Fictional Characters as Story-Free Denoting Concepts

Abstract

Some realist views about fiction are non-objectual in that they see fictional characters not as objects but rather as properties or the like; notably, kinds, roles, or denoting concepts. The view centered on denoting concepts proposed in Orilia’s “A Theory of Fictional Entities Based on Denoting Concepts” (2012) is presented in this paper with further motivations and details and in relation to issues not previously dealt with from its perspective. This view differs from other proposals of this sort, such as those by Cocchiarella and Landini, for its flexibility in allowing for story-free fictional characters: they can migrate from one story to another. This migration is granted in two ways, one that relies on the preservation of salient common features and another grounded on an appropriate causal connection between stories, typically involving authorial intentions. A more detailed account of this connection and of its interplay with the preservation of salient features is elaborated. Moreover, the phenomenon of fictionally non-existent characters (as in fiction within fiction) is addressed. Finally, the presence in fiction of historical, plural, indeterminate, and identity-inconsistent characters is examined and analyzed in terms of denoting concepts.

1. Introduction

We commonly distinguish realist and anti-realist accounts of fictional entities, or ficta, the characters of fictional stories such as A Study in Scarlet and The Adventures of Pinocchio, e.g., Sherlock Holmes, Pinocchio, and the speaking cricket, of which we intuitively say that they do not really exist. In realist accounts, notably Meinongian (Castañeda 1979; Parsons 1980; Zalta 1983; Priest 2016; Berto 2013), or artifactualist approaches (Van Inwagen 1977; Thomasson 1999; Voltolini 2006), there are such entities. In other words, fictional terms, e.g., singular terms such as “Sherlock Holmes”, “Pinocchio”, and “the speaking cricket”, do refer to items in the ontological inventory, even though they are not flesh-and-blood concrete denizens of the spatiotemporal world. In contrast, in anti-realist approaches (Walton 1990; Sainsbury 2005; Everett 2013), there are no such entities, and thus, “Sherlock Holmes”, “Pinocchio”, and “the speaking cricket” do not refer to any items in the ontological inventory.

The above-mentioned realist views are objectual, in that they see ficta as objects (particulars, individuals), i.e., things of which something can be predicated but that cannot be predicated of other entities; for instance, paradigmatically, ordinary objects such as this chair or that stone. Other realist views, however, are non-objectual, in that they see ficta as having a predicable nature, i.e., as properties or the like: kinds (Wolterstorff 1980), roles (Currie 1990), types (Bonomi 2008; Lamarque 2003; Terrone 2017), denoting (or referential) concepts (Cocchiarella 1982, 1996, 2007; Landini 1990; Orilia 2012), and individual concepts (Glavaničová 2021; Stokke 2021).1

In this paper, I would like to focus in particular on the non-objectual view based on denoting concepts proposed in Orilia 2012; I aim at filling in significant details that I had left implicit and at further substantiating the approach by considering how it can tackle some important issues that I had neglected but should be appropriately illuminated by a theory of ficta worth its name. For ease of exposition, I shall dub the theory with a specific name, FDC. This is an acronym derived from Flexible Denoting Concepts. The motives behind this pick will be clear in a moment.

FDC, like the other views based on denoting concepts, has been largely ignored.2 Nevertheless, it seems to me it is worth some attention for at least the following reasons. Like the other non-objectual views, it has the advantage of exploiting for its explanatory purposes entities that are already endorsed independently of fiction, namely properties (or the like); as we shall see, denoting concepts are properties of properties. FDC is, in other words, a form of identificationism, like the views that identify numbers with sets of sets or properties of properties (see Orilia 2012, p. 588): ficta are identified with entities already present in the ontological inventory; accordingly, the ontology is not conflated with the postulation of additional entities, placed in the ontological inventory just because fiction exists, as is most clearly the case in artifactualist approaches. Additionally, a view based on denoting concepts has these further notable advantages: the recourse to denoting concepts is especially well-motivated, since they can be seen, as explained below, as meanings of noun phrases in Montague grammar, which arguably provides the best approach to natural language semantics (Janssen and Zimmermann 2021). In particular, the denoting concepts with which ficta are identified are definite denoting concepts, i.e., denoting concepts that can work as meanings of definite descriptions such as “the tallest spy” or “the winged horse”.3 The identification of ficta with definite denoting concepts fits quite well with the fact that it is indeed with definite descriptions that we typically answer wh-questions concerning ficta, such as “who is Batman?” or “who is Pinocchio?” (Orilia 2012, p. 585).4 However, differently from the denoting-concept approach proposed by Cocchiarella and Landini, in which ficta are story-bound, FDC more flexibly allows for story-free ficta. These are ficta that can migrate from one story to another, such as, e.g., Odysseus, or Ulysses, taken to be both a character of Homer’s Odyssey and of Dante’s Divine Comedy. In contrast, a story-bound fictum cannot literally be a character of two different stories: at most, we can have in a story a counterpart of a character of another story; e.g., Dante’s Ulysses would only be a counterpart of Homer’s Odysseus. Indeed, FDC grants an extreme flexibility provided by two senses in which a fictum can migrate. One sense is based on the preservation of salient common features, pretty much as suggested in Wolterstorff’s kinds-based approach (1980, p. 149). The other sense is based on the existence of an appropriate causal connection between stories, typically involving authorial intentions. This second sense grants FDC the elbow room to take care of criticisms such as those leveled by Thomasson against Wolterstorff’s view, namely, in essence, that the preservation of features may well fail to ground the migration of ficta (see Thomasson 1999, pp. 59–60).

The rest of the paper is organized as follows. Section 2 offers a review of FDC that includes the leftover details; they regard the issue of fictionally non-existent characters and especially the nature of the causal connection between stories that allows for the just-mentioned second kind of migration of a fictum from one story to another. Section 3 concerns the previously neglected issues, namely historical, mythical, and legendary characters (Section 3.1), plural characters (Section 3.2), characters with indeterminate identity (Section 3.3), and characters with inconsistent identity (Section 3.4). Section 4 concludes by briefly summing up.

2. The Theory

I shall first lay down some background assumptions about fictional works and fictional stories by and large accepted in all current accounts of ficta (possibly with differences in details, which, however, are immaterial for present purposes). We shall then see how these assumptions are associated with a Platonist perspective, typical of non-objectual views, regarding stories and ficta (Section 2.1). Then, I shall clarify what denoting concepts are by insisting on their role as meanings of noun phrases in natural language semantics (Section 2.2). Next, I shall turn to how ficta come to be viewed as definite denoting concepts (Section 2.3). Finally, I shall dwell on how ficta are conceived of as story-free (Section 2.4).

