The Blank Slate

by Steven Pinker

  Like all conspiracy theories, the idea that language is a prisonhouse denigrates its subject by overestimating its power. Language is the magnificent faculty that we use to get thoughts from one head to another, and we can co-opt it in many ways to help our thoughts along. But it is not the same as thought, not the only thing that separates humans from other animals, not the basis of all culture, and not an inescapable prisonhouse, an obligatory agreement, the limits of our world, or the determiner of what is imaginable.28

  We have seen that perception and categorization provide us with concepts that keep us in touch with the world. Language extends that lifeline by connecting the concepts to words. Children hear noises coming out of a family member’s mouth, use their intuitive psychology and their grasp of the context to infer what the speaker is trying to say, and mentally link the words to the concepts and the grammatical rules to the relationships among them. Bowser upends a chair, Sister yells, “The dog knocked over the chair!” and Junior deduces that dog means dog, chair means chair, and the subject of the verb knock over is the agent doing the knocking over.29 Now Junior can talk about other dogs, other chairs, and other knockings over. There is nothing self-referential or imprisoning about it. As the novelist Walker Percy quipped, a deconstructionist is an academic who claims that texts have no referents and then leaves a message on his wife’s answering machine asking her to order a pepperoni pizza for dinner.

  Language surely does affect our thoughts, rather than just labeling them for the sake of labeling them. Most obviously, language is the conduit through which people share their thoughts and intentions and thereby acquire the knowledge, customs, and values of those around them. In the song “Christmas” from their rock opera, The Who described the plight of a boy without language: “Tommy doesn’t know what day it is; he doesn’t know who Jesus was or what prayin’ is.”

  Language can allow us to share thoughts not just directly, by its literal content, but also indirectly, via metaphors and metonyms that nudge listeners into grasping connections they may not have noticed before. For example, many expressions treat time as if it were a valuable resource, such as waste time, spend time, valuable time, and time is money.30 Presumably on the first occasion a person used one of these expressions, her audience wondered why she was using a word for money to refer to time; after all, you can’t literally spend time the way you spend pieces of gold. Then, by assuming that the speaker was not gibbering, they figured out the ways in which time indeed has something in common with money, and assumed that that was what the speaker intended to convey. Note that even in this clear example of language affecting thought, language is not the same thing as thought. The original coiner of the metaphor had to see the analogy without the benefit of the English expressions, and the first listeners had to make sense of it using a chain of ineffable thoughts about the typical intentions of speakers and the properties shared by time and money.

  Aside from its use as a medium of communication, language can be pressed into service as one of the media used by the brain for storing and manipulating information.31 The leading theory of human working memory, from the psychologist Alan Baddeley, captures the idea nicely.32 The mind makes use of a “phonological loop”: a silent articulation of words or numbers that persists for a few seconds and can be sensed by the mind’s ear. The loop acts as a “slave system” at the service of a “central executive.” By describing things to ourselves using snatches of language, we can temporarily store the result of a mental computation or retrieve chunks of data stored as verbal expressions. Mental arithmetic involving large numbers, for example, may be carried out by retrieving verbal formulas such as “Seven times eight is fifty-six.”33 But as the technical terms of the theory make clear, language is serving as a slave of an executive, not as the medium of all thought.

  Why do virtually all cognitive scientists and linguists believe that language is not a prisonhouse of thought?34 First, many experiments have plumbed the minds of creatures without language, such as infants and nonhuman primates, and have found the fundamental categories of thought working away: objects, space, cause and effect, number, probability, agency (the initiation of behavior by a person or animal), and the functions of tools.35

  Second, our vast storehouse of knowledge is certainly not couched in the words and sentences in which we learned the individual facts. What did you read in the page before this one? I would like to think that you can give a reasonably accurate answer to the question. Now try to write down the exact words you read in those pages. Chances are you cannot recall a single sentence verbatim, probably not even a single phrase. What you remembered is the gist of those passages—their content, meaning, or sense—not the language itself. Many experiments on human memory have confirmed that what we remember over the long term is the content, not the wording, of stories and conversations. Cognitive scientists model this “semantic memory” as a web of logical propositions, images, motor programs, strings of sounds, and other data structures connected to one another in the brain.36

  A third way to put language in its place is to think about how we use it. Writing and speaking do not consist of transcribing an interior monologue onto paper or playing it into a microphone. Rather, we engage in a constant give-and-take between the thoughts we try to convey and the means our language offers to convey them. We often grope for words, are dissatisfied with what we write because it does not express what we wanted to say, or discover when every combination of words seems wrong that we do not really know what we want to say. And when we get frustrated by a mismatch between our language and our thoughts, we don’t give up, defeated and mum, but change the language. We concoct neologisms (quark, meme, clone, deep structure), invent slang (to spam, to diss, to flame, to surf the web, a spin doctor), borrow useful words from other languages (joie de vivre, schlemiel, angst, machismo), or coin new metaphors (waste time, vote with your feet, push the outside of the envelope). That is why every language, far from being an immutable penitentiary, is constantly under renovation. Despite the lamentations of language lovers and the coercion of tongue troopers, languages change unstoppably as people need to talk about new things or convey new attitudes.37

