![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"})
Notes
1I am grateful to this journal's anonymous referee for helpful suggestions and to Andrew Lugg for extensive criticism and advice.
2According to Stroud, Moore betrays a ‘steadfast refusal or inability to speak or think in a “philosophical” or non-everyday way’ (Stroud, 1984, 100) and ‘resists, or more probably does not even feel, the pressure towards the philosophical project as it is understood by the philosophers he discusses’ (Stroud, 1984, 119).
3McGinn writes:
I am, therefore, in agreement with Stroud that something essential to the resolution of the problem of scepticism is missing from Moore's reply. However, as I see it the lack does not arise because Moore's proof belongs to a realm of discourse entirely disjointed from that in which philosophical illumination of our epistemic position is sought. Rather, it is simply that, while the conclusion of Moore's proof is at once one we are, in normal circumstances, entirely convinced of, and one that is in outright conflict with the conclusion of the philosophical sceptic, it does not provide any understanding of why our normal conviction is not a piece of unacceptable dogmatism.
4Ambrose thinks Moore's argument ‘is the sort used to convince the ordinary man of the existence of something in question’ (Ambrose, 1952 405). She claims he ‘dismisses the sceptic's conclusion by contradicting it but without countering his argument. The great importance of such a reply is to make one feel as the ordinary man feels, as though there is something ridiculous about it’ (Ambrose, 1952, 416). Thus, ‘Moore sets us on our way to seeing that like the impostors of the tale of The Emperor's New Clothes, the sceptic is wearing nothing at all, and that in fact the Emperor is naked’ (Ambrose, 1952, 417).
5Soames writes:
According to Moore, the problem with the sceptic is that he has adopted a philosophical theory about what knowledge consists in that is far too restrictive. The sceptic assumes that we can be certain about what knowledge is before we decide whether what we all ordinarily take to be paradigmatic cases of knowledge really are genuine. But this is backwards. Moore would say that one fundamental way to test any theory about what knowledge consists in is to determine whether it is consistent with what we all recognise to be the most basic and paradigmatic examples of knowledge. If the theory is not consistent with these examples, then Moore would insist that this result constitutes strong evidence against it.
6These four commentaries span the period 1952–2003. I use them as examples of reactions to Moore's proof without meaning to deny that there are many other accounts equally worthy of discussion.
7Stroud discusses these papers but only in a cursory way.
8As Soames puts it:
On [Moore's] view, both the sceptic and the philosopher who tries to provide the proof demanded by the sceptic accept an unjustified theory of what knowledge consists in. This diagnosis brings out the ironic nature of Moore's presentation. Would anyone who believed that a proof of the external world was needed be satisfied by Moore's proof? No. Anyone who demanded such a proof would have accepted the sceptic's restrictive conception of what knowledge is, and so would deny that Moore knew he was holding up his hand. What then was Moore's purpose in presenting his proof? It was to show there was no need for such a proof in the first place.
9Things presented in space include double-images, imaginings, dream-mages and hallucinations (Moore, Citation1959, 133, 136). These things are spatial inasmuch as they are perceived to have a certain shape and position in the visual field or in the world of thought. Things to be met with in space include human bodies, ‘the bodies of animals, plants of all sorts, stones, mountains, the sun, the moon, stars, and planets, houses and other buildings, manufactured articles of all sorts – chairs, tables pieces of paper etc.’ and shadows (Moore, Citation1959, 130). The term embraces objects of possible experience (whether actually experienced or not) and excludes things that cannot be experienced at all.
10However, it does not follow from the fact that an object is external to the mind that it is to be met with in space. Animal pains, for example, are external to the mind in Moore's sense – their existence does not imply that any human is having an experience – but they are not to be met with in space since they are not perceivable by human beings.
11Moore says that part of what is meant by calling something that is to be met with in space real, as opposed to hallucinatory or imaginary, is that it exists independently of its being experienced (1959, 144).