2.1. Fictional Works and Fictional Stories

Authors or story-tellers in their creative activity produce fictional works such as literary texts (novels, tales, epics, poems), movies, comics, etc., which express fictional stories, e.g., Conan Doyle’s A Study in Scarlet and R. L. Stevenson’s The Strange case of Dr Jekyll and Mr Hyde (I shall use the same name for both the story and the work that expresses it, and I shall conveniently take advantage of abbreviated names such as SIS for A Study in Scarlet and SCJH for The Strange case of Dr Jekyll and Mr Hyde). A story is a proposition, typically a complex conjunctive proposition made up of many conjuncts. At least in the case of literary texts, these conjuncts correspond to the sentences that sequentially compose the text.5 Precisely which proposition is expressed by a certain work may well be a matter of interpretation, possibly involving an ideal reconstruction (see, e.g., Parsons 1980, p. 180).

There is a paratextual relation6 that links a proposition to a story and that we express with locutions such as “according to the story, …”, “in the story, …”, “it is the case, or true, in the story that …”, and the like. A proposition P is linked by this relation to a story S, when either (i) S explicitly asserts P, e.g., P is one of the conjuncts that make up S, or (ii) P can be inferred from S plus all the background assumptions of both a conceptual and encyclopedic nature, which we can legitimate on the basis of what we know about the historical circumstances of the production of the story (e.g., what was taken to be common knowledge, when the fictional work was produced, to what extent the story-teller intended to endorse this common knowledge, etc.) (see, e.g., Currie 1990). Thus, the paratextual relation is an entailment relation: it is the case in story S that P, or more formally, in(S, P), if, and only if, (B & S) entails P, where B is the conjunction of all the appropriate background assumptions. Thus, for example, according to SCJH, Mr Utterson is a lawyer, as this is explicitly asserted in the story, and also, Mr Utterson is a human being, and thus, e.g., has a liver and a heart, as such things are entailed via the background assumptions. The entailment relation in question, it should be noted, should be paraconsistent rather than classical, since there are inconsistent stories, and we do not want to say that every proposition whatsoever is true in them.7

As usual, propositions are taken to be possible sentence meanings, with a compositional complexity that reflects the complexity of the sentences that express them. That is, a proposition is taken to have constituents corresponding to the syntactic constituents of the sentences expressing them. It is worth illustrating this with a simple example that is most relevant for present purposes. The sentence “the president of Italy is Sicilian” has two syntactic constituents, namely the subject term “the president of Italy” and the predicate term “is Sicilian”, which are appropriately composed to generate the sentence. Correspondingly, the expressed proposition has two constituents appropriately composed (via a predication relation), and these constituents are the meanings of the two above-mentioned syntactic components of the sentence. The meaning of the predicate is assumed to be a property, Sicilian, and the meaning of the subject term is assumed to be a definite denoting concept, [the president of Italy]. For reasons that we shall see, the proposition expressed by the sentence in question is appropriately represented thus: [the president of Italy](Sicilian), where the constituent [the president of Italy] features as a property attributed to another property, namely Sicilian.

The notion of a constituent of a proposition offers a sense in which a denoting concept can be said to occur in a story, which will be useful in the following. We shall say that an item x occurs in a story S when x is a constituent of a proposition P such that in(S, P). Thus, for example, if S is a story according to which the president of Italy is Sicilian, that is, in(S, [the president of Italy](Sicilian)), then [the president of Italy] occurs in S, since [the president of Italy] is a constituent of the proposition [the president of Italy](Sicilian).

FDC espouses a Platonist perspective, according to which propositions exist independently of human activities, in particular the thinking activities of the story-tellers who produce fictional works. However, once such works are produced, and the corresponding propositions are thus thought of and entertained by the author and possibly by a more or less vast audience, these propositions become fictional stories and more precisely novels, tales, and so on. When this happens, we may say that the author creates a story, although, strictly speaking, the story already exists as a proposition in a mind-independent Platonic realm of abstract entities.8 Once a certain proposition becomes a fictional story, it is also the case that certain definite denoting concepts become fictional characters, and, in particular, fictional characters of the story in question, or equivalently in the story in question. Just like the story, such denoting concepts exist as properties independently of human activities, but in virtue of the fact that, through appropriate human activities, a certain proposition has acquired the status of a fictional story, the denoting concepts in question acquire the status of fictional characters, more specifically of characters of the story in question: the story generates the characters, as we may put it. When this happens, we may say that the author of the story has created the characters, although again, as in the case of stories, strictly speaking, the character already existed, as a denoting concept in the Platonic realm, ready to work as meaning of a definite description.

2.2. Denoting Concepts as Meanings of Noun Phrases

Denoting concepts are properties of properties that function as meanings of noun phrases such as “every cat”, “some dog”, “the tallest spy”, and the like (Cocchiarella 2007, sect. 2.3; Orilia 2010a, sect. 4).9 I represent these meanings thus: [every cat], [some dog], [the tallest spy], and similarly for other noun phrases of this sort. Here, dog, cat, tallest spy, etc., are the properties expressed by “dog”, “cat”, “tallest spy” and so on; let us call them the property components of the denoting concepts in question. Thus, e.g., tallest spy is the property component of the denoting concept [the tallest spy]. Note that the property component may be a conjunctive property, a property resulting from the conjunction of other properties, e.g., winged and horse, which results from the conjunction of the properties winged and horse. When the property component is conjunctive, the conjuncts are, let us say, internal properties of the denoting concept. For example, the denoting concept [the winged and horse] (the meaning of “the winged horse”) has the properties winged and horse as its internal properties. This terminology will be useful below.

