Links & What I’ve Been Reading, November

Media recommendations

Here are the highlights from what I’ve been reading, watching, and listening to in the last month. Before that, some announcements: 

First, thanks to everybody who came to the Fitzwilliam San Francisco meetup. I had a great time. 

Second, if you are a curious and analytical person between the ages of 16 and 19, you should apply to the Asian Spring Program on Rationality, taking place in February in Taoyuan, Taiwan. If your friends, children, or loved ones match that description, you should send them the application form. I am a former instructor on the (non-Asian version of) the program, so perhaps I’ll be back again. 

Blogs

28 things Ava Huang has learned in her time on Earth. And: Is the essence of love annoyance? 

Those of you in the East Coast American time zone (or a sleep schedule close to it) may benefit from attending the Wolfram Science Winter School. 

Congratulations to the new cohort of Emergent Ventures winners.   

Anton Howes on how to be a public historian. 

The life of Milton Friedman encompassed more interesting and consequential policy debates than perhaps any other economist since Keynes. In contrast, Craig Palsson asks, where are the major economic policy debates of today? This is relevant to a paper I linked in a previous post, Robert Solow on why there is no Milton Friedman today.  

I am apparently being people watched. 

Gavin Leech on the acts of early genius. I’d be interested to know whether people find reading about prodigies to be more depressing or inspiring. The great Walter Pitts made major contributions to logic and pioneered neural networks in computer science before he turned 18 (while he was homeless!).  

Elsewhere, Gavin has restored a remarkable book about existential risks from 1930 (!), and written about the nonmonetary payoffs to starting a company. 

The best things Rebecca Lowe has been reading and listening to, parts one and two.

An anonymous Reddit user has shown that there is a specific deck in the card-based video game Magic: The Gathering which will unleash infinite damage if and only if the twin prime conjecture is true. It has even been shown that Magic: The Gathering is Turing complete. We’ve moved beyond proving computational completeness, and someone has created an entire programming language within Magic: The Gathering. Truth really is stranger than fiction.

I found the above from Scott Alexander’s links for November. He also links to some hilarious polling of the American public about their opinions on various historical empires. Ancient Carthage has 13% net favourability (higher than Biden!).  

You’ve probably seen at least one of the various efforts to collect and present information about how, by most objective measures, the world is improving, such as Gapminder, Human Progress, and Our World in Data. The latest one is the Rational Optimist Society, who have just launched and are looking for feedback and volunteers. 

Maxwell Tabarrok’s assorted links on why the internet is good. 

Henry Farrell on the life of the Irish political scientist Tom Garvin.

Nabeel Qureshi on what he learned working at Palantir.

What is on the bookshelf of a veteran journalist covering China? 

Sam Kriss against lists of books. Thankfully, I don’t think link roundups and curated reading lists of non-fiction suffer from the problems mentioned in this piece.  

Nicholas Decker on his eleven favourite economics papers. I don’t think jumping in and immediately reading these papers is an especially good way to learn economics, as the title implies.   

Book review of the Eighth Day of Creation, an account of the origins of molecular biology. 

Scott Alexander’s review of the progress studies conference. Beatriz Geitner’s review of the progress studies conference. I showed up to the progress studies conference late due to some travel chaos, so I’d feel like something of a faker if I wrote up a review too. Nevertheless, the part of the event I did catch was excellent, and I’m grateful to Jason Crawford and the organisers for inviting me.

Dynomight on why you should display as many digits as possible in your scientific papers.

Terence Tao on how AI will change mathematics. 

Source.

Books

Robert Hazen, Genesis: The Scientific Quest for Life’s Origins. This was the best popular science book I’ve read in a long time. It covers, among other topics, whether hydrothermal vents are a plausible location for life’s origin, and the schism in the field over it. It’s always nice when a book like this is written by a major figure in the field – the book is partly a scientific memoir. I spoke with a few origins of life researchers this year, and this book seems to be widely respected as the best popular treatment. 

Michael Huemer, Paradox Lost: Logical Solutions to Ten Puzzles of Philosophy. Who doesn’t love a good paradox? In this book, Huemer gives his opinion about ten famous philosophical paradoxes, including the Raven paradox, the Sleeping Beauty problem, and the Liar paradox. Michael Huemer is brash and overconfident in just about everything he does, so be wary if he’s ever the only person you’ve read on any topic. Unlike the covers of his self-published books, which feature the highly questionable choice of Calibri font, this one actually looks nice, and would make a great gift.    

