by Ted Honderich

Professor Searle replies to this paper and others in 'Why I Am Not a Property Dualist' (
Journal of Consciousness Studies, 2002, which is also on his home page at the University of California, Berkeley. Ted Honderich's paper was originally titled 'Consciousness, Neural Functionalism, Real Subjectivity', and published in the American Philosophical Quarterly.

Abstract:  Neural Functionalism is the doctrine that our human conscious events are neural events standing in relations to input, other neural events, and output. It is unswallowable since we have the conviction that our conscious events come to more than that. However, our grip on this fact is insufficient for philosophical purposes. We need an articulated conception of our conscious events. John Searle, despite his signal contribution to recent Philosophy of Mind, does not provide a conception which is inconsistent with Neural Functionalism. 13 characterizations of consciousness offered by him fail to provide a satisfactory conception. Such a conception will depend on a satisfactory account of real subjectivity. Such an account was unsuccessfully attempted in Honderich, "Seeing Things," Synthese, 98:1, Jan. 1994. Better is needed for progress in the Philosophy of Mind.  


Functionalism and Cognitive Science are sometimes understood in philosophically unambitious ways. For example, they are taken as coming to no more than such ideas as that desires, whatever else needs to be said of them, are typically effects of things perceived and causes of behaviour, or such ideas as that trying to define desire entirely without reference to behavior would be futile. Here there is little more than truism. Functionalism and Cognitive Science are also understood differently -- as giving the true account of all of the nature of our own and other conscious events. So understood, Functionalism and Cognitive Science seem certain to follow Behaviourism into the honorable past of the Philosophy of Mind.1 

That is, there seems no hope for this ambitious proposition: a complete account of conscious events is given by saying they are no more than the terms of certain causal or logical relations, no more than relata of input, other "inner" events, and output, with nothing more said of the relata. This proposition surely commits us to the absurdity that anything whatever that stands in the right relations is a conscious event. As it seems to me, much cajolery and bluster but nothing like an adequate reply has been made to this reductio ad absurdum argument.2 

The ambitious proposition surely also commits us to a second absurdity, somewhat concealed by usages but no less troubling. Although we are to identify conscious events as the terms of relations, we are somehow to understand that these terms in themselves contribute nothing to the reality of consciousness -- it doesn't matter what they are. The reality is all a matter of the relations. But these are indeed mere relations, not events or anything else. So in fleeing consciousness as ghostly stuff, Functionalists and Cognitive Scientists seem to have reduced it to yet less than ghostly stuff. Its reality seems really to evaporate. 

The insistence that conscious events are not just relata in the above way, but have some other particular nature, is consistent with several things. It is consistent with the proposition that our human conscious events are nothing but neural events standing in the specified relations to input, other "inner" events, and output. This proposition, about the conscious events of which we know most, can be called Neural Functionalism. It is a second ambitious proposition, perhaps more popular than the official Functionalism and Cognitive Science already put aside. It is this proposition which I shall be concerned with in what follows. 

It has been supposed to be less unswallowable than eliminative materialism or "nothing-but" materialism, the absolute denial of consciousness avoided even by many contemporary brain-mind "Monists" or "Identity Theorists".3 In fact Neural Functionalism seems no more swallowable than the mentioned materialisms. This is because nothing is added to the account of a conscious event itself, taken as a wholly neural relatum, by the central insistence of Neural Functionalists on variable or multiple realization -- the insistence that a non-neural event, say a silicon event or whatever, would be the same conscious event in virtue of standing in the same relations. I do not assign an additional property to a screwdriver by noting that its role could also be performed by something else, say a strong finger and a tough fingernail.4 

What is it that makes the idea that conscious events are no more than neural relata seem unswallowable? (Here and hereafter conscious events are to be understood as human conscious events, and neural relata are to be understood as wholly neural relata.) Well, what makes the idea seem unswallowable is the conviction that conscious events are different from neural relata. They either have both neural-causal and also other properties, or they have only the other properties. Which of these we say is unimportant. It is not a matter of the facts, but of how we choose to distinguish events -- as in the case, say, of the burning of a match. There too we can speak of one event with several properties, including temperature-change and so on, or several simpler events, one of which is just temperature-change. 

Most of us have the conviction that conscious events are different in the given way from neural relata. We have it because each of us has a grip on the nature of his or her consciousness. That is, we have an insufficiently conceptualized recollection of events of consciousness. More particularly, if we cannot do anything like observe a thought or feeling in having it, but only its object, we can immediately afterwards remember the nature of the thought or feeling. This insufficiently conceptualized recollection is analagous to an ordinary awareness of shape or colour, which also falls short of being an articulated conception. 