[W]hen I say that anything which I perceive is a soap-bubble, I am implying that it is external to my mind, I am, I think, certainly also implying that it is also external to all other minds: I am implying that it is not a thing of the sort such that things of that sort can only exist at a time when somebody is having an experience. I think, therefore, that from any proposition of the form ‘There's a soap bubble!’ there does really follow the proposition ‘There's an external object!’‘There's an object external to all our minds!’.
12Here I take issue with Crispin Wright's claim that ‘[n]othing particularly consequential emerges’ from Moore's ‘exasperating slow ruminations on what it means to describe objects as ‘external’, or ‘outside our minds’ or ‘presented in space’ or ‘to be met with in space’’ (Wright, 2002, 330).
13In particular, Moore says in response to Ambrose that she ‘wrongly … assumes that my refutation was intended to be a refutation of “Nobody knows that there are external objects,” whereas it was only intended to be a refutation of “There are no external objects”’ (Moore, Citation1952, 673–4). He makes a similar point in a letter to Malcolm (Malcolm, 1977, 174).
14Baldwin thinks sceptical challenges are beside the point because ‘the question of whether or not [Moore] is then dreaming is relevant, not to the issue of whether or not there is [a] hand before him, but to the issue of whether or not he knows that there is a hand before him’ (Baldwin, 1990, 292). He argues that at the end of the day Moore confuses the question of whether external things exist and the question of whether we know there are. However, this overlooks the fact that Moore says his proof that his hands exist is not rigorous unless he knows the premise ‘Here's one hand and here's another’. Sceptical objections challenge Moore's claim to know this premise and in so doing challenge his claim to have provided a legitimate proof of the existence of external things.
15Moore's suggestion that it is possible to know a proof is rigorous without knowing precisely what it means to be rigorous comports well with his more general view, stated in ‘A Defence of Common Sense’, that we can understand concepts for which we have no adequate analysis (1959, 37).
16Soames suggests a further criterion of rigour can be derived from Crispin Wright's notion of ‘transmission failure’ (Soames, 2003, 21). However, this is far from clear. Wright says transmission failure affects the cogency or persuasiveness of Moore's proof, not its rigour, and in fact says Moore's concept of proof as ‘valid argument from known … premises … seems unexceptionable’ (Wright, 2002, 331). Moore himself clearly distinguishes rigour and cogency. He acknowledges that showing his proof is rigorous does not remove all the objections that may be raised against it. I argue below that Moore's response to the lingering objections to his proof applies to Wright's criticism as well. See fn22 and the accompanying text.
17Incidentally, this shows that Ambrose is wrong to suggest that Moore overlooks the fact that the sceptic's doubts do not arise from a lack of the information presented in his proof.
18Stroud, for example, says: ‘that we do often prove things and come to know them in this way in everyday and scientific life seems to me undeniable’ (Stroud, 2003, 85) even though he thinks sceptical challenges undermine the philosophical significance of these proofs. Even absolute idealists accept proofs like Moore's as far as they go; they only deny that they yield absolutely true conclusions. Moore might argue that this view amounts to trying to have cake and eat it too. He might also urge that the arguments idealists give to show that ‘Here's a hand and here's another’ is not wholly true only challenge his claim to have fulfilled the conditions of his proof rather than the method of proof itself. Moore's case against idealism is beyond the scope of his paper. However, see Hylton (Citation1990, 106–66) for an excellent discussion.
19The fact that Moore devotes his time to discussing whether he has fulfilled the requirements of this sort of proof rather than defending the practice of pointing to things to establish they exist suggests that he views sceptical challenges in this way.
