Paul R Goldin
University of Pennsylvania
The Date of the Zuozhuan and the Hermeneutics of Emmentaler
Panel: The Historiography of Spring and Autumn
AAS Convention, Chicago, 24 March 2001

Paul R Goldin

The following is a position paper composed for this panel at Bruce's kind invitation. It is not a research paper or a rigorous study; the allotted time permits me to state my views, but not to defend them.

The date of the Zuozhuan is a significant scholarly issue for two reasons. First, the Zuozhuan is a monumental work covering a period that is among the most poorly understood in all of Chinese history. Scholars are eager to know more about how they can use this source. Second, there are few internal clues in the text that researchers can exploit to establish firm absolute dates. To be sure, passages are routinely provided with specific dates, but therein lies the problem: there is usually no external confirmation, so it is not clear whether the accounts are contemporaneous with the events that they describe, or whether they were written at a later time. Compounding this ambiguity is the possibility that the text may quote or incorporate genuine ancient documents.

There are two general camps, which for the sake of convenience I shall call "Chunqiu" and "Zhanguo." The Chunqiu view is that the Zuozhuan is a primary document from Chunqiu times and thus can be used as a source for Chunqiu history. In practice, this point of view comes in two forms: a strong form claiming that the entire text - or at least the overwhelming majority of it, excluding postulated interpolations - dates from the Chunqiu; and a weak form claiming that the received text may be the product of a Warring States redactor, but that the text still contains large sections of genuine Chunqiu material.

The Zhanguo view, by contrast, holds that the text was compiled in Warring States times and conveys a retrospective and romanticized image of Chunqiu history. According to this view, the Zuozhuan is still vitally important to the intellectual history of the Warring States and Imperial eras, but is not much more appropriate as a source for Chunqiu history than, say, the Sanguo Yanyi for the Three Kingdoms.

Any interpretation of the Zuozhuan must deal with a substantial number of passages that can only be considered "errors." These include: prognostications that history does not confirm until long after the end of the Chunqiu period; prognostications that history subsequently refutes - again, long after the end of the Chunqiu period; mistaken astronomical information that must reflect later calculations rather than contemporary observations; and outright anachronisms. (In the last category, Bruce frequently - and rightly - points to the absurd story about the casting of iron vessels inscribed with a penal code). These issues have been discussed at length by eminent scholars, so I need not rehearse the details here. Yang Bojun concludes on the basis of the prognostications that the text must have been compiled between the years 403 and 389. The magnitude of the error in certain astronomical data, similarly, suggests a date of c365.

I think these passages are devastating to the Chunqiu view. Taken singly, any one of them might be dismissed as inconclusive; but collectively, they are compelling because they all point in the same direction. Moreover, it is rarely stressed that these are the only passages in the entire text that can be dated directly. The point is not that there happen to be a few odd passages incompatible with the Chunqiu theory. All the datable passages in the entire text are from no earlier than the 4th century, whereas no proponent of the Chunqiu view has ever identified a single passage that must antedate the Warring States. The score is about 20-0. [Note 1]

Chunqiu advocates usually sidestep this problem by declaring these passages to be interpolations, and then dispensing with them entirely. This is what I mean by the "hermeneutics of Emmentaler." As more and more of these alleged interpolations are discovered and promptly removed from consideration, the image of the text that emerges is that of a great wheel of Swiss cheese, with Zhanguo bubbles and Chunqiu interstices. One cannot identify a passage as an "interpolation" simply because it is inconvenient to one's theories about the date and composition of a text. There must be some linguistic or philological protocol. But these are rarely offered; nor are we often told how and why a later writer goes about surreptitiously interpolating things like prognostications that history eventually proves untrue.

