In this post I should like to briefly present Alexius Meinong's critique of idealism. Since this critique is not well known among philosophers, I think that it will be most fruitful to present it in a succinct way. But in addition, I should also like to present some counter-arguments to Meinong's critique, with a view ultimately to steelman his critique and to place it on sturdy ground.
I should note at the outset that the idealism with which I will be concerned in this post is metaphysical idealism, which is the view that reality is at bottom mental. This thought is usually cashed out with the expression that reality is essentially composed of ideas. Let's get right to it and present the argument in deductive form, before unpacking it in greater detail:
P1: If idealism is true, then everything is an idea.
P2: All ideas are existing objects.
C1: Therefore, if idealism is true, everything is an existent object. (P1,P2)
P3: But some objects do not exist.
C2: Therefore, idealism is not true. (C1,P3)
P1 is just a description of idealism, so it needn't detain us. P2 should be fairly uncontroversial, at least to the philosophical mainstream. It has been standard doctrine throughout the history of the subject that ideas and minds are existing objects. So with both of these premises on board, C1 follows by modus ponens.
P3 is where the trouble lies. For it is a corollary of C1 that such objects as unicorns exist. Now, one might think that this can't be correct, because the idea of a unicorn is surely different than a unicorn itself. But this distinction is not available to the idealist, since he considers all of reality to be composed of ideas. So for the idealist, the idea of a unicorn just is a unicorn. And since the idea of a unicorn is an existent object, a unicorn must also be an existent object.
But surely unicorns don't exist? Maybe some idealists would be willing to bite the bullet on this and say that unicorns actually do exist. But there is a further trouble in store. For we also have an idea of the non self-identical round square, which is an existing object. But like we said earlier, the idea of the non self-identical round square just is the non self-identical round square. So, under idealism, the non self-identical round square exists. But it is a truth of reason that such an object cannot possibly exist. Therefore, C2 follows by modus tollens.
Now, of course, there is a way for the idealist to counter this argument: all he need affirm is that all ideas are nonexistent objects. Then he would be in no danger of affirming the existence of such preposterous objects as the non self-identical round square. It would now appear that the idealist is on safe ground.
But this is only an appearance; for the above argument need only be slightly tweaked to deal with this new variant of idealism. To wit:
P1: If idealism is true, then everything is an idea.
P2: All ideas are nonexistent objects.
C1: Therefore, if idealism is true, everything is an nonexistent object. (P1,P2)
P3: But some objects do exist.
C2: Therefore, idealism is not true. (C1,P3)
I don't think we need to spend too much time exploring how this new argument is supposed to work. For under the new variant of idealism, such commonplace objects as trees, dogs, and chairs would be considered as nonexistent objects. This might not seem objectionable to one with a proclivity towards mereological nihilism, but even the mereological nihilist affirms the existence of fundamental particles. But the idealist of this variety must deny the existence of even these. So it would seem that this new variant of idealism also runs into serious difficulties as well.
But what if the idealist wanted to allay the criticism by distinguishing between existent and nonexistent ideas? This sounds good at a first glance, but this idea also runs into many problems. For one thing: how are we supposed to make this distinction in a reliable way? One possibility is to define those ideas that are actually thought of as the existent objects, and all those ideas that are not thought of as the nonexistent objects. But this runs into two difficulties. Firstly, all the problematic objects mentioned earlier (viz. unicorns, the non self-identical round square, etc.) have actually been thought of by many minds. So these ideas must be counted as existent objects. But quite apart from that, I'm not sure that the notion of an 'idea that no mind has thought of' is even coherent. For surely an idea just is something thought of by a mind; so what sense is there is supposing that some ideas are not thought of by any mind?
Perhaps we might want to effect this partition by holding that all perceptual ideas are existent objects, while all conceptual ideas are nonexistent objects. (Briefly, an idea is said to be perceptual if it is based in some way on the senses; while an idea is said to be conceptual if it is merely based upon the relations of concepts, with no attendant sense-data). But this doesn't work either; for many clearly nonexistent objects stand in perceptual relations. For instance, many people have dreams or hallucinations of such fantastic beasts as unicorns, and I would not be surprised at all if some people have had perceptual experiences of such absurd objects as the non self-identical round square.
So to conclude, it would appear that idealism falls into the same sorts of problems as many of the mainstream metaphysical theories on offer (including empiricism, materialism, phenomenalism, reism, process philosophy, etc.) And that is the trap of reductionism. In other words, all of these philosophies try to reduce the panoply of items in reality down to a single type of item, call it 'X'. But this invariably leads to trouble because X will inevitably have various properties that cannot possibly apply across the board. It is my studied opinion (and I wholeheartedly agree with Meinong on this) that any adequate metaphysical theory will not be reductionist, and will have to account for all the various types of objects in reality.
