Odyssey

I’m currently (as of spring 2025) participating in the Oxford cohort of the Odyssey fellowship reading group, which is the spritual successor to last year’s Polaris fellowship. The goal of the reading group, as I understand it, is to speedrun an intellectual history of the western world, with a special emphasis on the ideologies that have contributed most to economic progress. Thus we start with Christianity, then move on to the Enlightenment and the scientific revolution, and then the American counterculture and the tech economy it birthed.

This document collects my notes on the Odyssey readings so far. I’ll aim to update it weekly as I progress through the curriculum.

Preamble (orienting to the present) #

Robin Hanson, “This is the dream time” #

“We live in the brief but important ‘dreamtime’ when delusions drove history.” The claim is that before agriculture and industrialization, we didn’t suffer from so many delusions as we do now. Our behavior and beliefs were well-adapted to our environment because that environment was stable for long enough that cultural evolution could converge. But for the last few centuries, progress has been too fast for cultural evolution to keep up. New delusions emerge faster than memetic selection can stifle them.

If all goes well and we survive this era, progress and change will have to slow down at some point. Technology will plateau, giving future civilizations the time that they need to orient correctly to what’s going on around them and reach true beliefs. Only for a few brief centuries will humanity have been too dazed and bewildered to understand what was going on.

I have three objections to Hanson’s view of history. (1) In the long run, I expect artificial minds to take over much of the business of understanding reality from biological minds, so we the humans might never need to recover from our delusions. (2) It’s unclear to me how delusional paleolithic people really were compared to us. They believed in all kinds of magic and mythology, even if their beliefs about tigers, cassava, and other features of their environment were more accurate than our beliefs about core features of the modern environment. (3) Will intellectual and technological progress ever slow down to the point where cultural evolution can keep pace? This is also unclear to me.

If this really is the Age of Delusions, one immediate consequence is that EMH goes out the window. Almost no one is lucid enough to understand what’s going on around us, so it’s little wonder that great opportunities are going unseized. Recognizing this should substantially reduce your bias against trying new things.

Peter Thiel, Zero to One #

I read several chapters from this book, but the only one that really grabbed my attention was chapter 6, “You are not a lottery ticket.” Thiel thinks attitudes toward the future can be ordered along two axes: definite-indefinite and optimistic-pessimistic. The former is about whether you have a specific view on how the future is going to look. The latter is about whether you think the future is going to be better or worse than the present.

Optionality is intrinsically good only on an indefinite view. If you know what the future is going to look like, you should just do the thing that will pay off in the future. It’s only if you have no clue what’s coming that you want to do something neutral or diversify or keep your options open.

Thiel claims much of our civilization’s decline is due to indefinite attitudes. Twenty-first century Americans expect the future to be better than the present, but they have no opinion on how it’s going to get better. Hence the rise of finance and law, the two easiest ways to make money if you have no idea how to create new wealth. Hence also the US elite education system’s preference for well-roundedness over specialization. How did we end up like this? What are the historical drivers of indefiniteness?

(1) Liberalism. Thiel picks on Rawls and Nozick in particular for putting process above purpose. But really, I think you can make this critique of philosophical liberalism writ large. By design, liberalism is neutral with respect to the good. It is actively allergic to prescribing goals for society or naming specific ways that we could make things better for ourselves.

(2) The postwar economic expansion happened without the Baby Boomers having to lift a finger, which made an entire generation complacent. They then passed this complacency on to their children by encouraging them to be generalists, not teaching them to think for themselves about where our civilization is headed, etc.

(3) Maybe also something to do with the concept of randomness. When I have time, I think it would be interesting to reread Ian Hacking’s intellectual histories of chance(1) (1)E of P and T of C in light of Thiel’s thesis. Did we become indefinite because we started reifying the concept of probability?

Christianity, Part 1 #

Notes on these texts coming soon…

Emmanuel Carrère, The Kingdom #

David Bentley Hart, The New Testament #

Christianity, Part 2 #

Framing #

Part 1 was about the content of Christianity; part 2 is about the dynamics of its spread. Now that we know what Christianity says, we want to know how most of the western world became convinced of it. The great moral achievement of Christianity is that it tells the powerful they should give consideration to the interests of the weak. But this ideology is clearly inexpedient for the elite, and we tend to think that the elite (at least partly) control which ideologies can spread and become dominant. So what happened?

