views updated


We think and talk about things that do not existor so it seems. We say that Santa Claus lives on the North Pole and that unicorns are white. We admire Sherlock Holmes or judge him to be more clever than J. Edgar Hoover. People search for the Northwest Passage and the Fountain of Youth. They dream about lottery winnings and fear disasters that do not materialize. A childless couple hopes for a daughter. So, according to Alexius Meinong and others, there are things that do not exist. Even to deny that Santa Claus or the Fountain of Youth exists, we must be able, it seems, to identify what it is whose existence we are denying.

Bertrand Russell's rejection of this line of thought is well known. Sentences containing expressions that appear to denote nonexistents are to be paraphrased, in accordance with his theory of descriptions, with ones that do not. Russell shifted the emphasis from thoughts and other intentional attitudes that appear to have nonexistents as objects to the language in which those thoughts and attitudes are expressed. Many later analytic philosophers shared with Russell a distaste for what they saw as Meinong's bloated universe. Even those who rejected his theory of descriptions often assumed that apparent references to nonexistents can somehow be paraphrased away. But there have been few serious attempts since Russell's to show how this can be done, and the task has proven to be much more difficult than it once seemed. Since 1970 several very different sophisticated realist theories were developed, some of which claim not that there are nonexistent objects, but that the entities in question exist (Woods, Van Inwagen, Parsons, Routley, Wolterstorff, Thomasson). These have been countered by a new generation of antirealist theories, many of them based on notions of pretense or make-believe (Evans, Walton, Yablo, Kroon; also see Currie).

Many discussions after 1970 have focused especially or primarily on one variety of purported nonexistents: characters and other objects in fiction and mythology. Posits of failed scientific theories (Vulcan, ether, phlogiston), sought after marvels and wished-for children, the golden mountain, the round square, and the present King of France are often treated along the way, although the issues they involve are not entirely analogous to those concerning fictions. Fictions are in some ways especially compelling, and also especially puzzling. We speak easily and elaborately about fictional characters as though they were ordinary people, describing Sherlock Holmes as a detective who lives on Baker Street, speculating about Hamlet's motivations, and recounting the amorous adventures of the various Don Juans. Yet when pressed in certain ways, we readily deny that there are such things as fictions. Parents assure their frightened children that there really are not any goblins or monsters or ghosts like the ones in storybooks, and they confess to having lied about Santa Claus.


What sorts of things are nonexistents, if there are such? Some take descriptions of Sherlock Holmes as a man and a detective at face value, understanding characters to be people, to possess the same kinds of ordinary properties that real people do, and to differ only in lacking existence. The golden mountain is, literally, golden and a mountain, according to Meinong, and the wished-for child is a child. Such literalists, as Kit Fine (1982) calls them, usually accept that, unlike existing objects, most nonexistents are incomplete (Holmes neither has a mole on his back nor lacks one) and some are impossible (fictional time travelers, the round square).

Literalism threatens to get out of hand, at least as far as fictions are concerned. Not only do we readily describe Holmes as a person and a detective, we are also prepared to say, in much the same spirit (that is, speaking within the story), that he and other characters exist. Macbeth's dagger may be a mere figment of his imagination, but Macbeth himself is real; he exists. (We do not, in a comparable spirit, describe the childless couple's wished-for child or the golden mountain as existing.) In general, we are prepared to assert what we take to be true in a story, or fictional. It is fictional both that Holmes is a person and that he exists. If Holmes is literally a person, it is awkward to deny that he literally exists. But fictional statements that do not involve fictitious particularsstatements such as "There are ghosts" and "Julius Caesar was warned of the ides of March" understood literally and straightforwardly, obviously may fail to be true. Why should "Holmes is a person," not to mention "Holmes exists," be different?

The most obvious alternative to literalism, in the case of fictions anyway, is to treat statements like "Holmes is a person and a detective" as elliptical, as short for "It is fictional (true in the story) that Holmes is a person and a detective." Of course, we still seem to have an entity on our handsHolmes. He (or it) is not literally a person or a detective, but he is such that it is fictional that he possesses these attributes. And it is fictional that he exists, which does not have to mean that he literally exists. Holmes may possess other kinds of properties as well, ones that do not consist in something being fictional of him: He is a fictional character, and (on some accounts) was created by Conan Doyle, and is admired by millions of readers. Abandoning literalism in this way removes the embarrassment of an incomplete Holmes, and we need not worry about running into inconsistent fictional objects. Holmes is not such that fictionally he has a mole on his back, nor is he such that fictionally he does not have a mole on his back, even though, it is fictional that he either does or does not have one there. A character may, be such that it is fictional both that she is both P and that she is not P, but that does not mean that the character really does possess incompatible properties.

