Learning Center
Plans & pricing Sign in
Sign Out



									Mental Terms and Empirical Psychology
Brandon N. Towl
Washington University in St. Louis

Abstract: In this paper I raise a problem for Kim‘s (1998) more recent solution to the
mind/body problem, his ―reduction by functionalization‖. Though this theory solves
certain problems, such as causal-overdetermination and mental/physical correlation, it
makes mental terms unnecessary for the explanation of behavior—they become either
eliminable or trivial. But if psychology is the conjunction of theories that use mentalistic
terminology and explains behavior, then there is no such conjunction of theories.
Reduction by functionalization makes no room for empirical psychology.

I. Introduction
         In this paper I raise a problem for Kim‘s (1998) more recent solution to the
mind/body problem, his ―reduction by functionalization‖.1 Though this theory solves
certain problems, such as causal-overdetermination and mental/physical correlation, it
makes mental terms unnecessary for the explanation of behavior—they become either
eliminable or trivial. But if psychology is the conjunction of theories that use mentalistic
terminology and explains behavior, then there is no such conjunction of theories.
Reduction by functionalization makes no room for empirical psychology.
         Here I take psychology to be the conjunction of theories that (1) use mentalistic
terminology and that (2) aim to develop true explanations of behavior.2 Examples of
mental terms include intentional descriptions, or sensory descriptions, or both. Examples
of intentional descriptions include such verbs as ―thinks‖, ―believes‖, ―hopes‖, ―fears‖,
and so on, that mark propositional attitudes. Examples of sensory descriptions are
usually descriptions of some sort of conscious state, such as ―seeing red‖ or ―being in
         Analysis of mental terms hopefully yields definitions and clarifications of these
terms. But besides the analysis of our mental terms, we can ask whether these terms
name unique mental properties. And if such terms do name mental properties at all, then
it is natural to ask what the relationship is between these mental properties and non-
mental physical properties. If there is no such relationship, mental properties would exist

  I do not have a proposed solution to the problem for Kim; my aim here is simply to show the apparent
  This is the same definition used by Jaworski (2002).

somehow separate from physical properties. And then dualism would be true. The
mind/body problem is a search for such a relationship.
           In the next section, I review a position recently defended by Jaegwon Kim (1998,
though similar versions can be found in earlier works by David Armstrong and David
Lewis). If we accept Kim‘s position, however, the demands of explanations of behavior
make mental terms either eliminable or trivial. Hence, it appears that we do not have a
science of psychology as I have defined it.

II. Kim’s Reduction by Functionalization
           Jaegwon Kim has recently defended a position on the mind-body problem that he
calls ―reduction by functionalization‖. On this approach, mental properties are identical
to physical properties (though the identity here is only nomically necessary), while
mental terms are simply non-rigid designators for these properties. This position has
much in common with the approaches of Armstrong (1968) and Lewis (1966, 1972,
           The first step of reduction by functionalization is to construe mental properties not
as intrinsic properties, but rather as extrinsic properties, characterized relationally (1998
p. 25). Let B be a set of first-order (―base‖) properties, P1…Pn, that are non-mental. F is
a second-order property over set B of first order (base) properties iff F is the property of
having some property Pi in B such that D(P), where D specifies a condition on members
of B. Functional properties, then, are those second-order properties over B whose
specification D involves causal/nomic relations (1998, p.20).
           To construe a mental property M as a functional property is to say that having
mental property M is just having some property with such-and-so as its typical causes and
such-and-so as its typical effects. The term for M is thus defined in terms of M‘s
causal/nomic relations to other properties. Being a realizer of M is defined as being one
of the base properties P in B that meets specification D.
           Properties, or at least those properties that science considers, are typically
individuated by their causal powers.3 If mental properties are second-order properties,
then every instance of a mental property is either identical to an instance of a first-order

    See Shoemaker (1980); also Fodor (1987, p. 38) and Kim (1993, pp. 348-349).

property, or part of an instance of a first-order property, according to the causal
inheritance principle:

