by Ted Honderich 

-- The Determinism and Freedom Philosophy Website --

This a piece mostly about one by Searle in an issue of the Journal of Consciousness Studies. The issue carried the title Mind the Gap. Persons who have not had the satisfaction of riding on the London subway may not know that that title is an echo -- of an awful instruction by loudspeaker to passengers stepping off the train at some stations where there is an empty space between the train and the platform. My title gives another warning. This piece too has now been published in the Journal of Consciousness Studies. Searle's piece, by the way, reappeared in his Royal Institute of Philosophy Annual Lecture in February 2001.

Abstract: (I) John Searle's conception of consciousness in the 'Mind the Gap' issue of the Journal of Consciousness Studies remains short on content, no advance on either materialism or traditional dualism. Still, it is sufficiently contentful to be self-contradictory. And so his Biological Subjectivity on Two Levels, like materialism and dualism, needs replacing by a radically different conception of consciousness -- such as Consciousness as Existence. (II) From his idea that we can discover 'gaps', seeming absences of causal circumstances, in our experience of deciding and acting, Searle is led to the positing of a self and to mysterious causing. (III) In fact philosophers of determinism and freedom over three centuries have concerned themselves with what are now termed 'gaps'. Searle's advance is a useful terminological one. Compatibilist philosophers of freedom, contrary to what is said, have not missed any point at all. A successor to both Compatibilism and Incompatibilism is needed. (IV) Searle's previous account of deciding and acting in Biological Subjectivity on Two Levels does indeed fail because of its epiphenomenalism. (V) The culmination of his paper, his preferred hypothesis now about deciding and acting, is that down-up causation is true of it but not left-right causation. Quantum Theory as often interpreted doesn't work down-up but does work left-right. The hypothesis is entirely in the tradition of the Incompatibilist and Libertarian philosophers of determinism and freedom, whom Searle has joined, but is factually incredible. 


John Searle is not only the king of American stand-up philosophers. No one has been stronger and clearer-headed in resisting the illusion that consciousness is yet less than ghostly stuff -- just causal or logical relations expressed in computer programmes or the less formal propositions of this or that Functionalism (1984, 1992, 1997, 2000a). In his paper in the recent 'Mind the Gap' issue of the Journal of Consciousness Studies, however, he offers us other ideas (2000b). They have to do with consciousness and freedom. 

Let us think about them in terms of an example of a typical sequence of experiences: (a) you see the dining room as you come in, (b) you then consider reasons for and against making your customary little speech at the end of the dinner, (c) you form the inactive intention not to -- i.e. you decide against, (d) later you remember seeing the room as you came in, (e) when the moment comes for a speech, you actively intend to remain seated, and do so, and (f) you go on in this active intention, concentrating on your pudding, until it is too late.1

I: Consciousness as Biological Subjectivity on Two Levels, etc.

Searle starts out his paper by telling us again that each of these various experiences, like all of consciousness, is a real biological phenomenon, an ordinary part of the physical world, but that it is also subjective, qualitative, and unified (pp. 3-4). These possibly conflicting descriptions of consciousness do not answer the question they prompt us to ask, and the question most of us want well-answered anyway. It is whether consciousness has only neural or electrochemical properties, and in particular whether Neural Functionalism2, which is Functionalism applied to humans and other animals, is true. Searle's possibly conflicting descriptions do not give us a view or argument about that, as can readily be shown.

It is no help being told, for example, that consciousness is subjective in the sense that it has a first-person ontology, that it exists only as experienced by some 'I', some human or animal that has it (p. 4). If an 'I' is a person or the like as standardly conceived in cool philosophy, which is to say a persisting body and/or an internally-related sequence of conscious events, then conscious events can clearly be wholly neural events. This is so since these events with only neural properties can have exactly the given dependency. If on the other hand the 'I' is something on the way to a Cartesian ego, or any other sort of thing beyond our usual ken, a matter of some traditional dualism, then the dependent items will presumably have some non-neural character. 

But the main point is that until we really hear about the 'I', we are left not knowing what character of subjectivity these conscious events are supposed to have. Is what we have so far not too close to empty talk? Certainly it burkes rather than answers questions.3

Searle does no better by telling us again that conscious events are higher-level brain events caused by and realized in lower-level brain events (p. 4). There is a little uncertainty introduced by the lower-level items being described in a down-to-earth way as 'neuron firings' or as 'neurobiological' and the upper-level items described a little more airily as 'features of the system of neurons that constitutes the human brain'. But you have to suppose that the message is the familiar one drummed in about levels, by way of several examples, not about two kinds of things in any other sense. Certainly no other kinds are mentioned here, let alone clarified. So our two kinds of events, the conscious ones and the ones down below, are to be understood as like the liquidity of water as against the molecules, the solidity of the piston as against its molecules, and the running of the engine as against movement of the piston, and so on.

But then the trouble is that very many wholly neural events are exactly higher-level effects of lower-level neural events. Any textbook of neuroscience illustrates the fact endlessly. Even neuron firings are higher-level events with respect to various lower-level events. So this talk about levels in characterization of consciousness does not distinguish conscious events. It does not let us know whether conscious events are wholly neural events or not.

As in the case of the possibly conflicting descriptions of consciousness as subjective and so on, you can't actually decide whether this is really the reductionism or materialism, including Neural Functionalism, to which Searle says he is profoundly opposed, or really the property-dualism to which he says he is also profoundly opposed, or something merely inchoate. Or maybe a determination not to get involved philosophically -- involved logically with certain questions that are real but not open to commonsensical answers? He is surely being hopeful, by the way, in thinking that his doctrine is controversial because it isn't materialism or dualism, and so doesn't satisfy expectations (p. 3). There might be another reason or two, my old buddy.

