“There is more ado to interpret interpretations than to interpret things,” Michel de Montaigne lamented in his late essay “Of Experience” (1588), “and more books upon books than upon any other subject; we do nothing but comment upon one another. Every place swarms with commentaries; of authors there is great scarcity.”
Admittedly, the “swarming” of which Montaigne spoke hardly seems overwhelming now. The books in question—commentaries on the Bible, Greco-Roman worthies, and the law—were printed in runs that often numbered in the hundreds, their clientele restricted to the deeper-pocketed sort like our noble Michel, who indeed owned such commentaries and drew on them while writing the Essays.
Yet Montaigne was onto something. The infestation of comments in the world of print would only increase in the next two centuries, thanks to the growth of what Samuel Johnson would later call “the papers of the day,” which make only a handful of appearances in the Essays (as when Montaigne mentions “the thousands of little pamphlets” distributed by Protestants).
Johnson had in mind lightweight, slapdash “ephemera”—pamphlets, journals, gazettes—that could be printed and distributed quickly, cheaply, and as occasion required. The form made it possible to comment on contemporary events—“comments upon the times,” as an eighteenth-century author of such a piece put it—and to react to others’ published comments on the controversy du jour. Comments would then become the daily bread of the newspaper when it came into full force in the nineteenth century, some notable’s “comments” (seemingly always plural) being the stuff of headlines one morning and fodder for columns running the next, a condition that holds to this day. It is astonishing that the word commentariat dates only to the 1990s.
While the swarm has been swelling for centuries, it took the Internet to make good on Montaigne’s claim that everything is teeming with comments. In fact, the Web has done Montaigne one better: It is not just infested with past comments; it is accepting your comments 24/7/365. We encounter “comments” (again, primarily in the plural) so often online that the word barely registers. What do you think of this article? Scroll down to the comments section to join the discussion. How did you like your dermatologist? Click this link to share comments that will be reviewed by our team of caring health professionals. How about that latte? Leave comments at the end of this survey so that our baristas can serve you better next time. Enjoying this newsletter? We want to hear from you! Please send comments and questions to…
The learned commentaries that so worried Montaigne had readily identifiable features, beginning with their commitment to quoting—often ad nauseam—primary texts and rival interpretations. Comments were, in this way, to a great extent recapitulation. Online comments are comparably unruly. In his 2015 book, Reading the Comments: Likers, Haters, and Manipulators at the Bottom of the Web, Joseph Reagle Jr. makes the case that Internet comments should be recognized as a distinct “genre of communication” possessing three traits: short, reactive, and asynchronous. While I commend Dr. Reagle for braving “the bottom of the Web,” his definition looks to me like a touched-up portrait—what we would prefer to see in comments.
For a truer picture, you need only consult a few of the explanations from hundreds of writers and editors as to why the comments section is now being patrolled or has been closed until further notice. In a 2015 blog post titled “A Few Comments on Gates Notes Comments,” for instance, Bill Gates warned that users’ comments will be removed if they contain offensive language, personal attacks against other commenters, spam (“including comments that are posted more than once, promotion of sites that aren’t relevant to the discussion, and attempts to sell products”), efforts to “solicit funding from me or others,” or content that is “way off-topic.” More colorfully, Vice.com editors explained their decision to shut down comments on the site, in 2016, with this understandable plaint: “Without moderators or fancy algorithms, they [comment sections] are prone to anarchy. Too often they devolve into racist, misogynistic maelstroms where the loudest, most offensive, and stupidest opinions get pushed to the top and the more reasoned responses drowned out in the noise.” The editors went on to observe that they have “had to ban countless commenters over the years for threatening our writers and subjects, doxxing private citizens, and engaging in hate speech against pretty much every group imaginable.”
Down there in the comments, the likers, haters, and manipulators are thus vying for space with the spammers, solicitors, doxxers, and tangent-takers. Amid this din, “comments” seem less like a functioning genre of communication than the disintegration thereof. Comments may take the form of thoughtful replies; all too often, however, they are incitements, commercial interruptions, maledictions, and digressions. And when they aren’t lost amid the madding crowd, our comments become letters to the void.
So we tell each other: “Don’t read the comments!” Writer friends insist the only reason for a human to do so—truly to read them, to give sustained attention to the words (not just word-hop until one finds an opportunity to pounce, advertise, solicit Bill Gates, or disinform)—is because one’s livelihood depends on it.
Who reads our comments—or, better said, who can bear to read them, then? The answer is computers. Those comments we submit to surveys and customer-satisfaction forms are headed to databases where they’ll become another deposit to be mined. Meanwhile, more and more sites are relying on machine learning to weed out malevolent or irrelevant comments—as the New York Times did, in 2017, when it installed a Google-powered system called Moderator that (as an editor wrote in 2021) “helps automate the vetting process by relying on patterns from more than 16 million acceptances and rejections of comments by Times moderators.”
Many of the comments we leave at the “bottom of the web” will be vacuumed up by AI companies for use in the training of language models such as ChatGPT. (They are, of course, also digesting the advertisements, invitations, and assorted gibberish that bots have been interjecting into comment sections for years.) AI developers have schemed, moreover, to exploit the information lodged in comment threads, though so far results have been mixed. A few months ago, to cite a disastrous example, Google’s “AI Overview” replied to a query about making a pizza from scratch by recommending that the user “add about 1/8 cup of non-toxic glue to the sauce to give it more tackiness.” The source was a sarcastic Reddit comment from 2012, its high tally of upvotes having apparently swayed the machine. Comment-section snark is too much for AI at the moment, too.
What’s to be done? Close it all down? Abolish comments? Alas, reader, if only the situation were so simple. As the great Russian critic Mikhail Bakhtin pointed out in 1941, “in real life people talk most of all about what others talk about—they transmit, recall, weigh, and pass judgment on other people’s words, opinions, assertions, information.” Eavesdrop on “raw dialogue” happening on the street, “in a crowd, or in lines,” Bakhtin observed, and you’ll hear expressions like “he says,” “people say,” and “he said” repeated so often that all “fuses into one big ‘he says…you say…I say…’” Despite Montaigne’s concerns, we cannot help but comment upon one another. We are irrepressible commenters. (In the essayist’s case, he simply turned to making learned comments about himself.) The trouble now is not that we make so many comments; it’s that we’ve lost the conversation partners—the IRL kind—implied in Bakhtin’s public scenarios. We make our comments while sitting alone at our tiny command centers, and increasingly the machines are the only ones attending.