Imperial Statistics (NP: Wei through Chen dynasties)

A forum for guides, lessons and sharing of useful information.
Salmoneus
MVP
MVP
Posts: 2676
Joined: 19 Sep 2011 19:37

Re: Roman Emperor Statistics

Post by Salmoneus »

I think the general historical assessment is that the stricter and more impersonal your rules of succession are, the less likely you are to get good rulers (because you're making the system effectively more random with regard to personal characteristics)... but the more likely you are to have stability. And while in the short run it's good to have good rulers, in the long run it's much more important to have stability. Also, the stricter system is less likely to give you really awful rulers (spoiled sons are often not great, but megalomaniacal social climbers who'll stop at nothing in their search for power are worse). And since in most of history the ruler's job is primarily just to not fuck things up, mediocrity is generally preferred.

Rome's best years, the Five Good Emperors, came when they had what seemed like the best of both worlds: succession based on some sort of merit, but also strict succession, thanks to the old emperor picking the best candidate and making that pick clear before death (in fact in several cases officially making the heir co-emperor so that nothing could stop their accession). This is also what Diocletian tried to do with the tetrarchy. But this system also has several flaws: it goes bad very quickly as soon as one emperor picks the wrong successor (who then goes on to keep picking wrong successors); it collapses into chaos when an Emperor dies too soon to pick an heir (more common than with sons, because emperors naturally want to have sex, whereas they're inherently averse to nominating an heir in case the heir gets impatient; also, a fully hereditary system can extend inheritence laws to account for a lack of a direct son, albeit controversially); and it means continually trying to promote people who don't necessarily have a strong power base (sons tend to have a power base because they grow into having one by being heir to the throne since childhood), which means that eventually you'll have one who is just outright rejected by the powerbrokers...
User avatar
Torco
sinic
sinic
Posts: 295
Joined: 14 Oct 2010 08:36

Re: Roman Emperor Statistics

Post by Torco »

That's like the default position in historiography, yeah, but often such consensuses (consensa?) are wrong. I was literally taught in uni that modernity consisted on, amongst other things, the invention of politics out of a supposed premodern attitude that social structure was just what it was and that people basically followed it blindly up until the renaissance or something like that, which is ludicrous. But this critique of the general historical opinion only goes so far: civil wars are, I grant, not generally good for polities, but the concept of stability seems too continuous for what is probably a discrete effect: if there are no riots in the streets, widespread destruction of property, teenage red guards lynching their teachers, civil wars and presidents fleeing the palace in helicopters you probably have all of the good things stability is going to give you. "old emperor says who is next emperor" is pretty free, as these things go, without being so free that you start having civil wars over it. Then again, too much stability can be bad: we're likely gonna not cock things up and not make waves and just keep doing what we're doing ourselves into a +5° world.

If you're going to have relatively free succession, though, it's probably better if there's some legitimized system for making the call, if only so that the decision is only made once. Still, the fact remains that there are extremely effective empires with and without formal rules of hereditary succession, so it's safe to say it's not an essential piece of the puzzle. more like having two chambers of congress: it has pros and cons, but ultimately it doesn't affect viability.
which means that eventually you'll have one who is just outright rejected by the powerbrokers...
luckily, they can quit, marry the american and leave the last guy's daughter to it.
Salmoneus
MVP
MVP
Posts: 2676
Joined: 19 Sep 2011 19:37

Re: Roman Emperor Statistics

Post by Salmoneus »

So, I thought I'd add a small update here for comparison: Emperors of the Qin, Early Han, Xin and Later Han Dynasties.

