![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
In this paper I aim to defend David Rosenthal's higher-order thought (HOT) theory of consciousness against a prominent objection. The central claim of HOT theory is that a mental state is conscious only if one has the HOT that one is in that state. In broad outline, the objection is that HOT theory is unable to account for cases where the relevant HOTs are false. I consider two variants of the objection, corresponding to two kinds of false HOT: those that merely misrepresent their targets, and those which lack targets altogether. I argue that a satisfactory response to the objection involving the latter, targetless, kind of HOTs, is to deny that one is in any conscious state in virtue of such HOTs. I show how this response is superior to Rosenthal's own response, and defend it against objections. I also argue that my account of targetless HOT cases may be generalized to cover cases of misrepresenting HOTs.
Notes
Notes
[1] By ‘thought’ Rosenthal means an occurrent, assertoric mental state with propositional content (Rosenthal, 2002, p. 411). The specific form of this propositional content is discussed below.
[2] In footnote 16, Rosenthal makes a qualification to this claim, allowing that in special cases, one can be aware of the inference in question.
[3] As Rosenthal notes, the example he discusses is not precisely an example of what psychologists call the ‘tip-of-the-tongue phenomenon’, since the latter involves having conscious access to partial information (Rosenthal, 2000a, p. 204).
[4] The objection is discussed in Balog (2000a), Byrne (1997), Gennaro (2004), Kriegel (2003), Levine (2001), Matey (2006), Neander (1998), Seager (1999), and Weisberg (2008). Most of these theorists also express the thought that this is a particularly damaging objection (see, e.g., Kriegel, 2003, p. 119).
[5] This example is discussed by Byrne (1997).
[6] Kriegel (2003, p. 119) and Weisberg (2008, p. 164) draw a distinction between different kinds of misrepresenting HOTs along these lines.
[7] Such theories trace their origins to Evans (1982, p. 225), and various forms of the theory have been defended by Carruthers (2000), Goldman (2006), Gordon (1996), and Nichols and Stich (2003). See Goldman (2006, p. 238) for the similarities between Nichols and Stich's and Gordon's views, which he classifies together as ‘redeployment accounts’.
[8] Carruthers (2005, p. 110) presents an argument of this sort. Indeed he seems to think that this protects his version of HOT theory from any objection from false HOTs.
[9] This condition was introduced into the philosophical literature by Patricia Churchland (1986, p. 228). A historical survey is provided in McGlynn and Schacter (1989).
[10] It is sometimes thought that Anton's syndrome patients in fact hallucinate the things they claim to see. In that case the patient's HOT may be accurate with respect to the content of his visual experience. But whether the HOT counts as roughly accurate concerning the attitude of the visual state will depend on complex issues concerning the relation between perception and hallucination. On a disjunctivist account of perception, whereby perception and hallucination have little in common other than subjective indistinguishability, the HOT might be thought to be substantially in error, and thus ‘targetless’, on my account. On disjunctivism see Haddock and MacPherson (2008).
[11] Byrne (1997) seems to assume that this is the only serious option available to the HOT theorist.
[12] On Levine's (2001, pp. 108–109) account of the objection (also found in Gennaro, 2004), there are three live options for the HOT theorist (aside from denying that the false HOT can occur at all). Recall that Levine hypothesizes a false HOT situation in which I have the HOT that I am having a visual experience of green, when in fact I am visually experiencing red. He suggests the three options available to the HOT theorist are (1) that my consciousness is greenish, (2) that my consciousness is reddish and (3) that there is no consciousness at all. The two accounts I am considering here correspond to options (1) and (3), in Levine's scheme. Levine's option (2), by contrast is arguably best seen as a solution to the problem of misrepresenting HOTs, as discussed in section 4.
[13] See Rosenthal (1990); for a more recent account of the distinction, see Rosenthal (2002, pp. 407–408).
[14] Creature consciousness can be subdivided further into transitive and intransitive varieties, but this is irrelevant for present purposes.
[15] If Kim's view of the metaphysics of events is correct, then mental state tokens, as I am defining them, are events. Nothing here depends on this theory of events, however.
[16] My discussion of the problem with Rosenthal's account here may be seen expanding on Kriegel's observation that “if M1 is not a state x is in, it can hardly be a conscious state x is in” (2003, p. 120).
[17] Another exception is Gennaro (2004, p. 58) whose more positive discussion unfortunately conflates the account with the rather different idea that HOTs are infallible.
[18] I am grateful to an anonymous referee for suggesting this switching scenario.
[19] Note that Weisberg himself is not actually criticizing Rosenthal in this paper, but rather paraphrasing (in a revealing way, I think) an objection made to Rosenthal by Kriegel (2003). Note also that, strictly speaking, one ought to distinguish seeming to be in some mental state, and seeming to be in some conscious mental state, in virtue of a HOT. Presumably, in order for one to seem to be in a conscious mental state in virtue of a HOT, the state's property of being conscious ought to feature in the propositional content of the HOT. This distinction makes little difference in the present context, however.
[20] This kind of argument may lie behind Rosenthal's frequent claims, in support of his account of targetless HOTs, to the effect that “a HOT's accompanying its target is subjectively indistinguishable from a HOT's occurring in the absence of that target” (2004, p. 41).
[21] Note that this is not simply the claim that our self-attribution of mental states is fallible concerning those mental states. It is a claim about the relation between the self-attribution of mental states and specifically conscious mental states.
[22] One might say that on their view consciousness is a matter of the way the world, rather than our minds, appears to us.
[23] I would like to thank Uriah Kriegel for emphasizing this objection to me.
[24] Similarly Dennett, for all his talk of ‘first-person operationalism’, and his denial of ‘real seemings’ over and above judgement (1991, p. 364), ultimately concedes that “when people generalize or theorize about what it's like, they cede their authority [concerning consciousness] altogether” (2007, p. 263).
[25] It is true there is a difficult problem here about whether the conceptual, propositional contents of HOTs can truly capture the apparently analogue or non-conceptual properties that seem to be involved in qualitative properties such as piercingness. Rosenthal's own solution to this problem is that “the mental properties of our sensations appear ultimately homogeneous to us simply because the way we are conscious of them smooths them out, so to speak, and elides the details of their particulate, bit-map nature” (1999/2005a, p. 174).