2. A Bit More Stage-Setting

Our thesis, if you like, is that, while Vulcan exists, Vulcan and Vulcan do not.Footnote ⁶ But it will probably make things a bit less confusing if we switch to using non-homonymous running examples. So, for the remainder, we’ll follow much of the literature in using Arthur Conan Doyle’s fictional detective, Sherlock Holmes, as our main example of a creature of fiction, while retaining “Le Verrier’s Vulcan” as our main example of a purported creature of failed scientific theorizing. And, for reasons that will become apparent, we will use the name “Thor,” on its “mythical” usage, where it purports to refer to a god of thunder, as our main example of a creature of myth. But in a more colloquial, and “non-squinty” sense, if you will.Footnote ⁷

We turn next to the arguments for Ecumenical Realism.Footnote ⁸ The first, having mainly to do with the relationship between belief and make-belief, is from Nathan Salmón (by way of Ben Caplan). The second, from David Braun (again by way of Caplan), appeals to the relationship between elements of fictional storytelling and elements of mistaken theorizing. The third is inspired by a view of the metaphysics of stories presented in Sam Cowling and Ley Cray’s Philosophy of Comics: An Introduction (Cowling & Cray, Reference Cowling and Cray2022), as well as the feud between Stan Lee and Steve Ditko over who created Spider-Man.Footnote ⁹ We maintain that these arguments for Ecumenical Realism do not make for a strong case, individually or collectively. Finally, we consider what we see as an important threat to our preferred view.

3. Arguing for Ecumenical Realism: Nathan Salmón’s and Ben Caplan’s Attitudes

In “Mythical Objects,” Salmón argues for Ecumenical Realism:Footnote ¹⁰

Here, Salmón alleges a close parallel between fiction and myth and suggests that the onus is on the fictional realist who does not also countenance creatures of myth, like us, to point to a salient difference. Caplan (Reference Caplan2004, pp. 333–334) also reads Salmón this way and presents Salmón’s argument as follows:

In support of the crucial premise (P2), Caplan says:

Caplan argues that, if (P2) is false, then there is an ontologically significant difference between attitudes of belief and make-belief. But there is not. So, if authors create creatures of fiction, then myth-makers create creatures of myth. But that is just (P2).

We deny (P2), and we deny that there is no ontologically significant difference between belief and make-belief in these sorts of cases. Consider again Doyle and Le Verrier. We think it is plausible to suppose that Doyle’s acts of make-belief were creative in a way that Le Verrier’s were not. That is, Doyle’s acts of make-belief involve creative intentions, which play an important role in bringing works of fictions (and their creatures) about.Footnote ¹¹ In central cases of myth-making, and particularly the cases of failed scientific theorizing, the right kinds of creative intentions on the part of theorizers who believe their theories seem to be completely lacking. Footnote ¹² This difference, we maintain, is ontologically significant.

We think there are further related, ontologically significant differences between attitudes of belief and make-belief. First, consider belief. Doxastic voluntarism, the view that we can form beliefs “at will,” is widely regarded as false.Footnote ¹³ But contrast that with “quasi-doxastic voluntarism,” as we might call it, which is, roughly, the view that we can engage in acts of make-belief at will. Quasi-doxastic voluntarism is plausibly true—we are free to pretend as we like. Acts of make-belief are under our control in a way that beliefs are not.

Second, make-belief is active, in the sense that engaging in make-belief often involves imaginative transformation of one’s environment. We can select whatever we like to serve as a “prop in a game of make-believe,” and then proceed to imagine whatever we wish about it.Footnote ¹⁴ When authors engage in acts of make-believe, they become literal content creators. In contrast, belief often involves comparatively passive comprehending, considering, and accepting of pieces of “pre-packaged” information.

We think that these differences underwrite the capacity for authors to use acts of make-belief to initiate novel practices of imaginatively transforming environments and inviting others to make-believe in a similar manner, thereby creating new “targets” for attitudes. Typical fictional stories are tools for engaging in just this sort of practice, which is (typically) established by the story’s author, as is plausibly the case in Doyle’s stories involving Holmes.

In contrast, scientific theorizing, like Le Verrier’s, seems to involve belief as a result of trying to follow the evidence gathered (from extra-fictional reality) in order to discern proper explanations for (extra-fictional) natural phenomena. No creative transformations of the relevant sort are established. And no practice of engaging in a “game of belief” is initiated by Le Verrier’s hypothesizing the cause of perturbations in the orbit of Mercury.

In short, whereas make-belief is active and creative and can initiate a distinct sort of imaginative practice, belief is passive and uncreative, and does not initiate any relevant practices in an analogous way. In this sense, Doyle’s attitude of make-belief is itself powerful in a way that Le Verrier’s attitude of belief is not. This difference is ontologically significant.

