Paloma Atencia-Linares is Research Associate at the Institute for Philosophical Research, UNAM, México. She works mainly on issues concerning pictorial representation, perception and fiction. She is the editor of the British Journal of Aesthetics together with Derek Matravers.
This week at The Junkyard, we're hosting a symposium on Greg Currie’s recent book: Imagining and Knowing: The Shape of Fiction. See here for an introduction from Greg. Commentaries and replies appear Tuesday through Thursday.
* * *
Commentary from Paloma Atencia-Linares.
Greg Currie’s recent book Imagining and Knowing, is an engaging (and imaginative) non-fiction book from which we learn—or which seriously persuades us—that we should be cautious of overestimating the power of fiction to convey knowledge. This is not to say that Currie repudiates the value of fiction as a source of knowledge. On the contrary, because he takes it very seriously, he invites us to be more thorough when endorsing certain views.
There are many things that are worth discussing in this book, but I’ll focus on just two: one very general about the overall project, and another more specific and detailed about the purported difference Currie claims there is between emotions in real-life and emotions in (fictional) stories.
The first part of the book—which is meant to focus on the first term of the book’s title, i.e. “imagination”—takes a step towards answering the question that connects it with the second part of the book—which, in turn, zeroes in on ‘knowing.’ The question is: “How can we learn from an engagement with something [fiction] that depends so crucially on what can seem like the antithesis of knowing—imagining?” (p.15). In order to do this, Currie first tries to establish that fiction actually depends so crucially on imagination, or that there is indeed an essential connection between imagination and fiction. I have my reservations about Currie’s view on the link between imagination and fiction, but even if we grant this connection in the way he proposes, it does not seem that this is what makes it difficult to acquire knowledge from fictional works. After all, it is compatible with Currie’s view that imagination—and fictional utterances, in particular—are present and important in various non-fictional texts and practices (scientific, political, moral etc.) and it is not clear how or whether, to the extent that imagination is constitutive of these utterances, and therefore is also involved in these practices, it hinders in any way the pursuit of knowledge. Nothing Currie says about the functioning of the imagination shows or makes us think that imagination itself affects the acquisition of knowledge, let alone that it is its antithesis. It is true that, unlike the case of beliefs, there is no essential connection between imagination and truth. However, imagining x is neither incompatible with believing x nor with x being true. I don’t expect Currie to disagree with any of this; in fact, he acknowledges that imagination sometimes facilitates (true) beliefs. But if this is so, this may raise a programmatic worry that can be stated as a modified paraphrase of other skeptical views Currie considers: ‘Whatever the problem is, it is not specially to do with the imagination itself because, even if there is an essential connection between fiction and imagination, the problems with the possibility of conveying knowledge may have more to do with the conventions and contingencies of the practice of (traditional) fiction than with something intrinsic to the functioning of the imagination.’[1] In fact, my sense is that most of the problems for fictions to convey knowledge that Currie identifies in the second part of the book have to do with the former rather than the latter.
Moving to the other issue. Currie claims that the conditions of appropriateness of emotions directed at the real world and emotions in fictions are different—at least in important respects. In particular, he claims that while real-life emotions are constrained by the rule of truth—or the rule of the best evidence for the truth (RET), emotions in fiction are governed by the rule of representational correspondence (RRC). So, my response of admiration of Greta Thunberg is appropriate if she really is a courageous young woman, but my response of admiration of Batman cannot be based on whether he is really courageous or not—there’s simply no fact of the matter, only facts of the fiction; hence, my emotional response is appropriate only if the story—the comic or the film—represents Batman as being courageous. Furthermore, Currie claims that emotions of real-life are not subject to the rule of representation (RRC): I can appropriately admire Greta on the basis of a journalist’s testimony of her courage. The testimony is certainly a representation, but it counts as evidence for the truth, so RRC is overruled by RET. This is not the case of emotions in fiction; they are not subject to RET, as Currie puts it “the rule of representation that does apply to fiction is not a rule of evidence (…) how the story represents the situation is not evidence for how things are in the story, they are constitutive of it” (p.63 my emphasis).
Interestingly, Currie mentions that this last statement and presumably also the application of the RRC, as opposed to RET, “is a general truth about stories, fictional and non-fictional” (fn.7, p.63). This is surprising. Certainly, non-fictional stories present people and situations in certain ways according to how the author thinks they are; some of these representations are accurate and some are not and, as competent readers/viewers, we are able to identify what the intended response is. Now, two things require clarification. First, presumably, at least in many nonfictional works, there are indeed some facts of the matter and, while the way the story represents the situation is constitutive of the story itself, it is not constitutive of how the world that this story purportedly represents, is. So, why is RET not applicable here? And, if RET is applicable to regular cases of testimony, why is it not to non-fictional stories? What justifies this asymmetry? It’s certainly true that many non-fictional stories cannot count as reliable sources, but some of them presumably do—I’m sure most people’s emotional assessment of Greta Thunberg is based on news stories, and this is not because they think that’s the story’s intended reaction, but because they take these stories as the best evidence they have on Thunberg’s character.
