20th World Congress of Philosophy Logo

Philosophy of Mind

To Resurrect a Ghost:
In Defence of Psychological Dualism

Donald V. Poochigian
University of North Dakota

bluered.gif (1041 bytes)

ABSTRACT: Cartesian dualism has largely been replaced by empirical theories of the mind. Central to this development is Gilbert Ryle’s criticism of an immaterial ‘ghost’ inhabiting the material ‘machine’ of the body. A metaphysical self is incredible, and even if it is credible, both it and its manifestation in phenomenal experience are unknowable by others. Failure of this approach occurs when it is realized that existence of the physical is just as incredible as existence of the metaphysical. Free will is also inconceivable without the assumption of a metaphysical self, it being the ‘ghost in the machine’ after all. As for consciousness, it is presupposed by the empirical. What counts as physical manifestations of mind are the effects or causes of phenomenal experience. Without this criterion the individual is a unity, it being impossible to separate the psychological since effectively it encompasses every aspect of the individual. Additionally, it is in phenomenal experience that the empirical is observed, and observation is the basis of empirical verification. To advocate the scientific method of intersubjective verification while denying the existence or significance of the phenomenal is inconsistent. At root, the mental attributes are ontologically distinct. Limited to only one ontological substance, empiricists either redefine or exclude troublesome attributes, commiting the error of confusing distinct kinds of substances. Dualism can accommodate all of the properties of mind in a single coherent theory by acknowledging these kinds of substances.

bluered.gif (1041 bytes)


Perhaps the most influential recent philosopher of mind is Gilbert Ryle. Beginning with his criticism of Cartesian dualism, empiricism has come to dominate current thinking on the nature of mind. Basic to this evolution is Ryle's position is that "the dogma of the Ghost in the Machine" confuses one logical with another. Mind is physical rather than metaphysical, there being no other way to account for knowledge of other minds.

What Ryle and other empiricists fail to recognize is that there is no category error in accepting a "ghost in the machine." Such an error occurs only if setting out to resolve a particular problem, that of other minds. But psychological dualism is concerned with more than this, the issue of other minds being only an aspect of a more comprehensive theory of mind. To focus simply on other minds as Ryle does, is to ignore or deny the other mental features encompassed by a dualist theory.

Three aspects are contained in dualism, each presupposing an ontologically different kind of substance. Free will is accounted for by a metaphysical self, consciousness by phenomenal events, and knowledge of other minds by empirical events. Ryle accuses dualists of misconstruing one kind with different kinds, when it is he who commits the category error of misconstruing different kinds with one kind.

What lies at the basis of the metaphysical self is free will. This is an ontological notion which does not apply to anything understood as a set of observable characteristics, although a metaphysical self is manifested by such characteristics. That this is so can be seen by considering the nature of an institution, which is a set of characteristics. Harm done to an institution is only blameworthy when intrinsically wrong such as stealing, or instrumentally injurious to individuals such as members or employees or shareholders. Why is because what is done to an institution is a characteristic of an abstraction, a way of understanding, not a thing or object. It is a hollow shell of linked properties with nothing inside, nothing which has these properties. What provides this "something" is the theoretical assumption of a metaphysical self, a "ghost in the machine" which is linked to the phenomenal and physical manifestations of the self by functions.

Functions identify the appearances of the metaphysical in the physical realm. Specifically an instance of something is understood as an observable effect of the metaphysical essence of that thing. Such manifestations are infinite in extent, appearing as simples and complexes in an unending process of reduction and construction. A function determines the circumstances in which an instance occurs, linking metaphysical essence with physical instance. Essence and instances together constitute a person.

This might be denied by arguing that although humans are understood as merely sets of properties just as institutions, they are evaluated differently because of definitional norms. Yet it seems that if a person is viewed as only a concatenation of properties, there is no sense of responsibility towards that individual.

Objection might be made that this is so only if someone is not understood as sentient.

But why does sentience matter? Suppose pain means a set of observable events, then whether or not something is understood to experience pain would make no difference in moral assessment.

What difference does it make if this set of physical events occurs or not when there is no sensation of pain? More interesting is the epiphenomenal assumpton of phenomenal events as the natural product of physical events. Would it seem wrong to cause phenomenal pain in this circumstance, when it is a property which understood as a whole with other properties constitute the individual? In answer it seems not wrong. Pain here is simply an event that appears and disappears in the universe. No one suffers it, it merely happens.

On the assumption of a metaphysical self activity is traced to a first cause. Such a thing being immaterial, it might be thought that it is incapable of material effect. It is difficult to comprehend, though, why many believe there is a mind/body problem. Causation is only necessarily applicable to empirical body, which is extended in space, and not to phenomenal mind, which is not extended in space, just if there is something in the ontological character of matter that is the source of cause and effect. There is no such thing on examination, only sequences of occurrences. Location in a sequence is a temporal matter, not one of space, and as such is as much a feature of the phenomenal as of the material. This being so there is no reason that phenomenal events cannot be combined with empirical in a causal sequence. Folk psychology certainly assumes this, mind thought to cause body and body to inform mind.

