The Value of Literary Narratives: Subjective Knowledge and Moral Education, Papers of Sculpture

Download The Value of Literary Narratives: Subjective Knowledge and Moral Education and more Papers Sculpture in PDF only on Docsity! Project MUSE" Scholarly journals online Journal of Aesthetic Education, Vol. 42, No. 3, Fall 2008 ©2008 Board of Trustees of the University of Illinois Simulation, Subjective Knowledge, and the Cognitive Value of Literary Narrative SCOTT R. STROUD Introduction Literary narrative holds the power to move individuals to thought, reflection, action, and belief. According to a longstanding view of literature, it is this impact on the reader that leads to literary narrative being valued so highly in our culture and in others. What exactly is the value of literature? Humanists such as Peter Lamarque and Stein Olsen1 argue for this value in terms of perennial themes that literature allows one to “grapple with,” as opposed to cognitivist accounts that discuss the value of literary narrative in terms of conveyed truths or knowledge. One particularly important posi- tion that explains the cognitive value of literature can be labeled the “Sub- jective Knowledge Theory” (SKT). Such an account of literature places its value in the subjective perspectives or experiences the narrative opens up to the reader. Thus, one can gain knowledge of “what it is like” to be caught in the throes of jealousy, endangered by an opposing army, or captivated by the lure of magical prowess. Lamarque and Olsen, however, argue that SKT is fundamentally flawed in its analysis of knowledge derivable from the lit- erary narrative and in terms of its account of literary practice. I wish to defend a version of SKT in this article. In order to do so, I make and defend the following claims. First, Lamarque and Olsen’s critique of SKT relies on a conflation of two distinct types of knowledge—the knowl- edge provided by the simulated subjective experience and its appropriation/ generalization to forms useful for the reader’s real life activities. I will argue that the former of these outcomes is due strictly to the aesthetic features of the literary narrative, whereas the second involves a non-necessary psycho- logical reaction to the text (that is, judgments of the veracity or usefulness of the text for one’s own purposes). By conflating both of these into the cogni- tive knowledge derivable from the text, Lamarque and Olsen’s critique can Scott R. Stroud is an assistant professor in the Department of History and Philosophy at the University of Texas, Pan American. His research interests center on the connection between ethics and aesthetics, and his recent work has explored pragmatism as a way of connecting these two areas of value theory. 22 Stroud objectivity involves an inclusion of multiple perspectives, whereas the subjective experience of perspective is ultimately irreducible.8 If Nagel’s analysis is correct, then literature seems to offer a potent source of an im- portant sort of knowledge—that of what experience is like from a certain individual’s perspective. Thus, SKT in its main permutation argues that lit- erature is cognitively valuable because it allows access to points of view that are outside of our normal subjective reach. By letting us “see” the world through the eyes of another subject, we may grow in terms of what we know about the world, people’s experience of it, and how we should act in terms of this information. Lamarque and Olsen provide a comprehensive argument against such a reading of literary narrative and its value. I will detail three of their im- portant objections to SKT here, as these form the heart of the opposition to subjective knowledge being a valid type of knowledge or a cognitively use- ful sort of knowledge. First, Lamarque and Olsen object to subjective knowl- edge being a legitimate type of knowledge. They argue that it is difficult to conceive of the experience itself as being a sort of knowledge; instead, it is more likely to be the case that knowledge arises from an experience.9 If this is the case, knowledge can be putatively gained using alternate means. In- stead of reading a story about a fictional traveler to the Andes, one can read a travel diary or nonfiction report of such a trip. According to Lamarque and Olsen, experience lacks cognitive value because it simply is not a type of knowledge we commonly recognize. A second objection concerns the dis- tinction between truth and falsity that is claimed to be required by any no- tion of “knowledge.” If subjective experience is to be a sort of knowledge, then there must be ways of telling “true” knowledge apart from apparent (false) knowledge.10 A common way this is done is by some sort of evidence analysis. This will not work in the case of narratives, however, as narratives tend not to provide any evidence outside of the constructed story to substan- tiate any implied claims the description may have to accuracy in represent- ing the world or subjects in it. In other words, what is to say that a particular narrative description of a subject’s experience is accurate? Nothing in the text vouches for this evaluation, and