Springer Philosophical Studies: An International Journal For Philosophy in The Analytic Tradition

A Nonreductivist Solution to Mental Causation
Author(s): Amie Thomasson

Source: Philosophical Studies: An International Journal for Philosophy in the Analytic
Tradition, Vol. 89, No. 2/3, The American Philosophical Association Pacific Division
Meeting 1997 (Mar., 1998), pp. 181-195
Published by: Springer
Stable URL: http://www.jstor.org/stable/4320818
Accessed: 15-07-2016 08:13 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
Springer is collaborating with JSTOR to digitize, preserve and extend access to Philosophical Studies: An
International Journal for Philosophy in the Analytic Tradition
This content downloaded from 129.82.28.144 on Fri, 15 Jul 2016 08:13:26 UTC
All use subject to http://about.jstor.org/terms
AMIE THOMASSON
A NONREDUCTIVIST SOLUTION TO MENTAL CAUSATION
(Accepted 31 July 1997)
ABSTRACT. Nonreductive physicalism provides an appealing solution to the

nature of mental properties. But its success as a theory of mental properties has
been called into doubt by claims that it cannot adequately handle the problems
of mental causation, as it leads either to epiphenomenalism or to thoroughgoing
overdetermination. I argue that these apparent problems for the nonreductivist are
based in fundamental confusion about causation and explanation. I distinguish two
different types of explanation and two different relations to which they appeal:
causation and determination. I argue that these types of explanation do not compete
with one another, nor do these relations jointly result in overdetermination. In
closing I develop a nonreductivist solution to mental causation which avoids both
the hazards of epiphenomenalism and of overdetermination and so demonstrates
a way to save nonreductive physicalism from the problems of mental causation.
Many are those tempted by some version of nonreductive physi-

calism. The attractions of such a view are obvious: The nonreduc-
tivism should ensure that mental properties are preserved in all of
their distinctive character, while the physicalism should ensure that
we nonetheless offer a view of mind compatible with a scientific
world-view. But this appealing solution has been brought into doubt
by arguments that the nonreductive physicalist faces an unpleasant
dilemma regarding mental causation.
Either mental properties have causal efficacy or they don't. If they
don't, then we seem to be left with a kind of epiphenomenalism about
the mental. But epiphenomenalism has unpleasant consequences:
First, it seems it would force us to give up our intuitions that our
beliefs, desires, and so on do have a causal impact on the world.
Secondly, if all mental properties lack causal powers, then there
seems little reason to postulate them; they would fail tests such as
Alexander's dictum: "To be real is to have causal powers". But the
other horn of the dilemma, that mental properties do have causal
powers, is hardly more palatable. If we allow higher-level mental
Philosophical Studies 89: 181-195, 1998.

? 1998 Kluwer Academic Publishers. Printed in the Netherlands.
182 AMIE THOMASSON
properties to have causal powers, we seemed threatened with a

thoroughgoing overdetermination of many sorts of events which
are caused both by instantiations of mental properties and of phys-
ical properties. Worse still, if we allow that mental properties may
have causal effects at the physical level, we risk violating the causal
closure of the physical. Fear of facing this dilemma has substan-
tially dimmed the initial appeal of nonreductive views of mental
properties.'
One way out now acknowledged even by Jaegwon Kim, who
develops such arguments against nonreductive physicalism in great
detail, is to embrace functionalism.2 As a functionalist one can main-
tain the view that mental property types are not reducible to physical
property types, and yet allow that instances of mental properties are
identical with instances of physical properties. If mental and phys-
ical property instances are identical, instances of mental properties
will have all of the causal powers of the physical property instances
(thus avoiding epiphenomenalism), and yet there will be no danger
of overdeternination since there is but one entity at work as the
cause.
The functionalist way out, however, is unlikely to fully satisfy
many nonreductivists. For there seem to be at least two good reasons
to resist the idea that mental property instances are identical to their
physical bases: unlike physical property instances, they seem to
have an essential qualitative nature, and in some cases at least they
seem to have essential semantic or contentual features which also
depend in part on external ("wide") social and physical context.
Those impressed by these features of our mental life will want to
resist the ontological reduction of mental property instances to the
physical, and thus, it seems, be left with the original dilemma.
The purpose of this paper is to examine the prospects for a differ-
ent solution, one that the preserves the distinctive status of the mental
by maintaining the irreducibility of mental property instances, as
well as mental property types, to their physical basis. The alleged
difficulties such views face with mental causation, I will argue, are
merely apparent problems based in fundamental confusion about
causation and explanation. Thus it would be a mistake to abandon
such nonreductive physicalist views because of fears about handling
mental causation.
A NONREDUCTIVIST SOLUTION TO MENTAL CAUSATION 183
Closer consideration of these issues shows the way to an account

