This week’s achievement is that I finished the Romance of the Three Kingdoms. This is Moss Roberts’s translation of 三国演义 Sānguó Yǎnyì, attributed to Luó Guànzhōng, one of the four great Míng novels. The Míng novelists had time on their hands and a dense writing system, so their novels are tomes; this one is over 1400 pages.

The title is a bit of a spoiler— the three kingdoms don’t settle down as such until chapter 98 (of 120). But the overall story is how the unified 汉 Hàn empire broke up, a dozen or more generals taking power in various regions, finally consolidating in a threefold division:


  • Wèi in the north, where 曹操 Cáo Cāo claimed to act for the Hàn emperor. His son Cáo Pī deposed the puppet emperor in 220 and seized the throne.
  • 蜀汉 Shǔ Hàn, declared in the southwest by 刘备 Liú Bèi the next year, claiming to continue the Hàn line.
  • in the southeast, where the cautious or indecisive 孙权 Sūn Quán did not declare himself emperor until 229.

If you look at the map, you’ll notice that the kingdoms don’t extend to the southern coast. The book has a wide geographical scope, but very little happens south of the Yangtze. This is because, at the time of the book, the far south was non-Chinese territory, unhealthy and occupied by barbarians.

For Western readers, the book is a little like all of these:

  • Herodotus’s account of the Persian war, concentrating on the intrigues and cruelties of dozens of monarchs
  • Malory’s Morte D’Arthur, where brawny knights ride around beating the hell out of one another
  • Lord of the Rings, which after all is the moralized story of a war, with light supernatural incursions
  • Shakespeare’s history plays— national drama turned into the literal kind
  • The Three Musketeers, also a story of a cunning but ultimately amoral prime minister, and a relatively lowly group which has sworn to defend each other and the King

What three kingdoms?

The book covers a span of 96 years, beginning in 184 with the rebellion of the Yellow Scarves; the emperor asks for assistance in putting down this bandit rebellion, and many of the later contenders for the realm were at first allies in this common struggle. It ends with Wèi’s conquest of the two southern states, and the usurpation of Wèi by the Sīmǎ clan, reuniting China under the 晋 Jìn dynasty. “The empire, long divided, must unite; long united, must divide. Thus it has even been,” the novel proclaims.

The novel is closely based on history, especially the Sānguózhì, written soon after the Jìn unification. But a thousand years had produced more histories and commentaries, and above all the development of legends and plays based on the Three Kingdoms. The novel is far longer than the Sānguózhì and, though it tones down the supernatural elements of the plays, retains what had become the received interpretation of the period: Cáo Cāo was a cunning traitor, Liú Bèi was virtuous hero and an ideal Confucian ruler.

(Why did popular culture, and thus the novel, favor Liú Bèi? He was a natural favorite of southerners, and this achieved a particular resonance during the Sòng dynasty, when the Jurchens had conquered the north and a valiant resistance from the south was a matter of Chinese pride. The Jurchens played along by favoring Cáo Cāo, which helped turn him into a villain.)

Now, how do you reduce a sprawling century of war and intrigue to a novel people can actually follow? Well, in part, you let it sprawl. Roberts’s notes include a list of “Principal Characters” that runs to 115 names, and there are literally hundreds more minor characters. Whenever there’s a major expedition, Luó doesn’t just tell you the generals, but their lieutenants and the defenders of key towns. It’s the opposite of most fantasy novels, which reduce a major war to no more than a dozen personalities, ideally all related. But it’s effective— you’re not going to remember all those names, but you get a picture of the immensity of the historical canvas that few other works can provide.

The novelist excels at lapidary little portraits of the characters. Though they’re mostly defined by action, you get a strong sense of personality from even the minor characters.  (I also have to say, not a word is wasted. The book is long but nothing is irrelevant to the overall story.)

Heroes and villains

But beyond that, to make the story manageable, you identify heroes and villains. The majority of the book concerns the developing struggle between Cáo Cāo and Liú Bèi (who is normally called by his style, 玄德 Xuándé). As Roberts points out, the quintessential Cáo Cāo moment has him smiling coldly as he elucidates some point of strategy, or executes a rival; the prototypical Xuándé by contrast is weeping over the state of the kingdom, or the loss of a general.

