Signalling Theory, Twitter, and the Fake Blue Tick

Twitter—ahem "X"—is a perfect illustration of signalling theory, as Elon Musk has taken a useful honest signal and turned it into a dangerous, dishonest one that's damaging our democracies.

Thank you for reading The Garden of Forking Paths. This article is free, but if you value my writing, please consider upgrading to a paid subscription. If you might consider tipping a waiter, how about a writer who so carefully prepares your delicious brain food?

Reminder: the first Zoom lecture / Q&A for paid subscribers—on “Trump and the Rule of Law”—will be held on this Sunday, September 3rd at 3pm ET, 8pm UK time.

---

Elon Musk, I regret to inform you, has unleashed hordes of fiddler crabs and fangblennies on the internet.

No, those aren’t weird British slang words that I’ve picked up during my twelve-years marooned in the British Isles. They’re a pair of deceptive species that engage in a practice now rife on The-Social-Media-Site-Formerly-Known-as-Twitter: dishonest signalling. And like the fangblennies, the users that Musk has empowered come with a dangerous bite.

This is bad news for all of us. But to understand what the hell I’m talking about, you first need to understand signalling theory. Once you do, we’ll return to the festering hellscape that is Elon Musk’s Twitter, to lament the rise of the fangblennies.

Is honesty always the best policy?

Signalling theory helps us understand the shortcuts we use to navigate our world.

We are all the byproduct of a long, unbroken chain of organisms that survived to reproduce. As my friend Zachary Blount, an evolutionary biologist, once pointed out: when you meet someone, you can be certain about precisely one fact about them: that all of their direct ancestors survived long enough to reproduce. That’s it. Everything else we infer about them is a guess, based on past experience and social cues.

Those cues are signals—and we’re not the only ones who use them to quickly size up another individual who moseys into our view. Are they friendly? Are they attractive? Are they going to kill me? Will they agree to watch the Great British Bake Off with me?

These are the kinds of questions that demand answers—and fast—in the natural world, though they persist in our cocktail parties and our Tinder dates. Nature, always clever, has developed useful shortcuts to help us find the answers to those questions, if only we know where to look.

Signalling can be classified on two axes.

1. Honest vs. Dishonest

2. Costly vs. Costless

An honest signal is a cue that accurately depicts an intended meaning. For example, some poisonous frogs have coloring on their backs that shows predators that they are, indeed, toxic. (The technical term for this is aposematism, and it means advertising to a potential predator that it’s not a good idea to eat them).

But honest signals can also be positive, such as advertising reproductive fitness, as with the tail feathers of a peacock, which are accurately correlated with their desirability as a mate. What matters for a signal to be honest is that the conveyed information matches the truth.

Humans have plenty of honest signals. As Jonathan R. Goodman of Cambridge University notes, hunter-gatherer societies have long used honest signals in selecting partners.

Among the Meriam people of Australia, for example, some hunters are better at catching sea turtles than others. Because they must actually produce a turtle, and because anyone who falsely boasts about being a good turtle hunter is quickly exposed as a fraud, it functions as an honest signal for a hunter’s ability to provide for their future family. A study of the practice showed that the honest signal was effective: good turtle hunters got more social recognition, found partners earlier, had more children, and so on.

In modern life, we use broadly honest signals as shortcuts. For example, finding out that someone has a PhD in particle physics from Harvard or Oxford is typically an honest signal that they’re likely to be smart (or at least booksmart). No signal is perfect in modern society, but some are usually pretty safe bets.

By contrast, a dishonest signal is deceptive. Nature is full of them. Spiders disguise themselves as something harmless, only to lure their prey closer (as with the astonishing “bird-dropping spider” of Australia and New Zealand, which camouflages itself as exactly what its name suggests).

Or, there’s the fiddler crab, which has one comically enormous claw to stave off predators. At first, that’s an honest signal: don’t mess with the giant claw. But if a fiddler crab loses a fight with a rival, it will also lose its big claw. In its place, a similarly sized, but much weaker claw grows back—making the defeated crab an easy target.

But here’s the clever bit: the strong claw and the regrown weak claw are indistinguishable, so a fiddler crab can effectively trick possible rivals into avoiding a fight. That’s a dishonest signal. It relies on deception.