of mental causation that provides an appropriate causal role for
mental properties without giving up ordinary principles about causa-
tion or abandoning the idea that the physical world is ontologically
and causally primary. If it can avoid the alleged problems of mental
causation, nonreductive physicalism may after all provide a viable
and appealing understanding of the mental.
I. CAUSATION
There are several different relations indifferently and equivocally

referred to as relations of 'causation'. It was to avoid this equivoca-
tion that Aristotle distinguished no fewer than four 'causal' relations,
carefully isolating efficient causation from material, final, and formal
causation.3 Different sorts of explanation may appeal to different
underlying relations. Thus there may be more than one explanation
of an event without these explanations being in competition, and
there may be more than one such relation involving the event with-
out its being causally overdetermined. Indeed far from holding that
explanations involving different types of cause compete, Aristotle
held that a complete explanation of an event is one which appeals to
all of these types of cause.
I will begin by considering explanation, and thereafter turn to
examine the different ontological relations underlying the different
types of explanation. Consider the following: Upon pulling his sock
out of the dryer, John notices the following state of affairs: The sock
is pink. Annoyed at this, John turns to his roommate, Tom, to ask
for an explanation. Tom might offer two different explanations of
that state of affairs. The natural explanation might be something
like this: the sock is pink because it was washed with a new red
sweatshirt. But, if he is feeling recalcitrant, Tom might offer instead
the following explanation: the sock is pink because its component
matter has a certain microstructure that preferentially reflects those
wavelengths of light making up pink. Or, consider a parallel case:
These cookies are hard. Why are they so hard? Explanation 1: They
were overcooked. Explanation 2: Their molecules are very tightly
bonded.
In neither case do the two explanations compete with each other.4
Indeed so far are they from competing that anyone who offered these
184 AMIE THOMASSON
as alternative explanations would be thought to be making a bad joke.

They can both be true without there being any tension between them,
or any need to ask which is the unique true explanation. They each
provide a different kind of explanation. The first explanation in each
case (the sock was washed with a pink sweatshirt, the cookies were
overcooked) appeals to the causal factors in the past that brought
about the present state of affairs. The second explanation in each
case (the sock is pink because its matter absorbs light of all other
wavelengths, the cookies are hard because their molecules are very
tightly bonded) appeals to the underlying physical basis for the state
of affairs at this moment. Call these explanation, and explanation2
respectively.
These two types of (non-competing) explanation refer to two
different types of relation. Although on occasion these may both
be indifferently referred to as 'causal', it is extremely important to
distinguish them. (There may be still more types of relation appealed
to in common discussions of causation, but distinguishing these two
will prove sufficient for present purposes). We can distinguish them
preliminarily as follows. Let us call the relation appealed to in the
first explanation of each case 'causation', which is at least typically
a diachronic relation between an earlier event and a later one that
it brings about. The relation appealed to in the latter explanation
of each case I will call 'determination'.5 As I shall use the term,
'determination' refers to a synchronic relation between higher-level
entities and the lower-level entities that contribute to their being the
way that they are. Concrete higher-level entities that are materially
constituted by physical entities, for example, are largely determined
by their physical bases.6 (Aristotle's efficient cause seems closest
to what I have labeled 'causation', while his material causation
provides a central case of what I call 'determination'.) A particular
state of affairs (the sock's being pink, the cookie's being hard) may
be both caused to be a certain way and determined to be that way.
These relations do not exclude each other, nor does the presence of
both of them constitute causal overdetermination, since only one of
them is causal. Armed with this distinction, let us return to examine
the arguments that the nonreductivist faces troubles with mental
causation.
II. KIM'S ARGUMENT AND RESPONSE TO KIM
The toughest criticisms of nonreductive physicalism have come from

