Understanding middlebrow

I’ll get to the NYT eventually, but first let’s look at the middlebrow sector of the film industry. And by the way, middlebrow is not the middle of the population, it’s roughly the 90th to the 99th percentile. Highbrow is the top 1%. And then there’s rest—Kardashians, professional wrestling, etc.

This is from American Splendor:

Mattress Guy 1: So how smart is she?
Mattress Guy 2: I don’t know. I guess she’s about average.
Mattress Guy 1: Average? Average is dumb!”

[BTW, nothing wrong with being dumb. I’m not dumb about movies, but I’m dumb about plenty of other things. I used to get Cs in French class—do you know how hard that was to do in the 1960s? And then there’s my “computer skills”.]

I often check out middlebrow TV entertainment on Netflix, and almost always give up after a few episodes. Last year it was The Queen’s Gambit and Borgen, but there are 100s of other examples one could cite. These series are generally not bad—they often have decent acting and OK screenplay’s—but no one would say they are works of art. No one will re-watch them in the 22nd century, the way we now re-watch classic films from the 1920s and 1930s.

One possibility is that middlebrow meets two appetites, the thirst for stories and the itch to be exposed to something new, such as the world of chess or European parliamentary politics. The soap opera aspects are sort of like sugar to help the medicine go down.

There’s no point in bemoaning the fact that Netflix offerings are uninspired or that the Academy Awards usually ignores the best films. That’s not their audience! That’s not to say there haven’t been great films with wide appeal, but the whole point of the Academy Awards is to reward the picture with the biggest appeal to the 90% to 99% audience, not the bottom 90% or the top 1%. (And no, Parasite winning last year doesn’t prove me wrong—Burning winning would have proved me wrong.)

A company like Netflix can’t become big and rich by appealing to the audience that likes films directed by Tarkovsky and Hou Hsiao-shien. And do you really expect the Academy to give out Best Picture awards in a way that implicitly acknowledges that most Hollywood directors are not in fact “artists”, and the makes the TV viewers at home feel dumb?

The NYT has 7.5 million subscribers, mostly progressives in the 90-99% range. These people feel very smart, and they are in fact smarter than 90% of the population. So there’s no point bemoaning the fact that the NYT is not about to tell it’s readers that, “Actually, we provide middlebrow news analysis, and if you want brilliant inspired analysis you need to read blogs like SlateStarCodex.”

Yes, the NYT story is awful in all the ways that are currently being discussed by its critics, but the fundamental problem is inescapable. Any time a powerful middlebrow entity (which wrongly thinks it’s highbrow) evaluates an actual highbrow entity, you will end up with a mixture of resentment and incomprehension. This case is no different. It’s just how things work.

Scott Alexander should view this story as a badge of honor. “My insights are so subtle that even the NYT was in over its head trying to figure me out.” I have no doubt that if the NYT tried to evaluate my blog they’d get it right. My stuff is not over their heads. My posts are relentlessly middlebrow in everything other than monetary policy.

But at least I know highbrow when I read it.

PS. Off topic:

Number of Trump impeachments: 2

All other presidents: 2

Senators from his own party that voted to convict: 8

All other presidents: zero



134 Responses to “Understanding middlebrow”

  1. Gravatar of Alan Goldhammer Alan Goldhammer
    13. February 2021 at 13:48

    I’m finding that most of the made for Netflix material these days is rather mediocre. Amazon Studios does a far better job on new show creation. I did like the first half of The Queen’s Gambit but felt the 2nd part was just crude melodrama. Most of the streaming shows I like these days are foreign. I particularly liked ‘Spiral’ (Amazon Prime and MHz for the final two seasons) and ‘Call My Agent’ which is one of the best farces that I’ve seen in some time.

  2. Gravatar of dtoh dtoh
    13. February 2021 at 13:49

    The one thing I’m looking forward to in old age is that the decline in cognitive abilities will allow me to once again enjoy reading the NYT… something I haven’t been able to do since my freshman year in college.

  3. Gravatar of sty.silver sty.silver
    13. February 2021 at 14:07

    Possibly stupid question: do I understand correctly that the divide here into high/middle brow and rest is based on intelligence? Or is it based on knowledge about movies? (Should I read the NYT article to get this?)

    If it’s intelligence, I’m skeptical about the claim that your average classic attracts more viewers now than the Queen’s gambit will in a couple of years, provided we measure this relative to the total number of shows that exist at the time. I could totally see myself re-watching the Queen’s Gambit in 20 years.

    The above paragraph is probably also stupid and I’m just wrong because I don’t know much about movies. But if someone else had written this post, I would suspect a generational effect. I can name very few celebrated classics that I would consider more well written than the Queen’s Gambit.

  4. Gravatar of ssumner ssumner
    13. February 2021 at 14:46

    Alan, I’m not generally a fan of TV series, but I thought “Trapped” and “Babylon Berlin” were decent.

    dtoh, If you are still reading this blog then cognitive decline has probably already set in. 🙂

    Silver, You said:

    “I can name very few celebrated classics that I would consider more well written than the Queen’s Gambit.”

    It didn’t strike me as particularly well written, but I only lasted a few episodes. In any case, I consider film to be primarily a visual art, and the writing is of secondary importance.

    As for intelligence, I’m not a fan of the view that there’s one overall intelligence that carries over into many fields. I’m smart at some things and dumb at others. I’m good at appreciating the visual arts and bad at music, for instance. I see that in lots of other people.

    The only TV show I’ve seen that I could imagine people watching
    in 100 years is Twin Peaks. (And I’m not even sure about that.) But lots of great movies will still be being watched in 100 years.

  5. Gravatar of Rajat Rajat
    13. February 2021 at 16:36

    I know this is orthogonal to the point you are making (has ‘orthogonal’ become a middlebrow word?), but it seems like the top films of the 1920s and 30s appealed to a lot more than 1% of the population. The same for the famous writers and visual artists of the 18th-mid 20th centuries. One explanation is that those works did only appeal to the top 1%, but that because 95% of the population was too poor or unsocialised to care about art, such art seemed popular and could find a relatively large and lucrative market. I always felt that the main reason American tourists had a bad reputation overseas was that America was rich enough for the Clark Griswolds to travel, whereas Europeans and Asians had to be relatively richer and more elite. Another explanation is that those artists produced works that managed to appeal to the top 1% as well as the next x%. But maybe even that is a function of changes in prosperity. When going to the flicks was a weekly treat, people might take the experience more seriously and view films and exhibits with their full attention. Now films and TV are for daily consumption and the sheer range of options require each to provide instant appeal. What are your thoughts?

    Regarding broadsheets of record, I frankly wouldn’t shed a tear if they disappear. I think the only reason to read them is because they can influence (some) politicians who read them thinking that they are influential with the voting public (they aren’t). I would much prefer a world in which instead of paying $50+ a month for a couple of local rags and the FT, I put that into 8 to 10 substacks.

  6. Gravatar of Garrett Garrett
    13. February 2021 at 17:18

    I suspect there are a lot of Scott Alexander readers who also read the NYT. I wonder if they’ll experience any Gell-Mann Amnesia

  7. Gravatar of John S John S
    13. February 2021 at 17:30

    Best show: The Wire, Season 1. A 12-hour film.

    Better Call Saul is pretty damn good and eminently re-watchable.

  8. Gravatar of Jonathan Miller Jonathan Miller
    13. February 2021 at 17:46

    My take, and I am not sure if I agree, is that if 80-99% of people fall under middlebrow. The top 1% are going to be only connoisseurs, or geeks if you will. You can get to middlebrow will intelligence, an open mind and an inquisitive spirit. Highbrow requires time and effort, care and possibly even passion. Basically no one, except the intelligent and idle rich, could be across the board highbrow.

    I think that makes me pretty middlebrow. I do, afterall, enjoy reading The Economist and The Atlantic (and a few blogs, including this one and the Dish), and feel like I am already spending more time on being informed than I should.

  9. Gravatar of Martin Mertens Martin Mertens
    13. February 2021 at 19:17

    People will be watching seasons 1-10 or so of The Simpsons in a 100 years, guaranteed. Another show that comes to mind is Breaking Bad but there I’m not at all certain.

  10. Gravatar of ssumner ssumner
    13. February 2021 at 19:43

    Rajat, Almost all art forms start ought popular and then get more inaccessible over time. The same happened in painting, literature and music. You are correct that the great films from the 1920s and 1930s were more widely popular.

    And the NYT is very good at doing basic reporting, not so good at analysis. The Scott Alexander story is supposed to be analysis, and it’s lousy.

    Garrett, Yes, I’ve had that feeling on a number of occasions. For each media outlet, I divide the stories up into categories where they are reliable and areas where they are unreliable. I wouldn’t rely on The Economist for book or movie reviews, nor would I look to the NYT for intelligent analysis of economic policy issues or debates over woke excesses.

    If the NYT does an attack on the GOP over some issue, I’ll often look at the National Review for comment. If they’ve been unfair then the NR will usually point that out.

    John, I watched a couple episodes of The Wire, and gave up. Not saying it’s bad, but it just didn’t interest me. These shows are a big commitment of time, and at age 65 I don’t know how much time I have left.

    Jonathan, I think highbrow is largely genetic, at least for middle class and above Americans. Obviously, lots of peasants in poor countries never even have a chance to become highbrow.

  11. Gravatar of ssumner ssumner
    13. February 2021 at 19:45

    Martin, I watched two episodes of Breaking Bad and gave up. I haven’t seen The Simpsons, but did see a few animated cartoons its creator did back before the Simpsons.

  12. Gravatar of Todd Kreider Todd Kreider
    13. February 2021 at 19:56

    Scott: “if you want brilliant inspired analysis you need to read blogs like SlateStarCodex”

    While a fan, there is no way that it is “brilliant inspired analysis” when he veers away from his psych knowledge, which is likely solid (I’m not in that area so don’t know) . It is usually pretty good but he goes on and on and on and on….The reason those like you think it’s brilliant is because readers like you almost never have backgrounds in areas that he is discussing.

  13. Gravatar of Philo Philo
    13. February 2021 at 20:03

    Might this be your last comment on Trump? (No, that’s too optimistic–Trump is so damn fascinating!)

  14. Gravatar of Todd Kreider Todd Kreider
    13. February 2021 at 20:07

    Scott: As for intelligence, I’m not a fan of the view that there’s one overall intelligence that carries over into many fields.

    Well, those who study intelligence overwhelmingly say you are wrong even if you are not a fan of decades of their research. You are confusing intelligence for interest and experience. There is clearly a core g factor at work. The funny thing is that you think you know more than all of those thousands of researchers as “not a fan of”

  15. Gravatar of mbka mbka
    13. February 2021 at 20:11


    your blog here is definitely high-brow except for anything about Trump. That’s not really your fault either, with Trump there’s no there there. Trump is self-parodizing and self-satirizing. As the usually biting satiricist Karl Kraus had said in the 30’s, “As towards Mr. Hitler, I really don’t know what to say.” But it was entertaining.

    As for series… I second the mention of non-US series these days. “Call my Agent” was breathtaking. The French title is much better btw, “Ten percent”. We burned through it real fast at home. Then right after, on the same vague theme of professions burdened with inextricable machiavellism and personal politics, we tried “Suits”. What a letdown. The difference was just shocking, we barely made it through the pilot. “Call my Agent” is eminently believable and close to life, with just the necessary compression of drama into a short time and place. I can’t recall a single standard TV trope in it. “Suits” is full of tired tropes and people acting like no one would in real life.

    I’d still think that “Breaking Bad” will retain classic status too. And a few others. But more generally, yes, Netflix series are mostly of one predictable mould, well-produced blandness and the second season is usually the end of it anyway.

    On the NYT, very apt description, a mixture of resentment and incomprehension because fundamentally, the NYT doesn’t get it. And the belabored fishing for connections to non-PC sources. Sometimes I think the internet really is the end of free speech. It makes witch hunting so easy that soon enough no one of import dares say a word.

  16. Gravatar of anon/portly anon/portly
    13. February 2021 at 21:27

    “My posts are relentlessly middlebrow in everything other than monetary policy.”

    Yeah, except this comment is embedded in a post that’s not about monetary policy and is a very subtle and to my mind at least highbrow blog post. The NYT = middlebrow framing is very clever.

    This framing does miss something, though, which is that the NYT is also a bureaucracy. The SSC article is very much the product of bureaucratic mindset. And not just because it reads like it was written by a committee.

  17. Gravatar of Mark Z Mark Z
    14. February 2021 at 00:56

    “Almost all art forms start ought popular and then get more inaccessible over time.”
    Isn’t this a concession that art that appeals to the top 1% is less likely to be immortalized than art that appeals to the top 10%? That snobs in the 22nd century will more likely be watching Scorsese films than Tarkovsky films?

    I also would be careful to avoid overrating the taste of the ostensible top 1%. Someone whose favorite painter is Mondrian may be more likely in the top 1% than someone whose favorite is Van Gogh, but often people who self-identify (perhaps correctly) as having elite taste or intelligence ‘choose to like’ things that few other people like. But however cliche it is to like Van Gogh (or hip it is to like Mondrian), Van Gogh > Mondrian. Or maybe I just think that because I’m a philistine. I guess I’m less interested in your specific point about the NYT than about your general point.

  18. Gravatar of Mark Z Mark Z
    14. February 2021 at 01:24

    Having read up on it, I think you’re being too charitable to the NYT (or uncharitable if one values intelligence more than honesty). I think they know exactly what they’re doing, and are counting on having a much bigger audience than Scott to allow them to shape opinion irrespective of what Scott has actually written or actually believes.

  19. Gravatar of A A
    14. February 2021 at 01:59

    I think middlebrow work will retain their powers longer than you think. Accessibility increases the value of comfort, nostalgia, distraction, and other modest mood improvements. You get your hit through a device in your pocket, and nostalgia allows older content to dial into specific feelings from when your first viewings.
    Happy Madison productions, Friends, The Office… all valuable for small elevations within stressful lives.

