Read Aloud the Text Content

This audio was created by Woord's Text to Speech service by content creators from all around the world.

Text Content or SSML code:

**If We Postulated Fictional Objects, What Would They Be?** If we are to postulate fictional characters at all, it seems advisable to postulate them as entities that can satisfy or at least make sense of our most important beliefs and practices concerning them. Often theories of fiction are driven not by an independent sense of what is needed to understand talk and practices regarding fiction, but rather by a desire to show how fictional characters may find their place in a preconceived ontology of possible, nonexistent, or abstract objects – to demonstrate one more useful application of the ontology under discussion, or to provide catchy and familiar examples. Instead of starting from a ready-made ontology and seeing how we can fit fictional characters into it, I suggest that we begin by paying careful attention to our literary practices so that we can see what sorts of things would most closely correspond to them. I thus begin by discussing what sorts of entities our practices in reading and discussing works of fiction seem to commit us to, and I draw out the artifactual theory of fiction as a way of characterizing the sort of entity that seems best suited to do the job of fictional characters. **What Fictional Characters Seem to Be** Fictional objects as I discuss them here include such characters as Emma Woodhouse, Sherlock Holmes, Hamlet, and Tom Sawyer - characters who appear in works of literature and whose fortunes we follow in reading those works.¹ In our everyday discussions of literature we treat fictional characters as created entities brought into existence at a certain time through the acts of an author. If someone contended that George Washington was a great fan of Sherlock Holmes, we might object that in Washington's time there was no Sherlock Holmes - the Holmes character was not created until 1887. The term “fiction” derives from the Latin *fingere* meaning "to form,” and this linguistic root is still evident in our practices in treating fictional characters as entities formed by the work of an author or authors in composing a work of fiction.² We do not describe authors of fictional works as discovering their characters or selecting them from an ever-present set of abstract, nonexistent, or possible objects. Instead, we describe authors as inventing their characters, making them up, or creating them, so that before being written about by an author, there is no fictional object. In short, on this view fictional characters are a particular kind of cultural artifact. Like other cultural objects, fictional characters depend on human intentionality for their existence. Like other artifacts, they must be created in order to exist, and they can cease to exist, becoming past objects. It is primarily in its treatment of fictional characters as ordinary cultural artifacts rather than as the odd inhabitants of a different realm that the artifactual theory differs most markedly from other ways of characterizing fictional objects. It is also their place as cultural artifacts that makes fictional objects of broader philosophical interest, for the ontology of fiction can thus serve as a model for the ontology of other social and cultural objects in the everyday world. It may help to locate the artifactual theory in conceptual space by briefly contrasting it with other views of what fictional objects are. Some of its advantages vis-à-vis these other theories only show up when we attempt to overcome the problems of developing identity conditions for fictional objects and handling reference to and discourse about them. Nonetheless, a brief comparison should help elucidate the differences between this theory and other treatments of fiction. **Meinongian Theories of Fiction** The most popular and well-developed theories of fiction that have been available are those broadly construable as Meinongian theories, including those that take fictional characters to be either nonexistent or abstract entities, such as those developed by Parsons, Zalta, and Rapaport. Neither Meinong's theory nor contemporary Meinongian theories are devised specifically as theories of fiction; they concern the wider realm of nonexistent objects generally. 18 Nonetheless, much of the motivation for and many of the applications of Meinongian theories of nonexistent objects concern fictional objects. Many different theories have been devised that may roughly be labeled Meinongian; despite their differences, they typically share certain fundamental characteristics captured by the following principles: 1. There is at least one object correlated with every combination of properties. 2. Some of these objects (among them fictional objects) have no existence whatsoever. 