But it is philosophically unsatisfactory to do no more than insist that we have a grip on the nature of consciousness, which grip establishes a difference between conscious events and neural relata. We need an account of the difference between conscious events and wholly neural relata. Indeed, we need more than that. We need a conception of the nature of consciousness, say as good as our conceptions of the nature of events generally, the realm of the physical, sentences, truth, and the past. Not having such a conception is philosophically unsatisfactory, since inexplicitness always is philosophically unsatisfactory. The want of a conception leaves Neural Functionalists in a better position to doubt or deny that our grip on consciousness does establish a difference between conscious events and neural relata, and leaves us in a position that must be less than truly confident.5 

My first concern in this paper is to consider whether we get a conception of the nature of consciousness, larger than and thus inconsistent with that of Neural Functionalism, by a certain means. Can we get such a conception without getting into deep and murky philosophical water? The attraction of Neural Functionalism for philosophers of a certain tenor of mind, philosophers among whom I number myself, has been exactly that it stays out of such water, stays away from what can be called real or fundamental subjectivity. In the words of John Searle, can we get a larger conception of consciousness by attending only to "humble and obvious truths about the mind", and, as he should have added, certain unobscure philosophical and psychological conceptions and doctrines?6 In his The Rediscovery of the Mind, he gives an exemplary guide to 13 of these. We can hardly do better than consider what he offers. 


(1) The first of the 13 claims is that conscious events are indeed physical events.7 I take this, of course, to be distinct from the claim we are in effect examining, that they are the sub-class of physical events that are neural relata. Searle is right to take conscious events as physical, but what does this come to in his understanding? What are physical events? He says that the Cartesian conception of them, in terms of res extensa, is too narrow, because electrons are physical and they are only points of mass-energy.8 Perhaps he suggests, partly as a consequence of electrons so conceived, that physical events are those within the domain of current science.9 This philosophically popular but unreflective idea makes for trouble. At least it makes for an unnecessary dispute about a possible contradiction. This is so because current science and the contemporary scientific world-view are claimed by many not to have conscious events within their domains. 

The dispute is avoided by those of us who agree that conscious events are physical, but make use of the most ordinary conception of the physical, related to but superior to that of Descartes. Here, roughly, physical things are either space-occupants having secondary properties, or space occupants in causal or other nomic connection with space-occupants having secondary properties.10 

One of the virtues of the conception is its not sharing the indeterminacy of the one tied to current science -- as already implied, there is no fixed and agreed demarcation of current science, let alone of the contemporary scientific world-view. Another virtue of the ordinary conception is its not sharing the relativity and transience of the conception based on current science -- or, as might be added, the emptiness of a conception tied to completed or final science, things of which we are unaware. Searle, who rightly agrees that conscious events are in space, might better have adopted the ordinary conception of the physical. 

But the main point to be made about conscious events being physical is yet simpler. We are looking for a conception of conscious events which makes a difference between them and neural relata. Evidently we do not get even a part of it in the claim that conscious events share something, physicality, with neural relata. It does not matter how physicality is conceived. This first truth about consciousness is of no help to us. 

(2) Do we get the difference when Searle insists on something else, that conscious events are brain events caused by wholly neural events in the brain? When he insists, more fully, that conscious events are higher-level biological events of the brain caused by lower-level wholly neural events in the brain?11 

The answer is again clear, and can be given quickly. We evidently do not get a difference between conscious events and neural relata in the specific fact, first, that conscious events are brain events. So are neural relata. Nor do we get the difference in the second fact that conscious events are biological. So are neural relata. And we do not get a difference between conscious events and neural relata when conscious events are said to be higher-level effects of lower-level neural events, as solidity is an effect of H2O molecules in lattice structures. Just this, being higher-level effects, is true of neural relata and most of their constituents. It is true, for example, of the transmission of an action potential down the axon of a neuron. Any textbook of neuroscience illustrates the fact. 


(3a) We may seem to arrive at a difference from Neural Functionalism when it is said that conscious events are in a certain category of things, those which have a special mode of existence or ontology. What this comes to is that each conscious event is or only exists as somebody's conscious event -- for each conscious event there is a "first person", an "I", upon which it is dependent. I have this special relation to my conscious events. This personal dependency of conscious events gives us a first sense in which they are subjective rather than objective.12 

(3b) It is said to be a consequence of this fact of personal dependency that a conscious event is in a second sense subjective: it is not equally accessible to any observer. Someone else may not be able to tell if I am thinking of Baden.13 

(3c) The fact that conscious events, or rather intentional conscious events, are personally dependent or have the first-person mode of existence, is said to have another consequence, although one not spoken of in terms of subjectivity. It is that my information about the world is perspectival. It may be said to be from a point of view.14 

(3d) Finally the fact of the first-person mode of existence is said to involve or have as a consequence that a conscious event is subjective in a third sense. It has a what-it-is-like or what-it-feels-like aspect.15 

How are we to take the point (3a) that each conscious event is subjective in the sense of being dependent on a person? What is said is vague. If a person is understood in a standard way, say in terms of one or both of a persisting body and an internally-related sequence of conscious events, then the point will not give us the different account of conscious events for which we are looking. This is so since it is consistent with Neural Functionalism. Advocates of that doctrine can and will accommodate the point that a conscious event, as conceived by them, is dependent on a person. They will say, with reason, that a conscious event as conceived by them is dependent on a persisting body. They will take the mentioned relations within a sequence of mental events ("inner events" as distinct from input and output) to be nothing other than a part of their own doctrine. 

