*Do* we need Shakespeare?

Megan McArdle asks: “Do We Need Shakespeare?“, and she offers some theories about why we might that don’t rely on “Because we’ve always done it that way,” including “What remains is a sort of stubborn belief that people ought to study literature because it is somehow good for them” and “Maybe the best argument you can make for English class is that it offers a way for people like myself, and many thousands of future English teachers, to find out that they like English class.”



Still, I’m not so sure. I began imagining reasons almost immediately, but most reduced to, “Because we’ve been doing it that way for a long time.” Which can be reduced to “path dependence.” Teaching a writer who seems incomprehensible at first glance and requires experts to decipher also raises the status of (some) teachers and professors, who have knowledge that can’t be readily accessed by every day people. That Hansonian reason, however, isn’t real good reason why we should choose Shakespeare plays over some other means of teaching English.

Let me try to develop an alternate possibility that will likely make many people unhappy. I’ve begun to think that education is really about cultivating a relatively small elite who really push forward particular domains (which is a variant of McArdle’s comment about the thousands of future English teachers). In other words, mass education doesn’t matter nearly as much as intensely educating a small number of very high skill people, but those people probably aren’t identifiable in advance. This idea isn’t purely mine, and I’ve been thinking about it explicitly since reading Joel Mokyr’s The Enlightened Economy, in which he writes:

It is important to stress that the Industrial Revolution was the creation of an elite, a relatively small number of ingenious, ambitious, and diligent persons who could think out of the box, and had the wherewithal to carry out their ideas and to find others who could assist them. This is not to return to the heroic interpretations of the Victorian hagiographers such as Samuel Smiles and credit a few famous individuals with the entire phenomenon […] Even these pivotal people were a minority, perhaps a few tens of thousands of elite workers, well trained through apprenticeships supplemented sometimes by informal studies.

Wow: Something as big a deal as the Industrial Revolution may have been driven by a small number of people. I’ve also read a lot about the early computer industry and the early development of integrated chips, and that too seems to have been driven by a small number of physicists and mathematicians, with particularly important companies like Fairchild Semiconductor and later Intel starting off with tiny workforces. Most of the world didn’t matter much to the development of those industries, even though those industries are now so large that a large part of the workforce spends our time in front of glowing screens that show executed code most of us don’t understand and can’t write. Computers and the Internet are the biggest stories of our age, possibly excepting global warming and mass extinction, yet many of us aren’t substantially participating and don’t care to.

What gives?

The unpleasant answer may be that most of us don’t matter that much to the process. By the same token, most people who learn to despite reading from being made to read Shakespeare may never be good readers, writers, or thinkers—but they’re not the ones who push the world forward, intellectually speaking. Instead, those of us who go on to realize that, say, “
Shakespeare’s Genius Is Nonsense: What the Bard can teach science about language and the limits of the human mind
” are the ones who matter, at least in this domain.

Most Westerners are uncomfortable with outright elitism, but I’d ask: How many of us really work as hard as we can at a given domain? In “How A Slight Change In Mindset: Accelerated My Learning Forever,” Tristan de Montebello observes that few of us really throw ourselves into learning. Most of us learn as much as we need to to survive and do okay and reproduce, but not much more than that. You can tell as much via behavior.

That may be true in language as a domain as well. I read more than the vast majority of people I know, and yet there are people who read and write much more than even I do.

To return to Shakespeare, I’d also argue that sometimes complex, weird, or seemingly outdated works force us to read closer and more carefully than we might otherwise. Shakespeare makes contemporary readers work harder to understand what the writer means—which is ultimately a useful and under-used skill. Just look at most Internet forums: since the 1980s and Usenet, going forward all the way to today with Reddit, we have numerous places where people gather online and utterly fail at basic reading comprehension (this is one reason I spend little time posting there and much more posting here).

Under this theory, reading someone like Shakespeare is akin to lifting weights: a 500-pound deadlift may not translate directly into 500 pounds of force in a game, but it sure translates more force than a guy who can’t deadlift 500 pounds.

