How is it possible for things that do not exist to not be the same? How can one differentiate nonexistent entities? How can I know the difference between ghosts and werewolves if neither exist?

  • 5
    As a flagrantly non-constructive comment. That question is awesome. Commented Jun 17, 2012 at 23:19
  • 4
    They didn't exist seperately, until the Word seperated them. To name something (even a fictional no-thing) is a partly magical act.
    – Marxos
    Commented Sep 3, 2012 at 19:55
  • 2
    A related question would be how to discern if an entity exists or not. Think gravity for example.
    – nakiya
    Commented Dec 9, 2014 at 5:15
  • 1
    As an aside, there's an analogy here in computing: If two values aren't defined, are they the same? Eg if an employee record and a sales invoice are both null (ie neither exist), are they the same value? Answer: not generally, because they have different Types. Comparing the two is probably meaningless. The comparison (Employee_record==null AND sales_invoice==null) may be true, but the comparison (Employee_record==Sales_invoice) will be false, or give an error. You know enough meta info about the nature of the items to determine their non-equivalence even if they have no value. Commented Dec 12, 2014 at 12:50

12 Answers 12


Ultimately, it all comes down to what you mean by "exist". Werewolves and ghosts do indeed exist, as fictional objects. And thus they can be distinguished from each other within the fiction, even though neither one exists in the real world.

If you wish to pursue this, I'd recommend you look into Husserl's notion of "regional ontologies", which attempts to capture the different types of existence from a phenomenological perspective.


I just want to add the SEP links to Non-Existent Objects and Fictionalism, for further reference.


I have to disagree with the "it depends on what it means to exist" answer on two points: The original question does not come down to what it means to exist, and werewolves and ghosts do not exist. Existence may not be a simple fact of the matter, nor easy to elucidate property, but it is manifestly true there are no werewolves or ghosts. If there were, by which I mean if being part of some universe of discourse were sufficient to establish existence, then among other things, the original question would be un-askable.

I think that what OP is asking is: is it possible for one thing with no properties to be different than another thing with no properties? The use of "possible" here is left wide open: If OP is asking this question metaphysically, conceptually, physically, inherently, relationally, etc., then there are going to be different approaches to answering it.

We might get some insight on the question if we keep Leibniz's Identity Laws (the Identity of Indiscernibles and the Indiscernibility of Identicals) in mind. If werewolves and ghosts have all their intrinsic properties in common, they are the same thing, and being that they both have 0 properties, they have all their properties in common. However, being that they also have none of their properties in common, because all of their properties are a sum of zero properties, and having no properties in common means having a sum of zero properties in common, I'd have to say they belong to the empty set.

A way to possibly undermine my conclusion above would be to argue that they are different insofar as I have been able to count two instances of them, so they must have some properties that they don't share. If you're inclined, I recommend looking to Georges Rey's work on intentional inexistence (2003, 2005, 2008) and Sainsbury (2008) on "reference without referents".

  • 2
    If you want to get technical, you will find it difficult to really prove anything exists outside of your own mind. We wade through a world of ideas, and these ideas we ascribe properties to. Unicorns and pizzas and ghosts and the concept of pi all share in that they are ideas and with properties, capable of existing and not existing. However, the fact that something exists or not does not automatically nullify it's other properties. In fact, existence itself is not a property; it describes a relationship between the subject (me, the thinker) and the idea (unicorns/pizza/etc).
    – stoicfury
    Commented Aug 19, 2011 at 0:03
  • 1
    Fortunately, my argument doesn't rest on positing the existence of anything. On the contrary, I say that they specifically don't exist, so my position is not going to be particularly susceptible to generalized appeals to idealism. Commented Aug 19, 2011 at 3:03
  • 4
    I disagree with your interpretation of the question; the OP seems to be operating under the assumption that anything lacking existence is thereby lacking all properties. Ghosts and werewolves have properties, stipulated in the fictions in which they exist, and thus can be distinguished on the basis of those properties. Commented Aug 19, 2011 at 7:05
  • 1
    To "exist in fiction" is to "be real in non reality". This is a self-contradicting proposition, and thus werewolves and ghosts are members of the empty set (aka, the law of the excluded middle need not apply). The fictions in which we engage may have properties, but the things described in them do not. If they did, they wouldn't be fictions (barring lies). Commented Aug 20, 2011 at 8:56
  • 3
    They have properties within the fiction. Thus we can distinguish between them when talking about them in the context of the fiction, even though we are not going to meet either one in reality. There is a large (and growing) literature on fictionalism; you can find an overview in the SEP article on the subject (plato.stanford.edu/entries/fictionalism) Commented Aug 20, 2011 at 14:06

I tend to agree with Michael on this one. An analogy might be helpful.

Take numbers. Does the number 1 exist? Can you find and show me 1? How do you know 1 is different from 0? Different from 2? I've never encountered a 1 or a 0 in real life. I've seen a representation of it, perhaps, but never the thing itself.

