[Return to Dmytro Sepetyi’s Personal Page]

Interactionism vs Epiphenomenalism:

Unclosing the Causal Closure of the Physical

© Dmytro Sepetyi

(The paper is copyrighted in the State Service of Intellectual Property of Ukraine)

There are two kinds of mind-body dualism: epiphenomenalism and interactionism. The difference is concerned with the thesis that physical reality is causally closed: if any non-physical things or properties exist, they cannot influence the course of physical events. The causal closure thesis is often appealed to by materialists, and some dualists agree − they are epiphenomenalists. Conversely, interactionists contend that 1) there is no weighty reason why the non-physical cannot influence the physical and that 2) the mind is non-physical, and it interacts with (influences and is influenced by) physical processes in the brain. So, is the thesis about the causal closure of the physical tenable?

Interactionism and the Causal Closure Thesis

Interactionism is the most common-sense view. We all really think that the mіnd interacts with the body. It seems as obvious as anything that:

1) when my body is injured somewhere, I feel pain; I have images, feel smell and taste − these sensations have physical causes (most immediately, some processes in the brain); I hear speech (air vibration) or read a text (physical configurations of ink on paper), and this influences my thoughts;

2) when I want to do something (motvated by my feelings and thought) I move my body, − it isn’t just that the body moves; it is that I (the conscious self) move it.

1) is admitted by both epiphenomenalists and interactionists; the controversy is about 2). Epiphenomenalists admit causal relations only in one direction − from body to mind, not vice versa: it is just an illusion that my will (motivated by my thoughts and feelings) moves my body; all movements of my body are completely determined by purely physical causes. Interactionists (as we all in real life, even epiphenomenalists when not philosophizing) think that causal relations are bilateral: I (thinking, feeling and willing conscious self) really move my body because I want it.

Are there good reasons to think that it is an illusion? Many philosophers believe that there are: physical reality is causally closed; the non-physical (if it exists) cannot possibly influence the physical.

But why it cannot? Arguments to the point are rather thin on the ground, and not very difficult to answer. John Beloff finds two arguments that “have often been put forward by sophisticated critics”, and argues that neither stands up to examination:

“The first, which troubled even Descartes, is that, if mind and matter have nothing in common, how can they even interact? Now the implicit assumption behind this objection can only be some such principle or axiom as: if A and B are cause and effect then A and B must have something in common (over and above their belonging to the same causal sequence). The question then arises: is such a principle a logical necessity, a necessity of thought? Or is it a universally valid empirical truth? Now, so far as I can see, no logical necessity is involved. For example, if an event A never occurred without being preceded by some other event B, we would surely want to say that the second event was a necessary condition or cause of the first event, whether or not the two had anything else in common. As for such a principle being an empirical truth, how could it be since there are here only two known independent substances, i.e. mind and matter, as candidates on which to base a generalisation? To argue that they cannot interact because they are independent is to beg the question.” (Beloff 1994)

Another objection to interationism is advanced in the form of the demand to explain how the mental can causally influence the physical. The simple reply is that on the fundamental level of causality, including any causality within the physical realm, no answer can ever be given to the how-question construed as the question about “the mechanism” of the causal influence; all we can do is to refer to a fundamental law of nature to which the causal influence at issue accords. Likewise, the epiphenomenalist dualist needs to presume that there are special laws of nature that bridge the physical with the mental, so that physical events in the brain can produce mental states. And nothing prevents the interactionist dualist from supposing that such psychophysical laws are a two-way traffic road.

Sometimes it is argued that if nonphysical mental events influenced physical events in the brain, it would violate the physical laws of the conservation of energy and momentum. However, this objection fails to take into account the possibility that the mind influences physical processes in such a way that there are changes in the distribution of energy and momentum without any change in their total amount.

