13 Ganz

1
Chinese Science
Number 13
1996
An annual journal dedicated to the study of traditional and modern East Asian
science, technology, and medicine in the Chinese tradition
Editor: Benjamin A. Elman
Center for Chinese Studies, University of California, Los Angeles
Los Angeles, California 90095-1487, USA
FAX: (310) 206-3397
E-Mail: Elman@history.ucla.edu
Managing Editor: Richard Gunde, Center for Chinese Studies, UCLA
Production Editor: Leslie Evans, Center for Pacific Rim Studies, UCLA
EDITORIAL COMMITTEE
Mario Biagioli, History of Science, Harvard University
Francesca Bray, Anthropology, University of California, Santa Barbara
Christopher Cullen, History, SOAS, London University
Joseph W. Dauben, History, Lehman College, CUNY
Keizo Hashimoto, History of Science, Kansai University
Yi-Long Huang, Institute of History, National Tsing Hua University
Catherine Jami, Centre National de la Recherche Scientifique, Paris
Dieter Kuhn, Institut fr Sinologie, Universitat Wrzburg
Angela Ki Che Leung, Institute for Social Sciences and Philosophy, Academia Sinica,
Taipei
Dun Liu, Institute for the History of Science, Academia Sinica, Beijing
Boying Ma, History of Medicine, Shanghai Medical University
Georges Mtaili, Musum National dHistoire Naturelle, Paris
Willard Peterson, East Asian Studies, Princeton University
Laurence Schneider, History, Washington University, St. Louis
Nathan Sivin, History and Sociology of Science, University of Pennsylvania
Sharon Traweek, History of Science, UCLA
Donald B. Wagner, Copenhagen
Norton Wise, History of Science, Princeton University
THE SCOPE OF CHINESE SCIENCE
Copyright 1996. Chinese Science is published by the International Society for the
History of East Asian Science, Technology, and Medicine. Numbers 110 (1975-92)
were published irregularly. Beginning with Number 11 (1993-94), Chinese Science has
been published annually at the University of California, Los Angeles, sponsored by the
International Studies and Overseas Programs office and the Centers for Chinese Studies
(Editorial Office) and Pacific Rim Studies (Production Office).
This journal includes studies of science, technology, and medicine in early and contem-
porary East Asia. Any study based on research in Chinese primary sources or artifacts (or
Korean or Japanese sources or artifacts if the theme is not exclusively Korean or Japa-
nese), that throws light on the work of scientists, physicians, and technologists in the
Chinese tradition will gladly be considered.
SUBMISSIONS
Potential contributors are encouraged to correspond with the editor before submitting
manuscripts in order to ensure that their work falls within the purview of Chinese Sci-
ence and to simplify the preparation of final copy. All contributions except invited re-
views are refereed. The editor does not referee. Both Pinyin and Wade-Giles transcrip-
tions are accepted. Guidelines on the preparation of manuscripts and on submission of
manuscripts in the form of computer files are available on request.
SUBSCRIPTIONS
SUBSCRIPTIONS (three issues), post paid anywhere (surface mail): Individual rate:
FF330.00 or US$60.00. Students or retired: FF220.00 or US$40.00. Institutions:
FF410.00 or US$75.00. SINGLE BACK ISSUES: FF160.00 or US$30.00. Note: Mem-
bers of the Society (which includes all individuals holding a subscription) may purchase
back issues at a special rate: FF 110.00/US$20.00 for single issues. An entire set of back
issues (nos. 112) is available to subscribers at FF 1320.00/US$240.00. Send orders or
inquiries to: ISHEASTM/SIHSTMAO, Institute dExtrme-Orient, 52 rue du Cardinal
Lemoine, 75231 Paris Cedex 05, France. Fax: 33/1/44271854. Subscriptions will remain
in force until terminated, and will be billed in advance for each three issues. Subscribers
are requested to notify the editorial office of changes in address, and to pay promptly the
invoice enclosed in issues for which the subscription is about to expire.
Payment of subscriptions also enrolls individuals (including students and retirees) as
members of the Society.
NOTICE TO LIBRARIES
Libraries are reminded that we charge a fee (FF 125, US$25) for processing forms, in-
cluding claims against issues not yet published. Please note that Chinese Science is pub-
lished annually beginning with Number 11.
BACK ISSUES
Issues of Chinese Science normally remain in print for three or four years. At the mo-
ment sufficient copies of Numbers 512 are still available. Only a few copies of Num-
bers 14 remain. Out-of-print issues can be supplied in unbound photocopy on special
request.
ISSN #0361-9001
Contents
A FESTSCHRIFT FOR NATHAN SIVIN, Part II
Editors Introduction 5
Finite and Infinite in Greece and China
GEOFFREY LLOYD 11
The Development of Hindu-Arabic and Traditional Chinese Arithmetic
LAM LAY YONG 35
The Use of the Twenty-eight Xiu as a Day-Count in Early China
MARC KALINOWSKI 55
The Evolution and Decline of the Ancient Chinese Practice of Watching
for the Ethers
HUANG YI-LONG AND CHANG CHIH-CHENG 82
Medicine and the Changes are One: An Essay in Divination Healing
with Commentary
JUDITH FARQUHAR 107
The Published Writings of Nathan Sivin: An Annotated Bibliography 135
REVIEWS
Hashimoto Keiz, Catherine Jami, and Lowell Skar (eds.), East Asian Sci-
ence: Tradition and Beyond
reviewed by Yuan-ling Chao 143
5
Editors Introduction
Number 13 (1996) of Chinese Science represents the third issue of the journal
produced at UCLA. More importantly, however, it represents the second volume
of solicited festschrift articles in honor of Nathan Sivin, the founder and editor
of the journal from 1975 to 1992, first at MIT and then at the University of
Pennsylvania. In this issue we are pleased to present two comparative essays on
ancient Chinese mathematics and mathematical thought, one by Geoffrey Lloyd
and the other Lam Lay Yong, an article on ancient Chinese calendrical studies
by Marc Kalinowski, an account of late imperial developments in traditional
Chinese physics by Huang Yi-Long and Chang Chih-cheng, and finally a
thoughtful essay on traditional Chinese medicine by Judith Farquhar.
The current issue concludes the festschrift for Nathan Sivin, begun in last
years issue which was focused on Technology and Culture in Chinese His-
tory. The present issue also demarcates more fully the many fields in the his-
tory of Chinese science, medicine, and technology to which Nathan Sivin has
made so many lasting contributions. For the members of the International Soci-
ety for the History of Chinese Science, Technology, and Medicine and for our
subscribers at large, we have included an annotated bibliography of Nathan
Sivins writings in these diverse research areas, which Richard Gunde, our
Managing Editor, has compiled and annotated in his usual professional way.
Fortunately, two recent collections of Sivins articles (some revised, some new)
entitled Science in Ancient China (#12 in the bibliography), and Medicine, Phi-
losophy, and Religion in Ancient China (#13) appeared in 1995 and now make
Nathans oeuvre more accessible than in the past.
Reviewing his impressive list of publications, I am reminded of the breadth
of Nathan Sivins scholarly writings, which are informed by meticulous histori-
cal research, personal experiences in the field, and a scientific acumen that goes
beyond the normal boundaries of our self-restricted domains between the hu-
manities, social sciences, and natural sciences. The campus maps that adminis-
tratively demarcate our colleges and universities into discrete intellectual zones,
sectors, and buildings, which are then subdivided further into conceptually dis-
crete floors for departments in those buildings, are rarely evident in Sivins aca-
demic work. As a result, his reading audience has included scientists, historians,
anthropologists, and physicians, in addition to several generations of graduate
students anxious to educate themselves by reading the best research available in
Chinese studies.
Sivins pioneering studies of ancient Mohist optics (with A. C. Graham) and
its penchant for geometry (when the field assumed that the Chinese had no ge-
ometry), early imperial Chinese mathematical astronomy and the demise of the
6 Chinese Science 13 (1996)
cosmos (where we learned that Chinese calendrical specialists were more than
just ideological hacks and political opportunists), medieval Chinese alchemy
and its religious meaning (where we learned that Daoism was more than just the
arm-chair philosophy of Laozi and Zhuangzi), Jesuit tactics in late imperial
China and their manipulation of Copernicus (when we had thought the Jesuit
agenda was premised on scientific interchange and not Catholic dogma), and the
role of ritual practice in curing by doctors and healers in Chinese popular culture
(when we thought that medical efficacy was simply the victory of the right
drugs or the most advanced medical technology) all move easily, but not effort-
lessly, across the technical boundaries of modern mathematics, chemistry, medi-
cine, and astronomy to the social, political, and cultural systems of meaning that
informed those fields in China before they broke free in the early twentieth
century, for good and bad, from their premodern conceptual moorings.
In the annotated bibliography, there are important biographies (my own fa-
vorites are on Shen Gua and Wang Xishan) in Western-dominated dictionaries
of science, accounts of Chinese alchemy in encyclopedias of religion and in
tomes on time, and thoughtful if ironic reviews of the scientific revolution de-
bates of Why in Europe? and Why not in China? in conference volumes,
various festschrift for others, and edited volumes. And of course, there are the
important collaborations between Nathan Sivin and so many other pioneers in
the field, which for example, enabled Sivin to publish his work on alchemy with
Joseph Needham and others in one of the greatest collectanea (congshu ) of
our century, Science and Civilization in China. Sivins collaborative work with
Shigeru Nakayama was highlighted in last years issue (see Nakayama,
Working With Nathan Sivin: Four Decades), and his current partnership with
Geoffrey Lloyd on science in China and ancient Greece has already led to a
number of seminars, talks, and articles (see one in this issue) comparing
Warring States China to City-state Greece without the Eurocentric blinders
that have constricted all such earlier attempts.
From his first book on Chinese alchemy published in 1968, to his study of
Han mathematical astronomy in 1969, and his 1972 co-edited volume (with
Nakayama) on Chinese Science, Sivin began a publishing endeavor that has had
international significance. But equally important, I think, Sivins work forced
American scholars of China for the first time to take seriously the history of
science there. We forget the degree of skepticism that Joseph Needhams proj-
ects initially had provoked in the 1950s and 1960s. Many in Europe and Amer-
ica had dismissed the great embryologists foray into the history of Chinese sci-
ence as a dead-end, a project they felt revealed Needhams wishful thinking
about China.
The consensus in those days, for example, drew on heroic accounts of the
rise of Western science to demonstrate that China had had no science worthy
of the name. Both Western scholars and Westernized Chinese scholars and sci-
entists had so essentialized European natural studies into a universalist ideal that
when Chinese studies of the natural world, her rich medieval traditions of al-
Editors Introduction 7
chemy, or pre-Jesuit mathematical and astronomical achievements in China
were discussed, they were usually treated dismissively and tagged with such
epithets as superstitious, prescientific, or irrational to contrast them with
the triumphant objectivity and rationality of the modern sciences.
We need not use this more happy occasion of a festschrift to beat a dead
horse. Some of us can still recall, however, the shrill orientalist-like refrains
about China that bedeviled a generation of our best post-war graduate students.
Many were so convinced that because China had had no industrial revolution
and had never produced capitalism, therefore the Chinese could never have pro-
duced anything like science. Few besides Needham, his collaborators, and Sivin
ever stopped to consider what the rich archives in Taiwan, China, and Japan
might yield if someone bothered to go through them.
If we recall the situation as it was then, something our current graduate stu-
dents are fortunately spared, then we can see how significant the contributions
of Joseph Needham in England and Nathan Sivin in the U.S. were to beginning
the unraveling of a consensus of poorly informed scholarly opinion about not
only the failed history of Chinese science but also about the victorious his-
tory of European science. Both histories were pieces in the larger global narra-
tive of Western success and Chinese failure that was taken as given. As the evi-
dence of a rich tradition of natural studies in many diverse fields accrued in vol-
ume after unrelenting volume of the Science and Civilization in China project
after 1954, it became harder and harder to gainsay it all as superstition or irra-
tionality or just plain inductive luck, though some still tried.
Nathan Sivin and others in the 1970s slowly persuaded us that as humanists
and social scientists we were foolishly overlooking Needhams discovery of
China as one of the great non-Western domains of human wisdom (along with
the Islamic and Indian traditions) in all aspects of science, technology, and
medicine. As Chinese science grew in respectability among the younger schol-
ars who now make up the International Society worldwide, the romanticized
story of European science, whether capitalist or socialist in genre, slowly unrav-
eled under the onslaught of Thomas Kuhn and his historicist successors. No
longer reified and automatically placed on a priestly/scientific pedestal as the
successor to our pre-1900 religions that had all failed, the story of Western sci-
ence became more complicated, more tragic, and more tense. The boyhood
dreams of invention among Europeans and the shrieks of eureka after new
discoveries were revisited. Newtons mystical fascination with alchemy, Coper-
nicus devout religiosity that informed his astronomy, and the deadly serious but
ungentlemanly competitive nature of priority in science and invention all
challenged earlier politically neutral and culturally neutered fictions about the
scientific revolution so long taken for granted. Scientists had been mostly
men, after all, each caught up in a world of social, political, economic, and cul-
tural concerns, which they could never completely leave out of their scientific
equations. Their equations, moreover, had yielded political, social, economic,
and cultural consequences that few of them had foreseen while toiling in the
laboratory.
The history of Western science was decisively refracted when it was viewed
through the lenses of Chinese natural studies, technology, and medicine. I sus-
pect that the history of modern science can never again be what it once was, the
history of a select group of early-modern Europeans who supposedly discovered
the natural laws of the universe and then passed them on with good will and
cheer in the name of progress to the rest of the world. The tragic uses of tech-
nology in two world wars and the post-colonial nightmares in the aftermath
make that narrative difficult to maintain uncritically.
In closing, let me stress a more positive legacy of the demystification of
modern science. The study of Chinese science, technology, and medicine, to
which Nathan Sivin contributed so much, did not take such demystification as
its goal. Rather, Sivin and others granted fundamental priority to research in
areas where the received wisdom was suspect, based as it was on careless
speculation about banal generalities. Sivin like Needham searched beneath the
surface of self-satisfied discourses about science and the self-serving appeals
to Greek deductive logic upon which they were conveniently based. As a conse-
quence of such pathbreaking research, humanists and social scientists at the end
of the twentieth century have had to take seriously the history of science as an
account of a domain of learning and practice that has influenced all peoples and
civilizations since ancient times. The largest archive of premodern records for
the study of nature remains in China. As Nathan Sivin has shown us, by better
understanding Chinese science, technology, and medicine, we are more percep-
tive about ourselves as well. We still have a long way to go.
Benjamin A. Elman, Editor
A FESTSCHRIFT FOR NATHAN SIVIN
Part II
[Blank]
11
Chinese Science 13 (1996): 1134
Finite and Infinite in Greece and China
Geoffrey Lloyd
[Geoffrey Ernest Richard Lloyd is Master of Darwin College, Cambridge; since
1983 he has also been Professor of Ancient Philosophy and Science. He was
born in 1933 and educated at Charterhouse and Cambridge (Kings College,
gaining BA, MA, PhD). He has published extensively in the area of Greek phi-
losophy and science. His most recent books are The Revolutions of Wisdom
(1987), Demystifying Mentalities (1990), and Methods and Problems in Greek
Science (1991).]
* * *
re the notions of finite and infinite used in similar or in different ways in
ancient Greek and Chinese thought? It is typical of a certain nave ap-
proach to the problems of comparing Greek and Chinese science to ac-
cept the question posed in such brutally simplistic and vague terms, and then to
proceed to answer it by developing either the thesis of a basic similarity or the
antithesis of a fundamental contrast between the two ancient societies. Indeed
this may happen without due consideration being given to how the term
infinite itself is to be understood, whether in the loose sense (where it stands
merely for the immense or the indefinitely large) or in a strictly defined mathe-
matical one (e.g., uncountability, where it is a property of infinite sets that they
are neither increased by addition nor decreased by subtraction). Further ambi-
guities in the Greek terms will be examined in due course.
The aim of this paper is to take this topic as illustrative of the tasks of com-
parative history of science. As announced in 1993 in the Newsletter for the His-
tory of Chinese Science, Professor Sivin and I have embarked on an ambitious
and wide-ranging exploration of Chinese and Greek science from circa 300 to
circa +200that is, down to the period when Buddhist and Christian influences
came to be a major factor in China and Greece respectively. But both of us are
very conscious of how superficial most comparative work to date has been.
It cannot be just a matter of recording certain similarities or differences be-
tween the ideas produced by these two ancient civilisations, let alone one of
chalking up points on either side, as in those studies that seem chiefly preoccu-
A
pied by the question of who did or thought what first. Why the Chinese should
have produced the ideas they did, and why again the Greeks theirs, have to be
addressed and those questions cannot begin to be answered without a sustained
analysis of the conditions under which scientific work was conducted.
Why were the problems defined in the ways they were (often very different
ways from those we expect if we do not liberate ourselves from the straitjacket
of our categories of mathematics, physics and so on)? What expectations
were entertained for an adequate solution? Whom were the scientists trying to
convince and how did they hope to do so? What was the institutional framework
within which they worked? How were they recruited and organised (if they were
organised)? What were their motivations or aspirations, what, indeed, their
sources of livelihood: how did they make a living, or did they need to?
In our collaborative explorations Professor Sivin has begun a radical reex-
amination of the Chinese materials, while I have been concentratingas will be
obvious from my essay hereon the Greek side. Yet each of us has found that
the juxtaposition of one anothers data brings to light problems that risk re-
maining invisible to anyone working within just a single tradition. New issues
come to the fore, especially to do with the relations, in each case, between the
science produced and the society which produced it. Our hope is that the agenda
we have set ourselves provides a firmer basis not just for doing comparative
history of science, but for doing history of science.
My own forays into comparative studies are profoundly indebted to Profes-
sor Sivins inspiration and willingness to collaborate, and this contribution to the
articles in his honour can serve as no more than a very meagre token of what I
feel I owe to him.
The finite and the infinite are, potentially, a vast topic, and of course there
can be no question of attempting a comprehensive analysis here. Rather I shall
focus on how to define the problems to be investigated and the most fruitful line
of attack to be adopted in their investigation. This is all the more important in
that much of what passes as orthodoxyand not just on the Greek side, but also
on the Chineseis open to challenge and certainly has to be challenged if we
are to fulfil the primary task of clarifying the issues.
We may begin with a critical examination of a number of sweeping generali-
sations that are to be found in the literature, starting with the simplistic theses I
mentioned at the outset. I shall develop an argument in three stages. In the first
the emphasis will be on the imagined contrasts between Greece and China, and
in the second the focus will shift to points of comparison. This will clear the
ground for an attempted radical redefinition of the explananda in my final sec-
tion, where I shall argue that it is not so much this or that concept or use of the
finite or infinite that is important: rather the interest lies in the light the ways
they are used can throw on the aims, presuppositions and interactions of Greek,
Geoffrey Lloyd: Finite and Infinite in Greece and China 13
and Chinese, thinkers and on the nature of the philosophical exchanges culti-
vated in each society.
I
Thus the first stage of the argument might begin with a set of common assump-
tions about contrasts. Some might claim that it is obvious that there is a massive
difference in emphasis between ancient China and Greece in that the latter toler-
ated, and even cultivated, the infinite in a whole lot of contexts where the Chi-
nese either deliberately excluded it or, more often, never even considered it.
There were plenty of ancient Greeks who considered the universe to be spatially
infinite and quite a few who believed that there was an infinity of worlds. Both
ideas come in different forms. Thus for the atomists, both matter (the atoms) and
the void (where matter is not) are infinite. Again for the Stoics, who denied void
within the cosmos, there is infinite space outside it.
1
Again, infinite worlds were sometimes conceived as succeeding one another
temporally, though others held that at any given time there is an unlimited num-
ber of worlds. In quite what form infinite worlds were maintained by some of
the earlier Presocratic philosophers, such as Anaximander, is disputed.
2
In the
4th century a pupil of Democritus named Metrodorus is said to have claimed
that it is as absurd to believe that one ear of corn would be produced in a great
plain as to hold that just one world would be, in the infinite void. That just sug-
gests a very large number of worlds. But our source goes on to make the argu-
ment that they are indeed infinite in number, since the principles, aitia, from
which they are formed (that is, atoms and the void) are infinite.
3
The topic generated a good deal of speculation about what the other worlds
may be like. In the 5th century, Anaxagoras, who held that matter is infinitely
divisible, believed that the cosmos contained an immense variety of seeds, and

1
See, for example, Stobaeus I 161.8ff, Sextus Empiricus, Against the Professors X
3, Simplicius, On Aristotles On the Heavens 284.28ff, translated in A. A. Long and
D. N. Sedley, The Hellenistic Philosophers, vol. 1, Translations of the Principal
Sources, With Philosophical Commentary (Cambridge: Cambridge Univ. Press, 1987),
section 49, pp. 294ff. See further below.
2
The issue is discussed at some length by W.K.C. Guthrie, A History of Greek Phi-
losophy, vol. 1, The Earlier Presocratics and the Pythagoreans (Cambridge: Cambridge
Univ. Press, 1962), pp. 106ff, who sets out the principal ancient Greek and Latin sources.
Cf. also the sceptical views expressed by G. S. Kirk, J. E. Raven and M. Schofield, The
Presocratic Philosophers (Cambridge: Cambridge Univ. Press, 1983), pp. 122ff.
3
Aetius I 5 4, quoted by Hermann Diels and Walther Kranz, Die Fragmente der
Vorsokratiker (3 vols., 7th edition, Berlin: Weidmann), vol. 2, section 70 A 6, pp. 231f.
in fragment 4 he describes the formation of humans and other animals in situa-
tions other than the one we are familiar with:
And [we must suppose that] men have been formed and all the other animals that
have life; and the men have settled cities and cultivated fields as with us, and sun
and moon and the rest as with us; and the earth grows all sorts of produce for
them, the most useful of which they gather into their houses and use. This is my
account of the separating off, that it must have taken place not only where we
live, but elsewhere also.
4
The contrast, here, between as with us and elsewhere is ambiguous: Anax-
agoras could be thought to refer not to another world, but just to another part of
the earth. But Democritus, as reported by Hippolytus, leaves no doubt that he
has other worlds in mind:
There are innumerable worlds of different sizes. In some there is neither sun nor
moon, in others they are larger than in ours and others have more than one. These
worlds are at irregular distances, more in one direction and less in another, and
some are flourishing, others declining. Here they come into being, there they die,
and they are destroyed by collision with one another. Some of the worlds have
no animal or vegetable life nor any water.
5
Spatial infinity was sometimes maintained with an argument, versions of which
were put forward by both the Stoics and the Epicureans,
6
though the original
may go back to pre-Aristotelian Pythagoreans. Archytas, in the 4th century, is
reported to have asked: If I were at the extremity, say at the heaven of the fixed
stars, could I stretch out my hand or staff or could I not? It would be absurd to
think you could not do so. But that means that there will be either body or place
outside the supposed extremityan argument that can be repeated for whatever
extreme point is postulated.
7

4
Translation from W.K.C. Guthrie, A History of Greek Philosophy, vol. 2, The
Presocratic Tradition from Parmenides to Democritus (Cambridge: Cambridge Univ.
Press, 1965), p. 314. Anaxagoras was also famous, or rather notorious, among the
Greeks, for having proclaimed that the sun and moon themselves are incandescent
stones: the ancient evidence is set out by Guthrie at pp. 307ff.
5
Hippolytus, Refutation of the Heresies I 13 2, in Diels and Kranz, Die Fragmente
der Vorsokratiker, vol. 2, section 68 A 40, p. 94. Translation from Guthrie, A History of
Greek Philosophy, vol. 2, p. 405.
6
See Simplicius, On Aristotles On the Heavens 284.28ff, Lucretius, On the Na-
ture of Things I 958ff, translated in Long and Sedley, The Hellenistic Philosophers, vol.
1, at pp. 295 and 44f respectively.
7
See Simplicius, On Aristotles Physics 467.26ff, in Diels and Kranz, Die Frag-
mente der Vorsokratiker, vol. 1, pp. 430f, where Simplicius claims to be drawing on the
Temporal infinity, in turn, was even more common, in that some who
thought the world spatially finite held nevertheless that it is eternal. Here too
different positions were adopted. Some held that the universe has neither begin-
ning nor end, others merely that it has no end. Aristotle took the first view and
argued that any supposed beginning to the cosmos must itself have a prior be-
ginning, leading to an infinite regress of beginnings and so to the denial of any
first one. Plato presents the second option, in his cosmological dialogue the Ti-
maeus, where it is emphatically stated that the cosmos came to be, though it will
not be destroyed. Already in antiquity, however, some thought that that state-
ment that the cosmos came to be was only introduced for the sake of the narra-
tive exposition in the dialogue, and did not represent Platos own belief.
8
So far I have taken cosmological examples. But Greek mathematics provides
many more examplesso it could be arguedwhere the infinite is accepted and
used readily enough. This could be said to be the case in arithmetic, with the
number series, including such famous arguments as Euclids proof of the infin-
ity of primes (Elements IX 20), and in geometry, where the standard Greek view
is that geometrical space is infinitely divisible.
Some of these Greek ideas no doubt find parallels in ancient Chinese
thought. But the classical Chinese were evidently less inclined to try to prove a
strict spatial infinity and less prone to speculate about an infinite number of
worlds separated from ours in either space or time or both. So at a first stage it
might seem that the contrasts between Chinese and Greek thought on this topic
are overwhelmingand to those to whom they do, that no doubt would be the
moment when speculative explanations would be invoked to account for those
differences. But that would be premature.
II
At a second stage of reflection much of the above picture, both of the Greeks
and of the Chinese, has to be modified. Let me deal first with Greek physics,
then with Greek mathematics, before turning to some of the Chinese materials.
Both the belief in spatial infinity and that in infinite worlds can, it is true, be
found. But two points need to be made even in this regard. First, there is a lexi-
cal point that is relevant to the whole of our analysis of Greek thought on the

4th century historian Eudemus. Translation from Guthrie, A History of Greek Philoso-
phy, vol. 1, p. 336, who discusses other Pythagorean ideas on time and the unlimited.
8
For Aristotles own view, see Physics VIII chaps. 5 and 6, e.g. 258b10ff, On the
Heavens I chaps. 1012. For Platos statement, see Timaeus 28b, and for the ancient
dispute over that, see Aristotle, On the Heavens 279b32ff and the other texts collected
and discussed by A. E. Taylor, A Commentary on Platos Timaeus (Oxford: Clarendon
Press, 1928), pp. 67ff.
subject. The Greek word often translated infinite is apeiron, which combines
a negative alpha with the term peras meaning limit. But apeiron often carried
the weaker sense of indefinite. So maintaining apeiroi kosmoi sometimes im-
plied no more than a belief in a large number of worlds, an indefinite number in
fact. However it has also to be said that in other contexts, infinite in the strict
sense is the right translation of apeiron, and not just in some of the mathemati-
cal work. Archytas argument, for instance, suggests that the notion of an ex-
treme point is self-contradictory.
Secondly, it might be argued that those who made most use of the notion of
the infinite in cosmological and physical speculation were the atomists, Leucip-
pus and Democritus in the 5th century, Epicurus and his followers in the 4th
and later centuries. But atomism, it might further be claimed, was never the
dominant tradition in Greek philosophy and science, which is represented,
rather, by Plato and Aristotle in the mid 4th century, and then by the Stoics,
from the late 4th century onwards. Plato, Aristotle, and such Stoics as Zeno of
Citium and Chrysippus disagreed on a number of issues, but all were continuum
theorists and all teleologists.
The assumption that many of the Greek cosmologists made was that the
sphere of the fixed stars marked the limits of the cosmos and this was shared by
most of the astronomers. Otherwise why would Archimedes have tried to cal-
culate how many grains of sand it could hold? Of course Archimedes interest is
in developing a mathematical notation to express very large numbers, and it is
for that purpose that he conducts his thought experiment with the grains of sand
idea. Indeed he sets out by denying both that the grains of sand would be infinite
and that they exceed any expressible number.
9
But practising astronomers too,
down to Ptolemy in the +2nd century and beyond, assumed that the so-called
fixed stars bound the cosmos, and while that did not rule out the Stoic view of
infinite space outside the cosmos, for someas for Aristotlecosmos and uni-
verse are alike bounded and coterminous.
The first-stage presentation of Greek thought on the infinite might be criti-
cised as seriously misleading in that many Greek philosophers, including some
of the most prominent ones, fought shy of the notion of the infinite in many of
its forms. Aristotle might be cited as a prime case in point.
10
True, he allows, as

9
Archimedes, Sand-Reckoner, II 216.2ff, translated in T. L. Heath, The Works of Ar-
chimedes Edited in Modern Notation, with the Method of Archimedes (Cambridge: Cam-
bridge Univ. Press, 1912), pp. 221ff.
10
This was the view expressed by Melbourne G. Evans, The Physical Philosophy of
Aristotle (Albuquerque: Univ. of New Mexico Press, 1964). Contrast, for example,
Jaakko Hintikka, Time and Necessity: Studies in Aristotles Theory of Modality (Oxford:
Clarendon Press, 1973); Wilbur R. Knorr, Infinity and Continuity: The Interaction of
Mathematics and Philosophy in Greek Antiquity, in Norman Kretzmann (ed.), Infinity
and Continuity in Ancient and Medieval Thought (Ithaca, N.Y.: Cornell Univ. Press,
we have noted, that the universe is temporally infinite. But he opts firmly for a
spatially limited universe. The heavens form a sphere with the heavenly bodies
moving with eternal circular motions in the superlunary region. The earth is at
the centre of the entire system, and there can, in this picture, be no question of a
plurality of worlds. In On the Heavens Aristotle first devoted three chapters to
showing that there can be no infinite body, neither simple nor composite,
11
and
then proves that there cannot be more than one world.
12
This he does in part on
the basis of his analysis of natural motions, which he claims must be either rec-
tilinearto or from the centre of the universeor circular. That consideration
also provides him with one of his arguments to show that the earth itself is
spherical, but in this context he is concerned to establish that there can be only
one centre and only one circumference to the world as a whole.
13
Moreover, so far as the infinite in mathematics goes (to which I shall be re-
turning) Aristotles view is that the mathematicians do not need the number se-
ries actually to be infinite: all they needso he believed, wrongly, as has re-
cently been claimed
14
is the idea that that series can be indefinitely extended.
Even when it comes to the geometrical and physical applications of the notion
of the continuum, where Aristotle is clear that it is a matter of infinite divisibil-
ity, the infinite in question, there, is, in both cases, potential, never actual. You
can cut a line or a stick wherever you like. But Aristotle resists the idea (that he
nevertheless records) that it is sensible to think of an actual division everywhere
of what can potentially be divided anywhere.
There is a famous argument in On Coming-to-be and Passing Away that may
give an atomists point of view.
15
Suppose that magnitudes are infinitely divisi-
ble. Then let them be divided everywhere. What have you left? You cannot be
left with nothing, nor with what is dimensionless such as a point (since in both
cases summing nothings, or what is dimensionless, gives nothing). Nor yet can
you say you are left with a magnitude, because ex hypothesi you are supposed to
have carried out the division everywhere. That may have been an argument that
led the atomists to the conclusion that magnitudes (of any sort, or at least physi-
cal ones) must be constituted from indivisible atomic quanta.
16
But Aristotle
simply denies that it could ever be the case that a magnitude is divided

1982), pp. 11245; Richard Sorabji, Matter, Space and Motion (London: Duckworth,
1988).
11
Aristotle, On the Heavens I chaps. 57, 271b1ff.
12
Aristotle, On the Heavens I chaps. 89, 276a18ff.
13
Aristotle, On the Heavens 276b7-21, cf. II chap. 14, 297a8ff, on the sphericity of
the earth.
14
Hintikka, Time and Necessity, pp. 118ff, on Aristotle, Physics 207b27ff.
15
Aristotle, On Coming-to-be and Passing Away 316a14ff.
16
The answer to the question of how far Leucippus and Democritus had been antici-
pated by earlier views postulating atomic quanta depends on the interpretation of the
evidence for Zeno and for the Pythagoreans: see further below, Note 44.
everywhere. That is just something potential, never actual, even though there
is some difficulty in squaring that view with Aristotles insistence elsewhere that
what is potential must be thought of as actualisable and that must mean at some
point actualisedthe idea that has been dubbed the principle of plenitude.
17
A similar potentially tricky deployment of the distinction between potential
and actual infinities appears also in Aristotles critique of Zeno. In the so-called
stadium argument,
18
reported by Aristotle at Physics 233a21ff and 239b9ff,
Zeno had claimed (1) that to cross a stadium a runner must touch infinitely
many points in the ordered sequence 1/2, 1/4, 1/8 and so on, traversing on each
occasion half of the total space remaining to be crossed; (2) that it is impossible
to touch infinitely many points in a finite time; therefore (3) the runner cannot
reach his goal. But Aristotle replies that the finite time itself, like the space to be
traversed, is infinitely divisible: so (2) can be denied, and the conclusion (3)
also. But while he believes that that is a sufficient argument ad hominem, to
answer a questioner who raises the point at issue in (2), he comes back to the
problem in a later passage, Physics 263a11ff, that reveals his continuing worries
over the distinction between potential and actual infinities:
So when someone asks the question whether it is possible to traverse infinite
things, either in time or in distance, we must reply that in a way it is but in a way
it is not. For if they exist actually, it is not possible, but if potentially, it is; for
someone in continuous movement has traversed infinite things incidentally, not
without qualification; for it is incidental to the line to be infinitely many halves,
but its essence and being are different.
19
Aristotles difficulty is as before: the continua of space, time and motion are all
infinitely divisible, for that is what it is to be a continuum. Yet they can be infi-
nitely divisible only in potentiality. That move gives Aristotle his reply to those
arguments of Zeno, and yet that leaves the relationship between the potential
and the actual, in these cases, unclear, given Aristotles usual insistence that the
potential corresponds to what can indeed be actualised.
But as a source of qualifications to the contrasts suggested between China
and Greece in my first-stage argument, the Stoics are just as important as Aris-
totle. Outside the cosmos, as noted, they hold that there is infinite void. But so
far as the cosmos itself goes, they too hold that there is just the one, even though

17
Arthur O. Lovejoy, The Great Chain of Being: A Study of the History of an Idea
(Cambridge, Mass.: Harvard Univ. Press, 1936); cf. Hintikka, Time and Necessity, chap.
5.
18
This analysis is based on that of Kirk, Raven and Schofield, The Presocratic Phi-
losophers, pp. 269ff. The so-called Achilles argument, reported by Aristotle at Physics
239bff, similarly depends on the assumption that space is infinitely divisible.
19
Physics 263b3ff, translation from Kirk, Raven and Schofield, The Presocratic
Philosophers, p. 271.
it is subject to periodic conflagration.
20
Furthermore they are continuum theo-
rists in a strong sense. The cosmos is finite and there is no void within it. Unities
come in different forms (the unity represented by a living organism is greater
than that of inanimate substances) but everything forms part of a single inter-
connected whole. That different parts of that whole resonate with one another is
the basis of their doctrine of sumpatheia, sympathy, which is made to do some
explanatory work in a number of different contexts, in biology, in medicine and
in physics.
21
More remarkably still, from the point of view of those looking for
parallels between Greece and China, the Stoics held that pneuma permeates the
whole cosmos: as such it could be thought to be the nearest the Greeks got to
qi.
The very fact that Stoic pneuma is so badly misrepresented by their oppo-
nents (Plutarch, Galen, Alexander of Aphrodisias, all the way down to Sim-
plicius and Philoponus in the +6th century) is interesting and indicative. The
questions that the opponents keep pressing include how many elements there
are, what the relation between fire and the rest is, how breath can pervade eve-
rything, and what the relation is between its different modalities or manifesta-
tions, in inanimate substances, in plants and in animals (tenor, phusis, psyche).
But it is clear that the Stoics worked rather with a notion of active and passive
principles, not with one of four static elements. At that point a general similarity
with yin and yang might suggest itself, even though the Stoics never developed
the system of correspondences with which those principles eventually came to
be associated in China.
22
Long and Sedley remark that, for the Stoics, four dis-
tinct elements . . . are not, as they are in Aristotles cosmology, permanent fea-
tures of the Stoic universe, but the basic qualifications of matter throughout the
duration of each temporally limited world-order.
23
That is correct in general
terms, although one may question just how far those elements are indeed
distinct. Even so, the more important difference from Chinese yin-yang is that

20
See, for example, Nemesius 309.5ff and other evidence collected by Long and
Sedley, The Hellenistic Philosophers, vol. 1, section 46, pp. 274ff, and section 52, pp.
308ff.
21
See, for example, Alexander, On Mixture 223.25ff and other texts discussed by
Long and Sedley, The Hellenistic Philosophers, vol. 1, section 47, pp. 280ff.
22
See, for example, Joseph Needham, Science and Civilisation in China, vol. 2,
History of Scientific Thought (Cambridge: Cambridge Univ. Press, 1956), pp. 253ff;
John B. Henderson, The Development and Decline of Chinese Cosmology (New York:
Columbia Univ. Press, 1984), chap. 1, pp. 158; Angus C. Graham, Yin Yang and the
Nature of Correlative Thinking (Singapore: Institute of East Asian Philosophies, 1986);
Angus C. Graham, Disputers of the Tao: Philosophical Argument in Ancient China (La
Salle, Ill.: Open Court, 1989), pp. 319ff.
23
Long and Sedley, The Hellenistic Philosophers, vol. 1, p. 286. The Stoic evidence
for elements and unities is collected at pp. 280ff.
they, yin-yang, are essentially functional and relational, and as such not basic
qualifications of matter.
24
Then there is a further feature of Greek philosophy that can be taken to sug-
gest a resistance to the notion of infinity at least in certain contexts. This is the
recurrent idea, found for example in Plato, Aristotle and Plotinus, that the infi-
nite is not intelligible: whatever has to be called apeiron is, as such, no proper
subject of inquiry and understanding. There are texts in Aristotle that say as
much in so many words. Thus in the Physics 187b8f he puts it that the infinite
qua infinite is unknowable: what is infinite in multitude or size is unknowable in
quantity, and what is infinite in form is unknowable in quality. He uses a simi-
lar argument in the Metaphysics 1036a8ff, 1037a27ff, to suggest that matter, as
indeterminate, is unknowable in itself, and again at Physics 207a14ff, 25ff, he
attacks Melissus for having asserted that the whole is infinite, repeating the
point that the infinite as such is unknowable.
Plato too equates what can be known with what is limited or determinate. In
a late and rather neglected dialogue, the Philebus, that draws on Pythagorean
ideas, especially in the fundamental contrast it uses between limit and the un-
limited, Plato makes it clear that the world as a whole, and its parts, can be un-
derstood only insofar as it, or they, manifest limit. True, there is also what is
there to be limited, the as-yet-to-be-determined, that of which the limit is the
limit. But it is the determinate or what has been determined that is there to be
known.
25
Finally the +3rd century neo-Pythagorean Plotinus was to go so far as to
make explicit what had been implicit already in the Pythagorean Table of Oppo-
sites reported by Aristotle in the Metaphysics 986a22ff, namely that the infinite
is evil, kakon (Enneads VI 6 1).
But it is not just Greek philosophy that provides evidence to suggest that
certain important qualifications need to be entered before we accept the idea that
Greek thinkers tolerated and used the notion of infinity with equanimity. Reser-
vations also need to be expressed with regard to our earlier, first-stage, charac-
terisation of Greek mathematics.
First it is well known that what is called, most misleadingly, the method of
exhaustion avoids infinite processes. The method, usually thought to have been
developed by Eudoxus in the 4th century, depends on the principle expressed
in Euclid Elements X 1: Two unequal magnitudes being set out, if from the
greater there be subtracted a magnitude greater than its half, and from that
which is left a magnitude greater than its half, and if this process be repeated
continually, there will be left some magnitude which will be less than the lesser

24
See Nathan Sivin, Yin Yang and the Five Phases, in G.E.R. Lloyd and Nathan
Sivin, Tao and Logos: Comparing Ancient Chinese and Greek Science (forthcoming).
25
Plato, Philebus 16c17a, 25a.
magnitude set out.
26
Used, as for example in Elements XII 2, in the investiga-
tion of the ratios of circles to their diameters, it allows the inscription of regular
polygons in a circle such that their area can be made to approximate as close as
one likes to the area of the circle. But the area is, precisely, never exhausted.
This gives a proof procedure that is rigorous, but involves no breach in the con-
tinuity axiom. There is no suggestion that the circle can be identified with the
infinite-sided regular polygon inscribed within it. The Greek preference for the
method of exhaustion is thus evidence both of their demand for rigour and of
their avoidance of infinite processes wherever possible.
When, as in Archimedes Method, the continuity axiom is breached at least
insofar as a curvilinear area is there treated as made up of the indivisible line
segments it contains, this is recognised as disqualifying the procedure as demon-
strative: it can be no more than heuristic. Archimedes is clearly on the defen-
sive, not just because of the use of indivisibles, but also because he treats plane
figures as having determinate centres of gravity and as balanced around a ful-
crum. Judged by the usual standards of Greek demonstrative rigour, the method
falls short. But it is equally remarkable that, if not demonstrative, it is at least
accorded heuristic status, for it was, after all, in conflict with the common Greek
assumption of the infinite divisibility of geometrical magnitudes.
Archimedes also provides us with most of our not very extensive evidence
for the summing of infinite series in Greek mathematics. The lack of a clear
distinction between converging and diverging series had caused difficulty al-
ready in the days of Zeno of Elea.
27
Nor perhaps should we be surprised that the
examples of Greek mathematicians confidently summing infinite series are
rare,
28
when we reflect that, expressed in Greek, the notion of taking the limit

