in that case would appear to be a version of emergentism. So suppose instead that the powers of the mental property are included among those of its physical base property, as on the Shoemaker-Wilson subset view. If, in response to the as-such problem, non-reductive physicalists were then to help themselves to a powers ontology, they could not avoid collapsing the mental property into its base property, resulting in the identity theory. On a powers ontology, a property is its "associated" powers. If the shared power just is the mental property and also just is the physical property, then the mental property is the physical property. Here then, via the powers ontology, is another route to the frequent charge that non-reductive physicalism is unstable (Kim 1993, ch. 17; Crane 2001; O’Connor and Wong 2005): it must abandon either physicalism for emergentism or non-reductivism for the identity theory.

V. Mental qualities

I return now to the identity theory and the powers-based solution to the as-such problem. In the next section I'll consider an objection, but I first want to look briefly at mental qualities.

I take qualities to be properties, so on a powers ontology, all qualities are powers,[11] and this includes mental qualities, or qualia (Tye 2007). On the identity theory, any given quale will be some physical property that is unproblematically efficacious with respect to behavior. But when a mental quality is efficacious, is it efficacious qua mental? Suppose that what makes a mental quality mental is its qualitative feel. So when a quale is efficacious, is it moving the human body as a quale, in virtue of its qualitative feel? There are two strategies for answering this question.

One strategy ascends to second-order properties and says the qualitative nature of the quale is some intrinsic property of it: call it Q. The as-such question will then concern the efficacy of the quale as Q, i.e., in virtue of having Q. But a powers ontology can't permit this. First, Q will, like all properties, have to be a power. But I argued in the last section that all of the powers a property "has" are just the property itself. Second, a less direct argument for the same conclusion asks how Q can account for the qualitative nature of the quale. It's hard to see how Q could do this unless Q itself had a qualitative nature. To block a regress, it seems we must not permit an even higher-order qualitative property for Q to have. Rather, Q just is its qualitative nature. But if we can say that about Q, why not say it about the original quale and be done with it? This leads to the second strategy, which is to say that the qualitative nature of a quale is just the quale itself. It won’t be surprising that this is the option I favor: a quale doesn’t have a phenomenal feel, it is a phenomenal feel. But then the as-such problem dissolves, since the causal efficacy of the quale just is its efficacy qua mental, as a quale.

VI. Too much power?

I turn finally to a worry about the solution presented here: while it delivers power for the mental as such, it appears to deliver too much power.[12]

A non-mental example will illustrate what I have in mind. Consider again my house key. Its shape is a power, including the power to open a lock of a certain sort and the power to cut open cardboard boxes (as usual, I leave out most of the relevant power partners). Now on the view defended here, each of these elements of the causal profile is made true by the property itself. But this seems to have the absurd consequence that when my key opens a lock, the shape of my key is efficacious as a box-cutter. Indeed, every element of the shape’s causal profile is "activated" when the key opens my lock. And this looks like the wrong result.

The same worry applies when we turn to the mental. On the identity theory, a mental property's causal profile will include, not just what's distinctively mental, but elements that do not mention the property's manifestations in behavior. What these non-mental elements of the profile might be is an empirical issue, but for the sake of discussion, suppose the profile of a given mental property includes its ability to dissolve a certain chemical when suitable partners are present. So when this mental property is causally efficacious with respect to behavior, its mental causal profile is "activated", but so is the chemical part of its profile, each having its ontological correlate in the efficacious property itself. Qua mental, the property produces behavior, but also qua chemical dissolver. And again, this looks wrong.

I see no option here but to accept these results. When a property, mental or otherwise, is efficacious, every element of its causal profile is, in the relevant sense, active. This means that in the key example,

(S) The shape of my key, when it's efficacious in opening my lock, is efficacious as a box-cutter.

But while I insist (S) is true, I think I can explain why it seems false, for it's tempting to infer the following from it:

(i) On this occasion, the power partners represented in the box-cutting part of the shape's causal profile are present.

(ii) On this occasion, the shape was involved in cutting a box.

(iii) Any property similar enough to this shape to cut a box would also be able to open my lock.

