SWIF Philosophy of Mind, 12 January 2001 http://www.swif.uniba.it/lei/mind/forums/carruthers.htm

Forums Forum 2

Précis of
Carruthers, Peter. Phenomenal Consciousness. Cambridge: Cambridge University Press, 2000.
ISBN 0521 78173 6

Peter Carruthers

University of Maryland, College Park


Most contemporary philosophers of mind think that mental states are physical states of the brain, characterised in terms of their causal roles; and many hope that our common-sense conception of the mind can be incorporated smoothly into science. These are beliefs and hopes which I share. But philosophers such as Nagel (1986) and McGinn (1991) have argued that consciousness – particularly phenomenal consciousness, or the sort of consciousness which is involved when one undergoes states with a distinctive subjective phenomenology, or ‘feel’ – is inherently, and perhaps irredeemably, mysterious. And many would at least agree with Penrose (1994) and Chalmers (1996) in characterising consciousness as the ‘hard problem’, which forms one of the few remaining ‘final frontiers’ for science to conquer. Yet there have also been a plethora of attempts by philosophers and psychologists at explaining consciousness in natural terms. These debates have attracted a great deal of interest, both throughout the academic community and amongst the wider public.

This book reviews and contributes to these debates, with the overall objective of defending a particular kind of naturalistic (scientifically acceptable) explanation of phenomenal consciousness – namely, dispositionalist higher-order thought theory. My view is that phenomenal consciousness consists in a certain sort of intentional content (‘analog’, or fine-grained), held in a special-purpose short-term memory store in such a way as to be available to higher-order thoughts about the occurrence and nature of those contents; and that in virtue of such availability (given the truth of some or other form of ‘consumer semantics’) all of those contents are at the same time higher-order ones, acquiring a dimension of seeming or subjectivity. While the problem of phenomenal consciousness may indeed be hard, it is by no means insuperable; indeed, I claim to have provided a solution to it within the pages of this book.

Chapter 1 Assumptions, distinctions, and a map

In this opening chapter I lay out some of my background assumptions, introduce a number of important distinctions, and outline the direction which the discussions of later chapters will follow.

  1. Physicalism and naturalism

  2. In this section I briefly explain and defend two default assumptions - physicalism and naturalism - which form the background to the problem of phenomenal consciousness. It is these assumptions which appear to be challenged by the very existence of phenomenal consciousness, as we shall see in chapters 2 and 3.

  3. Functionalism and theory-theory

  4. The assumptions in section 1 above relate to the metaphysics of the mind. In this section I say something about how I take the mind to be conceptualised, or conceived of. I assume that some kind of functionalism - viz. a form of theory-theory - provides the best account of the way in which we conceptualise mental states. Again the position is not entirely mandatory, and again some of the main challenges come from considerations having to do with phenomenal consciousness. But the advantages of functionalism as an account of the mind (viz. its metaphysical neutrality – hence allowing interactive dualism to be a conceptual possibility – and its solution to the problem of other minds) mean that it should not be given up lightly.

  5. Some distinctions: kinds of consciousness

  6. There are a number of different notions of consciousness and/or a number of different kinds of use of the term ‘conscious’ which need to be distinguished carefully from one another. Failure to draw the right distinctions, and/or failure to keep the different notions apart, has vitiated much work in the area.

    The most basic distinction is between creature-consciousness (or perceptual-consciousness) and state-consciousness. The latter then admits of a number of conceptually-different kinds, including phenomenal-consciousness and various forms of access-consciousness (first-order and higher-order). The project of the book is to reductively explain phenomenal consciousness in terms of higher-order access-consciousness.

  7. A route map: the tree of consciousness

This section lays out all the important theoretical positions in respect of the explanation of phenomenal consciousness, ranging from ‘no-explanation’ theories, through neurological explanations, first-order representationalist theories, and a variety of higher-order theories. These are related to the sequence of discussion in the book.


Chapter 2 Perspectival, subjective, and worldly facts

Many have alleged that phenomenal consciousness can neither be accommodated within a physicalist world-view, nor reductively explained in physical terms. In this chapter I confront some of these ‘mysterian’ arguments, concentrating on those which are more metaphysical in nature. If it is to be possible to provide a naturalistic explanation of phenomenal consciousness, as I intend, then all of these arguments must be flawed.

