Keju, China’s difficult civil service test
aeon.co
aeon.co
In ancient Korea it had a different effect and tended to fix certain classes into the "classes from which the bureaucracy could be reliably drawn from". They still had the academies, but it tended to be the same class/castes who attended and passed the exam. In theory it was open, but in practice it worked this way [1].
Some of the preserved academies still exist and are major UNESCO sights. [2]
Modern South Korea still uses an examination system for civil servants [3], but has lost the caste system so in an interesting way has ended up more like the ancient Chinese concept. It tests civics, history, data analysis, etc. and I believe there are different tests for different levels of seniority and are required for promotion. It's common for test takers to have spent several years in dedicated study after university to prepare for the exam [4].
A semi-equivalent might be like a bar exam in other countries.
1 - https://www.amazon.com/History-Korea-Antiquity-Present/dp/07...
2 - https://whc.unesco.org/en/list/1498/
3 - https://en.wikipedia.org/wiki/Republic_of_Korea_public_servi...
Huh? Why would you expect a flatter distribution? The concentration of surnames is easily explained by natural development over time.
So while Japan has a couple of hundred thousand surnames in circulation, Korea has in total about 350. However the great majority of the pop belongs to one of the top eight or so surnames: Kim, Lee/Li/Ri Park/Pak, Cho, Kang, and a couple of others.
Your link suggests that what was for sale was noble identity - a place in a specific and unique noble family, with formal documentation that you belonged to it - not a license to use the surname, which everyone was already free to use.
> It became increasingly common for successful merchants too to take on a last name. They could purchase an elite genealogy by physically buying a genealogical book (jokbo) — perhaps that of a bankrupt yangban [noble] — and using his surname.
> By the late 18th century, forgery of such records was rampant. Many families fiddled with theirs: when, for example, a bloodline came to an end, a non-relative could be written into a genealogical book in return for payment. The stranger, in turn, acquired a noble surname.
But again, what you're acquiring there isn't a "noble surname". It's noble legal status, and you end up with whatever surname the noble family uses by coincidence. No one participating in this system cared whether they ended up with the name 金 or with 網切, so this isn't a good explanation for why 金 is so common.
The link also dwells on the much greater variety visible in Japanese surnames. It doesn't point out why that is, but we do know the answer - they are more variable for the extremely obvious reason that the concept of a surname in Japan is much more recent, dating to the late nineteenth century.
Dr. Mark Peterson[1] who has his own great channel[2] and a video on this topic (in Korean)[3] explains it as mostly just happening as a clerical matter both during the Choson period by simply registering a name to align with prestigious families and to hide your caste status and then later during the Japanese occupation since it was demanded as part of the growing Japanese empire.
In practice, it's less confusing than it sounds. Koreans will use contextual associations such as other family members, titles, friend or business associations, and historic clan hometowns to discern things.
"Oh, you are a Kim also? Which one?"
"I'm from the Jeonju Kims."
"Oh haha! I'm from the Incheon Kims. I know another Jeonju Kim, are you that one?"
"No that's CEO Kim he's my distant uncle. I'm just the normal Kim. My wife is so and so."
"Ahh, okay."
(completely made up example that's probably wrong but meh)
And after that everybody is clear on who is who. This kind of thing is often included in some introductory stuff about what age everybody is so you can get your honorifics sorted out as well.
1 - https://youtu.be/4VVCbApZ4bg
This explains a lot, neat
The thing about socio-political systems is that their innovativeness doesn't last forever. Keju was innovative in the pre-industrial age because it provided a path to social mobility for those not born into hereditary nobility. Meanwhile European monarchs in the Early Modern era relied on nobility, landed gentry, and the clergy in administering their realms, none of which had academic competency as a requirement to entry.
Of course, by the 19th century this has fallen dreafully behind as Keju curriculums were never updated to fit an evolving world. Confucian Classics were no better than the Bible or the Quran at guiding the development of elecrical generators, excavating coal, synthesizing chemicals, or smelting steel. At some point in the future, our current model of measuring academic competency will be seen as rigid and obsolete too.
Standardized examinations aren't unique to China. The cut-throatness is a byproduct of China's large population base combined with limited possibilities of upward mobility due to its status as a developing economy not aligned with the most advanced economic bloc on the planet.
>Keju was so precocious that it pre-empted and displaced an emergent society. Meritocracy empowered the Chinese state at a time when society was still at an embryonic stage. Massive resources and administrative manpower were poured into Keju such that it completely eclipsed all other channels of upward mobility that could have emerged.