Humans are masters of dishonest signalling. On the extreme end, this is where psychopathic serial killers thrive, luring unsuspecting victims with superficial charm (our more sophisticated version, perhaps, of the dishonest signaling of the creepy anglerfish). But other dishonest signals are more banal, like knock-off Rolex watches, which cost $5 but fool plenty of people into thinking they’re more than a thousand times more expensive.

At what cost?

The second axis of classification is therefore about cost. Because the fake Rolex only costs $5, it’s also known as a mostly costless signal. It doesn’t require substantial resources to create the signal. (Paul Gonzalez, the infamous LA Dine and Dash Dater, who would take women to fancy restaurants to make them think he was rich, then ditch them with an expensive bill, mastered the art of a dishonest, costless signal, though he did end up paying the price by going to jail).

By contrast, buying a yacht to show wealth is a costly but honest signal, an example of what the economist Thorstein Veblen called “conspicuous consumption,” an attempt to buy social prestige by spending large amounts of cash.

In nature, the same terms apply. A peacock’s tail feathers may accurately portray mating fitness, but the elaborate plumes slow the birds down, which comes with serious downsides—including the risk of death from a predator—and consumes lots of food energy to produce. Put the two axes together and, in the peacock’s case, you’ve got an honest, but costly signal.

On the other end of the spectrum, consider a harmless frog that has markings on its back to make it appear similar to a poisonous frog. This form of deceptive signalling (known as Batesian Mimicry) is widespread in butterflies, too. But it’s both dishonest and costless: they can deter predators, but don’t need to grow some outlandish resource-intensive plume that makes them sluggish; the color pattern on their bodies is automatic and comes with few downsides.

Twitter, the Blue Tick, and Signalling Theory

In the pre-Elon Musk days, Twitter developed an opaque, somewhat arbitrary mechanism for allocating blue ticks for “verified accounts” (full disclosure: I had one). This was a weakly honest, costless signal (it was free, if you were awarded one). Sure, there were plenty of weirdos and cranks who had them affixed to their accounts, but they did loosely (though not always) correlate with accounts of those who were:

• Real people

• Influential in some way

Of course, there were plenty of influential people who didn’t have blue ticks on Twitter, and the system was guilty of arbitrary elitism. Nonetheless, it did provide a concrete benefit for processing information on the site.

When you came across an account from a recognizable name—such as a prominent journalist from the BBC—and you saw a blue tick next to their name, you could be reasonably assured that it was the journalist, and not some random imposter.

This was important, not just for checking the veracity of news, but also for safety. After all, if a verified weatherman warned that a tornado warning was in effect and they had a blue tick next to their name, you could have reasonable confidence that you’d be wise to heed that advice and seek shelter.

In social terms, the fact that the blue tick was perceived as a broadly honest signal was precisely what made it valuable.

Elon Musk blew that system up, destroying the old verification system and handing out blue ticks to any idiot who would pay him $8/month.

It soon went predictably wrong, as fake “verified” accounts swarmed the platform, creating major economic shockwaves for companies that were impersonated by playful hoaxsters. One, impersonating the pharmaceutical giant Eli Lilly, tweeted “Insulin is free now,” which wiped billions off the company’s value.

This is where the fangblennies come in. “What’s a fangblenny?” you ask. Great question!

This is a fangblenny.

These sneaky, mean little liar fish are imposters. They’ve evolved to look like the nice little bluestreak cleaner wrasse, a gentle, generous helper fish that nibbles on parasites that adorn the bodies of their larger underwater compatriots. When everything goes to plan, the cleaner wrasse gets a multi-course meal. The bigger fish gets rid of parasites. Win-win.

Unless, of course, the larger fish gets tricked by the fangblenny. When that happens, the fangblenny doesn’t do any parasite cleaning. Instead, the fangblenny takes a bite out of the bigger fish, which has let its guard down after mistaking it for a wrasse. Even more astonishing, the fangblenny bite contains an opioid venom, which dulls the pain and confuses the larger fish, giving the imposter time to escape. (I warned you it was a sneaky, mean liar fish).

But the fangblenny’s deception only works when it’s rare. If the fangblennies become too common, the larger fish learn the lesson and stop falling for the deception. The dishonest signal no longer works.

That has now happened to Twitter (sorry, “X”), where people have learned that blue ticks have completely lost their value. We’re no longer duped by fake accounts so easily, because, like a school of too many fangblennies, they’re everywhere and they’re not fooling anyone.

In the more formal language of signalling theory, what Musk did was to turn a weakly honest signal that was at least somewhat socially valuable into a totally dishonest signal that was more costly for users ($8 more, to be precise) but socially useless.