Granted, there is an attenuated sense in which both fiction-making and myth-making are creative in the same way; as Salmón points out, both are “made up.” That is, both Le Verrier and Doyle create texts of some sort. However, there is an important difference in the creative acts of Doyle and Le Verrier: in addition and related to those differences already mentioned, Doyle’s acts of text-creation were plausibly backed by (something like) “character-directed” creative intentions—like those involved in producing Holmes—while Le Verrier’s acts of text-creation had a different aim entirely: Le Verrier sought a factual explanation of perturbations in the orbit of Mercury. Instead of intending to create the text of a fictional story, largely unfettered by the vicissitudes of extra-fictional reality, Le Verrier intended to create a text that contained an accurate explanation of a real-world phenomenon.Footnote ¹⁵

In our view, entirely different processes which are only, at best, superficially similar are at play in Doyle’s and Le Verrier’s cases. Given Doyle’s aims, and the world’s cooperation, Doyle succeeded, and Holmes exists. That is, Doyle’s aim was to create a fiction, and he succeeded in doing this, because all that is required for his success is that he manages to engage his imagination in the right way. In so doing, he thereby created Holmes. But given Le Verrier’s aims and the world’s lack of cooperation, Le Verrier failed; Vulcan does not exist.Footnote ¹⁶ That is, Le Verrier’s aim was explanatory, and the world did not cooperate by providing a hitherto undiscovered planet satisfying his theory. While the similarity is superficial—both Doyle and Le Verrier produce texts seemingly involving “figures” resembling characters—the differences in the processes run deep.

We conclude, then, that there is a stark ontological difference between the following:

• • activities that lead to the creation of fiction, and the attendant invitations to make-believe, and

• • activities that lead to the construction of mistaken theories, and the attendant invitations to believe.

And these differences have their source in differences between the attitudes themselves. On the basis of this stark ontological difference, we believe (P2) is false, and Salmón’s argument for Ecumenical Realism is unsound.

4. An Objection and an Important Distinction: Fictional Storytelling Versus Mistaken Theories

We think the right contrast to draw in order to bring out our point above is between cases like Doyle’s, on the one hand, and Le Verrier’s, on the other. But it might seem that we have cherry-picked our examples to suit our purposes. Things might look very different if we compared a case of fiction to a more “standard” myth, as opposed to a failed scientific theory, for example. In a way, that is right. But we do not think that is a knock against us. We think that is best seen as a knock against what passes for “myth” in the literature.

Contrary to Salmón (Reference Salmón1998 (Reprinted in (Reference Salmón2005b)), Reference Salmón, Campbell, O’Rourke and Sheir2002, Reference Salmón, Brock and Everett2015, Reference Salmón2023) (and others who follow him in this—“Salmónians”), we think it is deleterious to assimilate failed scientific theories with all of the more colloquial cases of myths. Whereas Salmónians see the relevant “fault line” as dividing fiction on the one hand, and myths of all sorts, plus failed scientific theories, on the other, we think it is more natural to draw the division elsewhere. We think the Salmónian view is in some ways too crude, insofar as it has a distorting effect on the present debates. For one, myths themselves are but one strain of folklore that also includes legends and fables.Footnote ¹⁷ For another, colloquially at least, myths do not include scientific theories. While Salmónians are free to use “myth” however they like, we think it would be more natural to categorize these works according to whether they have more purely fictional ends, or more (perhaps quasi-) scientific explanatory aspirations, at least insofar as we can.Footnote ¹⁸

First, the (relatively) easy cases: Doyle’s fiction is a pretty paradigmatic case of fiction. Mistaken scientific theories, including Le Verrier’s, on the other hand, paradigmatically possess scientific explanatory aims. These are just our previous, “cherry-picked” cases. But where does that leave folklore in general and more “colloquial” myths?

Folklore bears some hallmarks of fictions, while also bearing some salient features of mistaken scientific theories in some cases. For example, the use of “Thor” first appeared in the context of a myth with some overtly quasi-scientific explanatory aims. But many cases, including the case of the “Thor” myth, are rather complicated and messy. It is plausible that the ancient pagan originators of the “Thor” myth sought both to entertain—much like modern fictions—and to explain—much like modern scientific theorizing. Many myths, legends, and fables plausibly have other aims as well—including moral instruction, strengthening social bonds, imparting life lessons, fomenting a sense of group identity, and so on. We can see the aims of fictions like Doyle’s and scientific theories like Le Verrier’s as more focused and specific, in contrast to the “cultural Swiss army knives” found in folklore. We consider the separation of storytelling into explanatory projects, on the one hand, and the production of fanciful stories, on the other hand, a major milestone in intellectual history. But to the extent that the aims of fiction and the aims of science were jumbled together in some folklore, including the “Thor” myth, it is somewhat difficult to say whether the “Thor” myth enjoys a status more like Doyle’s fiction, or more like Le Verrier’s failed scientific theory. We surmise, though, that the intention to explain played a central role in the creation of the “Thor” myth. Central enough, at any rate, to (more or less) safely say the Thor of pagan mythology does not exist, as thunder is not properly explained by invoking a god.Footnote ¹⁹

