Existence refers to the possibility of validly dereferencing concepts.
For a reference to exist means that the referent occupies some non-zero amount of space. If this is the case, we will also say that the concept can be validly dereferenced : it refers to something, as opposed to nothing.
Existence is a property only of references: existence means that a reference can be validly dereferenced. As non-referential things cannot be dereferenced, it is meaningless to ask if non-referential things exist. As an example, consider the following question: “Does the apple exist?” . This is a reasonable question if the term “apple” is being used as a reference (or a linguistic variable); in that case, the answer is true if it is possible to find the apple in the world. This is not a reasonable question if it is asking whether an apple-in-the-world exists, since under that reading of the sentence, the phrase apple-in-the-world presupposes existence.[35] In either case, it is interesting to note the facility with which we dereference references, especially since this process happens somewhat unconsciously.
There are at least two kinds of existence and nonexistence: necessary and contingent. For example, as necessity and contingency apply to nonexistence, there are things that do not exist because they are not valid conceptual forms (such as colorless green things) and things that do not exist because they happen not to exist (such as the people on mars). The former concepts do not exist as a matter of necessity: they cannot exist in virtue of their conceptual (or linguistic) construction. A popular example of things that cannot exist are self-referential concepts, which have an impossible deep structure. The latter kind of concepts, or contingently non-existent concepts, simply happen not to exist: there is no reason in principle why they could not.
Both necessary and contingent existence, however, assume a potentially perfect correspondence between references and the things to which they refer. The reality of the situation is often less black and white. A more subtle kind of existence (or lack thereof) occurs when the reference does not correspond exactly to its referent. For example, objects might exist in a different way than their associated concepts exist. Similarly, the question of existence is often not a determination of if concepts and objects correspond, but how concepts and objects correspond. For example, how does our percept ‘apple’ correspond to the object apple?
One way in which objects and concepts differ is that objects do not exist as physically singular entities in the way that our conceptions of them do. So to the extent that a particular conceptual partition into atomic entities prevents other partitions or points of view, that point of view cannot reflect the totality of reality. For example, if we see people as individuals, it may block our view of them as collections of cells, or as parts of a larger social organism. The world, since it is capable of being conceptually partitioned in any number of ways, cannot be entirely understood on the basis of only one of these partitions.
[35] To even speak of things like apple-in-the-world is problematic, because in virtue of their being spoken about they become referential expressions: that is the nature of language. The ontological argument for the existence of God is a wonderful example of this issue. However, there may be an unspoken immediacy of experience about which one cannot ask, and of which it is incorrect to conceive of as either existing or not existing (as it is not referential): of course, we could not talk about it if there were.