Until recently I hadn’t read Sven Birkerts’s 1993 book The Gutenberg Elegies, despite it being about a topic that I am fairly invested in: “the Fate of Reading in an Electronic Age,” as the subtitle has it. I had a vague impression from my graduate school days that the book was reactionary and anti-theoretical, the sort of work that complains about postmodernist theorists as though they were advocates rather than diagnosticians. I assumed it was full of paeans to reading as a sacred, arcane art that proved that the souls of certain people (literature undergraduates, humanist professors) were deeper and richer than the addled dupes who watched television. Later on, I saw it as an Ur-text for critics who wanted to insist that the internet was making people stupid and that phones and screens were emptying people of their capacity to have “real” conversations or “authentic” personalities.
This impression was not too far off. The Gutenberg Elegies does indeed contain a lot doom mongering about how “literature” is dying and kids don’t read books or know how to pay “deep” attention anymore, and it's also shot through with autobiographical passages about Birkerts’s own adolescent reading practices that read to me like inadvertent cautionary tales. But its arguments for what in print culture should be defended and preserved from contamination by screen culture — a struggle he dubs “the reading wars” — remains influential and worth examining more closely.
Birkerts is preoccupied with the materiality of text to the point that he suggests that using a word processor instead of a typewriter makes one less capable of writing a good sentence. And while this prejudice frequently veers into a full-on fetishization of print, it also sustains a focus on how the formal conditions of communication shape not only the content of what is communicated (à la McLuhan's "medium is the message") but also the relationships between the parties communicating. Reading War and Peace on your phone is different from listening to it in your car is different from reading a Penguin paperback. The experience is not just different for you, but it also posits a Tolstoy with a different set of capacities and aims. Similarly, printing out all your email would suggest a different set of intentions and responses for all parties involved. How text is circulating, what sort of permanence it has, how it is indexed, how transferable it is — all this changes every aspect of the relations between readers, writers, and texts.
Birkerts's concern with this ever-shifting context leads him to some bathetic proclamations — for instance, he likens hypertext’s interactive permissiveness to Nietzsche’s death of God — but it also prompts some interesting speculation about how different forms of media structure different sorts of subjectivity, sociality, and emotionality. But all of his insights are frustratingly abstracted from political economy. He rarely connects the evolution of media or technology with capitalism’s incentives, so he ends up defending the act of reading against ghosts and straw men who seem perversely determined to destroy it for no apparent reason. Birkerts ends up seeming angry at the future itself rather than the forces that have circumscribed it.
So throughout The Gutenberg Elegies, there are eloquent but intrinsically limited observations like this reflection on the difference between reading on the page and on a screen:
The page is flat, opaque. The screen is of indeterminate depth — the word floats on the surface like a leaf on a river. Phenomenologically, that word is less absolute. The leaf on the river is not the leaf plucked out and held in the hand. The words that appear and disappear on the screen are naturally perceived less as isolated counters and more as the constituent elements of some larger, more fluid process.
Has the fact that I typed that out on a computer and you are now reading it on a screen made those leaves more barren? Have they slipped your grasp? Have you already opened your Twitter app and forgotten all about Sven Birkerts? So many questions without answers. For a more down-to-earth assessment of the transition from print to screen, we can turn to “The TikTok War,” business analyst Ben Thompson's recent post about the app being potentially banned. Here the shift in medium is understood in terms of what sort of advertising each enables: “It is always tricky to look at the analog world if you are trying to understand the digital one,” he notes. “When it comes to designing products, a pattern you see repeatedly is copying what came before, poorly, and only later creating something native to the medium.” Already we are worlds away from what Birkerts is willing to notice — that under our socioeconomic conditions, text, even precious literature, is a “product,” and it is designed in particular ways to circulate in markets. The distinctive feature of a medium is primarily what can differentiate it in the broader market for information, entertainment, experience, connection — whatever cultural demand it is trying to meet or create. What is different between print and screen begins with the differences in how they are sold, not with some abstract phenomenological differences about how the letters are perceived.
Those phenomenological differences may seem fixed, but they are malleable: The experience of using a medium can be radically changed through design emphases. Thompson claims that “while mediums change, humans remain the same,” but that is overessentializing: It is more that mediums and audiences are in a state of dialectical interplay, mutually changing each other in response to the overriding and organizing economic incentives and outcomes. Thompson argues that content on mobile screens didn’t thrive and take on its distinctive forms until new modes of monetization were worked out: “Consider text: given that newspapers monetized by placing advertisements next to news stories, the first websites tried to monetize by — you guessed it — placing advertisements next to news stories. This worked, but not particularly well; publishers talked about print dollars and digital dimes, and later mobile pennies … What changed was the feed, something uniquely enabled by digital. Whereas a newspaper had to be defined up-front, such that it could be printed and distributed at scale, a feed is tailored to the individual in real-time — and so are the advertisements. Suddenly it was print that was worth pennies, while the Internet generally and mobile especially were worth more than newspapers ever were.”
This highlights in far plainer terms the nature of the problem with text on screens: the feed, or what Raymond Williams in his work on television called “flow.” This is not intrinsic to screens or networks, but emerged from their potential when it proved profitable. The feed could be seen as a proactive attempt by a broadcaster (or a platform) to control an audience’s attention, to be able to direct or at least disengage someone’s focus according to rhythms the broadcaster controls. The rhythms of doomscrolling and refreshing are different from those of channel flipping and commercial breaks, but they still impose themselves on the thought process of the viewer as a form of external control, demanding a certain acquiescence.