20Thus Moore can agree fully with McGinn when she writes:
It is impossible to use our conviction that the sceptic's conclusion is false as a legitimate ground for dismissing scepticism, for it is clearly part of our ordinary grasp of the concept of knowledge that personal conviction is never sufficient to warrant the affirmation of a knowledge claim exposed as doubtful … What the sceptic shows … is that, by the principles that ordinarily govern the entering of a knowledge claim, we cannot claim to know any external fact. It is, therefore, contrary to our ordinary ways of proceeding to insist, in the face of the sceptic's objection, that one is entirely warranted in claiming that one is not merely convinced but that one really does know that the proposition expressed by ‘Here is one hand and here is another’ is true. Thus, the trouble with [a]…‘hasty dismissal’ of the sceptic's conclusion is that it is, from the point of view of common sense itself, unacceptable: knowledge claims cannot be warranted or established as correct merely by the insistence that one really does know … common sense cannot on its own provide a philosophically satisfactory reply to the sceptic … [scepticism cannot be answered] … [b]y the bare assertion of our common-sense outlook.
21In this sense Moore is remaining faithful to Descartes's method of doubt. As he says in ‘Certainty’:
even if ‘I know that p’ can be sometimes properly used to express something from which ‘I know with absolute certainty that p’ does not follow, it is certainly also sometimes used in such a way that if I don't know with absolute certainty that p, then it follows that I don't know that p. And I have been and shall be only concerned with uses of ‘know’ of the latter kind, i.e. with such that ‘I know that p’ does entail ‘I know with absolute certainty that p’. And similarly, even if there are proper uses of the word ‘certain’, such that a thing can be ‘certain’ without being ‘absolutely certain’, there are certainly others (or at least one other) such that if a thing is not absolutely certain it cannot truly be said to be certain; and I have been and shall be concerned only with uses of ‘certain’ of this latter kind.
It is worth noting that Moore does not deny that the sense in which ‘Here's one hand and here's another’ is certain may be different from the sense in which necessary truths are certain (1959, 236–7). Also, according to his usage, to say he is certain of p implies the truth of p and is not to be understood as ‘I feel certain that p’ which is compatible with the falsehood of p (1959, 238).
22Wright, for example, claims that Moore's proof is a case of ‘transmission failure’ because Moore does not show to the sceptic's satisfaction that ‘Here's one hand and here's another’ is the sort of proposition that may be justified by appeal to experience. As he sees it, Moore's claim ‘Here's one hand and here's another’ rests on the implicit premise that ‘My experience is in all respects as of a hand held up in front of my face’ (Wright, Citation2002, 336) but, Wright contends, this premise provides justification for ‘Here's one hand and here's another’ only given ancillary information that there ‘is indeed a material world whose characteristics are mostly, at least in the large, disclosed in routine sense experience’ (Wright, 2002, 337). To avoid begging the question against the sceptic, Wright thinks ‘Moore would need independent information that experience of the kind he is having is unlikely to occur unless there is a material world’ or ‘independent information – perhaps testimony from a by-standing benevolent demon! – that there is a material world’ (Wright, 2002, 337). I take Wright to be challenging Moore's claims (to be discussed shortly) that his proof stands even in the absence of a general argument of the sort Wright demands and that ‘Here's one hand and here's another’ is known without proof. In the end, Wright thinks Moore misses the force of traditional sceptical worries that experience might be deceptive and not up to the job of showing external things exist. I return to this latter question later in the paper.
23Moore does not think it impossible to prove his premise in every case. Someone who thought Moore has prosthetic hands could prove ‘Here's one hand and here's another’ by inspecting the hands close up, touching them and pressing them. However, this is not the situation Moore takes his audience to be in (1959, 149).
24Moore does not say here whether he views his premise as unprovable in principle or as unprovable pending a satisfactory theory of the evidence on which it rests. In ‘Four Forms of Scepticism’ he suggests a non-deductive argument might be given to establish the certainty of propositions of this sort, though whether he would consider such an argument a proof is unclear. The main point here is that Moore thinks he knows his premise even though he cannot prove it.
25It is perhaps in virtue of these difficulties that Ambrose, McGinn and Stroud conclude that Moore is confused about the force of sceptical challenges and why Soames is prepared to view Moore's text as somehow ironic. Similar sceptical concerns underlie Wright's criticism of Moore's proof (Wright, 2002, 342–7).