3. Although they do not exist, they (in some sense) have the properties with which they are correlated. The first principle is sometimes known as a "comprehension principle,” ensuring a multitude of nonexistent objects. Meinongian theories differ with respect to which properties count in principle one. Parsons's theory, for example, limits properties to simple, nuclear properties such as "is blue" or "is tall"; Zalta's theory permits so-called extranuclear properties (such as "is possible" and "is thought about") and complex properties. Meinongians also vary with respect to how nonexistent objects "have" their properties according to the third principle; for views like Parsons's, there are two kinds of property (nuclear and extranuclear), but only one kind of predication, enabling "have" to be read straightforwardly, as (in this theory) nonexistents have their properties in the same way as real objects do. For views like Zalta's or Rapaport's, there are two modes of predication; nonexistent objects have properties in a different way than their real counterparts. Although ordinary objects exemplify their properties, nonexistent objects "encode” the properties with which they are correlated (Zalta) or have them as "constituents" (Rapaport).22 Meinongian theories of fiction resemble the artifactual theory in that both allow that there are fictional objects, that we can refer to them, that they play an important role in experience, and so on. Moreover, Meinongians are largely to be credited with showing that consistent theories of fiction can be developed and with undermining the paradigm according to which there are only real entities (a paradigm Parsons refers to as the "Russellian rut”). But there are also important differences between the artifactual theory and any such Meinongian theory of fiction. First, the theories differ with respect to where they apply the word "exists"; I am willing to claim that fictional characters exist; the Meinongian (by principle two) grants them no existence whatsoever. But because the Meinongian famously maintains that there are such objects, that we can think of them, refer to them, and so on, this difference is largely linguistic.23 A deeper difference between the theories regards how many objects they say there are. Unlike the Meinongian, I do not employ any kind of comprehension principle and so do not claim that there is an infinite, ever-present range of nonexistent (or abstract) objects. In the artifactual theory, the only fictional objects there are those that are created. This points to a further difference between this view and that of the Meinongian: In the artifactual view, fictional objects are created at a certain point in time, not merely discovered or picked out. According to the Meinongian, fictional characters are merely some of the infinite range of ever-present nonexistent or abstract objects - namely, those that are described in some story. If an author writes of a character, she or he is merely picking out or referring to an object that was already available for reference. Authors can then be said to discover their characters or pick them out from the broad range of objects available, but not to bring these objects into existence. They can be said to make these objects fictional, for an abstract or nonexistent object does not become fictional until it is written about. Nonetheless, the object remains the same; it simply bears a new relation to contingent acts of authoring.24 As Parsons writes: "I have said that, in a popular sense, an author creates characters, but this too is hard to analyze. It does not mean, for example, that the author brings those characters into existence, for they do not exist. Nor does he or she make them objects, for they were objects before they appeared in stories. We might say, I suppose, that the author makes them fictional objects, and that they were not fictional objects before the creative act." 25 In short, the only kind of creation permitted in Meinongian accounts is the author's taking an available object and making it fictional (by writing about it in a story). This, it seems to me, is not robust enough to satisfy the ordinary view that authors are genuinely creative in the sense of creating new objects, not merely picking out old objects and thereby making them fictional. By contrast, in the artifactual theory, authors genuinely bring new characters into being that were not around before - they invent their characters rather than discovering them. In short, the Meinongian might be said to offer a top-down approach to fiction that begins by positing an infinite range of nonexistent or abstract objects and then carves out a portion of those (those described in works of literature) to serve as the fictional characters. In contrast, the artifactual theory attempts to take a bottom-up approach to fictional characters by treating them as constructed entities created by authors and depending on ordinary objects such as stories and a competent audience. As we see in later chapters (Chapters 5, 4 and 7 respectively) there are also many differences between Meinongian theories and the artifactual theory regarding identity conditions for fictional objects and how reference to and discourse about fictional objects are handled. Some shortcomings of the Meinongian view include an inability to genuinely treat fictional characters as created entities and consequent difficulties in offering adequate identity conditions for fictional characters (especially identity conditions across texts). 