First, a short rant:
Spoiler:
Wikipedia articles for the Roman and Chinese empires really are night and day! Roman articles are generally closely cited, frequently offer opposing viewpoints, attempt to describe the emperors as complex people, but also freely acknowledge when not enough is known to have any firm opinion on someone. Chinese articles read like fanfiction. Almost everyone is either a valorous hero (though sometimes lead astray by evil eunuchs) or a supervillain; supervillains are constantly motivated by jealousy, fear and shame, and you can tell they're villains because they challenge the established ruling class in some way. When a good person does a bad thing, it has to be explained as the result of supervillainous influences (usually eunuchs or women, or occasionally warlocks), or occasionally a failure to control base desires (in turn usually due to the influence of eunuchs, women or warlocks). You can tell which things are bad, because they're actions contrary to the interests of people whose descendents later become powerful. (so X, a supervillain, will accuse Y of a crime, even though Y is a saint; fortunately, Y's daughter will later marry an emperor, at which point it will correctly be realised that Y was innocent all along. Oddly, nobody whose offspring don't marry an emperor is ever wrongly accused. Many actions are described as simply "inexplicable", by which the article writers mean that a Good Person did something that opposed the interests of another Good Person, despite the apparent absence of eunuchs being involved. Occasionally one article will have a sentence or two that looks like scholarship, which just makes the rest look more ridiculous. There are generally few if any citations. Different articles touching on the same events or people will sometimes deliver completely contradictory accounts, without any mention of the existence of rival views. Of course, this isn't entirely the fault of the wikipedia editors - a lot of this is relying on only one or two non-contemporary sources, which are themselves wildly self-contradictory (one source gives three totally different explanations of the same person's genealogy, in different chapters, with no acknowledgment of the contradictions). But even so, you kind of have to feel, if this is the sort of information Chinese people have about ancient history, what on earth is their knowledge of more politically-controversial modern issues like!?
[similar sensationalist sources exist for Roman emperors, particularly those written by Christians, who tended to sort everyone into Good and Bad depending entirely on their attitude to Christians - but the difference is that in the West those sources are widely recognised as non-credible. In China the equivalent sources seem to be regarded as unchallengeable dogma...]
Oh, and a second rant:
Spoiler:
For fuck's sake will you please stop killing everybody. How did China ever accumulate a large population? Everybody is either murdered or forced to commit suicide. Like, EVERYBODY. Woman orders an unusual vegetable for lunch? She and everyone connected to her must commit suicide (real example!). Man writes a prayer for general prosperity on a figurine of a turtle? 10,000 people must be tortured to death (real example!). Every time a woman adopts a child, she seems to have to have the biological mother killed. One emperor is singled out for his unusual habit of NOT murdering everyone who helped him come to the throne. I mean, Rome could be brutal - people were killed in wars, and occasional purges, and of course Roman law could be harsh at the best of times. But it was at least based on some sort of rationality and process, as opposed to the general 'this child blinked while I was speaking so everyone who grew up in the same village as him must be put to death' attitude the Chinese seemingly had at the time (not a real example - it was only the blinking child who had to be killed on that occasion!). Ugh! You could get so much more done if you didn't spend half your day commanding people to commit suicide, damnit! And the obsession with killing people for possible witchcraft! Sure, Romans did that very occasionally, but in general they very quickly got out of the habit of just assuming that witchcraft existed, let alone that every accusation was accurate. In China it seems like that was a lesson every emperor had to learn individually, and only after commanding many people to commit suicide...
.


Anyway, those frustrations aside...

Mean age at accession: 17
Mean age of 'natural' death: 31
Mean reign: 13 years
Quartile reigns in years: 3, 13, 19 (29)
Immediate succession by biological child: 45%
Eventual succession by biological child, niece or nephew: 58%
Survival: 64%
Most common route to throne: primogeniture (24%)


As you can see, China in this era looks in many ways more than monarchic England than imperial Rome - even Late Rome. Here's a few headlines:

- Chinese monarchs were young - 17 on average (median 15). That's compared to 24 among the Anglo-Saxons, the lowest I've found so far. And less than half the age of the Romans, who on average came to the throne at 40 (44 for Early Empire, 34 for Late Empire). Specifically, around 40% of Chinese Emperors (from these dynasties) were children (under 13), 30% were teenagers (under 20), and 30% were adults.

- despite their young age, they didn't have very long reigns: 13 years, the same as Late Roman emperors (who were twice their age). Albeit a little longer than the 8 years for Early Rome.

- however, one interesting thing pops out: adult Chinese weren't that unlike adult Europeans. The average Chinese emperor who took the throne as an adult was 34 years old, and reigned for 15 years; that compares very closely with Late Romans overall (34, 13) and with Tudors and Stuarts (32, 16 years), and not that badly with the English overall (27, 17). So Chinese monarchs did have similar reigns to europeans, provided we exclude the child-emperors (I didn't bother excluding children in this way with romans or english, because there were so few as a percentage that I doubt it would have made much difference).