Another interesting case to test Currie’s view can be non-fictional stories that include (purportedly) fictional sections—according to Currie, stories are a “mixture of fiction and non-fiction” (p.21). One work that may illustrate this idea is Errol Morris’ The Thin Blue Line. In this film, Morris introduces sequences that consist of re-enactments that simulate crime-scenes. The purpose of these re-enactments is to vividly show the discrepancies in the accounts of witnesses, suspects, and police and, ultimately, to make a case for the claim that Randall Adams was very likely sentenced to life in prison for a murder he did not commit. When viewers see these reenactments, given the purported injustice that they seem to show, they may feel surprised, angry or frustrated. Following Currie, these sections should be considered fictional utterances and hence, emotions generated under their influence should be governed by RRC. But this doesn’t seem to be right. Despite the full dramatization of events that may or may not have occurred in the precise way depicted, and the overall invitation to imagine the content, these sequences play an evidential role—in fact, they were used in court to support Adams’ innocence. This evidential role, moreover, is not only at the level of the facts presented, but at an emotional level as well: it seems appropriate or rational that someone who previously had negative feelings towards Adams, revise them in the light of this (fictional?) evidence. If this is correct, emotions in response to fictions are sometimes also subject to the rule of evidence.
Admittedly, although Currie explicitly claims that the reality test or rule of evidence does not apply to emotions in fictions even when the objects of such emotions exist (p.61), he also acknowledges that “considerations of truth sometimes do and should constrain the responses of the spectator” (p.64). But this, he suggests, should happen when we have reasons to think that emotional reactions generated within the fictional context can be carried-over to the real life domain with negative consequences, e.g. when we’re asked to sympathize with questionable characters or situations (fictionalized versions of real people), take attitudes to ethnic groups contrary to our beliefs, or exploit our emotional reactions to overlook aesthetic flaws and believe a work is aesthetically better where it is not. But the example I provided is different. In this case, it seems there is good reason to apply the rule of evidence in lieu of the rule of representational correspondence because the emotional reaction within the fictional context generates potential positive consequences when carried over to the real-world. If this is so RET and RRC do not only not neatly distinguish emotions in the real-world and in fiction, RET might be much more pervasive than what Currie is willing to admit.
[1] I say “traditional fiction” since media such as videogames may facilitate knowledge given different contingencies and conventions governing such practice. This seems to be the case if we look at the empirical evidence compiled, for example, in Plass, J.L., Meyer, R.E and Homer, B. D. (2019) Handbook of Game-Based learning, The MIT Press.
Response from Greg Currie.
Thanks to Paloma Atencia-Linares for comments that link the separate parts of Imagining and Knowing. She queries my way of connecting the later parts of the book, on knowledge, with the first part, on imagination. I suggested that claiming to get knowledge from fiction is made problematic by fiction’s dependence on imagination, echoing what I take to be a widespread view with at least some plausibility. A good deal of work has been done recently (by Kind and Williamson for example) to improve the epistemic reputation of the imagination. But I did, by considering the mechanisms for forming conditional belief (Chapter 8), try to identify relevant ways in which the imagination might lead to ignorance and error. So I don’t agree that “Nothing Currie says about the functioning of the imagination shows or make us think that imagination itself affects the acquisition of knowledge.”
Paloma thinks that the problem lies not with the imagination but with the contingencies of history (her italics):
…the problems with the possibility of conveying knowledge may have more to do with the conventions and contingencies of the practice of (traditional) fiction than with something intrinsic to the functioning of the imagination.
We should not be misled by the fact that plenty of fiction has been epistemically worthless or of negative value. If we haven’t learned much perhaps that’s because we don’t have many of the right kind of fictions.
It’s common ground in the debate over the cognitive value of fiction that we learn at most from a small subset of fictions. Friends of learning from fiction may think that, taking all the fictions and all the exposures to fiction, the net result is a positive one for learning. That strikes me as an impossible calculation—not that Paloma is advocating it. A second route avoids summation and identifies particular works for celebration, ignoring the rest. I tried in Part Three to address this strategy by providing “quality proof” arguments: ones that could not be deflected by saying they apply, at most, to works lacking the right kinds of merits. Another way would have been to speak directly to those fictional works so often cited as epistemically valuable: Shakespearean drama and the great novels of the last two centuries are the usual suspects. I didn’t. I was trying to create a framework within which that sort of discussion might take place; we have a lot of commentary designed to persuade us of the merits of certain works, where it is hard to distinguish the contribution of the work from that of the advocate.
In chapter 4 I asked when emotions in response to fiction are appropriate. The answer, I said, is distinct from that for the emotions of “real life”. Paloma takes me to be saying that emotions in response to non-fictional stories are, in respect of appropriateness, like emotions in response to fictional ones. In both cases it depends on how the subject is represented—in which case it would be appropriate for me to be deeply admiring of Stalin while reading a party-hack’s biography.
What I said was that fictional and non-fictional stories are alike in that “things are in a story as they are represented to be”; how things are according to the biography is one thing, how they really are is another. It doesn’t follow that fictions and non-fictions are alike in the appropriateness conditions for emotions. But I agree that more needs to be said on this topic. Cases get complicated when fictions invoke real people, as in Stacie Friend’s excellent example of prosecutor Garrison in JFK. I said there is a sense in which it is appropriate to admire Garrison in the context of the film, though admiring him in real life would be crazy, given the facts about his behaviour. Is there any sense in which it would be appropriate to admire Stalin while absorbed in the biography? (Assume for simplicity that you know what a monster Stalin was.) No, you will say. But I think you will say the same about admiring him when he appears in a party hack’s admiring novel. Is it all right, then, to admire Garrison in JFK? If so, is that because Garrison was so much less bad a person? What about the novel that imagines, tragically, the useful life Stalin might have lead? Feeling admiration for Stalin-so-represented does not seem obviously wrong—not as wrong anyway as admiring him as-represented-in-the biography or as-represented-in-the-hack’s-novel. Paloma’s comments made me realise the inadequacy of what I have said on this complex topic.