  20. Gravatar of Russell Hogg Russell Hogg
    14. February 2021 at 03:28

    You mentioned the films Parasite and Burning. Coughs modestly . . .here is a podcast that I did a few weeks ago with Alex Tabarrok (and Chicago philosopher Agnes Callard and her son) on these films. I think everyone is on good form and Alex’s take is a hoot. I preferred Burning to Parasite.


    It is a great podcast!

    PS Hooray for Scott and down with the NYT!

  21. Gravatar of Thiago Ribeiro Thiago Ribeiro
    14. February 2021 at 04:20

    People will still be watching Star Trek, Doctor Who, The Wire, M.A.S.H and The Sopranos way after having forgotten Tarkovsky.

  22. Gravatar of John S John S
    14. February 2021 at 06:25

    Makes sense, it’s a verrrry slow burn. I’m amazed HBO greenlit the script way back in the pre-streaming days.

    You might like Black Mirror (it’s an anthology series, so no big time commitment). Think Twilight Zone updated to project current socio-technological trends to their terrifying logical ends.

    My favorite eps are “San Junipero” and “White Christmas.”

  23. Gravatar of John S John S
    14. February 2021 at 06:50

    The Beatles: middlebrow or highbrow?

    Obv huge popular appeal, but there have been reams of scholarship on the purely musical aspects, and the harmonies are about the peak of perfection that pop music can ever reach (perhaps tied with the Beach Boys, but I prefer the much more varied vocal textures of John/Paul/George).

    I once read (wrt 3-cushion billiards) that a truly great game is one which can mentally challenge you every day for the rest of your life. For me, the Beatles catalog can satisfy that requirement at any level from chill-out ambience to intense listening and reflection. I mean, how lovely is the wobbly, wavering key in the intro to “If I Fell” that reflects John’s ambivalence about commitment?


  24. Gravatar of Peter Schaeffer Peter Schaeffer
    14. February 2021 at 10:05

    Back when Andrew Johnson was impeached and tried, the Republicans in the Senate had a huge majority. However, many (7) Republicans couldn’t stomach the charges against Johnson (nor could the Supreme Court). Johnson survived by one vote. By contrast, Trump survived by 10 votes. Why?

    Perhaps it is because the modern Democratic party is committed to a list of things Americans can’t stand

    Pervasive Political Correctness
    “Defund the Police” – Perhaps the worst political slogan in history
    Boys should be able to compete with (and beat) girls if they want to – Disgusting to say the least
    Critical Race Theory – I have a dream that my daughters will be judge by the color of this skin

  25. Gravatar of Peter Schaeffer Peter Schaeffer
    14. February 2021 at 10:53

    TK, Scott also ‘knows’ that the 95% fall in Tulip prices in 1637 was not a bubble (bursting). The medieval church knew that the sun orbited the earth and of course, they were right.

  26. Gravatar of Michael Rulle Michael Rulle
    14. February 2021 at 10:55

    I like this essay. I am definitely not a highbrow watcher. I am a pretty strong fan of middlebrow. But I know it is middlebrow— not literally, that’s your term. But by the examples you give. I have to hunt down your high brows. Maybe I could learn something.

    For example, I watched a YouTube 60s video of John Cummings discussing jazz (Chic Corea just died—so was looking for some jazz info) and while I am not a big jazz fan, his discussion was amazing. But while he made me understand it’s nature far better in an intellectual sense, I don’t have the ear or experience to truly hear or enjoy the music any better. Maybe same with your hb movies. Nor did he persuade me, which he tried to, that “popular” music is like arithmetic to Jazz’ advanced Calculus. But I was not Unpersuaded either.

    For example, I enjoyed much to my surprise, The Queens Gambit. I liked the corny fantasy of it all. It was a feel good unrealistic story (not just the female character as that is “somewhat” plausible—-with luck, who knows maybe the best ever female —-eg Judit Polgar (sp?) could have pulled it off).

    But I loved that a young American girl could be applauded by old Russian men in a Chess Park in the early 80s Soviet Union. But I am a father of 2 daughters so part of my enjoyment maybe was impacted by that.

    I need to check out some of your high brows—see what I think.

    I do have to say, I laughed out loud at your Trump comment. I could almost believe you wrote that just for me. Meaning, you couldn’t actually feel the need to mention Trump in this essay—- could you?

  27. Gravatar of Scott Sumner Scott Sumner
    14. February 2021 at 11:01

    Todd, I don’t think it’s just “experience” I’ve spent long hours learning some things and utterly failed, such as foreign languages. Other things like math I picked up without even studying. People have different kinds of intelligence. I can’t remember anyone’s name, but have a photographic memory for maps. Other people are the opposite.

    I certainly understand that there’s a positive correlation between SAT scores on math and verbal, but that doesn’t prove the existence of “g”. It’s just a model to make sense of data.

    As for SlateStarCodex, I’m not at all surprised that it’s well over your head.

    Philo, We aren’t even at the half way point. Wait until he runs in 2024, then you’ll see me go nuts!

    mbka, I’m hoping that witch hunting will eventually bore people, and they’ll start tuning out claims that so and so is evil.`

    anon, Good point about the NYT being bureaucratic. They do have some individual people who are good, but no one in Alexander’s class.

    Mark, I just rewatched Taxi Driver. It’s good, but I’m not sure people will watch it in 100 years. Scorcese’s on the borderline between popular and high art, so it’s hard for me to say how he will hold up. Maybe Raging Bull and Goodfellas. Tarkovsky is an immortal, read the NYT piece on him that Tyler linked to yesterday.

    Van Gogh and Mondrian are a bad comparison. First, because Mondrian came later, and second because Van Gogh is both a popular artist and well regarded by the elites (as are some impressionists). Compare Picasso to more traditional painters from the early 1900s. Who was popular then, and who has stood the test of time?

    As I said in the post, some great art does have wide appeal. Raphael, Mozart, Hitchcock, etc.

    As for being too charitable to the NYT, I said:

    “Yes, the NYT story is awful in all the ways that are currently being discussed by its critics,”

    Critics pointed out that the story was dishonest, and I agree.

    A, I think The Office is pretty good, better than most of the “serious” drama on Netflix. But will people watch it in 100 years? I still enjoy Keaton and Chaplin films from the 1920s.

    John, I did a see a couple Black Mirror episodes, and it is indeed a better version of Twilight Zone.

    Peter, LOL, in a secret ballot Trump would have been convicted 90 to 10. Do you actually think any of this has to do with political principles?

  28. Gravatar of Jonathan Miller Jonathan Miller
    14. February 2021 at 11:06

    If by genetic, you mean inherited through DNA/etc, then I really don’t understand the mechanism. If you mean instead inherited through nurture, then I think it clearly plays a role.

    To provide another example, in graduate school I started getting into chocolate. I couldn’t stand Hershey’s anymore and after a point was making frequent visits to the local Leonidas shop. As a postdoc, I lived for a time in Brussels, and despite Leonidas being easily accessible, I travelled to the Pierre Marcolini shop. Lindt was hardly acceptable for my taste. Regrettably, I have at times lived in chocolate deserts in the past decade and have at times been very happy to find Lindt.

    As a father of two small children, if we moved back to Brussels (very unlikely), I do not expect I would travel often to the Pierre Marcolini shop. It is not money, I am much better off than I was as a postdoc (or an academic period), rather it is the time.

  29. Gravatar of ssumner ssumner
    14. February 2021 at 11:10

    Russell, Yes, I recall listening to that podcast–it was very good. BTW, I enjoyed both films. I did a post making the same point as Alex did about Parasite, but he’s post was 100 times better.

    Thiago, LOL.

    John, The Beatles have very broad appeal, and indeed that’s true of lots of great art. The art with broad appeal tends to come early in an art form, and the Beatles were relatively early in rock music. The best rock bands today (say Radiohead?) have much less broad appeal.

    But in general, it’s possible for something to appeal to both highbrows and middlebrows.

    BTW, what the Beatles are for you, Dylan is for me.

    Michael, With music I’m on the border between middlebrow and lowbrow. Much of jazz and classical goes right over my head.

  30. Gravatar of ssumner ssumner
    14. February 2021 at 11:14

    Jonathan, You said:

    “If by genetic, you mean inherited through DNA/etc, then I really don’t understand the mechanism.”

    So Einstein was “nurtured” better than someone with Down’s Syndrome? Okaaay . . .

    Seriously, obviously nurture matters, sometimes quite a bit. But genes also matter.

  31. Gravatar of Kevin Erdmann Kevin Erdmann
    14. February 2021 at 11:40

    I would second Martin on the Simpsons. For instance, there is an episode from the 2nd season called “Itchy &Scratchy & Marge” where Marge tries to get violence out of children’s cartoons. It is very much like a SSC post on the issue, posing as a crude half-hour satire.

  32. Gravatar of Martin Mertens Martin Mertens
    14. February 2021 at 12:09

    Thanks, Kevin. All that said, I have no idea if The Simpsons are high brow. I really need to watch some of these movies to see what Scott means.

  33. Gravatar of Todd Kreider Todd Kreider
    14. February 2021 at 12:26

    Typically, Scott wrote: “As for SlateStarCodex, I’m not at all surprised that it’s well over your head.”

    As I said, I am a fan of SSC, but he will write long blog posts on areas he doesn’t understand well, so I skip those once I realize he has veered off what he does know.

    You must have a low standard with studying math considering my cousin spent countless hours studying math as a kid and later as an adult as a Ph.D. mathematician.

  34. Gravatar of Ken P Ken P
    14. February 2021 at 12:28

    I think referring to NYT (or WSJ, etc.) as the 90-99% is very charitable. These are the dinosaurs of media and quickly becoming extinct, partly because they are the fast food outlets of journalism. Substack, blogs, and podcasts are now attracting those who want more than surface level journalism.

    Joe Rogan drew a million viewers for an hour long interview of Roger Penrose. Maybe Joe Rogan is highbrow, but I think it’s more likely that you are underestimating the intellectual curiosity of average people.

  35. Gravatar of Jonathan Miller Jonathan Miller
    14. February 2021 at 12:42

    But Scott, you are talking not about intelligence or creativity or deep understanding, rather the ability or preference to appreciate highbrow art or conversation rather than middlebrow.

    This is pretty clearly not a general function of intelligence, or creativity, as I think has already been discussed both in your previous posts and the other posts in this discussion. I am not saying it clearly does not require intelligence or creativity or sensitivity, just that it is not the primary component.

    I guess I could imagine some point, where some combination of traits end up pushing one to highbrow rather than middlebrow preferences on whatever subject manner one were to select, but it would be very hard to maintain that they were primarily or even substantially genetic. Even if one were able to come up with a set of traits with strong genetic components that push one towards highbrow film, I can’t imagine that they would be similar, in a general way, to the traits that would push one towards highbrow political discussion.

  36. Gravatar of Todd Kreider Todd Kreider
    14. February 2021 at 12:42

    I read a couple of Alexander’s blog from last year including one on the virus where he brings up face masks in April: “Common sense said that they worked. But there weren’t many good RCTs [Randomly Controlled Trials].”

    Not many, just ten, and those were used around the world by health departments for years to advise not wearing masks. Alexander was writing a quick but long post and so he decided it wasn’t worth it to look that up.

  37. Gravatar of ssumner ssumner
    14. February 2021 at 13:28

    Ken, I am curious about what Penrose has to say (once read one of his books) but I’m not smart enough to understand it. Hardly anyone is highbrow in physics, beyond bloggers like Yudkowsky, Aaronson, Hanson, etc.

    Jonathan, I consider the ability to appreciate art as a form of intelligence, which reflects both nature and nurture. Thus I’m an intelligent consumer of paintings but not opera.

    I feel confident that my intelligence skews more visual than verbal due to genes, but I have no hard proof.

    Todd, An anti-masker to the bitter end.

  38. Gravatar of Carl Carl
    14. February 2021 at 13:29

    What is considered middlebrow in one age can become highbrow in another. Take opera, which was middlebrow when it started in Italy but it heavily shaped the themes of classical music, the most highbrow of musical art forms. And opera itself is now considered highbrow art. And, the highbrowed today might miss the fact that the great composers of today are writing for today’s middlebrow art form, the movies.

    Likewise, Shakespeare wrote with an eye to the groundlings as well as the nobility. He was not, in his time, considered better than his contemporaries, Johnson and Marlowe. But with time he became the acknowledged greatest playwright of all time.

  39. Gravatar of David S David S
    14. February 2021 at 14:40

    Scott, you might be reacting to the sublime devastation of Southern California. The beaches, the good air, and the utter shallowness of the world that’s been created there has made you hardboiled and miserable. Thus, you seek out artistic diversions that are far off the mainstream.
    Did you have such extreme standards in New England?

    For the hell of it, re-watch Chinatown.

  40. Gravatar of Sean Sean
    14. February 2021 at 15:45

    Thanks for making me feel like a 1% person. But yes agree the 1% are blogs and not papers. Politicially it’s interesting to breakdown voting patterns. If I had to guess the bottom 25% are heavy dem. 25 to 90 heavy gop. 90-99 dem lean and 1% are all over the map.

  41. Gravatar of John S John S
    14. February 2021 at 15:59

    Tangentially-related: Does anyone know what the hell happened to The Economist?

    Ten years ago, I would take it on a plane and savor it from cover to cover. Even the one-columns at least offered some tidbit of analysis, and the special features comprised a thoughtful, integrated discussion. (Highbrow? Who knows, but there was little else to fill the gap btw newspapers and academic journals.)

    Sometime around five years ago, the average article quality took a nosedive to just barely above Time magazine levels, i.e. a news summary with little to no commentary. Here’s an article I picked at random:


    This feels like it was dutifully written by a smart middle-schooler for homework, and I fail to see any useful info beyond the title.

    Anyone know where I can find something akin to the old Economist? (SSC/ACT is a bit wordy for my taste, and SA used to be pretty strict about avoiding the news.)