- importantly, however, age of accession had much less impact than you might imagine on the length of reigns: 13 years for children, 15 for adults (actually more like 12.5 vs 15.5). This is probably largely the result of demographics, and the brutality of infant and young adult mortality. The average emperor aged under 10 only lived to 14; aged under 13, to 18. If we put the 'child' threshold at 18, then children lived to 23, while adults lived to 46. But if we put it only a few years later, under-20s lived on average to 24 while over-20s lived to a whopping 52. As a result, lowering the age of accession may allow for occasional much, much longer reigns, but it may not lead to an actual increase in AVERAGE reign. Indeed, it will probably do the opposite! We've actually already seen this phenomenon in Rome: early Romans on average acceeded 10 years later than Late Romans, but also on average had 'natural' deaths 12 years later. [the effect is partially concealed because of the brutality of Rome: Early Roman Emperors "should" have had average reigns of 20 years, but in fact they were only 8 years, due to the sky-high risks of unnatural deaths].

- speaking of which: murder was very rare among the Chinese monarchs (even though, or perhaps because, murder BY chinese monarchs was ludicrously through the roof). I haven't actually registered a single case of murder - although wikipedia does allege a few. With Roman emperors, I decided not to listen to rumours unless there seemed to be compelling evidence or historical consensus; with China, it seems that the historical consensus is always just to accept the wildest rumours, but none of the allegations persuaded me. You may want to explain why your saintly, perfect child-emperor suddenly died and why it must have been all the fault of the evil eunuch supervillain, but let's be honest he had no obvious motive, there's no evidence other than rumour, and the reality is that children in that era (and most eras) did just randomly die a lot, so if we're going by science rather than seeing history as nothng but a morality play for our instruction, we ought to assume the kid just died naturally, OK?

- another thing that suprised me was the lack of primogeniture. In theory, descent was by primogeniture, but in reality this was only the case about a quarter of the time. There seem to have been three big reasons for this. One was meritocracy. It was common for emperors to have a horde of children, and to pick the one they thought best suited for the task of ruling an empire (or, you know, the one a warlock correctly told them would be a magical saviour sent by heaven). [I suspect marital dynamics were also an issue: Chinese emperors had countless concubines and sometimes multiple wives, which could give personal and political incentives for an emperor and his officials to prefer a son who was not his eldest].This happened surprisingly rarely in Rome, although there were examples - and basically not at all in England, at least after the Normans arrived, at which point primogeniture became an almost fixed rule. The second big factor was simply what we've discussed: child emperors dying young. Many simply didn't have children of their own yet - or, if they did, they were so young that they could command little authority (it is probably easier to justify ignoriing primogeniture when you're picking which random baby to enthrone than if you're trying to disinherit a 30-year-old with his own friends and army). And the third thing? Intrigue...

- no, seriously. Court intrigue was a huge issue, to an extent not seen in Rome. Specifically, the locus of political plotting was different. A huge number of roman emperors gained that position simply by being proclaimed as such - by the Senate, by local officials in Rome or Constantinople, by a general, by the praetorian guard, or simply by a unit of professional soldiers. None of those people or their equivalents held anything like the same power in China. In the dynasties in question, I only found one emperor proclaimed by any similar body - and that was a brief child emperor, not even usually counted as an emperor by the Official History, who was proclaimed by some peasant rebels. [I mean, obviously a bunch of other people were proclaimed in this way, but only one was put on the throne successfully. And I'm also excluding here the category of rebel generals who proclaimed themselves emperor directly by means of conquest. "Proclaimed" generals in my sense are those in which some force declares an individual to be the emperor, and this is accepted, without that individual actually leading an army in battle to enforce his authority]. But that doesn't mean that there were no emperors by proclaimation in China. On the contrary, there were a whole bunch of them - in these dynasties, about a quarter, by my count. But the proclaimers were different: these emperors were proclaimed by forces inside the palace. Generally, by some combination of eunuchs and dowagers. Unlike the external proclaimations in Rome, which tended to promote popular, known individuals (otherwise how would anyone know to proclaim them?), Chinese internal selection was able to appoint obscure, powerless candidates - most often young children (sometimes infants) who were relatively distant relatives of the recent emperors. These emperors would, at least for some time, be useful puppets for those who selected them (unlike Roman proclaimees, who almost inevitably either immediately dominated or immediately struggled against those who had put them in power). Choosing a distant relative - in a few cases people who had even been raised as commoners - allowed the palace power structures to circumvent any established networks of inherited personal support that closer relatives might have had.