26
Translation from T. E. Heath, The Thirteen Books of Euclids Elements (3 vols.,
2nd edition, Cambridge: Cambridge Univ. Press, 1926), vol. 3, p. 14. See Wilbur R.
Knorr, The Evolution of the Euclidean Elements (Dordrecht: D. Reidel, 1975), pp. 256f,
271ff, who argues that the stipulation more than half may suggest an origin in relation
to the anthyphairetic approach to incommensurables; cf. David H. Fowler, The Mathe-
matics of Platos Academy: A New Reconstruction (Oxford: Clarendon Press, 1987).
27
Thus in the arguments (Fragments 1 and 2) attacking the many on the grounds that
they must be both so small as to have no size and so large as to be unlimited
(Simplicius, On Aristotles Physics 141.8), there is an ambiguity in the term
unlimited. Most ancient commentators took it to mean indefinitely large. But that
had not been shown. Zeno had argued that there is an infinite number of parts: but the
sum of aconverginginfinite series may be finite. It is controversial whether Zeno
himself is under an illusion on the point, or whether it is rather just his opponents who
may be. But either way the conclusion could easily have been resisted by an appeal to the
distinction between converging and diverging series.
28
Thus Knorr, Infinity and Continuity, p. 125 n. 30, cited Archimedes, On the
Quadrature of the Parabola, Proposition 23, as unique: but that is to ignore other cases;
cf. Tohru Sato, A Reconstruction of The Method Proposition 17, and the Development
(peras) of any series that is limitless (apeiron) certainly ran the risk of appearing
flatly self-contradictory.
So much for some of the reservations that should be expressed with regard to
our initial characterisation of Greek thought. But equally on the Chinese side,
any thesis that would have it that Chinese thought in general was reluctant, even
incapable, of dealing with the infinite, encounters considerable counter-
examples. It is true that the evidence for some uses of infinity in Chinese
thought is late or unreliable or both. That applies especially to cosmology,
where many invocations of the notion of the infinite, such as, for instance, the
idea of infinite worlds, betray clear Buddhist influence,
29
and so fall outside the
period with which I am chiefly concerned. Again for the so-called xuanye,
cosmology, we have a mid +4th century source, Ge Hong, who reports the view
that the heavens are empty, immensely high and far away, and without bounds.
But the writer does not inspire confidence in that he states that the books that set
out this view were all lost, even though he claims to be reporting what a late
Han librarian, Qi Meng, said its earlier masters had taught.
30
However, in other contexts we are on firmer ground. In a variety of areas of
Chinese thought we have good evidence that recurrent procedures of different
types were used with some confidence. One such text comes in the paradoxes
reported towards the end of the Tianxia chapter (33) of Zhuangzi.
31
They begin
with a set of ten directly attributed to Hui Shi himself, and several of these evi-
dently challenge conventional notions of space and time. Thus the first states:
The ultimately great has nothing outside it. . . . The ultimately small has noth-
ing inside it, and the second: The dimensionless cannot be accumulated, yet
its girth is a thousand li.
32
The contexts of these remarks are not recorded either
here or in our other ancient evidence for Hui Shi, and modern scholarly inter-
pretations diverge radically.
33
But Hui Shis own paradoxes are then followed
by others, including: if a stick a foot long is cut in half every day, it will not be

of Archimedes Thought on Quadrature, Part II, Historia Scientiarum no. 32 (1987), pp.
75142 at pp. 114ff.
29
See, for example, Jacques Gernet, Space and Time: Science and Religion in the
Encounter Between China and Europe, Chinese Science no. 11 (1994).
30
Joseph Needham, Science and Civilisation in China, vol. 3, Mathematics and the
Sciences of the Heavens and the Earth (Cambridge: Cambridge Univ. Press, 1959), p.
219, cites Ge Hong from the Taiping yulan 2, 7a, and Jinshu chap. 11, 2a. I am grateful
to Nathan Sivin for stressing to me the difficulties of accepting this report.
31
Zhuangzi 33: 70ff.
32
Translation from Graham, Disputers of the Tao, p. 78.
33
With Needham, Science and Civilisation, vol. 2, pp. 19091, compare Jean-Paul
Reding, Les Fondements philosophiques de la rhtorique chez les sophistes grecs et chez
les sophistes chinois (Bern: Peter Lang, 1985), pp. 350ff, 435ff, and Graham, Disputers
of the Tao, pp. 78ff.
exhausted in ten thousand generations.
34
Once again we are at a loss for the
precise context, but what is beyond doubt, in this instance, is that we have a
clear appeal to a recursive bisection principle, similar in that respect at least to
those attested for Zeno of Elea.
Our evidence from the Chinese mathematical treatises is both richer and
more complete, even though, to be sure, not without its problems of interpreta-
tion. Thus in Liu Huis commentaries on the Nine Chapters recurrent procedures
are used both in the investigation of the area of the circle in chapter 1 and in that
of the volume of the pyramid in chapter 5.
35
In chapter 1 he first establishes that
the area of the circle is given by the formula, A = 1/2 c times 1/2 d, where A is
the area, c the circumference and d the diameter. He then proceeds to an ap-
proximation of the ratio between the circumference and the diameter (that is ).
In this a hexagon is first inscribed in the circle, then a dodecagon, and the pro-
cedure of doubling the sides of the inscribed polygon is then continued to give
increasingly accurate approximations to the area of the circle itself. In chapter 5,
similarly, Liu Hui proceeds by inscribing in the pyramid to be determined, fig-
ures that approximate closer and closer to it (see Appendix).
In chapter 1 he remarks: the finer it cuts, the smaller the loss [or error,
shi]; if one cuts it and further cuts, until one reaches what one can no longer cut,
then it coincides [fits] with ( he ti) the circle and there is no error.
36
Here
Liu Hui may envisage the convergence of the perimeter of the inscribed polygon
and the circumference of the circle, though it is clear from the subsequent dis-
cussion that it is recognised that with any determinate-sided polygon there is a
remainder. Precisely which sections of the ensuing investigation of the value of
the circumference-diameter ratio are Liu Huis own work and which the work of
later commentators is disputed,
37
but their efforts are directed at better and better
approximations and never claim an exact value.
Moreover in the otherwise similar investigation in chapter 5, Liu Hui uses a
rather different formula for his result. The smaller they are halved, the finer are
the remaining [dimensions]. The extreme of fineness is called subtle ( wei);
that which is subtle is without form ( xing). When it is explained in this
way, how could one get a remainder?
38

34
Zhuangzi 33: 78.
35
In tackling the problems addressed in this section I have benefited a great deal
from an extensive exchange of correspondence with Dr. Donald Wagner.
36
Chap. 1: 103104. Translation adapted from Li and Du, Chinese Mathematics, p.
68. Cf. Chemla, Mthodes Infinitsimales, p. 41 n. 15.
37
See especially Donald B. Wagner, Doubts Concerning the Attribution of Liu
Huis Commentary on the Chiu-Chang Suan-Shu, Acta Orientalia no. 39 (1978), pp.
199212, at pp. 206ff.
38
Chap. 5: 168. Translation adapted from Wagner, An Early Chinese Derivation,
p. 173. Cf. Karine Chemla, Mthodes Infinitsimales en Chine et en Grce Anciennes:
les limites dun parallle, in J. M. Salanskis and H. Sinaceur (eds.), Le Labyrinthe du
Both the similarities with, and the differences from, Greek procedures are
alike remarkable. In neither case does Liu Hui proceed, as the Greeks would
normally have done, via an indirect proof.
39
The Greeks would have demon-
strated that, for example, the area to be determined cannot be either greater or
less than a given area and so must be equal to it.
40
Liu Huis attack is direct, via
the securing of closer and closer approximations.
Again the Greeks, as we have seen, generally thought that the area to be de-
termined is, precisely, not exhausted, for Eudoxus principle allows the insertion
of a further rectilinear figure between any rectilinear figure and the circumfer-
ence to which it approximates. True, there is one notable Greek exception,
namely Antiphon, who is reported to have claimed that the circle is at some
point (pote) exhausted by the inscribed polygons.
41
But his suggestion was
rejected by philosophers and mathematicians alike as being in conflict with the
principles of geometry, that is to say the continuum assumption. Aristotle curtly
notes (Physics 185a16f) that whereas other attempts to square the circle need
refuting, it is not the business of the geometer to refute that of Antiphon.
42
Liu Hui, for his part, is clear, as we noted, that with any polygon with a de-
terminate number of sides, its perimeter falls short of the circumference of the
circle. Both here and in the pyramid investigation he envisages iterations that
can be continued indefinitely. It is interesting, however, that there is a certain
variation or hesitation in the strength of the claims made at the end of the in-
quiry. In chapter 1 you continue until no further cuts are possible and there is
no error, while in chapter 5 the conclusion is a question: how could one get a
remainder? Moreover the mathematical results obtained in the two cases differ.
In chapter 5 the formula for the volume of the pyramid is exact. In the circle
division, while the formula A = 1/2 c times 1/2 d is also exact, that is no use
until a value for the length of the circumference is obtained: in fact the numeri-
cal value for the circumference-diameter ratio is no more than approximate,
though it can be made increasingly accurate.

continu (Berlin: Springer, 1992), pp. 38f, who translates the final phrase: comment
obtiendrait-on un reste?
39
This point has been made by Chemla, Mthodes Infinitsimales, pp. 42ff.
40
As in Euclids investigation of the circle in Elements XII 2 (cf. also Archimedes,
On the Measurement of the Circle) and the discussion of the pyramid at Elements XII 3.
41
As reported in Simplicius, On Aristotles Physics 54.20ff (note pote da-
panomenou at 55.6), on which see Ian Mueller, Aristotle and the Quadrature of the Cir-
cle, in Norman Kretzmann (ed.), Infinity and Continuity in Ancient and Medieval
Thought (Ithaca, N.Y.: Cornell Univ. Press, 1982), pp. 14664 at pp. 154ff; Knorr,
Infinity and Continuity, pp. 130ff.
42
The contrast is with Hippocrates of Chios, at least according to the ancient com-
mentators, Themistius 3.30ff; cf. Simplicius, On Aristotles Physics 54.12ff, though it
is doubtful if Hippocrates was guilty of any fallacy; see G.E.R. Lloyd, The Alleged
Fallacy of Hippocrates of Chios, Apeiron no. 20 (1987), pp. 10328.
The question then arises as to whether Liu Huis expressions imply that he
has a clear grasp of the notion of taking the limit of an infinite series, and we
have to stress that much of the background we need to answer that is lacking.
We simply do not have good contemporary texts that could throw light on the
extent to which a technical vocabulary had been developed in contemporary
mathematics. The key question is the expression there is no error, which can
be taken either looselythere is no appreciable error, the error can be dis-
countedor strictlythere is no error since what is envisaged is the strict coin-
cidence of the infinite-sided polygon with the circle. On the one hand there is no
talk of infinite sides and no clear Greek-style definition of limit. On the other
the term wei is clarified in chapter 5 and not just for use in that investigation
since it is applied elsewhere in the commentary. What Liu Hui invokes in the
pyramid determination is an extreme of fineness that is without form. That
expression too can be taken strictly or loosely, but offered as an explanation for
why there is no remainder may suggest Liu Huis confidence that the formula of
the volume of the pyramid (that is, one third length times breadth times height)
is exactas indeed it is. If the variation in the expressions used in the two cases
is mere variation, for stylistic purposes, that would be compatible with the hy-
pothesis that Liu Hui is in complete control, on both occasions, of procedures
that imply the taking of a limit to an infinite series.
In the absence of more determinate evidence we have to suspend judgement
on that point. Nevertheless, for our purposes here, the upshot of this rapid sur-
vey is clear. It is obvious that Liu Hui is as much at home as any Greek geome-
ter with the deployment of indefinitely recurring procedures, and it is possible
that he was a good deal more at home than most with the notion of convergence
to a limit.
III
So at the second stage in our argument we can say that good grounds have been
identified for revising quite drastically a number of the generalisations about
Greece and China propounded at the first stage in the discussion. However, a
more radical approach to the issues needs to be adopted. What both lines of ar-
gument so far developed have in common, and where both seem flawed is in
this: they focus on similarities and differences in concepts or theories (between
Chinese and Greeks) and then strive to identify, on the Chinese or on the Greek
side, what the majority of thinkers, or the most important ones, maintained in
most contexts, and they then attempt to use what has been identified as the Chi-
nese or Greek view as the basis for an argument insisting either on the contrasts
(as in stage one) or on the similarities (stage two). Butquite apart from the
risky, even delusory, character of any generalisation to what the Chinese or the
Greek view isthat may miss the point.
On the Greek side, first, it is easy to see that on this subject, as on so many
others, there was no uniformity, let alone orthodoxy. Almost any position, any
use of infinity, can be attested, and most theses that were put forward by ancient
Greek thinkers were challenged by (other) Greek thinkers themselves. So it is no
wonder that both the claim that the Greeks espoused the infinite with equanim-
ity, and the claim that they fought shy of it, have some evidence that can be
cited in apparent support. But that does not get us anywhere.
But more importantly, secondly, the Greeks often maintained whatever po-
sition they took up on these issues in direct opposition to the alternative. First,
they were generally conscious of the alternative, but second, and again more
importantly, they often generate arguments for their own positions by way of
refuting those alternatives.
It is not as if Plato and Aristotle simply ignored the opposition: they used it.
Aristotle, especially, develops atomistic arguments then to put them downas
we have seen in the passage from On Coming-to-be and Passing Away cited
before. The same dialectical confrontations recur in the Hellenistic period,
though the quality of the polemic varies. Thus Epicurus sometimes proceeds by
simply denying the oppositions case or dismissing it out of hand, though there
are also plenty of arguments against rival views in him and in one of our fullest
later sources for Epicureanism, the +1st century Roman poet Lucretius. Again I
have remarked that neither Galen nor Alexander is exactly fair to the Stoic view
of matter when they attack it. However, the point remains that there continues to
be a heavy use of whatever opposition you face, not just for destructive pur-
poses, to show that they are mistaken, but also for constructive ones, to build up
your own position. The alternatives, indeed, were often seen as mutually exclu-
sive and exhaustive: that allowed arguments to proceed by exclusion. It cannot
be the case that the universe is finite/infinite: so it must be infinite/finite. It can-
not be the case that matter, space and time are infinitely divisible: so they must
be constituted by indivisibles. Or vice versa.
43
The dialectical confrontations get an added twist with the Hellenistic Scep-
tics. Dilemmatic arguments presenting antinomies were used already, long be-
fore, notably by Zeno of Elea, who aimed to defeat the assumption that motion
is possible whatever view was taken on how space and time were composed.
44

43
I explored the use of this and other types of polar arguments in G.E.R. Lloyd,
Polarity and Analogy: Two Types of Argumentation in Early Greek Thought
(Cambridge: Cambridge Univ. Press, 1966).
44
This remains the case, even though Zenos presentation of the dilemma was not as
crisp as Paul Tannery, Pour lhistoire de la science hellne (Paris: Gauthier-Villars,
1930) thought when he envisaged Zeno attacking motion first on the assumption that
space and time are infinitely divisible, then on the assumption that they consist in strict
indivisibles. The second limb of the reconstruction faces two difficulties: (1) that it con-
flicts with how the ancient commentators took Zeno, (2) the notion of strict quanta of
time is otherwise not attested before the late 4th century. But even if Zeno did not use
But the Hellenistic Sceptics held that on every theoretical issue, relating to un-
derlying reality or hidden causes, both sides were wrong.
45
One should, as they
said, suspend judgement. But the way they proceeded was, precisely, to match
every positive (dogmatic) thesis to its antithesisto leave not just one of
them, but both, undermined. Isostheneia, equal strength on either side, is what
they called this: but if equally strong, equally weak.
Here then, in stage three of our discussion, we rejoin some familiar topics on
the Greek side, at least, namely the antagonistic strands that are such a feature of
Greek thought, together with the recurrent concern for foundational questions
and the ultimate justification for a position.
46
In the latter regard, both atomic
and continuum theories addressed the issue of the ultimate constitution of mat-
ter, space and time. It is striking that continuum theorists tend to be continuum
theorists across the boardand so too were atomists.
47
That is, when a contin-
uum view was adopted, it was adopted not just for physical matter, but also for
space and time and indeed for geometrical magnitudes: all were treated as infi-
nitely divisible. Similarly the atomists often deny all these modalities of con-
tinua, and postulate atomic quanta, physical, spatial, temporal, geometrical, in-
stead. Mixtures of the two types of position are rare, if not (as Furley suggested)
quite unknown: one possible exception is Democritus, who may have combined
physical atomism with an assumption of a geometrical continuum, though the
point is certainly controversial.
48

the idea of strict quanta of time, the arguments called the Arrow and the Moving Rows
attack motion on the basis of some looser notion of the component elements of space and
time. See, for example, David J. Furley, Two Studies in the Greek Atomists (Princeton:
Princeton Univ. Press, 1967), pp. 71ff; G.E.L. Owen, Zeno and the Mathematicians, in
Logic, Science and Dialectic (London: Duckworth, 1986), pp. 4561 (originally in Pro-
ceedings of the Aristotelian Society no. 58 [19578], pp. 199222). The further sugges-
tion, developed by Tannery and others, that Zenos target was a specific Pythagorean
doctrine according to which numbers, points and atoms are identified, is nowadays dis-
counted for lack of reliable primary evidence.
45
Some of the primary texts are collected in Long and Sedley, The Hellenistic Phi-
losophers, vol. 1, sections 6872, pp. 438ff, and cf. the essays collected in Malcolm
Schofield, Myles Burnyeat and Jonathan Barnes (eds.), Doubt and Dogmatism (Oxford:
Clarendon Press, 1980), and in Myles Burnyeat (ed.), The Skeptical Tradition (Berkeley:
Univ. of California Press, 1983).
46
I have discussed aspects of this in G.E.R. Lloyd, Demystifying Mentalities
47
As has recently been argued by David J. Furley, The Greek Cosmologists, vol. 1
48
The ways in which these positions may have been reconciled have been discussed
by Kate Meakin, Pre-Platonic Ontology of Mathematics, Ph.D. dissertation, Cam-
bridge, 1990.
I am tempted to argue, therefore, that the plurality of theoretical positions
different Greeks adopted on this set of issues is theall too predictable
outcome of their fondness for the dialectical exploration of abstract ideas. That
in turn relates (1) to the rivalry between those who claimed special knowledge
and (2) to their sense of how to present their cases in the best possible light in
relation to the audiences they were hoping to persuadeoften general audi-
ences, indeed, present at open debate between rival views set out by contending
parties. Advocacy, practised extensively in the Greek law-courts, comes to be a
prominent feature of many Greek intellectual exchanges, when opposing
masters of truth set out to challenge conventional wisdom and each others
replacements for it.
Standard, conventional or common-sense opinions thus get to be over-
turned and repudiated, in ancient Greece, across the board, and on some un-
likely issues, by people who claimed to know better. The sensible assumptions
that the earth is flat, the heavens a dome above it, physical objects solid, are
confronted with all sorts of rival theories proposed by speculative thinkers of
varying degrees of fame and respectability. As for the fine structure of time and
space, common assumptions did not really tell people quite what to think. You
certainly did not need to have a view on that to order your life according to the
cycle of the seasons and the passing of the years. But once Greek speculative
thought began in earnest with the Presocratic philosophers, then just about every
thesisand its contradictoryare examined, and quite a few are seriously en-
tertained. On that line of argument, it is not at all surprising that, once Greek
philosophy got going, then sooner or later infinite worlds, atomic quanta of
time, even indivisible lines, would find proposers and secondersas well as
opponentsin the to and fro of abstract debate.
This argument, on the Greek side, does not, to be sure, resolve the problems,
so much as redefine them, setting a new agenda, though now one that is investi-
gable in terms of the types of examination of the social framework of science
that Sivin and I aim to undertake. The explanandum then becomes not why this
or that idea of the infinite was promulgated in this or that context, but rather
what the social and intellectual conditions were that allowed such abstract de-
bate to develop.
But what about China? Does not the complexity of the data we uncovered
suggest that here too an analogous point applies? If so, our task should not be to
attempt to explain some supposed Chinese readiness, or reluctance, to espouse,
or to shun, the infinite as such, but rather to see what light our study of the ac-
tual, complex, diverging views on the infinite can throw on the aims and pre-
suppositions of the thinkers concerned, whether in mathematics or in philoso-
phy.
Two tentative suggestions, in that regard, may be followed up briefly in con-
clusion, the first to do with the balance between theoretical interests and practi-
cal applications in Chinese mathematics, the second to do with the conduct of
philosophical exchanges in China more generally.
First, it has often been claimed that Chinese mathematics is essentially
pragmatic, if not indeed directly practical, in aim, and this has been a standard
way of contrasting Chinese mathematics with the dominant Greek tradition rep-
resented by Euclid or Archimedes. However, our analysis of the use of recurrent
procedures suggests that qualifications need to be recognised. First, to take one
example of many, the exploration of the ratio of circumference to diameter is
certainly carried far beyond any possible practical application. For practical
purposes a value of 3 or 3 1/7 is perfectly adequate. But the commentaries on
chapter 1 of the Nine Chapters pursue the analysis to the point where the area of
a polygon of 192 sides is being determinedand even beyond.
49
Yet we must
also acknowledge that the aim of that exploration is not a proof, but the best
approximation possible.
Yet that is not to say that nothing like proofs are ever attempted in Chinese
mathematics. Liu Huis preliminary study of the area of the circle shows that the
formula he states is correct, namely that the area is equal to half the circumfer-
ence times half the diameter. True, the form of the proof differs from the style
of axiomatic-deductive demonstration favoured by Greek mathematics in the
Euclidean tradition. However, the fact that axiomatic-deductive demonstration
does not figure in Liu Hui nor in any other classical Chinese mathematical text
should not blind us to the further point, that what we have in Liu Hui, in this
context as in others,
50
are perfectly proper procedures to verify correct results
proving them, indeed, in just that sense.
Moreover on the question of pragmatics, Liu Hui himself notes, at the end of
his pyramid investigation, that the object [called] bienao has no practical use;
the shape [called] yangma can be long or short, or broad or narrow. Neverthe-
less without the bienao there is no way to investigate the number [that is the
volume] of a yangma; and without the yangma there is no way to know [the
other shapes connected with the cone and the truncated cone]. These are primary
in application.
51
Two points are striking about this remark. First, Liu Hui clearly recognises
that his study goes beyond what is of direct practical use. But then, secondly, he
registers a certain defensiveness on the point. It is as if he feels some obligation
to justify himself here, as if his assumption, or that of other students, is that

49
Chap. 1: 1046. As noted before, the problem of which parts of this discussion are
the work of Liu Hui, which that of later commentators, is disputed.
50
On different styles of mathematical proof, see Wagner, An Early Chinese Deri-
vation, pp. 164ff, and Karine Chemla, Resonances entre dmonstration et procdure,
Extrme-Orient Extrme-Occident no. 14 (1992), pp. 91129.
51
Chap. 5: 168. Translation adapted from Wagner, An Early Chinese Derivation,
p. 182.
mathematical investigations should prove themselves to be, directly or indi-
rectly, of practical use. In that latter regard, the contrast can still be remarked
between Liu Huis views and those of those Greeks who rated mathematics the
higher the more abstract, the further removed from practice, it was.
52
My second, final, more general point relates to debate. I have stressed else-
where the antagonistic characteristics of much Greek speculative thought, and
they are certainly in evidence in discussions of infinity. Face to face confronta-
tion, often before a general audience, the public who in some cases adjudicated
who had won the contest,
53
is, at once, typically Greek and profoundly un-
Chinese.
But that is not to say that there is no criticism, sometimes indeed hard-hitting
criticism, of rivals in Chinese philosophy and science. Two examples relating to
our topic can be used to illustrate this. We noted before that the evidence for the
xuanye idea of a boundless heaven is late. In Han times, two main cosmological
ideas were current, in both of which the world is thought of as finite. In the
Gaitian system, described in the Zhou bi suanjing of the 1st century, the
heaven is a circular canopy set over a central earth. In the Huntian view the
heavens are a complete sphere, with half of this invisible below the earth at any
time. Each view was associated withindeed closely tied toobservational
techniques, and they are better considered not so much as cosmological theories,
as rather ways of doing astronomy.
54
Although the two are sometimes combined, the period of the rise of the
Huntian view is characterised by texts that criticise Gaitian. Thus the +10th
century encyclopaedia Taiping yulan preserves an account of a discussion from
the lost work Xin Lun by Huan Tan ( 40 to +30). In this the Gaitian view is
duly set out and then criticised from a Huntian perspective, on the grounds, for
example, that it could not account for the equality of day and night at the spring
and autumn equinoxes, and with a direct appeal to the actual experience of the
movements of the suns shadow in the White Tiger Hall where the purported

52
This is a prominent and recurrent theme among those who considered themselves
Platonists, such as Proclus, in his commentaries on the Timaeus and on Euclids Ele-
ments I. But they could and did cite texts from Plato himself as their ancient authority,
for example Socrates remarks on the so-called propaedeutic studies in the Republic (e.g.
525aff). Thus it is claimed at Republic 530bc that the chief usefulness of astronomical
study lies, precisely, in its value in training the soul in abstract thought.
53
This is the case, for example, with what are called sophistic epideixeis, or display
speeches, examples of which are extant in our Hippocratic Corpus and which are referred
to in On the Nature of Man, chap. 1.
54
For a recent discussion of these two, see Christopher Cullen, Astronomy and
Mathematics in Ancient China: the Zhou Bi Suan Jing (Cambridge: Cambridge Univ.
Press, forthcoming).
discussion took place.
55
While that is still a far cry from Greek debates aiming
to establish by argument that the heavens must be finite, or again infinite, it il-
lustrates an openness to alternatives and a recognition that they are just that,
potential competitors.
As a second example we may revert to the way in which the topic of the in-
finite provided resources for paradox in China, as it also did in Greece. We
mentioned the dicta attributed to Hui Shi and others reported in the Tianxia
chapter of Zhuangzi. However the reaction that Hui Shi provoked, in that dis-
cussion, is interestingly different from the reception that later Greek philoso-
phers accorded to Zeno. Aristotle, at least, as we saw, wanted to defeat Zeno by
argument, the effect of which was to claim that Zeno had ignored a distinction
that he, Aristotle, maintained to be important.
The reaction to Hui Shi in Zhuangzi is milder, if condescending, and it pro-
ceeds not by an attempted refutation but by lamenting the waste of his talents.
He had many formulas, his writings filled five carts, but his way was eccentric.
. . . Hui Shi day after day used his wits in disputation with others, but it was
only in the company of the disputers in the world that he distinguished himself
as extraordinary. . . . What a pity that Hui Shis talents were wasted and never
came to anything, that he would not turn his back from chasing the myriad
things. He had as much chance of making his voice outlast its echo, his body
outrun its shadow. Sad, wasnt it?
56
Where much Greek effort continued, right down to the times of the last great
Aristotelian commentators, Philoponus and Simplicius, in the +6th century, to
be devoted to destructive criticism of rivals, there was a greater sense, in China,
of the need to develop a common language. What this text in Zhuangzi helps to
bring out is a related difference in the manner of dealing with disagreement. The
reaction to Hui Shis speculations in this chapter is not to confront, let alone
refute him, but rather to pour scorn on his wasted effort. From the earliest
Presocratic philosophers onwards, the preferred Greek style of reasoning, on the
finite and infinite as on most other topics, was rather to attempt to disprove op-
ponents by argument. But that, we can see, had the effect of prolonging dispute,
rather than resolving it, and may thus have been one factor militating against the
formation of an orthodoxy in Greece. The Chinese, by contrast, from Han times
on especially, worked hard to secure one, and largely succeeded in elaborating
an orthodoxy that served, as Sivin has shown, at once to legitimate imperial rule
and to establish the role of the learned within the order of things.
57
Of course the topic of the infinite, in its various manifestations is only one of
many subjects that can be studied to help to throw light both on the social and

55
Taiping yulan 2, 6b7a, discussed in Cullen, Astronomy and Mathematics, who
speaks of an age of polemic when the Huntian system comes on the scene.
56
Zhuangzi 33: 6987.
57
Sivin, Yin Yang and the Five Phases.
professional institutions of science and philosophy in China and Greece, and on
the effects of those institutions on the character of the science and philosophy
produced. Moreover in the process there is much in the topic of the infinite, in
its deployment in both China and Greece, that remains puzzling, and it is cer-
tainly as well to recognise that many of the specificities, of why this or that
thinker proposed this or that use of the infinite in this or that context, are beyond
the range of anything that could be called a hard-edged explanation. We shall no
doubt never be able to explain Aristotle or Euclid or Hui Shi or Liu Hui in full,
in the sense of explaining all the complex processes that contributed to their
formulating the problems in the ways they did, and to their reacting to their
predecessors in the precise manner in which that happened. But what we can
hope to do is to use such specific topics as we have adumbrated here to explore
the goals and the styles of argument of different Greek and Chinese thinkers at
different junctures. The further investigation of those broader issues is, indeed,
the subject of that larger collaborative project on which Professor Sivin and I are
now engaged.
Appendix
Liu Huis problem is to find the formula for the volume of the figure called
yangma, a pyramid with rectangular base and one lateral edge perpendicular
to the base. He first shows that a yangma plus a bienao (pyramid with right
triangular base and one lateral edge perpendicular to the base) form qian-
du (right prism with right triangular base) (Figure 1 sets out the shapes, Figures
2 and 3 show the result of adding the bienao BACE to the yangma BFECD:
from Donald B. Wagner, An Early Chinese Derivation of the Volume of a
Pyramid: Liu Hui, Third Century AD, Historia Mathematica no. 6 [1979], pp.
16488, at pp. 166ff).
qiandu yangma bienao
Figure 1.
But the volume of a qiandu is given as half length times breadth times
height. So Liu Hui has to show that one yangma equals two bienao in order to
give the formula for the volume of the yangma (namely one third length times
breadth times height). This he does by first di-
viding the yangma and the bienao as in Figures
4 and 5. The bienao in Figure 4 (BACE) is di-
vided into two qiandu (AGIJML and ILMJCP)
and two bienao (BGIL and EPML).
The yangma in Figure 5 (BDFEC) is divided
into one box (HILKNDRO), two qiandu
(ILORCP and KLONFQ) and two yangma
(BHILK and LOPEQ). But the sum of the two
qiandu in the bienao (Fig. 4) can be seen to equal
half that of the box and the two qiandu in the
yangma (Fig. 5). So he must now show that the
remaining bienao of the bienao are half the re-
maining yangma of the yangma. But they can be
subdivided in exactly the same way as the origi-
nal bienao and yangma were. Again part of the
components thus distinguished can be seen to
fulfil the requirement of the proportion, one
yangma equals two bienao, leaving an even
smaller remainder.
B
A
D
C
F
E
Figure 3.
B
D
C
F
E
Figure 2.
L
M
I J
C
P
E
B G
A
Figure 4.
If the process is continued, the series
converges on the formula one yangma
equals two bienao. While there are disa-
greements on points of detail between
Wagner, An Early Chinese Derivation,
Li Yan and Du Shiran, Chinese Mathe-
matics: A Concise History, translated by
John N. Crossley and Anthony W.-C. Lun
(Oxford: Clarendon Press, 1987), pp. 70ff,
and Karine Chemla, Mthodes Infini-
tsimales en Chine et en Grce Anciennes:
les limites dun parallle, in J. M. Salan-
skis and H. Sinaceur (eds.), Le Labyrinthe
du continu (Berlin: Springer, 1992), pp.
3146 at pp. 32ff, they adopt the same
overall interpretation of this text.
F
Q E
Figure 5.
N
O
P
C
D
R
K
H
L
I
B
35
The Development of Hindu-Arabic and
Traditional Chinese Arithmetic
Lam Lay Yong
[Lam Lay Yong is Professor of Mathematics at the National University of
Singapore. She is an Effective Member of the International Academy of the
History of Science. Her latest research paper is Zhang Qiujian suanjing (The
Mathematical Classic of Zhang Qiujian): An Overview.]
* * *
The Basics
rithmetic is a branch of mathematics with which all of us are familiar.
Irrespective of the countries we come from, we are taught the Hindu-
Arabic numerals and their place value notation at a very early age.
Thereafter we learn how to add, subtract, multiply, and divide with the
numerals. We are also taught how to write the common fractions and how to
add, subtract, multiply, and divide with them. It was from these basic techniques
that arithmetic developed. Arithmetic is an important subject in our schools as
its applications have become essential in our everyday life; from the
man-in-the-street to the scientist, government officer, and businessmanall of
us need to know arithmetic.
That arithmetic is useful is clear, but the history of the beginnings of the
subject is still obscure. In recent decades, there have been some studies on
Arabic texts related to early arithmetical procedures, and these have helped to
shed some light. When we compare this knowledge with what we know about
the arithmetic of ancient China, what is immediately and impressively apparent
are certain strikingly similar features. The aim of this paper is to draw attention
to these similarities and to show how vital they were in the development of
arithmetic.
The Hindu-Arabic numeral system on which arithmetic is built possesses the
intrinsic property of a place value notation which has ten as base. The digits of a
numeral are arranged from left to right in descending order of ranks. Initially,
A
the system employed only nine signs which represented the numbers from one
to nine. For example, the number sixty-five thousand three hundred and ninety-
two was written in the form 65392 so that the signs representing six, five, three,
nine, and two were written in a horizontal line from left to right. The place
occupied by each digit indicated the rank of that digit; in this example, the ranks
of the digits 6, 5, 3, 9, and 2 were ten thousands, thousands, hundreds, tens, and
units, respectively. A number such as sixty-five thousand three hundred and two
was indicated by the appropriate signs for 6, 5, 3, and 2 which occupied their
due places, while the place which represented the tens rank was left vacant. This
blank space was called sunya in India and sifr in the Arab world, both words
meaning empty.
1
A tenth sign, which was usually in the form of the zero symbol
(occasionally it was a dot), was introduced later to occupy this vacant place.
The rules for manipulating the Hindu-Arabic numerals and the processes of
calculation were known in the West through Latin translations of Arabic
manuscripts from the twelfth century. The procedure of computation was called
algorism. This word is generally accepted as associated with the name of
Muhammad ibn Ms al-Khwrizm of the ninth century. The Latin text of
Cambridge University Library Ms. Ii.vi.5, dating from the thirteenth century, is
generally accepted as a copy of an earlier Latin translation of an Arabic
manuscript based on the arithmetic text by al-Khwrizm. The incomplete
manuscript has only sixteen pages
2
of which the first four and a half pages are
devoted to a description of the Hindu-Arabic numerals with detailed
explanations of the place value notation. Except for the last part on fractions and
two short paragraphs on halving and doubling and casting out nines, the rest of
the manuscript is concerned with the operations of addition, subtraction,
multiplication, and division.
An English translation of the part describing the division of 46468 by 324 is
reproduced below so as to give the reader some idea of how the process of
division was described and performed.
3
And know that division is similar to multiplication, but this is done inversely,
because in division we subtract and there we add, i.e., in multiplication is its

1
Karl Menninger, Number Words and Number Symbols: A Cultural History of
Numbers, translated by Paul Broneer from the revised 1958 German edition (Cambridge,
Mass.: MIT Press, 1969), pp. 400401; Tobias Dantzig, Number: The Language of
Science (London: George Allen & Unwin, 1930, 3rd ed. 1947), pp. 2930; Louis C.
Karpinski, The History of Arithmetic (Chicago: Rand McNally, 1925), p. 41.
2
The manuscript occupies folios 104r-111v, which were previously numbered as 102
109 in A Catalogue of the Manuscripts Preserved in the Library of the University of
Cambridge, vol. 3 (Cambridge: Cambridge Univ. Press, 1858), pp. 500501.
3
John N. Crossley and Alan S. Henry, Thus Spake Al-Khwrizm: A Translation of
the Text of Cambridge University Library Ms. Ii.vi.5, Historia Mathematica vol. 17, no. 2
(May 1990), pp. 11819.
Lam Lay Yong: Hindu-Arabic and Traditional Chinese Arithmetic 37
model. But when we wanted to divide forty-six thousand and four hundred and
sixty-eight by three hundred and XXIIII, we first put eight on the right side, then
we put six toward the left which is sixty, then IIII which is four hundred, then six
which is VI thousand, then IIII which is forty-thousand. Of these places the last
toward the left will be IIII, and the first of them toward the right eight; after this
write under them the number by which you are dividing, and write the last place
of the number by which you are dividing, which is the form of three and is three
hundred, under the last place of the upper number which is IIII, provided that it
is less than that which is above it: and if it were more than it, we shall move it by
one place and put it under the six. After this we shall put in that place, that comes
after the three, the form of two which is XX, beneath the six; then we shall put in
that place, which is beneath the IIII, likewise IIII and this will be their form:
46468
324
After this to begin with let us write one in the column of the first place of the
number by which you are dividing, above the upper number that you are dividing
which is four. And if we had put it beneath the IIII, it would be equally appropri-
ate. Let us multiply it (i.e., one) by three, and we shall subtract it (i.e., 3) from
that which is above it, and there will remain one. Then let us multiply it by two
and subtract it (i.e., 2) from that which is above it, which is VI, and there will
remain IIII. After this let us multiply it again by IIII and subtract it (i.e., 4) from
that which is above it, which is IIII and nothing will remain; and we shall put a
circle in its place. Next move the beginning of the number by which you are
dividing, i.e., IIII, beneath VI and there will be two beneath the circle and III
beneath IIII. Then write in the column of the lower number something in the
column [sic, should be row] of the one, i.e., IIII, and multiply it by three and
there will be XII; and subtract it (i.e., 12) from that which is above three which is
XIIII and there will remain II; after this multiply also IIII itself by the two that
follows the three and there will be VIII, and subtract it (i.e., 8) from that which is
above it, that is XX and there will remain XII, i.e., two above the II and one
above the three. Again multiply IIII by IIII which comes next on the right, and
there will be XVI; and subtract it (i.e., 16) from that which is above it which is
CXXVI and there will remain above the IIII a circle and above the two, one, and
above the three, one. Again move the number by which you are dividing, i.e.,
IIII, beneath VIII, and there will be two beneath the circle and three beneath the
one; next write three in the column of IIII above the upper number that you are
dividing in the row of IIII and one (i.e., 143), multiply it (i.e., 3) by three, and
there will be IX; and subtract it (i.e., 9) from that which is above three which is
XI and there will remain two above the three. Multiply also the three by the two
that follows the three and there will be VI, and subtract it (i.e., 6) from that
which is above the three, which is XX; there will remain XIIII. Once more
multiply the aforesaid three by IIII that follows the two and there will be XII, and
subtract it from that which is above it, which is CXXVIII . . . ;
4
and there will
remain six above the IIII, and three above the two, and one above the three. And
there will come out for us what is owed to one from it and this will be CXLIII
and CXXXVI parts from CCCXXIIII parts of one (i.e., 143
136
324
). And this is
their form:
143
136
324
Vogel
5
points out that this lengthy description was equivalent to the following
steps or algorism:
1 1 14 14 14 14 143 143 143 143
46468 14068 14068 2068 1268 1108 1108 208 148 136
324 324 324 324 324 324 324 324 324 324
The section on division in the manuscript begins with a general description
of the process of division which is illustrated by the above example and
followed by the division of 1800 by 9. Only one method of division is
employed.
In the same vein, only one type of addition method, one type of subtraction
method, and one type of multiplication method are described in the manuscript.
6
There are examples to illustrate the multiplication and subtraction methods, but
there is only a general description for the addition method.
Vogels summary of the steps for the subtraction of 3211 from 6422 and the
multiplication of 2326 by 214 are shown below:
7
6422 3422 3222 3212 3211
3211 211 11 1

4
It has been pointed out in Crossley and Henry, Thus Spake Al-Khwrizm, p. 119,
that the text has CXXVIII instead of the correct number CXLVIII.
5
Kurt Vogel, Mohammed ibn Ms Alchwarizi's Algorismus (Aalen: Otto Zeller
Verlagsbuchhandlung, 1963), p. 47.
6
For an English translation, see Crossley and Henry, Thus Spake Al-Khwrizm, pp.
11017.
7
Vogel, Mohammed ibn Ms, pp. 4546.
2326 428326 428326 492226 496486 497764
214 214 214 214 214
Each procedure begins from left to right and accuracy in the computation
depends on the correct positioning of the digits of one numeral relative to those
of the other. In the case of multiplication, note that the multiplier, 214, like the
divisor in the above example, is shifted place by place from left to right. Its first
position is such that its units digit, 4, is directly below the first digit from the left
of the numeral to be multiplied. This first digit of the upper numeral is
multiplied by each digit of the lower numeral commencing from the left. It is
erased when the product is written in its place.
8
The multiplier is then shifted to
the right by one place so that its units digit, 4, is directly below 3. The first digit
from the left of the lower numeral, 2, multiplies 3 and the product is added to
the digit directly above 2, so that the upper numeral is changed to 488326. Next,
1 multiplies 3 and the product is added to the digit directly above 1, so that the
upper numeral is changed to 491326. Finally, 4 multiplies 3, 3 is erased and the
upper numeral is changed to 492226. The multiplier is shifted to the right by
one place so that 4 is now directly below 2. The process continues till all the
digits of the upper numeral are multiplied; the multiplier is then erased leaving
the result.
The Kitb al-Fusl f al-Hisb al-Hind, written in Damascus in AD 952953
by Ab al-Hasan Ahmad ibn Ibrhm al-Uqldis, is, according to Saidan, the
earliest extant Arabic work of Hindu-Arabic arithmetic and shows this system
at its earliest stages and the first steps in its development.
9
The manuscript
claimed to have a collection of all past knowledge on arithmetic, a clear
exposition of what was currently known about the subject, adaptations of and
improvements on methods using paper as a medium, and other new
contributions.
10
The book therefore has several methods of addition, subtraction,
multiplication, and division; what is significant is that the first method of each
operation
11
is the same as that described in the Cambridge manuscript.
12

8
The product is in bold print.
9
A. S. Saidan, The Arithmetic of Al-Uqldis (Dordrecht: D. Reidel, 1978), p. xi.
10
Saidan, The Arithmetic of Al-Uqldis, pp. 3536.
11
Saidan, The Arithmetic of Al-Uqldis, pp. 4659.
12
The translated passage from Saidan, The Arithmetic of Al-Uqldis, p. 49 shows how
al-Uqldis began his description of the procedure for multiplication:
We start with the multiplication of whole numbers. Multiplication of rubies is
to be started with; that is multiplication of numbers from one to ten, equal or une-
qual. These must be memorized and learnt. Then comes multiplication of two places
It was quite common for authors in Islam writing on the four fundamental
operations of arithmetic to begin with a detailed description of these original
methods.
13
In his Kitb f Usl Hisb al-Hind (Principles of Hindu reckoning),
Kshyr ibn Labbn (fl. ca. 971ca. 1029), after introducing the numerals,
launches into a detailed description of the same procedures for addition,
subtraction, multiplication, and division.
14
In the second part of his book, he
gives a different set of methods based on sexagesimal numbers.
In his book, al-Uqldis pointed out that this first set of methods for
performing the four fundamental operations of arithmetic was done on a
dustboard called the takht.
15
Levey and Petruck also mention the use of the same
device for Kshyr ibn Labbns arithmetic. The dustboard was a writing
surface on which the reckoner spread dust.
16
This enabled him to write numbers
with a stylus or with the tip of his finger; manipulation depended on rubbing out
digits.
17
This first set of methods of how to add, subtract, multiply, and divide with
Hindu-Arabic numerals is of historic importance because it is not only the
earliest known in Islam, it is also the earliest known in the history of
arithmetical procedures performed through the Hindu-Arabic numeral system.
The detailed records of the Arabic arithmeticians reveal that the concept of the
numeral system and its accompanying methods of operations were learned step-
by-step as a package. From this foundation, they adapted, modified, and
improved the procedures for a paper and ink medium.
18
The ancient Chinese performed their addition, subtraction, multiplication,
and division using rods made from animal bones or bamboo. Rods were used for
reckoning since the Warring States period (475221 BC) and possibly earlier,
and through their use a numeral system was developed. The numeral system
possessed the same intrinsic property as the Hindu-Arabic numeral system: it

by one, like multiplying 74 by 9. We draw up the nine under the seven, like this:
7 4
9
.
Now we multiply 9 by 7 and insert the units of the result in place of seven, and
the tens after it, thus having
634
9
.
Then the nine is shifted and set under the four. It is multiplied by four; the units
are inserted in place of four, and the tens are added to the next place. The result is
6 6 6
9
which is the result of multiplication.
13
Saidan, The Arithmetic of Al-Uqldis, p. 13.
14
Martin Levey and Marvin Petruck, Kshyr ibn Labbn: Principles of Hindu
Reckoning (Madison: Univ. of Wisconsin Press, 1965), pp. 4860.
15
16
Levey and Petruck, Kshyr ibn Labbn, pp. 45, 41.
17
18
used a place value notation with ten as base. The nine signs which represented
the numbers from one to nine were as follows:
A rod numeral representing a number such as sixty-five thousand three
hundred and ninety-two would appear as . The rods
representing the digits five and nine which occupied places whose ranks were
thousands and tens respectively were rotated through ninety degrees so that a
horizontal rod was turned into a vertical one and a vertical rod was turned into a
horizontal one. The numeral system had a place value notation with ten as base
through this ingenious device of rotating the rods of the digits which occupied
alternate places. The rotated nine signs which occupied places whose ranks were
tens, thousands, hundred thousands, and so forth, were as follows:
A numeral representing a number such as sixty-five thousand three hundred
and two would appear as : the place that represented the tens
rank was left blank and was called kong (empty).
Among the extant literature, the Sun Zi suanjing (The mathemati-
cal classic of Sun Zi) (ca. 400 AD) has the earliest description of how
multiplication and division were performed with the rod numerals. Comparing
this description of the detailed procedures with those described in the Arabic
texts of the earliest known method of multiplication and division leads to a
startling discovery: the step-by-step procedures are the same in spite of the
different media and the vast difference in the dates of the records.
The Sun Zi suanjing gives a general description of the rod numerals and how
to multiply and divide with them. It also provides specific problems to illustrate
the multiplication and division process.
19
A translation of a problem on
multiplication is reproduced here. Models of rod numerals
20
are constructed to
show the different stages of the performance.

