Mitt Romney's rapid phrase-onset repetition

« previous post | next post »

Mitt Romney sometimes exhibits a rapid repetition of phrase-initial function words, often intermixed with um and uh. This behavior was especially frequent in  the third presidential debate (10/22/2012). Here's an example from the beginning of his first response:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

um uh this is obviously an area of great concern to the entire world
and to America in particular,
which is to see
uh a- a complete change in the- the- the- the structure and the- um the environment in the Middle East.

Just the last phrase:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

uh a- a complete change in the- the- the- the structure and the- um the environment in the Middle East.

A few phrases later in the same answer, we get

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

uh we see in- in uh uh uh uh in Libya

From his last response, an hour and 20-odd minutes later:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

I- I met a young woman in- in-   in- in Philadelphia
who's coming out of sch- out of college, can't find work.

The last example involves repetition of a sequence of three (well, 2.5) syllables, with substitution of one content word for another — "out of college" replacing "out of sch[ool]" — but most of the examples involve repetition of (part or all of) a phrase-initial monosyllabic function word, like "I" or "the" or "in", which is not abandoned but would be retained (as one copy) in an edited transcript.

The period of the repeated syllables is about 250 milliseconds on average, in the examples that I've examined.

These onset-repetitions are not constant in Romney's speech — they were much less common in the first two presidential debates, and even in the third debate, he sometimes delivered a long sequence of phrases with none of them at all. But repetition-rich regions can be found throughout his recorded performances, as in this passage from the Republican debate held in Charleston SC on 1/17/2012, addressing Newt Gingrich:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Let me tell you, one of- one of the things I find amusing
is- is listening- is listening to how- how much credit is taken in Washington
for what goes on on Main Street.
uh I- I- uh I mean- y- y- you- mister speaker
it wss- it was- you- you talk about all the  things you did with Ronald Reagan
and- and- and  the rev- the Reagan revolution
and the jobs created during the Reagan years and so forth
I mean I looked at the Reagan diary
you're mentioned once in Ronald Reagan's diary
and- and it's- and in the diary  he- he says you had an idea
in a meeting of- of young  Re- uh uh uh congressmen
and it wasn't a very good idea and he dismissed it.
That- that's the entire mention
and I mean he mentions George Bush a hundred times,
he even mentions my dad once,
so I- I- I- th- there's a sense that Washington  is pulling the strings in America.

It's plausible that the disfluency-free segments are well-rehearsed chunks of stump speech, with the onset-repetition disfluencies appearing in more extemporized passages and increasing with higher physiological arousal.

Everyone has these onset-repetitions sometimes, and the phenomenon doesn't seem to be especially salient to listeners — thus few people to have commented on this aspect of Romney's speech, although it has been a frequent feature of his interviews, debates and speeches in the past. For that reason I've avoided called it "stuttering", since stuttering is something that listeners expect to notice immediately.

As evidence that Romney's performance in the third debate was an especially rich source of examples of this phenomenon, I give you the results of this morning's Breakfast Experiment™ — my counts for both candidates in the whole debate:

Seconds Words False starts Um Uh Disfluencies/min Disfluencies/100W
Romney 2560.5 8819 357 9 171 12.58 6.09
Obama 2356.8 6842 63 0 68 3.34 1.91

Methodological notes: Phases where more than one person spoke at the same time were excluded entirely from the tally; "words" includes false starts and filled pauses; a sequence like "in the- in the" counts as just one "false start" (and usually the published transcripts would omit both of the first two words in such cases); possessive 's was split off as a separate word, but other contractions were left intact.

Next time: a verbal quirk that Obama exhibits in this same debate.

[N.B. A typographical error in my transcription file caused one of Obama's turns to be attributed to Romney. Fixing the mistake transferred 132.5 seconds, 367 words, 2 false starts, and 2 uhs from Romney to Obama. The result of the fix is to slightly increase the difference in disfluency rates: before it was 12.05 disfluencies per minute and 5.89 disfluencies per 100 words for Romney, vs. 3.43 per minute and 1.97 per 100 words for Obama. The new values can be found in the table above.]



11 Comments

  1. Dan Lufkin said,

    October 28, 2012 @ 11:18 am

    If you look at the photos of the windshield of the Citroën that Romney was driving in that crash in France and recall that the police had him DOA at the hospital, he was lucky to get away with a little disfluency.

  2. Victor Mair said,

    October 28, 2012 @ 11:45 am

    Why do people do this? Here are some of the reasons I can think of:

    1. gives them time to think (pause time-fillers / extenders); Chinese speakers often use nàgè nàgè nàgè nàgè (but usually pronounced nèige nèige nèige nèige) 那个那个那个那个 ("that that that that") or nàme nàme nàme nàme 那么 那么 那么 那么 for this purpose; someone I know once uttered this amazing sentence: "Nàme nèige rén zěnme nàme nèige?" 那么那个人怎么那么那个? ("How can that person be so that wayish?")

    2. to prevent others from breaking in or talking over the speaker

    3. out of nervousness, hesitation, or uncertainty

    4. a sheer stutter, which would give it a purely neurological or neuromuscular basis

    5. for emphasis

    I believe that "Obama's 'is is'" (http://languagelog.ldc.upenn.edu/nll/?p=4269) and his other types of repetition sometime function in these ways as well.

    When I was an undergraduate at Dartmouth (long ago!), there was a professor of Classics who used the syllable "oink" for these purposes (someone actually told me that it was a Greek interjection or pause marker!). Being from Ohio farm country, it was all I could do to keep myself from laughing throughout his lectures. About five years ago I encountered another academic who used the same pause-filler in his lectures.

