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

by Erard, Michael.

  But the electronic hardware didn’t come with instructions about how people should use it. In the same way that e-mail etiquette has shakily coalesced, the standards about how to speak on the radio took time to develop. Faced with a solitary microphone, not the shining faces of a real live audience, the earliest radio performers (many of whom were veteran actors) experienced “mike fright.” Radio also highlighted the weaknesses of speakers and was a merciless test for the speakers of the “rough-and-ready, catch-as-can school” (wrote the Saturday Evening Post in 1924) who couldn’t employ devices (the facial expression, the shifted posture) that kept the attention of physical audiences. In the first stage show broadcast by New York’s WJZ in February 1922, Ed Wynn, a comedian, was so struck by the silence that followed all his jokes that his performance began to falter. The producer ran through the building, assembling in the studio everyone he could find—electricians, janitors, telephone operators—to give Wynn a real audience. Once he heard their responses and saw their faces, Wynn was able to keep performing.

  One impact of these technologies of the voice was that people developed new standards of speaking for them. In the age of recording and broadcasting, a “good speaker” had a different style of voice. T. H. Pear, a British psychologist, described the speechifying of the preelectric era as taking place in halls with “bad acoustic qualities” as “bored or fidgety” audiences sat in uncomfortable chairs and were occasionally woken up by a speaker shouting, “Ladies and gentlemen!” In such settings the style of speaking was slow, overarticulated, and often “childishly simple.” Important words were SHOUTED! And important thoughts. Were broken. Into pieces. By dramatic. Pauses.

  Microphones gave speakers the ability to speak at a volume more suited to face-to-face conversation. Early radio orators didn’t grasp this at first—until they discovered that booming oratory could blow out a microphone. Speaking at a normal volume meant speaking at a normal pace, since the speaker didn’t need to pause as often to gather huge breaths. It also made sentences sound more connected, uninterrupted by gulps of air. Robert West, in his radio handbook, So-o-o-o You’re Going on the Air!, called the preelectric orators “leather-lunged word hurlers who depended on stentorian power to carry their voices to three counties at one time.” Once electromagnetism displaced sound waves in the air as the vehicle of the voice, the elocutionary standards of the nineteenth century were rendered obsolete.

  The elocution manuals had provided elaborate instructions about when to pause and how long to do it. It was a mechanical necessity—the orators needed a space in which to take a breath. Yet that same pause may have given the speaker a moment to plan what to say next. When the need to project the voice disappeared, so did the luxury of the pause. In this way the utilitarian pauses of oratory might have become the useless hesitations of the electrically amplified public speaker. It was in these hesitant moments that the “ums” are likely to have been spoken and recognized.

  Provided with new opportunities to hear real speaking, people began to prize umlessness. James Winans, a professor of speech communication at Cornell University, was one of the first to extol the conversational style and help transform “oratory” into “public speaking.” His textbook Public Speaking, published in 1915, was also the earliest textbook I found that explicitly advised against saying “uh” and “um.” (A few textbooks quoted only Oliver Wendell Holmes’s Urania.) Though the speaking style that he recommended was informal, it was less interactive than actual conversations. “Hesitation is especially annoying when the gaps are filled with urs and uhs,” he wrote. “Grunting is no part of thinking.” To emphasize his point, he quotes Holmes’s poem, which dictated against those “dreadful urs.”

  Hesitation was a common handicap of the unrehearsed speaker, who also erred by speaking too quickly. Winans urged the speaker not to confuse pauses—which he said provide natural boundaries between topics—with hesitation. “We pause to think; we hesitate because we cannot think,” Winans said bitingly. “Nothing is more tiresome to an audience than a hesitating, halting delivery.”

  Teachers of public speaking as far back as Quintilian had cautioned the speaker against hesitating. They advocated a style that would later be labeled masculine, controlled, and strong. Now the hesitant style had a name, and, thanks to the technology, it had a symptom. No longer referred to vaguely as “stammering,” hesitant speaking could be prevented if speakers followed this simple directive: don’t say “uh” or “um.”

  This admonition lodged in the culture of language with remarkable speed and emerged as a key standard for speaking in the electric age. It’s not clear whether other cultures share the umless aesthetic that Americans acquired in the early twentieth century. But we can safely suppose that in other parts of the world, technologies would also have made “uh” and “um” perceptible by throwing the fleeting sounds of interruption up to our ears so they were impossible to ignore. Recording, amplifying, and replaying human speech allowed it to be sent across time and through space. This not only transformed captured sound into a commodity, it enabled ways of listening to the aspects of the human voice that had once vanished in the air, overriding the natural filters that kept the “uhs” and “ums” of daily life beyond the limits of people’s conscious attention.

  People must have heard pause fillers so frequently that by the late 1920s, “uh” and “um” seemed unhygienic. In 1928, a public-speaking professor, Wilmer E. Stevens, published the first rating scale for grading student speeches or judging speech contests that included the absence of pause fillers as a sign of prized fluency. “Consider the flow of the speaker’s words,” the scale reads. “Notice tendency to block, repeat, stutter, or to say ‘ah.’”