thus nothing in literature qua aesthetic object accounts for such judgment of “true” subjective knowledge.11 A third objection raises a fundamental dilemma in SKT—either the subjective expe- rience is too unique to be useful outside the confines of the text, or the sub- jective experience is useful for other such situations and hence is not strictly a property of the literary narrative.12 Lamarque summarizes this important objection in the following manner: “Either the moral lesson is too close to the work, tied too specifically to the characters and incidents in the work, in which case it cannot function as an independent generalizable moral prin- ciple, or the moral lesson is too detached, too loosely connected to the spe- cifics of the work to be perceived as part of the literary context or meaning that the work expresses.”13 Thus, Othello either teaches us only about the Simulation, Knowledge, and Narrative 23 experience of a jealous individual in that specific context, or it “teaches” us about a general notion of jealousy by using extraneous reader extrapolation. This latter sense of “teach” is obviously rejected by humanists as involving an external purpose and power in terms of analyzing the literary narrative. One does not read Othello for such a general lesson, as such a lesson could be conveyed in other fashions and in other texts (through religious sermons or political propaganda, for instance). Lamarque and Olsen put forward two other arguments against any notion of SKT that involves “conceptual enrichment” via subjective experi- ence. First of all, such a modification of a reader’s concepts would not be unique to literature qua aesthetic object. As they point out, one can get Ib- sen’s general theme about art and the value of life through any one of a num- ber of his plays.14 What is valuable about any given play if another play can teach that same new concept or lesson? For that matter, one could even get this enrichment through some other type of text (perhaps an ethnographic description of an artist and the quality of his or her relationships). Accord- ing to this argument, SKT does not do justice to literature qua aesthetic ob- ject in such talk of conceptual enrichment. A second objection to this ver- sion of SKT is that it fails to distinguish between the recognition of a concept in a fictional situation and the adoption of that concept by an agent in real life situations.15 Again, the appeal is made to why individuals partake in literature—is it merely to change their conceptions? Most individuals do not go to literary narratives with this goal, and the narratives themselves do not contain any argument for such a change in the reader’s concepts. Indeed, Lamarque and Olsen find that the lack of a distinction between recognition and adoption dooms any subjective knowledge account since it will have failed to explain the lack of argumentative or illocutionary force such a text must have to change a reader’s concepts. Why should such “knowledge” result in the change that such proponents of SKT say will happen? Nothing in the text pushes for such an adoption nor seems to warrant such a change in the actual reader. What is common to these objections to SKT is that they conflate two types of knowledge that can be gained from the reading of literature—that of the subjective experience of a character’s perspectival existence with the knowl- edge of values and/or beliefs that may be useful or desirable in the life (or moral education) of the reader. One must notice that the challenge issued by Lamarque and Olsen is to find the moral claim or lesson in the literary nar- rative. This, of course, is unlikely, since what is in the narrative are fictional elements of that specific fictional sequence. Othello does not discuss jealously in general, nor should it—it is a story about particular individuals in particular situations. What opponents of SKT demand is that the subjective experi- ence be shown to contain the morally educative claims or content that sup- posedly benefit the reader of the narrative. Frankly, the proponents of vari- ous versions of SKT have not done much to resist this conflation. Catherine 24 Stroud Wilson, in discussing the distinction between a “‘deep’ way and a ‘shallow’ way of ‘knowing what x is like,’” indicates that the former way of know- ing necessarily entails the adoption of certain concepts—“if he [the reader] understands what it is like to be Newland Archer in the ‘strong sense’, his philosophical conceptions will be necessarily affected.”16 Gregory Currie, discussing how imagination as simulation operates in literature, also con- flates the two, making such claims as “Fictions can help here by inviting us to imagine ourselves more committed than we really are to our values and then to see ourselves, in imagination, flourishing as a result,” and that “engaging in imaginative simulations of hypothetical situations can bring about changes in our values.”17 It is obvious that critics of the SKT approach would question how it is that general knowledge is contained in the specific fictional account of a story. The moral learning that can come from liter- ary narratives