  1. Perspectival and ‘myness’ facts

  2. Nagel (1974) is often credited with putting the problem of phenomenal consciousness on the map. In this section I consider his main arguments. I show that they fail, and that their failure results from conflating together notions which should be kept distinct - in particular, many of the arguments involve a conflation of sense and reference.

  3. On facts and properties

  4. This is a vital section of the book. In it I argue that we can and should distinguish between two different notions of ‘fact’ and of ‘property’. One conception is thin, and is arguably required for semantic theory. On this conception, facts are just true thoughts, and properties are functions from possible worlds to extensions. And on this conception, the identity and distinctness of properties can be discerned a priori by means of thought-experiments. But the other conception is thick, and is arguably required for accounts of change, and to serve as the relata in scientific laws. On this account, what properties there are in the world is ultimately a matter for science to answer, and questions of identity and distinctness of properties cannot be resolved a priori. Moreover, it is this ‘thick’ notion of property which is relevant to questions of reductive natural explanation.

  5. Necessary identities

  6. This section discusses Kripke’s (1972) famous modal arguments against physicalism. It argues that they do not succeed. Token-identities are not ruled out, since all the arguments for their contingency rely on conceivability-experiments. Type-identities are not ruled out, since there may be no unitary ‘thick’ mental properties common across the imagined worlds. And claims that mental properties may be physically constituted are left untouched, which is all that is strictly required for reductive explanation.

  7. Logical supervenience

This section confronts one of the main arguments in Chalmers (1996), that phenomenally conscious states do not supervene on the physical facts in the way required for them to be physically constituted. It shows that the argument turns crucially on the ‘thin’ conception of properties as functions from worlds to extensions, and should thus be rejected by any would-be reductive naturalist.


Chapter 3 Explanatory gaps and qualia

In this chapter I continue my review and rebuttal of ‘mysterian’ arguments concerning phenomenal consciousness, focusing particularly on those which are epistemic in nature, having to do with possibilities of explanation, knowledge, or understanding.

  1. Cognitive closure

  2. This section discusses and criticises McGinn’s (1991) view that consciousness, while not intrinsically (metaphysically) mysterious, must always remain mysterious to us, because the answers to the questions we can frame here are cognitively closed to us.

    One general moral to emerge is that any explanation of phenomenal consciousness should be top-down and incremental in nature. (Somehow the rumour has got around, and become entrenched, that the problem of phenomenal consciousness is the problem of explaining how subjective properties can be constituted by processes in the brain; and most proposals on the market attempt to relate brain processes directly with properties of phenomenal consciousness.) In contrast, reflection on general scientific methodology suggests that we should initially seek our explanations in terms of the level immediately below our target – in this case intentional or computational psychology – which we will then, in turn, attempt to relate to the level below that, and so on until we ultimately reach processes which can be described in terms of the operations of neurons in the brain.

  3. Explanatory gaps

  4. This section discusses and criticises Chalmers’ (1996) argument that there is an unbridgeable ‘explanatory gap’ between phenomenal consciousness and all other worldly phenomena. I agree with him that reductive explanations normally work by specifying lower-level mechanisms for fulfilling some higher-level function. And I agree that we at least have available to us purely-recognitional concepts of phenomenally conscious states. But I disagree with the conclusions which Chalmers draws from these facts. His mistake is to assume that a given property or state can only be successfully reductively explained if the proposed mechanisms are what we might call immediately cognitively satisfying, in the sense that they mesh with the manner in which those states are conceptualised. While the ‘explanatory gap’ is of some cognitive significance, revealing something about the manner in which we conceptualise our experiences, it shows nothing about the nature of those experiences themselves.

  5. The knowledge argument

  6. This section outlines and criticises Jackson’s (1982, 1986) famous ‘knowledge argument’. My diagnosis is this: the knowledge-argument only seems compelling because we covertly read the ‘complete knowledge’ component of the argument in the thick sense (that is: ‘Concerning every worldly – thickly individuated – fact about colour vision, Mary knows the truth of a thought representing it’), but then we take the claim about Mary’s incomplete knowledge of colour-experience in the thin sense (that is: ‘Concerning some conceptual representation of colour-experience, Mary does not know its truth’). The argument commits a fallacy of equivocation.

  7. Inverted and absent qualia arguments

This section evaluates a variety of inverted-spectrum arguments for the conclusion that our experiences possess intrinsic qualia (non-representational, non-relational, properties). Mere conceivability arguments are easily dismissed, but natural-possibility arguments due to Shoemaker (1981) and Block (1990) are more challenging. They can, however, be decisively answered if intentional content can be narrow as well as wide. This is left over for discussion in the next chapter.