What pre-empted and displaced an emergent society was the creation and re-creation of unified imperial dynasties where emeperors were despots who were occasionally enlightened at best. Wealthy merchants existed in every Chinese Dynasty and "Ten Great Merchant Guilds" dominated the economic landscape of the Ming and Qing Dynasties. The upper bound of society was not limited by Keju, but by China's position as an unchallenged hegemon in its sphere of influence until the industrial revolution.
This represents a misunderstanding of objective of the test. The subject matter of the test is sort of irrelevant (other than cementing a common cultural foundation for the educated classes). The test functions primarily as an IQ test of sorts. Talented individuals end up rising while lazy/stupid ones do not. It's not about building skills
In the modern world it's not really all that different. Children in China study crazy hard for the gaokao, but if you speak to anyone who's gone through the system they've virtually all forgotten the history and multivariable calculus they had to study. People primarily learn relevant skills "on the job"
The UK didn't have the industrial revolution because they started to teach children algebra
India did the same. But, India being India, implementation and enforcement were quite haphazard and chaotic compared to China.
Still, China is not quite as unique as the author thinks: India too has massive competition for civil service jobs, heavy emphasis on rote learning and passing incredibly difficult exams, and an increasingly authoritarian leader.
So, let’s go out, vote, and get others to vote.
I believe it used to be that way a few years ago, but since the pandemic the ratio inverted such that most US colleges are now test-optional.
Or, lets take CalTech with a similar size enrollment (in the thousands): As of October 2022, Caltech has 46 Nobel laureates to its name awarded to 30 alumni (26 graduates and 4 postdocs), including 5 Caltech professors who are also alumni.
> IITs are prestigious institutions. They’re not OKish institutions in the slightest.
They aren't. They are a filter for a few thousand out of 1.4 billion people. Yet, all the IITs put together produce nothing compared to the output of one US University. > 99% end up writing Java code for Wall Street or work for ad-tech. I doubt they are pushing the boundaries of science.
> Or, lets take CalTech with a similar size enrollment (in the thousands) [...]
Caltech student population is a little over 2,200, mostly graduate students.
UC Berkeley is over 45,000, about 3/4 undergraduates.
They don't have "similar enrollments".
For more sobering examples: Universiti Malaya (UM) and City University of Hong Kong ranks higher than Brown University. University of Copenhagen, University of Waterloo, and Rice University are all outside of top 100, so I guess we can call them OKish (all according to QS 2024).
Indian education system is built by the British, they needed some locals to do clerical work, who repeat what the masters said is knowledge (remember and and faithfully reproduce), follow orders. The white sahibs are replaced by brown sahibs, the education system is the same, what British called ICS is renamed to IAS (top govt official in a district), but its the same thing.
"The map below is a scheme of subway lines in the city. Each line is marked by a different color and is associated with a different letter. Stations are numbered by their location on the line(s). The numbering of the stations progresses from the beginning of the line to its end. Lines always begin at the northernmost station, except for lines G and M which begin at the westernmost station. Codes associated with some of the stations have been written below them. A station can have more than one code associated with it, as some stations have more than one line passing through them. Which of the following routes would be the least efficient (passes through the most stations)?"[1]
If only someone could devise a workable sociopath filter...
[1] https://www.jobtestprep.com/civil-service-exam-nyc-free-prac...
[1] I’m pretty shocked about how bad the women situation is, especially in the 15-19 age group here https://www.statista.com/statistics/282119/china-sex-ratio-b... :-/
“Since the introduction of the examination system … scholars have forsaken their studies, peasants their ploughs, artisans their crafts, and merchants their trades; all have turned their attention to but one thing – government office. This is because the official has all the combined advantages of the four without requiring their necessary toil …”
It refers to a very difficult exam that existed for a time in Chinese society that the author feels is analogous to systems re-emerging in Xi’s modern China. I hope someone will correct me if I’m wrong. To be honest, I normally dislike comments like this that just paraphrase the article, but I read most of it and he repeats himself a lot. He could really use an editor to help him be more succinct.
I understand not wanting to spend the time detailing your research showing that China's technological growth lagged after introduction of the Keju in a teaser article. Repeating "I found this data, it supports my conclusion, and China's doing the same thing today" isn't a good alternative.
Maybe the book has a better editor.