We take these things to be different from each other by virtue of their properties. For example,

1 > 0

But how can we prove it? The law of trichotomy is that for two entities a and b , a is either less than b, greater than b, or equal to b. This law does not hold for all sets of numbers. For the sets of real numbers, integers and rational numbers it is taken as an axiom. It cannot be proven unless under certain conditions without falling into circular reasoning.

So how does a mathematician differentiate these nonexistent entities? The example I use here are a certain group of numbers and the operations that can be performed upon them. Michael has called this a "context" and in math we call them groups, fields, and rings. Such differentiation is done very carefully. A mathematician will commonly attempt to prove (1) existence and (2) uniqueness for this very reason. Both are necessary to make distinctions between these abstract entities.

Now that we might see a way to differentiate between entities that don't exist in the real world but can still be thought of, the question for me is are there entities that neither exist nor can be thought of? And how would we ever know?


Obviously you can't differentiate things that do not exist, given that there is nothing to differentiate. However you can differentiate descriptions.

(1) "Ectoplasm" was supposed to be a substance or spiritual energy "exteriorized" by mediums.

(2) "Phlogiston" was supposed to be a fire-like element called phlogiston, contained within combustible bodies and released during combustion, which has negative weight.

Clearly if there were such things as ectoplasm and phlogiston, we could differentiate them. There aren't, but we can still differentiate their descriptions.


Question 1. "How's it possible for things that don't exist to not be the same?"

Simple. Upon reading a fictional novel, can you not make a separation of the figures within? And since you cannot prove, beyond a shadow of doubt, the existence of those human constructs also means that all of them can be put in the same classification of fictional characters.

Question 2. "How can one differentiate nonexistent entities? "

See my statement above.

Question 3. "How can I know the difference between ghosts and werewolves if neither exist?" Can you not notice a clear separation between the actual objects of reality, and those from a fictional narrative? Because if you can't ... then you're truly lost. And I'd seek help if I were you. Good day


Reasoning/judging rely on language rather than on physical existence of objects (something you can see/touch/smell etc). That is why you can talk about ghosts, elves or even microbes that maybe you have never seen. Same thing about simple expressions as "hi" and "bye". All of theses expressions are elements in your communication system, signifiers that correspond to concepts and have different values. Actually this linguistic organization is determinant of the delimitation of objects of thought, so that in some culture there may be no difference between green and blue and both are considered the same color. I guess there's a problem with the lack of definition of existence in your question.


I think this goes back as far as Russell and his theory of definite descriptions (“On Denoting” 1905) which in part tried to deal with non-existent entities and to preserve the law of the excluded middle (vis. statements about non-existent entities seem to be neither true of false). If you have a concept to which nothing existent corresponds then that concepts extension is said to be the null set. So the extension of the concepts ghost and werewolves are equivalent- they are both “non-differentiable” members of the same null set.

  • Well, considering non-real entities goes back significantly further than Russell. And staying in the phil. language domain, the extension may be identically empty but the intension is different.
    – virmaior
    Commented Apr 21, 2015 at 22:20
  • @virmaior. You could not have put the point more clearly or concisely. GT
    – Geoffrey Thomas
    Commented Jul 18, 2018 at 6:44

Q. How can I know the difference between ghosts and werewolves if neither exist?

Because knowing is something that takes place in either the brain or mind, and the brain/mind is capable of creative thinking about imaginary things that do not materially exist.

Q. How is it possible for things that do not exist to not be the same?

Because imaginary or real, all things have properties that are universals rather than particulars (sometimes called the essence)- so a man may be tall in relation to a child, but short compared to a giraffe. Tallness is not to be found in the man himself.

Unicorns, though imaginary share universals with horses, and partly with antelopes, or similar horned animals. Werewolves share universals with men and wolves.

Sameness is yet another universal, maybe equal to a set of universals denoting similarity, so it can fail to apply to any pair of particulars, real or imaginary, or even particular types.

Q. How can one differentiate nonexistent entities?

Either by an act of pure imagination, or by real world modelling, which also requires some imagination. For example one might model the Three Billy Goats Gruff with real life goats of various sizes and attempt to sieve them to find the largest.

Even coding a virtual reality to model the nonexistent entities could work, in principle, but someone needs to imagine their properties first.

And as for impossible entities, as opposed to imaginary entities, such as the 3 sided square mentioned in earlier answers, we just have to think of the illusions in pictures by Escher to see that there are ways in which the impossible becomes imaginable.

