A year or so ago, I listened to a Tin House podcast, a recording of a fascinating lecture on "defamiliarization" by Anthony Doerr, who usefully lays out a number of the themes of his talk in an interview in Fugue (and don't miss this great handout):

I argue to my students that (in most cases) verbal repetition has a blunting, even soporific effect. When a writer writes that, say, a character has her “heart in her mouth” or “a surge of adrenaline” or her “eyes sparkle,” then a reader, seeing combinations of words he has seen thousands of times before, glosses over the phrase, rather than seeing a vivid image. Over time a reader gets “habituated” to commonly-seen combinations of words like sidelong glances, and glinting eyes, and “a chill ran up my spine.”

This is true of phrases, and it’s true of narrative structures, too. Popular narrative structures which have been repeated often enough to be familiar can also have the same blunting, sleepy, familiar effect. How many evil villains are physically scarred? How many films end in a kiss? How many protagonists have a wise old grandfather? And this is fine! I’m not suggesting that this isn’t perfectly acceptable. I’ll go see the new Spiderman movie, or the new James Bond....A lot of care is taken so that the viewer does not get shaken up in any significant way. So familiar sentence constructions and familiar stories offer something safe and comfortable and sometimes our brains crave safe and comfortable.

But I do think that the role of art is to show us the familiar world in an unfamiliar way—to shake us up. The guy I always quote when I get asked about this stuff is an old Russian commisar named Viktor Shklovsky, in an essay he wrote called “Art as Technique.” “Art exists,” Shklovsky says, “that one may recover the sensation of life; it exists to make one feel things, to make the stone stony. The purpose of art is to impart the sensation of things as they are perceived and not as they are known.”

Writers like us—writers trying (and usually failing) to make art—are trying to use words, maybe the most used and familiar elements of daily life, and we’re trying to combine them to create transcendent aesthetic structures. We’re trying to employ language in ways that helps a reader see life in some “defamiliarized” way. Always, for me, art is slightly strange. Strangeness is what helps us crack apart our old eyes and see the world in a slightly new way. ...

I'm glad Doerr takes care to note, following Shklovsky, that cliche influences not only the reception and effect of language, but also the apprehension and effect of structure, and indeed of all "techniques" or "devices" of literary construction. What's most striking to me about Shklovsky's famous essay, from which our current ideas about "defamiliarization" largely derive, is Shklovsky's novel theory that the function of art is to restore to human experience what is lost in the mind's relentless drive toward efficiency—toward the most economical use of its budget of mental energy and attention. This is so striking because Shklovsky's assumptions about the way the mind works, in 1923, turned out to be so congruent with contemporary cognitive psychology.

The thrust of Shklovsky's prescription—the defamiliarization or "enstrangement" (making strange) of that to which we have become inured through "automatization"—is most compelling in the context of his factual assumptions about the mind. Here's what he says leading into the passage Doerr quotes:

In the process of algebrizing, of automatizing the object, the greatest economy of perceptual effort takes place. Objects are represented by one single characteristic (for example, by number), or else by a formula that never even rises to the level of consciousness....

Automatization eats away at things, at clothes, at furniture, at our wives, and at our fear of war.

If the complex life of many people takes place entirely on the level of the unconscious, then it's as if this life had never been.

And so, in order to return sensation to our limbs, in order to make us feel objects, to make a stone feel stony, man has been given the tool of art.

Shklovsky runs together at least two distinct mental processes, but it would be shabby to blame him. "Algebrization," the use of physical tokens (bits of sound, marks on paper) to express abstract, general concepts referring to broad classes of particulars certainly has something to do with "the economy of perceptual effort." But when it comes to the perception of the actual world (as opposed to the mental representation of the world imagined in a text), that economy perhaps has more to do with "habituation"—diminishing sensitivity to repeated or unchanging stimuli.

Habituation (a.k.a. "adaptation" or "extinction" or "acclimatization" or "fatigue" or "stimulatory inactivation"), is an incredibly general psychological process. Habituation's why you hear the air conditioner click on, but you soon don't hear the hum; why the water feels frigid but then feels "fine"; why last year's tired silhouettes don't grab the dandy's eye. Habituation's why a luxury car ultimately contributes very little to our sense of well-being. You just stop noticing how great it is. We notice change. At any given moment, much of what you "see" in your visual field isn't what's out there as it is right now; it's somewhat stale information retrieved from memory. Why continually "re-draw" the unchanging elements of a scene? The fact is that our brains, as big and beautiful as they are, have fixed computational limits, and simply can't handle the load of extravagantly total attentional presence. We couldn't possibly get along without the economies of habituation. We don't have the bandwidth, processing power, or memory to be awake always to everything. But habituation has obvious costs, as Shklovsky so poignantly observes. Habituation is zombification. Insofar as we are creatures of routine, we literally stop noticing our lives.

