Another working paper is online. Title above, downloads, abstract, contents, and introduction below.
Academia.edu: https://www.academia.edu/28941555/Notes_on_the_Unreality_of_Fictional_CharactersSSRN: http://ssrn.com/abstract=2848422
* * * * *
Abstract: We know that characters in fictions are not real but even as critics, we tend to talk about fictional characters in the same terms we use for real characters. This obscures the mechanisms by which fiction works. By looking at several Shakespeare plays and comparing them it becomes obvious that both characters and plot are subordinate to some as yet obscure higher level of organization. We need to develop a language and concepts for dealing with that higher level.
C O N T E N T S
Introduction: They’re Not Real! 1
A. C. Bradley: Character in Shakespeare (Othello) 3
Norman Holland: Cues for the Reader 5
Analyze This: Three Shakespeare Plays 9
Much Ado about Simulation 12
Godshalk: Shakespeare and the Problem of Literary Character 17
Theory of Mind, Not! 20
Introduction: They’re Not Real!
This is one of the topics I took up in my first year of blogging at The Valve, back in 2006, the unreality of fictional characters . By that I mean that we tend to talk about them in pretty much the same way we talk about real people. We’re interested in and comment on their feelings, desires, and motives, we probe the depths of their minds, sometimes using psychoanalytic theory. These days we even worry about their Theory of Mind, as it’s called, and how they probe one another’s minds.
Yes, of course, we know they’re not real. But that’s an observation we make by stepping outside our theories and analytic methods. As Graham Harman has observed here and there, what matters is what our theories can account for, not what we as individuals may also know . To a first approximation we talk about characters in fiction in pretty much the same way as we talk about people in histories and biographies, or newspapers and magazines, or neighborhood gossip. And perhaps that’s the point, gossip, we’re a gossipy species and it turns out we like to gossip about imaginary humans.
That’s find and good for readers, but not so good for critics, who are supposed to be figuring out how fictions work. As William Flesch remarked in Comuppance, in a footnote about literary Darwinists:
…they treat literary characters as motivated by the same things that motivate real humans, rather than as representations to whom real humans react. It’s our reactions that psychology can analyze, not the actions of literary characters. 
Right. But that’s tough to do. While there’s a school of reader response criticism–actually, schools–they’ve done relatively little empirical work (except, perhaps for some of the psychoanalysts, such as Norman Holland and others) on how real readers respond. For the most part, reader response criticism has been about idealized readers and their communities.
While I would love to analyze the reactions of readers to fictional characters, particularly in real time, it’s not clear to me that doing so would fully satisfy my misgivings in this matter. In “Much Ado About Simulation” I point out how our reactions to certain actions in Much Ado About Nothing depend on what we know prior to those actions. The characters are what they are do what they do independently of what we know and when we know it, but our knowledge affects how we react to their actions.
That patterning is important. The issue is not simply about characters as individuals, but their interaction. Is it not their interactions that tell us about them? That same issue is at the center of the previous piece, “Analyze This,” where I compare configurations of characters in Much Ado About Nothing, Othello, and The Winter’s Tale. What emerges in that comparison is a distinction between characters and the functions they play in the plot. For example, in Much Ado, the function of Deceiver is enacted by Don John; that function is taken by Iago in Othello. The Deceiver fools the Protagonist into thinking that his Beloved has been unfaithful. There is no independent Deceiver in Winter’s. In that play, the Protagonist deceives himself.
Those characters are what they are. We can analyze them to our heart’s content. But the only way to see what Shakespeare is doing with them, how he deploys them, is to analyze them in context. Not only that, but to look at the three plays together. That’s when you notice that two plays have Deceivers, one does not. At this point we’re certainly talking about these characters as creatures of the dramas they enact, puppets at Shakespeare’s bidding, if you will, but we’re still not where I want to be, conceptually.
Those two sections flesh this out in more detail, but the problem remains. It’s about something that happens between fictional characters, as a function of overall patterning. Whatever it is that we ‘discover’ when we analyze these characters in the same terms as we use for real people, THAT is subordinate to this overall patterning. That patterning is related to the plot, obviously, but cannot be reduced to it. In fact, whatever THAT is, both plot and character are subordinate to it. How do we talk about THAT?
That’s the direction of this set of remarks. In the first section I look at A. C. Bradley’s discussion of Othello’s character, which is pretty much in terms we would use in describing real people, though he doesn’t go so far as to imagine what Othello might have been doing in his life prior to the actions represented in the play – a practice L. C. Knights lampooned in his well-known essay, “How Many Children Had Lady Macbeth?” This, of course, is something fans do all the time; not only that, but fans will create their own stories using their favorite characters, thus giving them life outside the originating fictions. After that I take a look at Norman Holland’s 1968 account of how it is that readers are justified in projecting (real) life into literary characters and, by implication, how it is that psychoanalytic critics are justified in plumbing the unconscious depths of imaginary minds.
That brings us to the two sections I’ve already discussed, “Analyze This” and “Much Ado About Simulation”. The next section is simply a pointer to an online article in which William Godshalk looks at discussions of Shakespeare’s characters from variety of thinkers ranging from Bertrand Russell to contemporary critics. I then recap the previous discussion and conclude with an appendix in which I rant against Theory of Mind.
 And here’s my first post on the subject, which generated a fair bit of comment: http://www.thevalve.org/go/valve/article/fictional_characters_1_a_c_bradley/
 For example: “Yet it is not a question of whether philosophers are personally 'aware' of this, but of whether their philosophies sufficiently account for it.” Graham Harman, Immaterialism: Objects and Social Theory (Cambridge, UK: Polity Press, 2016), 28.
 William Flesch, Comuppance: Costly Signaling, Altruistic Punishment, and other Biological Components of Fiction (Cambridge, Massachusetts and London: Harvard University Press, 2007) 231.
 FWIW, you can find the complete essay, which I have not read, online at the Internet Archive as part of a collection, Explorations (1949), URL: https://archive.org/details/explorations030001mbp