Might we then take the point as having to do not with a person but with a Cartesian self or ego? Clearly that "substance-dualism" would be anathema to Searle, who is already enraged by the idea that he is advocating "property-dualism". More generally, to turn to a Cartesian self or ego would take us well outside a ring of unobscure philosophical and psychological doctrines which we are contemplating and to which we are trying to confine ourselves. But if we do not go outside we seem to need to conclude that in this point about personal dependency we get no different conception of a conscious event. 

What of the claim (3b) that conscious events are subjective in the sense of not being open in principle to all observers? Well, our purpose now is to find an account of the nature of conscious events. Do we get it by way of the epistemological claim? It seems we do not.16 What we need is not an account of how conscious events are known but an account or part of an account of what is known. 

As for the claim (3c) that my information about the world is perspectival, what this comes to is that conscious events are related to objects by way of only some of the properties of those objects. (Hence the possible alteration of truth-value when co-referring terms are substituted in statements about consciousness.) That consciousness involves a point of view in this sense, but, so to speak, the world has no point of view, is true enough. But it remains clear, I think, that asserting the perspectival nature of conscious events does not provide and does not carry much promise of a distinct view of conscious events. It is surely no part of Neural Functionalism that in being conscious of something I am conscious of all its properties. 

It is the idea or ideas (3d), that a conscious event has a what-it-is-like or what-it-feels-like aspect, associated with the work of Thomas Nagel, that may seem most promising here.17 One feels that such an aspect is something of which Neural Functionalism cannot give an account. But there is a prior question. Does the idea of such an aspect make a significant contribution to any conception of consciousness? It may be part of what was earlier called our grip on the nature of consciousness, our insufficiently conceptualized recollection of events of it, and it does indeed seem to be no part of Neural Functionalism. But I doubt that talk of a what-it-is-like or a what-it-feels-like aspect contributes anything to any conception of consciousness, for the following reasons. 

Searle remarks that we can wonder what it feels like to be a dolphin, but not a shingle on a roof. (He would say the same, it seems, of what it would be like to be a dolphin as against a shingle.) The remark indicates what seems to be true: that to speak of what it is or feels like to be something -- that is, what it is or feels like to be anything -- is surely to speak elliptically of what it is or feels like to be anything that is conscious. It is difficult or impossible to attach any other sense to speaking of what it is or feels like to be anything. But then the words give no analytic advance. There is circularity. The definiendum appears essentially in the definiens. Speaking of what it is or feels like to be anything has been heuristically useful in persuading some philosophers in the tradition of Penelope's Wooers18 actually to return to a subject-matter -- consciousness. But it does not offer or contribute to a conception of that subject-matter. 

If, incidentally, we direct our attention to what it is like to be a particular thing, say a dolphin or a bat, as distinct from what it is like to be anything, there is the further disability that we are then inevitably thinking of distinctions between kinds of consciousness, and not about what we are after, which is the distinction between conscious events generally and other things. Finally, there is the familiar trouble, if we attach ourselves to the idea of a what-it-feels-like aspect as distinct from a what-it-is-like aspect, that many conscious events seem to lack such an aspect of feel or feeling. Pure thought seems to lack qualities of the suggested kind.19 

(4) Consciousness is said to involve two kinds of unity, the first existing at a time, the other existing over short stretches of time. My experiences at a time, say experiences of three objects I now perceive, are parts of just one conscious event. This is said to be the feature of consciousness which Kant spoke of in terms of the transcendental unity of apperception. Distinct from this "vertical" unity is "horizontal" unity. For example, it is said, my awareness of the beginning of my spoken or thought sentence continues after the beginning is past. This unity may involve not only the mentioned iconic memory but also short-term memory.20 

Certainly there are these two facts of unity about conscious events, and they are rightly included in what are called gross structural features of consciousness. It is as clear, I think, that these facts by themselves do not give us a distinct conception of the nature of consciousness. They are facts that can be claimed to be accommodated by any conception, and in particular Neural Functionalism. No doubt there have been philosophers who have moved from reflection on precisely the unities to a distinctive conception, involving a self or ego or the like, but Searle has no inclination to follow them. To follow them, it certainly seems, would take us into the deep and murky water which Neural Functionalism has the attraction of avoiding. 

So we do not get consciousness explained by way of just the unities. As it seems to me, something as disappointing is the case with each of the following mentioned features of consciousness. 

(5) Conscious events can be divided up in terms of a finite set of types or modalities, beginning with the five types of events related to the five senses. (6) Conscious events involve a figure-ground distinction, as shown by Gestalt psychology. They also involve (7) a pervasive familiarity, (8) overflow, which has to do with the way in which thoughts spill over or connect with other thoughts, (9) a center and a periphery, which has to do with selective attention and levels of attention, (10) boundary conditions, concerning the spatio-temporal-socio-biological location for me of my conscious events, (11) mood, and (12) a pleasure-unpleasure dimension.21 

In each of these cases, as it seems to me, it is left open to Neural Functionalists to fit the fact or facts into their account of the mind. For the most part these are familiar facts. In them, we do not get what is needed. Is there more hope in something else? 