Following Montague’s approach to natural language semantics (Janssen and Zimmermann 2021), to properly account for the compositionality of meanings, when these noun phrases occur in subject/predicate sentences such as “every cat is astute”, “some dog is pretty”, or “the tallest spy is Russian”, these sentences express propositions in which the property expressed by the predicate is the item of which the property of properties expressed by the noun phrase is predicated: [every cat](astute), [some dog](pretty), [the tallest spy](Russian). This works appropriately, given how these properties of properties are defined, namely, by exploiting the language of predicate logic enriched with the variable-binding lambda operator, “λ“, read as “being such that … “. By means of it, we can define new complex properties on the basis of already given properties. To pick a classical and simple example, the property of being a bachelor is defined as follows: [λx man(x) & adult(x) & ~∃y(married(x, y))], which can be read thus: being an x such that x is man and x is adult and it is false that there is a y such that x is married to y. Similarly, [every cat] is defined as [λf ∀x(cat(y) → f(x))], i.e., being an f such that, for every x, if x is a dog then x is f (i.e., property possessed by whatever is a cat); [some dog] is defined as [λf ∃x(dog(x) & f(x))], i.e., being an f such that there is an x such that x is dog and x is f (i.e., property possessed by something that is a dog); [the tallest spy] is defined as [λf ∃x(tallest spy (x) & ∀y(tallest spy(y) → x = y) & f(x))], i.e., being an f such that there is exactly an x such that x is tallest spy and x is f (i.e., property possessed by the unique thing which is tallest spy).

When properties are represented via the lambda operator, a logical principle called lambda conversion is assumed: [λx A(x)](t) ↔ A(t/x).10 This equivalence grants the expected connection between the complex property defined via the lambda operator and the properties that contribute to defining it. For example, by lambda conversion, [λx man(x) & adult(x) and ~∃y(married(x, y))](t) is equivalent to man(t) & adult(t) and ~∃y(married(t, y)), and this reflects the fact that if we say of someone, t, that he has the property of being a bachelor, we mean precisely that t is a man, that t is adult, and, finally, that t is not married. Analogously, to assert that every cat is astute, i.e., [every cat](astute), amounts to asserting ∀x(cat(x) → astute(x)), i.e., for every x, if x is a cat, then x is astute. Lambda conversion grants this, once [every cat] is defined as [λf ∀x(cat(y) → f(x))], for, by lambda conversion, [λf ∀x(cat(x) → f(x))](astute) is equivalent to ∀x(cat(x) → astute(x)). Similarly, lambda conversion grants that some dog is pretty, i.e., [some dog](pretty), is equivalent to ∃x(dog(x) & pretty(x)), that is, there is an x such that x is dog and x is pretty; and, finally, it grants that the tallest spy is Russian, i.e., [the tallest spy](Russian), is equivalent to ∃x(tallest spy(x) & ∀y(tallest spy(y) → x = y) & f(x)), that is, there is exactly one x that is tallest spy and x is Russian.11

FDC presupposes a form of descriptivism, according to which proper names are definite descriptions in disguise, and therefore they have definite denoting concepts as meanings (Orilia 2010a). Thus, for example, “Socrates” is short for the definite description “the Socrates” and thus expresses the definite denoting concept, [the Socrates], where “Socrates” expresses a property that is meant to individuate one object in particular; we may assume it is a property such as baptized as Socrates at a certain specific time in a certain specific place; such a property is nominal in that it entails the property of being called “Socrates” (Orilia 2010a, sect. 7.9; Orilia 2018).

A definite description may or may not denote, or refer to, one single entity; for instance, “the president of the US” refers to Donald Trump, whereas “the golden mountain” does not refer to anything. Analogously, we can say that the denoting concept expressed by a definite description also refers, or fails to refer, as the case may be, depending on whether or not the property component of the denoting concept is, or is not, uniquely exemplified: [the president of the US] refers to Donald Trump, since he is the unique individual who exemplifies the property of being president of the US. In contrast, [the golden mountain] does not refer to anything, since nothing exemplifies the property of being a golden mountain. Correspondingly, we say that the president of the US exists and that the golden mountain does not, which can then be understood as the claims that [the president of the US] refers to something and [the golden mountain] does not refer to anything, respectively. Let us say, in general, that a denoting concept that refers to something, e.g., [the president of the US], is referring, and that a denoting concept that does not refer to anything, e.g., [the golden mountain], is non-referring. More formally and in brief, we can represent these propositions as follows: E*([the president of the US]) and ~E*([the golden mountain]). The former is the proposition expressed by “the president of the US exists” or, equivalently, “the denoting concept [the president of the US] is referring;” the latter is the proposition expressed by “the winged horse does not exist” or, equivalently, “the denoting concept [the winged horse] is non-referring”.12

Furthermore, two definite descriptions may happen to be co-referring, i.e., they may refer to the same entity; for example, “the winner of the 2024 US presidential elections” and “the winner of the 2016 US presidential elections” both refer to Donald Trump. Correspondingly, we can also say that two denoting concepts may be co-referring. This happens when their property components are uniquely exemplified by the same entity. For instance, Donald Trump uniquely exemplifies the property of being winner of the 2024 US presidential elections and also uniquely exemplifies the property of being winner of the 2016 US presidential elections. Hence, we say that the winner of the 2024 US presidential elections is the winner of the 2016 US presidential elections, which can be understood as the claim that there is one single entity referred to by both denoting concepts, i.e., [the winner of the 2024 US presidential elections] and [the winner of the 2016 US presidential elections]. More formally and in brief, we can represent this proposition as follows: IS([the winner of the 2024 US presidential elections], [the winner of the 2016 US presidential elections]). This is the proposition expressed by the identity sentence “the winner of the 2024 US presidential elections is the winner of the 2016 US presidential elections”.13

These notions can be usefully relativized to stories. For example, according to SIS, Sherlock Holmes exists, and according to SCJH, Dr Jekyll exists. Moreover, according to SIS, Sherlock Holmes is the best friend of Watson, and according to SCJH, Dr Jekyll is Mr Hyde. Hence, we can say that the denoting concept [the Sherlock Homes] is referring in SIS: in(SIS, E*([the Sherlock Homes])). Similarly, the denoting concept [the Dr Jekyll] is referring in SCJH: in(SIS, E*([the Dr Jekyll])). And, furthermore, the denoting concepts [the Sherlock Homes] and [the best friend of Watson] are co-referring in SIS, and the denoting concepts [the Dr Jekyll] and [the Mr Hyde] are co-referring in SCJH: in(SIS, IS([the Sherlock Homes], [the best friend of Watson]) and in(SCJH, [the Dr Jekyll], [the Mr Hyde]).