         If a second-order property F is realized on a given occasion by a first-
         order property H (that is, if F is instantiated on a given occasion in virtue
         of the fact that one of its realizers, H, is instantiated on that occasion), then
         the causal powers of this particular instance of F are identical with (or are
         a subset of) the causal powers of H (or this instance of H). (p. 54)

Every instance of F at time t, then, is identical to the instance of H (or part of the instance
of H) that is F‘s realizer at time t. Any mental term that refers to some mental property,
construed as a second-order property, also refers to some first-order property for a given
individual (or set of individuals) at a time (or span of time).
         What do we gain with such an identity? Kim sees two main benefits.4 First, if
mental properties are identical to physical properties, we have a clear explanation of
mental/physical correlation. Suppose that science discovers that pain in humans is
correlated with some neural state such as the firing of C-fibers. Why does this correlation
hold? That is, why is it that when a human instantiates ―having C-fibers firing‖ they also
instantiate ―being in pain‖? Why is it that pain is correlated with C-fiber firing, but not
A-fiber firing? And why is it that C-fiber firing is correlated with pain, but not itches or
tickles? These are questions that Kim believes a good mind/body theory should be able to
answer (see Kim p. 95). Finding a bridge law that holds between pain and C-fiber firing
is not adequate to the task; such a law may state the correlation, but it does not explain it.
         An identity statement, however, would answer the above questions. If pains just
are identical to C-fiber firings in humans, then the fact that they always appear together is
not mysterious. For Kim, asking why pains and C-fiber firings always seem to co-occur
in humans is like asking why it is that, whenever Larua Bush shows up, the President‘s

  Kim‘s account is not the only account that attempts to solve these problems. For example, Yablo (1992)
provides a solution by considering physical properties to be determinates of mental properties, and Bennett
(2003) considers certain counter-factual accounts. My goal here is not to defend Kim‘s account against
competitors, however, but to show that Kim‘s account has some advantages that make it seem attractive as
a solution to the mind/body problem by itself.

wife also shows up: ―there is no better, or conclusive, answer than ‗[Laura Bush] is the
President‘s wife‘.‖
           Second, identifying mental properties with physical properties solves the problem
of causal overdetermination. There is reason to think that every physical event has a
physical cause (and that this physical cause is sufficient for the effect). If every mental
event is accompanied by a physical event, then the mental event and the physical event
compete for causal relevance. If the mental event and the physical event are different
events, then we have a case of causal-overdetermination, which seems implausible. It
appears, then, that if mental properties are not physical, then mental properties are not
causally relevant to behavior.
           We avoid the conclusion that mental properties are not causally relevant to
behavior if we are willing to deny that mental events and physical events are really
different events. Identifying mental properties with physical properties gives us a way of
saying how mental properties can cause physical events: because mental properties just
are physical properties, relative to a system.
           Mental terms do not name unique mental properties, then. They are abbreviations
of definite descriptions that refer to physical properties. Kim suggests that we think of
such mental terms not as predicates that name unique mental properties, but rather as
mental designators or concepts.5 Mentalistic terminology is not unified in virtue of
mental terms referring to single properties, either universals or tropes. Rather mental
terminology is unified in virtue of mental terms expressing some sort of conceptual unity
(p. 110).

III. Explanations Using Mental Terms
           Above, I said that psychology was the conjunction of theories that (1) use
mentalistic terminology, and (2) aim to develop true explanations of behavior. But it
seems that, given Kim‘s analysis, doing (2) means giving up on (1). First, I want to
consider explanations of single, particular cases of behavior, and then move to general
explanations of so-called ―patterns‖ of behavior.

    See also Heil (1992).

