Fat tails

How the internet made us both smarter and dumber

Over here at The Pursuit of Happiness, I don’t devote quite as much time to politics as I did at my old blogs (The MoneyIllusion and Econlog.) That’s not because I’m no longer following the news, and not because I don’t care about what’s going on. Rather I’d point to a few other considerations:

1. Unlike when I was younger, politics is no longer any fun. Almost all the news is depressing. Why spend your time wallowing in misery, especially in a blog devoted to happiness.

2. Political debate is often destructive—it makes one dumber. I notice that the blogs of intellectuals are generally smarter than the twitter feed of the same people. That’s probably because people tend to be more political on twitter.

3. Other people cover politics much more effectively than I can. Pundits like Richard Hanania and Matt Yglesias make most of the political points that I would have made, but more intelligently, more humorously, and to a much wider audience. What do I have to add that would be useful?

4. I’m not likely to convince anyone to change their mind. If I point to a Trump outrage, people will point to a similar action done by a previous president. This is where my “Less Wrong” post comes into play—the previous presidential outrages were often less wrong, even if qualitatively similar. More importantly, the atrocities now come by the boatload, not just here and there. But how does one prove that in a blog post? Would I have to cite a hundred outrages?

Hanania provides a good example of the sort of post I like. To the question of which is worse, left wing or right wing cancel culture, he basically answers “both”:

Most analysts fall into the trap of simply declaring one better or worse than the other. I think a more useful approach would stress that each is worse in its own ways, and nothing forces us to make an apples-to-apples comparison.

I’d say something similar about political commentary. Is it worse or better than in the 20th century? I’d say both. The internet has led to much fatter tails on the distribution of quality. It has raised up insightful pundits that previously would have been unheard, while at the same time dramatically reducing the average quality of political news coverage. The twitter debate over recent events is almost a textbook example of how electronic communication combined with strong emotions often leads us to provide bad takes.

I first noticed this back in the early 1990s, when email communication started leading to office feuds that never would have occurred if people were still meeting face to face and trying to resolve an issue in person. It’s not just that impersonal electronic communication leads to more anger, I find that people don’t even understand what’s happening—why the disputes are getting more vicious. They don’t realize how cold and rude the message they are sending will seem to the recipient. Amina Green pointed out that email is actually an inhuman form of communication. And here’s Astral Codex Ten:

[Tim Berners-Lee] built a cold and inhuman Web—so why would we be shocked that the online world became a cold and inhuman one?

For this reason, I increasingly gravitate toward pundits that usually (not always) employ a calm, dispassionate, above the fray approach. At times, that can come off as cold and unemotional, at least to some readers, but when it comes to politics that’s exactly what you want.

I am also increasingly interested in how our biases shape our view of the world, often without our even knowing what’s going on. (And to be clear, I’m just as guilty as the next guy.) In a comment thread for my recent post Less Wrong, Peter Gerdes said:

Wittgenstein showed that the word game got applied to a wide variety of family resemblance type concepts. However, I think you could still reasonably maintain that this didn't go so much to how each of us understood games just how we mapped it to words. You could insist that what was going on is that we each would (subpersonally) have a bunch of some relatively ground up clear concepts that -- thanks to limited bandwidth and difficulty coordinating -- we all kinda grope for the word game when describing. So something about the dictionary is fuzzy but that's a step away from saying the fuzziness and associational character is intrinsically part of how we think (tho I feel introspection makes it clear I don't think everyone did).

And I responded:

Good comment. I believe most people vastly underestimate the extent to which we all see the world in different ways--so much so that you might say we all live in different worlds. This misunderstanding leads, for instance, to people assuming that those holding different political views are bad people. After all, if they see things in the way that I do then how could they advocate such horrible policies?

Again, I’m often guilty here as well.

Another cognitive bias is to put too much weight on individual anecdotes. We tend to underestimate just how big and complex the world actually is. (This is a point I was trying to make regarding China.)

The OC Register has a recent article with this odd graph:

I was born in Michigan and raised in Wisconsin, and I honestly cannot see any difference between the two in terms of quirkiness. Neither state seems all that weird. Given that most Americans have probably never even visited Wisconsin, I wonder if their impression is based on some minor quirk, like Packer fans wearing a slice of cheese on their heads?

And why is Alaska #2? I don’t watch much TV, but a google search told me that there was a popular TV show about Alaska called Northern Exposure. According to google:

The location is remote, the people are weird, and Joel just wants to return to New York.

The least weird people I’ve ever met were Mormons. And yet Utah is viewed as the 4th weirdest state. Could that be based on a tiny handful of Mormons practicing polygamy?

Back in the 1990s, if I mentioned Singapore then most people would cite the American teenager that was flogged for some non-violent crimes. If I mentioned Bhutan, many people would mention their government’s promotion of “Gross National Happiness”. But Singapore is actually a very successful place, and Bhutan has a horrible human rights record that involves the ethnic cleansing of its Nepali residents. Popular anecdotes don’t tell us very much.

I get that it’s “only human” to see political issues in emotional terms. I get that it’s only human to think in terms of anecdotes rather than dry statistics. But it’s also only human to be living in poor failed societies, and we struggled for millennia to get to the stage where a modest fraction of the world now has a comfortable standard of living with relatively little political violence. It would be a shame to lose all that progress because we give in to our only human instincts.

Maybe artificial intelligence will rescue the human race. But now I’m reading that the only way to make AI safe is to give them human attributes such as the maternal instinct. Does that mean they’ll eventually lose the calm dispassionate tone that we now associate with artificial intelligence?

PS. If you insist on political commentary, here’s my thought for the day. For 8 months there’s been a whirlwind of activity in Washington. But activity doesn’t always equal effectiveness. There’s still this thing out there called REALITY. Putin won’t necessarily do our bidding. Neither will Netanyahu. China is too big to bully. Epstein files won’t go away. Some MAGA people really do believe in free speech and free markets. Tariffs won’t magically create manufacturing jobs with low inflation just because you think they will. The budget deficit isn’t going anywhere and Social Security cuts are unpopular. Farmers and builders don’t like all their workers being deported.

PPS. Here’s something else to think about:

Killing suspected terrorists? How is that different from killing enemy soldiers?

Killing suspected drug dealers on the high seas? How is that different from killing suspected terrorists?

Killing suspected drug dealers on the streets of NYC? How is that different from killing suspected drug dealers on the high seas?

Killing suspected murderers on the streets of NYC? How is that different from killing suspected drug dealers?

Killing suspected rapists on the streets of NYC? How is that different from killing suspected murderers?

Killing suspected bank robbers . . .

. . .

. . .

. . . something something jaywalkers . . .

Of course the answer is that some things are less wrong than others.

And one can play this game for almost any hot button issue: “Firing a worker for saying X? How’s that different from firing a worker for saying Y?”

You can instantly boost your IQ by 5 points simply by refusing to play these games.

This is from one of my favorite essays, explaining the slippery slope:

If once a man indulges himself in murder, very soon he comes to think little of robbing; and from robbing he comes next to drinking and Sabbath-breaking, and from that to incivility and procrastination." — Thomas de Quincey (1785-1859) "A Second Paper on Murder Considered as One of the Fine Arts,"

PPPS. Enquiring minds wish to know whether the $50,000 bag of cash was reported on Mr. Homan’s tax return.

PPPPS: For fans of fat tails, here’s a beaver shot:

Photo by The New York Public Library on Unsplash