A Materialist Theory of the Mind

D.M. Armstrong

• Publisher: Routledge
• Year: 2002
• ISBN:
  9781134856343
• Edition: 2
• Topic: Filosofía
• Subtopic: Historia y teoría filosóficas

Breaking new ground in the debate about the relation of mind and body, David Armstrong's classic text - first published in 1968 - remains the most compelling and comprehensive statement of the view that the mind is material or physical.
In the preface to this new edition, the author reflects on the book's impact and considers it in the light of subsequent developments. He also provides a bibliography of all the key writings to have appeared in the materialist debate.

Part One: THEORIES OF MIND

1: A CLASSIFICATION OF THEORIES OF MIND

THERE are many possible ways of classifying theories of mind. The classification to be put forward here is based upon different conceptions of the relationship of mind to body.

Some theories of mind and body try to reduce body to mind or some property of mind. Such theories may be called Mentalist theories. Thus according to Hegel and his followers, the Absolute Idealists, the whole material world is really mental or spiritual in nature, little as it may appear so. According to Leibniz, material objects are colonies of rudimentary souls. These are both mentalist theories. It may be plausibly argued that Bishop Berkeley and his philosophical descendants the Phenomenalists, who hold that physical objects are constructions out of ‘ideas’ or sense-impressions, are putting forward mentalist theories of matter.

In opposition to these mentalist theories, we have Materialist theories which try to reduce mind to body or to some property of body.

Between mentalist and materialist theories we find two sorts of compromise theories. In the first place, there are Dualist theories which treat mind and matter as two independent sorts of thing. In the second place, we have theories like Spinoza’s, which treat mind and matter as different attributes of the same underlying stuff, or Neutral Monism, which holds that mind and matter are different arrangements of a single sort of stuff.

However, if we consider the tradition of modern analytical philosophy, within which this book is written, we find that many of these views are not living intellectual options. Some analytical philosophers have accepted Phenomenalism. But most have taken the common sense view that physical objects are not mental in nature, nor are they attributes of, nor constructions out of, something neutral in nature. For this reason, the only theories of mind and body that I will actually examine in this first part of the book are those that accept the irreducibly physical nature of physical things. I have two reasons for ignoring Phenomenalism. In the first place, its intellectual credit has been shaken in recent years. In the second place, I have already said what I have to say in criticism of Phenomenalism in Perception and the Physical World (Routledge, 1961, Chs. 5 and 6).

Having in this way limited the field of theories to be examined, a new classification is required for those theories that remain. We may distinguish between Dualist theories of mind and body, Attribute theories and Materialist theories.

A Dualist theory is one that holds that mind and body are distinct things. For a Dualist a man is a compound object, a material thing—his body— somehow related to a non-material thing or things—his mind. There are two main types of Dualist theory.

In the first place, we have Cartesian Dualism. For the Cartesian Dualist the mind is a single non-material or spiritual substance somehow related to the body. Although the term ‘Cartesian’ refers to Descartes, and we find this view of the mind and body expounded by Descartes in his Sixth Meditation, the term, as I use it, is not to be restricted to the exact theory put forward by Descartes. It is to be applied to any view that holds that a person’s mind is a single, continuing, non-material substance in some way related to the body.

(Since this is the first time that the term ‘substance’ has been used in this work, and since it will be used frequently in the future, it will be convenient to say a word about the notion of substance here. Locke conceived of substance as the unknowable substratum of objects. Those who came after him often rejected the doctrine of such a substratum, but the unfortunate effect was not to revise Locke’s conception of substance, but to give the whole notion of substance a bad odour. But when I present the Cartesian view as the view that the mind is a non-material substance, I do not regard the Cartesians as necessarily committed to a doctrine of an unknowable mental substratum. I understand by a substance nothing more than a thing that is logically capable of independent existence.)