When the only criterion for verification is “having eight dollars,” well, that’s not a signal worth paying attention much attention to. Or, as Jonathan R. Goodman summarized it succinctly: “If having a Ph.D. were reduced to spending US$15 a month on a subscription to a university, we wouldn’t take it seriously.”

The Return of the Blue Tick as an Honest Signal

Elon Musk eventually managed to achieve something extraordinary: he made blue ticks honest signals again.

How? By being such a toxic, juvenile, conspiratorial blowhard that he created a sorting mechanism. It turns out the kind of people most eager to pay Elon Musk $8/month for a now-pointless digital signal were fellow travellers in Muskland, the cryptobros who “just want to ask questions” about race and the influencers who don’t care that Musk is a vector for Russian propaganda and vile right-wing disinformation.

Suddenly, the blue tick was again an honest signal: that you were an asshole with 8 bucks.¹

In fact, it has become such an honest signal for being an asshole, that Twitter has given its paying users the ability to hide their blue tick so they’re not embarrassed by it. In a few months, Musk turned a reputation-enhancing status symbol into something that turned you into a pariah so much that you needed to hide it. Elon Musk, Secret Genius, strikes again.

But there was another kind of social media behavior that again followed the patterns of the natural world, a phenomenon known as Mullerian Mimicry. In this extraordinary biological trick, two or more poisonous creatures who share a common predator evolve to look like each other.

It’s ingenious protection, because if a predator encounters one species and learns it’s toxic, the predator will automatically (and correctly) decide not to try to eat the other species that looks similar. For example, these two species of butterfly (one species on the left and one on the right) have evolved to look like each other, warding off common predators. If you see these markings—or some close variation of them—you know to stay away.

The same has happened on Elon Musk’s Twitter. Every account is different, of course, but when you see an account with an $8 blue tick, you click on it, and you see an American flag emoji with MAGA and #TRUMPWON written out while the photo is of a young man holding a large fish or a large gun, Mullerian Mimicry has allowed us to quickly deduce that this isn’t just an asshole with 8 bucks, but a dangerous asshole.

They’re helping us out by wearing the same toxic spots.

The Real Cost of Musk’s Blue Ticks

Signalling theory has one more lesson to teach us, and it’s arguably the most important one. We, as humans, take our social cues from what we experience most often. If you surround yourself with kind, compassionate, decent people, well, that’s likely to rub off on you and shape your worldview. But if you surround yourself with racists, it’s inevitable that you’ll start to view racism as normal.

That matters in the post-Musk social media hellscape of “X” because Musk hasn’t just restored accounts of neo-Nazis and racist bigots who normalize violence—he’s amplified them, so long as they pay him $8/month. For those of you who aren’t on Twitter/X (good for you!), it may be news to you that the most visible comments on most viral posts are often now extraordinarily vile and hate-filled.

Musk’s revised algorithm recently showed me a video in my feed—who knows why—of a cyclist being run over by a car. Now, the top comments are always provided by users who are Muskian blue ticks, by design. When I looked at the top comments to that horrifying video, I was shocked to see that all were celebrating the driver of the car and saying the cyclist probably deserved it.

Musk’s decisions will continue to corrode our social fabric. Because of his behavior, the blue tick system has become a self-sorting mechanism for the worst among us, pushing their views to the top of our social media public square. He has designed a system that uses an algorithm to pump the most extreme, awful opinions into the social media feeds of millions. It’s not about “free speech,” as he claims. It’s about selectively promoting vile, hateful speech, which is now happening regularly, all because they paid $8 to a billionaire.

And that, I hope we can agree, is an honest signal: that Elon Musk, like a fiddler crab or a fangblenny, should not be trusted.

---

Thank you for reading The Garden of Forking Paths. If you enjoyed this article, and think I should be paid for my research and writing, please consider upgrading to a paid subscription. The annual subscription is $4/month. I have no corporate sponsors, no grants to write for you—so I rely exclusively on reader support.

1. There are obviously exceptions to this. (Apologies especially if you’re a kindly reader who has a blue tick! Please don’t unsubscribe! Forgive me!). The sorting mechanism under Musk is strongest in the United States, where, given the polarized political environment and Musk’s politics, a blue tick has mostly, but not always, become an honest signal as I described. In other contexts, people have decided to pay $8/month for other reasons, and the sorting mechanism is weaker.