Since I moved to NYC last month, I’ve been seeing a lot of concerts.

Some of these shows feature the best jazz musicians in the world, playing for crowds of less than 30 people. I notice in these environments how welcomed it is to play weird sounds. It’s almost like the weirder or more novel the noise is, the more encouraged it is. It’s as if there’s this unspoken rule—the better the musicians, the stranger it’s expected to get.

I’ve also seen a handful of jam band concerts. At these shows, I notice that the part I enjoy most is clearly when the band is trying to imitate Phish, and specifically, the guitar player is trying to play exactly like Trey Anastasio. Despite enjoying these parts of the show the most in the moment, I also find it unsatisfying.

I don’t want to hear some band try to play Phish’s music, even if they could do it better than Phish—I want to see bands do their own thing.

Similarly, Phish is a largely improvisational band known for their jams. There are many jams the band has made in the past that fans hold in high regard, but fans don’t want to see Phish recreate their most legendary jams, nor other bands play them, or even composed parts that sound like them. This gets at something really interesting about how we think about art—we’re typically not actually looking for “better” versions of things we already love.

Most fans of music are not trying to hear something better, or a just-improved version of the category they like, but rather wanting to hear something novel.

If you go on YouTube, you can see an endless supply of musicians with such technical proficiency that it will make your jaw drop—they can play any style with such ease and perfection. But nobody cares about these musicians. Because music isn’t about playing fast, or perfectly, or being able to recreate what other people can play—it’s about creating new sounds. It’s wild how you can be technically better than every guitar player from the ’60s and ’70s, including the best players like Jimi Hendrix or Eric Clapton, and still not have anyone want to listen to you.

And then I realized, at these jazz shows, the musicians are not trying to produce the best-sounding noise. They’re actually not even best thought of as musicians at all. Rather, they are a kind of gold miner. And the tiny avant-garde jazz show is akin to being deep in the mines, with the band, pounding away at the rock, in search of a new nugget. And like in mining, it’s not the avant-garde jazz artist who will profit off it—rather, to the extent they discover new, interesting noises, those will then get incorporated by a musician one notch more popular or mainstream than them, until they incorporate it only for it to be taken and then incorporated into another musician one notch more mainstream, until it finally becomes part of the musical zeitgeist. It’s this weird ladder of innovation where each rung takes something weird and makes it a little more digestible.

Have you ever listened to the self-titled 1970 album by Emitt Rhodes? If not, I highly recommend you listen to it—it sounds amazing, but it’s more interesting for its lack of interest. 

Emitt sounds exactly like Paul McCartney. Many songs on here would be considered good Beatles songs, and this album as a whole would be considered a top-tier Paul solo album and among the best solo albums produced by a Beatles member. And that’s exactly the problem.

Emitt is trying to sound just like Paul, and while we notionally want to hear music that sounds great, we actually prefer something else altogether. This album was dismissed, despite it sounding amazing, because it’s pastiche. 

Sam Bankman-Fried once argued that it’s preposterous to think Shakespeare was the best writer of all time, simply due to population statistics.

The argument goes, there are so many literate people alive today that it’s implausible that someone at that time would be the best writer of all time. And then when you factor in how people today are way better educated, get to learn from reading everyone else’s stuff, have more free time to actually write—it seems even more unimaginable to think nobody’s topped Shakespeare yet.

Most people get really upset reading this argument, and there are huge numbers of people mocking Sam or writing counter-arguments, which basically say, no, you don’t get it, I can, using my own taste, determine that Shakespeare is better and you are a monster autist with no soul and wrong. 

Unfortunately, very few people actually grappled with Sam’s argument, which is so strong that you would need an incredibly strong explanation to rebut it.

On the other hand, those rare voices in support of Sam gladly point out, in literally everything we can objectively measure, things like running speeds, or weights lifted, or speed crossing some land, all the best at X are in the present, and those like Shakespeare, born in the 16th century, are far, far, far behind from being the leader in anything.