- this is a really interesting point to me because it suggests a paradox: Roman Emperors were more powerful than most Chinese Emperors in part because they had less power. What I mean by that is that the institution of the Roman emperor was more checked by rival institutions - the senate, the people of rome, the powerful governors and provincial and municipal civil servants, the generals, and the (professional, well-paid, increasingly non-Italian) soldiers themselves. The emperor's personal power was needed in order for his institution to maintain dominance - weak emperors weakened the institution, and were removed either to save the institution or by the enemies of the institution. The institution of the Qin/Han emperor, by contrast, appears to have had few if any limits on its power - other than the occasional (annoying but very rarely dangerous) rebellion, it was able to steamroller any opposition. But as a result, the emperor himself was much less crucial for the functioning of the system. His institution could keep things running without him (until, of course, it couldn't anymore, and the Han collapsed). The people around him - wives, mothers, in-laws, officials, eunuchs - had little incentive to strengthen the emperor personally, and instead tended to prefer and create weak and disengaged emperors, which increased their own power. And because the palace institutions had the power to maintain the imperial office without external interference (i.e. they could appoint emperors), and likewise could largely appoint their own replacements, they and their factions were able to build up multi-generational institutional power. [By contrast, certain Roman palace officials - the emperor's slaves - did become famous as powers behind the throne - but because they could not control who sat on the throne, they were swiftly swept away by the next emperor's new broom...] Thus, the Roman imperial office was always fragile, but Roman emperors themselves were generally despotic; by contrast, Chinese emperors were very often personally weak (and in some cases literal babies), yet the imperial office had totalitarian power.

- a bunch of emperors did still get overthrown by rebellions - 15%, compared to around 21% for Rome. However, whereas this was a constant, or at least sporadic, threat in Rome, Chinese overthrows occured exclusively around the collapse of the Qin, Xin and Later Han dynasties (three of them occuring within a few years as the Later Han was established). Rebels could only imperil an emperor when the imperial system itself was in crisis. On the other hand, another 15% were overthrown by court intrigue. This arguably happened sometimes in Rome (though usually only by the intervention of external agents, like a plotter getting the praetorian guard or the senate to back them) - but not often enough that I bothered to make a category for it when I was doing this for Rome.

- going with the above themes: it's notable that despite the strong dynastic succession, brothers almost never succeeded one another. This is probably for three reason: firstly, emperors were frightened of their brothers, and very often had them murdered, or sent away to remote offices - this was rarer in Rome (pace Caracalla), partly I think because Roman emperors felt they had more personal authority and were less worried that they and their brothers might be interchangeable; and secondly and thirdly, succession by brother directly opposed two preferences of the palace power structure - that is, for younger emperors (hence a strong preference to 'move on' to the next generation when possible), and for emperors more removed from any possible personal power base (hence the tendency to declare that heavenly omens insisted on some random fifth cousin being the next emperor).

- however, it's notable that several tendencies I've talked about here also look like, as it were, the 'end point' of the drift from the Early to the Later Roman Empire, as the hereditary principle (and totalitarianism) were gradually established. At some point I'll have to go back and do this for Byzantium to see how those trends continued. [tangent: but without running the numbers, reading about the chinese emperors really reminds me of the dynamics of the turkish sultans...]



....aaaaaanyway, hopefully someone can get something useful out of all this rambling!
User avatar
Creyeditor
MVP
MVP
Posts: 4604
Joined: 14 Aug 2012 19:32

Re: Roman Emperor Statistics

Post by Creyeditor »

Interesting as usual. I was never really enthusiastic about the idea of a monarchy for conworlding purposes. I kind of assumed that it was the default. If a conculture of mine does not have a dedicated political system, they are probably a boring monarchy. But now that you have shown the variation inside different monarchies, I might just try it. I especially like the idea of a "palace oligarchy"-like system with a weak emperor and stronger "wives, mothers, in-laws, officials, eunuchs" with constant plotting. This might also avoid the 'evil Grand Viziers/Eunuch" trope if there is simply a larger group of people in power.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Salmoneus
MVP
MVP
Posts: 2676
Joined: 19 Sep 2011 19:37

Re: Roman Emperor Statistics

Post by Salmoneus »

Our latest update: Chinese emperors of the (Cao) Wei, Jin, (Liu) Song, (Southern) Qi, (Unmodified) Liang*, and Chen dynasties. That's nearly four dozen emperors, over nearly 400 years.