Rodney Stark, The Rise of Christianity #

Chapter 1 tries to estimate the Jesus movement’s growth rate from 40-300 CE and speculates about the dynamics of conversion. If there were 1k Christians a decade after the crucifixion, and 5M Christians around 300 CE, assuming a uniform growth rate gives you 40% growth per decade.(2) (2)Uniform proportional growth is a sensible first-pass approximation, but presumably the growth rate picked up near the end of this period, when Christianity was in the ascendant, and pagans wanted to jump on the bandwagon. Stark doesn’t try to model this. That seems quite fast to me, but it’s about the same as Mormonism’s growth rate in the twentieth century. Stark therefore argues that you don’t have to believe in mass conversions to explain the rise of Christianity. Some combination of natural increase (more on this in chapter 5) and person-to-person proselytizing would have been enough. The upshot is that one can give a purely natural (as opposed to supernatural) account of Christianity’s rise, and the rest of Stark’s book attempts to do just that.

Chapter 2 tries to reconstruct the class composition of the early church. “If the early church was like all the other cult movements for which good data exists, it was not a proletariat movement, but was based on the more privileged classes.” This would be surprising ex ante if all one knew about the early church was the contents of the Gospels. Why would a religion that viewed wealth as inherently evil appeal to the comfortable classes? But Stark points out that cults are generally most appealing to the wealthy. (1) If you’re truly living hand to mouth, you don’t have time to adopt a demanding new set of religious practices. (2) Unless you’re at least somewhat educated, you probably don’t understand the doctrines of your current religion, so you won’t see any doctrinal reason to convert. (3) Adopting a deviant new religion is less risky if you’re rich and well connected. Compare the forebearance with which the US government treated Mormonism to the brutal suppression of the Taiping Heavenly Kingdom in the same era. Stark notes that the Romans’ persecutions against Christians were weirdly half-hearted, which makes sense if you assume early Christians disproportionately came from high-status backgrounds.

Chapter 4 argues that the 165 Antonine Plague and the 215 Plague of Cyprian—which was probably a smallpox epidemic—propelled Christianity’s rise. Wherever plague broke out, pagans appear to have focused on saving their own lives, not on caring for the sick. Even the great Roman physician Galen high-tailed it out of Rome during the Antonine Plague. On the other hand, it’s well attested that Christians tended to each other and to sick pagans during the plagues. Many of these valiant Christian nurses died in the process, but they were undeterred because unlike the pagans, they believed that God would reward them with eternal life for their acts of charity toward the sick. Even though there were no effective pharmaceutical defenses against the plague, Stark still estimates that nursing could have reduced the Christians’ mortality rate by more than two-thirds relative to the pagans’ rate, mostly because Christian invalids weren’t left to die of thirst and starvation. The Christians’ lower mortality rate would have looked like a sign of divine favor, and it meant that Christian communities stayed intact while pagan social networks were shattered.

Chapter 5 finally addresses demographic explanations for the rise of Christianity. The Romans had crazy anti-fertility norms. Infanticide against girls was so widespread that there were “140 males per 100 females in Italy, Asia Minor, and North Africa.” Abortion was widely practiced, and because it usually involved administering poisons or else surgery in a pre-antiseptic age, it was a major cause of death among Roman women. Contraception (including withdrawal and anal sex) appears to have been common. It was a commonplace among Roman authors that nearly all marriages were unhappy. Roman men, after spending their first three decades in the military, were uncomfortable relating to their wives, and Roman women, who were often married right around puberty to men more than twice their age, would have been equally distant from their husbands. The result of these Roman fertility norms was population decline begining in the late Republic and continuing for at least three centuries.

Early Christians rejected pretty much all of these norms. They treated infanticide and abortion as murder, denounced contraception as onanism, and celebrated marital harmony instead of viewing marriage as an unpleasant obligation. Their only anti-fertility norm, as far as I can tell, was to discourage widows from remarrying. All of this made Christian fertility much higher than Roman fertility—probably well above replacement rate. So even if rates of conversion had been negligible, the Christians eventually would have outgrown the pagans through shear force of demographics.

Stark’s answer to the framing question is that some elites voluntarily accepted Christianity near the start, and once they did, Christianity’s pro-social, pro-fertility norms caused them to proliferate faster than the pagans.