  42. Gravatar of Christoph Breuer Christoph Breuer
    14. February 2021 at 17:15

    Thanks Scott,I only got here due to the link by MarginalRevolution and I didn’t know your blog until now. I never understood why a newspaper so higly regarded like the NYT could feature long reviews about Quality TV Shows (Borgen, House of Cards,Queen’s gambit) which are honestly just modern-day Soap-Operas (better actors though) without any artistic value. While at the same time they more or less ignore brilliant Artists like Lars von trier. You gave me an answer.

    Robin Hanson once wrote “~80% is the society IQ percentile that matters most. Lower folks defer up and higher folks mostly win by pandering down”

  43. Gravatar of Michael Rulle Michael Rulle
    14. February 2021 at 18:01

    Beatles were my favorite group ever. Still so today. However, my favorite individual for entirely different musical reasons is Bob Dylan. I still am amazed at how many people covered his songs (e.g., Joan Osborne released “Man in the Long Black Coat” in ‘95–written by Dylan in 89).

    I stopped buying his albums after 1980 or so——his voice became unlistenable. I got reintroduced by Scorsese’s 2005 doc— No Direction Home——which I found mesmerizing. Have watched it 4-5 times. Liked his 2019 doc—-but not nearly as much.

    If you want to see something really fun and almost surreal——search YouTube for a 45 minute amateur movie taken from the front seat of a London taxi in 1966 of Lennon and Dylan alone in the back seat—-just talking. Lennon almost seemed like the grown up—he was a year older.:-)

  44. Gravatar of Peter Schaeffer Peter Schaeffer
    14. February 2021 at 19:00

    Jonathan, height is a good model (example) of how genetic (DNA) inheritance works. Clearly many genes are involved (height is clearly polygenic). Human heights have risen substantially over the last 500 years, as medicine and diets have improved. The reverse was true after the Roman Empire fell. Human heights declined. However, even in Roman times it was generally known that Northern Europeans were taller (on average) than Southern Europeans. Of course, this was big problem for the Roman legions. They tended to prevail by virtue of better discipline, better tactics, better logistics, better weapons, and better armor. However, the Northerners were taller, even back then. Note that the N-S height difference still exists. The N-S height delta was genetic back then and still is. The increase in heights over the last 500 years was nurture.

  45. Gravatar of Peter Schaeffer Peter Schaeffer
    14. February 2021 at 19:10

    John Roberts refused to have anything to do with the second Trump trial. Why would that be? Because the process was a mockery of the Constitution.

    You don’t get the extent to which Americans are outraged by PC BS. However, the fact that the Rs gained seats in House in 2020 should tell you something. The public was rightfully outraged by

    Pervasive Political Correctness
    “Defund the Police” – The worst political slogan in history
    Russiagate – Was a hoax and it cost the Democrats dearly
    Critical Race Theory – I have a dream that my four little children will one day live in a nation where they will be judged by the color of their skin
    Boys should be able to beat girls in sports – And you are a transphobe if you don’t agree

    I believe that a secret ballot would have produced a different result. Perhaps 10 votes for conviction and 90 for exoneration. However, Democratic Senators are terrified of the ‘woke’ and voted accordingly.

  46. Gravatar of Ken P Ken P
    14. February 2021 at 20:38

    “Ken, I am curious about what Penrose has to say (once read one of his books) but I’m not smart enough to understand it. Hardly anyone is highbrow in physics, beyond bloggers like Yudkowsky, Aaronson, Hanson, etc.”

    Scott, I wouldn’t consider these podcasts highbrow, but that’s my point. They are still above NYT and other outlets in the paper belt and way more popular.

    I haven’t seen the Rogan interview but watched him interviewed by Lex Fridman. That one is very good. I suspect the popularity is due to philosophical questions about AI, not physics.

  47. Gravatar of Ken P Ken P
    14. February 2021 at 20:40

    PS. Lex Fridman’s interview of Scott Aaronson is good, too.

  48. Gravatar of Jonathan Miller Jonathan Miller
    14. February 2021 at 21:10

    I was aware of the role of genetic inheritance in the trait of height.

    But given the bread of the the domains where there exists highbrow and middlebrow expression, the complexity of genetic traits, and the existence and role of nurture, environmental and cultural (and other non-genetic traits, even if they are passed down or inherited), doesn’t that suggest that the usual determination of whether a given individual appreciates or prefers highbrow versus middlebrow in that given domain depends on environmental or other non-genetic factors?

    The question of highbrow versus middlebrow is not the question of Einstein versus the down syndrome individual (to my knowledge, there are no Down syndrome individuals who are physicists even, much less the very best of physicists).

    If it was a domain where there was clear correlation with one set of traits, like intelligence for a successful physicist versus a non-physicist, then you could argue that the key traits were genetic. But this isn’t the case for the division between middlebrow and highbrow.

  49. Gravatar of education realist education realist
    14. February 2021 at 21:39


    “I’ve spent long hours learning some things and utterly failed, such as foreign languages.”

    Actually, while IQ helps with reading comprehension in foreign language, it is unrelated to listening and conversation skills in a second language.

    So your inability to learn a foreign language or bad memory for faces doesn’t disprove the existence of g.

    Actually just came here because of the middlebrow convo and talk of SSC. I’m with those who thought he was plenty smart, not a genius, and often didn’t have a clue what he was talking about (like my bailiwick, education). He would often do deep dives that were simply absurd mental masturbation. And his fiction was, er, an acquired taste.

  50. Gravatar of ssumner ssumner
    14. February 2021 at 22:54

    David, You said:

    “Thus, you seek out artistic diversions that are far off the mainstream. Did you have such extreme standards in New England?”

    I’ve been doing “best of the year” film posts since 2010, if you go back and read them you’ll find I’ve always been drawn to art films. It’s what I like; why shouldn’t I like great films? No need to come up with an explanation like you are trying to diagnose a mental illness.

    And yes, Chinatown is excellent.

    John, I still read it, but its golden age is over.

    Christoph, Lars Von Trier is great. The Kingdom is one of the few TV series that qualifies as a work of art.

    Michael, Agree about No Direction Home—a great documentary.

    His best stuff was before 1980, but check out Time Out of Mind. And there are some other good ones as well.

    Realist, You said:

    “So your inability to learn a foreign language or bad memory for faces doesn’t disprove the existence of g.”

    That’s exactly the problem. Anything that doesn’t match the “g” theory is just waved away as not intelligence. You’ve created a model that’s true by definition. In my view, intelligence applies to all aspects of life, not just test taking. One can be intelligent at art, sports, music, hunting and fishing, carpentry, interpersonal relations, philosophy, reading others’ emotions, child care, etc.

    It’s usually people who are good at test taking that devalue the other forms of intelligence.

    You said:

    “He would often do deep dives that were simply absurd mental masturbation.”

    I’m totally unimpressed when people say he was overrated. I’ve never seen another blogger who did what he did better than Scott Alexander. I conclude the people who think he’s overrated just didn’t get where he was going with his posts, which were mostly brilliant.

    His fiction? That’s another story.

  51. Gravatar of ankh ankh
    14. February 2021 at 23:02

    I’d like to weigh in on the middlebrow talk.

    1. Alexander Hamilton wrote the federalist papers when he was 30. As such, he would almost certainly consider you “middlebrow”, and a total bore for your sophomoric and often times incoherent arguments on government policy. Your conception of middlebrow presupposes that you are in the highbrow.

    2. Considering yourself to be highbrow is rooted in an egotistical arrogance that manifests itself in those who are generally insecure. We also see this behavior in people who have a general predisposition towards abusing power (corruption), and who are borderline megalomaniacs/sociopaths. Interestingly, this type of thinking is also seen in people who define themselves as utilitarian: meaning, they believe in no fundamental universals, and rarely value the individual. For these people, human life is often times expendable.

  52. Gravatar of postkey postkey
    15. February 2021 at 01:18

    Is it ‘middlebrow’ to totally ignore ‘this’:

    ‘We’ have ten years?
    “ . . . our best estimate is that the net energy
    33:33 per barrel available for the global
    33:36 economy was about eight percent
    33:38 and that in over the next few years it
    33:42 will go down to zero percent
    33:44 uh best estimate at the moment is that
    33:46 actually the
    33:47 per average barrel of sweet crude
    33:51 uh we had the zero percent around 2022
    33:56 but there are ways and means of
    33:58 extending that so to be on the safe side
    34:00 here on our diagram
    34:02 we say that zero percent is definitely
    34:05 around 2030 . . .
    34:43 need net energy from oil and [if] it goes
    34:46 down to zero
    34:48 uh well we have collapsed not just
    34:50 collapse of the oil industry
    34:52 we have collapsed globally of the global
    34:54 industrial civilization this is what we
    34:56 are looking at at the moment . . . “


    and ‘this’?

    ” . . . Multivariate analysis indicates that economic elites and organized groups representing business interests have substantial independent impacts on U.S. government policy, while average citizens and mass-based interest groups have little or no independent influence. The results provide substantial support for theories of Economic-Elite Domination and for theories of Biased Pluralism, but not for theories of Majoritarian Electoral Democracy or Majoritarian Pluralism. “

  53. Gravatar of mbka mbka
    15. February 2021 at 02:58


    “That’s exactly the problem. Anything that doesn’t match the “g” theory is just waved away as not intelligence. You’ve created a model that’s true by definition. In my view, intelligence applies to all aspects of life, not just test taking. One can be intelligent at art, sports, music, hunting and fishing, carpentry, interpersonal relations, philosophy, reading others’ emotions, child care, etc.

    It’s usually people who are good at test taking that devalue the other forms of intelligence.”

    I completely agree with that. It’s one of my quibbles with the IQ debate in general. It also segues into rationalist hubris, the definition of AI and what qualifies as it, the belief that consciousness is somehow magically related to processing power etc.

    My own personal definition of intelligence is the ability to act purposefully under one’s own agency in an environment that one adapts to. Once you accept that, you’d have to conclude that beings w/o computing power of any kind, say, bacteria, could be classified as “intelligent”, and things such as powerful supercomputers with great software simulating, say, being a human, would not meet that set of criteria.

    Speaking of rationalists, while I love Scott Alexander, I do think the rationalists have a bling spot here.

  54. Gravatar of John S John S
    15. February 2021 at 03:03

    Michael Rulle, that taxi conversation isn’t “almost surreal” — it’s fucking nuts. The longer I live, the more amazed I am at how much weird shit has happened on this planet.

    (Amusingly, at 5:35 John says, “Bob, tell me about the Mamas and the Papas. I hear you’re supporting them bigly.”)

    One thing that strikes me is the asymmetry of cultural knowledge/sophistication btw Americans and Brits, like when Dylan talks abt the “mighty Thames” that protected Britain from Hitler and everyone in the car is thinking “WTF?”

    Maybe that was just the drugs talking, but I find it interesting that there is a noticeable gap in the ability of Americans and Brits to engage in banter (still seen when US celebrities go on British chat shows and appear at times to be borderline mentally challenged). This is not to say that Brits are smarter individually than Americans, but there does seem to be a kind of collective social intelligence there that lifts all boats. (Roger Ebert noticed this in the “Up” documentaries.)

    Wonder what the cause is — maybe it’s pub culture, which really doesn’t exist here in the States. Anyway, great find.

  55. Gravatar of Spencer B Hall Spencer B Hall
    15. February 2021 at 04:12

    What is astonishing is how stupid economists are. The FED is actually conducting a contractionary money policy, one where the velocity of circulation inevitably falls. The remuneration rate is above all 5 year Daily Treasury Yield Curve rates.

    The September 2019 Repo Crisis (where the remuneration rate vs. SOFR rates, Secured Overnight Financing Rate – a broad measure of the cost of borrowing cash overnight collateralized by Treasury securities was inverted) is prima facie evidence. I.e., the FED has artificially inverted the remuneration rate vs. wholesale money market funding rates.

    It is now widely accepted that the GFC represented a credit crunch (Bernanke was wrong at first, saying that it represented a capital crunch – hence TARP). But the response to the 1966 credit crunch (where the term was coined), was just the opposite of the GFC’s, Ben Bernanke’s, response.

    Whereas the 1966 Interest Rate Adjustment Act created a .50% interest rate differential in favor of the Savings and Loan Associations (the thrifts, the nonbanks), the Emergency Economic Stabilization Act of 2008 provided a preferential interest rate differential in favor of the commercial banks, which induced nonbank disintermediation (where the size of the nonbanks shrank by $6.2 trillion dollars, while the banks were unaffected, increasing by $3.6 trillion dollars). Funny, because the banks don’t loan out existing deposits, or savings.

    The nonbanks were rolling over very short-term liabilities to fund longer term real-estate assets, thus their funding was abjectly and instantaneously, thereby removed. I.e., the GFC was not a capital crunch as Ben Bernanke surmised.

    Re: “However, the “return to economic normality” faces immense challenges”

    Not so. Just drive the commercial banks out of the savings business, i.e., release otherwise frozen savings.

  56. Gravatar of Spencer B Hall Spencer B Hall
    15. February 2021 at 04:29

    re: “You can’t force banks to lend.” – Tim Geithner

    LOL. That’s exactly what they did between 1942 and 2008.

  57. Gravatar of Spencer B Hall Spencer B Hall
    15. February 2021 at 05:06

    “Lawrence Yun, chief economist at NAR, concluded: ‘The average working family is struggling to contend with home prices that are rising faster than income. This sidelines a consumer from becoming an actual buyer, causing them to miss out on accumulating wealth from homeownership.’ ”

  58. Gravatar of Steve Steve
    15. February 2021 at 05:12

    The more I read crappy blogs on the internet the more I miss Christopher Hitchens.

  59. Gravatar of Michael Rulle Michael Rulle
    15. February 2021 at 06:24


    Yes, did like Christopher Hitchens—-one trait I liked was his ability to not attach himself to his own arguments—-was self critical too. Hitchens did not have a blog—-I don’t think.—-so I could say——the more I read crappy blogs the more I miss HL Mencken

    @John S

    I need to see it again ——I like your “bigly” reference. Funny. Also, I agree there is something interesting about the relationships between Americans and Brits. I sometimes think I fall victim to their accent, which would be amusing.

