Metametastasis

The news hasn't really been about the news for quite some time.

Most of us learned this when we couldn't help noticing how much news coverage was not about any event but about how people were responding to it. Yes, I mean the clickbait articles from engagement-farming outlets that cherry-pick some 100-follower Twitter account to conveniently represent the views of half of America, but I don't just mean that. I'm writing this at dawn on Sunday, August 11th, and these are your top fold articles on Fox News and the New York Times. If you want to know what happened this weekend, you'll have to do some digging. If you want to know what crucial swing state voters, retired cops, Senate democrats, black men, Trump fans and a maturing American subculture think about those events, however, America's largest news organizations stand ready to serve.

It isn't as if the reactions of certain people aren't sometimes newsworthy on their own. Especially in a hotly contested election with only a few states legitimately up for grabs, changes in the polled behaviors of swing state voters are obviously worthy of coverage. Yet for the most part, these types of articles are assigned, written, and designed with the self-evident purpose of telling you how other people have chosen to think about something. It doesn't take very long to realize that those "other people" are usually distributed into two clearly demarcated groups. The first are the wise, intelligent, virtuous, and thoughtful citizens who look and act very much like you, dear reader. The second are the dastardly, ignorant, foolish, and dare I say treasonous cretins who do not resemble you at all. It is our great fortune that the typical demographic traits and tribal membership of any publication's usual audience nearly always align with those of the admirable group. Those who would not dare to fill their minds with its propaganda invariably hail from one basket of deplorables or other.

This is part of a broader transition from news to fiat news that has taken place gradually over the last several decades and suddenly over the last fifteen or so since social networks emerged. Fiat news is coverage which presents opinion as fact. The goal of such coverage is not to give you facts about events and the world but to tell you how to think about them. Telling you what smart, virtuous people and vapid, sinister people think about the world is one of the most powerful ways to guide you how to do so. When we believe that everybody else believes something - especially those whose opinions we are inclined to find credible - it will affect how we think. There's a reason why every publication produced by people who want to change the world does this now: it works.

But again, this isn't new. For at least 15 years, like clockwork, every news story of any significance has been at first a news story, then a story we have to be carefully taught to understand correctly, then a story in which we have been told how good people and bad people are reacting to it.

But something is changing in this pattern.

The meta is metastasizing.

Around noon on August 10th, the above image made the rounds on social media - first on Twitter, then on Reddit, Instagram and Facebook. It was the creation of a medium-sized and obvious parody Twitter account. The account was a pretty typical right-wing parody of left-of-center "pronouns in bio" accounts, complete with the obligatory Ukraine and LGBT rainbow flags. It is executed cleverly enough, if pretty much in line with a few dozen other accounts with a similar gimmick. The image created is an intentionally obvious and goofy example of the silliness some AI image generation packages tend to produce when rendering groups of humans. There are extra fingers, extra limbs, odd contortions, silly lettering - all the telltale signs of a late-2023 AI-generated image. The intent of the parody is to play with the idea that the Harris campaign is fudging their crowd numbers and doing a lousy and amateurish job of it. It's good parody.

But it got caught up in its own wake.

There is some historical context here. An important part of former President Trump's narrative among his supporters has long been that the media intentionally suppressed his popularity. Trump himself has happily promoted this view. On many occasions, he frequently exaggerated the size of the crowds at his rallies and loudly complained that the fake news media purposefully ignored them. After his loss in the 2020 election, one of the most common refrains from the Stop the Steal crowd was the belief that the outcome was impossible given the relative infrequency and lack of energy at Biden campaign events. I think it's provably true that Trump regularly exaggerated the size of certain of his events. I also think that we could say the same of any event promoter in history, even if Trump did so in more flamboyant fashion than most. I also think that it is harder to prove the claim of media suppression in its coverage, but it doesn't seem unreasonable. Trump was historically unpopular among western journalists. I think it's very likely that all three of these things are true at once, but what is important is that the belief in this created a context layer for the way the socially networked hive mind processes the meaning of events about the 2024 campaign.