In an alternative to this strategy, developed by Edward Zalta (1988), Holmes is not a person in the sense that J. Edgar Hoover is; he does not exemplify personhood. Unlike Hoover, Holmes exemplifies properties such that of being a fictional character. Holmes bears a different relation, which Zalta calls encoding, to personhood, to being a detective, and to the other properties attributed to him in the Sherlock Holmes stories. J. Edgar Hoover, by contrast, is not the kind of thing that encodes properties.

Abstract Object Theories

If Holmes is not a person, what is he? What sort of thing is fictionally a person (or encodes personhood)? Realists who are not literalists usually understand fictions to be abstract entities of one sort or another, and to have whatever ontological standing the abstract entities in question do. Some take properties like being a person and a detective to constitute (rather than characterize) fictions, and so identify Holmes with the class of properties or conjunction of properties attributed to him in the stories. Some construe fictions as abstractions of other sorts: "theoretical entities of literary discourse" (Van Inwagen 1977), "kinds" (Wolterstorff), or "abstract artifacts" (Thomasson 1999). Zalta (1988) has fictions exemplifying abstractness. Different abstract-object theories give different answers to a battery of tricky questions about the identity and individuation of fictions and other nonexistents or nonactuals. Are they Platonic entities that are (some even say "exist") necessarily and eternally (Parsons 1980), or are they created when, for example, the relevant story is written or when they are thought about (Van Inwagen 1977, Thomasson 1999)? Do they cease to be if the story is destroyed and forgotten? If characters in different unrelated stories happen to have exactly the same characteristics attributed to them, are they identical? Are undifferentiated characters in a single fiction distinct from one another (the individual sheep in a fictional flock, for instance if nothing is said about any of them apart from the others)? Can the same character appear in more than one story if the characteristics attributed to it in each of them are not exactly the same? If so, by virtue of what are the characters identical?

The apparent fact that readers admire Holmes, or care about characters in stories, poses an awkward challenge for abstract-object theories. Do readers admire and care about abstract entities, be they properties or classes or theoretical entities or abstract artifacts? This is certainly not how readers themselves think of their experiences. It hardly helps to claim that Holmes is a person in the sense that he "encodes" personhood. He belongs to an ontological category fundamentally different from that of the usual objects of admirationMahatma Ghandi, Abraham Lincolnwhich exemplify personhood and do not encode any properties at all.

The antiliteralist might deny that people do, literally, admire or care about Holmes or Willy Loman or Desdemona, just as he denies that it is literally true that they are persons (or that they exemplify personhood). But then what is the reader's relation to them? Does the reader imagine admiring or caring about these abstractions, or is it true in an extended fiction that he admires them? Does the reader imagine of an abstract object that it is a person, one that he admires and cares about? That would be quite an imaginative feat!

Similar worries arise simply with the notion of fictionality and infect purported nonexistents of other kinds, as well as fictions. Fictional propositions are commonly characterized as propositions that appreciators or readers are to imagine, or ones that works of fiction invite them to imagine. If it is fictional that Holmes is a person, readers are to imagine of this abstract object that it is a person (or to imagine of something that encodes personhood that it exemplifies personhood instead). To imagine this would be to imagine a blatant impossibility. If a wished-for daughter is not actually a daughter or a person but an abstract entity, does the childless couple wish, futilely, of this abstraction, that it is a daughter and a person?

Pretense Theories

Pretense or make-believe theories return to a more intuitive understanding of statements like "Holmes is a detective," without embracing literalism. The speaker pretends to refer to an ordinary existing person and to attribute to him, in the ordinary way, the ordinary property of being a detective. Within the scope of the pretense, everything is normal. Yet nothing is actually referred to, and what is said, understood literally, is not true. This is pretense, yet with a serious purpose. The speaker does actually assert something by engaging in the pretense, very likely something about the Sherlock Holmes stories. Pretense theorists need to give some account of what is asserted, though it may be asking too much to expect an exact literal paraphrase. Part of the point of speaking in pretense to make a serious assertion may be to express something that is difficult or impossible to express literally (Yablo 1998).

Some philosophers find pretense accounts of "Holmes is a detective," "Hamlet hesitated," and the like plausible, but draw a sharp line between these statements and statements such as "Holmes is a fictional character" and "Holmes is smarter than any real detective." In the Sherlock Holmes stories it is presumably fictional that Holmes is a detective; readers are to imagine that this is so. In saying "Holmes is a detective," speakers are playing along with the fiction, pretending to assert of a person they refer to as "Holmes" that he is a detective. But it is not fictional in the stories that Holmes is a fictional character. So, it is claimed, to say "Holmes is a fictional character" is not to play along with the fiction; the speaker must really be referring to something by means of "Holmes," not just pretending to, and attributing to the thing referred to the property of being a fictional character.