The levels idea is not so short on content that it is not open to a refutation. Before it is filled in, if that happens, you can already see it will be wrong. We have it that there is something called higher-level activity of the brain, also known as consciousness, and there is also something called lower-level activity, also known as neuron-firings and the like. So think of the particular bit of higher-level activity that is (a) in our example above -- your seeing the dining room as you come in. It goes with a causal circumstance or sufficient cause for it, some simultaneous lower-level activity -- say La for short. Think also of some other lower-level activity later on, Ld, causing its simultaneous bit of higher-level activity, (d) your remembering seeing the room as you came in.

Epiphenomenalism threatens when it's added in, predictably enough, that La or maybe something wholly neural including La is also a causal circumstance for the later Ld -- this later event, as just remarked, being cause of the simultaneous remembering. The earlier seeing is left entirely out of the story. That perceptual experience, bizarrely, is no part at all of a full explanation of the remembering. To deal with this, you may be tempted traditionally to declare, as Searle always has, that the earlier Laone thing -- like the molecules and the water and all that -- and so to declare that the earlier consciousness does cause the later after all. No dualism and so no epiphenomenalism.  and the visual experience are

But the temptation overlooks something, the large proposition about the mind-body relation firmly asserted in this theory, and of a general sort that is certainly needed. The proposition is indeed that the earlier La causes the simultaneous perceptual consciousness that goes with it. In which case they certainly can't be one thing. Causation, whatever else it is, is a dyadic relation. No account of it does or can say otherwise. It doesn't matter, of course, that Searle's effect is also said to be 'realized' in the cause, or within the stuff of the cause, or in any other secondary sense 'one' with the cause. The commonest cases we have of one thing realized in another are such that the realized thing (say temperature) is an effect of and therefore numerically different from the other thing (say a lighted match).

So this two-level monism or identity theory of consciousness and its relation to the brain will always be a contradiction. You can't have the left-right or across-time causation you want and try to save the theory from epiphenomenalism by declaring it to be a monism -- and also have the down-up causal connection at a time without making the theory a dualism. You get the same disaster, of course, if you think not of seeing a room and remembering the seeing, but of (b) considering reasons for and against a speech and (c) deciding against. If this is what is emerging as the conception of consciousness in philosophy, cognitive science and neuroscience (p. 3), we better all give up and go bird-watching.

The whole doctrine of Biological Subjectivity on Two Levels as we first have it (pp. 3-4) is a kind of fundamentalism in Searle's paper, but several things happen to it, or somewhere around it. We are first reminded (pp. 4-6) that all of our consciousness at a time, called our conscious field, can be divided into parts or bits, and that neuroscience does this, and then looks for the neural correlates. There is only slow progress here, we are told. So neuroscientists and the rest of us should start thinking of all of somebody's consciousness at a time as a unity, and concentrate on just the differences between the conscious brain and the unconscious brain -- presumably absolutely general differences in consciousness between anytime when he or she is conscious and anytime when they are not. That is, with respect to consciousness itself, we should take 'the unified field approach'.

Well, you might say it would be a pity, particularly at the moment, if neuroscience gave up much of its main endeavour, neural localization with respect to parts of consciousness. It's doing pretty well, except in one or two Californian mystery-labs. So you can wonder if it really should spend more time -- in effect is spends a lot already -- on the neural situation for any consciousness. But you might also wonder, a lot more, about something else.

Are we to suppose there is something new and different for philosophers in the unified-field approach to consciousness? Seems so. It was announced to a good conference in Tucson before the 'Mind the Gap' paper and then also in the Royal Institute of Philosophy Annual Lecture for 2000-20014. The approach, on reflection, is in fact the assumption that consciousness has a general nature you can think about. The approach cannot be just the reminder, of course, that the bits of our consciousness at a time come together or bound up together, not in separate theatres or whatever. That the taste of the coffee is somehow in the same show with the feel of the shirt on your back. Indeed that obscure old Kantian fact isn't an approach in the relevant sense.

But the assumption that consciousness has a general nature is already an assumption made by almost all philosophers now at work on the subject. So isn't the unified-field proposal to philosophers just talk about what they are doing already? You aren't put off that disappointing thought about the proposal, certainly, when you hear that the unified-field approach to consciousness is such as to have a certain consequence -- that you may become completely conscious when you wake up from a dreamless sleep, even if you're not thinking or feeling much in particular, not conscious of various bits and pieces. What has that consequence is just that consciousness has a general nature.

Am I maybe missing out some further content in this unified-field stuff -- some proposal or advance, some move beyond a recommended concern with the general nature of consciousness? Is there something of an actual conception of that nature on offer? Could be. It could be in some additional sentences you can't miss, 

There is a metaphor we are advised to take seriously, something that goes beyond what we have on hand, the more or less literal talk of a conscious field. We are to think of consciousness not as a house made up of separate bricks, but as a great open field, pre-existing, with bumps or hillocks or ravines coming and going. There may be some homespun poetry in that, but it's not a lot of help to me. It doesn't do more for me than those other images of consciousness. The stream, the proscenium arch, the bottomless pit, the space behind the eyes, and so on. 

Nor does the metaphor do anything to deal with the earlier obscurity in Biological Subjectivity on Two Levels. It does not individuate that doctrine, make it something different and definite -- any more than the literalness of the unity-of-conscious-field addition did. No doubt Professor Dan Dennett and his reductionist forces can accomodate into their theories a bump or two and a couple of hillocks (1991). It is worth adding, too, that the literalness of science is a condition to which philosophy should aspire whenever possible. Maybe that will be the condition of the coming book by Searle of which the paper under consideration is an anticipation.