must be strictly separated from the type of knowledge that is subjective experience; while the latter can be a useful instigation or means to the former, its status as knowledge does not depend on its necessary con- nection to moral improvement or conceptual change. What will become evi- dent in the following sections is the distinction between these two types of knowledge and how they relate to one possible cognitive value of literary narrative. In order to start such a story, an account must be given as to how literature provides subjective knowledge. Simulation and Subjective Knowledge I now wish to provide an account of how one gets subjective knowledge (knowing what it is like to occupy x’s position) from a literary narrative. Af- ter such an account is provided, the previously mentioned objections to SKT will be addressed. It is my contention that subjective knowledge is gained mainly through the mechanism of cognitive simulation. Theories about how simulation works abound in the research on literature, but they originated in the debates concerning folk psychology. Robert Gordon, in his work in this area, gave the first sustained treatment of simulation as the predomi- nant mechanism by which individuals predict the reactions of others.18 Simulation is postulated as an alternative explanation to persons possess- ing extensive (and unrealistically accurate) theories of how others will react; this is labeled as the “Theory Theory” because it pictures individuals pre- dicting the reactions of others by applying the basic parameters and rules of the theory to a given human’s situation. Due to the impossibly complex nature of such a nomothetic theory, the more appropriate account is that of simulation theory, according to such authors. Some have used this notion of simulation to account for the value individuals take from reading literary narratives. For instance, Susan Feagin finds that simulating the emotions of characters in a story is an integral part to appreciating such a work.19 She provides a detailed account of how simulation plays into the empathetic and Simulation, Knowledge, and Narrative 27 simulation account can clearly gain are steps one and two—it does explain how one can be involved in a specific story and can learn things about cer- tain types of subjective experiences. If one conflates steps two and three, then one can make such objections as Lamarque and Olsen offer. If one does not make this conflation, but instead adequately describes the distinction be- tween these two steps and the consequent types of knowledge, then one has given a defensible account of cognitive literary value. Can this new, modi- fied analysis of SKT meet the objections levied in the previous section? Lamarque and Olsen initially object to subjective knowledge being a type of knowledge at all. What underwrites their claim? It seems to be the assumption that such knowledge must be able to be stated in determi- nant propositional form. Thus, an experience of a situation may give rise to knowledge (“This fire hurts when one touches it”), but the experience itself does not count as knowledge. The only explanation for their reasoning is that “knowledge” involves propositional statements that can be true/false (their second objection), whereas experiences are not something that can be judged in terms of truth or falsity. In my account, however, the experience of what it is like to be x in situation y is a type of knowledge—subjective knowledge. There is a difference between being told about standing on a beach in Bermuda and actually being on a beach in Bermuda. The differ- ence in “feel” between these two experiences must involve some different knowledge on the part of the individual, or a very good description would produce all the same qualia as the actual experience. Indeed, one could see how different propositional accounts that some take as the only candidates for “knowledge” could actually be effects of the sort of experiential knowl- edge one has—what one expresses as known would differ based upon what they know about this situation in a deeper sense. One who has been to the beach in question would be able to say more things about it, and in greater detail, than one who has merely heard about such a beach. Additionally, one who has had simulated experience of that beach will at the least have different things to say about it than one who has read a geological account of that location. While this example of knowledge is not earth shattering in its implications, it still should count as something that one can use in describing some aspect of her world and its experience, either actual or possible. Knowing what it is like to be on a beach is useful insofar as it, like the geological account, allows one to further discuss, think about, and de- scribe some aspect of the world—in this case, that of a subject standing in a certain location. Of course, one could object that this is merely a difference in experience but not in knowledge. It is clear that the difference in the former, however, amounts to or effects a difference in the latter. If one is asked about the sand and its composition on a certain