Niko McCarthy, Xander Balwit, Origins. This is a beautifully printed collection of essays from the recently launched biology magazine Asimov Press, a copy of which the editors were kind enough to gift me. My favourite essays were Wendi Yan on how artemisinin was discovered in Mao’s China, and Tom Ough on practical paths to blocking fomite transmission of disease. 

Matt Taddy, Business Data Science. I was sceptical of this book, as I am of any book that has the word ‘business’ in the title. But this is actually a great textbook, covering many topics in machine learning, advanced regression, and ‘big data’, for readers familiar with the R programming language. It was nice to work up to the final chapter on neural networks and deep learning, and the bibliography is helpful too. The book has no exercises, but does have worked examples where you download the data and play around with it yourself. I’ll be re-reading this one.  

Michael Breen, The New Koreans: The Business, History and People of South Korea. Michael Breen is an interesting fellow: after graduating from the University of Edinburgh, he moved to South Korea in 1982 – when it was still a military dictatorship – and learned fluent Korean. He married a Korean woman, rose to become president of the Seoul Foreign Correspondents Club, and even interviewed Kim Il-Sung (who lived until 1994). I strongly recommend this book if you have a Korean partner, or a lot of Korean friends. My favourite passage was about how you can actually buy land in the Demilitarized Zone!

I checked with the government and was told there is no private ownership in the DMZ... I found that only applies to the North Korean part of the DMZ. My real-estate agents in the towns of Shin Cheorwon in Gyeonggi Province and Haean in Gangwon Province told me a different story about the southern part. One said that in his rough estimate, only 30 to 40 percent of the southern sector is publicly owned, while 50 to 60 percent is privately held, and 10 percent of undetermined ownership... Moreover, the land [in the DMZ], although it couldn't be used for anything now, was being traded. For example, in November 2007, a Mr. Shin Joong-hyun donated 39,300 square metres of DMZ land he had bought in 1975 to the National Trust. If you want to buy some, here's what I found you can expect to pay: Rough land and hills, 2,000-3,000 Korean Won (KRW) per 3.3 square metres; potential farmland, KRW 10,000. (That's an average of US $2 for 36 square feet of rough land and US $9 for farmland). 

I assume there’s not enough liquidity for this to be possible, but using the price of DMZ land to infer the probability, by decade, of Korean reunification would make for an epic economic history paper. And, regarding whether Koreans are “the Irish of the East”:  

I came across a book of quotes by writers and other travelers waxing lyrical about foreign lands in which the Korea section was notable for being a) short and b) bitchy. Beatrice Webb, the socialist, was quoted saying the Koreans were “horrid”. The writer James Kirkup didn't seem to regard them very highly either. Western missionaries had dubbed the Koreans the Irish of the East, but he considered this an “insult to the Irish”. 

This book had all sorts of titbits about contemporary Korean society that I didn’t know about, such as the sheer scale of imprisonment of conscientious objectors evading conscription: 

The other shadow over the picture of religious freedom [in South Korea] is the jailing of young male Jehovah's Witnesses who reject compulsory military service. Even democratic administrations have made no effort to accommodate the rights of these pacifists. Eighteen thousand have been imprisoned since 1953.... Currently there are about six hundred Jehovah's Witnesses behind bars, giving South Korea the dubious distinction of being the world's leading jailer of conscientious objectors... Remarkably, this is a nonissue for Koreans.  

Overall one of the ways in which Koreans come across as most different from their neighbours culturally is the remarkable ease with which they adopt new religions: 

The total number of people who claimed affiliation to a religion increased during [a] seventeen-year period from 17 million to 82.1 million. That means that while only 42 percent of the population was religious in 1985, by 2002 it was 170 percent... As the Koreans believe the Jews punch above their weight in education and business in the world and want to find their secret, the Talmud is widely taught in popular accessible form as an education tool. 

Another great fact I learned from this book is that Syngman Rhee, who returned from 40 years of exile to become the first president of the Republic of Korea, received a history PhD from Princeton, which was supervised by none other than future president Woodrow Wilson!

