Kant argued that existence isn't a predicate; and presumably a similar argument would show that it is neither a property.

But if ontologically we believe all that there is are bundles of properties; it seems we must accept that then existence must be a property - otherwise it is the sole exception.

What general problematics assert themselves in ontological theories that ground themselves in such bundle theories (that is where existence is taken to be a property)?

One possibility suggests itself, which is a possibility that can't be found in substance based ontologies; this is are there entities that lack the property of existence, but have other properties?

This is paradoxical, when one translates the word 'there' as a word of existential import; but it is notably I suppose the position of Meinongonism.

  • Generally speaking, if Kant says something, it is not necessarily false.
    – Daniel
    Sep 7, 2015 at 1:02
  • Existence (in reality) is not the only non-predicate, so is uniqueness or possibility or coherence. What is common to them is that they "do not add a thing to the concept of a thing" in Kant's words. The way they function is that first we have to form a concept, and only then investigate if it exists, or is unique, or possible, or even meaningful. Whichever answer obtains makes no difference to the concept.
    – Conifold
    Sep 8, 2015 at 0:07

3 Answers 3


I don't think that Kant's existence-argument poses a special problem for bundle theories.

The thrust of Kant's argument is that the following two questions are separate and independent: the first, whether x exists, the second, what is x like, i.e. what properties x possesses. Existence is, according to Kant, neither a property, nor an entity of any kind.

Bundle theories pertain to the second question, what x (in this case, a particular entity) is like. Bundle theories hold that a particular is a bundle of properties, without an additional substance or substratum.

The question, what is a particular like is, according to Kant's argument, independent of the question whether the particular exists or not. Kant's famous example was: a hundred thalers possesses the same properties, whether it is imaginary or real. And it is immaterial to the point of the example whether the hundred thalers is a substance, a property bundle or whatever.

So if Kant was right, there is no use for an existence-property, whether we believe in bundles or in substances. There are other theories which do support existence-properties, but these theories are unrelated to the question of bundles.


If you fall far enough back down the scale of 'bundle theories', you are reduced to a place where existence is, in fact, not only a property, but an accident. Philosophically, we do not like this notion, but it has become inescapable.

Both in quantum physics, and in many psychodynamic theories in psychology, only the 'tendency to exist' can actually exist. (And I will not play the word game about whether the tendency then exists or only tends to exist. We are already well beyond the point where language will help make any of this clearer.)

The probability of a particle being in a location can only be determined when another particle interacts with it. You need two complex vectors to have a positive dot-product. Otherwise the particle may, in fact, never need to exist, in order to have its effects on the universe. It might only increase the odds of other particles existing, or influence them to be in a slightly different location.

Physics, forced into this corner, does fall back on the wave equation, which makes intuitive sense only if you adopt a notion like the tendency toward existence as a pervasive substance. The probabilities accumulate as complex potentials, so in the absence of true existence, the particles can still accumulate and have influence, as a field effect might, which is basically an immaterial substance. We can get patterns of interference that create multiple photons at once, quite close together, even though until they had a reason to interact, they did not actually exist as photons, but only as potential photons.

At the same time, object-relational entities that are based in intersubjective communication can only 'tend to be' part of a person's thinking or motivation. You need aspects of the environment to draw them out, and they may lie dormant forever, and only be expressed by other bearers of the same complexes. That does not mean their presence in one individual who does not communicate them is irrelevant to the person who does. Like the particle, the odds accumulate even when they don't exist, and when precipitated, will be composite.

The panic in a crowd may be precipitated by one individual's action, but is already incipient in the crowd itself, and might not have come to the point of precipitation if the set of individuals present were not the same, or did not at least contain tendencies toward the same fears.

Of course, this is not the level of reality where most of us dwell mentally, so the billion-to-one chance that existence actually acts like an accident is not worthy of ordinary consideration. But in an absolute sense, we can no longer deny that this is the case.

Of course, strictly taken , examples do not answer your question. But I would contend that this model of a subtle substance correlating to the tendency to exist is not too accidental, and that we are driven to it by the structure of our inquiry.


First of all: anything with properties necessarily "exists." The question is a specific type of existence -- in actuality, as opposed to as an idea only. The idea behind the ontological was that, if you could conceive of God, then he existed as an idea, and one of the properties that idea would have is that it also existed in actuality. It's a cute argument, but you really should be able to tell why it doesn't make sense; "existing" is not one of the parts of an idea, it's an idea, that's just what it is. It's as though you're talking about people; and you say that this person has the property of having blonde hair, and this person has the property of having dark skin, and this person has the property of being a television. No, people can't have that property. The same way ideas can't have the property of "existing" in actuality -- although they might refer to something that does.

"Existing," isn't a property. If you want to talk about it as though it's a property, it isn't a regular property, like the ones that can be added and taken away from things -- you can't take an idea and turn it into a real thing, or a real thing and turn it into an idea (although you can make physical embodiments of ideas, and break them). It could not be added to or removed from a bundle of other properties. It would describe that bundle, and whether that bundle refers to a thing in reality, or to an idea. But it wouldn't be a part of that thing. Existence is not proper to me. I exist, and then everything else about me follows that in some inferior way.

Maybe you want to call that a property, but it's clearly different from all the others. It seems to me different enough, at least, that it's not something that can be ascribed to an idea, such that God, if he can be conceived, can magically transubstantiate from an idea to a real thing.

Alternatively, you can say that, when you're conceiving of that greater than which nothing can be conceived, and one of the properties you think of it having is that it exists, then great, you're thinking of it existing. But it's still just a thought -- it's just, you think of it as something that has the property of existing. You can't actually give it that property by conceiving of it as having that property. I could conceive of a Maserati that had the property of being in my driveway, if you want to call that a property, but I can't actually give the Maserati that property; I can only think of it as having that property. An idea can't have "existence" just given to it the way it can have "red" given to it.

(Woah, this went a lot longer than I intended, sorry).

  • In the paragraph where you talk about "the idea of the ontological", you're confusing Descartes' argument for the existence of God (which is sometimes classified as a cosmological one and sometimes as an ontological one) with Anselm's argument. Anselm's argument doesn't depend on the initial idea of our ability to conceive the idea but rather a definitional property of the idea (doesn't mean it will work either -- but it's a different structure).
    – virmaior
    Sep 7, 2015 at 2:13
  • The claim "Existing" isn't a property is a complicated thing to say, because the words "property" and "existing" both mean different things to different philosophers. The definitions you seem to be working from are common contemporary definitions, but the question seems to be asking something deeper.
    – virmaior
    Sep 7, 2015 at 2:15
  • I am in no way referencing any Cartesian philosophy. Anselm's Ontological Argument is a definitional one, but defines god as that than which nothing greater can be conceived, and such a definition necessarily depends on our ability to conceive God. I don't know what part of this you think I am confused about, but I assure you, I am not.
    – Daniel
    Sep 7, 2015 at 18:35
  • I'm not quite sure what the point of your second comment is... I didn't go through many definitions of "property," but I think I made a compelling point, particularly in reference to Kant's counterargument. Existence is not a predicate in the sense that it can be "given" to a thought, or "added" to a bundle as described in the question asked. There may be some definition of predicate or property that includes meta-properties like existence, but I think that would be a rather silly definition, and not very relevant to this discussion.
    – Daniel
    Sep 7, 2015 at 18:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .