Gallery of Philologists
Pseudo Dz-sya

Three Pigs

The following tale is told in the Lw-shr Chun/Chyou (LSCC 22F3, c0209):

Dz-sya went to Jin, and passed through Wei. He happened to read in their historical records that "The Jin army three pigs crossed the River." Dz-sya said, That is not right; it the correct reading is "on the cyclical day ji/hai. "Ji" and "three" are close, and "pig" and "hai" are similar.

When he reached Jin, he inquired about it, and [their text] read "The Jin army on the cyclical day ji/hai crossed the River."

This represents the earliest report, in any language or language tradition, of a conjectural emendation which was subsequently confirmed by the reading of a better manuscript. The story cannot be true of Dz-sya (in the early and middle 05c), or of anyone else at any time, since for one thing there were no chronicles for the separate states apart from that of Lu until the popularity of the Chronicle of Lu suggested their composition, in c0312. It belongs rather to the perceptions of the time it was written (the late 03c, and specifically, during the early years of the reign of the second Chin Emperor). But it is interesting that the concept of scribal error and learned emendation did exist at that time. This layer of the LSCC text probably derives from scholars of Confucian persuasion who were associated with the Chin Academy, and reflects their milieu. Their idea of the work of a scholar evidently included error correction. The story itself involves the presumption that the Wei archive record in question was a faulty scribal copy of the Jin record, not that it was an independent record of the same event, a not necessarily convincing scenario. The story is probably merely a pretext for presenting a rather neat textual emendation.

From this story, and from the two proceeding it in the chapter, whose common theme is the danger of misunderstanding something in an ancient tradition, LSCC proceeds to draw the following moral:

Statements often seem wrong but are right; they often seem right but are wrong. Right and wrong must be distinguished, and this was something that the sages were careful about. But how is one to be careful about it? If you trace back to the nature of things, and the nature of men, you will be able to understand it.

In other words, human and situational probability. We have noted elsewhere that the tone of the LSCC is not only frequently Confucian, but sometimes specifically Mencian. There happens to be a later Mencian passage which deals with the strategy of reading difficult texts, just as this one deals with the strategy of reading corrupt texts. It is in MC 5A4, from c0280, some sixty years earlier then the story above. It deals with the danger of overinterpreting as literal some poetic statement in the Shr:

Don't because of a word do violence to the phrase; don't because of a phrase do violence to the meaning of the passage. Use your intelligence to apprehend the meaning; this is what we call understanding it.

The canon of human and situational plausibility, deployed against the possibility of going astray in the interpretation of difficult or corrupt passages in earlier texts, is essentially the same. The passages even end with the same two words, literally "getting it."

The invented nature of the example can be seen in several ways. (1) The error corrected by "Dz-sya" would most likely have arisen by scribal miscopying, which in this case would imply a copying of the Jin record into the Wei record; this is not a likely scenario. Wei would have kept its own record, and the error in question is not plausible for an original entry. (2) In the supposed original, the record of Jin, the army in question is described as that of Jin. This is not the style of the one firmly extant example of a chronicle, the Chun/Chyou of Lu, which for its own forces uses rather "I" or "our" (wo). (3) The position of the time adverb between the subject and the verb is also aberrant. Genuine chronicles (that of Lu) or those possibly written in imitation of them in the 04c (the Bamboo Annals, which are thought to have included the actual chronicle of Jin) put the date, including any cyclical days, at the head of an entry; the date was the principle of arrangement.

If these objections hold, then the one case of a successful conjectural emendation in the early Chinese literature was in fact made on an invented text.


  • James Legge (tr). The Works of Mencius (Chinese Classics v2)
  • John Knoblock and Jeffrey Riegel (tr). The Annals of Lü Buwei. Stanford 2000

To Next

1 Dec 2005 / Contact The Project / Exit to Gallery Index Page