Links & What I’ve Been Reading, September

Here are the highlights from what I’ve been reading, watching, and listening to in the last month. If you know of something I might like, you can email sam (at) thefitzwilliam (dot) com.

If you’re looking for more articles to read, you should join Lynkmi, an experimental new link-sharing website created by my dear friend Oisín Moran. It’s like a niche version of Hacker News for my extended friend group; you can join the (short) waitlist here.

I am a Non-Resident Fellow at Progress Ireland, a new think tank in Dublin, aiming to improve policy in housing, infrastructure, and metascience. Please take a look. You can look at the founding essay here and donate here. 

The Institute for Progress is hiring for several roles, including an ‘assistant editor’ role for Santi Ruiz (I recommend his interview about how it was discovered that the 1979 anthrax outbreak was a lab leak). If you successfully apply and I referred you, they will give me some money, so please apply. 

The polymath Gavin Leech has a new personal newsletter; you should subscribe. 

Patrick Collison’s account of the Silicon Valley canon. Also, from Andrew Gelman, what should or shouldn’t be in the Bayesian canon? 

My friend Yudhister on neighbourhood semantics, a field of modal logic that generalises the concept of Kripke frames. I think he based this blog partly on notes I sent him from a university course I took about this. He seems to have understood it better than I did, and he’s still in high school. Whoops.    

Janan Ganesh, at the Financial Times, has elevated naval-gazing into a high art. 

Why you should read Ludwig Wittgenstein’s Tractatus, something I will probably never do. 

Henry Farrell’s syllabus for the new global politics. 

Scott Alexander’s links for September: “I’m not saying Satoshi Nakamoto was a CIA asset, but isn’t it weird that “Satoshi Nakamoto” is Japanese for “central intelligence”?”

Stephen Malina: what is special about the ‘energetic aliens’? And Stephen’s tier list of favourite internet writing. 

Book review of The Real North Korea. The author of this book has an interesting backstory: he grew up in the Soviet Union during glasnost and was chosen to study at Kim Il-Sung University, to advise the North Koreans on industrial policy. Now he speaks fluent Korean and has a teaching position at a university in Seoul.    

I hadn’t realised quite the extent to which high-rise buildings are much more common in Canada than in the United States. We often bemoan English-speaking countries’ inability to build physical infrastructure, but perhaps Canada is a partial exception to this.

My friend Luke Fehily wrote about how and why to reform the system of research funding in Ireland. 

Why do some people have seven houses? Because they’re not like you and me. 

Scott Aarsonon’s chronicles of wasted time. 

A list of the most commonly cited philosophers on the Stanford Encyclopedia of Philosophy. Unsurprisingly, David Lewis comes in at #1. What might come as a surprise is how infrequently certain globally influential philosophers are cited in English-speaking academia. For example, Michel Foucault is only #187 on this list. And in case you care: ranking philosophical rankings. 

In the past two months, I listened to the entire back catalogue of Drum Tower, the Economist’s podcast about China. I subscribed to Economist Podcast+ specifically for it, as only a few of the episodes are available for free. To pick out one particular episode I liked, they had a fun one about the Mongolian ethnic minority in Yunnan province, in southern China, which claims – almost certainly falsely – to be directly descended from Genghis Khan’s army.  

Dan Shulz had a superb conversation with Henry Oliver about late bloomers and the humanities. 

Paul Bloom and Philip Ball both appeared on Conversations with Tyler, self-recommending.

Very Bad Wizards finally discussed The Shining, and it did not disappoint. 

The Empire podcast had a mini-series on the rise of Ho Chi Minh and the Vietnam War. This was an excellent tie-in with Ken Burns’ series. 

Matt Clifford was on the ChinaTalk podcast to discuss the AI arms race, the Bletchley Summit, and various UK government efforts in AI safety. 

The Rewatchables discuss The Grand Budapest Hotel.  

Lant Pritchett reflects on his career in development economics.

I don’t listen to enough music podcasts, for the simple reason that I listen to podcasts on ~1.8x speed, which gives everybody squeaky voices. But I have been trying out Song Exploder, where a musician explains the background to one of their songs, with the song playing in full at the end of the episode (e.g. Laufey, Franz Ferdinand).  