Is there help for me in sentences to the effect that the unity and the subjectivity and the qualitativeness of consciousness are not distinct features of consciousness but different aspects of its one essential feature, the very essence of consciousness (p. 6). OK, but I'd actually like some unwrapping. I'd like to know what that essence is -- or what this talk comes to. Isn't that what philosophy is supposed to do? The Holy Trinity can properly remain a mystery, but can consciousness in a philosophy of consciousness? Let us avert our eyes for a while. 

It is not only Biological Subjectivity on Two Levels that runs into epiphenomenalism, of course. Another theory of consciousness, Neural Causation with Psychoneural Correlation, is more developed but about as traditional, and is also wrecked by making the mental inefficacious (Honderich 1988, pp. 154-163). This theory puts in place of the down-up causation at a time another relation of nomic or lawlike connection. The theory asserts about a neural event taken as a nomic correlate of a conscious event, say of (c) deciding not to make a speech, that the neural correlate is fixed by a causal circumstance including the earlier neural correlate of a relevant conscious episode, say (b) considering reasons -- but not affected by that conscious episode itself. 

To say much the same thing differently, the theory asserts about a neural property, correlated with a conscious property of the same event, that the neural property is fixed by a causal circumstance including another neural property of an earlier event with a relevant conscious property, but not that relevant conscious property itself. The nomic relation between the neural item and the simultaneous conscious item is many-one.

This attempt, which deals with other objections to all such theories, must indeed fail because of the contained epiphenomenalism. Again there is a complete explanation of the later conscious event or property, say the remembering or the deciding, that leaves out the relevant earlier consciousness, say the seeing or the thinking of reasons.

One alternative traditional theory, the Hypothesis on the Causation of Psychoneural Pairs, does escape epiphenomenalism. It takes the earlier causal circumstance as including both (b) and its neural correlate Lb, and takes its effect as in a sense indivisible -- the later psychoneural pair consisting in (c) and Lc. The second bit rules out a complete explanation of (c) by a causal line from the earlier causal circumstance by way of just Lc. But this takes a little philosophical bravery -- more than bravery if it includes thinking of (b) and (c) in terms of traditional dualism In fact there is no need to think of non-physical stuff. (Honderich, 1988, pp. 163-175)

All of these three traditional views persist in the idea, hitherto not thought about enough, that consciousness is something or other in the head. The same is true of traditional dualisms generally, and also traditional materialisms, including Neural Functionalism. The latter is not the illusion that consciousness is merely causal or logical relations, which illusion makes consciousness into yet less than ghostly stuff. Neural functionalism is rather the proposition that consciousness is real neural items in such relationships.

Is your being conscious anything actually in your head? Doesn't that mislocate it? Is it not a state of affairs -- more than a cranial state of affairs? 

Compare a radical view of consciousness that is worth attention not just on account of the particular embarrassments lately noticed -- a theory that hasn't grown up into a theory at all, epiphenomenalism, seeming mislocation -- but on account of the widely accepted hopelessness of traditional dualism, and the widely suspected hopelessness of traditional materialism, and thus the need for a really new start in the philosophy of mind, a paradigm-shift.

You can think, on the ground that what is physical is at least the exemplar of what exists, that consciousness will, so to speak, be of the same order or type as the physical. An arguable conception of consciousness will will make it a matter of something like physical things themselves. The two sorts of these are (1) space-occupiers that are generally perceived by all of us and (2) space-occupiers that are in causal or other nomic connections with space-occupiers generally perceived by us. Tables are of the first sort, atoms of the second. Our apparatus of sense-perception is part of the fuller story of the physical, as generally agreed since Locke wrote on primary and secondary properties, as is our conceptualizing part of that story. In short we make a contribution to a thing's being physical.

The radical view (Honderich 2000) is of the nature of only perceptual consciousness. It is prompted by what is arguably the most natural answer to such a question as this one: What is it for you to be perceptually aware of the room you are in? An answer is that it is for the room in a way to exist. In general, what it is to see or otherwise perceive something is for a thing somehow to exist in space and time. That is what seems to be true of your present awareness, isn't it? What is essential, of course, is that this idea about your awareness can be explicated, given literal sense. It must also be shown to be non-circular, of course. It cannot reduce to the poetry that to see the room is for the room to exist in your consciousness. 

The idea can be given literal sense. It can be explained that the room of your perceptual consciousness, out there is space and time, has a unique dependency not on perceivers generally, as in the case of the first sort of physical things, but on your brain in particular. It can be explained, also, that the room of your perceptual consciousness shares something with the first sort of physical things -- in short a dependency on the atoms.

Those two remarks are a main part of the explanation of what it is for a thing 'in a way to exist', but there is more to be said. The view of perceptual consciousness as a kind of existence gives an explicit account of the fact of subjectivity of this consciousness. What this comes to is that there are factual differences between your world of perceptual consciousness and mine when we are together in the same room, and, more important, between both our worlds and the room considered as within the perceived physical world. The first two involve points of view, and different points of view. Full and exact sense is given to the contrast between the subjective and the objective.

The radical view includes a clear proposition, further, having to do with what has always stood in the way of traditional dualism -- the mind-body problem. It is arguable that there can be standard causal connection between your perceptual world, on account of its being in space and time, and anything in the physical world, notably your body and brain. Also, the view does not fall victim to what has been preoccupying us, the curse of epiphenomenalism. Certainly my perceptual consciousness, on this view, is an ineliminable part of any explanation of my later remenbering, deciding or whatever. Finally, the view has in it some news about perceptual consciousness and that supposed fact of consciousness so turned over by so many philosophers, the supposed relation of intentionality. That is a relation between a thought or whatever and a thing, which dyadic relation can wonderfully persist despite the non-existence of the thing (Honderich, 2001b).