  Writing in 1931 about the psychology of public speaking, T. H. Pear ranted against “er,” saying it exemplified the halting, fragmented style of speaking that now seemed to be cropping up everywhere. “How much time is given, in many secondary schools, to the writing—and speaking of Latin by boys and girls who cannot speak English coherently in public for two minutes?” he wrote, then mimicked the educated gentleman who’s grown up into an average speaker who begins his speech with “I’m sure—er—we’re all very—er—grateful—to the—er—speaker…for his interesting and—er—suggestive—er…talk.” Pear was incensed. “Cannot the generation which has added flying to the accomplishments of the human race do better than this?” he ranted. “What is the reason? Nervousness? A foolish, misleading word. Self-consciousness? A little better, but is complete anaesthesia the highest aim of civilisation?”

  As the conventions of the public sphere invaded the private sphere, the dictates against “uh” and “um” in normal, spontaneous conversation also made inroads. Beginning in the 1920s, the arbiters of well-mannered society denounced “uh” and “um” as frequently as the arbiters of the well-spoken did. In 1928, in Manners: American Etiquette, Helen Hathaway cautioned against “ah” and “er” as “little mannerisms” and “peculiarities of speech.” “Well-educated and well-bred people are always the simplest in every way,” she observed. “Puncturing our sentences with ah’s and er’s, mincing our words, employing affected tones and gestures, are tricks annoying even to our friends and positively repellant to strangers.”

  Even in that most intimate of twentieth-century conversational settings, the therapist’s office, “uh” and “um” were frowned upon. In his 1959 Mannerisms of Speech and Gestures in Everyday Life, Sandor Feldman, a psychiatrist, noted the mannerism as “a crutch” among his patients, whose “moaning-like ‘er…er…er…’ is annoying, draws the conversation out to great length, and makes the listener, who is far ahead of the speaker, wait.” Unfortunately, Feldman was convinced of the link between “er” and anxiety: “One can say that whenever this mannerism is used for one reason or another there is anxiety present.”

  Given that other disfluencies, particularly restarted sentences and repeated words, indicate mental disorganization, cognitive load, and anxiety more rea
dily (as George Mahl and Heather Bortfeld, among others, have shown), it is curious that “uh” and “um” are considered the sine qua non of hesitant speaking and disorganized thinking. Though “uh” and “um” are more frequent, sentence repairs and word repetitions clog their share of sentences, too. So why did “uh” and “um” become the shorthand for anxiety and disorganization? Why does their presence make us panic in a way that other disfluencies don’t?

  It may be that “uh” and “um” became the emblem of a disorganized self. If the aesthetic of umlessness applied only to public speaking, one might say that orators were most concerned with pleasing their listeners by leaving out such disfluencies. Yet umlessness was taught as the norm for private and intimate spheres of life. The only rule left unwritten was “don’t say ‘um’ when you’re talking to yourself.”

  This persistence suggests other worries were attached to “uh” and “um” uniquely. For the last century, sociologists have argued that identity is neither essential nor fixed in complex industrialized societies. The modern identity, they say, is performed. Individuals must navigate a variety of settings, interact with many different people, and play diverse roles to achieve a shifting set of goals. The aesthetic of umlessness may have emerged as a response to this demanding modern conception of self. Perhaps saying “um” was a sign that a person couldn’t keep their social performances in order and thus wasn’t fit for this social world. “Um” represented a breakdown of the bureaucracy in your head.

  Umlessness isn’t a natural preference—left to our own devices, we naturally ignore most “ums.” People began to prefer umlessness in public speaking and conversation around the same time they began to value order, organization, planning, and efficiency in an increasingly complex and urbanizing society. People worked hard to keep control of how they performed socially, and they prized and admired others who did the same. We prefer in speakers what we prefer in our individual selves.

  Frederick Houk Law, a lecturer at New York University and head of the department of English at Stuyvesant High School, wrote an eight-part textbook, Mastery of Speech, that linked success in the dizzyingly complex social life of modernizing America with talking; he specifically mentioned “uh” and “um.” Published in 1918 and intended for men and women, Mastery of Speech gave how-to advice on a range of interactions: “how to dictate to a stenographer,” “how to talk to a private secretary,” “how to ask for a loan,” “how to speak to unfortunates,” “how to talk to public officials.” Law outlined what’s at stake in the hypothetical biographies of “Henry” and “Jim,” both of whom are equally privileged. But Henry is rich, while Jim is stuck in his mediocre job. The reason? “[JIM] IS A POOR SPEAKER.”

  Henry’s a good speaker because he “can talk to any man at any time, and he meets everyone in exactly the right way.” He can act as a boss, a colleague, a citizen, a bystander, a customer, a client, and a parent all in sequence, keeping track of these performances of a social self in his repertoire and shuffling the right one out at the appropriate moment. Saying “uh” or “um” signals that one doesn’t, in fact, have control over these performances. “Lack of fluency may be caused by bashfulness or by fear of your audience,” Law wrote. “You have somehow formed the habit of saying ‘Well, now,’ ‘and-a-a-a-a,’ ‘n-o-w,’ ‘erer-er-er,’ and various other expressions. Resolve firmly to break the habit and to speak fluently.” The message is clear: if you say “uh” or “um,” you’re doomed to live the life of Jim, not Henry.