Chapter 4 Naturalisation and narrow content

In this chapter I begin to survey the prospects for a naturalistic account of phenomenal consciousness, taking us through some of the initial options. Attention quite soon comes to focus on theories which employ some combination of causal role and intentional content – that is, theories which are both functional–boxological and representational in character. I then suggest that intentional content should be characterised narrowly for purposes of psychological explanation in general, and for deployment in proposed reductive explanations of phenomenal consciousness in particular.

  1. Neural identities and consciousness boxes

  2. This section argues that neural identities, even if true, cannot provide reductive explanations of phenomenal consciousness. It also argues that the postulation of a consciousness box - again, even if correct - cannot provide a sufficient explanation. Rather, the account will need additionally to advert to the contents of the box.

    At the end of this section the main desiderata for a successful explanation of phenomenal consciousness are set out. Such an account should (1) explain how phenomenally conscious states have a subjective dimension; how they have feel; why there is something which it is like to undergo them; (2) why the properties involved in phenomenal consciousness should seem to their subjects to be intrinsic and non-relationally individuated; (3) why the properties distinctive of phenomenal consciousness can seem to their subjects to be ineffable or indescribable; (4) why those properties can seem in some way private to their possessors; and (5) how it can seem to subjects that we have infallible (as opposed to merely privileged) knowledge of phenomenally conscious properties.

  3. Naturalisation by content

  4. This section briefly reviews various kinds of attempt to naturalise intentional content. It argues that intentional properties are already in good natural standing by virtue of figuring in scientific- intentional psychology. While there are issues concerning the integration of such psychology with the rest of science, we have good reason to believe that this can be done, and so good reason to believe that the property of intentionality is both natural and real. It is therefore fit to serve in a naturalistic explanation of phenomenal consciousness.

  5. Wide versus narrow content

  6. In this section the distinction between wide and narrow content is explained, and the coherence of a notion of narrow content is defended. Then it is briefly argued that the notion of content employed for purposes of psychological explanation should be (and is) narrow (these arguments are pursued more fully in other publications - see especially Botterill and Carruthers, 1999).

  7. Phenomenal consciousness and narrow content

This section rounds off the chapter by discussing ways in which narrow content can be deployed in reductive explanations of phenomenal consciousness, and by re-visiting the inverted spectra examples from chapter 3.


Chapter 5 First-order representationalism

This is the first of two chapters which assess the prospects for a naturalistic explanation of phenomenal consciousness in first-order representational (FOR) terms, focusing particularly upon the accounts presented by Dretske (1995) and Tye (1995). Part of the point of these discussions is to develop an account of first-order perceptual contents which can then be fed, as a component, into the higher-order theories to be discussed in chapters 7, 8 and 9.

  1. FOR theory: elucidation

  2. This section outlines Tye’s PANIC theory (PANIC is for Poised Abstract Non-conceptual Intentional Content). It focuses especially on showing how the theory can handle the case of bodily sensations like pain, and emotions like fear. Tye’s view that pains are best understood on the model of secondary qualities like redness is set out and defended - so the pain itself is what is represented in a state of feeling pain, just as redness is what is represented in a state of seeing red.

  3. FOR theory: defence

  4. In this section the main arguments supporting FOR theory are set out. Such theories explain the so-called ‘transparency’ of experience, they mesh with our intuitions regarding the consciousness of animals, they admit of plausible evolutionary explanation, and they can meet at least some of the desiderata for a successful theory, set out in 4:1.

  5. Non-conceptual content versus analog content

  6. This section tackles the proper characterisation of the contents of perception. Are they non-conceptual, as Tye would have it? or merely analog, but imbued with concepts? The section argues that the latter option is preferable. It also goes on to discuss the sense in which perceptual contents can be concept-involving, given that they can be non-judgemental.

  7. More varieties of FOR theory

It is manifest that FOR theories admit of more varieties than are actually represented in the published literature. One set of options comes from the different ways of drawing the contrast between belief and perception, reviewed in section 3 above. Another choice concerns whether the intentional contents appealed to by FOR theory should be individuated widely (externally) or narrowly (internally). Both Tye (1995) and Dretske (1995) endorse forms of externalism about content. But the considerations adduced in chapters 3 and 4 above make it seem likely that a FOR theorist both can and should appeal, rather, to narrowly individuated contents.