I would add to your premise summary "keju helped (allowed? permitted? enabled?) the Chinese state to dominate Chinese society".
I rely a lot on TTS screen readers, especially for longer articles. Listening and reading at the same time somehow makes my focus much sharper. Or I can just relax my eyes and listen.
1. transform a long and repetitive test into a shorter one, and/or
2. analyze an article or a webpage and give us a "score" that represents how well or not well it has been edited.
We shouldn' rely on someone like you to read the entire article and tell others that it's repetitive, and therefore boring to read in its entirety.
Imperial China's hegemony over its surrounding regions was not thoroughly challenged until the 19th century, and once it began to be challenged by nations with more advanced ideas and technologies, Keju did enter terminal decline along with the rest of the imperial system. Even without Keju, Early Modern China didn't have the recipes for scientific or industrial revolution to start there.
I love the output that you can get with LaTeX, but I just can't wrap my head around the syntax.
For some things you run a \command just inline with whatever you're doing (and if you want that command to only apply for a small part, you can surround it in an implicit environment with {\command whatever you were going to type}, for others you create an \environment{and write things inside it}. I guess in general it's weird to me that something like \Large can affect everything that comes after it, it feels vaguely like overuse of global variables in C.
Parameters/options to commands and environment directives... I guess sometimes use square brackets and sometimes use curly brackets? I could never figure this out.
It was useless in that the texts were vague and fictionalized, and people weren't asked to offer new interpretations of the texts or to apply them to modern circumstances. The tests were graded based on how elegantly students could reproduce the standard orthodox interpretations in elaborate essays.
They have some philosophical values that were progressive for the time they were written (before 300BC), but by the 1800's they were as useless for adapting to an industrialized modern world as Abrahamic scripture. The disconnect between Keju study materials and realistic administrative responsibilities of bureaucrats have existed for as long as Keju itself.
While Harvard’s admit rate might be above 4% in some years, there is more selection bias in the applicant pool, it would seem. The vast majority of people don’t bother to apply.
Ironically, I also suspect that despite that, the quality distribution of accepted candidates is roughly similar to most other universities too. Much of this personal suspicion stems from empirically utilizing online courses of theirs when I was in undergrad and observing students ask questions in recordings of course lectures. Of course they got into Harvard, but I was surprised how often many of them struck me as very normal in terms of ability to grasp material. I suppose the rest of it comes from some of the silly stuff I occasionally read in HBR. Heh.
Harvard and most top private schools are generous in financial support if you can’t afford it. Harvard is free for families making less the 85k. But you are right that might leave an empty middle where you don’t get much aid, and don’t have personal wealth.
https://college.harvard.edu/admissions/why-harvard/affordabi...
I do appreciate their efforts, but I bet that they decide who can and can't afford it, probably based on rigid criteria that inflexibly denies at least some folks in the middle who really can't afford it.
Edit: It is possible I am wrong, the Harvard website linked above at least gives the impression they work with folks in varied circumstances. My experience helping folks with aid programs has perhaps coloured my perspective on Harvards unfairly. I have no experience dealing with them. If you want to attend Harvard, by all means, please apply and give them a chance! It is worth the rhe shot.
Your suspicions are evidence of your suspicion? I'm not sure what you meant by "academic middle class": If you mean, in terms of academic skill, well of course - why would Harvard accept any but the best (and wealthiest and most powerful)?
> the quality distribution of accepted candidates is roughly similar to most other universities too
Recent experience with Ivy League undergraduates leads me to a believe that it's a pretty effing talented cohort. That statement may match the snobbery that many are (rightfully) suspicious of, but it's not that; they were just very impressive.
The article gives a figure of ~1M taking the provincial exam annually during the Ming dynasty, with the later Ming empire having a population in the range 160-200M. You are probably right about self-selection being greater for Harvard, then, given that just about anyone can, in principle, apply there.
[1] More recent literature demonstrates the political effects of Keju. In 1905, the Qing dynasty abolished Keju, dashing the aspirations of millions and sparking regional rebellions that eventually toppled China’s last imperial regime in 1911.
[2] A 2020 study shows that, for every doubling of successful Keju candidates per 10,000 of the population in the Ming-Qing period, there was a 6.9 per cent increase in years of schooling in 2010. [comparison shows a logarithm relation between Keju success and current schooling between regions.]
