The New York Times
The Stuff of Thought: Language as a Window into Human Natureby Steven Pinker
Every time we swear, we reveal something about human emotions. When we use an innuendo to convey a bribe, threat, or sexual come-on (rather than just blurting it out), we disclose something about human relationships. Our use of prepositions and tenses tap into peculiarly human concepts of space and time, and our nouns and verbs tap into mental models of matter and… See more details below
Every time we swear, we reveal something about human emotions. When we use an innuendo to convey a bribe, threat, or sexual come-on (rather than just blurting it out), we disclose something about human relationships. Our use of prepositions and tenses tap into peculiarly human concepts of space and time, and our nouns and verbs tap into mental models of matter and causation. Even the names we give our babies, as they change from decade to decade, have important things to day about our relations to our children and to society. By looking closely at our everyday speech-our conversations, our jokes, our legal disputes-Pinker paints a vivid picture of the thoughts and emotions that populate our mental lives.
Pinker takes on both scientific questions-like whether language affects thought, and which of our concepts are innate-and questions from the headlines and everyday life. Why does the government care so much about dirty words? How do lobbyists bribe politicians? Why do romantic comedies get such mileage out of the ambiguities of dating? Why do so many courtroom dramas hinge on disagreements about who really caused a person's death? Why have the last two American presidents gotten into trouble by the semantic niceties of their words? And why is bulk email called spam?
The Stuff of Thought marries the two topics of Pinker's earlier bestsellers: language (The Language Instinct, Words and Rules) and human nature (How the Mind Works, The Blank Slate). It presents entirely new material, while written in the style that made those books famous: lucid explanations of deep and powerful ideas, presented with irreverent wit, elegant style, and a deft use of examples from popular culture and everyday life.
The New York Times
The Washington Post
Unless you have a reasonably good background in linguistics, you'll find this excellent book much easier to read than to listen to. Olsher is not to blame; he reads clearly and at a (slightly rapid) conversational speed. Pinker aims for the educated lay reader, using wit and popular metaphor to clarify his meanings and bring abstruse linguistic concepts to life. But his sentences are dense; you need to reread them and think them through. And the jargon, though clearly defined, requires time and thought to absorb: "Though hypernyms are not really examples of polysemy the way metonyms are, their use in emotionally tinged speech is another illustration of how choice among words can make a psychological difference." Such sentences are followed by clarifying illustrations, but they require cogitation-work that is well rewarded by a deeper and more complex understanding of language as a window into the mind. The chapter on the semantics of swearing is particularly fun and enlightening. In every culture swear words concern gods, diseases, excretions and sex, and Pinker tells us why. A person with some knowledge of linguistic theory will enjoy this audio enormously; a person without it will be enriched and delighted by the book, but have great difficulties with the audio version. Simultaneous release with the Viking hardcover (Reviews, May 21). (Sept.)Copyright 2007 Reed Business Information
Los Angeles Times
- Viking Adult
- Publication date:
- Product dimensions:
- 6.34(w) x 9.26(h) x 1.55(d)
- Age Range:
- 18 Years
Read an Excerpt
Praise for The Stuff of Thought by Steven Pinker
ABOUT THE AUTHOR
Steven Pinker is the Johnstone Professor of Psychology at Harvard University. His research on language and cognition has won prizes from the National Academy of Sciences, the Royal Institution of Great Britain, the American Psychological Association, and the Cognitive Neuroscience Society. He has also received several teaching awards, eight honorary doctorates, and many prizes for his nine books, including The Language Instinct, How the Mind Works, The Blank Slate, and The Better Angels of Our Nature. His most recent book, The Sense of Style, was a New York Times bestseller. He has been named Humanist of the Year and has been listed among Foreign Policy magazine’s “The World’s Top 100 Public Intellectuals” and Time’s “The 100 Most Influential People in the World Today.” He is currently Chair of the Usage Panel of The American Heritage Dictionary and writes frequently for The New York Times, Time, The New Republic, and other publications.
Table of Contents
Praise for THE STUFF OF THOUGHT by Steven Pinker
ABOUT THE AUTHOR
Chapter 1 - WORDS AND WORLDS
Chapter 2 - DOWN THE RABBIT HOLE
Chapter 3 - FIFTY THOUSAND INNATE CONCEPTS (AND OTHER RADICAL THEORIES OF ...
Chapter 4 - CLEAVING THE AIR
Chapter 5 - THE METAPHOR METAPHOR
Chapter 6 - WHAT’S IN A NAME?
Chapter 7 - THE SEVEN WORDS YOU CAN’T SAY ON TELEVISION
Chapter 8 - GAMES PEOPLE PLAY
Chapter 9 - ESCAPING THE CAVE
Published by the Penguin Group
Penguin Group (USA) LLC
375 Hudson Street
New York, New York 10014
USA | Canada | UK | Ireland | Australia | New Zealand | India | South Africa | China
A Penguin Random House Company
First published in the United States of America by Viking Penguin, a member of Penguin Group (USA) Inc., 2007
Published in Penguin Books 2008
Copyright © 2007 by Steven Pinker
Penguin supports copyright. Copyright fuels creativity, encourages diverse voices, promotes free speech, and creates a vibrant culture. Thank you for buying an authorized edition of this book and for complying with copyright laws by not reproducing, scanning, or distributing any part of it in any form without permission. You are supporting writers and allowing Penguin to continue to publish books for every reader.
Grateful acknowledgment is made for permission to reprint excerpts from the following copyrighted works:
“This Be the Verse” from Collected Poems by Philip Larkin. Copyright © 1988, 2003 by the Estate of Philip Larkin. Reprinted by permission of Farrar, Straus and Giroux, LLC and Faber and Faber Ltd.
The Nurture Assumption: Why Children Turn Out the Way They Do by Judith Rich Harris (Free Press). Copyright © 1998 by Judith Rich Harris. Reprinted with permission.
THE LIBRARY OF CONGRESS HAS CATALOGED THE HARDCOVER EDITION AS FOLLOWS:
Pinker, Steven, 1954–
The stuff of thought : language as a window into human nature / Steven Pinker.
Includes bibliographical references and index.
1. Language and languages—Philosophy. 2. Thought and thinking. I. Title.
While the author has made every effort to provide accurate telephone numbers, Internet addresses, and other contact information at the time of publication, neither the publisher nor the author assumes any responsibility for errors or for changes that occur after publication. Further, publisher does not have any control over and does not assume any responsibility for author or third-party Web sites or their content.
There is a theory of space and time embedded in the way we use words. There is a theory of matter and a theory of causality, too. Our language has a model of sex in it (actually, two models), and conceptions of intimacy and power and fairness. Divinity, degradation, and danger are also ingrained in our mother tongue, together with a conception of well-being and a philosophy of free will. These conceptions vary in their details from language to language, but their overall logic is the same. They add up to a distinctively human model of reality, which differs in major ways from the objective understanding of reality eked out by our best science and logic. Though these ideas are woven into language, their roots are deeper than language itself. They lay out the ground rules for how we understand our surroundings, how we assign credit and blame to our fellows, and how we negotiate our relationships with them. A close look at our speech—our conversations, our jokes, our curses, our legal disputes, the names we give our babies—can therefore give us insight into who we are.
That is the premise of the book you are holding, the third in a trilogy written for a wide audience of readers who are interested in language and mind. The first, The Language Instinct, was an overview of the language faculty: everything you always wanted to know about language but were afraid to ask. A language is a way of connecting sound and meaning, and the other two books turn toward each of those spheres. Words and Rules was about the units of language, how they are stored in memory, and how they are assembled into the vast number of combinations that give language its expressive power. The Stuff of Thought is about the other side of the linkage, meaning. Its vistas include the meanings of words and constructions and the way that language is used in social settings, the topics that linguists call semantics and pragmatics.
At the same time, this volume rounds out another trilogy: three books on human nature. How the Mind Works tried to reverse-engineer the psyche in the light of cognitive science and evolutionary psychology. The Blank Slate explored the concept of human nature and its moral, emotional, and political colorings. This one broaches the topic in still another way: what we can learn about our makeup from the way people put their thoughts and feelings in words.
As in my other books on language, the early chapters occasionally dip into technical topics. But I have worked hard to make them transparent, and I am confident that my subject will engage anyone with an interest in what makes us tick. Language is entwined with human life. We use it to inform and persuade, but also to threaten, to seduce, and of course to swear. It reflects the way we grasp reality, and also the image of ourselves we try to project to others, and the bonds that tie us to them. It is, I hope to convince you, a window into human nature.
In writing this book I have enjoyed the advice and support of many people, beginning with my editors, Wendy Wolf, Stefan McGrath, and Will Goodlad, and my agent, John Brockman. I have benefited tremendously from the wisdom of generous readers who reviewed the entire manuscript—Rebecca Newberger Goldstein, David Haig, David Kemmerer, Roslyn Pinker, and Barbara Spellman—and from the mavens who commented on chapters in their areas of expertise: Linda Abarbanell, Ned Block, Paul Bloom, Kate Burridge, Herbert Clark, Alan Dershowitz, Bruce Fraser, Marc Hauser, Ray Jackendoff, James Lee, Beth Levin, Peggy Li, Charles Parsons, James Pustejovsky, Lisa Randall, Harvey Silverglate, Alison Simmons, Donald Symons, J. D. Trout, Michael Ullman, Edda Weigand, and Phillip Wolff. Thanks, too, to those who answered my queries or offered suggestions: Max Bazerman, Iris Berent, Joan Bresnan, Daniel Casasanto, Susan Carey, Gennaro Chierchia, Helena Cronin, Matt Denio, Daniel Donoghue, Nicholas Epley, Michael Faber, David Feinberg, Daniel Fessler, Alan Fiske, Daniel Gilbert, Lila Gleitman, Douglas Jones, Marcy Kahan, Robert Kurzban, Gary Marcus, George Miller, Martin Nowak, Anna Papafragou, Geoffrey Pullum, S. Abbas Raza, Laurie Santos, Anne Senghas, G. Richard Tucker, Daniel Wegner, Caroline Whiting, and Angela Yu. This is the sixth book of mine that Katya Rice has agreed to copyedit, and like the others it has benefited from her style, precision, and curiosity.
I thank Ilavenil Subbiah for the many examples of subtle semantic phenomena she recorded from everyday speech, for designing the chapter ornament, and for much else besides. Thanks also to my parents, Harry and Roslyn, and to my family: Susan, Martin, Eva, Carl, Eric, Rob, Kris, Jack, David, Yael, Gabe, and Danielle. Most of all, I thank Rebecca Newberger Goldstein, my bashert, to whom this book is dedicated.
The research for this book was supported by NIH Grant HD-18381 and by the Johnstone Family Chair at Harvard University.
WORDS AND WORLDS
On September 11, 2001, at 8:46 A.M., a hijacked airliner crashed into the north tower of the World Trade Center in New York. At 9:03 A.M. a second plane crashed into the south tower. The resulting infernos caused the buildings to collapse, the south tower after burning for an hour and two minutes, the north tower twenty-three minutes after that. The attacks were masterminded by Osama bin Laden, leader of the Al Qaeda terrorist organization, who hoped to intimidate the United States into ending its military presence in Saudi Arabia and its support for Israel and to unite Muslims in preparation for a restoration of the caliphate.
9/11, as the happenings of that day are now called, stands as the most significant political and intellectual event of the twenty-first century so far. It has set off debates on a vast array of topics: how best to memorialize the dead and revitalize lower Manhattan; whether the attacks are rooted in ancient Islamic fundamentalism or modern revolutionary agitation; the role of the United States on the world stage before the attacks and in response to them; how best to balance protection against terrorism with respect for civil liberties.
But I would like to explore a lesser-known debate triggered by 9/11. Exactly how many events took place in New York on that morning in September?
It could be argued that the answer is one. The attacks on the buildings were part of a single plan conceived in the mind of one man in service of a single agenda. They unfolded within a few minutes and yards of each other, targeting the parts of a complex with a single name, design, and owner. And they launched a single chain of military and political events in their aftermath.
