In other words, to deceive us, the novel triggers our metarepresentational capacity by alerting us to the source tags of certain representations (i.e., agent-specifying source tags pointing to policemen, garage-fellows, hotel pages, vacationists, salesgirls, the implied reader, Dick, Bill, etc.), whereas to undeceive us, it triggers our metarepresentational capacity by alerting us to the time tags of certain representations (i.e., source tags pointing to Humbert "then" and Humbert "now"). Let us consider some instances of the latter, using as a starting point Phelan's analysis of Lolita in Living to Tell about Lt.
To show that Lolita contains frequent shifts between the perspectives of the pre-"Confessions" protagonist and the protagonist who is now writing his "Confessions," Phelan turns to Humbert's description of his first intercourse with Lolita. This description is introduced by Humbert's claim that he is "not concerned with so-called 'sex' at all," but is inspired instead by a "greater endeavor": to "fix once and for all the perilous magic of nymphets" (134). What follows, however—and belies this claim of higher purpose—is a series of fragmented images highlighting Humbert's sexual "desire . . . and pleasure" as well as his "selfish violence and Dolores's pain."18 Attempting to convey what transpired between him and Lolita through an impressionistic mural that he (ever a creative soul) might have painted, Humbert conjures up a "catalogue of fragments"19 that includes, among other images, "a fire opal dissolving within a ripple-ringed pool, a last throb, a last dab of color, stinging red, smarting pink, a sigh, a wincing child" (135). As Phelan argues,
through the very act of telling his story, the effort of perceiving and misperceiving himself and Dolores, [Humbert] is changing his relation to the story as well as to himself, to Dolores, and to his audience. . . . [During] the first intercourse, he has seen her wincing, stinging, and smarting, and
11: Nabokov's Lolita
during his two years with her, he has seen the kind of suffering that led to her sobs in the night, but during those years, he refused to let those sights affect his behavior. . . . The first time Humbert gives the account of the intercourse, he succeeds in keeping his eyes averted from Dolores's pain [hence his claim that he was not concerned about 'so-called sex']. But [as the images that follow suggest] the act of telling leads him to begin to face much of what he had previously turned away from. The more he allows himself to see, the less he can pursue his exoneration, and so the motive for his telling shifts.20
It shifts to Humbert's increasing willingness to condemn rather than exonerate himself. In contrast to the self-exonerating Humbert—the one who has forced us to see his version of events as coming from other sources throughout the novel—the self-condemning Humbert is a reliable narrator. And it is by following his text—by uncovering, that is, the parts of Lolita that can be traced to this Humbert—that we are able to reconstruct the true story of the relationship between the man and the girl.
Of course, this "present-tense" Humbert, who begins to face Lolita's suffering and thus may regain (at least some of) his readers' trust, does not totally break with the "past-tense" Humbert, who had refused to register those sufferings. Humbert still regularly "reverts to the kind of rationalization that [he] has engaged in before"21 to justify his abuse of Lolita. One important effect of such parallel narratives is, as Phelan observes, to make Humbert the narrator (i.e., "present-tense" Humbert) more sympathetic than Humbert the character (i.e., "past-tense" Humbert):
Nabokov uses this present-tense story and the technique of dual focalization to add a significant layer to the whole narrative: the ethical struggle of Humbert the narrator. The struggle, at the most general level, is about whether he will continue to justify and exonerate himself or shift to admitting his guilt and accepting his punishment. . . . [And this struggle] becomes a significant part of our interest, even as it becomes increasingly painful to see what he sees about his past behavior.22
Phelan believes that "the story of Humbert's gradual move toward greater clear-sightedness is a move toward greater reliability along the axis of evaluation" and that by "the end of the narrative he has stopped trying to hoodwink both himself and his audience and has instead confessed to his crimes against Dolores and condemned himself for them."23 If we indeed allow ourselves (for not all readers do) to trust the "presently
tense" Humbert, we can go back and reread the story looking for the early and not-so-obvious traces of that slowly and painfully emerging reliable narrator. It is then that we realize that the two different appeals to our source-tracking adaptations often do their insidious work side-by-side, sometimes even in the same sentence. That is, the same sentence may prompt us to see certain representations—corroborating the "past-tense" Humbert's version of events—as issuing from other independent sources, while, at the same time, alerting us to the voice of the "present-tense" Humbert competing, so to speak, with the voice of the "past-tense" Humbert.
Let me illustrate this point by returning to (and now quoting in full) one already discussed instance of Humbert's using the minds of strangers to insinuate his view of Lolita into readers' consciousness:
Oh, I had to keep a very sharp eye on Lo, little limp Lo! Owing perhaps to constant amorous exercise, she radiated, despite her very childish appearance, some special languorous glow which threw garage fellows, hotel pages, vacationists, goons in luxurious cars, maroon morons near blued pools, into fits of concupiscence which might have tickled my pride, had it not incensed my jealousy. (159)
It is easy to see how on our first reading we subconsciously use the "testimony" of hotel pages and maroon morons to corroborate Humbert's vision of Lolita. Lacking though they must be that creative insight which allows Humbert to recognize a nymphet when he sees one, these men still cannot help feeling that there is something special about that "little limp" girl and respond accordingly by falling into "fits of concupiscence." We trace what is really Humbert's representation of Lolita as a little oversexed "daemon" to the minds of aroused multitudes and, for the time being, buy that representation wholesale.