  • Welcome to Philosophy SE! Must say I liked this answer, but it would be great if you could add some references; people appreciate the "further reading" opportunities.
    – christo183
    Commented Apr 13, 2019 at 16:07
  • @christo183 Thankyou, but I am not a philosopher as such, So I don't necessarily know where this fits in standard philosophy. But I take your point and will bear that in mind in future.
    – Jeremy C
    Commented Apr 13, 2019 at 19:06

In set theory, we can define our sets with any arbitrary characteristics. Any entity that has those characteristics is, indeed, in our set. If I define a werewolf as:

  • Large teeth
  • Claws
  • Human by day
  • Wolf by night
  • Changes during a full moon

I would not be unjustified in saying that werewolves, by definition, have all of these characteristics. That's not to say that there are any entities in reality that exist with these characteristics, but nonetheless werewolves are still defines by these tenets. Also, I'm certainly not a Platonist, so I don't believe that werewolves exist in another 'realm' simply because they have a concept, but they do, indeed, have defined characteristics, despite not existing in nature.

All things, not limited to physical items but conceptual models as well, have definite characteristics even if they do not or can not exist within reality. For example, a square with three sides has definite characteristics, namely it is square and has only three sides, yet it is inherently contradictory and cannot manifest in reality. We differentiate them based on these characteristics, and existence becomes irrelevant.

  • 1
    I think you're using "all things" in a confused way in your TL;DR. Wouldn't it be better just to say that concepts can also have categories?
    – virmaior
    Commented Apr 18, 2015 at 1:32
  • Would you agree with the edit I made? Particularly, the "square triangle" analogy?
    – Goodies
    Commented Apr 19, 2015 at 2:16
  • I don't see how something with three sides is a square. While I agree such a thing can never be instantiated, I think it cannot even be a coherent concept, because square as a concept contains four-sidedness.
    – virmaior
    Commented Apr 19, 2015 at 2:18
  • It surely isn't coherent, but that isn't the goal. We have two sets: a set in which all elements are squares and a set in which all elements are triangles. This object can be described as being in both sets, but that limits it from being within a set of 'possible objects' or 'coherent concepts.' It's quite likely that I'm merely misunderstanding set theory.
    – Goodies
    Commented Apr 19, 2015 at 2:24
  • 1
    I think you're just misunderstanding set theory there. The problem is that the determinant of whether something falls in the set of triangles is whether it has three sides and whether it falls in the set of squares is whether it has four sides of equal lengths and 90 corners. No object can fall within both sets because membership in each set excludes membership in the other set.
    – virmaior
    Commented Apr 19, 2015 at 2:25

do 'thing's exist if an individual thinks they exist? Ghosts and werewolves have a substance and a form but this may be socially constructed to be deemed to be real if an individual beleives this form to exsit. Similar to if god exists?

  • This is an interesting point -- is there any chance I could encourage you to unpack it a bit?
    – Joseph Weissman
    Commented Aug 25, 2011 at 22:35
  • It reminds me of the difference between "exists" and "exists in the mind/consciousness" Commented Aug 26, 2011 at 13:00
  • I guess the concept of werewolf/vampires exist, even if none have been discovered (yet, muahahaa) Commented Dec 12, 2014 at 12:41

I think the fictional entities line of thought is helpful, on a general level, but if you look too closely things start to get confused, and you suddenly have to develop an entire ontology.

Part of the answer that hasn't been mentioned yet in precisely this form, is that it is possible for us to have different propositional attitudes to fictional entities, in the actual world. I believe and know different things about Achilles than I do about Superman.


There are many formulations of ontology, but specifically to categorize "unseen" objects you may find Alexius Meinong usefull:


Briefly, he held the being (having the capacity to be thought of), of an object, was prior to it having existence. He categorized objects into three types of being:

Absistence - Impossible objects like square circles. Having a subset that has:

Subsistence - non-temporal entities such as mathematical objects. Of which an even smaller subset having:

Existence - material and temporal expression, those things you can actually shake a stick at.

  • Valuable comment but shouldn't 'subsistance' be 'subsistence' (as in the standard translations) ? Also 'absistance' (='absistence'') is new to me. I can't find it in Routley's Meinong's Jungle or in the Stanford article. This is probably just my ignorance but can you point me to a source for 'absistence' ? Best - GT
    – Geoffrey Thomas
    Commented Jul 18, 2018 at 7:03
  • @GeoffreyThomas corrected spelling, thanks. The wikipedia article cited explains absistence nicely. It may be omitted some places because it refers to objects outside of immanent ontology. I believe Meinong held that since, for example a square circle, can be named or spoken of, there should be a type of (non)being to categorize it.
    – christo183
    Commented Jul 18, 2018 at 10:18
  • Thanks - I see the 'English' term, 'absistence', was coined by JN Findlay. It doesn't strike me as particularly apt but at least I now know where it comes from. Findlay did useful work on Meinong decades back - I don't think his interpretation and terminology are quite leading edge nowadays. Thanks. Best - GT
    – Geoffrey Thomas
    Commented Jul 18, 2018 at 11:33

You must log in to answer this question.

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