Candidate Existents

The general existence question has no useful answer, so ontology proceeds case by case. This article walks the standard cases and asks of each what it would take for it to be on the list. No case is settled here. The purpose is to see what the arguments are arguments about, and why the same considerations keep recurring in different clothes.

Particulars

Start with the least contested category and give it its name.

The contrast is with whatever is repeatable, and the article after this one takes that up directly. What matters now is the test. Ask whether the thing in question could be wholly present in two separated places at the same time. The mug could not. Whether anything could is exactly the question that divides the two great camps in the theory of properties.

Ordinary middle-sized particulars are the paradigm of existence, and every ontology has to accommodate them somehow. Even so, they are not entirely uncontroversial. If the mug exists in addition to the particles arranged mug-wise, then the world contains one more thing than the particles, and that extra thing has to earn its keep. Some philosophers deny that it does, holding that strictly speaking there are only the particles, and that "there is a mug here" is a convenient way of reporting how they are arranged. The intermediate tier gives that view a full article under the heading of composition. Note only that even the safest case is not free.

Events

Sentences about things are not the only sentences that look like they demand something.

The pressure to admit events comes from a straightforward source. "The collision caused the fire" appears to relate two things, and if it does, those two things have to be somewhere on the list. "Priya crossed the road slowly" appears to describe something, the crossing, as having been slow. Donald Davidson argued that a great deal of ordinary talk works this way, and that treating causal statements as relations between events is what makes the logic of such sentences come out right.

Grant that there are events and a further question opens immediately: when are two descriptions descriptions of one event? Davidson proposed that events are identical when they have the same causes and the same effects. Jaegwon Kim proposed instead that an event is the exemplifying of a property by an object at a time, which makes events finer-grained: Caesar's dying and Caesar's dying violently come out as two events rather than one, because the properties involved differ.

The disagreement is not a quibble. On the coarse view there is one thing that is both a stabbing and a killing. On the fine view there are two, occupying the same place at the same time. Neither result is comfortable, and which discomfort is worse is what the argument is over.

Properties

The case for properties runs parallel and starts from the same kind of observation. Two mugs are both chipped, and that seems to be a fact about the world rather than about the word "chipped". A theory that recognises only the mugs has to say what the shared fact consists in.

Grant properties and the same follow-up arrives: when is the property this mug instantiates the same one that mug instantiates? The answer to that determines whether properties are repeatable entities, present wholly in each of their instances, or whether each thing has its own private property and the sharing is a matter of resemblance. That fork is the subject of the next article, and this one leaves it open.

What is worth noticing now is the shape of the dispute. Nobody denies that the two mugs are both chipped. What is denied, by those who deny anything, is that this fact requires an extra entity in addition to the mugs. The disagreement is about what the fact requires, not about the fact.

Numbers

The hardest of the standard cases.

The case in favour is that mathematics is our best-confirmed body of knowledge and it is full of existence claims. "There are infinitely many primes" is a theorem. It is true. Read at face value it says that there exist infinitely many things of a certain kind, and it is hard to see by what right one accepts the theorem and rejects what it appears to say. Sharpening that argument, and connecting it to the fact that physics could not be stated without mathematics, produces the indispensability argument, which the intermediate tier treats in full.

The case against is not that the theorem is false. It is an epistemological worry, and Paul Benacerraf stated the version everyone now works from. If numbers are the kind of thing they are usually taken to be, standing outside space and time and incapable of causing anything, then nothing about them can affect us. But our best account of knowledge requires some connection between the knower and what is known. So the account that makes mathematical statements come out true makes mathematical knowledge inexplicable, and the account that explains the knowledge struggles to make the statements come out true.

That is a dilemma rather than a refutation, and every position in the philosophy of mathematics is in part a response to it. Some accept the entities and look for an account of knowledge that does not require causal contact. Others reject the entities and rewrite the mathematics so that the apparent commitment disappears. The second move is the paraphrase strategy from the previous article, applied at full stretch.

The rest of the list

Several further cases are worth having in view, each contested for its own reason.

Holes are the awkward one, because a hole seems to be somewhere and to have a shape and a size, while consisting of nothing at all. Boundaries and surfaces raise the same difficulty in a different form. Corporations and universities appear to do things and to be answerable for them, without being identical to any particular group of people or any building. Fictional characters can be counted, compared, and got wrong, which is odd behaviour for nothing.

Each of these can be handled by admitting the entity or by rewriting the sentences. Holes might be accepted as entities, or "there is a hole in the bucket" might be rewritten as a claim about how the bucket is shaped. The second option is not obviously available for every sentence about holes, and testing exactly where it fails is the ordinary work of the subject.

Where this goes

The cases share a structure. Something in ordinary or scientific talk appears to require an entity. The question is whether the appearance survives a careful rewriting, and if it does, what the entity would have to be like, and how one would be told apart from two.

The next article takes the properties case and pushes it, because the fork it reaches is the oldest question in the subject and the one the rest of the tier keeps returning to.