This account of the nature of perceptual consciousness needs to have added to it and account of reflective consciousness, roughly thinking without perceiving. Conceivably this will be in terms of possible worlds. Also an account of affective consciousness, having to do with desire, emotion and the like. So the view is incomplete, and needs to be built upon. Still it is a contentful start, and not a non-starter. 

It is not put forward as the only possible truth about perceptual consciousness and related things. It assumes that there is an ultimate reality, something like Kant's noumenal world, to which perceptions and conceptions are directed. One result of the meeting is the traditional outlook that has in it the physical world in its two parts and also consciousness in the head. The view of consciousness as existence is in clear ways more satisfactory. It leaves consciousness in relation with the physical world, but a different relation. If the view subtracts something from our traditional conceptual scheme, it subtracts what can be argued to be an illusion, and also puts something back. The view is out of sight of traditional dualism. If naturalism is a strict restriction to the physical realm in thinking about what there is, this is near-naturalism.

The radical view may be too strange to get prolonged attention in the present state of the Philosophy of Mind, but let me remind you again that the familiar is not working. The radical view may not be right, or get attention in the future, but at the very least it illustrates the possibility of contentful reflection that takes us beyond materialism, dualism and of course the indeterminacy of Biological Subjectivity on Two Levels -- to which we now return.

II: The Consciousness of Deciding and Acting

From the unified-field approach to consciousness, and the field with the bumps coming and going, and the three-part essence, we are led on by our guide, although understandably he does not say so, to what is presumably a good reason for precisely not getting too serious or single-minded about the unified-field approach. This is a supposed big difference (p. 12) between two things. One is what we have just been considering, perceptual consciousness, as in (a) seeing the dining room. The other is the consciousness of considering reasons for an action, deciding on it, initiating it, and carrying it out -- say (b), (c), (e) and (f). What you then need, presumably, to the extent that you believe in the big difference, is precisely not a global approach. But forget about that.

With this consciousness of deciding and acting, the interesting point and a remarkable fact we can discover is that there are 'gaps' (pp. 6-8). There is a gap between stages (b) reasons and (c) decision, and between (c) and (e) active intention or volition, and between (e) and (f) carrying on forward. That is to say that whatever the facts really are, we ourselves do not experience the prior stage as being a causal circumstance or causally sufficient condition for the later. We do not 'experience the causal relation' that way (pp. 6-7). We feel, on the other hand, that we or our experiences or our efforts are in some sense causing the later stage. We have a choice in the matter, the later stage is up to us, we are active rather than passive. This, as some other philosophers say, is the phenomenology or felt nature of the the thing. In perceptual consciousness, on the other hand, there are no gaps. What we see, maybe a room, is experienced by us as causal with respect to the experience.

This account of the felt nature of deciding and acting, we are told, has been taken to raise a question about the explanation of decisions, intentions and actions (Nagel, 1986, p. 116). Indeed it has been so taken, by many philosophers. It has been taken to raise a question, of course, because if the supposed felt experience is true to the facts, then decisions, intentions and actions do not have ordinary causal explanations. They do not have the explanations of the kind assumed so far in this paper. We are now told by our tutor, however, that the gaps story, the seeming absence of ordinary causal explanations, can be true consistently with there being perfectly adequate explanations of decisions, intentions and actions -- perfectly adequate explanations of another type.

This other type of explanation involves reasons. It involves, secondly, something called a self in a person (p. 8). A self is an entity, certainly more than a Humean bundle of impressions, and indeed more than an agent, since it thinks before it gets into action. This other type of explanation, thirdly, involves a kind of causation other than the ordinary kind. (p. 7) To bring these together, the explanation of why you decided against a speech, didn't stand up at the right time, and went on with your pudding, is of this form: a self S performed act A because S was acting on reason R (pp. 8-10).

More questions are raised by this story of an explanation of our deciding and acting than our tutor supposes. The first question is best introduced by a brief account of our ordinary or standard account of causation -- ordinary or standard causation as assumed so far. The match was struck and it lit. The striking was the cause, a required or necessary condition for the lighting, and the striking together with other things was a causal circumstance for the lighting. That is, the circumstance necessitated the lighting -- if the match hadn't lit, we would have concluded that some bit of an assumed causal circumstance was missing. The main philosophical problem about this standard causation is exactly how the relation of necessitation is to be understood. Hume said the circumstance was related to the lighting only in that all things like the first are followed by things like the second. Other views have improved on this, usually by explaining necessitation in terms of conditional connections (Honderich, 1988, Ch. 1).

I introduce this sketch only to set aside clearly what Searle is not saying. As reported, he does of course allow that his supposed explanation of an action does presuppose some causal connection, is implicitly an explanation in terms of some causal relation. It is not ordinary or standard causal connection. What is it then?

Will an innocent and kinder reader than I interrupt and say that that challenging question is mistaken? Will she say that the clear relation in question is the ordinary relation between reasons and what they are reasons for? In fact the two relations of entailment and inductive support? This will not do, of course. These are relations between propositions, abstract entities, not at all explanatory in the sense that explanations of events are explanatory. That I believe p, and p entails q, does not at all explain that I believe q if in fact I do. In fact, with respect to many premises we believe, we fail to believe their logical consequences, but believe things inconsistent with them instead.

Reasons in another sense, as ordinary, are what you can call beliefs in propositions or propositional beliefs, and it is these that Searle has and must have in mind. It is these that with the aid of a self 'cause' things. So we have the question again. What is this causing?

