2019 By The Books
Last year at around this time, I published a review of 2018 as told by the books I’ve read throughout the year. Since then, I’ve changed my mind about some things, but I still believe that this is a much more fruitful type of review than the typical condensed news and celebrity obituaries way of looking back, no less because it’s unique to everyone.
All in all, 2019 saw a lot fewer books being finished (20 vs. 36), which I mostly blame on two tomes — Infinite Jest and Gödel, Escher, Bach — that both took me a long time to digest. That’s not necessarily bad (both were extremely rewarding books), but since me book wish list tends to grow much faster than I can read, this poses a serious problem for mortal beings.
Here we go:
23 Things They Don’t Tell You about Capitalism by Ha-Joon Chang
Against The Grain By James C. Scott
An intriguing and, certainly, unorthodox account of the history of the very early states. Scott makes an interesting case that the transition from hunter-gatherer tribes towards states — “civilization”, as some have it — has not been a history of progress. Rather, it was a large-scale enslavement operation — making use of coercive measures that long predated the state but scaling them up to hitherto unknown levels — without creating much stability. During their initial period, they were a valuable tool for expropriation, but the business model proved unsustainable, as its “citizens” fled to the periphery and the more flexible “barbarians” systemically exploited its weaknesses. What’s more, they brought into existence a good many diseases, drudgery and unprecedented degrees of subjugation.
Whether or not this is the best explanation of the (rare) available evidence I do not know. But in addition to offering a fascinating prehistoric account, it also indirectly makes the case for what I believe is socialist anarchism, a position I never really understood before. For the first time, I found myself thinking that it’s at least somewhat plausible, for which the author deserves credit.
The title of the book, by the way, comes from the decisive role that grain as a crop supposedly played for early state formation, especially its properties of being relatively predicable, harvest-wise, easy to divide and to store, thus making it an ideal object for taxation.
Gödel, Escher, Bach: An Eternal Golden Braid by Douglas R. Hofstadter
I had high expectations for this book, and I wasn’t disappointed. This might well be one of the best books I’ve ever read, and not just in philosophy — although I feel I will have to read it again to fully appreciate the message. At some point, I really want to write a longer reflection on it, but for now…
First thing you’ll notice (in the updated edition) is that Hofstadter seems to think almost no one understands what his book is about, which makes reviewing it a daunting enterprise. It’s not about Gödel, Escher and Bach (well, sort of, but they’re used for illustrative purposes only). It’s also not about how, deep down, music and math and art are really all the same at the core. My best take is that he attempts to understand what intelligence really is from a naturalist viewpoint. We know that we are all made of a handful of simple particles, who certainly experience no such thing as consciousness or meaning. So how do we get from matter to mind without invoking something like a soul? We often talk about emergence and higher levels of abstraction, but how well do we really understand what’s going on here? His starting point is the study of formal systems, in some sense the simplest area where these issues arise. Should we argue that these are merely arbitrary symbols that can be transformed into one another by equally arbitrary rules, as the logical positivists would have it? Do we have to ascribe some mystic quality to them, as Plato (and many of his successors) thought? Or how do we bridge the gap between scribbles on a piece of paper that may well be internally consistent to something that clearly has meaning (like, say, the integral of a function)?
How symbols attain meaning that lies strictly outside of the system (transcendence) is thus the crucial question that he explores at length — how different level are interwoven in such diverse areas as biology (the genetic code), music (the structure of fugues), language or computer science. As a simple example, think of a fractal: There is a purely formal (recursive) definition of it as a series, but the most natural way to think about them is by representing them visually as geometric structures that are repeated at larger scales. There is nothing that says “butterfly” in the first definition, and in fact, you could never guess that something resembling this would appear if you were to plot the series. And yet this is precisely what happens. Could something similar be true of neurons and their connections (the “formal definition” of a brain) and consciousness? Could we, by modelling this structure in a different environment (something that, loosely speaking, is what’s being done in deep learning these days), artificially create intelligence? Interestingly enough, while Hofstadter (anno 2019) is still very skeptical about the prospects of current AI research, he argued back in 1978 that it must be feasible.
The book is quite long and a bit dated at times (like the 70s-flavor pseudocode sections that look nothing like modern programming languages). It also suffers, especially towards the end, from an overly forced attempt to artistically include the above-mentioned interweaving of levels in the book, such as when the title of a chapter is the same as the first letters of every sentence in the dialogue it contains. But these are minor points, and I recommend it hands down.