(13) Searle distinguishes ersatz conceptions of intentionality from attempts to get hold of the real thing, and effectively despatches the ersatz conceptions. That is, he agrees that the intentionality (or representativeness, aboutness, meaningfulness, or directedness) of the contents of conscious events cannot be described just by asserting that there are causal relations between them and things in the world, including sorts of relations proposed in some doctrines of externalism or wide content. The enterprise of trying to naturalize intentionality -- trying to reduce it to no more than neural facts, bare causal relations having these facts as effects, and perhaps external or environmental facts working in some other way -- is futile.22 But what is it, then, for an event to be intentional? 

Certainly many philosophers seem to have supposed that a good account of intentional content will help, maybe more than anything else, in characterizing conscious events. Searle is surely among these philosophers, despite a discordant thought of his to which we shall come. To his credit, he is not among those other philosophers who appear almost to suppose that there is nothing more to the problem of consciousness than the problem of intentionality. 

He says at one point that the attempt to naturalize intentional content fails on account of leaving out its subjectivity.23 It is not clear which particular fact of subjectivity is in question. 

Do we get an account of the intentionality of content by way of seeing (3a) that each intentional conscious event is subjective in the sense of being dependent on a person? This cannot give or promise an account of intentionality that is distinct from the ersatz conceptions. The situation is the same as the one noticed earlier in connection with personal dependency used directly (as distinct from via intentionality) in the hope of finding a conception of conscious events.24 

Do we get an account of the intentionality of content by way of seeing (3b) that each conscious event is subjective in the second sense of not being equally accessible to any observer? This epistemological point seems to go no way towards characterizing the nature of intentionality. 

Do we get intentionality characterized in a fundamental way via the fact (3c) that intentional conscious events are perspectival or have a perspectival shape? I cannot see this. The familiar fact in question is agreed on all hands. From the point of view of a theory of intentionality it is pre-theoretical. All accounts will take themselves to accommodate it. 

Do we then get somewhere with respect to intentionality by way of the proposition (3d) that intentional conscious events have a what-it-is-or-feels-like aspect? The answer again seems to be no, for various reasons, several of which were in view earlier, when this third subjectivity was looked at in the hope of coming directly to a distinctive conception of conscious events. Another thought here might also have been mentioned a moment ago in connection with (3a) and (3b). It is the fact that all conscious events are taken (dubiously) to have the subjectivity in question, but not all conscious events are intentional. Hence there can be no hope of explaining intentionality by reference just to this subjectivity, if that is the idea. 

The project of characterizing intentionality in other than the ersatz ways, as a means to characterizing conscious events, does not seem much advanced either by the doctrine that intentionality is normative: that intentional notions set standards of truth, rationality, consistency, etc.25 Supposing that the doctrine of normativity is acceptable, it nevertheless seems not to help us. The doctrine offers further facts about a fundamental fact, that of intentionality, of which we still lack an explanation. The same sort of disappointment seems inevitable with respect to a second and more developed doctrine, that intentional content depends on what is called Background, initially identified as certain capacities abilities, and general know-how.26 

Let us at this point consider the discordant thought anticipated earlier. It is the insistence that we cannot study the intentionality of content except by studying consciousness, that intentionality can only be understood by way of its relation to consciousness, that intentionality is dependent on consciousness.27 As will be clear, this recommendation reverses the order of inquiry we have just been contemplating. That is, the discordant thought is that we come to a conception of intentional content by way of already having a conception of consciousness. So the recommendation goes against the strategy we have been contemplating. Explanans and explanandum are reversed. Do we have to choose between these strategies? Well, although this is puzzling, perhaps we need to become less traditional or one-directional, less worried about circularity,28 and allow the possibility of two-way traffic: (a) somehow or in some way the subject of intentionality is enlightened by the subject of consciousness and (b) somehow or in some way the subject of consciousness is enlightened by the subject of intentionality. 

But if we try to start out in the way lately recommended, by contemplating (a) that we can find out about intentionality by turning to consciousness, we face the annoying difficulty which has emerged in this discussion, that we seem to lack a conception of consciousness. Furthermore, if we make our way through the bundle of propositions (1, 2, 4-12) which we have about consciousness, and which we have not yet considered in our endeavour to clarify intentionality in the desired way, it is hard to be persuaded that any of them is of use. I leave the exercise to the reader. 

Let us return to (b), trying to clarify consciousness by giving attention to intentionality. We could not get going since we lacked a good account of intentionality, and one was not to be found in what was said of the connection between intentionality and subjectivity. What will now come to mind, no doubt, is recourse to a large account of the intentionality of mental content given by Searle in his book Intentionality, an account in ways dependent on a prior one of the derived intentionality of language. 

This account, to be brief indeed, takes an intentional conscious event to be a matter of (i) a content, which fixes (ii) a truth condition or other "condition of satisfaction", and of (iii) a pyschological mode, such that the event is a belief, a desire, or whatever, which mode determines (iv) direction of fit -- in the case of a belief, the belief must fit the world rather than the world be changed to fit the belief. 

This account is perhaps the best of its kind. But what is its kind? Well, it sets out what can indeed be called the logical properties of intentional conscious events. But it is hard to resist at least hesitation about it as a distinct account of the fundamental nature of intentionality. Perhaps this is best expressed by way of the thought that it is unclear what it is in the account that is inconsistent with a Neural Functionalist or naturalized reading of it. It seems Searle takes it that it does contain such a fact, but what is it? Could it possibly contain such a fact if it does not concern, as Searle explicitly allows it does not, the "ontological status" of the events in question?29 He says he gives that status elsewhere in saying, as we have heard with disappointment, that (2) conscious events are higher-level biological events of the brain caused by lower-level neural events. 