Each of (i)-(iii) is false in the example described. (iii) deserve some comment. Properties stand in internal similarity relations to one another, and these relations fall along various dimensions and come in degrees. The color sphere provides an example, where colors are arranged along three dimensions of similarity: hue, saturation, and brightness. What determines the color sphere are the properties themselves: a property does not "have", as second-order properties, a hue, saturation, and brightness. Nevertheless, we can speak of colors as similar along one dimension but not another. For example, two colors may be very close along the hue dimension, but one may be much brighter than another. Now imagine a causal sphere (or rather, hypersphere), where each of the many dimensions corresponds to a causal manifestation. The cutting open of a cardboard box would be one dimension, the opening of a lock another. It could turn out that two shapes are more similar along one dimension than another. For example, a "sharp shape" and a "dull shape" could be very close along the lock-opening dimension, but more distant along the box-cutting dimension. But when we say that the shape of my key, when it opens a lock, is efficacious as a box-cutter, this does seem to imply (iii). That is, (S) seems to imply that close similarity to the shape along the box-cutting dimension entails close similarity along the lock-opening dimension.

(To digress a bit further, I note that there are two ways to measure similarity along a causal dimension: (1) Measure it by similarity of manifestation, holding power partners fixed. So, for example, given the same partners, the dull shape cuts the box roughly, while the sharp shape cuts it smoothly. (2) Measure it by similarity of partners, holding the manifestation fixed. So, for example, the dull shape needs much force to cut the box, while the sharp shape needs little force to cut the same box. Note that either measure merely fixes the "slots" along the dimension into which properties fall. It's still the properties themselves--and they alone--that determine where they fall along the dimension. In any case, the choice between (1) and (2) doesn’t look crucial for present purposes.)

Returning to the main thread: I submit that (S) seems false because we notice (correctly) that (i)-(iii) are false, and, thinking that (S) entails them, we infer that it's false as well. But (S) doesn’t entail any of (i)-(iii). At most, expressing (S) in a normal context pragmatically implies them. But expressing a truth can pragmatically imply a falsehood. I have no account to offer here of how expressing (S) pragmatically implies (i)-(iii), though I suspect it's because our purposes in expressing as-such propositions are often as much epistemic (in particular, explanatory) as metaphysical. But I've set aside these epistemic aims in this paper: my concern is in securing the metaphysical efficacy of the mental as such.

Turn then to mentality. Here similar points apply, where the offending proposition is:

(M) When a mental property is efficacious with respect to behavior, it's efficacious qua chemical dissolver.

While I say that (M) is true, I claim that it seems false because it seems to entail the following:

(i*) On this occasion, the power partners represented in the chemical part of the mental property's causal profile are present.

(ii*) On this occasion, the mental property was involved in dissolving a chemical.

(iii*) Any property similar enough to the mental property to dissolve that same chemical would also be able to produce the same behavior.

But while each of (i*)-(iii*) is false, (M) doesn’t entail them; expressing (M) merely pragmatically implies them.[13]