A number of arguments against interactionist dualism can be found in David Chalmers’ book The Conscious Mind (1996). Although later Chalmers has admitted that the standard arguments against interactionism, including those he had advanced in “The Conscious Mind” (1996), carry little weight (see Chalmers 2003), I think that in the absence of better arguments it can be useful to discuss these. For it is probable that other philosophers deny interactionism on similar grounds.

1. Interactionism and science generally

Chalmers wrote that interactionism “requires a hefty bet” on the future of science that “does not currently seem at all promising”. In particular, it concerns physics: “physical events seem inexorably to be explained in terms of other physical events” and cognitive sciences: interactionism “suggests that the usual kinds of physical/functional models will be insufficient to explain behavior” (Chalmers 1996, p.156).

The plausible reply of the interactionist may be that the bet is quite realistic. Nothing in the development of science does guarantee, or even make highly probable, that all physical events inevitably and exhaustively will be explained in terms of other physical events. It just as well seems to many that the development of science will inevitably lead to an explanation of everything, including consciousnesses, in terms of physics – but Chalmers, despite this (and, in my opinion, quite rightly), argues that it is impossible.

As for the realism of the interactionist bet, I am to remark the following.

(1) Even if all physical events in the brain were really completely determined by other physical events, one could make a bet with confidence that science would not succeed in proving this. The brain is so complex a system that it is not, and very probably will never be, practically possible to calculate through all physical interactions in it and to compare the results of these calculations with what actually occurs in the brain. And this is the only possibility to prove that all physical events in the brain are really completely determined by other physical events.

(2) Likewise, irrespective of whether interactionism is true or false, we can make a pretty sure bet that cognitive science will never succeed in developing an adequate, functionally complete model of human behaviour. (An adequate, functionally complete model of human behaviour must meet the condition: it is possible to make a robot on its foundation and ascertain that in all essential aspects this robot behaves as a real human being. Or this may be a computer model – a virtual person in a virtual world, so that the virtual world quite adequately reproduces the variety and complexity of real-life situations, and we can from the outside assign any real-life situation we are able to imagine, and this virtual person would behave in all these situations as a real human person. I think that the complexity of this task exceeds very much all that cognitive sciences will ever be able to accomplish.)

(3) Even if such a model were created, this would not prove the falsity of interactionism. The only possible proof of the falsity of interactionism is the calculation and check as described in (1). It is logically possible (although, from the point of view of interactionism, very implausible) that those functions that in a human being are fulfilled by the mind can be fulfilled in a robot or a computer by something else. Even if a computer imitation of consciousness would be so good that its functioning could not be distinguished from the functioning of real consciousness, this would not mean that this imitation is consciousness (let us recollect J. Searle’s Chinese room). Interactionism does not assert that the functions that are fulfilled by the mind cannot in principle be realized differently, without the mind, by physical means. It only accepts the assumption that in the human being these functions are fulfilled by the mind.

On the above considerations, we may, with pretty good confidence, make a bet that the development of science will never refute interactionism – irrespective of whether it is actually true or not. An interactionist has even more reasons to make such a bet, because he believes (and has weighty reasons to believe) that interactionism is true.

2. Interactionism and quantum mechanics

Let us consider one interactionist hypothesis that Chalmers declines in The Conscious Mind. The hypothesis is that the mind interacts with the brain on the microlevel, filling (partially) the causal incompleteness that is left by quantum mechanics.

Quantum mechanics says that physical events at the microlevel are not unequivocally determined by the preceding physical events and physical laws, that in the same physical situation any of a range of possible microevents may happen with higher or smaller probability. Thus, one possible way the mind may act upon the brain is by influencing the distribution of probabilities of some microevents in the brain (this influence can be thought of as a special physical field dependent on the mind). Perhaps, the brain is organized so that a certain set of microevents, the probabilities of which are regulated by the mind, causes corresponding macroevents (in principle, the organization of a complex physical system may be such that even a single microevent may trigger one of several possible macroevents, or determine whether or not a certain macroevent will happen).