In this section two further sets of options are considered. One is whether a FOR theory of phenomenal consciousness should adopt a reductive, or rather a non-reductive, account of intentional content. The other is whether the intentional content of perception is best explicated in terms of informational (that is, causal co-variance) relations to the environment, or rather in terms of some or other form of what Millikan (1984) calls ‘consumer semantics’ – either teleosemantics, or some sort of functional or inferential role semantics. The latter alternative is defended in each case.


Chapter 6 Against first-order representationalism

This chapter sets out the case against all FOR accounts, of whatever variety. It focuses merely on the first-orderness of such theories, and the argument turns crucially on the real existence of non-conscious experience. The first two sections of the chapter are concerned to argue for the reality of non-conscious experience, from both common-sense and scientific perspectives. The final two sections then develop the argument against FOR theory, in the form of a trilemma. The upshot is that FOR theory fails because it cannot really explain the feel, or ‘what-it-is-likeness’, of phenomenally conscious experience.

  1. Non-conscious experience: the case from common-sense

  2. This section presents a common-sense argument for the reality of non-conscious experience, using examples of absent-minded activity and experiences undergone while sleeping. It also sketches an outline of the two-layered mind which seems supported by such arguments.

  3. Non-conscious experience: the scientific case

  4. This lengthy section contains some discussion of the now-familiar phenomenon of blindsight. But most of it is devoted to exposition of Milner and Goodale’s (1995) dual-function theory of vision, and some of the evidence in its support. On this view the parietal-lobe stream of visual analysis is to produce ego-centric representations for use in the on-line guidance of movement, whereas the temporal-lobe stream is to produce allocentric representations for conceptualisation, planning, and consciousness. The outputs of the parietal-lobe steam are not conscious, although they guide movement in just the sort of fine-grained way which fits one aspect of our intuitive idea of experience.

  5. A trilemma for FOR theory

  6. One horn of the trilemma is to dismiss the data. This is dealt with briskly. Another is to accept that the outputs of the parietal system aren’t phenomenally conscious, and to try explaining why they aren’t in purely first-order terms. The problem here is to explain why availability to concepts and conceptual thought should transform intentional contents from ones which aren’t phenomenally conscious into ones which are. I argue that no adequate explanation can be forthcoming here. The final horn of the trilemma is discussed in section 4.

  7. Non-conscious phenomenality?

The final option is to insist that the contents of the parietal stream are phenomenally conscious, but unknowingly so to their subjects because they are not access-conscious. This option is hard to believe, and various additional objections are raised against it.

Chapter 7 Higher-order representationalism: a first defence

In this chapter I take the first steps towards a defence of higher-order representational (HOR) accounts of phenomenal consciousness. I argue that these accounts have considerable explanatory advantages over first-order representational (FOR) theories, and that they have the resources to rebut a number of potentially-devastating objections.

  1. Overview and preliminaries

  2. This section gives a brief review of the tasks ahead, and reminds the reader of the different varieties of HOR theory (first sketched in chapter 1).

  3. HOR theory and qualia irrealism

  4. This section shows how HOR theory can explain desiderata (2) through (5) from 4:1 - explaining why people are tempted to believe (falsely) that our experiences posses intrinsic properties, which are ineffable, private, and known with certainty by the subject. The challenge of explaining the core defining feature of phenomenal consciousness - its subjectivity, feel, or what-it-is-likeness - is held over until chapter 9.

  5. Of animals, infants, and the autistic

  6. This section deals with a seemingly-powerful objection to HOR theory, namely that it entails that non-human animals (as well as human infants, and perhaps also autistic people) are lacking in phenomenal consciousness. It argues that the objection is entirely without force, since grounds for attributing phenomenal consciousness to animals are lacking; and since it is easy to explain why our intuitions that they must be phenomenally conscious are so powerful. (Roughly, the explanation runs: we correctly attribute perceptual experience to animals; but when we try to imagine those experiences from the inside, we inevitably imagine conscious experiences, because our own acts of imagining are phenomenally conscious.)

  7. Moral consequences?

This section tackles another seemingly-powerful objection to HOR theories, this time that they must withhold moral significance from animals. Here I argue that the conclusion does not follow. It may be that first-order - non-phenomenal - frustrations of desire are the most basic and appropriate objects of sympathy and moral concern.