So suppose we follow Shklovsky and Doerr in the idea that the point of art is to make us alive to what we've stopped noticing by making it strange—to fight the effects of habituation with defamiliarization. Okay. But notice that there is a difference between defamiliarizing our experience of the world and defamiliarizing our experience of reading. Shklovsky and Doerr both slide a bit too easily between the two ideas.

Doerr says that "a reader, seeing combinations of words he has seen thousands of times before, glosses over the phrase, rather than seeing a vivid image." I don't believe it! Habituation to a combination of words doesn't suggest that those words will fail evoke a "vivid image." That gets the logic of habituation wrong. Habituation to a combination of words suggests not that we become unaware, or become aware only hazily, of what the words denote. It just means that we stop noticing the words. We process them easily. They get out of our way. We go fast. If words are atoms, and phrases (some of them clichés) are molecules, then defamiliarization of the world occurs at the compound level. It's a matter of higher-order combinatorial juju. You can make something pulsing and surpassing strange out of an ingenious combination of shopworn tropes. Conversely, you can state the boring, obvious, and banal in startlingly original terms.

Shklovsky similarly fudges the distinction between defamiliarizing the world and refreshing the use of language and other elements of storytelling. He basically just assumes that the former follows from the latter. He begins from nice examples of Tolstoy's success in defamilarizing the familiar world. Narrate from the point-of-view of a horse! Replace generic abstract terms with naive description in simpler terms closer to the phenomena! And somehow he gets to the idea that "the literary work attains its greatest and most long-lasting impact" when it is "created by the artist in such a way that the perceiver, pausing in his reading, dwells on the text." And he arrives finally at the notion that the essence of poetic (i.e., literary) language is "impeded, distorted speech." But Shklovsky needs an argument, which he never delivers: that impeded, distorted speech, which causes the reader to stop reading and dwell on the text, is the best way to get the reader to regain sensation in her limbs, to feel the stoniness of the stone. Why would it be?

Shklovsky's surely right that the forces of automatization/habituation are indiscriminate. Sooner or later everything becomes old news and that changes our relationship to it. Thus, an artistic technique or device that once produced a striking and fresh effect sooner or later become familiar and loses oomph. If you want to achieve something like the original effect, you're going to have to use a different device. That is, for me, the central lesson of "Art as Technique."

Perhaps a more practically relevant lesson of Shklovsky's essay is that, by defying automated expectations, fresh language and innovative technique generally slow the reader down. And of course they do. The point of habituation is cognitive economy. It's of the nature of defamiliarizing language and technique to increase the reader's cognitive load. It makes reading harder. To me this suggests that defamiliarization is one technique for controlling, among other things, the readability and pace of prose. You'd think folks in writing programs might pay more systematic attention to methods of modulating the pace and intensity of prose, much as composers and conductors modulate the pace and intensity of music. You'd think.

Anyway, that defamiliarization bogs readers down a bit also suggests that artful writing skillfully manages the level of attention and effort readers must put forth to successfully navigate the text—with "load balancing," as electric companies and computer network people say. If the point of habituation is the economical deployment of limited cognitive resources, and the point of defamiliarization is to resist or undo habituation, it would seem to follow that attempting to defamiliarize on too many dimensions at once may overburden all but the most Herculean reader.

This is not at all to say that the reader must be spared all "difficulty". Nor is it to say, with Herbert Spencer, with whom Shklovsky specifically disagrees, that the pinnacle of style is "to so present ideas that they may be apprehended with the least possible mental effort" or "that in composition, the chief, if not the sole thing to be done, is to reduce the friction to the least possible amount." We have friction to thank for so much of what's best in life. But there's friction and then there's chafing. We don't expect readers to put up with three-point sans-serif fonts, and we shouldn't resent them for feeling sore and bailing out because we tried to "make it new" in every way at once.

We must pick our battles in the war on the walking dead. There's no reason to make readers sweat, to force them anxiously to wrestle with the controls, if it's in our power simply to set the automatic pilot and crash them blithely into the fullness of their lives.