Wrt to [2]: there were only 260 regions, that period ranged over more than 500 years, and the correlation is not particularly good. It seems extremely improbable that all other factors have been excluded (it's just linear relations for a handful of numbers they could find). The logarithmic nature is counter-intuitive: since the exam was very, very selective, you would expect the reverse: a small increase in candidates passing requires many more students below that level, and thus a super-linear increase in schooling.
[1] sounds very far-fetched. The linked article is paywalled, but the jump from reducing fairly small hopes of fortune to revolution is too big to accept from statistics that sound similar to those in [2].
The author seems to use it as support for his argument, but in my view, it just turns it into a house of cards. I stopped reading after the second article. Do these things bother you?
So I think the author misappropriated the paper to support their own argument. It's possible that the abolition of Keju increased regional unrest which contributed to higher incidences of rebellion in 1911, but itself did not "spark" rebellions on its own. The 1911 revolution began with the Wuchang uprising which was started by Westernized "New Army" divisions of the Qing Dynasty, an effort that began in 1895 and was not tied to Keju. It ended quickly because Yuan Shikai, the real commander of the remaining loyal New army in the North, realized he could make a powerplay by forcing the imperial court to abdicate the convince the revolutionaries to support him as president instead.
Nevertheless [1] appears to claim that ending the civil service test caused the fall of the Qing dynasty which is complete nonsense. I assume this is just an inept translation.
[2] The claim (in English) is incomprehensible; I assume the author's intention is that when more civil service jobs were awarded more Chinese studied in an attempt to pass the exam.
Overall, I believe the English text of the article is poor. I believe he is arguing the civil service test was all pervasive because over a million people took the test annually in an effort to get a very small number of civil jobs. The test itself could be considered an intelligence test because it required intelligence to memorize and elegantly recite the orthodox answers based on Confucian classics.
If we want to argue what is deficient about Chinese government or culture during the Ming/Qing period, there are an infinite number of other topics. I would argue overall the Civil Service exam is a net positive for China, not a negative. As an aside, I watch some modern Chinese melodramas. These TV shows seem to argue China's past was marked by corruption, infighting among officials, and poison poison poison. According to these TV shows, ancient Chinese politics consists of everyone trying to poison everyone else. There are TV plots that involve multi-factor poisons. By itself the poisons are benign, but when they are mixed via food & incense & medical prescriptions, they are deadly. That is they blame endless in-fighting for China's predicament. The TV shows often feature characters spouting Confucian euphemisms but have no qualms poisoning or killing women/children for political gain. In short, modern Chinese people think past Chinese politicians were unethical and downright evil.
Are there any other kind? any place/any time
People in the (American, and I dare say elsewhere too) tech world do give a shit about the college you went to, and the elite colleges do give a shit about extracurriculars, especially if you’re Asian or something, unfortunate as it is.
I'm really curious what the "or something" in your race statement means though.
As for “or something”, it is well known that the bar is very high for Asians and extracurriculars are needed to distinguish oneself (too many perfect SATs and GPAs), but maybe there are other demographics similarly affected that I don’t hear about much.
Not really. Nobody gives a damn what college you went to for the vast majority of jobs. They care about whether you went to college or not, regardless of the school. And even then, it's becoming more and more rare to find employers who are sticklers that applicants must have a degree.
At most jobs? Yeah, they are actually.
Posers aside, the rest of the world does not function this way and advice like yours really fucked my professional development up.
If you're just a regular type of dude in most of the US, none of this applies. Your Petco job does matter, your employer won't even look at your Github page, and talking about PRs will only invite clarification of whether you mean public relations or Puerto Ricans.
The SV-optimized path is not the only path to success in America. This country is itself entirely optimized for entrepreneurs. Tradesmen crush it and don't have to live in fear of AI obsolescence. The most successful people I know are all undereducated immigrants and ex-military types running boring businesses selling medical equipment, carpet and tile. Always boring shit people generally need, not drop-shipped dollar-store crap from Temu.
To be the guy positioned to sell N95 masks in 2019 never required a Github profile or elite education.
US Civil Service merit & fitness requirements were undermined by courts finding that because most examinations correlated to general intelligence as defined by IQ, they were discriminatory to various stakeholders who are either under-assessed by IQ due to cultural or disability related bias.
The easy path for employers is to utilize the experience + education provided by the candidate, and clean up any mess later. (or not)
I've never had any potential employer look at my GitHub. It was all about the resume, experience, and being able to explain that experience, as well as a technical coding assessment.
If anything, "grinding leetcode" would be the closest analogy to Keju (although not exactly identical of course).