Keith Jarrett, Treasure Island. Somehow, I had missed this 1974 piano-heavy jazz album, of which my favourite track is ‘The Rich (And the Poor)’. Keith Jarrett is one of my favourite pianists, and his 1975 Köln concert might be my favourite solo piano performance of all time. That concert also has an interesting backstory which has passed into jazz legend.  

Rebecca Pan, 我的心. Rebecca Pan (潘迪華) is a Hong Kong singer who sings in a mix of Cantonese and English. This album is best known for the cover of the Indonesian folk song Bengawan Solo, which appears on the soundtrack for In the Mood for Love. 

Art Pepper, Friday Night at the Village Vanguard, especially ‘But Beautiful’. Alas, I didn’t get a chance to visit the Village Vanguard club the only time I was in New York, but it’s certainly on my bucket list.   

Miles Davis Quintet, Live at the Olympia 1957. Recently unearthed recording of the First Great Quintet during the period that Miles Davis lived in Paris. This album wasn’t available when I assembled my Miles Davis listening guide; if it had been, I would have included this rendition of ‘What’s New?’.      

Frank Barry, The Irish Single-Currency Debates of the 1990s in Retrospect. Many economists (e.g. Joseph Stiglitz) believe that the introduction of the euro was a catastrophic mistake. This paper is a helpful overview of how Irish academics and policymakers thought about the decision to adopt the euro, from one of the influential anti-euro voices at the time. For most of its post-independence existence, Ireland has been in a currency union with Britain, and Britain of course declined to join the euro. Even if you are pro-euro, it’s still important to understand how such a consequential decision was made. The bottom line is that the arguments both for and against the euro were not very rigorous and the overall standard of discourse was disappointing. I almost never hear people mention how, just a few years after the introduction of the euro, the proliferation of credit cards and contact payments eliminated much of the claimed benefit of currency union.

Natalie Wynn, An accidental Galut? A critical reappraisal of Irish Jewish foundation myths. I thought I could get away with only reading one book about the history of Jews in Ireland, but Cormac Ó Gráda’s book on this topic has shortcomings, and I was pointed to this paper by someone whose MA was about the history of Zionism in Ireland. This is one of those papers where it felt like I was being vaguely browbeaten for something I wasn’t tempted to believe in the first place. This paper is, among other things, about the various tall tales of how a population of 5,000 or so Jews ended up in Ireland in the early part of the 20th century. One urban legend is that calls for ‘Cork, Cork’ were misheard as ‘New York, New York’. Another myth the author discusses is the idea that Irish Jews came from a very small area, or possibly even just a few shtetls, in Lithuania. In fact, Irish Jews came from a reasonably wide area within the Western Russian Empire.  

Andrew Gelman and Aki Vehtari, What are the most important statistical ideas of the past 50 years? A highly informative overview of some of the key developments in statistics in the last half-century. Ever since learning the amazing anecdote about how the t-distribution in statistics was developed at Guinness, I’ve wanted to write something about the history of statistics. On my reading list is Stephen Stigler’s 1986 book The History of Statistics; perhaps I’ll write up a full review (if I understand it!). If you’re not familiar, Andrew Gelman’s blog is a gem.  

Brian Skyrms, Zeno’s Paradox of Measure. If you crack open a popular maths book, it may tell you that Zeno’s paradoxes were resolved with the advent of calculus. Calculus gives an intellectually coherent account of how an infinite number of quantities can sum to a finite number. This is both historically misleading and also misunderstands what Zeno’s paradoxes actually are. The paradox is not about how distances are possible, it’s about how motion is possible. The less inaccurate historical account is that Zeno’s paradoxes were resolved in the late 19th century when mathematicians like Lebesgue and Peano developed measure theory. But arguably that doesn’t resolve the philosophically interesting part of the paradox either. This paper also contains a discussion of the famous Banach-Tarski paradox.  

UNIDO, Sri Lanka Final Report on Contact 78/34. I did not expect my life to unfold in such a way that I would be reading hundreds of pages of reports on meetings by Sri Lankan politicians in the 1970s. But here we are.

One of the topics I’m interested in is what advice rich countries have given poor countries about development, and whether it’s been any good. A small piece of that story is how, in the 70s and 80s, the United Nations Industrial Development Organisation started promoting the idea of ‘export processing zones’, or special economic zones, where you designate some area of land as having special rules, such as not requiring import and export duties. My current research is about the remarkably obscure story of how a small group of Irish men associated with Shannon Airport served as consultants to a wide range of developing countries and advised them on these kinds of export-oriented development strategies.

