Study Guide

Week Nine: Mind-Brain Identity Theory

1. Materials Assigned for the Week
2. The Central Point of This Week's Material
3. Other Concepts and Points you are Expected To Master This Week
4. Miscellaneous Comments and Clarifications
   1. Mind-Brain Identity and the Success of the Biological Sciences
   2. The Identity Theorist’s "Argument from the Advance of Science"
   3. The Identity Theorist’s "Argument from the Principle of Parsimony"
   4. Is There Any Conflict Between the "Science Argument" and the "Parsimony Argument"?
   5. A Common But Not Recommended Way to Distinguish Between Identity Theory and Functionalism

I. Materials Assigned for the Week

Reading:

Lecture 21: The Mind/Brain Identity Theory
Reading: Smart, Mental processes are physical, pp. 348-352

Exercises:

Review Questions: pg. 297 esp. #5
Problems for Further Thought: pg. 297

II. The Central Point of This Week's Material

Mind-brain identity theory rejects parts of both Behaviourism and Dualism. Against Behaviourism, it argues that mental states are indeed inner causes of behaviour. Against Dualism, it argues that these inner causes are not states of anything especially non-spatial, incorrigible, private or subjective; they are simply states of our physical brain - as spatial, corrigible, public and objective as any other part of our body. To explain the astonishing correlation of specific mental states and specific physical states, materialism replaces the analysis "X causally interacts with Y" (which requires two kinds of things) with the analysis "X is one and the same thing as Y" (which requires only one). Mind-brain identity theory claims that this simpler view is supported by scientific advances in recent neurology and related disciplines.

III. Other Concepts And Arguments You Are Expected To Master This Week

• The difference between materialism in the philosophical sense and materialism in the daily-life sense - which is basically to see that in the philosophy of mind materialism and physicalism mean the same thing
• The difference between a mental property and a physical property- again
• What the Principle of Parsimony is - especially as a principle used in modern science (it was originally a theological principle)
• What the special home for parsimony is in abductive arguments
• The difference between "X is identical to Y" and "X is correlated with Y" (even "X is constantly correlated with Y") - briefly, then, the difference between identity and correlation
• Why the doctrine of vitalism and the theory of dualism are claimed to be the same sort of scientific dodo, at least as far as mind-brain identity theory is concerned

IV. Miscellaneous Comments and Clarifications

IV.A Mind-brain Identity Theory and the Success of the Biological Sciences

The mind-brain identity theory is the first of the genuinely scientific theories we have met - in the sense that it is the first argument which explicitly depends on the results which practising scientists working on mental phenomena have so far achieved. This is a bit strong. Two qualifications are needed:

1. None of the theories of mind we have yet encountered would happily admit to being outright unscientific. For instance, Descartes thought there was nothing particularly unscientific about relying on the procedure of introspection to find out about the existence and character of minds. Indeed, the science of the mental pretty well had to be wholly introspective for Descartes, or else it wouldn't be accessing the right subject matter (that which is private and incorrigible); it is only the science of objective physical objects which should be based on public observation. For his part, Skinner claims that only behaviourism has ever met the strictest canons of scientific method (not to posit unobservable inner mechanisms).
2. while neuroscience is a regular physical-biological science, the theory of mind-brain identity is not exactly neuroscience. Mind-brain identity theorists no more wear laboratory white coats than do dualists or logical behaviourists. Instead, mind-brain identity theory is more a philosopher's "take" on the results of neuroscience for the distinctly and admittedly philosophical purposes of deciding how many kinds of things there really are in the cosmos. The "How many kinds of things?" sort of question is a question which no practising scientist will touch with a barge-pole except out of hours. As we know, though, it's meat and potatoes to any specifically philosophical activity such as 34.101 Knowledge and Reality.

For all that, Sober is basically right when he says:

"The Mind/Brain identity Theory isn't only a different theory from those surveyed so far; it also is a different kind of theory. The identity theory argues for its solution to the mind/body problem by describing the progress that science has made so far and by predicting that science will make in the future. The relationship of the mind and the body, say the identity theories, is something that science discovers by observation and experiment. It isn't something that armchair philosophy - introspection and linguistic analysis - can hope to resolve." (pp. 292-293)

In short, mind-brain identity theory is the first of the theories which pins its success on the success of practising science.