Roy Foster, On Seamus Heaney. Foster is Ireland’s great literary biographer. One day I will tackle his two-volume treatment of W.B. Yeats, but that day is not today. At only 200 small pages, this book is a light and breezy interpretation on postwar Irish poetry. In chapter five, the description of the lectures which Heaney delivered in his role as Oxford Professor of Poetry is vivid, and makes me lament that they were never – so far as I can tell – recorded. Here’s a section from Clearances, in memory of Heaney’s mother:

When all the others were away at Mass

I was all hers as we peeled potatoes.

They broke the silence, let fall one by one

Like solder weeping off the soldering iron:

Cold comforts set between us, things to share

Gleaming in a bucket of clean water.

And again let fall. Little pleasant splashes

From each other's work would bring us to our senses.

So while the parish priest at her bedside

Went hammer and tongs at the prayers for the dying

And some were responding and some crying

I remembered her head bent towards my head,

Her breath in mine, our fluent dipping knives—

Never closer the whole rest of our lives.

Papers

Leepold Aschenbrenner, Situational Awareness: The Decade Ahead. This report argues, among other things, that there’s absolutely no way the US government will allow a private AI lab to invent superintelligence. Nationalisation is inevitable, so we may as well get started now, and prevent an authoritarian regime from developing AGI first. I’m quite late to the party in only reading this paper now; there was an enormous amount of attention surrounding it after Leopold’s appearance on the Dwarkesh Patel podcast. Leopold’s career has subsequently launched yet further into the stratosphere, most recently being cited as an influence by Ivanka Trump (!). I don’t feel qualified to write an in-depth response to this. Here’s a brief review from Scott Aaronson. All of this does engender the feeling that I have more urgent things to be doing than, say, reviewing books about Seamus Heaney’s poetry.     

Sergi Oms, Elia Zardini, An Introduction to the Sorites Paradox. This book chapter is so badly written and unnecessarily complicated that it fills me with rage. The sorites paradox sounds kind of silly when you just say it directly (‘if you keep adding grains of sand one-by-one, at what point does it become a heap?’), but it is related to the genuinely interesting question of how to formalise vague predicates in language. This also reminded me of Gareth Evans’ legendarily dense one-page paper Can There be Vague Objects? (his answer: no). 

Kit Fine, The Non-Identity of a Material Thing and Its Matter. My friend Peter McLaughlin, who has read more analytic philosophy than just about anyone I’ve met, said of this paper that it’s so abstruse that he doesn’t even care what the correct answer is. Most people reach a point of apathy much sooner.  

David Lewis, The Paradoxes of Time Travel. Here, Lewis is arguing that time travel is not logically impossible. Does it matter whether time travel is merely physically (‘nomologically’) impossible, rather than logically impossible? Perhaps not, but if we restrict ourselves to only talking about things that “actually” matter, then the justification for how I spend a lot of my time will crumble.

Adam Elga, Self-Locating Belief and the Sleeping Beauty Problem. This paper popularised the now-famous Sleeping Beauty problem; see my talk about it here. A piece of academic lore I gleaned from the footnotes is that Elga learned of the problem from Ariel Rubinstein, who you will recognise if you’ve taken an intermediate microeconomics course as the source of the Rubinstein bargaining model. Another early person to write about sleeping beauties was Robert Aumann from Aumann’s Agreement Theorem. Some nice crossovers!  

Michael Titelbaum, Ten Reasons to Care About the Sleeping Beauty Problem. Mostly I find any kind of meta framing of an intellectual issue (about why someone would be interested in something) tiresome. I liked this paper, but the writing quality isn’t great; you would think that in a paper entitled ‘Ten Reasons…’, the section headers would be those reasons, right? Instead, the section headers are vague and meandering. This paper draws a link between decision theory and Sleeping Beauty that I hadn’t considered before. There’s a relatively widespread view that causal decision theorists should be ‘thirders’, while evidential decision theorists should be ‘halfers’. ‘Thirders’ should also be more likely to accept the many worlds interpretation of quantum mechanics. 

Sharon Ryan, The Preface Paradox. The preface “paradox” arises when an author believes every statement in a book, but nonetheless includes a preface saying that there must be errors which she failed to catch. How is it possible to believe, and be justified in believing, a series of statements, but also to believe that the conjunction of those statements is false? An obvious response to this is to say that we are necessarily less than 100% confident in each belief, so, while each individual belief could be justified, when you multiply all the probabilities together, you’ll think that there’s less than a 50% chance they’re all correct. This view has lots of subtle issues, including how you get all the conditional probabilities to work out, and deal with highly interdependent beliefs.     

