In his paper ‘Proof of an External World’ Moore is concerned, as the title makes it abundantly clear, with providing a proof for the existence of the external world. He begins his essay with a quotation from Kant to the effect that Kant maintained that it was scandalous that the existence of such a world –‘things outside of us’- has had to be accepted merely on faith and not on some satisfactory proof. Kant thought it important to provide such a proof for giving a proof for the things outside of us was a task which lay within the domain of philosophy. Moore readily agreed, although he found Kant’s proof less than satisfactory or clear, as shall be evident in a moment. Moore set out to give one which is, or so he believed, able to remove the scandal.
Prior to going into the exposition of his ‘proof’ of an external world, Moore spent considerable effort to clarifying some conceptual confusions inherited from Kantian philosophy. According to Moore, Kant’s use of the proposition ‘things outside of us’ is ambiguous. It could be associated with an array of other meanings which may be irrelevant, confusing or unable to convey the meaning as intended by Moore. Perhaps, thinks Moore, the proposition would have been clearer if expressed as ‘things external to our minds’. But even this will not remove all ambiguities. Moore will then set out to remove the ambiguities of these expressions not by distancing himself from Kant but by returning to Kant. Kant himself had noted the ambiguity of the expression ‘outside of us’: it could mean one of the two following things: 1) something which exists as a thing in itself, and 2) something which exists as a mere outer appearance. By the first Kant meant ‘external in a transcendental sense’ and by the second, ‘empirically external objects’. In order to further remove any possible ambiguity Kant will distinguish the empirically external objects from transcendentally external objects, by designating the former as ‘things which are to be met with in space’ (p. 129).
‘Things to be met with in space’ is a designation which initially satisfies Moore and if he proves the existence of some such things as to be met with in space, then he has ipso facto proved the existence of an external world. Physical or material objects, such as chairs, stones, animals, bodies, etc, are examples of what the expression refers to. It can however be used to include things which properly speaking are not material objects. Moore’s example is ‘shadows’. Thus although shadows are not physical objects nor even a ‘thing’ in the material sense, Moore wants to retain and use the expression in this wide sense as to include shadows, since they clearly are things to be met with in space.
Wide as the meaning of this phrase may be is not as wide as the meaning of another expression used by Kant as equivalent/identical to the former expression. This other expression, to which Moore now directs his attention is ‘things presented in space’, or in Kant’s own words, an ‘empirical object is called external, if it is presented (vorgestellt) in space’ (p. 130). Moore is dissatisfied with this Kantian definition: for there are such things that are presented in space yet are not to be met with in space. Moore’s example comes from psychology: negative after-images. ‘If after looking steadfastly at a white patch on a black ground’ Moore is quoting from a psychological textbook, ‘the eye be turned to a white ground, a grey patch is seen for some little time’ (p. 130). After repeating the experiment several times himself, Moore concluded that these negative after-images although not to be met with in space, ‘can be quite properly said to have been ‘presented in space’’. Based on this observation Moore went on to argue that Kant was clearly mistaken to treat these expressions as equivalent or identical, for they were not as Moore’s examples show.
The difference between these two expressions can be rendered in this way: ‘things to be met with in space’ are perception conditioned activity, i.e. anyone provided certain conditions can perceive the object in question (the white patch on a black ground in this context) but only Moore could see the negative after-image. What under certain given conditions (normal conditions) can be perceived by others as well as the man in question are things to be met with in space, whereas negative after-images are ‘things’ which although can be presented in space, are nowhere to be met with in space.
Another way to put this distinction would be to say that things presented in space are not external to our minds: they are not logically independent of the perceiver. Moore brings two other concepts to clarify what he means by things not external to our minds: ‘double images’ of some object and ‘bodily pains’. These are examples of things which are presented in space but are not to be met with in space. Although it can be reasonably said that another person feels a pain like the one Moore is feeling, it is absurd, according to Moore, to suppose that ‘he could feel numerically the same pain’ (p. 132). For that would imply that pain is not internal but external to the mind, and therefore an object to be met with in space.
Moore then goes on to discuss a special class of things, ‘the electric light’, which does not properly belong neither to the ‘external things’ or things ‘to be met with in space’ nor things ‘presented in space’ in the sense Moore is using the expression (bodily pains, negative after-images, etc). Yet some philosophers are inclined to the view that this class of things is properly said to belong to things ‘presented in space’. Moore disagrees. After looking at an electric light for some time and then closing the eyes it happens that one sees a bright patch similar to the one seen with eyes open. The difference between these two kinds of ‘visions’ is that, in the language of psychology, the first are called ‘negative after-images’, images seen with eyes open, whereas the second are simply called ‘after-images’, images seen with eyes shut. Moore rounds the difference by saying – and simultaneously distancing himself from ‘other philosophers’ who may simply deny that these after-images are presented in space-, that although not easily noticeable there is nonetheless an important difference, which in Moore’s terminology is the difference between things being presented in a space, which is the case with after-images seen with eyes shut, and things being presented in space (without the article a) in the sense of things seen as part of physical space.
After making these distinctions Moore feels justified in drawing the conclusion that things which are presented in space (negative after-images) are not equivalent to things which are met with in space (physical objects). From this it also follows that the conception ‘presented in space’ is wider than the conception ‘to be met with in space’. Therefore it is not the case, as Kant had previously thought, that these two classes of things could collapse into each other.
However, argues Moore, things are more complicated than this. Although in one respect the conception ‘presented in space’ is wider than ‘to be met with in space’, in another respect, the reverse is the case. ‘For there are many “things” to be met with in space, of which it is not true that they are presented in space’ (p. 134). Considering the fact that Moore equates the conception ‘to be met with in space’ with an object, a thing that might be (visually) perceived, it does not follow that it is perceived. Moreover, such a thing (whether in the past, present or at all times) may simple not be perceived by anyone and thereby is not ‘presented in space’. Things to be met with in space do not immediately entail that they are presented in space. Prior to being presented in space they must be perceived in space. This is the case of things of possible experience.
This probably also explains the difference between the expressions ‘to be met with in space’ and ‘external to our mind’. Although Kant’s words suggest that he took them synonymously, they are not, according to Moore, synonyms, but two different expressions. The difference seems to be that the first expression is narrower than the second, for if Moore equates the first with objects being perceived then it clearly follows that there are things which are external to our mind and yet are not perceived by anyone. The example of animal pain prove Moore’s point: although animals experience pain which is not external to their minds, it certainly is external to our minds. Yet it cannot be met with in space.
There seems to be cases of ‘things’ which is difficult to classify as belonging to one or the other category. The examples of ‘mirror reflection’ or ‘the sky’ are not quite the same things as the examples hitherto given by Moore. Moore is clearly perplexed by these unclassifiable things but he goes on to dismiss them as irrelevant for the purpose of his discussion. One would have perhaps expected that the case of these dismissed examples have made Moore rethink his notions which he clearly intends to use as elements of his proof for the existence of the external world, but that is not the case.