Accepting that the immaterial and material can interact, does not the understanding of a metaphysical mind as a cause suffer from an infinite regress? Must it not be caused and this cause in turn be caused, and so on? In fact there is no necessity in this, infinite regress not being a difficulty for the homuncular ghost in the machine. Such a thing must be indivisible if it is basic, an indescribable simple without properties. Irreducible as a result, it is not contingent or dependent on anything else. Without a cause it is necessarily self-determining, escaping the problem of infinite regress.

And even if infinite regress is a difficulty for the homuncular theory, it is an equal difficulty for any empirical account of mind. All that is observable is infinitely reducible, which is why an empiricist account of mind is always incomplete. Advocates of artificial intelligence are invariably "almost there," but never quite get "there." No matter how an empirical conception of mind is constituted it is incomplete. Components are always reducible and it is always possible to identify indeterminate states that lie in between elements.

Still to be clarified is how the assumption of a metaphysical mind leads to anything other than determinism. Why should minds not function in wholly predictable ways? In answer it can be said that this is not how minds are understood. What is predictable is a machine rather than a mind, and what is unpredictable is diseased if biological and broken if mechanical. A mind can be diseased, but then it is no more in control of itself than what is wholly regular. Only when disposed is a mind free and healthy, understood as rational but imperfect. Orderly in general, it is not always and its inconsistency is unpredictable, anything not this way being mechanical. It is not an accident that Ryle understands the mental as dispositional, for this is the definitional character of mind.

That human beings are imperfect is not a matter of fact since imperfection is relative, depending on the range of what is understood as the same. Rather it is that humans presume themselves imperfect, generally following rules but not always. This is an a priori rather than a posteriori understanding.

Accepting a metaphysical self as necessary for free will, might this not be a remnant of a more ignorant time? Surely modern science shows that the universe is a supervenient whole where all is determined by the components which make it up. Perhaps free will is an archaic notion which should be either redefined or abandoned. No satisfactory sense of moral responsibility seems compatible with a complete causal determination of an agents's actions. Given this inability to redefine free will it might be thought that the concept should be abandoned all together, but matters are not as simple as this. Whether or not reality is understood as an architectonic whole is conditioned by concern with order or disorder.

Belief in an absolute metaphysic able to account for every aspect of the universe is no more reasonable than belief in an absolute geometry able to map every aspect of the universe. Often offered as strong evidence of the truth of the scientific conception of reality is that science "works." But things can "work" or be successful in different ways. Success is a function of the extent some objective is achieved. Different things can be more or less successful, depending on the objective by which this is measured. Science's success, then, depends on the standard or purpose it is evaluated against. This being so, what is the goal by which science is said to work? In answer it seems that of order since science explains by generalized rules.

What science is not good at is explaining disorder. For science disorder itself is an orderly event, the product of some law. Of course all instances to which a law applies are not the same. Such variation is either ignored in science or a law like explanation is sought for it. Variation will occur in the application of any law, though, requiring yet another law and so on infinitely. No ultimate account of difference can be provided by science as a result. Other accounts such as spiritualism are more suited to this task. Only if an account of order is sought is science successful, those who accept this as evidence of the truth of the scientific conception of reality revealing more about their own character than about the way the world is.


Determinism presumes a physical self and free will a metaphysical self. What is physical is observable, and anything observable is infinitely reducible. As such it is necessarily determined by its constituents. What is metaphysical is unobservable, and anything unobservable is irreducible. As such it is necessarily undetermined since it is without constituents.

Because it is infinitely reducible what is physical is contingent, dependent on something else for its existence. This is as true for the whole as it is of a part, so that the physical universe itself presupposes a metaphysical source. Being irreducible such a source in turn presupposes no source. It is self-determined in contrast to the physical which is other-determined. Understood as outside of consciousness, metaphysical simples are independent of the conditions of experience. Without components they do not change, and unchanging they are exterior to time. Occurrence can be eternal or not, and if not both genesis and demise can be spontaneous. This is possible because such things are outside the limits of the human ability to understand.

A cause constituting a relation rather than an object, the physical is a manifestation of the metaphysical. Relationship is by functions identifying the appearances of the metaphysical in the physical realm. Propertyless themselves these metaphysical simples are still distinguishable to account for the different things in experience. Distinction occurs in terms of observable effects, which are identified in functions. Specifically an instance of something is an effect in experience of the metaphysical essence of that thing. Such manifestations are infinite in extent, appearing as simples and complexes in an unending process of reduction and construction. A function determines the circumstances in which an instance occurs, linking metaphysical essence with physical instances.

