"Functionalism is no better than behaviourism as a theory of the mental."  Discuss. 


The quote in the essay title complains that as a theory of the mental, functionalism is no better than behaviorism.  The suggestion is that functionalism is just as deficient in some measure as behaviorism, and thus no reason to prefer functionalism over behaviorism.  One way to address this complaint would be to examine all the claims and criticisms (of some version) of each theory and evaluate whether functionalism scores better overall as a theory of the mental.  That, however, would take far too long for an essay of this sort.  So I will arbitrarily choose one particular area of contrast between the two theories, and presume that the essay's title complaint is that neither functionalism nor behaviourism provides an adequate explanation for the feel of subjective experience -- the "qualia" of consciousness.  I will conclude that in this, however, the title quote is wrong.  Functionalism, unlike behaviorism, can provide an adequate explanation for the subjective experience of the mental.


Broadly speaking, behaviorism is a psychological doctrine that demands behavioral evidence for any psychological theory.  In 1963, Sellars commented that someone qualifies as a behaviorist if they maintain "hypotheses about psychological events in terms of behavioral criteria"(2).  Such a doctrine holds that there is no knowable difference between states of mind unless there is a demonstrable difference in the behavior associated with each state.

The historical basis of behaviorism is the philosophical movement of Logical Positivism(3) dominant in the early third of the 20th Century.  Logical positivism argued that the meaning of statements used in science should be understood in empirical terms -- in terms of observations that verify their truth(4).  Behaviorism adapted this philosophy to claim that mental concepts must refer to empirically observable behavioral tendencies, and so can (and should) be translated into behavioral terms. 

Over the years, however, behaviorism has split into three rough families of argument.  "Methodological" behaviorism maintains that psychology is the science of behavior, not the science of mind, and thus does not even pretend to offer a "theory of the mental".  "Psychological" behaviorism (also known as "radical" behaviorism(5)) maintains that the causes and explanation of behavior are to be found in the external environment, and not in internal mental phenomenon.  Behavior thus can be understood without reference to mental events or to mental processes.  As a "theory of the mental," therefore, it treats the mind as a unit black box, knowable only via its externally observable behavior.  Such mental concepts as are entertained are treated as "theoretical fictions".  (A "theoretical fiction" is a computational or imagery device proposed by a theory with no pretense that the device is real.  An instrumentalist in the philosophy of science, for example, will regard the notions of electrons and quarks as theoretical fictions.)  "Analytical" behaviorism (also known as "philosophical" or "logical" behaviorism), on the other hand, maintains that all mental concepts can be translated into behavioral concepts, and that terms for mental concepts can and should be replaced by terms for behavioral concepts.  As a "theory of the mental", therefore, it also treats the mind as a unit black box, but does not permit mental concepts even as "theoretical fictions".  Analytic Behaviorism was the "orthodox" theory of the mental from about the 1930's to the dawn of computational cognitive science in the 1950's(6).  Given the context of this essay, I will hereafter assume that the behaviorism referenced in the essay title is intended to refer to Analytical Behaviorism, and disregard the others.

Analytic Behaviorism, as a theory of the mental, has no place for a representation of the environment as a determinant of behavior.  To many critics of behaviorism, the fact that the environment and one's learning history is represented internally by the subject seems to be more of a behavioral determinant than the subject's reinforcement history.  How a subject sees (classifies, interprets, represents) a stimulus appears more germane to the behavior elicited, than just the bare stimulus history.  Critics sometimes argue this point through the concept of "qualia".  Some experiences, it is argued, have characteristic "qualia" or presentationally immediate phenomenal qualities of experience.  Being in pain, for example, does not just involve the appropriate behavioral dispositions, it also involves (so the argument goes) a particular sort of "what it is like"(7) to experience pain.  Behaviorism must deny the reality of qualia, and the possibility of internal representations that can modify behavioral dispositions.  A philosophical zombie(8) is a hypothetical being that is indistinguishable from a normal human being except in that it lacks conscious experience, or qualia.  To analytic behaviorism, since there is no place for internal representations and denies the existence of quails, we are all philosophical zombies.  This seems to be wrong.

