1 In this paper, I use ‘self-awareness’ and ‘self-consciousness’ interchangeably.

2 I call the phenomenon denoted in (a) transitive self-consciousness because ‘to be self-conscious of’ is a transitive verb. ‘Self-consciously’ is not a verb at all, but an adverb, but I call the phenomenon it is used to denote intransitive self-consciousness mainly to mark it off from transitive self-consciousness.

3 Goldman, A. A Theory of Action (Princeton, NJ: Princeton University Press 1970), 96Google Scholar

4 It is possible, of course, for x to have M without being aware that she is having M, or without being aware of M as hers, but then x is not having M self-consciously. So in the sense of intransitive self-consciousness, M would be non-conscious.

5 To be sure, much more can — and should — be said about the relationship between intransitive self-consciousness and phenomenal consciousness. In particular, my claim that the former is a necessary condition for the latter must be defended more systematically. However, a full discussion of the issues involved will take us too far afield. In this paper, I am mainly interested in the nature of intransitive self-consciousness itself. I invite those readers who still feel that intransitive self-consciousness is not comfortably treated as a phenomenon of consciousness to take the argument of this paper as simply targeting the phenomenon of intransitive self-consciousness, whether or not it is related to consciousness. For a more extensive discussion of the role of self-awareness in consciousness, see Kriegel, U. ‘Consciousness as Sensory Quality and as Implicit Self-Awareness,’ Phenomenology and the Cognitive Sciences 2 (2003) 1–26CrossRefGoogle Scholar.

6 In a philosophical world in which eliminative stances with regard to qualia, content, or propositional attitudes normally remain outside the mainstream, eliminativism about primordial forms of self-awareness has curiously been the norm. This is bemoaned extensively by contemporary German philosophers who follow the analytic philosophy of mind, especially members of the so-called ‘Heidelberg School,’ e.g., Henrich, D. ‘Fichte's Original Insight,’ Lachterman, D.R. Contemporary German Philosophy 1 (1982) 15–53Google Scholar; Frank, M. ‘Mental Familiarity and Epistemic Self-Ascription,’ Common Knowledge 4 (1995) 30–50Google Scholar; Sturma, D. ‘Self-Consciousness and the Philosophy of Mind: A Kantian Reconsideration,’ Proceedings of the Eighth International Kant Congress vol. 1 (Milwaukee, WI: Marquette University Press 1995)Google Scholar. See also Kapitan, T. ‘The Ubiquity of Self-Awareness,’ Grazer Philosophische Studien 57 (1999) 17–44.Google Scholar

7 Similar views are propounded by Descartes (see Rodis-Lewis, G. Le probleme de l'inconscient et le cartesianisme [Paris: Presses Universitaires de France 1950]Google Scholar; Wider, K. The Bodily Nature of Consciousness: Sartre and Contemporary Philosophy of Mind [Ithaca, NY: Cornell University Press 1997])CrossRefGoogle Scholar, Cudworth (see Thiel, U. ‘Cudworth and Seventeenth-Century Theories of Consciousness,’ in Gaukroger, S. ed., The Ilses of Antiquity: The Scientific Revolution and the Classical Tradition [Dordrecht: Kluwer 1991])Google Scholar, Leibniz (see Gennaro, R.J. ‘Leibnitz on Consciousness and Self-Consciouness,’ in Gennaro, R.J. and Huenemann, C. eds., New Essays on the Rationalists [New York: Oxford University Press 1999]Google Scholar), and Kant (see Gennaro, R.J. Consciousness and Self-Consciousness [Philadelphia: John Benjamin 1996]CrossRefGoogle Scholar; D. Sturma; K. Wider), but the view goes all the way back to Aristotle (see Caston, V. ‘Aristotle and Consciousness,’ Mind 111 [2002] 751–815)CrossRefGoogle Scholar.

8 It is not required by (SA₁) that M be the only thing that M^* represents. It may be one among several elements figuring in M^*'s content.

