There are two distinct currents in the writing of novels that I would like to note in particular: the novel often described as “taut,” in which every word, sentence, paragraph, action, plot point, utterance, and the like has a central meaning utterly important to the meaning of the novel itself as a whole. Flaubert began this school in earnest, and it began somewhat after the other school, which a professor described as a “big bag of stuff,” containing a bit of everything and much that seems extraneous and wandering, though interesting. Dickens wrote such novels. The “big bag of stuff” school has never been my forte: 18th Century English novels like Clarissa and Pamela are a drag, and the hysterical realists who emulate them deserve the opprobrium they occasionally get. I generally prefer the Flaubert method of writers like Flaubert himself, Fitzgerald, Melville, and the like.

One novel that gets the balance nearly perfectly is Neal Stephenson’s Cryptonomicon, which succeeds in being pointed and yet digressive, and its meanderings are always illustrative of the characters and related, somehow, to the central plot—or, rather idea, which in the case of Cryptonomicon I can’t explain without including the ending. It’s an exceptions; John Barth’s The Sot-Weed Factor is another successful hybridized novel I like, which has characteristics of the big-bag novels without many of their faults. The temptation toward big-bag novels is clear, especially because the novel lacks a required form—as the Penguin Dictionary of Literary Terms & Literary says, the novel is “now applied to a wide variety of writings whose only common attribute is that they are extended pieces of prose fiction. But ‘extended’ begs a number of questions,” which it then goes on to enumerate. The problem with defining the novel is that the form itself arose as an original production and one major criterion for greatness continues to be originality, which becomes steadily harder to achieve as more novels are published. One could call this “contamination, as John Barth argues in The Friday Book.

If is by its nature a contaminated genre, then one of its chief progenitors is a sterling example of this general phenomenon. Don Quixote is a pastiche, and not just of allusion, but of poems, stories that would, on their own, qualify as “short stories,” and perceived history. Its eponymous protagonist acts as if stories are histories, and vice-versa; in Chapter VIII, a typical encounter whose broad outlines are repeated occurs with traveling shepherds. Quixote assures them he is a knight, and though they assume him mad, their own reasoning processes aren’t so different from his. These shepherds make questionable assumptions and use false heuristics as well—one says, “ ‘I think, Senor Vivaldo, that we are going to be well repaid for the delay it will cost us to see this famous funeral; for famous it must surely be, judging by the strange things that these shepherds have told us of the dead man and the homicidal shepherdess.” Are “strange things” enough to make a funeral worth attending, a film worth seeing, a text worth reading? Maybe, since the speaker implies that strange things can cause fame.

Fame itself lends some measure of reality to their perception, and their perception adds some measure of reality to the proceeding, as fame itself is an agglomeration of interested parties. I read once that a person is famous to the extent that more people know him or her than he or she knows. By that definition, Don Quixote (and, in italics, Don Quixote) has become very famous indeed; but even the funeral itself, within the text, becomes more important by way of its interest to the shepherds. The shepherds are astonished at Don Quixote, and “were likewise able to perceive the peculiar nature of his madness,” and yet his madness is like theirs, only to a greater degree. To be sure, quantity has a quality all its own, as Stalin infamously said, but nonetheless the principle remains even when the order of magnitude changes.

So it is with all novels: their parts reflect the wholes, in a recursive loop, just as perception can lead to changes in reality. The process is not perfect and doesn’t have a 1:1 correspondence—whether I “perceive” my computer levitating doesn’t make it levitate, and whether Don Quixote perceives King Author to be the figure made out in tales doesn’t mean he was. Yet when I perceive my computer levitating and use such an idea in a story that in turn becomes widely read as a metaphor for how working in a field one loves can make one accomplish more, or when Don Quixote perceives King Arthur to be a historical figure and then acts accordingly, our perceptions have changed and interacted with the real world—as fiction itself does. Umberto Eco writes in Reflections on The Name of the Rose, “However you choose to look at it, I arrived at scholarship by crossing symbolic forests inhabited by unicorns and gryphons […]” In Alan Lightman’s Einstein’s Dreams, different views of the reality of time affect different worlds that might or might not exist in different ways; in one such universe, “The world will end on 26 September 1907. Everyone knows it,” and they react accordingly. In another: “Suppose that people live forever.” In Don Quixote, one could have a false quote, a quote from Don Quixote in a different universe: “Suppose that Don Quixote believed himself to be a knight-errant.” He does, naturally, and its author or narrator says that details mean little, “providing that in the telling of [the story] we do not depart one iota from the truth.” One can’t depart from the truth of a made-up story.

