Intentions and Interpretations: Philosophical Fiction as ConversationSkip other details (including permanent urls, DOI, citation information)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. Please contact firstname.lastname@example.org to use this work in a way not covered by the license. :
For more information, read Michigan Publishing's access and usage policy.
Appeals to the actual author's intention in order to legitimate an interpretation of a work of literary narrative fiction have generally been considered extraneous in Anglo-American philosophy of literature since Wimsatt and Beardsley's well-known manifesto from the 1940s. For over sixty years now so-called anti-intentionalists have argued that the author's intentions – plans, aims, and purposes considering her work – are highly irrelevant to interpretation. In this paper, I shall argue that the relevance of the actual author's intentions varies in different approaches to fiction, and suggest that fictions are legitimately interpreted intentionally as conversations in a certain kind of reading. My aim is to show that the so-called conversational approach is valid when emphasizing the cognitive content of a fiction and truths it seem to convey, for example, in a philosophical approach to fictions which contain philosophical purport using Sartre's fictional works as paradigmatic, and that anti-intentionalists' arguments against intentionalism do not threaten such an approach.
actual intentionalism, anti-intentionalism, conversation, fiction, hypothetical intentionalism, intention, interpretation, philosophy
Appeals to the author's intention in order to legitimate an interpretation of a work of literary narrative fiction have generally been considered extraneous in Anglo-American philosophy of literature since Wimsatt and Beardsley's well-known manifesto from the 1940s. For over sixty years now so-called anti-intentionalists have argued that the author's intentions – plans, aims, and purposes considering her work – are irrelevant to interpretation. "Actual intentionalists," on their behalf, have claimed that the author's intentions are significant and may even determine the meaning of her work. In the broad anti-intentionalism versus intentionalism debate, it has become common to compare literary fictions to everyday conversations to defend one claim or another. Philosophers I call "aesthetic anti-intentionalists," such as Beardsley and Dickie, admit that the author's, or speaker's intentions are, at least to some extent, relevant in interpreting conversations in which they see a practical need for mutual agreement on the meaning of an utterance, but irrelevant in approaching fiction which allows interpretive freedom from its author because of its artistic nature. On the other side, many actual intentionalists, most notably Carroll, suggest that fictions are akin to conversations because of their cultural and linguistic nature, and thus they are to be interpreted as conversation from the author to her audience by appealing to the author's intended meaning.
In this paper, I shall argue that the relevance of the actual author's intentions varies in different approaches to fiction, and suggest that fictions are legitimately interpreted intentionally as conversations in a certain kind of reading. My aim is to show that the conversational approach is valid when emphasizing the cognitive content of the work and the truths it seems to convey, for example, in a philosophical approach to fictions which contain philosophical purport, such as Sartre's novels and plays, and that anti-intentionalists arguments against intentionalism do not threaten such an approach.
2. Intentions and Intentionalisms
The "intention" that the anti-intentionalists and intentionalists debate often remains woefully obscure. For Wimsatt and Beardsley, as it is well known, intention was originally, vaguely a "design or plan in the author's mind." The threefold distinction of intention I apply in this paper is based on Michael Hancher's and Jerrold Levinson's views. Using Levinson's apparatus for conceptual-economic reasons, there is, first, the "categorial intention": what sort of work the author wanted to write, for example, a short story or an essay. Secondly, there is the "semantic intention": what the author meant by the use of her words and sentences. Thirdly, there is the intention missing from Levinson's theory, the Hancherian "final intention": what the author wanted to do or cause by means of her work. The intentionalist arguments as they are known today can be, in turn, distinguished into narrower and broader versions. The narrower version, E. D. Hirsch's theory, which Monroe C. Beardsley labelled "The Identity Thesis," claims that the meaning of a work of fiction is the meaning the author intended in composing it, whereas the broader version argues that the fiction writer's intention is relevant to, or in some sense determines, the meaning of her work.
To begin with, it should be noted that the broader version of the intentionalist argument, which I argue for, does not hold that the correct interpretation of a work of fiction would be completely determined by the actual author. Rather, it holds that the author's intentions are relevant in interpretation. Noël Carroll clarifies well the aims of moderate actual intentionalism (which he calls "modest actual intentionalism"). As Carroll puts it, for a moderate intentionalist, a correct interpretation of a work of fiction is the meaning of the text compatible with the actual author's intention. Moderate actual intentionalism considers only those intentions important for which the text can give support. It does not lead to "Humpty-Dumpty-ism," in which the author could make her work mean anything just because she wills it so. Where the text can support several different interpretations, the correct interpretation is the one compatible with the text and the actual author's intentions. Thus, in moderate actual intentionalism the meaning of a work is constrained by the textual meaning or word sequence meaning and the best information about the author's intended meaning, where available. Best information, in turn, consists of evidence such as the art-historical context of the work, common beliefs of the contemporary audience, the author's public biography, her oeuvre, and the like.
The anti-intentionalist's main objection is that the author's intentions are irrelevant for the interpretation of works of fiction. For the anti-intentionalist, fictions are autonomous entities interpreted only by appealing to their semantic properties; references to the author are claimed to confuse the work and its origin. However, in philosophical approaches to fiction, the author's actual intention and our information of her are highly relevant and desirable in determining the categorial, semantic, and final meaning of her work. For instance, it has been often argued that Sartre used his novels and plays to convey his philosophy, perhaps for an audience broader than the readers of his philosophical works. On the other hand, it has been claimed that Borges, while he wrote philosophically significant works, one of them even applied as a mainstream thought experiment in the philosophy of art, did not intend his works to be works of philosophy. In such categorial debates in which one tries to determine the nature of a certain work, the author's intentions are relevant. An intentionalist interpretation of a fiction can solve questions such as whether the work is an oblique philosophical study dressed in literary form or a philosophical work of art. Naturally, both anti-intentionalists and hypothetical intentionalists allow appeals to the author's categorial intention – say, whether she meant her work to be an essay or a short story –, for anti-intentionalists and hypothetical intentionalists need to recognize the author's actual intention to compose a literary work as a licence to dismiss her semantic intentions concerning the work.