19
Qian Baocong (ed.), Suanjing shi shu [Ten mathematical classics]
(Beijing: Zhonghua shuju, 1963), pp. 28285. For an English translation, see Lam Lay
Yong and Ang Tian Se, Fleeting Footsteps: Tracing the Conception of Arithmetic and
Algebra in Ancient China (Singapore: World Scientific, 1992), pp. 3235, 3841.
20
These correspond to the lower Roman numerals that have been added in the passage.
Nine nines are 81, find the amount when this is multiplied by itself.
The answer says: 6561.
The method says: Set up the two positions: [upper and lower]
21
[i]. The upper 8
calls the lower 8: eight eights are 64, so put down 6,400 in the middle position
[ii]. The upper 8 calls the lower 1: one eight is 8, so put down 80 in the middle
position [iii]. Shift the lower numeral one place [to the right] and put away the 80
in the upper position [iv]. The upper 1 calls the lower 8: one eight is 8, so put
down 80 in the middle position [v]. The upper 1 calls the lower 1: one one is 1,
so put down 1 in the middle position [vi]. Remove the numerals in the upper and
lower positions leaving 6561 in the middle position [vii].
[v] [vi] [vii]
In Hindu-Arabic numerals, the above stages are as follows:
81 81 81 1 1
64 648 648 656
81 81 81 81 81
[i] [ii] [iii] [iv] [v]

21
How the two rod numerals should be placed in relation to each other has been stated
earlier in the general method; see Lam and Ang, Fleeting Footsteps, p. 32.
[i] [ii] [iii] [iv]
1
6561 6561
81
[vi] [vii]
The Sun Zi suanjing provides a description of the general method for
division (Figure 1) which is translated below:
In the common method of division (fan chu zhi fa ), this is the reverse
of multiplication. The dividend (cheng de , lit. product) occupies the middle
position and the quotient (chu de ) is placed above it. Suppose 6 is the
divisor (fa ) and 100 is the dividend (shi ). When 6 divides 100, it advances
(jin ) two places [to the left] so that it is directly below the hundreds. This
implies the division of 1 by 6. In this case, the divisor (fa) is greater than the
dividend (shi), so division is not possible. Therefore shift (tui ) [6 to the right]
so that it is below the tens. Using the divisor (fa) to remove the dividend (shi),
one six [is 6] and 100 is reduced to 40, thus showing that division is possible. If
the divisor (fa) is less than [that part of] the dividend [above it] (shi), it should
then stay below the hundreds and should not be shifted. It follows that if the
units of the divisor (fa) are below the tens [of the dividend], the place value [of
the digit of the quotient] is tens; if they are below the hundreds, the place value
[of the digit of the quotient] is hundreds. The rest of the method is the same as
multiplication. As for the remainder of the dividend (shi), this is assigned to the
divisor (yi fa ming zhi ) such that the divisor (fa) is called the denomi-
nator (mu ) and the remaining dividend (shi) the numerator (zi ).
The method of division is illustrated by a problem (Figure 2). This is
translated and explained by models of rod numerals.
If 6561 is divided among 9 persons, find how much each gets.
Answer: 729.
Method: First set 6561 in the middle position to be the shi (dividend). Below
it, set 9 persons to be the fa (divisor)
22
[i]. Put down 700 in the upper position
[ii]. The upper 7 calls the lower 9: seven nines are 63, so remove 6,300 from the
numeral in the middle position [iii]. Shift the numeral in the lower position one
place [to the right] and put down 20 in the upper position [iv]. The upper 2 calls
the lower 9: two nines are 18, so remove 180 from the numeral in the middle
position [v]. Once again shift the numeral in the lower position one place [to the
right], and put down 9 in the upper position [vi]. The upper 9 calls the lower 9:
nine nines are 81, so remove 81 from the numeral in the middle position [vii].
There is now no numeral in the middle position. Put away the numeral in the
lower position. The result in the upper position is what each person gets [viii].

22
How the two rod numerals should be placed in relation to each other has been stated
in the general method.
Figure 1 Figure 2
Figure 1 shows the general description of division and Figure 2 shows a
problem on division. Both excerpts are from Sun Zi suanjing; Qian Baocong
(ed.), Suanjing shi shu, pp. 28283; 285
[i] [ii] [iii] [iv]
[v] [vi] [vii] [viii]
The above steps are now shown in Hindu-Arabic numerals.
7 7 72 72
6561 6561 261 261 81
9 9 9 9 9
[i] [ii] [iii] [iv] [v]
729 729 729
81
9 9
[vi] [vii] [viii]
The similarity in the step-by-step procedure for multiplication and division in
the Chinese and Arabic texts is remarkable.
23

23
There is a slight variation in the multiplication methods: the first two lines of the
Chinese method are merged into a single line in the method of the Arabic manuscripts.
There is evidence in the Yang Hui suanfa (Yang Huis methods of computation)
that the Chinese also practiced the latter procedure. Lam Lay Yong, A Critical Study of the
Yang Hui Suan Fa: A Thirteenth-century Chinese Mathematical Treatise (Singapore:
Singapore Univ. Press, 1977), pp. 2324.
Let me summarize what has been discussed so far. The two arithmetics were
built on numeral systems with the same concept: they used a place value
notation with ten as base and thus depended on only nine signs. The digits of
each numeral in both systems were arranged horizontally from left to right in
descending order of their ranks. If a rod numeral did not have a digit of a
particular rank, the space representing that rank was left empty. The very early
Hindu-Arabic numerals also followed this convention. In the basic arithmetic
operations, both arithmetics used the same detailed procedures for multiplication
and division.
24
In view of such similar foundations, how did each arithmetic
evolve? I shall begin with the arithmetic of China.
The Early Evolution
With rods as tools of operation, arithmetic in ancient China grew from basic
computations. Its growth was stimulated by the demands of the practical needs
of society. The most important of the earliest mathematical writings that has
survived is the Jiu zhang suanshu (Nine chapters on the
mathematical art). According to Li Yan and Du Shiran, this work is
representative of the development of ancient Chinese mathematics from the
Zhou and Qin to the Han dynasties (c. 11th century BC220 AD).
25
In spite of
its early date, the mathematics is more advanced than that in the Sun Zi
suanjing. It also assumes that the reader is familiar with the rod numeral system
and the procedures of how to add, subtract, multiply, and divide with rods. The
Jiu zhang suanshu formed the basis for the development of Chinese
mathematics and had a powerful influence on its later development.
The book begins with four problems on finding the area of a rectangular
field, and then turns to fractions. This is an important section with explanations
of how to reduce a fraction to its lowest terms and how to add, subtract,
multiply, and divide with fractions. Furthermore, this section contains problems
on the averaging of fractions and the multiplication of mixed numbers. The
treatment of the common fraction in the Jiu zhang suanshu is valuable not only
because it is systematic and extensive but also because it represents the earliest

24
There is no existing description of how the ancient Chinese added or subtracted with
rod numerals. However, the procedure for multiplication involves addition and that for
division involves subtraction. If we infer the procedures for addition and subtraction from
them, these are then similar to those described in the Arabic texts. See Lam and Ang,
Fleeting Footsteps, pp. 4547; Li Yan and Du Shiran, Chinese Mathematics: A Concise
History, translated by John N. Crossley and Anthony W. C. Lun (Oxford: Clarendon Press,
1987), pp. 1213.
25
Li Yan and Du Shiran, Chinese Mathematics, p. 33.
known record of the manipulation of fractions using methods which are
generally still followed today.
The ancient Chinese were able to develop fractions because they used a
simple rod numeral notation to represent the difficult concept of a fractional part
in relation to its whole. This notation originated from the process of division;
there is a description of it in the Sun Zi suanjing (see Figure 1, above). To give
an illustration, let us consider the division of 6565 by 9. The procedure is the
same as the division of 6561 by 9 (described above) except for the last step
which appears as follows:
This shows that the answer is 729
4
9
and the notation for
4
9
is
.
The rod numeral notation for a fraction has a very simple form: the numeral
of the fractional part is placed above the numeral representing the whole. The
notation manifests that the positioning of the numerals relative to each other is
once again an essential factor in the art of reckoning with rods.
A cursory study of the Jiu zhang suanshu would reveal that after the four
fundamental operations of arithmetic, fractions and their operations are essential
knowledge for the development of arithmetic. Two-thirds of the two hundred
and forty-six problems are involved with fractions.
The rod numeral system proved to be an extremely useful invention and
played an important role in the advancement of Chinese civilization. Whenever
computation was needed, the rods would be deployed; the numeral system was
therefore known to a wide circle of people ranging from astronomers,
administrators, and engineers to Buddhist monks and commercial travelers. The
use of the system met the demands of such practical needs as finding the area of
a piece of land, the right amount in an exchange of goods, the amount of tax to
be paid, the interest on an amount of money borrowed, the distribution of goods
among a certain number of people, and so on. The rod numerals were also used
in scientific calculations such as those involving the celestial bodies or land
surveying. Furthermore, the numeral system allowed for computations in a
variety of measures.
The rod numeral system generated a series of mathematical methods:
addition and subtraction led to multiplication and division; the last led to a
simple notation of the common fraction which in turn led to methods on
fractions; the computation of the area of a square led to the search for a method
of finding the side of a square whose area was known, and, when that was
invented, it in turn was extended to the method of finding the cube root of a
number. The method of computing an amount to be received in a barter trade
established the Rule of Three (see below) which in turn generated other methods
in proportion. These went onthe existing treatises up to the sixteenth century
provide us insight into the continuous development of traditional mathematics
through the rod numeral system.
The exchange of goods in the ancient world led to the use of a method
commonly referred to as the Rule of Three or the Golden Rule. The Chinese
called it jin you ; the Jiu zhang suanshu devotes an entire chapter to it and
its applications. Problems involving partnership and sharing were very popular,
as were problems employing various methods of proportion and inverse
proportion. The subject of arithmetic arose from the needs of quotidian life and
in turn contributed to the advancement of the quality of life. Methods were
formulated and illustrated by problems of a practical nature. Such problems now
provide a rich source for the study of the socioeconomic aspects of life in
ancient China. The first problem of chapter six in the Jiu zhang suanshu is
translated here as an illustration.
Now there is a fair [way of] transporting millet: County A has 10,000 households
and [requires] 8 days journey to reach the destination; County B has 9,500
households and [requires] 10 days journey; County C has 12,350 households
and [requires] 13 days journey; County D has 12,200 households and [requires]
20 days journey. The 4 counties transport a total of 250,000 hu as tax and use
10,000 carts. It is desired that the contribution be based on the distances and the
number of households. Find the amount of millet and the number of carts from
each [county].
The answer says: County A: 83,100 hu of millet, 3,324 carts. County B: 63,175
hu of millet, 2,527 carts. County C: 63,175 hu of millet, 2,527 carts. County D:
40,550 hu of millet, 1,622 carts.
The method of jun shu (fair transportation) says: Divide the number of
households of each county by the number of days for the journey to obtain the
proportional parts. The proportional part for A is 125, for B and C 95 each, for D
61. Duplicate them and add up [one set] to form the fa . Multiply [each nu-
meral of the set] not added by the given number of carts to transport the millet
and let each [product] be the shi . Divide the shi by the fa to obtain the number
of carts for each [county]. When there are fractions, convert them to integers
[either zero or one] (shang xia bei zhi ). Multiply the number of carts
by 25 hu to obtain the amount of millet.
The Jiu zhang suanshu stands out among ancient mathematical works for its
extremely rich variety of subject matter.
26
There are problems concerning areas
and volumes of different shapes interweaved with the computation of
manpower; others deal with the subject of relative distance and relative speed. A

26
For a more detailed analysis of the Jiu zhang suanshu, see Lam Lay Yong, Jiu
Zhang Suanshu (Nine Chapters on the Mathematical Art): An Overview, Archive for
History of Exact Sciences vol. 47, no. 1 (1994), pp. 151.
whole chapter is devoted to what is called surplus and deficit, which involves
the Rule of False Position, and another chapter treats the right-angled triangle.
All computations were performed with rodsthe use of the rod numeral system
exhibited its innate potential as it provided the tool for ideas to be expressed and
thus for mathematics to be developed from stage to stage. Manipulations of rod
numerals on the board led to new techniques such as cross-multiplication and
other operations on rows and columns. The complex notion of what is now
called a linear equation was encapsulated in a notation on the counting board;
this in turn led to the expression of a set of simultaneous linear equations and
their solution, which in its turn led to the concept of negative numbers.
27
Since the solution of a problem was derived through the manipulation of rod
numerals, the method written in the text could only be at best a detailed
description of the rod numeral operations. The Jiu zhang suanshu and all other
mathematical treatises record numbers by means of a written number system.
Although this number system has some unique features
28
and served its purpose
of recording numbers accurately, it does not possess the place value notation.
The Chinese did not use it for computation.
We now turn our attention to the beginnings of arithmetic through the
Hindu-Arabic numerals. According to Saidan, prior to the introduction of the
Hindu-Arabic numeral system, addition, subtraction, and multiplication in the
Arabic countries depended largely on mental calculation and finger reckoning.
29
As for computation in terms of fractions, there were two major systems. The
first followed from the fact that the Arabic language contained single-word
names for nine fractions onlynamely, the unit fractions from one-half to one-
tenthand that the other fractions were expressed by combinations of these
nine words. The second system circumvented fractions by resorting to
submultiples in the various metrologies or to using the sexagesimal scale.
30
From the evidence of the three Arabic manuscripts mentioned above, after
the Arabs learned the step-by-step procedure of the first set of methods for
addition, subtraction, multiplication, and division, they automatically inherited
from the division method the notation for the expression of a common fraction.
The numeral for 729
4
9
was written as
31

27
Lam Lay Yong and Ang Tian Se, The Earliest Negative Numbers: How They
Emerged From a Solution of Simultaneous Linear Equations, Archives Internationals
dHistoire des Sciences vol. 37 (1987), pp. 23541.
28
Lam and Ang, Fleeting Footsteps, pp. 1219.
29
Saidan, The Arithmetic of Al-Uqldis, pp. 382, 4034.
30
31
See above for the notation of 143
136
324
in the Cambridge manuscript.
729
4
9
The initial notation for a fraction, such as
4
9
, was the same as that in rod
numerals with no horizontal line across.
32
With the use of this symbol,
al-Uqldis showed the various operations related to fractions; fractions were for
the first time freed from linguistic and metrological restrictions. The operations
are the ones we do today through the use of this symbol; they include the
handling of mixed numbers and improper fractions. Al-Uqldis and Kshyr
ibn Labbn also described the procedures for square and cube root extractions,
procedures which are structurally similar to those of the Chinese.
Both the Cambridge manuscript and the treatise of al-Uqldis began by
attributing the new numerals and arithmetic to the Indians. Saidan states that his
study was unable to answer in full how Hindu arithmetic reached the
Arabic-speaking world. He goes on to say, It seems plausible that it drifted
gradually, probably before the seventh century, through two channels, one
starting from Sind, undergoing Persian filtration and spreading in what is now
known as the Middle East, and the other starting from the coasts of the Indian
Ocean and extending to the southern coasts of the Mediterranean. He
continues, Whatever the case may be, it should be pointed out that Arabic
works give no reference whatsoever to any Sanskrit text or Hindu arithmetician,
nor do they quote any Sanskrit term or statement.
33
There are no descriptions of the Hindu-Arabic numeral system and the
fundamental operations of arithmetic among the early Hindu treatises. With the
exception of the Bakhshali Manuscript, whose date is controversial,
34
the
treatises do not use the Hindu-Arabic numerals to represent numbers. Rather,

32
Florian Cajori, A History of Mathematical Notations; vol. 1: Notations in Elementary
Mathematics (London: Open Court, 1928), p. 269; Saidan, The Arithmetic of Al-Uqldis, p.
60.
33
34
The manuscript, written on birch bark, was found at Bakhshali on the northwest
frontier of India in 1881. According to G. R. Kaye, The Bakhshali Manuscript (New Delhi:
Cosmo Publications, 1927, reprint 1981), p. 69, 7484, R. Hoernle suggested that the work
was written in the early centuries of our era, but Kaye himself put forward arguments to
support the dating to the twelfth century. Bibhutibhusan Datta and Avadhesh N. Singh,
History of Hindu Mathematics; part 1: Numeral Notation and Arithmetic (Bombay: Asia
Publishing House, 1935, single vol. ed. 1962), p. 61, dated it at ca. 200 AD and R. Sarkar,
The Bakhshali Manuscript, Ganita Bharati vol. 4 (1982), p. 52, dated it not later than the
fourth century A.D. See also L. V. Gurjar, Ancient Indian Mathematics and Vedha (Poona:
Ideal Book Service, 1947), p. 52; George G. Joseph, The Crest of the Peacock:
Non-European Roots of Mathematics ( London: Tauris, 1991), p. 241.
numbers are generally written in Sanskrit in a terse stanza form. The
Aryabhatiya, written by Aryabhata (b. 476 AD), contains a description of an
alphabetic notation for numerals.
35
In their discussion of the early Hindu
treatises, commentators and scholars have in general assumed that the medium
of computation is through the Hindu-Arabic numeral system. Datta and Singh
have pointed out that calculations were performed on a board called pati with a
piece of chalk, or on sand spread on the ground.
36
Besides the numerals of the
Bakhshali Manuscript, the earliest appearance in India of a symbol for zero in
the Hindu-Arabic numeral system is found in an inscription at Gwalior which is
dated 870 AD.
37
Datta and Singh probably had the Bakhshali Manuscript in
mind when they mentioned that from very early times (c. 200 AD) the Hindus
wrote fractions just as we do now, but without the dividing line.
38
According to al-Uqldis, the first set of arithmetical operations were
performed on the dustboard or takht; from then onwards and throughout the
middle ages, Arabic arithmeticians were kept busy modifying and extending the
operations to fit into their current mathematical system and to suit a paper and
ink medium.
39
During the period 1200 to 1600, the peoples of Europe
assimilated the Hindu-Arabic numeral system and its accompanying package of
computations in written form. This phenomenal adoption of a new numeral
system sprang not from a need for a new numeral system, but from a need to
learn computations through the new numeral system. Theretofore, the systems
of computation used in Europe had been restricted and difficult.
Knowledge of these computations generated new mathematical methods,
and indeed was essential to the progress of humankind. The early development
of these methods and their related topics were of a nature similar to those in the
Jiu zhang suanshu. Books on the new arithmetic grew exponentially in Europe:
from some thirty books in the year 1500 to a few hundred in the sixteenth
century and well over a thousand in the seventeenth century.
40
Since then
arithmetic has been continuously regarded as essential throughout the world
despite the advances in science and technology.

35
Kripa S. Shukla (ed.), Aryabhatiya of Aryabhata (New Delhi: Indian National
Science Academy, 1976), pp. 35; S. N. Sen, Aryabhatas Mathematics, Bulletin of the
National Institute of Sciences of India no. 21 (1962), pp. 298305.
36
Datta and Singh, History of Hindu Mathematics, p. 8.
37
Joseph Needham, Science and Civilisation in China; vol. 3: Mathematics and the
Sciences of the Heavens and the Earth (Cambridge: Cambridge Univ. Press, 1959), p. 10.
38
Datta and Singh, History of Hindu Mathematics, p. 188.
39
Saidan, The Arithmetic of Al-Uqldis, pp. 1314, 247.
40
Karpinski, The History of Arithmetic, pp. 6872.
The Essential Features
Three similar features in the early developments of Chinese and Hindu-Arabic
arithmetic have been identified. These were basic: the growth of both
arithmetics depended on them. First, both arithmetics were built on the
foundation of a numeral system which had a place value notation with ten as
base. Second, both arithmetics began with identical arithmetical procedures for
multiplication and division. Third, both arithmetics employed identical notations
for the complex concept of a common fraction.
The rod numeral system was the only numeral system of antiquity which
used a place value notation with ten as base.
41
Those who used such a system
needed to remember only nine signs; furthermore, the unique notation provided
the means to discover a series of new computations and mathematical methods.
The importance of this notation has been aptly described by Whitehead in his
reference to the Hindu-Arabic numeral system:
The interesting point to notice is the admirable illustration which this numeral
system affords of the enormous importance of a good notation. By relieving the
brain of all unnecessary work, a good notation sets it free to concentrate on more
advanced problems, and in effect increases the mental power of the race. Before
the introduction of the Arabic notation, multiplication was difficult, and the
division of integers called into play the highest mathematical faculties.
42
Both numeral systems followed the same convention of arranging the digits
of a numeral from left to right in descending order of their ranks, beginning with
the highest. The ancient Chinese evolved the system through the use of rods,
and each of the nine signs was formed by at most five rods. When a rod numeral
had no digit of a particular rank, the space representing that rank in the numeral
was naturally left blankthis created the concept of what we now called zero. It
is generally accepted that the nine signs of the Hindu-Arabic numeral system
originated from the first nine signs representing numbers one to nine of the
Brahmi numerals. However, it was not uncommon for peoples of different
countries to substitute the signs with symbols that they were more familiar or
comfortable with.
43
Should the concept of the two numeral systems be radically changed, not
only would their respective arithmetics collapse, but their applications would

41
42
Alfred N. Whitehead, An Introduction to Mathematics (New York: Oxford Univ.
Press, 1911, third printing 1961), pp. 3940.
43
Saidan, The Arithmetic of Al-Uqldis, pp. 186, 310, 360; David E. Smith and
Jekuthiel Ginsburg, Numbers and Numerals (New York: Teachers College, Columbia
Univ., 1937), p. 20.
also be eradicated. The enormous significance of the place value notation with
ten as base in the numeral system cannot be overemphasized.
The ancient Chinese evolved the rod numeral system for computation, and
the Hindu-Arabic numeral system was adopted by peoples all over the world for
the same purpose. No other numeral system had such a notation that enabled the
four fundamental operations of arithmetic to be performed with such relative
ease. The phrase relative ease is used only in a comparative sense; when the
new arithmetic was introduced into Europe, multiplication and division were
taught at university level and, according to Karpinski, for several centuries one
who could perform long division was considered an expert mathematician.
44
The notation of the numeral system had the potential for the human mind to
devise numerous ways of adding, subtracting, multiplying, and dividing. These
methods are found in the existing records on early arithmetical procedures in the
Islamic world and Europe.
45
Therefore one might not expect the same step-by-
step procedures for multiplication and division with the rod numerals and the
Hindu-Arabic numerals. And yet, the earliest methods in the Arabic texts are the
same as those described in the Sun Zi suanjing. These methods were described
against totally different cultural backgrounds separated by a span of more than
four centuries. This unexpected similarity has provided strong evidence that the
Hindu-Arabic numeral system has its origins in the rod numeral system.
46
Both arithmetics began with the same arithmetical procedures. History has
shown that they provided the foundation for the unleashing of a vast number of
methods which not only sustained the growth of arithmetic but also helped to
pave the way into the realm of algebra.
From the outgrowth of identical methods in division, the two arithmetics
used a simple, elegant notation to represent the complicated concept of a
common fraction. This notation allowed for the development of arithmetic to the
fullest, for the human mind could now compute not only with whole numbers
but also with fractions which were fully and simply expressed and were
independent of metrological units. A study of the struggle in the ancient
civilizations of Egypt, Mesopotamia, and Greece to depict the concept of a

44
Karpinski, The History of Arithmetic, p. 120.
45
See e.g., Henry T. Colebrooke, Algebra, with Arithmetic and Mensuration: From the
Sanscrit of Brahmegupta and Bhascara (London: John Murray, 1817); Georgij de
Hungaria, Arithmeticae Summa Tripartita. Facsimile with introduction by A. J. E. M.
Smeur (1965) (Nieuwkoop, Holland: B. de Graaf, 1499); Baldassarre Boncompagni, Liber
Abbaci di Leonardo Pisano (Rome: Tipografia delle Scienze Matematiche e Fisiche, 1857);
Louis C. Karpinski, Two Twelfth Century Algorisms, Isis vol. 3 (1921), pp. 396413;
Graham Flegg, Cynthia Hay, and Barbara Moss (eds.), Nicolas Chuquet, Renaissance
Mathematician (Dordrecht: D. Reidel, 1985); Robert Steele, The Earliest Arithmetics in
English (London: Oxford Univ. Press, 1922).
46
fraction and to invent ways to compute in terms of fractional parts
47
would
further serve to remind us of how, in Whiteheads words, a good notation could
set the human brain free to concentrate on more advanced problems. It can be
said that without the use of this notation to represent a common fraction, the
growth of both arithmetics would have been stunted.
Founded on these three core features, the Chinese arithmetic which had its
origins in antiquity was continuously developed until the seventeenth century.
As for the Hindu-Arabic arithmeticwe still cannot do without it. The progress
of humankind has very often depended on our ability to compute. In the
evolution of computation, the human race had chosen from the Warring States
period to retain two notations for their reckoning: one to express a numeral and
the other a common fraction. Through the use of these two notations, the science
of arithmetic arose. The language of mathematics is in terms of notations, and
these two notations can be said to be among the earliest, the most useful, and the
most powerful.

47
See e.g., Paul Benoit, Karine Chemla, and Jim Ritter (eds.), Histoire de Fractions,
Fractions dHistoire (Basel: Birkuser Verlag 1992), pp. 3168; Richard J. Gillings,
Mathematics in the Time of the Pharaohs (Cambridge, Mass.: MIT Press, 1972), pp. 2038;
Graham Flegg (ed.), Numbers Through the Ages (London: Macmillan, 1989), pp. 13162;
Thomas Heath, A History of Greek Mathematics; vol. 1: From Thales to Euclid (Oxford:
Clarendon Press, 1921), pp. 4145.
55
The Use of the Twenty-eight Xiu as a
Day-Count in Early China
Marc Kalinowski
[Marc Kalinowski is professor of Chinese religion and thought at the Ecole
Pratique des Hautes Etudes. He is the coordinator of the Mengxi bitan
translation project in Paris, and is now preparing a book on the formation of
Chinese calendrical astrology. Among his publications are Mythe, cosmogonie
et thogonie dans la Chine ancienne, LHomme 1271 (1996), Cosmologie et
divination dans la Chine ancienne (Paris, 1991), and La littrature divinatoire
dans le Daozang, Cahiers dExtrme-Asie, vol. 5 (19891990).]
* * *
he origin of the twenty-eight celestial lodges (xiu ) in China remains to
this day a much debated issue. The first complete list of these constella-
tions surrounding the sky may be dated from the second half of the fifth
century B.C.
1
In the astrocalendrical texts of the late Warring States (480222
B.C.) and the Former Han (206 B.C.8 A.D.) periods, the Chinese sidereal zodiac
takes its definite shape as an equatorial belt divided into 365.25 degrees (du )
grouped in twenty-eight unequally spaced lodges. Each lodge is defined by a
determinative star (juxing ) which indicates its first degree (chudu ).
2
The zodiacal position of a given celestial body is noted by the name of the
lodge in which it stands and, eventually, the degree it has reached within this

1
See the representation of the sky with names of the twenty-eight lodges inscribed
on a lacquerware object found in 1978 in the tomb of the Marquis of Zeng; Wang Jian-
min, Liang Zhu and Wang Shengli, Zeng hou Yi mu chutu de ershiba xiu qinglong
baihu tubiao [The twenty-eight lodges, Green Dragon and White Tiger design in the
tomb of the Marquis Yi of Zeng], Wenwu no. 7 (1979), pp. 4045.
2
For a general presentation of the system of the xiu, see Joseph Needham, Science
and Civilisation in China, vol. 3 (Cambridge: Cambridge Univ. Press, 1959), pp. 229
59; and Chen Zungui, Zhongguo tianwenxue shi [A history of Chinese astronomy]
(Shanghai: Renmin chubanshe, 1982), pp. 30584.
T
lodge. Thus, in Han times, the sun was to be found at the winter solstice be-
tween the end of Dou
8
and the beginning of Niu
9
. Fifteen days later, it had
moved by fifteen du eastwards, reaching the eighth du in N
10
. The same obser-
vation could be repeated every fifteen days, all along the twenty-four divisions
of the solar year.
3
Owing to the phenomenon of the precession of equinoxes, the
sidereal zodiac shifts westwards slightly every year. For example, the winter
solstice moved from Niu
9
to Dou
8
around Han times. It was still in Dou
8
during
the Tang dynasty (618907) and is now in Ji
7
. Thus, all the calendrical treatises
in the official histories contain a table giving the sidereal position of the sun for
the twenty-four solar periods (jieqi .
An important function of these tables was to regulate the adjustment of the
twelve lunar months (12 x 29.5 = 354 days) to the tropical year (365.25 days).
According to the lunar-solar standard of the Chinese calendar, the lunation con-
taining the winter solstice was counted as the eleventh month of the year, that
containing the spring equinox was the second month of the year, that containing
the summer solstice was the fifth month of the year, and that containing the
autumn equinox was the eighth month of the year. Thus, since the sun was
standing in Dou
8
at the winter solstice in Han times, lodge Dou
8
was defined as
the solar station (richan ) of the eleventh month. This applied also to
other months of the year, each of which was attached to a specific solar station
(or lodge).
4
Strangely enough, Han calendrical literature has no table giving sidereal lo-
cations for the moon. Though the length of the sidereal month (27.3 days) was
well known, the texts simply mention that the moon moves thirteen degrees
eastwards every day without assigning any sidereal location for its daily move-
ment.

3
To facilitate the identification of the lodges, we have added to their names numbers
as subscripts indicating their positions within the series (Jiao
1
, Kang
2
, ..., Zhen
28
). The
first complete list of zodiacal positions of the sun for the twenty-four solar periods is
found in the Three Revolutions Calendar by Liu Xin (c. 32 B.C.23 A.D.). According to
Liu Xins system, the solar station for the winter solstice was Dou
8
(12th du), the solar
station for the spring equinox was Lou
16
(4th du), the solar station for the summer solstice
was Jing
22
(31st du) and the solar station for the autumn equinox was Jiao
1
(10th du); see
Hanshu, Llizhi 2, in Lidai tianwen lli dengzhi huibian [A comprehensive collection
of the treatises on astronomy and calendar-making in Official Histories] (Beijing:
Zhonghua shuju, 1976), pp. 143132. For the zodiacal position of the winter solstice
during the Han dynasty and its variation in later times, see Li Jiandeng, Suicha zai
woguo de faxian, ceding he lidai dongzhi suozai de kaozheng [A study of the evolution
of the winter solstice position in relation with the discovery and calculation of the pre-
cession of equinoxes], Zhongguo tianwenxue shi wenji vol. 3 (1984), pp. 12437.
4
The keying of the twelve lunations to the twenty-four solar periods is described in
the Hanshu (Llizhi 2, Dengzhi huibian, p. 1409). For the role played by solar stations
in Chinese astronomy, see Chen Zungui, Zhongguo tianwenxue shi, pp. 72224.
Marc Kalinowski: Use of the Twenty-eight Xiu as a Day-Count 57
* * *
Such were the basic applications of the twenty-eight xiu sidereal zodiac in
early Chinese astronomy and calendar making. In this paper, we intend to draw
attention to the specific use of the twenty-eight xiu as a counting device for the
days of the month. It is well known that there were two standard ways of nam-
ing days in premodern China. One used numerals to indicate the position of
days within the month: day one for the new-moon (shuo , day fifteen for the
full-moon (wang ), day twenty-nine or thirty for the last day. The second
worked with the sexagenary cycle, using its sixty units as an ever recurrent day-
count, without any reference to the sun and moon at all: jiazi1 day, yichou2 day,
bingyin3 day, and so on.
5
A third system, based on the twenty-eight xiu and supplemental to the two
other systems, appears to have been of common use in calendars and almanacs
starting with the Southern Song (11271279) period. The first part of this article
is devoted to the description of this system, drawing on contemporary and later
sources. We will find that it had no relation at all with the actual movements of
the sun and moon and that it was simply carrying on the seven-day week of
Western origin current in Chinese calendars under the Tang dynasty.
In the second part of the article, we will show that during the same period
there concurrently existed another day-count by means of the twenty-eight xiu.
The procedure for counting in this system is thoroughly described in various
books influenced by Indo-Persian astrology that are included in the Buddhist
canon. Unlike the Southern Song system, lunation played an important role. To
each month was attached one lodge which represented the idealized sidereal
position of the moon on its fifteenth day. It was called wangxiu, full-moon
lodge. The other lodges were allocated to days within each month in an orderly
way, starting with the wangxiu and proceeding in both directions, adding or
subtracting one lodge for each day elapsed. The system does not seem to have
been used beyond this type of astrology popular in China during the Tang.
It is usually thought that the existence of a twenty-eight xiu day-count before
that period is highly hypothetical. The last part of the article is concerned with
tracing back the use of the twenty-eight xiu as a day-count to earlier times. Evi-
dence will be drawn from the divinatory texts in the Daoist canon, the official
histories, the diviners board from the Han dynasty, and hemerological manu-
scripts from Shuihudi. We shall demonstrate that a twenty-eight xiu day-count
flourished during the late Warring States and early Han period. This ancient
system operated in a way quite similar way to the one derived from Indo-Persian
astrology, the main difference being that the twelve monthly lodges represented

5
The number attached to each sexagenary unit indicates its position within the se-
ries.
the first day of each month (shuoxiu, new-moon lodge) instead of the fifteenth
(wangxiu).
The conclusions to be drawn from this study will lead us to consider if the
early use of the twenty-eight xiu as a day-count may not provide us with an ad-
ditional argument for the existence in China of a primitive lunar zodiac based on
a regular division of the celestial sphere in twenty-seven or twenty-eight aster-
isms following the twenty-seven or twenty-eight days of the sidereal month.
The Song System
The twenty-eight xiu day-count may be found in official calendars since the
Southern Song.
6
The compilers of the Xieji bianfang shu (1741)
devoted an entire section to explaining the system. It begins with a list of the
twenty-eight xiu and other associated elements, followed by a method for ap-
plying the system to the calendar. Finally, the section discusses the origins of
this day-count.
7
We find two features associated with the twenty-eight xiu in the list. Firstly,
there are the seven luminaries (qiyao ) used to name the days of the week:
the Sun (Su): Sunday; the Moon (Mo): Monday; Mars (Ma): Tuesday; Mercury
(Me): Wednesday; Jupiter (Ju): Thursday; Venus (Ve): Friday; Saturn (Sa): Sat-
urday. Secondly, there is a group of twenty-eight animals which includes the
twelve zoomorphic signs linked to the twelve branches since the Han dynasty.
8
These twenty-eight animals are themselves part of a larger group of thirty-six.
The following table lists the number representing the position of each animal in
the series of thirty-six (discussed below).
9

6
For a general outline of Chinese calendars and almanacs, see Richard J. Smith, A
Note on Qing Dynasty Calendars, Late Imperial China vol. 9, no. 1 (1988), pp. 12345;
and Chen Zungui, Zhongguo tianwenxue shi, pp. 15871669.
7
Xieji bianfang shu (Siku quanshu zhenben ed.), 1.24a28b. See also Xingli kaoyuan
(1713; Siku quanshu zhenben ed.), 1.40b42b and 5.8a9a.
8
Early evidence for the twelve zoomorphic signs is found in the Shuihudi manu-
scripts from the late Warring States (Marc Kalinowski, Les traits de Shuihudi et
lhmrologie chinoise la fin des Royaumes Combattants, Toung Pao vol. 72 [1986],
pp. 2089), and in the Fangmatan manuscripts from the early Han dynasty (He
Shuangquan et al., Tianshui Fangmatan Qinjian jiazhong Rishu shiwen [Transcription
of the Book of Days, version A, found at Fangmatan], in Archeological Research Insti-
tute of Gansu Province [ed.], Qinjian jiandu lunwenji [Lanzhou: Gansu renmin chuban-
she, 1989], pp. 34).
9
The translations given here are tentative. For the Chinese characters and more in-
formation, see Michael Loewe, Ways to Paradise: The Chinese Quest for Immortality
(London: George Allen and Unwin, 1979), p. 207; also Marc Kalinowski, Cosmologie et
divination dans la Chine ancienne: Le compendium des cinq agents (Wuxing dayi, VIe
Jiao
1
- Ju - Dogfish
2
Kui
15
- Ju - Wolf
20
Kang
2
- Ve - Dragon
3
(chen ) Lou
16
- Ve - Dog
21
(xu )
Di
3
- Sa - Badger
4
Wei
17
- Sa - Pheasant
22
Fang
4
- Su - Hare
5
(mao ) Mao
18
- Su - Cock
23
(you )
Xin
5
- Mo - Fox
6
Bi
19
- Mo - Crow
24
Wei
6
- Ma - Tiger
7
(yin ) Zui
20
- Ma - Monkey
25
(shen )
Ji
7
- Me - Leopard
8
Shen
21
- Me - Ape
26
Dou
8
- Ju - Unicorn
11
Jing
22
- Ju - Elk
29
Niu
9
- Ve - Bull
12
(chou ) Gui
23
- Ve - Sheep
30
(wei )
N
10
- Sa - Bat
13
Liu
24
- Sa - Stag
31
Xu
11
- Su - Rat
14
(zi ) Xing
25
- Su - Horse
32
(wu )
Wei
12
- Mo - Swallow
15
Zhang
26
- Mo - Deer
33
Shi
13
- Ma - Pig
16
(hai ) Yi
27
- Ma - Snake
34
(si )
Bi
14
- Me - Mongoose
17
Zhen
28
- Me - Worm
35
This threefold list seems to have been generally used under the Ming (1368
1644) and Qing (16441911) dynasties. It appears in several books on astrologi-
cal horoscopy and popular almanacs from this period.
10
Henri Dor mentions it
in the section of his Recherches sur les superstitions en Chine devoted to astral
cults.
11
In any event, the leading elements in the list are the twenty-eight xiu
with the animals and luminaries playing a supporting role.
This system is applied to the calendar using the sexagenary day-count.
12
The
jiazi1 day is connected to lodge Xu
11
, the yichou2 day to lodge Wei
12
, the
bingyin3 day to lodge Shi
12
, etc. The twenty-ninth day (renchen29) will again be
a Xu
11
day. After 420 days (fifteen xiu-cycles and seven sexagenary cycles),
lodge Xu
11
will return to its initial position on jiazi1.

sicle) (Paris: Ecole Franaise dExtrme-Orient, 1991), pp. 43747. The twelve
branches have been connected here to the twelve zoomorphic signs for more conven-
ience.
10
Yanqin tongzuan (Siku quanshu zhenben ed.), 1.6b7a; Zhang Guo
xingzong (Gujin tushu jicheng ed.), 567.2a2b; Qinxing yijian (Siku
quanshu zhenben ed.), 7a7b; Carole Morgan, Le tableau du boeuf du printemps: Etude
dune page de lalmanach chinois (Paris: Mmoires de lInstitut des Hautes Etudes Chi-
noises, vol. 14, 1980), p. 117.
11
Part 2, vol. 12 (Shanghai, 1918), pp. 11901210. Dor takes the names of the
planets (Wood for Jupiter, Fire for Mars, etc.) as referring to the five agents and therefore
does not identify the series as being that of the seven luminaries.
12
On the characteristics of the sexagenary day-count, see Needham, Science and
Civilisation, vol. 3, pp. 39698.
Therefore, the system consists of an ever-recurrent cycle of 420 days, with-
out any reference to the sun (year) or the moon (month) at all. The seven sexa-
genary cycles comprising the system are called the seven epochs (qiyuan
). Lodges connected to the first day of each epoch are:
First epoch: Xu
11
(Su - Rat
14
)
Second epoch: Kui
15
(Ju - Wolf
20
)
Third epoch: Bi
19
(Mo - Crow
24
)
Fourth epoch: Gui
23
(Ve - Sheep
30
)
Fifth epoch: Yi
27
(Ma - Snake
34
)
Sixth epoch: Di
3
(Sa - Badger
4
)
Seventh epoch: Ji
7
(Me - Leopard
8
)
Let us now examine the extant calendars in order to ascertain whether the
theory outlined in the Xieji bianfang shu is borne out.
13
The absence of the
twenty-eight xiu day-count in the late Tang and Northern Song calendars found
in Dunhuang suggests that the system was not in use during that period.
14
We
have chosen four calendars ranging from the Southern Song to the eve of the
Republican era. In each case, we have selected a day corresponding to the be-
ginning of a first epoch (jiazi1 day, lodge Xu
11
) in order to see if the system was
consistent throughout the entire period.
Calendar for the 4th year of Baoyou (1256)
15
day: 8th month, 6th day, jiazi1, lodge Xu
11
.
First epoch, 1st day.
Gregorian date: August 27, 1256. Sunday.
Julian day: 218 0051

13
The system of the seven epochs is also described in the Qinxing yijian 8a9a, and
in the Yanqin tongzuan 1.6b7a. Earlier evidence is provided by a Yuan dynasty alma-
nac; see Yan Dunjie, Ba Dunhuang Tang Qianfu sinian lishu [Postface to an almanac
dated 877 A.D. found in Dunhuang], in Zhongguo gudai tianwen wenwu lunji (Beijing:
Wenwu chubanshe, 1989), p. 248.
14
See the comparative table for the whole collection of Dunhuang calendars in Xi
Zezong and Deng Wenkuan, Dunhuang canli dingnian [Dating the extant Dunhuang
calendars], Zhongguo lishi bowuguan guankan vol. 12 (1989), pp. 2021; and in Huang
Yinong [Huang Yi-Long], Dunhuangben juzhu liri xintan [A new examination of
Dunhuang calendars], Xinshi xue vol. 3, no. 4 (1992), pp. 3637.
15
Da Song Baoyou sinian bingchen sui huitian wannian juzhuli yijuan (Wanwei
congshu ed.), 17a. For the conversion into Gregorian dates, we have used the Zhong-
guo-shi liri he zhongxi liri duizhao biao [Chinese historical chronology and its concor-
dance with the Western calendar] by Fang Shiming and Fang Xiaofen (Shanghai: Cishu
chubanshe, 1987).
Calendar for the 25th year of Zhizheng (1365)
16
day: 7th month, 1st day, dingsi54, lodge Liu
24
.
1st day of the current epoch: 5th month, 7th day, jiazi1, lodge Yi
25
(fifth epoch).
1st day of the closest first epoch: 10th month (intercalary), 10th day,
jiazi1, lodge Xu
11
.
Gregorian date: November 23, 1365. Sunday.
Calendar for the 60th year of Qianlong (1795)
17
day: 1st month, 5th day, wuzi25, lodge Xu
11
.
1st day of the current epoch: 59th year, 12th month, 11th day, jiazi1,
lodge Kui
15
(second epoch).
1st day of the closest first epoch: 59th year, 10th month, 10th day, ji-
azi1, lodge Xu
11
.
Calendar for the 1st year of Xuantong (1909)
18
day: 9th month, 18th day, jiazi1, lodge Xu
11
.
First epoch, 1st day.
Gregorian date: October 31, 1909. Sunday.
Verification
221 9951 218 0051 = 39 900 (420 x 95)
237 6611 221 9951 = 156 660 (420 x 373)
241 8611 237 6611 = 42 000 (420 x 100)
From the above data, it is clear that the twenty-eight xiu day-count has re-
mained unchanged since its adoption in calendars of the Southern Song period.

