Behaviorism and Identity Theory

By Susan J. Fleck.  December, 1992.

Essay Questions: Basis of Essay.

1.Philosophical Behaviorism has a fatal flaw; but it is also a theory that offers considerable insight.  Explain in detail what's good about Behaviorism, what insights it offers, and what factors motivated the theory.

2.Explain both the metaphysical and the empirical natures of modern materialism (the identity theory).  What advantages does the theory offer over the other theories?  What disadvantages do you think might accrue to it?

Bibliography note:  Most of the information for this paper came directly from lectures for a course taken in the Fall, 1992:  Philosophy 440: Philosophy of Mind.  California State University, Fullerton.  There are some references to one book (1,, pg.):  Edited by C.V. Borst: The Mind-Brain Identity Theory; Macmillan St. Martin's Press; London, 1970.

There are two main twentieth century, monist theories dealing with the philosophy of the mind in answer to the Cartesian Dualism problem - Behaviorism and Modern Materialism, or the Identity Theory.   Both of these theories have the advantage of bringing the mind from the 'realm of the hidden'.   This paper will explore what factors motivated these theories, their advantages in solving the mind / body dichotomy, and the problems with these views.  Since Modern Materialism was initially a response to the major flaw with Behaviorism, we will start with the earlier developed theory.

Background History

If we take a look at the history leading up to and fostering the idea of Behaviorism, then we can have a better understanding of it.  This will be a very short lesson in history relating to the philosophy of the nature of the mind from 1840 until around 1930.   August Comte (1840), known as the father of modern Sociology, is significant because of his attitude.  Until then, people tended to just theorize instead of gather empirical data.  He realized that in order to move social science up to the level of physical science, you must gather data first, and then make statements.  This view is Positivism.  By 1870 science had started to boom.  But you needed a lot of money for experiments and to support teaching science in universities.  Ernst Mach, physicist, borrowed ideas from Comte to show why the old way of theorizing was 'dead wrong' - why you must have empirical data to have a fundamental distinction between metaphysical speculation (philosophy) and scientific inquiry.  He wrote a philosophy of science to justify why money was needed.

The first developed theory of mind after Hume was from Thomas Huxley (1895), a theory called Epiphenomenalism.  "Consciousness is to be thought of as a mere byproduct of the operation of the brain."  It 'comes out' like 'smoke from a factory'.  Consciousness then is an emergent property, with no causal relations to the workings of the brain (the 'machinery').  This was the first seemingly successful way of getting over Cartesian Dualism, in that  (1) you do not have to be committed to any weird metaphysics, like monadology;  (2) there is no problem with causation - mental states are not actively involved in causal relationships with the body (mental states are not brain states);   and   (3) the mind is only passive - it does not make the brain do anything,  it is only conscious (aware) of what the brain does.  Consciousness is a substance that belongs to the brain, with emergent property.  Consciousness for the Epiphenomenalist is the product of an increasingly complex 'system';  e.g. a horse has consciousness, but maybe not a worm.  I.e. as creatures became more complex, at some point 'awareness' came along.

Then along came Einstein's 'awareness' of relativity (1905).  His theory caused a crisis in philosophy when he showed that space and time were not anything like what Newton thought, but were more like what Leibniz had thought.  Leibniz had argued that time and space are not real - they were idealized entities (time - out of change and motion) - they are relational.  Einstein forever changed the approach to the study of space and time.  Russell further demonstrated that metaphysics was a waste of time, but showed how symbolic logic could solve problems - how conceptual analysis will tell you what you mean when you say something.

This provided a basis for the Vienna Circle (1918 - 1927), a group of very famous people which met in Vienna.  Their purpose was to provide a manifesto for philosophy and science that would rule out as nonsense metaphysical and religious inquiry.  This evolved and they became known as 'Logical Positivists' as embodied in Rudolph Carnap's book Unity of Science.  Their fundamental weapon was the 'Verification Principle' which is a doctrine of meaningfulness:   A proposition is meaningful if it can be empirically confirmed or dis-confirmed.   Hume had said that for something to be a real idea, it had to have its origins in experience.  His attacks on the self and substance are the same kinds of attacks that this group carried out on e.g. God, goodness, badness,  the 'absolute'.  They tried to rid philosophy from all of those 'suspect' entities that had crept in over the years.

