Governed by alchemists

Why it matters what clever people learn

This is part one of a multi-part series on the history of transferable skills. You can read the introduction here and listen to me discussing it on the EconTalk podcast here.

Hong Xiuquan was a clever boy. He began studying classical texts when he was just seven years old, and within a few years he’d memorised the Four Books and the Five Classics, the Confucian texts at the heart of the Chinese curriculum. His teachers were so impressed by him that they worked without pay, and his family made sacrifices to allow him to study. His aim was to pass the civil service examinations and to be rewarded with a prestigious job for life.

But the exams were fiendishly hard. They required prodigious feats of memory, and also mastery of the stylised “eight-legged essay”. Even if you excelled at these tasks, there was no guarantee you would get a job. In the 200 years before Hong’s birth, China’s population had doubled, but the number of civil service jobs had stayed the same.

Four times, Hong travelled to Canton to take the civil service exams. Four times he failed. After his third failure, he had a physical and mental breakdown, seeing visions and hearing voices. After his fourth failure, he read a Christian tract which seemed to explain all of his visions. He realised he was the younger brother of Christ, and that he had been called to convert his fellow Chinese.

Over the next ten years, Hong travelled the Chinese countryside, proselytising to peasants and baptising converts. By 1853, his followers had grown to become a great army, and in March they achieved one of their greatest victories. They marched on the city of Nanjing, stormed its fortifications and massacred thousands of Manchu men, women and children inside. It was a major moment in the Taiping Rebellion, perhaps the bloodiest civil war in history.

Of course, the exam system did not cause the rebellion. But it did contribute to two of the conditions that made the rebellion possible. First, it practically guaranteed a large population of frustrated young men who would never get the job they wanted. Second, those who did pass the exam were experts in classical learning, but had little knowledge of the huge developments in learning in the rest of the world which were placing enormous pressures on the ancient Chinese state.

As the rebellion was sweeping across China, the British Parliament was debating civil service reform. Remarkably, despite all that was happening, China was sometimes cited as a model. In a debate in June 1853, a Cabinet minister, Earl Granville, pointed to China’s civil service exams as a model to emulate, arguing that they had allowed the Manchu to successfully rule China for so many years. This was barely three months after the slaughter of the Manchu at Nanjing.

In August 1853, even as news of the rebellion spread throughout Europe and was being eagerly debated in the British press, Granville doubled down, asserting that he had spoken to experts in China who had assured him that the exams had brought the Manchu ruling class many adherents and that it had developed a love of the literary character amongst the Chinese people.

In March 1854, an opposition peer challenged this narrative, noting that China’s current travails meant that perhaps it was not the best model. But he was a lone voice. And whilst it would be overstating things to say that the China model was a major influence on British reformers, it certainly served as a positive example.

So it was that after all these debates, in the very year that showed up the weaknesses of the Chinese system, Britain ended up endorsing two of its defining features: open competitive examination and a heavy emphasis on classical literary texts.

The 1854 Northcote-Trevelyan Report

Britain had good reasons for pursuing civil service reform. The service suffered from nepotism and corruption: some posts were awarded on the basis of family connections, whilst others were openly bought and sold.

The Northcote-Trevelyan Report was designed to address these issues. It was commissioned by William Gladstone, the Chancellor, and written by Stafford Northcote, a Conservative politician, and Charles Trevelyan, a Treasury official. It was published in 1854, and most of its recommendations were accepted in subsequent years, to the extent that it is now generally regarded as the founding document of the modern British civil service.

It proposed that civil servants should be recruited through one “competing literary examination” which would be the same for every department. In practice, the exams ended up closely following the curriculums of the elite universities, and assessing Latin, Greek, maths, English composition, geography and history.

The aim was to recruit “young men of general ability, which is a matter of more moment than their being possessed of any special acquirements” and it suggested that men who had passed these exams “would probably make themselves useful wherever they might be placed.” It did not only focus on recruitment: the report also made proposals for the career development of civil servants and the structure of the organisation they worked in. It drew a sharp distinction between intellectual and “mechanical” work, and proposed that the elite intellectual civil servants recruited by this system should not stay tied to one department but instead move around between them.