Or it could be argued that the answer is two. The north tower and the south tower were distinct collections of glass and steel separated by an expanse of space, and they were hit at different times and went out of existence at different times. The amateur video that showed the second plane closing in on the south tower as the north tower billowed with smoke makes the twoness unmistakable: in those horrifying moments, one event was frozen in the past, the other loomed in the future. And another occurrence on that day—a passenger mutiny that brought down a third hijacked plane before it reached its target in Washington—presents to the imagination the possibility that one tower or the other might have been spared. In each of those possible worlds a distinct event took place, so in our actual world, one might argue, there must be a pair of events as surely as one plus one equals two.
The gravity of 9/11 would seem to make this entire discussion frivolous to the point of impudence. It’s a matter of mere “semantics,” as we say, with its implication of picking nits, splitting hairs, and debating the number of angels that can dance on the head of a pin. But this book is about semantics, and I would not make a claim on your attention if I did not think that the relation of language to our inner and outer worlds was a matter of intellectual fascination and real-world importance.
Though “importance” is often hard to quantify, in this case I can put an exact value on it: three and a half billion dollars. That was the sum in dispute in a set of trials determining the insurance payout to Larry Silverstein, the leaseholder of the World Trade Center site. Silverstein held insurance policies that stipulated a maximum reimbursement for each destructive “event.” If 9/11 comprised a single event, he stood to receive three and a half billion dollars. If it comprised two events, he stood to receive seven billion. In the trials, the attorneys disputed the applicable meaning of the term event. The lawyers for the leaseholder defined it in physical terms (two collapses); those for the insurance companies defined it in mental terms (one plot). There is nothing “mere” about semantics!
Nor is the topic intellectually trifling. The 9/11 cardinality debate is not about the facts, that is, the physical events and human actions that took place that day. Admittedly, those have been contested as well: according to various conspiracy theories, the buildings were targeted by American missiles, or demolished by a controlled implosion, in a plot conceived by American neoconservatives, Israeli spies, or a cabal of psychiatrists. But aside from the kooks, most people agree on the facts. Where they differ is in the construal of those facts: how the intricate swirl of matter in space ought to be conceptualized by human minds. As we shall see, the categories in this dispute permeate the meanings of words in our language because they permeate the way we represent reality in our heads.
Semantics is about the relation of words to thoughts, but it is also about the relation of words to other human concerns. Semantics is about the relation of words to reality—the way that speakers commit themselves to a shared understanding of the truth, and the way their thoughts are anchored to things and situations in the world. It is about the relation of words to a community—how a new word, which arises in an act of creation by a single speaker, comes to evoke the same idea in the rest of a population, so people can understand one another when they use it. It is about the relation of words to emotions: the way in which words don’t just point to things but are saturated with feelings, which can endow the words with a sense of magic, taboo, and sin. And it is about words and social relations—how people use language not just to transfer ideas from head to head but to negotiate the kind of relationship they wish to have with their conversational partner.
A feature of the mind that we will repeatedly encounter in these pages is that even our most abstract concepts are understood in terms of concrete scenarios. That applies in full force to the subject matter of the book itself. In this introductory chapter I will preview some of the book’s topics with vignettes from newspapers and the Internet that can be understood only through the lens of semantics. They come from each of the worlds that connect to our words—the worlds of thought, reality, community, emotions, and social relations.
WORDS AND THOUGHTS
Let’s look at the bone of contention in the world’s most expensive debate in semantics, the three-and-a-half-billion-dollar argument over the meaning of “event.” What, exactly, is an event? An event is a stretch of time, and time, according to physicists, is a continuous variable—an inexorable cosmic flow, in Newton’s world, or a fourth dimension in a seamless hyperspace, in Einstein’s. But the human mind carves this fabric into the discrete swatches we call events. Where does the mind place the incisions? Sometimes, as the lawyers for the World Trade Center leaseholder pointed out, the cut encircles the change of state of an object, such as the collapse of a building. And sometimes, as the lawyers for the insurers pointed out, it encircles the goal of a human actor, such as a plot being executed. Most often the circles coincide: an actor intends to cause an object to change, the intent of the actor and the fate of the object are tracked along a single time line, and the moment of change marks the consummation of the intent.
The conceptual content behind the disputed language is itself like a language (an idea I will expand in chapters 2 and 3). It represents an analogue reality by digital, word-sized units (such as “event”), and it combines them into assemblies with a syntactic structure rather than tossing them together like rags in a bag. It’s essential to our understanding of 9/11, for example, not only that bin Laden acted to harm the United States, and that the World Trade Center was destroyed around that time, but that it was bin Laden’s act that caused the destruction. It’s the causal link between the intention of a particular man and a change in a particular object that distinguishes the mainstream understanding of 9/11 from the conspiracy theories. Linguists call the inventory of concepts and the schemes that combine them “conceptual semantics.”1 Conceptual semantics—the language of thought—must be distinct from language itself, or we would have nothing to go on when we debate what our words mean.
The fact that rival construals of a single occurrence can trigger an extravagant court case tells us that the nature of reality does not dictate the way that reality is represented in people’s minds. The language of thought allows us to frame a situation in different and incompatible ways. The unfolding of history on the morning of September 11 in New York can be thought of as one event or two events depending on how we mentally describe it to ourselves, which in turn depends on what we choose to focus on and what we choose to ignore. And the ability to frame an event in alternative ways is not just a reason to go to court but also the source of the richness of human intellectual life. As we shall see, it provides the materials for scientific and literary creativity, for humor and wordplay, and for the dramas of social life. And it sets the stage in countless arenas of human disputation. Does stem-cell research destroy a ball of cells or an incipient human? Is the American military incursion into Iraq a case of invading a country or of liberating a country? Does abortion consist of ending a pregnancy or of killing a child? Are high tax rates a way to redistribute wealth or to confiscate earnings? Is socialized medicine a program to protect citizens’ health or to expand government power? In all these debates, two ways of framing an event are pitted against each other, and the disputants struggle to show that their framing is more apt (a criterion we will explore in chapter 5). In the past decade prominent linguists have been advising American Democrats on how the Republican Party has outframed them in recent elections and on how they might regain control of the semantics of political debate by reframing, for example, taxes as membership fees and activist judges as freedom judges.2
The 9/11 cardinality debate highlights another curious fact about the language of thought. In puzzling over how to count the events of that day, it asks us to treat them as if they were objects that can be tallied, like poker chips in a pile. The debate over whether there was one event or two in New York that day is like a disagreement over whether there is one item or two at an express checkout lane, such as a pair of butter sticks taken out of a box of four, or a pair of grapefruits selling at two for a dollar. The similar ambiguity in tallying objects and tallying events is one of the many ways in which space and time are treated equivalently in the human mind, well before Einstein depicted them as equivalent in reality.
As we shall see in chapter 4, the mind categorizes matter into discrete things (like a sausage) and continuous stuff (like meat), and it similarly categorizes time into discrete events (like to cross the street) and continuous activities (like to stroll). With both space and time, the same mental zoom lens that allows us to count objects or events also allows us to zoom in even closer on what each one is made of. In space, we can focus on the material making up an object (as when we say I got sausage all over my shirt); in time, we can focus on an activity making up an event (as when we say She was crossing the street). This cognitive zoom lens also lets us pan out in space and see a collection of objects as an aggregate (as in the difference between a pebble and gravel), and it allows us to pan out in time and see a collection of events as an iteration (as in the difference between hit the nail and pound the nail). And in time, as in space, we mentally place an entity at a location and then shunt it around: we can move a meeting from 3:00 to 4:00 in the same way that we move a car from one end of the block to the other. And speaking of an end, even some of the fine points of our mental geometry carry over from space to time. The end of a string is technically a point, but we can say Herb cut off the end of the string, showing that an end can be construed as including a snippet of the matter adjacent to it. The same is true in time: the end of a lecture is technically an instant, but we can say I’m going to give the end of my lecture now, construing the culmination of an event as including a small stretch of time adjacent to it.3
As we shall see, language is saturated with implicit metaphors like EVENTS ARE OBJECTS and TIME IS SPACE. Indeed, space turns out to be a conceptual vehicle not just for time but for many kinds of states and circumstances. Just as a meeting can be moved from 3:00 to 4:00, a traffic light can go from green to red, a person can go from flipping burgers to running a corporation, and the economy can go from bad to worse. Metaphor is so widespread in language that it’s hard to find expressions for abstract ideas that are not metaphorical. What does the concreteness of language say about human thought? Does it imply that even our wispiest concepts are represented in the mind as hunks of matter that we move around on a mental stage? Does it say that rival claims about the world can never be true or false but can only be alternative metaphors that frame a situation in different ways? Those are the obsessions of chapter 5.
WORDS AND REALITY
The aftermath of 9/11 spawned another semantic debate, one with consequences even weightier than the billions of dollars at stake in how to count the events on that day. This one involves a war that has cost far more money and lives than 9/11 itself and that may affect the course of history for the rest of the century. The debate hinges on the meaning of another set of words—sixteen of them, to be exact:
The British government has learned that Saddam Hussein recently sought significant quantities of uranium from Africa.
This sentence appeared in George W. Bush’s State of the Union address in January 2003. It referred to intelligence reports suggesting that Saddam may have tried to buy five hundred tons of a kind of uranium ore called yellowcake from sources in Niger in West Africa. For many Americans and Britons the possibility that Saddam was assembling nuclear weapons was the only defensible reason to invade Iraq and depose Saddam. The United States led the invasion in the spring of that year, the most despised American foreign policy initiative since the war in Vietnam. During the occupation it became clear that Saddam had had no facilities in place to manufacture nuclear weapons, and probably had never explored the possibility of buying yellowcake from Niger. In the words of placards and headlines all over the world, “Bush Lied.”
Did he? The answer is not as straightforward as partisans on both sides might think. Investigations by the British Parliament and the U.S. Senate have established that British intelligence did believe that Saddam was trying to buy yellowcake. They showed that the evidence for the British intelligence officers’ belief at the time was not completely unreasonable but that it was far short of conclusive. And they revealed that the American intelligence experts had doubts that the report was true. Given these facts, how are we to determine whether Bush lied? It isn’t a question of whether he was unwise in putting credence in British intelligence, or of whether he made a calculated risk based on uncertain information. It’s a question of whether he was dishonest in how he conveyed this part of his rationale for the invasion to the world. And this question hinges on the semantics of one of those sixteen words, the verb learn.4
Learn is what linguists call a factive verb; it entails that the belief attributed to the subject is true. In that way it is like the verb know and unlike the verb think. Say I have a friend Mitch who mistakenly believes that Thomas Dewey defeated Harry Truman in the 1948 presidential election. I could truthfully say Mitch thinks that Dewey defeated Truman, but I couldn’t say Mitch knows that Dewey defeated Truman, because Dewey did not, in fact, defeat Truman. Mitch may think he did, but you and I know he didn’t. For the same reason I couldn’t honestly say that Mitch has admitted, discovered, observed, remembered, showed, or, crucially, learned that Dewey defeated Truman. There is, to be sure, a different sense of learn, roughly “be taught that,” which is not factive; I can say When I was in graduate school, we learned that there were four kinds of taste buds, though I now know, thanks to a recent discovery, that there are five. But the usual sense, especially in the perfect tense with have, is factive; it means “acquire true information.”
People, then, are “realists” in the philosophers’ sense. They are tacitly committed, in their everyday use of language, to certain propositions’ being true or false, independent of whether the person being discussed believes them to be true or false. Factive verbs entail something a speaker assumes to be indisputably true, not just something in which he or she has high confidence: it is not a contradiction to say I’m very, very confident that Oswald shot Kennedy, but I don’t know that he did. For this reason factive verbs have a whiff of paradox about them. No one can be certain of the truth, and most of us know we can never be certain, yet we honestly use factive verbs like know and learn and remember all the time. We must have an intuition of a degree of certitude that is so high, and so warranted by standards we share with our audience, that we can vouch for the certainty of a particular belief, while realizing that in general (though presumably not this time) we can be mistaken in what we say. Mark Twain exploited the semantics of factive verbs when he wrote, “The trouble with the world is not that people know too little, but that they know so many things that aren’t so.”5 (He also allegedly wrote, “When I was younger, I could remember anything, whether it had happened or not; but my faculties are decaying now, and soon . . . I will remember [only] the things that never happened.”)