How to Lie With Statistics by Darrell Huff
I honestly wonder who recommended that book to me — it’s 50% bleeding obvious and 50% junk science. If you want to read an even more bastardized version of statistics than Taleb’s, look no further.
Infinite Jest by David Foster Wallace
In Infinite Jest, Wallace tells a familiar story about an avantgarde filmmaker, his tennis academy, a nearby rehab facility and a wheelchaired Quebecois separatist that seek to use one of the filmmaker’s morbidly entertaining movies to further their political agenda. It’s everything you’d expect it to be (long, convoluted and hard to digest) and it took me several attempts to finish it, the first dating back to early this year. Even when I finally started liking it (after a few hundred pages), it still took me several weeks more to get through. But even though I really, really hated it at times, it didn’t let me go, and I still find thinking about the connections between the characters, the references I largely missed (and now discover thanks to a massive Wiki that has been created to help people make their way through the book), and the subtle reflections on the human condition that are hidden beneath the superficial silliness of the plot.
Knowledge And Decisions by Thomas Sowell
Many people had recommended Sowell to me during the last years, but it took me a while to finally read something written by him. That’s too bad, because he really is a great author. The first part of the book — which lays the theoretical foundations for an incremental approach to decision-making in social affairs (including, surprisingly, for things we usually only talk about in absolutes, such as justice), and explains the pervasiveness of “knowledge costs” (which differ vastly from person to person), is excellent, in particular because he doesn’t try to argue in favor of some specific ideology based on these insights. He merely stresses that no social system — be it communist, capitalist or anything in between — can’t escape these fundamental constraints.
The second half, in which these insights are applied to current affairs (that is, 1990s political talking points) is much less convincing. Until I got there, I didn’t understand why he was considerate a conservative by many, but his takes on discrimination, the death penalty and a host of other issues make that abundantly clear. It’s not that he doesn’t have anything valuable (or sometimes even original) to say about these topics, but his account is too one-sided.
Predictably Irrational by Dan Ariely
Of all the behavioral econ books I’ve read, this is probably the weakest. Part of this verdict surely stems from the fact that I’ve read my fair share of the field, there’s hardly anything new or surprising in here — the experiments are quite similar, and likewise for the conclusions. Two other things that disappointed me were the lack of curiosity about why people would behave as they do — isn’t irrationality a self-defeating strategy, likely to be punished severely by natural selection? Additionally, more than any other BE books, this one seems to offer a particularly inaccurate caricature of standard economics, which left me wondering if he’s not making his life too easy.
Stuff Matters: Exploring the Marvellous Materials That Shape Our Man-Made World by Mark Miodownik
“A thrilling account of the modern material world”, according to the Wall Street Journal, Miodownik sets out to shows to his readers that material science is anything but dry and boring. It was a quick and interesting read about a subject that I admittedly don’t devote much conscious thought to. However obvious some things seemed to me after he explained them (you can look through glass because visible light doesn’t have the correct wavelength to lift its electrons to a different energy level), I noticed how many of these things you just take for granted without ever thinking much about them. Not very deep, but a nice read.
The Agile Gene: How Nature Turns on Nurture by Matt Ridley
Ridley’s books are fun to read because I’m sitting relatively far at the bottom of the learning curve of his field, so the marginal gains in knowledge are large. Despite many fascinating stories about animal behavior that he uses to illustrate his story, the core message of is somewhat trivial: Nature and nurture aren’t exclusive concept, and we’re shaped by both to some extent. Indeed, I often struggle to understand why there is such a fierce debate between both sides, with ‘extreme determinism’ or ‘blank slate-ism’ being the milder terms that were coined to describe their respective opponents’ approach.
The Folly of Fools: The Logic of Deceit and Self-Deception in Human Life by Robert Trivers
A most bizarre book by who’s arguable the authority in the field. I found myself wondering again and again what the heck the editor was doing with the manuscript, as the final product is a collection of interesting facts, marred by an impressive lack of structure. Chapters are at best loosely connected, the order seems to be entirely random, and it’s hard to figure out what, if any, his key message(s) might have been.