IV.B The Identity Theorist's "Argument from the Advance of Science"

The deed is done using an argument more or less along the following lines. This argument is sometimes called the "Master" argument for materialism - which is of course a bit self-serving, since other versions of materialism don't use it (Functionalism doesn't for instance). Better to emphasise what makes it tick than who it serves.

The Argument from the Advance of Science

• Premise 1: Redefinition of mental states and processes:
  The mind is not an item of the cosmos to which we have a special sort of access. (Specifically, the mind is not to be defined as whatever is conscious.) Rather, the mind is an item in the cosmos which is the cause of a certain range of effects. (Specifically, the mind is to be defined as whatever causes behaviour.)
• Premise 2: "Science knows best" assumption:
  Modern science as practiced in the west is the sole arbiter of what the universe is really made of and what really causes what in the universe. (Don't listen to your phrenologist or psychic or economist or priest or novelist or historian, listen only to your physicist and chemist and molecular biologist and neurologist.)
• Premise 3: Lesson of recent scientific research:
  Scientists have now discovered, or soon will, that what actually causes our behaviour is our brain. (More accurately, they have discovered that what actually causes our behaviour are states and processes and events happening in our nervous system.)

therefore

• Conclusion 4: Materialist conclusion: The only scientifically respectable theory to hold anymore is that the mind = the brain.

That's the long version. Here's the short version:

• Premise 1: The mind = whatever entities turn out to cause our behaviour.
• Premise 2: What we've found actually causes our behaviour is our brain.

therefore

• Conclusion 3: The mind = the brain.

Although this "Science Argument" looks very simple, especially in its short form, there are several important ideas at work here. It is worth the effort, again, to "stretch out" some of these ideas.

Premise 1: The Mind is Whatever Inner Structural State Causes Our Behaviour

The first premise is this: "When we talk about a certain mental state, what we really mean to be talking about is whatever structure in a human being causes us to behave in the way we do." This redefinition of "mind" in terms of the structures in humans which cause humans to do what they do is absolutely vital, against two traditional foes.

On the one hand, it straightaway rules out of court anything like the traditional dualist's mind-set. Remember that Descartes did allow that states of the mind caused states of the body and vice versa; that is, that changes in our mental structure caused changes in our bodily structure. But he never defined the mind in terms of what the mind caused. He defined the mind exclusively in terms of the private, conscious, subjective experiences which only they can have. By the seemingly innocuous redefinition offered up in Premise 1, and by that redefinition alone, the identity theory tidily shoves all consideration of such subjective conscious experience into the back seat … way way back into the back seat.

On the other hand, this redefinition also straightaway rules out of court anything like the traditional behaviourist's mind-set. Remember that Skinner too allowed - indeed insisted on - causation as the key fact about mentality. But Skinner would never grant access to the "black box" as a cause of our behaviour, or even as a causal intermediary in our behaviour. He restricted proper science to studying only the cause and effect sequences between "outer" physical stimuli (from the environment) and "outer" physical responses (in our bodily behaviour); we were not to seek any "inner causes" for anything. By contrast, the identity theorist is quite happy to allow that the causes of our "outer" behaviour typically are thoroughly "inner" states. They are not, of course, the "inner states" so beloved of Descartes (immune to the Evil Genius, known from a first-person perspective, incorrigible and all the rest). But they are still "inner" in the important sense which Skinner denied science access to - they are inside rather than outside the epidermis. In one stroke, then, Premise 1 happily avoids the artificial restrictions of Skinner (on "outer" causes only) without falling back immediately into Descartes (on "inner" as accessible only to me). A nice bit of work. But a bit rough on the opposing views to find themselves simply defined out of the running!

Premise 2: The "Science Knows Best" Assumption

The second premise is this: "Science is the sole arbiter of what is real and what is not real". We need something like this "science knows best" at the second step because it is, after all, the physical scientists who claim to have discovered that what in fact causes our behaviour are states and processes in our brain. When you think about it, this really does mark a fundamental - and very recent - shift in intellectual history. We are not being invited to ask, say, Jane Austen or any of the other great novelists of human emotions and interactions for their insights anymore. We are not invited to ask the psycho-analysts with their apparatus of repressed desires and non-chemical explanations of the neuroses. We are not invited to ask historians or economists or sociologists or anthropologists or lawyers what they have discovered about the reasons why people do the nice and nasty things they do to each other.