Secondly, while the focus of this paper is on final, not semantic, intentions, the information about the author, such as the beliefs she has expressed elsewhere in her philosophical works, for instance, provide strong evidence for determining her semantic intentions and the textual meaning of her fictions, and in constructing and clarifying the thematic concepts, themes, and other "literary aspects" of her works. An intentionalist interpretation aims at revealing the author's intended meaning: what is expressed in the work. Again, it has to be noted that an anti-intentionalist may consider the author's (declaration of) semantic intentions helpful, but only if they support the interpretation the anti-intentionalist has formulated from the textual, or dictionary, meaning of the work. Nevertheless, moderate actual intentionalism holds that the semantic meaning of fictional utterances is correctly understood in the light of the author's intentions.
Thirdly, the Beckettian–Foucauldian question, what does it matter who is speaking, is different in literary and philosophical approaches to fiction. When a fiction is approached from a philosophical point of view, that is, focusing on its philosophical purport, the author's final intentions manifested in the work and our information about her are important in determining what the work conveys or suggests and which views can be attributed to the author. An intentionalist approach aims at solving whether she, say, "only" portrays a character who expresses certain kind of beliefs or uses the character as a mouthpiece for advancing claims; which of a fictional character's beliefs and views are about the fictional world and which are genuinely supported by the author; how reliable is the narrator, and so on.
3. Achieving Intentions
One of the main reasons for dismissing the actual author's intentions in interpretation has been that they are considered hard or impossible to reach. It has been argued that, for example, in some cases the author of a work is completely unknown, or that in many cases there is no information of the author's intentions at all, and still the fictions can be "understood." Cleanth Brooks, for one, ironically said to have read Andrew Marvell's poem "Horatian ode," not Marvell's mind. The strict version of the unavailability argument, "the metaphysical attack on intentionalism," as Denis Dutton calls it, advances that "intentions proper" are completely inaccessible. The so-called metaphysical anti-intentionalist claims that intentions consist of beliefs and desires, which are objects of mind, and argues that objects of mind cannot be discernible in public, for mind is by definition private. Hence, George Dickie and Kent Wilson go on jeering that even if there were an explicit statement about the author's intention, it would not help an intentionalist, because to understand the declaration the intentionalist would have to know the author's intended meaning in uttering the declaration.
As a moderate intentionalist I do not find the metaphysical attack threatening, for, as I see it, intentions manifest themselves in works. Even an insistent anti-intentionalist like Wimsatt has admitted that the author's intention might somehow "leak into and be displayed in the work." In his article "Genesis: A Fallacy Revived," Wimsatt explains that when talking about the intentional fallacy, he and Beardsley meant to say that "the closest one could ever get to the artist's intending or meaning mind, outside his work, would still be short of his effective intention or operative mind as it appears in the work itself and can be read from the work." On the other hand, even an insistent intentionalist such as E. D. Hirsch has admitted in his manifesto of strict actual intentionalism that there is no direct access to the author's intentions. What one can do is just use the most plausible assumption or hypothesis of what the author meant by her utterance.
Now, moderate actual intentionalism holds that when encountering an utterance, one tries to determine the speaker's meaning: what she meant by what she said. The access to the speaker's meaning is guided by the utterance meaning; the utterance meaning is a key for one's hypotheses about the speaker's intended meaning. Or as Stephen Davies puts it, the speaker's intentions are "successfully and publicly embodied only through their use." While intentions proper are inaccessible, the speaker's intention is usually recoverable from her utterance (including the context of the utterance and, typically, our information about her).
Besides the argument based on the inaccessibility of intentions proper, a common line of argument seems to advance the view that actual intentionalism would like to substitute literary works of art with the author's declaration of intention. Another conventional argument advanced against intentionalism says that the author is far from being a reliable witness for her work; that she may, for instance, manifest her aims in different ways in different situations. As Carroll points out, moderate actual intentionalism does not, however, aim at trading complete works with the author's "compact restatements" of her works. Although moderate actual intentionalism considers the author's statements about her intention valuable, it regards them with suspicion. Declarations of intention are not intentions proper but utterances, and therefore their meaning and reliability is subject to interpretation, as is the work itself. For a moderate actual intentionalist, the author's statements are rather useful keys for hypotheses about the meaning she intended for her work – when they are compatible with the meaning of the text.
Naturally, there are differences in the reliability between declarations uttered in different contexts. Declarations of intention differ from, for instance, Borges's playful interviews to Rand's serious The Romantic Manifesto. Further, the genres in which the declarations are made are more or less regulated by institutions, rule-governed social practices. A rough glance at two institutions, philosophy and literature with regard to the author's declaration of her intention clarifies the reliability of different declarations. Admittedly, the rules of the literary institution are vaguer than those of philosophy. For instance, fiction writers, as other artists, play with conventions and often transgress them. In some instances, the novelists' declarations could be counted rather as parts of their works than serious statements about their purposes.
Here, Genette's theory of paratexts is of use. According to Genette, paratexts, "peritexts" and "epitexts," are devices which help understanding a literary work. By peritexts Genette refers to textual material surrounding the work, such as preface, foreword, the author's notes, and the cover texts. When speaking of epitexts, Genette denotes more distant texts, for instance, the author's interviews, letters, diaries, and manuscripts. Furthermore, as Genette notes, in the literary institution the actual author does not compose the paratexts alone. For example, editors who compose book cover texts, advertisers, and the like partake in constructing the peritexts. And by taking part in the construction of peritexts, they easily take part in constructing the "meaning" and "purpose" of the work. Classifying all the assertions made in peritexts and epitexts as the actual author's genuine declarations of intention would be problematic, for the author's statement about intention might be, say, part of the work or an artistic performance.
The author's declarations made inside the literary institution generally help to construct a correct interpretation of her fictional work. However, it has been argued that sometimes they may misguide or bewilder the interpreter. An author may, for instance, intend her work to be ambiguous and encourage her audience to conflicting interpretations, to not manifest her thematic intentions, or perhaps to celebrate the death of the author. She may also make use of artistic role, or assume different, perhaps even inconsistent personae. Such instances led Beardsley to argue that when the author's statements about her work and the appreciators' observations of it radically conflict, appreciators are eager to turn toward the work and trust the internal evidence they get from it. Nevertheless, Beardsley's objection does not really bother a moderate actual intentionalist. When rejecting the author's statement about her intention as implausible, e.g. ironic, and appealing to the semantic properties of a work, one takes the semantic properties as better evidence of the author's intention than her statement. Moreover, one should keep in mind that, because of their "literary nature," authors' declarations ask for interpretation and judgment. They are not to be discarded outright as the anti-intentionalist demands, nor accepted straight away as the strict actual intentionalist might seek, but assessed critically in relation to the work.