In addition to that, it suffers from some other oddities. One of them is Trivers’ inability (or the editor’s, depending on your point of view) to leave out irrelevant personal trivia. Constant references to his buddy Huey Newton, former chair of the Black Panther Party, and most definitely not relevant researcher in the field of self-deception, stand out as particularly annoying. (Lowlight: He (favorably?) cites Newton’s advice how to deal with somebody who pretends to be dumber than she is in order to avoid working, namely, “physical or verbal assault”. You read that correctly.) Many of the other stories, while not quite as bizarre and obnoxious, make him look like a creep, too.
All of which is a pity, because analyzing why, against what we’d intuitively assume, evolution might favor self-deception, is a great theme for a book, and as the world’s foremost expert on the topic, he could have made a lot more out of it. But while the earlier chapter are still quite interesting, it gets worse and worse towards the end, especially the sections on war and religion. For example, he seems to believe that a US citizen who opposes US military interventions has definitely overcome the “us-vs.-them” fallacy. That’s not very deep, but — in a doubly ironic way — serves his purpose of demonstrating how pervasive self-deception can be well.
The Gene: An Intimate History by Siddhartha Mukherjee
My first thought when I started reading this: Why can’t everybody write like Mukherjee? Why is academic prose often either (a) informative but dry or (b) accessible but dumbed-down? This was a rare instance of a book that was a pleasant yet informative read from first page to last. I still think I don’t nearly know enough about genetics and biotechnology to really understand all the concepts he introduces and discusses, but at the very least, it’s a great place to get started. Highly recommended.
The Grand Chessboard by Zbigniew Brzezinski
Written much along the same lines of Samuel Huntington’s Clash of Civilizations, but without the verve and therefore much less controversial. The grand chessboard, on which the United States have set up their game, is Eurasia (Latin America, Africa and Oceania are completely ignored in his account), and the question the US faces is: which alliances, treaties, regional players should be strengthened or weakened to secure America’s position? The suggestions he makes are far from revolutionary and, in fact, often so broad that it’s not clear what exactly he has on mind. How this book managed to heap so much praise onto itself is honestly beyond me; from the little that I know, I feel that in international relations, rigorous testing of hypotheses has yet to be established as the norm.
The Master Algorithm by Pedro Domingos
It’s a great read if you are already familiar with aspects of machine learning; if not, it might feel overwhelming. Domingos introduces an impressive number of technical concepts (in a good way, but the sheer number will make your head spin), and I’ve found myself going back and forth to remind myself of their meaning. For someone who’s principally been studying deep learning, it provided a great overview of what he calls the other four tribes of machine learning, and does an excellent job to explain their strengths and weaknesses. I’m much less convinced that the road towards a general-purpose learner (the “Master algorithm”) consists mostly in “connecting the dots” of already existing techniques.
The Nurture Assumption by Judith Rich Harris
Harris was as unlikely a candidate as anyone could be to write one of the landmark books in child psychology. Having been dismissed from Harvard’s PhD program in 1960 for lack of originality, and working as a housewife for much of her adult life thereafter, she finally won the attention of the researchers with her 1995 article on group socialization development. Eventually, the article was expanded and turned into The Nurture Assumption, which attempts to show that parents matter much less (or differently) for child development than one would ordinarily expect. For example, parents tend to influence the politics of their children much more than they influence their personality (as long as they’re not downright abusive), which in turn is decisively shaped by the child’s peer group. The reason why we came to believe that parents play such a huge role is not only our everyday experience, but also because studies often fail to control for genetic influences.
Unfortunately, while I generally liked the book, the tone is a bit condescending, and some of her stylistic devices (like using the ending of the preceding sentence to start the next one) are irksome and annoying after a while. Advice on corporal punishment for children (“at the appropriate time, and on the appropriate part of the body”) is outdated and won’t help her to make her case either.
The Power Of Habit by Charles Duhigg
Did you know that habits are incredibly powerful? Like, it’s much easier to do sport if you pick a regular time and stick to it? Unbelievable as it sounds, Duhigg makes a powerful case for something that no one ever doubted in the first place. Habits define us in many important ways, and it seems like the only way to change them is to replace them by new (and more adequate) habits, following a circle of cue — routine — reward. While not exactly revolutionary, I think the book could indeed help you to overcome some routines you would like to get rid of, but you’re largely left to your own devices to figure out how. In addition to being way too long, many of the stories he tells are only marginally relevant for the theme (plus, you’ve probably heard about most of them before), and contribute little to a deeper understanding of the subject — think Malcolm Gladwell.
