"I'm no PhD."
So begins today's Library of Congress blog post by Director of Communications Matt Raymond, which breaks the news that the LOC has acquired every public tweet published since Twitter launched in 2006: billions of lines of text each numbering 140 characters or less. On hearing this announcement over NPR airwaves this evening, I experienced both curiosity and concern about the Library's plans to preserve this vast digital archive. My concern is less about privacy, however, and more about how the LOC and Twitter are framing the archive's implications for nothing short of democracy, history and intellectual inquiry of all kinds.
Consider Raymond's claims for the archive's scholarly value: "I'm no PhD, but it boggles my mind to think what we might be able to learn about ourselves and the world around us from this wealth of data. And I'm certain we'll learn things that none of us now can even possibly conceive." My quarrel with Raymond––and by extension the LOC––is that they present Twitter (and social networks more broadly) not as a serious object of study that demands the expertise of scholars (specifically scholars of media theory and emerging media) but as a "wealth of data" about, well, everything and anything but the media itself.
At the same time, I find myself taking issue with Twitter's framing of the digital preservation project as a testament to primarily the media's power. Expressing the company's excitement that "tweets are becoming part of history," Twitter suggests that tweets enter history––and thus create an "open" and "global" exchange of ideas––simply by entering a digital archive. Yet the archive in question, we should note, is national rather than global. And, more pointedly, tweets are both already in the historical record––thanks in part to the newly-launched Google Replay site––and, as a single body of text, utterly illegible. In other words, tweets about Justice Stevens's retirement (the chosen example of the LOC archive's potential importance to scholars) have very little to do with tweets about either the Copenhagen Climate Talks or Radiohead's "sliding-fee-scale" online release of In Rainbows.
My point here is simply that the significance of this archive––which prompted my own flutter of intellectual interest––is undercut by both the LOC's neglect of the tweet's particularities as a media and Twitter's implicit suggestion that the content of any particular tweet matters little to the historical and political significance of preserving all the tweets for all time.*
With these initial reflections then, I close by asking how, as critics and scholars, we might think about (or perhaps rethink and reframe) this archive of the super-ephemeral?
*I should note, following Twitter's lead, that "only after a six-month delay can the Tweets be used for internal library use, for non-commercial research, public display by the library itself, and preservation."
Surely there must be national mood swings -- think of 9/11, New Orleans flood, or national elections -- those should show up in collective twitter feeds, and not just as explicit content. I suspect more subtle things should show up as well.
"Tweets" are interesting both in terms of their content -- what an individual says about him/herself in "tweet" form, or an issue of the day -- and in terms of the aggregate actions of large numbers of tweeters -- some subjects receive much more "tweet" attention than others. Collecting the Twitter archive presumably makes it easier for scholars to learn all sorts of things about the "twitter" stream, as one stream of communication among many others. Of course, the form of the tweeting environment -- including third-party client-created environments -- also matters: who follows whom? Who retweets what? How frequently does a 140 character string get retweeted, and by what communities? With what changes or comments? How does a particular Twitter client change the way people tweet? Do people tweet differently from mobile devices vs. their computers? To which parts of the Web does a Twitter conversation (and set of links) send us? Etc. The "Twitter" form, though in a sense a branded or proprietary medium, seems no different to me than any other mode of communication, neither more nor less special. It seems entirely appropriate to me that the LOC should seek to record this form of public discourse, and likely that scholars could learn a lot from studying the tweet stream, among other vectors of discourse.
I appreciate these responses, as I am beginning a bit of research on Twitter that will explore how tweet feeds represent (or do not represent) "sustainability" as a concept and will investigate how "sustainable" Twitter itself is as compared to other, more media-intensive social networks. In this context, I appreciate your assessment, Lee, of twitter "streams"; any research in the new archive will certainly want to consider precisely the questions you raise about how individual tweets--via retweeting, following, listing, etc.--"bubble up" or cohere as a field of discourse and, in some cases, activism. So I agree that Twitter's digital archive offers a rich repository for scholars. However, I would disagree that the form of the tweet itself is largely irrelevant; this is precisely my grumble with how the LOC has framed the acquisition. The form of the tweet––its contours as a 140-character bit of text that can contain hyperlinks but no rich content (images, embedded videos, etc.)–––seems both worthy of study and crucial to research on any particular "stream."