Formal discipline and transferable skills

Many of the principles of the report were implicitly informed by an education theory known as formal discipline, which held that the mind is like a muscle that can be trained, and then turned to many different purposes. In order to train the mind, you need to study hard stuff like Greek. Translating Thucydides, in particular, was the equivalent of bench-pressing a particularly daunting barbell. If you spend your teenage years producing elegant renderings of the Melian dialogue, then any problem you face afterwards will be trivial. Your brain will be a razor-sharp tool capable of solving anything. The formal discipline theory was very popular in Britain in the mid-19^th century, and frequently used as a justification for the study of classical languages.

Stafford Northcote himself endorsed this argument, saying that his career success was attributable to the study of ancient Greek, and the reasoning skills it developed.

I attribute my own success, such as it has been, entirely to the power of close reasoning which a course of Thucydides, Aristotle…&c., engenders or developes [sic], and to the facility of composition which arises from classical studies. There is nothing that can compensate for the want of being able to follow out a train of reasoning, rejecting immaterial and irrelevant issues, and keeping close to the matter in hand.

Northcote’s argument is one we hear often today, except that Thucydides is replaced with “transferable skills” and “critical thinking”. In fact, if you read some of the comments on my previous post and in this tweet thread, you will see lots of people making arguments that are a kind of modern equivalent of Northcote’s: “But you don’t understand! I am really good at problem-solving – I can work through a problem, break it down into its basic parts, cut through the waffle and identify what is really important and relevant. It doesn’t matter what the topic is – that’s a skill I have that I am good at regardless of the domain.”

Obviously there is something seductive about this idea as it crops up in so many different eras, and people who have never heard of Northcote are saying something similar to him. And I kind of understand what both Northcote and the modern commenters are saying. This is probably the right moment to reveal (admit?!) that I have a GCSE in Ancient Greek and an A-level in Latin. I remember struggling through Greek unseen translation papers and wrestling with verb endings that contained so much meaning.

Incidentally, I started studying Greek at the same time as I started doing cryptic crosswords (I got interested in them after reading Enigma by Robert Harris). It did feel to me that they were kind of similar: you had to focus really intensely on just a few words, separating everything out to work out what it meant but then also putting it all back together to see how all the elements interacted with each other so you could derive some meaning from them. It also felt like all this practice was making me a better and more logical and more ordered thinker. I think if you’d shown me that Northcote paragraph when I was fifteen, I would have completely agreed with it.

But not all your fifteen-year-old intuitions are correct. Just because something feels like it might be true, doesn’t mean it is true. And because this idea has been around for such a long time, there is also quite a long history of experimental psychologists attempting to prove it – and we will consider all of that evidence in a later post.

Thomas Macaulay and “Cherokee and alchemy”

Other contemporary defenders of Greek and Latin took a different line. Thomas Macaulay, the famous historian and politician, was equally keen on Greek and Latin, but for reasons that might seem more baffling to modern eyes. Unlike Northcote, he didn’t think the substance of Greek and Latin was important. What was important about them was that everyone else thought they were important. Because everyone was trying to do well at them, the men who did best at them would therefore be the best men. In an 1833 speech in Parliament on the subject of Indian civil service reform, he defended entrance tests in Greek and Latin on these counter-intuitive grounds.

Whatever be the languages, whatever be the sciences, which it is, in any age or country, the fashion to teach, the persons who become the greatest proficients in those languages and those sciences will generally be the flower of the youth, the most acute, the most industrious, the most ambitious of honourable distinctions. If the Ptolemaic system were taught at Cambridge instead of the Newtonian, the senior wrangler would nevertheless be in general a superior man to the wooden spoon.