Has it sometimes been supposed that all that is needed for an explanation is a cause in the sense of a required or necessary condition? Certainly it has sometimes been supposed that a cause is no more than this -- with no causal circumstance being part of the story. (Anscombe, 1971) But of course if there actually is no causal circumstance, there is no real answer as to why a thing actually happened. The same is true if a cause is taken to be something that makes something else probable, again with no causal circumstance being part of the story. There is no real answer to the question of why the thing happened because, ex hypothesi, everything could have been exactly the same before and at the time of the thing happening, without the thing happening.

What answer is there, then, in Searle's story of deciding and acting, to the question of why something happened? Can you go on talking about causation without getting it clear, on the assumption that we all have some hold on it, and trust that you can sort out any difficulties later? In effect abandon the relatively clear if philosophically disputed concept we have, and gesture in the direction of something else? Not if things depend on what your talk comes to. Not in finished philosophy you can't. Not in the real thing.

Nor is there any reason to think that a satisfactory account of funny causing will be forthcoming. The whole history of several large parts of philosophy has failed to produce one. That particular part of the philosophy of mind that has told us of a self and its activity has certainly failed to produce one. Popper was not only being light-hearted but was pretty true to his and Eccles' stated views when he said a self might choose what hemisphere of a brain to light on and make dominant (Eccles & Popper, 1977, p. 507). He did not say anything of the causation involved. In the philosophy of determinism and free will, to which we are coming, a long line of philosophers has spoken of a self originating decisions and never been able to explain satisfactorily.

The objections to the idea of a self itself, if you can separate it from its activity, are too plain to require rehearsal. A proof of the self's existence (pp. 9-10), if it had no other weakness, would still be vitiated by these objections. We do have a kind of image of something elusive about us and of its activity, important with respect to the matter of determinism and free will, but an image is not philosophy. The philosophical idea of a self as an entity in a person has been a nonsense for several centuries, since Descartes.

What is a self supposed to be? Is it a person in a person -- a homunculus? Is it purely neural, or of some different stuff of all consciousness, or is it some third kind of thing, as with Popper and Eccles? For Searle, for several reasons, it does seem to need to be a third kind of thing. What kind? To revert to its activity, if an act of a self is not the result of a last condition completing a causal circumstance, why, as you might ask, does it go off? Also why does it go off then? No doubt Searle's book to come will make a self a real biological phenomenon, an ordinary part of the physical world, and will save any of its virtues and and escape its vices. Hold your breath if you can.

He responds to an old claim by adversaries of such ideas as his, by the way, that they reduce our deciding and acting to randomness. He writes of his explanations involving reasons, a self, and no causal circumstances: 'Is there an element of chance or randomness in any such explanation? Not at all. It is a traditional mistake to suppose that where you do not have causal sufficiency, you have randomness' (p. 9).

This confuses matters. If 'random' means 'standardly uncaused', the ideas evidently do make decisions and actions random. Their proponents do of course allow this if they are clear-headed. 'Random' can also mean 'without explanation'. It is arguable that the ideas do make decisions and actions into things without real explanation. Someone who levels a charge of randomness against Searle is best understood as challenging him to explain how on his theory decisions and actions do really have an explanation. In that challenge there is no mistake, traditional or otherwise.

One more word here on deciding and acting, Searle declares that consciousness is essential to it. (p.9) Presumably he knows this is agreed by all sane parties. How could it be otherwise? So he is really declaring that consciousness somehow conceived is essential to deciding and acting. The question, unanswered by him, is what consciousness is. That real gap in what he says deprives him of an actual proposition here, as it does elsewhere.

III: Free Will

Consider now what has been off-stage so far, the problem of determinism and freedom. As reported earlier, we are informed that the remarkable fact we can discover about our experience of deciding and acting is that there are 'gaps' between its stages. We are said not to experience a prior stage as being a causal circumstance for a later one. Indeed we are said to experience a prior stage as not being a causal circumstance for a later one. It is these supposed gaps that we have to mind if we are to make progress towards solving the problem of determinism and freedom.

It is hard to resist saying a word for the whole regiments of philosophers who have minded the gaps. That in considering the problem of determinism and freedom we need to mind the gaps, reflect on our own experience of deciding and acting, will come as no surprise to an undergraduate with a reading list for an essay to write. 

Bishop Bramhall (1676) said so in the 17th Century to Thomas Hobbes (c. 1650), who agreed about the need for the reflection but not about the bishop's conclusion. Since then there has certainly been no gap in the line of agreement by the philosophers of determinism and freedom on the need for the reflection. Hume (1748) joined theologians in thinking about it, and concluded differently than they. Scores of pairs of philosophers said after them that when you move your finger to the right, you feel you could have moved it to the left -- feel that you could have done otherwise. It is very educative to spend some time with Vol. 2, pp. 316-327, pp. 446-453, and Vol. 1, Chs. 20-25, of Mortimer's Adler's The Idea of Freedom. The latest introductions to the problem of determinism and free will could not possibly leave out our supposed consciousness of freedom (McFee, 2000, Ekstrom, 2000). It is fully and admirably discussed by Magill (1997). 

In short, what is recommended by our new guide to the problem of determinism and free will is a terminological advance. My fellow-workers over a couple of centuries have not missed something out, as indeed Searle quickly indicates in passing (pp. 7-8) when he is not seeming to unveil reality to us.

As for the question of whether our alleged experience tells us the truth, which Searle slides towards assuming, that is of course another matter. It has been denied by every determinist since Hobbes, to go no further back. It is possible that our alleged experience is an illusion of human nature, like many well-known ones. It is possible that it is an illusion contributed to by the dominant culture in which philosophers of our tradition have grown up. It is almost certain that the question will not be settled by the experience itself but by the wider considerations of philosophy and science. But more on the experience and its content later.