2.3. Ficta as Definite Denoting Concepts

There are usually various definite denoting concepts that are referring in a given story, and surely some of them are co-referring in it, as the above examples indicate. Let us look at all the denoting concepts that are referring in a certain story as ideally subdivided into distinct sets such that each set contains precisely all the denoting concepts that are co-referring, according to the story; let us call them character sets (Orilia 2012, p. 583). For example, for SIS there will be a character set containing [the detective called Sherlock Holmes] as well as any other denoting concept that is referring in SIS and co-referring in SIS with [the detective called Sherlock Holmes], e.g., [the best friend of Watson] and [the detective who lives in 221B Baker Street in London]; that is, a character set such as this: {[the detective called Sherlock Holmes], [the best friend of Watson], [the detective who lives in 221B Baker Street in London], …}. Among the denoting concepts in any character set, one is maximal (Orilia 2012, p. 584) in the sense that its property component puts together all that is said in the story; the property component of this maximal denoting concept can be seen as a vast conjunctive property that recapitulates the whole story. For example, according to SCJH, there was a Mr Utterson who is a lawyer and who is a man and … who is dreary and who is somehow lovable and … who is such that London was startled by a crime of singular ferocity … and who is such that a maid-servant living alone in a house from the river had gone to bed upstairs about eleven, and …14 There is then a corresponding conjunctive property that somehow recapitulates the whole story, namely: mister and Utterson and lawyer and … dreary and somehow lovable and … such that London was startled by a crime of singular ferocity … and such that a maid-servant living alone in a house from the river had gone to bed upstairs about eleven and … This is the property component of a maximal denoting concept that is referring in SCJH: [the mister and Utterson and lawyer and …]. We may conveniently call this the Utterson maximal denoting concept.

According to FDC, given an appropriate interpretation of the story, some of the internal properties of a maximal denoting concept are salient properties. They are “properties that can be somehow identified as ‘essential.’ Intuitively, they are the properties that truly characterize a given character” (Orilia 2012, p. 584). Consider the denoting concept that has all and only these salient properties as internal properties, which we may call a salient denoting concept; it is assumed to be co-referring with the maximal denoting concept in question so that they are both in the same character set. In this character set, it is, we may say, a privileged member, as the story in question makes it salient (given a certain interpretation). FDC takes fictional characters to be denoting concepts of this sort, i.e., salient denoting concepts (Orilia 2012, p 584).15 In particular, they are fictional characters of the stories that make them salient (and possibly of other stories, as we shall see). For example, as regards the Utterson maximal denoting concept, assume for convenience that only these properties are salient: mister, Utterson, lawyer, man, dreary, somewhat lovable. Then, the denoting concept [the mister and Utterson and lawyer and man and dreary and somewhat lovable] is a denoting concept made salient by SCJH, and thus FDC counts it as a fictum, and in particular, a fictum of SCJH, its Utterson character, we may say. Its internal properties are its salient features. In general, the internal properties of a fictum, as so understood, are the salient features of the fictum in question.

What we have seen so far are ficta that, according to a story, exist; for example, according to SIS, Sherlock Holmes exists, and according to The Adventures of Pinocchio, the speaking cricket exists. However, an important addition should be made, since there are also stories that seem to speak about entities that, according to such stories, do not exist.16 For example, according to Macbeth, Macbeth believes that there is Banquo’s phantom sitting at his place at the banquet’s table, although nobody at the banquet but him sees this phantom. Moreover, according to Macbeth, Macbeth sees a dagger before himself, with the handle toward himself, of which he suspects that it is only “a dagger of the mind, a false creation, proceeding from the heat-oppressed brain”. Many other examples are provided by cases of fiction within fiction, that is, stories in which a character makes up a story which in turn has its own characters. For instance, in Boccaccio’s Decameron, there are various narrators who tell stories, e.g., Filomena, who narrates a story, according to which Elisabetta da Messina falls in love with Lorenzo, a boy from Pisa.

Clearly, according to Macbeth, neither Banquo’s phantom nor the dagger seen by Macbeth exist. Similarly, according to Decameron, Elisabetta da Messina and Lorenzo do not exist. Yet, we may say that they are characters, whether of Macbeth or Decameron. They are fictionally non-existent ficta, we may say, in contrast with fictionally existent ficta such as Sherlock Holmes and Pinocchio. From the point of view of FDC, fictionally non-existent ficta are denoting concepts that are made salient by the relevant stories, just like fictionally existent ficta. They are generated by the relevant story just like the latter, except that in their case the character set is made up of denoting concepts that are co-referring from the perspective of a certain mental attitude of a character or from the perspective of a story within the story. For example, in Macbeth, the denoting concepts [the phantom of Banquo] and [the one sitting at Macbeth’s seat at the banquet table] are co-referring, according to Macbeth’s belief. Similarly, in Decameron, the denoting concepts [the Lorenzo] and [the lover of Elisabetta] are co-referring, according to Filomena’s tale. And thus, let us suppose, Macbeth generates a salient denoting concept such as [the phantom of Banquo and sitting at Macbeth’s desk and …], and Decameron generates a salient denoting concept such as [the Lorenzo and lover of Elisabetta and …], which count as characters of Macbeth and Decameron, respectively, albeit fictionally non-existent characters of such stories. To fix a convenient terminology, we may say that the denoting concepts in these examples involving fictionally non-existent ficta are relatively referring according to the stories in question, as they are referring according to mental attitudes or stories within the stories.

There are two important points worth being remarked upon, which regard the relation between ficta as so understood and the fictional terms that we typically use as seemingly referring to ficta.

Here is the first point. We may expect that among the salient features of ficta there are the properties mobilized by the fictional terms typically used to refer to such ficta. For example, we may expect that the nominal property Sherlock Holmes is one of the salient features of the Sherlock Holmes character of SIS referred to by “Sherlock Holmes”, and that speaking and cricket are among the salient features of the speaking cricket character of The Adventures of Pinocchio referred to by “the speaking cricket”. However, this need not be necessarily assumed. For example, we might view the property of being called Sherlock Holmes as not essential to the Sherlock Holmes character of SIS. Be this as it may, surely the denoting concept expressed by the fictional term is, according to the story in question, co-referring with the salient denoting concept which is the fictum. Thus, for example, if the Sherlock Holmes character of SIS is the denoting concept [the clever and detective and …], then, according to SIS, this denoting concept is co-referring with the denoting concept [the Sherlock Holmes], expressed by the fictional term “Sherlock Holmes”.