If, instead of learning Greek, we learned the Cherokee, the man who understood the Cherokee best, who made the most correct and melodious Cherokee verses, who comprehended most accurately the effect of the Cherokee particles, would generally be a superior man to him who was destitute of these accomplishments. If astrology were taught at our Universities, the young man who cast nativities best would generally turn out a superior man. If alchymy were taught, the young man who showed most activity in the pursuit of the philosopher’s stone would generally turn out a superior man.

Macaulay’s speech was lauded by his peers. The subsequent speaker in Parliament said it was “a speech which, he would venture to assert, had never been exceeded within those walls, for the development of statesmanlike policy and practical good sense.” Macaulay was also influential in the civil service reforms of the next few decades.

Given Macaulay’s influence, it is worth exploring the implications of what we might call his “philosopher’s stone” thought experiment. He is quite literally arguing that it’s acceptable to get teens and young adults to spend years of their lives studying things which are completely false or practically useless, as long as those things are prestigious and fashionable. All that matters is that you become proficient in the thing everyone wants to be proficient in. Whether that thing is true or useful is irrelevant.

There is of course a grain of truth to his argument. If you get all your industrious and ambitious youths – in modern terms, high agency types – to compete in the study of something, then it is very likely that the winners of that competition will end up in charge of the society’s powerful and elite institutions. It kind of doesn’t matter what the competition is: if a society sets it up as the most important thing, the winners of that competition will take all.

But if the competition is in knowledge that is wrong or useless, will those competition winners do a good job of running the institutions and the society? Surely not! Cherokee and alchemy would clearly not have been good preparation for running the major nineteenth-century institutions, nor would they be good preparation today. Macaulay’s flight of fancy was meant to show why it didn’t matter what you studied, just that you studied hard and competed with others, but I think in its extremism it does exactly the reverse: it reminds you that yes, it does matter what the kids are thinking about! As I’ve written before, in a different context: high-stakes exams don’t just identify elites; they determine what the elite knows.

Societies where youths are encouraged to study the arts of war are different to ones where youths are encouraged to study the arts of rhetoric. Societies where the elites have studied engineering are different to ones where the elites have studied law. Societies where religious leaders do the teaching are different from ones where scientists do the teaching. Societies where teens aspire to run factories are different from ones where teens aspire to run banks. Stuff matters!

You could argue that if the exam winners are smart and diligent and possessed of general thinking skills, it doesn’t matter if what they have studied is true or false or useful or useless. They will be able to use their general smartness and diligence to rapidly master whatever new body of knowledge is needed for the job, even if it is very different to what they studied when they were younger. I think this underestimates the difficulty and the time needed to master significant new bodies of knowledge, and I will explore that more in future posts.

But it is also unlikely to happen because once elites are in power, I don’t think they will be psychologically motivated to investigate whether the knowledge they were taught is true and useful. Elites clearly have a vested interest in legitimising the system that gave them power. People selected for mastery of a particular body of knowledge have every incentive to defend the importance of that knowledge, and to continue selecting their successors in their own image.¹

One might argue that if the knowledge being taught really is false or useless, a counter-elite will develop with more useful and true knowledge, and use those skills to take over the institutions. But if the established elite is full of smart and diligent people, and they have control of the levers of power, then it is going to be quite hard for a counter-elite to take over. What is more likely is stagnation, decline, and eventually being overtaken by other societies whose youths learn stuff that is true and useful. Of course, that might take a long time. It might take generations. But surely it will happen in the end.

Given some of the dubious reasoning behind its development, it is worthwhile considering whether Northcote-Trevelyan really deserves its place in British history as a landmark of successful reform – or whether perhaps it created more problems than it solved.

1. The other interesting psychological aspect of Macaulay’s argument is that he explicitly defines educational value in terms of what is fashionable. This made me think of René Girard’s theory of mimetic desire: we often desire things because other people desire them, creating self-reinforcing cycles of prestige and competition. These can be channelled towards socially productive ends, but they can also lead to intense and violent conflicts.