However, I shall not press the questions, since there is no need to do so. There is no need in our present endeavour -- finding a characterization of the nature of conscious events. This is so since, as already remarked, and as is very widely accepted, not all conscious events have the property of intentionality. It is a fact too often put aside in the contemporary philosophy of mind. Its importance for us now is simply that there was in fact little hope in contemplating that we could explain conscious events by way of the property of intentionality, and reflections on that property. There are paradigmatic conscious events which surely lack the property, notably sensations and moods.30 

It is true, I think, that the enumeration of the 13 features or groups of features of consciousness, and what is said of them, is likely to increase one's ordinary confidence about the nature of consciousness, to tighten one's grip on it. The procedure, partly because it includes good reminders, strengthens the conviction. This is no little achievement. 

But, to come to the first principal conclusion of this paper, reflection on ordinary truths about the mind and certain relatively unobscure philosophical and psychological conceptions and doctrines does not give us a distinct conception of the nature of consciousness. We do not get a conception which is distinct from Neural Functionalism.31 Rather, we get things consistent with it, or, in the case of (3d), something not useful. We have no sufficiently conceptualized ground or major premise, so to speak, for arguments against Neural Functionalism. None of the 13 features of consciousness gives us the needed conception. As we have seen, this is so in particular with what Searle most often relies on in summarizing his views: (2) the feature of consciousness that conscious events are higher-level biological events of the brain caused by lower-level wholly neural events.32 So too with what is said of subjectivity, intentionality, and so on. It is evidently as true that we do not and cannot get a distinct conception from the sum of the 13 features. 

Searle and like-minded philosophers are certainly not like the drunk he mentions, brother to the Functionalist and Cognitive Scientist, who loses his car keys in the dark bushes, but looks for them under the streetlight because the light is better there. These philosophers look in the right place, but, being worn out by trying to get the drunks into the bushes, and perhaps being a little attracted themselves to the streetlight and the activity under it, leave the keys in shadow. 


Searle, having earlier tried to approach consciousness by the indirect means of reflection on the intentionality of language33, does in The Rediscovery of the Mind follow the policy of what can be called Mental Realism. That is, he reflects directly on the reality of conscious events in themselves rather than being diverted by some more tractable fact pertaining to them. Functionalism and Cognitive Science have in fact been the prime examples of being diverted in this way. Another notable example has been the attempt to specify consciousness through no more than certain logico-linguistic properties of the part of language under which it falls.34 

But a distinct conception of consciousness will be informed by our experience, our common possession of consciousness, to a yet greater extent than Searle's reflections. It will also be owed to a determination not to succumb too soon to the proper demand for literalness, clarity and explicitness, thereby missing the reality on which we have a grip, and which needs to be respected in our reflections on the kinds of subjectivity mentioned above as well as unity, intentionality, and so on. Such a conception does not recoil from mystery, but attempts to dissipate it. It does not avoid deep and murky water but tries to recover something clear from it. Hence, I think, any such conception of consciousness is of a certain family of conceptions which includes Brentano's. He spoke of consciousness as activity which has reference to a content, or activity which is directed upon an object.35 

A distinct conception which is also adequate will have escaped the metaphor, obscurity, and inexplicitness of Brentano's view and others. It will also have avoided the disaster of discovering a true homunculus within a person, whatever tolerance of mystery entered into its history or construction. It will in the end have put not only a sufficiently conceptualized conception, but one that is literal, clear, explicit, and coherent, in place of our grip on consciousness. 

If we want to try to work our way towards a distinct and adequate conception of a conscious event, we can start with the idea that each conscious event has two properties, one of them its content. According to my lights, this has to be conceived in such a way as to avoid phenomenalism -- content has to be conceived in such a way as to preserve what seems to be the fact of the givenness of the world. It is not something subjective that I am ordinarily aware of. With respect to the other property or part of a conscious event, it is tempting to speak of the subject or the subject-part. This is evidently not a person in the sense of one or both of a persisting body and an internally-related sequence of conscious events, or indeed a person, self or ego in a Cartesian or any other sense. Still, there is the promise that this conception will be or will include an account of what seems to be the relation of real or fundamental subjectivity, as distinct from the subjectivities having to do with personal dependence, unequal access, and a what-it-is-like or what-it-feels-like aspect. 

The relation between the two properties or parts of a conscious event certainly seems to be one of interdependence. For several reasons, one of them having to do with avoiding phenomenalism -- the intruding of subjective objects of awareness into consciousness -- the relation is not one to be described as awareness of content. It is also a relation wholly distinct from another with which we have been concerned. That is the relation into which many but not all conscious events enter, the relation of intentionality between content and object.36 It is difficult to avoid the feeling that the second relation will really be enlightened in the end only by way of the first. 

I do not really have a distinct and adequate conception of a conscious event to offer, despite a considerable struggle to reach one.37 The lack of one does not make me doubt that it is what we need to pursue above all in the philosophy of mind. 