Julia Staffel, Three Puzzles about Lotteries. Many people are tempted to say that being ‘justified’ in believing something or having an ‘outright’ belief just means your confidence in it is over a certain threshold. But, if this is so, then, for a lottery where the probability of winning is below a certain level, for each ticket, you would be justified in believing that that ticket will lose. Yet you wouldn’t be justified in believing that all of the tickets will lose, because you also know that one of the tickets is going to win. This is a parallel problem to the preface paradox, involving moving between individual beliefs and conjunctions made up of those beliefs. I am personally fine with saying that it’s ‘probabilities all the way down’, but it does take some effort to get this view to work. Staffel’s paper is about one particular resolution to these kinds of problems, which is to say that justification is not about absolute degrees of belief, but about ratios of degrees of belief under different possible outcomes. 

Robert Nozick, Newcomb’s Problem and Two Principles of Choice. Nozick is perhaps best known for his libertarian screed Anarchy, State, and Utopia. This is the paper in which he popularised the Newcomb problem, about which I probably will write more at some point. Newcomb’s Problem started its life as a brain-teaser told by the physicist William Newcomb. What I find funny about this is that Newcomb was not, so far as I can tell, famous for any other reason. He succeeded in creating a puzzle so interesting that it immortalised his name forever.

Music 

The great tenor saxophonist Benny Golson has passed away. Here he is in 1959 playing ‘hard bop’ with Curtis Fuller and Tommy Flanagan. For some reason, Spotify has uploaded the title of the first track in Japanese. I think Golson’s most famous composition is the jazz standard I Remember Clifford.         

Speaking of recently departed jazz legends, Quincy Jones died a few days ago. You’ve Got It Bad Girl is a superb album, although he is more famous as a producer. The amount he influenced 20th century music is truly astonishing.   

…in fact, another great jazz musician, the pianist Charles Bell, has also recently died. Here is Richard Brody reviewing his criminally underrated career: “Bell took a day job, teaching music in Pittsburgh public schools, but his original methods were considered dubious and he was fired”.

Herbie Hancock, Man-Child. I love Herbie.   

Gil Scott-Heron, Reflections. ‘B’ Movie is a hilarious satire of Reagan.   

Ornette Coleman, Free Jazz. Is there a reason to listen to this album, other than historical interest in the development of jazz improvisation? To give a sense of just how avant-garde this recording is: The left and right ears of your headphones play two separate, uncorrelated, tracks. I listened to this on a turbulent flight, and I’m not sure which was more unpleasant. 

Podcasts

Seamus Heaney appeared on Desert Island Discs in 1989 to discuss his eight favourite songs. Heaney had an utterly delightful and melodious way of speaking; here he is answering a question about his childhood:

Our family came very quickly, one after the other… There were nine of us, I think probably six of us born inside eight years… There was no menace, other than the menaces that are in the imagination. The dark, and the trees, and the scuttling of wild things on the ceiling at night, and so on.     

Here he is on how Northern Irish people can even tell Catholics and Protestants apart: 

It was extremely subtle. It was a gram of psychological difference in the weighting of the way people behaved with each other. When, shall we say, a Protestant neighbour came to the door to collect milk, there would just be a faint delicacy of decorum, more decorum, observed.   

And on the various attempts for him to be made the spokesman of the Catholic side in the Troubles: 

I haven’t been a spokesman for, I think, any cause. Anytime I have spoken, I have spoken, usually impatiently, and out of genuine personal unease, and a desire to clear my own slate, in order not to deceive either the British or the Irish. I don’t think I have a programme. I have a set of dispositions and affections, and a desire, as far as possible, to conduct myself honestly. 

A simply marvellous recording – makes me homesick. 

How did the Hanseatic League affect mediaeval European trade? 

An overview of the political career of Benjamin Disraeli. Disraeli was 69 years old when he started his main term as prime minister; it is interesting how Britain once had a political tradition of elder statesman figures becoming prime minister after long careers in the Commons. One of the figures mentioned in this podcast as a precursor to Disraeli as a Jewish member of parliament is the economist David Ricardo (he of ‘comparative advantage’ fame). Ricardo represented the Portarlington constituency in Laois, Ireland. Very few people seem to be aware that, despite its vanishingly small Jewish population, Ireland technically had a Jewish parliamentary representative as far back as the 1810s. 