16
See Zhang Peiyu, Shilun xin faxian de sizhong guli canjuan [A tentative discus-
sion of four ancient calendars recently unearthed], Zhongguo tianwenxue shi wenji vol. 5
(1989), pp. 12325. Since this calendar is incomplete, we have chosen a day from its still
extant part and calculated the corresponding first day of the first epoch.
17
See Chen Zungui, Zhongguo tianwenxue shi, p. 1620. Since Chen only reproduces
the first month of this calendar from the Beijing Observatory collection, the first day of
the first epoch has been calculated according to the data appearing in his reproduction.
18
This is a calendar from the Maspero collection in the Library of the Socit Asi-
atique in Paris (Fonds Henri Maspero no. 1399).
The question can be asked: when and how was this system elaborated? Accord-
ing to the authors of the Xieji bianfang shu, its origin is to be found in astrologi-
cal texts brought into China from the West during the Tang period. Before
turning to this literature, let us first review the two other elements associated
with the twenty-eight xiu in the Xieji bianfang shu.
The Thirty-six Animals
Besides its strictly calendrical function, the series of twenty-eight animals
also appears in several mantic treatises influenced by Indo-Persian astrology.
19
In certain texts, we may find the animals included in a larger group of thirty-six
animals. But, in all of these sources, there is always a list marking the usual
connections among animals, lodges and luminaries.
The series of thirty-six animals is itself linked to the calendar through the
twelve branches and the twelve zoomorphic signs connected to them since Han
times. With each branch are associated three animals and, in each of these
groups, there are corresponding zoomorphic signs: zi = rat (zoomorphic sign),
bat, swallow; chou = bull (zoomorphic sign), crab, unicorn; and so on.
The earliest mention of the thirty-six animals is found in the Baopuzi
of Ge Hong (284363 A.D.).
20
The reference consists of a list of twelve days
named by using the twelve branches. Each day is associated with two or three
animals representing spirits that threaten the adept in his travels in mountains,
forests and other wild places. These animals are virtually the same as those in
later series. The Wuxing dayi (late sixth century) gives the full set of
thirty-six animals, with variants quoted from different sources, indicating that
several versions of the series still existed at that time.
21
This book, as well as the
Baopuzi, agree in assigning to the thirty-six animals the function of dividing the
day into three periods (morning, noon, evening). In the Mohe zhiguan
, a Buddhist text of the fifth century, we also find the complete set of
thirty-six, but associated with the twelve hours of the day (three animals per
hour).
22
There is no evidence in these sources of a connection between the ani-
mals and the twenty-eight xiu.
However, a close connection between the animals and the xiu is made in
divination, and more specifically in the structure of the diviners board (shi )
used in the Liuren method of calendrical astrology. One of these boards,

19
See above, footnote 10. An earlier reference might be the Yanqin doushu sanshi
xiang , since it is attributed to the famous Tang dynasty diviner Yuan
Tiangang . Two Song dynasty copies of that book are still available today.
20
Chap. Dengshe (Xin zhuzi jicheng ed.; Taibei: Shijie shuju, 1978), p. 79.
21
See Kalinowski, Divination et cosmologie, pp. 1078 and 43747.
22
Taish no. 1911, juan 8c, p. 115.
dating from the Six Dynasties (222589 A.D.), has been preserved to this day.
23
On the square-plate of the instrument, we find three groups of inscriptions: the
thirty-six animals on the outer band, the twenty-eight xiu on the middle band,
and the stems and branches on the inner ring. The square is divided diagonally
into four sections, so that there are nine animals, seven xiu, and five stems and
branches in each section, forming a three level pyramidal structure. Let us ex-
amine one of these sections (eastern quadrate):
24
(southeast corner)
Fish
1
Leviathan
2
Jiao
1
Dragon
3
Kang
2
chen (branch 5)
Badger
4
Di
3
yi (stem 2)
Hare
5
Fang
4
mao (branch 4)
Hedgehog
6
Xin
5
jia (stem 1)
Tiger
7
Wei
6
yin (branch 3)
Wildcat
8
Ji
7
Fu
9
(northeast corner)
It is clear that the connection between animals and xiu has been made by
setting apart the eight animals located next to the corners of the square-plate (1-
9,10-18,19-27, 28-36). This is all the more justifiable since, in the Wuxing dayi
and in a Tang commentary to the Mohe zhiguan, the distribution of the thirty-six
animals on the subdivisions of the day is also explained in relation to the posi-
tions they occupy on Liuren boards.
25
We may conclude that the series of thirty-six animals has taken a definite
shape around the fourth or the fifth century. It was used to name periods
(Wuxing dayi) or hours (Mohe zhiguan) of the day. The mantic function of the
animals was closely related to the Liuren method since they appear on Liuren
diviners boards, together with the twenty-eight xiu whose inclusion on the
square-plate of these instruments goes back to the Han period (see below). In-

23
For a general presentation of the Liuren mantic method and diviners boards, see
Marc Kalinowski, Les instruments astro-calendriques des Han et la mthode Liuren,
Bulletin de lEcole Franaise dExtrme-Orient vol. 72 (1982), pp. 309419. The Liuren
board from the Six Dynasties is described on pp. 36871.
24
The translations of animal names are based on Loewes identifications (Ways of
Paradise, p. 207). We have leviathan (jing ), hedgehog (wei ) and wildcat (li )
instead of dogfish (jiao ), fox (hu ) and leopard (bao ) in the threefold list used in
the twenty-eight xiu day-count.
25
Kalinowski, Cosmologie et divination, p. 438 and pp. 59697 (notes 5 to 7).
Zhiguan fuxing , Taish no. 1912, juan 8c, pp. 4078.
creasingly, it became a habit to associate the xiu with the animals facing them
on the Liuren boards, forming a twofold series similar to the ancient twelve
brancheszoomorphic sign correlations. This series was transmitted through
divination using the full set of thirty-six animals which were finally incorpo-
rated into popular almanacs when the twenty-eight xiu day-count came into ef-
fect.
26
The Seven Luminaries
The seven-day week as we know it in the Western calendar since Roman
times was introduced into China during the Six Dynasties via the Silk Road.
27
It
appeared in Tang dynasty calendars under the name seven-luminaries day-
count (qiyao liri ) and half of the Dunhuang almanacs between 798 to 993
used it.
28
The success on the seven-luminaries day-count was such that the com-
pilers of astronomical tables of the early Song period included a section ex-
plaining how to calculate the days of the week.
29
Let us take three samples cho-
sen from among Dunhuang almanacs:
30

26
Among the four calendars quoted above, only the last one (first year of Xuantong)
mentions animals and luminaries. The others merely indicate the lodges day-count, al-
though the threefold list bringing together lodges, animals and luminaries is found in
much earlier astrological literature.
27
On this matter, see Ye Delu, Qiyao li ru Zhongguo kao [An examination of the
introduction of the seven luminaries calendars into China], Furen xuebao vol. 11 (1942),
pp. 13757; and Zhuang Shen, Miri kao [An examination of the Sogdian week],
Zhongyang yanjiuyuan lishi yuyan yanjiusuo jikan (1960), pp. 271301.
28
Xi Zezong and Deng Wenkuan (Dunhuang canli dingnian, pp. 2021) give
thirteen occurrences of the seven luminaries day-count in a list of thirty-four calendars
covering that period. The survey made by Huang Yi-Long (Dunhuangben juzhu liri
xintan, pp. 3637) specifies fifteen occurrences in a list of thirty-eight calendars for the
same period. The two calendars from the Wei dynasty (fifth century) also found in Dun-
huang do not refer to the luminaries.
29
Songshi, Llizhi 2 (Denzhi huibian, p. 2477). This section has been studied by
Yan Dunjie (Ba Dunhuang Tang Qianfu sinian lishu, p. 248). Another method of cal-
culating days of the week has been preserved in the Japanese print of the Xiuyao jing.
According to the study made by Yano Michio (see footnote 32, below), this method was
inspired by the Jiuzhili , a Chinese text on Indian astronomical calculations com-
piled in the eighth century.
30
Since the Dazhong and the Duangong calendars are incomplete, the dates given
here have been recalculated on the basis of the data actually mentioned in the manu-
scripts. The conversion table for Gregorian dates is the same as the one used previously
(see footnote 15, above). Huang Yi-Long has pointed out that, in some of the Dunhuang
calendars, the seven luminaries day-count does not fully match the Western week
(Dunhuangben juzhu liri xintan, pp. 2122).
Calendar for the 12th year of Dazhong (858). S 1439v
day: 9th month, 6th day, jiazi1, sun-day.
Gregorian date: October 16, 858. Sunday.
Calendar for the 2nd year of Duangong (989). P 2705
Calendar for the 3rd year of Yongxi (986). P 3403
Gregorian date: June 6, 986. Sunday.
Unless we suppose that this similarity between the computation of the Chi-
nese and Western seven-day week is pure coincidence, we must admit that the
two systems were technically connected at least from the middle of the Tang
period.
Knowing that the combination of the seven luminaries with the sixty sexage-
simal units makes a cycle of 420 days divided into seven periods of sixty days
each, we can take the first day of each period and draw up a table similar to the
one used in the twenty-eight xiu day-count.
First epoch (jiazi1, Xu
11
day): Sun-day (Sunday)
Second epoch (jiazi1, Kui
15
day): Jupiter-day (Thursday)
Third epoch (jiazi1, Bi
19
day): Moon-day (Monday)
Fourth epoch (jiazi1, Gui
23
day): Venus-day (Friday)
Fifth epoch (jiazi1, Yi
27
day): Mars-day (Tuesday)
Sixth epoch (jiazi1, Di
3
day): Saturn-day (Saturday)
Seventh epoch (jiazi1, Ji
7
day): Mercury-day (Wednesday)
It is clear that the use of the twenty-eight lodges day-count in Southern Song
calendars is a mere duplicate of the seven-day week notation already current in
Tang times. This was achieved by keying both cycles together once and for all.
The Sun-day was connected to middle lodges Xu
11
, Fang
5
, Xing
25
and Mao
18
, the
Moon-day to lodges Wei
12
, Xin
6
, Zhang
26
and Bi
19
, and so on. The jiazi1 day fal-
ling on a Sun-day was chosen as the beginning of the cycle and, thus, lodge Xu
11
became the first day of the first epoch. That is why the four jiazi1Xu
11
days
taken as examples (above) of the twenty-eight xiu day-count in post-Song cal-
endars fall without exception on a Gregorian Sunday. Moreover, since the days
of the week (luminaries) are not mentioned in Southern Song, Yuan and Ming
calendars, we may follow Yan Dunjies conclusion that the late twenty-eight xiu
day-count was understood not only as a duplicate of but also as a substitute for
the seven luminaries system of foreign origin.
31
The Twenty-eight Lodges Day-Count
in Indo-Persian Astrology
The Chinese books on Indo-Persian astrology found in the Buddhist canon in-
clude several sections on the method of connecting the twenty-eight xiu to days
of the month. It is through texts such as the Modeng qiejing
(Mtaga-avadna; second or third century A.D.), the Xiuyao jing (759
and 764) and the Qiyao rangzai jue (865), that Indo-Persian horo-
scopic methods spread into China. They enjoyed great popularity during the
Tang and deeply influenced the later evolution of Chinese divinatory tech-
niques.
32
According to these records, each month (lunation) was connected to one
specific lodge called the full-moon lodge (wangxiu) whose function was to
mark the idealized position of the moon on the fifteenth day of the month (full-
moon day). These twelve wangxiu were as follows:
33
1st month, fifteenth day: Yi
27
2nd month, fifteenth day: Jiao
1
3rd month, fifteenth day: Di
3
4th month, fifteenth day: Xin
5
5th month, fifteenth day: Ji
7
6th month, fifteenth day: N
10

31
Yan Dunjie, Ba Dunhuang Tang Qianfu sinian lishu, p. 248.
32
All these books are contained in the Tantric section of the Chinese Buddhist
canon: Modeng qiejing (Taish no. 1300); Xiuyao jing (Taish no. 1299); Qiyao rangzai
jue (Taish no. 1308). For a general presentation, see Yabuuti Kiyosi, Indian and Ara-
bian Astronomy in China, Silver Jubilee Volume of the Jinbun Kagaku Kenkysho
(Kyoto: 1954), pp. 585603. A more detailed analysis may be found in Zen Hash,
Matogaky no tenmon rekis ni tsuite [On astronomy and ephemerides in the Modeng
qiejing], in Bukky daigaku (ed.), Tygaku rons (Tokyo: Bukky daigaku, 1952), pp.
171213; Yano Michio, The Hsiu-yao ching and its Sanskrit Sources, in G. Swarup, A.
K. Bag and K. S. Shukla (eds.), History of Oriental Astronomy (Cambridge: Cambridge
Univ. Press, 1987), pp. 12534; Yano Michio, The Chi-yao jang-tsai cheh and its
Ephemerides, Centaurus vol. 29 (1986), pp. 2835.
33
Modeng qiejing, juan 2, pp. 40910; Xiuyao jing, juan 1, p. 388 and juan 2, p.
394.
7th month, fifteenth day: Shi
13
8th month, fifteenth day: Lou
16
9th month, fifteenth day: Mao
18
10th month, fifteenth day: Zui
20
11th month, fifteenth day: Gui
23
12th month, fifteenth day: Xing
25
Lodges connected to other days of the month could easily be found by
counting one lodge per day before and after the wangxiu. For example, in the
first month (Yi
27
), the fourteenth day would be a Zhang
26
day, the thirteenth a
Xing
25
day, the sixteenth a Zhen
28
day and the seventeenth a Jiao
1
day.
34
It seems that there is no relation at all between the wangxiu system and the
one seen in calendars beginning with the Southern Song. Firstly, the wangxiu
day-count is related to astronomical events. In the Xiuyao jing, the list of the
twelve wangxiu is preceded by a small section explaining that the system was
conceived in remote antiquity at a time when a conjunction of the sun and moon
(new moon, first day of the month) took place in Lou
16
on the day of the spring
equinox. From this initial situation, repeated observations of the moons posi-
tion on the next twelve full-moon days led to the establishment of the corre-
sponding twelve full-moon lodges.
35
Since the lunation containing the spring
equinox represents the second month of the year according to Chinese calendri-
cal standards, the list comes out just as it has been quoted above: Jiao
1
for the
second months full moon, Di
3
for the third months full-moon, etc. Thus, these
twelve wangxiu were taken to represent the primeval and idealized sidereal po-
sitions of the moon on the fifteenth day of each month and applied to calendrical
day-counting.
Secondly, unlike Southern Song calendars, there is no continuity here in the
distribution of the lodges from one month to the other because of the keying of
the day-count on the twelve wangxiu. In other words, the system does not con-
sist of a continuous cycle of days. On the one hand, the civil month may be of
twenty-nine or thirty days in the Chinese calendar. On the other hand, since the
gap between two consecutive wangxiu is not regular, the lodge connected to the

34
See Xiuyao jing, juan 2, pp. 39495. The Xiuyao jing (juan 1, p. 391; juan 2, p.
395) and the Rangzai jue (juan 1, p. 428) give several examples showing that the system
actually worked that way.
35
Xiuyao jing, juan 1, p. 388. According to Chinese astronomical tables, the suns
position at the spring equinox was in Lou
16
(4th du) during the Former Han period
(Hanshu, Llizhi 2; Dengzhi huibian, p. 1431). Starting from the Later Han calendrical
systems, this position has moved to Kui
15
due to the precession of equinoxes. Both the
Sifen calendar (85236 A.D.) and the Jingchu calendar (237444) locate the spring equi-
nox on the fourteenth du in Kui
15
(Hou Han shu, Llizhi 3, Dengzhi huibian, p. 1532;
Jinshu, Llizhi 3, Dengzhi huibian, p. 1632).
last day of a given month is not necessarily next to the one connected to the first
day of the following month. For example, Niu
9
, the first day lodge of the elev-
enth month (wangxiu Gui
23
), will be separated from the last day lodge of the
tenth month (wangxiu Zui
20
), Wei
6
or Ji
7
, by two or three positions.
Thirdly, the use of the wangxiu in the Xiuyao jing and the Qiyao rangzai jue
is mainly astrological. The twenty-eight lodges day-count appears to be closely
related to the other aspects of Indian lunar and planetary horoscopy, as in the
method for finding the fate lodge (mingxiu ) governing an individuals
birthday: If a person was born on the twenty-eighth day of the eleventh month,
the count would start from the full-moon lodge Gui
23
and Dou
8
would be the fate
lodge of that person.
36
Nevertheless, the diffusion into China of Indo-Persian astrology may have
played a role in the formation of the late twenty-eight xiu day-count. The fact is
that the Song system is based on the seven-luminary week which was introduced
into China via the same Central Asian routes as Indo-Persian astrology.
37
Fur-
thermore, a tendency to normalization already appears in the commentary writ-
ten between 759 and 764 by Yang Jingfeng to the Xiuyao jing.
38
He
notes first how much the wangxiu system differs from the most elementary pro-
cedures of Chinese astronomy and calendar-making. He then suggests a revised
version of the system and puts together a complete list of the lodges connected
to every day of the month for one year. Each month has thirty days and the
lodges follow each other without interruption from the first day of the first
month to the last day of the last month, resulting in a distortion of the original
wangxiu day-count.
39
Though the Yang Jingfeng table bears no similarity to the
Southern Song system, the idea of a continuous day-count which characterizes
this system is nevertheless present.

36
Qiyao rangzai jue, juan 1, p. 428. Other examples can be found in the Xiuyao jing
(juan 1, p. 391 and juan 2, pp. 39495). See also the Lingtai jing , a related as-
trological treatise preserved in the Daoist canon (Daozang, fasc.136, 3a3b).
37
The seven luminaries tradition is well documented in the astrological books men-
tioned in this article. We have seen above that a section on calculating the days of the
week appears in a Japanese print of the Xiuyao jing. We find several instances where
luminaries are connected to days, but never as a proper day-count. In some cases, one
day may be related to two or three luminaries at a time for astrological purposes (Xiuyao
jing, juan 1, p. 392 and juan 2, p. 399). There is also a table of correspondence between
luminaries and days of the month (Xiuyao jing, juan 2, p. 399).
38
On the evaluation of Yang Jingfengs commentary, see Yano, The Hsiu-yao
ching and its Sanskrit Sources, p. 126.
39
Xiuyao jing, juan 1, p. 388. Nevertheless, calculations of day-lodges made by
Yang elsewhere in his commentary show that he was not relying on his own recon-
structed table but was following the wangxiu system. Yanos study of the Kakusho edi-
tion of the Xiuyao jing suggests that Yang Jingfengs table was corrupted in later trans-
missions of the text (The Hsiu-yao ching and its Sanskrit Sources, p. 129).
The Han System
The question of the possible existence of a twenty-eight xiu day-count prior to
the Tang dynasty has been raised on several occasions, but the lack of data and
documents showing how this day-count really worked has prevented us from
reaching a definite conclusion. Recent archeological findings, however, have
provided enough evidence to suggest that such a system did indeed exist as early
as the third century B.C.
The Liuren Treatises in the Daoist Canon
The Huangdi longshou jing , a mantic text in the Ming Daoist
canon, constitutes, together with the two texts appended to it, the oldest source
of our knowledge of the Liuren divinatory method.
40
We have seen that the in-
strument used in this method played a key role in the connection of the twenty-
eight xiu to the series of thirty-six animals. Despite some later additions, these
books may be considered as reflecting the state of Liuren computation tech-
niques during the Six Dynasties and Tang periods.
41
In chapter four of the Longshou jing, entitled Method of Calculation of Lu-
nar Lodges (yuexiu ), we find detailed information on how to connect the
twenty-eight lodges to days of the month using a Liuren board. The chapter is
followed by a commentary indicating the position of the yuexiu for the twelve
months of the calendar year:
The first month, Shi
13
is connected to the first day, Bi
14
to the second day, Kui
15
to
the third day, Lou
16
to the fourth day, and so on following that cycle. After all the
days of the month have passed, one arrives at lodge Bi
14
. The second month, Bi
14
is connected to the first day. After all the days of that month have passed, one
reaches lodge Lou
16
. The third month, Wei
17
is connected to the first day. The
fourth month, Bi
19
is connected to the first day. The fifth month, Jing
22
is con-
nected to the first day. The sixth month, Liu
24
seventh month, Zhang
26
is connected to the first day. The eighth month, Jiao
1
is
connected to the first day. The ninth month, Di
3
tenth month, Xin
5
is connected to the first day. The eleventh month, Dou
8
is con-
nected to the first day. The twelfth month, N
10
is connected to the first day. If
there is an intercalary month, the first day lodge will be the one following the
new-moon lodge (shuoxiu ) of the preceding month. For example, (since the

40
Huangdi longshou jing (Daozang, ce 135), Huangdi jinkui yuheng jing
(Daozang, ce 135), Huangdi shou sanzi xuann jing
(Daozang, fasc. 135).
41
For a presentation of these texts, see Marc Kalinowski, La littrature divinatoire
dans le Daozang, Cahiers dExtrme-Asie no. 5 (198990), p. 91; and Kalinowski, Les
manuscrits astro-calendriques des Han, pp. 396401.
first day of the first month is a Shi
13
day) the first day of an intercalary first
month will be a Bi
14
day.
42
This system is technically close to the Indo-Persian one: the two differ mainly in
the choice of the mark-point lodge. In the Xiuyao jing, it is the wangxiu (full-
moon lodge, fifteenth day of the month); in the Longshou jing this role is played
by the shuoxiu (new-moon lodge, first day of the month).
Therefore, we may compare the two systems by converting the full-moon
(fifteenth day) lodges of the Xiuyao jing into new-moon (first day) lodges. The
conversion has been made on the basis of the twenty-seven lunar lodges zodiac
(excluding Niu
9
) used in the Xiuyao jing.
43
Longshou jing Xiuyao jing
1st month, first day: Shi
13
Shi
13
2nd month, first day: Kui
15
Kui
15
3rd month, first day: Wei
17
Wei
17
4th month, first day: Bi
19
Bi
19
5th month, first day: Jing
22
Shen
21
6th month, first day: Liu
24
Gui
23
7th month, first day: Zhang
26
Zhang
26
8th month, first day: Jiao
1
Jiao
1
9th month, first day: Di
3
Di
3
10th month, first day: Xin
5
Xin
5
11th month, first day: Dou
8
Dou
8
12th month, first day: N
10
Xu
11
Differences are very minor since there are only three discrepancies with a
gap of one lodge each time. Is it possible therefore that the Liuren practitioners
were simply using the Indo-Persian system after having (for some unknown
reason) converted the wangxiu into shuoxiu just as we have done here? Further
evidence will show that the Liuren treatises in the Daoist canon were in fact
following a tradition which goes back to the Han dynasty diviners board and
hemerological methods.

42
Longshou jing 1.3a. The same set of lunar lodges is mentioned in the Jinkui yu-
heng jing (17a).
43
See Yano, The Hsiu-yao ching and its Sanskrit Sources, pp. 12829. Yano
proves quite convincingly that references to the twenty-eight xiu in the Xiuyao jing are a
later addition by Yang Jingfeng.
The Hanshu Evidence
In the History of the Former Han Dynasty (Hanshu ) we find a brief
mention of what may be a calendrical use of the twenty-eight lodges. The text is
about a fire which broke out in a building within the imperial palace in 46 B.C.
The precise dating of the event is given as follows: (Year two of Chuyuan),
fourth month, yiwei32 day, lunar lodge (yuexiu) Kang
2
. . .
44
Knowing that in
the year in question, yiwei32 was the eleventh day of the fourth month, we can
calculate the lodge connected to that day using the shuoxiu system (first day of
the fourth month is Bi
19
). The result will be Jiao
1
instead of Kang
2
in the Hanshu,
giving an error of one lodge only. Another noteworthy point is the use of the
expression yuexiu (lunar lodge) which also appears in the Longshou jing quota-
tion translated above.
The Liuren Diviners Board from the Former Han
An early model of a Liuren diviners board unearthed in 1977 at Fuyang,
Anhui province in a tomb dated 165 B.C. provides us with further evidence of
the existence of the shuoxiu system at the beginning of the Han dynasty.
45
The
round-plate of the instrument has the twenty-eight xiu inscribed on it and, next
to some of them, numbers from one to twelve indicating the twelve months. In
other words, we have here a list of twelve lodges connected in a specific way to
the twelve months. Previous studies have identified this series as representing
the twelve solar stations (richan) which, according to the lunar-solar standard of
the Chinese calendar, mark the sidereal positions of the sun attached to the
twelve months of the year.
The first text that establishes a fixed connection between lodges and months
of the year is the Yueling (Monthly Ordinances) compiled in the third century
B.C.
46
To each month is connected one lodge indicating the relative position of
the sun among the stars during that month. As we have indicated earlier, the
solar stations play an important role in Chinese astronomy and the calendar.
They command the adjustment of the twelve lunar months (12 x 29.5 = 354
days) to the solar year (365.25 days) and provide the celestial mark points for

44
Hanshu 75 (Beijing: Zhonghua shuju ed., p. 3175).
45
The original report may be found in Wenwu no. 8 (1978), pp. 1231. For a study
of this instrument, see Kalinowski, Les instruments astro-calendriques des Han, pp.
32352.
46
The earliest version of this ritual almanac is found as the first section of each of
the first twelve chapters of the Lshi chunqiu (c. 239 B.C.) named the Twelve Records
(shier ji). See the recent translation of the Huainanzi version in John S. Major, Heaven
and Earth in Early Han Thought: Chapters Three, Four, and Five of the Huainanzi (New
York: State Univ. of New York Press, 1993), pp. 21768.
the division of the zodiac into twelve mansions (suici ) and of the tropical
year into twenty-four solar periods (jieqi) of approximately fifteen days each.
The astrocalendrical treatises in the Huainanzi and the Hanshu also
specify the notations for the twelve monthly lodges.
47
For the whole period cov-
ering the end of the Warring States and the Former Han dynasty, we may add to
this group the series mentioned in the Wuxing zhan , a text belonging to
the Mawangdui manuscripts unearthed in 1972.
48
Now, if all these series, in-
cluding the one appearing on the round-plate of the Liuren board, are compared
to one another, it will be noticed that they are all different, even though the vari-
ants bear upon no more than three or four lodges in each list (see below).
It is thus astonishing to find that the twelve monthly lodges inscribed on the
Liuren board from Fuyang are identical with the twelve new-moon lodges
(shuoxiu) used in such Six Dynasties Liuren treatises as the Huangdi longshou
jing. In other words, as early as Han times, this series might have been taken as
having a double function, namely, that of solar stations (richan) regulating the
course of the sun within the year and that of lunar lodges (yuexiu) commanding
the movement of the moon within the month. For example, at the beginning of
the first month, while the sun was standing in Shi
13
(richan), the daily motion of
the moon was supposed to start. On the first day (shuo, new moon, conjunction
day), it was assumed that it dwelled also in Shi
13
(shuoxiu). Then it moved one
lodge a day until the end of the month. The second month, the moon resumed its
cycle, taking the solar station of that month (Kui
15
) as a starting point. And so
on, month after month, until the end of the year, exactly as it is described in the
Longshou jing section on the calculation of lunar lodges. The existence of such
an artificial and idealized system under the Han dynasty is perfectly plausible.
In any case, this was exactly the way the system was understood later on since
in the Tang dynasty text Xiuyao jing, the link between solar motion and the
twenty-eight xiu day-count is clearly established.

47
Huainanzi (Xin zhuzi jicheng ed., pp. 4950); Hanshu, Llizhi 2 (Dengzhi
huibian, pp. 143132). In the Hanshu, the monthly lodges are found in the table giving
the sidereal positions of the sun for the twenty-four solar periods. Therefore, two lodges
are connected to each tropical month, one to its first day (jieqi initial period), the other
to its fifteenth day (zhongqi middle period). Comparisons to be made below will show
that the twelve first-day lodges are much closer to other sets of monthly lodges in late
Warring States and early Han literature (see below, footnote 52).
48
An excellent edition of the text is found in Yamada Keiji (ed.), (Shin hakken
Chgoku) kagakushi shiry no kenky: yakuch hen [Studies on documents related to the
history of science recently found in China: translations and commentaries] (Kyoto: In-
stitute for the Humanities, 1985), pp. 144. Mention of the twelve monthly lodges is
found on pp. 12 and pp. 3637.
Shuihudi Manuscripts (I): The Twelve Monthly Lodges
The hemerological texts (Rishu , Book of Days) dating from the end of
the Warring States period unearthed in Shuihudi in 1976 offer conclusive evi-
dence of a twenty-eight xiu day-count in early China.
We find three complete sets of monthly lodges in these manuscripts (Rishu I
and Rishu II).
49
They appear in hemerological tables giving auspicious positions
for the twelve months of the year. We have translated the section corresponding
to the first month in each table to give an idea of the context in which the
monthly lodges were used.
First month Shi
13
/ Xu stands on (first) branch zi / Ying stands on
(second) branch chou / Jian stands on (third) branch yin / . . . / Yong
stands on (twelfth) branch hai [Rishu I, slip 1r1, transcription p. 180].
50
First month Shi
13
/ lodge Xin
5
is highly inauspicious / lodges Zhang
26
and Yi
27
will bring death / lodges Wei
12
and Shi
13
are highly auspicious / . . . / Xuange
is attached to lodge Yi
27
[Rishu I, slips 47r158r1, transcription pp. 187
88].
51
First month Shi
13
/auspicious for making sacrifices / forbidden for building
houses or moving into [a new] one / marrying a woman will bring unrest / a
child born in this month will become an official [Rishu II, slips 8011071,
transcription pp. 23738].
Again, what strikes us here is that these three series of monthly lodges (for the
full list, see group B below) are identical with the twelve full-moon lodges
found on the Fuyang diviners board and in the Liuren treatises of the Daoist
canon. In other words, it may be said that this particular set of twelve lodges
belonged to a common technical environment and that its usage was not re-

49
References to the manuscripts are based on the slip numberings in Shuihudi Qinmu
zhujian [Bamboo slips from the Qin tomb at Shuihudi] (Beijing: Wenwu chubanshe,
1990). For a presentation of the Shuihudi Rishu, see Marc Kalinowski, Les traits de
Shuihudi, pp. 174228. A general bibliography on the Rishu is included in Liu Lexian,
Shuihudi Qinjian Rishu yanjiu [A study of the Books of Days in the Qin dynasty slips
from Shuihudi] (Taibei: Wenjin chubanshe, 1994).
50
The names connected here to the branches are part of the hemerological system
known as Jianchu (founding-evicting). See Kalinowski, Les traits de Shuihudi,
pp. 19799.
51
On the Xuange (Dark spear) system, see Kalinowski, Les traits de Shuihudi, p.
219; and Zhang Mingqia, Qinjian Rishu xuange bian jiexi [An analysis of the Dark
spear system in the Books of Days from the Qin dynasty], Qin Han shi luncong vol. 4
(1989), pp. 197204.
stricted to the practice of the Liuren method but applied also to the whole of
early Han hemerological procedures.
Let us now put all the sets of monthly lodges in the late Warring States and
early Han period together, dividing them into two groups: group A consisting of
series representing independent traditions and group B made up of those which
are identical with the full-moon lodges tradition in the Liuren treatises in the
Daoist canon.
Group A
Yueling: Shi
13
(1st month), Kui
15
(2nd month), Wei
17
(3rd month), Bi
19
(4th month), Jing
22
(5th month), Liu
24
(6th month), Yi
27
(7th month), Jiao
1
(8th month), Fang
4
(9th month), Wei
6
(10th month), Dou
8
(11th month), N
10
(12th month).
Wuxing zhan: Shi
13
(1st month), Bi
14
(2nd month), Wei
17
(3rd month), Bi
19
(4th month), Jing
22
(5th month), Liu
24
(6th month), Zhang
26
(7th month),
Zhen
28
(8th month), Kang
2
(9th month), Xin
5
(10th month), Dou
8
(11th
month), Xu
11
(12th month).
Huainanzi: Shi
13
(1st month), Kui
15
and Lou
16
(2nd month), Wei
17
(3rd
month), Bi
19
(4th month), Jing
22
(5th month), Zhang
25
(6th month), Yi
27
(7th
month), Kang2 (8th month), Fang4 (9th month), Wei6 (10th month), Niu9
(11th month), Xu
11
(12th month).
Hanshu: Wei
12
(1st month), Kui
15
(2nd month), Wei
17
(3rd month), Bi
19
(4th month), Jing
22
(5th month), Liu
24
(6th month), Zhang
26
(7th month),
Zhen
28
(8th month), Di
3
(9th month), Wei
6
(10th month), Dou
8
(11th month),
N
10
(12th month).
52

52
See above, footnotes 46 to 48, for the references. The Wuxing zhan series is the
one that is mentioned at the beginning of the manuscript in relation to the twelve months
(Kagakushi shiry no kenky, pp. 12). The other occurrence of the series in the manu-
script (Kagakushi shiry no kenky, pp. 3637) has Lou
16
instead of Bi
14
for the second
month and N
10
instead of Xu
11
for the twelfth month. Concerning the Hanshu series, we
are quoting here the series of twelve lodges connected to the first day (jieqi) of each
tropical month (see footnote 47) since it appears to be quite similar to the other sets listed
under group A (three or four differences at the most). The twelve fifteenth day lodges
(zhongqi) are as follows: Shi
13
(1st month), Lou
16
(2nd month), Mao
18
(3rd month), Jing
22
(4th month), Jing
22
(5th month), Zhang
26
(6th month), Yi
27
(7th month), Jiao
1
(8th month),
Fang
4
(9th month), Ji
7
(10th month), Niu
9
(11th month), Wei
12
(12th month). The group A
series shows differences ranging between five and ten in number.
Group B
Huangdi longshou jing, Huangdi jinkui yuheng jing, Han dynasty Liuren
diviners board, Shuihudi Rishu manuscripts: Shi
13
(1st month), Kui
15
(2nd
month), Wei
17
(3rd month), Bi
19
(4th month), Jing
22
(5th month), Liu
24
(6th
month), Zhang
26
(7th month), Jiao
1
(8th month), Di
3
(9th month), Xin
5
(10th
month), Dou
8
(11th month), N
10
(12th month).
53
The series from Group A are all different. They differ also from the Group B
series. The variants are relatively minor and do not start before the sixth month
of the year. Consequently, lodges attributed to the first five months (Shi
13
, Kui
15
,
Wei
17
, Bi
19
, Jing
22
) are almost identical in both groups (with the exception of the
first month lodge in the Hanshu, Wei
12
, and the second month lodge in the
Wuxing zhan, Bi
14
). We may then assume that what we have here are records of
a related nature based on observations of the suns sidereal positions made at
about the same time, between the fourth and the first centuries B.C.
54
The dis-
tinctive feature of the twelve lodges in group B is that, besides their function as
solar stations (richan) regulating the adjustment of lunations to the course of the
sun within the year, they were also used as new-moon lodges (shuoxiu) con-
nected to the first day of the month and commanding the distribution of other
lodges within each month. In the first place, that this set was transmitted without
any change during nearly a thousand years testifies to its specific use, in sharp
contrast with the fluctuating nature of the notation of sidereal positions of the
sun in Chinese astronomical tables. Secondly, since there is an obvious continu-
ity between the Han dynasty Liuren diviners board and the Liuren treatises in
the Daoist canon, it is most likely that the twenty-eight xiu day-count described
in those books was already present in earlier forms of the method. Thirdly, a
closer reading of Shuihudi manuscripts shows that such a day-count was actu-
ally in use among hemerologists.
Shuihudi Manuscripts (II): The Twenty-eight Xiu Day-count
The unearthing of the Shuihudi Rishu manuscripts has led scholars to recon-
sider the possibility of the existence of a twenty-eight xiu day-count during the

53
On the Longshou jing and the Jinkui yuheng jing see above, footnote 42; on the
Fuyang diviners board see above, footnote 45.
54
There has been much debate concerning the dating of the astronomical contents of
the Yueling. We follow Yabuutis suggestion that these observations might have been the
result of the astronomical activity which took place in the fourth century B.C.; Yabchi
Kiyoshi [Yabuuti Kiyosi], Chgoku no tenmon rekih [Astronomy and calendar-making
in China] (Tokyo: Heibonsha, 1969), pp. 4648.
early Han dynasty.
55
The arguments provided to support this hypothesis, how-
ever, have been inconclusive. The similarity between the twelve monthly lodges
in the manuscripts and on the round-plate of the Fuyang diviners board has not
been compared with other sets of monthly lodges in Han literature. The Liuren
treatises in the Daoist canon have remained unexamined.
56
The only reference to
a twenty-eight xiu day-count in later times draws on the system described in the
Qing dynasty Xieji bianfang shu with no mention made of the Indo-Persian as-
trological traditions preserved in the Buddhist canon. As regards the manu-
scripts themselves, the studies have focused on the complete series of twelve
monthly lodges, especially the one related to the Xuange system, which has re-
ceived much attention, but disregarding other references to the lodges in both
manuscripts.
57
Nevertheless, in at least two instances, we find definite evidence
of the use of the twenty-eight xiu as a day-count.
The first instance is a list of the twelve months with the days corresponding
to the position of Xin
5
, the monthly lodge connected to the first day (wangxiu)
of the tenth month.
58
Since this month marked the beginning of the calendar
year in late Warring States and the early Han periods, it was probably seen as
more significant than the others.
59
In the translation which follows, we have
added, between brackets, the days corresponding to the Xin
5
positions as they
may be calculated using the twelve monthly lodges found in the manuscripts
(see group B above). For example, the shuoxiu of the tenth month being Xin
5
itself, lodge Xin
5
will fall on the first day of that month. The shuoxiu of the
eleventh month being Dou
8
, lodge Xin
5
will fall on the 26th day of that month
(and not on the 25th).