It is important to be sensitive to the full implications of this. In everyday life, when we ask each other for reasons and explanations of why we did whatever we did, we typically do not come up with accounts of neural firings in our brains, and we wouldn't understand someone who did. For example, I explain why I went to university on the bus Friday week by saying: "I had to get to class and my car broke down". I do not explain my unusual bus-tripping behaviour by saying: "My brain went into state 123456". Modern materialism insists that we really should look elsewhere than these everyday commonplaces for our explanatory apparatus. Such pedestrian explanations as "I had to get to class and my car broke down" are very well in their way, and they may get us through the night all right, but they don't tell us what's really going on, and what actually caused the behaviour we explain so glibly. To do that job - admittedly not a job we often have occasion to do - nothing less than the sorts of "My brain went into state 123456" will deliver the goods. For deep and powerful explanations, we need to go to those people who are in the business of giving deep and powerful explanations of things. And these people, identity theorists claim from the explosive advance of science in the last two centuries, are the real custodians and practitioners of the scientific method. When a chemist or a physicist or a biologist tells us something in the pages of Science or Nature, well, then we are finally getting somewhere. Philosophers are in intellectual backwaters if they don't heed the same advice.

Premise 3: Mental - Physical Identity as a Scientific Hypothesis

The third premise is this: "Science discovers that what in fact causes our behaviour are states and processes of the central nervous system". Suppose that when we pay attention to such scientists - nowadays they are mostly neurologists and neurophysiologists - we learn that they mostly discover that stimulation of a certain part of the brain, say the c-fibres, plays the causal role associated with pain. ["c-fibres" is a made-up cover term as far as philosophers are concerned.] Bodily damage typically brings about c-fibre stimulation, and c-fibre stimulation, via the motor centre in the brain, brings about typical pain-behaviour.

Conclusion 4: Mental-Physical Identity as a Philosophical Conclusion

The final conclusion is this: "The structural item which we call the 'mind' is really nothing over and above the brain" and "The structural property which we call a particular 'mental state' is really nothing over and above a particular state of the brain". Note that in order to draw this specifically philosophical conclusion from the strictly neuroscientific evidence - which is all that the practice of neuroscience ever purports to deliver - the "Science Argument" requires two very different kinds of premises:

1. A distinctly philosophical premise: specifically, an analysis of the concept of pain, say, and all other mental concepts in terms of things, objects, entities, structures (states and properties of them) that cause certain sorts of behaviour in humans; Plus
2. A distinctly scientific premise: specifically a piece of empirical research into the brain and its structural parts, research which shows which parts of the brain cause which sorts of human behaviour.

Identity theory needs both. It is not a purely philosophical theory - it uses science to deliver the empirical goods. But neither is it a purely scientific theory - it uses philosophy to set out what those empirical goods mean in a full theory of the mind. This is exactly what Sober is pointing out when he says that identity theory is "not merely a different theory than those surveyed before, but a different kind of theory".

IV.C The Identity Theorist's "Argument From the Principle of Parsimony"

The operative principle of this argument about scientific method is simple (pp. 294-297). When push comes to shove between two scientific explanations of something - or more accurately, between two research programmes for explaining something - choose the simpler one.

How does this argument look "stretched out"?