Enough has been said of a distinct and adequate conception of consciousness to make it possible to draw the second principal conclusion of this paper. It has to do with a question which may come to mind. Despite what has been said and implied of such a conception of conscious events, is it possible that it too can be seen or anticipated as something that can be or will be open to being Functionalized -- that is, shown to be consistent with Neural Functionalism? Will it fare no better than the 13 features of consciousness already considered? The answer seems to be no. 

Part of my reason for saying so is that Neural Functionalism, if we put aside its contained proposition that our conscious events are neural, is simply not a conception of the nature of a conscious event itself. Neural Functionalism takes an event as conscious, so to speak, as a black box. Unlike a distinct and adequate conception, it is an account of only the event's external relations. 

Will it be said that the proper conception will also be fundamentally a matter of a relation, even if a relation internal to an event? And hence that the proper conception will be part of a New Functionalism? One reply is to assert again that the relations with which Neural Functionalism concerns itself, and which define it, hold between only the content of a conscious event and these other things: input, the contents of other conscious events, and output. So with other Functionalisms, including the first ambitious conception of Functionalism noticed at the beginning of this paper. 

That is not all. Suppose, as presumably is true, that the relation between the content-property and the subject-property of a conscious event is either nomic or logical. That is no invitation to talk of any Functionalism. What will be essential to to any adequate idea of such an event will not be that bare fact but, so to speak, the uniqueness of the nomic or logical relation, its unique character.38 Inquiry into that character -- say inquiry prompted by the conclusion mentioned above, that the relation is not awareness -- will be alien to the sources of all Functionalisms in analogies to Turing Machines and their successors, and also the history and continuing impulses of all Functionalisms. It is not too much to say that the strength of these doctrines lies in precisely the avoidance of the mystery and deep water with which a disinct conception of consciousness tries to contend. 


Two matters remain, the first having to do with the question of the relation between consciousness and the brain, the mind-body question. This has not been our main concern so far -- we have been mainly concerned, rather, with the question of the nature of consciousness. But evidently accounts of the nature of consciousness may contain or entail answers to the mind-body question. It is common to judge accounts of consciousness by way of their upshots for the mind-body question. In particular it is taken to be bad news for an account of consciousness if it entails what is called dualism. 

Searle notices that the terms "dualism" and "monism" or "materialism" or "Identity Theory", and variants of them, are no longer of much use.39 One reason is that typical "monisms" are in ways property-dualisms. Another is that some "dualisms" are no more dualistic than typical "monisms". He also strenuously insists that his own view can only be mistakenly or misleadingly characterized as either materialist or property-dualist. What is to be said of this insistence? 

Certainly he is right to object to the names "property-dualism" and "materialism" for his view in so far as the names carry the implication of a possible division of what there is into two domains, the domain of consciousness and the domain of the physical, such that nothing is in both domains. Rather, to recall (1) above, conscious events are physical -- the domain of consciousness is included in the domain of the physical. He is also right to disavow the name "property-dualism" if it carries implications of support for particular doctrines of introspection, privileged access, or inward observation and so on which he rejects. He is as right to resist the name "property-dualism" if it carries the implication that the view so named nevertheless somehow requires or countenances more "substances" than particles in fields of force, or that consciousness is "stuff" in a particular sense which he disavows, or that the mind-body problem is insoluble, which of course he does not believe.40 Finally, he would be right to say that his proposition (2) about two levels of events does not make him into any kind of dualist. What matters is whether the characters or natures of events at the two levels are different. 

All of that is worth insisting on, given the deplorable tendency in the contemporary philosophy of mind to conduct argument by sticking on undefined labels. But none of what he insists on takes away from two other things that are true. 

He initially and thereafter declares himself to be against reductionism, and this is in part the insistence that our conscious events have properties other than those assigned to them by Neural Functionalism.41 Plainly, if this had turned out really to be the demonstrated burden of his view, we would have had a reason for speaking of the view as property-dualist. (This conclusion is unaffected, by the way, if we take up a way of speaking of these matters closer to Searle's usually preferred one, and thus get rid of events. Then the reason for speaking of a property-dualism would simply be that the view was that there are wholly neural-causal properties or features of the brain and there are properties or features that require another characterization.) But of course there is something else to be said. It is one thing to announce and insist that conscious events in themselves are not merely neural relata. It is another to give a conception of them which bears this out -- a distinct conception of them. Our inquiry issued in the conclusion that Searle fails to provide this. None of the features (1) to (13) succeeds in this, and neither does the sum of these features. Thus the account actually given is consistent with Neural Functionalism. 

My third principal conclusion is therefore be that the view as actually set out, or as it turns out under examination, is not property-dualist in just the sense in which it is announced to be. There is a difference between the headline and the story. This is perhaps the main explanation of the conflicting responses to the view which Searle reports. 

It seems clear -- a fourth conclusion -- that the distinct and adequate conception of consciousness of which I was speaking will in intent and in fact be correctly describable as entailing a property-dualism. That will not make it different from some contemporary "Identity Theories", including Davidson's Anomalous Monism. Like them, also, it will in certain senses be quite as correctly describable as "monist".42 What is most important is that the adequate conception will not be unswallowable. It will not be among the theories without a future. 

