Against incomplete story worlds

Go ahead and wonder

Henry V, Act IV. Pistol captures a French soldier on the field of battle. Speaking no French, he says "Come hither, boy; ask me this slave in French what his name is." The boy comes hither and negotiates a ransom; he speaks good French. How and when did this English kid become fluent in this foreign tongue? "Did he learn to speak the lingo from Prince Hal, or from Falstaff in London, or did he pick it up during his few weeks in France with the army?" So asks Ellen Terry in her 1932 lecture “The Children in Shakespeare’s Plays.” Terry, a Shakespearean actor, has been telling of the boy's adventures, which began in Henry IV Part Two; her questions come at the end of her tale, and Terry admits to being "curious about everything that concerns this child for whom I have a great affection." Terry toured this lecture around England and America, and you can imagine sitting in the audience, and listening to her, and warming to her expression of love for one of Shakespeare's characters, and thinking of the characters you have loved in the novels you have read and the plays you have watched.

Meanwhile L. C. Knights, eventually to be King Edward VII Professor of English Literature at the University of Cambridge, is standing in the wings vomiting. Knights complained in a 1933 essay that Shakespeare criticism was treating Shakespeare's characters like real people, and that this was a terrible mistake. He would prosecute this error, and Terry’s lectures would be his exhibit A: “To her the characters are all flesh and blood.”

The allegation, of course, is ridiculous. It goes without saying that Terry was not delusional. She knew fiction from non-fiction and did not think Shakespeare was describing any real flesh-and-blood boy. True, Terry did imagine, when watching the play, that there was a real, flesh and blood French-speaking English boy on that battlefield; but Knights and everyone else did that too. Where is the mistake? Is Knights just talking out of his ass?

Actually Knights’ complaint is not exactly that critics are mistaking fictional characters for real people. His accusation is that critics think that the main aesthetic interest of Shakespeare’s plays is in the characters he created. But Shakespeare was a poet! His poetic use of language is at least as important. And Knights is surely right on this last point. His criticism still feels unfair, though. It’s not like Shakespeare’s characters are all cardboard cutouts useful only for hanging some nice iambic pentameter on. The characters are interesting, even if the language is too. For all that’s been said, the critical writing that so upset Knights sinned only by omission: it failed to attend to something aesthetically important. How bad is that, really, if what it did attend to (the characters) was also aesthetically important? One cannot do everything. Shakespeare’s plays are so stuffed with goodness, it is weird to get all worked up about critics writing about only some of the good things, when there is more than a lifetime’s worth of material there. Knights should have saved the ink he penned his complaints with, and used it to write his own lectures, focusing on what he thought was neglected—the poetry of the plays. Knights does admit, after spending sixteen pages mapping out for us the history of the alleged error, that “what is wanted...is a detailed examination of a particular play” illustrating his favored approach. His pencil was tired, though, and he writes with sadness that “anything approaching a complete analysis is precluded by the scope of the present essay.”

But maybe the problem is not that critics care about the characters when they should care about the poetry. Maybe the problem is that they care about the characters in the wrong way. Maybe Knights was arguing that the questions critics ask about the characters are (often) bad questions. There may be good questions about the boy in Henry V, but (one might think) Terry’s “how did he learn French?” is not one of them. Another (allegedly) bad question gives Knights’ essay its title and has itself come to be shorthand for this thesis: “How many children had Lady Macbeth?” If you find yourself wondering how many children the Macbeths’ had, the idea goes, you are making a mistake.

What mistake? Knights does not say (and indeed never really states the thesis under discussion). But we can try out some hypotheses. Maybe the mistake is asking these questions while you are watching the play. To switch media, if you are watching Titanic, and Jack and Rose have just been thrown from the boat into the North Atlantic, and at that moment your whole mind is taken up with wondering how Jack developed his artistic talents, then yes, you are making a mistake. Certainly at any point in a play (or film or novel), some questions are bad questions to ask, because asking them then will distract you from what comes next (in Titanic, of course, Jack’s supreme sacrifice). The right questions are those that will prepare you for what is coming (will they both fit on that door?). But all this stuff about timing is a red herring. The “don’t ask that now” mistake cannot be Terry’s mistake. When she wonders how the boy learned French, she is onstage lecturing, not watching the play. And Knights thinks her questions are bad “in themselves,” whenever they are asked.

So, again, why is it a mistake to ask (e.g.) how many children Lady Macbeth had? Two more ideas are worth discussing. Maybe the question is bad because the answer is unknowable. It is like asking how many grains of sand there are in the universe, or what Mary Shelly’s fifth word was: some number is that number of grains, and some word is that fifth word, but short of divine revelation no one can know what they are.