The second point has to do with fictional terms that can be used to formulate identity sentences that are true according to the relevant story. The paradigmatic case is offered by “Dr Jekyll” and “Mr Hyde”, as we use them to claim that, according to SCJH, Dr Jekyll is Mr Hyde. In such cases, FDC grants that there is one single character, since the denoting concepts expressed by the two singular terms are co-referring according to the relevant story and thus belong in one single character set. In such a set, there is a certain salient denoting concept, which counts as the single fictum that is referred to by both singular terms. Thus, for example, [the Dr Jekyll] and [the Mr Hyde] are co-referring, according to SCJH; hence, they belong in one character set. This set includes as a member a salient denoting concept which is a single Dr Jekyll/Mr Hyde character, referred to by both “Dr Jekyll” and “Mr Hyde”, or, more precisely by “the Dr Jekyll of SCJH” and “the Mr Hyde of SCJH” (Orilia 2012, p. 584).

2.4. Ficta as Story-Free

Ficta, as conceived by FDC, are story-free: they are capable of migrating from one story to another, or, in other words, can be present in more than one story. FDC acknowledges two senses in which a fictum can be present in a story. The primary sense is simply the fact that the fictum in question is a denoting concept that is referring, or at least relatively referring, in the story. For example, as we saw, the denoting concept [the mister and Utterson and lawyer and man and dreary and somewhat lovable] is a fictum of SCJH, and this denoting concept is referring, according to SCJH; or, as we also saw, [the phantom of Banquo and sitting at Macbeth’s desk and …] is a fictum of Macbeth, and this denoting concept is referring, according to Macbeth’s belief in Macbeth. Hence, in this primary sense, these ficta are present in SCJH and in Macbeth, respectively (Orilia 2012, p. 586). A fictum that is primarily present in a story may well happen to be (relatively) referring and primarily present in another story. For example, imagine that a story-teller, independently of Doyle, produces a story, SIS*, very similar to SIS. In such a case, SIS and SIS* make salient, we may assume, the same Sherlock Holmes character, a certain denoting concept that is referring in both stories (Orilia 2012, p. 587).

Let us now move to the second sense in which a fictum may be present in a story. FDC acknowledges that in producing a new story an author may have in mind a character of a previous story in such a way that this author somehow makes that character also a character of the new story. When this happens, the character of the previous story is present in the second sense at issue, secondarily present, let us say (Orilia 2012, p. 587). Accordingly, we do not have the creation of a new character in this case, but rather the migration of an old character into a new story. As we shall see, this secondary presence may or may not be accompanied by a primary presence of the old character in the new story. In any case, for this secondary presence to take place, there must be a special relation of authorial character connection, as we may call it, involving a previous story S, a denoting concept DC occurring in S, a new story S’, and a denoting concept DC’ occurring in S’: the occurrence of DC’ in S’ is character-connected by the author of S’ to the occurrence of DC in S. I shall clarify with some examples when this relation happens to subsist and why this subsistence makes it the case that a character of a previous story becomes also a character of a new story.

First of all, there are the most typical cases, provided by series of stories, such as the Sherlock Holmes series by Conan Doyle; here, the author of the old story is the author of the new story, and the denoting concept occurring in the old story is the denoting concept occurring in the new story. For example, there is this relation of authorial character connection between SIS, the denoting concept [the Sherlock Holmes], occurring in SIS, The Hound of the Baskervilles (HB), and again [the Sherlock Holmes], which is also occurring in HB: the occurrence of [the SH] in HB is character-connected by the author of HB, namely Conan Doyle, to the occurrence of [the Sherlock Holmes] in SIS, also authored by Conan Doyle. This happens because, in producing HB, (i) Conan Doyle uses the very same singular term, “Sherlock Holmes”, which he had used in producing SIS, and (ii) he clearly does this with the intention of making the Sherlock Holmes character of SIS also a character of HB. In cases of this sort, the character in question can be assumed to also be primarily present in the new story. Thus, for instance, the Sherlock Holmes character, i.e., (let us assume) the denoting concept [the clever and detective and …], made salient by SIS, is referring in the new story just as it is referring in SIS; it is co-referring with [the Sherlock Holmes] in both stories.

Next, there are many cases that differ from those just considered, only because the new story has a new author. For instance, Nicholas Meyer uses “Sherlock Holmes” in writing the novel The Seven Percent Solution, with the intention of making the Sherlock Holmes character of SIS also a character of his story, thereby providing a story in which (i) the denoting concept [the Sherlock Holmes] occurs, and (ii) this denoting concept is co-referring with the denoting concept made salient by SIS, namely, [the clever and detective and …].

Furthermore, there are cases in which the denoting concept occurring in the old story is different from the one occurring in the new story, but nevertheless, the two concepts are sufficiently connected to grant the secondary presence of a character of the old story in the new story. An example is offered by the denoting concept [the Odysseus], which occurs in the Odyssey, and the denoting concept [the Ulysses] occurring in Alfred Tennyson’s Ulysses. In this case, Tennyson does not use the same singular term used by Homer. Nevertheless, there is a historical connection between the two names and Tennyson appears to have the intention of making the Odysseus character of the Odyssey also a character of Ulysses, even though in the latter, the character is attributed Ulysses rather than Odysseus. But even so, this character has the same salient features and thus is also primarily present in Ulysses. That is, we may assume, there is a salient denoting concept, say [the Achaean and valiant and warrior and astute and …], made salient by Odyssey, co-referring with [the Odysseus] in the Odyssey, which is co-referring with [the Ulysses] in Ulysses.

So far, we have seen cases in which the character of the old story is secondarily present in the new story and also primarily present in the new story. We have secondary presence without primary presence when there is authorial character connection but no preservation of salient features. This happens paradigmatically with a story whose author has the explicit intention of connecting to a previous story, and in particular to a character of such a story, so as to make it also a character of the new story, while, however, failing to attribute to such a character in the new story some of its salient features. An example could be Thom Eberhardt’s movie Without a Clue (WAC). This is a comedy featuring a rather stupid Holmes, who only seems a clever detective because Watson secretly grants him the solution when he is called to solve a crime. In this case, in producing Without a Clue, Eberhardt uses the term “Sherlock Holmes” just like Doyle and thus the denoting concept [the Sherlock Holmes] occurs in both SIS and WAC. However, in SIS, [the Sherlock Holmes] is co-referring with the denoting concept [the clever and detective and …], made salient by SIS, whereas such a denoting concept is not co-referring with [the Sherlock Holmes] in WAC. Thus, the SH of SIS is not primarily present in WAC (Orilia 2012, p. 587).