Fischer Black, Noise. Fischer Black was one of the sources of the famous Black-Scholes equation in derivates pricing. This paper is one of the most influential ever in finance, and Black wrote it while working at Goldman Sachs. It gives an overview of the influence of the idea of statistical noise on different fields in economics. I’d recommend this paper if you’ve read Nassim Taleb and you’re looking for something less vague to read about the importance of randomness in finance.

W.V.O Quine, The Ways of Paradox. Influential attempt to define and understand what a paradox is. He discusses a fun one called the Berry paradox, which arises from sentences like ‘The least number not specifiable in less than nineteen syllables’ (which has just been specified in 18 syllables). PS: Matt Yglesias’ conclusion from studying philosophy at Harvard is that people should read more Quine!

William Lycan, What, exactly, is a paradox? This paper is more modern and readable than Quine and covers much of the same material. 

Ronald Lee, The Demographic Transition: Three Centuries of Fundamental Change. I don’t think it’s widely known that, while, most of the time, mortality declines precede fertility declines in the demographic transition (producing high population growth), there are exceptions to this. France and, to a lesser extent, the United States, saw large fertility declines before mortality declines or modern birth control. This is a review paper from the Journal of Economic Perspectives, and I again lament that I can’t find an equivalent of the JEP for more fields.          

Stephen Wolfram, What is ChatGPT… And Why Does it Work? This is a short ebook that is also available for free online. This covers largely the same ground as the 3Blue1Brown series – transformers, softmax, vector embeddings, the attention mechanism, classifiers, etc. I enjoyed it. 

Francois Flueret, The Little Book of Deep Learning. Another short ebook available for free. This book is a good deal more technical than Wolfram’s, and I would certainly recommend reading something more introductory first. I can’t say I understood most of Flueret’s book, but it’s a good resource and I’ll be revisiting it.  

Anil Ananthaswamy, Why Machines Learn: The Elegant Maths Behind Modern AI. This is one of the better popular science books I’ve ever read, and it has the rare characteristic of actually discussing the mathematics of its subject matter. The writing style is very American – take from that what you will. I’ve also been looking for materials on the history of AI, and this had a lot on that front.    

Geoff Dyer, But Beautiful: A Book About Jazz. The first novel I’ve read in an embarrassingly long time. But Beautiful consists of fictionalised accounts of well-known stories about jazz musicians, and brief biographical sketches inspired by photographs, with some non-fiction reflections about jazz and its influence at the end. Geoff Dyer seems to be doing for jazz here what Benjamín Labatut did for physics in When We Cease to Understand the World. How many novels have a ‘selected discography’ section of what songs you should listen to while you read it? Truly excellent. 

David McCullough, The Path Between the Seas: The Construction of the Panama Canal 1870–1914. If you only read one book about the Panama Canal, this should certainly be it. It’s a terrible shame this book hasn’t been adapted into a film, as the story has so many colourful characters and touches on so many themes in French, American, and Central American history. I started reading this book because, last year, I came into possession of a French Panama Canal Company bond, i.e. one of the physical pieces of paper which financed the canal’s construction in the 1880s. It’s a long story – I’m working on a full essay about it. 

Michael Palin, Sahara. Another ‘popcorn’ book of Michael Palin’s travelogues. I thought the most interesting part was the part where he talked with members of the Sahrawi Arab Democratic Republic, the group which claims Western Sahara as a country independent from Morocco. It’s striking just how few people are involved; Wikipedia says that the entire Sahrawi ethnic group only has about 600,000 members. I was hoping to go to Senegal this year, but it didn’t work out; I am mostly saving more serious Africa reading until a trip there is on the horizon.  

Tobias Harris, The Iconoclast: Shinzo Abe and the New Japan. Shinzo Abe is sometimes described as the second most important East Asian politician this century. This book is truly excellent, and I recommend it if you have even the slightest shred of interest in Japanese politics. If you’re not familiar, Abe was the longest-serving Japanese prime minister ever, and pushed for many important reforms, including a more expansionary monetary policy from the infamously conservative Bank of Japan. Something interesting about this book is that it only covers up to July 2020, i.e. slightly before Abe resigned as prime minister, and before he was assassinated. It can be interesting to read books written about certain figures before some transformative or well-known event involving them occurred. I also previously linked to Jordan Schneider’s highly entertaining conversation with the author on Chinatalk. 