• Premise 1: Our mental terminology, let us propose, refers to what we shall call "mental events". This says nothing about what such so-called "mental events" really are in themselves - whether they are special Cartesian non-bodily events or whether they are simply electrical and chemical events in the bodily brain.
• Premise 2: Our physical terminology, let us propose, refers to what we shall call "physical events". This likewise says nothing about what physical events really are in themselves.
• Premise 3: Recent scientific experimentation has shown that there is a perfect one-to-one correlation between the occurrence of specific mental events (e.g. feeling pain) and the occurrence of specific physical events (e.g. c-fibres firing in my brain).
• Premise 4: The theory of mind-body dualism claims that the mental event of feeling pain and the physical event of c-fibres firing are two entirely different kinds of events and that their observed correlation is due to the fact that events of one kind causally interact with events of the other kind.
• Premise 5: The theory of mind-brain identity claims that the mental event and the physical event are perfectly correlated because they are one and the same event - there exists only one sort of event in the universe, physical events; there exist no causal interactions between such physical events and events of any other radically different kind.
• Premise 6: The Principle of Parsimony says that all other things being equal between two competing explanations, we should prefer explanations which minimize the number of entities, processes and events which they require to be postulated.
• Restatement 7: The theory of mind-body dualism requires us to postulate two different kinds of entities (Cartesian-type minds and Cartesian-type bodies) plus as well two different kinds of processes (body-body causal interactions and mind-body causal interactions).
• Restatement 8: The theory of mind-brain identity requires us to postulate only one kind of entity (physical bodies) and as well only one kind of process (body-body causal interactions).

therefore

• Conclusion 9: The theory of mind-brain identity is more parsimonious than the theory of mind-body dualism.

therefore the best explanation we have so far is that

• Conclusion 10: The theory of mind-brain identity gives a better account of the observed correlation between feeling pain and c-fibres firing than the theory of mind-body dualism.

The first five steps here "stretch out" Sober, pp. 294-295: the part where we set up some phenomenon (here a surprisingly perfect experimental correlation between two events). The last five steps "stretch out" Sober pp. 295-297: parsimony provides a clear-cut way of preferring one explanation of this correlation over another. In short form:

• Premise 1: It is a surprising fact that mental events and physical events are perfectly correlated with each other.
• Premise 2: Both the theory of mind-body dualism and the theory of mind-brain identity explain this fact, but mind-body dualism requires us to postulate a larger number of kinds of entities and kinds of processes to do so.
• Premise 3: All other things being equal, a simple (parsimonious) explanation of something is better than a complex explanation.

therefore

• Conclusion 4: The scientifically parsimonious theory to hold, given the present alternatives, is that the mind = the brain.

You should recognise the Principle of Parsimony being used here as just one more of several similar principles we have frequently employed in abductive arguments already. All abduction arguments basically start by setting out a surprising fact - in this case, the surprising fact that there is a perfect (or near-perfect) correlation between occurrences of being in pain and occurrences of c-fibres. If some theory T were true, then that surprising fact would be explained, or better explained than if theory T were not true, or if theory S were true rather than theory T were true. Here our theory T is mind-brain identity theory and theory S is mind-body dualism. And the Principle of Parsimony simply sets out one sort of reason we use (there are several) for preferring one theory over another.

All other things being equal, a parsimonious theory explains better than a florid theory.

Hence if theory T explains some surprising fact, and theory S also explains that surprising fact, the surprising fact is better explained by the theory which postulates the fewer number of entities, processes and events.

This Principle of Parsimony - or "Occam's Razor" as it is sometimes called from its theological days - won't provide the Last Word, of course. [Almost nothing does in philosophy, as you should be beginning to appreciate.] For a start, there is that key little qualifier: "All other things being equal, prefer a parsimonious theory". This qualifier requires us at the least to count into the assessment any auxiliary assumptions and background assumptions made by the competing explanations. One of those background assumptions will be the subject of the next unit - can a materialist science of human nature make proper room for the phenomenon of free will and human responsibility? If it can't, that fact will have to be factored into any decision. We will have at least one other preference principle to apply:

All other things being equal, a theory which explains a larger number of surprising facts is better than a theory which explains a smaller number.

Then simply counting and comparing the number of entities and processes and events required by each explanation won't decide matters. At the least any counting will be more complicated.

Identity theory uses a smaller number of resources, but it also rules out a larger number of phenomena as explainable. Dualist theory uses a larger number of resources, but rules out a smaller number of phenomena as unexplainable. Parsimony gets complicated!

IV.D Is There Any Conflict Between the "Science Argument" and the "Parsimony Argument"?

There may be. In particular, there may be a subtle conflict between putting both arguments to the service of the same theory, the mind-brain identity theory. For one of the arguments picks out the mind as something which causes so and so, while the other contrasts causation with the notion which it recommends for the mind, namely identity.

Warning!

This section is optional and for advanced readers only.