In contrast, there are many myths that lack an overtly quasi-scientific explanatory aim.Footnote ²⁰ For example, the myth of Sisyphus’s main aim is apparently moral instruction, with no overt quasi-scientific explanatory aspiration. In our view, myths that mainly aim to entertain or morally instruct (and include other typical elements of fictions) are, by and large, sufficiently fiction-like, so as to involve fictional characters. We think it would be beneficial to separate cases of mistaken (proto-) theorizing in which explanatory aims figure centrally, whether they occur in a mythical setting like the pagan invocation of Thor, or an overtly scientific setting, like Le Verrier’s, from more “story-like” myths, like the myth of Sisyphus. Of course, many cases will be even more difficult than these to categorize.Footnote ²¹

Inspired by David Braun’s (Reference Braun2005, p. 615 and passim) discussion of these matters, we will classify those bits of folklore with overt quasi-scientific explanatory aims, together with failed scientific theories, as “mistaken theories.” We will call the acts that produce such mistaken theories “mistaken theorizing.” So, the supposed god of thunder and the hypothesized intra-Mercurial planet are both purported creatures of mistaken theorizing, whereas Holmes is a creature of fiction. And given the aims of the relevant myth, the myth of Sisyphus is more fiction-like than the myth of the god of thunder. We think that, as a result of the myth of Sisyphus’s closer kinship with fiction, it is plausible to suppose that Sisyphus is a “creature of myth” that is more or less just like Holmes in the relevant respects.Footnote ²²

We are now able to characterize Ecumenical Realism more carefully as the conjunction of fictional realism, realism about creatures of folklore (regardless of explanatory aim), and realism about creatures of mistaken theorizing (regardless of whether the theorizing occurs in a mythologizing or in a scientific setting). We can contrast Ecumenical Realism with our preferred “sectarian” view, Sectarian Realism. According to Sectarian Realism, creatures of fiction, like Holmes, exist, as well as creatures of pieces of folklore that lack relevant explanatory aims, like Sisyphus. However, creatures of mistaken theorizing, like the intra-Mercurial planet and the purported god of thunder, do not exist. Note that, on our division, myths in particular, and folklore more generally, are split between those works with overtly science-like explanatory aims on the one hand (together with mistaken scientific theories), and those myths and bits of folklore with more “fictional” aims (together with, well, fictions), on the other. Mistaken theories versus fictional stories, on our scheme, instead of fictions versus myths-plus-failed-scientific theories, according to the way the Salmónians slice the pie.Footnote ²³

5. Arguing for Ecumenical Realism: David Braun’s Elements of Storytelling

In “Empty Names, Fictional Names, Mythical Names,” David Braun also argues for Ecumenical Realism:

Like Salmón, Braun is also arguing that there are creatures of mistaken theorizing because mistaken theorizing is similar in important respects to fictional storytelling. And we again follow Caplan’s lead in presenting Braun’s argument in a parallel manner (but updated with Braun’s preferred terminology, mistaken theorizer, which we also prefer):

While Salmón seems to hold that mistaken theorizers are responsible for the existence of creatures of mistaken theorizing, Braun’s overall view is more nuanced. It might be that, in some cases, the activities of others, in part as a reaction to the activities of mistaken theorizers, bring about creatures of mistaken theorizing.Footnote ²⁴ However, setting such cases aside for now, we see the important difference between Braun’s argument and Salmón’s in terms of how they support the crucial premise. While Caplan sees Salmón’s as relying on a parallel between the propositional attitudes of belief and make-belief, we might instead defend the crucial premise (P3) in Braun’s argument as follows:

We might add on behalf of the proponent of this sort of argument, though Braun himself does not, that (i-iii) seem to capture everything important about the acts of fictional storytelling relevant to the creation of creatures of fiction by authors. But since all of these important aspects are also present in other forms of storytelling—those of the mistaken scientist or the folklorist with explanatory aims, in particular—we should conclude that these mistaken theorizers also manage to create creatures of mistaken theorizing. But if that is so, then (P3) is true.

In response, we reject (P3). That is, we reject that, if authors create creatures of fiction, then activities of mistaken theorizers create creatures of mistaken theorizing. We have already noted some important differences as well as some superficial similarities between fiction and mistaken theorizing: All cases involve the production of a text, and the text produced is intended to be believed, or make-believed. And we agree that (i-iii) capture some important similarities across these diverse phenomena.

However, we continue to think that there are important underlying differences. In order to bring these out in the context of the present argument, we will consider a series of cases involving the same orthographic string. Consider the following two lines:

Suppose something orthographically similar to these two lines is produced entirely accidentally—let your imagination run wild. We think the accidentally produced string would not even be a text, or perhaps would only be a “text” by courtesy or “in an extended sense.” We think it is clear that names and predicates are not used here, nor is anything truth-apt produced.Footnote ²⁶ And we think this is because no relevant intentions were involved in the production of this “text.” And so no story is created or produced either, to say nothing of characters.