[why those? Because this is essentially the continuation of Han as a state, dynastically if not geographically. This makes it a bit neater than splitting off into the various rebel or invader kingdoms - although of course those kingdoms in some cases occupied the geograpical heartlands of China, were sometimes more powerful, and ultimately ended up winning the war].

So, as anyone who knows Chinese history might have expected, this epoch was a little more chaotic than the preceding Qin/Han era. That's not prima facie obviously going to be reflected in reigns, though, I guess: geographical division doesn't have to mean instability. Except that for China in this era one major reason for division was instability, and division in turn (though inter-dynastic meddling) exacerbated instability.

The headlines:

Mean Age at Accession: 26
Mean Age of "Natural" Death: 44
Mean Reign: 8 years
Quartile reigns: 2, 4, 11 (22)
Immediate succession by child: 36%
"Perpetuation rate": 55% [percentage of emperors who succeed in having any son, grandson or nephew eventually become emperor]
Survival to a natural death: 38%


In other words, compared to the Qin/Han epoch, the following epoch saw reign lengths plummet: the major decline of the mean (13>8) conceals the total collapse of the median (13>4).

Excitingly (!), reigns in this epoch in China were almost exactly the same as in the tumultuous "Pagan" (Augustus to Constantine) Rome, where the mean was 8, and the quartiles were 1, 5 and 13, with the average upper-quartile reign being for 20 years. Why is that similarity exciting? Because although both periods were famously unstable, and these reign lengths prove it, almost nothing else is the same about the two systems. Early Roman succession was almost entirely non-biological, and the average Emperor acceded at the age of 44 (in other words, the average Chinese emperor would already be dead by their coronation day if they'd followed the Roman route to the throne). And yet the outcome is so similar!

And again: both Wei-Chen and Pagan Rome emperors shared the intimidatingly poor prospect of only an exactly 38% survival rate.

It's bizarre, to be honest, that six different stats (three quartiles, the mean, the mean of the upper quartile, and the survival rate) should be so near-identical. Of course, these stats are all related underlyingly, but I'd have expected them to be a little less corrolated!

-----------

What was going on behind these stats? Well, the big thing is that the 'palace intrigue' thing got worse and worse. 32% of emperors in this period were removed by internal intrigue (usually fatally). The previous distinction between intrigue and rebellion also becomes more blurred: rebellions (now not simply peasant uprisings, but military mutinies) often resulted in a change in power structure within the court (i.e. a rebel general assumes de facto power), which could often trigger intrigues and counter-intrigues. Outright conquest - marching on the capital, killing the emperor and declaring yourself emperor - becomes very rare, only happening near the end of this period; usurpation, on the other hand (where someone already in the power structure declares themselves emperor) was much more common. [a common pattern seems to have been a rebel taking de facto control, replacing the emperor with someone else, then deciding that wasn't a great improvement and taking over themselves]. It's also significant that the people involved in the intrigue were much less likely to be eunuchs, or even women, and were more likely to be essentially provincials (by which I mean: some of them were senior civil servants at the time of the intrigue, but they were usually not political insiders, but people who had forced their way into power). [intrigue involving relatives still continued, of course].


A fascinating element of this that I hadn't expected is the symbolic role of dowager empresses. In a very real sense, it almost seems that in this era the true ultimate 'authority' in China was NOT the emperor, but the dowager. Repeatedly, dowagers demoted Emperors to lower ranks as part of a change in power; at least twice an emperor was removed from power for having allegedly plotted against (or wanted to murder with a sword) the dowager. This is weird because there's no other sense that the dowager was or ought to have been more powerful than the emperor; they only seem to serve as a constitutional font of legitimacy. The paradox, indeed, is that although dowagers were indeed frequently directly political powerful in the Han era, when they did not seem to have this ultimate authority (and had to have everything done in the name of the emperor), they rarely if ever appear to have wielded overt political power in the Wei-Chen era (even though everyone else had things done in their name).