  After the pioneering work of George Mahl and Freida Goldman-Eisler in the 1950s, it took some time for the study of pauses, pause fillers, and other disfluencies to be seen as valuable. But once sociologists, psychologists, linguists, speech pathologists, and communication theorists started to plumb the depths of “um,” they explained the universe of pause fillers in myriad ways. Some said an ummer was searching his memory for a word. Others concluded that the ummer was uncertain. In the view of some, the ummer was trying to decide what to say or how to say it. For others, it was a way to keep a place in a conversation, while others thought it was a way to signal that one wanted to take the next turn. Others said the opposite, that the ummer wanted to give up the turn. Meanwhile, in the popular mind, umming was simply a bad habit, akin to spitting or picking one’s nose.

  Some communication theorists argued that the umming public speaker discredited himself and his message; others found evidence (usually by surveying college students in one’s own speech classes where umlessness is insisted upon, which means one is surveying a biased sample) that it discredited the speaker but didn’t interfere with the message itself. Some theorists think that “um” invited others to speak, or requested help finishing a sentence, or invited a listener to think about what the speaker was about to say, or was simply another way of being polite.

  One of the most far-reaching conclusions about “um” comes from Herb Clark, a psychologist at Stanford, who submits “uh” and “um” to exhaustive exegesis. A slim man in his midsixties, Clark keeps an electric kettle and a row of tea boxes for visitors to his office. Because of what he studies, it’s hard not to notice how he speaks: he’s a fast talker, not given to much umming, though his sentences often shift midway through an idea. The effect isn’t distracting. When I transcribed our interviews, I realized how deliberately he places his “ums” and his repetitions, often adding a commentary, sometimes as brief as a single word, on his own pause.

  Which turns out to be the heart of his argument about “uh” and “um.” Clark argues that when people talk, they produce two parallel streams of information he calls “tracks.” One is the “primary track,” where the subject of the conversation—the football game, a novel, dinner—travels. The other is the “collateral track,” interjected with the speaker’s commentary on his or her primary track. Editing comments (“sort of”), moments of self-consciousness (“what was I saying?”), even dramatic asides (“as if I cared”) all belong to the collateral track. The collateral track coordinates the primary track in one person’s speech but also coordinates the communication between two speakers.

  Clark likes to use the metaphor of a waltz to describe this communication—that is, as a dance in which participants provide subtle cues to each other about how and where to move. When someone forgets a word in the middle of a sentence and says, “I’m trying to think of this word but it’s not coming to me,” she’s coordinating the conversation by explaining why her primary track has faltered. When she repairs and then apologizes for interrupting herself, she’s coordinating as well. For most of us, this happens at a fairly unconscious level—we’re just trying to keep the conversation going—but someone like Clark, practiced at controlling his collateral track, can manage this subordinate conversation with great consideration for his listener. Far from being distracting, his speech seems to embrace the listener in warmth.

  For Clark, “uh” and “um” are valuable tools in this waltz-like communication, and they function more like words than conversational grunts. Not only do they sound as if they’re being used deliberately, they seem to have two distinct meanings. With Jean Fox Tree, Clark’s former student, now a psychologist at the University of California, Santa Cruz, Clark examined about 170,000 words of transcribed conversations between adults, 5,000 words of calls to telephone answering machines, 2.7 million words of telephone conversations, and thousands of words of storytelling. Clark and Fox Tree discovered that “uh” and “um” always preceded some notable pause or delay in both the monologues and conversations. They figure that speakers are deliberately signaling the upcoming delay in the sentence, and doing so with “uh” or “um.” As Clark likes to put it, the “uh” isn’t the problem; it’s the solution to the problem.

  Clark and Fox Tree also argued that “uh” and “um” have two different meanings. In their study, “uh” appeared more often before a short pause, “um” before a longer pause. Not only do speakers have control of “uh” and “um” (as demonstrated by the fact th
at some people can learn to stop saying them) but their internal speech monitors could deftly foretell how long a delay was likely to be. That formed the basis of whether or not to use “uh” or “um.”

  Whether or not one agrees with Clark and Fox Tree’s perspective, the notion of conversation as waltz suggests a reason that some novice public speakers say “um” too much. They may be using the word to try to interact with their listeners, and since they get no feedback they can use, they keep saying “um” to no avail.

  From the listeners’ perspective, what’s annoying about this is that there’s no way to signal, we’re with you, keep going, you’re doing fine. A flood of “ums” is like a continual cry for help by someone who can’t help him-or herself. In a 1988 Psychology Today article about speaking as communication, psychologist Michael Motley offered a way of circumventing this frantic flood drawn from his public speaking classes. He begins a conversation with the speaker and tells the rest of the audience to leave the room. Once the speaker is confident in the conversation he or she is having with Motley, the rest of the audience returns until the speaker is interacting with all of them.