Searle says our supposed experience of deciding and acting, whether it it is true to the facts, and what is then to be said of the brain, is the problem of determinism and freedom. It is the interesting problem (pp. 11-12). The subject of moral responsibility, taken in its relation to determinism, is not of interest. It is in fact a red herring, as he let us know in his Royal Institute of Philosophy lecture. He would say the same of the connected subject of the justification of punishment taken in relation to determinism. 

So too of the subject of our personal standing in a more general sense (Kane, 1996). In his R.I.P. lecture he also said the same of 'predictability' -- by which he may have meant the subject of our feelings about the future, and in particular our personal hopes. He would of course add other red herrings: the relation of of determinism to gratitude and resentment and other personal feelings, finely delineated by Peter Strawson (1968), and the relation of determinism to our confidence that our beliefs constitute knowledge, and its relation to moral principles and indeed political commitments.

This is remarkable. Our interest in our experience of deciding and acting, the supposed gaps, would be of a different order, indeed by comparison minimal, if it did not connect with the red herrings. That the problem of determinism and freedom has persisted for at least three centuries has mostly certainly not been owed to a little conflict between our alleged impression about our deciding and acting and our large conviction of naturalism.

Given the fact that determinism seems in ways to threaten moral responsibility and so on, and Libertarianism seems in ways to preserve moral responsibility and so on, there has been every reason for philosophers of determinism and of Libertarianism to concern themselves with these matters. In the first place, these matters give some content to the Libertarian's very thin idea of our origination of our deciding and acting -- some content, as might be added in the present context, to talk of funny causing. Against the objection that no adequate content has been given to talk of a decision's coming about non-causally but non-randomly, it can be said that it comes about in such a way as to preserve a certain moral responsibility of the agent.

With respect to these matters -- the red herrings -- philosophers have of course divided into Compatibilists and Incompatibilists. The first are said to maintain that determinism is logically consistent with freedom, as indeed they may as a result of defining freedom as voluntariness, this being acting in accordance with an agent's own desires, personality, character etc. Incompatibilists are said to maintain that determinism is logically inconsistent with freedom, as indeed they may by defining freedom as including origination. In fact the two regiments are greatly better characterized as restricting or committing themselves to different families of attitudes -- holding people resonsible and crediting them with responsibility in one way rather than another and so on.

Searle says that Compatibilism simply misses the point about the problem of free will (p. 11). If he quickly qualifies this into the humdrum proposition that it misses the point about the problem of free will as he has defined it, the impression remains that Compatibilism is yet another red herring, as indeed it was announced to be in the Royal Institute of Philosophy lecture.

Searle is of course a simple Incompatibilist, and inclined to be a simple Libertarian. He has fallen into the line of philosophers descended from Bramhall and no doubt his predecessors. What is the content of Searle's dismissal of Compatibilism as missing the point? What point? It has to be the proposition that determinism is inconsistent with origination, inconsistent with our existence if our gap-feelings are true. The impression is given that here is a fact, and to consider something else is irrelevant.

This is bluff. It is quite as open to any of his adversaries to declare that Incompatibilism misses the point. What point? The proposition that determinism is compatible with voluntariness, compatible with our existence as it seems to be in the light of general neuroscience and despite any gap-feelings. The proposition of consistency is precisely as much a fact as the previous proposition of inconsistency. Will someone suppose that the point the Compatibilist is missing is that there is this particular argument for indeterminism and libertarianism -- our supposed gap-impressions in our deciding and acting? No doubt, but absolutely mistakenly. Moreover, there are two sides in this introspection-contest. 

Many Compatibilists have of course said, with John Stuart Mill (1878, pp. 578-585), that in fact we have no experience of 'gaps' in our deciding and acting -- supposed absences of causal circumstances. That is not the content of our experience at all. What we have is experience that our deciding and acting is up to us, in accordance with our own desires, personality, character, etc. There is an absence of compulsion, constraint, or any such 'external' causation.

They might have added something else in analysis of our experience, of course. It is that we take our deciding or whatever, say (c) in the example, to have the distinction of being the cause in the set of conditions that makes up a causal circumstance for something. This is what is special about deciding and acting as against perceiving. This, together with the absence of compulsion, is what lies behind our confidently giving our reason (p. 9, p. 10) in explanation of why something happened. We do not need Searle's alternative. Nor do we have to think, of course, that in thinking about whether determinism is true, we should pay a lot of attention to our 'consciousness of freedom' as against, say, neuroscience.

He must try harder. Better must be done about determinism and freedom. To my mind, and now to some other minds, both Compatibilists and Incompatibilists are demonstrably mistaken (Honderich, 1988, Chs. 7-8). It is clear enough that each of us has or can have attitudes of the kinds mentioned above that are consistent with determinism, and also attitudes of these kinds that are inconsistent with determinism. There is what amounts to a behavioural proof of this. It may be that the attitudes inconsistent with determinism are in some way deeper, indeed that they have a unique truth in or about them (Honderich, 2001, pp. 397-9). It cannot be that what needs to be said about this is the very little indeed that has been said in the line from Bramhall to Searle. It may be that what needs to be said will be said in connection with some conception of consciousness as different from materialism, dualism and indeterminateness as the one mentioned earlier in connection with a kind of existence.

IV: Libertarianism with Neural Determinism

Section IV of Searle's paper is more about the previous matter of explaining decisions and actions than it is about free will. We get additional thoughts on the explanation of what happens when we consider reasons, decide, get an action going, and carry it through (pp. 13-21). In effect this Hypothesis 1, as it is called, is elaboration of the doctrine of Biological Subjectivity on Two Levels.