So did Bush lie? A strong case could be made that he did. When Bush said that the British government had “learned” that Saddam had sought uranium, he was committing himself to the proposition that the uranium seeking actually took place, not that the British government believed it did. If he had reason to doubt it at the time—and the American intelligence community had made its skepticism known to his administration—the sixteen words did contain a known untruth. Defense Secretary Donald Rumsfeld, speaking in Bush’s defense, said that the statement was “technically accurate,” and National Security Advisor Condoleezza Rice added that “the British have said that.” But note the switch of verbs: Bush didn’t state that the British had said that Saddam sought yellowcake, which would be true regardless of what Saddam did; he stated that they had learned it, which could be true only if Saddam had in fact gone shopping. The logic of factivity, then, is what Bush’s critics implicitly appeal to when they accused him of lying.
Lying is an impeachable offense for a president, especially when it comes to the casus belli of a terrible war. Could semantics really be that consequential in political history? Is it plausible that the fate of an American president could ever hinge on fine points of a verb? We shall return to that question in chapter 4, where we will see that it depends upon what the meaning of the word is is.
Words are tied to reality when their meanings depend, as factive verbs do, on a speaker’s commitments about the truth. But there is a way in which words are tied to reality even more directly. They are not just about facts about the world stored in a person’s head but are woven into the causal fabric of the world itself.
Certainly a word meaning depends on something inside the head. The other day I came across the word sidereal and had to ask a literate companion what it meant. Now I can understand and use it when the companion is not around (it means “pertaining to the stars,” as in a sidereal day, the time it takes for the Earth to make a complete rotation relative to a star). Something in my brain must have changed at the moment I learned the word, and someday cognitive neuroscientists might be able to tell us what that change is. Of course most of the time we don’t learn a word by looking it up or asking someone to define it but by hearing it in context. But however a word is learned, it must leave some trace in the brain. The meaning of a word, then, seems to consist of information stored in the heads of the people who know the word: the elementary concepts that define it and, for a concrete word, an image of what it refers to.
But as we will see in chapter 6, a word must be more than a shared definition and image. The easiest way to discover this is to consider the semantics of names.6 What is the meaning of a name, such as William Shakespeare? If you were to look it up in a dictionary, you might find something like this:
Shakespeare, William (1564-1616), n.: English poet and dramatist considered one of the greatest English writers. His plays, many of which were performed at the Globe Theatre in London, include historical works, such as Richard II, comedies, including Much Ado about Nothing and As You Like It, and tragedies, such as Hamlet, Othello, and King Lear. He also composed 154 sonnets. [Syn.: Shakespeare, Shakspeare, William Shakspere, the bard]
And the definition would typically be accompanied by the famous engraving of a doe-eyed balding man with a very small mustache and a very big ruff. Presumably that is not too far from your understanding of the name.
But is that what William Shakespeare really means? Historians agree that there was a man named William Shakespeare who lived in Stratford-on-Avon and London in the late sixteenth and early seventeenth centuries. But for 250 years there have been doubts as to whether that man composed the plays we attribute to him. This might sound like the theory that the CIA imploded the World Trade Center, but it has been taken seriously by Walt Whitman, Mark Twain, Henry James, and many modern-day scholars, and it rests on a number of damning facts. Shakespeare’s plays were not published as serious literature in his lifetime, and authorship in those days was not recorded as carefully as it is today. The man himself was relatively uneducated, never traveled, had illiterate children, was known in his hometown as a businessman, was not eulogized at his death, and left no books or manuscripts in his will. Even the famous portraits were not painted in his lifetime, and we have no reason to believe that they resembled the man himself. Because writing plays was a disreputable occupation in those days, the real author, identified by various theories as Francis Bacon, Edward de Vere, Christopher Marlowe, and even Queen Elizabeth, may have wanted to keep his or her identity a secret.
My point isn’t to persuade you that William Shakespeare was not the great English poet and dramatist who wrote Hamlet, As You Like It, and 154 sonnets. (Mainstream scholars say he was, and I believe them.) My point is to get you to think about the possibility that he wasn’t, and to understand the implications for the idea that the meanings of words are in the head. For the sake of argument, imagine that forensic evidence proved beyond doubt that the Shakespearean oeuvre was written by someone else. Now, if the meaning of William Shakespeare were something like the dictionary entry stored in the head, we would have to conclude either that the meaning of the term William Shakespeare had changed or that the real author of Hamlet should be posthumously christened William Shakespeare, even though no one knew him by that name in his lifetime. (We would also have to give full marks to the hapless student who wrote in an exam, “Shakespeare’s plays were written by William Shakespeare or another man of that name.”) Actually, it’s even worse than that. We would not have been able to ask “Did Shakespeare write Hamlet?” in the first place, because he did by definition. It would be like asking “Is a bachelor unmarried?” or “Who’s buried in Grant’s Tomb?” or “Who sang ‘Hey, Hey, We’re the Monkees’?” And the conclusion, “William Shakespeare did not in fact write Hamlet,” would be self-contradictory.
But these implications are bizarre. In fact we are speaking sensibly when we ask whether Shakespeare wrote Hamlet; we would not be contradicting ourselves if we were to conclude that he did not; and we would still feel that William Shakespeare means what it always meant—some guy who lived in England way back when—while admitting that we were mistaken about the man’s accomplishments. Even if every biographical fact we knew about Shakespeare were overturned—if it turned out, for example, that he was born in 1565 rather than 1564, or came from Warwick rather than Stratford—we would still have a sense that the name refers to the same guy, the one we’ve been talking about all along.
So what exactly does William Shakespeare mean, if not “great writer, author of Hamlet,” and so on? A name really has no definition in terms of other words, concepts, or pictures. Instead it points to an entity in the world, because at some instant in time the entity was dubbed with the name and the name stuck. William Shakespeare, then, points to the individual who was christened William by Mr. and Mrs. Shakespeare around the time he was born. The name is connected to that guy, whatever he went on to do, and however much or little we know about him. A name points to a person in the world in the same way that I can point to a rock in front of me right now. The name is meaningful to us because of an unbroken chain of word of mouth (or word of pen) that links the word we now use to the original act of christening. We will see that it’s not just names, but words for many kinds of things, that are rigidly yoked to the world by acts of pointing, dubbing, and sticking rather than being stipulated in a definition.
The tethering of words to reality helps allay the worry that language ensnares us in a self-contained web of symbols. In this worry, the meanings of words are ultimately circular, each defined in terms of the others. As one semanticist observed, a typical dictionary plays this game when it tells the user that “to order means to command, that to direct and instruct ‘are not so strong as command or order,’ that command means ‘to direct, with the right to be obeyed,’ that direct means ‘to order,’ that instruct means ‘to give orders’; or that to request means ‘to demand politely,’ to demand [means] ‘to claim as if by right,’ to claim [means] ‘to ask for or demand,’ to ask [means] ‘to make a request,’ and so on.”7 This cat’s cradle is dreaded by those who crave certainty in words, embraced by adherents of deconstructionism and postmodernism, and exploited by the writer of a dictionary of computer jargon:
endless loop, n. See loop, endless.
loop, endless, n. See endless loop.
The logic of names, and of other words that are connected to events of dubbing, allay these concerns by anchoring the web of meanings to real events and objects in the world.
The connectedness of words to real people and things, and not just to information about those people and things, has a practical application that is very much in the news. The fastest-growing crime in the beginning of this century is identity theft. An identity thief uses information connected with your name, such as your social security number or the number and password of your credit card or bank account, to commit fraud or steal your assets. Victims of identity theft may lose out on jobs, loans, and college admissions, can be turned away at airport security checkpoints, and can even get arrested for a crime committed by the thief. They can spend many years and much money reclaiming their identity.
Put yourself in the shoes of someone who has lost his wallet, or inadvertently divulged information on his computer, and now has a doppelgänger using his name (say, Murray Klepfish) to borrow money or make purchases. Now you have to convince a bureaucrat that you, not the impostor, are the real Murray Klepfish. How would you do it? As with William Shakespeare, it comes down to what the words Murray Klepfish mean. You could say, “ ‘Murray Klepfish’ means an owner of a chain of discount tire stores who was born in Brooklyn, lives in Piscataway, has a checking account at Acme Bank, is married with two sons, and spends his summers on the Jersey Shore.” But they would reply, “As far as we are concerned, ‘Murray Klepfish’ means a personal trainer who was born in Delray Beach, gets his mail at a post office box in Albuquerque, charged a recent divorce to a storefront in Reno, and spends his summers on Maui. We do agree with you about the bank account, which, by the way, is severely overdrawn.”
So how would you prove that you are the real referent of the name Murray Klepfish? You could provide any information you wanted—social security number, license number, mother’s maiden name—and the impersonator can either duplicate it (if he stole that, too) or contest it (if he augmented the stolen identity with his own particulars, including a photograph). As with picking out the real Shakespeare after his familiar biographical particulars had been cast into doubt, ultimately you would have to point to a causal chain that links your name as it is used today to the moment your parents hailed your arrival. Your credit card was obtained from a bank account, which was obtained with a driver’s license, which was obtained with a birth certificate, which was vouched for by a hospital official, who was in touch with your parents around the time of your birth and heard from their lips that you are the Klepfish they were naming Murray. In the case of your impostor, the chain of testimony peters out in the recent past, well short of the moment of dubbing. The measures designed to foil identity theft depend upon the logic of names and the connection of words to reality: they are ways to identify unbroken chains of person-to-person transmission through time, anchored to a specific event of dubbing in the past.
WORDS AND COMMUNITY
Naming a child is the only opportunity that most people get to anoint an entity in the world with a word of their choosing. Apart from creative artists like Frank Zappa, who named his children Moon Unit and Dweezil, traditionally most people select a prefabricated forename like John or Mary rather than a sound they concoct from scratch. In theory a forename is an arbitrary label with no inherent meaning, and people interpret it as simply pointing to the individual who was dubbed with it. But in practice names take on a meaning by association with the generation and class of people who bear them. Most American readers, knowing nothing else about a man other than that his name is Murray, would guess that he is over sixty, middle-class, and probably Jewish. (When a drunken Mel Gibson let loose with an anti-Semitic tirade in 2006, the editor Leon Wieseltier commented, “Mad Max is making Max mad, and Murray, and Irving, and Mort, and Marty, and Abe.”)8 That is because of another curiosity of names we will explore in chapter 6. Names follow cycles of fashion, like the widths of ties and the lengths of skirts, so people’s first names may give away their generational cohort. In its heyday in the 1930s, Murray had an aura of Anglo-Saxon respectability, together with names like Irving, Sidney, Maxwell, Sheldon, and Herbert. They seemed to stand apart from the Yiddish names of the previous generation, such as Moishe, Mendel, and Ruven, which made their bearers sound as if they had a foot in the old country. But when the Murrays and Sids and their wives launched the baby boom, they gave their sons blander names like David, Brian, and Michael, who in their turn begat biblically inspired Adams, Joshuas, and Jacobs. Many of these Old Testament namesakes are now completing the circle with sons named Max, Ruben, and Saul.
Names follow trends because people in a community have uncannily similar reactions to the ones in the namepool (as parents often find when they take a child to school and discover that their unique choice of a name was also the unique choice of many of their neighbors). A name’s coloring comes in part from the sounds that go into it and in part from a stereotype of the adults who currently bear it. For this reason, the faux-British names of first-generation Americans became victims of their own middle-class respectability a generation later. In a scene from When Harry Met Sally set in the 1970s, a pair of baby boomers get into an argument about Sally’s sexual experience:
HARRY: With whom did you have this great sex?
SALLY: I’m not going to tell you that!
HARRY: Fine. Don’t tell me.
SALLY: Shel Gordon.
HARRY: Shel. Sheldon? No, no. You did not have great sex with Sheldon.
SALLY: I did too.
HARRY: No, you didn’t. A Sheldon can do your income taxes. If you need a root canal, Sheldon’s your man. But humpin’ and pumpin’ is not Sheldon’s strong suit. It’s the name. “Do it to me, Sheldon.” “You’re an animal, Sheldon.” “Ride me, big Sheldon.” It doesn’t work.
Though postwar parents probably didn’t have great sex in mind, they must have recoiled from the name’s nebbishy connotation even then: beginning in the 1940s, Sheldon, like Murray, sank like a stone and never recovered.9 The reaction to the name is now so uniform across the English-speaking world that humorists can depend on it. The playwright Marcy Kahan, who recently adapted Nora Ephron’s screenplay to the British stage, notes, “I included the Sheldon joke in the stage play, and all three actors playing Harry got a huge laugh of recognition from it, every night, without fail.”10
The dynamics of baby naming have become a talking point in newspapers and conversation now that the fashion cycles have accelerated. One of the most popular American names for baby girls in 2006 was unheard of only five years before: Nevaeh, or “heaven” spelled backwards. At the other end of the curve, people are seeing their own names, and the names of their friends and relations, becoming stodgy more quickly.11 I don’t think I ever felt so old as when a student told me that Barbara, Susan, Deborah, and Linda, some of the most popular names for girls of my generation, made her think of middle-aged women.