Cáo Cāo in general continually moves up: at first one of many generals, he achieves a special position when he rescues the boy emperor Xiàn. He becomes prime minister, and makes the boy into his puppet; yet he retains till his death a scruple against actually seizing the throne himself. The novel always gives him his due as a master strategist; he’s often called a tyrant and usurper, but from his own point of view he is simply doing his best for the realm and reducing one rebel after another. He defeats his major rival in the north, Yuán Shào, and attempts to conquer the south as well— but never succeeds in doing so.

Xuándé, by contrast, goes through a series of elevations, betrayals, and reversals. In the first chapter he finds two allies, the lordly Guān Yǔ and the irascible Zhāng Fēi; they make an oath in a peach garden to support each other as brothers and to maintain the Hàn. Xuándé himself is a remote member of the Hàn imperial house, the Liú, though so long as there is an emperor he supports him. He gains and loses several cities— always being known as a just and kind ruler, but never quite as strong as his adversaries. In early years he is allied with Cáo Cāo, but the intrigues of the warlord conflicts pulls them apart.

Xuándé’s fortunes change— and the novel really hits its groove— when he meets 诸葛亮 Zhūgě Liàng, styled Kǒngmíng. At this point you may well ask, what the heck is a style? It’s an extra name given to males at the age of 21, normally by the parents. Men of the same generation would address each other by their style. In TK the style is used mostly for Xuándé and his associates, apparently a subtle way of indicating that they’re the good guys.

Enter Obi-Wan

A hero needs a wise, curmudgeonly mentor, and that’s Kǒngmíng. He’s living in a hut as a Dàoist, and Xuándé has to come three times to ask for him before Kǒngmíng even deigns to speak with him. (Cáo Cāo would never show that level of courtesy to anyone.) Fortunately they hit it off, and Kǒngmíng becomes his military advisor and eventually prime minister. One of the pleasures of the book is watching Kǒngmíng ply his tricks: an ambush here, a conflagration there, a seemingly vacated camp, a betrayal that it turns out the master has foreseen. Few epics are as concerned with the pithy details of military strategy.

(You’d expect the master to be an old bearded guy, but in fact Kǒngmíng was in his twenties when he met Xuándé.)

The first order of business is an alliance with Sūn Quán against Cáo Cāo. The latter had recently beaten Yuán Shào, consolidating his hold on the north of China; he advanced to the south with an army of 200,000 men. The southern general, Zhōu Yú, had about 50,000. To conquer the south Cáo Cāo had to cross the Yangtze; he had his men on ships, and as they weren’t used to the water the ships were bound together. An enemy commander approached with his fleet to defect. However, the defection was a ruse: his ships were dummies filled with oil and kindling. The sailors set them afire and escaped, letting the wind blow them into the fleet at Chìbì (Red Cliffs). The result was a disaster, and Zhōu Yú and Xuándé heavily harrasssed the retreating northern army. In history, this is a key battle (if you’ve read the PCK, I used it there), effectively ending Cáo Cāo’s hopes of unifying China. The novel improves it by having Guān Yǔ capture and disdainfully release the fleeing Cáo Cāo, and by having Kǒngmíng summon the winds which blew the fire boats into his fleet.

Next Kǒngmíng engineers a new territory for Xuándé— the city of Jīngzhōu. Technically this was part of Sūn Quán’s territory— Kǒngmíng explains that Xuándé needs to borrow it for awhile while they conquer Shǔ to the west, but this sets up a southwest-southeast conflict that would be fateful later. Xuándé proceeds to conquer Shǔ while simultaneously fending off northern attacks by Cáo Cāo. This is a little tricky inasmuch as Shǔ has a lord of its own, Líu Zhāng, a fellow member of the Liú clan; events are engineered such that Xuándé takes over the province without having to break with Líu Zhāng. (Not a few times, Kǒngmíng urges Xuándé to some morally questionable behavior. Though we’d all prefer virtuous monarchs like Xuándé, sometimes we also want a canny bastard like Kǒngmíng on our side.)