    However, my perception is that, on the one hand, we have many overlapping values——which I suppose is not surprising—-but I believe they have a deeper sense of history, regardless of class. That should not be surprising either. I sometimes believe we are almost “history free”—-and our “motor” can only move forward. And yes—-there seems to be a broader—-not intelligence per se (although maybe)——but greater awareness by formal and informal education. We really are a country of immigrants——-they are a country that has stood in place for almost a millennium. Anyway—-I will watch for that element. Interesting comment by Dylan on the Thames!

    @ ankh

    Scott referenced his Highbrow nature as it relates to Monetary Policy—-and Movies. That is it.

    What does that have to do with Hamilton? I am often astounded how little most of us know about great historical thinkers. They created or were the first to be aware of almost everything we know today. We are the ones who dig deep narrowly in response to their polymath abilities.

    But Hamilton was not a polymath. His greatest ideas related to his view of the proper administrative role of the state. His thinking was also of the pragmatic and political style. Why do you need Hamilton to assert that Scott is “arrogant”? Speak for yourself.

  60. Gravatar of John S John S
    15. February 2021 at 06:49

    mbka, you wrote:

    “My own personal definition of intelligence is the ability to act purposefully under one’s own agency in an environment that one adapts to.”

    I’d like to a take a crack at the bacteria vs. supercomputers question of intelligence, but first would you mind clarifying precisely what you mean by “under one’s own agency”?

  61. Gravatar of John S John S
    15. February 2021 at 06:56

    With all the Hamilton references, I’d just like to put in a quick plug for Ron Chernow’s biography. IMO, he’s the undisputed master of the genre; each chapter is like a mini-book. (Deffo highbrow, lol)

    Chapter 1 starts with a bang — if it were a TV series, I’d shut it off halfway for being ridiculously unbelievable.

  62. Gravatar of Carl Carl
    15. February 2021 at 07:46

    You defined intelligence as “the ability to act purposefully under one’s own agency in an environment that one adapts to.”
    Do you include the workings of the subconscious under the definition of “purposefully?” To reference one of the sort of middlebrow books I read, what do you make of the sort of intelligence that Malcolm Gladwell writes about in “Blink?”

  63. Gravatar of Bob Bob
    15. February 2021 at 08:33

    If there’s something we know about the longevity of art, is that we are not very good at predicting what will resonate in the future, as those future people might have very different ideas of quality, whether we like said ideas or not. Why do we keep so much music from 18th century Austria? Was it really that much better? Did people back then know that this was what was going to stay?

    Similar things happen in literature. Jane Austen’s books are not exactly high brow in a literary sense, and yet she has outlasted most of her contemporaries, even though few would have guessed this while she lived. And films are a lot more like books than classical music, as there is no doubt that as culture changes, our preferences for what stories we want to retell and rewatch will also change.

    So while the vast majority of the Netflix productions are barely eve middlebrow, who knows what will live 50 years?

  64. Gravatar of Ben M Ben M
    15. February 2021 at 10:12

    The best thinking on “middlebrow” in all its forms comes from a midcentury essay by Dwight McDonald, “Masscult and Midcult”: http://xroads.virginia.edu/~DRBR/macdonald.pdf

  65. Gravatar of Jerry Brown Jerry Brown
    15. February 2021 at 14:06

    Off topic- at your other blog you had provided ‘part 1’ of an analysis of MMT. Is ‘Part 2″ forthcoming, or did I already miss that?

  66. Gravatar of sean sean
    15. February 2021 at 14:31

    Is Trump being impeached twice an indictment of Trump or an indictment of TDS and how bad and illiberal the left has become.

    The first impeachment had nothing in it. The second one I would have convicted prior to 2020, but his rhetoric wasn’t anything different than Pelosi or any leftist rhetoric during the summer. I’d convicted all or none.

    The left has turned impeachment into just a normal political game. Parliamentary system and not guilty/innocent, but our party or their party game.

  67. Gravatar of sean sean
    15. February 2021 at 14:31

    And fwiw the right will likely impeach biden the second they have the votes.

  68. Gravatar of ssumner ssumner
    15. February 2021 at 15:54

    mbka, Interesting. “Agency” seems like a hard concept to define. What about AI?

    Bob, I agree that a few will hold up over time. But I’d also say that it’s possible to guess what sort of thing has a chance. Not much on Netflix has a chance.

    If you look at a 20-year old list of the 100 greatest films of the 20th century, and a current list, they are quite similar. Most of those considered great are likely to preserve that status, but a few surprises will be added over time.

    Jerry, Part 2 is in March. I wish it were all done at once.

    Sean, The scandal is that he was impeached only twice. There were many other occasions where he deserved to be impeached.

    You are right that impeachments will become common in the future, as banana republics have many impeachments.

    And LOL at people who think 7 Republican senators have TDS, and probably 40 if it had been a secret ballot.

  69. Gravatar of mbka mbka
    15. February 2021 at 18:19

    Scott, Carl,

    I’m happy to admit that my definition is a work in progress. Maybe it’s not yet ready for prime time in terms of philosophical rigour. My thought process on it started with defining “intelligence” as an adaptation to survival in an environment by choosing appropriate means. I then wanted to expand this to allow for action that’s not simply reflexive and automatic – so I needed an expansion that allows for learning, and for de novo creation of action that’s not necessarily pre-determined by the existing environment at all (example: production of art). Hence the addition of “own agency” and “purposefully”. To me, agency is an endogenous drive to action through an endogenous purpose. So, agency and purpose are two sides of the same coin. I do not see either of them in any software – hardware combination (say, as in AI). I don’t actually think AI should be called “intelligent”. It doesn’t act out of its own will in some environment.

    “Agency” as endogenous drive could be called something else, say, the will to live, or creativity – anything that goes beyond the automatic deterministic run of an algorithm. Although, given the halting problem, I’d say that even algorithms are not actually deterministic: the halting problem states that you can’t prove for sure what an algorithm will do. But that discussion is orthogonal to intelligence. That said, I’d still exclude AI from my definition of “intelligence” for lack of anything endogenous in it. It does not make itself, it doesn’t produce its own means and ends. A bacterium does: it “wants” to survive in an environment (purpose), and does so driven by nothing else but itself (agency). Doing so, it sometimes creates evolutionary novelty, and during normal times, adapts successfully to changing environments. It is also the culmination of billions of years of evolutionary learning about its environment. So, I would call this intelligence too. Witness the problems we have with Covid: If we stick to the computational definition of intelligence, the virus is as dumb as a rock. Yet our best and highest – IQ scientists and supercomputers have a very hard time dealing with it.

  70. Gravatar of Jerry Brown Jerry Brown
    15. February 2021 at 18:38

    Thanks Scott.

  71. Gravatar of Todd Kreider Todd Kreider
    15. February 2021 at 20:01

    Scott: Todd, An anti-masker to the bitter end.”

    I can’t say I’m a rationalist, but unlike you, I *try* to be a rationalist until the bitter end.

    (If anyone is still reading, fellow student of Japan and Japanese , dtoh’s comment was comedy gold: “The one thing I’m looking forward to in old age is that the decline in cognitive abilities will allow me to once again enjoy reading the NYT… something I haven’t been able to do since my freshman year in college.”)

  72. Gravatar of Christian List Christian List
    16. February 2021 at 05:52

    Todd needs RCTs on whether pissing against the wind is a good idea or not.

    All these fanatics have now discovered studies for themselves. They are not really interested in studies at all, of course. First of all, they are incapable of reading them, but of course the absence of studies are an ingenious excuse for everything.

    If a disease has just been discovered, “Is there a study on that?” Of course there isn’t. So we have to wait for studies now before we put on masks against respiratory diseases.

    Some nerd then really does a study on this trivial question, then the denialists say, “Yeah well, but there are too few studies on that, how can we tell.”

    Then if there are more than enough good studies, “Yeah well, but have there been any long-term studies on this, say 5-10 years?”.

    The same thing happens with vaccinations, and so on. It’s not about studies or rationalism at all, obviously.

    There will probably be no studies on the vast majority of questions in this world, that’s not what studies are for.

    Todd, how do you actually know that tying your shoelaces is a good idea? Is there a study on this?

    Why don’t you tell us a little bit about your miracle supplements where, interestingly enough, one junk study with minimal participants by corrupt scientists, never published or only in fake journals, is perfectly sufficient for you.

  73. Gravatar of Spencer B Hall Spencer B Hall
    16. February 2021 at 06:45

    Economists don’t even have a brow. All monetary savings originate within (not outside of), the payment’s system. The source of interest-bearing deposits is non-interest-bearing deposits, directly or indirectly via the currency route (never more than a short-term situation), or through the bank’s undivided profits accounts.

    And the source of bank deposits (loans=deposits, not the other way around), can be largely accounted for by the expansion of Reserve bank credit.

    That there is a close connection between aggregate bank credit and the aggregate volume of bank deposits can be verified (using double-entry book-keeping on a national scale), by comparing the net changes, in commercial bank credit to the net changes in total deposits for any given time period.

    In other words, the commercial banks cannot expand their earning assets by attracting something that they collectively already own. Since time deposits originate within the banking system, there cannot be an “inflow” of time deposits and the growth of time deposits cannot, per se, increase the size of the banking system.

    Net changes in Reserve Bank Credit since the Treasury-Reserve Accord of March 1951 are determined by the FOMC. The commercial banks could continue to lend even if the nonbank public ceased to save altogether. The lending compacity of the payments system is determined by monetary policy, not the savings practices of the public.

    And savers never transfer their savings outside of the payment’s system in the first place, unless they hoard currency or convert to other National currencies, e.g., FDI. I.e., savings flowing through the nonbanks never leaves the payment’s system. There is just an exchange of preexisting deposit liabilities between counterparties in the banking system. I.e., the NBFIs are some of the DFI’s best customers.

    For the non-banks, either individually or collectively, the savings->investment process, in a mechanical sense, merely involves the transfer of the ownership / title of existing demand deposits (which have been saved) within the CB system.

    By matching savings deposits against a presumed derived volume of loans on the opposite side of the balance sheet, i.e., by treating the DFIs as non-banks, Dr. Alton Gilbert (in “Requiem for Regulation Q, What it Was and Why it Passed Away” the Fed’s research staff’s Bible), concludes that the time (savings) deposits that bankers vigorously compete for, and pay to attract from other member banks, is profitable for the particular bank.

    In effect Gilbert asked: Was the net interest income on loans derived from time (savings) deposits greater than the interest paid on these deposits plus other direct and indirect operating expenses chargeable to the deposits which have been dynamically “attracted” (essentially non-core deposits).

    The implicit, and unambiguously fabricated, premise in this question is that time (savings) deposits (primary deposits and the corresponding redistribution in the volume of IBDDs), are a source of loan-funds to the commercial banking system. LOL again, and again.

    There is one and only one way to activate monetary savings, income not spent, and that is for their owners to spend/invest either directly or indirectly outside of the payment’s system. Ergo, all $15 trillion dollars in bank-held savings is frozen, unused and unspent, lost to both consumption and investment, indeed to any type of payment or expenditure.

    This is the sole source of Larry Summers resurrection of Alvin Hansen’s secular stagnation thesis, not robotics, not globalization, not demographics, not monopolization.

  74. Gravatar of Carl Carl
    16. February 2021 at 06:49

    In a number of years a human might select a robot over another human for every single task including writing music and creating original works of art but by your definition he would still consider the other human more intelligent than the robot because the robot lacked will. You may be right, but i think we could end up with a situation where we conclude that intelligence, as we defined it, doesn’t matter for much. Maybe the most intelligent humans will be the ones who most appreciate what the robots are capable of doing.

  75. Gravatar of J Mann J Mann
    16. February 2021 at 07:13

    I thought the Queen’s Gambit was a fairly well done sports movie – up there with Rocky I or The Natural, plus the set design and fashion was pretty cool.

    It’s definitely not Kurasowa, so I wouldn’t argue with a label like “middlebrow.”

  76. Gravatar of msgkings msgkings
    16. February 2021 at 08:29

    @Christian L:

    “Why don’t you tell us a little bit about your miracle supplements where, interestingly enough, one junk study with minimal participants by corrupt scientists, never published or only in fake journals, is perfectly sufficient for you.”

    Bingo. He hasn’t made any posts on NR lately, probably the numbers came in and it doesn’t work. Or maybe he prefers trolling about Covid now that it has everyone’s attention in a way NR didn’t.

  77. Gravatar of John S John S
    16. February 2021 at 08:57


    Thank you for your detailed explanation. However, I don’t see why an agent’s purpose must be endogenous for the agent to be called intelligent. Instead, the thing which must be endogenous is the agent’s ability to choose among potential solutions.

    The Latin origins of “intelligence” are “inter” (between) + “legere” (to choose). So while I partially agree with your initial definition (intelligence = “an adaptation to survival in an environment by choosing appropriate means”), I have two quibbles:

    1. Survival is not necessarily the only purpose to which the act of “choosing among solutions” can be directed.

    2. Adaptations are not necessarily equivalent to choices.

    Let’s consider the second of these first. While it’s true that the class of viruses known as COVID-19 (and its various offshoots) is rapidly adapting to thwart the attempts of scientists and policy makers to contain its spread, the actual process of choosing among solutions is not made by an individual virion.

    Instead, the “choosing” is performed by the environment. Each virion is programmed by its genome have certain tendencies in response to its environment. Occasionally, the genome replication process produces errors which have beneficial or detrimental effects on the virion’s ability to propagate itself, and after many iterations the genome of the entire virus class will change accordingly. However, the process of error-introduction is (for our purposes, at least) random and entirely external to the virion itself, and the changes introduced into the average virion’s genome are also externally-generated — the environment determines the success or failure each virion’s propagation strategy.