    But maybe I'm way off base with these remarks. Perhaps rapid phrase-onset repetitions are a different category altogether from what I've been calling "pause-fillers" in this comment.

  3. ALEX MCCRAE said,

    October 28, 2012 @ 1:15 pm

    @Mark L.,

    For me, the results of your "Breakfast Experiment" are a bit of a shocker, and a true revelation, considering that Obama is often mildly critiqued for his occasional halting manner of public speaking, predicating his sentences, (or sentence breaks), with one, or perhaps a series of "uh"s, or "um"s.

    Anecdotally, when Obama is parodied on satirical political sendup skits on SNL, both by his former spoofer, Fred Armisen, and current excellent newbie Jay Pharoah, this Obama speech quirk is played up with even exaggerated frequency.

    In your analysis of the Obama/ Romney 3rd presidential debate, it is strikingly significant that Romney's disfluencies exceeded Obama's by roughly two-thirds. I would have never guessed that there would have been such a divergence in numbers between the two, with Romney being the major offender, by far. Romney's huge total of false starts was also a bit of a shocker, as well.

    So do Romney's frequent hesitations, and 'stutter-steps'* reflect perhaps uncertainty in what he is about to say, or as some have suggested earlier, is it almost an automatically triggered 'device' to allow him a smidgen of time to quickly structure, in his noggin, what he wants to articulate; being extra cautious in not blurting out some bone-headed comment that he would sorely regret, post debate?

    I sense in some of Obama's largely partisan political rallies, out on the hustings, that he, at times, intentionally uses the repeat word 'device', eg. …the….the…the…the……, as kind of an emotional build up to delivering an anti-GOP zinger. Almost like a little lead-in vocal drum roll. I feel it also gives his manner of speech, and delivery a more folksy, conversational, colloquial air, which hopefully further endears him to his listening audience.

    *I'm not implying clinically diagnosed stuttering here.

  4. Lazar said,

    October 28, 2012 @ 3:59 pm

    Obama has, of course, got his "is is" and his pretty frequent "um"s, but I think his most distinctive verbal tick is his heavy use of "you know", often as a very quick monosyllable.

    [(myl) Yes, that's the one I've got queued up to write about. There are 21 examples in debate #3 — not a lot compared the phenomena discussed in this post, but more than enough to qualify as a verbal tic.]

  5. Dick Margulis said,

    October 28, 2012 @ 4:20 pm

    " thus few people to have commented on this aspect of Romney's speech"

    It may just be that right-wing commentators are more prone to pick on Obama's involuntary verbal tics than left-wing commentators are to pick on Romney's involuntary verbal tics.

    [(myl) Perhaps, though there was plenty of unfair and inaccurate attention paid to George W. Bush's alleged "Bushisms". As far as I know, even comedians caricaturing Romney don't pick up on this feature — as I said, it doesn't seem to be very salient.]

  6. Barry Ross said,

    October 28, 2012 @ 11:55 pm

    Could this repetition be related to the speed of Romney's speech? From your chart it looks like he gets about 205 words out per minute versus 173-4 per minute for Obama.

  7. Acilius said,

    October 29, 2012 @ 9:03 am

    @Professor Mair:
    Considering that the counts are from televised debates, I'd suspect that your #2 is likely to be heavily represented among the results.

  8. Tom Saylor said,

    October 29, 2012 @ 2:26 pm

    Is it at all significant that Romney's disfluencies occur in prepositional phrases that begin with "in" and have place names in their complements? Was he maybe having trouble recalling his prepared geographical references?

    @Victor Mair: "When I was an undergraduate at Dartmouth (long ago!), there was a professor of Classics who used the syllable "oink" for these purposes (someone actually told me that it was a Greek interjection or pause marker!)."

    When I was a graduate student at Princeton (also long ago) there was an eminent professor of ancient philosophy, German born, who inserted the syllable "swine" into virtually every sentence he spoke, in the manner of a verbalized comma. Classics grad students called it a "porcine particle."

  9. Joseph Lee said,

    October 30, 2012 @ 12:21 am

    The rate of disfluencies might reflect the power the central processing Unit of the computer (brain) some has higher hertz some has lower hertz.

  10. Alexander said,

    October 30, 2012 @ 8:13 am

    @Tom Saylor
    Though the pronunciation was more like "svine", and there was a claim, consistent with this, made by other past graduate students, that it was a contraction of "it's [so] fine". A friend once asked him what was up with that, and he responded, "It's just a, svine, filler." Interestingly, it was *not* associated with pausing. It often happened in very fast speech, and even in the morphological breaks of compounds, e.g. "Middle [svine] Ages".

  11. Sarah Kuhn said,

    November 6, 2012 @ 5:54 am

    @Victor Mair: "When I was an undergraduate at Dartmouth (long ago!), there was a professor of Classics who used the syllable "oink" for these purposes (someone actually told me that it was a Greek interjection or pause marker!)."

    @ Tom Saylor: "When I was a graduate student at Princeton (also long ago) there was an eminent professor of ancient philosophy, German born, who inserted the syllable "swine" into virtually every sentence he spoke, in the manner of a verbalized comma. Classics grad students called it a "porcine particle.""

    And when I was an undergraduate at Princeton, I went to the first meeting of a class by eminent sociologist Daniel Bell. I enrolled in the course because I thought I'd never heard someone so fluent and erudite. I came to understand that his "filler" was not "um" or "ah" but the phrase, "in this fundamental sense." He used it many times over the course of each lecture. In fact, one day I stopped in at his office and found his secretary transcribing a talk he had given. I asked her, "what do you do with 'in this fundamental sense?'" Her laugh of recognition told me all I needed to know.

RSS feed for comments on this post