In the second place, we have what may be called ‘Bundle’ Dualism, the term ‘bundle’ echoing Hume’s notorious description of the mind as a ‘bundle of perceptions’. This form of Dualism characteristically arises out of reflection on the difficulties of Cartesian Dualism. When the great Empiricist philosopher, David Hume, turned his gaze inward upon himself, he found that he could discern no continuing spiritual principle within himself. In his discussion of the nature of the self in his Treatise of Human Nature, Bk. I, Pt. IV, Sect. 6, he says:

For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe any thing but the perception. (p. 252, ed. Selby-Bigge, Oxford University Press, 1888)

Hume is arguing here that there is no continuing object in the mental sphere corresponding to the body in the physical sphere. Nobody ever observes such a spiritual principle within himself. All that observation of what goes on in our minds reveals is a succession of what Hume calls ‘perceptions’: that is, perceptions, sensations, emotions, thoughts and so on. So the ‘Bundle’ Dualist takes the mind to be a succession of non-physical particulars or items distinct from, although related to, the body.

Although ‘Bundle’ Dualism is closely linked with Hume’s name, it is not absolutely clear whether or not Hume himself was a ‘Bundle’ Dualist. His view of the mind certainly fits our definition, but it is legitimate to doubt whether he holds a Materialist theory of the body. For sometimes he seems to hold a view of physical objects similar to that of Berkeley, making them nothing but our perceptions ‘of’ them. This would imply that minds and bodies are both constructed from perceptions. However, at other times Hume’s view seems to be less radical, and closer to common sense. Then he talks as if a man were a material thing somehow united to a bundle of non-material items: perceptions or experiences.

There is another important way of classifying Dualist theories which cuts across the distinction between Cartesian and ‘Bundle’ Dualism. This classification is based on the particular nature of the relationship thought to hold between the mind and the body: it is the distinction between Interactionist and Parallelist theories.

The difference between Interactionist and Parallelist theories may be brought out by considering the causal relations between (i) a room and its thermostat; (ii) a room and its thermometer. A room and its thermostat act upon each other. A rise in the temperature of the room brings about changes in the thermostat; the changes in the thermostat in turn affect the room, bringing back its temperature to a certain level. If this action and reaction did not occur, the thermostat would not be acting as a thermostat. The Dualist who is an Interactionist thinks of body and mind as related like room and thermostat. The body acts on the mind, the mind reacts back on the body.

On the other hand, although a room acts upon its thermometer, a rise in the temperature of the room causing a rise in the mercury of the thermometer, the thermometer does not react back upon the room. (In fact it does so to a very small extent, but we may ignore this point.) If the thermometer affected the temperature of the room, as a thermostat does, it would be no use as a thermometer. Now the Dualist who is a Parallelist thinks of body and mind as related like room and thermometer. The body acts on the mind, but the mind is incapable of reacting back on the body in any way at all.

There is a still more extreme form of Parallelism according to which not only is the mind incapable of acting on the body, but the body is also incapable of acting on the mind. Instead they run in two parallel series like the two rails of a railway line, or two perfectly synchronized clocks. However, I do not propose to consider this form of Parallelism explicitly. My reason is that I do not think that it is seriously considered by present-day thinkers. Here we concern ourselves only with those theories of Mind and Body which are living options for the thought of our time.

It is sometimes argued that if we accept a Uniformity theory of causation, for which causation is nothing but regular sequence, then the distinction between Interactionism and Parallelism disappears. And it is true that acceptance of a Uniformity theory would blur the distinction between the mild and the extreme forms of Parallelism. But it is clear that, even on a Uniformity theory, there is a distinction between the way a room’s tempera ture is related to its thermostat, and the way it is related to its thermometer. The uniformities involved in the first sort of relation are much more complex than the uniformities involved in the second sort. Now the relations of the room’s temperature to thermostat and thermometer respectively are the same as the relations of body to mind envisaged by Interactionism and Parallelism respectively.

I said that the distinction between Interactionism and Parallelism cuts across our original distinction between Cartesian and ‘Bundle’ Dualism. Descartes himself was an Interactionist, although he did think that there was difficulty in conceiving of the interaction of spiritual and material substance. His successors, however, gave up the doctrine of Interaction, and so became Cartesian Parallelists. In his Dualist mood, Hume believed in the interaction of mind and body. Later ‘Bundle’ Dualists, however, became sceptical about the possibility of such interaction, and so we have ‘Bundle’ Dualists who are Parallelists. This is the important position known as Epiphenomenalism, that finds its classical exposition in T. H. Huxley’s paper ‘On the Hypothesis that Animals are Automata and its History’ (to be found in his Methods and Results, Macmillan, 1894). For Huxley, consciousness is thought of as a mere by-product of the operation of the brain. We do have experiences which are, in themselves, something more than the workings of the brain, but they are incapable of influencing the operation of the brain in any way. Experience is, in Huxley’s phrase, the smell above the factory, or, as the American philosopher Santayana put it more poetically, the motes above the stream. It is clear that Epiphenomenalism will have a great appeal to anybody who is sympathetic to Materialism, but who still thinks that there is something irreducible about mental states with which Materialism cannot deal.