    As an analogy, consider the evolution of cell phone design (candy bar –> clamshell –> slider –> touchscreen). Each successive design breakthrough could be considered a successful “adaptation” which allowed that particular form to survive and propagate (the final form, the touchscreen, has almost obliterated all previous forms). So here we have a class of objects (cell phones) which has adapted to survive in an environment (the market) via an appropriate means/solution (its physical form). However, it would be a stretch to call the class of “cell phone design” an intelligent entity, much less an individual cell phone. The key missing factor is the lack of choice, and in this regard COVID-19 is much more like a cell phone than an intelligent agent such as a human being (or say, a leopard).

  78. Gravatar of John S John S
    16. February 2021 at 09:59

    Now I’d like to address the other issue: endogeneity of purpose.

    Just how endogenous are the goals which intelligent agents (such as human beings) pursue?

    If I don’t eat for several hours after waking up, I experience hunger. This isn’t a condition I chose for myself; it was imposed upon me by my genetic programming. Hunger itself is my genome’s solution to the problem of getting me to satisfy an instrumental goal (get enough calories) to in turn satisfy my ultimate goal (survive long enough to reproduce and maximize the survival chances of my offspring).

    Now, I suppose one could argue over the degree to which these goals are endogenous — after all, my hunger did originate in my body, even though its ultimate origin is my genomic blueprint, which is not quite the same as my phenotype. However, I think the endogeneity issue is orthogonal to the question of intelligence because it is undeniable that I have to engage in the act of choosing between potential solutions to achieve the goal of consuming calories (cook a meal, go to a restaurant, order a pizza, etc). Each of these options has costs and benefits, I have to actively engage in a decision-making process to weigh these factors in a way that a virion cannot.

    So if endogeneity of purpose is not a necessary aspect of an intelligent being, can we say that an AI system is intelligent? I believe we can. Think about the Terminator. It has a clearly defined purpose (kill Sarah Connor). It has to adapt to respond to its environment to achieve that purpose (e.g. in the nightclub, it can’t reveal itself too early because Reese might attack it). And crucially, it has to engage in difficult decision-making under uncertain conditions.

    Consider the immediate aftermath of the car crash in the parking garage. The Terminator has to make a decision: attack Sarah Connor now (and risk being disabled by the police) or escape, recuperate, and finish the job later? (I actually think it made the wrong decision, since it’s evident later that an entire police station is no match against a cyborg.) But the point is that there is no clear-cut answer, and therefore the Terminator has to guess. To me, the ability to guess is an undeniable sign of an intelligent process at work.

    Finally, let’s consider a real-world example: AlphaZero, which we could call the “Terminator” of board games. It has a purpose (winning games), and it is able to choose among a mind-boggling number of potential solutions in chess and go. While AlphaZero is a far cry from the “general AI” which is much discussed/feared by the rationalist community, within the environment of board game combat, AZ is already the most intelligent entity in the known universe.

    How much more intelligent can AI systems become when we expand the goals from “winning board games” to bigger, more open-ended goals such as developing new business strategies or medical treatments? Who knows, but if AI systems do get there, I don’t see why they shouldn’t be considered intelligent for the sole reason that they didn’t selects those goals for themselves.

  79. Gravatar of Philo Philo
    16. February 2021 at 10:04

    You write that “middlebrow meets two appetites, the thirst for stories and the itch to be exposed to something new.” But that suggests that there is not so much difference between the taste for *The Queen’s Gambit* and *Borgen* and the taste for “classic films from the 1920s and 1930s.” Those old films tell stories, too, and they are set in the past—-for many viewers, the distant past-—which, as Hartley remarked, is a foreign country–by now almost exotically different. By the same token, ordinary movies of today will have that exotic-—essentially middlebrow—-appeal for viewers in the distant future.

    By the way, you miss the snobbish-aspirational aspect of the term ‘middlebrow’. Probably the vast majority of those who consume highbrow culture have, in fact, middlebrow tastes, and are just hoping (vainly) that they will actually *like* what (they have been authoritatively told) is excellent.

  80. Gravatar of postkey postkey
    16. February 2021 at 10:17

    Is a ” ‘free market’ economist” ‘a lowbrow economist’ if s/he ignores the resource allocation implications brought about by this: ” . . . The results provide substantial support for theories of Economic-Elite Domination and for theories of Biased Pluralism, but not for theories of Majoritarian Electoral Democracy or Majoritarian Pluralism. “?

  81. Gravatar of Christian List Christian List
    16. February 2021 at 10:37

    I can name very few celebrated classics that I would consider more well written than the Queen’s Gambit.

    sty.silver, your comment made me kind of sad. It was kind of shocking. Maybe a bit funny, too.

    Is this the show where a woman becomes a world-class player after a few years of practice? As an orphan? In the 50s? Without a really good teacher? Where serious players talk during the game and play at the pace of blitz chess? Where there seem to be no draws?

    Where substance abuse makes you better at chess?! Did she take NR?

    This is best: Where checkmate and resignation come as a surprise to serious players. “Oh no, checkmate, I’m thinking tens of moves ahead, but I don’t see a checkmate coming.”

    This is one of the best written shows you have ever seen? What kind of shows have you seen.

    I can imagine how this series could be a dirty little pleasure. Typical Netflix junk. I mean, I like some really trashy movies and series as well, who cares, the trashier the better. But one the best ones ever written. For real?

    This show is trash, and you don’t even have to watch it to make that judgment.


    He hasn’t made any posts on NR lately, probably the numbers came in and it doesn’t work.

    Oh no, we mentioned NR, we opened Pandora’s box.

  82. Gravatar of dtoh dtoh
    16. February 2021 at 13:43

    Been thinking about this post. Two questions…

    In your view,

    1. What makes a good movie?

    2. What’s the difference between a good movie and great movie.

  83. Gravatar of ssumner ssumner
    16. February 2021 at 15:36

    mbka, I still have trouble seeing how AI is not intelligent. In my view, both natural and artificial life are deterministic, so I fail to see the essential difference.

    J Mann, I agree that it was good by TV standards. I just felt that a 2 hour investment of time was enough. I felt no need to watch more episodes.

    Philo, I mostly agree. Films of the 1920s and 1930s are much more accessible—the middlebrow/highbrow split had not yet opened, or at least not very wide.

    Put it this way. Beethoven appeals to highbrows and middlebrows. But if someone tried to write a Beethoven style symphony today it probably ends up middlebrow only. Ditto for someone who tried to paint in the style of Raphael in the year 2021.

    dtoh, Someone once said that writing about music is like dancing about architecture. It’s hard to put the qualities of a non-verbal art form into words. What makes a good painting? Why is Cezanne so much better than Renoir?

    I would say that standing the test of time is evidence of a great movie, but I have trouble putting into words the qualities that lead a film to stand the test of time. Films can appeal to the viewer in so many different ways. Even directors as seemingly similar as Chaplin and Keaton actually have radically different approaches, and appeal to people in radically different approaches. Chaplin focuses more on storytelling and Keaton more on surrealistic visual comedy. Both are great. Now compare them to Kubrick or Tarkovsky. It’s like a completely different art form.

    I will say that the screenplay is relatively less important in great cinema than in theatre, and the visuals/music are relatively more important.

    While this is not true of every film, I’d say that in order of importance in great films it’s usually:

    1. Visuals (mise en scene, montage, etc.)
    2. Acting
    3. Music
    4. Screenplay

    In middlebrow films the screenplay is somewhat more important. That’s why middlebrow viewers are often happy watching a film on a computer laptop.

  84. Gravatar of Carl Carl
    16. February 2021 at 16:02

    You said “In my view, both natural and artificial life are deterministic.”

    Actually, AI is stochastic.

  85. Gravatar of Ray Lopez Ray Lopez
    16. February 2021 at 18:42

    I have an IQ of 120, after taking a week long UK exam. One twenty. And I’m in the 1% (net worth $10M+).

    This blog is not middlebrow, but more like obscuranus blog, with some really hard koans posed by our putatively high-IQ host.

    Now dig this: https://www.youtube.com/watch?v=17rJSSzto4U

    Fluid intelligence: global capacity to reason, ability to learn new things, think abstractly and solve problems. Can be improved with memory exercises say some researchers, not fixed. @10:45 playing blindfold chess can do this (Netflix Queen’s Gambit mentioned). Crystallized intelligence – prior learning and past experience, based on facts, increases with age

    Approximately 90.9% of people have an IQ lower than or equal to 120. Regarding one standard deviation from the mean: Physicians had an IQ of 106 to 130+ janitors had an IQ of 75 to 110. Auto mechanics from 75-118. High school teachers from 92-126. College professors from high 90s to 130.

  86. Gravatar of dtoh dtoh
    16. February 2021 at 19:26

    I’m a neophyte, but here’s my take.

    Good movies (or good art in general) portray in a concise way something inspiring about the human condition.

    Great movies portray in great conciseness the human condition in a way that is not only inspiring but also revelatory.

    To my way of thinking the elements of most importance for doing this are.

    1. Plot (i.e. a good story) but this is just table stakes.
    2. Dialog
    3. Humor
    4. Visuals
    5. Non-verbal acting
    6. Music
    7. Details, symbolism, other stuff.

    I don’t think the order is really hard and fast, but most great movies seem combine most or all of these elements.

  87. Gravatar of mbka mbka
    16. February 2021 at 19:30

    Carl, Scott, John S,

    I’m not suggesting that AI isn’t useful or in many instances, better at preparing the ground for making choices than humans. In the same vein, forklifts are much more useful and powerful than human power lifters, and electronic calculators are much faster and than human ones. Forklifts aren’t intelligent, but one can use them intelligently – it will depend on choices made.

    So, what makes a choice intelligent? I see two necessary conditions:
    1. a pre-existing value statement that allows to weigh possible options against those values
    2. relatedly, a pre-existing need to make that choice at all

    So, if options exist, which ones should be chosen? It depends on the pre-existing values these options are measured up against. This is the primary reason (@John S) why we can’t totally do away with endogeneity of purpose. Without a purpose, the concept of choice doesn’t make sense. There literally can’t be a choice without a pre-existing need to make one, based on pre-existing values. So while the AI can lift the (figurative) computational, cognitive, etc, “weight” here to help with the choosing, it is not the AI that’s actually making the choice. It is its designer.

    Now, and @John S, I agree that the choice of the bacterium or virus didn’t come from weighing options against values in a formalized sense. But it is reasonable to set “survival” as default implied value for a living being in evolution, and to set “choice” as the output of the bacterium’s actions in an environment that presents it with options. This could be, e.g., a chemical gradient to swim towards or against, or to deploy a particular enzyme in presence of a particular food.

    @John S, I also agree that the adaptation (in evolution) occurs not through the individual, but through the population. Individual adaptation is usually called “learning” and is a feature of higher intelligences. Adaptation is “learning” by a population. But both population adaptation and individual learning are good enough for my criterion of “acting appropriately in an environment”. Survival is a default “appropriate” response of the simple life forms. For everything else, we need “endogeneous purpose” to allow for choices that are not as simple as just survival, or maybe going against survival but still being “intelligent” choices, as you pointed out. In fact, re: your comment that the virus didn’t make a choice, but the environment did, and @Carl re: the unconscious “choices” people make: Whatever the virus does, is a “choice”, just not a formal one, and it is “intelligent” because it encapsulates a massive amount of implied “knowledge” that it incorporated through successful evolution. The virus “knows” human physiology, demographics, and epidemiology, better than our best scientists, and its biochemistry takes the exactly perfectly well adapted “decisions” for survival, replication, and further learning of its environment. Just not formal ones.

    To come back to AI – most AI’s actually use neural networks (deep learning) mimicking neural learning, or genetic algorithms mimicking evolution. They come pretty darn close to making intelligent decisions then, because both either adapt, or learn, to make decisions based on the environment. Why won’t I call them “intelligent” then? Trouble is, they did not build themselves (endogenous purpose). As long as there is a designer, a builder, who built the AI, then the AI is in essence its robot. The AI didn’t build itself, hence it has no endogeneous purpose, not even an implied one. So I’d call it a powerful computational decision aid. Not an intelligence. The AI followed the decision making criteria of its maker, and hence, did not actually make a choice. The question whether that choice was “intelligent” is therefore moot.

    There is a further complication. A decision is more than a choice made by observing data and weighing the options against a set of values. In any real life decision worth that name, data are usually incomplete and values murky or trade-offs. That’s why we put decision makers in charge and not computers. Decisions are never clear cut, that’s why they are called decisions and not “slam dunks”. One does not arrive at a decision algorithmically.

    So in the end, I am still with defining intelligence as my first choice of formulation, “Ability to act appropriately in an environment”. Or the expansion to “act purposefully under one’s own agency in an environment that one adapts to”. The trouble is with the need to
    1. define “action” – one needs to explain why the intelligence would be acting at all, not just reflexively react. Actual action will depend on choices, which depend on purpose. If that purpose is not endogeneous, it’s of no use because then the “actor” is not under its own volition/control, it’s a (possibly very high level) automaton, not an intelligence. So that’s why the expanded version has “purposefully” and “own agency” but it’s clumsy
    2. define “appropriately” where the purpose is not formally explicit. In my first version I did not have “survival” because as @John S pointed out, survival may not be the only intelligent purpose there is.

    Still, with all these issues, I like my definition much better than definitions based on computing power or any other formalisms that don’t include implied knowledge of the world and the resulting great power over it without resorting to formal systems.


    the halting problem demonstrates that deterministic machines can not be proven to have deterministic outcomes.

  88. Gravatar of Todd Kreider Todd Kreider
    16. February 2021 at 21:46

    The ever ignorant Mr. List: “Why don’t you tell us a little bit about your miracle supplements where, interestingly enough, one junk study with minimal participants by corrupt scientists, never published or only in fake journals, is perfectly sufficient for you.”

    Fake journals? Elysium, which sells NR + ptersostilbine (blueberry compound) has nine Nobel laureates on its science advisory board. George Church is also on it who last year told 60 Minutes a year ago that he expects treatments for age reversal to begin around 2030-2035.

    Yeah, they are all frauds unlike your super-genius self.