In 2020, I started following the sport of long-distance triathlon. Sometime between 2020 and 2024, performances in the sport of triathlon got way better. A few things happened—there became much more money in the sport, so many more people started competing in it; those already interested could now invest much more intensely into pursuing it. In parallel with this, platforms like Strava and YouTube became very popular, as well as there being bigger races in which all the pros would attend. This meant the knowledge transfer became much more widespread and rapid, and so all the good ideas quickly spread and were adapted by everyone else. On top of this, with increased interest and money, the technology also rapidly advanced. This is what normal progress looks like, when everyone is trying to achieve the same goal and the goal stays constant over time—you can literally see it getting better.

In contrast to this, I think of random data points like how Jamaica does better than the US at the Olympics in sprinting, even though there are a larger number of the same genetic groups in the US than Jamaica. It’s argued that because sprinting is so important in Jamaica, everyone is tested in it for talent, and there is no higher calling, so it attracts the best. In the US, there are many people with athletic potential who were never tested and simply became fat, or those with incredible talent, but chose to play a sport like football instead, so their running limits were never discovered or developed. Sometimes what looks like exceptional talent is really just showing you how shallow the pool is.

In the 2024 Paris Olympic Games, Kristen Faulkner won the gold medal in the women’s road bike race, despite only beginning to cycle in 2017 in her spare time while working as a venture capitalist. Does this suggest that Kristen is so ungodly great at cycling, or that there are relatively few women who pursue road cycling at the highest level? It’s notable how sometimes what looks like being the best in the world at something is really just showing you that not enough people are trying to be the best in the world at that thing.

I believe that most pursuits in life follow a direction of progress—where nearly everything gets better over time as we collectively invest more resources into pursuing the same aim, just better. 

I don’t think there is a coherent idea of making progress in art.I think most art is not about producing something that is better than before, but something that is novel and great. And that’s a crucial difference that changes how it evolves.

I believe, as Scott Sumner and Holden Karnofsky have argued in the past, that art involves discovery, and when a new medium is created, there is more opportunity to discover the low-hanging fruit of that medium.

I think what many people miss from this is that once a particular era in art is over, people stop trying to refine it.

This means that once the era of novels like Anna Karenina is over, we don’t stop creating better versions of Anna Karenina because humans are incapable, or there was something uniquely special about Tolstoy, but because the most talented and obsessive artists have no interest in writing a better version of Anna Karenina. Instead, they want to create their own, new thing. And once the low-hanging fruit has been plucked, the artists need to go deeper and deeper into the gold mines, hoping to find something new and interesting. They’re all playing a different game entirely.

This is what I think many people in the SBF Shakespeare argument miss. The fact we don’t make better music than Bach that sounds like Bach, or music better than the Beatles that sounds like the Beatles is not because we can’t, but because nobody with talent is aspiring to do this. 

I will note that my favourite era of music in history is around the years of 1966–1975. You’ll note that when you see rock ’n’ roll bands today, most dress like they are still in this era. But when you see gold miners trying to innovate new music, they always dress like they are in the present. Are they trying to recreate someone else’s sound and identity, or do their own thing? The clothes reveal a lot.

As a huge movie fan, it seems clear to me that movies were much higher quality in the past. Although, I will note that TV shows in the early days of TV were terrible, and only recently became relatively passable, although still nowhere close to as great as movies peaked at. This isn’t because we lost the ability to make great movies—it’s because the energy that used to go into making great movies is now sucked up by making efficient bland hollywood flicks.

In theory, since so many of the best films of the past were done as passion projects, inexpensively, and with small teams, it should be possible for many filmmakers today. However, I’d argue that the infrastructure isn’t around to support small-budget movies of artistic brilliance as in the past.

People are somewhat limited by their zeitgeist. If you’re not in the time of classical, or of disco, you can’t really make it. Because you need to be consuming others, constantly talking about it, learning from others, trying to impress others in it, etc. If you’re in the wrong time, it’s hard to have the right inputs for it. You can’t just decide to make great disco music in 2024—you needed to be there when it was happening.