I know what noneists and abstractionists say the difference is, I just don't grasp the difference. Noneism and abstractionism are two approaches to objects like numbers, fictional character, hypothetical objects that don't exist, etc. The basic problem is sentences like

(1) Sherlock Holmes smokes a pipe.

This sentence seems to be true, but how can it be true if Sherlock Holmes does not exist? A lot of philosophers deny that (1) is literally true; They explain why we agree with the sentence by some other criteria. Noneists and abstractionists both claim that (1) is literally true but they have different ways of explaining how it can be true.

The abstractionist says that Sherlock Holmes does exist, but only as an abstract object. That is, abstractionists distinguish between existence and actuality. Sherlock Holmes exists but is not actual. Sentence (1) is true because it makes non-actual claims about non-actual objects, but those claims are true.

The noneist claims that Sherlock Holmes does not exist but that there are things that don't exist and that there are true statements about those things.

Let's call statements about the ontological status of an object an ontological statement. Then the noneist and the abstractionist seem to disagree on ontological statements about fictional objects but agree on all other statements. I'm having a hard time seeing any real difference between these two views other than ontological terminology. What is the difference between saying that fictional objects aren't actual vs. saying they don't exist, if in both cases, the exact same set of non-ontological statements are true of those objects?

  • 1
    My lack of understanding is even deeper :-) I do not understand the benefit of problematizing the sentence "Sherlock Holmes smokes a pipe." with such pretentious terms like noneist and abstractionist. - There is an imagined figure named Sherlock Holmes and the story imagines that this figure smokes a pipe. All together, the figure, the pipe and the activity to smoke a pipe are imagined, neither is real.
    – Jo Wehler
    Nov 8, 2023 at 17:52
  • Humans love a good story.
    – Scott Rowe
    Nov 9, 2023 at 0:02
  • 1
    In practice, they have to use different technical means to support their ontological claims. For noneists, it is a modalized Meinongian ontology with some revisionary logic attached, and for abstractionists it is a more traditional variant of Meinongianism. Both have (claimed) advantages and problems, and also implications elsewhere. So pragmatically, the debate is over which formalism is technically superior, matches intuitions better, and better fits broader theories of abstract objects and the like.
    – Conifold
    Nov 9, 2023 at 9:42
  • @Conifold, I don't see how that makes them different. The modal Meinongian says that the non-existent object exists and has its properties in a world (which need not be a possible world) while the abstractionist says that it exists and has its properties in some sort of abstract scenario or context. The two seem isomorphic. Nov 9, 2023 at 14:52
  • Reicher on SEP thinks that a form of abstractionism is superior, approvingly commenting that it "fits well into a general ontology of abstract artefacts... and does justice to the intuition that fictitious objects... are literally brought into being through human acts of creation". Noneism gets for conclusion a long paragraph-full of "difficulties" instead.
    – Conifold
    Nov 9, 2023 at 21:10

5 Answers 5


The difference is largely a matter of terminology. Where one group says that Sherlock Homes exists and the other that he does not, they are using the word 'exist' in different ways. Apart, perhaps, from objecting about your terminology, neither group would deny the essence of any common sense statements you would make about Sherlock Holmes, such as:

I can imagine Sherlock Holmes. Sherlock Holmes was introduced to the world by Sir Arthur Conan Doyle. Several actors have played Sherlock Holmes. I cannot touch Sherlock Holmes. Sherlock Holmes may be imagined by different people as having different attributes. No atoms ever formed Sherlock Holmes. Sherlock Holmes does not have a current physical location.

And so on. Essentially, what noneism and similar schools of thought amount to are efforts to find a formal way of accounting for the differences between types of object such as Sherlock Holmes and other types of objects. To do that you must define your terminology very precisely, and the different schools of thought adopt definitions that differ in sometimes quite nuanced ways. The task is very difficult, since a particular approach might lead to apparent contradictions which might take considerable ingenuity to resolve, and different philosophers naturally have conflicting ideas for the best way to go about it.

Take your example 'Sherlock Holmes smokes a pipe'. You ask how can that be true if Sherlock Holmes doesn't exist? There are at least two ways around that conundrum. One is to be clear that Sherlock Holmes was an imaginary character created by an author, so you could replace the sentence with 'The character Sherlock Holmes in the stories written by Sir Arthur Conan Doyle was said by the author to smoke a pipe'. There you have noted that a fact about Sherlock Holmes is really a fact about what Conan Doyle wrote. One other solution to the problem is to say that Sherlock Holmes does exist as a type of notional object with notional attributes, one of which is a propensity to smoke a pipe. These are not arguments about whether Sherlock Holmes 'really' exists, but about how to use words to think about the nature of Sherlock Holmes' existence in a helpful and rigorous way that is free from contradictions.


These two theories have different ontologies. What this means is that their entities may be different and there exist different semantics/meanings. It all depends on their definitions.

For example @Jo Wehler, from his comment could be a noneist: for him Holmes does not exist. In the ontology of noneist imaginary things do not exist but nevertheless they are entities, in the sense that claims can be made upon them: for example although Holmes does not exist, the claim he is smoking can be true: in a picture for example.

For me, Holmes exists - even fictious - as an abstract entity. I can see a picture of him, a film, read a book have an imaginary conversation etc. In this sense I am more close to abstractionism. In abstractionism, even imaginary (abstract) things exist in it's ontology and thus every claim can be made upon them.

Nevertheless, we agree in more things than you can imagine.

I'm not an expert on the field - but in order to make a piece in your mind - I believe that there exists a kind of "isomorphic equivalence" between them at least in the "logic" part but not in the metaphysical one.


