Um-- Slips, Stumbles, and Verbal Blunders, and What They Mean

by Erard, Michael.

  Contents

  Title Page

  Dedication

  Epigraph

  A Note to the Reader

  Introduction

  1 The Secrets of Reverend Spooner

  2 The Life and Times of the Freudian Slip

  3 Some Facts About Verbal Blunders

  4 What We, uh, Talk About When We Talk About “uh”

  5 A Brief History of “Um”

  6 Well Spoken

  7 The Birth of Bloopers

  8 Slips in the Limelight

  9 Fun with Slips

  10 President Blunder

  11 The Future of Verbal Blunders

  Appendix A: Recommended Reading

  Appendix B: A Field Guide to Verbal Blunders

  Appendix C: Slips Versus Disfluencies

  Acknowledgments

  About the Author

  Copyright

  for my parents,

  who gave me the slip of gifts

  I want some hot poppered butt corn—I mean cot buttered bop corn—that is—corn buttered pop butt, or rather cuttered pot born, I mean—oh well, gimme some peanuts.

  Thesaurus of Humor, 1940

  If you are a young man, for instance, will it not be from small pointers that you will conclude that you have won a girl’s favor? Would you wait for an express declaration of love or a passionate embrace? Or would not a glance, scarcely noticed by other people, be enough? A slight movement, the lengthening by a second of the pressure of a hand?…

  So do not let us underestimate small indications; by their help we may succeed in getting on the track of something bigger.

  Sigmund Freud

  A Note to the Reader

  As you read this book, you may find that you begin to notice how people actually speak. That reaction is perfectly normal. You may also become more aware of your own speaking. This, too, is normal.

  Your heightened awareness may also make you want to point out to your friends, loved ones, colleagues, and even random talkers their every verbal blunder. This is not recommended. For the sake of harmony, point out only the most interesting ones.

  What is an interesting verbal blunder? You’d be surprised.

  Introduction

  Transcribers at the Federal News Service in Washington, D.C., encounter bucketloads of verbal blunders every weekday morning. Their small office contains six desks with computers and dictaphones and a shelf filled with almanacs and dictionaries. On a wall hangs a whiteboard scrawled with the unfamiliar words, names, and acronyms of this news cycle. Today it’s “Tamiflu,” “LIHEAP,” and “III-MEF.” Next to a government spokesman’s name, someone’s scribbled a heart—he sounds like he’s good-looking, one of the transcribers teases. Across the hall in another small room, dozens of Marantz tape recorders on shelves are prepared to catch audio feeds from around the city and the country. At ten o’clock, the world of government and business begins to talk in earnest, and the words from White House briefings, federal agency press conferences, congressional hearings, and corporate speeches begin to flow to the recorders, now alive and humming, or through audio streamed through the Web. Once a cassette contains about four minutes of speaking, the news director carries it to the transcribers’ office and lines it up on a low cabinet with other tapes. A transcriber will bring the tape to his or her cubicle, put it in the transcribing machine, and start typing the segment. There’s no talking in the room, only the sounds of fingers on keyboards, feet on foot pedals, and tapes rewinding. For eight hours a day the transcribers turn speech into text, spinning these ordinary words, Rapunzel-like, into a commodity posted on Web sites and sold to hundreds of clients.

  The transcripts have to be readable, so the transcribers generally clean up people’s speaking, as instructed by an in-house style guide that cautions, “Don’t type ‘um,’ ‘ah,’ ‘er,’ or partial words.” The style guide also stipulates that transcribers should “clean up a false start or starts consisting of only one or two words if the omission of the words does not affect the meaning.”*1

  There is one exception to those rules: “DO NOT clean up major policymakers, including the president,” the style guide says, “since not only what they say but how they say it often makes the news.” When George H. W. Bush was president he gave a speech commemorating Pearl Harbor Day but mistakenly gave the date as September 7, so the Federal News Service (FNS) transcribers kept September 7. (The White House corrected the error to December 7.) Alice Tate, a cherub-faced woman with red hair who is the supervisor of the news transcription team and has been an FNS employee for twenty years, mentioned that President George W. Bush had recently praised the efforts of a national counterism director.