  Finally, language itself could not function if it did not sit atop a vast infrastructure of tacit knowledge about the world and about the intentions of other people. When we understand language, we have to listen between the lines to winnow out the unintended readings of an ambiguous sentence, piece together fractured utterances, glide over slips of the tongue, and fill in the countless unsaid steps in a complete train of thought. When the shampoo bottle says “Lather, rinse, repeat,” we don’t spend the rest of our lives in the shower; we infer that it means “repeat once.” And we know how to interpret ambiguous headlines such as “Kids Make Nutritious Snacks,” “Prostitutes Appeal to Pope,” and “British Left Waffles on Falkland Islands,” because we effortlessly apply our background knowledge about the kinds of things that people are likely to convey in newspapers. Indeed, the very existence of ambiguous sentences, in which one string of words expresses two thoughts, proves that thoughts are not the same thing as strings of words.

  LANGUAGE OFTEN MAKES the news precisely because it can part company with thoughts and attitudes. In 1998 Bill Clinton exploited the expectations behind ordinary comprehension to mislead prosecutors about his affair with Monica Lewinsky. He used words like alone, sex, and is in senses that were technically defensible but which deviated from charitable guesses about what people ordinarily mean by these terms. For example, he suggested he was not “alone” with Lewinsky, even though they were the only two people in the room, because other people were in the Oval Office complex at the time. He said that he did not have “sex” with her, because they did not engage in intercourse. His words, like all words, are certainly vague at their boundaries. Exactly how far away or hidden must the nearest person be before one is considered alone? At what point in the continuum of bodily contact—from an accidental brush in an elevator to tantric
bliss—do we say that sex has occurred? Ordinarily we resolve the vagueness by guessing how our conversational partner would interpret words in the context, and we choose our words accordingly. Clinton’s ingenuity in manipulating these guesses, and the outrage that erupted when he was forced to explain what he had done, show that people have an acute understanding of the difference between words and the thoughts they are designed to convey.

  Language conveys not just literal meanings but also a speaker’s attitude. Think of the difference between fat and voluptuous, slender and scrawny, thrifty and stingy, articulate and slick. Racial epithets, which are laced with contempt, are justifiably off-limits among responsible people, because using them conveys the tacit message that contempt for the people referred to by the epithet is acceptable. But the drive to adopt new terms for disadvantaged groups goes much further than this basic sign of respect; it often assumes that words and attitudes are so inseparable that one can reengineer people’s attitudes by tinkering with the words. In 1994 the Los Angeles Times adopted a style sheet that banned some 150 words, including birth defect, Canuck, Chinese fire drill, dark continent, divorcée, Dutch treat, handicapped, illegitimate, invalid, man-made, New World, stepchild, and to welsh. The editors assumed that words register in the brain with their literal meanings, so that an invalid is understood as “someone who is not valid” and Dutch treat is understood as a slur on contemporary Netherlanders. (In fact, it is one of many idioms in which Dutch means “ersatz,” such as Dutch oven, Dutch door, Dutch uncle, Dutch courage, and Dutch auction, the remnants of a long-forgotten rivalry between the English and the Dutch.)

  But even the more reasonable attempts at linguistic reform are based on a dubious theory of linguistic determinism. Many people are puzzled by the replacement of formerly unexceptionable terms by new ones: Negro by black by African American, Spanish-American by Hispanic by Latino, crippled by handicapped by disabled by challenged, slum by ghetto by inner city by (according to the Times) slum once again. Occasionally the neologisms are defended with some rationale about their meaning. In the 1960s, the word Negro was replaced by the word black, because the parallel between the words black and white was meant to underscore the equality of the races. Similarly, Native American reminds us of who was here first and avoids the geographically inaccurate term Indian. But often the new terms replace ones that were perfectly congenial in their day, as we see in names for old institutions that are obviously sympathetic to the people being named: the United Negro College Fund, the National Association for the Advancement of Colored People, the Shriners Hospitals for Crippled Children. And sometimes a term can be tainted or unfashionable while a minor variant is fine: consider colored people versus people of color, Afro-American versus African American, Negro—Spanish for “black”—versus black. If anything, a respect for literal meaning should send us off looking for a new word for the descendants of Europeans, who are neither white nor Caucasian. Something else must be driving the replacement process.