But that can lead to seeing screens as destroying attention and focus rather than constituting it. The point is that different mediums have different economic potentials that lead to different ways of structuring audiences. It seems unreasonable to me to say that “advertising revenue is a reasonable proxy for attention,” as Thompson assumes. Rather ad revenue is a means for constituting attention and focus in ways that have been found to be profitable. Advertising doesn't follow where human attention naturally wants to go; rather ad money restructures human attention so that it can be more efficiently captured and measured.
Reading print doesn’t necessarily involve the same external imposition of temporal control that we associate with watching screens. For there to be a rhythm to reading print, the reader must work to sustain it or deliberately surrender to it. Birkerts celebrates this as a unique form of character building. His core belief is that reading printed books is necessary for developing both interiority and a sense of autonomy over one’s attention span. It produces a “specialized and self-directed inwardness,” he claims, a “warm and safe environment, one that I had complete control over.” Getting lost in a book is an accomplishment of internal focus rather than a hijacking by way of intrusive images, jump cuts, new distractions being forced in. It invests us in a particular illusion of our autonomy that is rooted in sustained focus. (Being mesmerized by a slot machine is maybe not so different than being lost in a page-turner that you can’t put down.) This immersiveness allows Birkerts to claim that that books provide “ a place I can repair to to release the private, unsocialized, dreaming self. A book is solitude, privacy.” It is as if there is no question of being persuaded into new or strange beliefs or practices by the content of a book. The content is just a pretext for self-involvement.
The effort of immersion, in Birkerts's view, creates a depth for the self apart from the idiocy of everyday life. “Devoted readers,” Birkerts argues, “formed a good part of their essential selves through interaction with books. That is, they somehow founded their own inwardness, the more reflective component of their self, in the space that reading opened up.” Implicit in this is that less devoted readers are relatively devoid of the capacity to reflect, and inhabit a barren and diminished inward space. “The time of reading, the time defined by the author’s language resonating in the self, is not the world’s time, but the soul’s,” he writes. “I don’t know how else to define the soul in secular terms except as a kind of self-consistent condensation of self. Reading makes this self more present.” Illiterate people are presumably without souls.
But at the same time, Birkerts conflates this sense of control with the feeling of being completely cared for. Responding to a Robert Coover essay in which he claimed that hypertext tended toward “freeing the reader from domination by the author,” Birkerts responds that submitting to such domination is “the point of writing and reading.” Like BDSM, it can be understood as expressing control through the choice of conditions under which one relinquishes it, a surrender to a constructed helplessness rather than an involuntary one. Reading, Birkerts insists, “offers readers something they want: a chance to subject the anarchic subjectivity to another’s disciplined imagination.”
This is a good way to understand algorithmic recommendation as well. TikTok’s For You page, for instance, takes your “anarchic subjectivity” and evens it out into a coherent experience. It claims to anticipate your needs while centering the authority of your tastes, even as it removes control over what you see, forcing you to submit. When reading is engrossing (as when algorithmic feeds are compulsively compelling), it can feel as though the reading material is opening us up even as it is dimming and narrowing our focus. Birkerts insists this narrowing is actually a scooping out of interior depth. But it may be that interiority itself — the sense that a book or feed is “for you” — becomes an alibi for how attention can be channeled, manipulated, externally controlled.
Unlike print, the feed is characterized not by a depth immersion but by routinized distraction that posits an immersion that you just lost, retroactively. Another way of putting that is that the feed immerses us in the experience of continual novelty, so that it requires continual distraction to make us believe that we had been "really" paying attention — a feeling we know only through its loss, which we come to enjoy as the thing itself. The feed teaches us to enjoy attention as already spent rather than actively absorbed. We come to prefer scrolling to landing.
But it seems wrong to assume that this distractibility can’t also a ground a sense of self. Birkerts states categorically that screens are “antithetical to inwardness.” This is because, in his view, “inward experience, including all aesthetic experience, unfolds in one kind of time; electronic communications, of their very nature, depend upon — indeed create — another. The time of the self is deep time, duration time, time that is characterized by our obliviousness to it … All circuit-driven communications, by contrast, are predicated on instantaneousness.”
That seems persuasive enough, but instantaneousness also can produce an obliviousness to time, can also be the “time of the self,” unfolding in a series of novel stimuli organized for you personally. Almost everybody knows what it is like to lose track of time scrolling through a feed. And the “place” we occupy when this happens can also be conceived as an inwardness, can also feel deeply personal and private even though it depends on our being placed under constant and intimate surveillance.
What feeds don’t do is promise an experience of self that is “unsocialized.” When our “anarchic subjectivity” is disciplined and tamed by algorithms rather than an author, there is not necessarily a corresponding sense that we’ve escaped the demands of other people and elevated ourselves above the everyday concerns of the soulless nonliterary people. Rather, our feeds may be filled with a cacophony of other voices that have been harmonized around the algorithms’ concept of our self — of what we want from other people, of what we need to hear, of who we’d like to be like or what we’ll want to condemn. Consuming feeds creates interiority as an explicit space of socialization, albeit without the direct presence of others or the requirement of reciprocity.
Reading print doesn’t grant readers autonomy; it instead fosters a dependency on how it makes us feel autonomous, in control of how we are being controlled. After all, we really want to believe our autonomy is autonomous. The interiority facilitated by algorithmic sorting is not about a sense of independence or exemption. It’s not oriented toward a transcendent aloofness — toward feeling unique and superior because of the idiosyncratic nature of your escapism. It instead implants social dynamics like mimetic desire and status seeking in that interior space, more practical forms of the vicariousness and didacticism that has always characterized reading fiction. It tells you that you are special not because you can reject other people for a book, but because the algorithmic proxies for other people are making themselves at home within you and feel comfortable there.