Now, suppose an orthographic twin is intentionally produced, but with no other intention than to create just such an orthographic shape. Perhaps a monolingual Japanese artist produces the string because the artist finds the shapes involved particularly interesting or aesthetically pleasing. We also think, in such a case, no names are used, and nothing truth-apt is produced. Note that, in this case, intentions are present—creative intentions, including the intention to create an artwork—but even if some sort of “text” is created (we are less sure what to say in this case), no names, etc., are present. Note, too, that the intentional production of the mere orthographic string does not result in a story, and certainly not a story with a protagonist.

Now, suppose another orthographic twin is produced, but this time, it is backed by an intention that is a bit richer: the intention now is to produce two lines of English text, but that is it: no further plans aside from the concomitant intentions to produce syntactical noun and verb phrases and so on, resulting in a text that seems truth-apt. Perhaps a linguistics professor creates it as an example in a class on morphology. Suppose, further, the professor does not even understand English, but knows enough to produce a string of English expressions. Crucially, there is no intention to tell a story of any sort. There are, again, however, creative intentions—the intention to produce an example for class, or to simply create two lines of English text.

Notice in this case, each of Braun’s (i-iii) appear to be present. It seems to us, though, that no story of any sort is produced, even if a text is (intentionally) produced via creative intentions. So the presence of (i-iii) is perhaps sufficient for the creation of some sort of artifact—a text—but, crucially, it seems insufficient to produce a story or characters.

Recall, however, that Braun’s crucial premise relies on similarities between acts of storytelling that take place in the process of the creation of fictional characters and in mistaken theorizing. So maybe (i-iii) alone are not sufficient to create characters, but, perhaps, they are sufficient in the context of acts of storytelling. Footnote ²⁷

Consider yet another orthographic twin: Suppose this one is produced as a lie. We think a lie can count as a story—at least insofar as a mistaken scientific theory can.Footnote ²⁸ So, this twin is a product of an act of storytelling backed by an intention that the resultant text is to be believed. In such a case, the truth-apt text produced is false. Furthermore, creative intentions are again at play—the intention to create a story, along with the intention to create a lie. But note that, here, there is no intention to create characters. If the liar wishes the audience to believe the liar did thus-and-so, then the liar does not create a story involving some fictional character according to which the fictional character does thus-and-so. Rather, it is a story according to which the liar does thus-and-so.Footnote ²⁹

In this case, perhaps unlike the previous ones, there is an act of storytelling involving (i-iii), but, we think, no character is created. And this is because creative intentions of the right sort are not present. Mere intentions to create something—an orthographic string, a text, or even a story—are not sufficient to create characters. What is missing, we submit, is the intention to create characters. Footnote ³⁰

To see this, consider, finally, the same orthographic string, again, but now in the context of its actual creation. The string was created by @VeryShortStory and posted to X (née Twitter) as a work of fiction. The author intended the story to have a fictional protagonist, so, in our view, in virtue of creating the story, @VeryShortStory also created the parts of that story—the parts that were not imported, at least.Footnote ³¹ But the non-imported, created parts of the story include its plot, the situation we are invited to make-believe, and, we think, the story’s protagonist. Crucially, the latter is brought about, in our view, by @VeryShortStory’s intention to create a fictional story with a fictional protagonist.

And this is the crucial bit, because it seems the only relevant variables in our examples were the intentions that backed the production of the relevant orthographic strings. But the missing intention in all but the last case is: the intention to create characters. It is no coincidence, we think, that in each of the prior cases in which the relevant intentions are absent, what is also absent are characters.

We also think, in cases of mistaken theorizing, these “character-directed” creative intentions are largely or altogether absent. Instead of attempting to produce a text for use as a fictional story, Le Verrier, as with certain folklorists with explanatory aims, sought to produce a text with real-world explanatory aspirations. When the world is not cooperative, those aspirations are not met.

In our view, ultimately, the problem with Braun’s argument is that it attempts to locate the important ingredients involved in character creation in the wrong aspects of acts of storytelling. We think, instead of Braun’s (i-iii), the presence of character-directed creative intentions is what is crucial.Footnote ³² And that crucial element is lacking in central cases of mistaken theorizing, including failed scientific theories and myths or other bits of folklore with overarching explanatory aims.

We conclude that the argument from (P1) and (P3) to (C2) for Ecumenical Realism is unsound. Despite the similarities between acts of storytelling and acts of mistaken theorizing, there are important and ontologically significant differences between them. One important difference is that acts of storytelling typically involve character-directed creative intentions in a way that acts of mistaken theorizing do not. While we agree with Braun that further activities of, for example, commentators on mistaken theories introduce further complications, we think the argument considered here, which focuses on parallels between acts of storytelling and acts of mistaken theorizing, does not provide sound support for Ecumenical Realism.Footnote ³³

6. An Objection and a Reply: Inadvertent Character Production

Above, we appealed to creative intentions of the right sort as the ontologically significant difference-maker. But are such intentions always required in order to bring about fictional characters? Spoiler alert: No.