Edward Watts, The Final Pagan Generation #

In the late 300s, a large number of elite Roman men chose to become Christian priests or ascetics. They came from wealthy provincial families that had raised and educated them to pursue lucrative careers in the imperial administration, and their parents were generally either pagans or lukewarm, barely converted Christians. Why did they turn away from the career path expected for men of their class? (1) The church’s rapid rise over the fourth century meant that it needed more capable administrators. Churches now had large holdings of land that the previous generation of middle-class bishops didn’t know how to manage. Ecclesiastical courts had become much powerful, so the bishops who oversaw them now needed to have legal training. (2) Athanasius’s Life of Anthony was wildly popular and convinced lots of pious aristocrats to become ascetics. (3) Many of these elite young men who went into the church were genuinely very zealous Christians. According to John Chrysostom, he and his peers rejected the path expected of them because they felt “the things of this world are in their nature no better than dreams.” They cared only about the waking life that awaited in the kingdom of heaven.

Where did this generation of bishops and monks find the courage to buck society’s expectations? Watts suggests that they created a tight community among themselves, within which Christian devotion was valued above the worldy forms of status and achievement that their class had traditionally strived for.

They supported one another when unsympathetic friends and family members pushed them to return to [the conventional path]. And some of these men began to write extensively so that they could better articulate the alternative value structure that guided them. As these ideas spread among young elites, they diluted the power of the social incentives that the empire used to guide elite conduct (pg 164).

One key point I see Watts adding to Stark is that Christianity took over the hearts of the Roman elite one funeral at a time. The first generation to convert around the time of Constantine did so rather tepidly. They kept caring about civil service jobs and estate management just like the generations before them had done. But their children took the new faith much more seriously and allowed it to radically reorient their values.

Enlightenment #

Francis Bacon, The New Atlantis #

I notice that Bacon’s ideal enlightened society of Bensalem is very pious. Bacon makes a big deal of how the Bible was miraculously revealed to the Bensalemites, and now they are all extremely devout. Almost every time one of them gives a speech, it begins with some prayer or invocation of God. In fact, Bensalem is so Christian that even the Jews there “acknowledge that Christ was born of a virgin and that he was more than a man.” I think this points to a key difference between the English and French enlightenments. You could never imagine Voltaire writing anything like this. In his ideal society, the priests would all have been hanged forthwith and the Bibles burned. My best guess at why the French enlightenment had more of an anticlerical, antireligious edge is that the counter-reformation had made the Catholic church much more reactionary and hostile to science than the Anglican church by the time of the Enlightenment.

Bacon’s ideal society is very well governed. State capacity is high. Immigration laws are enforced. Public officials refuse bribes. The state encourages high fertility by paying for a banquet in honor of any man who has thirty simultaneouly living descendants. Obviously a well-functioning state would have appealed to Bacon, a former civil servant, but it appealed to plenty of other enlightenment thinkers as well. Remember Montesquieu’s dream that separation of powers would allow the various parts of a government to check and correct each other, to keep each other honest. This gels with the Newtonian idea, which became a core Enlightenment idea, of nature as lawful and non-arbitrary.(3) (3)See also Hume’s rejection of miracles. Why shouldn’t the exercise of state power be as predictable as the motion of the planets?

By far the best part of New Atlantis is the speech that the Father of Salomon’s House gives at the end. Bacon sums it all up well when he has the Father say:

The end of our Foundation is the knowledge of causes, and secret motions of things; and the enlarging of the bounds of human empire to the effecting of all things possible.

This is the mission of science in a nutshell. To understand why things happen in the world and to use that causal understanding in order to achieve all things possible. So what is possible in our world? Bacon offers twelve pages worth of predictions, many of which are spot on. He correctly guessed that it is possible to desalinate water at scale, to artificially extend human life, to extract heat from deep underground, to transmit voices over vast distances, and to achieve human flight.(4) (4)He even predicted the invention of Soylent: “some of the drinks are such as they are in effect meat and drink both.” Bacon didn’t quite bat a one-hundred. He thought it should be possible to build perpetual motion machines, and he seems to have missed that the key application of cheap on-demand heat would be engines that convert it into work.