Notes

References

Antony, L. 1991. The Causal Relevance of the Mental: More on the Mattering of Minds. Mind and Language 6: 295-327.
Antony, L. and J. Levine. 1997. Reduction with Autonomy. Philosophical Perspectives 11: 83-105.
Armstrong, D. M. 1983. What is a Law of Nature? Cambridge University Press.
Blackburn, S. 1990. Filling in Space. Analysis 50: 62-5.
Boyd, R. 1980. Materialism with Reductionism: What Physicalism Does Not Entail. In N. Block, ed., Readings in Philosophy of Psychology, Vol. 1. Harvard University Press.
Burge, T. 1993. Mind-Body Causation and Explanatory Practice. In J. Heil and A. Mele, eds., Mental Causation. Clarendon Press.
Campbell, K. 1990. Abstract Particulars. Basil-Blackwell.
Crane, T. 2001. The Significance of Emergence. In C. Gillett and B. Loewer, eds., Physicalism and Its Discontents. Cambridge University Press.
----- 2008. Causation and Determinable Properties: On the Efficacy of Colour, Shape, and Size. In J. Hohwy and J. Kallestrup, eds., Being Reduced: New Essays on Reduction, Explanation, and Causation. Oxford University Press.
Davidson, D. 1980. Mental Events. In Essays on Actions and Events. Oxford University Press.
Fodor, J. 1974. Special Sciences (Or: The Disunity of Science as a Working Hypothesis). Synthese 28: 97-115.
Heil, J. 2003. From an Ontological Point of View. Clarendon Press.
Heil, J. and D. Robb. 2003. Mental Properties. American Philosophical Quarterly 40: 175-96.
Honderich, T. 1982. The Argument for Anomalous Monism. Analysis 42: 59-64.
Horgan, T. 2001. Causal Compatibilism and the Exclusion Problem. Theoria 16: 95-116.
Kim, J. 1988. Explanatory Realism, Causal Realism, and Explanatory Exclusion. Midwest Studies in Philosophy 12: 225-39.
----- 1993. Supervenience and Mind. Cambridge University Press.
----- 1998. Mind in a Physical World. MIT Press.
----- 2005. Physicalism, or Something Near Enough. Princeton University Press.
Lepore, E. and B. Loewer. 1987. Mind Matters. Journal of Philosophy 84: 630-42.
Levin, J. 2010. Functionalism. In E. Zalta, ed., The Stanford Encyclopedia.
McLaughlin, B. P. 1989. Type Epiphenomenalism, Type Dualism, and the Causal Priority of the Physical. Philosophical Perspectives 3: 109-35.
Marmodoro, A. 2010. Do Powers Need Powers to Make Them Powerful? From Pandispositionalism to Aristotle. In A. Marmodoro, ed., The Metaphysics of Powers: Their Grounding and their Manifestations. Routledge.
Martin, C. B. 2010. The Mind in Nature. Oxford University Press.
Mulligan, K. 1998. Relations--Through Thick and Thin. Erkenntnis 48: 325-53.
O’Connor, T. and H. Y. Wong. 2005. The Metaphysics of Emergence 39: 658-78.
Oddie, G. 1982. Armstrong on the Eleatic Principle and Abstract Entities. Philosophical Studies 41: 285-95.
Pereboom, D. 2002, Robust Nonreductive Materialism. Journal of Philosophy 99: 499-531.
Pettit, P. 1992. The Nature of Naturalism. Proceedings of the Aristotelian Society, Supp. Vol. 66: 245-66.
Robb, D. 1997. The Properties of Mental Causation. Philosophical Quarterly 47: 178-94.
----- 2001. Reply to Noordhof on Mental Causation. Philosophical Quarterly 51: 90-4.
----- 2005. Qualitative Unity and the Bundle Theory. The Monist 88: 466-92.
----- 2012. The Identity Theory as a Solution to the Exclusion Problem. In S.C. Gibb, E.J. Lowe, and V. Ingthorsson, eds., Mental Causation and Ontology. Oxford University Press.
Robb, D. and J. Heil. 2008. Mental Causation. In E. Zalta, ed., The Stanford Encyclopedia of Philosophy.
Seager, W. ms. Generalized Epiphenomenalism.
Shoemaker, S. 1980. Causality and Properties. In. P. van Iwagen, ed., Time and Cause. Reidel. Reprinted in Shoemaker 2003.
----- 2001. Realization and Mental Causation. In C. Gillett and B. Loewer, eds., Physicalism and Its Discontents, Cambridge University Press. Reprinted in Shoemaker 2003.
----- 2003. Identity, Cause, and Mind, Expanded Edition. Oxford University Press.
----- 2007. Physical Realization. Oxford University Press.
Simons, P. 1998. Farewell to Substance: A Differentiated Leave-Taking. Ratio 11: 235-52.
Sinnott-Armstrong. W. ms. Downward Mental Causation.
Smart, J. J. C. 1959. Sensations and Brain Processes. Philosophical Review 68: 141-56.
Sosa, E. 1984. Mind-Body Interaction and Supervenient Causation. Midwest Studies in Philosophy 9: 271-81.
Tye, M. 2007. Qualia. In E. Zalta, ed., The Encyclopedia of Philosophy.
Unger, P. 2006. All the Power in the World. Oxford University Press.
Williams, D. C. 1966. The Elements of Being. In Principles of Empirical Realism. Charles Thomas.
Williams, N. E. 2010. Putting Powers Back on Multi-Track. Philosophia 39: 581-95.
Wilson, J. 2009. Determination, Realization and Mental Causation. Philosophical Studies 145: 149-69.
-----. 2011. Non-reductive Realization and the Powers-based Subset Strategy. The Monist 94: 121-54.
Yablo, S. 1992. Mental Causation. Philosophical Review 101: 245-80.