Still, I’m treating the argument that Shakespeare-is-good like a lawyer and trying to come up with the best possible argument, rather than arguing from first principles. I’m not fully convinced we need Shakespeare, as opposed to some other writer or group of writers, as a necessary component of teaching English.

A discussion of a weird grammar quirk: tense and Mark McGurl’s The Program Era

In Mark McGurl’s excellent The Program Era: Postwar Fiction and the Rise of Creative Writing, he writes this about Vladimir Nabokov:

In fact, one of his best-known quirks was a scientific passion for a certain family of butterflies, the Blues.

The word “was” is interesting, because Nabokov’s quirks are still well-known, in the present. But Nabokov’s quirks happened in the past—he’s obviously dead. So there’s a moment of verb tense weirdness in this sentence, which might otherwise read something like, “In fact, one of his best-known quirks is the scientific passion Nabokov had for a certain family of butterflies, the Blues.”

There’s no particular point to this post other than a writer’s duty to notice language, and the opportunity to observe a specific example of language’s sometimes bizarre ambiguity.

(Those of you who are reading the last sentence and thinking about its own weirdness, be aware: that is intentional.)

Hey guys, read this

Referring to women or mixed-sex groups as “guys” or “you guys” apparently offends a fair number of people. This is interesting to me because I use the phrase a fair amount, especially in class—”Hey guys, now I want you to take out your papers and…”

The problem is that the phrase “you guys” is useful: what non-gendered term could replace it? “Ladies and gentlemen” is old-fashioned, verging on archaic, and “guys and girls” could be demeaning, and I can’t think of a good replacement. “People” or “hey people” is coarse. “You people” has historical/racial baggage of its own—almost enough to have a Coleman Silk problem.


Favorite words

Literary agent/blogger/spandex-clad crime fighter Nathan Bransford asked for favorite words on his blog, and I answered: that “my favorite is callipygian, followed closely thereafter by defenestrate.”

Later, he asked for least favorite words, and I settled for academic clichés like epistemological, trope, and destabilize.

Anyone else want to share?

The Stuff of Thought and Steven Pinker in Tucson

It’s sometimes harder to describe what comes naturally than it is what comes artificially. We learn to speak by virtue of being around adults who speak, and yet analyzing the languages humans have developed and what those languages represent is harder than it is for a toddler to intuitively learn them. Speaking develops with no schooling aside from the “school” of other humans—and yet its manifold distinctions are the subject of Steven Pinker’s The Stuff of Thought, a complex book that gives some answers leading toward still more questions as he tries to explain the paradoxical mysteries of consciousness and perception.

The subtext of The Stuff of Thought seems to be that language affects us more than we consciously realize and that our uses of language tends to occur in previously unexamined patterns that, once perceived, can be better used to our advantage. Such bold statements take some explaining: language reveals a great deal about us, Pinker argues, including theories of causation embedded inflectionally in some languages and syntactically in English. Some examples are simple: “John threw the ball” indicates who acted on what, and in that model it is difficult to misinterpret what is being said and who is doing what to what. Throw in prepositions and other spatial features encoded in language, however, and it becomes steadily harder to grasp precisely why “A sad movie makes you sad, but a sad person is already sad,” even if we understand the difference without being told the rule. The Stuff of Thought is a guided tour through what we didn’t know that we know. “I am exploring my sexuality; you are promiscuous; she is a slut,” and while all three phrases or words might describe the same fundamental behavior, and yet each has very different and apparent shades of meaning, from positive to pejorative.