beach, perhaps she will have an an- swer ready at hand. If she is asked what it is like to recline on such a beach 28 Stroud (on which she has not been), she will not be able to give a specific answer. Perhaps she will “infer” from other experiences she has had of beaches, but that is just my point: having been there (on a beach) gives one the ability to answer the question. That ability can be described as having access to some sort of knowledge about the item in question. If one cannot answer such questions, one surely cannot say she “knows” about the experience of the blinding sun on that beach. Being able to reflectively think about and dis- cuss a certain topic is due to that experience, and thus the experience is inte- grally bound up with the issue of knowledge of some aspect of the world. Thus, experience of something, while not a cognitive object itself (that is, a proposition or claim), can be closely connected to things that are of cognitive value (that is, those statements or claims that do describe some aspect of the objective or subjective world). Simulation, while not the exact experience in question, can come close to such an experience, thus giving one a more complete notion of what it would be like to actually be having that experience in a nonsimulated fashion. The focus on propositionality seems secondary to the real issue—both the touching of fire and standing on the beach give one something to know and to put in propositional form. It is not unreasonable to see that “something” as knowledge, and one can see how this is separate from any of the specific determinant propositional statements one can construct about the fire touching example as well (since there is no one statement that exhausts the “knowledge” of such a situa- tion). The different experience is connected with a difference in knowledge since it is also connected to a different, perhaps richer, conception of some part of the world. As for the second objection, it is unreasonable for critics to demand that texts provide an internal delineation of truth and falsity concerning their views. Argumentative texts, such as works of science, give claims and rea- sons for such claims but leave it to the reader to accept such arguments or to verify/falsify them outside of the text. Within a text, there is no way to tell a bad argument from a good (sound) argument—one must necessarily bring in external elements (experience, experiments, other sources, etc.). One can always ask of any argumentative text, what makes this (claim, argument, evidence) true or correct? This is obviously true in terms of the Sokal hoax, as well as in cases where readers blindly trust the statistical analyses of so- cial science writers. Given that scientific/argumentative texts fail to live up to an internal determination of their truth value (as required by this objec- tion to the noetic quality of literature), it is unfair to demand that a literary narrative must go beyond the mere provision of subjective knowledge to the actual proving that this experience actually conforms to reality. The read- er may go outside the text (say, one about whalers) and ask actual sailors about their experience for confirmation, but this is not a failing on the part of the text. What the literary narrative gives one is a simulated experience Simulation, Knowledge, and Narrative 29 of what it would be like to occupy a different subjective position; whether this coincides with the experiences of existent sailors is not the real issue. Is the reader gaining an understanding of what it would be like to be in that subject’s position? I argue that the reader is gaining such a new experience, and such an experience can be described in linguistic terms (although not determinately/exhaustively); thus, it should be counted as a cognitive gain. As individuals acting in life, we have only a finite amount of choices that we can make; each choice precludes (in practice) other choices (at the very least the negation of the choice made). One can become a lawyer, but such a career choice (and all its subchoices of school, place of work, etc.) limits one’s ability to do something else (such as be a brain surgeon). Thus, one will practically never be able to experience certain ways of being, and per- haps some ways of being are very risky to experience (the occupation of a U.S. soldier in World War II, for instance).27 Literary narratives that involve the simulation of subjective experience can be a valuable way to make up for this inherent deficiency because they allow one to experience what such a perspective or life would be like (for example, the life of a soldier as de- scribed in James Jones’s novel The Thin Red Line). Such an experience will involve gaining access to some of that character’s beliefs and/or emotions that are part and parcel of her subjective experience.28 This is what is meant by “subjective knowledge” of a subject’s perspective, and it can be gained through actually living such experiences (a practically limited option) or by simulating such perspectival experiences. One could still object that in “experiencing” some perspective through simulation, one does not have a guarantee that the narrative instigating this experience is accurate in its representation of this experience. In other words, how do we really know that the simulative experiences of the sol- diers in Saving Private Ryan are accurate, or that they approximate the sort of experience real soldiers went through? To answer this challenge we must always insist on parity in what we expect of argumentative texts and narrative texts. We ought not to expect narrative texts to uphold a more rigorous standard than that to which our argumentative texts are held. In this case, one sees the disparity in standards when one asks whether the claims of a certain book, such as a scientific treatise, are true or count as knowledge. Take the case of error: an argumentative text claims x, and x is not the case about the world. The scientific treatise (and its author) could either be mistaken or misleading and deceptive, but either way we know x is false because of some extra-textual appeal. The book clearly claims x, yet we are not convinced of its truth because of real life experiments (including the disciplined ones of experimental science), the testimony of others, and the recorded claims of knowledge we do not doubt. All of this is brought to bear when we determine some claim in a book is wrong, no matter what sort of evidence the author presents in its defense. We detect mistakes and 32 Stroud If one sees this separation, then the knowledge that comes from simulation can be said to be directly derived from the aesthetic qualities of the literary narrative being read (description, plot, characterization, etc.) and would not depend upon the outcome of disputations concerning the argumentative warrants within the text, the conditions under which a reader will adopt a value, and so forth. Lamarque and Olsen also argue that the modification of concepts is not unique to literature and, thus, cannot count as a unique sort of knowledge that literature provides. This objection seems related to the earlier ones concerning the nature of “subjective knowledge.” First of all, my account does not require the clarification of concepts to count toward the gaining of knowledge; instead, one is at least gaining knowledge (which can be thought about and talked about) of what it would be like to be in a certain subjective position. Concepts are not needed in such an account, and any demand that concepts be included would indicate a bias toward knowledge being determinately propositional. Not all knowledge can be narrowed down to clear, limited propositions that can be conclusively called “true” or “false.” Instead, my account allows one to discuss such simulated experi- ences but does not let that simulation itself be exhausted solely by senten- tial utterances—indeed, such actual utterances may be the effect of having knowledge but not the knowledge itself. To have the same simulated expe- rience, and to know about that experience, one would really have to read the text and actually simulate the experience for oneself. Otherwise, one can tangentially approach it in a piecemeal fashion through discursive recollec- tion. What one gets from a text is not a concept (“justice”) or a claim (“justice is for the weak”) but the experience of what it would be like to be a specific individual suffering a specific injustice. This is a cognitive gain; it can be discussed and recounted in a linguistic fashion and is something that the reader would not cognitively possess unless she actually had that exact ex- perience or read that exact fictional account (which involves imagination in its simulative function). The last objection Lamarque and Olsen broached was the distinction between recognizing a concept and adopting that concept. This can be ad- dressed similarly to my discussion of the previous objection concerning the usefulness of what is gained from the text—if one stops at step two, then proponents of SKT can readily grant this distinction. Recognizing a value or concept in a work may come as a direct result of simulating the experience of a character from his or her point of view. This is subjective knowledge insofar as it gives the reader some describable notion of what it is like to be that person with those values (or concepts) in that situation. Step two does not necessitate the adoption of such recognized values or concepts. Proponents of SKT face difficulties due to their insistence that generalized moral learning comes directly from the simulative experience and that it is Simulation, Knowledge, and Narrative 33 the knowledge that is gained. If the distinction is made between steps two and three, then Lamarque and Olsen’s objection can be granted—it simply points out the difference between steps two and three.33 Without the insis- tence that adoption is a necessary part of the simulated experience, their objection disappears. To this point I have argued that literary narrative does hold cognitive value in terms of its provision of knowledge. This knowledge, which is unique to experience, is available in literature in the only form practically accessible to a person given the limitations on experiences and perspectives he or she can actually have. This subjective knowledge is valuable in terms of expanding one’s base of experience and occurs through the aesthetic properties of the literary object in their direct interaction with the reader. The process by which it is argued that this occurs is simulation. Now I pro- ceed to the next step, that of the appropriation of values, actions, beliefs, and so forth from the specific fictional situation to the actual lived situation in which the reader finds or may find herself. Identification and the Appropriation of Literary Values One important divide in modern narrative theory is between those who find the best analysis of narrative in a purely structuralist account and those that utilize a phenomenological description of narrative in terms of the reader’s life projections. The former concentrates on the comprehension of the nar- rative, looking at events in “story” time manipulated in “discourse” time by the narrator, whereas the latter argues for the phenomenology of under- standing the narrative in light of the reader’s future plans and actions.34 The former approach will largely eschew issues of audience identification, whereas the second approach demands such an identification (or judgment) by the audience in the very act of understanding the narrative.35 The ac- count I am giving attempts to chart a middle path between these schools of thought so as to answer the charges against SKT: the audience gains new knowledge from simulations involved in the comprehension of the text, and these textual features can be forces (though not necessitating ones) toward identifying or not identifying with certain characters and their values.36 This section will first delineate what is meant by identification and then re- view some of the criticisms against such a notion of reader-text interaction. The section will conclude with some remarks as to how this notion of identi- fication can avoid such criticisms and be an accurate account of some of the moral knowledge we gain from moral texts. What is meant when it is said that one “identifies” with a character in a literary narrative? It seems to imply that we somehow share the “iden- tity” of the character. What I mean by identification is the process by which a reader finds a character depicted in a novel to instantiate values and or 34 Stroud strategies of action (or belief) that are seen as worthy of acceptance by the reader. I do not mean this in merely a descriptive sense of noticing similari- ties; instead, the sense of identification used here is that in cases of persua- sion and belief change. Thus, a judgment is made in such a case of identi- fication about portrayed values and/or strategies that result in the reader changing, reinforcing, or modifying her own held values or strategies of action. Notice that the reader does not have to already hold these values— what is identical is the value/strategy extrapolated from the character and the value/strategy that the reader judges as desirable (whether the reader presently holds it or not). Walter Fisher, discussing the process of identifica- tion at work in persuasive situations, argues that we identify with an account (and its author) or we treat it as mistaken. We identify with stories or accounts when we find that they offer “good reasons” for being accepted. . . . Reasons are good when they are perceived as (1) true to and consistent with what we think we know and what we value, (2) appropriate to whatever decision is pending, (3) promising in effects for ourselves and others, and (4) consistent with what we believe is an ideal basis for conduct.37 Thus, the aspects of a character and how he or she behaves in the situations in the narrative can lead the reader to identify with those general values or strategies that can be extrapolated from such accounts. I would disagree with Fisher’s claim that the reader identifies with the author of a narrative, as there are many cases of works with an unknown or composite author. I believe the insight that Fisher has right is that the reader can and does make judgments concerning the desirability of the moral lessons to be learned from characters described in a work; these moral lessons take the form of “good reasons” for action and/or belief (as I have been describing them, values and/or strategies for action). We see in the account provided by the literary narrative some type of “reason” for adopting or holding a certain type of value (valuing moral purity over earthly rewards, for instance) or a specific strategy for action (such as “always strike first when it comes to dealing with treacherous opponents”). What makes this represented experi- ence in the literary narrative possessive of a good reason for a certain value? If it reflects values and strategies we find reasonable (or could find accept- able upon the simulation of it), and if it accords with general values and desires we already hold (such as the general wish to be successful in our projects or relationships, etc.), then it can be said to contain good reasons in Fisher’s sense. If a reader gains the simulative experience of a character, say Conrad’s Marlow, she will recognize certain