Jaegwon Kim's arguments that the nonreductivist faces insuperable
problems with mental causation. Kim argues, first, that the nonreduc-
tive physicalist should choose the second horn of the dilemma about
mental causation: That mental properties do have causal powers. For,
he says, following Alexander's dictum, we should only postulate
entities that have new causal powers (which no physical thing can
deliver).7 So the nonreductive physicalist must accept that mental
properties have causal powers. Kim plausibly suggests that the most
likely place to look for the causal powers of mental properties is
in their power to cause instances of other mental properties; these,
at least, should present the easier cases than purported cases of
downward (to the physical) or upward (to the social) causation.8 So,
for some instances of mental properties M and M*, M causes M*.
But, since the nonreductive physicalist also insists that all mental
properties are physically realized, she must also admit that M* is
physically realized by some physical property P*. It is here that the
purported trouble arises:
Now we seem to have two distinct and independent answers to the question, Why
is this instance of M* present? Ex hypothesi, it is there because an instance of M
caused it; that's why it's there. But there is another answer: it's there because P*
physically realizes M* and P* is instantiated on this occasion. I believe these two
stories about the presence of M* on this occasion create a tension and must be
reconciled.9
From there Kim proceeds to reject several views about how these
views could be reconciled. M can't be jointly causally responsible
with P* for M* since P* is alone sufficient:
... every instance of M* must have some physical base that is by itself sufficient
for M*; and this threatens to preempt M's claim to be the cause of this instance of
M* 10
It can't be just that M is physically realized in P, which causes P*,

since this leaves M no causal work to do in bringing about M* and
so leaves mental properties in violation of Alexander's dictum. And
so, Kim concludes, the only way out is to say that M is causally
responsible for M* by causing P*. In other words, the only way to
make sense of cross-level causation is by resorting to downward
186 AME THOMASSON
causation. But that would leave P* overdetermined - caused by both

M and P. It seems that M is bound to be the loser in this competition.
This unsavory conclusion derived from Kim's claim that there
are two distinct and competing explanations for the presence of
M* on the occasion in question: Because M caused it or because
P* physically realizes it. These explanations, Kim claims, "create a
tension and must be reconciled". But do they? It seems to me that
they do not. The word 'because' is being used equivocally; these
cases offer two different types of explanation. In the first case, "M*
is there because M caused it" expresses a causal relation between
M and the M* it brings about. This is an explanation,, parallel to
the explanation that the sock is pink because it was washed with
a red sweatshirt. In the latter case, "M* is there because P* phys-
ically realizes it" expresses a relation of material constitution and
determination between P* and M*. This is an explanation2, parallel
to the explanation of the sock's pinkness in terms of its microstruc-
ture. Just as there is no tension between the two explanations of the
sock's color, so is there no tension between these explanations of the
mental property. They are different types of explanation appealing
to different relations: M caused M*, while P* determines M*.
The initial plausibility of Kim's argument derives only from our
mistaken tendency to confuse causation and determination, or to
think of determination as a causal relation. Despite his care at some
points not to label it such, Kim clearly is tempted to think of physical
realization this way:
When P is said to "realize" M in system s, P must specify a microstructural

property of s that provides a causal mechanism for the implementation of M in s
The tendency to think of physical realization, one relation involving

what I have called 'determination', as a causal relation might make
it seem that there is an irreconcilable tension between these two
explanations. But if we bear in mind the parallel cases where we are
less likely to be led astray than in the labyrinths of the mind-body
problem, we can see quite clearly that the problems lies not with
mysteriously competing explanations of mental events but merely
with failing to distinguish different types of explanation and the
relations they describe.
III. THE "LAYERED" SOLUTION
If the foregoing is correct, then the widely accepted arguments based

on purported troubles with mental causation give us no reason to
abandon nonreductive physicalism. Yet the question remains of how
a thoroughly nonreductive physicalist could offer a positive resolu-
tion to the difficulties of mental causation without falling into epiphe-
nomenalism or violating principles about causation. The details
cannot be completely worked out here, but the following sketch
should demonstrate how such a solution could be drawn out.
The solution is based on the very view Kim argues against, which
he refers to as a "layered" view of the world.12
It is generally thought that there is a bottom level, one consisting of whatever

microphysics is going to tell us are the most basic physical particles out of which
all matter is composed ... As we ascend to higher levels, we find structures that
are made up of entities belonging to the lower levels, and, moreover, the entities
at any given level are thought to be characterized by a set of properties distinctive
to that level.'3
One common tenet of such layered views is that causation occurs