However, there are good reasons for taking certain sort of declarations as reliable, for instance those uttered by so-called philosopher-novelists. By the term I mean an admittedly marginal and very heterogeneous group, writers such as George Santayana, Ayn Rand, Jean-Paul Sartre, Simone de Beauvoir, Umberto Eco, and Iris Murdoch, who are recognized members of both the institutions of philosophy and literature. Now, there are institutional bases for formulating the intuitively accepted and broadly used interpretative practice to decipher a philosopher's literary fictions in the light of her philosophical studies into a conversational approach. The philosopher-novelist's intentions, beliefs and views concerning the philosophical purport of her fictions are recoverable with an epistemically respectable degree of warrant from her declarations and philosophical studies. The reason for this is that in their discourse, philosophers follow certain dialectic norms which ensure that their conversation serves its purpose. Generally, philosophers are assumed to commit themselves to the truth of the propositions they express, provide reasons for their claims, to be honest and to believe in what they assert.
As declarations, philosophical studies also, or especially, demand interpretation when solving what the author intended. Nevertheless, while no declaration, not even a testimony, is necessarily true, there are reasons for taking some declarations to be more reliable than others. For instance, because of his institutional status as a philosopher, Sartre's philosophical studies provide very strong evidence for a moderate intentionalist approaching his fictions from a conversational philosophical point of view. There is no need to stay at hypotheses of the author's aims when there are clear indications for attributing the intentions to the actual author.
4. Conversational Approach to Fiction
Rather than a legitimative factor, the institutional assurance is an indication for a conversational approach. Instead, what makes the conversational approach legitimate is the interpreter's aim. Now, it has been often argued that works such as Sartre's Nausea and Rand's Atlas Shrugged offer at least two readings: one that treats them as literary works, and other that emphasizes their philosophical characteristics, say their views, arguments, and the way the philosophical points are put forward, conveyed, suggested, or implied. These readings are made possible because such works are considered written in both aesthetic (or entertaining) and philosophical aims in mind: they are intended as philosophical fiction.
Philosophical approaches to fiction have been objected to in various ways. Some theorists hold that literary and philosophical readings exclude each other, some that a philosophical reading does great injustice to literary works of art, some that a philosophical approach to fiction is a sort of a category mistake. A common complaint about philosophical approaches is that they treat works of fiction "schematically," failing to understand the aesthetic or literary qualities of them. It has been argued that a "merely" philosophical use of literary works – say, when used as examples in moral philosophy – does not take them seriously as artworks, and often fails to see that aesthetic qualities of literary works also have a philosophical role. The critics have reminded us that even though works like Nausea and Atlas Shrugged are generally taken to advance philosophical claims, they are (also) works of art.
For example, Lamarque and Olsen argue that if a work of fiction is read "as a piece of moral reasoning," the reader does not approach it from a "literary stance," and thus the evaluation of the work will differ from a literary reading. Lamarque and Olsen criticize theories that accept or exhort the philosophical reading of fiction for being philosophers' theories of literature and about the nature of philosophy, not literature. According to them, philosophical approaches to fiction are not concerned with literature as a "separate and independent practice" but rather limit themselves to genres or works which are thought to provide examples for philosophy. Thus, the theories do not clarify the "role and function of literature." Elsewhere, Lamarque objects to the conversational approach to fiction by claiming that philosophers who "import extraliterary paradigms, even linguistic ones, as models of literary interpretation […] are in danger of losing sights of the specificity of these literary functions."
It should be noted that generally philosophical approaches to fiction are not philosophical theories of the literary institution or criticism, but theories of the philosophical function of certain genres or works of fiction. Moreover, the argument of ignoring the literary qualities of a fiction actually questions strictly propositional approaches which violently reduce works of fiction to the author's assertions, failing to see the relevance of fictional narrator(s) and the overall design of the work. However, when a work is taken as a philosophical fiction, an interpretation concentrating on, say, its argumentative aspects and the truth conveyed through the work, especially when there is institutional indication or other reasons for such an interpretation, a conversational reading is highly appropriate, relevant, and, as I shall show, even necessary.
The difference between literary and philosophical approaches is their primary aim and emphasis: the literary interpretation of, say, Sartre's Nausea is in the first place interested in, among other things, explicating and appreciating the thematic content of the work, whereas the conversational philosophical approach is primarily interested in how Jean-Paul Sartre, by means of depicting Roquentin's feelings, attitudes, and thoughts, for example, illustrates the experience of existential angst. Nonetheless, while a work can be approached primarily as a literary or philosophical work, emphasizing different aspects of the work, these readings need not exclude others. One can, for example, approach a work as a literary work of art and simultaneously pay attention to the strategies by which the work conveys philosophical views; one may observe, say, both the way illustrations convey genuine suggestions on philosophical issues and approach the illustrations from a literary viewpoint, paying attention to their literary qualities. Roughly put, if the readings were exclusive, it would take two critics to review a philosophical fiction: one who does the aesthetic job and another who does the philosophical. Nevertheless, as Carroll puts it, one's aesthetic satisfaction does not forestall one's "conversational interests" in works of art. Or as Lamarque himself has later suggested the other way around, while one can read a work of philosophy from a literary point of view, giving prominence to, say, structural or rhetorical aspects of the work, the literary focus on philosophical narratives does not "preclude or supersede or make redundant the focus of a philosopher on the text's arguments, conclusions, validity, and truth." Naturally, Lamarque's claim can be used vice versa.
Finally, philosophical approaches to fiction have been objected to by an institutional argument which claims that philosophy and literature are distinct social practices, and for that reason works of philosophy and works of fiction allow different criteria for interpretation and assessment; that fiction should not be approached from a philosophical point of view at all, for advancing philosophical views is not its (primary) purpose. I do not really find such an argument sound. Although philosophy and literature are different practices, they can be and often have been conjoined. Plato, Kierkegaard, Sartre, de Beauvoir, and Rand, to mention a few, used literary fiction as means for presenting their philosophical views. Their works are both artworks and philosophical works, and can be approached from different viewpoints and with different primary interests. (Sometimes the demarcation line between the two practices is really vague as in, say, Nietzsche's, Bataille's, and Blanchot's works.)