During this time, Cáo Cāo becomes more and more tyrannical, is named to higher and higher offices (finally becoming King of Wèi) and reduces what little power emperor Xiàn retains— even killing his wife when Xiàn attempts to incite a rebellion. Cáo Cāo dies in 220, and his son Cáo Pī finally deposes Xiàn and declares himself emperor of a new Wèi dynasty. In history Xiàn lived peacefully as a duke, but in the book it’s rumored he’s been murdered, which leads Xuándé to declare himself emperor of Shǔ Hàn.

History makes the novelist’s task difficult here… with a superhuman mentor, why shouldn’t Shǔ Hàn go on to victory after victory? The narrative dilemma is solved by having Xuándé ignore Kǒngmíng’s good advice— which is to maintain the alliance with Wú. The problem is that Wú had fought a war to get Jīngzhōu back, and during this Guān Yǔ is killed and a defector assassinates Zhāng Fēi. Xuándé feels he must avenge his oath brothers, and carries on the war without Kǒngmíng. It goes well for awhile, but then he loses a major battle and dies, just three years into his reign. That’s what you get for ignoring the master.


Now Luó has lost his main character and his main antagonist, and we’re only up to chapter 85. But he is seeking a thematic unity, so the book goes on, though the pace quickens. Xuándé is succeeded by his son Líu Shàn— who however inherits really none of Xuándé’s fine points, except for his deference to Kǒngmíng, who runs the government.

Kǒngmíng wages a major campaign against the Mán, the southern barbarians. This is a strange interlude. Westerns are sometimes purveyors of “Orientalism“— positing eastern states run by ugly primitive peoples who have a rude warrior culture, where the women are more sexually available, but who are easily mastered by Westerners. Well, the Mán are treated exactly that way. Theirs is a land of “rampant pestilence”; they have access to shamanistic arts and can control wild animals; women and men “mixed freely and coupled without parental prohibition”. In this section the novelist feels free to include more supernatural and fantastic elements on both Hàn and Mán sides. (It also features the only female warrior in the story.) Kǒngmíng has to capture the Mán king seven times before he submits.

(The Mán are not a real group— “蛮 Mán” is a geographical term, the barbarians to the south. The people to the south of the ethnic Chinese were, then as now, a mixed bag: Tai, Miao-Yao, Mon-Khmer, Austronesian. The campaign reaches into Yúnnán, which even today isn’t entirely sinified.)

He then makes several campaigns against the Wèi. He is full of tricks and victories, and yet makes two major mistakes. In one campaign he trusts a rash general who loses a major city (and with it the chance to threaten the Wèi capital). And in another he is outmaneuvered, becomes sick, and dies (234, or chapter 105).

As Liú Bèi is succeeded by a lesser son, Kǒngmíng is succeeded by a lesser general, Jiǎng Wǎn. He too attempts to conquer Wèi, meets with initial success, and fails to seal the deal. But things are not static in the northland. Cáo Cāo’s successors are each weaker than the last, and the general Sīmǎ Yì accumulates more and more power.

The end is anticlimactic, but thematically apt. A eunuch, Huáng Hào, was his favorite official, and he abandoned the careful and virtuous administration of Kǒngmíng. The realm is described as descending into poverty as corrupt officials rise and emperor Líu Shàn cares about little but pleasures organized by Huáng Hào. Jealous of the general, the eunuch succeeds in sidelining him. As a result, the Wèi are able to push into Shǔ Hàn, and a detachment takes difficult back roads to reach the capital, Chéngdū. Líu Shàn surrenders, in 264— chapter 118.

Now all that’s left is for the current Sīmǎ leader, Sīmǎ Yán, grandson of Sīmǎ Yì, to depose the last Wèi emperor, Cáo Hùan, in 265, declaring the Jìn empire. In 280 Jìn forces walk over the similarly declined kingdom of Wú, and China is united again.

The book thus makes a neat thematic circle: from unity to disorder to unity. Moreover, though the northerners triumph, the line of Cáo Cāo does not: as it had usurped power from the Hàn, it was in turn usurped by the Jìn.

(History marches on, of course. The Jìn did not last long— they lost northern China only thirty years later, and this time the period of disunity would last nearly three hundred years.)

Linguistic sidebar

Would it help if all these names had meaning?