What does the Ramayana actually say?  

The basics of mitochondria and bacteriophages. There are ten times as many phages as there are bacterial cells on Earth – amazing! Both this podcast, and an essay from the Asimov Press booklet, discuss the association between bacteriophages and the country of Georgia. Phage therapy was discovered in Georgia, and for decades a significant fraction of all experts in the field lived in, or were from, Tbilisi. The Soviet association hindered bacteriophage research, which reminded me of Scott Alexander’s great essay about how poorly Russian scientific discoveries were communicated to the West: An Iron Curtain Has Descended Upon Psychopharmacology.    

A new largest prime number has been found. By a retired Nvidia engineer self-funding compute on H100 chips in order to get a better first-principles understanding of mathematics, no less! ‘The World’s Largest Prime Number’ by Michael Perusse is now worthless, and I’ll have to find a new joke gift for my mathematical loved ones. 

Nate Silver on gambling, Sam Bankman-Fried, and his criticisms of effective altruism. 

My friend Julia on underrated policy ideas for the UK. 

Matt Levin and Katie Greifeld interview John Collison, the president of Stripe. 

Rick Rubin chats with Tyler Cowen about classical music, with a focus on the Russians. 

Some insight into the songwriting of Crosby, Stills, Nash and Young. 

The economist Larry Summers joined the board of OpenAI in its hasty reconstruction following the drama of Sam Altman being fired. Here he is being interviewed by Joe Walker about what he’s learned from the experience. I have to respect Tyler Cowen’s wordplay here: “better an AI Summers than an AI winter”! [Edit: This is funnier than I thought. After this post was published, I found out that the person Tyler “overheard” this pun from was Joe Walker himself!]

A podcast I enjoy, and wish I listened to more, is the Partially Examined Life. They split up the episodes and clutter the feed with other podcasts in such a way that makes deciding to listen unnecessarily psychologically painful. In any case, here they are on Timothy Williamson’s influential ‘knowledge-first’ approach to epistemology. And here is Williamson as a crochet puppet.  

The Marginal Revolution podcast on price controls and oil shocks. 

The Empire podcast on the Mau Mau uprising against British rule in Kenya.

The Rest is History on the presidential campaign of Eugene McCarthy and America in 1968. 

Henry Oliver and Nabeel Qureshi on why literature requires the fuller engagement of the soul. 

Films

Ken Burns, The Vietnam War. I have now reached episode nine of this series. It reaches a low point around episode five, hence my previous harshness. But so much happened in 1968 that it would be difficult to fail to make it interesting (the documentary is chronological, and dedicates over two hours to 1968). I suppose at some point I must have learned that Henry Kissinger most likely encouraged South Vietnam to withdraw from the Paris peace talks to make Lyndon B Johnson look bad and improve Nixon’s election chances (and, implicitly, lengthen the war…). But I had failed to register how completely insane this is. If 1968 were a normal year, this would have been the biggest scandal of the year, and everybody today would know about it. A presidential candidate with no formal power was committing treason – and LBJ knew about it! I suppose it’s a bit late for me to only be getting upset about this now, but Watergate really was just the tip of the iceberg…  

Todd Haynes, The Velvet Underground. The first (and only) documentary about Lou Reed’s avant-garde rock band with John Cale from the 1960s. The flashing images and sounds gave me a headache which lasted several hours. The experimental format is fitting for the subject matter, but I can’t quite say it’s my thing.  

David Bickerstaff, Van Gogh: Poets & Lovers. This was a digitised version of the London National Gallery’s Van Gogh exhibition, covering the 1888–1890 period during which he lived in the South of France, especially the town of Arles. I dislike how most galleries overload you with more artworks than anyone could realistically process. I prefer the format of art historians curating a small selection and talking through the selection and relevant background, while the works are presented one at a time. It’s nice to see art in person, but it seems most people (including myself) do it prematurely, before they have any idea what they’re looking for. Something about this film which made me chuckle is that the Van Gogh experts interviewed pronounce his name in three different ways, and this is never commented upon.    

The Hospital at Arles

From YouTube, I enjoyed Grant Sanderson on where facts are ‘stored’ in large language models, and how he animates his videos using his Manim Python library. Also, recommendations of the best books about Python, and the Masego Tiny Desk Concert.