Literary fiction is communication. Fictions are linguistic entities, they consist of words and sentences, and they are composed for an audience to read. The critical questions are, first, what sort of communication is fiction, and second, from whom to whom? Actual intentionalism suggests that art is human action and artworks artefacts made with an aim in the author's mind. Moreover, it advances that, as both aesthetic and linguistic objects, fictions are used to express thoughts and ideas; that they are consciously made objects the author has intended her audience to understand. Carroll, for one, argues for the conversational nature of fiction by claiming that people do not approach works of art as "codes to be deciphered" but as actions and "action-products" made by rational agents who act intentionally.
In turn, anti-intentionalists and hypothetical intentionalists admit the communicative nature of fiction but they deny the conversational nature of fiction. Even though many insistent anti-intentionalists allow intention-appealing in conversations, for example when encountering misspeaking or malapropisms, ambiguous instructions, and written manuals and the like, they strictly reject the relevance of the author's intention in interpreting fiction. Moreover, they often make an ontological distinction between the realm of life, say, conversations and actions, and the realm of art.
While the "aesthetic anti-intentionalists" consider it natural to appeal to what we know of a person, such as her public biography, to supply clues or to help in constructing hypotheses about her intended meaning in everyday discourse, aesthetic anti-intentionalists take all references to the actual author as illegitimate evidence for literary interpretation. Aesthetic anti-intentionalists object to the conversational approach by claiming that linguistic artworks and conversations are different in their communicative nature. Peter Jones, for instance, has argued that conventionally, when classifying a work as a novel, that is, a literary work of art, the readers are conceded "a certain degree of freedom in its interpretation." As Jones sees it, in everyday discourse there is commonly a practical need for speakers to agree on the interpretations of utterances, say, "Sir, you really cannot stop your car at the motorway," whereas in art the practical agreement is given a low priority.
The debate on the legitimacy of the conversational approach to fiction derives from the different standpoints from which the debaters approach the subject. Dickie and Wilson, for example, are interested in an "aesthetic" reading of literary fiction, and therefore object to the conversational approach; from an aesthetic point of view, literary fictions are more than plain "messages" from the author to the audience; they are works of art. Carroll, in turn, pays attention to the cognitive function of fiction, and therefore argues for the conversational approach; while philosophical fictions are artworks, they are used in truth-claiming; they are philosophical literature or literary philosophy. Now, if one accepts that literary fiction is both art and a medium that may be used in conveying knowledge, both sides are correct.
I claim that there are different approaches to fiction allow different criteria for interpretation. As I see it, intentionalism and interpretive pluralism can be conjoined: there can be both a true interpretation and several plausible, legitimate, or apt, interpretations depending on the way a fiction is approached. Stephen Davies, for one, has suggested that there could be several legitimate interpretative approaches to works of fictional literature, and that the "pursuit of truth," or "the truth-targeting interpretation" could be only one of these. As Davies sees it, the truth-targeting interpretation would aim at revealing the author's intentions concerning the meaning of the work and be assessed in terms of truth, while the other approaches, for example, a psychoanalytic or a Marxist interpretation, would be judged for, say, aptness, appropriateness, and suitability to an audience or theory. In the truth-targeting approach, the author's intentions would determine the utterance meaning, and an interpretation of utterance meaning would be true if it recognized the author's intentions. Thus, Davies suggests that the truth-targeting approach to literature would not differ from interpreting conversation.
However, as Davies also argues, while literary interpretations are not "unconcerned with truth," the "truth-conditions" for literature and conversations differ in ways about the "dissimilar functions" these two have. For him, conversation aims principally "at the communication of information," and this is executed only when the participants aim at understanding what they mean by their utterances. Literature, in turn, is in Davies's view a "(sophisticated) form of entertainment not only allowing for but encouraging the exploration of variants of meaning." Davies suggests that such a view of literature is appropriate "given that our interest in literature is typically motivated by purposes somewhat unlike those giving point to our concern with meaning in ordinary discourse."
Davies's account clarifies the question of the legitimacy of the conversational approach. For him, the conversational approach to literary fiction is doubtful given that the purpose of the interpreter differs somewhat from the interpreter's concern with meaning in ordinary discourse. I agree. Literary fiction, as other arts, allows (at least some) freedom in interpretation. Nonetheless, there is a type of interpretation in which the interpreter's purposes do not differ much from the interpretation of conversation. For instance, the conversational philosophical approach – like any approach interested in truths claimed in or suggested by a fiction – is driven by the same purpose as interpretation of any linguistic communication: what did the author mean by her utterance. A strict distinction between different interpretative practices, intentional conversations and intention-irrelevant fictions, is not reasonable in, say, a philosophical approach to a fiction. After all, beside its central aesthetic function, fiction also has other social functions, such as authors' aim at changing readers' beliefs.
For the actual intentionalist, the cognitive goal in interpreting conversations is to figure out what the speaker intends to say. For example, Carroll, leaning on Grice, suggests that on a plausible theory of language, the meaning of an utterance is explicated according to the speaker's intention. Further, Carroll claims that literary interpretation is "roughly analogous," or "on a par," with interpretation of everyday conversation. Similarly, for Robert Stecker, textual works have meaning similar to "other linguistic utterances," and reference to the actual author's intentions plays a pertinent role in determining their meaning.
The legitimacy of a conversational approach to fiction varies in different interpretations. In the conversational philosophical approach to fiction, people have, whether explicit or implicit, interests similar to those they have in interpreting conversations. Philosophical fictions, such as Sartre's works, are conversation from, in the end, the actual author to the actual reader. The cognitive goal in interpreting such works from a conversational philosophical point of view is to figure out what the actual author intends to say by her fictional utterances, such as claims and dialogue exhibited. In the conversational philosophical approach, one encounters Daviesian communication of information, for one is interested in the author's intended philosophical meaning, for example, what Jean-Paul Sartre is doing by means of depicting Roquentin's experience. This is not to say that in the conversational philosophical approach the work is reduced into a philosophical treatise and assessed by the standards of philosophy but rather taken as an artwork that is used to convey philosophical views. Neither does the conversational philosophical approach dismiss the special ways fiction may contribute knowledge, such as illustrating issues from different standpoints.
5. Fiction as Philosophical Conversation
Austinian and Gricean oriented aesthetic anti-intentionalists have objected to the conversational approach further by claiming that the author's mode of speaking releases her from the commitments of "assertive" discourse. They have argued that literary use of language "aestheticizes" language and thus obviates the relevance of the author's, here final, intention in interpretation. The objection is complex, and it can be roughly divided into three parts: first, that the novelist's mode of speaking completely disallows the author's genuine assertions in fiction, second, that even though there are author's assertions in fiction, it is difficult, or impossible, to say which utterances belong to the actual author and which to the narrator or a fictional speaker (or which beliefs belong to the actual author and which to the implied author), and third, that there is no epistemic principle for distinguishing the genuine assertions made by the actual author from those derived by the reader.