       Suppose we want to explain a given instance of behavior, say, someone‘s raising
her arm. Part of the explanation for this person‘s raising her arm might go like this: she
desired to ask a question, she believed that in order to ask a question she had to be
recognized by the speaker, and she also believed that raising her arm would make the
speaker recognize her. Here the explanation of the behavior is in mental terms, such as
―belief‖ and ―desire‖. These beliefs and desires, once functionalized, refer to specific
physical properties.
       Either we know what the physical realizers of the mental properties are in this
particular example, or we do not. First let us consider the case where we do know what
the physical realizers are for each of these mental properties. In this case, it seems that
using the mental terms adds nothing new to our explanation. The causal story is already
given by the laws of the physical realizers. No new information is added by saying that
those realizers are referred to by certain mental terms; the mental terms are eliminable.
       Next, let us consider the case where we do not know what the physical realizers
are. This case might seem like a case where mental terms play a role. But, if we do not
know what the physical realizers are, we do not know to what, specifically, the mental
terms refer. We know what second-order properties our behaving system has, but
explanation of a single case requires knowing what the first-order property is as well.
Consider Kim‘s brief example of a medical condition (pp. 107-108). Suppose that a
certain medical symptom, such as joint pain, could be caused by two very different
diseases, such as lupus and rheumatoid arthritis. Suppose also that Mary has painful
joints, and that tests indicate that she either has lupus or rheumatoid arthritis. Do we have
an explanation for Mary‘s experiencing joint pain in this case?
       It would appear that we do not. We have two candidate explanations in this case,
and we do not yet know which one is the true explanation. There is not a single
explanation here that uses the disjunctive property ―lupus or arthritis‖. In the particular
case of Mary, the joint pain is either caused by her having lupus, or it is caused by her
having arthritis. And we do not have the full explanation of her joint pain until we know
which she has. Likewise with the explanation of a behavior: if we know only that it was
caused by physical property P1, or P2, or P3, we do not know enough to constitute a full
explanation of the behavior. We only have a set of possible explanations. Thus, without

knowing what physical properties are identical to our mental properties in this case, we
do not have a full explanation yet. A full explanation seems to require knowing what the
physical realizers are. But then, again, the mental terms are eliminable.
       So mental terms are eliminable as far as explanations of singular instances of
behavior are concerned. But could mental terms feature in explanations not of singular
instances of behavior, but for certain classes of behavior? Some authors (Fodor 1975;
Pylyshyn 1984) argue that mental terms feature in explanations of ―patterns‖ of behavior.
Consider Pylyshyn‘s (1984) example of the pedestrian who has just witnessed an
automobile accident. The pedestrian rushes to a nearby phone booth and dials a 9 and a
1. What will this person do next? Most of us would guess that she would dial another 1,
with overwhelming likelihood. Why? Because of a systematic generalization holding
between seeing those kinds of events, certain kinds of background knowledge, the
resulting intentions, and that action.
       Pylyshyn further holds that we will not discover this generalization if we focus on
the neurophysiology of the pedestrian. A given physiological explanation only links one
way of learning the emergency phone number to one way of coming to know that an
emergency occurred to one sequence of neural events and resulting muscular contractions
producing the behavior (nonintentionally described). A description at the level of
psychology, on the other hand, will capture this generalization despite the variety of ways
in which we can learn the emergency phone number and execute the movements that
compose the behavior of dialing a 1.
       Notice, however, that capturing the generalization is not the same as explaning it.
After all, the sentence ―all bachelors are unmarried males‖ captures a generalization as
well, but it is not an explanation of why there are such unmarried males. It is a definition
of what it is to be a bachelor. If we were to ask for an explanation of why there were
unmarried males, it would be a kind of joke to say ―because there are bachelors‖.
       Recall that the ―unity‖ of mental terms is conceptual unity, not the unity of an
underlying property. Even if we find what causal powers are had in common by all the
realizers of some mental property, this does not mean we have found a unique property
that figures in explanations. If the behaviors that we are interested in explaining are
simply one of the sets of behaviors given as an effect in our specification D, then it

follows that mental property F causes the behavior by virtue of the definition of F. But if
some mental property F causes a given behavior by definition, we do not have an
explanation for the behavior—or if we do, it is a trivial one.
        It would appear, then, that mental terms are either eliminable in a science of
psychology, or else trivial. In fact, the elimination of mental terms does occur if we look
at the history of psychology. Consider the following example from Kornblith (cited in
Pereboom 2002):