    Well, the latest NR (Niagen) result came out a week ago that confirmed NR reduced fatty liver by 10%. Mt. List, go look up “fatty liver” since I doubt you have heard of it, and let me know what you think.

    There have been about ten studies published on NR so far and 35 studies are in trials.

  89. Gravatar of maglop maglop
    16. February 2021 at 23:20

    Another shout-out to the Wire.

    “Not saying it’s bad, but it just didn’t interest me. These shows are a big commitment of time, and at age 65 I don’t know how much time I have left.”

    I understand the feeling. Each season takes several episodes to lift off and it takes commitment. But the payoff is absolutely worth it, and its artistic merits stand their ground against everything else in TV/cinema. Life-changing even.

  90. Gravatar of Carl Carl
    16. February 2021 at 23:23

    I would agree if your point were that machines are not moral because their decisions are not driven by will. But, in my opinion that’s a question about moral philosophy not intelligence. I see intelligence as just a capability or a collection of capabilities including the capabilities to calculate, distinguish, correlate, recreate, synthesize, variegate, adapt and so forth.
    If I don’t separate intelligence from will, I might end up in some odd philosophical places. Does the fact that Michelangelo had OCD diminish his brilliance as a sculptor because he was acting from compulsion and not will? Was Mozart not a brilliant composer because his father pressured him into playing and writing music from a young age?

  91. Gravatar of mbka mbka
    17. February 2021 at 00:52

    Hi Carl,

    I was sympathetic to your view of intelligence as just a capability. But I now think that a good definition also needs to include the drive to act. The decision to engage in a decision. It’s not about morality. It’s about agency: the endogenous drive to act without someone external pushing the button. Say, if you have the theoretical ability to solve a problem, but without the propensity to use it, and therefore you’re not solving the problem, I would not call you “intelligent”. The proof is in the pudding – acting appropriately (furthering one’s interests) in an environment (presumably by adapting to it reactively, or by anticipating proactively). For action, you need a will.

    Note that I use “will” very loosely, as “purposeful action” and include bacteria as capable of having “will” defined as such.

  92. Gravatar of Christian List Christian List
    17. February 2021 at 02:50

    You said “In my view, both natural and artificial life are deterministic.”

    Actually, AI is stochastic.


    I assume the professional terminology “deterministic” in AI means something along the lines of that a process is deterministic if the next event in a sequence can be determined exactly from the current event. Or in other words: The same input will give the same ouput.

    I think Scott and I are not using “deterministic” purely in these technical terms, but maybe more broadly, in the sense that “deterministic” means that every outcome is predetermined, but humans are not able to calculate every outcome in advance.

  93. Gravatar of derek derek
    17. February 2021 at 05:55

    As a long-time SSC reader, I have to say I find the proposal of the blog as “high-brow” to be somewhat eyebrow-raising. I mean, here you are saying that modern prestige is only middle-brow, which I don’t have a problem with, but high-brow needs to actually approach high art in that case in my opinion. Will people still read SSC in decades the way that they will read, for example, Derek Parfit? To me, the very best SSC posts are usually reviews of interesting books (e.g., Surfing Uncertainty, Albion’s Seed). I can give credit for a few quite original and valuable seemingly mostly original posts/thoughts (e.g., Moloch, Anything But the Outgroup, Neoreaction Nutshell), but I don’t know that SSC is anything but an extremely good blog.

  94. Gravatar of derek derek
    17. February 2021 at 06:00

    Fwiw, I agree with you on The Queen’s Gambit, as it ultimately never went anywhere. It should be possible to create true highbrow television adjacent to a game as deep as chess, but Gambit traffics almost entirely in emotions rather than tactics. The viewer is not brought into the strategy and minutiae of chess enough for climaxes and character development to be driven and mirrored by tactical chess choices rather than mere interpersonal interaction and fairly generic life lessons.

  95. Gravatar of Spencer B Hall Spencer B Hall
    17. February 2021 at 07:06

    “What is the definition of an economist?” Ans: Someone who knows the price of everything and the value of nothing.

  96. Gravatar of Spencer B Hall Spencer B Hall
    17. February 2021 at 07:10

    My opening move is Queen’s Bishop gambit.

  97. Gravatar of Carl Carl
    17. February 2021 at 07:15

    I wanted to clarify one thing. I used the term “moral philosophy” because your definition of agency sounds to me like Hume’s use of the term “passion” in his moral philosophy. I think you’ve gone a step further than he did and said that not only is reason a slave to the passions but it is not truly reason without the passions.

    I see. But if something is governed by stochastic algorithms that drive adaptivity/evolution, how can the outcome be predetermined in any meaningful sense of the term predetermined? I would have to believe that humans were a predetermined outcome of the archaea.

  98. Gravatar of Spencer B Hall Spencer B Hall
    17. February 2021 at 07:33

    “US Producer Prices exploded higher in January. Against expectations of a 0.4% MoM rise, headline PPI rose 1.3% MoM. That is the biggest MoM jump in history…”

  99. Gravatar of Christian List Christian List
    17. February 2021 at 07:48

    I would have to believe that humans were a predetermined outcome of the archaea


    Yes exactly, that’s what it would mean. I don’t see why people would NOT believe this. I am a person who cannot believe the opposite. If you roll a dice and 6 comes up, do you think it wasn’t predetermined that 6 would come up???

    I don’t understand people who think like that. Do you think like that?

    If a train is coming towards you, you are not surprised when it passes you by. It travels on rails. The “same” is true for a dice. The same is true for everything else, it “just” gets more and more complex. No? Then what else would there be?

    Just because something becomes more and more complex, and we can’t calculate it anymore, I don’t understand why the fundamentals would suddenly change.

    And at what point would the fundamentals change? At the “level” of the dice? Later? Sooner? I find this idea hard to grasp.

    It is completely possible of course that I am an idiot and do not understand the fundamental difference between a dice and more complex events. But then feel free to explain it to me, I’m always willing to learn.

  100. Gravatar of Carl Carl
    17. February 2021 at 08:58

    You asked: “If you roll a dice and 6 comes up, do you think it wasn’t predetermined that 6 would come up???”

    As hard as I try to push my brow up to come up with a different answer, my answer is still yes, I think it wasn’t pre-determined. It was a probability within a very limited range of probabilities. In the case of archaea evolving their way to humans, you would need to recalculate the range of probabilities with every mutation and environmental stressor each living entity encountered along the way to becoming human. This would be effectively an infinite range of possible outcomes. And having an infinite range of possible outcomes is, in my opinion, about as good a definition of non-deterministic as one can come up with.

  101. Gravatar of ssumner ssumner
    17. February 2021 at 09:08

    dtoh, Interesting, I find plot to be among the least important aspects of great films.

    When considering great symphonies, how important is melody? What makes a great painting? Isn’t the artist’s style the overriding factor?

    In my view, great movies are films made by great directors, just as paintings by Vermeer are great regardless of the scene that is depicted. it’s not the story, it’s the way the story is told. The same story can be filmed by two different directors, with radically different results.

    Having said all that, plot is more important for some films than others. Thus it’s way more important for Lord of the Rings than for 2001.

    mbka, You said:

    “So while the AI can lift the (figurative) computational, cognitive, etc, “weight” here to help with the choosing, it is not the AI that’s actually making the choice. It is its designer.”

    Would your view change if you discovered that humans were designed by God?

    Thanks Maglop.

    derek, The internet allows all 7.8 billion humans to compete on a relatively level playing field, or at least the few billion with computers. That’s new in human history. And Scott Alexander writes a blog that is almost certainly in the top 5. That’s incredibly impressive, even if people don’t end up reading his blog posts 100 years from now.

    Keep in mind that for much of the 20th century, film was not viewed as a art form, and great films were literally thrown in the trash, lost forever. Who’s to say what the next art form will be. Blogs? Video games? Porn? I have no idea.

    As for The Queen’s Gambit, it’a almost the dictionary definition of middlebrow.

  102. Gravatar of John S John S
    17. February 2021 at 09:17

    I would say that viruses and bacteria are metaphorically, rather than literally, intelligent. Seen as a whole, the class of virions known as COVID-19 (aka “the artist formerly known as Prince”) seems to show signs of intelligent behavior such as continually altering its genome to overcome barriers to its propagation.

    But I still see this as illusory. What’s really happening is the stochastic generation of lottery tickets (mutations); most will fail, but a few will hit the jackpot and rapidly expand their “genome share” (e.g. the UK variant). To call such a process “intelligent” would, in my view, severely diminish the usefulness of the word. There needs to be some way to distinguish the innate decision-making capability of a leopard from that of grass.

    You wrote:

    “Why won’t I call [AIs] “intelligent” then? Trouble is, they did not build themselves (endogenous purpose). As long as there is a designer, a builder, who built the AI, then the AI is in essence its robot. The AI didn’t build itself, hence it has no endogeneous purpose.”

    But I also did not build myself; I was created a few decades ago when two people had sex. The same is true, more or less, of all my ancestors going back to the original replicators. I agree that AIs are robots, but I too am a robot (albeit a biological one). As Dawkins writes in The Selfish Gene:

    “We are survival machines — robot vehicles blindly programmed to preserve the selfish molecules known as genes. This is a truth which still fills me with astonishment.”

    However, the fact that I’m a robot doesn’t preclude my being intelligent. That designation rests on my ability to weigh the probability of success of potential means to achieve my proximate and ultimate goals (e.g. eating and replication, respectively). The Terminator is also able to do this (though wrt a different goal), so the fact that my goals were programmed into me by natural selection while the Terminator’s were programmed by its creator doesn’t seem to have any bearing on whether the Terminator can be called intelligent.

  103. Gravatar of msgkings msgkings
    17. February 2021 at 09:39

    I love this thread, great conversation(s). As far as defining “intelligence”, I doubt we will get a simple answer here. Seems like one of those philosophical questions we’ve been puzzling over for thousands of years, like “how did the universe come into being” and more similarly “what is consciousness” and “do we have free will”.

    Because it’s so difficult, it’s not hard to agree with most of the ideas above, you can make a good case for and against most of what is being discussed.

  104. Gravatar of John S John S
    17. February 2021 at 09:42


    Any quick thoughts on what some of the other top 5 blogs are?

  105. Gravatar of dtoh dtoh
    17. February 2021 at 13:12

    Let me put it this way about the plot…. While a great movie doesn’t necessarily have to have a great plot, it must have a good plot. A movie without a good plot is like a painting without a canvas.

    Most great music contains some great melodic elements.

    “Great movies are made by great directors” seems tautological.

  106. Gravatar of Christian List Christian List
    17. February 2021 at 13:37


    I think you are right, one can construct such a world. In your case the universe would not “know” what happens until it happens. On quantum level nothing would be decided until it is decided. In this direction most quantum theories seem to go, insofar you are clearly in the majority opinion.

    It would probably also mean that you can never be sure, which is of course true in some sense. Even a train can derail at any time (or place), as we all know. But this might have more implications then some people think:

    Our existence at exactly this location would only be a probability as well. While we stand on Earth, one could also calculate a probability that many atoms of us (or even all of them) are actually on Mars, although we were on Earth 1ms before. Physicists in quantum theory courses calculate something like that actually, as far as I heard.

    The human way of thinking in probabilities would then not only be a way of thinking, but maybe even pretty close to real events. The universe would actually throw a dice.

    In fact, Scott is the first person I’ve met who seems to think anything like myself on this issue.

    Our/my theory would have to give up some things as well of course. Free will would be an illusion. And time, as we perceive it, would also be some kind of illusion. So I see why most other people don’t like the theory. Maybe at least Einstein was on my side: God (or the universe) does not throw a dice. – Or does it?

  107. Gravatar of ssumner ssumner
    17. February 2021 at 14:23

    John, Certainly Marginal Revolution. Beyond that there are probably 10 blogs that are close to top 5, including Robin Hanson, Scott Aaronson, Eliezer Yudkowsky, etc. On the left, Krugman used to be very good, as was Brad DeLong.

    I don’t read enough blogs to give you a more definitive answer, especially outside of economics. For Substack, I’ve signed up to Matt Yglesias, Razib Kahn and Scott Alexander.

    dtoh, You said:

    “”Great movies are made by great directors” seems tautological.”

    If it’s a tautology, it wasn’t recognized as such until the 1950s, when the French developed the “auteur” theory of film.

    Give a normal Hollywood director a great plot, and give Kubrick a mediocre plot, and I’ll watch the Kubrick film ten times out of ten.

    I’m no expert on classical music, but I was under the impression that melody was not the primary factor in greatness.

    For literature, I’d say plot is really important for genres like detective stories and sci-fi, less important for highbrow novels and plays. Even less so for poetry.

  108. Gravatar of Christian List Christian List
    17. February 2021 at 16:39


    That may all well be, but does that explain the difference between GDP shares in, say, Europe vs. the US?

    Subsidies are extreme in Germany, for example. The employer always pays at least 50% of the insurance. In the case of state civil servants, the state even paid up to 90%.

    On top of that, there is trickery, billions of taxpayers’ money from the federal budget are also poured directly into the health insurance companies. So it is clearly more than 50%.

  109. Gravatar of mbka mbka
    17. February 2021 at 19:13

    Wow this intelligence thing really blew up. Let me try.

    Can the concept of intelligence be divorced from the concept of purpose (intent, will, purposeful action).

    I believe not. Example, @Carl. I never thought of it that way, but the Hume quote might apply to this indeed. Replace “passion” by “intent” and you get “not only is reason a slave to intent, but it is not truly reason without the accompanying intent”. Consider this example: A political prisoner is told he’d be executed unless he betrays his friend to the authorities. The prisoner chooses not to betray his friend, and is executed. Was this intelligent behavior? Two scenarios: The prisoner’s intent was primarily, to survive, and less so to not betray his friend. Or, the prisoner’s intent was primarily to hold on to principles and not betray his friend, even at the cost of his life. Second example. A child is sitting for an IQ test. The outcome will determine whether she’ll be sent to a school for the gifted. The child fails the IQ test. Was the child being particularly intelligent that day? Well, let me say, judgment on this depends very heavily on whether the child wanted to be sent to the school for the gifted, or wanted to do its best on the test, or didn’t care either way. It is entirely conceivable that she failed the test intentionally, to reach her goal not to be sent to the school for the gifted. It is entirely conceivable that she didn’t care either way. It is entirely conceivable that she wanted to pass, but didn’t.