The Signal And The Noise by Nate Silver
Personally, I wasn’t confronted with a lot of new material in this book. Silver tells a few interesting stories from different areas where prediction matters (earthquakes, economics, weather,…), but it feels very much like a mash-up of Tetlock’s Superforecasting. The best chapters were on chess and poker, but for someone who has fascinated pollsters and forecasters (e.g. for correctly calling all 50 states in the 2012 US election), this is a little disappointing. I also struggle to understand why these days, every single book that even remotely deals with predictions has to include a chapter that introduces and defends Bayesianism, as if it were some magic wand that we choose to ignore at our peril.
The Singularity Is Near by Ray Kurzweil
As Douglas Hofstadter has it, it’s almost impossible to tell the factual from the fictional in this book. Realistic predictions, based on extrapolation of decade-long trends, are mired with fantastical sci-fi scenarios (immortality, time travel, crushing the speed of light) for which there is little to no evidence in anything we are currently doing in AI. Kurzweil is a visionary, and an intelligent one at that — he’s done a lot of research into different areas of science that he reports in the book — but don’t make the mistake to assume his predictions are the result of a sober analysis. While not a realistic account of current affairs, its far-fetched visions might be productively used to inspire some more down-to-earth futurist technologies, or as a starting point to engage in reflections about the profound changes the future is going to bring.
The War At The End Of The World by Mario Vargas Llosa
To my knowledge, Vargas Llosa happens to be the only Nobel laureate I ever met in person, so there’s always something special about his writings for me. This one is a recount of a late 19th century civil war in the state of Bahia, just after the creation of the first Brazilian republic. I had never heard of Canudos (the settlement where the battle raged) before, and it’s a fascinating story in and by itself. By contrast, I found (some of) the characters a bit stale, and lacking subtlety.
Towards an Imperfect Union: A Conservative Case for the EU by Dalibor Rohac
Rohac is well-known in libertarian circles, and before picking up his book, I was surprised why he was going to argue for anything from a conservative perspective. Having finished it, I’m not much wiser — was this just a marketing gag? Putting aside this matter, the points he raises are important ones, indeed: Sure there is a good deal wrong with the EU, both on the systemic and on the day-to-day level, but what better alternatives can we reasonably hope for? Libertarians (and conservatives, to a lesser extend) who imagine the break-up of the EU as the dawn of a golden age of free trade, economic development and peace do not nearly think enough about how they will achieve this transition, or how likely it is that this vision will win against its nationalist or socialist competitors. As of today, it has led to many questionable alliances between far-right figures and free marketeers.
What’s missing from the book are concrete proposals as to how one could reform the “imperfect union” in a pragmatic, feasible way. Here, Rohac doesn’t go much further than the people he criticizes: He might suggests the EU stops doing this and does that instead, without explaining how to ensure these measures are indeed likely to be implemented. Interestingly enough, he also puts a lot of the blame on the ECB, which (following Milton Friedman) he criticizes for being too close to the Bundesbank spirit and resisting quantitative easing, whereas other libertarians often point to the ECB’s loose monetary policy as a major culprit for sub-optimal economic performance.
Zen and the Art of Motorcycle Maintenance: An Inquiry Into Values by Robert M. Pirsig
I had seen this book mentioned as their absolute favorite by some people whose opinion I highly respect, so I went along and gave it a try. In the beginning, it was almost like a road trip documentation, and I seriously wondered how it was going to be different from the thousands of bike stories out there. But it gets rather philosophical soon, as the main character recounts his own story through the eyes of his own past self, referred to as Phaedrus.
As with most books that attack the Big Questions, it has the potential to inspire readers without offering any specifics. The “inquiry into values” revolves around a discussion of Quality (always capitalized), which turns out to be not so novel a concept after all. My superficial takeaway: The author urges us to see an intellectual side to even the most mechanical of tasks, if we are to prevent alienation from the world we inhabit. But this doesn’t do it justice, so I fear you’ll have to take a look yourself.
Up next:
Descartes’ Error by Antonio Damasio
Ethical Intuitionism by Michael Huemer
The Big Picture by Sean Carroll
Chance and Necessity by Jacques Monod
As always, if you’d like to recommend something yourself, don’t hesitate to do so in the comments!