55
See, for example, the debate between Zhang Wenyu (an advocate of the twenty-
eight lodges day-count) and Wang Shengli (opposed to it): Zhang Wenyu, Yunmeng
Qinjian Rishu chutan [A preliminary examination of the Qin dynastys Books of Days],
Jiang Han luntan no. 4 (1987), pp. 6873; and Wang Shengli, Yunmeng Qinjian Rishu
chutan shangque [Some remarks on the article by Zhang Wenyu], Jiang Han luntan no.
4 (1987), pp. 7680. A good summary of the question is made by Liu Lexian, Shuihudi
Qinjian Rishu yanjiu, pp. 7886.
56
The similarity between the twelve monthly lodges in the Liuren treatises contained
in the Daoist canon and in the Han archeological documents has been pointed out in my
1986 paper on the Shuihudi manuscripts (Les traits de Shuihudi, pp. 21016).
57
There are nine occurrences altogether in the manuscripts. On the Xuange system,
see above, footnote 51.
58
Rishu I, slips 83v491v4, 92v3, 93v294v2, transcription p. 223; Rishu II, slips
9521062, transcription p. 239. Both versions are identical.
59
The beginning of the calendar year was set to the first month (spring) as a result of
the Taichu reform in 104 B.C. The use of the tenth month (winter) as the beginning of the
year is well documented in the manuscripts since we find several instances where the
twelve months are enumerated starting from that particular month.
Tenth month, 1st day (shuori), Xin
5
[1st]
Eleventh month, 25th day, Xin
5
[*26th]
Twelfth month, 23rd day, Xin
5
[*24th]
First month, 21st day, Xin
5
[21st]
Second month, (1)9th day, Xin
5
[19th]
Third month, (1)7th day, Xin
5
[17th]
Fourth month, 15th day, Xin
5
[15th]
Fifth month, 12th day, Xin
5
[12th]
Sixth month, 10th day, Xin
5
[10th]
Seventh month, 8th day, Xin
5
[8th]
Eighth month, 5th day, Xin
5
[5th]
Ninth month, 3rd day, Xin
5
[3rd]
The results are quite satisfactory since the reconstructed connections present
two variants (eleventh and twelfth months) with a gap of only one day each
time.
60
The second instance consists of a list of the twelve months and, under
each month, one day with its corresponding lodge.
61
The translation below in-
cludes reconstructed data in brackets as in the preceding case.
First month, Xu
11
, (day missing) [?]
Second month, Bi
14
, 27th day [*28th]
Third month, Jiao
1
, 13th day [13th]
Fourth month, Fang
4
, 14th day [14th]
Fifth month, Ji
7
, 14th day [14th]
Sixth month, Jing
22
, 27th day [27th]
Seventh month, Xing
25
, 28th day [28th]
Eighth month, Zhen
28
, 28th day [28th]
Ninth month, Kui
15
, 13th day [13th]
Tenth month, (lodge missing), 14th day [?]
Eleventh month, Shen
21
, 14th day [14th]
Twelfth month, Dou
8
, 21st day [*27th]

60
There are some obvious errors in the text which have been emended here. For the
twelfth and the first month, we have replaced erri sanri (second day third day) and erri
yiri (second day first day) by erxun sanri (second decade third day = 23rd) and erxun yiri
(second decade first day = 21st), following the way days are designated in this section
(xun decade for ten, erxun second decade for twenty). We have also provided the
characters xun (decade) which are obviously missing for month two (jiuri ninth day
instead of xun jiuri first decade ninth day) and month three (qiri seventh day instead
of xun qiri first decade seventh day).
61
Rishu II, slips 8931003, transcription pp. 23940.
Leaving aside the two untestable months (the first and tenth), we obtain
eight correct answers out of ten, which again is quite satisfactory. The gap of six
days for the twelfth month may be due to a confusion between the characters
one ( ) and seven ( ) used in the manuscripts.
Conclusion
Despite these few discrepancies, it is unlikely that such a conformity is purely
fortuitous. The two cases just mentioned, as well as other evidence provided in
this paper, testify quite convincingly to the existence in early Han hemerology
and divination of a twenty-eight xiu day-count. The system operated on the basis
of a set of twelve lodges attached to each month of the year representing its first
day or new-moon lodge (shuoxiu). The distribution of lodges on the other days
of the month was made by counting one lodge per day, starting from the new-
moon lodge. Thus, the first month, monthly lodge Shi
13
was connected to the
first day, lodge Bi
14
was connected to the second day, and so on until the end of
the month. The same procedure was repeated every month, from one new-moon
day to the next.
The choice of the twelve new-moon lodges was closely related to observa-
tions of the suns sidereal positions in the course of the year. The similarity be-
tween all series of monthly lodges in pre-imperial and early Han literature
shows that the notion of new-moon lodges commanding the distribution of the
twenty-eight xiu within the month and that of the solar stations (richan) regu-
lating the adjustment of lunar months to the seasonal year belong to a common
set of practices grounded in the Chinese calendar-making tradition.
The early shuoxiu system was then transmitted without much change until
the Six Dynasties through practitioners of the Liuren divinatory method. During
that period, the twenty-eight xiu were progressively connected to the series of
thirty-six animals following the distribution of these elements on the Liuren
diviners board.
In parallel with this evolution, a new system was brought into China together
with Indo-Persian astrology. Current under the Tang dynasty, its novelty mainly
consisted in the use of full-moon lodges (wangxiu) as starting points for the dis-
tribution of other lodges within the month. Besides the fact that the system was
based on these twelve wangxiu connected to the fifteenth day of each month and
that its application was chiefly horoscopic, there was no big difference with the
shuoxiu day-count. The compilers the eighth century Xiuyao jing seem not to
have been aware of the existence of an early native system since Yang Jingfeng
does not refer to it in his commentary.
62

62
See above.
From the Southern Song period onwards, the twenty-eight xiu day-count was
totally reshaped, and the regulating role of the twelve monthly lodges aban-
doned. As a result, the system freed itself from the constraining reference to the
motion of sun and moon. The twenty-eight xiu were associated with the sexage-
nary cycle by means of the seven luminaries day-count applied to Chinese cal-
endars during the Tang dynasty. The Song system continued in use into later
dynasties. It was fully described in the eighteenth-century Xieji bianfang shu
and still survives today in contemporary almanacs all over East Asia.
In conclusion, several remarks may be made. Firstly, that the Song dynasty
was a period of profound changes in ways of thought, institutions and social
behavior, leading Chinese civilization onto the road to modernity, is an assump-
tion common to many scholars. In that respect, the evolution of the twenty-eight
xiu day-count is in full agreement with conclusions drawn from studies carried
out in other fields of social practices and technical knowledge. The break be-
tween ancient systems and the one applied to Southern Song calendars is char-
acterized by a rationalization of the procedure through which lodges were con-
nected to days. By liberating itself from theories imposed by Han dynasty astro-
calendrical beliefs, the system becomes more sensible, coherent and simple. We
may equally note a process of acculturation to foreign influence since the new
system was built upon the Western seven-day week. Finally, if we follow Yan
Dunjies conclusion that the late twenty-eight xiu day-count was understood not
only as a duplicate of but also as a substitute for the seven luminaries system,
we may consider this substitution as part of a more general trend towards the
revival of native methods apparently free from any alien element.
63
Secondly, we have seen that the twelve full-moon lodges in astrological
books of Indo-Persian origin found in the Buddhist canon, once converted into
new-moon lodges, show only few differences with those used in the Han dy-
nasty system. Instead of interpreting this phenomenon as the result of a direct
influence exerted by one system on the other, we are more likely to conclude
that both systems were initially conceived on the basis of a similar set of solar
stations. On the other hand, the close concordance between the early Chinese
twelve monthly lodges and those found in later Indo-Persian astrological books
not only applies to the group B set, but also to all late Warring States and early
Han series listed in group A. It may be noted in particular that lodges attributed
to the first four months (Shi
13
, Kui
15
, Wei
17
, Bi
19
) are basically the same in all
series.
64
On the other hand, the Xiuyao jing indicates that the twenty-eight
lodges day-count was conceived at a time when a conjunction of the sun and
moon (new moon, first day of the month) took place in Lou
16
on the day of the
spring equinox. Now, the last calendar which gives Lou
16
as the solar station

63
See above, footnote 31.
64
Compare the list in the Xiuyao jing (above) with the early Han series of group A
and group B.
corresponding to the spring equinox is the Three Revolutions Calendar (Santong
li) by Liu Xin (c. 32 B.C.23 A.D.).
65
Starting with the Later Han, the vernal
point is already given as being in Kui
15
owing to the precession of equinoxes.
Unfortunately, the procedure through which the translators of the second- or
third-century Modeng qiejing shifted from the Indian twenty-seven or twenty-
eight nakshatra to the Chinese twenty-eight xiu remains a subject of great con-
troversy.
66
What can be said here is that the conversion was perhaps made on
the basis of early or middle Han astrocalendrical tables.
Thirdly, the existence of a system connecting the twenty-eight xiu with days
of the months in the late third century B.C. may provide an unexpected contri-
bution to the debate on the formation of the sidereal zodiac in China. There is
indeed little evidence in pre-imperial and early Han literature supporting the
view that the sidereal revolution of the moon in 27.3 days was initially related to
the division of the celestial sphere into twenty-eight parts. It has even been sug-
gested that other astronomical cycles like the sidereal revolution of Saturn in
approximately twenty-eight years might have been at the origin of this divi-
sion.
67
Added to this, the absence in these sources of any reference to a proper
lunar zodiac divided into twenty-seven or twenty-eight sections of equal length
following the twenty-seven or twenty-eight days of the sidereal month, makes
the question all the more puzzling.
68
In that respect, the twenty-eight xiu day-
count used by late Warring States and early Han hemerologists is an interesting
exception since it draws precisely on the idea of a lunar zodiac whose twenty-

65
See above, footnote 3.
66
On this subject, see Zen Hash, Matogaky no tenmon rekis ni tsuite, pp. 182
88.
67
Zheng Wenguang, Zhongguo tianwenxue yuanliu [The origins of Chinese astron-
omy] (Beijing: Kexue chubanshe, 1979), pp. 8391.
68
A recent attempt to show the existence in early Chinese astronomy of an equally
spaced lunar zodiac may be found in David Nivison, The Origin of the Chinese Lunar
Lodge System, in Anthony F. Aveni (ed.), World Archeoastronomy (Cambridge: Cam-
bridge Univ. Press, 1989), pp. 20318. Professor Nivison shows convincingly that the
regularities underlying the unequal distribution of the xiu on the equatorial belt might
provide an interesting ground for the study of the formation of the system as it appears in
late Warring States and early Han sources. However, the hypothesis that the xiu system is
a distorted version of a primitive zodiac, mathematically based on the regular combina-
tion of various solar as well as lunar cycles (the two solstitial halves of the year, the four
seasons, the twelve solar mansions, the twenty-four fortnightly periods, and the sidereal
month of twenty-eight days; see pp. 2058, points one to five) is highly controversial.
The demonstration does not bring new evidence in favor of the historicity of such a re-
constructed model in early times. In my view, the only positive element indicating a
regular spacing of the xiu in existing literary sources and archeological devices is pre-
cisely their possible use as a day-count in the Shuihudui manuscripts (pp. 2078, point
five).
eight parts represented the daily lodgings of the moon among the stars, as was
also the case in ancient Indian lunar astrology.
69

69
See Roger Billard, Lastronomie indienne: Investigation des textes sanscrits et des
donnes numriques (Paris: Ecole Franaise dExtrme-Orient, 1971), pp. 1516. On
evidence showing the use of a system of twenty-seven lodges equally spaced in the Xi-
uyao jing, see Yano, The Hsiu-yao ching and its Sanskrit sources, pp. 12829.
82
The Evolution and Decline
of the Ancient Chinese Practice
of Watching for the Ethers
Huang Yi-Long and Chang Chih-cheng
AUTHORS NOTE: Research for this article was sponsored by the Yang
Kuang-hsien and his Anti-Christian Activities Project (NSC 82-0301-H-007-
022) at the Kuo-ko-hui (National Science Council), and a Special Re-
search Fund grant from Tsing-hua University. The authors appreciate their
support. The authors would also like to extend their thanks to Mr. An Shuang-
cheng of Beijings Number One Historical Archive, who kindly allowed us to
read the section of his manuscript translation of the The Archive of Secret
Manchu Memorials which deals with the Kang-hsi Calendar Case.
[An earlier version of this article, in Chinese, was published in Tsing-hua
hsueh-pao 23.2 (1993): pp. 12547. The revised, English version is published
here with permission.]
* * *
The Development of the Practice of Watching for
the Ethers in Ancient China
ou-chi (watching for the ethers) was a method used to calculate
the seasons.
1
It embodied the ancient Chinese concept of the unity of
Heaven, Earth and Man. The practice of hou-chi involved burying

1
Previous research on hou-chi is very scantyDerk Boddes is the only article to
treat the subject. See Bodde, The Chinese Cosmic Magic Known as Watching for the
Ethers, in Sren Egerod and Else Glahn (eds.), Studia Serica Bernhard Karlgren Dedi-
cata, Sinological Studies Dedicated to Bernhard Karlgren on his Seventieth Birthday,
October Fifth, 1959 (Copenhagen: Ejnar Munksgaard, 1959), pp. 1435. Boddes major
contribution appears in his discussion of hou-chi as it was practiced before the Ming.
The present article, however, will focus on the debate over hou-chi during the Ming and
Ching periods.
H
Huang Yi-Long and Chang Chih-cheng: Watching for the Ethers 83
twelve musical pitchpipes of graduated lengths in a sealed chamber, and filling
the pipes with ashes produced by burning the pith of a reed (Phragmites com-
munis, var. longivalvis).
2
The ancients believed that when the sun entered the
second fortnightly period of any given month (chung-chi ), the earths chi
(seminal force) would rise and expel the ashes from the pipes.
Among those who have studied the history of Chinese musical harmonics
and calendar-making, the hou-chi practice has long been a controversial topic.
This has been a result of the difficulty of integrating the practice with the results
of actual calculations. Though the practice appeared at least as early as the time
of Ching Fang (7937 B.C.) of the Western Han dynasty (206 B.C.8
A.D.), detailed descriptions of its operation are few. The few examples include
Tsai Yungs (13392) Yueh-ling chang-ch (Punctuated
commentary on the monthly ordinances),
3
Ssu-ma Piaos Hs Han chih
(Sequel to the records of the Han), written during the Chin dynasty
(265420), and Li Chun-fengs L li chih (Monograph on
musical harmonics and the calendar), which appeared in the History of the Sui
Dynasty, written in the Tang.
4
According to these documents, very few people
were successful in practicing hou-chi. Hsin-tu Fang of the Northern
Chi dynasty (55177) was one of the exceptions.
5
In the history of the hou-chi practice, Hsin-tu Fang and his wondrous
method played an important role.
He knew how to use pitchpipes to watch for the ethers and observed the appear-
ance of the clouds. Once, while conversing, he pointed to the sky and said, The
fortnightly period of the early spring has arrived. His interlocutor thereupon
checked the pitchpipes, only to discover that the ashes had already flown free.
[Hsin-tus] monthly calculations were always correct. He made a wheel of
twenty-four fans and buried it so as to predict the twenty-four fortnightly peri-
ods. With the arrival of each fortnightly period, a single fan moved while all the

2
The twelve musical pitchpipes of traditional China were the six yang l
huang-chung , tai-tsu , ku-hsien , jui-pin , i-tse , and wu-yeh
) and six yin l (ta-l , chia-chung , chung-l , lin-chung ,
nan-l , and ying-chung ). The length of each pipe is determined by the so-
called san-fen sun-i-fa (the method of one-third decrease and increase). By
reducing the length of a pitchpipe of known length and pitch by one third, one produces
a pitchpipe five tones higher. By increasing the original pitchpipes length by one third,
one produces a pitchpipe three tones lower.
3
Bodde, The Chinese Cosmic Magic, pp. 1819.
4
Hou Han-shu [The history of the eastern Han dynasty] (Peking: Chung-
hua shu-ch, 1975), pp. 301516; Sui-shu [The history of the Sui dynasty]
(Peking: Chung-hua shu-ch, 1975), pp. 39497.
5
Sui-shu, pp. 39497.
others remained immobile. This method duplicated the results obtained by pitch-
pipes and ash.
6
According to this text, Hsin-tu not only employed the traditional pitchpipe
method, he also employed a system of twenty-four mechanical fans to verify
the result. As a result it was said that whether he used flying ash or mechanical
fans, Hsin-tu never failed in his tests.
Hsin-tu Fang also offered explanations of the phenomena related to the ex-
pulsion of ashes from pitchpipes recorded in Ssu-ma Piaos Hs Han-chih. He
believed that if blown free from the pitchpipes by the proper chi, the ashes
would land close to the mouth of the pipe rather than far away; if moved by
breath, the ashes would be scattered in all directions and a bit further [from the
mouth of the pipe]. Here, breath (kou-chi ) refers to the breath of the
person conducting the test. Hsin-tu also explained the observation that ashes
moved by drafts from peoples clothes are [found] together by noting that if
drafts from clothing arose in the chamber, there would be whirlwinds at both
ends; the ashes would be moved from the pitchpipes by these whirlwinds, and
therefore would [be found] together. In order to avoid untoward influences on
the test, Hsin-tu requested that observers calm their spirits, move in accordance
with propriety and avoid carelessness.
Because of the specificity of Hsin-tu Fangs description, the results of his
observations created a tremendous surge of interest in a procedure the theoreti-
cal foundations of which were fascinating to contemporaries. Supporters of the
practice found they had gained real moral support, and gradually the hou-chi
practice became a part of the lore of the ancients.
After a stable development during the course of the Northern and Southern,
Sui, and Tang dynasties, the hou-chi practice became in the Sung dynasty the
subject of considerable interest. The most influential or significant of those who
wrote on it at this time were Tsai Yan-ting (113598), Shen Kua
(103094), and Chu Hsi (11301200). Tsai advocated the use of
hou-chi in obtaining the standard tone.
7
Shen compared hou-chi with acu-
puncture to support his belief in chi.
8
Chu employed the analogy of the pitch-
pipes as

6
Sui-shu, p. 394.
7
Tsai Yan-ting, L-l hsin-shu [New book on pitchpipes], in Ssu-ku
chan-shu, vol. 212 (Taipei: Shang-wu yin-shu-kuan, 1983), 2: p. 3.
8
Shen Kua, Meng-hsi pi-tan chiao-cheng [Collated and verified
writings of Meng Hsi] (Shanghai: Shanghai ku-chi chu-pan-she, 1987), 7: p. 325.
Analogies between hou-chi and acupuncture continued to be made during the Ming
dynasty. For instance, Tao Wang-ling once wrote On the winter solstice, the
yang-chi rises to within nine tsun of the ground. Only the huang-chung pitchpipe can
be reached, therefore only the huang-chung pitchpipe responds. Other months corre-
spond with the lengths of the pitchpipes. This resembles the use of acupuncture in the
a proof in his discussion of lutes.
9
The above discussion should make it clear that in ancient Chinas learned
circles the hou-chi practice was generally accepted.
The Appearance of Doubts
Throughout the historical development of the hou-chi practice, controversies
surrounded every element of the technique. The location of the experiment, the
quality of the soil, the selection of materials, the placement of the pitchpipes
(bases or tops aligned? buried or not?), the means of filling the pitchpipes with
ash (ought the ash to be placed on light silk fabric?) and so forth were all hotly
contested. Not only that, there was also a constant production of new ideas. All
of this may well have been the inevitable consequence of repeated experimental
failures.
The most stridently expressed doubts did not appear until after the midpoint
of the Ming dynasty. An unabashed critic, the li-hseh scholar Wang
Ting-hsiang (14741544), reckoned that the chi of Heaven and Earth
rises and falls. . . . If you say, The yang pitchpipe responds before the summer
solstice, and the yin pitchpipe responds before the winter solstice, well, that
means that in one year both yin and yang have risen and neither has fallen! He
therefore concluded that the hou-chi practice was nothing but Tsou Yens
and Ching Fangs ridiculous talk. Furthermore, he denied the possibility
of employing pipes of various lengths to induce chi. He declared that there is
no little nook into which chi does not penetrate, and it cannot be constrained. Is
it possible that chi will be constrained within [a pipe of] nine tsun? Is it possi-
ble that chi will linger in a small space and take a month and more to produce a
response?
10
Liu Lien , a chin-shih and censor of the Cheng-te period
(150621), was another important opponent of the hou-chi practice. He was
particularly critical of Hsin-tu Fang, of whom he once wrote:

treatment of disease. By probing with needles of different lengths, the chi will be
reached. Chang Hsan , Hsi-yan wen-chien-lu [Records from the
experiences of Hsi-yuan], in Chung-hua wen-shih tsung-shu
[Collectanea of Chinese history and literature], part 5 (Taipei: Ta Chung-kuo tu-shu
kung-ssu, n.d.), 51: p. 20.
9
Chu Hsi, Chu-tzu ta-chan [Complete works of master Chu], in Ssu-pu
pei-yao [Essentials of the Ssu-pu] (Taipei: Chung-hua shu-ch, 1965), 66: pp.
3031.
10
Wang Ting-hsiang , Wang Ting-hsiang chi [Collected works
of Wang Ting-hsiang] (Peking: Chung-hua shu-ch, 1989), 40: p. 711.
During the Northern Chi, Hsin-tu Fang observed the appearance of the clouds.
Once, while conversing, he pointed to the sky and said, The fortnightly period
of the early spring has arrived. His interlocutor thereupon checked the pitch-
pipes, only to discover that the ashes had already flown free. [Hsin-tu] made a
wheel of twenty-four fans and buried it so as to predict the twenty-four fort-
nightly periods. With the arrival of each fortnightly period, a single fan moved
while all the others remained immobile. I say that chi was in the earth, [but] it
was without form, and pitchpipes had to be used to observe it. If he could detect
the arrival of chi by observing the appearance of the clouds, why did he bother
with pitchpipes and ashes? If fans could be used instead of pitchpipes in observ-
ing it, the pitchpipes must be unreliable!
Liu reckoned that if the observation of clouds or mechanical fans could be
used to detect the arrival of chi, continuing to use pitchpipes for hou-chi was
unnecessary. He went further, condemning Hsin-tu Fang as a treacherous and
cunning person who dared to employ bewitching absurdities to cheat his master
and confuse the people: he ought to have been killed!
11
Chi Pen , a student of Wang Yang-mings , also criticized the
hou-chi practice. He believed that from the jui-pin [pitchpipe] on, the chi
ought to descend, not ascend, as is said. Otherwise, the chi of Heaven and Earth
would be always rising and never descending! His argument closely resembled
Wang Ting-hsiangs.
12
He also expressed some doubts based on a topographi-
cal approach:
The shape of the earth has high points and low points, [but] there is no before
or after with chi. Does chi first come to rest in the low points of the earth, ar-
riving later in the high points? If we follow the practice, when watching for the
ethers there is no need to measure the diameters of the pitchpipes, so long as their
lengths are proper. As long as the standard note (chung-sheng ) was deter-
mined, the chi always responded.
As to how to determine that standard note, Chi advocated training blind people
with perfect pitch for the task.
13
Hsing Yun-lu , a chin-shih of the Wan-li period (15731620)
who served as Surveillance Commissioner of Shensi province, was also
profoundly skeptical of the results obtained by Hsin-tu Fang, as well as by Mao

11
Liu Lien , Yueh-ching yan-i [The original meaning of the Clas-
sic of Music] (1550), 1: pp. 1516.
12
Chi Pens opinions are cited in Chu Tsai-ys Yueh-l chan-shu. See Chu Tsai-
y , Yueh-l chan-shu [Complete musical harmonics], in Ssu-ku
chan-shu , vol. 213 (Taipei: Shang-wu yin-shu kuan, 1983), 5: p. 31.
13
The term standard note refers to the note produced by the huang-chung. As was
written in the L-shih chun-chiu, the huang-chung note, midway between the turbid
and the clear tone, was the foundation of music (5: p. 7).
Hsi-cheng and his son. Hsing argued that The motion of the fan and
the flight of the ashes are both mechanical [matters]. People trigger the me-
chanicsthese are connected to the mouth of the pitchpipethen, instantly and
without fail, the fan starts to move and the ashes fly up. [They] performed these
fraudulent maneuvers to fool their masters.
14
In those days, the Astronomical Bureau sent officials to Shun-tien
prefecture every year at the time of the li-chun fortnightly period to prac-
tice hou-chi.
15
Hsing violently attacked what he saw as deceiving others
through dishonest communication by mere sleight of hand. Up to now, wrote
Hsing, officials from the Astronomical Bureau have visited Shun-tien prefec-
ture during li-chun and other festivals in order fraudulently to use mechanics to
cause ashes to fly. They then tell people: The ashes have flown! and report to
their superiors: The ashes have flown! Whom are they cheating? Heaven?
16
Hsing Yun-lu was not the only one to complain of the fraud perpetrated by
officials charged with astronomical matters. In Chu Yun-mings Yeh-
chi (Notes from a commoner) we find a detailed description: In the early
years of the present dynasty, the officials who dealt with astronomy still prac-
ticed hou-chi and measured the [length of the] suns shadow. . . . It is said that
later hou-chi became a mere formality. By stuffing ash into a hole which was
[in fact] the mouth of a tunnel, the officials were able to make the ashes fly up
whenever [they liked] by pouring a soup of boiling lime into the other end of the
tunnel.
17
The addition of water to lime produces heat and carbon dioxide. This is the
soup of boiling lime mentioned above. A secret tunnel dug under the pitch-
pipes, along with knowledge of a chemical reaction, thus permitted the officials
to produce a shower of ashes at will.
With the arrival of Chu Tsai-y (15361611), criticism of the prac-
tice of hou-chi reached a new level. Chu, styled Po-chin , was a sixth
generation descendant of the Ming emperor Jen-tsung ; his father, Hou-
wan , had inherited the title of Prince Kung of Cheng , and Chu
Tsai-y, soon after his birth, became the designated heir. Tsai-y was an avid
scholar and specialized in musical harmonics and mathematics.
18
In his L-l

14
Hsing Yun-lu , Ku-chin l-li-kao [Investigations of musical
harmonics and calendars past and present], in Ssu-ku chan-shu, vol. 787, 33: p. 6.
15
Li Tung-yang and Shen Shih-hsing , Ta-Ming hui-tien
[The comprehensive codes and statutes of the Ming dynasty] (Taipei: Hsin-wen-feng
chu-pan kung-ssu, 1976), 233: p. 9.
16
Li Tung-yang and Shen Shih-hsing, Ta-Ming hui-tien, 233: p. 9.
17
Chu Yun-ming , Yeh-chi [Notes from a commoner] (1874 edition
from the collection of the Academia Sinica), 1. This passage does not appear in Li Shihs
Li-tai hsiao-shih edition of the same work.
18
For the biography of Chu Tsai-y, see Tai Nien-tsu , Chu Tsai-y: Ming-
tai ti ko-hseh ho i-shu ti ch-hsing [Chu Tsai-
ching-i (Essentials of musical harmonics), Chu asserted that all of this
talk of watching for the ethers and flying ashes was unorthodox:
The hou-chi method went unmentioned in the Classics, appearing instead in
apocryphal works (wei ). Hsin-tu Fangs mechanical fans are especially ab-
surd. Mencius said, It is better to give up books than to place ones complete
faith in them. A Confucians first priority ought to be the investigation of things
so as to understand principle thoroughly (ko-wu chiung-li ). But hav-
ing been unwittingly misled by unfounded statements for a thousand years, in-
vestigations have gone to who-knows-where.
19
Chu also had idiosyncratic opinions about what he saw as superstitious attitudes
towards hou-chi. An instance of this is the vividly drawn passage below, from
Chus L-hseh hsin-shuo (New studies of musical harmonics):
The inventors of hou-chi were a frivolous lot. . . . Those who have fallen under
its spell nowadays truly revere it. Why? One might well ask: In making a like-
ness, what does the painter find easy, and what does he find difficult? And the
answer is: Painting supernatural beings is easy; painting dogs and horses is dif-
ficult. Why is this? In making a likeness, one wants there to be a resemblance
[to the original]. Since no one can verify [the appearance of] supernatural beings,
[painting them] is easy. One can verify [the appearance of] dogs and horses, so
[painting them] is difficult. The [case of the] musician who gathers millet to
make woodwinds is like that of the dogs and horses, while [the case of] watching
for the ethers and verifying the tones is like that of the supernatural beings. Long
ago, crazed actors and bewildered blind men invented this unfounded practice in
order to confer magical power upon their techniques. They thereby fooled their
silly audiences and made it quite impossible for others to question them. All
through history they were believednone doubted. This is making the difficult
easy and the easy difficult: is it not ridiculous?
20
Chu not only criticized the gullibility of earlier scholars confronted with hou-
chi, he also carried out rigorous experiments himself, driven by a profound
scientific spirit. For instance, traditionalists all preferred bamboo from Mount
Chin-men when it came to making pitchpipes,
21
but after a meticulous
analysis, Chu wrote in his L-l ching-i:

y: A superstar of science and art in the Ming] (Peking: Jen-min chu-pan-she, 1985);
Chen Wan-nai , Chu Tsai-y yen-chiu [Studies of Chu Tsai-y]
(Taipei: Kuo-li ku-kung po-wu-yuan, 1992).
19
Chu Tsai-y, Yueh-l chan-shu, 5: p. 40.
20
Chu Tsai-y, Yueh-l chan-shu, 22: p. 30.
21
The use of bamboo from Mount Chin-men in the practice of hou-chi became
popular no later than the Chin dynasty. During the Ching dynasty, Ma Kuo-han wrote in
the afterword to his Mei-tzu hsin-lun [New discourses of master Mei]:
According to chan 42 of the Y-lan, Juan Chis I-yang-chi is cited to the
Mount Chin-men is also known as Pitchpipe Mountain (l-kuan shan ).
Nowadays it is located in Yung-ning county, in Honan . Although
this area produces bamboo, it is only the small bamboo with widely spaced joints
that is usedthe large bamboo is useless. [In particular,] bamboo with close-set
joints which is not round, and bamboo the ends of which are uneven are useless.
Sweet bamboo is the best, particularly if its joints are widely spaced. Finding
natural pitchpipes, whose inner and outer measurements naturally fit the pat-
ternthat is precious indeed! However, one must first know the correct pattern;
only then can one recognize what is suitable and what is not.
22
Some time later, after many comparative experiments, Chu discovered that the
old favorite Mount Chin-men bamboo was inferior to the Nan-pi-kuan
bamboo of Y-hang county, Chekiang province. This is a good example of
the sort of perfectionism Chu Tsai-y brought to his experimental procedures.
The results of Chus own experiments with hou-chi are recorded in his L-
l cheng-lun (Unbiased discussions of musical harmonics), which
he composed late in life. He wrote:
In days of old it was said that whenever pitchpipes were made, reed ash from Ho-
nei , black millet from Mount Yang-tou in Shang-tang and
bamboo from Mount Chin-men in I-yang were all indispensable. In fact,
all of these materials can be easily obtained. The place called Ho-nei lies within
my fief. Shang-tang lies to the north, and I-yang, to the south; it is about three or
four days travel to either one [from Ho-nei]. In the eighth year of the Wan-li
reign, the keng-cheng year, I ordered my subordinates to collect several
tens of thousands of each of the three, specifically, single grains of mature black
millet, double grains of small black millet, and bamboo with widely spaced
joints. I planted them myself. The millet has come to occupy a ching of land;
the bamboo has become a forest. Up to the present time, all of the bamboo and
all of the black millet harvested at my manor are the result of those plantings.
However, the [quality of the] soil is not adequate, so my harvest has never
matched that of the original lands. I once followed Tsai Yan-tings directions,
and buried [pipes] whose finely graduated lengths ranged from five to ten
tsunall in all there were some 384in rows at varying depths: no ashes were
blown free. I am beginning to think that all who believed in that business are mo-
rons or blind men.
23
The above narration makes it clear that not only did Chu give over lands to the
large scale cultivation of black millet, reeds and bamboo, he also zealously fol-

effect that the bamboo of Mount Chin-men could be used to make musical pipes. There-
fore, Yang Chan and Mei Tzu both cited Juan Chi. (Ma Kuo-han , Y-han
shan-fang chi-i-shu , 4: pp. 264849.) Yang Chan, Juan Chi and
Mei Tzu all lived during the Chin dynasty.
22
Chu Tsai-y, Yueh-l chan-shu, 5: pp. 1, 11.
23
Chu Tsai-y, L-l cheng-lun [Unbiased discussions of musical har-
monics] (Wan-li edition from the collection of Academia Sinica), 1: p. 1.
lowed Tsai Yuan-tings dictum to cut more bamboo. Between the outer limits
of five and ten tsun, he had 384 pitchpipes cut so that the length of each was
just slightly greater than its predecessor. All this was in order to perform more
precise and complete predictions.
Thus, Chu Tsai-ys criticism of the hou-chi method was solidly founded
on the results of his careful measurements and observations. None of his fellow
criticsneither those who went before him, nor those who succeeded himwas
able to reach the mark that he set.
The Calendar Case of the Kang-hsi Reign
and the Hou-chi Debate
Though such Ming intellectuals as Liu Lien, Hsing Yun-lu and Chu Tsai-y
raised their voices in opposition to the hou-chi practice, traditional attitudes did
not suffer a mortal blow. Only with the introduction of Western learning during
the late Ming (13681644) and early Ching (16441911) did the formerly sta-
ble position the practice had maintained in Chinese society become less sure.
This was the result both of hou-chi being dragged into the debates about the old
and new calendrical methods, and of the intervention of officials in the matter.
24
Johann Adam Schall von Bell (15921666; Chinese name Tang Jo-wang
) assumed the directorship of Chinas Astronomical Bureau during the
Ming-Ching transition.
25
Beginning in the second year of the Shun-chih
reign (1645), Schall reinstituted the yearly excursion to Shun-tien prefecture to
watch for the ethers during the five days preceding the onset of the li-chun
fortnightly period. Perhaps as a proleptic gesture to silence possible mutter-
ings, which could have led to undesirable confrontations, the Jesuit sent an offi-
cial from the Calendrical Office (Li-ko ), one from the Clepsydra Office
(Lou-ko-ko) , and such local officials as timekeepers (ssu-chen )
to perform the traditional operations.
However, perhaps because the operations of hou-chi were unverifiable,
these officials did not normally bother to make actual measurements and instead
the timekeeper simply submitted a false report, stating that the chi had mani-
fested itself. The day before the arrival of the li-chun fortnightly period, the

24
Boddes The Chinese Cosmic Magic fails to make any mention of this.
25
Huang Yi-Long, Tang Jo-wang y Ching-chu hsi-li chih cheng-tung-hua
[Adam Schall and the legitimation of the Western calen-
dar in the early Ching], in Wu Chia-li and Yeh Hung-sa (eds.), Hsin-
pien Chung-kuo ko-chi-shih [New perspectives on the history of
Chinese science and technology] (Taipei: Yin-he wen-hua shih-yeh kung-ssu, 1990), pp.
46590.
pitchpipes were put away and a report made to the effect that some or all of the
ashes had flown.
26
These activities surrounding the li-chun fortnightly period must, as they be-
came customary, have come to be related to the folk customs surrounding the
arrival of spring, called welcoming the spring (ying-chun ). In those
days, officials and commoners gathered at the time of li-chun to make prepara-
tions for the spring rituals, the ultimate goal of which was the stimulation of
agricultural yields. Using earth, they sculpted a spring ox (chun-niu )
and flanking peasant figures (known as mang-shen ) who were depicted
flailing the ox with whips.
27
The Astronomical Bureau, charged with making the
yearly calendar, had the formal responsibility of ensuring the precise timing of
li-chun.
When Adam Schall assumed the directorship of the bureau, he deliberately
forced out those astronomers who had been trained in traditional Chinese or
Muslim astronomy. However, he had underestimated the tangled intertwining of
astronomy and yin-yang numerology. The old methods numerologists had used
in telling fortunes were undermined when Schall changed the sequence of the
tzu ( Orionis) and shen ( Orionis) constellations, transposed lo-hou
(Rahu) and chi-tu (Ketu) and obliterated the imaginary tzu-chi
star. The reaction to this was an intense, recriminatory outcry from con-
servatives.
28
Yang Kuang-hsien (15971669), in order to uphold tradi-
tion, brought a suit against the Jesuits in 1664this was the so-called Calendar
Case. In the midst of this conflict, the attitudes of the Catholic astronomers
toward hou-chi came to be the focus of the attacks of the Chinese conserva-
tives.
29

26
Man-wen mi-pen-tang [The archive of secret Manchu memorials],
chan 150, memorial dated the ninth day of the second month of the fourth year of the
Kang-hsi reign (25 March 1665). For a description of this volume, see An Shuang-
cheng , Tang Jo-wang an shih-mo [The beginning and con-
clusion of the case of Adam Schall], Li-shih tang-an 3 (1992), pp. 7987.
27
Chuang Chi-fa , Sui-tzu i-chou hua chun-niu: chien-chieh yuan-tsang
Chun-niu mang-shen-tu [On the Spring Ox
in the I-chou year: a brief introduction to the Spring Ox and Mang-Shen graph in our
museum], Ku-kung wen-wu yueh-kan vol. 2, no. 11 (1985), pp. 4954; Nakamura Ky
, Ritsushun gyji fsh (part one) , Ritsumeikan bungaku 509
(1988), pp. 43454.
28
Huang Yi-Long, Ching-chu chin-tien-chien chung ko min-tsu tien-wen-chia
ti chan-li chi-fu [The waxing and waning
of the power of various ethnic astronomers in the early Ching Astronomical Bureau],
Hsin shih-hseh vol. 2, no. 2 (1991), pp. 75108.
29
Huang Yi-Long, Ching chien-chi tui tsu, shen liang hsiu hsien-hou tzu-hsu ti
cheng-chih in Yang Tsui-hua and
Huang Yi-Long (eds.), Chin-tai Chung-kuo ko-chi-shih lun-chi
In the second month of the fourth year of the Kang-hsi reign (1665), Yang
Kuang-hsien attacked Schall in a memorial to the throne:
To date, in establishing the calendar, numbers have been used to calculate it,
[celestial] phenomena to measure it, clepsydra to verify it, and chi to investigate
it . . . . Therefore, the Lou-ko officials undertook an examination, determined
the measurement of the clepsydra ought to be one hundred units per day, placed
pitchpipes in the hou-chi chamber, ascertained the exact moment of the flight of
the reed ashes, and employed it to determine whether the predicted time agreed
with the naturally occurring moment of the fortnightly periods. . . . Nowadays,
[Schall] relies only upon his own calculations and has abolished those offices
that used this old system . . . . [When] the pitchpipes used by the Lou-ko Office
are abolished, and no consideration is given to their flying ashes, even if people
go so far as to violate the hou-chi in its very chamber and celestial aberrations
appear, who will dare to speak up? Thus will Schall deceive the whole world in
order to present his new method.
30
Responding to the charges in court, Schall declared that this business of
flying ashes has long been in desuetude, a fact which can be demonstrated by
consulting the memorial made by Li Tien-ching under the Ming dy-
nasty. In that memorial, Li accounted for the reed ashes. He also claimed that
[since] the earth is [by turns] hard and yielding, dry and moist, and uneven as
well, it is difficult to produce consistent results. This amounted to a confession
that Schall himself had never performed hou-chi. In addition, he slyly asserted
that hou-chi actually had nothing to do with the measurement of seasons: its
true use was the observation of harvests in certain provinces. However, when
Prince Kang of Ho-shou (Giyeu, also known as Chieh-shu
or ) (164597) asked whether Schall had officially reported this, and why

[Science and technology in modern China] (Taipei: Chung-yang-yen-chiu-yuan chin-tai-
shih yen-chiu-so, 1991), pp. 7193; idem, Ching chien-chi tui ssu-y ting-i chi tsun-
fei ti cheng-chih: she-hui tien-wen-hseh ti ko-an yen-chiu
: [The early Ching debates over the definition
and existence of the four reminiscences: a case study in the social history of astronomy],
Tzu-jan ko-hseh-shih yen-chiu vol. 12, no. 3 (1993), pp. 20110 and vol. 12, no. 4
(1993), pp. 34454; idem, Chung-hsi wen-hua tsai Ching-chu ti chung-tu y tou-
hsieh: I Tang Jo-wang so-pien min-li wei ko-an yen-chiu
[Conflict and compromise between
Eastern and Western cultures in the early Ching dynasty: a case study of the popular
calendar made by Adam Schall], paper presented at the International Symposium on the
Occasion of the 400th Anniversary of the Birth of Johann Adam Schall von Bell, S.J.
(15921666), St. Augustin, Germany, May 1992.
30
Man-wen mi-pen tang, chan 149, memorial dated the thirteenth day of the first
month of the fourth year of the Kang-hsi reign (27 Feb. 1665); Man-wen mi-pen tang,
chan 150, memorial dated the ninth day of the second month of the fourth year of the
Kang-hsi reign (25 March 1665).
he had failed to set up pitchpipes in all fifteen provinces to check the time of
their harvests, Schall was at a loss. All he could manage to say was that he was
old and sick, and had erred in his earlier testimony.
It had been in order to avoid the familiar accusation that fortnightly periods
calculated in accordance with Western methods are not verifiable through heav-
enly motions that Schall had posited a relationship between hou-chi and the
prediction of harvests. In ancient China it was believed that when the sages
ruled, timely winds came in season; when worthy men were at the court, the
eight winds dispersed. As a result, during the fortnightly periods of the two
equinoxes, the winter and summer solstices and the four initial fortnightly peri-
ods of each season (ssu-li ), people observed the winds. The varying wind
directions enabled them to predict the events of the coming year. This is illus-
trated in Ta-Ming tien-yan y-li hsiang-i tu-shuo
(The illustrated omens in the Tien-yan calendar of the Ming dynasty) recom-
piled by Y Wen-lung under the Hsan-tsung emperor (r. 1426
36):
[If] the wind comes from the chien direction, severe frost will lay things
waste; [wind] from the kan direction [means] the year will be very cold, and
the northern barbarians will invade; [wind] from the ken direction [means]
the year will be bountiful and there will be a full harvest of the five grains;
[wind] from the chen direction [means] that the chi will leak and nothing
will come to fruition; [wind] from the hsn direction [means] the year will be
windy and many pests will die; [wind] from the li direction [means] that
drought will harm the animals; [wind] from the kun direction [means] that
cold weather will last into spring, floods will rise in the sixth month, and people
will be melancholic; [wind] from the tui direction [means] that wars will
break out and frost and cold will produce calamities.
31
This practice was very similar to the astro-meteorology authorized by the
Church. Having protected himself in advance from the attacks of fellow priests,
who might have accused him of superstition,
32
Schall sent officials to the obser-
vatory to watch for the winds on the fortnightly periods of the two equinoxes,

31
Y Wen-lung , Ta-Ming tien-yuan y-li hsiang-i tu-shuo
[The illustrated omens in the Tien-yan calendar of the Ming dynasty]
(Taipei Central Library, rare book no. 6478), 7: pp. 34.
32
Huang Yi-Long, Yeh-ssu-hui-shih tui Chung-kuo chuan-tung hsing-chan shu-
shu ti tai-tu [Jesuit attitudes toward tradi-
tional Chinese astrology and numerology], Chiu-chou hseh-kan vol. 4, no. 3 (1991),
pp. 523. A translation of this paper by Dr. Catherine Jami appeared in the conference
volume for the Temps et espace dans la rencontre de la Chine avec lEurope aux XVIIe
et XVIIIe sicles conference held in Paris in October 1991.
two solstices and four seasonal fortnightly periods.
33
However, when Schall had
to face an interrogation in court, he unwisely conflated watching for the ethers
with watching for the winds and had to admit somewhat embarrassedly that he
had perjured himself.
Rather unhappy because of Schalls neglect of hou-chi, the regents wrote:
Ou Chi-wu (the Erudite in the Clepsydra Office) confesses that It is a
tradition to set up pitchpipes in order to watch for the ethers on li-chun.
[However], all the pipes were collected and put away on the day before li-chun
and reports were made and real measurements never took place. This confession
makes it clear that the timing of li-chun has never been verified. Whenever fort-
nightly periods are reported, it is important also to report the harvest and condi-
tion of peoples lives. When the pipes are pulled out early, how can we verify the
action of spring chi on li-chun? This matter ought to be thoroughly investigated
and reported through memorials.
34
After a grim interrogation, Schall altered his confession, admitting that though
the spring chi was determined by celestial phenomena, he had sent officials out
to observe the earths chiquite a different matter. Moreover, Schall stated
that the pitchpipes had been pulled one day early because of the presence of chi
floating in the air, a chi neither hot nor cold, which had prompted the officials
to declare that the spring chi had manifested itself. Obviously this is mere
blather. An inspection of the Hui-tien (The comprehensive codes and stat-
utes) shows that there is no statute that justifies the practice of pulling up the
pitchpipes the day before li-chun. Schall was severely reproached because of
all this.
35
Adam Schall was not the only official under whose direction the results of
hou-chi were falsified. During the period of the Calendar Case, the same prac-
tices were perpetuated under Schalls successor, Chang Chi-chun . Ac-
cording to the 1664 calendar, li-chun fell on the nineteenth day of the twelfth
month; the officials sent to Shun-tien prefecture pulled out the pitchpipes on
the eighteenth day. Not only did Chang fail to report or correct this, he reported
that he had personally witnessed the arrival of spring and the removal of the
pitchpipes from the earth on the eighteenth day.
36

33
We have seen, for example, a memorial dated 22 June 1645 from the collection of
the Beijing Number One Historical Archive recording Schalls orders to astronomical
officials to predict the winds on the summer solstice. We are grateful to Mr. An Shuang-
cheng for providing us with this material.
34
Man-wen mi-pen-tang, chan 150, memorial dated the ninth day of the second
month of the fourth year of the Kang-hsi reign (25 March 1665).
35
Man-wen mi-pen-tang, chan 150, memorial dated the twenty-fourth day of the
second month of the fourth year of the Kang-hsi reign (4 April 1665).
36
Man-wen mi-pen-tang, chan 150, memorial dated the ninth day of the second
month of the fourth year of the Kang-hsi reign (25 March 1665).
On the first day of the fourth month of the following year (26 April 1665),
Minister of Personnel A-ssu-ha determined which officials had been
sent to Shun-tien to perform hou-chi, and suggested that they be removed from
office and judged by the Board of Punishments. Some of the officials involved
had already died (e.g., Pan Kuo-hsiang ) or been removed from office
for other crimes (e.g., Chu Kuang-hsien ); the others were all pardoned
because of a general amnesty declared on the fifth day of the third month (31
March 1665).
37
The only official who had to face the Board of Punishments was
Chang Chi-chun, who was ultimately pardoned, along with Schall, thanks to
the same general amnesty that had saved his fellow conspirators.
38
The power of the Catholic astronomers was dramatically reduced as a result
of Yang Kuang-hsiens accusations. High ranking astronomers were either dis-
missed or decapitated. After declining the offer several times, Yang accepted the
post of director of the Astronomical Bureau in the ninth month of 1665.
39
Having accused Adam Schall of faking the hou-chi calculations, it was now
incumbent upon the new director to apply hou-chi correctly in measuring the
fortnightly periods. This was especially so since the applicability of hou-chi

37
The combination of a severe earthquake in Peking and the appearance of two com-
ets with tails twenty to thirty degrees long prompted the emperor to declare an amnesty.
Huang Yi-Long, Yeh-ssu-hui-shih Tang Jo-wang tsai hua en-jung kao
[A study of the imperial honors won by the Jesuit Adam
Schall in China], Chung-kuo wen-hua 7 (1993). This paper was presented at the Interna-
tional Conference in Honor of the 400th Birthday of Adam Schall and the Introduction of
Catholicism into China, Taipei, October 1992.
38
Man-wen mi-pen-tang, chan 152, memorial dated the first day of the fourth
month of the fourth year of the Kang-hsi reign (15 May 1665); chan 152, memorial
dated the twelfth day of the fourth month of the fourth year of the Kang-hsi reign (26
May 1665). Chang Chi-chun was demoted to the position of vice director in the eighth
month of the same year. This was a result of his failure to report promptly the damage to
a concave simplified instrument (chien-i ) that resulted from an earthquake that hit
Peking in the third month. Only after reports by Yang Kuang-hsien and the Manchu
Director had been submitted did Chang memorialize to request repairs. Thus his demo-
tion and censure for sloppiness and negligence in handling things. Yang Kuang-hsien
, Pu-te-i [I cannot do otherwise], in Wu Hsiang-hsiang (ed.),
Tien-chu-chiao tung-chuan wen-hsien hs-pien [Supplement to
the Christian documents coming to the East] (Taipei: Hseh-sheng shu-ch, 1965), ser.
2, vol. 3, pp. 12551300.
39
Huang Yi-Long, Tzu-jih chih cheng y Kang-hsi li-y
[Debates over divination and the Kang-hsi Calendar Case], Tsing-hua hseh-pao (New)
2 (1991), pp. 24780. This article has been translated into Japanese and English. The
former, entitled Takujitsu no arusoi to Kki rekigoku, , trans.
It Takeyuki , Chgoku: shakai to bunka 6 (1991), pp. 174203. The latter,
under the title Court Divination and Christianity in the Kang-hsi Era, has been trans-
lated by Nathan Sivin, Chinese Science 10 (1991), pp. 120.
had been one of the principal issues in the confrontation between Chinese and
Western astronomies. When officials from the Astronomical Bureau were un-
able to verify the chung-chi of the twelfth month of 1665 with any precision,
Yang Kuang-hsien memorialized that the hou-chi method had long been lost,
and asked permission to search for people whose skill in calculations and
making measuring tools for hou-chi will enable them to assist your humble
subject. He also asked that the Board of Rites be ordered to secure bamboo
tubes from Mount Chin-men in I-yang, black millet from Mount Yang-tou in
Shang-tang and reeds from Ho-nei.
40
After these materials had been provided
him, Yang followed the ancient methods in performing hou-chi and personally
measured the fortnightly periodsin vain. In time, Yang applied for funds from
the Board of Works to have a hou-chi chamber built in front of his private resi-
dence. Again, he personally attended to the operations of hou-chi, and again he
was unsuccessful. Nor did his search for hou-chi experts produce any results.
Under the circumstances, Yang did not dare falsify his results. Ultimately, he
was not even able to perform the old hou-chi rituals.
41
All of these failures lent weight to the arguments of Ferdinand Verbiest
(162388) and others who, by the end of 1668, had begun to undermine the
verdict in the Calendar Case. In his Pu-te-i pien (I cannot help re-
butting), Verbiest refuted Yangs criticism of the Western methods. When it
came to hou-chi, Verbiest wrote that everyone knows that using reed pipes
and flying ashes is no way to measure the fortnightly periods. Only Yang did
not realize this. As a proof, Verbiest provided four arguments.
First: on the chun-fen fortnightly period, the sun intersects the equator. The date
is always the same in all countries. The use of flying ashes in watching for the
ethers depends on the earths chi, which may be cold or hot, dry or damp. Dif-
ferent as the chi of the earth in different countries may be, the timing of the
chun-fen fortnightly period is everywhere the same.