only within a level; there is no upward or downward causation.
Layers are not connected by causal relations but by relations of deter-
mination, dependence and material constitution. Provided that she
adheres firmly to this principle, the nonreductivist can avoid Kim's
problems and find a way out of the dilemma of mental causation.
Suppose that there are different levels of properties, where the
higher level properties are dependent on (ultimately exclusively
dependent on) lowest level, purely physical, properties, but not
reducible to them. Mental properties are placed among the higher
level properties; each mental property instance could be said to
depend on the particular physical property instance in which it is
realized. We can add, moreover, that each mental property instance
is ultimately exclusively dependent on and fully determined by the
physical, though perhaps not only that physical property instance on
which it depends - it may also depend on and be partially determined
by features of the physical and even social context surrounding it,
provided that the social context in turn is ultimately exclusively
dependent on the physical. This ensures the primacy of the physical
and the dependence of the mental on it. Nonetheless, according to this
view, in virtue or their semantic, contentual or qualitative features
188 AMIE THOMASSON
or their dependence on the surrounding social context, mental prop-

erties at the level of both type and token cannot be identified with
the physical properties upon which they depend.
Now suppose that not only the most fundamental properties, but
also the most fundamental causal relations (and objects) are at the
basic microphysical level. Nonetheless, just as there may be higher-
level objects and properties that depend on but are not reducible
to fundamental objects and properties, so may there be higher-
level causal relations that depend on lower-level causal relations
without being reducible to them. This dependence seems plausible.
Consider a higher-level causal relation such as: The speculation of
investors caused the stock-market crash. We are unlikely to credit
this causal story unless there were also relations grounded in funda-
mental physical interactions between the speculators and traders - if
there were no interactions via phone calls, faxes, or what have you,
we would rule out these speculations as irrelevant to the crash. Insist-
ing that such higher-level causal relations are themselves dependent
on fundamental causal relations should mitigate the impression that
mysterious new powers are being brought in at the higher levels:
these causal relations, like the events they connect, are ultimately
dependent only on entities at the basic physical level.
Then the nonreductivist's answer to the question. "Does the
mental have causal powers?" should be "Yes: there are distinct,
dependent, but not reducible causal relations involving instances of
mental properties". This solution avoids the first horn of the dilemma.
It is not epiphenomenalism because it allows there to be causal rela-
tions among higher-level entities (though these relations, like the
properties, are dependent). Thus we need not give up intuitions that
our mental properties enter causal relations with our actions, that the
decisions of traders enter causal relations with the stock market and
so on.
Moreover, mental properties so conceived satisfy Alexander's
dictum that to be real is to have distinct causal powers. For an
instance of a mental property M may have the power to cause M*.
Since M* is a higher-level property, it does not appear at the basic
physical level, and so no causal relation of the form P causes M* can
obtain at a lower level. Nor can it obtain across levels, since there is
no cross-level causation. M* may be dependent on, determined by,
even materially constituted by physical properties, but not caused by

them. So M has a causal power that no physical property has, namely
the power to cause M*. And so M satisfies Alexander's dictum.
But allowing mental properties to have causal powers in this sense
does not lead to causal overdetermination. P* is not overdetermined
because M does not cause P*. We have no need to posit a downward
causal relation between M and P* because there is no tension between
explanations of M* in terms of its causation by M and determination
by P* and so no need to route M's causation of M* via P*. And so
P* has only one cause - a physical property P. Thus there is also
no violation of the causal closure of the physical, since downward
causation is ruled out on this view.
One might fear that even if P* is not overdetermined, M* is
overdetermined by being caused by both M and P* (in which it is
physically realized). But only M causes M*, while P* determines
it. M* is no more overdetermined by its relations to its cause and
physical basis than the cookie's hardness is overdetermined by the
cooking and the tight bonding of molecules.
Even those who agree this far may still fear that there is some
overdetermination going on. It may be agreed that M* is not over-
determined via M and P*, yet, it might be said, the real competitor
for the cause of M* is P, not P*: The causal story that seems to
compete with M causing M* is the one that runs across the physical
level. Thus in the cookie analogy the real causal competitor for the
higher-level story told simply in terms of overcooking might be that
the prolonged heating causes the emulsion bonds in the dough to
break down, causing the evaporation of water molecules and the
replacement of weak emulsion bonds with strong rigid bonds among
remaining fat and carbohydrate molecules.14
But if we carefully maintain the initial distinction between causa-
tion and determination, this threat also may be held at bay. We may
tend to casually, and harmlessly, speak of such relations as "causal"
in ordinary conversation, but if we respect the philosophical distinc-
tion between causation and determination, we can see that there
is equivocation at work in our desire to call P the "cause" of M*;
the same equivocation which the distinction between causation and
determination was meant to eradicate. The prolonged heating causes
the breakdown of the emulsion bonds and their replacement with
190 AMIE THOMASSON
strong rigid bonds among molecules. And it is by thinking transi-