[another oddity: there were at least three imperial princes purported to have severe learning difficulties - and all from different dynasties, so it's not even a genetic thing! Of course, since chinese history largely reads like bad fanfiction, it's possible that this is a narrative device rather than a medical issue in some or all of these cases**]


On the other hand, something non-odd, by Western standards: primogeniture has definitely become the expectation in this era. It was common in the Han, but always seemed a bit haphazard. It's still not absolute in the Wei-Chen epoch, because a bunch of emperors reportedly debated making a junior son heir (and there seems to have been a more fixed rule that sons of minor concubines didn't count); but they were almost always dissuaded, apparently with the explicit reasoning that primogeniture was best as a general rule and the rule shouldn't be undermined. Only a few seem to have actually gone through with violating primogeniture, and usually for seemingly specific reasons (son has learning disability, or emperor wants son-of-dead-eldest-son to inherit over his junior uncle). That said, the lynchpin of succession in 'normal' times was still selection by the previous emperor, via appointment as crown prince during the emperor's own life (oddly there aren't any clearcut cases of sons killing their fathers to inherit, perhaps because of combination of patricide taboo and general instability of ruling families - they probably knew they'd be deposed if it were proven); although, as mentioned, empress dowagers seem to have had some authority to subsequently demote crown princes (or even emperors), at least in theory.

However, despite this ideal of primogeniture, the instability of the era ensured it was much rarer than might be expected. Only 36% of emperors were sons of their predecessor (29% firstborns), but the real problem is when you look for sequences of succession: only once in nearly 400 years did a firstborn son inherit, die of natural causes while emperor, and be succeeded by a firstborn son. Only once! So although dynasts in this era had evidently internalised the idea of primogeniture, and kept trying to return to it, constant outbreaks of bullshit kept getting in their way...




Meanwhile, not everything was awful! Life expectancy had improved compared to previous centuries: the average reign ended at the age of 33, up from 30 under the Qin-Han (not as the result of a skew - all quartiles rose a similar amount). Actual 'natural' life expectancy rocketed up from 31 to 44 (perhaps indicating that some Qin-Han deaths had not actually been natural? Or perhaps reflecting improved material conditions, despite the instability?).

Of course, both these measures are still comparatively shit. With one exception, our European datasets [Early and Late Rome, and three eras of English monarchs] all had average age at reign end between 47 and 55, and average age of natural death between 50 and 64. The exception is the Saxon kings, who actually had a younger age of natural death than the Chinese (38) but still an older age at end of reign (37). The difference is partly because Saxons, like Chinese, came to the throne younger, while Romans (particularly early Romans) were, as it were, pre-selected for longevity by their advanced age at ascension; but clearly that's not the whole issue, as that also applies to the Normans/Plantagenets, who had the same accession age (26) but a 'natural' life expectancy 11 years longer (and age at end of reign a whole 22 years older).


----------------------

I thought I'd add two more measures that might be interesting to conworlders: "how long will this king reign?" and "how much of a century is made up of reigns of which lengths?"

For the former:
- 100% of monarchs (whose age is known) lived to at least 10. [mostly because those who died younger than that rarely became monarchs]
- 73% lived to 20
- 60% lived to 30
- 42% lived to 40
- 22% lived to 50
- 2% lived to 60 [although bear in mind sample size is too small to take this too seriously]

[this compares to Qin-Han figures of 82, 64, 48, 33, 12 and 9%]

But we can also put this another way: the chances of surviving to the next decade:
- 100% made it to 10 (82% for Qin/Han)
- of those who made it to 10, 73% survived to 20 (78% for Qin/Han)
- of those who made it to 20, 82% survived to 30 (76% for Qin/Han)
- of those who made it to 30, 70% survived to 40 (69% for Qin/Han)
- of those who made it to 40, 53% survived to 50 (36% for Qin/Han)
- of those who made it to 50, 10% survived to 60 (75% for Qin/Han)

I wouldn't put TOO much weight on the specific numbers due to small sample sizes - particularly for the last two lines (and the first line is more about the number of child emperors than their life expectancy). But both eras have fairly similar patterns and ranges: in general we can say that over about 700 years of Chinese history, there was generally a 70-80% chance of surviving the next decade, but a sudden drop-off in terms of making it from 40 to 50.