Look back at each argument in turn.

The "Argument from the Advance of Science" makes the mind-brain identity theory essentially a theory based on which items in the universe provide us with the most satisfactory set of causes for some phenomenon we are interested in:

• Premise 1: The mind = whatever causes behaviour.
• Premise 2: Science has found out that the brain is actually the cause of our behaviour.

therefore

• Conclusion 3: The mind = the brain.

Here, the identity theorist bases the claim that the mind is the same thing as the brain on the fact that the brain is the best candidate we have found for explaining the causation of behaviour.

Oddly enough, this is not at all how it works in the identity theorist's "Argument from the Principle of Parsimony". There, remember, we are really presented with alternative scientific theories and asked to choose between them (before all the empirical facts are in). Parsimony is recommended because the history of science shows that parsimony has been a very successful strategy for picking which of several theories will turn out to provide fruitful research programmes in the long run. In this case, the two theories are rival explanations for the constant conjunction of mental events and physical events: there is a perfect one-to-one correlation between feeling a certain sort of pain and my flesh being punctured by a knife, say. Theory One is standard Cartesian dualism:

The reason why X and Y are constantly conjoined is because X causes Y.

Pain is constantly correlated with puncturing my flesh because the physical process of puncturing my flesh causes the mental event of feeling pain. Theory Two is the newcomer mind-brain identity theory:

The reason why X and Y are constantly conjoined is because X is the very same thing as Y.

Pain is constantly correlated with puncturing my flesh for exactly the same reason that Invercargill, for instance, is constantly correlated with the southernmost city of New Zealand: not because one causes the other, but because one simply is the other (we have only one entity here, not two). Now choose between Theory One and Theory Two. Parsimony recommends Theory Two.

Here is the odd bit. In the second argument causation and identity are opposed to each other: dualism postulates causation in order to explain what identity theory explains more simply by merely postulating identity. In the first argument, however, causation and identity are not notions opposed to each other in any way, certainly not as rival explanatory notions. Indeed, we are supposed to find out what the mind is really identical to precisely by finding out that it rather than some other thing is the actual cause of some phenomenon. The question is whether the same theory can recommend itself for both reasons. Usually you can't have your cake and eat it too. Can you in Philosophy?

[Maybe you can, here. If the causal relations are limited to the likes of brain-firings and the blushing of cheeks, then there will be no overlap with insisting that the relations between feeling pains and brain-firings is not causation but identity. Advanced students in 34101 may want to tickle the idea a bit. I'd suggest others run for your lives.]

IV.E A Common But Not Recommended Way to Distinguish Between Identity Theory and Functionalism

Sober explains an important distinction invented by identity theorists, the difference between a "type" and a "token" (see especially pp. 298-299). Unfortunately, he does all this in Lecture 22 on Functionalism, which is the subject for Week Ten. Even more unfortunately, when he proceeds to employ this "type" / "token" distinction to mark off Week Nine's theory of mind-brain identity from Week Ten's theory of functionalism, your eyes are guaranteed to roll right back into your heads.

Here are some examples in advance to ease into things.

Proceed at your own risk!

This is pretty advanced stuff. Come back to it next week - if ever. I have put it here because there will be nothing about "types" and "tokens" in the next Study Guide.

Start off with a type - token distinction that has nothing to do with theories of the mind:

These count up to be five specific items (tokens) of the same type of thing, a refrigerator.

Here is a type-token distinction for a mental state or event:

Five specific, countable, cases (tokens) of the same type of thing, a certain type of experience.

Now a type-token distinction for a well-known neurological phenomenon:

Five specific, dateable, countable samples (tokens) of the same type of neurological firestorm.

Finally, a type-token distinction for a made-up brain state or event which I shall suppose is the one wanted for a Physicalist account of elation:

Five more specific, datable, countable examples (tokens) of a single type of neurological firestorm.

With the notions of types and tokens socked away, one way to mark off the difference between identity theory and functionalism is simply to count up how many and how strict are the identities which each proposes. [However, a much better way in the long run will be simply to toss out the whole "type" / "token" business as applying very interestingly to functionalism and instead to try to understand the new view in terms of the distinction between "structure" and "function" which it invents as much more illuminating. This is the approach I take in the Study Guide for Week Ten.]