The Behavior Answer

Knowing this background, then, we can see how theories of Behaviorism came about when they did.  The first view of this was in 1914 by John Watson and it is unknown what he said, but by 1920 the theory had been absorbed and transformed by Gilbert Ryle.  Take the example:  How would you know what is the meaning of "I am in pain"?  The verificationists would ask someone about their pain and would observe them.  For them, the meaning of ascriptions of mentality reside under the conditions for which you confirm or dis-confirm.  The obvious candidate for this is behavior.  Bob is in pain must mean that Bob is 'pain-behaving'.  The central thesis of this theory is that we can give an account of the nature of all mental states and processes in terms of the physical behavior and tendencies or dispositions to behave of human beings.

Gilbert Ryle attacked Descartes' conception of the mind as a spiritual object.  According to D. M. Armstrong - for Ryle, the mind was not something behind the behavior of the body, it was simply part of that physical behavior.  For example, my anger with you is not some modification of a spiritual substance which somehow brings about aggressive behavior; rather it is the aggressive behavior itself - the cussing, the hitting, etc.  Thought is not an inner process that lies behind, and brings about, the words I speak and write - it is my speaking and writing.  "The mind is not an inner arena, it is outward act" (1, 70).

It is not necessarily true that a Behaviorist  must also be a 'logical positivist', although many are.  C.V. Borst points out a close parallel to Logical Phenomenalism, where physical objects can be analyzed into a set of statements about sense-impressions.  In both cases the logical tool involved can be seen as Bertrand Russel's notion of the logical construction.  - "that A's are logical constructions of B's if and only if all statements about A's can be replaced, without loss of meaning, by concatenations of statements about B's."  Minds can become logical constructions out of behavior (physical out of sense-impressions), somewhat as nations might be logically constructed out of individual citizens.  "But in no case was any real entity to be postulated over and above the constituent entities"  (1, 15).

Defense of Behaviorism Theory

There are a lot of 'pluses' with this theory.  (1) It is compatible with scientific knowledge of humans and brains.  (2) It accounts for the close unity we think there is between mind and body.  It is somewhat intuitive to say 'The mind is just the body in action'.  When we talk about how a person thinks, or believes, we are talking about how a human behaves.  (3) It is a monistic theory which eliminates mental substance, which makes it an O.K. theory for materialists.  (4) The causal role for mentality is eliminated  (this will be discussed later - as part of the major flaw with this theory).   (5) This provides a positive clinical role for psychologists.  For example, depression to a behaviorist is just a set of dispositions, and a set of learned responses (patterns of behavior).  You just need to alter your behavior, e.g. go dancing, to eliminate the depression (which can not be a mere feeling - a mental quality that Descartes might have ascribed).  'Mental' illness is empirical and treatable.

Behaviorism is the first theory of mind that removes the mind from the 'realm of the hidden'.  An account and analysis of a mental concept, e.g. anger or love, can be construed without any reference to a mental state or without much introspection.  Mental concepts have a set of dispositions, a set of learned patterns or conditions.  One remnant of Cartesian Dualism is the 'privileged access' doctrine which is maintained with Behaviorism.  This doctrine says that I experience my own mental states from a privileged position, and others can experience my mental state only indirectly.  This privileged access is only with respect to dispositions.  E.g. a student has no idea how Bob would react if he is offered money for an 'A' (even though we may theorize he would immediately go to the racetrack based on past behaviors - just kidding of course!), -- but Bob knows his reaction from self knowledge of disposition and beliefs.  Feeling is antecedent to the act.

Although Behaviorism is empirical, Behavioralists recognize that different people have different reactions from each other; that there is no behavior common to all cases.  But they deny a need for commonality.  A Cartesian says that feeling is common to behavior, but a Behavioralist does not recognize feelings.  There are certain resemblances among these behaviors when deciphering a mental concept such as anger.   So we have ended up with dual vocabularies; we do not talk about brain states, we talk about feelings.   Another empirical idea is that with Behavioralists there are no reports of inner processes.  The statement "I have a headache" is not a report but is instead a piece of behavior.  Behaviorism is reductionist of sorts, where the entity is eliminated.  But it is impossible to provide translations of statements of mental states to statements of bodily behavior and disposition.  This is especially obvious in the example where one is deducing what the next number is in a series e.g. 1,3,6,10...  Behavioralists do not want to recognize this mental process of 'illumination'.