Chapter 8 Dispositionalist higher-order thought theory (1): function

By this time I have argued that some form of higher-order representational (HOR) theory of phenomenal consciousness is to be preferred to any more modest first-order (FOR) approach. It remains to adjudicate between the different forms of HOR account. In the present chapter I deploy a variety of functional and evolutionary considerations to argue that dispositionalist higher-order thought (HOT) theory is greatly preferable to both actualist HOT theory, on the one hand, and to higher-order experience (HOE) theory on the other.

  1. Higher-order experience (HOE) theory

  2. In this section I critically examine (HOE) theories, of the sort defended by Armstrong (1968, 1984) and Lycan (1987, 1996). These are ‘inner sense’ models of phenomenal consciousness. They postulate a set of inner scanners, directed at our first-order mental states, which construct analog representations of the occurrence and properties of those states. I argue that inner-sense theories are functionally and evolutionarily implausible by comparison with higher-order thought (HOT) accounts. The basic objection is that such inner scanners would have to be computationally complex, but that there are no plausible explanations of why they might have evolved, or of what they might be for.

  3. Actualist HOT theory

  4. This section critically examines the form of actualist HOT theory proposed by Rosenthal (1986, 1993). On this account an experience of mine is conscious if and only if it is actually causing an activated higher-order belief that I am undergoing that experience, and causing it non-inferentially. The main difficulty for this account is the objection from cognitive overload - what is to explain the huge number of higher-order beliefs which would have to be caused by any given phenomenally conscious experience, given the richness and detail which is standardly present in such experience? This objection is developed at some length. But it depends upon an assumption of experiential richness, challenged by Dennett (1991). This is left as an issue to be returned to in chapter 11.

  5. Dispositionalist HOT theory

  6. In this section I develop and defend a dispositionalist form of HOT theory, according to which the conscious status of an experience consists in its availability to HOT. As with actualist HOT theory, in its simplest form we have here a quite general proposal concerning the conscious status of any type of occurrent mental state, which becomes an account of phenomenal consciousness when the states in question are experiences (or images, emotions, etc.) with analog content (narrowly individuated); thus:

    Any occurrent mental state M, of mine, is conscious = M is disposed to cause an activated belief (possibly a non-conscious one) that I have M, and to cause it non-inferentially.

    In contrast with the actualist form of HOT theory, the HOTs which render M conscious are not necessarily actual, but potential. So the objection now disappears, that an unbelievable amount of cognitive space would have to be taken up with every conscious experience. There need not actually be any HOT occurring, in order for a given perceptual state to count as phenomenally conscious, on this account.

    This theory is elaborated, and it is shown how phenomenal consciousness admits of a smooth and plausible evolutionary explanation on this account.

  7. Dispositional theory and categorical experience

This section confronts an obvious challenge to dispositionalist HOT theory. When I am subject to a conscious experience, there is something actually taking place in me which constitutes my state of phenomenal consciousness. How, then, can the conscious status of my experience consist merely in the fact that I am disposed to have an appropriate HOT about it if circumstances should demand? For this is not something which is actually happening, but merely something which would happen if certain other things happened. The section argues that there are three dissociable strands in this complaint, in fact. Two are dealt with easily and briskly in the remainder of the section. The third is addressed at some length in chapter 9.

Chapter 9 Dispositionalist higher-order thought theory (2): feel

In this chapter we come to the crux. I examine how the competing higher-order accounts can explain the defining feature of phenomenal consciousness – namely its subjective feel, or ‘what-it-is-likeness’ – and I give a final adjudication between them on this ground. I argue that each of the two forms of higher-order thought (HOT) theory – in contrast with higher-order experience (HOE) theory – can advance essentially the same (fully successful) reductive explanation of phenomenal consciousness. In which case, given the strength of the arguments which we were able to deploy in chapter 8 on behalf of dispositionalist forms of HOT theory, it is the latter which should emerge as the overall winner.

  1. HOE theory and feel

  2. This section argues briefly that the weaknesses of ‘inner sense’ theory are not compensated for by any advantage in explaining the subjectivity of phenomenally conscious experience. On the contrary, such theories face the problem of ‘excess content’ - if there really were the inner scanners postulated by such theories, one would expect that conscious experience should be factorable into two distinct aspects, one provided by our first-order senses, and one provided by our inner senses. But this expectation is not borne out.