  Tate left “counterism” intact. “Sometimes he kind of stumbles,” she says of Bush, “and we have to make a decision—since we don’t type partial words—is a word a combination of words he’s messed up?” Bush isn’t the only contributor to a document circulated in the office, the “duh files,” which contains many of the gaffes, flubs, solecisms, and bloopers that fall into transcribers’ laps. There was Howard Dean trying to say “incentivize,” which came out like this: “What that does is incense providers to do more than just park kids in front of the television.” There was an Atlanta reporter asking Secretary of Defense Donald Rumsfeld, “Is our current situation such that the harder we work, the behinder we get?”

  More common verbal blunders don’t make the duh files—they’re excised from transcripts, scrubbed out for the sake of readability. Another veteran transcriber, Cate Hagman, says that she’s stopped hearing “uh” and “um” in people’s speech. Because FNS cuts them from transcripts, she automatically wipes them from what she hears, too. Other verbal tics still irritate her, though. “Transcribing makes you very sensitive to verbal tics. ‘Like,’ ‘you know,’ ‘I mean.’ Our parents were right—we should drop them from our speech, but you’d be surprised to see how many responsible people put them in.” Surely the politicians say them unconsciously—just as the ordinary listener unwittingly lets them pass by.

  On the other side of the country, a team of robotics researchers at SRI International, a research institute in Menlo Park, California, paid attention to the fillers and interruptions of ordinary speaking with another purpose in mind. In the early 1990s, they developed a robot named Flakey, a three-foot-tall black box designed to roam around the SRI offices, like a large, mobile, single-minded filing cabinet that could move out of the way of walking humans and navigate obstacles. It could also understand spoken commands to retrieve objects from various places in the institute. The problem was, however, that if you ordered Flakey to go on a mission, you didn’t know for several seconds if the robot had comprehended what you said. It would sit there, humming. So you might repeat yourself, giving Flakey two commands to process and perhaps confuse, making it stall.

  This didn’t make for the smoothest human-machine relationships. Liz Shriberg, a speech researcher who was studying how “uh” and “um” function in human communication, suggested that her colleagues program Flakey to say “uh.” One could say, “Flakey, go to so and so’s office and get me the folder,” and the robot would say “uh,” in order to signal that it had heard the command and was processing it. The pause filler was an effective signal to the human that the robot wasn’t, in fact, broken down, and in so doing, eased the interaction between human and machine. (Even more effective would have been an “uh” with the proper intonation, Shriberg said. Otherwise, Flakey’s monotone “uh” sounded like an oddly long beep.)

  As nearly as anyone can estimate, humans have been using language
more or less in the form that we know it for about one hundred thousand years. They’ve probably been making verbal blunders for about as long. When you speak, you hesitate, you. Stop. You also pereat, repeat—you repair, you fix, you repair your sentences, you utter the wrong words and confuse them in the most headbutted ways, you say “uh,” you linger on “um.” But this is normal. Our ordinary speech is notoriously fragmented, and all sorts of verbal blunders swim through our sentences like bubbles in champagne. They occur on average once every ten words, by some accounts. So if people say an average of 15,000 words each day, that’s about 1,500 verbal blunders a day. Next time you say something, listen to yourself carefully. You st-st-stutter; you forget the words, you swotch the sounds (and when you type, you rverse the lttres—and prhps omt thm, too). The bulk of these go unnoticed or brushed aside, but they’re all fascinating, as much as for why they’re ignored as why they’re noticed.

  Some of these verbal blunders are “slips of the tongue” (as when Howard Dean said “incense” or George Bush said “counterism”); far more frequent are the unsmooth moments that the transcribers at the Federal News Service clean up. Called “speech disfluencies,” these include fillers like “uh” and “um,” repeated words, repeated sounds, or repaired (and restarted) sentences. If I’m speaking off the top of my head then I might, uh, alter the the syntax in the middle of my sentence, and that’s not all, because interruptions—look, over there!—and rep rep repetitions and changes in vocal INTENSITY THEY HAPPEN A LOT along with streeeetched vowels and ratesofspeechspeedingup and. Getting. Slow. Er.