Some philosophers would wish to use the term ‘Epiphenomenalism’ to cover any form of the ‘thermometer’ view of the mind, whether the mind is conceived of as a bundle of items or as a spiritual substance. They would restrict the term ‘Parallelism’ to what I called the extreme form of Parallelism: the ‘railway-line’ view. But the term ‘Epi-phenomenalism’ clearly suggests that the mind is conceived of not merely as totally passive but also as a mere collection of phenomena. So I restrict the term to the intersection of ‘bundle’ and ‘thermometer’ Dualist theories.

This completes our classification of Dualist theories. In sharp opposition to any form of Dualism we have Materialist or Physicalist theories of mind. For a Materialist, man is nothing but a physical object, and so he is committed to giving a purely physical theory of the mind. There are two types of Materialist theory. In the first place, there are various forms of Behaviourism. The Behaviourist denies that the mind is any sort of object, or collection of objects, arguing that to have a mind is simply to behave physically in a certain way, or to have tendencies to behave physically in a certain way. On this theory, the difference between a living man with a mind, on the one hand, and a corpse or a stone, on the other, is simply that the living man behaves, or has tendencies to behave, in a different way from the corpse or the stone.

Behaviourism originates with the psychologist J. B. Watson (see his Behaviour, New York: Holt, 1914 and his Behaviourism, University of Chicago Press, 1924). In recent times the American psychologist B. F. Skinner has defended a Behaviourist theory of mind (see his Science and Human Behaviour, New York: Macmillan, 1933). Turning to the philosophers, it is very easy to interpret Gilbert Ryle’s The Concept of Mind (Hutchinson, 1949) as a defence of Behaviourism, although this interpretation is denied by some philosophers. It seems to be denied by Ryle himself in the course of the book, but I think that this is only because he gives the word ‘Behaviourism’ a particularly narrow meaning. Some philosophers, of whom I am one, think that Wittgenstein’s very important, but very difficult, Philosophical Investigations (Blackwell, 1953) also expounds a Behaviourist account of the mind. But in some quarters one risks being torn to pieces for interpreting the book in this way.

The second form of Materialism is what the American philosopher Herbert Feigl has called the Central-state theory of the mind. Mental states are identified with physical states of the organism that has the mind, in particular, with states of the brain or central nervous system. Such a view has always been attractive to many psychologists, but until recent years most philosophers have thought that there were obvious and conclusive objections to this sort of theory. But in the last decade or so, the Central-state theory has been revived by philosophers in a very interesting way. The revival is associated with such names as Herbert Feigl, Paul Feyerabend, U. T. Place, Hilary Putnam and J. J. C. Smart.

Before the modern revival, the most conspicuous defender of a Central-state theory was Thomas Hobbes. But his defence of the theory lacks philosophical sophistication, and in any case he seems to waver between a Central-state theory and Epiphenomenalism. The object of this book is to defend a particular form of the Central-state theory of the mind.

Instead of speaking of the Central-state theory, some philosophers speak of the Identity theory, that is, the theory that the mind and the brain are identical. The label is less explicit than ‘Central-state theory’, although it is briefer. In any case, the term seems sometimes to be applied to Attribute as well as to Central-state theories.