I have assumed that in Thom Eberhardt’s case, there is conscious authorial intention to make a previous character also a character of a new story. But perhaps there can be an authorial character connection that generates a secondary presence without primary presence, even without conscious authorial intention. Let us imagine the following case.17 Tom intends to memorize the text of Doyle’s A Study in Scarlet and then recite it. Unfortunately, his memorization attempt is not fully successful, and he gives voice to a text only similar to Doyle’s text. As a result of the changes in the text, we hear of a Sherlock Holmes with many salient features that differ from the ones of the original Holmes. Tom, we should admit, has inadvertently created a new story, say A Study in Scarlet II, which arguably still has the Holmes of SIS as a character: the new story is such that the occurrence of [the Sherlock Holmes] in it is character-connected by Tom to the occurrence of [the Sherlock Holmes] in SIS. However, Tom did not have the intention of transferring a previous character to a new story, as he was not even aware of making up a new story.

3. Some Problems of Fiction

FDC has been already appealed to in accounting for the most crucial cases involving the identity of ficta across different stories, analogous to those we have considered above, as well as others with an alleged fission or fusion of characters (Orilia 2012, sect. 9). Here, I consider additional problems of central importance in testing a theory of ficta.

3.1. Historical, Mythical and Legendary Characters

A crucial question about fiction is whether real objects can be characters of a story; non-fictional characters, we may say, of which we intuitively say that they in fact exist or existed, e.g., London in SIS or Napoleon in War and Peace. As these examples suggest, the issue arises most specifically with singular terms, especially proper names, that we normally take to be referring in their everyday use outside of fiction and that are also used in producing a story. “London” is a paradigmatic example; we normally take it to refer to a big city in the UK, and it was also used by Doyle in realizing the text of SIS.18 Is then this very city a non-fictional character of SIS? Or is it rather the case that, among the fictional characters of SIS (in a realist perspective about ficta), there is also a fictional London of the same nature as Holmes and Watson? There seem to be conflicting intuitions about this. For example, Castañeda (1979) has forcefully argued for the former option, whereas Bonomi (1994) has defended the latter. Maybe it is appropriate to pursue both ways. Let us see how it can be done from the perspective of FDC.

On the one hand, the intuition that real objects can be characters of stories can be pursued in FDC on analogy with the treatment of the secondary presence of old characters in new stories. That is, we can assume that the author’s use of a singular term that in everyday use refers to a real object can establish a significant connection to previous uses of this term; a connection that makes it the case that the real object at issue becomes a character of the story. We know that the name “London” refers to London, a certain city, a real entity, the real London, we may say. As that name occurs in SIS, we presumably assume by default that even then the name refers to that very city, which can then be considered a character of SIS, albeit not a fictional character. This is substantiated by the fact that we know that Doyle knew that “London” referred to London and presumably intended to use that name precisely to refer to London and think of it as the place where the characters that he invented live and act. Thus, we may say, there must be a relation of authorial character connection in which, instead of a previous story, there is as a relatum some proposition typically expressed in ordinary discourse by recourse to the singular term in question. For example, the proposition that we express by saying: “Hyde Park is the largest green area in London”. Let us call this relatum the anchoring proposition. The other terms in the relation would be the story, its author, and the denoting concept expressed by the singular term, first as occurring in the anchoring proposition, and then as occurring in the story. Thus, for instance, the occurrence of [the London] in SIS is character-connected by Doyle to the occurrence of [the London] in some anchoring proposition. This makes it the case that the real object to which [the London] refers is a character of SIS, one that has a secondary presence in SIS. London can then be considered a character of the story but not a fictional character of the story: rather, we may say, a historical character. The story is, inter alia, about the real object.

We could say that this secondary presence of a real object in a story precludes the generation of a fictum. But we could also think that the alternative intuition should not be ruled out and that the story also generates a fictum, namely a certain salient denoting concept, which we may call a historical fictum. If so, for example, “the London of SIS” is ambiguous, as it may have either a historical or a fictional interpretation. With the former, it denotes the real London; with the latter, it denotes a fictum, say [the London*], where London* is a conjunction of salient features. Given the historical interpretation, it is true that

(1)

The London of SIS is the real London.

However, given the fictional interpretation, (1) is false.

Myths and legends can be dealt with similarly. We take for granted that Apollo does not exist, and so “Apollo” is non-referring. However, Ancient Greeks believed otherwise, and we may assume that in Greek myths, or in stories such as the Iliad, that name is used by story-tellers who took it to refer to a real entity. For this reason, e.g., “Apollo” was used by them just like Doyle used “London”. That is, there was a relation of authorial character-connection involving an anchoring proposition and a denoting concept, say [the Apollo]. Nevertheless, there is of course, in this case, no real entity that is a character of their stories, since “Apollo” is not in fact referring. However, as in the case of “London”, we may still assume that the stories generate a fictum, a certain denoting concept, say [the Apollo*], where Apollo* is a conjunction of salient features; we may call it a mythical fictum. Thus, e.g., “the Apollo of the Iliad” is ambiguous just like “the London of SIS”. In contrast with the latter, however, it denotes nothing in its historical interpretation. On the other hand, in its fictional interpretation, it denotes the fictum [the Apollo*]. Given the latter interpretation, we may assume, for example, that this is true:

(2)

The Apollo of the Iliad is the Apollo of Madeleine Miller’s The Song of Achilles.

Here, the assumption is that [the Apollo*] has a primary presence in the Greek myths and a secondary presence in Madeleine Miller’s novel (and presumably also a primary presence).

Consider legends now. In their cases, we do not know whether they speak of real entities or not. For example, we do not know whether Robin Hood really existed or not. Let us assume that in the first Robin Hood story, apparently Piers Plowman by William Langfold, the author takes the name “Robin Hood” to be referring to a real man. In this sense, the name is used just like “London” by Doyle. That is, there was a relation of authorial character connection involving an anchoring proposition and a denoting concept, say [the Robin Hood]. However, we do not know whether there is a real entity that is a character of Piers Plowman (PP). In any case, we may still assume that the story generates a fictum, a certain denoting concept [the Robin Hood*] (where Robin Hood* is a conjunction of salient features); we may call it a legendary fictum.19 Thus, again “the Robin Hood of PP” is ambiguous; in its historical sense, it may denote a real man or it may denote nothing; in its fictional sense, it denotes the fictum [the Robin Hood*]. Given the latter interpretation, we may assume, for example, that this is true:

(3)

The Robin Hood of PP is the same as the Robin Hood of Sir Walter Scott’s Ivanhoe.