Return to (b) in the example, the bit of higher-level activity of the brain also known as consciousness, specifically your considering reasons for and against making a speech. Also (Lb) the bit of simultaneous lower-level activity, also known as neuron firings and the like. Lb causes the consciousness. Later on there is (c) your consciously deciding against a speech, along with the simultaneous lower-level activity Lc. Lc causes this consciousness.

Searle, having now considered the consciousness of deciding and acting in particular, as distinct from perceptual consciousness and presumably any other part of consciousness, has been belatedly struck by the epiphenomenalism objection. Why it is not taken to apply in connection with perceptual experience, remembering, and of course all of consciousness, is entirely unclear. At any rate he first contemplates that there is his reasons-and-self explanation of (c) the deciding, but rightly sees that this cannot save the picture from epiphenomenalism. There remains a full explanation of the deciding that leaves the reasons out. This is the explanation that Lb without the aid of (b) standardly caused Lc, and Lc standardly caused the deciding.

V: Libertarianism with Some Neural Indeterminism

You will remember the contradictoriness of Searle's two-level identity theory. Is it saved from contradiction by something now added to it? This is that the talk of an upper and a lower level is of course not to be taken in a paralysingly literal way -- the upper level being above the lower in the sense that the paint is on the surface of the table. 'Consciousness is no more on the surface of the brain than liquidity is on the surface of the water. Rather the idea we are trying to express is that consciousness is a system feature. It is a feature of the whole system and is -- literally -- at all the relevant places of the system...' (p. 16). 

Is the intended proposition that consciousness is pervasive or ubiquitous in the brain the same proposition that it is a system-feature? Consider first just the proposition that it is pervasive or ubiquitous. Does that save us from the contradiction? No. If there is a causal connection, there are two of something, which remains true if they occupy the same space, as do certain common properties of a multitude of things. 

Take now the utterance that consciousness is a system-feature of the brain, where that expresses something other than the pervasiveness proposition. In this connection we get the example of a rolling wheel, spoken of as a system, and the molecules in it -- the molecules in which the system entirely consists. There is 'system causation', about which you will be hearing more later, which is the system's effect on a molecule of it. Whatever else is to be said of this, it will be clear that it is no help with the contradiction. It is explicitly asserted here that there is causal connection between a higher level or system-feature, also known as consciousness, and a lower level or element-feature -- in which case there are two things where another part of the same theory says there is only one.

In passing, and without reference to the contradiction, notice that the addition of the description of consciousness as a system does not help us at all with the question of whether we are being told that conscious events are wholly neural events. Putting aside the vagueness of 'system', it is evident, partly from the wheel analogy, that this talk does not distinguish between a wholly neural system and something else.

There is something larger for us to end with. Since Hypothesis 1 is epiphenomenalistic, our guide now reluctantly prefers and recommends something else, Hypothesis 2. He regards it as at least an empirical possibility. It is the most salient item in his paper. It has prompted the selfless labour of this response of mine more than anything else.

It subtracts the standard causation of the neural Lc by the neural Lb. It keeps the standard causation by the neural Lb of the consciousness that goes with it, (b), and the standard causation by the neural Lc of its conscious partner (c). This is how we escape epiphenomenalism. To repeat, there is a causal circumstance or sufficient causal condition in the neurons for the simultaneous considering of the reasons, and a causal circumstance in other later neurons for the later deciding, but there is no causal circumstance in the earlier neurons for the later neurons. There is down-up causation but there isn't left-right causation.

To my mind this proposition is not merely intellectual unattractive but fantastic. Whatever lesser reasons may lead you in the direction of the thing -- certainly a conviction about the evolutionary advantage of consciousness doesn't commit you to this -- it seems you should not lose sight of its being as good as a factual and empirical absurdity. Searle mentions, furthermore, the customary indeterminism-interpretation of Quantum Mechanics as entering into the story of the absence of the left-right causation. Whatever may be added to it, this is the idea of true randomness. So what we have got in the brain is Quantum Mechanics working from left to right but not from down to up. Quantum Mechanics doesn't work upwards, just sideways.

So, the neurons and consciousness constitute a kind of machine at any time or over any period of time, but over no period of time are the earlier and later neurons such a machine. That seems to be not theoretical courage, but theoretical suicide.

There is another reason for saying so. Return again to the example at the beginning of an ordinary sequence of experiences. We are to distinguish the experience or consciousness of deciding and acting from other experience or consciousness. Say remembering your seeing something. So while there is no left-right causation but just Quantum Mechanics between various related neural events -- Lb and Lc , and Lc and Le, and Le and Lf -- there is left-right causation and no Quantum Mechanics between the other related neural events La and Ld. The neural story of remembering is dramatically different from the neural story of deciding and acting. This is dreamland, maybe even Graceland.

Will someone who read the degree in Philosophy, Politics and Economics in Oxford in about 1960 now ask what canons or principles of inductive or scientific proceeding justify a conclusion about theoretical suicide? It seems to me there is a good answer. We get those principles from clear cases of good and bad inductive or scientific practice. It is bad practice to suppose that with a kind of experience, deciding and acting, the brain is causal in one direction (upwards) and not causal in another (rightwards). It is bad practice to suppose that in a sequence of experiences the brain jumps back and forth between being left-right causal and not being left-right causal.

