On the False Epistemological Overconfidence of Social Media

A typical Twitter exchange might go like this:

“Coal kills _____ millions of people per year.” (Possibly a link to a related article, which only .01% of people who see the tweet will actually read in its entirety.)

And then someone like Jordan Peterson will reply or re-tweet an argument along the lines of, “Such deaths are offset by development. Coal pulls poor nations out of poverty.”

Both tweets will get significant likes, re-tweets, and approving comments. Such engagement creates social proof and lends authority to their respective authors, but there is nothing really being argued here. This exchange doesn’t even rise to the level of a flame war. It’s just sound bites that poorly approximate a dialogue, but without any of the work of having to substantiate the position beyond the superficial or any attempt to understand the opposing side.

The negative replies in the comments that shed light on the obvious holes is the aforementioned arguments, can easily be ignored or the user blocked without any consequence such as being called out for being wrong or refusing to debate (unless another high-profile user does so).

Twitter also has the ill-effect of reducing complicated issues to simple binaries, or creating a false consensus about controversial or complicated issues or issues in which no such consensus exists. Seldom is something newsworthy and there is a consensus. The element of the unknown is part of what makes it newsworthy and controversial to begin with. You will never see the headline “Experts agree murder is bad!”

Like with Covid vaccine effectiveness, I was under the mistaken impression, probably due to an over-reliance on twitter, that the ‘media consensus’ in late 2020/early 2021 was that the vaccines were ‘95% effective’. Despite Biden saying so, there was no such consensus: there were plenty of mainstream sources that purported they were only 40-70% effective at stopping Covid, or doctors expressing some doubt.

It leads one to underestimate the strength of the opposing side. ‘Owning the libs’ outside of Twitter, is not as easy as Ben Shapiro makes it seem, nor is as simple as reciting a Thomas Sowell quote. People who spend their lives researching such issues have seen and heard all the sound bite rebuttals to their arguments, yet outside of Twitter such arguments carry little currency, so all that prevails are sound bites.

Peterson, Sharpio, etc. have created artificial, carefully curated environments, whether it’s twitter, softball interviews, or ‘owning’ campus SJWs, in which they are impervious to substantive criticism of their ideas. Not just to pick on ‘the right’, but this also applies to left-wing personalities such as Krystal Ball, Jimmy Dore, and Cenk Uygur. If they had to debate and defend their beliefs against actual experts, they would almost certainly get chewed up and spat out. They would encounter arguments and terminology they had not seen or prepared for. So obviously, they are not going to do that. I agree with Dr. Peterson about global warming and overpopulation, but he’s not a good enough debater or knowledgeable enough to sufficiently defend his beliefs against experts in those fields. Look at what happened when he debated Slavoj Zizek, in which to the majority outside observers Peterson lost. Or his debate against Sam Harris about the meaning of ‘truth’, which also went poorly.

I’m sure everyone has seen those infamous YouTube videos in which the late economist Milton Friedman soundly refutes anti-capitalism objections by university students during a Q&A session, one featuring future NYTs columnist David Brooks, but as smart as Dr. Friedman was, likely his arguments would have been met with much more forceful objections by actual economists and experts than students.

Moreover, in spite of the considerable engagement and positive feedback on Twitter, such as tens of thousands of likes, retweets, and hundreds of approving comments, such arguments outside of Twitter tend to be much more poorly received. They are popular because Twitter is an echo-chamber, for both sides. On Reddit , blogs, or any site in which the mean IQ of its users is above 100 or so, the holes become as obvious as Swiss cheese, and called out as such. Those Thomas Sowell quotes tend to not hold up too well to even the slightest scrutiny by anyone who spends a few minutes thinking about it, much less actual economists. Thus, you cannot get a good bearing of what constitutes good or bad beliefs or arguments, because inflated social proof makes it nearly impossible to differentiate between good or bad, correct or wrong beliefs (unless the opinion is so overtly wrong or bad that it gets ‘ratioed,’ which means tons of negative comments).

Bongino, Shapiro, Levin, etc. and others are popular because they use mass marketing to attract huge audiences receptive to such sound bites, but this resource is scarce, as you have dozens of pundits and hosts competing for this same audience, much like how infomercials compete for air time for an viewership that is getting older and not growing in size. Such apparent popularity is thus not organic, but manufactured through extensive marketing.

The reason why ‘smart’ content goes viral but low-information content by the likes of Shaprio , Bongino, Levin, and others does not, is in part because smart people have larger social networks, such as measured by followers and social clout (except for celebrities, there is a positive correlation between IQ and the size of one’s social network). A Substack blog post may go viral organically on 4chan, Reddit, Twitter, or Hacker News, but you will never see a Bongino or Shaprio clip on any of those sites, ever (except on Twitter when re-tweeted by its creator). That because such arguments tend to be repulsive to anyone who is at least smart or perceptive enough to see the obvious holes and lack of originality, which is not a high bar, hence no virality.

To bring this back, content that seems outwardly popular and viral, such apparent popularity is likely not organic and in no way predictive of the quality of the argument brought forth. Second, issues in the social sciences are complicated and cannot be reduced to sound bites without losing the nuance and necessary details; relying on social media can provide a false sense of epistemological certainty, made worse by inflated social signals/proof. Rather, you have to seek sources in which arguments are received on their merit, not the popularity of who originated them.