26The other propositions are:
I am at present, as you can all see, in a room and not in the open air … I have clothes on, and am not absolutely naked; I am speaking in a fairly loud voice, and am not either singing or whispering or keeping quite silent … there are a good many people in the same room in which I am; and there are windows in that wall and a door in this one.
It is clear that Moore considers these propositions to be of the same kind as ‘Here's one hand and here's another’. The proposition that he has paper in his hand, for example, implies his hand exists and he says in ‘Proof of an External World’ that denying he is standing up is just as absurd as denying ‘Here's one hand and here's another’ (1959, 146–7). In claiming these propositions imply the existence of a world external to his mind, Moore means, as before, that they are propositions ‘from which nothing whatever about the state or condition of my own mind followed’ (1959, 242). He also notes that these propositions are contingent, meaning that their denial is not self-contradictory and each might have been false, and claims that they are propositions for which he has ‘the evidence of [his] senses’, that is, ‘when I made each, I was seeing or hearing or feeling things … such that to see or hear or feel those things was to have evidence (not necessarily conclusive evidence) for part at least of what I asserted when I asserted the proposition in question’ (1959, 243).
27He adds that:
in the case of assertions such as I made, made under the circumstances under which I made them, the charge would be absurd. On the contrary, I should have been guilty of absurdity if, under the circumstances, I had not spoken positively about these things, if I spoke of them at all … Would it not sound rather ridiculous for me now, under these circumstances, to say ‘I think I've got some clothes on’ or even to say ‘I not only think I have, I know that it is very likely indeed that I have, but I can't be quite sure’? For some persons, under some circumstances, it might not be at all absurd to express themselves thus doubtfully … But for me, now, in full possession of my senses, it would be quite ridiculous to express myself in this way, because the circumstances are such as to make it quite obvious that I don't merely think that I have, but know that I have. For me now, it would be absurd to say that I thought I wasn't naked, because saying this should imply that I didn't know that I wasn't, whereas you can all see that I'm in a position to know that I'm not.
28Soames (2002, 22) makes much of the fact that in Some Main Problems of Philosophy Moore argues that Hume's epistemology reduces to absurdity since it precludes knowledge of the external world. However, Moore acknowledges in the relevant chapter of that book that this argument will be deemed question-begging by his opponents and he even suggests that it may be possible to give a more convincing defence of our knowledge of the external world (1910, 1953, 120 and 126). However, even if his argument there is dogmatic, it must be remembered that this book comprises lectures written three decades before ‘Proof of an External World’ and ‘Certainty’. Even more to the point, in his reply to Ambrose in 1951 Moore says explicitly that he cannot refute scepticism by pointing to examples of knowledge in the way that he refuted idealism by pointing to external things:
I do not think I have ever implied that it [‘Nobody knows for certain there are any material things'] could be proved false in any such simple way; e.g. by holding up one of your hands and saying ‘I know that this hand is a material thing; therefore at least one person knows that there is at least one material thing'.
29In this connection I might mention that Wright also thinks Moore fails to recognise that even if it is granted that perceptual knowledge of things such as hands is non-inferential, it does not follow that the relevant perceptual states are indistinguishable from delusional states and can be taken to justify claims about the existence of external things. According to Wright, Moore presupposes, rather than defends, our ordinary practice of ignoring sceptical possibilities (Wright, 2002, 345–7). Unfortunately, he does not consider Moore's responses to this sort of objection and to the sceptical arguments underlying it.
30Moore says, for instance, that ‘to say “Jones last night was only dreaming that he was standing up, and yet all the time he had the evidence of his senses that he was” is to say something self-contradictory’ (1959, 247).
31In ‘A Defence of Common Sense’ Moore answers the question of whether he really knows propositions concerning the external world to be true by saying ‘I think I have nothing better to say than that it seems to me that I do know them, with certainty’ (1951, 44). However, he is not suggesting that there is no better response possible and in ‘Proof of an External World’ and ‘Certainty’ (written approximately fifteen years later) he is attempting to provide one.