Don Quixote continually emphasizes the “truth” in a way that’s merely ostentacious rather than clever. The book contains a note referencing the fictional layers that Umberto Eco mocked at the beginning of The Name of the Rose, but aside from hyperbole, there is little if any strong sense of mockery here: “He who translated this great history from the original manuscript left by its author, Cid Hamete Benegeli, states that when he came to the chapter dealing with the adventure in the Cave of Montesinos, he found in the margin, in Hamete’s own handwriting, these words: […]” The novel lets us count the layers of narrative contamination: Don Quixote is the principal actor, who is contaminated by Cid Hamete Benegeli, who is contaminated by (potentially) the translator, who is contaminated by Cervantes himself. Given such uncertainty, the need to draw more attention to Hamete’s uncertainty doesn’t have the effect of allaying uncertainty, as the plea for “how it is impossible for me to believe that Don Quixote lied.” Rather, by calling attention to the possibly fictive nature of Quixote’s adventures, he increases their uncertainty, like someone who guiltily overexplains an absence to a lover. Indeed, the very use of “contamination” so many times and in so many subtly different ways expands it the point of near meaninglessness, like Don Quixote’s constant citation of Romance as a drive to defend his numerous acts of folly.

Furthermore, much of the nature of “truth” in Don Quixote depends on personal reputations rather than any attempt at external verification. Don Quixote is believable “since he is the truest gentleman and noblest knight of his age and would not utter a falsehood if he were to be through with arrows.” In an age with no other gentlemen and no other knights, it isn’t difficult to be the truest and noblest—or the least true and least noble, especially without external checks and balances. If I pronounced myself a Ph.D. and proclaimed myself the truest doctor of the age, and by implication my work the most correct, others would correctly look askance at me: it generally takes the verification and seal of others who represent an institution as well as a large body of work to “prove” myself the finest doctor in the land. Conceivably, however, my work could still be the best, even without the external verification, but it would be harder for others to prove. Don Quixote lacks those proofs by others, and yet in his mind, he is still following their examples—and at bottom, he is testifying for himself, and others are believing him because he of his self-created status, not because of a widely agreed upon status. Cid Hamete Benegali is one flimsy shield against such charges—so flimsy, that he will not testify on Quixote’s behalf in Chapter XXIV, despite the myriad of other far more ridiculous events than the relatively benign one described in a chapter concerning “A Thousand Trifling Matters:” “[…] I would state that if the episode has the appearance of being apocryphal, the fault is not mine, and so, without asserting that it is either false or true, I write it down.” But sophisticated readers should assume such things, and understand implicitly such contamination; it, like the many others of its kind, should be assumed by the reader, rather than stated. Instead, it’s used as a form of paralipsis in drawing attention to the fictionality of the world by arguing over or testifying about its fictionality.

Perhaps this is a reflection of the contamination of Don Quixote by history and by legend, and the standards of truth each implies, as well as the standards of translators and others whose standards might be lower still. It is hard to believe Cid Hamete Benegali if he has accepted Don Quixote’s account of himself simply by the account itself; such tautological reasoning is no more persuasive than Don Quixote’s reasoning about the truth of historical romance. Yet perhaps this is besides the point: in a contaminated narrative, what matters is that characters believe and what it causes them to do, not what they believe. Arguing whether the ghosts are real or fake in Henry James’ The Turn of the Screw is of less importance than what those ghosts cause the Governess to do. Although I had not previously realized it, the same general principle animates the novel I spent most of this morning editing, which is tentatively titled A Winter-Seeming Summer’s Night.

Don Quixote still believes in the Romance narrative he lives, and he can only live through misunderstanding the nature of fact and fiction. Cid Hamete Benegali seems to believe Quixote. And yet, all this is contained in a chapter entitled “A Thousand Trifling Matters,” in which Sancho Panza marvels, “ ‘Is it possible that a man who can say as many wise things as you have just said could have told the nonsensical and impossible tale that you did of the Cave of Montesinos? Well, well, we shall see.’ ” Given that Sancho Panza believes them nonsensical, as does Cid Hamete Benegali (“in Hamete’s own handwriting”), we have bookends of disbelief around an event not so different than the many other. Such sections make literal the belief in Romance and demonstrate faulty reasoning more efficiently than the LSAT—for example, a group in white going to pray for rain causes “Don Quixote [to imagine] that this must be some adventure or other” only to have him “strengthened in this belief” by further misinterpreting what he sees. In the second half, he becomes more deeply enmeshed in both the reality of his unreality and in the reality outside the novel, further straining the epistemological ropes pulling his arms in each direction. This is because Quixote doesn’t accept standard explanations for truth. Don Quixote and Don Quixote are both quite famous, and they’re famous for exemplifying and defying the epistemological models we have imposed on the past. In defying them, they nonetheless have others apparently upholding them, but neither matters half so much as the end result: Quixote’s adventures fueled by his belief, and the contaminated beliefs of others. Too bad they never infect me, as I see Quixote as irritating above everything else.