Chalmers raises two objections to this:

First, “the theory contradicts the quantum-mechanical postulate that these microscopic ‘decisions’ are entirely random and in principle it implies that there should be some detectable pattern to them” (Chalmers 1996, p. 157).

The plausible reply of the interactionist may be that the postulate is but a postulate. There are no reasons to contend that microscopic "decisions" are always entirely random. It may be that in the usual conditions (without influence of the mind) microscopic "decisions" are really random, but in the conditions of the presence of a mind they cease to be purely random, because their statistical distribution is influenced by the mind. Randomness can be regarded as a manifestation of the incompleteness of the physical causal determination; thus, we may suppose that this incompleteness leaves a place for a non-physical causality by the mind.

Second Chalmers’ objection is that if this hypothesis is correct, then the influence of consciousness on the statistical distribution of microevents in the brain must be such that this essentially influences the behaviour of a person, so that if the statistical distribution of microevents were not influenced by the mind but were really random, the behaviour would be essentially degraded. This supposition “is testable in principle, by running a simulation of a brain with real random processes determining those decisions” – for example, on some computer model. Chalmers suggests that “to hold that the random version would lead to unusually degraded behavior would be to make a bet at long odds” (Chalmers 1996, p. 157).

The plausible reply of the interactionist may be that it is hard to see how Chalmers calculated the odds. It is a pity that the supposition on the bet is testable only in principle, but hardly will ever become checkable in practice: an exact simulation of a brain with really random processes is so difficult that it is hardly realistic to expect that it will ever be accomplished.

3. “Subtraction” of the phenomenal component

Yet one argument of Chalmers is as follows:

“all versions of interactionist dualism have a conceptual problem that suggests that they are less successful in avoiding epiphenomenalism then they might seem... Even on these views, there is a sense in which the phenomenal is irrelevant. We can always subtract the phenomenal component from any explanatory account, yielding a purely causal component.” (Chalmers 1996, p. 157)

For example, if we suppose, following J. Eccles (1989, pp. 191-192), that there are interactions in the brain between physical elements and psychical units, psychons, then all that we need to describe and explain these interactions is their causal dynamics; the fact that psychons have, besides such dynamics, subjective experiences “is explanatorily irrelevant” (Chalmers 1996, p. 158).

Supporters of Eccles’ theory could object that it is the intrinsic nature of psychons to be bearers of phenomenal properties (states). However, even if this is so, these properties are irrelevant to explaining the interactions of psychons with physical elements. For such an explanation, psychons’ intrinsic nature is of no use to us; all we need is the causal dynamics of their influences on physical processes. Even if this causal dynamics is, somehow, a manifestation of psychons’ latent intrinsic nature, this nature is irrelevant to the explanation of the influences on physical processes; the intrinsic nature could be any other, or there could be no intrinsic nature at all – all that matters is that there was the same causal dynamics. If in the case of a phenomenal zombie (an exact physical and behavioural copy of a human being with no phenomenal mind) the same causal dynamics is realized by some other elements and processes that have no subjectivity, then, as far as the explanation of physical processes (behaviour) is concerned, there is no difference between the zombie and the human being.

We can imagine as well that all physical elements, or everything in nature, have this internal phenomenal-subjective aspect, and all physical interactions are actually accomplished at this level, as interactions between mental processes, whereas the physical picture of the world and of causality captures only their causal dynamics, leaving the internal nature off-screen. Chalmers even confesses that for him personally such an interpretation (panpsychism) is the most attractive. He remarks that this is not quite epiphenomenalism; nevertheless, as far as explaining physical processes is concerned, it is equivalent to epiphenomenalism.

Thus, according to Chalmers, there is no way to escape epiphenomenalism or such “almost epiphenomenalism” that allows the possibility that, on some deep latent level of the “intrinsic nature”, mental processes have a causal influence, but this is irrelevant to the explanation of really observable (physical) causal dynamics.