In naming a baby, parents have free rein. Obviously they are affected by the pool of names in circulation, but once they pick one, the child and the community usually stick with it. But in naming everything else, the community has a say in whether the new name takes. The social nature of words is illustrated in Calvin’s presumably ill-fated attempt to pass a physics exam:
The way in which we understand “your own words”—as referring only to how you combine them, not to what they are—shows that words are owned by a community rather than an individual. If a word isn’t known to everyone around you, you might as well not use it, because no one will know what you’re talking about. Nonetheless, every word in a language must have been minted at some point by a single speaker. With some coinages, the rest of the community gradually agrees to use the word to point to the same thing, tipping the first domino in the chain that makes the word available to subsequent generations. But as we shall see, how this tacit agreement is forged across a community is mysterious.
In some cases necessity is the mother of invention. Computer users, for instance, needed a term for bulk e-mail in the 1990s, and spam stepped into the breach. But many other breaches stay stubbornly unstepped into. Since the sexual revolution of the 1960s we have needed a term for the members of an unmarried heterosexual couple, and none of the popular suggestions has caught on—paramour is too romantic, roommate not romantic enough, partner too gay, and the suggestions of journalists too facetious (like POSSLQ, from the census designation “persons of opposite sex sharing living quarters,” and umfriend, from “This is my, um, friend”). And speaking of decades, we are more than halfway through the first one of the twenty-first century, and no one yet knows what to call it. The zeroes? The aughts? The nought-noughts? The naughties?
Traditional etymology is of limited help in figuring out what ushers a word into existence and whether it will catch on. Etymologists can trace most words back for centuries or more, but the trail goes cold well before they reach the actual moment at which a primordial wordsmith first dubbed a concept with a sound of his or her choosing. With recent coinages, though, we can follow the twisted path to wordhood in real time.
Spam is not, as some people believe, an acronym for Short, Pointless, and Annoying Messages. The word is related to the name of the luncheon meat sold by Hormel since 1937, a portmanteau from SPiced hAM. But how did it come to refer to e-mailed invitations to enlarge the male member and share the ill-gotten gains of deposed African despots? Many people assume that the route was metaphor. Like the luncheon meat, the e-mail is cheap, plentiful, and unwanted, and in one variant of this folk etymology, spamming is what happens when you dump Spam in a fan. Though these intuitions may have helped make the word contagious, its origin is very different. It was inspired by a sketch from Monty Python’s Flying Circus in which a couple enter a café and ask the waitress (a Python in drag) what’s available. She answers:
Well, there’s egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam; spam bacon sausage and spam; spam egg spam spam bacon and spam; spam sausage spam spam bacon spam tomato and spam; spam spam spam egg and spam; spam spam spam spam spam spam baked beans spam spam spam, or Lobster Thermidor: a Crevette with a mornay sauce served in a Provençale manner with shallots and aubergines garnished with truffle pâté, brandy and with a fried egg on top and spam.
You are probably thinking, “This sketch must be stopped—it’s too silly.” But it did change the English language. The mindless repetition of the word spam inspired late-1980s hackers to use it as a verb for flooding newsgroups with identical messages, and a decade later it spread from their subculture to the populace at large.12
Though it may seem incredible that such a whimsical and circuitous coinage would catch on, we shall see that it was not the first time that silliness left its mark on the lexicon. The verb gerrymander comes from a nineteenth-century American cartoon showing a political district that had been crafted by a Governor Elbridge Gerry into a tortuous shape resembling a salamander in an effort to concentrate his opponent’s voters into a single seat. But most silly coinages go nowhere, such as bushlips for “insincere political rhetoric” (after George H. W. Bush’s 1988 campaign slogan “Read my lips: No new taxes”), or teledildonics for computer-controlled sex toys. Every year the American Dialect Society selects a “word most likely to succeed.” But the members of the society are the first to admit that their track record is abysmal. Does anyone remember the information superhighway, or the Infobahn?13 And could anyone have predicted that to blog, to google, and to blackberry would quickly become part of everyone’s language?
The dynamics of taking from the wordpool when naming babies and giving back to it when naming concepts are stubbornly chaotic. And as we shall see, this unpredictability holds a lesson for our understanding of culture more generally. Like the words in a language, the practices in a culture—every fashion, every ritual, every common belief—must originate with an innovator, must then appeal to the innovator’s acquaintances and then to the acquaintance’s acquaintances, and so on, until it becomes endemic to a community. The caprice in the rise and fall of names, which are the most easily tracked bits of culture, suggests we should be skeptical of most explanations for the life cycles of other mores and customs, from why men stopped wearing hats to why neighborhoods become segregated. But it also points to the patterns of individual choice and social contagion that might someday make sense of them.
WORDS AND EMOTIONS
The shifting associations to the name for a person are an example of the power of a word to soak up emotional coloring—to have a connotation as well as a denotation. The concept of a connotation is often explained by the conjugational formula devised by Bertrand Russell in a 1950s radio interview: I am firm; you are obstinate; he is pigheaded. The formula was turned into a word game in a radio show and newspaper feature and elicited hundreds of triplets. I am slim; you are thin; he is scrawny. I am a perfectionist; you are anal; he is a control freak. I am exploring my sexuality; you are promiscuous; she is a slut. In each triplet the literal meaning of the words is held constant, but the emotional meaning ranges from attractive to neutral to offensive.
The affective saturation of words is especially apparent in the strange phenomena surrounding profanity, the topic of chapter 7. It is a real puzzle for the science of mind why, when an unpleasant event befalls us—we slice our thumb along with the bagel, or knock a glass of beer into our lap—the topic of our conversation turns abruptly to sexuality, excretion, or religion. It is also a strange feature of our makeup that when an adversary infringes on our rights—say, by slipping into parking space we have been waiting for, or firing up a leaf blower at seven o’clock on a Sunday morning—we are apt to extend him advice in the manner of Woody Allen, who recounted, “I told him to be fruitful and multiply, but not in those words.”
These outbursts seem to emerge from a deep and ancient part of the brain, like the yelp of a dog when someone steps on its tail, or its snarl when it is trying to intimidate an adversary. They can surface in the involuntary tics of a Tourette’s patient, or in the surviving utterances of a neurological patient who is otherwise bereft of language. But despite the seemingly atavistic roots of cursing, the sounds themselves are composed of English words and are pronounced in full conformity with the sound pattern of the language. It is as though the human brain were wired in the course of human evolution so that the output of an old system for calls and cries were patched into the input of the new system for articulate speech.
Not only do we turn to certain words for sexuality, excretion, and religion when we are in an excitable state, but we are wary of such words when we are in any other state. Many epithets and imprecations are not just unpleasant but taboo: the very act of uttering them is an affront to listeners, even when the concepts have synonyms whose use is unexceptionable. The tendency of words to take on awesome powers may be found in the taboos and word magic in cultures all over the world. In Orthodox Judaism, the name of God, transcribed as YHVH and conventionally pronounced Yahweh, may never be spoken, except by high priests in the ancient temple on Yom Kippur in the “holy of holies,” the chamber housing the ark of the covenant. In everyday conversation observant Jews use a word to refer to the word, referring to God as hashem, “the name.”
While taboo language is an affront to common sensibilities, the phenomenon of taboo language is an affront to common sense. Excretion is an activity that every incarnate being must engage in daily, yet all the English words for it are indecent, juvenile, or clinical. The elegant lexicon of Anglo-Saxon monosyllables that give the English language its rhythmic vigor turns up empty-handed just when it comes to an activity that no one can avoid. Also conspicuous by its absence is a polite transitive verb for sex—a word that would fit into the frame Adam verbed Eve or Eve verbed Adam. The simple transitive verbs for sexual relations are either obscene or disrespectful, and the most common ones are among the seven words you can’t say on television.
Or at least, the words you couldn’t say in 1973, when the comedian George Carlin delivered his historic monologue arguing against the ban of those words in broadcast media. In a conundrum that reminds us of the rationale for unfettered free speech, a radio network that had broadcasted the monologue was punished by the Federal Communications Commission (in a case that ultimately reached the Supreme Court) for allowing Carlin to mention on the radio exactly those words that he was arguing ought to be allowed to be mentioned on the radio. We have a law that in effect forbids criticism of itself, a paradox worthy of Russell and other connoisseurs of self-referential statements. The paradox of identifying taboo words without using them has always infected attempts to regulate speech about sexuality. In several states, the drafters of the statute against bestiality could not bring themselves to name it and therefore outlawed “the abominable and detestable crime against nature,” until the statutes were challenged for being void for vagueness. To avoid this trap, a New Jersey obscenity statute stipulated exactly which kinds of words and images would be deemed obscene. But the wording of the statute was so pornographic that some law libraries tore the page out of every copy of the statute books.14
Taboos on language are still very much in the news. While sexual and scatological language is more available than ever on cable, satellite, and the Internet, the American government, prodded by cultural conservatives, is trying to crack down on it, especially within the dwindling bailiwick of broadcast media. Legislation such as the “Clean Airwaves Act” and the “Broadcast Decency Enforcement Act” imposes draconian fines on broadcast stations that fail to censor their guests when they use the words on Carlin’s list. And in an unscripted event that shows the unavoidable hypocrisy of linguistic taboos, the Broadcast Decency Enforcement Act was passed on the day in 2004 that Vice President Dick Cheney got into an argument with Senator Patrick Leahy on the Senate floor and Cheney told the senator to be fruitful and multiply, but not in those words.
No curious person can fail to be puzzled by the illogic and hypocrisy of linguistic taboos. Why should certain words, but not their homonyms or synonyms, be credited with a dreadful moral power? At the same time, no matter how illogical it may seem, everyone respects taboos on at least some words. Everyone? Yes, everyone. Suppose I told you there was an obscenity so shocking that decent people dare not mention it even in casual conversation. Like observant Jews referring to God, they must speak of it at one degree of separation by using a word that refers to the word. An elect circle of people are granted a special dispensation to use it, but everyone else risks grave consequences, including legally justifiable violence.15 What is this obscenity? It is the word nigger—or, as it is referred to in respectable forums, the n-word—which may be uttered only by African Americans to express camaraderie and solidarity in settings of their choosing. The shocked reaction that other uses evoke, even among people who support free speech and wonder why there is such a fuss about words for sex, suggests that the psychology of word magic is not just a pathology of censorious bluenoses but a constituent of our emotional and linguistic makeup.
WORDS AND SOCIAL RELATIONS
In recent years the Internet has become a laboratory for the study of language. It not only provides a gigantic corpus of real language used by real people, but also acts as a superefficient vector for the transmission of infectious ideas, and can thereby highlight examples of language that people find intriguing enough to pass along to others. Let me introduce the last major topic of this book with a story that circulated widely by e-mail in 1998:
During the final days at Denver’s Stapleton airport, a crowded United flight was canceled. A single agent was rebooking a long line of inconvenienced travelers. Suddenly an angry passenger pushed his way to the desk and slapped his ticket down on the counter, saying, “I HAVE to be on this flight, and it HAS to be first class.” The agent replied, “I’m sorry, sir. I’ll be happy to try to help you, but I’ve got to help these folks first, and I’m sure we’ll be able to work something out.” The passenger was unimpressed. He asked loudly, so that the passengers behind him could hear, “Do you have any idea who I am?” Without hesitating, the gate agent smiled and grabbed her public address microphone. “May I have your attention, please?” she began, her voice bellowing through the terminal. “We have a passenger here at the gate WHO DOES NOT KNOW WHO HE IS. If anyone can help him find his identity, please come to the gate.” With the folks behind him in line laughing hysterically, the man glared at the agent, gritted his teeth, and swore, “[Expletive] you!” Without flinching, she smiled and said, “I’m sorry, sir, but you’ll have to stand in line for that too.”
