blog.kennypearce.net

On Moore's Alleged Refutation of Idealism

I've just finished reading G.E. Moore's paper, "The Refutation of Idealism." The paper was originally published in Mind in October 1903 and reprinted in Moore's Philosophical Studies in 1922, but I've got the version reprinted again in Colin Turbayne's edition of Berkeley's Principles with critical essays, and I'll be citing page numbers from there.

Moore's target in the essay is not Berkeley directly (it is, of course, Berkeley in whom I am most interested here), but what he calls "modern Idealism." The modern Idealism described by Moore seems to be a sort of immaterialist panpsychism; that is, Moore claims that these Idealists hold the "Reality is spiritual" and, in particular, "chairs and tables and mountains ... are in some sense neither lifeless nor unconscious ... When the whole universe is declared to be spiritual, it is meant not only that it is in some sense conscious, but that it has what we recognise in ourselves as the higher forms of concsiousness. That it is intelligent; that it is purposeful; that it is not mechanical; all these different things are commonly asserted of it" (pp. 57-58, emphasis original). Now this is very different from Berkeley's view, and even Schopenhauer does not ascribe to the universe consciousness, or at least not intelligence and "higher" consciousness (he does say that the universe wills which may require some degree of consciousness). But Moore says that "modern Idealists" assert all of this. The only philosopher he specifically identifies in this connection is A.E. Taylor (p. 64). He also mentions Hegel (p. 71) and Bradley (p.77) but is ambiguous as to whether these people are supposed to count as modern Idealists. (I have some minimal familiarity with Taylor as a Plato scholar, but no idea other than what Moore says as to his own philosophical views; Hegel is on my "to read" shelf, and of Bradley I am utterly ignorant, though I think I have heard of him before.) Moore explicitly contrasts these modern Idealists with Berkeley and Mill on p. 74 (though he is actually saying that something that will be easily admitted about Berkeley and Mill will take some convincing to admit about modern Idealists, so the similarities are here more important than the differences).

There is, nevertheless, good reason for Moore's inclusion in Turbayne's collection. Moore believes that the modern Idealists are bound together more by a type of argument than by a type of position. The Idealistic arguments are of the form "Since A is B, and B is C, and C is D, it follows A is D," where the first step, "A is B," is "esse is percipi" and the last step, "A is D," is "Reality (i.e. being, esse) is spiritual" (p. 60). Interestingly, one of the intermediate steps is supposed to be "esse is percipere" (p. 62). That is, by a chain of inferences of this sort, Idealistic arguments begin from "to be is to be perceived (or thought, or experienced)," through "to be is to perceive (or think, or experience)" to "Reality is spiritual." Moore will attempt to refute this line of reasoning by attacking the claim that esse is percipi, which, of course, originates from Berkeley (Principles 3), and therefore justifies the essay's conclusion in this volume.

Moore begins by examining the meaning of esse is percipi. He argues that what is meant here is that esse includes percipi. That is, the predicate 'exists' is to be defined as 'is perceived' plus some additional x. Being perceived is a necessary but insufficient condition for existence.

So far so good. Moore seems to have in mind the problems of dreams, etc., in denying that Idealists mean esse to be simply identical with percipi. Instead, he thinks that there are some additional required characteristics which he does not specify. (On what these characteristics might be, see my "The Ontological Status of Dreams in Berkeleian Metaphysics," forthcoming in The Dualist 13 (2006).)

Moore's next task is to argue that this is not so. His discussion is rather long and if you want the whole thing you should be reading his article and not my blog, but the crux of the argument is quite interesting: Moore argues that Idealists cannot consistently hold that blue is simply identical with the sensation of blue, since they recognize it as the content of the sensation and a sensation involves consciousness plus a certain content, and the two can be separated (at least we can consider them separately). Because all of our sensations are modes of consciousness, they must have certain commonalities, certain properties of consciousness qua consciousness. Likewise, there must be some difference between conscious of blue and conciousness of yellow. Blue is the object of our sensation, rather than being strictly identical with the sensation. What we have in this sensation, what we can certainly know, is that there exists the awareness of blue. The Idealist denies that this entails the existence of blue, but here, Moore thinks, he is inconsistent: awareness of blue is not, at core, any different than awareness of my own mental states. If awareness of blue does not justify the conclusion that blue exists, then awareness of mental states does not justify the conclusion that mental states exist.

Now, Moore's target is a very broad one. He hopes to refute all modern Idealists, despite their widely divergent views. Furthermore, these modern Idealists don't hold positions as close to Berkeley as the position that I would like to defend. For this reason, we can't fault Moore if he hasn't really refuted the Berkeleian position. However, it is indeed true that he hasn't.

I believe that Berkeley holds, as Moore does, that the object of a mental experience is distinct from the experience itself. In Berkeley's taxonomy of ideas, there are perceptions, thoughts, and volitions. The very same content (e.g. blue) can be the content of a perception, a thought, or, in some cases, a volition. That is, I can perceive something blue, I can think of something blue, or I can will something to be blue. In all these cases, the content may be the same. Thus the content is, indeed, something distinct from the particular experience of it.

However, to say that this entails that blue is mind-independent (which Moore doesn't say, so it's not clear what he thinks he's proving) fundamentally misunderstands what sort of thing blue is. Blue is not a range of wavelengths of light. That might be part of a physical explanation of blue, but it isn't blue. Blue isn't a Lockean simple idea either, because Berkeley doesn't believe in those (simple ideas are formed by abstraction, so Berkeley doesn't think they exist). What blue is is, quite simply, a characteristic certain ideas may have. When we speak of blue, we are referring to something we can see or imagine. Anything not seen or imagined is not blue. One might suppose (probably correctly) that wavelengths of light detected by instruments but not actually viewed by humans can show that possibility of blue under certain circumstances; that is, if a person looked where the instrument was pointed, he would see blue. This does not amount to mind-independent blue, because the wavelength of light, or the readout of the oscilloscope or spectrometer or whatever is simply not blue. Blue is a color, and colors are conceptually inseparable from the sense of vision.

Blue is thus a type of mental experience, and our awareness of blue is indeed treated in parity with our awareness of other mental experiences. The illicit inference Moore thinks he sees is not there.

What follows for Berkeley is, of course, that if there really were a physical world as described by realists, it would be unknowable, irrelevant, and ultimately incomprehensible to human minds. Why? Because it is part of the description that the physical world is external to the mind. It follows that it is not made up of ideas. Ideas are (by definition) the direct objects of the mind. It follows from this that our mind cannot contain the physical world, so we can't think of it or experience it. (I have just sketched in a few lines a whole argument which Berkeley works out at great length in several places - if it seems to be lacking rigor, blame my brevity.) This conclusion Moore has not refuted.

The next paper in the collection is W.T. Stace's "The Refutation of Realism" (first published in Mind in April 1934) which I hope to read before the end of the break if I have a chance.