Here we go again. Reporters and Republicans have been making a big deal out of two slips of the tongue that occurred in quick succession during yesterday's rally introducing Joe Biden as Barack Obama's running mate. The ABC News version, for example, is Obama Misspeaks, Calls Biden 'The Next President'; Biden Calls Obama 'Barack America':
When introducing his running mate, Obama said, "So let me introduce to you the next president – the next vice president of the US of America, Joe Biden."
And then when it was Biden's turn to speak, the Delaware senator called the presumptive Democratic nominee "Barack America" instead of Barack Obama.
"My friends, I don't have to tell you, this election year the choice is clear. One man stands ready to deliver change we desperately need. A man I’m proud to call my friend. A man who will be the next president of the United States, Barack America,” Biden said, per ABC News' Sunlen Miller.
Since we know that journalists' transcriptions are not always reliable, let's go to the recordings.
Senator Obama's slip was very simple and clear, and more or less as reported, if you expand ABC's "US" to "United States".
So let me introduce to you [0.417]
the next president [0.886]
the next vice president [0.230]
of the United States of America [1.535]
As the pause durations (indicated in square brackets) indicate, he spent about half a second noticing his slip and deciding how to deal with it.
One thing that I didn't hear from any of the talking heads commenting on the rally, or see in any of the text-journalism reports this morning, is a simple point about word-sequence frequency. Current Google counts are 3.5M for "the next president", more than 100 times greater than the 30.6K for "the next vice president". But over the past 19 months, Senator Obama must have heard the phrase "the next president of the United States" thousands of times more often, because it will have been used several times a day, at every rally and event in a busy campaign schedule, while the phrase "the next vice president" will never or hardly ever have come up.
Senator Biden's slip was a little less crisp, and its causes were a little more complex.
… a man I'm proud to call my friend [0.836]
a man who will be the next president of the United States [0.324]
Barack Americ- [4.252]
You know [1.529]
you know you learn a lot of things being up close with a guy …
This time, it's not as simple as raw word-sequence frequency. Obviously the phrase "Barack America" is not especially common — or wasn't before this event. But there's still a phrase-frequency effect, because the conventional political introduction for presidential candidates is often given in the form "…the next president of the United States of America" — as Barack Obama did in the slip-up discussed above. Biden left out the "of America" part, and as a result, the parts of his brain associated with the word America were all primed and ready to go.
In fact, at the very start of Biden's speech, he used the full form of the conventional introduction.
and I'm proud to stand firm [0.424]
with the next president of the United States of America [0.318]
And it's not just the conventional political introduction that did that for him — according to the transcript of his remarks , he used America or American 32 times in 1667 words. In normal text, that's the sort of rate (about 1.9%) associated in general word-frequency lists with words like in and to. So Senator Biden's neural network for America wasn't just activated, it was throbbing.
Finally, "America" has a fair amount in common with "Obama" from the point of view of sound. They're both polysyllabic words that start with a vowel, end with a schwa, and contain only open syllables. When the -ri- syllable of "America" is fully reduced, as it often is in fluent speech — and was, for example, in the passage from the start of Biden's speech just cited — both words are three syllables with second-syllable stress.
My impression is that we're likely to be hearing and reading a lot more about slips of the tongue by Joe Biden, a fluent speaker who sometimes seems to put his words out before all his neural activation levels have quite finished spreading, so to speak.
This is all explanation after the fact, but it's based on one of the most thoroughly studied areas in psycholinguistics, where theories are based on extensive collections of natural errors, as well as many studies of errors created in experimental settings.
Derek Smith has put a useful review on the web ("Speech Errors, Speech Production Models, and Speech Pathology"). For the present discussion, the most relevant work is based on a model proposed by Gary Dell in his widely-cited paper "A spreading-activation theory of retrieval in sentence production", Psychological Review, 93(3): 283-321, 1986:
The data to be accounted for by the theory are facts about speech errors, or slips of the tongue. [...] Freud (1901/1958) argued that slips, in addition to revealing unconscious motives, may provide “insight into the probable laws of the formation of speech” (p. 71). Meringer and Mayer (1895), who were among the first to draw attention to slips as a data source, hoped to discover the “mental mechanism in which the sounds of a word or sentence … are linked” (p. 10). More recently, Fromkin (1973) remarked that “speech error data … provide us with a window into linguistic processes and provide, to some extent, the laboratory data needed in linguistics” (pp. 43–44). Each of these researchers recognized that the inner workings of a highly complex system are often revealed by the way in which the system breaks down. The theory presented here specifies the kinds of breakdowns that are likely to occur, the constraints on the form of slips, and the conditions that precipitate them. In addition, the theory will deal with the tendency for errors to create meaningful combinations of items, the tendency for similar items in similar environments to slip with one another, and the influence of changes in the speaking rate on errors.
We've discussed the speech errors of politicians (and other people as well) many times over the years:
"You say Nevada, I say Nevahda" (1/3/2004)
"Weisbergism of the week" (4/27/2004)
"A CNN-ism?" (6/18/2004)
"Non-Bushism of the day" (9/27/2004)
"Gibson scores a 'Bushism', with an assist to Kerry" (10/9/2004)
"Ceci n'est pas un Bushism" (10/15/2004)
"The way the cookie bounces" (12/20/2004)
"The Eternal General of the United States?" (5/5/2007)
"Quotes from journalistic sources: unsafe at any speed" (7/9/2005)
"And they're just as ignorant as it used to do" (10/19/2005)
"Don't read it as something more than it's not" (10/29/2005)
"Has George W. Bush become more disfluent?" (11/17/2005)
"Trends in presidential disfluency" (11/26/2005)
"Presidential self-repair" (11/28/2005)
"The [sic]ing of the president" (1/12/2006)
"An editorial conflict of interest at Slate?" (6/22/2006)
"Productivity at the Bushism mine" (7/2/2006)
"Keep on truculent" (2/4/2007)
"'Republicans and Democratics': Hypercorrection or speech error?" (6/7/2007)
"A kinder, gentler speech error" (6/8/2007)
"An experimental control" (11/26/2007)
"Hoping to be haunted by legitimacy" (3/30/2008)
With respect to speech-production blunders all across the political spectum, our perspective has been consistent, . Everyone commits speech errors, even professional talking heads, and anyone who makes a big deal about particular examples is either a fool or a hypocrite. Since fatigue, stress, and complex ideas all promote speech errors, you can depend on political rhetoric to provide plenty of occasions for foolishness and hypocrisy.