Cáo is ‘class, company’, but mostly just a surname. Cāo is ‘manage’. Sīmǎ means ‘manage-horses’.

Liú has an old meaning ‘kill’, but it’s also mostly a surname. Bèi is ‘perfect, ready’; Xuándé is ‘deep virtue’.

Sūn is ‘grandson’, Quán is ‘power, authority’.

Zhūgě is ‘various-bean’ (possibly a transliteration of a foreign name?); Liàng is ‘bright’; Kǒngmíng is ‘great light’.

All right, maybe it wouldn’t help, but I thought it was interesting.

(Roberts occasionally translates names. E.g. instead of Shǔ Hàn and Wú, he has the Riverlands and the Southlands.)

It’s personal

For Herodotus, every political action in the Persian wars was driven by personal quarrels among the rulers. TK is almost as personal, but it has a greater awareness of logistics and statecraft. After conquering a city, the generals (even the villains) “reassure” the populace— they forbid looting, reestablish order, and confirm minor officials in their posts. The characters are aware that armies are supported by farmers, and that oppressive policies will ultimately undermine warmaking; Confucian morality also, of course, dictated that the virtue of the ruler translated directly into the strength of the state. Generals are very conscious of supply— Cáo Cāo wins his major battle against Yuán Shào by finding and destroying his grain supplies.

There’s a vivid depiction of the relationship between ruler (or general) and his lieutenants. Typically there’s a threat, scouts rush in with the news, and the ruler asks “Who will volunteer to ride out against them?” To lead a successful military action was a quick route upward; on the other hand, losing a battle was grounds for execution.

Or the ruler would ask for advice. If the ruler was truly undecided, it was safe to offer advice to make war or retreat or even surrender; but it was dangerous to offer advice the ruler didn’t want to hear. An advisor whose advice was continually rejected would sicken and withdraw— and that was if he was lucky; leaders felt free to execute someone advocating a contrary plan. (Other leaders could intercede for you and sometimes the leader relented. Xuándé and Kǒngmíng are unusual for listening to all sides without shouting for the executioner. Even they have to execute a few underlings, mostly for treason or inexcusable defeat, but they feel bad about it.)

One of the charming bits is the grandiose titles given to certain generals— e.g. General Who Conquers the South. The wordiness is not present in the Chinese: it’s just 征南将军 Zhēng nán jiāngjūn ‘conquer-south general’.

As TK depicts it, war consists of enormous armies marched to a spot and deployed, then champions riding out from the ranks and engaging in combat with lances. If your man won, your men would be jazzed and rush the demoralized enemy. I assume this is mostly for narrative convenience— as in Malory, personal bravery and strength counted for a lot, but most of the real battle undoubtedly consisted of lowly soldiers whacking at each other. But at least you get a lot of commanders, so it’s not just a personal struggle between the three kings.

Interestingly, it was an accepted principle that a general in the field could disobey his lord’s orders from afar. This undoubtedly reflected a premodern situation where the home base does not have adequate information.

Death and cruelty

Resisting the enemy to the death was highly admired— the enemy general might even bury you with honors. On the other hand, you could change sides honorably, especially if you immediately made a name for yourself by attacking your former overlords. To be caught doing this, however, was unhealthy— the leader would not only execute you but kill all your family.

There’s a fair amount of cruelty and butchery. The normal process for firing people is to behead them, but if the ruler is really annoyed he’ll have them carved up. One of the Cáo emperors, tiring of his empress, issues an edict “giving her permission to commit suicide”. As in Herodotus, I think these things are intended to shock the original readers as they do us. Chinese statecraft posited an absolute monarchy, but one mitigated by Confucian humaneness; tyranny was as despised there as elsewhere.

On the other hand, sometimes a different attitude comes through. One woman, in order to avoid a remarriage, cuts off her own nose and ear; her devotion is praised. Another woman, refusing to surrender to conquerors, bashes her head in against a pillar. There’s a very weird scene where a hunter receives Xuándé and, having no game at hand, butchers his wife. Rather than punishing this psychopath, Xuándé weeps at his sacrifice and rewards him. I don’t think any of these are intended as ‘normal events’— they’re above and beyond the call of duty. (You can of course find similar stories in Western tradition, such as the spy in Herodotus who cuts off his nose in order to make a feigned betrayal more convincing, and there’s a Catholic tradition of self-mutilation.)