values and/or strategies for action resident in the accounts of him in that fictional narrative. If one recognizes these values as something that may be useful or beneficial for one’s life projects (resoluteness in accomplishing tasks, for instance), then one may judge that they would be useful values to hold. One may make Simulation, Knowledge, and Narrative 37 that certain values and/or strategies evident in the characters in the narrative are identical with her present self-conception (what she wants to be and wants to do) or with a more desirable self-conception (what she sees now as more valuable to be or to do). This last step comprises the stage of identification and can be seen as separate from the provision of subjective knowledge through simulation alone. The concept of identification has come under harsh criticism from Noël Carroll. I will conclude by examining some of his criticisms and the respons- es that I believe a proponent of the account given previously can make. Car- roll’s first objection deals with the conception of identification itself—he ar- gues that supposing that a reader takes up the exact identity of the character with which he or she is identifying is ludicrous. If this were the case, the reader (or audience) would run from the enemies on screen or on the print- ed page instead of amusedly reading onward.45 Instead, Carroll argues that it is better to say that the reader assimilates parts of the character of which she approves.46 At another point Carroll objects that sharing a value with a person is not the same as identifying with that person.47 The proponent of identification, I would argue, is in no sense committed to this literal “iden- tification” of one person with another in terms of their becoming identical. The crucial meaning that this concept gets at is that something important to the character in the story is seen as being equally important to the reader. This is a mimetic relationship, so taking notice of this similarity causes a similar value/strategy to appear in the reader (the gaining of a new value/ strategy) or the reinforcement of a held one (the new strengthening of it from the experience of the character). It is not a mere recognition that per- sons a and b hold the same value c—it is the gaining of new reasons to hold the value one holds with new strength or to start holding that new value as one’s own since it accords with one’s desired self-conception. One does appropriate only some things from a character, but these are vital aspects of the character in terms of how he copes with important situations/issues or in terms of what he values. Another objection Carroll raises against identification concerns the reader’s endorsement of only certain characters. If identification means the taking on of a certain character’s values, etc., why don’t we identify with all the characters? If the answer is because some characters accord better with our values, then Carroll argues that this would preclude any type of learning from this identification (since we only endorse those characters who harmonize with the values we hold).48 First of all, one can respond that we may simulate all characters that are presented in such a way as to require secondary imaginings of the reader, but we are not required to make the judgments involved in identification concerning all characters given. Second, there is room for identification with characters who do not share our current values via the mechanism of higher-level values/self- conception. A reader may learn through a literary narrative that certain ways of treating 38 Stroud friends do not coincide with her higher-level value of having many close friends or with her self-conception of making people feel valued. The iden- tification of a reader’s desired value conception with the conception pro- vided in the narrative is what occurs in such cases of moral learning, not the mere reflection of the current individual’s commitments in the characters of which they approve. In all, I believe that this three-step account of literary narratives and their effects on readers does justice to the demands of critics of SKT, as well as to the insights of those who insist that we learn from literature. What we learn directly from our encounter with the aesthetic properties of the literary nar- rative will be subjective knowledge, derived most often from the process of simulation. Perhaps there are other ways of gaining cognitive value from literature, but I have attempted to argue in this study that simulation offers a path for one important way of accomplishing this. What we can learn in- directly from these specific simulated situations are values and/or strategies that may be of benefit to our moral projects that we either currently hold or those that we come to see as ones in which we should be engaged. This kind of account defends SKT and its analysis of subjective knowledge and does justice to the distinction Lamarque identifies between the literary dimension of a narrative and its moral dimension.49 The error he identifies is the con- flation of the latter with the former, a mistake that many proponents of SKT can easily be