Here, the assumption is that [the Robin Hood*] has a primary presence in PP and a secondary presence in Ivanhoe (and presumably also a primary presence).

3.2. Plural Characters

Stories do not only speak of single individuals, they also speak of groups of individuals, like families, teams, armies, etc., i.e., entities with components or members: in the language of mereology, we may say they are mereological sums of individuals. It is thus reasonable to admit that stories generate plural characters, as we may call them (see, e.g., Parsons 1980, pp. 190–94; Schnieder and von Solodkoff (2009, pp. 146–47)). From the perspective of FDC, a plural character, just like other characters, can be taken to be a definite denoting concept. Its being plural consists in the fact that its property component is a property that applies to a group. Examples of properties of this sort are Kennedy family, Iowa basketball team, British army, Italian Parliament, etc., or even properties expressed by plural nouns such as policemen who arrested Tess, to use an example that will come up below. In contrast, an individual character is a denoting concept such that its property component is a property that applies to individuals, such as detective who lives in London.

Anti-realists about fiction have argued that plural characters constitute a problem for the realist. In essence, the problem is this. Surely for there to be a group, there must be the individual members of the group. On the basis of this correct principle, the anti-realist charges that in facing a story that talks about a fictional group, the realist should acknowledge as fictional characters of the story the individual members of the group, and not only the group. However, often and typically, the realist is in no position to do it, because the story does not characterize and differentiate the members of the groups. For example, Everett (2013, p. 191) considers Thomas Hardy’s Tess of the d’Urbervilles (TU), which tells us that 16 policemen arrested Tess, without giving us any further information about such policemen. And then he claims that the realist should admit that Hardy created 16 fictional policemen, and not only a fictional group, namely the group of policemen who arrested Tess. However, urges Everett, for there to be such 16 distinct characters, there should be features that differentiate them, but there are no such features, for they are characters of a certain story and thus the differentiating features could only be found in that story, which, unfortunately, is silent about them. One could add that the problem is compounded when the number of members of the group is not specified, which typically happens in fiction. For example, in Tolkien’s The Lord of the Rings, one reads of Sauron’s army made up of orcs, and not only is nothing said about the individual members of the army, but nothing is said about how many there are. Thus, following Everett’s line, the realist should be trapped in this contradiction: on the one hand, she should admit that there is a definite number of fictional orcs created by Tolkien, since any group has a definite number of members; but, on the other hand, she should say that there is not such a number, since the story is silent about it.

However, there is no reason why the realist should admit that, once a story generates a fictional plural character, it also eo ipso generates fictional members of such characters. Let us look at the issue from the point of view of FDC. There is surely a denoting concept, [the Sauron’s Army], which is referring, according to Lord of the Rings (LOR), and which is co-referring with other denoting concepts, so that the story generates as a character a salient denoting concept, say [the army and led by Sauron and made of orcs and …], which “the Sauron’s army of LOR” can be taken to denote. Nevertheless, there is no denoting concept such as [the component of Sauron’s army who is such and such], which is referring, according to LOR. However, this is what we would need for this story to generate as one of its characters a fictional member of Sauron’s army. Similarly, according to Hardy’s TU, there is a referring denoting concept such as [the policemen who arrested Tess] and other co-referring denoting concepts, wherefrom a salient denoting concept working as a plural character is generated, but there are no denoting concepts that are referring according to TU and that could generate fictional members of the group of policemen.

Surely, we should grant that, according to LOR, there are individual orcs, distinct from one another, who are members of Sauron’s army. This follows from the general principle invoked by Everett and the fact that, according to LOR, Sauron has an army of orcs. Similarly, according to TU, there are exactly 16 distinct policemen who collectively arrested Tess. Analogously, since any human being has a definite weight and, according to SIS, Holmes is a human being, it follows that, according to SIS, there is a weight which is Holmes’s weight. But it does not follow that there is a certain weight, w, such that, according to SIS, w is Holmes’s weight. Similarly, even though LOR and TU have plural characters such as an army of orcs or a group of policemen, it does not follow that there are corresponding individual orc or policeman characters.

3.3. Indeterminate Identity of Characters

According to Everett (2005, 2013), there are fictional stories that convey cases of indeterminate identity, and they are problematic for realists about ficta, since they force them to admit ontic indeterminacy; indeterminacy in the world,20 not merely within a fiction, even if involving fictional objects. Everett argues for this on the basis of a plausible principle regarding ficta, which goes as follows (in my terminology):

(EP)	Ficta A and B of story S are identical if, and only if, according to S, A is B.

For example, as an instance of the principle, we get that the ficta Mr Hyde and Dr Jekyll of SCJH are identical if, and only if, according to SCJH, Dr Jekyll is Mr Hyde.21 This is indeed a plausible principle, but, as we shall see, what Everett actually uses are variants of (EP), which are not as plausible. Everett provides two examples of stories involving indeterminate identity. Let us consider them in turn.

The first example is Vladimir Nabokov’s Pale Fire (PF) (Everett 2013, p. 209). According to it, there is a poem written by the poet John Shade and presented with a critical apparatus written by Shade’s academic colleague Charles Kinbote; this looks more and more like an unreliable narrator so that readers end up suspecting that Kinbote and Shade are in fact the same person. The story depicts a world that is like our ordinary world in that people seem to have a determinate identity but does not determine whether Shade is the same person as Kinbote or not. In sum,

(1)	It is indeterminate that, according to PF, Shade is the same as Kinbote.

Although Everett is not quite explicit on this, a claim such as (1) is to be understood as follows:

(1a)	It is not the case that, according to PF, Shade is the same as Kinbote, and it is also not the case that, according to the story, Shade is not the same as Kinbote.22

Everett classifies this as Type-A indeterminacy. As he sees it, when there is this indeterminacy, there is ontic indeterminacy for the realist, since the realist should then admit this:

(2)	It is indeterminate that the Shade of PF is the same as the Kinbote of PF.

To argue for this, Everett really uses this variant of (EP):

(EP′)

It is indeterminate that ficta A and B of story S are identical if, and only if, it is indeterminate that, according to S, A is B.