9 Some may take the question whether there are two distinct states involved or only one State to be a wholly arbitrary matter. On this view, what counts as one State or two states is up to us. And on this view, the difference between (SA₂) and (SA₃) is insignificant, if real at all. It seems to me, however, that what counts as one State or two states is not generally up to us. Thus, there is a psychologically real difference between believing that p and q and believing that p and that q. The former involves holding a Single conjunctive belief while the latter involves holding a conjunction of belief s. That this is a real difference is manifest in the fact that a person may believe that p and that q without believing that p and q. (For a demonstration of this, see Williams, J.N. ‘Inconsistency and Contradiction,’ Mind 90 [1981] 600–2CrossRefGoogle Scholar.) Moreover, as we will see below, (SA₂) leads to certain problems avoided by (SA₃). If so, the difference between them must be very real indeed.

10 D.M., Armstrong A Materialist Theory of the Mind (New York: Humanities Press 1968)Google Scholar, and ‘What is Consciousness?’ in Block, N.J. O., Flanagan and Guzeldere, G. eds., The Nature of Consciousness: Philosophical Debates (Cambridge, MA: The MIT Press 1997)Google Scholar; Carruthers, P. Language, Thought, and Consciousness (Cambridge: Cambridge University Press 1996)CrossRefGoogle Scholar; Dennett, D. Content and Consciousness (London: Routledge and Kegan Paul 1969)Google Scholar; W.G., Lycan ‘Consciousness as Internal Monitoring,’ Philosophical Verspectives 9 (1990) 1–14Google Scholar, and Consciousness and Experience (Cambridge, MA: The MIT Press 1996); Rosenthal, D.M. ‘Two Concepts of Consciousness,’ Philosophical Studies 94 (1986) 329–59CrossRefGoogle Scholar, ‘A Theory of Consciousness,’ in Block et al., eds., The Nature of Consciousness, and Consciousness and Mind (Oxford: Oxford University Press forthcoming); Gulick, R. Van ‘Inward and Upward — Reflection, Introspection, and Self-Awareness,’ Philosophical Topics 28 (2001) 275–305.CrossRefGoogle Scholar

11 Although this is debatable. Locke is commonly taken by current-day proponents of (SA₂) to have anticipated their view, but there is nothing in his writings to rule out (SA₃) and quite a bit to suggest it (see footnote 19 below). As for Kant, Rosenthal, D.M. ‘Consciousness and the Mind,’ lyyun 51 (2002) 227–51Google Scholar, interprets him as a proponent of (SA₂), but Gennaro, Consciousness and Self-Consciousness interprets him as a proponent of (SA₃), or something very close to it. (According to Gennaro, Kant holds that M^* is a non-conscious part of the same mental State M is a conscious part of.) Descartes, by contrast, was clearly more in line with the (SA₃) view (see Rodis-Lewis).

12 Brentano, F. Psychologe from Empirical Standpoint, Rancurello, A.C. Terrell, D.B. and McAlister, L.L. trans., McAlister, L.L. ed. (London: Routledge and Kegan Paul 1973 [1874])Google Scholar