26 Other problems arise for Meinongian treatments of fictional discourse, notably in handling fictional discourse about real individuals. Thus, despite the merits of Meinongian theories in offering a consistent and well-developed view of fictional characters, I argue that the artifactual theory provides a better conception of them overall. The main difference underlying the two theories and responsible for the advantages I claim for the artifactual theory lies in a fundamental difference in approach, as the Meinongian sees fictional characters as part of a separate realm of abstract or nonexistent objects, disjoint from and dissimilar to that of ordinary objects, and in the artifactual theory their similarities and connections to entities in the ordinary world are taken as fundamental. **Possibilist Theories of Fiction** Whether in an attempt to provide a complete account of fiction or as motivation for an ontology of possibilia, an attempt is often made to locate fictional characters among unactualized possibilia. Fictional characters have long provided some of the most appealing examples of merely possible entities and have often been used in arguments in favor of postulating unactualized possibilia. Kripke, for example, uses Sherlock Holmes as an example (which he later retracts) of an entity that "does not exist, but in other states of affairs he would have existed," and Plantinga treats the view that "Hamlet and Lear do not in fact exist; but clearly they could have" as one of the most persistent arguments in favor of unactualized possibilia.27 And at first glance, it seems plausible that, even if there is no actual person who has all of the properties ascribed to Hamlet in the play, surely there is some possible person exhibiting all of those properties, making Hamlet a member of another possible world. This is a fundamentally different approach to fiction from that of the artifactual theory, because in the artifactual theory fictional objects are not possible people but actual characters. Although it is a tempting way to accommodate fictional characters, and fictional characters may provide fun (purported) examples of mere possibilia, major problems arise if we try to identify a fictional character with that merely possible individual exhibiting all and only those properties ascribed to the character in the story. First, as has been frequently acknowledged, there seem to be simply too many possible individuals that fit the bill, and no means to choose among them.28 For the descriptions provided in literary works fail to completely specify what the characters described in them are like, leaving indeterminate a wide range of properties such as, typically, a character's blood type, weight, diet, and mundane daily activities. Thus we run into trouble immediately if we try to identify characters with possible people, for the features of a character left open by the story could be filled out in an infinite variety of ways by different possible people. Selecting any one as identical with a particular character seems hopelessly arbitrary. On the other hand, if the character is described as bearing incompatible properties, making it an impossible object, we have not too many possibilia to do the job, but too few. A further problem arises in that possibilist views, like Meinongian views, give us no way of accounting for the created status of fictional characters. Even if we could find a single candidate possible detective to identify with Sherlock Holmes, this would be a possible man with the property of being born in the nineteenth century, not of being created by Arthur Conan Doyle. Finally, possibilist theories, Meinongian theories, or any theories that base the identity of a character on the properties ascribed to it eliminate the possibility that there can be more than one story about a single character. For if the character is ascribed even a single different property, it is a different character. Thus these views provide no means to admit that the same character may appear in different stories, sequels, or even slightly altered new editions or translations of an old story. Perhaps it is because of such problems that this view has been far less popular among those working seriously with fiction. Indeed Kripke and Plantinga both, after considering it, reject this view. In light of these problems it seems that possible objects are not candidates well suited to do the job of fictional characters. Because this view of fiction seems hardly able to get off the ground, I do not spend much time discussing it.29 **Fictional Characters as Objects of Reference** Other views of fiction consider fictional objects mere objects of reference that we must postulate to make sense of a certain kind of literary discourse. Such views are developed by Crittenden, who treats fictional objects as "grammatical objects," and by van Inwagen, who considers fictional objects to be the "theoretical entities" referred to in works of literary criticism. These views parallel the artifactual view in many important respects, and the differences between such theories and the artifactual theory lie less in direct conflicts than simply in the artifactual theory's filling in areas left blank by the other theories. Nonetheless, there are also important differences of approach between these theories and the artifactual theory. **Working within a broadly Wittgensteinian view of language,** Crittenden postulates fictional objects as (mere) objects of reference, or grammatical objects. Although he takes fictional names as referring to certain objects, was not created until 1887. The term “fiction” derives from the Latin *fingere* meaning "to form,” and this linguistic root is still evident in our practices in treating fictional characters as entities formed by the work of an author or authors in composing a work of fiction.