Behaviorism is clearly a monist theory that does not believe in mental substance.  But it is a positive view not only because it takes the mind out of the realm of the hidden, but because it treats mental concepts as if they do more than just 'be a mental state'.  The concepts are more complex. For example, the 'Other Minds Problem' - How can I ever know there are other minds in the world?  I observe in certain cases that there are certain 'associations of behavior' to certain behavior and then I build a bridge to certain mental states through inference (causal relationship).  I.e. we can know there are other minds by observing behavior.

Major Flaw With Behaviorism

But there is a major flaw with Behaviorism which causes it to fail as an answer to our Mind / Body dichotomy.  It takes the behavior occurant state out of the picture.  I.e. we can not say 'he got a drink because he was thirsty."  Armstrong points out how Behaviorism is a "profoundly unnatural account of mental processes."    It is more natural to think of someone's speech and action as the expression of his thought, not as identical with his thought.  We think of thought as something distinct from the actions - it may lie behind the behavior and bring it about, but it is not to be identified with it (1, 72). 

Modern Materialism: Identity Theory

Therefore we need to search further for the solution.  We need a theory of mind which - (A) will not be dualistic, because you cannot maintain causal interaction between two distinct things; and (B) will have to agree with Behaviorism in that you can not keep mentality hidden.   Modern Materialism, or the Identity Theory of Mind, is the logical response to the flaw in Behaviorism.  The central thesis of this theory is that the Mind and the Brain are identical.  Mental states and processes are the same as brain processes - this is 'one step ontology'.  This removes the flaw in Behaviorism because it enables you to give a causal explanatory role to mental states.  It recognizes the logical tie between mental states and behavior and allows you to say "I drank because I was thirsty."  Now that we have discovered it advantage, let us examine this theory a little more closely.

In the 1950s U.T. Place and J.J.C. Smart advanced this theory.  They felt that self-knowledge could be understood in terms of physical processes and that everything in the universe should be explained in terms of physical except sensations. Smart says that the mental should not be any different than other things in the world.  Physics should be able to explain what thirst is, what causes images of the house you grew up in, etc.  By looking at the set of laws that govern everything else in the universe, they wanted to bring the occurrence of experiences into the web of physical theory.  So Materialism came about to solve the flaw with Cartesian Dualism by moving away from the two-substance world view.

So what is this theory?  This theory was put forward as a metaphysical empirical hypothesis.  Just as Einstein solved the physical dispute about space and time, this hypothesis says that we will eventually understand so much about the brain:  It says  "the mind is the brain, and we will prove it."   Brain research scientists will eventually put the philosophers of mind out of business!   These philosophers who subscribe to this theory feel that brain science will eventually get to the point where it no longer makes sense to talk about a non-physical substance (like the mind).  This solves Cartesian Dualism in that it is all physical. 

The originators of the theory say you may speculate about the physical constitution of a particular substance, for example, lightening.  You discover it is identical to electrical discharge.  They figure we will eventually find the 'identity' in the brain;  there is nothing necessary about the mind being the same as the brain - it just is; it is contingent - a one-way reductive identity.  We have two different ways of talking about something.  In reality there is only one set of entities.  The idea here is that the identity is reductive.  What can be said about brain states, can also be said about mind states.  E.g. "I am thirsty" can be replaced by 'mind state 504'.  We have a vocabulary that is inclusive about our brain states.  We say 'I am thirsty', we do not say 'I have mind state 504'.  But eventually we will discover how to identify and make these connections about what is going on in our brain.

Rorty: Eliminative Materialist

Some have given this theory a bad name.  For example Rorty is very radical and is known as an Eliminative Materialist.  He says that if this theory is true then we should stop talking about the mind and only talk about the brain.  Sensations and sensation-talk would disappear from a scientifically oriented language.  But it is very difficult to stop talking about the 'mental' - it is hard to take the mental out of the 'hidden'.  As Berkeley pointed out, this will probably not happen in practice and we would be permitted to 'speak with the vulgar' so long as we were to 'think with the learned' (1, 21).  The more 'normal' Materialist would not give up our way of talking, but just realize that we are getting the mental out of the hidden - i.e. there is no important distinction between the 'inner' and the 'outer'.

Materialists admit that there are grounds to dispute the theory, and that while we should be able to produce neural maps that will tell us everything that is going on in the brain (i.e. to map the brain), that we still may not be able to understand how Bob's thoughts of Cleveland function on a physio-chemical level.  Not only does it sound unromantic to discover how e.g. music is composed, but it may not be 'reducible' to be able to determine the mechanics of how this is done in the brain.  According to Borst, it is generally acknowledged that it is practically impossible to settle the dispute between the Identity Theory and the rival theories of Epiphenomenalism and Psycho-physical Parallelism (1, 28).  But since the latter doctrines involve the view that none of our mental states can be causally effective on behavior, the Identity Theory has greater strength because it allows causal efficacy to mental occurrences.   So while there is plenty of evidence, in its favor, we are not yet in a position to prove the Identity Theory.

