Which metaphor we choose determines how we regard normal features of spoken language like ah, um, and er.
If we choose the prescriptive, hot dog metaphor, then those are “filler” words. They’re ugly, evil and disgusting. They should be ruthlessly eliminated! Nobody wants a hot dog with filler - we’d all prefer a sausage with “all meat.”
If we choose the descriptive, driving metaphor, then those just are “filled pauses,” which like filled potholes, perhaps aren’t any worse than empty potholes (pauses).
The first viewpoint is common in public speaking textbooks. For example, in discussing pauses on page 302 of the 8th (2004) edition of Stephen E. Lucas’s The Art of Public Speaking he proclaims that:
“Most important, do not fill the silence with ‘uh,’ ‘er,’ or ‘um,’ These vocalized pauses, as they are called, are always annoying, and they can be devastating. Not only do they create negative perceptions about a speaker’s intelligence, but they often make a speaker appear deceptive.”
The second viewpoint began to get widely noticed after Michael Erard wrote an article in the January 3, 2004 issue of the New York Times titled Just like er words not um throwaways. Then, in 2007, he published a book titled Um - slips, stumbles, and verbal blunders, and what they mean. This July in Slate he added An Uh, Er, Um Essay in praise of verbal stumbles. Mr. Erard has an MA in linguistics, and linguists tend to be descriptive, not prescriptive.
Erard’s latest essay prompted a long (over 1600 word) rant by Marsha Hunter on August 16th asking Is Um an Honored Part of Speech? She claimed it was not, and pointed out that it doesn’t appear in scripted (written) speech - novels, plays, or films. She further claimed:
“....People do speak without saying um, and we’ve all heard them: politicians, teachers, professors, lecturers of every stripe, talking heads, our friends, perhaps even you, esteemed reader. A fair percentage of speakers express themselves fluently and smoothly with no ums whatsoever....”
Should filled pauses be eliminated, or just reduced to some acceptable level? Some speech coaches acknowledge that having a few is not fatal. In a blog post on September 12th Simon Raybould said that:
“It’s not about how many times the presenter does, or doesn’t use a particular filler word – if it’s not a problem to the audience.”
What level is unacceptable? On page 54 of her book The Theory That Would Not Die (2011) Sharon Bertsch McGrayne noted that the renowned Cambridge University professor Harold Jeffreys (reportedly an appalling lecturer) was once counted mumbling er 71 times in 5 minutes (14.2 per minute). An article by William H. Stevenson, III in the February 2011 Toastmaster on Cutting Out Filler Words pointed out that Caroline Kennedy had been widely criticized when she used 65 fillers in 5 minutes (13 per minute). A 2008 web article by Felicia Slattery began with a client’s comment that four or five within a minute was too many.
Conversely, what level is acceptable? In a comment on Marsha Hunter’s post I pointed out that professors certainly do use fillers (as was noted by Erard in his 2004 article). Twenty years ago Schacter, Christenfeld, Ravina, and Bilious examined how professors at Columbia University really lectured, and they found one to six ums per minute. See “Speech Disfluency and the Structure of Knowledge’, Journal of Personality and Social Psychology, Volume 60, page 362, 1991. Google the title as a phrase, and you’ll find a free .pdf to download.
The bar chart shows Schacter et al’s results for how frequently professors said uh during introductory lectures (blue) and interviews (pink). (Click on it for a larger, clearer version). During lectures the rate ranged from 1.0 to 6.5 per minute, and depended on the department. During interviews (conversation) it was higher, and ranged from 4.4 to 5.8 per minute.
I think that a rate of 2 filler words per minute would be acceptable. What do you think?