² We do not describe authors of fictional works as discovering their characters or selecting them from an ever-present set of abstract, nonexistent, or possible objects. Instead, we describe authors as inventing their characters, making them up, or creating them, so that before being written about by an author, there is no fictional object. Thus, if we are to postulate fictional characters that satisfy our apparent practices regarding them, it seems that we should consider them to be entities that can come into existence only through the mental and physical acts of an author - as essentially created entities. Once we begin to treat fictional characters as created entities, a further issue arises. Do they simply need to be created at some time, by someone, or is the identity of a fictional character somehow tied to its particular origin in the work of a particular author or authors taking part in a particular literary tradition? Unexamined intuitions may provide no clear answer to this question, but our goal is to draw out a view of fictional characters that corresponds as closely as possible to our practices in studying fictional characters. Such critical practices provide grounds for taking the latter view, that a particular fictional character not only has to be created but is necessarily tied to its particular origin. Suppose that a student happens on two literary figures remarkably similar to each other; both, for example, are said to be maids, warding off attempts at seduction, and so on. Under what conditions would we say that these are works about one and the same fictional character? It seems that we would say that the two works are about the same character only if we have reason to believe that the works derived from a common origin - if, for example, one work is the sequel to the other, or if both are developments of the same original myth. Literary scholars mark this difference by distinguishing "sources" drawn on by an author in composing a work from coincidentally similar characters or works, mere "analogs." If one can show that the author of the latter work had close acquaintance with the earlier work, it seems we have good support for the claim that the works are about the same character (as for example in the Pamela Andrews of Richardson's and Fielding's tales). But if someone can prove that the authors of the two works bore no relation to each other or to a common source but were working from distinct traditions and sources, it seems that the student has at best uncovered a coincidence – that different individuals and cultures generated remarkably similar analogous characters. So it seems that if we wish to postulate fictional objects that correspond to our ordinary practices about identifying them, fictional characters should be considered entities that depend on the particular acts of their author or authors to bring them into existence. Naturally, the process of creating a particular character may be diffuse: It may be created by more than one author, over a lengthy period of time, involving many participants in a story-telling tradition, and so on. But the fact that the process of creating a fictional character may be diffuse does not disrupt the general point that, whatever the process of creation for a given character may be, for coming into existence it depends on those particular creative acts. Such a requirement not only is consistent with critical practices in identifying characters but also (as we see in Chapter 5) is crucial to treating characters as identical across different sequels, parodies, and other literary developments. Once created, clearly a fictional character can go on existing without its author or his or her creative acts, for it is preserved in literary works that may long outlive their author. If we treat fictional characters as creations invented by authors in creating works of literature, and existing because of their appearance in such works, then it seems that for a fictional character to be preserved, some literary work about it must remain in existence. And so we have uncovered a second dependency: Characters depend on the creative acts of their authors in order to come into existence and depend on literary works in order to remain in existence.³ Here again the question arises: Does a fictional character depend on one particular literary work for its preservation, or does a fictional character need only to appear in some literary work or other to remain in existence? It certainly seems that a character may survive as long as some work in which it appears remains. If we could not allow that the same character may appear: in more than one literary work, or even slightly different editions of a work, then we would be unable to account for literary critical discourse about the development of a character across different works, and we would even be unable to admit that readers of different editions of *The Great Gatsby* are discussing one and the same Jay Gatsby. In short, we would be left postulating many characters in cases in which there seems to be but one. So it seems we should allow that one character may appear in more than one work, and if it can appear in more than one work, it must remain in existence as long as one literary work about it does. Thus even if "A Scandal in Bohemia" should exist no longer, the character Sherlock Holmes can go on existing provided that one or more of the other works in which he appears remains in existence. So, although a fictional character depends on a literary work for its continued existence, it depends only on the maintenance of some work in which it appears.