  Linguists are familiar with the phenomenon, which may be called the euphemism treadmill. People invent new words for emotionally charged referents, but soon the euphemism becomes tainted by association, and a new word must be found, which soon acquires its own connotations, and so on. Water closet becomes toilet (originally a term for any kind of body care, as in toilet kit and toilet water), which becomes bathroom, which becomes restroom, which becomes lavatory. Undertaker changes to mortician, which changes to funeral director. Garbage collection turns into sanitation, which turns into environmental services. Gym (from gymnasium, originally “high school”) becomes physical education, which becomes (at Berkeley) human biodynamics. Even the word minority—the most neutral label conceivable, referring only to relative numbers—was banned in 2001 by the San Diego City Council (and nearly banned by the Boston City Council) because it was deemed disparaging to non-whites. “No matter how you slice it, minority means less than,” said a semantically challenged official at Boston College, where the preferred term is AHANA (an acronym for African-American, Hispanic, Asian, and Native American).38

  The euphemism treadmill shows that concepts, not words, are primary in people’s minds. Give a concept a new name, and the name becomes colored by the concept; the concept does not become freshened by the name, at least not for long. Names for minorities will continue to change as long as people have negative attitudes toward them. We will know that we have achieved mutual respect when the names stay put.

  “IMAGE IS NOTHING. Thirst is everything,” screams a soft-drink ad that tries to create a new image for its product by making fun of soft-drink ads that try to create images for their products. Like words, images are salient tokens of our mental lives. And like words, images are said to have an insidious power over our consciousness, presumably because they are inscribed directly onto a blank slate. In postmodernist and relativist thinking, images are held to shape our view of reality, or to be our view of reality, or to be reality itself. This is especially true of images representing celebrities, politicians, women, and AHANAs. And as with language, the scientific study of imagery shows that the fear is misplaced.

  A good description of the standard view of images within cultural studies and related disciplines may be found in the Concise Glossary of Cultural Theory. It defines image as a “mental or visual representation of an object or event as depicted in the mind, a painting, a photograph, or film.” Having thus run together images in the world (such as paintings) with images in the mind, the entry lays out the centrality of images in postmodernism, cultural studies, and academic feminism.

  First it notes, reasonably enough, that images can misrepresent reality and thereby serve the interests of an ideology. A racist caricature, presumably, is a prime example. But then it takes the concept further:

  With what is called the “crisis of representation” brought about by… postmodernism, however, it is often questioned whether an image can be thought to simply represent, or misrepresent, a supposedly prior or external, image-free reality. Reality is seen rather as always subject to, or as the product of, modes of representation. In this view we inescapably inhabit a world of images or representations and not a “real world” and true or false images of it.

  In other words, if a tree falls in a forest and there is no artist to paint it, not only did the tree make no sound, but it did not fall, and there was no tree there to begin with.

  In a further move… we are thought to exist in a world of HYPERREALITY, in which images are self-generating and entirely detached from any supposed reality. This accords with a common view of contemporary entertainment and politics as being all a matter of “image,” or appearance, rather than of substantial content.

  Actually, the doctrine of hyperreality contradicts the common view of contemporary politics and entertainment as being a matter of image and appearance. The whole point of the common view is that there is a reality separate from images, and that is what allows us to decry the images that are misleading. We can, for example, criticize an old movie that shows slaves leading happy lives, or an ad that shows a corrupt politician pretending to defend the environment. If there were no such thing as substantial content, we would have no basis for preferring an accurate documentary about slavery to an apologia for it, or preferring a good exposé of a politician to a slick campaign ad.

  The entry notes that images are associated with the world of publicity, advertising, and fashion, and thereby with business and profits. An image may thus be tied to “an imposed stereotype or an alternative subjective or cultural identity.” Media images become mental images: people cannot help but think that women or politicians or African Americans conform to the depictions in movies and advertisements. And this elevates cultural studies and postmodernist art into forces for personal and political liberation:

  The study of “images of women” or “women’s images” sees this field as one in which stereotypes of women can be reinforced, parodied, or actively contested through critical analysis, alte
rnative histories, or creative work in writing and the media committed to the production of positive counter-images.39

  I have not hidden my view that this entire line of thinking is a conceptual mess. If we want to understand how politicians or advertisers manipulate us, the last thing we should do is blur distinctions among things in the world, our perception of those things when they are in front of our eyes, the mental images of those things that we construct from memory, and physical images such as photographs and drawings.

  As we saw at the beginning of this chapter, the visual brain is an immensely complicated system that was designed by the forces of evolution to give us an accurate reading of the consequential things in front of us. The “intelligent eye,” as perceptual psychologists call it, does not just compute the shapes and motions of people before us. It also guesses their thoughts and intentions by noticing how they gaze at, approach, avoid, help, or hinder other objects and people. And these guesses are then measured against everything else we know about people—what we infer from gossip, from a person’s words and deeds, and from Sherlock Holmes-style deductions. The result is the knowledge base or semantic memory that also underlies our use of language.