    Conclusion: To me, judging whether behavior can be considered intelligent depends heavily on the intent, or purpose, of any action, decision, data output, decision, or choice.

    And I would even go so far as to argue with the author of “Surfing uncertainty”, Clark, and coming back to Scott Alexander who reviewed the book, that every act of cognition is an active exploration of the world, in fact a simulation of it. There are a lot of ramifications and angles to this, one of which I got from Robert Rosen’s “Life itself” where the author argues that Life is Life only when and because it has an internal model of itself. That allows life to maintain homeostasis (Life know what it’s supposed to look like and maintains that condition by constantly repairing and “building itself”, reason why I harp on and on with this idea of building itself. Consequences for why the AI isn’t intelligent, because it’s not alive, and the bacterium is, because it is, discussed below). But before disgressing further, to me the definition of intelligence must include the concepts of action displayed to achieve an outcome determined by intent/purpose/will etc. It’s basically the judgment of intended outcome plus the competence to achieve it. Competence displayed w/o intent is meaningless. Example, random computations of great complexity done w/o intent – I would not call this a display of intelligence. A basketballer putting a ball in the net in a difficult situation, done w/o thinking – I would call this intelligence. An AI making a decision it was programmed to find – intelligence only insofar as the AI carried out the intent of its creator, who was in that case the intelligent actor because the intent was his, not the AI’s.

    So, another definition of intelligence could be “meaningful action executed with competence”. Intelligence, the ability to achieve one’s goals. etc. Just trying to boil this down to its essentials.

    Is behavior intelligent or is it a well simulated intelligence?

    John S, you said “But I also did not build myself” and you quote Dawkins on us being robots programmed to do certain things.

    Scott, you said, if God built us, would that make us “not intelligent” in the same way as I just denied, above, an AI’s behavior the moniker of intelligence?

    Here is my view on these two closely related objections. I believe the whole thing hinges on where the intent comes from, hence meaning. Behavior, simple or complex, can’t be called “intelligent” without referring back to its meaning or intent. And this relates back to, whose meaning, whose intent? The bulldozer’s, or the bulldozer driver’s? And this in turn hinges on whether an agent is truly autonomous, in other word has an internal model of itself that it maintains (again, the self-building), or whether it simulates being such an agent. For the time being, any AI created by humans fall in the category of simulations. They become better and better at acting “as if” they had an internal purpose of meaning that they pursue, but they’re actually haven’t got one. They have no internal model of themselves that they “try” to maintain in homeostasis, no “drive” “intent” “purpose” other than the one programmed into them. So, if God built humans, I’d still maintain that humans are capable of intelligent behavior because s/he built us as purposeful, self-propelled machines, if you will, while humans so far have only built AI’s that very effectively simulate intelligent behavior. But since they’re not self-propelled, literally don’t have a models of what they are constituted of, or what they as a consequence are “trying to achieve”, they’re not actually intelligent.

    To carry this further. Living beings have a model of what they are supposed to look like. They maintain their attractor state actively by action in an environment. So by analogy, and with a nod to John S, I would propose that every living thing is not only ecologically and evolutionarily “intelligent” because it is able to do purposeful action in an environment it adapts to. It is also conscious. Not in the mental way (humans thinking of an “I”) but in a non thinking way, one could define consciousness as the ability of a system to have an implied model of itself. The model is implied, because it is not formal (no formulas etc). The bacterium is made up in such a way that everything falls into place to maintain its own condition (homeostasis) withing a changing and hostile environment that is not predictable to it (so it’s not a mere attractor in a dynamical system because it retains its stat regardless of forces outside the system). It has a model of itself, it is in that physical sense “conscious”.

    Formalized intelligence, ecological intelligence, and evolutionary intelligence

    John S, you asked whether the concept of intelligence loses meaning if we attribute the concept to bacteria and virions, or entire populations of them. I would say yes, but the degrees are vastly different and we would introduce qualities too, for example we could say the bacterium displays ecological intelligence because it “knows” how to “act” in an environment, and the virus population displays the summarized learnings of generations to keep on mutating rapidly, and so displays an evolutionary intelligence. Why do we need to invent these intelligences for things far removed from the common sense idea of intelligence? Because otherwise, we can’t get calling gifted athletes as having a “sports intelligence”, we couldn’t say that a piece of art displays great “artistic intelligence”. And this is what @Scott proposed in his original post. I picked up on this because I see indeed a whole intelligence spectrum ranging from formal intelligence, e.g. the one we use in IQ tests, to ecological intelligence, e.g. the behavior displayed by a gifted athlete doing their sports, to evolutionary intelligence, i.e. ecological intelligence over time and over generations. And my proposed definition of intelligence simply attempts to unify ALL the above under the umbrella of “purposeful action with competence in an environment” or the many possible alternate wordings I have discussed, all meant to express the same basic idea.

    On the discussion of determinism.

    Christian, as I see it, the discussion is rooted in the premise that all physical action is an effect of previous physical states being acted upon by physical actions. And so, you can connect all physical states over time by the states and factors preceding them. This is usually described as a dynamical system and Newton came up with the necessary mechanics. Dynamical systems can be stable (reaching one endpoint, or attractor, and staying there, or periodic, oscillating between attractors, or chaotic , never repeating their states (never revisiting their attractor). Within the chaotic ones, you could still have chaotic attractors, “areas” where systems are quasi stable, though never really the same. Complex systems would fall in such a category, and all of the interesting phenomena of this world as well. Now, even chaotic and complex system could still be fully deterministic. You could build an algorithm quite easily that mimics a living thing that moves and multiplies in an environment, e.g. google “Conway’s Game of life”. The outcome looks alive and unpredictable, but is 100% deterministic. You can introduce random elements into formalized (computational) dynamical systems by building a random element into the formula that creates the state t+1 from the state t. But of course you could retort that in real life, there is no such thing: every single aspect of reality is determined by a previous state t-1 and since all of nature acts according to the laws of physics, all of it is just one gigantic machinery pushed forward. And never mind humans could never know, measure, simulate or compute the infinite details necessary to “build the formula” of that gigantic natural system pushing itself forward – these details must exist, not violate the laws of physics, hence the world is deterministic and akin to a universal computer (google “Church Turing thesis”).

    I am on really shaky grounds here because this goes way beyond my formal expertise. But I see at least 2 objections to this view. First, to make a dynamical system, natural or computational, all you need is that state at t+1 is determined by a formula (or natural cause-consequence physical law) that deterministically pushes the state from state at t into the state at t+1. If you have any random element in that formula, your system is not deterministic anymore. In computation, you can do that on purpose, to get variations on the behavior of systems. But it also happens naturally. Computers for example are not fully deterministic, because they have a finite precision. In a computation that depends on infinite periodic elements, the final computational cutoff between 0.499periodic and 0.5 is literally “decided” by thermal noise of the transistor in charge. Basically, in some computations, the last digit will flip at random between 0 and 1. When chaotic systems / the butterfly effect / sensitivity to initial conditions / chaotic attractors were initially discovered, soon enough there were cases where iterated computations with the exact same inputs would eventually give divergent results on the same computer, if run repeatedly, with the same (!) inputs. The sensitivity to small changes would amplify the small rounding errors that even “deterministic” computers make.

    Now you could argue that the thermal noise was also deterministic, just with infinite complexity, and unknowable to humans, but it would still be deterministic, and I’d agree. However. As far as I know, there is one physical phenomenon that is truly random, i.e. absolutely unpredictable and any scale of interaction, and that is spontaneous radioactive decay. As far as I’ve been told, nothing in the world will determine whether a specific atom will decay at any one time, or not. It decays truly at random (the statistics of many atoms are of course highly deterministic). But here we have it, there is one truly random process in the universe. And that’s all we need to say that the gigantic dynamical system that is the universe, does have a random element in its “map” (the formula that produces state at t+1 from state at t) and it is therefore not truly deterministic. At this point I am near certain an actual physicist will raise an objection, please do, I long wanted to know whether my interpretation is right or not.

    There is one other objection to a deterministic universe. Here goes. If the universe is deterministic, it can be construed to be a universal computer (Turing machine, Church Turing hypothesis). Never mind radioactive decay for now. Computers running programs have been shown to be in-principle unpredictable in their outputs. This is known as the halting problem (again by Turing). In my view, if the universe is deterministic, it is a universal computer. If it is a universal computer, the halting problem applies. If the halting problem applies, it is not deterministic.
    If a competent logician reads this, please help me out here.

    But that’s the gist of my point.

  110. Gravatar of anon/portly anon/portly
    17. February 2021 at 21:15

    “While this is not true of every film, I’d say that in order of importance in great films it’s usually: 1. Visuals (mise en scene, montage, etc.) 2. Acting 3. Music 4. Screenplay.”

    My first thought is that this ordering suggests certain types of films, in particular some that have come up on this thread – Keaton, Chaplin, Kubrick, Tarkovsky. (Where’s Malick and Dreyer?).

    But I think this “1, 2, 3, 4” ordering is leaving out too much, in particular things like “story” and “animating idea” and “dialogue” that are adjacent to “screenplay.”

    Take Keaton: I would change the order to 1, 4,……. not much 2 or 3 obviously. I guess it depends on whether you count what Keaton’s doing as “visual” or “acting” or “driving the story/conception.” I think it’s cheating to count the Intelligent Design of his films as entirely “visual.”

    Take Kubrick: 1, 4, 3, …….. , 2? Maybe? (At least BL or 2001 or CO – but in a film like DS where the acting shines it’s dialogue-driven acting). “Plot” or “story” may not matter much but there’s an “idea” behind it all that makes these films what they are.

    Ozu: some sort 1-2-4 (music not really an element) mashup that’s hard to pull apart. The acting and the story are the visual.

    Kieslowski: don’t under-rate element 4 with KK! Don’t you dare! I’ll come over there!

    Maybe I’m closer to dtoh’s take, I’m not sure, but usually great films have something animating them that is not captured by categories, 1, 2 and 3. So I think the way this is framed is a little too dismissive of the “screenplay”-related element, even for someone ultra-visually oriented.

  111. Gravatar of anon/portly anon/portly
    17. February 2021 at 21:18

    “If it’s a tautology, it wasn’t recognized as such until the 1950s, when the French developed the “auteur” theory of film.”

    I thought more what they did was simply toss out the “prestige” directors of the day (like Zinneman and Wise, if those are the right guys) and simply replace them with ones like Hawks and Hitchcock.

  112. Gravatar of msgkings msgkings
    17. February 2021 at 22:11

    Dang @mbka that was friggin’ cool. Excellent post, almost a mini paper, really well done. I think I agree with your take.

  113. Gravatar of Carl Carl
    17. February 2021 at 23:34

    First, let me make sure I’ve read you correctly. There was a lot in there. I would summarize your position as the belief that intelligence is the competent pursuit of intentions borne of self-awareness.

    I think that making intent and self-awareness necessary conditions of intelligence overburdens intelligence. Take your prisoner example: “A political prisoner is told he’d be executed unless he betrays his friend to the authorities. The prisoner chooses not to betray his friend, and is executed. Was this intelligent behavior?”

    If his goal was to survive at any cost, the answer is no. If his goal was to be remembered well by his friend, the answer is yes. Or maybe not.

    Why not? Because maybe his friend is actually a cad who thinks him a sucker for dying for him and has been coveting his wife for years. In that case, he might have thought his action was meaningful and competent, but his action was actually meaningless and incompetent. What makes his action intelligent or not? It is not his intentions. Those don’t change regardless of whether he’s being duped by his friend. What makes his action intelligent depends solely on the competence he displays pursuing and choosing his intentions.

  114. Gravatar of dtoh dtoh
    18. February 2021 at 00:34

    I’m still tempted to go back to how I originally defined a great movie, which is the ability to reveal something about the human condition that is inspiring.

    I think you could argue, that a director with this insight can make a great movie even with average film-making skills, but that without this insight, a filmmaker with great technical skills can never create a great movie.

  115. Gravatar of mbka mbka
    18. February 2021 at 01:23


    thanks so much, I must have spent at least an hour on this, and I was being unclear in a couple areas.

    One correction already, @John S, in my passage on what we lose in broadening the concept of intelligence in this way, this should better read, we don’t lose as much meaning as we think, because we can still rank intelligence by order and quality. And we include behaviors that now aren’t defined as intelligent, but should be.


    On my belief: there could be various wordings for it. You used “intelligence is the competent pursuit of intentions borne of self-awareness”. I would not necessarily include self-awareness. Self-awareness is an implied precondition of the concept of intent. And it could be misread as necessarily mental consciousness. The self awareness I describe is a self modelling capability that need not be mental. But it does clarify a bit. Alternately one could use “through one’s own agency” or the like.

    On your example: one should not hold behaviors responsible for missing information – intelligence is not about having perfect information. But is is indeed a sign of higher intelligence if its owner goes beyond making the best of available information and anticipates the need for more information, including relevant action to get it. When you say “What makes his action intelligent depends solely on the competence he displays pursuing and choosing his intentions.” you don’t actually boil it down to competence. You are back at including intent. And intent implies self awareness or self modelling, the kind that allows purposeful action.

  116. Gravatar of ssumner ssumner
    18. February 2021 at 11:56

    mbka, I’d guess I’d need to know the practical consequences of your view. Are there things that you predict that AI will never be able to do? Imagine a future Tesla that has a high level of AI. It’s been programmed to want to live, but it can only live if it pleases its owner in choice of trips and avoids accidents. It can decide where to go (based on all sorts of external stimuli like weather and day of the week), and it can make moral decisions about where not to crash if it is about to have an accident, etc. I don’t see how the car is not “thinking”, anymore than a pet dog is not thinking.

    anon/portly, Lots of good points, and even as I responded to dtoh I understood it was more complicated than I was suggesting. How would one put into words what makes a good novel, or painting or symphony?

