Unforeseeable Sentences
Last month, professor and media theorist Jay Bolter wrote a short post called ‘Browsing culture’ that rebuts a refrain commonly heard from fans of physical books. He writes, ‘One of the criticisms one hears over and over about the digitization of the library is that it will eliminate the element of serendipity…Yet one wonders whether these critics have ever gone online to look for anything.’ He contends that the web, as indicated by its primary interface, is structured around browsing. Google, YouTube, Amazon, and various digital databases funnel us through never-ending rabbit holes of associated links.
I think about this topic a lot, as libraries have been a constant site of joyful, non-directed browsing since I was a child. My mother was an avid reader, and we spent hours wandering amidst the shelves of our local library. In college, I spent hours nestled in the stacks reading and writing. In grad school, I loved browsing the philosophy section with no particular author in mind, checking out books I’d never heard of. As a professor, I continue this practice, often finding new class resources by scanning the spines surrounding a book I’ve searched for online.
I also think about whether my students have this same experience after having search engines and Wikipedia on hand for most of their school-age lives. Anecdotally, I rarely see students leaving the library with books, nor even browsing the aisles. Our university’s first floor is now dominated by computers, conference rooms, an enormous whiteboard, and a Starbucks. I’ve had students buy books or DVDs for assignments when they were freely available at the library. My perspective may be slightly distorted since I teach in the Art Department, but my students rarely use the library as a primary source.
I understand Bolter’s perspective. Remediation, a book I adore and one that he co-authored, reminds us that digital media is novel not so much in its formal configurations or content, but in the way it re-presents old media: ‘What is new about new media comes from the particular ways in which they refashion older media and the ways in which older media refashion themselves to answer the challenges of new media.’ I think Bolter is saying that search engines, digital libraries, and online stores remediate the physical library, that ‘for better or worse, serendipity is a fundamental feature’ of their design. The access, process, and form may have changed, but browsing is central part of our online experience.
The problem with Bolter’s perspective is that it conflates serendipity with algorithm. The content portals he groups together have different functions that ultimately drive toward similar aims. Google is trying to surface meaningful (and personalized) search results, YouTube is trying to keep you watching videos, and Amazon is trying to keep you adding items to your shopping cart. Each uses sophisticated and proprietary recommendation algorithms to ensure a frictionless consumer experience, one meant to keep you buying products or watching advertisements. And there’s the key difference: while serendipity is agent-driven, subject to chance and environment, the recommendation algorithm is product-driven, subject to computation and capital.
Of course libraries are not sites of disordered accumulation (second-hand bookstores fill that role). The Dewey decimal systems imposes some ‘algorithmic’ consistency to a library’s collection, assuring that any serendipitous wandering into the film studies section will not be truly driven by chance. Likewise for the themed displays that libraries erect to attract customers to certain parts of their collection. Or even the librarians who are happy to serve as an ad hoc recommendation engine. But rarely do these groupings or classifications mask an underlying profit motive. I imagine a publishing rep has at some point wandered a university library murmuring, ‘Customers who liked The Lord of the Rings also like The Wizard of Earthsea,’ but I also imagine they were quietly escorted out.
And presuming that Google, Amazon, or YouTube one day distill serendipity to a pristine algorithmic experience, we still lack a compelling interface with which to browse, a digital remediation of our hands, feet, and eyes. Let’s even strip away all the other attributes that make library books compelling—typography, the texture of pages, marginal scrawls, date stamps, covers, the bustle of imperfect silence, the feeling of reading among other readers—and reduce books to uniform spines set in identical type listing their names. I can still browse dozens of titles at a glance, scanning in rows, columns, or haphazardly, far faster than I can browse through Amazon, whose entire interface is designed to distract and entice. And when I pull that book from the shelf and flip through its pages, I have a fundamentally different experience than clicking the ‘Look inside’ logo and browsing the pages the publisher has granted Amazon the privilege to reproduce.
I do not mean to romanticize books or dispense with digital tools. My library’s online database is indispensable, Amazon makes for a handy research partner, and YouTube occupies far more of my viewing time than television. But the browsing culture promised by the early web, when hypertext unquestionably changed our linear experience of text and other media, is not the culture that most corporations seek. Amazon wants to provide you the best possible means to search within Amazon. Facebook wants your news feed to be so comprehensive and personalized that you’ll never leave their confines. YouTube keeps tabs on your viewing habits to ensure you’re seeing the most relevant ads. And even Google, who ostensibly scrape the entire web with their search engine, prefer you log in to your Google account while searching so they can subtly sculpt your search results over time. None of these companies are remediated libraries—they’re a remediated concierge, and you’re not their employer.
In the culture of the recommendation algorithm, serendipitous browsing is not only a pleasurable experience but also a political act. Michel de Certeau said it best in The Practice of Everyday Life:
In the technocratically constructed, written, and functionalized space in which the consumers move about, their trajectories form unforeseeable sentences, partly unreadable paths across a space…[T]he trajectories trace out the ruses of other interests and desires that are neither determined nor captured by the systems in which they develop.
Browsing is the unforeseeable sentence we write as we wander the library aisles, pass our fingers over countless bound voices, and choose.