The plausible reply of the interactionist may be as follows:

1) There is a great difference between epiphenomenalism and the supposition that the mind influences processes in the brain and human behaviour. Although we can abstract causal dynamics from its bearer (the mind) and imagine some other bearer of the same dynamics (or even pure dynamics without any bearer), the important fact is that actually it is the mind that is causally efficient with the human beings. The possibility of abstracting causal dynamics from its bearers does not make those bearers causally inefficient.

2) Dualism-interactionism (though not the theory of psychons that just introduces unneeded mediators between the mind (self) and physical processes in the brain) has obvious advantages over panpsychism: it does not attribute (proto)mental properties to all physical elements while having no reasons for this and not the slightest idea as to what these elements’ mental states are (or can be) like (what is it like to be an electron?) and what these states have to do with physical processes. Dualism-interactionism admits the interaction between the mental and the physical only if the following conditions are satisfied:

– the mental (consciousness) is obviously present or we have weighty reasons to suppose its presence (in the case of human behaviour and, probably, the behaviour of higher animals);

– we know or conjecture the character of these mental states and processes, their content and dynamics;

– the relationship between the mental states and physical processes (behaviour) is known or may be conjectured by the analogy with the relationship between our own mental states and our own behaviour.

3) The supposition about the influence of phenomenal properties (mental, subjective states) on physical processes (behaviour) is irrelevant for causal explanations only in cases of assuming the existence of some phenomenal properties while having no idea as to what these properties (states), their phenomenal structures and dynamics are – i.e., in cases like panpsychism or the theory of psychons. However, this supposition is explanatorily very fruitful if we (introspectively) know or (by making a projection from introspection) conjecture about the phenomenal structures and dynamics, and know how they are correlated with the structures and dynamics of physical processes – and this is the case with the relationship between the human mind and behaviour.

The human mind has its own phenomenal structures and dynamics that we reveal introspectively. We know (assume in all real explanations and predictions of human behaviour) that these structures and dynamics guide behaviour, and we also approximately know what forms of behaviour correspond to certain structures and dynamics of the mind. As an example, we can take any behaviour aimed at the realization of our conscious desires and decisions. Structurally the richest example is an expression of our thoughts in speech – oral or written. I say (or write) what I think, because I think it and want to say (or write) it.

In fact, the supposition about the influence of the mind on behaviour always was and is the only efficient way to explain (understand) and predict human behaviour.

Objections to Epiphenomenalism

For me, epiphenomenalism seems a very extravagant theory in which nobody can believe seriously. Nevertheless, Chalmers, even after he has raised his estimation of interactionism and admitted that the traditional arguments against it are weak, continues to consider epiphenomenalism as one of the three (along with interactionism and russellianism) acceptable alternatives to materialism.

Epiphenomenalists recognise that the phenomenal mind (subjective experiences, thoughts, desires, and awareness) is non-physical, but they deny its influence on physical events. They accept the thesis about the causal closure of the physical. It only seems to us that our subjective sensations, emotions, thoughts, and desires cause our behaviour, but this is merely an illusion. In fact (as epiphenomenalism contends), our behaviour is entirely determined by purely physical causes that simultaneously generate our corresponding sensations, emotions, desires, and thoughts (including the illusion that our consciousness influences or causes our actions, the physical movements of our bodies).

I will marshal some known arguments against this view and develop some of them.

1) The theory of epiphenomenalism is practically unacceptable: all our conscious practical activities are necessarily based on the assumption that our desires, plans, and ideas can influence our behaviour (direct our actions toward the achievement of our desired purposes) and in this way change the physical world accordingly.

