Of happiness and despair we have no measure

In Keywords, Raymond Williams famously proclaimed culture to be “one of the two or three most complicated words in the English language.” His attempt to disentangle the different meanings of “popular” suggests some of the same complexities:

Popular culture was not identified by the people but by others, and it still carries two older senses: inferior kinds of work (cf. popular literature, popular press as distinguished from quality press); and work deliberately setting out to win favor (popular journalism as distinguished from democratic journalism, or popular entertainment); as well as the more modern sense of well-liked by many people, with which of course, in many cases, the earlier senses overlap. The sense of popular culture as the culture actually made by people for themselves is different from all these … The range of senses can be seen again in popularize, which until C19 was a political term, in the old sense, and then took on its special meaning of presenting knowledge in generally accessible ways. Its C19 uses were mainly favorable, and in C20 the favorable sense is still available, but there is also a strong sense of ‘simplification’, which in some circles is predominant.

In other words, it is not necessarily self-evident what someone means in evoking “popularity” and how it should appropriately be determined. For instance, in this recent post, after assessing a range of online-famous YouTubers, meme stars, and would-be main characters, Ryan Broderick wonders whether “virality is decoupling from popularity.” That initially struck me as fairly cryptic, since at first glance those words seem synonymous. His point presumably is to tease out an emerging distinction between the two, but the post doesn’t offer specific definitions of “viral” or “popular,” or make the case that they were ever actually “coupled.”

It seems easier upon reflection to make the opposite case, that “virality” has always connoted a kind of illegitimate popularity secured through hacking algorithmic systems and exploiting the ludicrous scale of social media platforms. Virality is engineered through media systems that have been developed to manufacture rather than reflect consent, to produce the fads demanded by the logic of captive audiences. It does not indicate “real” popularity. (You might call it “coordinated inauthentic behavior.”)

This would explain why internet culture remains somewhat marginal. It doesn’t reflect the zeitgeist so much as loopholes in interface and algorithmic design. “We know MrBeast is a big deal,” Broderick claims. (Do “we”?) “Though, it’s still a little hard to actually feel it irl.” (Isn’t that a small blessing? Who would want to feel that?) What Broderick is noting is that MrBeast is viral without being “actually popular.” He has amassed aggregate online metrics, but these don’t reflect consensus or cultural significance. Like any other fad, he testifies mainly to the significance of fads in general and not the significance of anything he specifically does. (By the same token, fashion trends only signify that trends happen, serving to indicate the vitality of the culture industry branch that exists to produce them and not anything about the public mood.)

Online metrics are also intrinsically unreliable, as the periodic anxiety about “fake” clicks, likes, and followers suggests. Generative models, this Axios roundup notes, will serve to intensify this anxiety; it cites a “cybersecurity” executive who asserts that “bots are used to manipulate public opinion by rigging the virality algorithm.” Which algorithms are unrigged, though? Every “virality algorithm” is inherently a position on what should be popular and what should be shadowbanned.

The Axios item claims that “bots that simulate human engagement online” are “making it tougher to use social media chatter as a bellwether for public opinion,” but why should “social media chatter” be taken for public opinion in the first place? It is not as though social media platforms manifest an ideal speech situation or a Habermasian public sphere; they are machines engineered to produce audiences with certain characteristics, not capture their given proclivities. Bots, for that matter, are no different from any other kind of PR, which also simulates human engagement and the semblance of a “public opinion” where there isn’t any.

Public opinion, like virality, is made and not found; manipulation brings it into being. Virality reflects the same process, only it produces a public without an opinion — it refers to things known for being known, things that circulate through inertia, “trending.” What does it mean to have an opinion about Hawk Tuah? Broderick asks, “Do we just know her name and see her on our screens?” Yes, that’s all it is. There is always a Hawk Tuah in some guise or other at any given time, and it never means anything more than attention is being measured.

The momentum of memes seems like it has to mean something, seems like it has to directly correlate with “the circulation of social energy” (to borrow from Stephen Greenblatt) in some decipherable way, and if you devote enough time to tracking YouTube celebrities and TikTok ephemera you will eventually crack the code and become rich. That may be so, but resolving “social energy” into profitability seems like a means of obviating culture rather than understanding it. “Money is only one kind of cultural capital,” Greenblatt argues. Social media metrics are only another kind.

For some reason, Broderick seems to believe that millennials were the first generation to confront these media popularity paradoxes: “the very idea that mass appeal had to be accurately reflected back at us online and vice versa was an entirely millennial idea. A neurotic need to know, and quantify, exactly what everyone else was seeing and doing.” That is true only in the sense that millennials were “online” and previous generations weren’t. But every culture looks to understand itself in some way, to locate what is essential amid the churn of economically driven trends and novelties.

“Viral” vs. “popular” seems like another version of the mid-20th century concern with “mass culture” and whether it could be conflated with a “true” popular culture or folk culture or organic culture. And in a larger sense, these kinds of paradoxes inhere in the concept of representation in general, in contrasts between media and “real life,” if not between thought and things in themselves. Is geist real? Can you measure it, or is its existence predicated on being unobserved, a silent ceaseless weaving until it emerges fully formed as a “vibe”?

The poisoning implicit in the term “virality” aligns it with the old assumptions about mass culture, which reflected an industrial demand for audience conformity and submission. Essay after essay in a collection like Mass Culture: The Popular Arts in America (1957) hammers home how the masses are brainwashed into believing they enjoy what they are forced to accept as compensation for their deskilled and preprogrammed lives. In an essay on “The Public Arts,” Gilbert Seldes writes that

The social reverberation produced when millions of people follow the same entertainment or receive the same communication at one time is something different from the imitation of a royal mistress’s hairdo — the diffusion is immeasurably greater, the penetration deeper. The physical reduplication of comic books and phonograph records, the velocity of radio and television, the availability of the motion-picture film, and the way the various entertainments support one another create another kind of contagion: the public mind is crammed with details about them, so that the true significance of “the mass media” becomes, not their appeal to the mass audience, but their own dimensions, the size and weight and speed and force that the mass media possess.

One could add social media to the list of mass-produced “contagion” without complicated any of the rest of its claims. The true significance of what Broderick calls “online gutter culture” is not in the particular nature of its appeals to a mass audience but what it demonstrates about the size and power of tech companies.