Bryan Caplan, Build Baby Build: The Science and Ethics of Housing Regulation. I love non-fiction graphic novels, and I wish that more academics would communicate their ideas in this format. Having said that, this book was a bit of a disappointment compared to Open Borders, Caplan’s previous graphical novel. It’s a shame that he was unable to retain Zach Weinersmith as the artist for this book – I don’t like the new guy as much. There’s not much new material here for people familiar with YIMBY arguments and the work of scholars like Ed Glaeser (on the other hand, there were a lot of interesting anecdotes in Open Borders even for people who are reasonably familiar with the economics of migration). I would have liked to see a more serious discussion about why so much new construction is ugly, and how the most beautiful and functional cities were constructed historically. The extent to which increasing the supply of housing is unpopular, even among the people it would help the most, really is astonishing. Harkinson 2018 finds that 42% of homeowners, and 35% of tenants, support banning new construction in their areas. Only 28% of homeowners, and 59% of renters, support allowing the housing supply in their city to increase by 10%. I just signed a lease on a new flat in a particularly overheated market, so these issues are particularly salient to me right now. As with many books aimed at a general audience, if you’re not reading the endnotes, you’re not getting much (most?) of the value.

Erwin Schröodinger, What is Life? The Physical Aspects of the Living Cell. This is a short and influential book which is close to my heart. Watson and Crick credited it as one of the influences on looking for the DNA double helix. It’s based on a series of lectures that Schröodinger gave in Dublin, in a lecture theatre I spent many evenings in as a teenager. I once went to a conference inspired by this book, but to my shame had never actually read it. The book is modern and philosophical, and the prose is crisp. My Cambridge University Press version also includes his essays ‘Mind and Matter’ and ‘Autobiographical Sketches’, the latter of which gives some more context about his time in Ireland.

Mike Newell, Donnie Brasco. This is the first mafia film I’ve seen not from Scorcesse or Ford Coppola. I thought it was hilarious how this film focused on the underappreciated lower-middle-class gangster. This film, I learned, popularised the slang term ‘fugazy’ (fake). Strong recommend.

Michael Wood, The Story of China. This is a six-part documentary series about the history of China which I watched with some friends. Supposedly Xi Jinping is a fan. Michael Wood really does look delighted to be there in every scene. I would love to watch a much more detailed and longer version of this series. If Ken Burns were to direct a 10-part documentary about the Mao era, I think it would become one of my all-time favourite films.   

Sofia Coppola, Lost in Translation. I thought this was excellent. The scene where the Japanese director’s words are clearly being unreliably translated was my favourite, and Tokyo is used very well as a setting. Some people think that this film is mocking Japanese people or culture, but I thought those elements were clearly poking fun at the protagonist and his preconceptions. It’s striking that Scarlett Johansson gave such a nature performance when she was only 17 (!) – another data point toward the view that almost everybody who is distinguished or famous was doing remarkable things at young ages. 

Akira Kurosawa, The Bad Sleep Well. To my shame, this is the first Kurosawa film I’ve seen. It shows its age in certain ways, including in how blatant the exposition is. I thought the opening scene with the photographers was a masterpiece, but later parts of the film drag. Eventually I’d like to work more systematically through the films using Donald Richie’s companion to Kurosawa. I also recommend this Reddit user who has written in-depth reviews of every Kurosawa film. This is one of the many films which is said to be partly based on Hamlet, which is a description that is used… incredibly loosely. Is having themes of familial betrayal, indecision, and murder enough to be ‘based on Hamlet’? 

Tomm Moore, Ross Stewart, Wolfwalkers. This was an absolutely delightful animated film set in Kilkenny (in the middle of Ireland) during the Cromwell era. Although the film is aimed at children, it has a more nuanced treatment of environmentalism and the tradeoff of environmental destruction than the vast majority of films for adults. Ireland in recent years has been strong in cinema; it’s nice to see that translated to animated films.    

Finally, from YouTube, I can recommend Andrej Karpathy’s one-hour talk about large language models, and WelchLabs on the significance of AlexNet.