However, it is not obvious that we should accept (EP’). Let us see how FDC can assess this matter. Even if the identity of Shade and Kinbote is unsettled within the story, this does not prevent us from individuating two distinct salient denoting concepts, say, [the Shade*] and [the Kinbote*], where Shade* and Kinbote* are two distinct conjunctions of salient features. These are then two distinct characters generated by the story; they are, respectively, the ficta Shade and Kinbote of PF. Hence, (EP’) and (2) should be considered false: Shade and Kinbote are distinct characters, rather than examples of ontic indeterminacy. Surely, if we accept (1a), we should also grant both that (i) it is not the case that, according to PF, [the Kinbote] and [the Shade] are co-referring, and (ii) it is not the case that, according to PF, they are not co-referring. That is:

(3)	~in(PF, IS([the Kinbote], [the Shade]) & ~In(PF, ~IS([the Kinbote], [the Shade]).

However, the second conjunct of (3) cannot hinder the fact that the two denoting concepts, [the Kinbote*] and [the Shade*], are distinct; this distinctness is indeed supported by the first conjunct of (3).

The other example is Haruki Murakami’s The Wind-Up Bird Chronicle (Everett 2013, p. 210) (WBC). It features Toru and his wife Kumiko as protagonists and also involves a woman in a hotel room. There are several episodes that look like distorted nightmares with no fact of the matter as to what is exactly going on, so the story depicts a world very different from our ordinary world, with ontic indeterminacy, one may say. In particular, let us concede, according to the story, it is indeterminate that Kumiko is the same person as the woman in the hotel room:

(4)	According to WBC, it is indeterminate that Kumiko is the same as the woman in the hotel room.

In Everett’s terminology, this is Type-B indeterminacy. As he sees it, when there is this indeterminacy, again there is ontic indeterminacy for the realist, since the realist should admit that:

(4a)	It is indeterminate that Kumiko is the same as the woman in the hotel room.

To argue for this, Everett really appeals to this other variant of (EP):

(EP″)

It is indeterminate that ficta A and B of story S are identical if, and only if, according to S, it is indeterminate that A is B.

However, again, this variant is not as plausible as the original. Let us look at the matter through the lens of FDC. We are supposed to grant (4a) as a result of Murakami’s creative activities, which succeed in conveying a strange world with indeterminacy. However, from the perspective of FDC, this means that, according to the story in question, it is indeterminate that two denoting concepts, [the Kumiko] and [the woman in the hotel room], are co-referring. Thus, more formally (using, like Everett, the symbol “∇” for “indeterminately”), the claim is:

(4b)	in(WBC, ∇(IS([the Kumiko], [the woman in the hotel room]).

Now, even if (4b) is true, it is a truth that involves two distinct denoting concepts, which presumably contribute to two distinct salient denoting concepts such as [the Kumiko*] and [the Woman in the hotel room*], where Kumiko* and woman in the hotel room* are two different conjunctions of salient features. And these two salient denoting concepts are the two ficta in question, namely, ficta Kumiko and the woman in the hotel room of WBC. Thus, (EP’’) should not be accepted and (4a) should be reckoned as false. It is rather the case that the two ficta in question are definitely distinct. The ontic indeterminacy of Murakami’s story, if there really is such a thing, does not result in indeterminacy in the real world.23

3.4. Inconsistent Identity of Characters

Everett (2005) has also argued that another problem arises for realists about ficta, in view of stories involving characters with an inconsistent identity; that is, stories according to which there are characters A and B such that it is both the case and not the case that A is identical to B. The problem is that the realist should admit both that A is identical to B and also that A is not identical to B, thereby acknowledging a contradiction in reality and not just in the relevant story. Everett (2005, p. 214, n. 11) has concocted a story of his own, Dialethialand, according to which two characters, Jules and Jim, are both identical to and distinct from each other. But there is in fact a real and sophisticated piece of literature that appears to convey a story involving an inconsistent identity, namely Un drame bien parisien by Alphonse Allais (UDBP).24 The story features two characters, Raul and the Templar, such that it seems to be the case both that Raul is the same as the Templar and also that they are distinct. Accordingly, the following paratextual statement seems to be true:

(1)	According to UDBP, the Templar is the same as Raul and the Templar is not the same as Raul.

Everett would then claim that the following contradiction is true:

(1a)	The Templar of UDBP is the same as the Raul of UDBP and the Templar of UDBP is not the same as the Raul of UDBP.

Everett thinks he can get this result by appealing again to his principle (EP). As a matter of fact, given that, according to UDBP, Raul is the same as the Templar, it follows, by that principle, that the Templar of UDBP is the same as the Raul of UDBP. However, that principle does not allow us to infer that the Templar of UDBP is not the same as the Raul of UDBP from the fact that, according to UDBP, Raul is not the same as the Templar. To make this inference we need this further version of (EP):

(EP‴)

Ficta A and B of story S are not identical if, and only if, according to S, A is not identical to B.

This is what Everett really appeals to. However, at least from the perspective of FDC, this is not plausible. Let us see why.

Presumably, since, according to the story, the Templar is the same as Raul, the story generates a single Templar/Raul character, a certain salient denoting concept, which is internally both Raul and Templar. Hence, it is true that the Templar of UDBP is the Raul of UBDP, i.e., the first conjunct of (1a) is true, and, accordingly, the second conjunct is false. In the light of (EP‴), it may be thought that this second conjunct should rather be considered true as well, thereby generating a contradiction, for it is also the case that, according to the story, the Templar is not the same as Raul. However, this is only the claim that, according to the story, the denoting concepts [the Raul] and [the Templar] are not co-referring, and these two denoting concepts are not the characters in question. As noted, there is just one character which is the Templar/Raul character (presumably inconsistent, in that it has internally (i) the property of being identical to the Templar and also the property of not being identical to the Templar, as well as (ii) the property of being identical to Raul and also the property of not being identical to Raul).25

4. Conclusions

The theory of fictional entities presented in Orilia (2012) is a non-objectual and identificationist approach to ficta that has specific advantages in terms of ontological economy, since it identifies ficta with entities, definite denoting concepts, that are very well motivated from the point of view of natural language semantics. Many details of that approach, however, had been left implicit, and crucial problems for a theory of ficta had not been treated. Such details have been developed here and these problems have been addressed. In particular, an account of the causal connection between stories that can grant the migration of a fictum from one story to another has been provided. Moreover, the phenomenon of fictionally non-existent characters has been addressed. Finally, the presence in fiction of historical, plural, indeterminate, and identity-inconsistent characters has been examined and analyzed in terms of denoting concepts. Hopefully, all this may contribute to rendering this theory an alternative worth some attention in the current debate on the nature of fictional entities.26