Response to
Liu Xiaogan (1)
Foreword to Don Munro, The Concept of Man in Early China (repr 2001)

The Author's Note in the 2001 reissue of Don Munro's 1969 book The Concept of Man in Early China (to which we have responded separately) directly precedes a longer Foreword by Liu Xiaogan. This Foreword takes pains to ward off what it sees as the challenges to Don's classic work arising from our book and from Lionel Jensen's Manufacturing Confucianism:

"Although Munro's book has met no serious attack over the years, and no one book has superseded his arguments, two books should be mentioned here because their scope overlaps with problems Munro tackles. Moreover, their assumptions and approaches pose a kind of challenge to Munro's methods and conclusions" [xv]

Xiaogan also made these points in a lecture on "The Challenge of Brooks and Jensen to Sinology," delivered at the Chinese University of Hong Kong in October 2001. Xiaogan has told us that the text of the Foreword is substantially the same as the contents of the lecture. We thus here reply to both, adding some material that may be useful to other readers of TOA. For reader convenience, our response has been divided among several pages. This is the first.

Liu. [xviii] The second book I want to discuss is E Bruce Brooks and A Taeko Brooks's The Original Analects. In addition to providing an alternative translation of the Analects, it proposes a chronological regrouping of all its sections and chapters.

Brooks. The actual situation is not quite that exciting. We find that most Analects chapters (17 out of 20, or 85%) and most sayings within Analects chapters (388 out of 530, or 73%) require no regrouping to be in what we consider to be their original order of composition. These sayings and chapters occupy the same relative position both in the traditional text and in our reconstruction of the original text. The degree of actual rearrangement is thus modest. We admit that the implications are profound.

Liu. According to historical records, the Ch'i version of the Analects contained two chapters more than the Lu version, which is considered the master copy of the received text. The state of Ch'i was conquered in 221 BC, twenty-eight years after the conquest of the state of Lu. The authors propose that this twenty-eight year period "implies a rate of growth" for the Analects, namely a constant rate of two chapters for every twenty-eight years. Based on this rate, they develop an astonishingly precise chronology for the work from 515 to 249 BC.

Brooks. Our Chi Analects suggestion is offered (see our p202, and note the subhead "Another Route") as one among many ways of getting into the question of the history of the text; in this case, a way of opening up the possibility that (1) the text was accumulative and not static, and that (2) its earliest portions might go back to the time of Confucius. The rough guess of 0515 for an Analects start year was fine-tuned in the following argument, using other evidence, to 0479. Xiaogan appears not to have read that later argument, or to have appreciated the experimenter's device of constructing an initial rough hypothesis which is then modified by further scrutiny.

There is nothing particularly "astonishing" about the number 0515, or the number 0479. They are just years. The second of them is the year of Confucius's death. If the date of the first Analects material were firmly known, it would be in one or another year. What counts is whether there are grounds for assigning a text to one year, or its vicinity, rather than to another. The reference to Confucius in the third person in LY 4 (and on stylistic grounds, LY 4 is probably the earliest Analects chapter) suggests that it was not written down in 0515 or in any other year during Confucius's lifetime, but rather after his death. 0479 is the terminus a quo for all such years. It is an intrinsically likely year for the writing down of some of Confucius's most memorable remarks by his surviving disciples.

It was by this sort of reasoning that we fine-tuned the initial rough computation.

Liu. But the foundation and reasoning for their chronological stratigraphy is suspicious. Where is the necessary connection between the growth of the Analects and the sequential conquest of Ch'i and Lu?

Brooks. There is no necessary connection. There is no necessary connection between any two historical events. History is orderly, but not determinate. It may be that a connection can be observed. One thing may be suggestive evidence for the other.

Liu. Can we find evidence that shows the Analects to have had a constant rate of growth?

Brooks. A constant rate of growth is not what our book reports as a conclusion. The text is not an organism with its own DNA; it is a repository for activity occurring elsewhere. The question should then be phrased this way: Is there evidence for a constant rate of intellectual activity on the part of the sponsoring Confucian school? We judge that Confucian school activity would consist in part of responses to ideological and political crises in the world around it. That is, it is likely to be in part a function of the quietness or energy of the times. As stock speculators are continually rediscovering, there is no way such things can be precisely predicted, or subjected to rule. We may well imagine the Confucian establishment in Lu to have been continually active, just as it was no doubt continually in existence, but the level of that activity need not have been constant. Whether constant or not, it will have had an average rate, and what we are attempting in this part of our exposition is to see if we can get some idea of that average rate.