Beardsley, who could roughly be placed among the supporters of the first version of the argument, has insisted that illocutionary actions, such as claims exhibited by the author, are actions of the speaker of the work. Further, Beardsley argues that the actual author only represents the actions, while the author of the performances is the speaker, who, as a fictional entity, does not exist outside the work nor has intentions outside it. And as the speaker does not have intentions outside the work, Beardsley claims it is futile to look for them in the life of the actual author. Aesthetic anti-intentionalists have found it hard to explain other than "aesthetic functions" of fictional works because they exclude references to the actual author. Nonetheless, there cannot be irony, satire, or parody without reference to actual beliefs and attitudes; also, by dismissing references to the context of the work, they have ignored the fact that the same utterance gains different meanings in different contexts uttered. The truth is, however, that authors do things with words (of their characters). Authors make, for instance, ethical, historical, political, theological, aesthetic, and philosophical points in their works; they write didactic pieces, satires, and investigations of different sorts. There are genuine illocutionary actions embedded in works of fiction, both the author's assertions and especially assertions conveyed by means of fictional utterances, illustration, thematic statements, and the work as a whole. It is, as Carroll puts it, a "commonly known, openly recognized, and frequently discussed practice in our literary culture."
Whereas aesthetic anti-intentionalism maintains that the author's mode of speaking rules out genuine assertions, the conversational philosophical approach considers them oblique, indirect, implicative, or suggestive. In the conversational philosophical approach, a fiction is interpreted in similar way to fictional utterances, such as "as if" scenarios, in assertive discourses from everyday conversations to philosophy: paying attention to the way the fictional utterance is used as means to an end. As Carroll suggests, the author's mode of speaking does not exclude a conversational interpretation; rather, it encourages one, because the author's actual views are often "implicit" or "implied," and "secured through oblique techniques," conveyed by implications, allegories, presuppositions, illustrations, and so on. In the conversational philosophical approach, one is trying to figure out why the author has constructed a certain sort of character and put words in the character's mouth "in terms of the contribution it makes to the point of the character as an element in the overall design of the work."
Aside from the historical questions about institutions, Plato's dialogues work here as a rough analogy. When approaching them from a philosophical viewpoint, one pays attention to how the characters – who, while they might have real world counterparts, are fictionalized – reason their claims, and why, in some instances, become convinced of Socrates' arguments in the denouement. Hence, Carroll asks, why take for granted the philosophical status of the Athenian playwright's works, fictionalized they are, and yet insist that novelists cannot attribute utterances to their characters? Further, Carroll asks that if the illocutionary representations in the Plato–Socrates case can be legitimately treated intentionalistically, would it not be justified to treat the Dostoyevsky–Alyosha case similarly? Here, I want to remind one of the institutional issues: in Dostoyevsky's case, one has in one's hands hypotheses of the actual author's, a novelist's, intentions derived from the work and biographical information such as diaries whose plausibility and "literary nature" one has to weigh; in the case of, say, Sartre, a philosopher-novelist, one has both hypotheses derived from the work and declarations with an institutionally solid guarantee.
Of course it is also possible that the explicit or implicit philosophical points made in fictions are "merely" constituents of fictional characters or the theme of the work, or that the work expresses some beliefs which cannot be attributed to the actual author. Jerrold Levinson, for one, argues that the parts or elements in novels which Carroll calls nonfictional, such as the philosophy of history in Tolstoy's War and Peace, and which seem to be put forward by the actual author, must first be attributed to "an implied speaker" or a (Beardsleyan) narrator. As Levinson sees it, the "nonfiction" parts are uttered by the implied speaker, a fictional character who describes her (fictional) world. For Levinson, these "essayish portions" cannot logically be attributed to the actual author, for she does not belong to the fictional world and believe in the fictional characters and events, as the implied speaker does. Further, Levinson claims that distinguishing the "narrative and essayistic portions" of a novel to those that pertain to the actual author and those that pertain to an implied speaker the narrator(s) threatens the "artistic integrity" of a work. As Levinson sees it, such a distinction would make the work neither fiction nor nonfiction.
Here, I propose a position between Carroll and Levinson. The conversational philosophical approach does not break the artistic integrity (Levinson) nor destroy the literary aspects (Lamarque) of a work of fiction. It does not literally tear a work into two parts, those that are genuinely asserted by the author and those that are put forward to make-believe. "Essayish" parts, for example, are always part of the fictional story, no matter whether the author asserts or only expresses the assertions they contain. While the conversational philosophical approach is interested in the author's beliefs expressed in the works, and especially the essayish parts where available, it does not attribute these beliefs and parts to the actual author only.
Peter Swirski, for one, argues that although Rabelais's first, i.e. categorial, intention was to write fiction, to tell a story for the audience to make-believe, one could say that he had also "assertive intentions" or that he advanced non-fictional ideas, for example, criticized the church. The literary institution cannot prevent authors from advancing genuine points in their works. Rather, the institution determines how the readers are expected to react to them: to make-believe the propositions and to attribute the points made in the work in the first place to the narrator or the fictional speaker in question. While the first entity, to which the performances, such as utterances, in fiction are attributed, is fictional, the performances may be "in the second place" genuine, that is, also supported by the author. Fictional utterances and assertions conveyed by them can be applied as the actual author's assertions.
Furthermore, hypothetical intentionalists, who rely on the methods of actual intentionalism but maintain hypotheses considering the intentions of the actual or a fictional or a postulated author, should recall that because constructed authors are fictional entities, they cannot make claims outside their world, which is – fictitious. If a fiction is said to claim, put forward, or suggest something, or to express beliefs, the claims and beliefs need necessarily to be attributed to the actual author, taken that there is no reliable authorial declarations or other evidence which would contradict the beliefs expressed, for claims, in order to be genuine claims, need to be put attributed to an human agent.