⁴ The dependence of a character on a literary work forces us to address a second question: If a character depends on a work of literature, what does a work of literature depend on? When can we say that a literary work exists? Because characters depend on literary works, anything on which literary works depend is also, ultimately, something on which characters depend. As ordinarily treated in critical discourse, a literary work is not an abstract sequence of words or concepts waiting to be discovered but instead is the creation of a particular individual or group at a particular time in particular social and historical circumstances. Thus, as with characters, it seems that literary works must be created by an author or authors at a certain time in order to come into existence. Like a character, it also seems that a work of literature depends rigidly on the acts of its particular author to exist, so that, even if two authors coincidentally composed the same words in the same order, they would not thereby have composed the same work of literature. One way to see the essentiality of a work's origin to its identity is by observing that literary works take on different properties based on the time and circumstances of their creation and creator. By virtue of originating in a different place in literary, social, and political history, at the hands of a different author, or in a different place in an author's oeuvre, one and the same sequence of words can provide the basis for two very different works of literature with different aesthetic and artistic properties. The same sequence of words appearing in *Animal Farm* could have been written in 1905, but that literary work could not have had the property of being a satire of the Stalinist state, a central property of Orwell's tale. If the same words of *Portrait of the Artist as a Young Man* were written by James Joyce not in 1916, but instead after *Ulysses* came out in 1922, that work would lack the property of exhibiting a highly original use of language, which *Portrait of the Artist* has. Two mysteries based on the same sequence of words written in 1816 and today, both ending with "the butler did it," might have the property of having a surprise ending in the former but not the latter case. A screenplay with the same sequence of words as Oliver Stone's *Nixon*, if written in 1913, could have the properties neither of being about (the real) Richard Nixon, nor of being a sympathetic portrayal of the main character, nor of being revisionary and speculative. Similar cases could be brought to bear to show that a wide variety of aesthetic and artistic properties central to discussions of works of literature—being a work of high modernism, a parody, horrifying, reactionary, exquisitely detailed, an updated retelling of an old story—depend on the context and circumstances of creation, so that literary works may be based on the same series of words but have different aesthetic and artistic properties. In at least some cases, these properties seem essential to the literary work, e.g., being a satire seems essential to *Animal Farm* considered as a work of literature. For that reason, it seems that a literary work is best conceived not as an abstract sequence of words but as an artifact that had to be created in those original circumstances in which it was created. Like fictional characters, literary works, once created, can clearly survive the death of their author; indeed the great majority of literary works we have today persist despite the deaths of their authors. But does a literary work, once created, always exist, or can a work once again cease to exist even after it is created? If we take seriously the view that literary works are artifacts created at a certain time, it seems natural to allow that, like other artifacts from umbrellas to unions to universities, they can also be destroyed. It would surely seem bizarre to claim that all of the lost stories of past cultures still exist as much as ever. On the contrary, one of the things that is often lamented about the destruction of cultures, be they ancient Greek or Native American, is the loss of the stories and fictional worlds they created. We treat literary tales as entities that can cease to exist, that at times take special efforts and government projects to preserve (e.g., by recording the oral folktales of Appalachia), or that may be destroyed by a temperamental author burning unpublished manuscripts. Treating works of literature as entities that may be destroyed—at least if all copies and memories of them are destroyed—seems a natural consequence of considering them to be cultural artifacts rather than Platonistic abstracta. Yet certainly there are many who do not share the intuition that literary works may cease to exist after being created. The idea that literary works, if they exist, must exist eternally (once created) seems to me to be a hangover of a Platonism that assimilates all abstract entities to the realm of the changeless and timeless, and in particular a consequence of viewing literary works roughly as series of words or concepts that can survive the destruction of any collection of copies of them. To the extent that it is a hangover of Platonism, this position should lose its appeal if one accepts the earlier arguments that literary works are, instead, artifacts individuated in part by the particular circumstances of their creation. Apart from a lingering Platonism, one feature of our language might incline some to the view that literary works cannot cease to exist: We often speak not of destroyed or past works, but rather of lost works, as if all that were missing was our ability to find these (still existing) works of ancient, careless, or