If mental states are identical with behavioural dispositions, as behaviourism holds, then any two people with the same behavioural dispositions will be in the same mental states.  There do seem, however, to be circumstances in where two different mental states can be associated with identical behavioural dispositions, or the same mental state with two different behavioral dispositions.  Behaviourism cannot account for this.  Behaviorism appears unable to distinguish between real mental states and pretend mental states, since both arguably involve precisely the same behavioural dispositions.  If the mental state of being in pain is defined in terms of the behavior we would normally associate with such a state, then how to characterize a person who exhibits that behavior without being in pain?  Behaviorism would seem to deny this possibility.  So an actor who displays all the associated behavior of pain, would be considered to actually be in pain.  Pretense is impossible.  The reverse scenario involves Super-Spartans(9) - hypothetical beings with mental states identical to our own, but who lack normal behavioural dispositions.  Stab a Super-Spartan, and though he will feel pain, he will have no disposition to exhibit normal pain behavior.  Behaviourism must say that Super-Spartans are also impossible.  This seems to be just as wrong.


Behaviorism views the mind as a unit black box, treating only the inputs (stimuli) and outputs (behaviors), and regarding mental states as operators translating inputs into outputs.  Functionalism, on the other hand, builds a theory of the internal workings of the black box, positing functional roles (mental states) that mediate the translation of inputs (stimuli) to outputs (behavior).  Functionalism, as a philosophy of mind, is the doctrine that mental states are defined by the functional roles they play in the complex economy of causal interactions within the network of roles of which they are a part.  Behaviorism takes the identity of a mental state to be determined by its relations to the organism's inputs (stimuli) and outputs (behavior).  Functionalism takes the identity of a mental state to be determined by the organism's inputs (stimuli) and outputs (behavior), but also by its causal relations to other mental states. 

Thus functionalism solves one of the problems that plagues behaviorism.  For behaviorism, a belief that it is raining has to be defined in terms of behavioral dispositions to stay out of the rain or carry an umbrella, and so forth.  But this approach runs into the difficulty that my disposition to stay out of the rain will not manifest itself if I also have the desire to go "singing in the rain".  Behaviorally defined mental states thus seem to require an infinite regress of other mental states.  Critics argue that the holism that this regress requires can never be satisfied -- that one cannot get away from mental states being a part of the determinants of behavior.  Functionalism, however, embraces this holism, defining any one mental state by the role that it plays in the larger functional network.

Functionalism, as a theory of the mind in contemporary philosophy, developed largely as an alternative to both the type-identity theories of mind and behaviourism.(11)  Functionalism is a theoretical level between the physical implementation and behavioral output.(12)  Functionalism thus plays its role at Dennett's "Design Level".(13)  Functional roles are defined independently of any physical structure realizing the role.  Thus functionalism solves the multiple realizability problem plaguing the type-identity theories.  Functional roles are also defined independently of the higher level of Dennett's "Intentional Level"(ibid). Thus functionalism seeks to explain and understand intentional concepts by positing a sub-intentional network of roles that realize the intentional concepts.

Over the years functionalism, like behaviorism, has divided into a number of different varieties.  "Machine-state Functionalism" was first proposed by Hilary Putnam in the 1960's.(14)  It was inspired by the work of the early computer scientists -- most notably Alan Turing.(15)  Any system that possesses a mental life is simply a complex Turing Machine instantiating a certain machine table.  Each mental state (a thought, e.g.) is actually a machine state that arises in the course of that program.  "Psycho-functionalism," most notably associated with Jerry Fodor,(16) views psychology as employing the same sorts of irreducibly teleological or purposive explanations as the biological sciences.  Psychological states have the same kind of functional and teleological roles as do biological organs.  Psycho-functionalism can be characterized as "a posteriori" functionalism or "scientific" functionalism.  We will learn that functionalism is true only as a result of empirical investigation into the workings of the mind.  "Analytic Functionalism," most closely associated with David Lewis(17) is, like analytic behaviorism, mostly concerned with the meaning of mentalistic terms.  Mental terms are defined by the theories of the mental in which they occur, and not by any extrinsic features of the world.  Such functional definitions, therefore, are claimed to be analytic or a priori, rather than empirical.  The central idea of analytical functionalism is that the essential functional roles of the mental are to be discovered by analysis of common sense knowledge of mental phenomena - folk theories of the mental.  Daniel Dennett(18) and William Lycan(19) are more recent advocates of this branch of functionalism.  As with behaviorism, I will assume that it is "Analytic Functionalism" that is the intended target of the essay's title quote, and will ignore the other varieties.

