What's the Problem With Meinongianism?
Meinongians say that some things do not exist. In other words, existence is a property that befalls only some of the things there are. It follows that by 'existence' these Meinongians do not mean the trivial property that every thing whatever has. What else do they mean? Maybe they mean by 'existence' being in space or time, as Meinong sometimes does. Or maybe they mean an alleged primitive property of certain things. At any rate, I have no objection to this except that I'd rather not use the word 'existence' for that. But I can't really say that ordinary usage is on my side, given that a) ordinary quantification is almost always restricted (though not always in the same way), and b) there is hardly an ordinary usage of 'existence' at all. So far, Meinongianism is utterly trivial. It merely holds that some objects lack a certain property.
Next, Meinongians say that whenever somebody is in a mental state we would ordinarily describe as a state of fearing some F-thing or searching some F-thing or thinking about some F-thing, then there is an F-thing the person fears or searches or thinks about. This, rather than the completely uninteresting question of how to use the words 'being' and 'existence', I take to be the main point of Meinongianism. Meinongianism is a theory of intentionality.
I find a modest version of the Meinongian theory rather attractice (as I explained in this entry): Whenever somebody is in a state we would ordinarily describe as a state of fearing (etc.) some F-thing, and it is possible that there is an F-thing, then there is an F-thing the person fears (etc.). However, it is characteristic of Meinongians that they do not impose that restriction to possible things. For example, full-blooded Meinongians accept that there is a round square people talk about in discussions of Meinongianism, even if round squares are impossible because 'round' implies 'not square'. Since it is unclear whether 'round' does imply that, let 'round*' mean 'not square' and take the Meinongian to claim that there is a round* square.
What does the Meinongian mean by this claim? Does he really mean that there is something, let's call it RS, which is both round*, that is, not square, and square? No. Interpreting someone as consciously asserting an outright contradiction should always be a means of last resort. This even holds for people (like Richard Sylvan) who insist on asserting outright contradictions: Charity demands that they either misunderstand their own assertions or the concept of a contradiction.
Fortunately, there are alternative interpretations of what the Meinongian means. The trick is to reinterpret his predications: 'RS is not square' is compatible with 'RS is square' if, for instance, it only means that according to a certain story, RS is not square; or if it means that RS has the property of being not square if existing rather than just the property of being not square (Terence Parsons); or if it means that RS somehow encodes the property of being not square, e.g. by containing it as a part, rather than just having it (Ed Zalta).
So the Meinongian can save the letter of the Meinongian theory of intentionality only by interpreting that letter in a deviant way. When I'm thinking about the round* square, there isn't really a round* square I'm thinking about. Instead there is some ersatz thing that in one way or another represents itself as being round* and square (by 'encoding' these properties, or by being round*-if-existing and square-if-existing -- it follows that whatever 'existing' means, ersatz objects do not exist). Meinongianism is ersatzism about impossibilia. More precisely, it is ersatzism about non-actual objects of intentionality.
I don't see any good reason to doubt that some such ersatz construal of intentional objects is feasible. In particular, many of Lewis' arguments against ersatzism about possibilia (in general) do not apply here because arguably the realm of the intentional comprises fewer things than the realm of the possible.
Then Meinongianism is after all not a very eccentric ontological position. The real question is not whether it is ontologically acceptable but whether it provides a good theory of intentionality.
This is perhaps the clearest exposition I have found about why Meinongianism is not as problematic as some contemporary philosophers claim it to be. Great post!