2) One of the most interesting and absurd conclusions that follow from epiphenomenalism concerns the significance of consciousness in biological evolution and human history. John Beloff formulated it as follows:

“For an epiphenomenalist, it can be only a brute fact that consciousness supervenes when the cortex of the brain is appropriately innervated. There is no conceivable reason why this should happen for it serves no purpose that would favour it from an evolutionary standpoint. Nothing whatsoever that makes a difference to what goes on in the real world follows from the supervenience of the mental upon the cerebral. We might just as well have evolved, therefore, as totally insentient automata… Moreover, if, at some future time, we were to make contact with intelligent aliens from another planet, we would have no grounds whatever for assuming that they, too, were conscious, no matter how knowing or sympathetic they might appear to us. For, given the fact that consciousness arose during the course of evolution on this planet, no inferences could be drawn with respect to evolution on some other planet – it being a sheer fluke that we ourselves happen to be conscious. Hence, if epiphenomenalism is true, we are forced to conclude that, but for this one unaccountable freak in our evolutionary origins, the whole of human history could have proceeded exactly as it has done but without anyone, anywhere, ever being aware (in the full sense of awareness) of anything that ever happened!” (Beloff 1994, p. 514)

Chalmers discussed a similar objection, “that if consciousness is epiphenomenal, it could not have evolved by natural selection” (as it cannot influence an organism’s survival in any way), and wrote that epiphenomenalism “has a straightforward reply”: “there are fundamental psychophysical laws associating physical and phenomenal properties” (Chalmers 2003, p. 128).

I think this straightforward reply is unsatisfactory in a number of respects.

First, in J. Beloff’s formulation, the argument is not only about the possibility that consciousness arises in the process of evolution by natural selection (in fact, Beloff does not directly deny such a possibility). The argument is much more substantial: it covers not only biological evolution but also “the whole of human history”. Its main point is: “Nothing whatsoever that makes a difference to what goes on in the real world follows from the supervenience of the mental upon the cerebral.”

Applied to human history, this means that everything would happen in exactly the same way as it has actually happened, if people (subjectively) felt nothing, had no desires, did not think, and had no awareness.

Phenomenal zombies (exact physical copies of human beings), who feel nothing, have no desires, do not think and have no awareness, would write scientific and philosophical treatises, novels and verses, symphonies and pictures – exactly as we do. A zombie whom other zombies would “name” (without being aware of doing it) "Plato" would write all the same philosophical works, with not even a single word or letter being different; these works would be “read” and “discussed” throughout more than two millennia by many other human-looking zombies although this "Plato" and all these numerous readers-zombies would understand absolutely nothing. A zombie named "Einstein", having no idea of what it is doing and what is happening around, would leave fanciful configurations of ink on sheets of paper, – and if we could compare these papers with the actual Einstein's manuscripts, we would find out, to our surprise, that they are absolutely identical. Other zombies would unconsciously print books on which covers we would see (if we could see them) the title “Relativity: the Special and General Theory. By Albert Einstein”; these books would be outwardly and inwardly exactly the same as the actual Einstein's books that really have been printed. In theatres, zombies-actors would unconsciously play roles in performances that would be attended by zombies-spectators; these spectators, although understanding and feeling absolutely nothing, would cry as if deeply moved and shout “Bravo! Encore!” as if delighted – exactly as in real theatres...

If “fundamental psychophysical laws associating physical and phenomenal properties” really exist, then it is absurd to think that if these laws and, hence, consciousness, did not exist, everything would happen as described in the previous paragraph.

Second, epiphenomenalists have no reason to believe that such laws exist. Since consciousness does not play any role in the process of evolution, its emergence may be something accidental that happened only in the evolution on the Earth.

We do not know what factors are responsible for the existence of consciousness. So if consciousness does not play any role in the processes of evolution and history, then in slightly different conditions the evolution could occur absolutely without consciousness and lead to the same results – something very like the zombie world described above. So J. Beloff was perfectly right that if on some faraway planet there are beings who have developed a very advanced civilization, culture, and science (perhaps even more advanced than ours), then, if epiphenomenalism is true, we have no reason to think that they have consciousness. Suppose these beings have a slightly different organisation of the brain; in this case, even if there are laws of nature that associate consciousness with physical processes occurring in the human brain, it would not mean that these laws also generate consciousness in these beings.