40
Ta Ching Sheng-tsu jen-huang-ti shih-lu [Veritable rec-
ords of the Kang-hsi emperor] (Peking: Chung-hua shu-ch, 1985), 18: pp. 910.
41
Nan Huai-jen (Ferdinand Verbiest), Hsi-chao ting-an
[Resolved cases of the Kang-hsi reign] (Vatican Library, ARSI Jap. Sin. II 67), pp. 47,
6263; idem, Pu-te-i pien [I cannot refrain from disputing], Tien-chu-chiao
tung-chuan wen-hsien hs-pien, ser. 1, p. 13. On the various versions of the Hsi-chao
ting-an, see Willy Vande Walle, Problems in Dating the Writings of Ferdinand Verbi-
est: The Astronomica Europea and the Xi-chao ding-an, in Fu-jen Catholic University
(comp.), Nan Huai-jen shih-shih san-pai chou-nien kuo-chi hseh-shu tao-lun-hui
[International Conference in Honor of the
300th Anniversary of the Death of Ferdinand Verbiest (16881988)], (Taipei: Fu-jen ta-
hseh chu-pan-she, 1987), pp. 23752; Huang Yi-Long, Kang-hsi chao she-chi li-yu
ti tien-chu-chiao Chung-wen chu-shu kao
[An investigation of Chinese works by Catholics related to the Kang-hsi Calendar Case],
Shu-mu chi-kan vol. 25, no. 1 (1991), pp. 1227.
Second: since the chun-fen period is determined by the intersection of the sun
with the equator, it remains the same for ten thousand yearsit is unchanging.
The earths chi [on the other hand] is different every year in every country dur-
ing chun-fen. If the earths chi is used to calculate chun-fen, the date will
change every year.
Third: chun-fen is a mere date, and during the period surrounding it the earths
chi is more or less the same. It is difficult to differentiate [the fine differences in
chi]. Moreover, fortnightly periods such as chun-fen arrive at a specific mo-
ment during the day, while the earths chi remains nearly constant throughout
the day and is therefore difficult to differentiate. How can one use the earths chi
to measure the precise moment of the chun-fen fortnightly period?
Fourth: [the quality of] the earths chi depends on the earths contours: it varies
depending on proximity to mountains, rivers and lakes. Furthermore, wind, rain,
clouds and fog all affect the earths chi. The timing of chun-fen [on the other
hand], depends completely on the intersection of the sun and the equator, which
occurs far from the earth. How is this related to the earth? How can one use the
earths mercurial chi to measure the unchanging chun-fen?
A careful examination of this passage shows that Verbiest has criticized hou-
chi from a rational, scientific perspective. He believes that the earths chi var-
ies depending on where it is found and that is it changeable. Therefore, it is
useless in determining the timing of the fortnightly periods.
In the tenth month of 1668, the Kang-hsi emperor ordered the Board of
Rites to determine once and for all whether people who could perform hou-chi
could be found, and whether Yang Kuang-hsien himself possessed this ability.
42
Clearly, the recently enthroned emperor had grave doubts about Yangs calcula-
tions. In Yangs response, a note of embarrassment is quite audible: Though
Ssu-ma Chiens Shih-chi recorded the size of the pitchpipes,
their usage has long been lost. As to the search for people who are capable of
performing hou-chi, it has produced no one. I myself am ill and incapable of
handling [this matter]. The emperors reply was a reminder that Yang had a
responsibility as director of the Astronomical Bureau and ought not to shirk it.
He urged his subject to persist in his search for those people whose skill in cal-
culations would enable them to carry out hou-chi.
43
In the second month of 1669, Yang was dismissed from the Astronomical
Bureau. The cited reason was that having assumed the position of director of

42
Sheng-tsu jen-huang-ti shih-lu, 27: p. 11. Article dated the wu-tzu day of the tenth
month of the seventh year of the Kang-hsi reign (28 Nov. 1668).
43
Sheng-tsu jen-huang-ti shih-lu, 27: p. 18. Article dated the ping-chen day of the
eleventh month of the seventh year of the Kang-hsi reign (24 December 1668).
the Astronomical Bureau, he was unable to correct the errors in the calendar.
44
Shortly before stepping down, Yang noted that there were three matters I have
not been able to bring to completion because of ill health. He requested that his
Manchu counterpart, Director Ma-yu of the Astronomical Bureau, present
to the court a memorial in which Yang had written that all these [matters] are
official bureau business which, [I,] Kuang-hsien, was unable to complete; they
should be clearly reported and assigned to others. Please remove me from office
for my negligence so that I may rest and quietly recuperate. Hou-chi was the
first of the three things mentioned in these memorials: clearly this was one of
the foci of the disputes of the time.
45
On the fifteenth day of the sixth month of the eighth year of the Kang-hsi
reign (12 July 1669), Verbiest was authorized to take over the business of the
Astronomical Bureau under the unofficial title of Chih-li li-fa
(Orderer of the calendar).
46
On the twenty-ninth day, the Board of Rites memo-
rialized to urge a resolution of the previous hou-chi episode in the Astronomi-
cal Bureau:
Nowadays Yang Kuang-hsien has been removed from office; he never found
anyone capable of performing hou-chi. Moreover, when the Council of Delib-
erative Officials [earlier] this year made a decision concerning the calendar, Ver-
biest stated that the practice of hou-chi is traditional; and hou-chi and calen-
drical calculations are not related. Therefore, the practice of hou-chi in Shun-
tien prefecture ought to be abolished. Since hou-chi is irrelevant to calendrical
calculations, Yang Kuang-hsiens hou-chi experiments in other places and his
search for hou-chi experts should also be abandoned.
The memorial was approved and respectfully followed on the first day of the
seventh month (28 July 1669).
47
The long-standing practice of hou-chi had
been officially proscribed.
Official and Civilian Attitudes towards Hou-chi
after the Kang-hsi Calendar Case
Though the verdict of the Calendar Case had been overturned, and Catholic as-
tronomers such as Adam Schall had been vindicated, an attack on the Western
methods soon appeared. In his Chen li yen (True words about the cal-

44
Sheng-tsu jen-huang-ti shih-lu, 28: pp. 89. Article dated the kung-wu day of the
second month of the eighth year of the Kang-hsi reign (8 March 1669).
45
Nan Huai-jen, Hsi-chao ting-an, pp. 4950.
46
Nan Huai-jen, Hsi-chao ting-an, pp. 1718; Huang Yi-long, Ching-chu chin-
tien-chien, pp. 75108.
47
Nan Huai-jen, Hsi-chao ting-an, pp. 4950.
endar) of 1673, a commoner of Wu county named Yang Ching-nan
, claimed that there were errors in the official calculations of the li-chun and
li-chiu fortnightly periods and the intercalary month.
48
His arguments repli-
cated those used by Yang Kuang-hsien in his polemic against Adam Schall.
Yang Ching-nan was the scion of a line of bookish men which had gone into
decline. One of his fathers brothers (tang-shu ), Yang Ting-shu ,
had been one of the key figures in the Fu-she . During the 1644 Calendar
Case, Ching-nan had assisted Yang Kuang-hsien. His attack on the Jesuits can
therefore be seen as part of the conflict over the New (Western) and Old
(Chinese) calendars.
Having endured the hardships of the earlier case, Verbiest was extremely
concerned, alarmed even, about this new development. In responding to the new
accusations, his stance was firm and severe: he contended that only the punish-
ment of this mean person from Chiang-nan would guarantee that none would
follow in his footsteps. Once Verbiests memorial had been presented, Kang-
hsi ordered the Council of Deliberative Officials to investigate the accusations
made in Yang Ching-nans Chen-li-yen. It is worth noting that when Kang-hsi
ordered Ming-chu minister of the Board of Warand others to examine
this matter, he not only exhorted them to avoid all prejudice and judge the
rights and wrongs of the issue through open verification, he also particularly
reminded them that you are aware that the transmission of the hou-chi method
of flying ashes ceased long ago and is not reliable.
The new calendar calculated fortnightly periods and the intercalary month
by the fixed fortnightly period (ting-chi ) method instead of the tradi-
tional average fortnightly period (ping-chi ) method. Two results of
this were that the number of days in the fortnightly periods varied by one or two
days and the intercalary month differed from that which the traditional method
would have produced. These changes led to differences in the gods associated
with auspicious days. Some popularly produced almanacs and calendars, how-
ever, continued to be arranged in accordance with the traditional methods.
49
When Yang Ching-nan attacked what he called the mistakes in the official
calendar, he said that the li-chun, li-chiu and intercalary month and the other
fortnightly periods for this year are all erroneous. The almanacs available in the
bookstores show this. Each fortnightly period falls either early or latethey are

48
For descriptions of this matter, see Nan Huai-jen, Hsi-chao ting-an (Vatican Li-
brary, R.G. Oriente. III 231), pp. 91100 (this text appears also in Wu Hsiang-hsiang,
ed., Tien-chu-chiao tung-chuan wen-hsien; Huang Yi-Long, Yang Ching-nan tsui-hou i
wei shu-kao hsi-fang tien-wen-hseh ti pao-shou chih-shih-fen-tzu
[Yang Ching-nan: the last conservative intellectual
and his attack on Western astronomy], Han-hseh yen-chiu vol. 9, no. 1 (1991), pp. 229
45.
49
Huang Yi-Long, Chung-hsi wen-hua.
uneven and irregular. Only by means of the flying ashes can the calendrical
method and fortnightly periods be illuminated.
50
For the most part, what Yang knew of calendar-making was from books. He
confessed that according to my own careful analysis of what the books contain,
even the sages were transmitters rather than creators. Even the Seven Governors
[i.e., the sun, the moon, and the five planets visible to the naked eye] and Four
Reminiscences [the imaginary stars rahu, ketu, yueh-pei and tzu-chi] have a
predictable pattern which can be examined. I have also analyzed the revolution
of stars presented in the books, though I have never made use of the armillary
sphere to verify the motions of stars. As to the flying ashes, I dare not construct
a flying ash chamber in my private residence, and there is no other means to
check the evidence. It was no doubt the prohibition on astronomical studies by
private scholars that inhibited Yang and kept him from using the armillary
sphere or building a flying ash chamber.
51
When Yang Ching-nan brought his lawsuit against the Jesuits, the reputation
of the Western methods was at its apex. The environment had changed little
since Yang Kuang-hsiens pardon in 1669. To make matters worse, Yang
Ching-nans calendrical learning lacked a solid foundation. Despite all of this,
he launched his accusations with no more basis than his own firm convictions.
This suggests how deeply the belief in hou-chi was entrenched in the Chinese
mind. Ultimately, Yang Ching-nan was condemned to be flogged one hundred
times with a stick, a sentence which was converted to forty lashes with the bam-
boo and three years in prison; all printed copies of his book and the printing
blocks were to be burned.
Based on Kang-hsis communications to Ming-chu, we can see that the hou-
chi method had ceased to hold the high position of the method of the an-
cients. But it was not until the Chien-lung reign (173695) that the gov-
ernment clearly repudiated the practice. In the introduction to the Ssu-ku
chan-shu , the authorsChi Yun and othersmade the following
comments on Tsai Yan-tings L-l hsin-shu: Since hou-chi was unreliable
and the human voice untestable, what Tsai calls the standard note (sheng-chi
chih yuan-che ) was nothing but flowery language. These writers
also criticized Liu Chins L-l cheng-shu (Book on musical
harmonics), stating that discarding the discernible note to go chasing after the
misty, ungraspable chi amounts to [chasing] insignificant things.
52

50
Nan Huai-jen, Hsi-chao ting-an, p. 95.
51
Ching-yuan tiao-fa shih-lei [Cases and statutes of the Ching-
yuan period] (Peking: Chung-kuo shu-tien, 1948), 17: pp. 20b21a. From the eleventh
century on, private astronomical studies were officially prohibited. This statute was pre-
served in the legal codes of the Ming and Ching dynasties.
52
Liu Chin, L-l cheng-shu in Ssu-ku chan-shu, vol. 212, 2: p. 116.
Another book that attacked hou-chi was L-l cheng-i hou-pien
(Sequel to the exact meaning of the pitchpipes), compiled during
the Chien-long reign. The author stated quite frankly that hou-chi was the
mere fabrication of learned men, and does not exist.
53
The arguments mustered
by the author are persuasive. For instance, in a passage that criticizes the prac-
tice of burying pitchpipes to measure each months chung-chi according to the
direction of the twelve chen (the divisions of the day) the author writes:
Hou-chi is only tested within a chamber. Now that the west of the eastern cham-
ber is the east of the western chamber, there is very little floorspace; if we expect
the chi to enter a certain pipe, the chi must have knowledge; only if the pipe is
sentient will it be moved when the chi arrives. The notion that the chi of the
twelve seasons can all be pursued by their bearings is nonsense.
Moreover, when refuting the notion that with the birth if the first yang, the
earths chi ascends, the author referred for the first time to the theory of the
earths sphericity and argued that since the earth was round, the first yang must
be born in the center of the earth. He further pointed out that
the radius [of the earth], from its center to its surface, is 12,000 li [6911 km]
. . . . If the yang is created on the hsiao-hseh fortnightly period and
reaches the surface of the earth during the winter solstice fortnightly period, then
the yang-chi will have moved 12,000 li in the space of a month, which is
400 li per day, not some small distance. The length of the huang-chung pitchpipe
is nine tsun [32.2 cm], while that of the tai-tsu pipe is eight tsun [28.6 cm].
The difference is one tsun. If, on the winter solstice, ashes fly from the nine-
tsun pitchpipe, the ashes in the eight-tsun pitchpipe should not wait until rain-
fall (y-shui ) to fly. If all the various pitchpipes can let fly their ashes on
the winter solstice, it is not necessary to use the flight of ashes to verify harmon-
ics.
The author points out here that the chi which is able to move several hundred li
in one day cannot possibly trigger the ejection of ashes from but one of a num-
ber of pitchpipes only a few tsun in length. The argument resembles that of
Wang Ting-hsiang.
L-l cheng-i hou-pien also contains refutations of those past instances of
hou-chis verification.
The Sagacious Ancestor of the Benevolent Emperor [i.e., Kang-hsi ] tested
it in accordance with the ancient methodsthey all failed. Heaven does not
change, neither does the tao . Therefore the tao of heaven and earth is con-

53
Yun Lu , Chang Chao , et al., Lu-l cheng-i hou-pien
[Sequel to the exact meaning of the pitchpipes] in Ssu-ku chan-shu, vol. 218, 120: pp.
3239.
stant, and there is nothing that was verified in the past that cannot be verified in
the present. In the past, the rulers neglected mathematics and harmonics, turned
such matters over to officials, and rarely verified them in person. The historical
records simply echoed previous documents without realizing their errors. Even if
[hou-chi] has occasionally been verified, this is not usual. As a result, it is misty
and unreliable.
Here the truth or falsity of the hou-chi practice is rationally evaluated and
Kang-hsis unsuccessful tests are mentioned to demonstrate its unverifiability.
Although hou-chi had been officially discarded during the Kang-hsi and
Chien-lung reigns, its appeal by no means vanished completely from the popu-
lar mentality. For instance, Hu Yen-sheng , an official who served as
magistrate of Ting-tao county, was a stubborn supporter of hou-chi. Hu,
styled Chu-hsan , was an expert in music and mathematical harmonics.
After his retirement, he set down his thoughts on mathematical harmonics in
Yueh-l piao-wei (An explication of the subtlety of mathematical
harmonics), a book which he dedicated to the Chien-lung emperor during a trip
the emperor made to south China. In this book, Hu discussed hou-chi at
length.
54
He contended that hou-chi appeared in the Yueh-ling chu
[Commentary to the monthly ordinance] and its methods were documented in
the Hs Han-chih and Sui-shu [History of the Sui dynasty]. How can such a
thing be misty?
In response to the comment that in the past hou-chi had often proven un-
verifiable, Hu expressed his belief that this was [because] the instruments were
not sophisticated enough, [because] the calendar had minute errors, or [because]
the fortnightly periods were not in harmony. In other words, he believed that if
the tests were inconclusive it was either because the procedure or the instru-
ments were faulty, not because of any problem with the theory itself. As to the
proper methods, only the knowledgeable can perform them; only the clever can
describe them. The gist of Hu Yen-shengs text is consistent with the attitudes
of those who loyally supported hou-chi.
Conclusion
The first concrete description of the hou-chi procedure to appear in ancient
China was that of Tsai Yung (133192), in the Eastern Han (25221
A.D.). It was not, however, until Hsin-tu Fang produced positive results through
his mechanical apparatus that hou-chi gained a wider acceptance, ultimately

54
Hu Yen-sheng , Yueh-l piao-wei [An explication of the sub-
tlety of mathematical harmonics], in Ssu-ku chan-shu, vol. 220, 2: pp. 2943.
becoming one of the greatest frauds in the history of Chinese science.
55
Despite
the consistency with which tests of hou-chi failed to lend support to the system,
the practice had great appeal because it beautifully united Heaven, Earth and
Man even as it forged a relationship between units of measurement, units of
mathematical harmonics, earthly statecraft and the heavenly fortnightly periods:
people seldom dared to question its validity. Even that great advocate of
investigating things so as to acquire knowledge (ko-wu chih-chih ),
Chu Hsi, and so scientifically puritanical a man as Shen Kua both followed the
custom and affirmed its validity.
Between the impossibility of verifying hou-chi and peoples unwillingness
to surrender it, the loss of ancient methods came to be used as an excuse and
failures came to be attributed to an inability to develop precise enough tech-
niques. People also suggested an analogy between the failure to verify hou-chi
and improper governance, thereby providing a good explanation for any fail-
ures. For instance, the Li-chi i-shu (Subcommentary to the Records
of Rites), written by Hsiung An-sheng during the Chin dynasty (265
420) contains the following passage: [If the ashes] are slightly moved, the chi
is in harmony; violent motion means the ruler is weak and the ministers power-
ful; no movement indicates a severe ruler.
56
Though the Kao emperor of the
Sui (598618) criticized this system as unreliable,
57
Chiao Hung of the
Ming still believed in the subtlety of the method and argued that this practice

55
This resembles the fabrications known as ying-huo shou-hsin (Mars
enters the Hsin Constellation) and as wu-hsing lien-chu (the conjunction of
the five planets). See Huang Yi-Long, Hsing-chan, shih-ying y wei-tsao tien-hsiang: i
ying-huo shou-hsin wei li [Astrology,
verification and forging celestial phenomena: examples from Mars entering the Hsin
Constellation], Tzu-jan ko-hseh-shih yen-chiu vol. 10, no. 2 (1991), pp. 12032; idem,
Five-Planet Conjunctions in Ancient Chinese History, trans. Edward L. Shaughnessy,
Early China no. 15 (1990), pp. 97112.
56
Hsiung An-sheng , Li-chi i-shu [Subcommentary to the Rec-
ords of Rites], in Yu-han shan-fang chi-i-shu, vol. 2, 2: p. 1061.
57
After the Kao emperor had defeated the Chen dynasty in 589, he sent Mao
Shuang , Tsai Tzu-yan , and Y Pu-ming to perform hou-chi
according to the ancient methods. The three men found that the timing of the appearance
of chi and of the flight of the ashes were different. The emperor then asked Niu Hung
about this; Nius response was: when half the ashes had flown, that meant harmo-
nious chi. When all the ashes had flown, that meant severe chi. When none of the ashes
had flown, that meant chi in decline. When the harmonious chi responds, the state is at
peace. When the severe chi responds, the ministers are indulgent. When the chi in de-
cline responds, the ruler is a tyrant. The emperor rejected Nius conclusions, pointing
out that when the ministers were indulgent and the ruler a tyrant, the state would cer-
tainly not be at peace. Given so great a variety of response in the pitchpipes within a
single year, is it likely the ruler and ministers can be so changeable? Niu found it diffi-
cult to respond to this. Sui-shu, 16: p. 394.
seems partial and almost false, but its warning to the emperor holds a mirror up
[to government]. Therefore, I prefer to believe it.
58
Wang Wen-ching was another advocate of the system:
The sages used the Way of the Gods (shen-tao ) to set up education. There-
fore the ministers who use the theory of the harmonious chi, the severe chi and
the declining chi have to practice it every month to persuade the ruler, while the
ruler uses the practice to illuminate his statecraft. This [practice] has attained the
standard of Heaven and Earth and must not be doubted on account of minute in-
cidents. There are things under Heaven which can be known but cannot be spo-
ken, which can be doubted but cannot be known.
59
Ironically, despite the overwhelming number of failures reported by those
independent scholars who tried to perform hou-chi, official astronomers went
on mechanically reproducing the flight of ashes year after year. Their fraudulent
practices, performed at the time of li-chun, had become a tradition no later than
the Ming dynasty. The reason for the disappearance of hou-chi must no doubt
be attributed to the divergence between calendrical calculations and celestial
motions, a divergence which engendered a transformation in expectations about
and usage of hou-chi, from establishing the correct dimensions of the pitch-
pipes to measuring the fortnightly periods.
Not until after the midpoint of the Ming dynasty did the scholarly commu-
nity begin to cast doubt upon hou-chi. This was due to both the influence of a
flourishing field of mathematical harmonics and the appearance of an ethos of
practical learning (shih-hseh ).
60
A member of the royal house, Chu
Tsai-y, was the most outspoken critic of hou-chi in its long history. To test the
validity of hou-chi, he followed the ancient methods, and his procedures were
scientific. For instance, he carefully chose the black millet, reeds and bamboo he
used in his experiments, and his trial of many lengths of bamboo method was
similar to the trial-and-error method employed in modern experimental science.
Based on the solid foundation of his own experiments, Chu Tsai-y severely
criticized the contradictions to the Confucian dictum ko-wu chih-chih that hou-
chi represented. So deeply planted was hou-chi in the minds of the Chinese,
however, that Chus experiments had little influence on society in general, and

58
Wang Wen-ching , Ku-yueh yuan liu, Yueh-chih-kao ,
[The origins of ancient music and an investigation of music] (Academia Sinica, manu-
script copy), 4: p. 29.
59
Wang Wen-ching, Ku-yueh yuan liu, Yueh-chih-kao, 5: p. 28.
60
Ko Jung-chin , Yin-lun (Introduction), in Chen Ku-ying
, Hsin Kuan-chieh , and Ko Jung-chin (eds.), Ming-Ching shih-hseh
ssu-chao-shih [The history of the practical studies trend during the
Ming and Ching] vol. 1 (Chi-nan: Chi-lu shu-she, 1989), pp. 111.
official astronomers continued to use their mechanical fakery to fabricate results
during the li-chun of each year.
At the beginning of the Ching, the Jesuit Adam Schall assumed the direc-
torship of the Astronomical Bureau. Though he thought hou-chi laughable, he
concocted false reports of hou-chi predictions year after year throughout the
twenty years that he held his post. In the early part of the Kang-hsi reign, Yang
Kuang-hsien initiated the Calendar Case, in which Western religion and learning
were put on trial. Hou-chi was cited as an example of the incommensurability
between Western methods and celestial phenomena. Yang took over the Astro-
nomical Bureau when Catholics who had occupied high-ranking positions in the
bureau had been removed or executed. When he was obliged to take up the di-
rection of hou-chi, his failure opened the way to Verbiests attacks and the offi-
cial proscription of hou-chi.
Verbiests refutation of hou-chi was entirely rhetorical. Persuasive as it may
sound, it was insignificant when compared with earlier criticism based on theo-
retical concerns. By comparison with Chu Tsai-ys experimental and theoreti-
cal critiques, Verbiests was far too mild. Despite this, it is a historical reality
that Verbiests criticism led to the formal rejection of the hou-chi methods. We
can see this as a side effect of Verbiests vindication of the Jesuit position in the
Calendar Case.
While traditional hou-chi had nearly evaporated by the end of the Kang-hsi
reign, during the Chien-lung period people like Hu Yen-sheng remained to
defend it. Such examples attest to the vitality of the hou-chi practice and its
persistence in the Chinese mind. Even in a childrens book published in 1904 by
Yunnan Official Bookstore, hou-chi was mentioned: The first yin is born on
the summer solstice, therefore the days become gradually shorter. The first yang
is born on the winter solstice, therefore the shadow of the sundial becomes
longer. On the winter solstice, the reed ash flies. On li-chiu , the leaves of
the wu-tung tree (Stercelia plalonifoliu) fall.
61
Even nowadays, during the waning of the twentieth century, some still at-
tempt rational explanations of the hou-chi practice. In an article entitled
Report on a Hou-chi Experiment Performed during the Summer Solstice of
1989, Liu Tao-yuan claimed that his experiments indicated that the
correspondence of the chi and the flying ashes was a case of physical oscilla-
tion related to periodic changes in the gravitational force. Lius experiments
are clearly a fabrication.
62

61
Cheng Yun-sheng and Tsou Sheng-mai (eds.), Yu-hseh ku-shih
chiung-lin [The jade forest of childrens educational stories] (Yun-nan
kuan shu-ch, 1904), 1: p. 14.
62
Liu Tao-yuans paper was presented at the 1989 Conference on the History of
Physics in High School Education, held in Chengdu, Szechuan Province. We are grateful
to Mr. Tai Nien-tsu for providing us with this information.
Despite a conflict between theory and experimental results, hou-chi re-
mained an element of the Chinese intellectual experience for two thousand years
because men were fascinated by the falsifications of a few charlatans. Truly this
is an extraordinary occurrence in the history of science.
Translated by Pingyi Chu (Academia Sincia, Taipei) and Samuel Gilbert
(Department of History, UCLA)
107
Medicine and the Changes Are One:
An Essay on Divination Healing
with Commentary
Judith Farquhar
AUTHORS NOTE: I am indebted to Alison Greene and Hong Liyi for intrepid
and discriminating research assistance. My warmest thanks to them and to Dr.
Zhao, Hospital Director Li Mingzhong, Dr. Niu Wenquan, and Foreign Affairs
Director Shi Changxiang, all of whom were of invaluable assistance during the
field research on which this article is based. Gratitude also to Margie Hattori,
Jim Hevia, Ann Stewart and Jing Wang, who critically read the manuscript and
provided useful advice, and to Faith Wallis who advised on medieval European
astrology. Field research was supported with a grant from the Henry Luce
Foundation and the Curriculum in Asian Studies at the University of North
Carolina.
[Judith Farquhar is Bowman and Gordon Gray Associate Professor of
Anthropology at the University of North Carolina, Chapel Hill. She is the author
of Knowing Practice: The Clinical Encounter of Chinese Medicine.]
* * *
ccasionally in the course of anthropological research one encounters a
text that demands to be read beyond the confines of an ethnographic
project. The essay I present and comment on in this article is such a text;
it draws on complex historical resources, makes an argument for epistemologi-
cal legitimation of divination, and demands close technical reading. More
importantly, it was written for me. In its direct address to me as a visiting
anthropologist, it claims an audience broader than the Chinese communities in
which its author has spent his life. My comments will clarify some of these
elements in the essay where straightforward translation is inadequate to do so,
and I will conclude with remarks on commentary itself as a way of pursuing
ethnographic and historical study of the indigenous Chinese sciences.
O
But first I want to acknowledge the specific inspiration for this project
provided by Nathan Sivin and his book on Chinese medicine. The history and
the commentary presented in this work contain, I will argue, important lessons
for any reading of Chinese cultural materials.
By the time his study of traditional Chinese medicine (Zhongyi )
appeared in 1987, Sivins book in manuscript form had long served me as any
good history of science should: Traditional Medicine in Contemporary China
(TMCC) was an indispensable guide to the particular historiographic challenges
presented by Chinese medicine. It was also a source of informed and considered
views on historical materials and a systematic reference work on technical
terminologies. The book did not, however, set out to provide a continuous
narrative history of something called Zhongyi.
There are good reasons for this. As Sivin himself pointed out, what he calls
classical medicine in the book refers not to the theory and practice of a
coherent group, but to the records left by the most literate and scholarly
representatives of several traditions. . . . It is almost exclusively [the] elite
literary tradition that has survived in modern China (1987: 2223). And he
proceeds to gloss Zhongyi as precisely this literary heritage, noting that many
other traditions of healing associated with less literate or scholarly groups have
been written out of the modern specialty of traditional Chinese medicine. How
should a history of an essentially modern objecta clinical and theoretical
specialty purged of superstition and selectively assembled from textual
materials that have come to us by diverse and sometimes unknown pathsbe
written?
This is not to say that a history of Chinese medicine is impossible. Others
have produced very ambitious narratives on this subject in Western languages
(Wong and Wu 1932; Unschuld 1985, 1986) and the history of Zhongyi is
taught in China with a growing number of comprehensive textbooks and
specialist studies (Beijing College of Traditional Chinese Medicine 1979; Jia
Dedao 1979; Ren Yingqiu 1982; Cui Xiuhan 1983; Si and Gong 1988; Kong
Jianmin 1988; Zhao 1989; Fu Weikang 1990). Sivin himself continues to write
on the social history of medical practices in various eras, while working with
younger historians who are providing us with very rich social and cultural
contextualizations of healing specialties (e.g., Chiu 1986; Keegan 1986; Hanson
1995, forthcoming; Skar 1995, forthcoming). But TMCC should not be seen as a
mere preliminary to the real history that is eventually to appear in English.
The very structure of the book is historiographically instructive, I believe,
realizing certain social relations of scholarship which I have sought to emulate.
Let me first explain what it is I find exemplary in Sivins study of Zhongyi,
and then follow his example with a microcosm of my own, presenting and
commenting on an essay about divination medicine written in 1993 by a doctor
Judith Farquhar: Medicine and the Changes Are One 109
practicing in a county town of Shandong province.
1
With this exercise I hope,
first, to demonstrate that Chinese medical writing can be read on its own
(translated) terms, and second, that history and ethnography present ethical and
political challenges to which our writing would do well to respond more
explicitly.
TMCC as Exemplar
Why is TMCC organized the way it is? Why should Sivin have chosen a 1972
textbook that is clearly sketchy, shallow, over-politicized, and (if relied upon
uncritically) clinically misleading to anchor the book as a whole?
2
Even though
The Revised Outline of Chinese Medicine (ROCM) forms only the second half of
the book, it is nevertheless this text that organizes the whole: the commentary
provided in Sivins introductory study both relies on citations of descriptive
sections from ROCM for its coherence and takes problems presented by that
work as the occasion of its useful clarifications.
One clue to the privilege this work is accorded appears in the Preface to
TMCC: So much misinformation crowds the shelves of European and
American bookshops that a Chinese account is badly needed (p. xx). This
remark implies that a Chinese account will somehow by its very nature avoid
that particular category of misinformation. We can detect here a certain
nominalism: Chinese medicine in a sense can be nothing other than the
medicine primarily practiced in China by Chinese people, and what these
practitioners say about their field should thus be accorded definitional priority.

1
In keeping with standard anthropological practice, and because divination medicine
is a potentially sensitive topic, I disguise the identity of the author of this unpublished
essay, as well as the identities of his patients.
2
I will not here consider the practical exigencies that may have led to Sivins
authorial decision to use The Revised Outline of Chinese Medicine (ROCM) in this way.
(He discusses the conceptual advantages of the work in TMCC pp. 3336.) Though the
time required for translating a technical medical work systematically (in my view an
invaluable contribution of the TMCC project), and the relative scarcity of modern
textbooks of Zhongyi in North American and European libraries when the project was
begun, may have also been important determinants of the books structure, I wish to
emphasize a different reading of it. Once published, a book speaks the authors
intention in a way that is somewhat removed from the author himself. The final decision
to allow a work to appear under ones name expresses a willingness to adopt the role of
the-author-who-intended-this. However complex and constrained the process of making
a book has been, its final form can be held to account as a meaningful structure in
relation to many contexts beyond that of whatever lies in the heart of the historical
person whose name adorns its cover. And of course authors write in full awareness of
these reading practices, adjusting their product to them as they work.
Thus in discussing Porkerts (1974, 1983) systematizing and purifying approach
to Chinese medicine, Sivin explicitly decides against presenting an idealized
picture of an unchanging art (p. 33n).
Moreover, the book tacitly acknowledges the historians indebtedness to
contemporary categories: like a genealogy which begins with the living and
traces the particular, contingent and not always proper liaisons that have
engendered the present, a history begins with and frequently returns to those
objects of interest (from bodies of knowledge to institutions) that populate the
contemporary world. When it comes to topic formation, modern categories rule
even the most historicist methods. This being the case, it is an important gesture
on the part of a North American scholar to allow a Chinese Communist
textbook published in 1972 (the product of a tragic era, p. 36) to delineate the
contours of a scholarly study in English.
Of course, as readers of Chinese Science know, no text or author (including
himself) escapes Sivins criticism completely, and the commentary of TMCC
often takes issue with interpretations of terminology and tradition that appear in
ROCM. Commentary is itself an interesting function on which to reflect,
however. A hermeneutic style of commentary, which Sivin eschews (p. xix),
would restate the import of the text in more abstract, generalizing, or analytic
terms; it would deliver up a deeper or truer meaning than the original authors
themselves saw fit to explain. Sivin instead adopts the stance of translator,
adapting a foreign linguistic medium to allow the chapters of ROCM to
summarize themselves (p. xix). Translation of this kind involves extended
discussions of terminology, for which TMCC is justly admired. And it requires
that certain linguistic and logical assumptions built into Chinese and absent in
English be elucidated. At the same time, an appropriate ground must be cleared
in the target language, English, allowing readers to bracket their assumptions
about bodies, nature, and language so that translations from the Chinese can
become legible.
In such a project commentary claims to extend and enable rather than restate
or replace the original text. All this is part of the work of translation, which
negotiates complex historical, political, and logical conditions inhering in both
source and target languages.
3
In this work translators tend to cast themselves as

3
For interesting discussions of translation as a theoretical problem in itself see
Benjamin 1968 [1955]; Asad 1986; Latour 1988: 6567; Latour 1993; Haraway 1991:
19596; Niranjana 1992; Eoyang 1993; Lydia Liu 1993. See also Behar 1993 for a
feminist ethnographic exploration of the concept. One of the most significant conditions
for translating Chinese medical writing into English is the pervasiveness of an empiricist
epistemology in North American middle-class common sense, including our assumptions
about medicine and bodies (Foucault 1973, 1977; Gordon 1988; Good 1994). I have
elsewhere discussed the implications for scholarship on Chinese medicine of a naive
the humble servant of both the original text and the reader. The commentary
claims only to return the reader to an appreciation of the powers of the original
text itself.
But Sivin is historian as well, of course, promising in his Preface to convey
what [ROCM] suggests about the past and future of the millennial Chinese
medical tradition (p. xix). As he makes very clear, no handbook of medicine
read in isolation from its scholarly environment could suggest much about a
tradition, especially when that tradition is characterized as both singular and
millennial. Allowing any text drawn from the huge archive of Chinese medicine
to exemplify so much requires the work of more than a translator or commen-
tator, it demands that the historian at least represent Chinese medicines larger
past while suggesting its future.
Interestingly, much of this labor in TMCC is done in footnotes which
evaluate Chinese sources and relate them to each other, as well as in an
introductory discussion of Chinese patterns of writing, reading, and education
that cites roughly 200 premodern works (pp. 2130). In other words, TMCC is a
history which places the archive and the process of reanimating it at the center
of consideration. It is, thus, neither a social history of medical institutions nor a
history of disease, a fact which must disappoint many medical scholars more
interested in comparison than in change. Sivins emphasis on sources and their
intertextual relations tends to return us to an even more serious reading of
ROCM: what his presentation makes most clear is that this little textbook
belongs to an ancient and always active tradition of writing and rewriting. We
can now see that the Revised Outline neither naively reproduces nor violently
erases its literate past. TMCC hovers between philology and a Foucauldian
archaeology, turning away from grand narrative history to do the hard work of
letting Chinese medicine speak in another language.
An Essay on Divination Medicine
The translation exercise which follows cannot claim even these sensible goals.
The speaker in question is neither an institution (as in ROCM) nor a
committee claiming to represent one. Dr. Zhao, the essays author, is a self-
taught licensed private practitioner who sees patients in a middle-sized county
seat in Shandong province. I visited him several times in the summer of 1993
and again in 1995. He speaks less for Chinese medicine as a whole than for a
specialty he calls Changes studies (yixue ). Thus the aspect of his
practice Dr. Zhao emphasizes is divination, a specialty in which he claims

empiricist approach to scientific representation (Farquhar 1994b; on representation in
science, see the papers in Lynch and Woolgar 1990).
expertise, in a town where fortune-telling that draws on the Zhouyi ( , Book
of Changes) and other Chinese mantic traditions is at least a newly popular
hobby.
But Changes studies are presently quite marginal to the preoccupations of
the Chinese medical establishment. Whatever interests may be speaking through
Dr. Zhaos essay, they are not those of mainstream Chinese medicine; as a
private (getihu ) practitioner and a somewhat eccentric rural intellectual,
he invokes elements from the Chinese medical and philosophical archive which
do not appear in textbooks like ROCM and its officially sanctioned successors.
Even so, writers in contemporary China must position themselves in relation to
the language and logic of grand public campaigns and state policy; at the same
time a businessman like Dr. Zhao must consider the wishes and practical
concerns of his clientele. These concerns can be seen in the essay I reproduce
below; to make them and other local interests more visible, my footnotes extend
and contextualize some of Dr. Zhaos references.
* * *
TRANSLATION
Comments on the Utilization of Changes Studies
in Syndrome Differentiation
People have long seen the Zhouyi, or Book of Changes, as a work of divina-
tion (shi ) with a very dense superstitious coloration, hence it has not been
widely used by intellectuals.
4
I particularly enjoy Changes studies, however,

4
Superstitious (mixin ) is here used in the special sense it has acquired in
public discourse since 1949 (see Anagnost 1994). In popular usage the term refers to all
kinds of religious and magical beliefs and practices. Dr. Zhao is aware that controls have
lessened over many of the activities considered within Maoist discourse to be feudal
superstitions, once expected to wither away under socialism. In his town there is an
active Christian congregation and a growing market for religious goods used in domestic
worship, all officially superstitious. Still, he here acknowledges the stigma that still
attaches to divination practice and classic Chinese works of pseudo-science (a Maoist
term). A dogmatic Maoist approach would classify the Zhouyi and divination practices
associated with it as part of the dross that must be discarded in the nationalist search for
what is valuable in Chinas traditional culture.
especially what I have read in reports and periodicals published by contemporary
Changes specialist Professor Liu Dajun, who posits connections between
Changes studies knowledge and enterprise development, as well as science and
technology.
5
He makes this precious age-old historical classic better serve the
human race. I find this deeply inspiring.
In accord with the ancient way, in which medicine and the Changes are
one, I became both a medical worker and a practitioner of Changes studies,
[believing that] they ought to be interrelated, and Changes studies knowledge
should be made a resource for Chinese medical syndrome differentiation.
6
I dis-
covered that the book Yuanhai ziping was particularly penetrating in
its discussion of the generation and checking relations, control and transforma-
tion relations, and opposing and harmonizing (chonghe ) relations of Five
Phases theory;
7
in particular it coordinates the Heavenly Stems (jia , yi ,
bing , ding , wu , ji , geng , xin , ren , gui ) and the Earthly
Branches (zi , chou , yin , mao , chen , si , wu , wei , shen
, you , xu , hai ) into their least common multiple, 60, making a peri-
odic series commonly called the Jiazi wreath, which is used to record
years, months, days, and hours in a clear progress; according to the year, month,

Even so, he overstates the books neglect. It has been my impression as an ethnogra-
pher in several parts of China during the 1980s and 1990s that rural intellectuals in
particular have maintained a hobbyist approach to The Book of Changes, and that it has
been preferred as a focus of such activity to all other Chinese classics. Few would have
dared to practice divination for pay, however; this would be seen as swindling the
masses. Diviners and fortune-tellers practicing on street markets in Dr. Zhaos town
denied charging fees for their services; in conversations with me (a foreigner assumed
correctly to have close ties with the local government) they presented their practice as a
hobby they pursued while peddling trade goods.
5
Liu Dajun is a professor in the Department of Philosophy at Shandong University.
He is the chief editor of Zhouyi yanjiu , a journal for Changes studies
established in 1988, and author of many studies of the Book of Changes. In keeping with
the 1980s rhetoric of nationalism, modernization, and scientization, he argues that the
Zhouyi is a work of naive materialist mathematics showing strong commonalities with set
theory and number theory and (by implication) historically outranking modern
mathematics by virtue of its great age. It is on this basis that he develops the links to
science and technology mentioned by Dr. Zhao.
6
See Smith 1991, for various relationships between Changes divination and medical
practice in premodern China. Though Dr. Zhao here implies that the unification of the
Changes and medicine would be an innovation, it is perhaps more accurate to see his
position as an attack on an arbitrary division that has been introduced between medicine
and other indigenous scientific traditions in the twentieth century.
7
This Song dynasty work by Xu Sheng is one of many similar works that have been
recently republished in cheap paperback editions. In the early 1990s these works began
appearing prominently at bookstalls in Dr. Zhaos town. For discussions of Five Phases
theory in medicine, see Porkert 1974, Sivin 1987, Liu Yanchi 1988, and Farquhar 1994a.
See Figure 1 for the modern conventional arrangement of the Five Phases and their
generation and checking relationships.
day, and hour of a persons birth this generates four couples of stems and
branches [for each birth time], commonly called the Eight Characters.
8
Although
this is used mainly to tell peoples fortunes, it takes the generation and checking,
opposition and harmony, control and transformation relations of Five Phases
theory as its theoretical foundation; thus it is not hard to see that this book arises
from a dialectical materialist impulse, and is not a superstitious work.
9
Its just
that its ancient characters are simple, its reasoning is deep and subtle (shenao
), and its dialectical relations complex, making it hard for the novice to enter
its gates, so it has not been accepted by many readers.
In Chinese medical theory the Five Phases are used to explicate illness syn-
dromes (lunzheng ), and the five kinds of substancesWood, Fire, Earth,
Metal, and Waterare divided up to stand for the five visceral systems of the
body, i.e. the Lung (Metal), Liver (Wood), Kidney (Water), Heart (Fire), and
Spleen (Earth) systems. [See Figure 1] Hence when I differentiate syndromes and
deploy therapy, I combine this with the patients eight characters to make a com-
parative analysis. I discovered that, inferring the Five Phases from a persons
eight characters,
10
they really do reflect (fanyingchu ) a persons charac-
ter (xingge ) just as [they indicate] the excesses and debilities of ones vis-

8
Dates expressed in terms of the sixty stem-branch couples are fundamental to many
systems of divination. See Smiths (1992, 1993) extensive studies of divination systems
and almanacs, which are widely used to select auspicious days for important activities.
Throughout his essay Dr. Zhao treats stem-branch divination as an integral part of the
Changes studies tradition. But as Richard Smith reminded me in a preliminary reading of
this paper, the Zhouyi itself does not refer to a stem-branch system. Rather, like other
divination systems in China, Dr. Zhaos technique is conventionally associated with the
Changes, lending it an ancient pedigree and the (nowadays problematic) prestige of a
Confucian canon.
9
This declaration that Five Phases theory is a foundation for the numerology of the
Book of Changes amounts to a bid for legitimation of divination practice. Throughout the
course of Chinas twentieth-century crisis of nationalism and science, scholars of
medicine have been culling the literature to develop the essence and discard the dross
(this became a pervasive Maoist slogan) of traditional medical practices. In this project
the field of medicine has developed its own characteristic problems of legitimation. Thus
many kinds of knowledge that could be dismissed as superstition, such as Five Phases
theory, have been depicted by some authors as reflecting a naive dialectical materialist
impulse among the ancient Chinese masses, a result of their empirical exploration
of the natural world. This language of scientific Marxism allows ancient correspondence
systems to appear as theoretical foundations of effective healing, i.e. service to the
people, an unassailable good. Here Dr. Zhao borrows the legitimacy of a scientistic and
nationalistic argument in the world of medicine, using it as if it were unquestionable to
provide a foundation for his argument that Changes divination is not superstitious but
scientific.
10
Actually he does not infer the connections, he classifies them following tables of
correspondence between the Ten Heavenly Stems and the Five Phases to be found in
Yuanhai ziping. Smith 1991 emphasizes the inseparability of divination from yinyang
and Five Phases logic throughout most of its history.
ceral systems and the Chinese medical diagnosis. At times when I have had great
uncertainties about the diagnosis and treatment of a difficult case,
11
I have ana-
lyzed the patients eight characters and gotten a satisfactory therapeutic outcome
as a result. Below I describe how I have used Changes studies (the Yuanhai zip-
ing method) in combination with the diagnosis of disease.
CASE 1: Qi depletion in Spleen and Lung Systems
12
Sufferer: He X X, male, born September 17, 1946 at about 10 AM. Resident of X
county, Y village, a farmer.
Chief complaint: Constrictive pressure in chest (xiong men ),
13
shortness of
breath, abdominal swelling more marked after eating. Duration more than 2
months.