13 Brough, J.B. ‘The Emergence of an Absolute Consciousness in Husserl's Early Writings on Time-Consciousness,’ Mand and World 5 (1972) 298–326CrossRefGoogle Scholar; Frank, M. ‘Mental Familiarity’; Gurwitsch, A. Marginal Consciousness (Athens, OH: Ohio University Press 1985)Google Scholar; D., Henrich ‘Fichte's Original Insight’; Husserl, E. Phenomenology of Internal Time-Consciousness, Churchill, J.S. trans., Heidegger, M. ed. (Bloomington, IN: Indiana University Press 1964)Google Scholar; Natsoulas, T. ‘The Case for Intrinsic Theory: I. An Introduction,’ Journal of Mind and Behavior 17 (1996) 267–86Google Scholar; Sartre, J.-P. La Transcendence de l'ego (Paris: Vrin 1937) and L'Etre et le Neant (Paris: Gallimard 1943)Google Scholar; Smith, D.W. ‘The Structure of (Self-)Consciousness,’ Topoi 5 (1986 149–56CrossRefGoogle Scholar and The Circle of Acceptance (Dordrecht: Kluwer 1989); Sokolowski, R. Husserlian Meditations (Evanston, IL: Northwestern University Press 1974)Google Scholar; K. Wider, The Bodily Nature of Consciousness; Zahavi, D. ‘Brentano and Husserl on Self-Awareness,’ Etudes Phenomenologiques 27–8 (1998) 127-69 and Self-Awareness and Alterity (Evanston, IL: Northwestern University Press 1999)Google Scholar

14 P., Carruthers Phenomenal Consciousness (Cambridge: Cambridge University Press 2000)Google Scholar; Caston, V. ‘Aristotle on Consciousness’; Gennaro, R. Consciousness and Self- Consciousness; J. Levine, Purple Haze: The Puzzle of Consciousness (New York: Oxford University Press 2001)Google Scholar; Thomasson, A.L. ‘After Brentano: A One-Level Theory of Consciousness,’ European Journal of Philosophy 8 (2000) 190–209CrossRefGoogle Scholar; Gulick, R. Van ‘Inward and Upward'; Kriegel, U. ‘Consciousness, Permanent Self-Awareness, and Higher-Order Monitoring,’ Dialogue 41 (2002) 517–40Google Scholar and ‘Consciousness, Higher-Order Content, and the Individuation of Vehicles,’ Synthese 134 (2003) 477-504. As you may notice, some philosophers who have defended (SA2) earlier on have recently turned to (SA₃). Usually, they still call their views Higher-Order Monitoring — in particular, Carruthers and Van Gulick do — but the way I will use the terms here, Higher-Order Monitoring refers only to (SA₂)-style theories.

15 Brentano, 153-4. The view is first introduced in Section 7 of chapter II (’Inner Consciousness’) in Book 2, which is entitled ‘A Presentation and the Presentation of that Presentation are Given in One and the Same Act,’ In this section, Brentano canvasses his conception of conscious experiences as self-representational.

16 For more on the relationship between Aristotle and Brentano, see Caston. Caston notes that an Interpretation of Aristotle along these lines is also to be found in a dissertation on the unity of mental life in Aristotelian philosophy, written under Brentano's supervision by one J. Herman Schell.

17 There is a long tradition of reading this passage in terms of perceptual modalities, or senses, rather than in terms of perceptual states, or activities. But that is not how Brentano reads it, and Caston argues in favor of Brentano's reading. My understanding is that the Greek word used by Aristotle, αισθησις, is ambiguous between sense and State. I am using here Caston's translation of the passage.

18 (1) is definitional; (13) is empirically true; and (15) is a methodological point I see no justification for rejecting.

19 As a historical side note, it is interesting to note that Freud appears to have held Brentano's view about conscious states — that they represent their own occurrence within the subject — but did not couple it with the Cartesian notion that all mental states are conscious. His position was that not all mental states were conscious, but those which were represented themselves (see Natsoulas, T. ‘Freud and Consciousness: I. Intrinsic Consciousness,’ Psychoanalysis and Contemporary Thought 7 [1984] 195–232)Google Scholar. This position is of course coherent, and in my opinion is in fact true, but it undermines Brentano's specific argument for his conception of conscious states as self-representational.

20 Leibniz may have anticipated this line of reasoning. Leibniz is perhaps the first philosopher to have recognized the existence of non-conscious states (’small thoughts’), and he offers the following way out of the infinite regress: ‘It is impossible that we should always reflect explicitly on all our thoughts; [otherwise] the mind would reflect on each reflection ad infinitum…. It must be that… eventually some thought is allowed to occur without being thought about; otherwise I would dwell forever on the same thing’ (quoted in Gennaro, ‘Liebniz on Consciousness,’ 355-6).

