Sunday, July 15, 2007

Passive Observation

Earlier this week, I took a look at the TED Talk presented by Jonathan Harris, creator of the programmed-art installations WordCount, 10x10, Phylotaxis, We Feel Fine, and several more, including his most recent project, Universe. Universe, like most of Harris' work, presents a more dynamic and aesthetically lively interface for encountering large samples of texts, such as news feeds from all over the world, collections of blog entries, or the British National Corpus.  No question Harris' projects stand apart from nearly everything else I've seen online where sizeable corpuses are rendered visually. I mean that these projects are created in such a way that they lead with artfulness, enriching data visualization with aesthetics.

If you play the TED Talk, you'll hear Harris talk about We Feel Fine, a project that gathers "feel" and "feeling" statements and funnels them back through the interface he designed.  He calls this process "passive observation." Subjects implicitly assent to We Feel Fine's use of their writing because they have published it online.  It is out there, available on the web. In this context, the "passive" allows for working with a large sample of texts.  It's not humanly possible to read every weblog entry published in a single day for instances of "feel" and "feeling."  I think of Harris's methods as well-aligned with Moretti's "distant reading." Passive observation, in effect, bears some correspondence to distant reading.  In much the same way "distant" is a term in need of recuperation, particularly in the humanities, so too is "passive," given that active and activity usually win the day.  Rare are the arguments against the active, against activity, against activism, against the act (as event?). Is passive in "passive observation" opposite of active? Not necessarily.  If by "active," we refer to the efforts it would take an army of readers to glance a few million blog entries for "feel" and "feeling," passive indicates a different way--aggregate, casual filter, pass over if, drifting attitude or manner (the sub-terrain of Burke's agency, which incorporates instruments).  Relaxed finding not so much bound to today's set of blog entries as a focal act or object of study but speculative and futuristic, open to an undetermined end.

Harris's most recent project, Universe, abstracts global news coverage. Universe is explained at length in the talk linked above.  The notion of constellations is central here.  Constellations of words, references, names, figures.  And although Harris's work serves generally as a relay to the texts, he does not seem concerned with writing or rhetoric. The writing that all of these projects piggyback is phenomenal, its constructedness and context is downplayed if not ignored altogether. Within Universe, Harris extends the forming of constellations to a "mythology of the world." Maybe we could hold up these projects alongside Barthes on the spreading and ripening of myth and its social geography; its discursiveness, its rootedness in select social strata, and its micro-climates:

Thus every myth can have its history and its geography; each is in fact the sign of the other: a myth ripens because it spreads. I have not been able to carry out any real study of the social geography of myths. But its perfectly possible to draw what linguists would call isoglosses of a myth, the lines which limit the social region where it is spoken. (149)

After all, these projects set out after text-based pattern. 

While I am enthusiastic about Harris's work, I view much of it with a faint wish at the back of my mind, a wish that it will one day be set loose from the gallery so that others might adapt and appropriate it (with credit where due).  Maybe I've mentioned it before, but why not have WordCount scale to any set of texts? Certainly Harris is under no obligation to share the back-end on any of these projects, particularly the attention-getters and recent releases.  But at a time when the only attention being systematically given by universities to large corpuses of texts is to march student essays through the criminal-infested by-ways of Turnitin.com, it's encouraging to think about some of the other ways these open-ended text-trajectories (i.e., student writing or writing in any field or discipline) might be read distantly or observed passively.  We Feel Fine for an entire curriculum or program, where "feel" and "feeling" can be aggregated and re-associated right along with [verbs of choice, perhaps "argue" and "arguing" for so much academic interchange].  Imagine a hybrid set of applications blending Harris's projects with MONK and with texts of the everyday not limited to news feeds, digitized literary archives, or national corpuses, applications, that is, scalable to whatever.

Bookmark and Share Posted by at July 15, 2007 12:40 PM to Methods
Comments