The story seems too good to be true, and is probably an urban legend.16 But its two punch lines make a nice teaser for oddities of language that we will explore in later chapters. I have already touched on the puzzle behind the second punch line, namely that certain words for sex are also used in aggressive imprecations (chapter 7). But the first punch line introduces the final world that I wish to connect to words, the world of social relations (chapter 8).
The agent’s comeback to “Do you have any idea who I am?” springs from a mismatch between the sense in which the passenger intended his rhetorical question—a demand for recognition of his higher status—and the sense in which she pretended to take it—a literal request for information. And the payoff to the onlookers (and the e-mail audience) comes from understanding the exchange at a third level—that the agent’s feigned misunderstanding was a tactic to reverse the dominance relation and demote the arrogant passenger to well-deserved ignominy.
Language is understood at multiple levels, rather than as a direct parse of the content of the sentence.17 In everyday life we anticipate our interlocutor’s ability to listen between the lines and slip in requests and offers that we feel we can’t blurt out directly. In the film Fargo, two kidnappers with a hostage hidden in the back seat are pulled over by a policeman because their car is missing its plates. The kidnapper at the wheel is asked to produce his driver’s license, and he extends his wallet with a fifty-dollar bill protruding from it, saying, “So maybe the best thing would be to take care of that here in Brainerd.” The statement, of course, is intended as a bribe, not as a comment on the relative convenience of different venues for paying the fine. Many other kinds of speech are interpreted in ways that differ from their literal meaning:
If you could pass the guacamole, that would be awesome.
We’re counting on you to show leadership in our Campaign for the Future.
Would you like to come up and see my etchings?
Nice store you got there. Would be a real shame if something happened to it.
These are clearly intended as a request, a solicitation for money, a sexual come-on, and a threat. But why don’t people just say what they mean—“If you let me drive off without further ado, I’ll give you fifty bucks,” “Gimme the guacamole,” and so on?
With the veiled bribe and the veiled threat, one might guess that the technicalities of plausible deniability are applicable: bribery and extortion are crimes, and by avoiding an explicit proposition, the speaker could make a charge harder to prove in court. But the veil is so transparent that it is hard to believe it could foil a prosecutor or fool a jury—as the lawyers say, it wouldn’t pass the giggle test. Yet we all take part in these charades, while knowing that no one is fooled. (Well, almost no one. In an episode of Seinfeld, George is asked by his date if he would like to come up for coffee. He declines, explaining that caffeine keeps him up at night. Later he slaps his forehead and realizes, “ ‘Coffee’ doesn’t mean coffee! ‘Coffee’ means sex!” And of course this can go too far in the other direction. In a joke recounted by Freud in Jokes and Their Relation to the Unconscious, a businessman meets a rival at a train station and asks him where he’s going. The second businessman says he’s going to Minsk. The first one replies, “You’re telling me you’re going to Minsk because you want me to think you’re going to Pinsk. But I happen to know that you are going to Minsk. So why are you lying to me?”)
If a speaker and a listener were ever to work through the tacit propositions that underlie their conversation, the depth of the recursively embedded mental states would be dizzying. The driver offers a bribe; the officer knows that the driver is offering him a bribe; the driver knows that the officer knows; the officer knows that the driver knows that the officer knows; and so on. So why don’t they just blurt it out? Why do a speaker and a hearer willingly take on parts in a dainty comedy of manners?
The polite dinnertime request—what linguists call a whimperative—offers a clue. When you issue a request, you are presupposing that the hearer will comply. But apart from employees or intimates, you can’t just boss people around like that. Still, you do want the damn guacamole. The way out of this dilemma is to couch your request as a stupid question (“Can you . . .? ”), a pointless rumination (“I was wondering if . . .”), a gross overstatement (“It would be great if you could . . .”), or some other blather that is so incongruous the hearer can’t take it at face value. She does some quick intuitive psychology to infer your real intent, and at the same time she senses that you have made an effort not to treat her as a factotum. A stealth imperative allows you to do two things at once—communicate your request, and signal your understanding of the relationship.
As we shall see in chapter 8, ordinary conversation is like a session of tête-à-tête diplomacy, in which the parties explore ways of saving face, offering an “out,” and maintaining plausible deniability as they negotiate the mix of power, sex, intimacy, and fairness that makes up their relationship. As with real diplomacy, communiqués that are too subtle, or not subtle enough, can ignite a firestorm. In 1991, the nomination of Clarence Thomas to the U.S. Supreme Court was nearly derailed by accusations that he had made sexual overtures to a subordinate, the lawyer Anita Hill. In one of the stranger episodes in the history of the Senate’s exercise of its power of advice and consent, senators had to decide what Thomas meant when he spoke to Hill about a porn star named Long Dong Silver and when he asked the rhetorical question “Who has put pubic hair on my Coke?” It’s presumably not what the Framers had in mind when they formulated the doctrine of the separation of powers, but this kind of question has become a part of our national discourse. Ever since the Thomas-Hill case put sexual harassment on the national stage, the adjudication of claims of harassment has been a major headache for universities, businesses, and government agencies, particularly when a putative come-on is conveyed by innuendo rather than a bald proposition.
These tidbits from the news and from the net show some of the ways in which our words connect to our thoughts, our communities, our emotions, our relationships, and to reality itself. It isn’t surprising that language supplies so many of the hot potatoes of our public and private life. We are verbivores, a species that lives on words, and the meaning and use of language are bound to be among the major things we ponder, share, and dispute.
At the same time, it would be a mistake to think that these deliberations are really about language itself. As I will show in chapter 3, language is above all a medium in which we express our thoughts and feelings, and it mustn’t be confused with the thoughts and feelings themselves. Yet another phenomenon of language, the symbolism in sound (chapter 6), offers a hint at this conclusion. Without a substrate of thoughts to underlie our words, we do not truly speak but only babble, blabber, blather, chatter, gibber, jabber, natter, patter, prattle, rattle, yammer, or yadda, yadda—an onomatopoeic lexicon for empty speech that makes plain the expectation that the sounds coming out of our mouths are ordinarily about something.
The rest of this book is about that something: the ideas, feelings, and attachments that are visible through our language and that make up our nature. Our words and constructions disclose conceptions of physical reality and human social life that are similar in all cultures but different from the products of our science and scholarship. They are rooted in our development as individuals, but also in the history of our language community, and in the evolution of our species. Our ability to combine them into bigger assemblies and to extend them to new domains by metaphorical leaps goes a long way toward explaining what makes us smart. But they can also clash with the nature of things, and when they do, the result can be paradox, folly, and even tragedy. For these reasons I hope to convince you that the three and a half billion dollars at stake in the interpretation of an “event” is just part of the value of understanding the worlds of words.
DOWN THE RABBIT HOLE
The discovery of a world hidden in a nook or cranny of everyday life is an enduring device in children’s fiction. The best-known example is Alice stumbling down a rabbit hole to find a surreal underworld, and the formula continues to enchant in endless variations: the wardrobe passageway to Narnia, the wrinkle in time, the subtle knife, Whoville in a speck of dust.1
In nonfiction as well, the revelation of a microcosm is a recurring source of fascination. In 1968, the designers Charles and Ray Eames made a film called Powers of Ten, which began with a view of galaxy clusters a billion light-years across, and zoomed by tenfold leaps to reveal our galaxy, solar system, planet, and so on, down to a picnicker asleep in a park, to his hand, his cells, his DNA, a carbon atom, and finally the atomic nucleus and its particles sixteen orders of magnitude smaller. This magnificent unfurling of physical reality can be seen in a companion book by the film’s scientific consultants, Philip and Phyllis Morrison, and the idea has recently been adapted to one of the most enjoyable ways to waste time on the Web: zooming smoothly from a photograph of the Earth taken from space through seven orders of magnitude of satellite photographs down to a pigeon’s-eye view of your street and house.
This chapter is about my own stumbling upon a microcosm—the world of basic human ideas and their connections—in the course of trying to solve what I thought was a mundane problem in psycholinguistics. It is a hidden world that I had glimpsed not by training a telescope on its whereabouts from the start but because it kept peeking out from under the phenomena I thought I was studying. By taking you through the layers of mental organization that must be exploded to make sense of the problem, I hope to offer you a view of this inner world.2
The rabbit hole that leads to this microcosm is the verb system of English—what verbs mean, how they are used in sentences, and how children figure it all out. This chapter will try to show you how cracking these problems led to epiphanies about the contents of cognition that serve as leitmotifs of this book. Why leap into the world of the mind through this particular opening? One reason, I confess, is personal: I simply find verbs fascinating. (A colleague once remarked, “They really are your little friends, aren’t they?”) But as every enthusiast knows, other people can’t be counted on to share one’s passion, and I like to think I have a better reason to introduce you to my little friends.
Science proceeds by studying particulars. No one has ever gotten a grant to study “the human mind.” One has to study something more tractable, and when fortune smiles, a general law may reveal itself in the process. In the first chapter I introduced four ideas:
• The human mind can construe a particular scenario in multiple ways.
• Each construal is built around a few basic ideas, like “event,” “cause,” “change,” and “intend.”
• These ideas can be extended metaphorically to other domains, as when we count events as if they were objects or when we use space as a metaphor for time.
• Each idea has distinctively human quirks that make it useful for reasoning about certain things but that can lead to fallacies and confusions when we try to apply it more broadly.
These claims may strike you as reasonable enough, but not particularly meaty—just four out of hundreds of platitudes that could be listed as true of our thought processes. In this chapter I hope to show that they are more than that. In solving the problem of how children learn verbs, each of these hypotheses served as a puzzle piece that took a long time to find but then fit perfectly into its slot, together completing an attractive picture of the whole. This offers some confidence that the themes of this book are real discoveries about the mind, not just innocuous comments about it.
My plan is as follows. First I will take you on a plunge from the intergalactic perspective to the quark’s-eye view, showing how a general curiosity about how the mind works can lead to an interest in verbs and how children learn them.
Then we will bump up against a paradox—a case in which children seem to learn the unlearnable. Isaac Asimov once wrote, “The most exciting phrase to hear in science, the one that heralds new discoveries, is not ‘Eureka!’ (I found it!) but ‘That’s funny. . . .’” The following section presents a discovery—the mind’s ability to flip between frames—that was the crucial opening to solve the paradox.
The remaining parts of the solution bring us face to face with two of the basic concepts in our mental inventory, moving and changing. The same line of reasoning, applied to other verbs, illuminates the other major elements our thoughts are built from: the concepts of having, knowing, and helping, and the concepts of acting, intending, and causing.
From there we step back up to reflect on what it all means. We will consider whether the signs of intelligent design in the English language imply a corresponding intelligence in every English speaker—a question that will recur throughout the book as we try to use language as a window into human nature. I will then suggest an inventory of basic human thoughts, ones that will be unpacked in later chapters. Finally, I will show how design quirks in these basic thoughts give rise to fallacies, follies, and foibles in the way that people reason about the conundrums of modern life.
POWERS OF TEN
Let me now take you, in a few turns of the zoom lens, from a wide concern with human nature to a close-up look at how children learn verbs.
The first, galaxy-wide view is of the human mind and its remarkable powers. It’s easy for us humans, safe inside our well-functioning minds, to be jaded about the mundane activities of cognition and to attend instead to the extraordinary and the lurid. But the science of mind begins with a recognition that ordinary mental activities—seeing, hearing, remembering, moving, planning, reasoning, speaking—require our brains to solve fractious engineering problems.3 Despite the immense hazard and cost of manned space flight, most plans for planetary exploration still envision blasting people into the solar system. Partly it’s because of the drama of following an intrepid astronaut in exploring strange new worlds rather than a silicon chip, but mainly it’s because no foreseeable robot can match an ordinary person’s ability to recognize unexpected objects and situations, decide what to do about them, and manipulate things in unanticipated ways, all while exchanging information with humans back home. Understanding how these faculties of mind work is a frontier of modern science.