  3. Actualist HOT theory and feel

  4. This short section argues for the inadequacy of Rosenthal’s (1998) explanation of the feel of experience in terms of actual higher-order thought. It is pointed out, however, that actualist HOT theory could embrace the very same consumer-semantic explanation to be offered in section 3, but that the result would be theoretically ill-motivated in comparison with its dispositionalist cousin.

  5. Consumer semantics and feel

  6. This lengthy section constitutes the heart of the book. My claim is that the very same perceptual states which represent the world to us (or the conditions of our own bodies) can at the same time represent the fact that those aspects of the world (or of our bodies) are being perceived. It is the fact that the faculties of thinking to which experiences are made available can make use of those experiences in dual mode which turns them into dual-mode representations. This is because, in general, the intentional content of a state will depend upon the nature and powers of the ‘consumer-systems’, as Millikan (1984) would put it. The content possessed by a given state depends, in part, upon the uses which can made of that state by the systems which can consume it or draw inferences from it. And similarly, then, in the case of perceptual representations: it is the fact that perceptual contents are present to a system which is capable of discriminating between, and of making judgements about, those perceptual states as such which constitutes those states as second-order representations of experience, as well as first-order representations of the world (or of states of the body). And it is in virtue of this that those states acquire a subjective dimension, or feel.

    Note that on this account phenomenal consciousness is constituted by higher-order analog representations, or higher-order experiences (HOEs), just as HOE theory – or ‘inner sense’ theory – maintains. But there don’t actually need to be two physically distinct sets of representations to carry the two sets of perceptual contents, in the way that HOE theory supposes. Rather, dual content comes for free with the availability of perceptual contents to the mind-reading faculty, or with the availability of those contents to HOT. It is in virtue of the availability of first-order perceptual contents to a mind-reading system which understands the is/seems distinction and/or contains recognitional concepts of experience, that all of those first-order contents are, at the same time, higher-order ones.

    This account is elaborated at some length, and its virtues expounded.

  7. Elucidations and replies

This is another lengthy and important section. It further elaborates on dispositionalist HOT theory’s explanation of phenomenal consciousness by way of responding to a series of potential objections and counter-examples.

Chapter 10 Phenomenal consciousness and language

In this chapter I argue that the simple form of dispositionalist HOT theory of phenomenal consciousness defended in chapters 8 and 9 is preferable to three other similar but more elaborate accounts (due to Carruthers, 1996; Dennett, 1978; and Dennett, 1991 respectively). Each of these is a form of dispositionalist HOT theory, but each makes out a constitutive connection of some sort between phenomenal consciousness and language.

  1. Reflexive thinking theory and language

  2. In this section I contrast – favourably – the account outlined in chapter 8 with a rather more elaborate form of HOT theory of phenomenal consciousness, defended in some earlier publications of mine (e.g. 1996), which I refer to as ‘reflexive thinking theory’. On this account, experiences are conscious when they are available to acts of higher-order thinking which are reflexively available to further higher-order thinkings. This more demanding theory is shown to be ill-motivated for explanatory purposes. But I suggest that it may well be that reflexive thinking theory does, de facto, describe the (language-involving) structure of human consciousness, as I claimed in my 1996.

  3. Higher-order description (HOD) theory

  4. In this section I begin to explore a set of views – focusing mainly on Dennett – which are like dispositionalist HOT accounts, except that they define state-consciousness in general, and phenomenal consciousness in particular, in terms of the availability of a state to higher-order linguistic description. In the present section I examine and criticise Dennett’s early view (1978), which can be seen as a simplified, language-based, form of reflexive thinking theory.

  5. The Joycean machine

  6. This section examines Dennett’s (1991) theory of the conscious mind as a language-based Joycean machine. The theory is elaborated and its connections with evolutionary issues discussed.

  7. The independence of structured HOTs from language

In order for higher-order linguistic description (HOD) theory to be preferred to higher-order thought (HOT) theory as an account of the phenomenal consciousness of human beings, it has to be the case that all hominid thought (realistically construed, as involving discrete, structured, content-bearing states) involves natural language. More particularly, it has to be the case that there can be no structured HOTs except those which are formulated in language. The present section defends dispositionalist HOT theory by arguing, on a variety of grounds, that the human capacity for HOTs is independent of language.