These points are well known, and yet Chunqiu advocates still live and breathe among us, so their sense must be that the overall quality of the text is sufficiently Chunqiu-ish that the "error" passages may be disregarded as the manifestations of careless Warring States packaging. This would be a "weak form" of the Chunqiu view. My sense, on the contrary, is that the overall quality of the text is extremely fourth-century-ish. The language sounds like archaizing fourth-century writing - though I am aware that this matter has been hotly debated, without any irrefutable arguments on either side. One inadequately appreciated point concerns the use of the word dao in the Zuozhuan as an ethical term. This is rare in textual and palaeographic literature from before the Warring States. There are sporadic occurrences - one in the "Junshi," for instance, and Shi 245 refers to the dao of Houji - but in the Zuozhuan, this sense is attested far more than sporadically. The Fraser-Lockhart Index lists dozens of References under such categories as "good government," "the way, path of duty, reason," "principle," "general rule." We know now from the Guodian manuscripts (among other texts) that the dao was a crucial ethical and political concept in fourth-century philosophy, but there is not much evidence that it enjoyed this status before then.

Next, there are certain bizarre features of the narrative in the Zuozhuan that are not easily compatible with the Chunqiu thesis. Take the character of Lord Mu of Qin. In the Battle of Han (Xi 15) for example, he is portrayed as a paragon of virtue and forbearance; he attacks Jin only in order to punish its treacherous ruler, Yiwu; after capturing Yiwu, Lord Mu spares his prisoner and eventually returns him to his homeland; and in the aftermath of his victory, Lord Mu continues to treat the nation of Jin kindly, because his quarrel has been not with its people, but with its lord. His troops, moreover, are said to be possessed of great fighting spirit, and he commands them with insight and aplomb. Above all, he listens to his advisors.

Then, eighteen years later, in the Battle of Yao (Xi 32-33), Lord Mu plans an unsound campaign of conquest despite the pointed remonstrances of his ministers. Now he exemplifies all the commonplace characteristics of a doomed ruler in the Zuozhuan: he is overconfident, has no sense of ritual, and is greedy for territory. Of course, his forces are smashed and he is humiliated.

Lord Mu was not a perfect ruler - this is the same Lord Mu who forced those three good brothers to be buried alive with him when he died - but there is no hint in the account of the Battle of Han that he was the kind of ruler who would ignore the counsel of sage ministers in a vain attempt to seize a few scraps of territory. It is shocking that the same man should make all the shortsighted mistakes that, eighteen years earlier, he so wisely identified and so admirably avoided.

I think this difficulty is a consequence of the competing constraints on the author or authors of the Zuozhuan: the philosophical theory, on the one hand, that Heaven always helps the virtuous defeat the iniquitous; and the historical fact, on the other hand, that Qin defeated Jin in 645 but was defeated by the same enemy in 627. In the Battle of Han, the author is compelled to portray Lord Mu as a moral hero and Yiwu as a tyrant; the Battle of Yao is written as simply another episode in the ongoing struggle between right and wrong - but this time, Lord Mu must be depicted as the personification of impropriety. Neither of these passages tells us very much about the real Lord Mu.

One final, general comment about the battle scenes in the Zuozhuan: they read like the nostalgic chimeras of later ages, and not like forthright contemporary accounts. They are all about heroism, honor, and Heaven-ordained victory or defeat; they glorify individual valor and condemn ignominious folly, with little consideration of practical concerns such as strategy and logistics. Moreover, they never convey the horrors and atrocities of war: the reader is spared the gruesome sight of civilians raped and slaughtered, the cries of tortured prisoners, or even the inevitable stench of corpses decaying on the battlefield. As a genre, they are more like Mallory - or Homer, to follow Eric Henry's suggestion - than The Face of Battle, and are therefore highly suspicious as historical documents.

In conclusion, the Zuozhuan espouses fourth-century ideas in fourth-century language, and every datable passage in it must be assigned to the fourth century. I believe it is a fourth-century text.


Note 1. On the WSW E-mail list, Yuri Pines (Message 3088, August 8, 2001) cites two mistaken predictions (Xi 23, Jin will be the last of the Ji states to perish; and Xiang 31, Zheng will enjoy several generations of good fortune) that imply a date before the 4th century. Pines himself suggests a latest plausible date of 450. [Return to text]

All papers and abstracts posted on this site are Copyright © by their authors.

 Back to Panel Listing

20 Feb 2001 / Contact The Project / Exit to Conferences Page