32I follow Moore in taking the term ‘sensory experience’ to include veridical experience and dream experience. Only veridical experience is sensory evidence, however.
33Moore has first to reconstruct the sceptic's argument since he claims never to have seen ‘clearly stated any argument which is supposed to show [that I don't know for certain that I'm not at this moment dreaming]’ (1959, 247).
34Moore suggests this inference might be defended by showing that someone has had dream-images exactly like experiences undergone while awake. This would entail, he thinks, that it is at least logically possible to have a dream-image exactly like a waking experience and from this he thinks it might be inferred that it is logically possible for all of one's waking experiences at a given time to be exactly like the images in some dream. It is worth noting that even though he denies this argument would render the sceptic's reasoning conclusive – he says it would justify it only ‘to some extent’ (1959, 250) – and thinks, moreover, that any philosopher who relies on it to defend scepticism contradicts himself by claiming knowledge of past dreams, Moore does not reject the dream argument on these grounds and in fact grants the sceptic his unargued premise.
35The draft is published in Moore (1993, 195–6). While this material was deleted from the final version of the paper, I think it can be taken to reveal something of Moore's inclinations, however dissatisfied he was with his articulation of them.
36As Moore puts it, ‘even if it is a fact that no contingent proposition is ever known to be true, yet in no case does this follow from the mere fact that it is contingent’ (1959, 231).
37Moore does not say what the precise weaknesses are and I shall not speculate. It is worth noting, however, that in an editor's note, Casimir Lewy writes: ‘Moore was particularly dissatisfied with the last four paragraphs … and I believe that he was thinking primarily of these … when he wrote … that the paper contains bad mistakes’ (1959, 251). The four paragraphs in question deal with Moore's doubts about the logical possibility of his having his immediate memories and present sensory experiences while dreaming.
38I confine myself to the sections of Moore's paper dealing with knowledge of the external world. While Russell's arguments are not specifically directed at Moore, I shall frame them this way for ease of exposition. Moreover, I shall treat Russell's challenges to claims based on memories of the recent past and claims about the external world based on present sensory experiences together, even though Moore separates them, since his reply is the same in both cases. Finally, I shall consider Russell's dream argument and his objection that Moore may be the victim of a deception engineered by a scientist as one argument since Moore gives essentially the same reply to both objections.
39Moore does not deny that memory is fallible. He writes: ‘that we do sometimes, not only in dreams, but also in waking life, feel as if we remembered things which in fact never happened, I fully grant. That this is true I don't feel at all inclined to question’ (1959, 219). But what he is ‘inclined to question is that this fact is in any way incompatible with the proposition that I do now know for certain that I heard a sound like “Russell” a little while ago’ (1959, 219).
40Moore thinks Russell's objection assumes as a general principle that: ‘we can only know with certainty that so-and-so is true, if the “so-and-so” in question belongs to some class of which it is true that no member of that class ever “leads us into error”’ (1959, 216). Evidently this is Moore's reformulation of Descartes's claim that ‘it is wiser not to trust entirely to any thing by which we have once been deceived’ (1641] 1931, 145).
41Even here Moore is tentative about his conclusion. He says
[t]here may perhaps be something more than this simple fallacy in the arguments that because experiences of this sort are sometimes not preceded by the sound of which they feel as if they were a memory, and this is an experience of this sort, therefore this experience may not have been preceded by that sound, i.e. that I do not know for certain that it was so preceded. But I cannot see that there is anything more in it.
42Moore adds, however, that ‘it seems to me terribly difficult to say exactly what the sense is’ (1959, 226).
43The objections by Wittgenstein (1969) and Malcolm (1977) that Moore's claim to know ‘Here's one hand and here's another’ on the basis of sensory evidence is nonsensical are particularly pressing but they merit separate treatment both because of their complexity and because they are not addressed in any detail by Moore in his writings.