Or maybe no such laws exist: consciousness is a sort of peculiar spiritual virus that parasitizes on the physical processes in human bodies but somehow manages not to influence these processes. In this case, if some other planet is inhabited with living beings with precisely the same bodily organisation as human beings, and they create the same philosophical and scientific treatises, novels, verses, symphonies, and pictures, but (it has so happened) there is no such spiritual virus on that planet, then these beings are zombies with no consciousness.

Actually, there is no reason to even think that this accidental event peculiar to the evolution on the Earth has really happened – that Plato, Einstein and other people (with the one and only exception − you, my dear reader) have consciousness. If epiphenomenalism is true, then I (you) have no reason to think that anyone besides me (you) has consciousness: perhaps I am (you are) the only conscious being in the Universe. This is not a joke, and to see this, we will make a deviation to find out why it is reasonable to believe that other people have consciousness.

The Problem of the Existence of Other People’s Minds

Why is it reasonable to suppose that other people have consciousness? Maybe they all just behave as if they have consciousness but actually are phenomenal zombies.

We are not to consider as a proper reason a simple analogy of the kind: they are very similar to me in other respects, they behave like me; I have consciousness, therefore it is likely that they have it too. Such reasoning is obviously invalid. I have two feet and two hands, and I am a dualist. Would it be right to draw, by analogy, the conclusion that everyone who has two feet and two hands is a dualist? Also, there is no question of inductive inference (even if there were such a thing), because I am (you are) the only person of which I (you) directly know that he-or-she has consciousness; such a foundation is too narrow.

Newertheless, I think (and hope you agree) that there are very weighty reasons to suppose that other people have consciousness, and they consist of the explanatory job that the supposition does.

1) The hypothesis that other people have consciousness is the best of all ways known to us to explain and predict human behaviour in different situations. We cannot understand the behaviour of other people without this hypothesis. Really, you can profess the philosophical theory that all forms of human behaviour are just physical and chemical processes in human bodies, but when you need to understand or predict a concrete behaviour of a concrete person in a concrete situation, this philosophical theory will not help you at all. You will turn not to physical calculations or chemistry but to reasoning based on the assumption that the person has thoughts, feelings, desires, and conscious purposes. The behaviour of a person can be understandable and predictable for us only on the basis of (conjectural) knowledge about his feelings, desires, ideas, and thoughts.

2) If we deny the assumption that other people have consciousness, our explanations of their behaviour (in particular, their speech: why a person speaks as if he-or-she feels and understands) should parasitize on this assumption. We know too little about the relevant physical (chemical) processes in the body (brain) to really explain and predict human behaviour on this basis; therefore, all such "explanations" will be of the kind: in the body of this person, there are biochemical processes that determine behaviour as if he has consciousness, feels X, thinks Y, desires Z, etc. An explanation that only parasitizes on another explanation is obviously not a genuine explanation. The genuine explanation is the one it parasitizes on – in our case, an explanation based on the assumption that other people have consciousness. Admittedly, a non-genuine explanation, which explains nothing, may in principle be true, while a genuine one may be false. Although no guarantees of truth exist, it is reasonable to accept a theory that gives a genuine explanation rather than a theory that parasitizes on it.

3) Moreover, without the assumption that other people have consciousness, I cannot understand even my own consciousness. I know that my consciousness has not always been (from the very moment of my birth) as it is now. It has developed, beginning with some rudimentary ability and gradually forming a complex system of ideas expressible in language. I know that I have not invented all these ideas and language. I have learned language and the largest part of these ideas from other people.[1] I understand the meanings of words, statements, ideas, and theories, and I have learned these from other people in the process of communicating with them. But this is possible only if they themselves understand the meanings of words, sentences, ideas, and theories and can convey (communicate) their understanding to me. They would not be able to convey this understanding to me if they themselves understand nothing. And if they understand something, it means that they have consciousness.