Chapter 11 Fragmentary consciousness and the Cartesian theatre

I have argued for the superiority of higher-order thought (HOT) theory over higher-order description (HOD) theory: there is no reason to think that a theory of phenomenal consciousness should implicate natural language, and there is good reason to think that it should not. This leaves untouched Dennett’s (1991) arguments for a ‘multiple drafts’ approach to phenomenal consciousness, however, and in support of the radical indeterminacy of facts concerning the latter; together with his attacks on ‘Cartesian theatre’ models of phenomenal consciousness (of which both his own earlier 1978 theory and my sort of HOT theory are alleged examples). These arguments form the topic of this final chapter.

  1. Multiple drafts versus integrated contents

  2. This section defends the intuitive richness of phenomenally conscious experience against Dennett’s attacks (this is the issue held over from 8:2). It also argues positively that practical reasoning in relation to the perceived environment requires a set of integrated perceptual contents.

  3. Fragmenting the Cartesian theatre

  4. This section begins defending dispositionalist HOT theory against the charge that it is committed to a ‘Cartesian theatre’ conception of the conscious mind. Various distinct strands in this charge are distinguished, and a number of them are shown to be innocuous. Those which are more challenging are held over to sections 3 and 4.

  5. Time as represented versus time of representing

  6. This section demonstrates that dispositionalist HOT theory - like Dennett’s own theory - can maintain that time is represented in the brain, rather than being given by time of representing. However, the account is genuinely committed to the idea that there is an objective (albeit vague) time at which an experience first becomes conscious, which may be distinct from the time at which that state is experienced as occurring.

  7. Objective versus subjective time

Dennett develops an argument against the very coherence of the idea that there might be an objective, determinate, time at which an experience first becomes phenomenally conscious, as distinct from the subject’s representation of the time at which the experience first occurs (1991; Dennett and Kinsbourne, 1992). This section expounds and elaborates on that argument, and unpicks the assumptions on which it is based. The upshot, then, is that Dennett has no damaging objections to the dispositionalist HOT theory being proposed in this book.


The book concludes with a brief survey of the main course of the argument, and of its major premises. It closes with a slogan: a disposition to get higher makes consciousness phenomenal.


Armstrong, D. 1968. A Materialist Theory of the Mind. Routledge.

Armstrong, D. 1984. Consciousness and causality. In D. Armstrong and N. Malcolm, Consciousness and Causality, Blackwell.

Block, N. 1990. Inverted Earth. Philosophical Perspectives, 4.

Botterill, G. and Carruthers, P. 1999. The Philosophy of Psychology. Cambridge University Press.

Carruthers, P. 1996. Language, Thought and Consciousness. Cambridge University Press.

Chalmers, D. 1996. The Conscious Mind. Oxford University Press.

Dennett, D. 1978. Toward a cognitive theory of consciousness. In C. Savage ed., Minnesota Studies in the Philosophy of Science, 9.

Dennett, D. 1991. Consciousness Explained. Allen Lane.

Dennett, D. and Kinsbourne, M. 1992. Time and the observer. Behavioral and Brain Sciences, 15.

Dretske, F. 1995. Naturalizing the Mind. MIT Press.

Jackson, F. 1982. Epiphenomenal qualia. Philosophical Quarterly, 32.

Jackson, F. 1986. What Mary didn’t know. Journal of Philosophy, 83.

Kripke, S. 1972. Naming and necessity. In G. Harman and D. Davidson, eds., Semantics of Natural Language, Reidel.

Lycan, W. 1987. Consciousness. MIT Press.

Lycan, W. 1996. Consciousness and Experience. MIT Press.

McGinn, C. 1991. The Problem of Consciousness. Blackwell.

Millikan, R. 1984. Language, Thought, and Other Biological Categories. MIT Press.

Milner, D. and Goodale, M. 1995. The Visual Brain in Action. Oxford University Press.

Nagel, T. 1974. What is it like to be a bat? Philosophical Review, 83.

Nagel, T. 1986. The View from Nowhere. Oxford University Press.

Penrose, R. 1994. Shadows of the Mind. Oxford University Press.

Rosenthal, D. 1986. Two concepts of consciousness. Philosophical Studies, 49.

Rosenthal, D. 1993. Thinking that one thinks. In Davies and Humphreys, eds., 1993.

Rosenthal, D. 1998. State consciousness and what it’s like. Paper delivered to a cognitive neuroscience seminar, Corpus Christi, Oxford. Forthcoming in D. Rosenthal, Consciousness and Mind, Oxford University Press.

Shoemaker, S. 1981. The inverted spectrum. Journal of Philosophy, 74.

Tye, M. 1995. Ten Problems of Consciousness. MIT Press.

© 2001 Peter Carruthers