Braun (Reference Braun2005, pp. 610–12) considers a (perhaps actual) case of Doyle with inchoate creative intentions. Doyle may have lacked character-directed creative intentions when he began writing the Holmes stories. Perhaps he later came to have them, perhaps he did not. Either way, he may have generated Holmes stories prior to having character-directed creative intentions of the right sorts. Regardless, fictional realists should hold that Doyle’s activities brought about Holmes.

Stuart Brock (Reference Brock2010) discusses the (imaginary) case of an author who is an anti-realist about fictional characters, and scrupulously avoids forming intentions to create creatures of fiction. They may go even further by making manifest their explicit intention to not bring about any fictional characters. But, much like a carpenter who produces a table, though they happen to be a compositional nihilist, Brock’s discussion suggests that fictional realists should say that the anti-realist author also brings about fictional characters, despite their intentions.Footnote ³⁴

Zsófia Zvolenzsky (Reference Zvolenszky2016) discusses another (imaginary) case, in which Tolstoy mistakenly thought that Andrei Bolkonsky was a real person. Just as Tolstoy intended to include (the real) Napoleon as a character in War and Peace, he also intended Bolkonsky to be included as an “immigrant” in the fiction. As it happens, there is no real person that Tolstoy had in mind; Tolstoy is simply mistaken. Zvolenzsky argues (convincingly, by our lights) that Tolstoy inadvertently creates (better: produces) Bolkonsky as a character in the fiction.

Call cases like those described by Braun, Brock, and Zvolenzsky cases of inadvertent character production. We think fictional realists should accept the reality of inadvertent character production. This is an important notion. So, we think it is worthwhile to digress a bit to get clearer on how we are thinking about inadvertent character production.

7. Arguing for Ecumenical Realism: Creation of the Whole is Creation of the Parts

Authors of fiction create more than texts when they write. They also create stories. Stories that can be told and retold, changed, and adapted in various ways. For example, when Doyle wrote the text for A Study in Scarlet, he also created a story. That story has been told and retold many times and adapted to radio, film, and television. But what are stories? Cowling and Cray (Reference Cowling and Cray2022) suggest a “neo-idealist” theory of stories; according to it, stories are ideas for narrative manifestation. Footnote ⁴² When Doyle wrote A Study in Scarlet, he became the first person to have a particular idea; an idea that involved some combination of various narrative elements such as characters, settings, events, and themes. That idea was manifested in his text, and the text became the initial medium by which that idea was passed along to others. The same idea was later manifested on radio, film, and television. That idea is the story of A Study in Scarlet. Footnote ⁴³

A story, on this view, must involve various narrative elements, typically including characters. As we have argued, it is not implausible to think that when a story involving a certain character is created, then that character is also thereby created (assuming the character does not already exist). Or at the very least, that character comes into existence as a byproduct of the creation of the story and are inadvertently produced in the above sense. After all, the whole cannot exist without its parts.

Similarly, theorizers create theories. Theories can be told and retold, changed, and adapted in various ways. What are these theories? One might think, along the lines of Cray and Schroeder (Reference Cray and Schroeder2015), that theories, too, are ideas—ideas that involve, at the very least, some combination of explanatory elements including objects, features, events, and causes. It is not implausible to think that when a theory is created, its various components that are not already in existence must be created, or at the very least produced. After all, the whole cannot exist without its parts.

This all suggests the following argument:Footnote ⁴⁴

(P4) Authors produce creatures of fiction as parts of stories when they create those stories.

(P5) If authors produce creatures of fiction as parts of stories when they create those stories, then mistaken theorizers produce creatures of mistaken theorizing as parts of false theories when they create those theories.

(C3) So, mistaken theorizers produce creatures of mistaken theorizing as parts of false theories when they create those theories.

Let us think a little bit more about the parts that are necessary for the whole. Consider Doyle again. In order to have the idea, that is, the story A Study in Scarlet, he had to have ideas of its various narrative elements: a “Holmes-idea,” a “Watson-idea,” a “Lestrade-idea,” etc., and an idea of how the contents of those elements combined together. These ideas are literal parts of a larger whole idea that is A Study in Scarlet. Footnote ⁴⁵ Similarly, when Le Verrier came up with his theory about the perturbations in the orbit of Mercury, he had to have various ideas: perhaps these included a “Mercury-idea,” a “Vulcan-idea,” and an idea of how the contents of those elements fit together. Those ideas are literal parts of the larger idea that is Le Verrier’s theory.

One might think that the character of Sherlock Holmes just is the Holmes-idea. This certainly seems to fit with how some people think about fictional characters.Footnote ⁴⁶ But it does not fit with others. Consider a famous dispute between Stan Lee and Steve Ditko over the creation of Spider-Man. Stan Lee succinctly sums up the dispute as follows:

Clearly, Stan Lee thinks that having the idea of a character is sufficient for creating the character. Stan Lee’s thesis about creation would make a lot of sense if the fictional character just is the character idea. Ditko, on the other hand, thinks that something more needs to be done to create a character. In addition to having an idea, one must manifest the idea in a tangible way. That suggests, instead, that a fictional character cannot just be the character idea.Footnote ⁴⁸ Rather than recapitulate Ditko’s defense of his view, we think it’s better to just let Ditko speak for himself, as it were:

We are somewhat sympathetic to Ditko’s position. For example, it does not seem implausible to think that Sherlock Holmes is what the Holmes-idea is an idea of, rather than the idea itself. But that’s harder to square with Lee’s view. And similarly, Vulcan, were it to have existed, would presumably be what the Vulcan-idea was an idea of. We grant that the Holmes-idea must have been created when the story of A Study in Scarlet was created. And similarly, the Vulcan-idea must have been created when the theory of certain orbital perturbations was created. But it does not follow that a creature of mistaken theorizing was thereby created, or that Le Verrier created anything relevant beyond the idea.

