Overlearning

I largely agree with this piece by Maarten Boudry called “The Fallacy Fallacy.” The central insight is quite correct: the kind of fallacy hunting that’s become popular in the era of online forums and social media has a way of turning you into a specific kind of stupid and credulous, so intent on finding error in the thinking of others that you end up captive to the idea that everyone but you is a rube. We’re all trapped in various mental mazes, we’re all subject to biases and blindspots. The trick is to neither ignore that possibility nor to fall too deeply in love with the idea that you’re the one who can pierce the veil and see the real world beyond; as the rationalists themselves demonstrate with some of their side quests, you can be deeply deluded even as you spend your days enumerating the delusions of others. You need to be critical about the reasoning of other minds, but first you need to be critical about your own reasoning, about yourself.

This is, it turns out, quite hard to do.

As you’ve no doubt deduced, this is all a lot of throat clearing before I tell you about a way other people are wrong. There’s a particular kind of intellectual error that gets almost no attention, or so it seems to me, perhaps because it’s error that masquerades so convincingly as wisdom. It’s a particular kind of error, not the error of the fool who learns nothing from experience but quite the opposite. It’s the error of the sharp, skeptical mind that learns too much, that overfits the curve, that extrapolates irresponsibly, that takes a genuinely correct observation and rides it so far past its destination that it ends up somewhere just as wrong as pure ignorance, only with more confidence. It’s not an error of the credulous, as so many of the canonical fallacies are, but rather an error of the converted. Call it overlearning. And to better understand what I mean, as an initial example, consider the much-maligned audiophile.

The audiophile world is a perfect laboratory for the overlearning phenomenon.¹ Audiophiles are stereo-equipment enthusiasts willing to spend extravagantly on audio gear (“hi fi”) in pursuit of sonic perfection, people who agonize over the warmth of a turntable’s tonearm or the “air” a particular amplifier lends to a violin’s upper registers. To outsiders, and especially to the technically minded, they’re an easy target, credulous consumers who’ve mistaken expensive placebo for genuine fidelity. For decades, the high-end audio industry has developed products that have elicited scorn on Reddit and Twitter: $500 speaker cables that measured identically to $15 ones, turntables priced like used cars, amplifiers with specifications that exceeded the limits of human hearing by orders of magnitude. Pre-internet, the audiophile world puttered along largely unseen the way many enthusiast communities have, outside of the public eye and thus protected from public scrutiny. But as the social media era has thrust us all together in a profoundly unnatural and uncomfortable way, the audiophile has become more and more a figure of scorn. Skeptics - many of them technically literate, and in fact, many of them correct - pushed back hard against extravagant audiophile claims. Blind listening tests frequently failed to distinguish “audiophile grade” from mid-fi gear. The emperor, in many cases, appeared to have no clothes.

That basic observation, I will reiterate, is largely correct! A lot of audiophile gear is ludicrously expensive, and a lot of what gets sold in that world has very dubious claims to improvements in sonic quality. Precisely because that central insight is correct, a lot of people have absorbed the point, and mocking audiophiles has become something of a Millennial meme, a classic anti-Boomer axe to grind for a younger, savvier generation. And then a lot of people overlearned that lesson.

The overlearned conclusion was that audio equipment quality is essentially a myth, that it’s all snake oil, all placebo, and that anyone who invests in a real stereo system has been conned. This is wrong in a way that becomes obvious the moment you actually sit in front of a decent system. Take someone who has spent years listening to music more or less exclusively through iPhone speakers and wireless earbuds (which is to say, take almost any person under the age of 35) and sit them down in front of a modest integrated amplifier, a CD player, and a pair of bookshelf speakers. A cheap but real setup, in other words, something that has been built with sound quality in mind, which you could assemble for $1,500 to $2,000, well below the entry point audiophiles would even consider serious. The superiority over the Spotify-through-Bluetooth experience will not be subtle. The soundstage opens up; instruments occupy distinct space; vocals have body and texture; bass is felt as well as heard. This isn’t a matter of imagination or expensive expectation but a straightforward consequence of playback hardware that was engineered to move air in a room rather than vibrate a tiny membrane pressed against an ear canal. Or you could see what’s really possible in the earbud space by spending a combined $600ish dollars on these earbuds and this portable CD player, finding a used CD at a thrift shop, and giving it a whirl. Do an A/B test with an iPhone, a streaming service, and Airpods. I promise you’ll hear a difference. I promise.

The correct lesson, that marginal returns in audio diminish sharply and that much premium pricing is irrational, was fun to make; it had a clear target, audiophiles, and they did not appear sympathetic to many people on social media, given that they tended to be older and were almost definitionally rich. And so over time the lesson grew and grew and grew, leaving behind its sensible form and becoming an exaggerated distortion, a blanket dismissal that threw out something real. The person who overlearned audiophile skepticism ends up no closer to good sound than the person who never questioned anything audiophiles said. They just feel smarter about it.