Snuggled in the middle of Humbert's ravings about the garage fellows' carnal wishes, is, however, a quiet observation that Lolita in fact had a "very childish appearance." Upon our first reading, the "past-tense" Humbert's assured tone as he divines the thoughts of strangers prevents our realization that with her "childish appearance" Lolita is unlikely to affect people the way that the "past-tense" Humbert claims she does. When we are rereading the novel, however, the description of Lolita as a mere child begins to sound like something that the "present-tense" Humbert might have written and thus "faced" (to use Phelan's insight) for the first time.
Of course, he must have faced it only askance. For the "present-tense"
11: Nabokov's Lolita
Humbert is at this point quite a way from finishing writing his "Confessions" and thus attaining that painful clear-sightedness with which Phelan credits him. Still, the sentence can be profitably read in terms of a tacit tension generated by the two source-monitoring strategies that compete for its overall meaning. When we pay more attention to the distributed sources of representation of Lolita as a nymphet, we are being sold, more or less, on Humbert's lie. When we focus instead on the time tags and think in terms of "Humbert then" vs. "Humbert now," we begin to perceive the text as telling us the truth in spite of itself.
I see the same tension sustained until the very end of Lolita. It is this tension that makes possible very different critical responses to the novel's closing sentences, in which Humbert speaks about "the refuge of art" as the only shared immortality that could be granted to him and his (still his!) Lolita. I have earlier read this sentiment as typical of Humbert's manipulation of his readers' source-monitoring ability. If Humbert wants us to think that "the minds of later generations"—a series of seemingly independent sources—will indeed unite him with "his" Lolita, his project of self-exoneration is apparently far from over, and manipulation and deceit go on. In contrast, Phelan, whose interpretation can be seen as geared more towar
d registering the time tags implicitly present in the novel (i.e., Humbert then vs. Humbert now), reads the same passage very differently. He considers it a "statement of noble purpose," pointing out that "the very last line shows that [Humbert] harbors no illusions about his own redemption: the implication of where he expects to spend eternity—in contrast to where he expects Dolores to spend it—is very clear."24
Is there a way to combine the two readings by trusting and distrusting Humbert at the same time? Sustaining such an ambivalent state of mind is generally challenging, as Dorrit Cohn observes in her analysis of a "historical pattern that recurs time and again in critical responses" to novels featuring unreliable narrators. As she puts it:
A first phase of their reception—sometimes lasting for decades—takes the narrator at his word, in a manner that makes for a fully concordant reading; and a second phase understands this same narrator as discordant— producing a reading that is itself at first received with surprise and disbelief, but that is before long widely accepted. I would propose that this second phase might ideally be followed by a third—one that is actually quite rare in practice: a self-conscious reading that understands the choices involved, a reading aware of the fact that there are choices involved, that the problems created by certain types of narrators—
narrators in whom one can spot incongruities in their evaluation of the events and characters of the story they tell—can be resolved in different
25
ways.
Though Cohn sees the third phase as "quite rare," I wonder if focusing on the ways in which a given text manipulates our cognitive predispositions may make it easier for us to sustain that challenging state of "selfconscious reading." Specifically in the case of Lolita, if we realize that the novel encourages us to gravitate now toward one type of source-monitoring and now toward another (sometimes switching between the two in the same sentence), can we maintain for some time that strange mental stance of simultaneously believing and disbelieving Humbert?
And if we can do it with Lolita, what about Nabokov's other novels, such as The Eye, The Real Life of Sebastian Knight, and Pale Fire ? For it seems that by tirelessly probing and teasing and stretching our tendency to monitor sources of our representations, Nabokov made the cultivation of a mental vertigo in his readers into his trademark as a writer. Will our reading experience change as we gradually articulate the ground rules of the cognitive games that his novels play with us? Will we start putting a premium on consciously prolonging and cultivating those moments of cognitive uncertainty when we both believe and disbelieve, know and don't know, see and don't see?
Though, of course, we are already doing this, or something very close to this, when we are reading fictional chronicles of mayhem and murder, lies and thievery (i.e., narratives more immediately accessible and less disturbing than Lolita). Nabokov's novels are sometimes called "metaphysical" detective stories.26 Let us turn now to the "plain" detective stories and see how the research into the workings of our metarepresentational capacity clarifies the affinity between the two and generally begins to explain the pleasure that we derive from being intensely aware that we are being lied to.