temperamental authors. This language practice, however, is easily explained without adopting the odd view that works of literature, once created, exist eternally despite even the destruction of the whole real world. The explanation is simply that, because a literary work does not require any particular copy to remain in existence, it is hard to be certain that there is not some copy of the work, somewhere, that has survived, and with it the work of literature. Who knows what may be lurking in the basement corridors of the Bodleian Library? A formerly lost sonnet of Shakespeare's was discovered there not so long ago. Unlike in the case of a unique painting, of which we can find the ashes, we can always hold out hope in the case of a literary work that a copy of it remains in some library, attic room, or perfect memory, so that the literary work might be “found” again. (This is reinforced by noting that, although we ordinarily speak of old or ancient works as lost, in the case of a modern manuscript burned by its author, we are more prone to count the work "destroyed" than merely lost.) But none of this speaks against the idea that, provided all copies and memories of a literary work are destroyed, never to be recovered, the literary work is gone as well—or, to put it another way, the literary work is then lost not in the sense in which sets of keys are lost, but in the sense in which an exploded battleship is lost, or a doctor can lose a patient. If we consider characters to be creations owing their continued existence to the literary works in which they appear, then if all of the works regarding a character can fall out of existence, so can that character. Thus it is a consequence of this view that if all copies of all of the works regarding some ancient Greek heroine have been destroyed, never to be recovered or recalled, then she has fallen out of existence with those works and become a “past" fictional object in much the same way as a person can become a dead, past, concrete object. If we take seriously the idea of fictional characters as artifacts, it seems equally natural to treat them as able to be destroyed just as other artifacts are. Thus fictional characters as well as the literary works in which they appear may fall out of existence with the literature of a culture. One objection that might be raised to the idea that both fictional characters and literary works may fall out of existence is that it seems we can still think of them, refer to them, and so on, even after their founding texts have all been destroyed. But this is no different than in the case of other perishable objects and artifacts: We may still think of and refer to people after they have died, buildings long since destroyed, civilizations long gone by. If fictional characters and literary works cease to exist, I am not suggesting that they then enter a peculiar realm of Meinongian nonexistence or that it is as if such objects never were, but rather that they become past objects just like the other contingent objects around us. The problem of how we can think of and refer to past objects is no small one but is not unique to fiction. Ordinarily, a literary work is maintained in existence by the presence of some copy or other of the relevant text (whether on paper, film, tape, or CD-ROM). It is in this way that the literature of past ages has been handed down to our present day. But even if printed words on a page survive, that is not enough to guarantee the ongoing existence of the work. A literary work is not a mere bunch of marks on a page but instead is an intersubjectively accessible recounting of a story by means of a public language. Just as a language dies out without the continued acceptance and understanding of a group of individuals, so do linguistically based literary works. A literary work as such can exist only as long as there are some individuals who have the language capacities and background assumptions they need to read and understand it. If all conscious agents are destroyed, then nothing is left of fictional works or the characters represented in them but some ink on paper. Similarly, if all speakers of a language die out, with the language never to be rediscovered, then the literary works peculiar to that tongue die out as well. Thus preserving some printed or recorded document is not enough to preserve a literary work - some competent readers are also required. If competent readers and a printed text survive, however, that is enough to preserve a literary work. In other cases, however, we speak of a work of literature as being preserved even if there are no printed copies of the text. In oral traditions, for example, the work is preserved in memory even if it is not being spoken or heard, and (as in *Fahrenheit 451*) it seems that a work could be preserved in memories during times of censorship, even if all printed copies of it were destroyed. So even if a literary work is typically maintained by a printed, comprehensible text, it seems that such is not necessary. A latent memory of the work (disposed to produce an oral or written copy of the work, given the appropriate circumstances) may be enough to maintain it in existence. Thus we can say that, for its maintenance, a character depends generically on the existence of some literary work about it; a literary work, in turn, may be maintained either in a copy of the text and a readership capable of understanding it or in memory.