Clinamen

Image Credit: “Sound Effects Box” by Mark Demers

On Friday, I posted the first chapter of a new project called Making Machines, a book/website that will combine (or “mashup”) works from the rhetorical tradition to build new theories and concepts. This first chapter combined George Campbell’s The Philosophy of Rhetoric and Cicero’s De Oratore in a mashup entitled “Sonic Eloquence.” Here’s a brief excerpt from the “mashup” essay:

“Together, Campbell and Cicero offer us a theory of sonic eloquence, a term that would help us consider eloquence not only in terms of words and meaning but also in the tones and musicality of the voice. These two texts offer detailed accounts of how the human voice persuades and even how it does so in a way that sneaks underneath our cognitive radars. Thus, sonic eloquence operates at various registers, at both conscious and unconscious levels. Their focus is on the human voice, but we could extend Cicero and Campbell’s discussions to other realms of sound. Sonic eloquence need not be confined to the conscious choice of the orator or even to human voices. Various frequencies might be persuasive, suggesting that sound itself can be eloquent or ineloquent, persuasive or unpersuasive. Thus, sound is not only a tool we use but is also a force that operates on us.”

That essay and its accompanying digital material – a brief discussion of how looping audio reveals sonic eloquence and a tool called Sound Arguments that examines audio in various ways – were generated over the course of a month. As I mentioned in a blog post on July 8, the rules were that I had to finish this chapter before the Making Machines bot spit out another pairing of texts. The new pairing appeared this past Saturday, and my next task is to combine Stanley Fish with Giambattista Vico. (The fact that the bot has generated two pairings of white guys says something about what we call “the rhetorical tradition,” and I’ve been wondering if it might be worth writing an algorithm that weighted the works rhetorical tradition differently. That weighting could account not only for race and gender of authors but also for geography or historical moment. If the rhetorical tradition skews west, white, and ancient, would it make sense to balance things out algorithmically?)

Generating this material over the course of a single month was an interesting experience, and (perhaps unsurprisingly) the self-imposed deadline was generative. I found myself “watching the clock,” making sure I’d budgeted enough time to read, write, design, collaborate, and code things. The arbitrary one-month deadline also meant that Eddie Lee and I had to release an alpha version of the Sound Arguments tool that we’ve been hatching. Even given it’s rough state, I’m particularly proud of this part of the project because it demonstrates the kinds of collaborations that can emerge in these “mashups.” In future chapters, I’m hoping to be able to team up with programmers and designers that can help me make various machines. Given that future chapters will be created during the course of three months instead of one month, I should have a bit more time to enlist the help of others.

Sound Arguments is still under construction, but I’m extremely happy with what Eddie and I were able to accomplish in just a couple of weeks. The Making Machines site explains the tool and offers some examples of what it can do for those interested in analyzing (or even composing with) audio.

There was one example that I wasn’t able to include on the site prior to the deadline, so I thought I’d share it here. As the site explains, Sound Arguments allows us to represent sound in multiple ways. It generates visual representations (spectrograms) and audio representations (a series of tones or “notes” for a given audio clip). I’m especially interested in the tones generated, but maybe that’s because I’m still learning to read spectrograms. The site contains some examples of these outputs, but here’s another example: Barack Obama’s 2004 DNC address.

Here’s the opening of Obama’s speech:

Download Clip

Here are the tones generated by Sound Arguments:

Download Clip

Here are those two audio files overlaid:

Download Clip

Now, we can go back and isolate the phrase “crossroads of a nation”:

Download Clip

And we can loop it:

Download Clip

And we can grab the Sound Arguments tones associated with that phrase:

Download Clip

If you go back to the initial clip above, it’s nearly impossible not to hear “crossroads of a nation” as a song. As the Radiolab episode I cite in the “Sonic Eloquence” chapter makes clear, this is largely because looping audio draws our attention to the music that lives in language. While it’s weird to think that Obama is “singing” here, it’s also difficult to ignore the song that’s playing right there in the middle of his sentence. Is this a rhetorical tactic? Or is it merely something that lives inside of speech, playing upon our reason and emotion in unpredictable ways? What are the rules of sonic eloquence, and how would we follow and/or teach them? These are some of the questions opened up by the first chapter of Making Machines.

This discussion of “opening up” questions is important, because I want to highlight that this is about as far as the “Sonic Eloquence” chapter goes. In the course of a month, I was (I hope) able to ask some interesting questions about sound an eloquence, questions that were actually exposed by the discussions of sound in Cicero and Campbell. Making Machines is about making theory and making digital objects, but it’s primarily about launching new directions for rhetorical research and rhetorical practice. So, I hope the chapter makes its own “sound argument” and that it finds plenty of textual evidence for tracking sonic eloquence through these two texts, but what happens with the concept “sonic eloquence” or the Sound Arguments tool is an open question that I hope others in the field will see fit to help me pursue.