tively that we then wish to say that the strong rigid molecular bonds
"cause" the cookies to be hard. But this move makes use of the
two senses of "causation": causation in the first case and determi-
nation in the second case. If we keep them distinct, the presence
of strong rigid molecular bonds cannot be properly said to "cause"
the cookie's hardness; instead, it forms the physical basis for that
hardness.15 Put more formally, in the case of a physical property P
purportedly causing mental property M*, M* is not overdetermined
by being caused by both M and P. For although P causes P*, since P*
does not cause M* (but determines it), P does not transitively cause
M*. It merely causes the physical event in which M* is physically
realized. And so M* has only one cause, namely M.
Nor is there non-causal overdetermination of M* or competition
among explanations of M*. Observations that both P* and M are (in
some sense) sufficient for M* should not mislead us into thinking
that M* must be overdetermined. As has been repeatedly observed,
one should not confuse sufficiency with causation.16 P* and M are
"sufficient for" M* in different ways: One is sufficient to bring it
about, the other is its sufficient physical basis. In short, as I have
argued above, the relations of causation and determination do not
exclude or compete with one another. M* is no more overdetermined
than the pink sock is overdetermined by being turned pink through
careless washing and determined to be pink by having a certain
microstructure; the explanation in terms of causal history does not
compete with that in terms of current physical basis.
If the foregoing is correct then the nonreductive physicalist can
after all provide a consistent answer to the dilemma of mental causa-
tion: Just as mental properties are dependent on but not reducible to
the physical properties that realize them, so are there causal relations
among mental properties that are dependent on but not reducible to
fundamental physical causal relations. By maintaining the ultimate
dependence of the mental on the physical, and of higher-level causal
relations on fundamental physical causal relations we preserve the
primacy of the physical, but do so without saying that all causal
powers are just the causal powers of fundamental physical entities.
IV. ASSESSMENT
The solution proposed has many virtues: It enables us to preserve

mental properties in all of their distinctive nature, to provide them
with an appropriate causal role, to avoid problems of overdetermi-
nation or violation of the causal closure of the physical, and at the
same time to preserve a scientific world view that sees the physical
world as fundamental.
But the problem of mental causation is indeed a thorny one and
no solution to it is likely to be entirely without cost. The solution
proposed above is no different in this regard. The main apparent
drawback is that it prevents us from accepting at face value ordinary
claims of physical-mental and mental-physical causation, since on
this view causal relations do not hold between levels. Thus, for
example, it cannot be literally true on this view that ingesting the
drug caused the headache to go away, nor can it literally true that
Julie's deciding to swing the baseball bat caused the ball to fly in an
arc and break the window.17
Cases of purported physical/mental causation such as medication
causing one's headache to go away are fairly easily handled as simple
cases of harmless "loose talk". In common parlance we may speak
of this all as causation, but what is really going on is a causal
relation (the medication causing the change in noradrenaline activity,
causing the resumption of normal blood flow to the cortex) and a
determination relation (the resumption of normal blood flow to the
cortex determines and is the physical basis for the ceasing of the
headache). This seems to me to provide an appropriate analysis
which preserves the idea that such claims are, in some sense, true,
provided the appropriate distinctions are preserved.
It is somewhat more difficult to provide a satisfying analysis
of cases of purported mental/physical causation on this model. Yet
there are mitigating factors which can make this easier to accept.
First, it should be noted that, although this picture rules out down-
ward mental causation of fundamental physical events, there can be
mental causation of higher-level events such as resignations, wars,
stock market crashes, professions of love, or acts of vandalism. For
these themselves, as full cultural events with meanings in the social
human world, likewise are based in the physical world but are not
mere fundamental physical entities. Thus, for example, Julie's deci-
192 AMIE THOMASSON
sion could be said to cause higher-level events such as the school's