There is more immediate context, too. A related narrative has emerged among Trump's supporters that the news media have wittingly or unwittingly made themselves partners in the Harris campaign's attempts to make her seem more popular and well-liked than the reputation that precedes her. This point of view did not spring from nothing. Since the earliest days of the Harris campaign, if there has been a line between news pages and opinion, it has been almost imperceptible.

Add in the context of continued widespread belief among Trump voters that the 2020 election was stolen, and you have a pretty good recipe for the zeitgeist. Both sides are working feverishly to create common knowledge of energy and excitement around their campaign. For the same reason sitcom producers always give you a laugh track, political campaigns want to give you videos, images, and sizzle reels of excited superfans experiencing joy, rapture, and rage about what Harris-Walz or Trump-Vance will do for them or to them over the next four years. The three weeks since we memory-holed a literal assassination attempt have been filled with battles over the authenticity, tricks, deception, and claims made about these speech audiences and rally crowds. Images from different events are being portrayed as being from the same one. Photographic tricks are hiding closed-off sections. On the off chance that you are tempted to interpret this article as at all implying that these kinds of shenanigans are above anyone, yes: AI, Photoshop, and other image-editing tools are already being put to good use to emphasize and deemphasize certain features of these events. If you are reluctant to believe that both campaigns have, are, and will use such tools during the next three months, I am really not sure what to tell you.

Now, I know it seems trivial and childish for the campaigns and surrogates to sit here bickering, arguing, lying, and insulting each other over how many people were at a rally. It is trivial and childish. It is also wildly important. The power of the crowd watching the crowd is how the Chinese Communist Party maintains control. It is how social networks create a similar panopticon in otherwise free societies. It is why Joe Biden isn't running for President any more. A narrative of virality is virality. This is the context into which that (obvious) parody image was brought into the conversation at around noon on Saturday. It began to go viral almost immediately, generating millions of views after its promotion by several high-follower accounts across Twitter, TikTok, Reddit, and other social media platforms.

Only a couple hours later, perhaps because the topic was hitting a nerve, another influencer introduced a similar claim. In this case, the writer argued that the Harris-Walz rally at the Detroit airport earlier that week was fabricated, too. The difference was that in this case, the event in question actually happened, and the photos were among those released by press and the campaign. As this post made its way around the world, the tale grew in the telling. The crowd should be reflected in various places on the plane's metal exterior, you see! There is a missing tail number! Citizens aren't allowed on the tarmac, you ignorant sheep!

By Sunday we would have the fully mediated meaning of this new symbol delivered to us from on high.

There's just one problem: the Detroit event absolutely happened. Not only did it absolutely happen, the evidence of it happening pretty much precisely as the purportedly AI-generated photos show it happening is extraordinary in its depth, non-partisan sourcing, and the ease with which it can be accessed. The event was livestreamed through multiple channels. There are camera phone videos from multiple vantage points and perspectives. There are feature articles in multiple publications with photographs from credited and known photojournalists. Like a lot of marketing collateral, a couple of the rather more, um, stylized pictures have definitely been run through the filter mill, but the pictures in question more or less authentically represent this event. Most importantly, the event was already a few days old, so by Saturday afternoon the underlying information was almost embarrassingly accessible.

I think the easy answer would be to call this a case of tribally influenced confirmation bias and move on. It would also be the wrong answer. That isn't what's happening here. People didn't come into this news event with a set of prior beliefs and devalue new information which disagreed with those beliefs. They skipped over the process of evaluating truth at all. If you explored the various threads about the Detroit event, you'd find hundreds of meticulously analyzed versions of that original parody post - look at these hands, fingers, and faces! It didn't matter that it was an entirely different photo posted by a parody account and never purported to be even remotely related to this event. In the place of a process for evaluating the truth of an event, confirmation biased or not, we chose the more seductive, more abstracted, more deeply symbolic context layer to determine its truthiness relative to our chosen worldview. This is not really about an event or a photo or whether it was AI-generated so much as it shows that Democrats like Kamala Harris are deceptive and will cheat (again!) to win. It is really about how the Harris campaign desperately wants to create the perception of the energy of a Trump rally.