The Analytic Functionalism view of "qualia" and the presentationally immediate phenomenal qualities of experience is that they have acceptably defined functional natures.  Unlike behaviorism, functionalism does not have to deny the reality of qualia (although some functionalists do(20)).  How a subject sees (classifies, interprets, represents) a stimulus simply is a particular functional role.  The phenomenal character of, say, pain or seeing red or remembering where the tiger was last reported, is one and the same as (is type-identical to) the functional role that plays an essential part in mediating between physical inputs (body damage, surface reflectance properties, verbal reports from others) and physical outputs (yelling "Ouch!", picking the ripe apple, going left rather than right).(21)  All the other mental states that might be involved (for a Super-Spartan say, a desire not to let you know I am in pain) are just other functional roles connected in suitably causal ways in the holistic network that is the conscious mind. 

There are two famous objections to functionalist theories of qualia: the Inverted Spectrum Argument and the Absent Qualia Hypothesis.  The inverted spectrum argument is based on the apparent possibility of two people sharing their color vocabulary and discriminations, although the colors one sees (their qualia) are systematically different from the colors the other person sees.(22)  Necessary to this argument is the stipulation that there is no discernible difference in the behavioral dispositions of the two subjects.  Thus, the argument proceeds, the two are in different phenomenal states with the same behavioral dispositions.  Hence, the qualia of sensory experience cannot be captured in functional roles.  But functionalists can argue in response that this scenario is either not conceptually possible or not metaphysically possible.(23)  It is open to the functionalist to reply that there would necessarily have to be some salient fine-grained functional differences between the two subjects, notwithstanding any admitted larger-scale functional identity between the two.  Seeing red, or green, just is a particular functional role.  If the two subjects differ in what they see, then necessarily they are exercising different functional roles.

The absent qualia hypothesis is the hypothesis that it is possible for there to be functional duplicates of conscious subjects that entirely lack qualia.  Two examples of this hypothesis dominate the literature.  There is the Zombie scenario(24) analyzed in detail by Chalmers(25).  And there is the China Brain suggested by Block(26).  The zombie scenario can be dismissed by the functionalist as both logically incoherent and metaphysically impossible(27).  The China Brain scenario can, however, be accepted by the functionalist.  Contrary to block's argument, and however strange it might seem, the China Brain does experience qualia and have beliefs.  That supposedly counter-intuitive response is explained by the disparity in size -- the individual Chinese citizen participating in the experiment need have no more awareness of the consciousness and qualia of experience of the China Brain than does any individual neuron in the human brain(28).


In other words, functionalism, unlike behaviorism, does provide a better explanation for the subjective experience of the mental.  In the view of functionalism, our minds are an anarchic network of interacting functions.  There is no Official Understander to give functional roles meaning, to make decisions, and to understand the results.  Consciousness, and the feel of subjective experience, is the result of the interaction of all those differing functions.  Although it may not be the best possible theory (there are lots of criticisms of functionalism in the literature that I have not touched on), it is clearly better than behaviorism as a theory of the mental.


Notes & References

(1)  Graham, George;  "Behaviorism" in The Stanford Encyclopedia of Philosophy (Fall 2010 Edition), Edward N. Zalta (ed.), URL=<http://plato.stanford.edu/archives/fall2010/entries/behaviorism/>.

(2)  Sellars, Wilfred;  "Philosophy and the Scientic Image of Man" in Science, Perception, and Reality, Routledge & Kegan Paul, New York, New York, 1963, pg 22.

(3)  Smith, L.;  Behaviorism and Logical Positivism: A Reassessment of Their Alliance, Stanford University Press, Stanford, California, 1986.]

(4)  Wikipedia contributors;  "Logical Positivism" in Wikipedia, The Free Encyclopedia, URL=<http://en.wikipedia.org/w/index.php?title=Logical_positivism&oldid=526844476>.

(5)  Rey, George;  Contemporary Philosophy of Mind: A Contentiously Classical Approach, Wiley-Blackwell, Oxford, England. 1997.

(6)  Bechtel, W. & Graham, G. (eds.);  A Companion to Cognitive Science, Wiley-Blackwell, Oxford, England, 1998.