A final matter. Is the prospect of a distinct and adequate conception of consciousness, if not the prospect of something unswallowable, nevertheless the prospect of something outrageous? I suspect that it will seem so to many, and I admit to sharing something of their feeling. It is the prospect of a view to the effect that there really are two different kinds of events that occur in our heads: there are two different kinds of events which are physical in the ordinary sense and hence involve space-occupants. Or, as we can as well say, there are two different kinds of physical properties of our brains. One kind of events or properties involves only neurons and the causal relations fundamental to Neural Functionalism. The other kind of events, events of consciousness, are such that each involves what may initially be spoken of as a subject-part and a content-part. It needs to be admitted that if such a view43 is not clarified by describing it as having to do with conscious stuff, it is not travestied either. 

I shall not now try to do a lot to alleviate the outrageousness, but only make some remarks. Several of them echo things Searle has to say about the conception of consciousness he promises but does not produce. 

One is that it is only such a conception, as it seems to me, that is true to our experience, true to the grip we have on consciousness, true to what we know when we do not withdraw in apprehension from the subject. A second remark is that such a conception depends on no dubious doctrine of introspection, but only the indubitable fact of recollection. A third is that it really does need to be kept clear that the events or properties in question are conceived of as physical in a strong sense, the ordinary sense. They are not other-worldly. A fourth and related remark is that nothing stands in the way of their entering into the explanatory web of nomic or lawlike connections. More particularly, nothing stands in the way of their being nomic correlates of wholly neural events. and of being causes or anyway causal and also effects. Fifthly and finally, it seems to me remarkable, indeed silly, to say that conscious events so conceived are inconsistent with the scientific world-view. Neuroscience has long been engaged in precisely the study of nomic connections of a kind between neural events and conscious events -- between neural events and the events on which we have always had a grip and of which we can hope for a distinct and adequate conception. 

University College London 


1. My thanks for comments on earlier drafts of this paper are due to Finn Collin, Tim Crane, Owen Flanagan, John Heil, O. R. Jones, E. J. Lowe, Simon Matthew, Gregory McCulloch, Alfred Mele, Paul Noordhof, Jane O'Grady, Ingmar Persson, Gabriel Segal, Michael Targett, and anonymous APQ reviewers. We are not in perfect agreement. 

2. The best-known and the best form of this argument is John Searle's Chinese Room Argument ("Minds, Brains, and Programs," Behavioral and Brain Sciences 1980, 3, pp. 417-424). I salute him for a signal contribution. 

3. E.g. Donald Davidson, "Mental Events", Essays on Actions and Events (Oxford: Clarendon, 1980), p. 214. For reflection on the usefulness of the label "Identity Theory", or rather the want of usefulness, see below -- the last section of this paper. 

4. For arguments for the unswallowableness or rebarbativeness, and also the incoherence, of Neural Functionalism, see my "Functionalism, Identity Theories, the Union Theory", in T. Szubka and R. Warner, eds., The Mind-Body Problem: The Current State of the Debate: (Oxford: Blackwells, 1994). 

5. I allow that we would at least get well forward, be in a better philosophical position with respect to opposition to Neural Functionalism, if we could give an explicit account of just one feature of consciousness that cannot be accommodated by Neural Functionalism. As it turns out below, we do not get this by restricting ourselves to humble and obvious truths about the mind and plain philosophical and psychological ideas. One of them is of particular interest in this connection. See note 19. 

6. John Searle, The Rediscovery of the Mind (London & Cambridge, Mass.: MIT Press, 1992), p. 17. All page references below, unless otherwise indicated, are also to this book. 

7. pp. xii, 13, 14-15, 28, 54, 91, 100 

8. Pp. 25, 86 

9. pp. xii, 3-4, 9, 85-91, 118 

10. See my A Theory of Determinism: The Mind, Neuroscience, and Life-Hopes (Oxford: Clarendon, 1988) or Mind and Brain (Oxford: Clarendon, 1990), in each case pp. 87-89. 

11. pp. 1, 14, 90, 111-112 

12. pp. 16-17, 19, 20, 55, 70, 94 

13. pp. 16, 21, 94 

14. pp. 70, 95, 131 

15. pp. 117-118, 132 

16. I take it Searle might agree that we do not, since he is very firm about distinguishing ontology from epistemology. See p. 18. 

17. "What is it like to be a bat?", Philosophical Review, 1974, reprinted in Nagel's Mortal Questions (Cambridge: Cambridge University Press, 1979). See also Timothy Sprigge, "Final Causes", Supplementary Proceedings of the Aristotelian Society, 1971. 

18. "Aristippus said that those that studied particular sciences, and neglected philosophy, were like Penelope's wooers, that made love to the waiting women." Francis Bacon, ed., Apopthegmes Old and New (London: Hanna Barret, Richard Whittaker, 1625) No. 189. [title spelling OK] 

19. Do we get, in what is said of a feels-like or is-like aspect of consciousness, a feature of it made explicit that cannot be accommodated by Neural Functionalism? (See note 5.) It seems to me that we do not. Given the circularity, insistence that a feels-like or is-like aspect cannot be accommodated by Neural Functionalism reduces to what we began with and allowed to be philosophically unsatisfactory: that our grip of consciousness, our insufficiently conceptualized relation to it, persuades us of a difference between neural relata and conscious events. 