Ultimately, though, we aren’t completely opposed to Lee’s “Fictional Idealism,” as we might call it.Footnote ⁴⁹ Maybe Sherlock Holmes is the Holmes-idea and maybe, generally, merely fictional characters are identical to character-ideas. So let us assume, for the sake of argument, that authors produce creatures of fiction as parts of stories when they create those stories. We deny that, if that is so, then mistaken theorizers produce creatures of mistaken theorizing as parts of false theories when they create those theories. That is, we deny (P5).

We think it is pretty plausible to suppose that Fictional Idealism is true, and Doyle creates Holmes by creating the Holmes-idea, because Holmes is the Holmes-idea. However, it seems implausible to us to think that Vulcan is identical to the Vulcan-idea. This is so for two main reasons.Footnote ⁵⁰

First, and perhaps most importantly, we think that, if Vulcan is produced by Le Verrier on something like the model whereby Doyle produced Holmes, as all of the arguments for Ecumenical Realism that we have considered suggest, then that is either because Le Verrier produced Vulcan by creating Vulcan, or Le Verrier merely produced Vulcan.

Suppose Le Verrier created Vulcan. Then Le Verrier had something like character-directed creative intentions toward the main “figure” in his theory, in much the same way that it is plausible to suppose that Doyle had character-directed creative intentions toward the main character of his story. But we think this is implausible for the reasons we have discussed. Chief among them is that Le Verrier likely harbored anti-creative intentions toward the main “figure” in his theory.

Suppose instead that Le Verrier merely produced Vulcan. But then Le Verrier produced the text he did in the context of successful engagement with a creative practice. But we do not think that is right, either, for the reasons we have discussed. Chief among them is that Le Verrier produced his text in the context of theorizing, a practice with overt explanatory aims, which is not creative in the relevant sense.

Our second main reason for thinking the analog of Fictional Idealism does not hold for mistaken theories (“Mistaken Theory Idealism”) is because we think that, while the Holmes-idea is apt to serve as the fictional character Holmes, Le Verrier’s Vulcan-idea is not similarly apt to serve as the titular figure in Le Verrier’s “Vulcan Theory.” We think that, in fictional cases, since we are just pretending that something is some way, it matters less what that “something” is, becasue practically anything can potentially serve as a “prop in a game of make-believe.”Footnote ⁵¹ But things are different in the case of theorizing. After all, Le Verrier hypothesized that Vulcan orbits the sun and causes perturbations in the orbit of Mercury. But whereas Doyle might have imagined that the idea he came up with smoked and consulted, it is not similarly plausible to suppose that Le Verrier thought that the idea he came up with orbits or perturbs. Similarly, it was thought by pagan myth-makers that Thor caused thunder. And while myth-makers thought something causes thunder, and that that thing was a god, presumably, they did not think that the idea they just came up with for what causes thunder caused thunder. So, even if Lee’s Fictional Idealism is plausible, it seems less plausible to hold a corresponding “Mistaken Theory Idealism.”

We would like to stress that our objection here is not the following:Footnote ⁵²

The problem is not that the Vulcan-idea, intrinsically, is the wrong sort of thing. Nor is the problem that agents should be able to determine “from the inside” that their psychological states and propositional attitudes predicate features of things that no reasonable person would believe those things can have. After all, such states are not plausibly transparent in that way. Rather, basically for the reasons Braun, Reference Braun2005 holds that Le Verrier did not refer to Vulcan, we think nothing “Vulcan-ish” can be the target of Le Verrier’s intentions in the way it would need to be in order for Le Verrier to have created it broadly in the way authors create characters. Supposing Le Verrier sought to name an intra-Mercurial planet, if there is one, and nothing otherwise, then nothing “Vulcan-ish,” including the Vulcan-idea Le Verrier did plausibly create, or any sort of “abstract artifact” of the kind Ecumenical Realists discuss, is apt to serve as a creature of Le Verrier’s mistaken theorizing.Footnote ⁵³

Separating cases, it seems implausible to hold that Le Verrier’s Vulcan is identical to the Vulcan-idea and it seems very questionable to hold that Thor is identical to the Thor-idea. We think the point generalizes: while it is plausible to suppose that fictional characters are closely tied to fictional character-ideas (perhaps by the tie of numerical identity!), it is implausible to suppose that creatures of mistaken theorizing are so closely tied to ideas. So we conclude that the argument from (P4) and (P5) to (C3) for Ecumenical Realism is unsound, as well, since (P5) is false.Footnote ⁵⁴

8. Immigration Issues

We have argued, in effect, for Sectarian Realism, according to which creatures of fiction exist, while creatures of mistaken theorizing, and many creatures of myth, do not. But the threat of sectarian violence is close at hand.