There is not a huge role for women in TK, but there are some notable exceptions. Perhaps inevitably, there’s a courtesan who helps drive a wedge between a particular lord and his main lieutenant, by allowing both to think she would marry them; less conventionally, she does this out of patriotism, because the lord has taken over the capital.

Perhaps the most intriguing woman is Sūn Quán’s daughter, who marries Liú Bèi to seal their alliance. She’s depicted as devoted to the military arts, and her retinue of a hundred women are all armed. Liú Bèi is frightened to enter her quarters, but she’s actually rather loyal to him… till her father steals her back.

Lady Sūn doesn’t actually fight, but there is one female warrior, the queen of the Mán tribe— Lady Zhurong, who “had five throwing knives stuck into the gear on her back and an eighteen-span spear in her hand.”

Against kings

TK, while exalting the dedication of those who upheld the Hàn dynasty, makes a strong case against monarchy. Monarchy never works, really. In Chinese history, it’s fair to say that no dynasty succeeded in establishing peace and prosperity for more than 150 years, and even that depended on the luck to have a sequence of able rulers in that period. None of the Three Kingdoms managed it.

The novel (and the official history) blame the eunuchs for the fall of the Hàn. Emperors trusted the eunuchs, as they had grown up with them and they generally had no outside loyalties— so they made trustworthy officials and secret police. They are depicted as duplicitous and corrupt. Ray Huang suggests that the real problem was the landowners. Imperial China is the only nation which, for two thousand years, directly taxed individual peasants. There was never a permanent nobility which interposed between the peasants and the central government. But estates would be consolidated and rich landowners appeared. In theory they owed the land tax as well, but it was not hard for rich men to evade taxes, often with the aid of the eunuchs in the capital, their natural allies against the scholar-officials. (China operated with a very lean bureaucracy, and there was no real possibility of surveying land or even keeping track of changes of title.) The missing taxes would be levied on the remaining small peasants. Eventually the burden would be intolerable, you’d get peasant revolts and then warlords, and the dynasty would fall.

The most pathetic figure in the book is the emperor Xiàn, the symbol of the Hàn for which everyone is fighting. He’s placed on the throne at eight, and at one point has to flee alone from one of the warlords. Cáo Cāo rescues him— but there is little pretense that he will be allowed to rule. He reaches adulthood— he’s 39 when he is finally forced to abdicate— but he’s a prisoner in the palace, and even his empresses aren’t left in peace— Cáo Cāo executes two of them.

But Cáo Cāo’s own descendants end up in the same sorry state. And for that matter, Liú Bèi’s son is a disappointment, and Sūn Quán’s descendants squabble for the throne and weaken the kingdom to the point where a takeover is easy. The book makes a case that the successful ruler has to have the benevolence of Confucius and the canniness of Kǒngmíng— but that this combination is vanishingly rare.


(Edit:) A common trope in adventure stories is that the hero suffers some terrible loss, and that’s their motivation for saving the world.

Interestingly, TK inverts this. Liú Bèi does suffer a loss— the death of his oath brothers. And it arguably leads him to his worst mistake and the one that ultimately dooms his kingdom: he blames it on Wú and ignores the greater threat of Wèi. (This is what makes him ignore Kǒngmíng’s strategic vision.)

Summing up

Should you read it? Of course! I recommend starting with a general history of China, such as the one I’m writing. But the way things work is this: when you begin investigating China, even the dynasty names are exotic, and the individual personalities and events have little meaning. But then you start to make connections, and things start to get interesting. I read TK not because it’s history (a good fraction of it is invented) but because it’s a classic of Chinese literature. But of course it illuminates Chinese history. Names like Cáo Cāo and Liú Bèi become touchstones of familiarity. Plus, the Chinese have been fascinated with this story for centuries— no less than 10% of the Yuán and Míng operas relate to the Three Kingdoms story— and you can now access it in movies, TV series, and video games.

Will you enjoy it? I think so, especially if you like any of the books I’ve compared it to: Herodotus, Malory, Dumas, Tolkien, Shakespeare. Or, so I hear, that George Martin guy.