Existent, Meaning, and Truth Sara Smollett
March 5, 2002

Sherlock Holmes: A Study in Existence, Meaning, and Truth

I wish to discuss the truth values of the following four propositions:

  1. Venus is a planet.
  2. Venus is a star.
  3. Sherlock Holmes is a detective.
  4. Sherlock Holmes is a pirate.

I propose that both the first and third propositions are true, that the second proposition is false, and that the fourth proposition is neither true nor false.

To arrive at this conclusion, I will first consider the distinction between Venus and Sherlock Holmes and, in particular, the question of the existence of Sherlock Holmes. I will then go on to discuss the truth values of propositions involving physically existent and fictional objects and the important distinctions between the two.

I believe that it is meaningful to say that Sherlock Holmes exists in some sense, but not in the same way that Venus does; namely, that Venus may be said to exist in the physical sense, whereas when Holmes is said to exist what is meant is that he exists in the fictional sense. As a result of this crucial distinction between types of existence, truth and falsity take on different meanings with respect to propositions about Holmes than they do with respect to propositions about Venus and other physical objects. I will introduce a third notion, definedness, to explain the difference.

I. Existence and Meaning

I have stated my belief that Sherlock Holmes exists. By this I mean that there is something, namely Sherlock Holmes, that is described and referred to by Sir Arthur Conan Doyle. I believe this is what existence entails. However, many philosophers, including Alexius Meinong, Bertrand Russell, and Terence Parsons disagree. While I recognize that there is an important difference between the existence of Sherlock Holmes and the existence of Venus, I believe that the word ``exists'' is used to refer to both types of existence.

To differentiate between the existence of Venus and the existence of Sherlock Holmes, I will use the terms physical existence and non-physical existence. To my way of thinking, Venus, Sherlock Holmes, unicorns, yellow pigs, frictionless planes, and numbers all exist in some way. I claim that in addition to physical objects, fictional, mythical, imaginary, hypothetical, and formal constructs can all be said to exist non-physically.

Of these many types of non-physical existence, I will focus on just one: fictional existence. I claim that Venus exists physically while Sherlock Holmes does not exist physically, but instead exists fictionally. For those who are bothered by the semantics of exists fictionally, I believe the Meinongian subsists, can be substituted without modifying the meaning of my discussion of the four enumerated propositions. However, I will continue to use the phrase exists fictionally to stress my belief that fictional existence is, in some way, a type of existence.

I believe that propositions involving properties of fictional characters such as Sherlock Holmes can be meaningful. I agree with Russell's assertion that the third proposition, that Sherlock Holmes has the property of being a detective, is not nonsense (Russell, 25). Neither is the proposition that Sherlock Holmes is one of the most well-known detectives (Parsons, 37). Parsons claims that ``Sherlock Holmes is a detective'' can be true without ``Holmes existing'' being true (Parsons, 41). But how can Sherlock Holmes have the property of being a detective without existing in some meaningful sense?

Sherlock Holmes does exist, at least in the sense that it is meaningful for him to have properties. While Holmes does not exist in the physical sense to which Parsons is referring, I claim that Sherlock Holmes exists fictionally and has properties within a fictional realm. I believe that often when we speak of existence we are looking for something more than the narrow notion of physical existence.

The important point I am trying to make, then, is not the semantic point that Sherlock Holmes exists, but that the existence of Sherlock Holmes is different from the existence of Venus. There is no physically existent Sherlock Holmes. What then, do I mean by saying he exists? I claim that Sherlock Holmes is a fictional character who exists in the possible world created by Doyle and defined in his literary works. Sherlock Holmes is assumed to exist within this fictional realm, and he has whatever properties he is defined to have (Companion, 361). The existence of Sherlock Holmes is relative to Doyle's fiction. This is what I mean by asserting that Holmes exists fictionally.

II. Truth Values

How does fictional existence differ from physical existence? One way is in its implications for evaluating truth values. Recall the four propositions from the introduction. The subject of the first two propositions, Venus, exists physically, while the subject of the second two propositions, Sherlock Holmes, exists fictionally. I believe that the reader will readily accept that the first proposition, ``Venus is a planet'', is true, while the second, ``Venus is a star'', is false. At first glance, one might say that the third proposition is similarly true and the fourth is similarly false, but I believe a closer examination will yield different results.

