eBooks, Surveillance and a New Aesthetics?

(Source: Instagram user hotdudesreading,
from PopSugar)

Perhaps five year ago when, at hiring and committee meetings, readings, or during the rare non-literary performance (as in the crazy Raymond Roussel play John Beer directed with such aplomb in Chicago), I would pull out my iPad and read from it, my use of this technology would spark comments and queries. I seldom saw people reading from an iPad, let alone another e-Reader, in public spaces, including on the MTA subway trains, and even less so on the El, though I did spot a number of early adopters on my flights commuting between Jersey City and the Windy City. My status as a tablet/e-book harbinger, however, did not result from any innovative thinking on my part; it was the direct result of living with an app developer who had bought two machines to test his work on, and I ended up with the more basic one. There was little likelihood that I would have gotten one so quickly otherwise.

Nowadays almost everyone I know has a tablet computer or an e-Reader, and many have both. "Sent from my iPad" pops up at the bottom of perhaps every fourth email. As a present a few years ago I received an updated iPad from C, and I do use it but far less frequently than I once did, for reading, drawing, animation, or anything else. I have not purchased an e-book or downloaded a free one in a while. Instead if I'm reading anything online, I primarily turn to my laptop (which I recently had to have replaced, because the prior one's logic board fried), or my iPhone (which I had to replace because I left the previous one in a US post office near Chelsea Market, and when I rushed back to find it, it was, unsurprisingly, gone). One of my animations (cf. below) expressed in a few images my thoughts about the e-reading experience, which I find far less enjoyable than codex (printed) books, but for many people today, these handheld computers are becoming the default. (I found it telling that at least one recent survey of college students suggests they prefer print books, which tracks my classroom experiences, a small group of phone-centric student readers notwithstanding.)

Still, convenience, physiological comfort, and to some degree costs, make strong arguments on behalf of e-books and e-devices. Rather than having to carry about heavy hardcover or even lighter but still bulky paperback books, as I have done for most of my life, you can store hundreds (thousands?) of books on many e-reading devices, which have increasingly shrunk in their physical dimensions to the point that some are not much bigger than the largest smartphones. (Compare an iPad Mini or smaller Kindle reader to one of the larger Android phones, or Apple's news gigantic iPhone.) On most e-readers you can enlarge print with a swipe or a button, obviating the need, as your eyes age, for squinting at tiny print. E-books also tend to be less expensive (as they should be) than codex books, and with most--all?--e-reading devices you can also easily read .pdfs of anything you can transform into that format, which today is anything you find on a laptop or tablet screen. A vast array of literature can be found free of charge on the web, such that an enterprising person could--and some have--cobble together from existing materials an ever-changing, personalized e-anthology.

Recently, Francine Prose published a short article, "They're Watching You Read," in the New York Review of Books, in which she describes reading William Makepeace Thackeray on her e-reader, and while doing so, she came across an article in the Guardian, "Ebooks can tell which novels you didn't finish," that shares something many readers may be aware of but probably are not thinking about too much, which is that some e-devices are carefully tracking their readers' habits, down to where they break off in reading before resuming, which passages they reread, and how far they get in terms of a book's full length. (Do all e-reading programs do this?) According to e-bookseller Kobo's stats, "less than half of British readers" or 44.4%, of Donna Tartt's immense Pulitzer Prize-winning novel The Goldfinch, finished the 800-page tome. With Solomon Northup's shorter memoir Twelve Years a Slave, returned to public notice by Steve McQueen's Academy Award-garnering film, only 22.8% managed to complete it. On the the other hand, according to Kobo, the most completed book, at 83%, was self-published author Casey Kelleher's Rotten to the Core. Although it did not make the Kobo Bestseller list, it held four-fifths of its readers to the final period. Kelleher has gone to "win a book deal with Amazon's UK publishing imprint." In fact, as Kobo's stats show, its bestseller lists (books purchased) and most completed books (read to the end) are quite distinct, as are the national reading trends it tracked (though the percentages aren't broken down according to gender, age or other readership categories):

Kobo also revealed that the people of Britain were most likely to finish a romance novel, with 62% completion, followed by crime and thrillers (61%) and fantasy (60%). Italians were also most engaged by romance (74% completion), while the French preferred mysteries, with 70% completion.

I wonder what American readers tend to finish the most. Fantasy? Nonfiction books? Something not listed above?

Even raising such questions underlines the basic fact that the collection of data is a form of surveillance. Neither Prose nor the Guardian report says whether the data collection is voluntary, but I imagine it is similar to most online data collection today, which is to say, you have to opt out, sometimes with great difficulty. But then it isn't just online sites that are collecting data. A few years ago on these pages, I raised an uproar over the warrantless wiretapping George W. Bush's government was engaging in, but as Wikileaks and other whistleblowers, including most importantly, I'd argue, Edward Snowden have certified, the surveillance of everyone, via a variety of increasingly networked means, is widespread. To put it another way, every networked device, as well as every data point collected by any means possible, becomes the means to follow, track, quantify, and monetize our existences. Everything we do that can be tracked can be commoditized, among other things. The most sinister aspects of this panoptic society remain to be reckoned with.