11
Dr. Zhao reemphasizes the problem of medical uncertainty below. For a classic
study of the problem of uncertainty in biomedical training, see Fox 1959.
12
In keeping with my practice in earlier publications on Chinese medicine, I
capitalize certain translated technical terms to indicate that their referents are quite
different from the objects denoted by English words like spleen and lung.
Note that the structure and style of the two case histories included here is quite
familiar from biomedical practice, though it is modified to accommodate three of the
four examinations (asking, looking, listening/smelling, and palpating) typical of
Chinese medicine. This format, with minor variations, is typical of hospital charts and
clinic records in traditional medical institutions as well as of published medical cases (an
important genre of traditional medical writing) in the textbooks and journals of
traditional medicine. Case histories in Chinese medicine frequently use the asking
examination as a history (in the biomedical sense) which emphasizes the patients
previous fruitless efforts to find adequate care in competing institutions of biomedicine
and traditional medicine.
13
This symptom is often reported in clinics. It refers to a sensation of discomfort and
constriction in the chest, and also indicates that the sufferer has worries and is not happy.
Kleinman has noted this symptom as a somatization of dysphoric affect in his work
with psychiatric cases in China (Kleinman 1977: 6, 1980: 13878, 1986: 83). Though I
wish to avoid the dualism of the notion of somatization here, there is little doubt that
both cases discussed herein would be diagnosed as emotional disorders (e.g., depression)
in the United States. Dr. Zhao wrote the document partly to demonstrate to me his
expertise in treating neurasthenia (shenjing shuairuo ), though he does not use
the term in the text. Kleinman and Kleinman note that 87%93% of the neurasthenia
cases they studied in 1980 Hunan fieldwork could be diagnosed as depression (Kleinman
and Kleinman, 1985: 436).
Figure 1. Schematic of five phasesvisceral system relations
[Liver]
Wood
[Heart]
Fire
[Kidney]
Water
[Lung]
Metal
[Spleen]
Earth
[Liver]
Wood
[Heart]
Fire
[Kidney]
Water
[Lung]
Metal
[Spleen]
Earth
Respiratory
symptoms
Digestive
symptoms
(elevated Heart
functions)
Spleen functions
harmed
(Spleen system
cannot transmit
and transform
foor and drink;
dampness
accumulates)
Lung system
doubly disabled
Figure 2. Five phasesvisceral systems pathology in Dr. Zhao's case 1.
Key
generation excessive generation
checking excessive checking
Spleen system cannot
generate Lung functions
Asking examination: Sufferer normally healthy. No history of infectious disease
in the family.
14
Two months ago, due to the entanglements of his responsibilities,
he became sad and worried. While assisting a neighbor with the plowing, he felt
pressure in his chest and abdominal swelling. He bought some Chest Opening
Qi Channeling pills,
15
and after taking them twice and experiencing some diar-
rhea he felt somewhat better. Within two days the symptoms returned. So he
went to a specialist clinic in the county hospital for treatment. In the course of
the examination there, they suspected a tumor in his chest and arranged for him
to go to the Provincial Hospital for surgery. But the patient was afraid to do this,
so he went to the Army Hospital in Z City for an examination. The results were
not consistent with those from the county hospital clinic; tumor was ruled out
and he was treated for allergic asthma. After 19 days without seeing any im-
provement he started going to doctors of Chinese and Western medicine in
nearby villages.
16
Many doctors and more than a month later, nothing had been
obviously effective; the above doctors had all taken a draining method to be the
key to their treatment, which rather depleted the patients strength (tizhi ).
17
Eventually he grew afraid to eat or drink anything (no matter how little he ate the
pressure in his chest, shortness of breath, and abdominal swelling became un-
bearable), and his family was very worried. Through a friend they heard that I
used Changes studies in conjunction with Chinese medical diagnosis and treat-
ment of disease, and that I had successfully treated many difficult syndromes.
18
They came to my clinic seeking treatment on March 2, 1993.

14
This observation is conventional for medical case histories, probably influenced
by state epidemiology, which tracks infectious diseases.
15
Chinese patent medicines such as these are widely available without prescription
in state-run and private pharmacies. In Dr. Zhaos town, they are most easily obtained in
small private and collective clinics, where one need not consult with the doctor to
purchase them.
16
This pattern of care-seeking is quite typical of patients with chronic illnesses in the
developed parts of modern China. See Romanucci-Ross 1969, Kunstadter 1976, and
Janzen 1978 for classic medical anthropology studies of patient strategies in pluralistic
medical fields elsewhere in the world.
17
The various draining (xiexia ) methods available in Chinese medicine are
recommended in cases of repletion where orthopathic (zheng ) qi has not yet been
negatively affected.
18
Under conditions of market reform, the number of private medical practices
(getihu doctors, mostly of traditional medicine) is increasing fast in towns like
Dr. Zhaos. Here he demonstrates his reliance on word-of-mouth advertising to build his
practice, which is not yet thriving, though he has a number of very loyal patients. I have
elsewhere discussed some of the strategic business concerns of getihu doctors as they
affect the nature of their medical work (Farquhar 1996).
Looking examination: The sufferers spirits wan and sallow (jingshen qiaocui
),
19
facial color without brightness, emaciated and weak with rather
marked shortness of breath. Tongue pale and tender, coating white and greasy.
Pulse: Left pulses sinking, fine, and weak; right pulses hollow, large, and
rapid.
20
[Discussion]: Noting that the patient had been sick for some time, and had un-
dergone therapy in many places without effect, and the patients thought proc-
esses were already seriously affected, I had to see this as a special case. So first I
worked out the eight characters of the patients birth date, analyzing his lunar
birth date of 1946, 9th month, 17th day, about 10 AM, as follows:
Year Month Day Hour
Stems bing
(Fire)
wu
(Earth)
wu
(Earth)
ding
(Fire)
Branches xu
(Earth)
xu
(Earth)
wu
(Fire)
si
(Fire)
These eight characters are excessive in relation to Fire and Earth phases and
they completely lack Metal, Water, and Wood phases. According to what it says
in the Yuanhai ziping, if the Five Phases and the Eight Characters are incomplete
the controlling and transforming functions of the visceral systems lose equilib-
rium, and its easy to get sick. With excessive Fire the personality is anxious and
irritable. Excessive Earth leads to over-eating and drinking and static retention of
fluids and dampness. This person lost regularity in the transmission and trans-
formation functions of the Spleen system due to sadness and worry.
21
Conse-

19
This is an interesting use of the term jingshen , spirit or spirits, though it is
not at all unusual in Chinese medicine. This term is used to translate the psyche part of
psychiatric, and in educated speech tends to refer to all manner of mental phenomena
and matters of consciousness. With a slight variation in pronunciation, however, it can
indicate bodily vitality and vigor. It is perhaps a self-conscious practice in Chinese
medical discourse to mainly use the term in the second sense, confining the appearance
of jingshen in its mentalist sense to Western medical contexts. This could be a sly
substitution of a holistic and materialist reading of the term for one that has been
excessively influenced by a cosmopolitan Cartesianism. Here Dr. Zhao uses the term
somewhat ambiguously, since he is worried about the patients thought processes (see
below) yet speaks of his spirits as observably wan and sallow.
20
See Sivin 1987 for an introduction to pulse diagnosis, and Kuriyama 1987, 1994
for historical considerations of pulse lore and its implications for the history of
perception.
21
The seven sentiments (qiqing ) are treated as physiological forces in
contemporary Chinese medicine. They are defined in the Abridged Dictionary of Chinese
quently he had abdominal swelling for a long time along with poor appetite be-
cause Earth was unable to generate Metal. Qi was depleted and breathing was
difficult as a result. The Lung system was depleted, losing regulation of its
clearing and descension functions, so there was oppressive pressure in the
chest.
22
The pale and tender tongue, and the fine, weak, hollow, and large pulse
images all suggest a qi depletion syndrome. This is a case of a root depletion
showing a branch repletion.
23
According to Eight Characters analysis, one
should restrain Fire and clear out Earth so that Lung Metal can regain generative
capacity and no longer be checked by Heart Fire; this strategy can be combined
with Chinese medical treatment principles to deploy the therapy.
24
Treatment principle: Restrain Fire and dredge Earth, treat root and branch si-
multaneously.
Prescription: Skullcap Heart-Draining Decoction together with Six Lords De-
coction.
25
Five doses.
Second examination:
Patient returned April 2 to be reexamined , having finished the medicine. The
pressure in the chest was gone, abdominal swelling much reduced, and patient
was eating more. But he still experienced shortness of breath on exertion, and he
was not sleeping well. I wrote the following prescription to treat the physical
exhaustion.

Medicine (Editing Committee of the Dictionary of Chinese Medicine, 1979) as follows:
Joy, anger, sorrow, thought, grief, anxiety, and fear are the seven types of emotional
(qingzhi ) activities which make up the conscious human response to external
events. Sorrow and anxiety are particularly likely to harm the Spleen system.
22
Here as elsewhere in Dr. Zhaos essay where he refers to the generation and
checking relations of the Five Phases, see Figure 1 for the visceral system affiliations of
the Five Phases and their normal modes of influence on each other. In this case the qi of
breathing loses its regularity when the Metal phase is weak because the Lung system is
affiliated with Metal.
23
This root-and-branch logic is particularly emphasized in complex cases of
repletion (shi ), i.e. obvious swelling or clear perceptions of pressure or fullness,
which fail to improve with time. Dr. Zhaos analysis here amounts to a criticism of the
shallow and inattentive medical care this patient has previously received, since all
previous treatment attempted to drain the repletion without noticing that it was caused
by a more fundamental depletion of qi, or more properly, a functional debility of the
Lung (Metal) system.
24
See Figure 2 for a summary of Dr. Zhaos strategy based on Five Phases mutual
production and mutual conquest relations.
25
This combination of formulae probably included the following drugs, in order of
relative importance: baical skullcap, goldenthread, Chinese foxglove root, anemarrhena,
and licorice. Translations for drug names here and below are taken from Shiu-ying Hu
1980; this source uses Wade-Giles romanizations, which I have converted to pinyin
where there is no English equivalent of the drug name.
Prescription: Spleen-Restoring Decoction along with Calming Stomach Pow-
der.
26
Five doses.
[Third examination]
By April 7 when the patient came to the clinic all his symptoms had cleared
up. Because he had been subjected to excessive use of Spleen-Stomach draining
medicines before coming to my clinic, resulting in damage to Lung system qi, I
prescribed Bolster Middle Supplement Qi Pills and Magnetic Pearl Pills to be
taken for half a month.
27
Half a month later the patients wife came by to report
that he was completely cured and to express her thanks.
CASE 2: Organic Neurological Syndrome (Visceral Systems Agitation Syn-
drome)
28
Sufferer: Yang X X, female born July 24, 1938, cadres wife.
Chief complaint: Heart palpitations, dizziness, insomnia, excessive dreaming; for
the last two months, constant sighing, tendency to weep.
29
Asking examination: The patient herself reported that in the winter of 1991 her
family had a dispute and quarreled. After being angry she took a chill and got
facial nerve paralysis, which was basically cured with three months of treatment.
In the spring of 1992 she experienced dizziness, heart palpitations, insomnia, and
excessive dreaming. She dosed herself with Xindan Anding (a tran-
quilizer) and Guweisu (a vitamin and glutamate compound)
30
and her
condition improved. But her previous symptoms returned when she became very
angry on another occasion. Sometimes she could not sleep for the whole night,
and felt extremely agitated. Took medicine from the County Hospital of Chinese
Medicine for six months without much effect.
31
In the fall of 1992 she again
manifested excessive sweating, frequent sighing, sadness and weeping; she felt
this time it was worse than before. Each time she heard of a renowned doctor she
would go to see him; in the last three months she has gone to no less than five or
six places with less than ideal results. Lately she heard that my clinic combines

26
This combination of formulae probably included the following drugs: ginseng,
baishu, huangqi, China-root, longan fruit, Chinese angelica, Chinese senega, sour jujube,
muxiang, cangshu, magnolia bark, orange peel, licorice, ginger, and jujube.
27
These are Chinese patent medicines of the tonic, or system-bolstering type.
28
The first term is a biomedical diagnosis; the second, in parentheses, is a Chinese
medical syndrome name.
29
It is typical of Chinese medical symptom lists to mix conditions North Americans
might consider to be psychological or at least non-medical (dreaming, weeping,
sighing) with terms that translate more directly into English as medical symptoms
(dizziness, heart palpitations).
30
Both of these are Western patent medicines.
31
This hospital provides the most high-status (and, for cadres and their families,
most reimbursable) traditional medical care in Dr. Zhaos area.
Changes [study with medical] diagnosis of disease, and, being very curious,
came to my clinic for treatment on October 16, 1992.
Looking examination: The patients eyes were dispirited, very tired looking,
occasional sighing, eyes filled with tears. Tongue enlarged and tender, little
coating, showing teeth marks along edges.
Pulse: Sinking, fine, and rapid, especially pronounced at the guan section.
32
[Discussion]: Because this patients condition had been rather prolonged, and
internal injury (anger harms the Liver system)
33
was rather serious, and be-
cause she had been to many clinics without marked improvement, this seemed to
be a special case. So I analyzed the eight characters of the birthdate as follows:
because the patient was born on the 24th day of the 7th month of the year 1938,
around 4:00 in the afternoon, the eight characters are:
Year Month Day Hour
Stems wu
(Earth)
geng
(Metal)
gui
(Water)
geng
(Metal)
Branches yin
(Wood)
shen
(Metal)
wei
(Earth)
shen
(Metal)
The patients Eight Characters have four Metals, two Earths, one Water, one
Wood, and no Fire.
Analysis: This Eight Character picture of 4 Metal, 2 Earth, 1 Water, 1 Wood, and
no Fire gives no method of restraining (zhi Metal, and Metal is [in this con-
figuration] already quite dominant (note that the illness appeared in the first
month of the fall). The Wood phase is weak in the fall and was being checked by
Metal, moreover Autumn Wood is classified with Dry Heat, and easily generates
fire (Liver Fire). Thus Liver [Wood] yins normal dispersion functions were
inadequate. In addition anger had harmed the Liver system producing insomnia,
dreaming, and dizziness.
34
Because [the Eight-Character picture] lacks Fire the
Heart system was weak and produced palpitations as well as an enlarged and
tender tongue image. The stem of the day, gui (Water), was being checked by
the branch character wei (Earth) below it;
35
Metal was dominant, but due to

32
The guan section is one of three positions on the surface of the wrist where the
pulse is palpated. See Sivin 1987: 31415.
33
This is a quote from the Plain Questions (5: 37, 67: 376) which (like many other
pithy formulas) is often repeated in medical work and functions as a rule of thumb.
34
The Liver system controls (among other things) the regular upward and outward
dispersion of qi. Irregularities in this function would produce the symptoms listed.
35
In keeping with the ziping systems characteristic methods, Dr. Zhao here accords
a special privilege to the stem and branch of the day. In addition he calculates influences
the lack of Fire was unable to generate Water,
36
hence the lack of coating on the
tongue; Wood was being checked by dominant Metal (the Liver is the firmness
visceral system, Wood seeks to ramify outward), the pressure was too great,
hence the frequent sighing. With regard to the sadness and weeping, it is espe-
cially easy for a patient with these Eight Characters to manifest yin depletion,
and the pulse image of sinking, fine, and rapid generally belongs to yin depletion
syndromes. Thus she should, according to Chinese medical theory, be treated for
Visceral Systems Agitation Syndrome. Because this case is classified as Liver
Wood dispersion functions weak, checked by Metal, with excessive pressure, and
because [Lung] qi is excessive producing oppressive constraint on Liver qi, one
ought to use a Restrain QiDisperse Liver method to treat it.
Treatment principle: Restrain Qi and Disperse Liver, Nourish Heart and Pacify
Spirits.
Prescription: Unfettering Powder with Licorice-Wheat-Jujube Decoction.
37
Five
doses.
Follow-up examination: Patient returned for follow-up October 21, 1992. Five
doses of the prescription had already led to a substantial reduction in symptoms.
Little sighing, no tendency toward weeping any more, only slight feelings of
dizziness or agitation. Still experiencing insomnia and excessive dreaming: this
[indicates that] the above prescription had dispersed Liver Wood, but because
Metal was so dominant in the Eight Characters, it was still hard to calm breath-
ing. I gave the patient a detailed explanation of her Eight Characters illness syn-
drome, and after hearing it her confidence was greatly increased, more com-
pletely matching her will with her treatment right up to the point of a complete
cure. Bearing the above syndrome in mind, I prescribed the following:
Prescription: Unfettering-Licorice-10,000 Harmonies Decoction (see Miraculous
Formulae of 1000 Experts, vol. 1 p. 498).
38
Five doses.

between stems and branches on the basis of their position in the conventional tabular
arrangement that I have reproduced here.
36
Needed to fill in for the Water that was diminished by virtue of Earths activity of
checking discernible in the day characters. This appears to be a thoroughly metaphorical
attribution of cause: How would metal generate water? Only, one presumes, through
melting, which would require fire. Within the terms of normal relations of influence
among the Five Phases, there is no particular reason why Fire would have to be present
for Metal to generate Water. This is, after all, its proper nature in the generation sequence
of the Five Phases.
37
Probably consisted of hares ear root, Chinese angelica, peony, baishu, China-root,
licorice, ginger, mint, wheat kernels, and jujube.
38
Dr. Zhao cites one of many compilations of herbal formulae that have become
available since 1976 in the Peoples Republic. This formula is not listed in standard
formulary reference works.
Third examination: Patient came to the clinic on October 26. The five doses
of medicine had been used up with excellent effectiveness. She herself felt that
her illness was just about entirely cleared up, with only a little insomnia remain-
ing. Her whole body felt comfortable and her breathing was deep and easy.
39
Because the illness course had been so long for this patient, and her spirits
(jingshen , qi, and Blood had been greatly depleted, I prescribed Heart
Bolstering Elixir along with Spleen Restoring Pills to be taken for the next half
month.
40
After half a month the patient returned to the clinic to say that she was
completely cured. I reminded her of the configuration of her Eight Characters
and advised her not to get angry and to particularly attend to self-care.
41
The
patient was deeply impressed and very satisfied.
The above two cases were both of patients with long disease durations, many
previous treatment attempts, and extensive internal damage, calling for rather
complicated therapies. Drawing on experience with many patients, I knew that
these patients were both cases of nerves, with uneven Eight CharacterFive
Phases control relations; after becoming ill their recovery process was slow, and
they had little confidence in doctors. The good points of combining Changes
studies with Chinese medical theory to evaluate syndromes (lunzheng are:
1. When the illness condition is explained in detail to the patient, it elevates
his or her faith in the doctor; this forges a good foundation for long periods of
taking herbal medicine.
42
2. By doubling the evaluation of the syndrome, combining the Eight Char-
acters with Chinese medical syndrome differentiation, [the physician] can have a
more considered plan and be more certain that when he prescribes a drug formula
it will clearly be effective.

39
See Kuriyama 1994 for an extensive consideration of the centrality of breath in the
history of perception and embodiment in China. The symptom mentioned here, whole
body comfort, is often mentioned in medical contexts as a sign of a true recovery from
illness; its connection with wholesome respiration is not accidental.
40
These are Chinese patent herbal medicines.
41
This is very common advice in clinics of Chinese medicine; a great many
disorders are associated with excessive anger, and patients are told not to get angry in a
solicitous tone that nevertheless suggests they should be able to control such feelings.
42
Dr. Zhao acknowledges here that herbal medicine therapies are inconvenient and
time-consuming, especially under hectic modern conditions. Patients and their families
must manage the preparation and administration of decoctions themselves, and those
who have no faith in the ultimate efficacy of the procedure would be unlikely to take the
trouble. Dr. Zhaos emphasis on a clear explanation is not universally shared in the world
of Chinese medicine, and is seldom taught in the formal training manuals of the field. He
may have arrived at this conclusion on his own or picked it up from articles in medical or
popular periodicals.
For the last five years Ive been using Changes studies to supplement Chi-
nese medicine theory, successfully treating many people with stubborn diseases.
I trust these two cases can serve as useful examples.
Above are several points derived from my experience of using Changes
studies and Chinese medicine together, but Changes knowledge is extremely
deep and abstruse. I have a great interest in continuing to pursue these studies in
the hope of attaining precious experience in the use of combined Changes and
medical knowledge that can be more widely useful to the masses.
When my American friend Judith Farquhar came to study my clinic, I was
extremely pleased, and felt an added stimulus and impetus to continue my study
of the Changes and of medicine. At the same time I was gratified to think that
Professor Farquhar might, in studying our countrys knowledge about the
Changes and medicine, find something of great value within it.
July 13, 1993
* * *
Were I to continue the impossible task of emulating TMCC here, this would
be the end of the paper, except for voluminous appendices. But I did, in fact,
find some things of great value in our countrys medicine as Dr. Zhao
presented it. Without attempting to upstage him, then, I will note a few
anthropologically and historically interesting issues raised by his work.
The Social Life of Texts
The final paragraph of Dr. Zhaos essay returns to some of his complex modern
purposes in adopting and explaining a method of knowing and healing with very
old roots. Presenting this text to Professor Farquhar and thereby invoking a
transnational and scientific relationship, he continues certain legitimating
gestures that are visible at the outset of the essay. Recall that the problem of the
Zhouyis dense superstitious coloration and the modern stigma attached to its
study opens the essay. The regionally prestigious name of Liu Dajun, a
philosopher who presents Changes studies as a Chinese treasure which can
contribute to Chinas modernization, is then invoked as a nationalist, modernist,
and scientific response to the stigma that Maoism and other Chinese modern-
isms have attached to divination practices. After a brief explanation of Changes
techniques, the text returns to this legitimating project, this time deploying a
Maoist idiom (still the language of government argumentation in China) to
assert that the book arises from a dialectical materialist impulse, and is not a
superstitious work. The arrival of a first world professor at Dr. Zhaos small
store-front clinic, avowing an interest in his methods of curing, can be read as
confirming almost beyond doubt the great value of his style of medical
practice. A foreigner, after all, presumably outranks most Chinese people in
terms of modernity and scientific cosmopolitanism.
Of course, the two successfully treated cases of nerves he discusses are
also meant to persuade. Clinical case histories reporting successful therapeutic
strategies are perhaps the most common form of writing and publishing in the
world of Chinese medicine, appearing in every professional periodical and
almost every textbook, as well as in specialized and popular books and
compilations. Even in these conventional venues, such texts have a purpose
beyond neutral description. Their clinical details contribute to ongoing debates
on the legitimacy of Chinese medicine, their successes advance a polemic.
At one level, of course, published cases inform practitioners of new and
interesting methods they may wish to adopt. But if this were their sole purpose
one presumes more cases would report useless, dangerous, or disastrous
interventions, and sound needed warnings. Dr. Zhaos use of heart-warmingly
successful cases in this essay is typical, however, and clearly indicates that more
is at stake than practical clinical communication. Doctors of traditional medicine
know that their field is often seen as a pseudo-science, at best a form of
psychotherapy, by a science-besotted world. The routine efficacy of their work,
their clear ability to help patients who suffer from real discomforts, continues to
persuade them that what they do has a practical material value that any
reasonable reader, however scientific, should be able to recognize.
Dr. Zhao of course faces a particular problem of legitimation, in that his
emphasis on Changes medicine is less widely accepted within China than the
mainstream secular tradition of classical medicine (Zhongyi). The result is
historically ironic in the Chinese context: the more ancient practices of
divination associated (however tenuously) with the canonical Book of Changes
have been denounced as superstitious; in the 1990s they must be re-legitimated
through their relation to a modernizing and scientizing Chinese medicine. Dr.
Zhao is well aware of the decades of voluminous arguments within the world of
traditional medicine that Zhongyi has scientific value and can modernize. Hence
his decision to link the often-questioned legitimacy of Changes medicine to the
well-established value of a broader, but newer, traditional medicine.
This legitimation strategy may be partly because his honored guest Professor
Farquhar was locally known to have an interest in Zhongyi. But it may also
indicate a genuine gap between cultural and epistemological worlds. From the
viewpoint of contemporary North American intellectuals, this tacticjustifying
study of the Changes with reference to traditional Chinese medicinemight not
solve the problem, especially for those who can only believe in medical methods
based on bioscientific principles and protocols. From Dr. Zhaos viewpoint,
educated as he was in the Maoist and nationalist medical discourses of the last
40 years in China, this clinical demonstration of the intimate connection
between the Changes and the Five Phases, between the Zhouyi and traditional
Chinese medicine, and this careful presentation and explanation of his clinical
work in Changes terms, persuasively legitimate his practice and its textual
sources.
The gap between the viewpoints of North American academics and rural
Chinese doctors is real; anthropological commentary classically attempts to
bridge such gaps and, through a contextualizing ethnography, produce
understanding.
43
Much could be said about the continuing reality of yawning
epistemological chasms in todays global politics (see, e.g., Asad 1986;
Harding, with commentaries by Farquhar, Kuriyama, and Cohen, 1994). Uneven
distributions of knowledge are easy to relativize, and they do not quite parallel
(apparently more objective) global inequities in access to minimal material
resources. But they are just as thoroughly imbued with the asymmetries of
power and the institutional supports that make it difficult to imagine any rapid
redistribution.
For present purposes, it is important to note that legitimating texts, of which
the present essay is an example, are often used to reconcile conflicting
knowledges and translate local argumentative idioms into terms that enjoy wider
currency, especially those of cosmopolitan science. They often fail, as anyone
knows who has tried to explain to a scholar of Chinese medicine in East Asia
why his professional papers are unlikely to be published in the United States.
44
My own present effort cannot bridge this great divide, but by turning to a
stubborn problem of language perhaps some contours of the epistemological
differences at issue can be better visualized. One could focus, for example, on a
central challenge posed by practices deriving from the Book of Changes, that of
the ontological status of its most important words: the trigrams and hexagrams
themselves, their names, and the terms for the heavenly stems and earthly

43
See Visweswaran 1994, for elegant summaries of new critical traditions in
anthropology. Visweswaran shows how realist representation has been challenged in
twentieth-century ethnography and critical theory, particularly focusing on fictions of
empathy, rapport, and understanding in both the disciplinary tradition and in her
own ethnographic work.
44
This point is made by Shigeru Nakayama in a recent paper. To explain the
stubborn and frustrating divides between Eastern and Western historians of science,
as well as among the scholars of various Asian nations, he developed the notion of local
citation groups which for a variety of reasons remain isolated from one another
(Nakayama 1995: 9092). Thus Japanese historians of Asian medicine tend to be
unaware of historiographical innovations in North America and Europe, and the work of
Chinese scholars of Chinese medicine is framed by state institutions that link historical
research to practical medicine. North American and European scholars in their turn
seldom take advantage of large Chinese-language and Japanese-language archives of
medicine and science. These constraints are more institutional, political, and economic
than they are epistemological, but they lead to the formation of national or regional
communities of discourse within which scholars relate intellectually chiefly to each
other.
branches which function in some similar ways. These terms do not represent:
they are not words standing for things, rather they are things in themselves.
Unlike ordinary nouns, they do not refer to something non-linguistic (concepts
and objects are the usual suspects); instead they act in the world of time and
persons. Used in a divination system, such terms position us whether we like it
or not (this is, after all, the nature of fate) in cosmic processes that transcend
human devising.
This problem has troubled Western translators and commentators throughout
the course of the Book of Changes history in Euramerican China studies. Most
experts have hesitated to consider the work outside of its use in divination or
magical thinking (Eber, in Wilhelm 1979: xvi; see also Legge 1963 [1899];
Needham 1956: 30445; Smith 1991, 1994: 10128). In other words, though
there has been a tendency to see the Book of Changes as a symbolic system
which represents the (non-linguistic) world (Wilhelm 1979: xxii; Smith 1991,
1994: 10128), no serious scholar of the Changes writing in English has been
willing to treat the work primarily as a book that merely symbolizes.
Approached from the point of view of an empiricist theory of language in
which words refer to or stand for things, the Book of Changes can only be
superstitious nonsense. Needhams distaste for the numerological character of
the trigrams and hexagrams is well known, for example. The standard
anthropological solution to problems presented by irrationalities of this kind is
to simply recount native beliefs while politely refraining from pointing out
how deeply they differ from what the anthropologists readers know, according
to their lights, to be true. This has also been a classic tactic of humanistic
sinology. Comparison is thus left to the reader, who may stigmatize or
romanticize these unfamiliar beliefs as she sees fit.
Perhaps there is more to be said about divination technologies and their
associated texts once the problematic character of their particular use of signs is
noted.
45
Dr. Zhaos use of Eight Characters analysis in his medical work, for
example, suggests that by combining several technologies of enumerating and
naming he can place the sufferer in time, locating the Changes of his illness
within the Changes of the cosmos. By mapping an Eight Character birthdate
over a Five Phases understanding of physiology, he not only fills time with
significant content for the client, he rationally structures an intervention on the

45
It is tempting for good reasons to compare the Chinese Book of Changes tradition
of divination language with that of the European and West Asian astrological tradition.
In both cases popularized and technically simple versions of astrological technology
survive as horoscopic classification, the Chinese animal signs and the signs of the
Western zodiac. But previous applications were much more complex. See French 1994
for the uses of medieval astrology in medicine.
basis of that significance.
46
Perhaps this move explains Dr. Zhaos emphasis on
giving a detailed explanation to his patients: he has gone from the identifica-
tion of the appropriate abstract, contentless markers of temporal position (the
stems and branches of the clients birthdate) to a connotatively rich account of
the long-term illness process, while simultaneously indicating a pathway for
therapeutic action. This is, after all, no small achievement.
But it is arrived at by treating certain words as natural rather than cultural
entities.
47
The relationship between Mr. Hes or Ms. Yangs birth and the stems
and branches which identify that moment is not treated as arbitrary, these words
are not mere labels assigned by fallible calendrical scholars. The original
numerical inscriptions from which the trigrams and hexagrams, stems and
branches, have classically been derived were not, after all, of human devising
(See Legge 1963 [1899]: 1418 and Smith 1991: 59 for the spontaneous origins
of the River Diagram Hetu and the Luo River Writing Luoshu ).
Similarly, the affiliations between the stems-branches and the Phases go beyond
classification to at least resonance (Needham 1956: 28591); in some contexts
the closeness of the connection narrows to an apparent identity.
One could argue that the logic of the Changes, in their basic form, is strictly
metonymic; each trigram, each hexagram, and by extension each stem and
branch takes its sole significance from its position in an ordered and minimalist
system of arbitrary signs.
48
The canonical history of the Changes as text in
China has brought content to these arbitrary arrays; this has generally been
presented as an unpacking of meanings already inherent to the lines, dyads,
trigrams, and hexagrams and, especially, in their relations to each other. To read
the Changes was to read the qualities of time and space themselves (Kuriyama
1987). But perhaps the historian must acknowledge the interpretation of the
Book of Changes as a labor of addition, as an imputation of metaphoric content

46
In a longer discussion it would be possible to indicate the Five Phases affiliations
of the herbal drugs included in Dr. Zhaos prescriptions. Interested readers could consult
the Zhongyi dacidian , Formulary and Materia Medica volumes (see
Editing Committee of the Unabridged Dictionary of Chinese Medicine, 1982, 1983) for
the character (xing and flavor (wei of each drug in the formulae noted in the case
histories, correlating these with Five Phases tables of correspondence to be found in
Sivin 1987: 209, Needham 1956: 26263, or any introductory Zhongyi text published in
China in the 1980s.
47
The nature/culture dyad once widely used in anthropology (following Levi-
Strauss) is useful here for comparative purposes despite its inappropriateness in the
Chinese context itself. A more classically Chinese phrasing of the distinction might see
the hexagrams as having a (macro)cosmic origin beyond (microcosmic) human devising.
48
My use here of the literary terms metonymy and metaphor derives from Barthes
1972, Jakobson 1956: 7682, and Levi-Strauss 1978.
to the almost purely metonymic, or as a bricolage of preconstrained signs hung
onto the now meaningless pegs of ancient diagrams.
49
We can perceive a labor of this kind in Dr. Zhaos case interpretations.
Conceivably he could treat his method of correlating stems and branches with
the Five Phases as a pure exercise in positioning the patient in abstract and
spontaneous cosmic processes. But in fact he does not treat these cosmic
processes as purely abstract. He notes in Mr. Hes case that with excessive Fire
the personality is anxious and irritable, excessive Earth leads to over-eating and
drinking and static retention of fluids and dampness. How easy it becomes,
upon reading this, to translate what we have experienced as irritable and
depressive personalities into a clinical picture of excessive Fire and Earth!
A similar metaphoric quality is evident in Ms. Yangs case, where Dr. Zhao
points out that Metal was dominant, but due to the lack of Fire, which can
liquefy metal, it was unable to generate Water, hence the lack of coating on the
tongue. In these remarks the common-sense characteristics of fire, earth, metal,
and water are necessary for the logic of his diagnosis; the fire in question is not
just the Fire that labels one position in an interconnected array of five
generalized phases; it is also the fire that flickers and burns, the fire that melts
metals to produce a fluid. Working through natural connections of this kind,
Dr. Zhao ultimately produces a rich medical narrative which connects the
arbitrary character of the patients birth with the lived world of prediction,
manipulation, and causation.
Another, more purely medical, level of metaphor might be noted here,
namely the spatial relations between certain viscera and the visceral systems of
function (Sivins term) within the body. Thus in Mr. Hes case the Lung
system functions of proper clearing and descension of qi were disordered
because the Spleen system (centered at a lower and more inward position in the
body than the Lungs) had been inappropriately drained in previous therapies.
The proper ordering of light and heavy substances, the smooth downward flow
of the heavy, and the upward flight of the light, was disturbed. The metaphor in
such narratives is to everyday mechanics or hydraulics. Though one would not
wish to understand all of Chinese medical logic in these terms, one can easily
imagine experiencing the body both in health and disease as a hydraulics of
eating, breathing, and eliminating, a system in which movement and rest,
pleasures and pains, would be readily comprehensible.
50

49
On bricolage, see Levi-Strauss 1966: 1636; Comaroff 1985: 19799.
50
There are, of course, many classical methods of correlating body processes to the
trigrams and hexagrams of the Book of Changes. Needham rather scornfully mentions
some of them (1956: 30445). Interestingly, Dr. Zhao does not appear to deploy these
correlations in his own practice of divination medicine, preferring at present to stick to
the stem-branch system.
Thus it is that Dr. Zhaos Changes medicine is able to bring significance to
illness in a particularly powerful way, at least in the two cases he presents. Case
1 is especially persuasive: an unusual configuration of stems and branches
which are all Fire and Earth connects (both metonymically and metaphorically,
in Five Phases generation-and-conquest terms, and in terms of the content of
specific phases) the patients (previously arbitrary) position in time to
experienced discomforts. Once Dr. Zhaos technique has translated time (the
birth characters) into pathology (Five Phases asymmetries), it can work on the
illness with the formidable armamentarium of Chinese medicine. The classic
question of divination is here answered in practice: if fate is ineluctable, why
bother to know it? For medicine, at least, it appears that any persons fate is
simply an array of qualitatively knowable forces in a particular configuration,
and the forces at the disposal of medicine are no less or more powerful than
those which converged at the patients moment of birth. In a perfectly
conventional sense, then, knowledge (of fate) can be power (to heal).
Medical anthropology has, almost from its inception, argued that meaning
is a universal human need (Comaroff and Maguire 1981; Good 1994: 5256),
and it is tempting to view Dr. Zhaos practice as one that brings the order of
significance to otherwise meaningless distress.
51
His practice claims to
reanimate an ancient text, an undying element of Chinas national essence,
applying it to reinscribe patients lives in both history and nature. But history
and nature are not really distinct in the Book of Changes, nor is this a
psychologizing medicine that distinguishes the patients understanding from his
or her organic disorder. No one, in other words, would think of these successful
therapies as a triumph of mind over matter. Rather they are translations of the
language (wen) of the cosmos which marks its tempos in human lives. Any gaps
we may imagine between writing and the world, texts and their social contexts,
are in this practice neatly closed.
References
Anagnost, Ann. 1994. The politics of ritual displacement, in Charles F. Keyes et al.
(eds.), Asian Visions of Authority: Religion and the modern states of east and
southeast Asia. Honolulu, Univ. of Hawaii Press: 22154.