Among these magnificent faculties, pride of place must go to language—ubiquitous across the species, unique in the animal kingdom, inextricable from social life and from the mastery of civilization and technology, devastating when lost or impaired.4
Language figures in human life in many ways. We inform, we request, we persuade, we interrogate, we orate, and sometimes we just schmooze. But the most remarkable thing we do with language is learn it in the first place.5 Babies are born into the world not knowing a word of the language being spoken around them. Yet in just three years, without the benefit of lessons, most of them will be talking a blue streak, with a vocabulary of thousands of words, a command of the grammar of the spoken vernacular, and a proficiency with the sound pattern (what tourist isn’t momentarily amazed at how well the little children in France speak French!). Children deploy the code of syntax unswervingly enough to understand improbable events like a cow jumping over the moon and a dish running away with the spoon, or to share their childlike aperçus like “I think the wind wants to get in out of the rain” or “I often wonder when people pass me by do they wonder about me.”6
To become so fluent in a language, children must have analyzed the speech around them, not just memorized it. We see this clearly when children say things that sound wrong to adult ears but that reveal acute hypotheses about how the ingredients of language may combine. When children make errors like “All the animals are wake-upped,” “Don’t tickle me; I’m laughable,” or “Mommy, why did he dis it appear?” they could not have been imitating their parents. They must have extracted the mental equivalent of grammatical rules that add suffixes to words and arrange verbs and particles in phrases.
The triumph of language acquisition is even more impressive when we consider that a talking child has solved a knotty instance of the problem of induction: observing a finite sample of events and framing a generalization that embraces the infinite set from which the events are drawn.7 Scientists engage in induction when they go beyond their data and put forward laws that make predictions about cases they haven’t observed, such as that gas under pressure will be absorbed by a liquid, or that warm-blooded animals have larger body sizes at higher latitudes. Philosophers of science call induction a “scandal” because there are an infinite number of generalizations that are consistent with any set of observations, and no strictly logical basis for choosing among them.8 There is no guarantee that a law discovered this year will continue to hold next year, no limit to the number of smooth curves that can connect a set of points on a graph, and, upon glimpsing a black sheep in Scotland, no strictly logical reason for choosing among the conclusions that all sheep in Scotland are black, that at least one sheep in Scotland is black, and that at least one sheep in Scotland is black on at least one side. As Mark Twain wrote, science is fascinating because “one gets such wholesale returns on conjecture out of such a trifling investment in fact.” Yet the returns keep coming. Philosophers of science argue that theories are not just peeled off the data but constrained beforehand by reasonable assumptions about the way the universe works, such as that nature is lawful and that simpler theories that fit the data are more likely to be true than complex ones.
As children learn their mother tongue, they, too, are solving an induction problem. When listening to their parents and siblings, they can’t just file away every sentence and draw on that list in the future, or they would be as mindless as parrots. Nor can they throw together all the words they have found in any order they please. They have to extract a set of rules that will allow them to understand and express new thoughts, and do it in a way that is consistent with the speech patterns used by those around them. The induction problem arises because ambient speech offers countless opportunities for the child to leap on seductive yet false generalizations. For instance, as children learn how to ask questions, they should be able to go from He ate the green eggs with ham to What did he eat? and What did he eat the green eggs with? But from He ate the green eggs and ham they should not be able to ask What did he eat the green eggs and? To take another example: the sentences Harriet appeared to Sam to be strong and Harriet appealed to Sam to be strong differ by only the curl of the tongue in a single consonant. Yet their meanings (in particular, who is supposed to be the strong one) are completely different. A child hearing one sentence should not generalize its interpretation to the other just because they sound so similar.
In cracking the code of language, then, children’s minds must be constrained to pick out just the right kinds of generalizations from the speech around them. They can’t get sidetracked by how sentences sound but must dig into the grammatical structure hidden in the words and their arrangement. It is this line of reasoning that led the linguist Noam Chomsky to propose that language acquisition in children is the key to understanding the nature of language, and that children must be equipped with an innate Universal Grammar: a set of plans for the grammatical machinery that powers all human languages.9 This idea sounds more controversial than it is (or at least more controversial than it should be) because the logic of induction mandates that children make some assumptions about how language works in order for them to succeed at learning a language at all.10 The only real controversy is what these assumptions consist of: a blueprint for a specific kind of rule system, a set of abstract principles, or a mechanism for finding simple patterns (which might also be used in learning things other than language).11 The scientific study of language acquisition aims to characterize the child’s built-in analyzers for language, whatever they turn out to be.
Language itself is not a single system but a contraption with many components. To understand how children learn a language, it’s helpful to focus on one of these components rather than try to explain everything at once. There are components that assemble sounds into words, and words into phrases and sentences. And each of these components must interface with brain systems driving the mouth, the ear, one’s memory for words and concepts, one’s plans for what to say, and the mental resources for updating one’s knowledge as speech comes in.
The component that organizes words into sentences and determines what they mean is called syntax. Syntax itself encompasses several mechanisms, which are tapped to different extents by different languages. They include putting words in the right order, enforcing agreement between elements like the subject and the verb, and keeping track of special words that have their fingers in two places in the sentence at once (such as the what in What do you want?—it serves as both the element being questioned and the thing that is wanted).
One of the key phenomena of syntax is the way that sentences are built around their verbs. The phenomenon goes by many technical names (including subcategorization, diathesis, predicate-argument structure, valence, adicity, arity, case structure, and theta-role assignment), but I’ll refer to it using the traditional term verb constructions.12
Most people already know something about verb constructions in the form of a dim memory of the distinction between intransitive and transitive verbs. Intransitive verbs like snore appear without a direct object, as in Max snored; it sounds odd to say Max snored a racket. Transitive verbs like sprain require a direct object, as in Shirley sprained her ankle; it sounds odd to say Shirley sprained. The transitive and intransitive constructions are the tip of an iceberg. English also has verbs that require an oblique object (an object introduced by a preposition), as in The swallow darted into a cave, verbs that require an object and an oblique object, as in They funneled rum into the jugs, and verbs that require a sentence complement, as in She realized that she would have to get rid of her wolverines. A book by the linguist Beth Levin classifies three thousand English verbs into about eighty-five classes based on the constructions they appear in; its subtitle is A Preliminary Investigation.
A verb, then, is not just a word that refers to an action or state but the chassis of the sentence. It is a framework with receptacles for the other parts—the subject, the object, and various oblique objects and subordinate clauses—to be bolted onto. Then a simple sentence held together by a verb can be inserted into a more inclusive sentence, which can be inserted into a still more inclusive sentence, and so on without limit (as in the old sign “I know that you believe you understand what you think I said, but I am not sure you realize that what you heard is not what I meant”).
The information packed into a verb not only organizes the nucleus of the sentence but goes a long way toward determining its meaning. We see this most clearly in sentences that differ only in their choice of verb, like Barbara caused an injury and Barbara sustained an injury, where Barbara is involved in the event in completely different ways. The same is true for Norm in Norm gave a pashmina and Norm received a pashmina. You can’t figure out what a sentence means by guessing that the subject is the doer and the object is the done-to; you also have to check with the verb. The entry for the verb give in the mental dictionary indicates in some way that its subject is the giver and its object the gift. The entry for receive says in some way that its subject is the recipient and its object the gift. The difference between Harriet appearing to Sam to be brave and appealing to Sam to be brave shows that the different schemes for casting actors into roles can be quite intricate.
A good way to appreciate the role of verb constructions in language is to ponder jokes that hinge on an ambiguity between them: same words, different constructions. An old example is this exchange: “Call me a taxi.” “OK, you’re a taxi.”13 According to a frequently e-mailed list of badly translated hotel signs, a Norwegian cocktail lounge sported the notice “Ladies are requested not to have children in the bar.” In The Silence of the Lambs, Hannibal Lecter (a.k.a. Hannibal the Cannibal) taunts his pursuer by saying, “I do wish we could chat longer, but I’m having an old friend for dinner.” And in his autobiography the comedian Dick Gregory recounts an episode from the 1960s: “Last time I was down South I walked into this restaurant and this white waitress came up to me and said, We don’t serve colored people here. I said, That’s all right. I don’t eat colored people. Bring me a whole fried chicken.”14
The constructions that a verb may appear in depend in part on its meaning. It’s no coincidence that snore is intransitive, snoring being an activity that one accomplishes without anyone’s help, and that kiss is transitive, since a kiss ordinarily requires both a kisser and a kissee. According to a long-standing assumption in linguistics (accepted both in Chomsky’s theory and in some of its rivals, like Charles Fillmore’s Case Grammar), the way that the meaning of a verb affects the constructions it appears in is by specifying a small number of roles that the nouns can play.15 (These roles go by many names, including semantic roles, case roles, semantic relations, thematic relations, and theta roles.) A verb with just an actor (like the snorer in snore) likes to be intransitive, naturally enough, with the actor as the subject. A verb with an agent and an acted-upon entity (like a kisser and a kissee) likes to be transitive, with the agent as the subject and the acted-upon as the object. And verbs that talk about things moving from place to place (like the verb move itself) also take one or more oblique objects, like a from- phrase for the source of the movement and a to- phrase for its goal.
Nonetheless, it has long been known that the fit between the scenario behind a verb and the constructions it may appear in is highly inexact. Ultimately it’s the verb itself, not the underlying concept, that has the final say. For instance, a given concept like “eating” can underlie both a transitive verb, as in devour the pâté (you can’t say Olga devoured), and an intransitive one, as in dine (you can’t say Olga dined the pâté). And in thousands of cases a verb refuses to appear in constructions that would seem to make perfect sense, given the verb’s meaning. Based on meaning alone, one would expect that it would be natural to say Sal rumored that Flo would quit, or The city destroyed, or Boris arranged Maria to come. But while these sentences are perfectly understandable, they sound odd to an English speaker’s ears.
In order for children to acquire an English speaker’s ears, they must somehow learn this whole system: what each verb means, which constructions it naturally appears in, and which roles are played by the various nouns that accompany it in a sentence. This is the rabbit hole that I invite you to explore—one that leads to the world of human ideas and the dramas they engage in.
Before we descend into this world, I owe you an explanation of what it means to claim that “you can’t say this” or “such-and-such is ungrammatical.” These judgments are the most commonly used empirical data in linguistics: a sentence under a certain interpretation and in a certain context is classified as grammatical, ungrammatical, or having various degrees of iffiness. 16 These judgments aren’t meant to accredit a sentence as being correct or incorrect in some objective sense (whatever that would mean), nor are they legislated by some council of immortals like the Académie Française. Designating a sentence as “ungrammatical” simply means that native speakers tend to avoid the sentence, cringe when they hear it, and judge it as sounding odd.
Note too that when a sentence is deemed ungrammatical, it might still be used in certain circumstances. There are special constructions, for example, in which English speakers use transitive verbs intransitively, as when a parent says to a child Justin bites—I don’t want you to bite. There are also circumstances in which we can use intransitive verbs transitively, as when we say Jesus died a long, painful death. And we all stretch the language a bit when we paint ourselves into a syntactic corner or can’t find any other way to say what we mean, as in I would demur that Kepler deserves second place after Newton, or That really threatened the fear of God into the radio people. Calling a sentence ungrammatical means that it sounds odd “all things being equal”—that is, in a neutral context, under its conventional meaning, and with no special circumstances in force.
Some people raise an eyebrow at linguists’ practice of treating their own sentence judgments as objective empirical data. The danger is that a linguist’s pet theory could unconsciously warp his or her judgments. It’s a legitimate worry, but in practice linguistic judgments can go a long way. One of the perquisites of research on basic cognitive processes is that you always have easy access to a specimen of the species you study, namely, yourself. When I was a student in a perception lab I asked my advisor when we would stop generating tones to listen to and start doing the research. He corrected me: listening to the tones was research, as far as he was concerned, since he was confident that if a sequence sounded a certain way to him, it would sound that way to every other normal member of the species. As a sanity check (and to satisfy journal referees) we would eventually pay students to listen to the sounds and press buttons according to what they heard, but the results always ratified what we could hear with our own ears. I’ve followed the same strategy in psycholinguistics, and in dozens of studies I’ve found that the average ratings from volunteers have always lined up with the original subjective judgments of the linguists.17
A PARADOX IN BABY TALK
Put yourself in the booties of a child who is in the midst of figuring out how to speak the language as it is spoken by parents, friends, and siblings. You have learned a few thousand words, and have an inkling (not conscious, of course) of the difference between subjects, verbs, objects, and oblique objects. The verbs keep coming in, and as you learn them you have to figure out how you can use them. Just knowing what a verb means isn’t enough, because, as we saw, verbs with similar meanings can appear in different constructions (like dine and devour, or hinted and rumored); you have to pay attention to which participants accompany the verb in the sentence.