In the elaboration of Hypothesis 2 a striking addition is made to the down-up causation of which we know. It is up-down causation. Maybe this gives us another direction in which Quantum Mechanics doesn't work. The elaboration of the hypothesis is associated with some mysterious neuroscientific research which purports to prove, against neuroscience generally,  that the mind gets ahead of the brain (Libet, 1975, 1978; Honderich 1984, 1986). I leave the reader to reflect on the possibility of avoiding contradiction in the first item, and to inform himself of the very surprising research.5


Adler, M. J. (1958), The Idea of Freedom (Garden City: Doubleday)

Anscombe, G. E. M. (1971), 'Causality and Determination', Inaugural Lecture (Cambridge: Cambridge University Press). Reprinted in Sosa, E. (1975), Causation and Conditionals (Oxford: Oxford University Press).

Bramhall, J. (1676), Works (Dublin: publisher unspecified)

Buser, P., Rougeul-Buser, A. (1978), Cerebral Correlates of Conscious Experience (Amsterdam: Elsevier)
Dennett. D. (1991), Consciousness Explained (New York: Little, Brown)

Eccles, J. C., Popper, K. R. (1977), The Self and Its Brain (Berlin: Springer)

Ekstrom, L. W. (2000), Free Will (Boulder: Westview Press)

Hobbes, T. (c. 1650, 1839-45), The English Works of Thomas Hobbes, Vol. 5, ed. Thomas Molesworth (London: Scientia Aalen)

Honderich, T. (1984), 'The Time of a Conscious Sensory Experience and Mind-Brain Theories', Journal of Theoretical Biology 110, 115-129.

Honderich, T. (1986), 'Mind, Brain and Time: Rejoinder to Libet', Journal of Theoretical Biology 118, 367-375

Honderich, T. (1988), A Theory of Determinism: The Mind, Neuroscience, and Life-Hopes (Oxford: Oxford University Press)

Honderich, T. (1995), `Consciousness, Neural Functionalism, Real Subjectivity', American Philosophical Quarterly, 32, 4

Honderich T. (2000) 'Consciousness as Existence Again', Philosophy of Mind: The Proceedings of the Twentieth World Congress of Philosophy, Vol. 9, ed. Bernard Elevitch (Bowling Green: Bowling Green State University, Philosophy Documentation Center). Also in Theoria, No. 95, June 2000.

Honderich T. (2001a), Philosopher: A Kind of Life (London, New York: Routledge)

Honderich T. (2001b forthcoming), 'Consciousness as Existence, and the End of Intentionality', Philosophy at the New Millenium: Royal Institute of Philosophy Lectures, Royal Institute of Philosophy Lectures, ed. Anthony O'Hear (Cambridge: Cambridge University Press) 

Hume, D. (1748, 1902), An Enquiry Concerning Human Understanding, ed. L. A. Selby-Bigge (Oxford: Clarendon)

Kane, Robert (1996) The Significance of Free Will (New York: Oxford University Press)

Libet, B. (1978) 'Neuronal vs. subjective Timing for a Conscious Sensory Experience', in Buser and Rougeul-Buser, 1978.

Libet, B. (1975) 'Subjective Antedating of a Sensory Experience and Mind-Brain Theories: Reply to Honderich', Journal of Theoretical Biology.

Libet, B., Freeman, A., Sutherland, K. (1999), The Volitional Brain: Towards a Neuroscience of Free Will (Thorverton: Imprint Academic)

McFee, G. (2000), Free Will (Teddington: Acumen)

Magill, K. (1997), Freedom and Experience (London: Macmillan; New York: St. Martin's)

Mill, J. S. (1878), Examination of the Philosophy of Sir William Hamilton (London: Longmans, Green)

Nagel, T. (1986), The View From Nowhere (New York: Oxford University Press)

Searle, J. R. (1984), Minds, Brains, and Science (Cambridge, MA: Harvard University Press.

Searle, J. R. (1992), Rediscovery of the Mind (Cambridge, MA and London: MIT Press)

Searle, J. R. (1997), The Mystery of Consciousness (New York: A New York Review Book)

Searle, J. R. (2000a), 'Consciousness', Annual Review of Neuroscience 2000, Vol. 23, pp. 557-78.

Searle, J. R. (2000b), 'Consciousness, Free Action and the Brain', Journal of Consciousness Studies, Vol. 7, No. 10, October 2000, 3-22. All page references in the text are to this article.

Strawson, P. (1968) 'Freedom and Resentment', Studies in the Philosophy of Thought and Action
(Oxford: Oxford University Press) 


[1] The distinction between (i) the two kinds of intentions, inactive and active, of which the second is what used to be called a volition, and between (ii) the active intention that begins a movement and the intentionality involved in carrying it forward, is laid out in Honderich 1988, Ch. 4. Except with respect to causation, there seems to be no disagreement between Searle and me on these matters of deciding and acting.

[2] Functionalism is the doctrine that conscious events are any events standing in certain relations. If (i) their consciousness is a matter of just the relations, as was commonly said, conscious events are made into yet less than the ghostly stuff of a kind of dualism. If (ii) conscious events are any physical things in certain relations, they can be silicon events in computers or events of other stuff in distant space. Neural Functionalism, as remarked below in the text, is the doctrine applying to us and other animals that our conscious events are neural events in certain relations. The same distinctions can be made with respect to what can be called Cognitive Science with Philosophical Ambition.

[3] That about a dozen related characterizations of consciousness by Searle are as indeterminate is argued in Honderich, 1995.

[4] To be published in the Institute's journal Philosophy.

[5] See also Libet, Freeman and Sutherland, 1999.


This material has been published in Journal of Consciousness Studies, Vol. 8, No. 4, April 2001, pp. 62-78, the only definitive repository of the content that has been certified and accepted after peer review. Copyright and all rights therein are retained by Imprint Academic. My thanks to Anthony Freeman for improving the text.

Review of Searle's book, Mind, Language and Society

HOME: Determinism & Freedom Philosophy

HOME: T.H. front page