Some characters are indigenous to stories, whereas other characters are immigrants. For example, Hulk and Spider-Man are both indigenous to Marvel Comics, first appearing in The Incredible Hulk #1 and Amazing Fantasy #15, respectively. However, David Letterman and Justin Trudeau are both immigrants to Marvel Comics; they are real people who appeared in issues of Marvel Comics.Footnote ⁵⁵ But some immigrants to fiction appear to immigrate from “realms” other than the extra-fictional world. For example, Alan Moore and Kevin O’Neill’s League of Extraordinary Gentlemen stars several characters who originated from 19th century fiction, including Brahm Stoker’s Nina Murray and Jules Verne’s Captain Nemo.Footnote ⁵⁶

What about Thor? Thor first appeared as a character in Marvel Comics in Journey into Mystery #83, and is a founding member of the Avengers. So, it would seem, fictional realists like us should countenance Thor. But it appears as if Thor is an immigrant from pagan mythology. If Thor is an immigrant from pagan mythology, then Thor must have existed as a part of pagan mythology. And if Thor existed as part of pagan mythology, then Thor must have been created by the original pagan myth-makers. So, one may be tempted to conclude, if fictional realists are welcoming towards immigrants, then they must be welcoming to creatures of mistaken theorizing, as well. We have argued, though, against somewhat similar cases for Ecumenical Realism. Moreover, given the explanatory aims of the Thor myth and the lack of creative intentions backing up our explanatory practices, we do not think that Thor was created or produced by those myth-makers.

One might respond to this challenge by denying that there are any immigrants at all. One might distinguish between, for example, Justin Trudeau—the Marvel fictional character—and Justin Trudeau—the real prime minister of Canada. Similarly, one might distinguish between Thor—the Marvel fictional character—and any putative creatures of mythology. If there are no immigrants at all, then there is no problem for our brand of Sectarian Realism.

But we do not think that is the right response to this anti-Sectarian challenge. To see why, note that practically everyone agrees on the following: Spider-Man is depicted as taking place in New York, Spider-Man is depicted as speaking English, and Spider-Man’s outfit is depicted as being red. Those who endorse the anti-immigration policy will claim, at best, that Spider-Man is depicted as taking place in New York by way of representations of a fictional city, Spider-Man is depicted as speaking English by way of representations of a fictional language, and Spider-Man’s outfit is depicted as being red by way of representations of a fictional color. But we think that Spider-Man is depicted as taking place in New York by way of representations of New York, Spider-Man is depicted as speaking English by way of representations of English, and Spider-Man’s outfit is depicted as red by way of representations of red. To claim otherwise seems overly complicated as well as more than a little absurd. Thus, we advocate open borders and welcome immigrants.Footnote ⁵⁷

Where does that leave Thor, so to speak? Thor seems to be an immigrant from bits of folklore that engaged in mistaken theorizing. And we doubt there are any such things. So, if Thor is an immigrant, where did he immigrate from? The Ecumenical Realist has a relatively easy time here—there was this character from myth who later became a character in a fiction. But that explanation is not available to us Sectarians. What to do?

Although we believe that there are immigrants in fiction, we do not think that Thor is an immigrant to Marvel Comics. Or, rather, we do not think that Thor is an immigrant in the relevant sense. What seems right to us is that Stan Lee and Jack Kirby, who created the Thor stories for Marvel, intended to write a fiction, and they were inspired by pagan mythology.Footnote ⁵⁸ Whether they intended to tell a story about an already existing creature of mythology is unclear. If they intended to create a character inspired by the pagan myths, then there is no problem here for Sectarian Realists. Thor would not, in that case, be an immigrant. Instead, Thor would be a creation of Lee and Kirby and would be indigenous to Journey into Mystery #83. But suppose that they really did fully intend to bring Thor from the realm of myths to the realm of Marvel Comics; suppose they intended to immigrate Thor from pagan myths to Marvel Comics. What could we say in that case?

Well, there are a lot of intentions involved in the creation of a story. In this case, it seems clear that Lee and Kirby intended to create and tell a fictional story. Perhaps they intended that story to be about an already existing creature of myth. But we do not think that there was any creature of myth around to be the target of their intentions. In that case, given our established fictional storytelling practices, their intention to create a story was a sort of dominating intention, and Thor—the fictional character—was a byproduct of that intention. So, even if they did not intend to create a fictional character, we think that a fictional character was nevertheless brought into existence, as in our analogy above involving tables. This, we think, is perhaps another example of character production in the absence of character-directed creative intentions, as we discussed above in replying to Braun’s argument for Ecumenical Realism.