Reducible to observable instances, though, a metaphysical self is something which it might be thought can be disregarded. There is a severe difficulty with this, all that occurs in experience succumbing to an infinite regress. Nothing can occur in experience without properties because all of experience is extended at least in time, and as such is composed of constitutent moments. Appearance being divisible into parts which are divisible into other parts and so on without end, there is a structure without a content. It may be a way to build a world, but there is nothing with which to build it. A conception of this sort is like the cartoon character Wile E. Coyote who after running off the edge of a cliff only falls when realizing he is walking on air. This process of regress can be halted solely by the metaphysical assumption of an irreducibly simple essence. Without an irreducible simple from which mind is built, there can be no account for consciousness. Only by assuming a metaphysical simple can the mind be provided a foundation, consciousness itself as a thing being necessarily a monad. No matter the observable manifestations of awareness, whether material or phenomenal, there must be something outside of experience to account for what is experienced.

Of course the assumption of a metaphysical self is a tenuous proposition since it is quite impossible to know if such a thing exists. As the presumed source of experience it cannot be experienced, an eye being unable to see itself afterall. Mysticism might be appealed to, but one can never be certain of being fooled in a mystical experience. Those without the requisite experience cannot verify its occurrence because they cannot know if it has occurred. And those with the requisite experience cannot verify its occurrence because they cannot know if their own experience is real. Additionally even if it is possible to know whether a substratum exists, does each property of a substratum have its own substratum, and since each property is infinitely divisible, does each subdivision have its own substratum? If so the universe is a very crowded place indeed.

Resolution of this last difficulty is to simply assume some things have substrata and others do not. Different individuals can be expected to have different opinions on this matter, animists for instance disagreeing with mechanists. An even more interesting contrast is between contemporary ethicists and philosophers of mind. Many ethicists are contractarians for whom free will is the essential human nature. Among philosophers of mind, though, an even larger proportion are engaged in an agenda of reducing mind to a determined structure, whether conditioned behavior or neural activity or a functional program. None of these opposites can convince advocates of the other of the error of their way because all judgments are arbitrary. They are so because both empirical and essential accounts of reality are fatally flawed. Yet there are no better alternatives.

A supervenient universe forms a solidly constructed edifice with no foundation, a fairy castle floating in the sky. Only by assuming an irreducible simple can a foundation be provided. Such a thing must be without properties and to be without properties is to be unextended. No combination of what is unextended can ever be extended, though, so that it is impossible to construct the physical from the metaphysical. Yet there is no basis for the physical without the metaphysical. Left unresolved is how to get from transcendence to imminence, which can be viewed as the mind/body problem.


Resolution of the mind/body problem occurs in the form of consciousness which is basic to psychology. It is basic in the sense that it determines the mental, any other aspect of the mental being what is understandable as a cause or effect of consciousness. Distinguishing consciousness is that it is phenomenal, which is indicated when considering that it is always possible to doubt someone else's consciousness, to imagine another not being conscious despite all appearances. An individual might be sleepwalking or a zombie or a cleverly designed machine after all. But it is never possible to doubt one's own consciousness, to imagine one's self not being conscious despite all experiences. To doubt is to be aware, and to be aware is to be conscious.

Perhaps it is because consciousness is phenomenal that it is essential to mind. Although real, it is immaterial, allowing it to be understood to provide a link between a metaphysical self and the physical world. Whether this is so or not, although consciousness is central in determining the mental, it is still not all there is to the mental. However the mental is understood, consciousness need not occur for psychological events. It is just that the kinds of things which can be psychological are those which are means to or products of consciousness. These are psychological even if occurring independently of a phenomenal experience.

Exactly what empirical causes and effects of consciousness are encompassed in the mental is where there is uncertainty. It is impossible to clearly delimit the content of any empirical criterion. Why is because the person is a unified being without the phenominal, making it impossible to separate the psychological. When understood as physical the psychological cannot be isolated from any aspect of the individual, since effectively it encompasses every aspect. Psychology is without any distinguishing character.

Only with Descartes' distinction between mind and body is some resolution offered for this. Physical manifestations of mind are distinguished as the products of phenominal events, which is why the phenomenal plays a central role in defining the mental. Ignoring the phenomenal and focussing on the physical alone leaves the psychological capricious. Which physical activity counts and why? Without the phenominal as a criterion, either all of the physical manifestation of the individual is psychological or the psychological is arbitrary. Ultimately, though, even this fails to provide a complete settlement.

Indeterminacy on this matter is a product of the infinity of causal sequences. There is no precise point in a chain of events beyond which is to go too far. All things in experience being contingent, a condition for their existence is presupposed. This condition in turn can always be understood as the "real" thing since it is more basic. Acceptance of this is mistaken because the condition for it can now be understood as the "real" thing and so on. With no way of avoiding this process, resolution is possible only by simply understanding mind as some one alternative. All concepts being potentially infinite, their extent is circumscribed by denomination in order to avoid indeterminacy. Each constituent conceptualization composes an autonomous representation of mind in this way, a different way of considering it. As such every one individually is a sufficient condition for the whole, mind being conceivable in terms of any one.

bluered.gif (1041 bytes)


Back to the Top

20th World Congress of Philosophy Logo

Paideia logo design by Janet L. Olson.
All Rights Reserved


Back to the WCP Homepage