Before considering the truth values of the latter two propositions, I wish to address the question of whether it makes sense to discuss the truth or falsity of propositions involving fictional characters at all. I believe that it does, and that it can be meaningful to say that a claim about a fictional character is true or false relative to the fictional world of that character. What do I mean by this relativity? I mean precisely that the truth values of propositions of the form ``Sherlock Holmes is a detective'' can be evaluated only if such propositions are understood to be shorthand for propositions such as ``Within the world defined by Doyle, the fictional creation Sherlock Holmes is a detective.'' That is, I assert that propositions may be fictionally true or fictionally false. G. Evans offers a similar theory in which propositions can be said to be make-believedly true of false (Companion, 148).

I will now consider the proposition ``Sherlock Holmes is a detective''. I believe that the proposition is meaningfully, fictionally true. Relative to the world defined by Doyle, Sherlock Holmes is a detective. Why is he a detective? Because Doyle says so, and his fiction is the definitive source of all truths about Sherlock Holmes. That Sherlock Holmes is a detective is indisputable because it is a given. I claim that a proposition is fictionally true if it is a given within the fiction or if it necessarily follows from propositions given within the fiction.

Analyzing the proposition ``Sherlock Holmes is a pirate'' is a bit more difficult. One might say that the proposition is false, because Doyle does not mention it and if it were true, surely it would have been mentioned. Because the proposition is undefined, it must be false. But consider another example: ``Sherlock Holmes wears underwear''. I don't believe Doyle ever mentions that Holmes wears underwear. If we take as true only that which Doyle says is true, then do we want to commit ourselves to saying that the proposition ``Sherlock Holmes wears underwear'' is false? Certainly not, as the negation ``Sherlock Holmes doesn't wear underwear'' isn't given to us by Doyle, so it cannot be true either. The proposition ``Sherlock Holmes wears underwear'' is neither true nor false. Then there must be something besides truth and falsity. We want to say that the truth value of such undefined propositions is undefined (Wall, 986).

Does this mean then that all propositions involving fictional subjects are either true or undefined? No; there are also fictionally false propositions. For example, the proposition ``Sherlock Holmes has never played the violin'' is false because it contradicts Doyle's statement that Holmes plays the violin. This leads to the following definition: A proposition is fictionally false if and only if it is inconsistent with that which is given in the fiction.

One might then argue that Holmes couldn't have been a pirate because he was too busy doing all of the things Doyle said he did to have time to be a pirate; that is, that the proposition ``Holmes is a pirate'' is inconsistent with the set of truths about Sherlock Holmes. If the proposition is inconsistent, then it is fictionally false, but while the claim that Holmes is a pirate seems unlikely, I do not believe that it will lead to a contradiction.

It is not just that the truth value of ``Holmes is a pirate'' is unknown, for the truth value of ``Venus is a planet'' was unknown at one point, but that it is unknowable. In fiction all truths and falsehoods are known because they are given or can be logically deduced from givens. Anything else is unknowable and has an undefined truth value. Holmes is no more and no less than what Doyle says he is. I remain convinced, then, that the truth value of the proposition in question is undefined.

In summary, I believe that Sherlock Holmes does exist, but in a different way than Venus exists. The difference can be stated as follows: Venus exists physically, whereas ``Sherlock Holmes'' refers to a a non-physical, fictionally existent character. While there can be true and false statements about both Venus and Sherlock Holmes, there is an important distinction to be made between evaluating the truth values of propositions about physical and fictional entities.

The proposition ``Sherlock Holmes is a detective'' is true by definition; it is a fact given to us by Doyle about his fictional creation and we accept his statements as authoritative. In contrast, ``Venus is a planet'' is not true by definition. It is true because the physical object Venus has the property of being a planet. Similarly, ``Venus is a star'' is a proposition that is false. In contrast, ``Sherlock Holmes is a pirate'' is a proposition that is undefined within the context of Doyle's stories. Because it is not inconsistent, it is not false, nor is it true, because it is not a given. Instead, the truth value of the proposition is undefined.


Kim, Jaegwon and Ernest Sosa, Ed. A Companion to Metaphysics. Blackwell Publishers, 1999. 145-50,171-73, 361.

Parsons, Terence. Metaphysics: An Anthology. Ed. Jaegwon Kim and Ernest Sosa. Blackwell Publishers, 2000. 36-44.

Russell, Bertrand. Metaphysics: An Anthology. Ed. Jaegwon Kim and Ernest Sosa. Blackwell Publishers, 2000. 23-35.

Wall, Larry, Tom Christiansen, and Jon Orwant. Programming Perl, Third Edition. O'Reilly, 2000. 986.

File translated from TEX by TTH, version 2.51.
On 4 Mar 2002, 22:03.