4) A traditional argument, in Karl Popper's formulation:

“If we talk to other people, and especially if we argue with them, then we assume (sometimes mistakenly) that they also argue: that they speak intentionally about things, seriously wishing to solve a problem, and not merely behaving as if they were doing so. It has often been seen that language is a social affair and that solipsism, and doubts about the existence of other minds, become self-contradictory if formulated in a language. We can put this now more clearly. In arguing with other people (a thing which we have learnt from other people), for example about other minds, we cannot but attribute to them intentions, and this means, mental states. We do not argue with a thermometer.” (Popper 1953, p. 105)

Criticisms of Epiphenomenalism Continued

If epiphenomenalism is true, all the above reasons for accepting the hypothesis that other people have consciousness are void.

1) An explanation of the behaviour of people by reference to their consciousness makes sense only if consciousness influences behaviour. But epiphenomenalism denies this.

2) Epiphenomenalism means that the parasitic explanation of human behaviour is true: people behave in certain ways not because they feel, think, and desire something but because the physical and chemical processes in their bodies cause such a behaviour that is as if it were caused by their sensations, thoughts, and desires. Yet these processes and behaviour would be exactly the same if people felt nothing, did not think and had no desires.

3) Epiphenomenalism means that our learning from other people does not depend on their having any ideas and understanding whatever: it has to do only with the physical movements of their mouths, tongues, lips, hands (holding and moving a pencil), etc., and all these movements occur automatically, quite independently of their consciousness. Therefore, the supposition that they have consciousnesses is needless. Whether they have it or not, they would speak and write exactly the same; so their having or not having consciousness does not influence my learning from them in any way.

4) Epiphenomenalism means that we talk and argue with other people not because we think that they understand us and argue with us, communicate their thoughts and arguments, etc. but because physical and chemical processes in our bodies (whether we want it or not) force our mouths to open and make our tongues and throats produce certain movements that cause certain vibrations of air (sounds)...

So if epiphenomenalism is true, I (you) have no reasons to think that other people have consciousness, and vice versa: if you think that you have good reasons to believe that other people have consciousness, you should recognize that epiphenomenalism is mistaken.

If epiphenomenalism is true, I (you) have no reasons to believe that other people have consciousness (although, certainly, there are some causes that make me (you) inclined to such a belief). Or, if I (you) suppose that other people have consciousness, there are no reasons to think that their behaviour corresponds to their consciousness so that they speak and write what they think and do what they want. Maybe Daniel Dennett, in his conscious mind, is a dualist and is sure that all his writings prove the existence of the soul and its influence on the body, while David Chalmers, in his conscious mind, is a materialist-eliminativist and believes that in his books he proves that consciousness does not exist. All these are but illusions produced by some physical processes in their brains, while in fact their hands write something quite different. Or perhaps it just seems to me, owing to some physical processes, that I am now writing about interactionism and epiphenomenalism, while actually my hand is writing some treatise on biochemistry, or perhaps I am (my body is) now sunbathing on a beach in Miami. If epiphenomenalism is true, I have no reason to think that the ideas that physical reality produces in my consciousness correspond to reality rather than that all these ideas are but illusions.

There is but one fundamental reason to believe that other people have consciousness and that my and their consciousnesses correspond with our behaviours – the assumption that our conscious minds influence our bodies allows us to explain or predict a person’s actions in various situations on the basis of some knowledge and guesses about what that person feels, thinks, and desires. But if there is no causal link from consciousness to behaviour, then there can be no genuine explanatory link. People would behave exactly as they do even if their consciousness was not correlated with their behaviour, or if it was exactly opposite to what I am prone to think about it judging by what they do, say, and write, or if people had no consciousness at all.

