Tuesday, November 10, 2009

Neurobiology of fictional characters in literature

The Egyptian writer, Alaa Al-Aswany, author of the very successful The Yacoubian Building, was interviewed in Harriet Gilbert’s World Book Club this morning on the World Service of the BBC. His answers to the questions were brief and succinct, indicating a man who has thought deeply about things and thus able to respond briefly and with assurance.

I was especially struck by his comment on the creation of fictional characters, which raises an interesting neurobiological problem, though I cannot figure out a good way to study it.

Al-Asnawy said that, in creating his fictional characters, a moment comes when they acquire their independence and he is no longer in control of them; he merely describes their actions. He put in brief and forceful words what I have often considered, in a somewhat vague way, to be true of some of the great characters in world literature, that the characters are independent of the author, though I never envisaged that there comes a moment, during the writing of the novel, when they acquire their independence. Think of Anna Karenina, or Emma Bovary, or of Julien Sorel in Scarlet and Black. Oddly enough, I have never thought in the same way of Shakespeare, because the power and beauty of his language somehow always reminds me that it is Shakespeare who is writing.

One of the fundamental operations of the brain leads to a capacity to distinguish between self and not self. Indeed, Immanuel Kant thought that this is an a priori with which we are born and into which all experienced is read. I presume that there must be a radical shift in the brain of an author when he or she realizes that the character is no longer their creation but has acquired an independence which they can only describe – a moment when the character becomes detached from the self and becomes the non-self, and when the author knows that the character he or she is creating is separate from them.

I am not sure that I am communicating this well; I am sure that Al-Asnawy would be able to describe it better than me. Nor have I figured out a good way of studying it. But that it must involve a potentially describable shift in neural activity seems very likely. And it shows the power of the arts to point the way to interesting experiments in neurobiology and neuroesthetics.


Phunicular said...

Starting from a somewhat trivial perspective, as soon as I have decided that a fictional character is arachnophobic, that character has acquired a level of independence from my personal arachnophilic sensibilities. I am no longer free to model that character by thinking how I would react to a spider; I have to use my memories of people with phobias, my estimation of how that would make my character behave.

This character modelling can be a fairly clinical, deliberate assessment of possible behaviours and selection of the behaviour that both advances my story and remains consistent with my character's personality. Such a deliberate assessment is slow. At the story level, it's hard to envisage different flows of the narrative if one must enumerate a wide range of reactions for each character at every stage. After working with a particular character for long enough, the character's behaviour becomes intuitive -- my mental model of that other personality is automated enough that I don't have to analyse each situation in detail. The character has made the transition from a "not me" that I consciously construct to a "not me" that is intuitive.

At that stage the character has achieved independence, not because it has become "not me" (it's always been other), but because I've automated the character's personality and I simulate the character's behaviour with very little conscious effort.

(I see this situation as very much like how we simulate and anticipate our friends and loved ones. You can't live with a person for a decade or two without developing an intuitive model for how they'll behave.)

I think the costs of having a character going independent are:
1. it's very hard to change that character, since I don't have direct access to the largely non-conscious process that models him/her.
2. the character can automatically assume aspects of my own personality or aspects of some real person I know without me realising that I've taken a short cut.
3. the character can expose stereotypes I hold of certain "types" of people - my naturally formed abstractions of character types.

I think the development of fictional character independence is merely an example of normal human skill acquisition, and that character independence is akin to an "unconscious competence" of imagining a particular character. If this is the case, then testing techniques applied to other learning processes and human competences could be applied to character creation. I would also expect that the ability to learn to model an independent fictional character would correlate with an author's social ability to predict the feelings and behaviours of associates, i.e., people who are the best at intuiting the behaviour of friends would be the people who find their fictional characters acquire independence faster.

Uno said...

There's some relationship between the topic of your Nov. 10 blogpost (or whatever they're called) about the neurobiology of fictional characters and the topic of your Jan 2 post about artistic creativity, the brain, and the idea that self censorship has to be absent at crucial points in the creative process. Well, in a way, that describes something about how a fictional character takes on a life of its own. As soon as it does so, then the 'author' is absolved of responsibility, there can be no self censorship. Improvisors and those who teach improvisation have long worked on letting this process happen. The actor becomes "inhabited" by such and such a character; in some cultures where trance was supposed to be part of the celebration proceedings, individuals became 'inhabited' by gods or spirits, losing their own inhibition in favour of the incoming entity. If you watch ventriloquists, they sometimes seem surprised at what their charge gets up to. Somehow, these examples all seem to have a common theme - that the fictional character seems to have volition of its own, and the author/host really feels as though they don't (even shouldn't) have to 'try' to make any of this happen, just 'allow' it to.

Margaret Bowker said...

A.S. Byatt said in conversation that her characters were part of a carpet, laid out before her that she had created, and not particularly separate from her; hopefully, I interpreted it correctly. I find the whole subject stimulating because I haven't really examined it before. But I suppose my characters begin to live and are separate from me when they speak. I think of a name, in the connection with a general story line and when that name gels, I make them speak, maybe a page of interaction with someone else and they come alive. Then they will tell me if they are not acting or speaking in character, when I read the chapters through, or record them and listen back to what I have written. Speech is very important in creating character, I feel, and if a character has gained independence, then he or she will have more chance of entertaining and engaging the reader.