If, with that general purpose, we go on to examine the Analects in detail, we find that the text, meaning its institutional sponsor the Lu Confucian school, responded with shock and outrage in LY 3 to something that looks very much like the Chi assumption of Kingship in 0342, and that it was unusually active in the vicinity of LY 14-15, which are crowded with exactly twice the usual number of sayings per chapter (sayings which deal in a distinctive and sometimes controversial way with issues known to be current in the late 04c). At other points in time, the school seems to have gone for long stretches without issuing very many new pronouncements. The text thus seems to have its excited moments, and also its calm ones.

These are some of the surface facts. There are interesting implications in them They suggest that the Analects was not solely produced by any steady-state philosophical impulse of its own, over the years, but that at least in part it accumulated in the more erratic pattern of a response mechanism. The Analects activity seems not to have been gratuitous and self-generated, but to have peaked at moments of engagement.

Liu. If the conquests of Chi and Lu had not taken place, can we suppose that the Analects would have continued to expand endlessly?

Brooks. Still on Chi and Lu? As it happens, there are test cases for this question. Let's look at a few.

The Gwandz material, which to judge from the Chi minister's name that was later attached to it, is a Chi product, did keep growing after the Chin conquest of Chi. Some of its high-numbered chapters are convincingly dated by many scholars to Han times. Here, then, is a case of the continued expansion of a text even when the original home locale of that text had ceased to exist.

Some of the Jwangdz material gets involved with the Analects in the vicinity of LY 18, and those Jwangdz chapters are thus probably of Warring States date. Sywndz (early and middle 03d) also refers to Jwangdz by name. Some of the Jwangdz material was this intellectually current in the middle 03c. But at the same time, the chapters beyond JZ 29 (the latest attested by quotation in the Lw-shr Chun/Chyou) would seem to be of Han date. Here is another case of continued text activity beyond the end of the Warring States period, when all states (and we do not know in what state, or states, the Jwangdz material was originally accumulated) had ceased to exist.

So also with the Mician art of warfare. Most of those chapters of the Mwodz can be related to the kind of escalation whose other, attacking half is reflected in the Sundz. That puts them in the late 04c and early 03c. But MZ 70, exceptionally, has a schematic directional-symbolism character in the style of the early Han Hwang/Lau correlative groupings. The implication is that the military Micians continued to be active, and that their text continued to grow, into the Han dynasty.

These three Warring States text-producing enterprises thus seem to have continued active past the highly disruptive Empire transition. It would then seem that while a school is supplying something that the government or the society of the time want, its associated text tends to grow. Until something stops it.

Calligraphic Separator

This "growth" question is indeed basic to the discussion. It may thus be worthwhile to explore it further, and we venture to do so, for those who may be interested. (Others may skip to the end of this page by clicking here). The questions in this supplementary section are our own:

Q. Why did not the Confucian texts continue to grow under the Empire?

A. Perhaps some of them did. But the general climate, at first, was one of strong Chin (and early Han) hostility to Confucian texts, and indeed to the whole antiquity-centeredness of the Confucians. The mere possession of Confucian books was made illegal in Chin (in 0210), and remained so until well into Han (0191). Such governmental discouragements tend to affect people living at the time, as some alive today can attest. Governmental hostility was not expressed in equal measure toward the useful sciences of government and warfare, or toward such transcendental writings as the Jwangdz. And it seems that texts of those types did continue to be available, and in some cases to have continued to grow, in Chin times and into Han.

Q. Did even these texts grow "endlessly?"

A. In a sense, yes, since the commentarial and translational literature continues to expand and dilate on such works as the Jwangdz even in our own time. But in more literal terms, the statecraft texts and also the classical Dauist texts, in fact all texts of whatever stripe and persuasion which have their beginnings in the Warring States and continue to be active under the Empire, seem to have become fixed as texts early in Han Wu-di's reign. It was then (or so it is widely thought) that official government favor was extended to the Confucian classics, and the route to official position was decreed to be a Confucian route. It would thus have been impolitic to spend time publicly constructing alternate schemes of public order. Here too, we see the cogency of the principle that texts and life have something to do with each other. We suggest that the same may be true of the Analects before 0249. The suggestion is not really extreme.