(7)  Nagel, Thomas;  "What it is Like to be a Bat" in The Philosophical Review, Vol. 83, No. 4 (Oct., 1974), pp. 435-450. URL=<http://www.jstor.org/stable/2183914>.

(8)  Wikipedia contributors;  "Philosophical Zombie" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Philosophical_zombie&oldid=525886965>.

(9)  Putnam, Hilary;  Mind, Language, and Reality. Cambridge University Press, Cambridge, England, 1975.

(10)  Levin, Janet;  "Functionalism" in The Stanford Encyclopedia of Philosophy (Summer 2010 Edition), Edward N. Zalta (ed.), URL=<http://plato.stanford.edu/archives/sum2010/entries/functionalism/>.

(11)  Wikipedia contributors;  "Functionalism (Philosophy of Mind)" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Functionalism_(philosophy_of_mind)&oldid=528239243>.

(12)  Marr, D.;  Vision: A Computational Approach. Freeman & Co., San Francisco, California, 1982.

(13)  Dennett, Daniel C. The Intentional Stance. The MIT Press, Cambridge Massachusetts. 1998. ISBN 0-262-54053-3.

(14)  Putnam, Hilary;  "Minds and Machines" (1960) reprinted in Mind, Language and Reality, Cambridge University Press, Cambridge, Massachusetts, 1975. 

"The Nature of Mental States" (1967) reprinted in Mind, Language and Reality, Cambridge University Press, Cambridge, Massachusetts, 1975.

(15)  Wikipedia contributors;  "Alan Turing" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Alan_Turing&oldid=530445657>.

(16)  Wikipedia contributors;  "Jerry Fodor" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Jerry_Fodor&oldid=529016296>.

(17)  Wikipedia contributors;  "David Lewis (Philosopher)" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=David_Lewis_(philosopher)&oldid=527490288>.

(18)  Wikipedia contributors;  "Daniel Dennett" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Daniel_Dennett&oldid=528400151>.

         Dennett, Daniel;  Sweet Dreams: Philosophical Obstacles to a Science of Consciousness, A Bradford Book, MIT Press, Cambridge, Massachusetts. 2006. ISBN 978-0-262-54191-6.

(19)  Wikipedia contributors;  "William Lycan" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=William_Lycan&oldid=523476592>.]

         Lycan, William G.;  Consciousness and Experience, A Bradford Book, MIT Press, Cambridge, Massachusetts, 1996. ISBN 978-0-262-12197-2.

(20)  Dennett, Daniel;  "Quining Qualia" in Mind and Cognition: A Reader, W.G. Lycan (ed.), MIT Press, Cambridge, Massachusetts, 1990.

(21)  Lycan, William G;  Consciousness, MIT Press, Cambridge, Massachusetts. 1987.

(22)  Wikipedia contributors;  "Inverted Spectrum" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Inverted_spectrum&oldid=436315508>.] 

(23)  Tye, Michael;  Ten Problems of Consciousness, Bradford Books, The MIT Press, Cambridge, Mass. 1995

Consciousness, Color, and Content, Bradford Books, The MIT Press, Cambridge, Mass. 2000. 

Hardin, C.;  Color for Philosophers, Hackett Publishing, Cambridge, Mass. 1993.

(24)  Kirk, Robert;  "Zombies" in The Stanford Encyclopedia of Philosophy (Summer 2012 Edition), Edward N. Zalta (ed.), URL=<http://plato.stanford.edu/archives/sum2012/entries/zombies/>.

(25)  Chalmers, David;  The Character of Consciousness (Philosophy of Mind), Oxford University Press, New York, New York, 2010.  ISBN 978-0-195-31111-2. 

"Does Conceivability Entail Possibility" in Conceivability and Possibility, T.Gendler & J.Hawthorne (Eds.) Oxford University Press, New York, New York, 2002. ISBN 978-0-198-25090-6.

(26)  Wikipedia contributors;  "China Brain" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=China_brain&oldid=521938777>.]

(27)  See my essay M05.docx.

(28)  Tye, Michael;  "Qualia" in The Stanford Encyclopedia of Philosophy (Summer 2009 Edition), Edward N. Zalta (ed.), URL=<http://plato.stanford.edu/archives/sum2009/entries/qualia/>.

            "Absent Qualia and the Mind-Body Problem" in The Philosophical Review, Vol 115 (2006), Pgs 139-168.


[Up] [Home] [Next]