Metametastasis is when it is the context of the thing - not the thing itself - that goes viral and transforms into new common knowledge.

Metametastasis is when no amount of fundamental truth is clear enough, no dose of reality strong enough to offset what we consider to be true at higher, more symbolic levels of abstraction.

Metametastasis is when it no longer matters whether a record of an event is authentic, at least to its virality or spread. The only thing that matters is how it fits the context, or how it can be slotted to some underlying truth we hold.

It's got other people worried, too.

Weinstein isn't saying he thinks the picture is or isn't fake. He's saying that we're gazing into the abyss of a major epistemic crisis. He's saying that we aren't ready yet to deal with a world in which we don't know what is real and what isn't. I think Weinstein's heart is in the right place. I think his thinking isn't too far behind. And I think he's a bit off base on three counts.

The first disagreement is pedantic, since he's trying to make a (fair) general point and I'm making a specific one. We absolutely did not have to be there to know in any practical sense. There are enough sources, enough human bodies and human institutions with names and reputations attached to the materials testifying to the basic facts of the event to know what happened as well as we know just about anything these days. That won't always be true, of course, which is why his general point still stands.

The second disagreement is more meaningful. If you thought this was a simple case of confirmation bias, you would be worried rightly about the epistemic problem. If it is becoming harder to know that anything is real, it might stand to reason that we would all simply revert to the things we think are more likely to be true. If we feel that our political opponents are cheats and liars, the uncertainty might just exacerbate our tendency to go with the feeling. That is obviously happening, but I don't think that's what happened here. There isn't a problem with how we seek to know. The problem is that we no longer have any reason to seek to know.

Metametastasis cannot be resolved by improving our access to the authenticity or truth of a thing. Maybe we could lean on blockchain-based or other methods to authenticate the origin of videos and images. Maybe we could improve AI detection algorithms. Maybe we could teach a generation of young Americans how to be skeptical again. None of it would help. Not really. Not with this. Every bit of that information was available in this case at trivial cost and with minimal effort. It simply wasn't accessed because in the socially networked age, the process of truth-finding has been replaced with the negotiation of the meanings of higher order complex symbols. Once we can say (or once we are told, or once we believe that everyone else believes) that an event is "about" something bigger and far more important, we can no longer unravel it by worrying about trivialities like whether it is "true" at some basic level.

The third disagreement is related, and less of a disagreement than a reaction to an implication that seems present in Weinstein's response and those of many others I have seen. I think Weinstein is right to call what we are facing a Cartesian Crisis. Whether we like it or not, it will be our lot to grapple with the untethering of our world and our experience of our world from anything like a grounded reality. He implies and others have suggested concerns that AI-generated images, audio, video, deepfakes, and other artificially generated sources of false realities often will be what produces these dislocations. It is a legitimate concern, and I think AI will be a vector of harm in many such cases to come. Yet the origin of the problem is not AI's capacity to create real-seeming things as much as the way that social networks are now mature, dense, and socially integrated enough to attach complex symbolic relationships and meanings to any statement or event in a matter of mere hours. Generative AI supercharges this, sure. But this is simply a computational version of what the media has been doing for a few years now with accelerating boldness: telling us how to think. Telling us what the events of our world really mean. Telling us what is real.

The media realizes today that social networks offer greater reach to traditional efforts to explain the events of our world to the unwashed masses. If you doubt it, go take a look at our measures of the linguistic markers of fiat news at fiatnews.com. What I think the media have yet to (fully) realize is that mature social networks also give them the power to assign symbolic relationships to facts and events. Social networks give them the power to mediate what everyone believes those symbols mean, wholly untethered to any sense of obligation to reality. That's when the real Cartesian Crisis sets in. Not when we realize that generative AI can effectively simulate reality, but when the media realize they can effectively define it. Elon's delusions about social networks allowing ideas to percolate from the bottom up notwithstanding, there is only one way to avoid this future, I think.

Facebook delenda est. X delenda est. Tiktok delenda est.