For instance, say you’ve heard load in a sentence for the first time, such as Hal is loading hay into the wagon. Say you have an idea of what the words mean, and from watching what’s going on, you can see that Hal is pitching hay into a wagon. A safe bet is to file away the information that load can appear in a sentence with a subject, which expresses the loader (Hal); an object, which expresses the contents being moved (the hay); and an object of into, which expresses the container (the wagon). You can now say or understand new examples with the same verb in the same construction, like May loaded some compost into the wheelbarrow. (Linguists call this the content-locative construction, because the contents being moved are focused upon in the object of the sentence.) But that’s as far as you go—you don’t venture into saying May loaded (meaning she loaded something into something else), or May loaded into the wheelbarrow.
So far so good. In a little while you hear load in a new construction, like Hal loaded the wagon with hay. Once again hay is being pitched into the wagon, and as far as you can see, the sentence has the same meaning as the familiar sentence Hal loaded hay into the wagon. You can add an addendum in your mental dictionary to the entry for load: the verb can also appear in a construction with a subject (the loader), an object (the container, such as a wagon), and an object of with (the contents, such as the hay). Linguists call this the container-locative construction, because now it’s the container that’s being focused upon.
As you continue to hoover up verbs over the months and years, you encounter other verbs that behave like load: they appear in two synonymous constructions but differ in whether it is the content or the container that shows up as the direct object:
Jared sprayed water on the roses.
Jared sprayed the roses with water.
Betsy splashed paint onto the wall.
Betsy splashed the wall with paint.
Jeremy rubbed oil into the wood.
Jeremy rubbed the wood with oil.
This is starting to look like a pattern (what linguists call an alternation), and now you face a critical choice. Do you keep accumulating these pairs of verbs, filing them away pair by pair? Or do you make a leap of faith and assume that any verb that appears in one of these constructions can appear in the other one? That generalization could be put to work by coining a rule that more or less says, “If a verb can appear in a content-locative construction, then it can also appear in a container-locative construction, and vice versa.” With this rule (which we can call the locative rule) in hand, you could hear someone say brush paint onto the fence and then surmise that brush the fence with paint is fine, without having actually heard it. Likewise, if you hear Babs stuffed the turkey with breadcrumbs, you can assume that Babs stuffed breadcrumbs into the turkey is also OK.
It’s a small step toward mastering the language, but a step in the right direction. English is crawling with families of constructions that admit verbs interchangeably, and if children can dig out the patterns and extend them to new verbs, they can multiply their learning speed by the average number of constructions per verb. This could be an important path to becoming a fluent and open-ended speaker of the language, as opposed to one who simply regurgitates a small number of formulas.
There is only one problem. When the locative rule is applied willy-nilly, it cranks out many errors. For example, if you apply it to Amy poured water into the glass, you get Amy poured the glass with water, which English speakers reject (as I’ve verified in questionnaires).18 You can also get into trouble when you apply it in the other direction, to verbs like fill: though the input, Bobby filled the glass with water, is fine, the output, Bobby filled water into the glass, is not (again, a survey bears this out).19 And these aren’t isolated exceptions. Many other verbs resist being fed into the maw of the locative rule. Here are four other unhappy campers, two of them verbs that like only the content-locative, and two that like only the container-locative. (Following the usual convention in linguistics, I’ve put an asterisk next to the sentences that sound odd to native speakers.)
Tex nailed posters onto the board.
*Tex nailed the board with posters.
Serena coiled a rope around the pole.
*Serena coiled the pole with a rope.
Ellie covered the bed with an afghan.
*Ellie covered an afghan onto the bed.
Jimmy drenched his jacket with beer.
*Jimmy drenched beer into his jacket.
That’s funny. . . . Why should the second sentence in each pair sound so odd? It’s not that the iffy sentences are unintelligible. No one could be in doubt as to the meaning of Amy poured the glass with water or Jimmy drenched beer into his jacket. But language is not just whatever set of ways people can think of to get a message across. Children, in the long run, end up with a fastidious protocol that sometimes rules out perfectly good ways of communicating. But why? How do children succeed in acquiring an infinite language when the rules they are tempted to postulate just get them into trouble by generating constructions that other speakers choke on? How do they figure out that certain stubborn verbs can’t appear in perfectly good constructions?
An equivalent puzzle arises if you invert the way you think about the problem and make the child the master and the language the slave. How did the English language come down to us with all those exceptional verbs, given that they should have been whipped into conformity by the first generation of children faced with learning them?
There are three ways out of this paradox, but none of them is palatable. The first is that we (and the hypothetical child we have been imagining) have framed the rule too broadly. Maybe the real locative rule is restricted to a subset of verbs sharing an overlooked trait, and children somehow figure out the restriction and append it as a codicil to the rule. But if there is such a trait, it’s far from obvious, because the verbs that submit to the rule and the ones that resist it are quite close in meaning. For example, pour, fill, and load are all ways of moving something somewhere, and they all have the same cast of characters: a mover, some contents that move, and a container that is the goal of the movement. Yet pour allows only the content-locative (pour water), fill allows only the container-locative (fill the glass), and load goes both ways (load the hay, load the wagon).
The second option is that children don’t coin these rules at all. Maybe they really do file away in memory just those combinations of verbs and constructions they have heard in the speech of their elders, and conservatively stick to just those combinations. Under this theory, they would be like Marvin in the eponymous comic strip on the following page.
Well, that would certainly solve the problem. Children would never be tempted to say pour the cup with juice or cover an afghan onto the bed because they would never have heard anyone else say things like that. Verbs would keep their privileges in perpetuity, because children would learn the constructions on a verb-by-verb basis, just as they learn the words themselves, each a unique combination of a sound and a meaning.
Some linguists have taken this hypothesis seriously, but it doesn’t seem to be right.20 For one thing, it would be surprising if children were that conservative, given that they have an infinite language to master and only a finite sample of speech to go on. For another, the English language seems to expand rapidly to accommodate new verbs in new constructions, suggesting that at least by the time they reach adulthood, speakers are not conservative verb-memorizers. Most Americans, on hearing the Britishism He hoovered ashes from the carpet (a content-locative), readily generalize to He hoovered the carpet (a container-locative). Likewise, when the container-locatives burn a CD (put songs onto it) and rip a CD (copy songs off of it) came into common parlance, the content-locatives burn songs onto the CD and rip songs from the CD followed closely on their heels (or perhaps the other way around).21
Do only adults make these leaps, or can they be seen in childhood, as children are learning the language? The psychologist Melissa Bowerman, like many psycholinguists, kept meticulous diaries of her children’s speech when they were small, recording and analyzing every anomaly. She showed that children really do use verbs in constructions that they could not simply have recorded from the mouths of their parents.22 Here are three examples of creative content-locatives, and three of creative container-locatives:
Can I fill some salt into the bear?
I’m going to cover a screen over me.
Feel your hand to that.
Look, Mom, I’m gonna pour it with water, my belly.
I don’t want it because I spilled it of orange juice.
I hitted this into my neck.
To ensure that these weren’t rare errors from unusual children, the psychologist Jess Gropen and I corroborated the finding in two ways. First, we sifted through online corpora of children’s speech, where we found similar errors.23 Second, we used a method for assessing generalizations called the wug test, after a classic study by the psychologist Jean Berko Gleason.24 Gleason showed children a cartoon of a little bird and said, “Here is a wug. Now there are two of them. There are two . . .”—at which point four-year-olds happily filled in the blank with wugs, a form they could not have memorized from adults. In our case we told children that mooping meant to move a sponge to a purple cloth, turning it green. Sure enough, the kids said we were mooping the cloth—a container-locative that they had never heard anyone use before.25 So much for Moderate Marvin.
There’s a third way out. Maybe children do make errors but are corrected by their parents and are thereby chastened into avoiding the offending verb in that construction forever after. This, too, is unlikely. Notwithstanding the widespread belief among psychologists that parents are responsible for everything that develops in their children, attempts to show that parents correct their children’s deviant sentences, or even react differently to them, have turned up little.26 Parents are far more concerned with the meaning of children’s speech than its form, and when they do try to correct the children, the children pay little heed. The following exchange is typical:
CHILD: I turned the raining off.
FATHER: You mean you turned the sprinkler off?
CHILD: I turned the raining off of the sprinkler.
And even if parents did occasionally raise an eyebrow at their children’s odd usages and the children did take heed, the effect would fall short of what we need to solve the problem. Many of the obstreperous verbs are rare, yet people have strong intuitions about what the verbs can and can’t do. People sense that they would never say They festooned ribbons onto the stage or She siphoned the bottle with gasoline, yet word-frequency counts show that these verbs are literally one in a million.27 It is unlikely that every English speaker uttered each of the obdurate verbs in each of the offending constructions at some point in childhood (or, for that matter, adulthood), was corrected, and now finds the usage strange on account of that episode.
We have a paradox.28 From the time they are children, people generalize. They avoid generalizing to certain words (at least as adults). It’s not because they were corrected for each overgeneralization. And there is no systematic difference between the words that allow themselves to be generalized and those that don’t. These four statements can’t all be true.
Why should anyone care about what seems like a small problem in a tiny corner of psycholinguistics? The reason is that the learnability of the locative construction is typical of many paradoxes in explaining language, where partial patterns, too seductive to ignore but too dangerous to apply, are ubiquitous. In Crazy English, the language maven Richard Lederer calls some of them to our attention:
If adults commit adultery, do infants commit infantry? If olive oil is made from olives, what do they make baby oil from? If a vegetarian eats vegetables, what does a humanitarian consume? A writer is someone who writes, and a stinger is something that stings. But fingers don’t fing, grocers don’t groce, hammers don’t ham, humdingers don’t humding, ushers don’t ush, and haberdashers do not haberdash. . . .
. . . If the plural of tooth is teeth, shouldn’t the plural of booth be beeth? One goose, two geese—so one moose, two meese? If people ring a bell today and rang a bell yesterday, why don’t we say that they flang a ball? If they wrote a letter, perhaps they also bote their tongue.29
Each of these oddities defines a scientific problem for linguistics and psychology. The one in the second paragraph—irregular plurals and past-tense forms—is tricky enough that I have written a book and many papers trying to make sense of it.30 Unfortunately, my favorite solution to that puzzle is of no help here. Irregular forms like teeth and rang are what linguists call positive exceptions: they exist, even though the usual rule, like “Add -ed to form the past tense,” fails to generate them. Children can learn them upon hearing them, one at a time. We also have a good idea as to how children use positive exceptions to preempt or block the rule-governed forms, so that they don’t say flang or meese. Conjugations and declensions are neatly organized into paradigms in which each verb ordinarily has a single past-tense form and each noun a single plural form. When the child hears Boggs flung the ball or Vern shot two moose, those irregular forms stake out their cells in a mental matrix and fend off the rival forms flang and meese (together with flinged and mooses).31
But verb-construction mismatches are negative exceptions: they fail to exist despite the fact that a rule does generate them. Children have no direct evidence from parents’ speech that these forms are ungrammatical. Not hearing them—the proverbial dog that didn’t bark—isn’t itself evidence, because there are an infinite number of perfectly grammatical forms that they also don’t hear, and they can’t very well exclude all of them or they would be confined to parrothood. Nor can they use some competing form to preempt them (in the way that flung preempts flinged and flang), because verb constructions, unlike conjugations, aren’t arranged into neat cubbyholes. The reason you can’t say pin a board with posters or coil the pole with a rope is not that each of these is repelled by some synonym guarding the grammatical turf in the way that flung repels flang and flinged. There is no verb that allows one to talk about covering a board by pinning posters onto it in the container-locative construction.
The verb-learnability paradox has attracted attention for another reason. About halfway down our plunge from a wide view of the human mind to the acquisition of locative constructions, I said that language acquisition is an example of the problem of induction—making valid generalizations about the future from limited data available in the present, whether they involve language acquisition by a child, learning by a computer, or theorizing by the scientist. The pickle we find ourselves in is common to induction of all kinds: how to back off from an overly general hypothesis in the absence of negative data.32 If you frame a conclusion too broadly, and don’t have complete corrective feedback from the world (say, you grow up thinking all swans are white, and never get to New Zealand, where you’d see black swans), you are in danger of never finding out that you are wrong. In this case, a hypothetical child is tempted to generalize that all verbs about moving something somewhere can be expressed in either of two English constructions. Yet somehow children grow into adults who generalize beyond the verbs they have heard while uncannily holding back from some of the verbs they haven’t heard. The locative construction (along with similar constructions) presents us with a paradox of a child seeming to learn the unlearnable, and thus became a focus of attention among linguists and computer scientists interested in the logic of learning in general.