But a key change of perspective is starting. To a pre-enlightenment thinker, the question What is physically possible in the world? called for a religious answer. A sea can part, a rock can gush water, and a virgin can conceive and bear a son, but we have no causal explanation for any of these phenomena, nor can we induce them to happen at our will. Bacon is suggesting that we can do better. The limits of possibility might be natural rather than supernatural, and if we observe nature carefully enough, we might learn to make things happen in our world that were once thought impossible.

Derek de Solla Price, Science Since Babylon #

This book’s grand narrative is that western science represents the synthesis of Greek and Babylonian science. In dSP’s stylized telling, Greek science is about geometric reasoning, the visual, and the abstract, while Babylonian science is about symbolic calculation, the verbal, and the concrete. The difference, as he sees it, is neatly summed up in “the attitudes of each civilization toward the square root of two. The Greeks proved it was irrational; the Babylonians computed it to high accuracy.” I recall Dunham making basically the same point about pre-Hellenistic Egyptian mathematics in Journey Through Genius. Egyptian scribes would fill reams of papyrus calculating the volumes of frustums of various dimensions, but they never made the characteristically Greek move of proving a general formula for the volume.

I wonder whether Babylonian science is making a comeback in the rise of scientific computing. If the theory of natural selection is Greek, genomics is Babylonian. If blackboard physics is Greek, computational physics is Babylonian. Maybe one could also cast symbolic AI and deep learning as the Greek and Babylonian approaches to AI, respectively.

Lawrence Principe, The Secrets of Alchemy #

This book reminds me of Arthur Koestler’s The Sleepwalkers. Principe and Koestler basically accept the modern scientific consensus against chrysopoeia and geocentrism, but they still take issue with whig history of science. In both cases, truth defeated error mostly by playing dirty and getting lucky.

In Principe’s story, two things happen in the mid-eighteenth century that tank alchemy’s reputation.(5) (5)The core of this story is told in chapter 4.

(1) A conceptual distinction is drawn between alchemy and chemistry, and this distinction gains widespread currency. Notice that “alchemy” is the same word as “chemistry,” just with the Arabic definite article slapped on the front. Understandably then, the two words had been used interchangeably all the way through the early modern period. But somewhere around 1700-20, alchemy was redefined as the subject concerned with transmutation, necromancy, black magic, and so forth, whereas chemistry was defined as that useful, respectable subject that Robert Boyle and Étienne Geoffroy studied.

(2) Enlightenment thinkers subject alchemy, thus defined, to all kinds of irrational abuse. They made guilt by association arguments, pointing out that alchemists studying chrysopoeia were (by definition) the colleagues of necromancers and fortune-tellers. A newly professionalized class of academic chemists also made ad hominem attacks on alchemists, whom they portrayed as liars, charlatans, and lowlifes. Principe says this was enough to strip alchemy of almost all its former legitimacy by the 1740s.

We know now that Enlightenment-era chemists were right; chrysopoeia really is chemically impossible. But they were certainly not epistemically entitled to reject alchemy. There was no decisive experimental evidence against transmutation, at least until Dalton’s measurement of atomic weights in the early nineteenth century. And until the Rutherford experiment in 1911, nobody had even a roughly correct explanation of why the elements are fundamentally different from one another. Tellingly, when belief in chrysopoeia had a mini-revival in the mid-nineteenth century, mainstream chemists didn’t refute it with evidence, but rather dismissed it as a religious craze. The chemists could do no better because they had collectively gotten Gettiered.

One point I take away is that it’s weird how often science uses unreliable modes of reasoning, but nevertheless ends up with correct conclusions. Kepler made three significant mathematical blunders in his derivation of the angular momentum conservation law, each of which would have ruined his result on its own. But as if by a miracle, Kepler’s mistakes cancelled out, and his second law turned out to be correct.(6) (6)See chapter VI in The Sleepwalkers Similarly, Gwern Branwen has investigated known errors in math papers, and he found that when a published proof contains an error—even a very substantial one—the theorem it was meant to prove still turns out to be true the majority of the time.(7) (7)The most famous example is probably Andrew Wiles’s first proof of Fermat’s Last Theorem. Enlightenment-era chemists rejected chrysopoeia mostly because it had bad vibes, and their view was vindicated by evidence that emerged two centuries later. There’s something to be explained in all of these cases. Maybe we could call it the unreasonable effectiveness of scientific practice.

Last updated 11 March 2025