⁹ In sum, it looks as if, if we are to postulate entities that would correspond to our ordinary beliefs and practices about fictional characters, these should be entities that depend on the creative acts of authors to bring them into existence and on some concrete individuals such as copies of texts and a capable audience in order to remain in existence. Thus fictional objects, in this conception, are not the inhabitants of a disjoint ontological realm but instead are closely connected to ordinary entities by their dependencies on both concrete, spatiotemporal objects and intentionality. Moreover, they are not a strange and unique type of entity: Similar dependencies are shared with objects from tables and chairs to social institutions and works of art. Artifacts of all kinds, from tables and chairs to tools and machines, share with fictional characters the feature of requiring creation by intelligent beings. But it might be thought that the way in which fictional characters are created does make them strange, for although one cannot simply create a table, toaster, or automobile by describing such an object, fictional characters are created merely with words that posit them as being a certain way. For example, because characters are created by being written about by their authors, Jane Austen creates the fictional character Emma Woodhouse and brings her into existence (assuming she did not exist before) in writing the sentence: "Emma Woodhouse, handsome, clever, and rich, with a comfortable home and happy disposition, seemed to unite some of the best blessings of existence, and had lived nearly twenty-one years in the world with very little to distress or vex her." But the fact that a character can be created merely through such linguistic acts should cause no peculiar difficulties for a theory of fiction. It has long been noticed that a common feature of so-called conventional or effective illocutionary acts such as appointing, resigning, adjourning, and marrying is that they bring into existence the state of affairs under discussion. Thus, for example, the celebrant of a marriage pronounces a couple husband and wife, a pronouncement that itself creates the couple's new status as husband and wife.10 More recently, it has been noticed that it is a common feature of many cultural and institutional entities that they can be brought into existence merely by being represented as existing. Searle discusses this general feature using money as the example. A dollar bill may read: "This note is legal tender for all debts public and private." But that representation is now, at least in part, a declaration: It creates the institutional status by representing it as existing. It does not represent some prelinguistic natural phenomenon. A contract, similarly, may be created simply by the utterance of words such as "I hereby promise to." Searle even cites as a general feature of institutional reality that institutional facts can be brought into existence by being represented as existing and can exist only if they are represented as existing. What I am suggesting is a parallel with fictional characters: Just as marriages, contracts, and promises may be created through the performance of linguistic acts that represent them as existing, a fictional character is created by being represented in a work of literature. If there is no preexistent object to whom Austen was referring in writing the words above, writing those words brings into existence the object therein described: The fictional character Emma Woodhouse. 13 Thus even the feature that fictional characters may be created not through hard labor on physical materials but through the utterance of words, rather than placing them in a peculiarly awkward situation, points again to their being at home among other cultural entities. Human consciousness is creative. It is that creativity that enables us to increase our chances of survival by formulating plans and examining scenarios not physically before us. It is also that creativity that enables the human world of governments, social institutions, works of art, and even fictional characters to be constructed on top of the independent physical world by means of our intentional representations.14 Nor are fictional characters alone in requiring certain forms of human understanding and practice for their ongoing preservation as well as creation. It has often been argued that works of art in general are not mere physical objects but instead depend both on some instantiation in physical form (in a performance, on canvas, in a printed copy), and – for their intentional properties such as expressiveness and meaning - on the intentional acts of humans.15 Similarly, cultural and institutional facts regarding money, contracts, and property are plausibly characterized as depending not only on certain physical objects like pieces of paper with a certain history, but also on maintaining forms of human agreement. For something to be money, it is not enough that it be a piece of paper with a certain history, it must also, both initially and continually, be accepted as what people collectively agree to count as money in a particular society.16 In sum, fictional characters are a particular kind of cultural artifact, one that depends on human creativity for both its creation and its continued existence. Similar to objects like social institutions or works of art, fictional characters do not belong to a separate, abstract realm but are entities dependent on human intentionality and practice, existing within the ordinary, cultural world we live in.