Nature does not go out of its way to befuddle us. If some phenomenon seems to make no sense no matter how we look at it, we are probably overlooking some deeper principle about how things work. This is exactly what happened in the paradox of learning locative verbs, and the missing principles are about the kinds of ideas that populate the human mind.
FLIPPING THE FRAME
Of the four apparent facts that can’t all be true at the same time—people generalize; they avoid some exceptions; the exceptions are unpredictable; and children don’t get corrected for every mistake—the most assailable is the one about the unpredictability of the exceptions, the one that says there is no way to distinguish the verbs that take part in an alternation from those that sit it out. Maybe we just haven’t looked hard enough. Often a linguistic pattern that at first seems haphazard turns out to have a stipulation that divides the sheep from the goats. For example, the mystery of why you can’t apply -er and -est to certain adjectives, as in specialer and beautifullest, was solved when someone noticed that the suffixes apply only to words that are monosyllabic (redder, nicer, older) or have at most an insubstantial second syllable (prettier, simpler, narrower). Perhaps there is also a subtle criterion that distinguishes the verbs enlisted into the locative construction from the draft dodgers—what the linguist Benjamin Lee Whorf called a cryptotype.33 If children’s rules became sensitive to that criterion, the paradox would vanish. The hidden criterion is unlikely to involve the sounds of the verbs, since in that regard they are pretty much alike; it is more likely to involve their meanings.
The breakthrough, in my mind, came in a paper by a pair of linguists, Malka Rappaport Hovav and Beth Levin, who at the time were working down the hall from me at MIT.34 Under the influence of Chomsky, linguists had tended to think of rules as operations that cut and pasted phrases, such as moving a prepositional object leftward into the position of the direct object, or moving the direct object rightward into a prepositional phrase.35 It was this mindset that made it seem so odd that the locative rule would care about the content of the verb, just as it would seem odd if your wordprocessing program announced that it refused to cut and paste words with some meanings while obediently doing so with others. But what if the rule transformed not the arrangement of the phrases in a construction, but something much more abstract, namely, the framing of events that goes into its meaning?
Imagine that the meaning of the content-locative construction is “A causes B to go to C,” but the meaning of the container-locative construction is “A causes C to change state (by means of causing B to go to C).” In other words, loading hay onto the wagon is something you do to hay (namely, cause it to go to the wagon), whereas loading the wagon with hay is something you do to the wagon (namely, cause it to become loaded with hay). These are two different construals of the same event, a bit like the gestalt shift in the classic face-vase illusion in which the figure and ground switch places in one’s consciousness:
In the sentences with the hay and the wagon, the flip between figure and ground is not in the mind’s eye but in the mind itself—the interpretation of what the event is really about.
Now, at first glance the difference between causing a thing to go to a place and causing a place to change by moving a thing to it may seem as rabbinical as the difference between whether the destruction of the World Trade Center consisted of one event or two events—a question of “mere semantics.” But as with the gigabucks at stake in the aftermath of 9/11, mere semantics can matter.
For one thing, this new understanding of the phenomenon is simpler and more elegant—not always a sign that a theory is true, but not something to be ignored either. When reconceived as a conceptual gestalt shift, the locative rule is no longer a matter of cutting and pasting phrases in complicated ways for no particular reason. It can now be factored into two very general and useful rules:
• A rule of semantic reconstrual (the gestalt shift): If a verb means “A causes B to move to C,” it can also mean “A causes C to change state by moving B to it.”
• A rule for linking meaning to form: Express the affected entity as the direct object.
In the content-locative (load hay onto the wagon), we have hay as the direct object, because the event is construed as something being done to the hay. In the container-locative (load the wagon with hay), we have the wagon as the direct object, because the event is now construed as something being done to the wagon. Other linking rules take care of how the other participants are expressed. One rule links the causal agent (the guy pitching the hay) to the subject. The other links miscellaneous participants to oblique objects, each getting a preposition suitable to its meaning. The preposition into means “to in,” namely, “to an interior portion of ”; onto stands for “to on”; with stands for “a means of changing something.”
Though we have replaced one rule with several, the picture as a whole is simpler, because, as we shall see, these rules get reused in different combinations all over the language. And satisfyingly, we can explain why the locative rule does what it does. The wagon participant has to switch from oblique object to direct object (as opposed to being pasted into any old position), because that participant has been reconstrued as “the affected entity,” and affected entities, whether they are things that change location or things that change state, are expressed in syntax as direct objects.
I promised you that the reason we are obsessing over locative constructions in a book about human nature is that it tells us things about the way humans think. One of these things was broached in the first chapter: the mind has the power to frame a single situation in very different ways. Here we see that this power is so pervasive that it isn’t just recruited in contentious squabbles like invading Iraq versus liberating Iraq or manipulating a ball of cells versus killing a young person, where no one is surprised that different perspectives are possible. Rather, it pervades the way we construe even the simplest, most concrete, and most innocuous events of everyday life, like putting hay in a wagon or breadcrumbs in a turkey.
Scrutinizing the locative construction not only shows that construing and reconstruing is a basic power of cognition, it exposes the elements that make up each construal, and some of their quirks. The gestalt-shift theory implies that the two locative constructions, contrary to first impressions, are not completely synonymous. There must be situations in which one construction truthfully applies and the other does not. That is indeed the case.
When one loads hay onto a wagon, it can be any amount, even a couple of pitchforkfuls. But when one loads the wagon with hay, the implication is that the wagon is full.36 This subtle difference, which linguists call the holism effect, can be seen with the other locative verbs: to spray the roses with water implies that they all got sprayed (as opposed to merely spraying water onto the roses), and to stuff the turkey with breadcrumbs implies that it is completely stuffed.
The holism effect is not an arbitrary stipulation tacked onto the rule, like a pork-barrel amendment on a spending bill. It falls out of the nature of what the rule does, namely, construe the container as the thing that is affected. And that, in turn, reveals an interesting feature of the way the mind conceives what things are and how they change. The holism effect turns out not to be restricted to the locative construction; it applies to direct objects in general. For instance, the sentence Moondog drank from the glass of beer (where the glass is an oblique object of from) is consistent with his taking just a couple of sips. But the sentence Moondog drank the glass of beer (where the glass is a direct object) implies that he chugged down the whole thing. Similarly, you might say He climbed up the mountain even if he thought better of it partway and came back down, but if you say He climbed the mountain you are suggesting that he reached the summit. Or think about the difference between the sentences in each of these pairs, which at first glance look synonymous:
Peter painted on the door.
Peter painted the door.
Betty put butter on the bun.
Betty buttered the bun.
Polly removed peel from the apple.
Polly peeled the apple.
In each pair, the second sentence, which expresses the affected entity as the direct object, implies that something was done to the whole thing, not just part of it: the door was entirely painted, the bun thoroughly buttered, the apple completely skinned.
But the holism effect is even more sweeping than that. It’s not so much a property of the direct object (which is just a position in a sentence) as it is a property of the concept that tends to be expressed as a direct object, namely, the entity being affected. In the examples we have been looking at, the affected entity happens to get expressed as the direct object because when a sentence contains a causal agent, the agent generally gets first dibs on the subject slot. But when the agent is not mentioned, the affected entity can be the subject, as in The ball rolled or The butter melted. And crucially, when the subject does accommodate an affected entity, it is interpreted holistically, just like direct objects. This is best seen in lovely pairs like these:
Bees are swarming in the garden.
The garden is swarming with bees.
Juice dripped from the peach.
The peach was dripping with juice.
Ants crawled over the gingerbread.
The gingerbread was crawling with ants.
The second sentence in each pair presents a sensuous image of an entity so saturated with stuff or bits that the mind blurs the two and apprehends the entire entity as doing what the stuff or bits ordinarily do: the garden swarms, the peach drips, the gingerbread crawls.
But why is the content interpreted as a whole in these constructions? The reason is that the English language treats a changing entity (a loaded wagon, sprayed roses, a painted door) in the same way that it treats a moving entity (pitched hay, sprayed water, slopped paint). A state is conceived as a location in a space of possible states, and change is equated with moving from one location to another in that state-space. In this way, locative constructions illustrate a second discovery in the hidden world down the rabbit hole, the ubiquity of metaphor in everyday language. The linguist Ray Jackendoff has explored the way in which many of the words and constructions used for motion, location, or obstruction of motion in physical space are also used for a kind of metaphorical motion, location, or obstruction of motion in state-space:37
Pedro went from first base to second base.
Pedro went from sick to well.
Pedro was at second base.
Pedro was sick.
The manager kept Pedro at first base.
The doctor kept Pedro well.
In the first sentence, Pedro’s body actually moved in space, but in the second, he could have been in bed the whole time; only his health moved, metaphorically speaking. Concepts of space seem to infect other concepts as well, as we saw in the first chapter when noting the way that people count and measure out events as if they were objects made of time-stuff. People also use space as a model for an abstract continuum when they speak of the rising or falling of their paycheck, their weight, or their spirits,38 or when they plot data points, representing anything whatsoever, on graph paper.39 Whether the ubiquity of metaphor is a revolutionary discovery about the mind, a banal fact about the history of a language, or something in between is the topic of a later chapter. My intent here is to show how the psychology of space sheds light on the holism effect, and tells us something about the psychology of concepts in general.
When the mind conceptualizes an entity in a location or in motion, it tends to ignore the internal geometry of the object and treat it as a dimensionless point or a featureless blob. The linguist Len Talmy notes that a typical preposition or other spatial term specifies a relationship between a figure and a place that is defined by some reference object.40 Usually the reference object is larger and more prominent, and the figure moves or is located relative to it. (The exception that proves the rule is Beatrice Lillie’s quip about the Queen Mary when she first saw it, “How long does it take for this place to get to London?”) And usually the reference object is specified in more geometric detail. It is conceptualized as having a certain number of dimensions along which it is stretched out: one dimension, like a stick or a string; two dimensions, like a sheet of paper or plywood; or three dimensions, like a couch or a watermelon. And it is conceptualized as having certain axes, parts, cavities, and boundaries that align with those dimensions.
So the figure being positioned and the place where it is said to be located are treated differently in language: the first is reduced to a dimensionless speck, whose internal geometry is ignored; the second is diagrammed, at least schematically. Take the English phrases on your hand, under your hand, and in your hand. Each picks out an aspect of the geometry of the hand, namely, its top, its bottom, and a cavity it can form. The choice of preposition depends upon that geometry: a marble can be in one’s hand if the hand is cupped and the marble is on the palm side, but the marble is on it if the hand is straightened or if the marble is on the back. The marble can’t be in one’s forearm or shin or trunk at all, since these are conceptualized as one-dimensional dowels. Compare this blueprinting to the stylized treatment of the figure being located. In these examples I have cited a marble, but the figure can have any shape or configuration whatsoever: it can be a marble, a matchstick, a matchbook, or a moth, and these can be upright, sideways, or upside down, and it can still be in or on or under your hand. To be sure, not all prepositions treat the figure as a blob or point: along and across, for example, require the figure to be elongated. But the most common ones are myopic about the figure being positioned.
This leads us to a deeper explanation of the holism effect. In the locative alternation, when the container (such as the wagon in load hay into the wagon) gets promoted to direct object, it is also conceptually reanalyzed as something that has been moved in state-space (from the “empty” slot to the “full” slot). And in this reconstrual, it gets compacted into a single point, its internal geometry obliterated. Wagons become loaded, flowerbeds sprayed, turkeys stuffed, not as arrangements of matter in space with niches and hidey-holes that may separately accommodate bits of matter, but as entities that are, taken as a whole, now ready for carting, blooming, or cooking. Indeed, the holism effect is a bit of a misnomer. We’re really talking about a state-change effect, and ordinarily the most natural way that an object changes state when something is added to it is when the stuff fills the entire cavity or surface designated to receive it. But if an object can be thought of as changing state even when it has stuff in just one part, then the container-locative may be used there, too. Thus we can say that a graffiti artist has sprayed a statue with paint even if he has colored just one part of it, because a single splotch is enough for people to consider it defaced.
and post it to your social network
Most Helpful Customer Reviews
See all customer reviews >