This line is not a sharp one, however, and in any case, it is not to be drawn in the place indicated. People often speak with tongue more or less evidently in cheek when what they are expressing is not fictional in an established work of fiction, not what a recognized work of fiction prescribes or invites them to imagine. We play along with established fictions in special or unusual or unauthorized ways, altering or extending them in various directions, in order to make serious points by engaging in pretense. Sometimes we improvise new fictions. The commentator who remarks that the Hardy Boys, still living at home and attending Bayport High, have turned 75, and that their publisher now equips them with cell phones, is speaking in pretense, although what he pretends to assert is not fictional in any of the stories or the series as a whole. He is, in effect, observing that the Hardy Boys stories have been published for 75 years, and that it is fictional in some of them that the brothers use cell phones. In explaining the tenets of a discredited scientific theory, we may convert it into a fiction, speaking as though we accept it as true. An example: "Vulcan is a planet in our solar system between Uranus and Neptune."

Pretense theorists propose to understand other kinds of apparent references to fictional objects and nonexistents as merely pretended, or at least as less than straightforwardly literal. Evaluating such proposals is not easy. Apparent references to wished-for children, failed scientific posits, and claims of existence and nonexistence often lack any apparent tongue-in-cheek flavor. But pretending, like other psychological states and processes, need not be explicit, conscious, or open to introspection. That people are engaging in pretense may be the conclusion of inferences to the best explanation. Moreover, there is room for adjusting or refining the notion of pretense, or replacing it with something weaker. Some pretense theorists prefer to characterize speakers as merely making as if referring to something.

In the end, what matters is the success of one or another variant of the pretense theory as a whole and how it compares to its competitors.

See also Existence; Fictionalism; Meinong, Alexius; Realism; Russell, Bertrand Arthur William.


Azzouni, Jody. Deflating Existential Consequence: A Case for Nominalism. Oxford, U.K.: Oxford University Press, 2004.

Currie, Gregory. The Nature of Fiction. Cambridge, U.K.: Cambridge University Press, 1990.

Donnellan, Keith. "Speaking of Nothing." Philosophical Review 83 (1974): 331.

Evans, Gareth. The Varieties of Reference. Oxford, U.K.: Oxford University Press, 1982.

Everett, Anthony, and Hofweber, Thomas, eds. Empty Names, Fiction, and the Puzzles of Non-existence. Stanford, CA: CSLI Publications, 2000.

Fine, Kit. "The Problem of Non-existence. I: Internalism." Topoi 1 (1982): 97140.

Howell, Robert. "Fictional Objects: How They Are and How They Aren't." Poetics 8 (1979): 129177.

Ingarden, Roman. The Literary Work of Art: An Investigation on the Borderlines of Ontology, Logic, and Theory of Literature. Translated by George G. Grabowicz. Evanston, IL: Northwestern University Press, 1973.

Kroon, Fred. "Belief about Nothing in Particular." In Fictionalist Approaches to Metaphysics, edited by Mark E. Kalderon. Oxford, U.K.: Oxford University Press, 2005.

Meinong, Alexius. "The Theory of Objects." In Realism and the Background of Phenomenology, edited by Roderick Chisholm. New York: Free Press, 1960.

Parsons, Terrance. Non-existent Objects. New Haven, CT: Yale University Press, 1980.

Quine, W. V. O. "On What There Is." In his From a Logical Point of View. New York: Harper, 1953.

Récanati, François. Oratio Obliqua, Oratio Recta: An Essay on Metarepresentation. Cambridge, MA: MIT Press, 2000.

Routley, Richard. Exploring Meinong's Jungle and Beyond: An Investigation of Noneism and the Theory of Items. Canberra, Australia: Research School of Social Sciences, Australian National University, 1980.

Russell, Bertrand. "On Denoting." Mind, n.s. 14 (1905): 479493.

Thomasson, Amie L. Fiction and Metaphysics. Cambridge, U.K.: Cambridge University Press, 1999.

Van Inwagen, Peter. "Creatures of Fiction." American Philosophical Quarterly 14 (1977): 299308.

Voltolini, Alberto, ed. Do Ficta Follow Fiction? Special issue, Dialectica 57 (2003).

Walton, Kendall. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, MA: Harvard University Press, 1990.

Wolterstorff, Nicholas. Works and Worlds of Art. Oxford, U.K.: Clarendon Press, 1980.

Woods John. The Logic of Fiction: A Philosophical Sounding of Deviant Logic. The Hague: Mouton, 1974.

Yablo, Stephen. "Does Ontology Rest on a Mistake?" Proceedings of the Aristotelian Society, supp. vol. 72 (1998): 229263.

Zalta, Edward N. Intensional Logic and the Metaphysics of Intentionality. Cambridge, MA: MIT Press, 1988.

Kendall L. Walton (1996, 2005)

About this article

Nonexistent Object, Nonbeing

Updated About content Print Article Share Article