51
It is still not clear to me whether Dr. Zhaos services really are needed by his
neighbors in the county town. His practice had few regular clients when I first visited
him in 1993. By the time I returned in the summer of 1995, he had moved to better (and
more expensive) quarters fronting on a busy street. He said business was steadily
improving and he had, in the intervening two years, had other opportunities to treat
illnesses with divination methods.
Asad, Talal. 1986. The concept of cultural translation in British social anthropology, in
James Clifford and George E. Marcus (eds.), Writing Culture: The poetics and
politics of ethnography. Berkeley, Univ. of California Press: 14164.
Barthes, Roland. 1972. Myth today, in Mythologies. New York, Hill and Wang: 109
59.
Behar, Ruth. 1993. Translated Woman: Crossing the border with Esperanzas story.
Boston, Beacon Press.
Beijing College of Traditional Chinese Medicine. 1979. Zhongyi gejia xueshuo
(Scholarly schools of thought in Chinese medicine). Shanghai, Shanghai Science and
Technology Press.
Benjamin, Walter. 1968 [1955]. The task of the translator, in Hannah Arendt (ed.),
Illuminations: Walter Benjamin, essays and reflections. New York, Schocken Books:
6982.
Chiu, Martha L. 1986. Mind, body and illness in a Chinese medical tradition. Ph.D.
diss., Harvard Univ.
Comaroff, Jean. 1985. Body of Power, Spirit of Resistance: The culture and history of a
South African people. Chicago, Univ. of Chicago Press.
Comaroff, Jean, and Peter Maguire. 1981. Ambiguity and the search for meaning:
Childhood leukemia in the modern clinical context. Social Science and Medicine
15B: 11523.
Cui Xiuhan. 1983. Zhongguo yishi yixian shuyao (Scholarly essentials of Chinese
medical history and documents). Yanbian, Yanbian Peoples Press.
Editing Committee of the Dictionary of Chinese Medicine. 1979. Jianming Zhongyi
cidian (Abridged dictionary of Chinese medicine). Beijing, Peoples Health Press.
Editing Committee of the Unabridged Dictionary of Chinese Medicine. 1982. Zhongyi
dacidian, zhongyao fence (Unabridged dictionary of Chinese medicine, materia
medica volume). Beijing, Peoples Health Press.
. 1983. Zhongyi dacidian, fangjixue fence (Unabridged dictionary of Chinese
medicine, formulary volume). Beijing, Peoples Health Press.
Eoyang, Eugene. 1993. The Transparent Eye: Reflections on translation, Chinese
literature, and comparative poetics. Honolulu, Univ. of Hawaii Press.
Farquhar, Judith. 1994a. Knowing Practice: The clinical encounter of Chinese medicine.
Boulder, Colo., Westview.
. 1994b. Scientific terms: Fashions in cross-cultural truths. Paper presented at
Translation, Transcreation, Transportation: A Workshop on the Interpretation of
Chinese Culture, Curriculum in Asian Studies, Univ. of North Carolina, Chapel Hill,
Feb. 1819.
. 1994c. Political economies of knowledge: Comment on Harding. Config-
urations: A journal of literature, science, and technology 2, 2: 33135.
. 1996. Market magic: Getting rich and getting personal in medicine after Mao.
American ethnologist 23, 2: 23957.
Foucault, Michel. 1973. The Birth of the Clinic: An archaeology of medical perception.
New York, Pantheon.
. 1977. Discipline and Punish: The birth of the prison. New York, Vintage.
Fox, Renee C. 1959. Experiment Perilous: Physicians and patients facing the unknown.
Glencoe, Ill., Free Press.
French, Roger. 1994. Astrology in medical practice, in Luis Garcia-Ballester (ed.),
Medicine from Salerno to the Black Death. Cambridge, Cambridge Univ. Press.
Fu Weikang (ed.). 1990. Zhongguo yixue shi (History of Chinese medicine). Shanghai,
Shanghai College of Traditional Chinese Medicine Press.
Good, Byron. 1994. Medicine, Rationality, and Experience: An anthropological
perspective. New York, Cambridge Univ. Press.
Gordon, Deborah. 1988. Tenacious assumptions in Western Medicine, in M. Lock and
D. Gordon (eds.), Biomedicine Examined. Dordrecht, Kluwer Academic.
Hanson, Marta E. forthcoming. External chaos, internal disorder: The Chinese medical
discourse on diseases of the south. Ph.D. diss., Univ. of Pennsylvania.
. 1995. Merchants of medicine: Huizhou mercantile consciousness, morality, and
medical patronage in seventeenth century China, in Keizo Hashimoto et al. (eds.),
East Asian Science: Tradition and beyond. Osaka, Kansai Univ. Press: 20714.
Haraway, Donna. 1991. Situated knowledges: The science question in feminism and the
privilege of partial perspective, in Simians, Cyborgs, and Women: The reinvention
of nature. New York, Routledge: 183202.
Harding, Sandra. 1994. Is science multicultural? Challenges, resources, opportunities,
uncertainties. Configurations 2, 2: 30130.
Hu, Shiu-ying. 1980. An Enumeration of Chinese Materia Medica. Hong Kong, Chinese
Univ. Press.
Huang Jitang. 1982. Zhongyixue shi (History of Chinese medicine) in Selected
Materials, vol. 2. Ed. by the Dialectics of Nature Research and Training Section,
Guangzhou College of Traditional Chinese Medicine.
Jakobson, Roman. 1956. Two aspects of language and two types of aphasic
disturbances, in Roman Jakobson and Morris Halle (eds.), Fundamentals of
Language. The Hague, Mouton: 5581.
Janzen, John. 1978. The Quest for Therapy in Lower Zaire. Berkeley, Univ. of California
Press.
Jia Dedao. 1979. Zhongguo yixue shile (Summary history of Chinese medicine).
Taiyuan, Shanxi Peoples Press.
Keegan, David J. 1986. The Huang-ti Nei-ching: The structure of the compilation, the
significance of the structure. Ph.D. diss., Univ. of California, Berkeley.
Kleinman, Arthur M. 1977. Depression, somatization, and the new cross-cultural
psychiatry. Social Science and Medicine 11: 310.
. 1980. Patients and Healers in the Context of Culture: An exploration of the
borderland between anthropology, medicine, and psychiatry. Berkeley, Univ. of
California Press.
. 1986. Social Origins of Distress and Disease: Depression, neurasthenia, and
pain in modern China. New Haven, Yale Univ. Press.
Kleinman, Arthur M., and Joan Kleinman. 1985. Somatization: The interconnections in
Chinese society among culture, depressive experiences, and the meanings of pain, in
A. Kleinman and B. Good (eds.), Culture and Depression: Studies in the anthropol-
ogy and cross-cultural psychiatry of affect and disorder. Berkeley, Univ. of Califor-
nia Press: 42990.
Kong Jianmin. 1988. Zhongguo yixue shigang (Outline history of Chinese medicine).
Beijing, Peoples Health Press.
Kunstadter, Peter. 1976. Do cultural differences make any difference? Choice points in
medical systems available in Northwestern Thailand, in A. Kleinman et al. (eds.),
Medicine in Chinese Cultures: Comparative studies in health care in Chinese and
other societies. Washington, D.C., Fogarty International Center: 35184.
Kuriyama, Shigehisa. 1987. Pulse diagnosis in the Greek and Chinese traditions, in
Yosio Kawakita (ed.), History of Diagnostics. Osaka, Taniguchi Foundation: 4367.
. 1994. The imagination of winds and the development of the Chinese
conception of the body, in Angela Zito and Tani E. Barlow (eds.), Body, Subject
and Power in China. Chicago, Univ. of Chicago Press: 2341.
Latour, Bruno. 1988. The Pasteurization of France. Cambridge, Mass., Harvard Univ.
Press.
. 1993. We Have Never Been Modern. Cambridge, Mass., Harvard Univ. Press.
Legge, James (trans.). 1963 [1899]. The I Ching (The Book of Changes). New York,
Dover Publications.
Levi-Strauss, Claude. 1966. The Savage Mind. Chicago, Univ. of Chicago Press.
. 1978. Introduction, in Roman Jakobson, Six Lectures on Sound and Meaning.
Cambridge, Mass., MIT Press: xixxvi.
Liu, Lydia. 1993. Translingual practice: The discourse of individualism between China
and the West. Positions 1, 1 (spring): 16093.
Liu Yanchi (tr. Fang Tingyu and Chen Laidi). 1988. The Essential Book of Traditional
Chinese Medicine. 2 vols. New York, Columbia Univ. Press.
Lynch, Michael, and Steve Woolgar (eds.). 1990. Representation in Scientific Practice.
Cambridge, Mass, MIT Press.
Marcus, George E. and Michael Fischer. 1986. Anthropology as Cultural Critique: An
experimental moment in the human sciences. Chicago, Univ. of Chicago Press.
Nakayama Shigeru. 1995. History of East Asian science: Needs and opportunities.
Osiris (series 2) 10: 8094.
Needham, Joseph. 1956. Science and Civilisation in China, vol. II. Cambridge,
Cambridge Univ. Press.
Niranjana, Tejaswini. 1992. Siting Translation: History, post-structuralism, and the
colonial context. Berkeley, Univ. of California Press.
Porkert, Manfred. 1974. The Theoretical Foundations of Chinese Medicine: Systems of
correspondence. Cambridge, Mass., MIT Press.
. 1983. The Essentials of Chinese Diagnostics. Zurich, Chinese Medicine
Publications, Ltd.
Ren Yingqiu (ed.). 1980. Zhongyi gejia xueshuo (Scholarly schools of thought in
Chinese medicine). Shanghai, Shanghai Science and Technology Press.
Romanucci-Ross, Lola. 1969. The hierarchy of resort in curative processes: The
Admiralty Islands, Melanesia. Journal of Health and Social Behavior 10: 2019.
Si Yuanyi and Gong Chun (eds.). 1988. Yishixue (Medical history). Hubei, Hubei
Science and Technology Press.
Sivin, Nathan. 1987. Traditional Medicine in Contemporary China: A partial translation
of Revised outline of Chinese medicine (1972). Ann Arbor, Center for Chinese
Studies, Univ. of Michigan.
Skar, Lowell. 1995. Ethical aspects of Daoist healing: The case of Song and Yuan
thunder rites, in Keizo Hashimoto et al. (eds.), East Asian Science: Tradition and
beyond. Osaka, Kansai Univ. Press: 22129.
. forthcoming. Ph.D. diss., Univ. of Pennsylvania.
Smith, Richard J. 1991. Fortune-Tellers and Philosophers: Divination in traditional
Chinese society. Boulder, Colo., Westview.
. 1992. Chinese Almanacs. Hong Kong, Oxford Univ. Press.
. 1993. Divination in Ching dynasty China, in R. J. Smith and D.W.Y. Kwok
(eds.), Cosmology, Ontology, and Human Efficacy. Honolulu, Univ. of Hawaii Press:
14178.
. 1994. China's Cultural Heritage: The Qing dynasty, 16441912. 2nd ed.
Boulder, Colo., Westview.
Unschuld, Paul U. 1985. Medicine in China: A history of ideas. Berkeley, Univ. of
California Press.
. 1986. Medicine in China: A history of pharmaceutics. Berkeley, Univ. of
California Press.
Visweswaran, Kamala. 1994. Fictions of Feminist Ethnography. Minneapolis, Univ. of
Minnesota Press.
Wilhlem, Richard. 1979. Lectures on the I Ching: Constancy and change. Trans. Irene
Eber. Princeton, Princeton Univ. Press.
Wong, K. Chimin and Wu Lien-teh. 1932. History of Chinese Medicine: Being a
chronicle of medical happenings in China from ancient times to the present. Tianjin,
Tientsin Press, Ltd.
Woolgar, Steve. 1988. Science: The very idea. New York, Tavistock and Ellis Horwood.
Xu Sheng. 1992. Yuanhai ziping (Divination profundities). Hainan, Sanhuan Press.
Zhao Hongjun. 1989. Jindai Zhong Xi yi lunzheng shi (History of modern debates on
Chinese and Western medicine). Hefei, Anhui Science and Technology Press.
135
The Published Writings of Nathan Sivin:
An Annotated Bibliography
Books
01 Chinese Alchemy: Preliminary Studies. Harvard Monographs in the History of Sci-
ence, vol. 1. Cambridge, MA: Harvard University Press, 1968. Chinese translation,
Taipei: National Translation Bureau, 1973. Heuristic studies and discussions mostly
focused on developing an analytical and critical approach to the history of alchemy.
Includes a bibliography of translations of alchemical treatises.
02 Cosmos and Computation in Early Chinese Mathematical Astronomy. Leiden: E.J.
Brill, 1969. Traces the changing roles of metaphysical assumptions in shaping the
mathematical techniques of early astronomy. See 21.
03 Chinese Science: Explorations of an Ancient Tradition. MIT East Asian Science Se-
ries, vol. 2. Edited by Shigeru Nakayama and Nathan Sivin. Cambridge, MA: MIT
Press, 1972. Includes an introduction and an extensively annotated bibliography (26)
by Sivin, and two articles (24, 25) co-authored by Sivin.
04 Science and Technology in East Asia: Articles from Isis, 19131975. Selected and
edited by Nathan Sivin. New York: Science History Publications, 1977. Includes an
introduction by Sivin (which summarizes the main issues in the study of East Asian
science, technology and medicine) and a reprint of 37.
05 Astronomy in Contemporary China. A Trip Report of the American Astronomy Dele-
gation. By ten members of the delegation. CSCPRC Reports, 7. Washington, D.C.:
National Academy of Sciences, 1979. Contains several contributions by Nathan
Sivin, including a chapter on the history of astronomy.
06 Science and Civilisation in China. Vol. 5, Part 4. Chemical Discovery. By Joseph
Needham, Lu Gwei-djen, Ho Ping-yu, and Nathan Sivin. Cambridge, Eng.: At the
University Press, 1980. Includes a section (The Theoretical Background of Elixir Al-
chemy, pp. 10822) by Sivin on theories of alchemy created by Chinese practitio-
ners. See 37.
07 Chgoku no Kopernikusu (Chinas Copernicus), trans.
Nakayama Shigeru and Ushiyama Teruyo . Selected essays by
Nathan Sivin, 1. Tokyo: Shisakusha. See 27.
08 Chgoku no renkinjutsu to ijutsu (Chinese alchemy and
medicine), trans. Nakayama and Ushiyama. Idem, 2. See 01.
09 Traditional Medicine in Contemporary China. A Partial Translation of Revised Out-
line of Chinese Medicine (1972) with an Introductory Study on Change in Present-
day and Early Medicine. Science, Medicine and Technology in East Asia, vol. 2.
Ann Arbor: University of Michigan, Center for Chinese Studies, 1987. Translates
discussions of theory, and how it bears on diagnosis and therapy, from Xin bian
Zhongyixue gaiyao . Includes an extensive introduction.
10 Contemporary Atlas of China. Nathan Sivin, consulting editor. Boston: Houghton
Mifflin, 1988.
11 Science and Medicine in Twentieth-Century China: Research and Education, ed. John
Z. Bowers, William J. Hess, and Nathan Sivin. Science, Medicine, and Technology
in East Asia, vol. 3. Ann Arbor: Center for Chinese Studies, University of Michigan,
1989.
12 Science in Ancient China: Researches and Reflections. Variorum Collected Studies
Series. Aldershot, Hants: Variorum, 1995. Contains nine previously published es-
says: five reprinted (17, 21, 46, 64, 69) and four revised (26, 27, 32, and a combina-
tion of 34 and 35).
13 Medicine, Philosophy and Religion in Ancient China: Researches and Reflections.
Variorum Collected Studies Series. Aldershot, Hants: Variorum, 1995. Contains nine
essays: three reprinted (60, 61, 65), two revised (41, 56), and four new: Comparing
Greek and Chinese Philosophy and Science (previews Sivins current research, and
his collaboration with G. E. R. Lloyd) (see also 74); Emotional Counter-Therapy (on
Chinese abstract thought about disorders arising from excessive emotions); The
Myth of the Naturalists (a fresh look at the origins of the doctrines of yin-yang and
the Five Phases); Taoism and Science (sweeping and magisterial examination of the
links between Daoismin its many guisesand scientific attitudes).
Articles and Essays
14 William Lewis (17081781) as a Chemist. Chymia, 8 (1963): 6388.
15 On Chinas Opposition to Western Science during Late Ming and Early Ching. Isis,
56 (1965): 2015. Comments on an article by George H.C. Wong in Isis, 54 (1963).
16 A Simple Method for Mental Conversion of a Year Expressed in Cyclical Characters
to the Corresponding Year in the Western Calendar. Japanese Studies in the History
of Science, 4 (1966): 13234.
17 Chinese Conceptions of Time. Earlham Review (Richmond, IN: Earlham College), 1
(1966): 8292. Reprinted in 12. On the cosmological background of the early Chi-
nese sense of time.
18 A Seventh-Century Chinese Medical Case History. Bulletin of the History of Medi-
cine, 41 (1967): 26773.
19 On the Reconstruction of Chinese Alchemy. Japanese Studies in the History of Sci-
ence, 6 (1967): 6086. See 01.
20 Chinese Alchemy as a Science (Abstract). Transactions of the International Confer-
ence of Orientalists in Japan, 13 (1968): 11729. See 01.
21 Cosmos and Computation in Early Chinese Mathematical Astronomy. Toung Pao
(Leiden), 55 (1969): 173. Reprinted in 12. See 02.
22 On the Pao pu tzu nei pien and the Life of Ko Hung. Isis, 60 (1969): 38890. Com-
ments on a review by Pierre Huard and Ming Wong of James R. Wares Alchemy,
Medicine, and Religion in the China of A.D. 320.
23 Chinese Alchemy as a Science. In Nothing Concealed: Essays in Honor of Liu Y-
yn, ed. Frederic Wakeman, Jr. Taipei: CMRASC, Inc., 1970: 3550. See 01.
24 A Systematic Approach to the Mohist Optics (ca. 300 B.C.). Coauthored by A. C.
Graham. In 03 (pp. 10552). Translates and analyzes optical propositions from the
Mohist Canon .
25 Man as a Medicine: Pharmacological and Ritual Aspects of Traditional Therapy Us-
ing Drugs Derived from the Human Body. Coauthored by William C. Cooper. In 03
(pp. 20372). Through a discussion of eight substances derived from the human
body, addresses the question of the mix of pharmacological, theoretical, and magical
factors that determined the adoption and retention of drugs in the Chinese pharmaco-
poeia.
26 An Introductory Bibliography of Traditional Chinese Science: Books and Articles in
Western Languages. In 03 (pp. 279314). Revised version in 12. Emphasizes publi-
cations, mostly since the 1970s, in science (excluding medicine) in traditional and
modern China. Entries, with detailed annotations, in some cases virtually amount to
reviews.
27 Copernicus in China. Studia Copernicana (Warsaw), 6 (1973): 63122. In special
series Colloquia Copernicana, 2. Japanese trans. in 07; revised version in 12. Ex-
plores how the injunction of 1616 against teaching Copernicanism constrained the
teaching of astronomy by Jesuit missionaries.
28 The Celestial Elders Canon of the Spirit Lights. Io, 4 (1973): 23238.
29 Li Shih-chen (15181593). In Dictionary of Scientific Biography, VIII. New York:
Charles Scribners Sons, 1973: 39098. On the famous Ming herbalist.
30 Next Steps in Learning about Science from the Chinese Experience. Proceedings,
XIVth International Congress of the History of Science (Tokyo and Kyoto, 1927
August 1974), I: 1018.
31 Kagakushi kenky ni okeru saikin no keik
(Recent trends in understanding the evolution of science). Chgoku no kagaku to
bunmei , December, 1.2 (1974): 24.
32 Shen Kua (10311095). In Dictionary of Scientific Biography, vol. XII. New York:
Charles Scribners Sons, 1975: 36993. Reprinted as Shen Kua: A Preliminary As-
sessment of his Scientific Thought and Achievements, Sung Studies Newsletter, 13
(1977, publ. 1978): 33156. Revised version in 12. Examines the career of the great
eleventh-century polymath and his impact on science, culture, and politics in tradi-
tional China. Includes an annotated bibliography (virtually a bibliographic essay in
itself), which has been expanded and revised in 12.
33 Introduction and Directory of Chinese Science. Chinese Science, 1 (1975): 151.
Directory contains information (including research interests and a list of publica-
tions) on 42 persons studying Chinese science.
34 Wang Hsi-shan (16281682). In Dictionary of Scientific Biography, vol. XIV. New
York: Charles Scribners Sons, 1976: 15968. Revised version in 12. See 35. An ex-
ploration, through the career of Wang Xishan, of the Western influence on Chinese
mathematical astronomy. With bibliography.
35 Wang Hsi-shan. Coauthored by Chaoying Fang. In Dictionary of Ming Biography,
13681644, vol. II, ed. L. Carrington Goodrich. New York: Columbia University
Press, 1976: 137982. See 34. Mainly biographical, with greater attention than 34 to
Wangs life outside science.
36 loge. Giorgio de Santillana. Isis, 67 (1976): 43943. Elegy to de Santillana (1902
74), historian of science.
37 Chinese Alchemy and the Manipulation of Time. Isis, 67 (1976): 51327. Reprinted
in 04 (pp. 10822). Concise reconstruction of Chinese alchemical theories. For a
fuller treatment, see 06.
38 Chgoku dent no gireiteki iry ni tsuite
(Ritual therapy in Chinese tradition). In Dky no sgteki kenky
) (General studies of Daoism), ed. Sakai Tadao . Tokyo: Kokusho
kankkai, 1977: 97140.
39 Further Comments on the Use of Statistics in the Study of Han Dynasty Portents.
Coauthored by Hans Bielenstein. Journal of the American Oriental Society, 97
(1977): 18587. Comments on a critical note by R. R. Newton on correlations in
Bielensteins paper ...the Portents of Tsien-Han-Shu.
40 Social Relations of Curing in Traditional China: Preliminary Considerations. Nihon
ishigaku zasshi , 23 (1977): 50532. Reprinted in History of Tra-
ditional Medicine: Proceedings of the 1st and 2nd International Symposia on the
Comparative History of MedicineEast and West, ed. Teizo Ogawa. Osaka: The
Taniguchi Foundation, Division of Medical History, 1986: 2146.
41 On the Word Taoist as a Source of Perplexity: With Special Reference to the Rela-
tions of Science and Religion in Traditional China. History of Religions, 17 (1978):
30330. Revised version in 13. Unravels the unreflective and confusing use of
Daoism in writings about Chinese science and religion.
42 Current Research on the History of Science in the Peoples Republic of China. Chi-
nese Science, 3 (1978): 3958. A survey of research on the history of science under-
way in the PRC as of late 1977. Includes annotated bibliography.
43 Report on the Third International Conference on Taoist Studies. Bulletin, Society for
the Study of Chinese Religions, Fall, 7 (1979): 123.
44 Science in Chinas Past. In Science in Contemporary China, ed. Leo A. Orleans. Stan-
ford University Press, 1980 (publ. 1981): 129. Survey of quantitative and qualita-
tive sciences in premodern China and their subsequent interaction with the European
premodern sciences in the 17th century and the penetration of modern science in the
19th and 20th. See 64.
45 The Roanoke ConferenceCritical Issues in the History of Technology. Roanoke,
Virginia, 1418 August 1978. Concluding Remarks on the Conference. Technology
and Culture, 21 (1980): 62132. Remarks on why the history of technology is not a
discipline and on issues discussed at the conference.
46 Why the Scientific Revolution Did Not Take Place in ChinaOr Didnt It? The Ed-
ward H. Hume Lecture, Yale University, Chinese Science, 5 (1982): 4566. Re-
printed in Transformation and Tradition in the Sciences, ed. Everett Mendelsohn.
Cambridge University Press, 1984: 53154, and in The Environmentalist, 5.1 (1984):
3950. Chinese trans. by Jun Wenren and Yu Gang in Kexue yu
zhexue , 1 (1984): 543; better trans. in Zhongguo kejishi tansuo
(Explorations in the history of science and technology in China), ed.
Li Guohao , Zhang Mengwen and Cao Tianqin . Shanghai:
Shanghai guji chubanshe, 1986: 97114; Japanese trans. by Yano Michio
in Ty no kagaku to gijutsu. Yabchi Kiyoshi sensei sju kinen ronbunsh
(Science and skills in Asia. Fest-
schrift for Professor Yabchi Kiyoshi [Yabuuti Kiyosi]), Kyoto: Dhsha, 1982:
25280. Reprinted in 12. Reflections on the assumptions behind the question.
47 Chinesische Wissenschaft: Ein Vergielch der Anstze von Max Weber und Joseph
Needham. In Max Webers Studie ber Konfuzianismus und Taoismus. Interpretation
und Kritik, ed. Wolfgang Schluchter. Frankfurt: Suhrkamp, 1983: 34262. Revised
English version in 51. An analysis of Benjamin Nelsons efforts to invoke medieval
scholasticism to rethink both Weber (Protestants and science) and Needham
(ecumenical science) on the significance of the history of Chinese science and tech-
nology for understanding the emergence of modern science in Europe, with sugges-
tions for resolving the scientific revolution problem.
48 Dragons and Toads. The Chinese Seismoscope of A.D. 132. Coauthored by Andr
Wegener Sleeswyk. Chinese Science, 6 (1983): 119. A consideration of the me-
chanical principles that would make possible the high sensitivity claimed for the
seismoscope of Zhang Heng .
49 A Directory of Scholars in East Asia Engaged in Research on Traditional Chinese
Science. Coauthored by Shigeru Nakayama. Chinese Science, 6 (1983): 3358. In-
cludes research interests and a list of major publications of 34 scholars.
50 Reflections on Nature on Trial. In Methodology, Metaphysics and the History of
Science. In Memory of Benjamin Nelson, ed. Robert S. Cohen and Marx W. Wartof-
sky. Boston Studies in the Philosophy of Science, vol. 84. Dordrecht: D. Reidel Pub-
lishing Co., 1984: 32330. Ruminations on the thinking behind the question of why
the revolution that gave rise to modern science did not take place in China.
51 Max Weber, Joseph Needham, Benjamin Nelson: The Question of Chinese Science.
In Civilizations East and West: A Memorial Volume for Benjamin Nelson, ed. E.
Victor Walter et al. Atlantic Highlands, NJ: Humanities Press, 1985: 3749. Revised
version of 47.
52 Zhongguo kexue jishu yixue shi zhanwang (The
prospects of the history of Chinese science, technology, and medicine). Daziran tan-
suo (Explorations of nature), 4.11 (1985): 1116. Transcript of a ple-
nary lecture delivered in Chinese at the Third International Conference on the His-
tory of Chinese Science, Beijing, 21 August 1984.
53 Chinese Science, Medicine, and Technology. Proceedings, Conference on Technol-
ogy and World History, Aspen, Colorado, 1415 June 1985: 3846.
54 Traditional Chinese Medicine and the United States: Past, Present, and Future. Bulle-
tin, American Academy of Arts and Sciences, May, 39.8 (1986): 1526. Informal
survey of American interest in and knowledge of traditional Chinese medicine since
ca. the seventeenth century.
55 On the Limits of Empirical Knowledge in the Traditional Chinese Sciences. In Time,
Science and Society in China and the West: The Study of Time V, ed. J. T. Fraser et
al. Amherst: University of Massachusetts Press, 1986: 15169. See 60. Explores an-
cient Chinese concernsmanifested especially, but not exclusively, in writings on
astronomyabout whether nature can be comprehended fully by rational, empirical
investigtion.
56 The First Neo-Confucianism: An Introduction to the Canon of Supreme Mystery
(Tai hsuan ching, ca. 4 B.C.). Coauthored by Michael Nylan. In Chinese Ideas about
Nature and Society: Studies in Honour of Derk Bodde, ed. Charles le Blanc and Su-
san Blader. Hong Kong University Press, 1987: 4199. Revised version in 13. Ana-
lyzes Yang Xiongs remarkable Canon of Supreme Mystery, a work of phi-
losophy and literature: its ideas, organization, use of language and imagery, and re-
lation to the Book of Changes.
57 A Supplementary Bibliography of Traditional Chinese Science: Introductory Books
and Articles in Western Languages. Chinese Science, 7 (1986): 3341. Attempts to
provide a concise but critical guide to someone who wants to learn about the Chinese
scientific tradition without knowing the Chinese language. It is meant to aid the be-
ginner. . . . No attempt has been made to fully list mediocre publications. Supple-
ments a similar list in 03.
58 Science and Medicine in Imperial China: The State of the Field. Journal of Asian
Studies, 47.1 (1988): 4190. For a Japanese summary with commentary by
Nakayama Shigeru see Chgoku kagakushi kokusai kaigi: 1987 Kyto Symposium
hkokusho 1987 Symposium (International
Conference on the History of Science in China. 1987 Kyoto Symposium Proceed-
ings), ed. Yamada Keiji and Tanaka Tan . Kyoto: Kyto daigaku
jimbun kagaku kenkysh, 1992: 15562. Survey with bibliography of publications
on science and medicine since ca. 1980, with discussion of trends.
59 Chinese Archeoastronomy: Between Two Worlds. In World Archeoastronomy: Se-
lected Papers from the Second Oxford International Conference on Archeoastron-
omy Held at Merida, Yucatan, Mexico, 1317 January 1986, ed. Anthony R. Aveni.
Cambridge University Press, 1989: 5464. Summary of then recent research, with
bibliography.
60 On the Limits of Empirical Knowledge in Chinese and Western Science. In Rational-
ity in Question: On Eastern and Western Views of Rationality, ed. Shlomo Biderman
and Ben-Ami Scharfstein. Leiden: E. J. Brill, 1989: 16589. Revised version (minus
most of the discussion of time concepts) of 55. Reprinted in 13.
61 A Cornucopia of Reference Works for the History of Chinese Medicine. Chinese
Science, 9 (1989): 2952. Bibliographic essay with accompanying annotated bibliog-
raphy of reference works published in China and Japan since ca. 1985.
62 Alchemy: Chinese Alchemy. In Encyclopedia of Religion, 1. New York: Macmillan,
1987: 18690. Reprinted in Hidden Truths: Magic, Alchemy, and the Occult
(Religion, History, and Culture), ed. Lawrence E. Sullivan, pp. 25360. Idem. Aims
and means, history, and relations with science and religion of Chinese alchemy.
63 East Asia, Ancient. In International Encyclopedia of Communications, II. New York:
Oxford University Press, 1989: 7178.
64 Science and Medicine in Chinese History. In Heritage of China: Contemporary Per-
spectives on Chinese Civilization, ed. Paul S. Ropp. University of California Press,
1990: 16496. Revised version of 44. Reprinted in 12. Italian translation: Scienza e
medicina nella storia cinese. In LEredita della cina. Turin: Fondazione Giovanni
Agnelli, 1994: 192223. Chinese translation: Zhongguo lishi shang de kexue he
yixue . In Meiguo xuezhe lun Zhongguo wenhua
. Beijing: Zhongguo guangbo dianshi chubanshe, 1994: 165192.
General introductory essay.
65 Research on the History of Chinese Alchemy. In Alchemy Revisited: Proceedings of
the International Conference on the History of Alchemy at the University of Gronin-
gen. 1719 April 1989, ed. Z.R.W.M. von Martels. Leiden: E. J. Brill, 1990: 320.
Reprinted in 13. Argues that the study of Chinese alchemy has been hindered by the
disciplinary blinders that have led scholars looking at it from different perspectives
to ignore what they might have learned by combining viewpoints.
66 Traditional Medicine in Contemporary China and Reflections on the Situation in the
Peoples Republic of China, 1987. American Journal of Acupuncture, 18.4 (1990):
32543.
67 Saiy kagaku to Ty kagaku no hikaku
(Comparing Eastern and Western science). Kokusai kryu , 56 (1991):
2429.
68 Change and Continuity in Early Cosmology. In Chgoku kodai kagaku shiron. Zoku
(On the history of ancient Chinese science, 2). Kyoto:
Institute for Research in Humanities, 1991: 343.
69 Over the Borders: Technical History, Philosophy, and the Social Sciences. Chinese
Science, 10 (1991): 6980. Reprinted in 12. Sivins lecture at the Sixth International
Conference on the History of Chinese Science (Cambridge, Eng., 28 Aug. 1990).
On the relation of Chinese science to the history of science in general, and to anthro-
pology and sociology.
70 Huang ti nei ching. In Early Chinese Texts: A Bibliographical Guide, ed. Michael
Loewe. Early China Special Monograph Series, 2. Berkeley: Society for the Study of
Early China and Institute of East Asian Studies, University of California, 1993: 196
215.
71 Introduction to East Asian Cartography. Coauthored by Gari Ledyard. In The History
of Cartography, II, pt. 1, ed. J. B. Harley and David Woodward. Cartography in the
Traditional East Asian Societies. University of Chicago Press, 1994: 2335.
72 State, Cosmos, and Body in the Last Three Centuries B.C. Harvard Journal of Asiatic
Studies, 55.1 (1995): 537. Explains why nature, state, and body were considered a
single complex in imperial Chinese views of state, society, and the cosmos; includes
an appendix with comparisons to ancient Greek cosmology and natural thought.
73 Text and Experience in Classical Chinese Medicine. In Knowledge and the Scholarly
Medical Traditions, ed. Don G. Bates. Cambridge University Press, 1995: 177204.
On the role in Han-dynasty medicine played by the transmission of written texts.
74 Comparing Greek and Chinese Science. In East Asian Science: Tradition and Beyond,
ed. K. Hashimoto et al. Kansai University Press, 1995: 2331. Slightly revised ver-
sion of the essay Comparing Greek and Chinese Philosophy and Science, in 13.
Review Essays
75 Blofeld, John, trans. The Book of Change (London, 1965). Harvard Journal of Asiatic
Studies, 26 (1966): 29098.
76 Needham, Joseph, et al. Science and Civilisation in China, IV, pt. 2 (Cambridge,
1965). Journal of Asian Studies, 27.4 (1968): 85964.
77 Needham, Joseph, et al. Science and Civilisation in China, IV, pt. 3 (Cambridge,
1971). Toung Pao, 57 (1971): 30620.
78 Needham, Joseph, et al. op. cit. Scientific American, January 1972: 11318.
79 Teng Ssu-y and Knight Biggerstaff. An Annotated Bibliography of Selected Chinese
Reference Works (3rd ed., Cambridge, MA, 1971). Isis, 64 (1973): 53438. Not
merely a review, but also contains a supplement to the Teng and Biggerstaff vol-
ume in the form of an annotated bibliography of fifteen essential additional Chinese
reference works on traditional science and technology.
80 Elvin, Mark. The Pattern of the Chinese Past (Stanford, 1973). Harvard Journal of
Asiatic Studies, 38 (1978, publ. 1979): 44980. Evaluates Elvins use of evidence
concerning science and technology.
81 Needham, Joseph, et al. Science and Civilisation in China, V, pt. 3 (Cambridge,
1976). Harvard Journal of Asiatic Studies, 41 (1981): 21935.
82 Some Important Publications on Early Chinese Astronomy from China and Japan,
19781980. Archaeoastronomy, 4.1 (1981): 2631. Annotated, with analytic discus-
sions.
83 Unschuld, Paul U. Medicine in China (3 vols., Berkeley, 19851986). Isis, 81 (1990):
72231.
84 Science, Religion, and Boundary Maintenance. Contemporary Sociology, 20.4
(1991): 52630. Review of four books on Mertons sociology of science.
85 Reference Books for the History of Medicine (pp. 5356) and Astronomy (p. 56).
Chinese Science, 10 (1991): 5356. Review of four Chinese reference works for the
history of Chinese medicine and two for the history of Chinese astronomy.
86 Ruminations on the Tao and its Disputers. Philosophy East and West, 42.1 (1992):
2129. Memoir of A. C. Graham and comments on his Disputers of the Tao (1989).
87 Simon, Denis Fred and Merle Goldman, eds. Science and Technology in Post-Mao
China (Cambridge, MA, 1989). Minerva, 30.3 (1992): 43239.
143
Reviews
Hashimoto Keiz, Catherine Jami, and Lowell Skar (eds.), East Asian Science:
Tradition and Beyond. Osaka: Kansai University Press, 1995. xii, 568 pp.
Yuan-ling Chao
[Yuan-ling Chao is assistant professor of history at Middle Tennessee State
University, in Murfreesboro, Tennessee.]
here are two fundamental aspects to the study of the history of science in
East Asia today. On the one hand the field has undergone considerable
change in both its scope and direction of research during the past few
decades. Yet, on the other hand, it continues to confront familiar challenges
such as breaking out of a Western-dominated discourse on science and
modernization. Both of these aspects are reflected in the conference volume of
the Seventh International Conference on the History of Science in East Asia (7th
ICHSEA), convened near Kyto, 27 August 1993.
The volume is divided into two parts. Part I consists of opening addresses by
Yabuuti Kiyosi , Ho Peng Yoke , and Yamada Keiji ;
public lectures by Nathan Sivin, Pierre-tienne Will, and Jeon Sang-woon
; and four panels organized around major themes. Part II consists of five
groups of individual papers which report on current research.
In the past several decades not only has there been an overall increase in the
number of scholars engaged in the research of science and medicine in East
Asia, andas pointed out by Ho Peng Yokein the number of conferences
available to scholars, but also in the number of scholars in the humanities and
social sciences who have turned to both the research and teaching of East Asian
science. This has been reflected in a gradual shift in emphasis to cultural and
social aspects of science and medicine and the context within which they
emerged.
Another important change, discussed by Yabuuti Kiyosi in his opening
address and suggested by the name of the conference itself, has been an
expansion of the scope of the field beyond the geographic borders of China to
the entire East Asian cultural region. However, from the papers collected in this
conference volume, it is apparent that most research still focuses on China and
Japan. Over half of the fifty-four papers of the panels in the conference volume
T
itself focus on China, and approximately one-fourth deal with Japan. Korea
receives attention in the lecture by Jeon Sang-woon and in two panel papers. In
his lecture Traditional Korean Science and East Asia: Science and Technology
drawn from East Asian Experience, Jeon highlights the experience of Korean
science within the East Asian natural world but anchored in its own historical
and cultural developments. The two panel papers on Korea are Park Seong-rais
Some Indices of the Rise of Modern Science in Korea, and Muhn
Joong-yangs The Late Emergence of the Traditional Knowledge of
Water Utilisation in Choson Dynasty and its Background. There is one paper
on early Tibetan medicine: Cai Jingfeng , The sMan-tsho-ba mDo and
the Arts of Keeping Healthy in Early Tibetan Medicine, which examines the
text The Buddhas Canon Explaining the Cultivation of Vitality. There is also
a comparative study of ancient metallurgy in India and China by Mei Jianjun
and Ko Tsun . Thailands need to prepare for the twenty-first century
is the theme of a paper by Rittaporn Itti. Obviously there is still room for the
development of research outside of China and Japan.
By looking at a more broadly conceived concept of modernity, Pierre-
tienne Will in his Modernisation Less Science? Some Reflections on China
and Japan Before Westernisation strives to reveal the fallacy of equating
Westernization and modernization, a fallacy that still dominates much of
contemporary scholarship. Will proposes a new way of defining modernization
which emphasize structural changes not necessarily resulting in the same end
product as in the West, and which do not necessarily involve the development of
science. In his examination two major periods in Chinathe Five Dynasties
(907960) and the Song (9601279), and the late Ming (ca. 15501644)
and the Qing (16441911)and the Tokugawa period in Japan, Will finds
that modernization as he defines it did indeed occur in these two countries prior
to the mid-nineteenth century.
Earlier studies of the history of science in East Asia were overwhelmingly
consumed with the famous Needham question: why did China initially surpass
and then later fall behind Europe in science and technology? It is largely
through the efforts of scholars such as Nathan Sivin that comparative studies of
China and the West have been able to break out of the confines of this
formulation. As noted by Yamada Keiji in his opening address, comparative
studies remain an important aspect of the study of East Asian science. However,
Sivin points out that many comparative studies have only served to produce an
accumulation of facts and dates without adding to our understanding of science
as a human endeavor. In his Comparing Greek and Chinese Science, Sivin
exhorts us to compare processes rather than isolated concepts. Comparing the
latter, in his view, is futile. Thus Sivins current, collaborative study of Greek
and Chinese science compares not concepts per se but rather the evolutionary
activity of natural philosophy and science in China and in the Hellenic and
Hellenistic world. Sivin and his collaborator, G.E.R. Lloyd, thus weave together
Reviews 145
the ideas, the use of these ideas, and the processes that produced them in a richly
textured fabric.
This theme of comparison is again taken up by the five papers in the first
panel, Comparisons and Exchanges between East Asian and Western Science.
However, for the authors of these papersall of which employ the method of
textual analysisthe West means the ancient Greeks, the Indian and Arabic
worlds, Renaissance Europe, and seventeenth-century Europe. As pointed out
by Catherine Jami in her introduction to the panel, the concept of Western
science was constructed only relatively recently.
Dealing with the problem of the idea of modernization in East Asia, the
second panel, under the theme Science, Technology, and Modernisation,
explored the issue of modernization and the usefulness of this concept in
studying the development of science and technology in East Asia. Yoshida
Tadashi in his paper Traditional vs. Modern in the Japanese
Context argues the need to break out of the confines of the dichotomy between
tradition and modern. Morris F. Low in his Beyond Modernisation:
Towards a Post-Modern History of East Asian Science and Technology
attempts to deconstruct the traditional accounts of scientific development in
Japan and instead looks at dissenting sciences which do not readily seem to
meet the traditionally accepted standards of Western science. Low believes that
only by this sort of deconstruction can we transcend the confines of ideas of
progress and modernization, and use diverse approaches in studying the
production of knowledge, techniques, and machines.
The theme of the third panel, The Future of Technology in East Asia,
corresponds to the intention of the conference itself, as reflected in the
conference volume title, not to be limited to discussions of past and present
developments in East Asian science, but also to explore how East Asian science
contributes to the future of science and technology. In his Proposal of Problems
and Introduction of Speakers, Nakaoka Tetsur reflects upon the
change in thinking in which traditional patterns once considered to have
inhibited economic and scientific growth are now seen as crucial factors for the
impressive development of East Asia. For example, Confucianism, long seen as
a hindrance to the development of a modern economy, is now viewed by many
as the basis of the spectacular development of capitalism in East Asian
countries. And capitalism, so this line of thinking goes, is itself a prerequisite for
the development of science and technology. The papers in this panel focus on
the future of science and technology in East Asia and how science and
technology of the future may relate to scientific and technological traditions. For
example, in the paper Chinese Characters, Language, and Computers, Yokoi
Toshio examines computer technology and the processing of
language, and discusses the search for word processing solutions for cultures
where Chinese characters are widely used.
The papers in the fourth panel, Ethical Aspects of Chinese Medicine,
provide some thought-provoking discussions on ethics in traditional Chinese
medicine. Concern for ethics in traditional Chinese medicine is not a new
phenomenon and indeed has been a major concern of physicians and of
historians of medicine alike. However, in order to understand the ethical aspect
of Chinese medicine, one must first recognize the plurality of Chinese medicine.
At one end of the spectrum was elite medicine with a coherent intellectual
tradition; at the other end were religious practices of healing. In between were a
wide variety of conceptions of healing and practices. These healing practices,
which differed in their social and intellectual contexts, show diversity not only
in their medical content but also in their ethics. The panel opens with a paper by
Murakami Yoshimi on how death was perceived in the Huangdi
neijing (Inner Canon of the Yellow Lord). This is followed by a
paper by Chang Chia-feng which examines the deep concern of the
early Manchu imperial family over smallpox and their efforts at preventing the
disease and the moral significance of those efforts. The most prominent strategy
of the Manchus was the establishment of the bidousuo (sites for
keeping smallpox at bay) which were used to segregate populations. This
segregation, however, was the reverse of quarantine methods used in the West.
Instead of isolating those who were contagious, the bidousuo sought to create a
shelter for the imperial family during outbreaks of smallpox. Thus the bidousuo,
unlike asylums in the West, were conceived as safe havens for those threatened
by the disease.
Marta Hanson looks at the flourishing of a new genre of medical works in
Huizhou in the mid-sixteenth and seventeenth centuries and examines their
origins. The surge in this new genre corresponded to social and economic
changes in the region which witnessed a flourishing economy, rising literacy,
and a growing population. As scholars found it more difficult to pass the
examinations and receive official appointments, many turned to medicine as an
alternative career because of its profession of benevolence. In looking at the
prefaces of medical texts that were published in this period, Hanson finds a
range of motives for becoming physicians often couched in ethical terms.
Frdric Obringer points out that traditional historical documentation of the
deaths of emperors and officials often contained cloaked moral and political
criticisms. Lowell Skar examines the opposite end of the spectrum of medicine
and looks at ethical aspects of Daoist healing. He focuses especially on the
Thunder Rites (leifa ), one of the most popular therapies in the Song and
Yuan period. In this version of popular medicine, the practitioner healed the
biological body through the rectification of moral and social ills.
Part II of the volume consists of five groups of papers. The first group looks
at how non-textual materials such as Japanese swords, rock paintings, stone
circles in the Jmon period (ca. 10,000300 B.C.), multi-storeyed timberwork
Reviews 147
towers, and other archaeological findings have enhanced our understanding of
science and technology in East Asia.
The second group of papers returns to the issue of traditional medicine by
examining the vitality and dynamism of medical traditions and their interaction
with influences from outside. Cai Jingfengs paper, a textual study of Tibetan
medicine and the cultivation of longevity, reveals the dynamic interaction of
three medical traditions: Indian, Chinese, and indigenous Tibetan. Christopher
Cullen critically assesses the discourses of two American missionary physicians,
Peter Parker and Edward Hume, who traveled to China in 1834 and 1905
respectively. His paper testifies to the plurality of healing practices in China and
points out that discourses on medicine are intimately related to discourses on
culture, religion, politics, and science.
The papers in the third group all center on the theme of employing modern
and quantitative analyses to study traditional disciplines. The use of quantitative
methods not only allows for a careful analysis of traditional texts, but also
serves as a bridge between the ancient and the contemporary.
The fourth group of papers deals with the theme Man, Numbers, and the
Cosmos: Conceptual and Political Approaches. The papers in this group use
textual analysis of ancient texts with the aim of illuminating the larger historical
context in which they are situated. For example, Sat Masayuki in
his paper on the idea of chronology in East Asia looks at the functioning of the
combined system of era names and the sexagesimal cycle within specific
political, theoretical, and sociocultural contexts. He shows that era names
symbolized both independence and sovereignty in East Asia and were
sometimes used as a political vehicle for declaring independence from the
cultural hegemony of China.
The final group of papers looks at the triangular relationship between China,
Japan, and the West from the mid-sixteenth century onward. The papers not
only explore the interaction among the three, but also the process of change
which ultimately altered the accepted relationship.
Most of the papers in this conference volume are quite short, ranging as they
do from two or three pages to twenty-one pages. These papers are thus meant as
introductions to or synopses of current research and developments in the field of
science, technology, and medicine in East Asia. As such, the conference volume
serves two purposes: it delineates the present lines of research undertaken by
scholars in different countries, and it provides thought-provoking questions for
future research.
The study of East Asian science, technology, and medicine, however,
continues to receive little attention from mainstream societies and institutions.
Many important journals still publish relatively few articles on non-Western
traditions. There remains the persistent problem of equating modernization and
Westernization and of imagining a strict dichotomy between traditional and
modern. Only by breaking out of this framework and examining change
within social, political, economic, and cultural contexts can one understand
scientific and medical developments in East Asia.
The editors of the conference volume have provided an errata but the work
is still marred by many typographical errors and solecisms which escaped their
notice. These errors, however, do not diminish the overall quality and
contribution of the volume. That contribution might be summarized as a
reiteration that the production of science, technology, and medicine in East Asia
is closely intertwined with the processes of cultural, social, economic, and
political production.

13 Ganz

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

13 Ganz

Uploaded by

Copyright:

Available Formats

1

You might also like