- The 'naïve view' of existence, according to which 'exists' is a genuine predicate, expressing a genuine property, is defended against the orthodox Russellian view, which maintains that to predicate existence of an object is really to say of some property that it is instantiated.
- It is argued that the orthodox view faces intractable problems, which the naïve view does not face, and that the latter fares better than the former in three specific contexts in which the notion of existence plays a central role: the cogito, essentialism, and the ontological argument.
- Existence, likewise, has the look of a highly general property of objects - certainly of all existing objects. It is a unitary property that holds of objects of many different categories - physical, mental, abstract; it does not vary its nature according to the metaphysical type of the object that has it.
- Yet, of course, such a view has been widely rejected, in favor of the thesis that attributions of existence to objects are really disguised statements about properties or predicates, to the effect that they have instances (the "Russell- Frege view").
- I argue against this view, first noting that the notion of having instances must be interpreted to mean having existent instances, thus re-introducing the concept of existence. For a property or predicate to be instantiated in such a way that the appropriate object can be said to exist we have to assume that the instantiation relation can only relate existent objects; we cannot allow, for example, that Sherlock Holmes instantiates the property of being an opium-smoking detective. I also suggest that the Russell-Frege view cannot handle attributions of existence to properties them- selves, as well as encountering problems with singular statements of existence.
- On the positive side, I defend the position that non-existent objects can be "quantified over" and said to have properties. Thus I interpret quantifier words like "some", "all" and "most" as not carrying, semantically, existential import - they merely function to specify how many. This leads me to revise the usual interpretation of the "existential quantifier": I re-name it the "partial quantifier" and assign the affirmation of existence to a predicate, as in "for some x is F and x exists".
- I also defend the metaphysical thesis that existent objects are not intrinsically mind-dependent but that non-existent objects are - that is, there is no non-existent object that has not been an object of thought. This leads me into the thorny thickets of possible and impossible objects, about which I adopt the highly alarming position that they exist. Here I am under unwelcome pressure from a number of directions and I try to resolve these pressures as smoothly as possible; these matters strike me as still very unsettled.
- What mainly moves me is the thought that there cannot be any definite individuated thing that fails to exist unless thought has conferred identity on that thing - for what else could confer identity on a non-existent thing? Sherlock Holmes had no identity before he was invented by Conan Doyle, so it makes no sense to suppose that there was a definite fact of his non-existence before he was created. Non- existence requires individuation3, and individuation4 must come from the mind if it cannot come from the world. But these matters are, I concede, very obscure.
Photocopy filed in "Various - Papers on Identity Boxes: Vol 10 (M1: Ma-Mc)".
Footnote 1: Taken from Oxford Scholarship Online.
Footnote 2: Taken from "McGinn (Colin) - Précis of 'Logical Properties: Identity, Existence, Prediction, Necessity, Truth'".
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2018
- Mauve: Text by correspondent(s) or other author(s); © the author(s)