20. pp. 129-130 

21. pp. 128-129, 132-141 

22. My critical view of one doctrine of externalism is given in "The Union Theory and Anti-Individualism", in J. Heil and A. Mele, eds., Mental Causation (Oxford: Oxford University Press, 1993). 

23. p. 50 

24. See above, p. 000 [para starting "How are we to take the point..."] 

25. pp. 51, 238 

26. Ch. 8 

27. pp. 18, 84, 227 

28. p. 83 

29. Intentionality: An Essay in the Philosophy of Mind (Cambridge: Cambridge University Press, 1983), p. 14 

30. The few philosophers inclined to say otherwise, I suspect, either have an inadequate idea of intentionality or manage to confuse the relation with another one, internal to a conscious event, about which something will be said below. Incidentally, the point that consciousness cannot be elucidated by intentionality since not all consciousness is intentional is distinct from the earlier point that intentionality cannot be explained just by subjectivity since not all consciousness is intentional but all is subjective. 

31. As implied earlier (p. 000), I take it that the Chinese Room argument ("Minds, Brains, and Programs," op. cit.) and also a further good argument in The Rediscovery of the Mind (Ch. 9), despite my admiration for them, require reinforcing by the provision of a positive and full account of the nature of consciousness. 

32. This account of the mind-body relation seems to me open to several questions. One is its seeming, despite Searle's good intentions, to suffer from the curse of much of the contemporary philosophy of mind, which is epiphenomenalism. See A Theory of Determinism or Mind and Brain, in both cases pp. 99-102. 

33. Intentionality 

34. R. M. Chisholm, Perceiving: A Philosophical Study (Ithaca: Cornell University Press, 1957), Davidson, "Mental Events", op. cit. 

35. "The Distinction Between Mental and Physical Phenomena", in his Psychology from an Empirical Standpoint, ed. O. Kraus and L. McAlister (London: Routledge and Kegan Paul, 1973) 

36. In connection with the two relations, although he does not speak explicitly of the first one, Searle writes as follows. "One can never just be conscious, rather when one is conscious, there must be an answer to the question `What is one conscious of?' But the `of' of `conscious of' is not always the `of' of intentionality." (p. 84; cf. pp. 130-131) That brevity, if that is what is is, is misleading. The "of" which has to do with a relation between subject and content is never the "of" of intentionality, which has to do with the relation between content and object. Every conscious event involves something which prompts use of the first "of". Only many conscious events also or in addition involve the second "of". Perhaps it is in the quoted lines, in so far as they may be taken to touch on the first "of", that Searle comes closest in a general way to the essential matter of real subjectivity. 

37. "Seeing Things", Synthese, 1994 

38. A similar point is to be made about intentionality. What other relations could hold between content and object than nomic ones? But the problem is the character of these particular nomic relations. The assertion of bare relations is no help. 

39. pp. 2, 4, 14-15, 25-26, 54-55 Cf. my "The Union Theory and Anti-Individualism", op. cit. 

40. His objections (pp. 100-105) to arguments for insolubility do not seem to me as efficient as they might be. One argument, assigned to Thomas Nagel, is that if matter explains consciousness, there has to be a necessary connection between the two, but in fact there is no conceptual connection. The objection to that is simply that the explanation depends not on a conceptual but only a nomic connection. Causal connection isn't conceptual. A second argument for insolubility, derived from Nagel and assigned to Colin McGinn, is that consciousness is stuff of which we are aware through introspection, and the brain by contrast is something of which we are aware through perception; as for an explanatory link, we could have no way of being aware of it -- there is no third kind of awareness. The objection to that, although evidently there are others, is that there is no need at all, if I am to explain C by B, for me to have the same kind of awareness of both C and B, and no need for me to have an explanation of the explanation. With respect to the second point, I don't fail to explain C by B (having shown that B is a causal circumstance or nomic correlate for C) because I have not filled out the story. All I need to do is establish the truth of a certain conditional statement about B and C. We would have no explanations, anywhere, if every explanation had to be explained in the given sense. See A Theory of Determinism or Mind and Brain, Ch. 1. 

41. There can be no doubt that Searle is officially opposed to the reductionism exemplified by Neural Functionalism -- whatever actually happens in the course of his book. (See pp. xii, 28, 112, 116, 116-118, 169, 199.) Hence, incidentally, his saying that "the mental is neurophysiological at a higher level" (Note 3, p. 253; cf. p. 161) is at least misleading. It does not go well, to take one example, with the anti-reductionist line that "The neurophysiology does indeed admit of different levels of description, but none of these objective neurophysiological levels of a level of subjectivity." (p. 169) Subjectivity, as we have seen, is taken by him as fundamental to consciousness. 

42. One of these theories is Davidson's Anomalous Monism. See "Mental Events", op. cit. 

43. One such view is the Union Theory, involving nomic connection between conscious and neural events or properties, set out in my A Theory of Determinism and Mind and Brain.