This is an example of how we “flip frames,” or understand an event in multiple ways depending on its context. In Newsweek, Lynne Spears—the mother of children famous for celebrity and fecundity, in that order—said of one who recently gave birth at 17, “But [despite] a situation that has fallen in her lap, she’s doing exceptionally well[…]” Notice the phrase, “a situation that has fallen in her lap,” as if the person involved had no agency and was struck by a meteor on her way to school. Then again, maybe the girl in question didn’t have as much agency as classical economists would believe; in Dan Ariely’s excellent Predictably Irrational, he discusses an experiment in which students who were aroused admitted to considerably risk taking in an inventory of potential sexual behaviors than those who were not.* The frame Lynne Spears uses betrays at least some idea of her “frame,” but if we’re not paying attention to her statement, we’re likely to miss it. Furthermore, to be fair, Lynne Spears might refer to her daughter’s choice long after conception, at which point it’s too late to remake the past and one must deal with the options at hand. Temporal ambiguity—a subject Pinker discusses in Chapter 4, “Cleaving the Air”—becomes essential, and nothing about what Lynne Spears said indicates the precise time period she meant. It turns out that such relativity is inherent in language, which applies imprecise spatial metaphors to time, leaving us with the uncertainty much celebrated by Deconstructionists.

Other chapters in The Stuff of Thought deal with metaphors, naming, and game theory, but to go into each would expand this post into a weak shadow of the book, rather than a pointer in its direction. Some extra discussion is warranted, though, and Pinker also discusses swearing and how it changes over time in Chapter 7, “The Seven Words You Can’t Say on Television,” and especially why so much revolves around excretion, sex, and religion. The power of the latter has declined in much of the West along with belief in a literal manifestation of God, and Pinker speculates that phrases like “go to hell,” or “damnit,” that are sufficiently innocuous to be broadcast on television, might have been more threatening when people believed they were Sinners in the Hand of an Angry God (sample: “The God that holds you over the pit of hell, much as one holds a spider, or some detestable insect, over the fire, detests you, and is dreadfully provoked”). Those around excretion and sex, however, still hold more power because they’re both vectors for disease, literally, and the latter can also be a vector for emotional disease, as many pop songs and novels about jilted love attest.

The good news is that Pinker visited Tucson on his tour for the paperback edition of The Stuff of Thought. The bad news for readers is that he hewed so closely to the material in it as to render the talk itself redundant. The points were identical and the examples to support generalizations merely less frequent and deep. But he did expand slightly on issues of swearing and “how to identify and quantify the material world,” and perhaps the most interesting part of his talk was not the talk itself but the audience’s reaction to his discussion of swearing. It’s fairly unusual to hear an impeccably dress professor speculate about the tabooness of words like “fuck” and “cunt,” and the audience tittered appropriately. Pinker can euphemize with the best, referring to “the gynecological-flagellative term for uxorial dominance” at one point in The Stuff of Thought. He leapt between high and low registers with relative ease, and I suppose after discussing the issues numerous times it becomes easier to keep one’s equanimity around swearing. At the end Pinker discussed using language and knowledge of what others know as a way to redefine relationships, expressing the dangers of being too blunt or not blunt enough, and suffering the consequences in the form of missed opportunities or social blunders. One might avoid the kinds of problems from Chapter 8, “Games People Play,” by refusing to feel awkwardness or by reducing one’s susceptibility to societal influence. But he never went that far, and some problems he presents leaves us with the implied answers or ameliorations, like a coyer version of Machiavelli in The Prince.

The sense of Pinker giving only a small taste of his book was reflected in the question and answer period: someone would ask a question, Pinker would begin to elaborate, and then refer the questioner to the relevant chapter. Materials as complex as his can’t easily be summarized and grokked, particularly because one of his book’s major virtues is the wealth of examples and metaphors he uses to describe the general principles he and others have derived from language itself. It’s also a drawback of this post: I’ve tried to give a general overview of Pinker’s ideas, but my own writings are at such a surface level that they can do no more than point to the book. Call it the difference between something like Lily Koppel’s The Red Leather Diary, which would’ve been better left a newspaper article and The Stuff of Thought, a book whose teachings are easier to describe than to apply. Pinker has accomplished a difficult task in synthesizing so much research, but its readers have the harder work of deciding what to do with what we’ve learned.

* I won’t give away the experiment design; for that, you’ll have to read Predictably Irrational.

%d bloggers like this: