Epistemological and Ontological Idealism

Arguments for idealism generally come in to forms. There are epistemological arguments that attempt to show that the world as it is known to us is essentially mind dependent. Kantian arguments are the most famous example of this sort of a approach. So too are some of Berkeley's arguments for idealism, such as the arguments presented with such force in the _Dialogues_ showing that the only empirically defensible concept of a material thing is one that identifies matter with a system of ideas. Ontological arguments for idealism, on the other hand, attempt to show that there is something about the nature of reality that requires that it be of a spiritual or mind dependent character. Thus McTaggart argues that the world must consist of substances and that the only candidate for substance that we have available is that which we find when we reflect on our own minds. Berkeley too presented an ontological argument for idealism when he argued that the very conception of a mind independent entity was incoherent. The famous master argument was not intended as a claim about merely knowable reality, it was intended, it seems to be a claim about reality as such whether known by us our not. This last claim of Berkeley's has always struck me as tendentious because it appears to attempt to prove a negative: How can we know that the only thing that is real is minds, unless we presuppose that our knowledge exhausts all that is real? And how can any non-omniscient being make that claim?
An answer to this last question can be made by those who endorse a non-realist theory of truth. To take a cheap and easy example, if verificationism is true, then the claim that there is a reality utterly independent of our mode of cognition is not false but meaningless. On these sorts of views the sentence, "There is a reality that is impossible for us to experience" has no cognitive content at all. And this brings out another feature of the distinction between ontological and epistemological idealism. Ontological idealism is perfectly compatible with a realist theory of truth. Thus Berkeley is perfectly happy claiming that, at least when we are "thinking with the learn'd," the literal truth of statements depends on the existence of some reality "out there" that corresponds to it. The difference between Berkeley and metaphysical realists is not with respect to what makes a statement true, but rather with respect to the sorts of things are truthmakers for propositions, to wit, minds and ideas.
Personally I have never really understood the various sorts of non-realist theories of truth. Coherentist and pragmatic theories of truth seem to me to be either obviously false, or really just disquised theories of justification or knowlege. Yet it does seem to me that any variety of idealism is going to be in one limited sense epistemic. Whatever claims about reality we may make, we can only really justify them with respect to the reality that we have some conception of. Kantian things in themselves can never be disproven. The best one can hope to do with respect to such hypothetical entities is to show that such an hypothesis is superfluous, not that necessarily, such things do not exist.
So my own argument for idealism, grounded as it is in the claim that existence is an essential property of consciousness, is an ontological idealism. My claim is that the epistemic claim of the cogito requires an ontological ground. I can only know I exist if there is a necessary connection between existence itself and consciousness or the self. Otherwise I may know I am thinking, and yet not recognize that this thinking must be existent thinking. So the essential premise, the pivotal claim is (1) Consciousness essentially exists. But this claim only gets us to idealism if we can also claim (2) Nothing else exhibits this characteristic of existence. And it is here that the epistemological caveat appears--I can only speak of what I can think of. I can confirm as a phenomenological datum that no perceived object exhibits the characteristic of consciousness. I can even go further and in Berkleian manner argue that no conceivable material thing has the property of existence. Its not just that in fact I am not perceiving material things. but I cannot even in imagination combine the property of existence (which I am aware of in the cogito) with the material things. This shows that at least of those concrete particulars I am aware of (minds and material objects) it is only the mind that exists. Not only do I affirm with Meinong that there are non-existent objects. I affirm the stronger claim that all material things are non-existent objects. But the epistemological limitation remains. The objects that I claim cannot exist are perceivable objects. I cannot by this method show that there is nothing else that also has this property. In fact I may have to admit that there are some things, universals for example, which actually do exhibit existence even though they are not of the character of consciousness or the mind. The idealist claim,though ontological, is thus limited. It is not reality as such, but only reality insofar as I can conceive of it (that is, the concrete particular reality, ignoring universals for the moment), which necessarily does not exist. This may be an "apparently outrageous view" as one reviewer called it, but it is also limited to reality as conceived. Kantian things in themselves may exist even if they are not conscious, but if there are any such "things" they also must not have the character of the material world that is actually conceived and thought of.