However, these figures don't reflect the natural rate of death, because they include all the unnatural deaths. So we could also look only at emperors who died seemingly natural deaths, for an alternative view (though sadly even less statistical validity due to the smaller sample size):
- 94% of those who made it to 10 made it to 20 (89% for Qin/Han)
- 82% of those who made it to 20 made it to 30 (75% for Qin/Han)
- 71% of those who made it to 30 made it to 40 (67% for Qin/Han)
- 80% of those who made it to 40 made it to 50 (25% for Qin/Han)
- 13% of those who made it to 50 made it to 60 (50% for Qin/Han)

What does this mean? I don't know. The early numbers clearly don't reflect real death rates because of the sampling bias (death rates in teens were far higher in reality, but those who survived teens were far more likely to become emperor than those who didn't). The last two lines are pretty unreliable due to sample size.

However, these figures may be useful to somebody as back-of-envelope figures if you're, say, plotting out a dynasty. If you know your emperor survives to at least 30, what are their chances of surviving to 40? Well, if they're Chinese emperors from the Qin to Chen dynasties, about 70%.

I'll have to go back to the earlier datasets to see if anything pops out in comparison.


---------------------


And then there's that second, more confusingly-stated question. What I meant by that was basically: we've talked about the median length of reign, but of course if you're dropped into China in a random year there's a disproportionate chance that the current emperor has an above-average reign length. [if we imagine one emperor who reigns 91 years, and 9 who reign 1 year each, the median reign with be only 1 year, but there's a 91% chance that a random year sees the reign of an above-median emperor!]. The median tells us something about the longevity of a given emperor, but not about the longevity of the emperor of a given year. The raw average is actually more useful in this respect, in that it will at least tell us how MANY emperors we expect to find in any given span of time, but it still doesn't fully answer the question.

So, here's the answer. If you are dropped into any given year of Chinese history in the Wei through Chen dynasties...
- there's a 15% chance the current emperor will not reign longer than 5 years (in total, that is, not from the given year onward)
- there's a 14% chance the current emperor will reign between 5 and 10 years
- there's a 17% chance the current emperor will reign between 10 and 15 years
- there's a 14% chance the current emperor will reign between 15 and 20 years
- there's a 20% chance the current emperor will reign between 20 and 25 years
- there's a 21% chance the current emperor will reign more than 25 years

This is incredibly important for understanding stability!

In short, 53% of emperors reigned for fewer than 5 years, but only 15% of years had such short-reigned emperors. Instead, 41% of years saw an emperor who would reign for more than 20 years. The shockingly brief median (and even average) reigns give a good sense of the prospects of any individual emperor, but give a very misleading idea of actual imperial turnover and its consequences for political stability on the ground.

Let's use broader categories, actually:
- ephemeral emperors (reigning 5 years or less) governed 15 years of each century
- average emperors (reigning 5 to 20 years) governed 45 years of each century
- long-lived emperors (reigning more than 20 years) governed 41 years of each century (even though there were only 5 of them in the entire ~360 years)

Really gives a sense of why "Good King X" and "Y the Great" and "Z the Old" tend to dominate popular perceptions of history: there's not many of them, but they take up almost as much time as all the other memorable monarchs put together!


Again, I'll have to look to see how this works in the other datasets as well.


Also, I need to think whether I can meaningfully capture anything about reign-clustering. What I mean by that is that two centuries could each have 15 years of ephemeral emperors... but the 'stability' of the government could be very different if one century gets those 15 years of ephemeral emperors all in a row, and another century gets them spread out evenly across the period.


Many other things to consider here, clearly...

--------------





*Wait, in addition to the Liang, the Former Liang, the Later Liang, the Northern Liang and the Southern Liang, there were TWO "Western Liang"s!? Ugh...


**seriously, I'd hoped that as we moved out of the Han we'd come to more solid history, but no, we're still getting lines like "he can be considered a super-extraordinary genius, and was defeated by wickedness" in lieu of serious assessments...
User avatar
Creyeditor
MVP
MVP
Posts: 4604
Joined: 14 Aug 2012 19:32

Re: Imperial Statistics (NP: Wei through Chen dynasties)

Post by Creyeditor »

Another big thank you for all the work you put into this. Especially because I am working on my conlang Kobardon, whose conculture is supposed to be be inspired by Rome, China, and Europe (intentionally broad). Guess the monarchic period will no longer be a mythological time with not a lot of details.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Post Reply