We have now mentioned the main forms of Dualist and Materialist theory. Some philosophers are dissatisfied both with Dualism and with pure Materialism. They try to get the best of both theories in the following way. They agree with the Materialists, against the Dualists, that a man with a mind is but a single substance. There is no question of the mind being a non-material substance, or a collection of non-material items, distinct from the body. But the compromisers argue that this single substance, the man, has further properties beyond those conceded by the Materialists. For a Materialist, a man is a physical object, distinguished from other physical objects only by the special complexity of his physical organization, and the special complexity of his physical capacities. He does not have any non-physical properties. But the theory we are describing argues that men, besides having physical properties, have further properties quite different from those possessed by ordinary physical objects. It is the possession of these unique properties that gives men a mind. I shall call such theories Attribute theories of the mind. Some modern philosophers have spoken instead of Double-Aspect theories, and have seemed to mean by the term what I mean by Attribute theories. But the term ‘Double-Aspect’ is unsatisfactory, because all it suggests is that mind and body are the one identical thing observed in two different ways, or from two different points of view. This might even be compatible with Central-state Materialism, and so the term blurs the distinction between a pure Materialism and a compromise theory.

Aristotle’s doctrine, put forward in De Anima, that the mind is the ‘form’ of the body is perhaps a version of the Attribute theory. Spinoza certainly held that mind is an attribute of the body. But he does not really fit into our classification, because he holds that everything has mental as well as material attributes. This means that he does not accept the framework within which our discussion of theories of mind is placed: the purely Materialist or Physicalist account of ordinary physical objects. Samuel Alexander’s theory in Space, Time and Deity (Macmillan, 1920) is an Attribute theory, although one that comes very close to Materialism. The late John Anderson, of Sydney University, held a somewhat similar view. (See his Studies in Empirical Philosophy, Angus and Robertson, 1962, especially the paper ‘Mind as Feeling’.) Locke’s case is an interesting one. Locke is a Cartesian Dualist, holding that the mind is a non-material substance. Nevertheless, on a number of occasions he says that it is not beyond the power of God, if he should please, to endow matter with consciousness. He seems to be saying that, although Dualism is true in fact, the Attribute theory of mind is a logically possible one. (I think it is very unlikely that he thought that a purely Materialist theory of mind is logically possible.)

Is the Thomist theory of the relation of mind and body an Attribute theory? The Thomist insistence that a man is a single substance points in this direction, but the assertion that the soul can continue to exist independently of the body, even if in an unnatural state, points back towards a Dualist theory. I am inclined to see in Thomism an uneasy and somewhat confused oscillation between an Attribute theory and Dualism, although I lack space here to substantiate the accusation.

I cannot see any clear and important way of sub-dividing Attribute theories.

It is important for the subsequent argument to realize that the object, or collection of objects, with which the Dualist identifies the mind is not in space although it is in time. It may then be objected to our classification that it has ignored the possibility that the mind is an object or collection of objects distinct from the body but nevertheless spatial. But to this view we may put a dilemma. Either this spatial mind is in the same place as part or the whole of the body, or it is not. If it is not, we have an eccentric and implausible form of physicalism. If mind and body are spatially coincident, we have not got a Dualism but an Attribute theory. For two spatial things cannot be at the same place at the same time. To attempt to speak of two spatial things with different properties at the same place at the same time is to speak of just one thing with both sets of properties.

This completes our scheme of classification. I hope it will cover most of those theories of mind and body that are seriously held today, always remembering the limitation to what I have called Materialist or Physicalist accounts of matter. But it will be wise to make some qualifications to prevent misunderstanding.

In the first place, it may well be possible to find theories of mind which stand on the border-line of our divisions. In particular there may be certain theories which could be classified indifferently either as Attribute theories, or as forms of Materialism.

In the second place, closely connected with the first point, there is the possibility of combining two or more of these different views of the nature of mind to produce a mixed theory. Thus, in recent philosophy there have been many triumphs for Behaviourist and Behaviourist-oriented analyses of psychological concepts. At the same time, those who acknowledged these triumphs were, very often, understandably reluctant to say that Behaviourism was the whole truth about the mind. This led to the idea of an account of mind which would combine Behaviourism with one of the other theories. A ‘Bundle’ Dualist, for instance, might admit that when we ascribe mental predicates to persons we are very often speaking only of the way they behave. On a certain occasion, for instance, what we mean by saying that a man ‘thought fast’ may simply be that he acted in the right way very promptly. But, the ‘Bundle’ Dualist would add, besides behaviour there is a suc...

Table of contents

• Cover Page
• Half Title Page
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Acknowledgements
• Preface To The Paperback Edition
• Introduction
• Part One Theories of Mind
• Part Two The Concept Of Mind
• Part Three The Nature of Mind
• Bibiliography
• Index