The noneist seems to be varying over theories of ∃ or E, i.e. either a quantifier or a predicate, whereas the abstractionist seems to vary over a propositional operator "it is actual that..." alongside the more common modal operators. The strength of the difference between noneism and abstractionism then relies on the strength of the differences between the some-quantifier, an existence predicate, and an actuality operator. If these are effectively "symmetrical" or "isomorphic" enough, the noneism/abstractionism distinction is weakened; if there are subtle, but (somehow) important dissimilarities between them, then the distinction in philosophical standpoints holds more robustly.

So to say:

  1. "Not some are" doesn't always (or usually) equal "some are not," i.e. the phrases can be disambiguated equivalently, but they need not be. "Does not exist" and "does exist not" seem more often symmetrical, although I can dimly imagine someone getting a distinction out of the difference. "Not actually" and "actually not" also seem interchangeable (c.f. "actually possible" vs. "possibly actual").
  2. Noneism-via-quantification then seems to have room to be differentiated from abstractionism-via-propositional-operators, whereas noneism-via-predication seems barely different from that abstractionism.

Alternatively, one can look at the issue as one of two mutually interpretible theories/models. Every nonexistent object countenanced by the noneist can be had to correspond to a non-actual object countenanced by the abstractionist, and so on (or so we might think). You might then wonder why we would go through the trouble of conceiving two different such theories, but perhaps this turns out to be a coherentist moment in the greater philosophical system (i.e. one of the methodologies of coherentism is the production of multiple systems, in multiple logico-metaphysical sublanguages, that are then shown, or not shown, to cohere, so as to coherentistically justify them or not).

Caveat: something in the back of my head is making me wonder how adequate the above really is to your question, maybe some peculiarity involving (merely) possible objects or fictional entities. "It is fictional that not..." doesn't necessarily seem (always?) equal to, "It is not fictional that..." or, "It is not possible that..." is distinct from, "It is possible that not..." So how are we to situate possibilism and quasi-fictionalism(?) relative to noneism and abstractionism, then?


The "abstractionists" are correct. The fictional character Sherlock Holmes is often represented smoking a (fictional) pipe. Not any (fictional) pipe, mind you, but a large, iconic meerschaum pipe, at least in the movies.

The pipe is part of his attributes, with the trench coat, the hat and Dr Watson. All fictional.

Such attributes are the mark of heroes and gods, those often represented, who the public is expected to recognise, fictional as they may be. So the pipe is indeed iconic: it helps identify Sherlock, like the club and the lion skin help identify Hercules. If you see a classic statue or painting of a big, muscular guy with a club in hand and a lion skin on his shoulders, you can safely conclude that you're in front of a representation of Hercules.

So the example of Sherlock and his iconic pipe is a good one because everybody knows that he smokes it, even though he is fictional, and moreover the pipe is part of how he exists, as the same well-known fictional character across many books, plays and movies. Without his pipe, he would not be the character Sherlock Holmes that we know.

Sherlock Holmes smokes a pipe, almost by definition.

  • In recent times the 'everybody' who knows of a particular character, sports figure, song lyric, advertising jingle... is rapidly shrinking, and breaking up in to multiple everybodies. We have lots of icons in computer systems alone that many folks have never seen the original. Do you know why it is called a "telephone call" ? Even fiction is getting more fictional.
    – Scott Rowe
    Nov 9, 2023 at 0:07
  • @ScottRowe why is it called a "telephone call"?
    – Olivier5
    Nov 9, 2023 at 7:08
  • Because people used to "call on" (visit) others, in person. So telephoning was marketed as a way to call on people by voice. Why do we "hang up" at the end? The phone used to have a cylinder that you held to your ear (the receiver) and it hung on a hook, which disconnected the line. (line?) Why do some people say, "ring off" instead of hang up? Because you used to turn the crank a couple times after hanging up to ring a bell so the Operator knew to disconnect your line. Knowing the history of technology seems important, to me.
    – Scott Rowe
    Nov 9, 2023 at 11:46
  • 1
    @ScottRowe Thanks. "Hang up" always confuses me, because of the phonetically similar but semantically very different "hang on".
    – Olivier5
    Nov 9, 2023 at 13:00
  • "End" makes more sense, but what should it be named that is better than 'call'? Conversation? We should find names for things that don't change with changing technology. A music 'album' (collection of songs or pieces) works, but LP, CD, MP3, are doomed to obsolescence. Actually, my 'phone' does lots of things, so it needs a new name now.
    – Scott Rowe
    Nov 9, 2023 at 18:51

A couple of thoughts: Can something that exists in the brain, whether by original thought or by influenced thought, such as reading a book or seeing a movie, be said to not really exist? When we see or experience a relationship with a "real" person, what exists in our brain are images of that person. Whatever the physical makeup of that image, it is still just an image or memory. Does it matter to our experience whether that set of images comes from fiction or personal interaction? The distinction seems to me to be one of semantic definition. There are some things that exist in the mind as a representation of reality, and some things that are a representation of fictional being. Most, but certainly not all of us can differentiate between the two. But the fact that not everyone is capable of recognizing the difference between the abstract and the real indicate that the difference is learned rather than innate. But isn't all experiential reality illusion? Just playing devil's advocate, but it appears that noneism and abstractionism are neither clearly defined terms if the difference between them is strictly semantic.

  • Welcome to SE. Your answer would be greatly improved if you addressed the question in the terms that it was asked. It is about noneism and abstractionism specifically. If you think those terms are unhelpful, you should explain why and how your way of answering it is more helpful.
    – Ludwig V
    Nov 19, 2023 at 10:51

You must log in to answer this question.

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