Mind-Brain Identity Theory

The theory of mind-brain identity proposes the strictest possible identity between minds and brains.

Each token of a mental state (my feeling elated right this instant, say) is identical to some token of a brain state (my neural net 12345 firing right this instant, say).

There are no dateable, countable, specific occurrences which aren't physical ones. So science tells us, or soon will tell us (see IV.B). As well,

Each type of mental state (being elated) is identical to one and the same type of brain state (neural net 12345 going off).

A type of thing isn't of course a dateable occurrence. The dateable occurrences are all tokens of a type. That's the difference between types and tokens. Mind-brain identity theory goes out on a limb and predicts that scientists will discover there is a one-to-one correspondence between types of mental states and types of neurological states, in addition to the token - token identity.

Functionalism

Functionalism disagrees fundamentally with such strict mind-brain identity theory. Here is a quick way into their point of disagreement. The functionalist takes quite seriously that entities from Alpha Centauri, say, may well suffer from pain and depression and hatred - for all that they are built very different from us and don't have brains perched on top of spinal cords. They may also be able to be elated from time to time - for all that they are built very different from us and don't have neural nets even, much less that kind of neural net designated 12345 by neurologists (any more than they have that kind of neural firestorm designated "petit mal seizure"). So it would be plain silly, according to a functionalist, to identity the type of occurrence feeling elated with the type of occurrence neural net 12345 firing. The first type of occurrence we may well discover is applicable to Alpha Centaurians. The second type of occurrence we can be almost certain in advance will never be applicable to anything except beings that have evolved on the third rock from the sun, since neurons and therefore neural nets have evolved because of the precise history and peculiarities of the ecosystems on that rock. If the type feeling elated is a possible type for other species than our own, while the type neural net 12345 firing is certainly unique to us, then feeling elated and neural net 12345 firing could hardly turn out to be one and the same type of occurrence could they?

Functionalism therefore loosens up on both of the identities which the mind-brain identity theory is committed to.

Each token of a mental state (my feeling elated right this instant) is identical to some token of some physical state of whatever physical structure scientists discover is the appropriate one.

The point of all these "somes" is that a functionalist is a materialist too. So whatever mental event tokens are going to found to be identical to, they will always be physical tokens - specific, dateable, countable, physical events rather than specific, dateable, countable ghostly events for instance. But that is as far as functionalism needs to change the tokens part of the story. After all, in human beings the obvious physical structure to go for is the human brain. So a functionalist can agree it may well turn out that scientists discover the required physical tokens, for human beings, are tokens of brain states, exactly as mind-brain identity theory tells us too. If so, then for human beings, we will go on and identify each token of a mental state with some token of a state of the brain - and maybe even with such a token of a physical state as my neural net 12345 firing right this instant.

It is with the types part of the story that the theory of functionalism outright denies the identity proposed by the theory of mind-brain identity. According to functionalism:

No single type of mental state will ever be identical to one and the same type of brain state.

In human beings maybe, a single type of mental state (e.g. being elated) may turn out to be identical to a single type of brain state (e.g. neural net 12345 going off). But at most that will be true of human beings only. The very same type of mental state - elation - may turn out to be identical to a very different type of physical event for Alpha Centaurians. Indeed, we can practically guarantee it will, seeing as Alpha Centaurians will almost certainly have no neurons nor brains made up of neurons nor networks of neurons firing in those brains. Instead, that type of mental event, scientists may discover, will be identical to a single type of physical event in the purple appendage sticking out at the bottom, say.

We have already found this to be a live option with computers by the way. It may readily turn out that computers are capable of the very same type of mental event too, elation. But in computers, that type of event could never be found to be identical with the type of physical event 12345 neural net going off. For computers just don't have brains made of neurons, with networks of such neurons firing at some times and not at other times. They are made of silicon and ice-cream. So the type, elation, in them will be identical with a type of silicon and ice-cream event (silicon getting overheated and jelly beans 72 and 38 melting, say).

Contents
Previous Week: Behaviourism and "Scientific Method"
Next Week: Functionalism

Content © 2000, 2001 Massey University | Design © 2000, 2001 Alun David Bestor | Any questions? Email the webmaster