suffering yet another act of vandalism, Julie's feeling avenged, or
Julie being wanted by the local authorities. Such higher-level causal
relations in turn depend upon lower-level causal relations such as
her neurons sending signals causing her muscles to contract, the bat
to swing, the ball to fly, and the glass to shatter.
Limiting the causal powers of the mental to higher-level events,
ruling out effects at the basic physical level, may not seem so bad
if we remember the alternatives. The choices among nonreductive
views seem to be the following: 1) Choose functionalism and identify
instances of mental properties with their physical bases, at risk of
losing the essential qualitative, semantic, or contentual features of
mental or 2) Remain nonreductivist at both type and token levels and
a) Treat the mental as epiphenomenal (clearly a less attractive option
for those who take intuitions about mental causation seriously) or b)
Accept mental-physical causation and its attendant thoroughgoing
overdetermination and violation of the causal closure of the physical
or c) Postulate mental causation, but limit it to higher-level events.
Once the alternatives are made evident, many nonreductivists might
find that the costs of positing only higher-level causal relations fade
in comparison to the costs of losing the distinctive status of the mental
or accepting overdetermination and violating the causal closure of
the physical.
The availability of a consistent reply to Kim's challenge should
at least establish that arguments such as Kim's give us no reason
to abandon nonreductive physicalism. Whether we have positive
reason to accept nonreductive physicalism and therewith embrace
the idea that higher-level properties such as mental properties can
enter into higher-level causal relations is a further matter. Ultimately,
the nonreductive physicalist will owe us an account of what these
higher-order causal relations are, under what conditions they obtain,
and so on. Likewise, she will owe us a more thorough analysis of
the relations of determination and causation. Only thereby can we
properly evaluate the nonreductivist picture. Nonetheless, I hope this
does enough to show that the project of a nonreductive physicalism
- even one that denies the identity of mental and physical property
instances - is far from dead, and to suggest a direction to take in
shoring up such a view. To give up a nonreductivism that insists
A NONREDUCrIVIST SOLUTION TO MENTAL CAUSATION 193
on the unique and distinctive status of mental properties because

of fears that it has no consistent reply to the dilemma of mental
causation would be to abandon an interesting and promising theory
prematurely.
Kim closes his second Postscript on Mental Causation with the
following assessment:
Ultimately, we are likely to face the following choice: either embrace the [func-
tionalist's] realization view and save mental causation, or insist on the unique and
distinctive status of mental properties, especially the qualia, but be prepared to
give them up as causal powers ... If you choose the former, you may lose what
makes the mental distinctively mental, and what good is it, one might ask, if
you save mental causation but end up losing mentality in the process? . . . If you
choose the latter, you may again lose the mental, for what good is something that
is causally impotent? ... We therefore seem to be up against a dead end. Perhaps,
that is what's really so intractable about the problem of the mind.18
What I have been suggesting is that there is another, perhaps more

appealing, option. For we can preserve the distinctive status of
mental properties (instances as well as types) and also preserve an
appropriate causal role for them, without giving up ordinary princi-
ples about causation or the idea that the physical world is primary.
If it can be made to work, there may be a path out of the dead end,
and the problem of the mind may not be so very intractable after
all.'9
NOTES
1 Since the purported problem arises by considering mental properties to be

higher-level properties, it would also apply to non-reductive views of other sorts
of properties such as economic and social properties. Thus the issue of how to
handle causation within a nonreductive physicalism reaches far beyond issues in
the philosophy of mind; what is at stake more generally is the ability to take any
kind of property to be both higher-level and causally efficacious. Nonetheless, as
it is in the philosophy of mind that this problem has received the greatest attention,
I will focus on that aspect of the problem here.
2 "Postscripts on Mental Causation" in Supervenience and Mind (New York:
Cambridge University Press, 1993), pp. 362-367.
3 See Physics 11.7; 22-25.
4 Steven Yablo develops a different sort of "no competition" argument based in
the idea that mental properties stand to physical in the relation of determinable
to determinate. See his "Mental Causation" (The Philosophical Review, Volume
101, No. 2, April 1992). Unlike his, the no-competition argument advanced here
does not rely on the idea that physical properties are properly seen as determinates
of mental properties.
194 AMIE THOMASSON
S "Determination", like "causation", is sometimes used in different ways to signify