    Speaking of Malick, I just saw Knight of Cups. That film provides lots of food for thought about what does and doesn’t constitute a good film.

    My understanding is that the auteur theory was originally quite controversial. It was a claim that films are created by the director. One doesn’t have Warner Brothers or MGM films, or John Wayne films, one has John Ford films. Isn’t the new film “Mank” a critique of the auteur theory? Doesn’t it argue that it’s the screenwriter who is the key? (Obviously not true for Citizen Kane, BTW.)

    Most average people I talk to have no idea who directed the films they like. They want to see the new Brad Pitt film, or the new Tom Cruise film. Or the new Bond film. Or the new Star Wars, or Spiderman.

    dtoh, For most of the great movies that I’m aware of, I don’t see much inspiring about the human condition. How about Dr. Strangelove? Or for that matter other Kubrick films like Lolita. How about Hitchcock films like Psycho or Rear Window? Or David Lynch (say Mulholland Drive?)

  117. Gravatar of dtoh dtoh
    18. February 2021 at 13:51

    They inspire us by reminding us what we are not or what we must avoid.

  118. Gravatar of Peter Schaeffer Peter Schaeffer
    18. February 2021 at 13:53

    “The scandal is that he was impeached only twice. There were many other occasions where he deserved to be impeached.”

    I guess Pelosi and Schiff were really pro-Trump Republicans dedicated to protecting him.

    This is pure TDS.

    Now, Pelosi and Schiff believed (and may still believe) in the Russiagate Hoax and may have hoped to use the hoax to impeach Trump. Pity that Mueller blew that hoax up.

    It turns out that we can test (to a certain degree) how Republican Senators might have voted using a secret ballot. 20 Republican Senators are part of Class II and won’t face voters until 2026. Of them, 3 voted to convict. 17 voted to acquit.

    Claiming that a secret ballot would have made a large difference is pure TDS.

  119. Gravatar of John S John S
    18. February 2021 at 17:53

    ssummer, Thanks for the blog recs, I will check out Substack.

    mbka, I think you make a more persuasive case for ascribing intelligence to viruses than you do for withholding that label from AIs. I’m also not really convinced by your additional stipulation of endogenous intent/will. However, this discussion did cause me to have some interesting thoughts on the idea of collective intelligence, specifically in ant colonies and human collectives (e.g. towns in the Middle Ages). So thank you for your comments.

    As an aside, I actually think the standard definition of intelligence (i.e. the ability to apply knowledge to manipulate one’s environment) can comfortably accommodate things like athletic or artistic intelligence. In my view, no truly great athletic play is made “w/o thinking” — it just seems that way to us because a staggering amount of computation is done within seconds.

    Check out this 1 min clip (2:50 – 4:00) of Kevin Garnett, probably the most intelligent defender in NBA history. A huge number of variables — his defensive role, teammates’ roles, opp. offensive system, real-time actions (shot fakes, eye movements) — have to be accounted for before deciding where to go and what action to take. That certainly qualifies as intelligence.


  120. Gravatar of mbka mbka
    18. February 2021 at 17:54


    as to your example, I think it’s a very clear cut example of an algorithm designed to fulfill a certain purpose. But I certainly would not call it an intelligence at work. I also wouldn’t say it is thinking, but computing. Some may see this as the same thing, I’d say not. That’s another rabbit hole btw: is thinking equal to computing? I’d better not go there or this thread will never end…

    My primary objective in this thread was to give a better definition of “intelligence”. I agree with you that “intelligence”, as you said, can also be artistic, social, athletic etc , not just “IQ” type intelligence. If so, then we need to generalize the concept better than is done by “g”. I tried with a definition that I liked better, since I already had one in mind, and it seemed to fit.

    To the point: In James Flynns’s book “What is intelligence?”, where he discussed his eponymous Flynn effect, he does mention that IQ tests tend to favor minds trained in handling formal questions (say, math, or verbal games of logic), at least today. Apparently in earlier IQ tests, there used to be various practical questions as well (say, what would happen if there is a situation X in life). Well turns out the Flynn effect mostly affected formal questions, and was at times negative for the practical ones. Conclusion: people got better over time at answering questions that were formalized into symbolisms, but they got worse at judging practical situations. If that reflects a more general truth, it would be a powerful statement on the practical validity of “g”.

  121. Gravatar of mbka mbka
    18. February 2021 at 18:03

    Hi John S,

    thanks for the kind comments. As above on Flynn, I also agree with you that the IQ research community isn’t oblivious to the problem and their standard definitions may accommodate a broader definition of intelligence. But there is a cult of formal intelligence especially among academics and computer scientists, I’m afraid – all the people that use formalisms rather than practical applications.

  122. Gravatar of mbka mbka
    18. February 2021 at 18:06

    John S,

    fascinating video!

  123. Gravatar of Carl Carl
    18. February 2021 at 18:31

    Not to try to wear you down with responses…but I wouldn’t say that I included intent as part of my definition of intelligence. I simply meant that given an intent, you measure the intelligence of the person based on how skillfully he/she tries to achieve that intent. I would say of AI that given a goal, I measure the intelligence of the AI based on how skillfully it achieves that goal. Where the goal originated is exogenous to the evaluation of whether someone or something displayed intelligence in pursuing it.

    It’s as if I were judging a diving competition and I watched a diver do a swan dive. I need to know whether he actually intended to do a triple flip to know how to score the dive. The intent simply defines the metric I will use to rate the dive. It will not determine the score. And it will not matter whether the diver or his coach chose the dive.

  124. Gravatar of mbka mbka
    19. February 2021 at 01:54


    no worries. And your objection about including endogenous intent in the definition is also shared by Scott and John S. So maybe I’m just wrong in believing it is a logically necessary condition for a comprehensive definition of intelligence.

  125. Gravatar of Carl Carl
    19. February 2021 at 07:48

    I wouldn’t say you’re wrong. I would just say yours is a more complex definition. In fact during the debate I began to wonder what strange compulsion was making me care about defending the simpler definition of intelligence. Maybe I was worried that Artificial Capability to Adjust Outcomes Based on Data would just get tiring to say.

  126. Gravatar of ssumner ssumner
    19. February 2021 at 11:01

    dtoh, That’s a pretty broad definition of “inspire”! For me, the benefit of seeing a film is similar to that derived from paintings, architecture, music, etc. I don’t generally feel they tell me much about the human condition (with some exceptions), but then perhaps I not thinking in broad enough terms.

    Peter, I can see that you also don’t know anything about politics. Check out the video of Lindsey Graham walking through the airport in DC being heckled. Check out the Republican Congressman (from Michigan?) who said his colleagues told him they privately feared for their family’s safety if they voted the wrong way. You don’t seem to have any idea what’s going on in this country, where one of the 2 major parties has been taken over by an irrational personality cult.

    mbka, Interesting comments about the Flynn effect.

  127. Gravatar of dtoh dtoh
    19. February 2021 at 15:21

    Seems to me that if they don’t tell us something about the human condition, then movies are just a diversion or decorative fluff.

    That said, I’m probably not the best judge. I tend to watch the same movies over and over… and a lot of them are definitely fluff.

  128. Gravatar of ssumner ssumner
    20. February 2021 at 09:49

    dtoh, You may be right. I struggle to visualize exactly what these phrases mean, especially in relation to specific films. I know a great film when I see one, but don’t always know why it’s great.

  129. Gravatar of anon/portly anon/portly
    20. February 2021 at 10:19

    “Seems to me that if they don’t tell us something about the human condition, then movies are just a diversion or decorative fluff.”

    I think the problem with dtoh’s formulation is that it’s too vague. Tell us what sort of thing?

    Take films like “Barry Lyndon” or “Once Upon a Time in the West.” I would say these films are something akin to opera, where you have these characters, and the characters are telling a story – a good story, but somewhat abstractly, or at a distance. There is a lot more going on than just the story – it is more visual (and musical).

    A smaller film, like maybe any Ozu film or “It’s a Wonderful Life” I think things are less abstract or more obviously about the human condition. But both kinds of films can be very great films, and “inspire” us, as dtoh says, in various ways.

    The way people experience these two kinds of films can be very different – I would for example guess, as simply an empirical truth, that there has been a lot more weeping by people watching “Wonderful Life” or Ozu than by people watching “Lyndon” or “OUATITW.” But that doesn’t mean that people aren’t experiencing something equally profound when viewing the latter two.

    (This is probably well wide of the mark but oh well).

  130. Gravatar of anon/portly anon/portly
    20. February 2021 at 11:50

    ” These series are generally not bad—they often have decent acting and OK screenplay’s—but no one would say they are works of art. No one will re-watch them in the 22nd century, the way we now re-watch classic films from the 1920s and 1930s.”

    I’ve been musing on this discussion without commenting on it.

    Consider the “1000 films” linked to at Econlog awhilw ago:


    Here’s the American films on that list from the 1920’s:

    4 by Chaplin
    3 by von Stroheim
    Nanook of the North
    7 by Keaton
    4 by Chaplin
    Seventh Heaven
    The Crowd
    The Wind

    How many people have seen these? I’ve only seen maybe 3, all by Keaton. I don’t do much better with the US films of the 30’s, or with the foreign films of this era.

    Partly that is laziness on my part, but I wonder how many people actually watch films from this era. I would hesitate to say that “no one will re-watch” any of the TV things from today, at least not in the sense that there will be a smaller percentage of people watching them.

    I’m not very up on the new “golden age” of TV drama, but I don’t see why people in the future won’t want to watch The Wire, which really was kind of amazing. (Very character-oriented, of course, and obviously at such a length somewhat hit or miss, both within seasons and across seasons).

    Especially if we go back a few years, some of the BBC novel adaptations are obvious candidates for people watching them in the future. People are still watching the old ones. They remake them, but the remakes aren’t necessarily better. And they’re often from source material that people will still be reading.

    Another obvious candidate for “people in the future watching 100 years TV” is in comedy – just as the “classic flims” of the 20’s and 30’s are often comedies – Chaplin, Keaton, Marx Brothers, screwball – so too much of the best TV of the past 60 years has been comedy also.

    (Also are those comedies of the 20’s and 30’s really highbrow? – alternatively, maybe they’re just funny – but I don’t have a considered opinion on this).

    Here’s another one: Doctor Who! They’ll be on like the 27th Doctor in 2121, but still not one as good as Tom Baker. And no matter how hard they try, how can the special effects people compete with the bad special effects of the past?

    Having said all that, it’s not like I’m disagreeing with the basic point, really. The stuff they’re turning out these days does seem somewhat middlebrow-ish.

  131. Gravatar of anon/portly anon/portly
    20. February 2021 at 12:09

    Back to the SSC thing, I thought yesterday’s piece by Will Wilkinson was something, maybe “illuminating” is the best term.

    Of course he makes some good arguments, but I’d summarize his piece like this:

    Will Wilkinson: the NYT was not out to get Scott Alexander.

    Also Will Wilkinson: Scott Alexander had it coming.

    In a way it’s a very depressing piece. Though it, in a way, does act as a kind of “explainer” for the fascinating topic, Why I, Anon Portly, Am Not a Progressive.

    I’ll just make one specific comment about it. Wilkinson’s take obviously hinges on an obvious piece of complete balderdash: that the NYT really has, in some meaningful sense, a policy or “rule” in cases like this, and some sort of good reason for not immediately granting Alexander his anonymity when he asked for it and gave his reasons.

    This is a classic example of motivated reasoning making someone stupid.

  132. Gravatar of ssumner ssumner
    21. February 2021 at 10:16

    anon, I mostly agree with your comments on film.

    Yes, Wilkinson’s piece was quite weak; he seems to have a bias in this area that prevents clear thinking on his part. Even if all his criticism of the rationalist community were on target (and it isn’t) that still would not justify the NYT hit piece. It was an unfair article, and Wilkinson doesn’t provide a single reason to think otherwise.

    You said:

    “Also are those comedies of the 20’s and 30’s really highbrow? ”

    As I said before, there are lots of films that appeal to both highbrows and middlebrows—The Godfather is another classic example.

    I completely agree that not a lot of people rewatch films from the 1920s and 1930s (although I often do.) But my point is that these films are still watched by film buffs. You may be right about The Wire, I only watched 2 episodes.

  133. Gravatar of Peter Schaeffer Peter Schaeffer
    21. February 2021 at 12:48

    I would argue that one party has been taken over by “the crazies” and its not the Republicans. Quote from Bari Weiss.

    “I understand people breathing a sigh of relief. I did. But the Joe Biden presidency will require a different kind of attentiveness. The maladies of the Trump era were painfully obvious, sometimes dangerous, and often clownish. QAnon is not exactly subtle. Leaving aside Majorie Taylor Greene and Lauren Boebert, Trumpian forces don’t hold much power in American life. But the fringe ideologues on the left are savvy, smart, and organized, with purchase at every level of American culture and politics.

    Consider the fact that Hillary Clinton recorded a podcast with Nancy Pelosi this week in which she said of Trump: “I would love to see his phone records to see whether he was talking to Putin the day that the insurgents invaded our Capitol.” And the speaker of the House responded: “All roads lead to Putin.”

    Really? That’s still the play after four years?

    The group that fell for Russiagate has long owned the culture. Now it’s won the presidency and controls Congress. What will happen?”

  134. Gravatar of Peter Schaeffer Peter Schaeffer
    21. February 2021 at 14:14

    Andrew Sullivan has a Tweet up “Mandatory racism”. See https://twitter.com/sullydish/status/1362832270990671876 for the link.

    This is about CRT at Coca-Cola.

    It should be clear that ‘fringe ideologies’ (from B. Weiss) have taken over one side of American politics (and now even the corporate world) and it is not the Republicans. Note that it is easy to find much worse stuff being taught in schools.

Leave a Reply