But back to Kobo: as the Guardian report and Prose both point out, e-devices allow publishers to know not only which books readers are purchasing but how slowly or quickly we move through them, whether we finish, and where we bog down. Prose goes on to talk about how this knowledge about readers might become public, but then, assuming the perspective of the published writer that she is, wonders about whether publishers may start using these data when to shape the writing the publish. She writes:

Since Kobo is apparently sharing its data with publishers, writers (and their editors) could soon be facing meetings in which the marketing department informs them that 82 percent of readers lost interest in their memoir on page 272. And if they want to be published in the future, whatever happens on that page should never be repeated. 

Will authors be urged to write the sorts of books that the highest percentage of readers read to the end? Or shorter books? Are readers less likely to finish longer books? We’ll definitely know that. Will mystery writers be scolded (and perhaps dropped from their publishers’ lists) because a third of their fans didn’t even stick around long to enough to learn who committed the murder? Or, given the apparent lack of correlation between books that are bought and books that are finished, will this information ultimately fail to interest publishers, whose profits have, it seems, been ultimately unaffected by whether or not readers persevere to the final pages? 

Other than Kobo, it appears most publishers are not (yet) willing to share the full data they have been collecting, but an earlier article along these lines made me wonder about not only about how publishers and readers might respond, which is the ultimate focus of Prose's article, but rather, within a specular regime of surveillance of writers' aesthetic choices how writers ourselves might start to respond as the data haul increases and we have an ever clearer picture of how readers are moving through books. (This also may be of great interest to the big data researchers now active in literary studies.) Publishers, agents and editors already push fiction and nonfiction writers toward certain themes and topics, as well as towards certain styles of writing that have proved, at least by empirical standards, to result in books that readers will purchase, whether they read them all the way through or not. With this new data, analogous to how we modify our behavior (or don't) based on the expectation of widespread surveillance, might some of us start to make even more specific aesthetic decisions based on the data, aware that it's always being collected and will be shared with us?

As things already stand most authors, whose work draws upon the prevailing discourses, frameworks, ideas, and rules of the era in which we are writing, already grasp the parameters for readership and publication, but this is often intuitive, with some recourse to empiricism. For example, most writers know an 800-page novel is going to be harder to sell than one that's 400 pages, though the example of Tartt, Foster Wallace, Catton, and others suggests that it's not impossible. Also, most writers know that language that is too baroque, too ornate, too lyrical, to subtle, and so no, will repel most, though not all, readers off, although the example of Cormac McCarthy's The Road, to give one fairly recent example, suggests that the right writer can write in a prose as finely wrought as that found in the Bible or Shakespeare's plays, and still sell books. (And that McCarthy novel is, I would argue, significantly more simple in its language than an earlier book of his, like Blood Meridian.) Most writers know that certain genres, and the conventions embedded within them--romance, mystery,  horror, science fiction, speculative fiction and fantasy, etc.--sell far more than realist literary fiction, though there are authors who successfully, in aesthetic and market terms, blend and bend all of these genres.

If an emerging or even established fiction writer, to take one example, not only wants to get her work into print, but to have readers engage with it and finish it, however, awareness of these data, let alone the specific data points assembled in descriptive--or prescriptive and proscriptive form--might begin to function like a template or paint-by-numbers guide. How to write? Shorter sentences and paragraphs; stock characters (young, upper middle class, physically abled, white) with certain winning descriptors (the default "blond(e)" or "sun-kissed"); prose so simple and shorn of rhetorical devices a second grader could read it without breaking a sweat; rigid Aristotelianism in terms of form; faster pacing; muted, middle-of-the-road politics; boilerplate, normative sexualities; and so forth. You could call this "e-data aesthetics," I suppose, and in some cases a writer may feel it benefits her work aesthetically. For others, of course, it could prove problematic, on multiple levels. If you know readers in general do not like complex syntax, to much figuration, and complex characters, but you are drawn to write such things, will you change to please (and gain?) readers?

These days such questions may appear farfetched, but you need only look at the increasingly rigid, money-driven template Hollywood employs with regard to mainstream filmmaking to recognize that once enough writers proceed from such premises, it will change writing dramatically. It may help certain writers reach a level of competency, but it may harm the talents and aims of others. This is not to say that won't be other mechanisms, including forms of technology, social and cultural attitudes, as well as economic criteria, and so forth, at play, that determine what people are writing, but "e-data aesthetics" are something we might want to start thinking a bit more about, since the data are only growing and the means to aggregate them are steadily growing too.