Allow us to elaborate. We like Cowling and Cray’s (Reference Cowling and Cray2022) view that stories are ideas for narrative manifestation—ideas made up of some combination of further ideas related to various narrative elements such as characters, settings, events, and perhaps themes. Let us consider how our fictional storytelling practices can help to bring the various elements of a story together into an idea for narrative manifestation. Sometimes, an author intends to create certain characters, or settings, or events. Other times, an author intends to immigrate certain characters, or settings, or events.Footnote ⁵⁹ Let us focus on the characters. When an author intends to create a character and produces the right sorts of elements, such as notes or drawings that would properly convey the idea that the author has for the character, then the author thereby creates the character. When there is an individual that an author intends to immigrate as a character, and the author produces the right sorts of elements that would properly convey the idea of that individual, then the author thereby immigrates that individual as a character. But when an author intends to immigrate an individual as a character, and there is no such individual to immigrate, then something else happens. Suppose the author has an idea that the author takes to be an idea of an already existing individual. And suppose that the author produces the right sort of elements that would properly convey that idea. And finally, suppose that this is all done while intentionally and successfully writing fiction, as part of the practice of fictional storytelling. Then, in that case, the author inadvertently produces a character. We believe that if Lee and Kirby really did intend to immigrate Thor from pagan myths, then, as a consequence of their overall authorial intentions, productions, and with the aid of our storytelling practices, they might have instead produced Thor as a fictional character.Footnote ⁶⁰

9. An Objection and a Reply Revisited

Recall, above, we considered an objection to our replies to Salmón, Caplan, and Braun. According to that objection, we cannot rely on creative intentions of the right sort to explain the differences between cases where fictional realists say a fictional character is created and cases where Ecumenical Realists say a creature of mistaken theorizing is created, too. In light of our discussion of Cray and Schroeder’s (Reference Cray and Schroeder2015) view of theories as ideas, and Cowling and Cray’s (Reference Cowling and Cray2022) extension of that idea to stories, we can say more about what we consider to be among the crucial differences between cases involving fictional storytelling versus those involving mistaken theorizing.Footnote ⁶¹

When Lee and Kirby created Journey into Mystery #83, they created a story. Following Cowling and Cray (Reference Cowling and Cray2022), suppose that the story is an idea. And, in line with the previous section, suppose that story-idea includes a Thor-idea. The Thor-idea may have existed previously, and Lee and Kirby may have pressed it into service as (perhaps) an idea of an (eventual) Avenger.Footnote ⁶² Now suppose, for relative simplicity, that Lee’s Fictional Idealism is true and creating a character-idea suffices for creating a character, even in the absence of appropriate drawings, and so forth. Then we are in a position to identify another way in which a sort of inadvertent production of fictional characters may occur, even while a parallel inadvertent production of a creature of mistaken theorizing would not. In Fictional Idealism, the Thor-idea is the sort of thing that is apt to serve as the fictional character. When it is pressed into service by Lee and Kirby as part of a story idea, it becomes a fictional character. In that sense, a fictional character is inadvertently produced. But as we have argued, “Mistaken Theory Idealism” is false; hence, the Thor-idea was not apt to be a creature of mistaken theorizing while it was embedded in the quasi-explanatory project of the pagan myth it originated in. So, whereas the Thor-idea might have been produced by myth-makers as part of their explanatory practices, no creature of myth was thereby produced.

We could, similarly, appropriate Le Verrier’s Vulcan-idea, and press it into service as a character in the context of fiction. In fact, we think something like this may be what actually happened in the case of Harl Vincent’s (Reference Vincent1932) story, “Vulcan’s Workshop.”Footnote ⁶³ So, on this proposal, perhaps Le Verrier created a Vulcan-idea,Footnote ⁶⁴ but his activities did not bring about anything that answered that idea; his uses of “Vulcan” in “Vulcan est une planète” were empty, and his utterance perhaps expressed and encoded a gappy proposition.Footnote ⁶⁵ But when that idea spread to Vincent, Vincent instead used it as a central character in a work of fiction that he created.

One thing that is nice about this story is that we can see how fictions can borrow from myths and false scientific theories. And we can see a sense in which certain creatures of fiction are immigrants from myths or false scientific theories. What fictions borrow are the ideas. Those ideas immigrate from mistaken theories to the fictions. Given Fictional Idealism, those ideas become fictional characters. But, crucially, those ideas were never the things posited to exist by mistaken theorizing, and, given theorizers’ typical intentions, there is nothing that their ideas are ideas of.

Though many antecedents need yet to be discharged, on this line, the ontology of fiction and mistaken theorizing may be quite similar. Ideas would serve as fictional characters in the context of fictional storytelling, while those same ideas would be inapt to serve as creatures of mistaken theorizing, due to (i) facts about the practices of fictional storytelling versus theorizing, (ii) the intentions of authors versus theorizers, and (iii) the unsuitability of ideas as objects of theorizing, even while they might be well-suited to playing the role of characters in fictions. In a way, then, perhaps even the Sectarian Realist may be able to strike an ecumenical chord.