Hallucination: Philosophy and Psychology

14. The Failure of Disjunctivism to Deal with “Philosophers’ Hallucinations”

14 The Failure of Disjunctivism to Deal with “Philosophers’ Hallucinations” Howard Robinson Abstract I restate the causal-hallucinatory argument against naive realism. This argument depends on the possibility of “philosophers’ hallucinations.” I draw attention to the role of what I call the nonarbitrariness of philosophers’ hallucinations in supporting this argument. I then discuss three attempts to refute the argument. Two—those associated with John McDowell and with Michael Martin—are explicitly forms of disjunctivism. The third, exemplified by Mark Johnston, has, I claim, disjunctivist features. None of these responses to the argument is plausible. 1 “Philosophers’ Hallucinations” Philosophers are (mainly, at least) interested in what I shall call “philosophers’ hallucinations .” These are not, as far as we know, hallucinations as they actually occur; they are, it is argued, the hallucinations that would occur if the perceptual system and brain were stimulated in just the way they are stimulated in genuine perception, but directly and not by the usual external objects. This would give, it is supposed, a hallucination indistinguishable to the subject from the corresponding perception, which is not the case, at least in general, for hallucinations as they actually occur. A belief in the possibility of such hallucinations is taken as grounds for rejecting naive or direct realism, as follows. (1) Possibility of Philosophers’ Hallucinations It is theoretically possible, by activating some brain process that is involved in a particular type of perception, to produce a hallucination that is subjectively indiscriminable from that perception. (2) Same Proximate Cause, Same Immediate Effect It is necessary to give the same account of hallucinations and perceptual experiences when they have the same neural cause. Thus it is not possible to say, for example, that 314 H. Robinson the hallucinatory experience involves a subjective image or sense-datum, but the perception does not, if they have the same proximate—that is, neural—cause. (3) Hallucinations do involve some subjective image or sense-datum. Therefore (4) Perception involves some subjective image or sense-datum. The main response to this argument from defenders of direct realism has been to adopt a disjunctivist theory of perception, which is supposed to show that one cannot argue from the nature of hallucination to the nature of perception, and so to undermine (2). Disjunctivism is not so much an argument against (2) as a simple denial of it. According to the disjunctivist, a proposition of the form S seems to see something F is essentially generic, being disjoined into either (a) S sees something F or (b) S is illuded that he sees something F. The state denoted by (b) does not figure in the state captured by (a), even under another description. This is a straight denial of (2).1 A contrast has now been set up between disjunctivism and the common-factor theory, which says that seeing and being illuded share a common component, and is the core claim in (2). It is worth pausing for a moment to recognize that, strictly speaking, the options are not a straightforward disjunction between disjunctivism and the commonfactor theory as so far set out. In particular, disjunctivists can hold that there is a common factor, provided that this is not such as to constitute the full phenomenal character of both hallucination and perception. There could be weaker common-factor claims. For example, one might hold that hallucination and perception share a common factor, but hallucination is that common factor plus something further (for example, a mental image) and perception is the common factor, plus something else 1. Disjunctivism is generally deployed to cope with hallucinations, but there seems often to be an ambiguity about whether it is supposed also to handle the kinds of perceptual relativities covered by the so-called argument from illusion. If it is, and if such “illusions” cover all cases of an object appearing other than it exactly is, then almost all—if not all—cases of perception will be cases of “being illuded.” This will be no help to the direct realist. If, on the other hand, disjunctivism does not touch these cases of illusion, then the direct realist still faces the challenge of accounting for them. My suspicion is that the disjunctivist would like to think that he has a cure-all that deals with all cases, but fears to look too closely at how it might be applied to nonhallucinatory cases. The Failure of Disjunctivism to Deal with “Hallucinations” 315...