21 Caston argues that this is precisely how Aristotle meant his argument. The word Aristotle uses is not ‘states,’ but ‘activities,’ which suggests an occurrent reading.

22 Adapted from Malcolm, N. ‘Thoughtless Brutes,’ in Rosenthal, D.M. ed., The Nature of Mind (New York: Oxford University Press 1991).Google Scholar

23 Another, less disturbing problem is that while premise (1) in the previous version of the argument was more or less definitionally true, premise (1) in the present version is not. Thus, according to proponents of the Dispositional Higher-Order Monitoring theory (e.g., Carruthers, Language, Thought, and Consciousness), what makes a mental State conscious is not that it is represented by an occurrent mental State, but that it is represented by a dispositional mental State. That is, they reject premise (1) of the current version of the argument. Their view may not be very plausible, but it is nowise ruled out by the present version of Brentano's argument.

24 This must be the case if Locke is to be interpreted as a proponent of (SA₂) at all. I am not sure that he was, but in any event it is clear that he did not countenance non-conscious states: ‘Whilst [the soul] thinks and perceives … it must necessarily be conscious of its own perceptions’ (Essay II, i, 12). This means that Locke was either oblivious to the infinite regress consequent upon holding Position 2, or held rather something like Position 3 and is wrongly appropriated by current-day Higher-Order Monitoring theorists.

25 In another sense, it refers to the fact that all or many of the subject's conscious states are united in a Single personal consciousness. But that is not the sense I am interested in here.

26 R., Aquila ‘Consciousness and Higher-Order Thoughts: Two Objections,’ American Philosophical Quarterly 27 (1990) 81–7Google Scholar; Byrne, A. ‘Some Like it HOT: Consciousness and Higher-Order Thoughts,’ Philosophical Studies 86 (1997) 103–29;CrossRefGoogle Scholar Carruthers, P. Phenomenal Consciousness; Dretske, F.I. ‘Conscious Experience,’ Mind 102 (1993) 263–83Google Scholar; Goldman, A. ‘Consciousness, Folk Psychology, and Cognitive Science,’ Consciousness and Cognition 2 (1993) 364–83;CrossRefGoogle Scholar Guzeldere, G. ‘Is Consciousness the Perception of What Passes in One's Own Mind?’ in Conscious Experience, Metzinger, T. ed. (Padborn: Schoeningh-Verlag 1995)Google Scholar; Levine, J. Purple Haze; R. Moran, Authority and Estrangement: An Essay on Self-Knowledge (Princeton, NJ: Princeton University Press 2001)Google Scholar; Natsoulas, T. ‘What Is Wrong with Appendage Theory of Consciousness?’ Philosophical Psychology 6 (1993) 137–54CrossRefGoogle Scholar; Neander, K. ‘The Division of Phenomenal Labor: A Problem for Representational Theories of Consciousness,’ Philosophical Perspectives 12 (1998) 411–34Google Scholar; Rey, G. ‘A Question about Consciousness,’ in Perspectives on Mind, Otto, H. and Tueidio, J. eds., (Norwell: Kluwer 1988)Google Scholar; Shoemaker, S. ‘Self-Knowledge and “Inner Sense.” Lecture II: The Broad Perceptual Model,’ Philosophy and Phenomenological Research 54 (1994) 271–90.CrossRefGoogle Scholar

27 The Higher-Order Monitoring theory comes in two varieties, depending on how the second-order State is construed. If it is construed as a perception-like State, we have a Higher-Order Perception theory; if it is construed as an intellectual thought, we have a Higher-Order Thought theory. Byrne presents the problem of targetless higher-order states for (Rosenthal's) Higher-Order Thought theory, whereas Neander presents it for (Lycan's) Higher-Order Perception theory. Levine develops the argument in a generic way, against both varieties of Higher-Order Monitoring theory.