TART 111
CONCEALING MINDS
ToM AND THE DETECTIVE NOVEL: WHAT DOES IT
TAKE TO SUSPECT EVERYBODY?
et us remind ourselves what a strange affair a typical detective novel is. Here is one of the masters of the genre, Dorothy Sayers, on the integrity of the craft:
There you are, then: there is your recipe for detective fiction: the art of framing lies. From beginning to end of your book, it is your whole aim and object to lead the reader up the garden; to induce him to believe some harmless person to be guilty; to believe the detective to be right where he is wrong and mistaken where he is right; to believe the false alibi to be sound, the present absent, the dead alive and the living dead; to believe in short, anything and everything but truth.1
In other words, we open a detective novel with an avid anticipation that our expectations will be systematically frustrated, that we will be repeatedly made fools of, and that for several hours—or even days, depending on how fast one reads—we will be fed deliberate lies in lieu of being given a direct answer to one single simple question that we really care about (i.e., who done it?). Ellen R. Belton observes that the reader of the detective story is motivated "by two conflicting desires: the desire to solve the mystery ahead of or at least simultaneously with the investigator and the desire not to solve it until the last possible moment in order to prolong the pleasures of the mystery situation."2 The desire "to prolong the pleasures of the mystery situation" rings immediately recognizable and true, and yet how can we explain this perverse craving? After all, what is so "pleasurable" about remaining in the dark for a long time about something sinister and threatening that you really, desperately, passionately want to know now? I do not think that many of us would find such a suspended state particularly delectable in real life.
One way to approach this question is to suggest that the enjoyment we derive from whodunits is akin to the enjoyment some people derive from watching/reading suspense thrillers: they get to experience the emotional thrill of danger, of chase, of relief, and then, perhaps, of a renewed danger, all the while remaining safe and warm and not at all threatened by a homicidal maniac posing as a kind next-door neighbor one has known for five years (hmm . . .). Moreover, we can stand being kept in the dark for three hundred pages because we know from our previous experience and from certain cultural conventions associated with this genre that ultimately the mystery will be fully explained. What makes suspense largely unpleasant in real life is that there is no guarantee that we will ever get a complete, or even a partially true, answer to any perplexing question. We can thus enjoy being lied to in the highly structured world of a murder mystery because it offers us a safe setting in which to relieve our anxieties about the uncertainties and deceptions of real life. Or, as Erik Routley puts it, it is the "matter of. . . assurance: it's . . . being allowed for a space to go out of the draught of doubt—that's what the detective story reader thanks his author for."3
I cannot argue with this explanation or with many other fine explanations put forth by literary critics and aficionados of the detective genre in the last hundred or so years. But neither can I pretend to be satisfied with them, for each of them feels incomplete once you start probing deeper. For example, the concept of "relieving-our-anxieties-about-reallife-deceptions" in a safe setting of the novel is useful because it allows us to make some immediate sense of the apparent paradox inherent in our interaction with detective stories, but it does not have any predictive capacity. Postulating that as readers we enjoy dwelling in a state of cruel uncertainty which we would by all means try to avoid in real life implies that, everything else being equal, we should derive pleasure from reading about any activity or about any state of mind that makes us anxious in reality. To a certain limited extent this is true,4 but it sets no boundary condition for its truth. It cannot predict or explain why reliving some of our numerous anxieties in fiction could be a pleasure and reliving others is a nightmare. Even more important, it does not explain why this experience should differ so radically from one reader to another, for plenty of people cannot stand whodunits and thus apparently derive no pleasure, to quote Routley again, from "being allowed" this particular literary "space to go out of the draught of doubt."
2: Reading a Detective Story
This part of the book develops an explanatory framework that can be used to address some of the very basic and yet at the same time very complex issues informing our interaction with detective novels. I approach the question of why some people may enjoy being lied to in the context of the detective narrative by arguing that, although any narrative engages our metarepresentational ability, whodunits tend to "work out" certain aspects of this ability in a rather focused way. I then speculate on the larger implications of this argument, considering the possibility that our genre designations, such
as "detective" or "romance," could be viewed as shorthand expressions of our intuitive awareness that certain texts engage one particular cluster of cognitive adaptations to a slightly higher degree than another. Finally, I check my argument about metarepresentationality and the detective novel against John Cawelti's warning about the dangers of reducing a literary text to psychological factors, and I discuss an important difference between a more traditional psychological approach to fiction and the one made possible in the context of cognitive framework.
-f 2 •
WHY IS READING A DETECTIVE STORY A LOT LIKE
LIFTING WEIGHTS AT THE GYM?
Poirot smiled at me indulgently. "You are Like the little child who wants to
know the way the engine works. You wish to see the affair . . . with the eye
of a detective who knows and cares for no one—to whom they are all
strangers and all are equally liable to suspicion. "
"You put it very well," I said.
"So I give you then a little lecture. The first thing is. to get a clear history
of what happened that evening—alivays bearing in mind that the person
who speaks may be lying. "
I raised my eyebrows. "Rather a suspicious attitude. "
—Agatha Christie, The Murder of Roger Ackroyd, 111
ringing in what we currently know about our metarepresentational
ability can begin to explain our strange hankering for being deceived again and again as part of our experience of reading detective novels. I suggest that detective stories "work out" in a particularly focused fashion our ability to store representations under advisement and to reevaluate their truth-value once more information comes in.1 They push this ability to its furthest limits, first, by explicitly requiring us to store a lot of information under a very strong advisement—that is, to "suspect everybody'—for as long as we can possibly take it and, then, as the story comes to an end, to readjust drastically much of what we have been surmising in the process of reading it.
WhyWeReadFiction Page 16