One should also keep in mind the distinction between the suggestions or reflections the author has made and the suggestions the reader has derived from the work. It has been claimed that there is no epistemic rule for telling when one is speaking of the author's reflections and when of reflections one has derived oneself. As literary fictions play with implication, suggestion, and fictional speakers, the distinction between said and derived is more vague than in philosophers' fictional dialogues, for instance. Thus, the question of the author's genuine assertions is not ontological but epistemological: how to recognize them from those the reader has derived herself. Unfortunately, there does not seem to be a general rule or an epistemological principle which could be applied when trying to decide, whether some (implicit) philosophical point belongs to the actual or to an implied author. Carroll notes that one may have to proceed on a case-by-case basis, relying "on the results of practical criticism (of a sort that at least countenances the applicability of intentionalist hypotheses)."
I agree with Carroll when he argues that it might be better to accept his proposed modus operandi than to conclude that because there is no epistemological principle for detecting the author's assertions straight away, actual authors do not perform illocutionary acts but only represent them. Moreover, the fact that an actual intentionalist interpretation may not always be able to verify its hypotheses about the author's actual intentions in using fictional utterances is just something to be accepted. What Plato, Berkeley, and Sartre intended to convey by their fictional dialogues is a question one has to find out by appealing to the work, such as the style of the narrative, and best information about the author, her aims, and beliefs.
Finally, one may attend an intrinsic philosophical reading of fiction by paying attention only to, say, the arguments the characters employ and the views they have. One can infer philosophical views from a fiction, whether supported by the author or not. Also, keeping in mind the heterogeneity, complexity, and aesthetic delicacy of literary fictions, it would be some sort of a didactic heresy to consider all explicit philosophical views expressed in fiction as the actual author's truth-claiming. Nonetheless, it is another thing to say something about the philosophical signification of a fiction, say, The Brothers Karamazov, than it is to say what the work philosophically means or genuinely suggests. If one is interested in the genuine philosophical meaning of a fiction or the philosophy conveyed through it, one is interested in the meaning given and philosophy conveyed by the actual author.
Literary and philosophical approaches to fictions allow different criteria. A conversational approach to fiction is legitimate when the interpreter aims at solving the nature of a work and especially the points conveyed by it. However, the fictive mode of speaking and the literary qualities of philosophical fictions sometimes make it difficult to distinguish the author's genuine, often implicit, views from those she intended for her characters only. The characters of a fiction might be just plain fictional characters talking just plain fictional talk, not mouthpieces of the author. For example, Roquentin's strongly Sartrean existentialism in Nausea might be just Roquentin's strongly Sartrean existentialism in a certain fictional world. Nevertheless, as I have argued, if the reader aims at solving the truths a fiction seems to convey, a conversational interpretation is not only valid but necessary.
Jukka Mikkonen is Researcher in the Department of History and Philosophy at the University of Tampere (Finland).
Published January 27, 2009
1. My view of the conversational approach is roughly based on Carroll's influential article "Art, Intention, and Conversation," in Intentions and Interpretations, ed. Gary Iseminger (Philadelphia: Temple University Press, 1992). For another view of "conversational" interpretation of literature, see Robert Stecker's "The Role of Intention and Convention in Interpreting Artworks," The Southern Journal of Philosophy, Vol. XXXI, No. 4 (1993), 471–489; ref. on pp. 471–475.
2. Wimsatt, William K. and Monroe C. Beardsley, "The Intentional Fallacy" (1946), in Philosophy Looks at The Arts, ed. Joseph Margolis (Philadelphia: Temple University Press, 1987), pp. 367–380; ref. on p. 368.
3. See Levinson, Jerrold, "Intention and Interpretation: A Last Look," in Intention and Interpretation, ed. Gary Iseminger (Philadelphia: Temple University Press, 1992), pp. 221–256; ref. on pp. 222 & 232–233; Hancher, Michael, "Three Kinds of Intention," MLN, 87, 7 (1972), 827–851; ref. on 829–836.
4. See Carroll, Noël, "Interpretation and Intention: The Debate between Hypothetical and Actual Intentionalism" (2000), in Beyond Aesthetics: Philosophical Essays (Cambridge: Cambridge University Press, 2001), pp. 197–213; ref. on pp. 197–198 & 200–201); Carroll, Noël, "Andy Kaufman and the Philosophy of Interpretation," in Is There a Single Right Interpretation? ed. Michael Krausz (University Park: Pennsylvania State University Press, 2002), pp. 319–344; ref. on pp. 321, 323, 326 & 328. See also Lintott, Sheila, "When Artists Fail: A Reply to Trivedi," British Journal of Aesthetics, 42, 1 (2002), 64–72; ref. on 66–67.
6. See Dutton, Denis, "Why Intentionalism Won't Go Away." Internet: denisdutton.com/intentionalism.htm. Originally published in Literature and the Question of Philosophy, ed. Anthony J. Cascardi (Baltimore: Johns Hopkins University Press, 1987). Read 20 February 2007.
7. See Lyas, Colin A., "Wittgensteinian Intentions," in Intentions and Interpretations, ed. Gary Iseminger (Philadelphia: Temple University Press, 1992), pp. 132–151; ref. on p. 133; Lyas, Colin A., "Anything Goes: The Intentional Fallacy Revisited," British Journal of Aesthetics, 23, 4 (1983), 291–305; ref. on 292–293.
10. See Hirsch, E. D. Jr., Validity in Interpretation (New Haven: Yale University Press, 1967), p. 99. Moderate actual intentionalism does not search for "facsimile pictures" of artists' beliefs, attitudes, experiences, emotions, and thoughts. On the complexity of mental events and the aim of moderate intentionalism see e.g. Wollheim, Richard, Painting as an Art. (Princeton: Princeton University Press, 1987), p. 44. For intentions and stages of artistic creation, see Livingston, Paisley, Art and Intention: A Philosophical Study (Oxford: Clarendon Press, 2005); pp. 47–48 & 165–169.
11. See Carroll, Noël, "Art, Intention, and Conversation," pp. 100–101); Carroll, Noël, "Anglo-American and Contemporary Criticism: Intention and the Hermeneutics of Suspicion," Journal of Aesthetics and Art Criticism, 51, 2 (1993), 245–252; ref. on 251; Carroll, Noël, "The Intentional Fallacy: Defending Myself," Journal of Aesthetics and Art Criticism, 55, 3 (1997), 305–309; ref. on 308. In turn, Colin A. Lyas argues that intentions should not be consiered private mental events distinct from verbal behavior but connected with it and "known only because they are so connected." See Lyas, Colin A., "Personal Qualities and the Intentional Fallacy," in Aesthetics: A Critical Anthology, ed. George Dickie, Richard Sclafani & Ronald Roblin. Second Edition. (New York: St. Martin's Press, 1989), pp. 442–454; ref. on p. 443.