Hi Allison. I might have been unclear in my reply, and I may have misunderstood your initial post. I do think that the form of Twitter -- its use of 140 characters, the temporal flow of tweets, etc. -- matters a great deal. What I was suggesting was that I didn't see anything wrong with what the LOC was doing or Matthew Raymond's framing of the acquisition, though I do recognize that Twitter's incentive is to cast Twitter as world-historically important, and may therefore tend toward self-serving hyperbole when describing why the acquisition matters.
I guess I don't understand what you mean when you write that "Raymond... present[s] Twitter... not as a serious object of study that demands the expertise of scholars ... but as a 'wealth of data' about, well, everything and anything but the media itself." Twitter, meanwhile, casts "digital preservation project as a testament to primarily the media's power." You write that "the significance of this archive... is undercut by both the LOC's neglect of the tweet's particularities as a media and Twitter's implicit suggestion that the content of any particular tweet matters little to the historical and political significance of preserving all the tweets for all time." I was trying to say in response to this that acquiring Tweets doesn't differ in kind from acquiring anything else: from a writer's papers to old manuscripts to video game consoles.
Libraries acquire archives of all types for all sorts of reasons and often think of these acquisitions in very narrow terms. Fortunately, scholars are very inventive in finding new uses to put those archives to. The value of archiving Twitter seems very great, even if the LOC and Twitter don't grasp its full importance.
Thank you for that clarification, Lee. In hindsight, I realize that my post may have been a bit muddled. I suppose I was just looking for the LOC to reflect a bit on some of the particularities of the media. While recognizing that the twitter acquisition has much in common with other digital archive projects and perhaps archives in general, and to Raymond's point does indeed offer a "wealth of data" for scholars, I am most intrigued myself with some of the new challenges/opportunities this archive will pose to researchers. How might researchers draw on database software for organizing particular tweet "streams" (to use your terms)? Does this kind of archive invite new forms of collaboration? Does it require many researchers to work together to apprehend how a "topic" (for lack of better words) has moved through / taken shape within Twitter? What kinds of research projects might this archive generate that are not linked to particular fields of content at all but rather to habits of writing, communicating, networking, agitating, etc. that are inchoate in the early twenty-first century? (I am thinking especially of the Twitter campaign that asked all users to set their "location" to Iran as a means of disrupting the Iran government's use of Twitter to track dissent and protesters during the last election.)
Again, I appreciate your own musings on the value / interest of the LOC archive and your questions about my own (perhaps overstated) quarrel with Raymond!
One of the most challenging aspects of archival work is anticipating research trends, or even sensing that there is some content out there that needs to be fixed for future use, but not knowing exactly why. In the 1980s, my colleagues and I collected some wacky pamphlets in the Soviet Union with a frivolously anachronistic Russian flag in red, white, and blue. And cheesy home-made posters from Leningrad that said they came from St. Petersburg... if you looked at that emphemera seriously, it was actually easy to see that something new was brewing...and something big as it turned out.... The fringe is sometimes the cutting edge, and emphemera is sometimes the first place to spot what's coming at us. I'm with LC...I'm sure there are signals buried in all those tweets...since we don't know exactly what they are, the net has to be cast out there in a wide swath...to mix up a few metaphors... ED
Thank you, Elena. Your example of the ephemera collection of "wacky pamphlets" and other objects that you developed with colleagues is wonderful! I could not agree more that the fringe is sometimes the cutting edge, and I must say that, above all, I am delighted by the new archive of Twitter ephemera.