different relations. Indeed it is occasionally used as a cover-all term to include
causation as well as the upward determination of higher-level entities by their
lower-level bases discussed here. What is required is simply a consistent usage
that clearly separates out the two meanings, which I have stipulated using the
familiar terms "causation" and "determination", but this is merely a terminolog-
ical choice.
6 Determination, however, is not the same as material constitution. A concrete
entity may be determined by other factors than its own material make up, as the
shape of a balloon may be partly determined by the surrounding air pressure,
or the meaning of an utterance may be partly determined by the relevant social
practices. Dependent abstract entities, such as fictional characters, although they
are not materially constituted, may also be determined by aspects of the stories in
which they appear. Thus while the paradigm cases of determination such as those
discussed below are cases where the determining entity materially constitutes the
determined entity, determination is in fact a broader relation that does not merely
hold between entities and their material bases.
I Leonard Clapp has pointed out that is not clear that Alexander's dictum demands
that these causal powers must be new, although Kim interprets it as such.
8 "The Nonreductivist's Troubles with Mental Causation", in Supervenience and
Mind, p. 351.
9 Ibid,p. 351.
10 Ibid, p. 352.
l Ibid, p. 343 (italics mine).
12 This may not be the only possible resolution; I am grateful to Michael Watkins,
Anthony Dardis, and Leonard Clapp for the clear formulation and helpful discus-
sion of other solutions.
13 Supervenience and Mind, p. 337.
14 The accuracy of claims regarding the physics of cookie dough cannot be
guaranteed.
lS There may be those who resist the above example by insisting that the cookie's
hardness is the strong rigid bonds among its molecules, so that in fact the physical
causal chain is the only one. But the issue here is not whether nonreductivisim is
true concerning hardness or mental properties - that must be argued separately,
and certainly there is a much stronger case to be made for nonreductive views of
mental properties than of a property such as hardness. The issue here is merely
whether, if we postulated a nonreductive view, it would encounter insuperable
difficulties with mental causation. The cookie analogy is simply intended to make
it clear how a reply to this objection could go for someone maintaining a nonreduc-
tive view, its success in this regard does not depend on accepting a nonreductive
view of hardness.
16 See, e.g., the introduction to Ernest Sosa and Michael Tooley, Causation
(Oxford University Press, 1993).
17 Thanks to James Garson and an anonymous member of the audience at the
1997 Pacific APA presentation of this paper for these examples.
18 Supervenience and Mind, p. 366-7.
19 This paper took shape as a result of discussions at the 1996 National Endow-
ment for the Humanities summer seminar "The Metaphysics of Mind". Thanks go
to the National Endowment for the Humanities and to John Heil and the seminar
members for many helpful discussions. Thanks also to Murat Aydede and Tim
Van Gelder for helpful comments on an earlier version of this paper.
Department of Philosophy,
Texas Tech University,
Lubbock, TX 79409-3092,
U.S.A.

Springer Philosophical Studies: An International Journal For Philosophy in The Analytic Tradition

Uploaded by

Copyright:

Available Formats

Springer Philosophical Studies: An International Journal For Philosophy in The Analytic Tradition

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Springer Philosophical Studies: An International Journal For Philosophy in The Analytic Tradition

Uploaded by

Copyright:

Available Formats

A Nonreductivist Solution to Mental Causation

Author(s): Amie Thomasson

A NONREDUCTIVIST SOLUTION TO MENTAL CAUSATION

(Accepted 31 July 1997)

ABSTRACT. Nonreductive physicalism provides an appealing solution to the

Many are those tempted by some version of nonreductive physi-

Philosophical Studies 89: 181-195, 1998.

properties to have causal powers, we seemed threatened with a

Closer consideration of these issues shows the way to an account

There are several different relations indifferently and equivocally

as alternative explanations would be thought to be making a bad joke.

II. KIM'S ARGUMENT AND RESPONSE TO KIM

The toughest criticisms of nonreductive physicalism have come from

It can't be just that M is physically realized in P, which causes P*,

causation. But that would leave P* overdetermined - caused by both

When P is said to "realize" M in system s, P must specify a microstructural

The tendency to think of physical realization, one relation involving

III. THE "LAYERED" SOLUTION

If the foregoing is correct, then the widely accepted arguments based

It is generally thought that there is a bottom level, one consisting of whatever

One common tenet of such layered views is that causation occurs

or their dependence on the surrounding social context, mental prop-

even materially constituted by physical properties, but not caused by

strong rigid bonds among molecules. And it is by thinking transi-

The solution proposed has many virtues: It enables us to preserve

sion could be said to cause higher-level events such as the school's

on the unique and distinctive status of mental properties because

What I have been suggesting is that there is another, perhaps more

1 Since the purported problem arises by considering mental properties to be

S "Determination", like "causation", is sometimes used in different ways to signify

You might also like