A second idea is that the questions are bad because they do not have answers. This idea is popular among philosophers, but (or maybe: and so) is tricky to explain. We do bump into questions-without-answers here and there. “When did the National League abolish the designated-hitter rule?” has no answer, because the rule is still in force. But the cases in fiction where (allegedly) a question has no answer are weirder. The baseball-rule question has no answer because, as it were, the conditions for the question to arise are not met. You cannot ask when something happened unless it has, in fact, happened. But in the fiction cases the relevant conditions appear to be in place. In Macbeth, the Macbeths do have children, and usually that is all that is needed for “How many?” to have an answer. Similarly, so-and-so knows French should be enough for “How did he come to know French?” to have an answer. So how can it be that in fictions these questions lack answers?

Kendall Walton, in Mimesis as Make-Believe, answers with an analogy. Grant that in reality, if Jones has children is true, then “How many children?” has an answer; Jones has N children is true, for some particular number N. Not everything, however, behaves like reality. If Smith believes that Jones has children is true, it does not follow that Smith believes that Jones has N children is true, for some particular number N. Someone’s “belief-world,” so to speak—the way-they-take-things-to-be—can be “incomplete.” In a belief-world it can happen that (i) a statement is true, (ii) that statement’s being true in reality is enough for a question Q to have an answer, and yet (iii) Q has no answer in the belief-world. If belief-worlds can be incomplete, then this kind of incompleteness is not incoherent, or even all that puzzling really; and so we should not be bothered when fictional worlds also turn out to be incomplete.

I agree that the hypothesis that story worlds are incomplete, and so that certain questions about them are bad because they lack answers, is not incoherent. It remains, however, to be shown that this idea is better than the alternative (the first idea): that story worlds are complete—for any statement S, either S is true in the story, or not-S is,—but in many cases we cannot know which it is.

Isn’t the idea that fictional worlds are incomplete obviously the better hypothesis? Fictional worlds are human creations, made up by authors and filmmakers, maybe in collaboration with the audience, as part of a social practice of fiction-making. Great as we human beings are, we just do not have it in us to create worlds so detailed that every question that can be raised about them has an answer. Shakespeare never gave a thought to how the boy learned French, and hardly anyone else did either, and you can easily and happily watch the play attentively—you can easily get a PhD in Henry V studies—without ever entertaining the question. There is no “pressure” for the question to have an answer. So...it doesn’t. Right?

Not so fast. This line of thought tempts when one looks at fictions and fiction-making from the outside, as it were; the corrective is to look from the inside, to think about what one does when one engages with a fiction, and what is required for those modes of engagement to make sense.

To engage with a fiction is to imagine what is fictional; when the actor on stage steps forward and says “Ecoutez: comment etes-vous appele?,” one imagines that an English boy says these words. But to engage with a fiction is to do more than just imagine a bunch of people and events. It is also to imagine responding to these imagined people and events. Some responses are intellectual: if you, like me (and Pistol), know no French, you might imagine wondering what the boy has said. Other responses are emotional: you might imagine fearing for the boy’s—or the French soldier’s—safety, you might imagine admiring the boy’s courage, and—especially having imagined the boy’s many adventures up to this point—you might imagine an affection for the boy growing within you. These responses can snowball, and bounce from one kind to another. In reality, loving someone generates curiosity about them: falling in love, you find yourself wanting to know things about your beloved that other people might think silly or too small to care about. Imagining loving someone (even a fictional character) is no different. This means—and here is the turn—that if imagining loving the boy is an appropriate response to Henry V, and the example of Ellen Terry proves that it is, then so also is a curiosity, borne, like Terry’s, of (imagined) affection, about everything that concerns the boy. If L. C. Knights thought that this curiosity exhibited a defect in one’s response to the play, a kind of over-zealousness in one's imagination, in fact the defect is in his pinched and cramped emotional range.¹

The rest is simple. If (as just argued) it is okay to (imagine) wondering how the boy learned French, then it must make sense to imagine wondering this. But it only makes sense to wonder about questions that have answers. If none of the possible answers to Q are true (as none of “the National League killed the designated-hitter rule on …” are true), then you are doing something wrong in wondering whether Q. What goes for real wondering about reality goes also, I presume, for imagined wondering about fictional worlds: if none of the possible answers to a question Q about a story-world are true-in-the-story, then you are doing something wrong in imagining wondering Q. Since (again) it is okay to imagine wondering how the boy learned French—since this can be part of an appropriate response to the play—it follows that the question of how the boy learned French must have an answer, even if no one does, or can, know what it is.

The argument generalizes beyond this case. Love rationalizes curiosity, but many other emotions do as well; and indeed curiosity about one thing can rationalize curiosity about another. (Imagining) wondering how Lady Macbeth became so ruthless is a good response to have to Macbeth, and wondering whether her experience as a parent played a role is a sensible extension of this wondering. None of this proves that the worlds of Henry V, or of Macbeth, are complete, but it does, I think, push much farther out the frontiers across which their story-world go all fuzzy.

1. Okay, this is rhetorical, not based on any real insight into Knights’ character.