15. See e.g. Barnes, Annette, On Interpretation: A Critical Analysis (Oxford: Blackwell Publishing, 1988), p. 81; Currie, Gregory, "Interpreting Fictions," in On Literary Theory and Philosophy – A Cross-Disciplinary Encounter, ed. Richard Freadman & Lloyd Reinhardt (London: Macmillan, 1996), pp. 96–112; ref. on p. 103; Davies, Stephen, "Author's Intention, Literary Interpretation, and Literary Value," The British Journal of Aesthetics, 46, 3 (2006), 223–247; ref. on 231.
16. See Nathan, Daniel O., "Art, Meaning, and Artist's Meaning," in Contemporary Debates in Aesthetics and the Philosophy of Art, ed. Matthew Kieran (Malden: Blackwell Publishing, 2006/2005), pp. 282–293; ref. on pp. 292–293); Downey, James, "A Fallacy in the Intentional Fallacy," Philosophy and Literature, 31, 1 (2007), 149–152; ref. on 150–151.
17. See e.g. Krausz, Michael, "Intention and Interpretation: Hirsch and Margolis," in Intentions and Interpretations, ed. Gary Iseminger (Philadelphia: Temple University Press, 1992), pp. 152–166; ref. on p. 152.
19. See e.g. Cioffi, Frank, "Intention and Interpretation in Criticism," in Collected Papers in Aesthetics, ed. Cyril Barrett (Oxford: Cambridge University Press, 1965), pp. 165–175; ref. on pp. 173–174.
20. See e.g. Grice, Paul, Studies in the Way of Words (Cambridge: Harvard University Press, 1989), p. 227. I do not think that the author's private or hidden meanings threaten an interpretation of the philosophical meaning of a work. As Stephen Davies suggests, a moderate actual intentionalist does not treat all of the author's intentions, as regards her work's meaning, as on a par: for a moderate intentionalist, fictions belong to the "public sphere"; only communications which are addressed to "appropriate" audiences are the proper subject of interpretation. See Davies, Stephen, "Author's Intention, Literary Interpretation, and Literary Value," 230; Davies, Stephen, "Interpreting Contextualities," Philosophy and Literature, 20, 1 (1996), 20–38; ref. on 25 & 30.
21. Naturally, "philosophical studies" is a complex concept. Although the paradigmatic form of a philosophical study is an article that follows the natural scientific IMRDC structure, the philosophical canon contains, for example, meditations, letters, and poems. Even though philosophical studies are written mostly using the assertive mode of speaking, they do not consist of genuine assertions only. Philosophical texts also make use of suggestions, hypotheses, fictions, overstatements and polemical claims, and so on.
23. See Beardsmore, R. W., "Literary Examples and Philosophical Confusion," in Philosophy and Literature, ed. A. Phillips Griffiths (Cambridge: Cambridge University Press, 1984), pp. pp. 59–73; ref. on p. 60; see John, Eileen, "Literary Fiction and the Philosophical Value of Detail," in Imagination, Philosophy, and the Arts, ed. Matthew Kieran & Dominic McIver Lopes (New York: Routledge, 2003), pp. 142–159; ref. on p. 144; see also Friend, Stacie, "Narrating the Truth (More or Less)," in Knowing Art – Essays in Aesthetics and Epistemology, ed. Matthew Kieran & Dominic McIver Lopes (Dordrecht: Springer, 2006), pp. 35–49; ref. on pp. 35–36.
25. Lamarque, Peter, "Literature," in The Routledge Companion to Aesthetics, ed. Berys Gaut & Dominic McIver Lopes (New York: Routledge, 2001), pp. 449–462; ref. on p. 457; see also Lamarque, Peter, "Appreciation and Literary Interpretation," in Is There a Single Right Interpretation? ed. Michael Krausz (University Park: Pennsylvania State University Press, 2002), pp. 285–306; ref. on p. 291.
27. Lamarque, Peter, "On Not Expecting Too Much from Narrative," Mind & Language, 19, 4 (2004), 393–408; ref. on 397; see also Lamarque, Peter, "Cognitive Values in the Arts: Marking the Boundaries," in Contemporary Debates in Aesthetics and the Philosophy of Art, ed. Matthew Kieran (Malden: Blackwell Publishing, 2006/2005), pp. 127–139; ref. on pp. 132 & 134.
28. Carroll, Noël, "Anglo-American and Contemporary Criticism: Intention and the Hermeneutics of Suspicion," 247; Carroll, Noël, "Art and Interaction," Journal of Aesthetics and Art Criticism, 45, 1 (1986), 57–68; ref. on 59–60. See also Stecker, Robert, Artworks: Definition, Meaning, Value (University Park: The Pennsylvania State University Press, 1997), pp. 168–170.
29. See e.g. Levinson, Jerrold, "Two Notions of Interpretation," in Interpretation and Its Boundaries, ed. Arto Haapala & Ossi Naukkarinen (Helsinki: Helsinki University Press, 1999), pp. 2–21; ref. on p. 18; Levinson, Jerrold, "Hypothetical Intentionalism: Statement, Objections, and Replies," in Is There a Single Right Interpretation? ed. Michael Krausz (University Park: Pennsylvania State University Press, 2002), pp. 309–318; ref. on pp. 317–318); Trivedi, Saam, "An Epistemic Dilemma for Actual Intentionalism," British Journal of Aesthetics, 41, 2 (2001), 192–206; ref. on 193–195; Nathan, Daniel O., "Art, Meaning, and Artist's Meaning," in Contemporary Debates in Aesthetics and the Philosophy of Art, ed. Matthew Kieran (Malden: Blackwell Publishing, 2006/2005), pp. 282–293; ref. on pp. 289–291.
