In fiction, readers engage with the characters, and wonder what they are up to. It may even be, as Jennifer Jenkins has proposed, that when we read fiction, theory of mind is a matter of constantly considering in our own minds the range of characters' next possibilities of action, including mental action. In this way, what characters do becomes recognizable, though it can also be satisfyingly surprising.
It turns out that writers have some of the same experience as readers, of finding that their characters do things that seem appropriate, but without the writer having—as it were—to pull the strings. Marjorie Taylor, Sara Hodges & Adèle Kohányi (2002-2003) published a study based on interviews with 50 fiction writers to explore this question. The writers were recruited by advertisements and word of mouth. They ranged from professional writers to people who had not yet published anything. All but four of them reported some experience of characters exhibiting apparently autonomous agency. Writers who had published their work had more frequent and more detailed reports of the phenomenon, which suggests that it is associated with expertise. As compared with the normal population, writers were more likely to have had imaginary companions as children, and they also scored higher on tests of empathy.
Taylor et al.'s conclusion about expertise is also supported by our analyses of 52 Paris Review interviews with very distinguished writers, including 14 Nobel Prize winners (Oatley & Djikic, 2008). We found that 30 of the 33 Paris Review interviewees who were asked a question about whether they made new discoveries in the course of writing said that they did, and these discoveries included characters behaving in ways the writers had not expected.
Taylor et al. cite E. M. Forster's (1927) Aspects of the novel (see our list of books with micro-reviews by clicking here). Forster had this to say:
Taylor, M., Hodges, S., & Kohányi, A. (2002-2003). The illusion of independent agency: Do adult fiction writers experience their characters as having minds of their own? Imagination, Cognition, and Personality, 22, 361-380.
Oatley, K. & Djikic, M. (2008) Writing as thinking. Review of General Psychology, 12, 9-27.
It turns out that writers have some of the same experience as readers, of finding that their characters do things that seem appropriate, but without the writer having—as it were—to pull the strings. Marjorie Taylor, Sara Hodges & Adèle Kohányi (2002-2003) published a study based on interviews with 50 fiction writers to explore this question. The writers were recruited by advertisements and word of mouth. They ranged from professional writers to people who had not yet published anything. All but four of them reported some experience of characters exhibiting apparently autonomous agency. Writers who had published their work had more frequent and more detailed reports of the phenomenon, which suggests that it is associated with expertise. As compared with the normal population, writers were more likely to have had imaginary companions as children, and they also scored higher on tests of empathy.
Taylor et al.'s conclusion about expertise is also supported by our analyses of 52 Paris Review interviews with very distinguished writers, including 14 Nobel Prize winners (Oatley & Djikic, 2008). We found that 30 of the 33 Paris Review interviewees who were asked a question about whether they made new discoveries in the course of writing said that they did, and these discoveries included characters behaving in ways the writers had not expected.
Taylor et al. cite E. M. Forster's (1927) Aspects of the novel (see our list of books with micro-reviews by clicking here). Forster had this to say:
The characters arrive when evoked, but full of the spirit of mutiny. For they have these numerous parallels with people like ourselves, they try to live their own lives and are consequently often engaged in treason against the main scheme of the book. They “run away,” they “get out of hand”: they are creations inside of a creation, and often inharmonious towards it; if they are given complete freedom they kick the book to pieces, and if they are kept too sternly in check, they revenge themselves by dying, and destroy it by intestinal decay (p. 64).What this means for theory of mind and fiction could be something like this. It is not just that we readers imagine ourselves into the minds of characters as we run the simulation which is the literary story. Writers do something of the same, and the possibilities suggested by the contexts of action take precedence over other forces in how the writer imagines the character. Our theory of mind, in other words, is theory-of-mind-in-certain-compelling-social-situations.
Taylor, M., Hodges, S., & Kohányi, A. (2002-2003). The illusion of independent agency: Do adult fiction writers experience their characters as having minds of their own? Imagination, Cognition, and Personality, 22, 361-380.
Oatley, K. & Djikic, M. (2008) Writing as thinking. Review of General Psychology, 12, 9-27.
6 comments:
I recall that in Anthony Trollope's Autobiography he describes walking in the woods and thinking about his characters and the situation he'd devised for them until he knew, or discovered, exactly what they would do. On a somewhat different level, perhaps, I recently watched an interview with mystery novelist Sue Grafton who described a writing rut she had fallen into and said she came out of it when she realized her main character, Kinsey Millhone, "was sulking" because Grafton had been censoring her.