Q. Why then did the Analects, in both its versions, stop growing with the extinction of those states?

A. Chi was brought to an end by the Chin Empire, and we have already noted the Chin Empire's hostility to Confucianism. For Lu, it may be relevant that at the time of the preliminary conquest of Lu by Chu in 0255/54, no less a foe than Sywndz (newly hired, and located in the Chu garrison capital at Lan-ling) was put in charge of the conquered territory. Sywndz's scorn for the ravings of the "Dz-sz and Mencius" gang, that is, the school of Confucius's lineal successor and that of its main branch, the Mencian movement, is made clear at several points in his preserved writings. Here is a very openly confessed hostility. Suddenly, in 0249, the hostile Sywndz is in charge of intellectual life in the area of the former state of Lu. It stands to reason that the best thing that the head of the Confucian enterprise in Lu could do, in the circumstances of 0249, was to get out of town while his own head was still on his shoulders. As it happens, this is precisely what Shr Ji 47 tells us happened. It notes that Dz-shvn, the 5th generation successor to Dz-sz, left Lu for Ngwei, and wound up as a high minister in Ngwei.

No fool he. On the other hand, Dz-shvn's son Kung Fu guessed wrong at the roulette table of incipient Empire. He served as a resident Confucian expert to the loser Chvn Shv, and perished when Chvn Shv himself went under, after a brief reign, in 0208. Fu was 57 at the time. Had Chvn Shv's own regime persisted, we might well expect to have had further additions to the Analects by the hand of Chvn Shv, who was the lineally proper inheritor and custodian of the text. Productive philosophical schools need an institutional base. The Analects school lost their base when Chu exterminated the Lu state, and with it any connection that the Confucian school had with that state (and it will have been close; we know that the Confucian headquarters building was very close to the Lu palace). Kung Fu lost his institutional base when Chvn Shv was killed. Kung Fu's lineal successor operated under the Chin and early Han official ban of the possession, let alone the production, of Confucian text. No less conducive institutional base can readily be imagined. When the possibility of an institutional resurgence presented itself, under Han Wvn-di, the prevailing tone at court was Dauist, with a distinct hostility to individual Confucians (the Shr Ji is full of good stories illustrating this intellectual situation). Confucian works were produced in this period, but they were not produced at court; rather in peripheral locations, and largely in the form of exegesis of what by then had become the Confucian classics, not the Analects. By then, the thread of Analects continuity had been broken, and the position the Analects had once held, as more or less definitive for Confucianism, had been irretrievably lost. Confucianism had become focused on the classics older than Confucius, and its growth in Han (as mirrored in the Palace Library catalogue composed at the end of Han by Lyou Syang) was expressed largely in terms of exegesis of those texts.

These are elementary facts, known to every student, including every second-year nonnative student, of early China. To ask, as though it presented a serious challenge, a question which is abundantly answered by elementary facts, is to trifle with the review process.

For a schematic idea of how we see the text formation process as operating, across the whole ideological spectrum and in the "exciting" conditions of political life in pre-Imperial China, consider TOA p4-5. An overview chart of similar type, though not including the Han part of the story, is posted elsewhere at this site. Some elements of that proposal, and some of the problems which that proposal for the first time explains, will be familiar to most readers of this page. A table of text events reaching into Han is available in our Prospects publication of 1994.

These conditions of state support or suppression, we conclude, along with questions of what is of vital intellectual concern in the elite culture itself, at a given moment in time, are among the conditions under which Warring States texts, and Han texts, either flourished or declined; either lived or died.

We think the resulting general picture is highly plausible. Details and local exceptions of course remain to be worked out. Our main effort is devoted precisely to working them out. But there is nothing in the overview so far that needs to disquiet, or even surprise, a knowledgeable reader.

[The Response resumes on the following page]:

To Response Part 2

22 December 2001 / Contact The Project / Return to Responses Page