30. See e.g. Beardsley, Monroe C., The Possibility of Criticism (Detroit: Wayne State University Press, 1970), pp. 32 & 34; Davies, Stephen, "The Aesthetic Relevance of Painters' and Writers' Intentions," Journal of Aesthetics and Art Criticism, 41, 1 (1982), 65–76; Olsen, Stein Haugom, "Authorial Intention" (1973), in The End of Literary Theory (Cambridge: Cambridge University Press, 1987), pp. 29–41; ref. on pp. 39–41; Olsen, Stein Haugom, "Modes of Interpretation and Interpretative Constraints," British Journal of Aesthetics, 44, 2 (2004), 135–148; ref. on 146–7; Goldman, Alan H., "Interpreting Art and Literature," Journal of Aesthetics and Art Criticism, 48, 3 (1990), 205–214; ref. on 206–8; Dickie, George & Kent Wilson, "The Intentional Fallacy: Defending Beardsley," 241–242 & 245; Wilson, Kent, "Confession of a Weak Anti-intentionalist: Exposing Myself," Journal of Aesthetics and Art Criticism, 55, 3 (1997), 309–311; ref. on 309; Lamarque, Peter, "Appreciation and Literary Interpretation," pp. 288 & 290; Nathan, Daniel O., "A Paradox in Intentionalism," British Journal of Aesthetics, 45, 1 (2005), 32–48; ref. on 37–38; Dickie, George, "Intentions: Conversations and Art," The British Journal of Aesthetics, 46, 1 (2006), 70–81; ref. on 71.
31. Jones, Peter, Philosophy and the Novel. Philosophical Aspects of Middlemarch, Anna Karenina, The Brothers Karamazov, A la recherche du temps perdu and the Methods of Criticism (Oxford: Clarendon Press, 1975), pp. 182 & 201. For critique of taking fictions as conversation, see Levinson, Jerrold, "Intention and Interpretation: A Last Look," pp. 245–246; Lamarque, Peter, "Appreciation and Literary Interpretation," in Is There a Single Right Interpretation? , ed. Michael Krausz (University Park: Pennsylvania State University Press, 2002), pp. 285–306; ref on p. 291; Livingston, Paisley, Art and Intention. A Philosophical Study (Oxford: Clarendon Press, 2005); ref. on pp. 150–151; and Lamarque, Peter, Philosophy of Literature (Malden & Oxford: Blackwell Publishing, 2008), pp. 129. See also Carroll's response to Livingston's critique in Carroll, Noël "Art, Mind, and Intention," Philosophy of Literature, Vol. 31, No. 2 (2007), pp. 394–404; ref. on p. 404 footnote number 4.
32. See Davies, Stephen, "Relativism in Interpretation," 9; see also Barnes, Annette, On Interpretation: A Critical Analysis, pp. 44–62 & 68–81; Stecker, Robert, "Moderate Actual Intentionalism Defended," Journal of Aesthetics and Art Criticism, 64, 4 (2006), 429–438, ref. on 430.
34. On the difference between the actual intentionalist's and hypothetical intentionalist's views of fiction as conversation, see Carroll, Noël, "Interpretation and Intention: The Debate between Hypothetical and Actual Intentionalism," p. 202 and Iseminger, Gary, "Actual Intentionalism vs. Hypothetical Intentionalism," Journal of Aesthetics and Art Criticism, 54, 4 (1996), 319–326; ref. on 321 & 324–325.
35. See Carroll, Noël, "Art, Intention, and Conversation," p. 97 & Carroll, Noël, "Interpretation and Intention: The Debate between Hypothetical and Actual Intentionalism," 202; cf. Kiefer, Alex, "The Intentional Model in Interpretation," Journal of Aesthetics and Art Criticism, 63, 3 (2005), 271–281; ref. on 273; Lamarque, Peter, "Appreciation and Literary Interpretation," 299; Olsen, Stein Haugom, "The 'Meaning' of a Literary Work," New Literary History, 14, 1 (1982), 13–32.
38. See Beardsley, Monroe C., "Intentions and Interpretations: A Fallacy Revived" (1982), in Aesthetics and the Philosophy of Art – The Analytic Tradition, ed. Peter Lamarque & Stein Haugom Olsen (Malden: Blackwell Publishing, 2005), pp. 189–199; ref. on pp. 190–192; see also Beardsley, Monroe C. , Aesthetics – Problems in the Philosophy of Criticism, pp. 238 & 240. For other brands of this view, see Booth, Wayne C., The Rhetoric of Fiction. Second Edition (Chicago: The University of Chicago Press, 1983/1961), pp. 73–75 and Foucault, Michael, "What Is an Author?" (1977; Qu'est-ce qu'un auteur, 1969), in Language, Counter-memory, Practice – Selected Essays and Interviews, ed. Donald F. Bouchard. Translated by Donald F. Bouchard & Sherry Simon (Ithaca: Cornell University Press, 1986), pp. 113–138; ref. on pp. 129–131.
39. Carroll, Noël, "Art, Intention, and Conversation," p. 108. See also Carroll, Noël, "Art, Practice, and Narrative" (1988), in Beyond Aesthetics: Philosophical Essays (Cambridge: Cambridge University Press, 2001), pp. 63–75; ref. on p. 66.
45. For fictional authors, see e.g. Nehamas, Alexander, "The Postulated Author: Critical Monism as a Regulative Ideal," Critical Inquiry, 8, 1 (1981), 133–149; ref. on 145–147 and Nathan, Daniel O., "Art, Meaning, and Artist's Meaning," p. 255. See also Stecker, Robert, "Apparent, Implied, and Postulated Authors" (1987), in The Death and Resurrection of the Author? ed. William Irwin (Westport: Greenwood Press, 2002), pp. 129–140. William Irwin, an actual intentionalist himself, argues that actual intentionalists also need an "actual intentionalist" author construct, "the urauthor," because it cannot ever reach "the real historical author." See Irwin, William, "Intentionalism and Author Constructs," in The Death and Resurrection of the Author? ed. William Irwin (Westport: Greenwood Press, 2002), pp. 191–204; ref. on pp. 193–194. See also Gracia, Jorge J. E., "A Theory of the Author" (1996), In Irwin (Ed.), The Death and Resurrection of the Author? ed. William Irwin (Westport: Greenwood Press, 2002), pp. 161–190; ref. on pp. 170 & 181. For different sorts of (fictional) intentionalisms, see Trivedi, Saam, "An Epistemic Dilemma for Actual Intentionalism," pp. 201–202, esp. footnotes 17 and 18, and Livingston, Paisley, Art and Intention, pp. 140–141.
47. See Juhl, P. D., Interpretation. An Essay in the Philosophy of Literary Criticism. (Princeton: Princeton University Press, 1986/1976), p. 13; Wollheim, Richard, Art and Its Objects. Second Edition (Cambridge: Cambridge University Press, 1980), pp. 190–191.