Objections To Identity Theory

Most of the objections to this theory are within three main categories.  One is related to Leibniz's Law of Identity.  This law basically says that if two objects are identical, then they have all of the same properties.  So if you could show something 'true' about brain events that are not 'true' of mental events, then you could prove this theory false.  In 1962 Shaeffer pointed out that the brain is in space, hence the mind is not; so how can they be identical?  This has been overcome by pointing out that to say the mind is the brain is to speak very loosely.  We are not saying there is an 'entity' of the mind.  But rather that e.g. Mental Event (1) equals Brain State (1).  States do not have physical location.  Processes and events do not have spacial location.  The really dumb part of this objection, is that the mind and brain are not two different entities.  Borst points out that these objections "have been based on the failure to appreciate the possibility of contingent, as opposed to logical, identities" (1, 25).

A second type of objection states, simply, that there is no such thing as a contingent, or reductive identity - all identities are a necessary truth.  Saul Kruepke (spelling?) wrote a paper "Naming a Necessity" which is a complex theory about semantics and language.  According to this view, since there is nothing 'necessary' about brain states equaling mind states, then the theory must be wrong.  When you attach a name to something, then it must be an identity.  This is a very rough statement of his view of the 'causal theory of names'.  Since we 'named' both the brain and the mind, they must therefore be two different identities.

The third type of objection has to do with observances that even though seemingly crucial portions of peoples' brains have been destroyed, e.g. due to accidents, yet none of the mental functions are disturbed.   Therefore, the Materialists speculated that 'neural maps' may be quite different for each person.  To overcome this problem they first devised a 'token, token Identity Theory' which said that specific mental states are identical to a wide variety of brain states.  But with this view,  we can not have a general science of the brain about which mental states are identical to which brain states.  So they came up with a 'type, type Mental States':  For a given type of mental state will always be identical to a given type of physical event.  That way, even if some of your brain is destroyed, all that is generalizable is that memory functions the same between individuals  (i.e. the same kinds of functions; the same kinds of chemicals, etc.). 

The big problem with this view is that it is very anthropocentic.  Borst gives this example:  Suppose we found creatures on some other planet that displayed behavior such that we would say that they had beliefs, preferences, etc.  But then we discovered that their brains, their 'central processes' were very different from ours.  Then, "if the identity of  the Identity Theory is to be taken seriously, then what belief, say, is depends on the actual physical nature of the underlying state" (1, 23).  We would be forced to ascribe a different psychological vocabulary to these creatures.   Instead of trying to force some arbitrary 'mental constructs' to the mental / brain problem, why is it not all right to just accept that the brain can be taken over by other portions of the brain?  I.e. that it is all part of the brain science, the study of how the brain actually works.

Conclusion?

So, are we getting closer to the solution to the Cartesian Dualism problem?  Although there seems to be something to the concept that if you change your behaviour you can change your 'attitude' (e.g. depression).  But what comes first, the thinking about changing your behavior -for rational results, or the actual behavior (how or why? on a whim?)?   I agree with Armstrong in that one's speech and actions are expressions of one's thoughts - not identical with the thoughts.  The Identity Theory seems much more intuitive to me.  I think part of the problem with this is semantics of 'what do we mean by' things such as - soul, psyche, mind, one's essence.  This, like most of philosophy's questions, points back to the fundamentals of universals and forms.  If you are primarily  a Platonist and have a dual world view which believes that the mind/soul is separate from the body - that the soul is good and seeks knowledge of the 'real' world of the forms (the 'real' world of god and heaven), and the body is evil and seeks pleasures of the physical world; then you may lean more towards the Behaviorists' view, or be stuck in an impossible dualistic view.   However, if you are an Aristotelian with a more cohesive one-world viewpoint, where  the terms mentioned above relating to a human would all be wrapped up in one 'real' human entity; then there is no dichotomy or need to figure out how separate entities interrelate, because they are not separate.  The Identity Theory would make the most sense to one who leans toward Aristotelian world views (which includes this author).  It will be interesting to watch the development of the sciences studying the brain, as well as the continuing arguments in philosophy about the nature of mind and thought.

*** end