All our ideas about physical reality (besides that it produces feelings and thoughts in our minds) also have to be recognised as groundless. All that you know about is your mind: what you feel, desire, and think. As far as external reality is concerned, if you accept epiphenomenalism, you have no reason to believe it to be this or that way, except that it produces your subjective mental states, including your ideas about itself. But you have no reason to think that the ideas that external reality produces about itself in your mind are (by and large) right, correspond with reality, or progress in that direction. Perhaps external reality produces in your mind nothing but illusions, including illusions about itself.

The rationality of our ideas about physical reality cannot be based only on the supposition that they are produced by this reality. It is based on

1) our ability to check our ideas by making relevant observations and experiments, – but if epiphenomenalism is right, then we cannot make observations and experiments on our will (to do so, our consciousness must be able to direct movements of our hands, head, eyes, etc.)

and

2) more than anything else, on the intersubjective process of exchanging and checking ideas, – but this process is possible only if we are able to communicate (orally and through writing) our ideas (the ideas of our consciousnesses) to one another, – and this contradicts epiphenomenalism.

Surely, an epiphenomenalist can believe that physical reality produces in his consciousnesses correct ideas about itself, but this would merely be a belief having no rational grounds. There is no reason at all why physical reality has to produce in our minds correct ideas rather than illusions. Epiphenomenalists have an extra reason to think that these ideas are illusions since epiphenomenalism means that the idea that our minds influence our bodies – although it is an idea that we are all very much disposed to believe and the denial of which seems an outrageous absurdity – is just a great illusion produced in our minds by physical reality.

David Chalmers wrote that, in his judgment, “there is no knock-down objection to epiphenomenalism” (2003, p. 129). I am not sure what he means by “knock-down objection”. But I think that the above arguments are sufficient to show that it is unreasonable to accept epiphenomenalism.

T. Nagel has wisely remarked:

“…to create understanding, philosophy must convince. That means it must produce or destroy belief, rather than merely provide us with a consistent set of things to say…” (Nagel 2008, p. Xi)

I do not deny that epiphenomenalism can be “a consistent set of things to say” (and this is the only sense I know in which “there is no knock-down objection” to it). I wonder if any of those philosophers who advanced and defended epiphenomenalism really believed that he writes what he writes not because he thinks what he thinks, and not because he wants to write it? Is this not a “knock-down objection”?

To use Chalmers’ own fair remark, “highly counterintuitive claims … need to be supported by extremely strong arguments” (Chalmers 2003, p. 110). The claim that our minds do not influence our behaviour is counterintuitive to the highest degree and is not supported by strong arguments.

References

Beloff, J. (1994). “The Mind-Brain Problem”, The Journal of Scientific Exploration, Vol. 8, No. 4, pp. 509-22.

Chalmers, D. (1996). The Conscious Mind, New York, Oxford: Oxford University Press.

Chalmers, D. (2003). “Consciousness and its Place in Nature”, In Stich, S. & Warfield, F. (eds.) Blackwell Guide to Philosophy of Mind, Blackwell, pp. 1-45.

Eccles, J. C. (1989) Evolution of the Brain: Creation of the Self. London and New York: Routledge.

Nagel, T. (2008). Mortal Questions, Cambridge University Press.

Popper, K. (1953) “Language and the Body-Mind Problem”, Proceedings of the 11th International Congress of Philosophy, Vol. 7, pp. 101-107.

Popper, K. (1996) Knowledge and the Body-Mind Problem. London and New York: Routledge, 1996.

Popper, K. (2008). “Language and the Body-Mind Problem”, in Popper, K. Conjectures and Refutations, London, New York: Routledge, pp. 395-402.

---

[1] Cf.: K. Popper:

“When reading Shakespeare, or hearing any of the great composers, or seeing a work of Michelangelo, I am very conscious of the fact that those works go very, very far beyond anything I could ever produce. But according to the theory of solipsism, only I exist – so that in dreaming these works I am, in fact, their creator.” (Popper 1996, p. 107.)

[Return to Dmytro Sepetyi’s Personal Page]