said to make when they claim that specific moral knowledge is part of the simulated experience from a specific character’s point of view. In- stead, my account leaves these two dimensions separate and claims that the first dimension (step two in my analysis) gives a reader subjective knowl- edge of what it is like to be in a specific type of situation, whereas the second dimension includes appropriation through identification with something one sees in a certain character as containing good reasons for future action and value-holding. Much of the debate concerning the cognitive value of literature gets muddled simply because too much is conflated in too hasty of a fashion. I have tried to argue that if one keeps important distinctions in place, one can still have SKT and generalized notions of identification with- out violating the practice of reading and benefiting from literary narrative. NOTES John Gibson, Peter Lamarque, Jeanette Bicknell, Pradeep Dhillon, and two anonymous referees for this journal are to be thanked for their helpful comments on earlier ver- sions of this article. 1. Peter Lamarque and Stein Haugom Olsen, Truth, Fiction, and Literature: A Philo- sophical Perspective (Oxford: Clarendon Press, 1994). 2. Noël Carroll, A Philosophy of Mass Art (Oxford: Clarendon Press, 1998). 3. Lamarque and Olsen, Truth, Fiction, and Literature, 225. See also the analyses of narrative time and plot in Wallace Martin, Recent Theories of Narrative (Ithaca, Simulation, Knowledge, and Narrative 39 NY: Cornell University Press, 1986), 81-83, and the discussion of story time and discourse time in relation to the connection of events in a narrative fashion in Sey- mour Chatman, “What Novels Can Do That Films Can’t (and Vice Versa),” in On Narrative, ed. W. J. T. Mitchell (Chicago: University of Chicago Press, 1981), 118. 4. Lamarque and Olsen, Truth, Fiction, and Literature, 368. 5. I do not make any claim that such experiences, etc., can be totally, exhaustively, or determinately captured in linguistic form, only that they can be talked about. 6. Dorothy Walsh, Literature and Knowledge (Middletown, CT: Wesleyan Univer- sity Press, 1969); Catherine Wilson, “Literature and Knowledge,” Philosophy 58 (1983): 489-96. 7. Lamarque and Olsen, Truth, Fiction, and Literature, 371. 8. Thomas Nagel, The View from Nowhere (New York: Oxford University Press, 1986), 5. 9. Lamarque and Olsen, Truth, Fiction, and Literature, 373. 10. Ibid., 371. 11. For a similar discussion of truth in literature, see Christopher New, Philosophy of Literature: An Introduction (New York: Routledge, 1999), 116-21. 12. Lamarque and Olsen, Truth, Fiction, and Literature, 378. 13. Peter Lamarque, “Tragedy and Moral Value,” in Art and Its Messages: Meaning, Morality, and Society, ed. Stephen Davies (University Park: Pennsylvania State University Press, 1997), 62. 14. Lamarque and Olsen, Truth, Fiction, and Literature, 384. 15. Ibid., 384. 16. Catherine Wilson, “Literature and Knowledge,” 494. 17. Gregory Currie, “The Moral Psychology of Fiction,” in Art and Its Messages: Meaning, Morality, and Society, ed. Stephen Davies (University Park: Pennsylvania State University Press, 1997), 54. 18. Robert M. Gordon, The Structure of Emotions: Investigations in Cognitive Philosophy (New York: Cambridge University Press, 1987). 19. Susan L. Feagin, Reading with Feeling: The Aesthetics of Appreciation (Ithaca, NY: Cornell University Press, 1996). 20. Gregory Currie, “The Moral Psychology of Fiction.” 21. Ibid., 51-52. 22. Gregory Currie and Ian Ravenscroft, Recreative Minds: Imagination in Philosophy and Psychology (Oxford, Clarendon Press, 2002), 51. 23. Ibid., 55. 24. This notion of simulation involving imagination running “off-line” processes is also used in Currie’s later work on simulation (see Currie and Ravenscroft, Recreative Minds, 67). In general, this latter work is very similar to the themes in his other article that I use to discuss simulation. As Currie and Ravenscroft put it, both are accounts “interested in the capacity to put ourselves in the place of another, or in the place of our own future, past, or counterfactual self: seeing, thinking about, and responding to the world as the other sees, thinks about, and responds to it” (8-9). 25. Ibid., 55. 26. Ibid., 53. 27. One could even make the argument that certain experiences are theoretically unavailable in real life to individuals—for instance, experiencing life as the op- posite sex or as a lifelong member of another racial or ethnic group. 28. One may object and argue that simulation requires knowledge of the character’s beliefs before the simulation can even begin, thus precluding the learning of any- thing about such a perspective and its experiences. First of all, not all the beliefs of a character are needed; one may learn (via simulation) what the character’s next belief or judgment probably will be, or what beliefs may be very important for him in that situation. Second, this objection does not address two other key elements that one can learn about: emotions/affective response and future action. Even if all beliefs are required before the simulation is run, one can still learn how