  As unavoidable as they are ineradicable, verbal blunders are rich with meaning. Until recently, they’ve been treated as meaningless filler, played for laughs, or used to reinforce stereotypes of people who were already disliked or mistrusted. Which is to say, we’ve missed other ways of listening to them that could tell us more about each other, about our era, and about our individual selves.

  Profuse and varied, verbal blunders exist in all the world’s languages, spoken and signed. They permeate our daily speaking, even the speech of the most golden-tongued. Omnipresent, they have unique tales to tell about the inner workings of language and mind.

  A verbal blunder is also an indelible mark of humanness. Difficult to predict, verbal blunders are one human behavior that cannot be simulated by computers. Flakey’s “uh” had to be programmed into him. Other animals that have sophisticated means to communicate, too, often blurt and stumble. (Zebra finches, for example, are used in stuttering research because some birds repeat parts of their songs.) But verbal blunders are native only to humans; they separate us from animals, because we’re the only ones with language. And to seal the connection, each individual human being blunders uniquely, in as many styles as there are of speaking, a fingerprint of the self that we leave floating in the air.

  A verbal blunder can be either a momentary loss of control over speaking, called a “slip of the tongue,” or one of the blurps and interruptions in what we think should be smoothly flowing talk, called a “speech disfluency.”

  A slip of the tongue is an inadvertent accident, as the time that Democratic presidential candidate Senator John Kerry, tired from campaigning, said “wasabi” instead of “Wahhabi.” Kerry hadn’t mistaken a Japanese spice for the fundamentalist Islamic sect. He’d planned to say “Wahhabi,” had said it many times before. But his plan went awry—when he reached to retrieve “Wahhabi” from memory, “wasabi” jumped out in front of his brain. (That both are non-English words probably helped, too). Such moments, which are also known as “speech errors” or often just “slips,” appear when the mental machinery that turns ideas into spoken (or signed) words crashes into itself. Replacing “Wahhabi” with “wasabi” was a slip that involved a whole word. Similar accidents can also occur with individual sounds (“black bloxes”), parts of words (“groupment,” “they sit there motionly”), phrases (“I have to smoke my coffee with a cigarette”), and even intonation (“on your look OUT—on your LOOK out—I mean—on your OUT look”). In these slips, pieces of language are swapped (“heft lemisphere,” “fuwtmeeving”), dropped (“tubbled”), blended (“momentaneous,” “slickery”), rushed (“telefathic fox”) or slowed (“a phonological fool”).*2

  Human behavior teems with what Bernard Baars, a cognitive scientist, calls “momentary control problems.” A concert pianist may have rehearsed for months but still plunks the wrong notes on opening night. A skilled baker puts in a tablespoon, not a teaspoon, of salt. The psychology of error is full of insights into the causes of air disasters, medical mistakes, friendly-fire casualties in warfare, automobile accidents, even walking into the other room and forgetting why you’re there. Because they represent the unpredictable human element in increasingly complex technical systems, errors have led to antilock brakes and robot surgeons and other designs that reduce the costs of human error. By contrast, humans already perceive slips of the tongue quite flexibly, and their costs are low—despite, of course, the brief mortification you feel when you ask the hostess for her delicious gahspetti recipe.†3

  Meanwhile, a speech disfluency is something that interrupts the continuity of speaking. Such smoothness is an ideal. Like glossy locks of hair or smooth, clear skin, this ideal is elusive, even unattainable, by most people. Listeners generally prefer when one word is followed promptly by another, and another. Speakers, too, generally intend to get their words in order in time. Yet they still say “uh” and “um.” They repeat words. They forget words and repeat sounds and syllables. With all of these they’ve disrupted the flow of speaking. Elongating a vowel, pausing for too long, even speaking arhythmically can be considered interruptions. But the dimensions of human communication are so wide and subtle, such wild moments don’t, in fact, impede communication when they crop up. At times they may even aid it. After all, successful communication doesn’t solely depend on the unimpeded downloading of information into your head. It’s not the information in the joke that makes you laugh, just as it’s not the information in your lover’s final phone call that makes you cry.

