An
article posted last week on Slate, Michael Erard makes the case for the usefulness of "ums" and "ahs" in speech, "disfluencies" that we typically discourage. Apparently, such pauses in the flow of speech engender a feeling of anticipation in listeners that can focus attention, if not overused. It reminds me, in a way, of musicologist Charles Keil's notion of "
participatory discrepancies," the ways in which being slightly "out of time" and "out of tune" in performance actually can make music feel
more, not less, groovy and danceable.
Unfortunately, it is far easier to understand a speech as an artifact that is to be delivered (without disfluency) rather than an utterance that is heard (with disfluency). In our culture, a public speech is commonly documented as a script that has gone through multiple drafts, can be scrolled on a teleprompter, and whose success depends on an accurate delivery. Even a transcript of a speech, after the fact, tends toward idealized representation, with the true messiness of non-sentences, false starts, and paralinguistic acts edited out and the dynamics of the speaking event largely ignored. Journalistic transcriptions of the annual State of the Union address are a good example of this sort of reduction. While the text of the speech is always offered for study, anyone who witnesses the event knows that a major part of its meaning is about the dynamics of delivery and response:
how things are said and
who in the room applauds, stands, sits, smiles, or scowls--and when. How is that behavior documented? Why don't we have good ways to represent it?
The institutionalization of
langue over
parole leads me to the problem of how we might recognize disfluency in the past. Imagine that you wanted to look back and really analyze the oratory of a public speaker and how his or her delivery was variously heard and understood by audiences. Could you do so? You can partially recover instances of speaking and hearing during the era of recording, using a combination of audio records, film, and contextual understanding of audiences from acoustics, linguistics, oral history, and social history. However, for 19th century public speakers like Henry Ward Beecher, or even early 20th century figures like Theodore Roosevelt (who apparently had a surprisingly high-pitched voice), such analysis is even more difficult. To account for how public speakers were heard before 1900 or so, you can only work with fragments: brief written descriptions (from letters and news reports) of how speakers may have sounded, or how audiences reacted; images and photographs, if you can find any, that might allow you to make something out of the body language and facial expressions of audience members. It's not promising.
It seems to me that the static nature of what we typically deem as "historical evidence" erases the dynamism of the events we would like such evidence to represent. I'm quite interested in thinking more about what, exactly, we archive as historical evidence, something that raises all sorts of questions about processes of interpretation and the limits of historical practice. I also think we can do much
now to more fully document contemporary performing and audiencing (
with its "ums" and "ahs") for the historians of the future.