Thank you, Rohan, for this comment. These instances from Trollope and Grafton are very interesting. The Grafton one, in particular, connects both to the quotation from E.M. Forster in this post, and to our earlier post about writer's block. Part of the purpose of fiction, as it seems to me, is to explore what it is to be human. Now, in the process of writing, is discovered a disunity within ourselves that had been only partially recognized: between an apparently controlling and individualistic consciousness (as represented by the writer wanting to get on with a plan) and a strong influence of the goals and idiosyncrasies of other people (as represented by fictional characters) in their predicaments.
I wonder . . . Is this so very different from the situation of a thinker who's working on this or that subject, and the ideas just won't fall into place like they should? They resist the thinker's will and desire and pull her thoughts in unanticipated and, perhaps, undesired directions.
It seems to me that any creative process that's a genuine exploration, whether of imaginary characters in a fictional world, mathematical objects in some abstract domain, concepts in some psychological model, and so forth, is going to have an integrity that "resists" the easy impulses of the creator. Where the domain is that of fiction, this resistance may be experienced as that of a character having his or her own will. When it's an intellectual domain, it's just, well, things aren't working out.
Thank you, Bill. I was not thinking about it like that, but now that you put it in this way, I am sure that you are right: that part of the issue is of the writer working with constraints that are difficult to reconcile, not unlike a mathematician. Constraint-juggling was the major metaphor that Flower and Hayes came up with from all their research on the writing process. I think that must be the centre of it. It's just that after reading Marjorie Taylor's paper, and with my own sense as a fiction writer who has experienced this phenomenon, Rohan's comment prompted the rather sudden intuition of something else, too: a split between an individualist self (the one that plans, tries to get a book finished, or tries to solve a problem in psychology) and another self, perhaps just as important, of the compellingly social kind. So that might mean a more problematic sort of constraint juggling. Or I am too influenced by the rather dramatic nature of Marjorie Taylor's idea of the autonomous agent, with its link to ideas of theory of mind?
Thanks for your remarks, Keith. Though I don't write fiction, I've heard anecdotes of characters "resisting" their authors for decades. I've also got decades of experience trying to wrestle my ideas into shape and, of course, sometimes it's so bad that I just have to give up. But it's only in the last 2, 3, 4, whatever, years that it's occurred to me that they might be aspects of the same phenomenon, hence my comment to your piece.
But your remark prompted me to think of the Freudian unconscious, which is likely to impinge on a novelist rather differently than on a mathematician or a sociologist, or even a literary critic. A character that embodies an aspect of one's unconscious (whatever we mean by "embody" and "unconscious") might well resist authorial control. Beyond this, of course, we commonly talk about ourselves as though the mind were full of agents, some of which obey us, and some of which do not (George Lakoff has an article about this*).
As for theory of mind, some years ago Kelvin Konner registered a protest against the term, suggesting that "This is fascinating stuff and something we need to understand. But a term such as ‘theory of mind’ simply stands in the way. It makes for catchy article titles but conveys no meaning" (Melvin Konner, Bad Words, Nature, 411, p. 743; 14 June 2001). I tend to agree with him. I have some faint hint of an idea of what might be going on in a situation where you ask a four-year-old what a doll knows about what's behind door number three; given generous helpings of such experimental observations, I can simply ignore the "theory of mind" term. But when we're dealing with adults reading or writing fiction and invoke TOM to explain something about imaginary characters, well, at that point it seems to that the term has taken on a life of its own and simply serves to paper over our ignorance.
I mean, if I'm going to use marginal concepts magnified by conceptual tap-dancing and hand-waving, I'd rather go with good old psychoanalysis than this new-fangled theory of mind gizmo.
* Lakoff, G. (1996). Sorry, I'm Not Myself Today: The Metaphor System for Conceptualizing the Self. Spaces, Worlds, and Grammar. G. Fauconnier and E. Sweetser. Chicago, University of Chicago Press: 91-123.
Thank you, Bill. I too have wondered about the relation of autonomous characters to Freudian ideas. The term I have used, "split," comes from psychoanalysis. But I am not sure what else to say in this direction. Although I used to be somewhat skeptical about the importance of theory of mind, I think I have come round. Certainly the term is now firmly established in developmental research. As you pointed out yourself, the idea derives from Piagetian perspective taking, and hence from being able to take the perspective of another. That seems an interesting idea to me.
Post a Comment