  Speech disfluencies are a hallmark of spontaneous, unscripted speaking and conversation, as in this snippet from an interview: “Let’s stop this for a minute and I’ll, um, ask you some questions and then we can, uh, w-well, we can talk a little bit about your participation and, kinda roll in th—in these group meetings and then maybe we could just…look at it a little bit from that point of view.” In truth, you can hear disfluent moments almost anywhere: newscasts, political debates, presentations at work, radio talk shows, town meetings, speeches, prayers, sermons, arguments. They happen because talking is an activity that takes place in time, and disfluencies are signs of the inevitable friction between thinking and speaking. It’s almost as if you can either think or speak—but rarely manage both simultaneously in a timely way. (It’s the rare person who can, reliably and consistently.)

  Laughing, crying, shouting, sighing, panting, yawning, coughing, throat clearing, spitting, belching, hiccuping, or sneezing aren’t verbal blunders. The slurred words of the drunk or the person who’s recently woken up aren’t either. Neither are disorders of speaking and language: aphasia, vocal cord tremors, or chronic stuttering. It’s worthwhile to read about how people experience these disorders, particularly stutterers, to put one’s own occasional stumbles into perspective. Blundering isn’t saying yes (“uh-huh”) or no (“uh-uh”) or keep going (“mmhmm”). And while there’s much to be learned from slang (“ain’t”), nonstandard dialects (like Ebonics), and regionalisms (“I’m fixin’ to go to the store”), they’re social phenomena, not verbal blunders. Verbal blunders arise from organic flashes in the brain. Simply put, verbal blundering is something that normal speakers speaking normally do. For that reason alone, verbal blunders, so frequent yet so misunderstood, deserve a closer listen.

  This book is a work of applied blunderology, and it began with my listening to speech in public and private. During the 2000 presidential campaign,
I became fascinated with how George W. Bush’s speech was portrayed, even fetishized, by the media and other observers, a fad that lasted into the early months of his presidency. Whether you sympathized with Bush’s politics or not, you had to recognize that the critics of his verbal style took gleeful liberties in diagnosing what his speech patterns meant. They were also as inaccurate about how people normally speak as they were unreflective about the standards that govern speech. Why do we care how Bush talks? Did (and do) we care more with this president than any other? If so, why? These questions weren’t discussed. Of course, Americans expect their political leaders to speak a particular way—though our leaders rarely do. This puzzling disparity has neither buried the ideal nor, apparently, hampered the reflexive talk of politicians when microphones are thrust in their faces. (In fact, it may have hampered the ambitions of many able politicians we’ve never heard of.)

  To most observers’ surprise, Bush’s speaking, blunders and all, turned out to be more of a political asset than a liability. In 2000 and 2004, around half of the American electorate seemed willing to accept his verbal blunders as an authenticity that they found lacking in smoother-tongued politicians. Perhaps they liked it for the same reason that public radio listeners love the shuffling voice of Ira Glass: the quirky casual, whether it’s intentional or spontaneous, can inspire more trust than the slick and polished. This seemed to be a remarkable moment in the public life of language in the United States. But no one stepped up with an explanation of what was going on and why. Eventually I decided to write the book myself. This is that book.

  Around the same time, I was working as a journalist and listening to myself in interviews I had taped. Now, I speak slowly, in a way that some would call saturnine; I make my points, but I take my time getting there, as much out of habit as preference. As a writer, I struggle on a daily basis to make my own writing as good as it can be, a form of self-expression in which I feel most comfortable. Yet my own talking resists improvement, which can be frustrating. I became curious: How much of my verbal style belongs to me? How much to my family or my community? How much of my perceptions of that style are shared by my listeners?