28 Quoting this very passage, Levine (109, n.24) comments: ‘But doesn't this give the game away?’ To all appearances, it does.

29 There is no Cartesian voodoo involved in this form of immunity to misrepresentation. It is simply an artifact of the fact that M₁'s occurrence is a condition of M₁'s representation of its occurrence. That is, the obtaining of the State of affairs M₁ purports to represent is a condition of M₁'s representation of this State of affairs, and this is why M₁ cannot misrepresent its own occurrence. So M₁ can fall to occur or it can represent its own occurrence while occurring, but it cannot represent its own occurrence while falling to occur. Recall, however, that M₁ can perfectly well misrepresent anything else it may represent (other than its own occurrence).

30 Goldman's (‘Consciousness, Folk Psychology’) argument against the HOM theory unfolds along these lines. It is not essential to this argument that it be coupled with an account of the difference between first-person knowledge and third-person knowledge. All is required is a commitment to the existence of such a distinction. Some philosophers may feel that the very distinction is an affront to a naturalist conception of knowledge. But this is clearly misguided: if there is a distinction between first-person and third-person knowledge, it must be possible to account for it in naturalist terms. For an attempt to explain the distinction in naturalist terms, see, e.g., Dretske, F.I. Naturalizing the Mind (Cambridge, MA: The MIT Press 1995)Google Scholar, Fernandez, J. ‘Privileged Access Naturalized,’ Philosophical Quarterly 53 (2003) 352–72.CrossRefGoogle Scholar

31 A fuller formulation of the argument would go as follows:

32 It is also quite common in the phenomenological tradition to accuse Brentano of falling prey to the same infinite regress he used in order to argue for his view. This is one of the main tenets of the Heidelberg School (see M. Frank), and is often emphasized by Zahavi (see ‘Brentano and Husserl’).

33 Furthermore, some proponents of (SA₂) accept the existence of self-representing states, but just do not think that all conscious states are such (e.g., Rosenthal, D.M. ‘Higher-Order Thoughts and the Appendage Theory of Consciousness,’ Philosophical Psychology 6 [1993] 155–66)CrossRefGoogle Scholar. This amounts, however, to an admission that there is nothing innerently mysterious about mental states that carry self-representational content.

34 Chalmers, D.J. ‘Facing Up to the Problem of Consciousness,’ Journal of Consciousness Studies 2 (1995) 200–19Google Scholar, gives a succinct expression of this suspicion: ‘It is common to see a paper on consciousness begin with an invocation of the mystery of consciousness, noting the strange intangibility and ineffability of subjectivity, and worrying that so far we have no theory of the phenomenon. Here the topic is clearly the hard problem — the problem of experience. In the second half of the paper, the tone becomes more optimistic, and the author's own theory of consciousness is outlined. Upon examination, this theory turns out to be a theory of the more straightforward phenomena — of reportability, of introspective access, or whatever. At the close, the author declares that consciousness has turned out to be tractable after all, but the reader is left feeling like a victim of a bait-and-switch. The hard problem remains untouched’ (211).

35 The distinction is noted by Sosa, E. ‘Experience and Intentionality,’ Philosophical Topics 14 (1986) 67–83, at 73.CrossRefGoogle Scholar

36 This difference shows up in the fact that the sub-argument against Position 1 only establishes an intermediate conclusion on the way to establishing Position 3, whereas the phenomenological argument is intended to establish Position 3 directly.

37 On top of this, the second reason offered in §V for rejecting Position 2 still applies: Position 2’ is inconsistent with the unity of consciousness.

38 I would like to thank Gary Bartlett, Elizabeth Vlahos, and Ken Williford, as well as two anonymous ref erees for the Canadiern Journal ofPhüosophy, f or helpful comments on earlier drafts of this article.