                In 1869, the term 'neuraesthenia' was introduced to designate a
        nervous disease that results in severe fatigue – a characterization that is at
        least fairly functional. The term was soon established worldwide, but "like
        most descriptive terms, where basic organic or psychological
        understanding was lacking, it tended to be overinclusive and a receptacle
        for many diverse conditions."6
                But when cures for neuraesthenia were sought, it was found that
        different sorts of causes had to be treated. Several distinct sorts of
        underlying causes were discovered for the dispositional features of this
        purported natural kind. As a result, the term 'neuraesthenia' became
        obsolete by about 1930.

        In the 19th century ‗neuraesthenia‘ was a mental term thought to refer to some one
condition: a particular nervous disease. This disease was characterized at least partially
in terms of its effects—it had a specification that included the ability to cause sever
fatigue. In each case, however, the term referred to a different underlying physical
property. As our science developed, the term was eventually abandoned for the more
detailed terms for the underlying physical properties. Although the term ‗neuraesthenia‘
grouped together many properties that were superficially similar, simply having a
dispositional term was neither necessary nor even useful for the science that studied and
treated this ―disease‖.

  L. Diamond, "Neuraesthenia," in Benjamin Wolman, ed., International Encyclopaedia of Psychiatry,
Psychology, Psychoanalysis, and Neurology VIII (New York: Aesculapius, 1977), pp. 27-8.

        Does the same thing happen with mental terms such as ―belief‖, ―desire‖, ―hope‖,
and the like? If there really are a number of unique physical realizers for the mental
properties referred to by these terms, it would seem so. Perhaps it is possible that such
terms do feature in generalizations in a way that is not trivial—that is, in a way that does
not depend solely on the definition of these terms. But I do not know how this would
        Thus Kim‘s particular view, reduction by functionalization, has some virtues, but
also has a consequence. If we endorse Kim‘s view, then mental terms are either
eliminable or else feature trivially in explanations. But then there are no psychological
explanations per se. The price of saving mental phenomena and making it part of the
physical world appears to be a kind of eliminativism with regard to mental terms.

Works Cited

Armstrong, D.M. (1968) A Materialist Theory of the Mind. London, Routledge. Second
Edition with new preface 1993.

Bennett, Karen (2003) "Why the Exclusion Problem Seems Intractable, and How, Just
       Maybe, to Tract It." Nous 37: 471-97.

Diamond, L. (1977). Neuraesthenia. in Benjamin Wolman, ed., International
      Encyclopaedia of Psychiatry, Psychology, Psychoanalysis, and Neurology VIII.
      New York: Aesculapius, 1977, pp. 27-8.

Fodor (1987). Psychosemantics. Cambridge, MA: MIT press.

Heil, John (1992) The Nature of True Minds. Cambridge: Cambridge University Press

Kim, Jaegwon (1998). Mind in a Physical World. Cambridge, MA: MIT press.

Kim, Jaegwon (1993). The Non-reductivist‘s Troubles with Mental Causation. In
       Supervenience and Mind. Cambridge, MA: MIT press.

Lewis, David (1966) ‗An Argument for the Identity Theory‘, Journal of Philosophy, 63,

Lewis, David (1972) Psychophysical and Theoretical Identifications. Australasian
       Journal of Philosophy, 50, 249-258.

Lewis, David (1980) Mad Pain and Martian Pain. In Lewis D. Philosophical Papers, Vol.
       1, Oxford, Oxford University Press

Pereboom, Derk (2002) Robust Non-reductive Materialism. Journal of Philosophy 99: 499-

Pylyshyn, Zenon (1984). Computation and Cognition. Cambridge, MA: MIT Press

Shoemaker, Sydney (1980) Causality and Properties. In van Inwagen (1980): 109-36.
      Reprinted in Shoemaker (1984): 206-33

Yablo, Stephen (1992) "Mental Causation." Philosophical Review 101: 245-80.


To top