² See pp. 60–1.

³ See p. 60.

⁴ Pope-hennessy, , Dame Una, Ealy Chinese jades, London, 1923, 13Google ScholarNott, S. C., Chinese jade throughout the ages, London, 1936 27–30 (repr. Rutalnd, Vermont, 1962)Google Scholar

⁵ See pp. 61 ff.

⁶ Needham, J., Science and civilisation in China, III Cambridge, 1959. 332–9.Google Scholar

⁷ Salmony, A., Archaic Chinese jades, Chicago, 1952, 69–70, reports Michel's theory but rejects his proposed use of the jade-form ts'ung as a sighting-tube combined with the discs; on Salmony's view of the discs as cord-fixers on silk bales see below, n. 66Google Scholar. Willetts, W., foundations of Chinese art, London, 1965, 57–9, receives Michel's view of the discs very favourably, although he too expresses doubts about the ts'ungGoogle Scholar. Hansford, S. H., Chinese carved jades, London, 1968, 65–70 considers that Michel's theory is 'by far the most plausible explanation so far propounded'; although he feels that ts'ung were not necessarily used with the discs, he concludes that several may well have been sighting tubes used by astronomers, perhaps even fitted with lenses as telescopesGoogle Scholar, Dohrenwend, D., Chinese jades, Ontario, 1971, 17 and 54 does not give an opinion on Michel's therory but still names the discs hsϋan chiGoogle Scholar. Loehr, M., Ancient Chinese jades, Harvard, 1975, 10–11. is highly sceptical; despite this he refers to the discs as hsϋn chi (but hsϋan chi in his fig. 232: on hsϋn/hsϋan, see n. 11)Google Scholar. Rawson, J. and Ayres, J., Chineses jade, London, 1975, 33, state that the view of discs as having an astronomical purpose ‘can be traced to a misunderstanding of a classical text’. They still label the Birtish Museum specimen cited by them as hsϋan chi. although this identification is precisely the root of the misunderstanding which led to the discs being taken as stronomical instruments in the first place. Two recently excavated specimens of three-lobed discs (see Part II and pl. VIII(a), (b)are labelled hsϋan chi in K'ao-ku, 1977, 3, pl. 4, 8 and 9, and an excavated toothed dise in the Shensi provincial museum is also so labelled (1980, private communication). The theories of Kuo Pao-chϋn, Na Chih-līang and Hayashi MInao on these discs are dicussed in Part II. p. 74 and nn. 62, 66, 92.Google Scholar

⁸ Watson, W., Ancient Chinese bronzes, London, 1977, 32.Google Scholar

⁹ Alone and in compunds, this word appears in many pre-Ch'in texts as a reference to an object which is clearly precious of the Chou dynasty (end of second millennium B.C.) when it is described as being used in ancestral invocation (Shang shu , 7, 7b. SPTK) and at the funeral of a ruler (ibid., 11, 9a); the texts may be serveral centruies later. The Tso chuan (c. fifth century B.C.) has several stories concerning pi, and the chou li (5, 37a, SPTK;? third century B.C.) says it is one of the jades used in laying out a corpse. It is not until the Shuo wen (la, 4a, SPTK; C. A. D. 100) that we are explicitly told that the pi is circular, although the Erh ya (2, 4b, SPTK; ? third century B.C.) links it with other discoid forms. However, the continuous use of discus of jade in burials form Neolithic times onwards (see Part II) up to and beyond the time of these later sources makes ti probable that they preserve names given to other forms are often somewhat questionable.

¹⁰ Ch'ϋ Wan-li , Shang shu shih i , Taipei, 1954, 2–4 lists agruments for this.

¹¹ l. 6b. SPTK. See also Legge, J., The chinese Classics, Hong Kong, 1865, III, 33,Google ScholarKarlgen, B., ‘ Glosses on the Book of Documents’ (Bulletin of the Museum of far Eastern Antiquities, 20, 1948, 77–9).Google Scholar The transliteration given here follows modern Peking pronunciation, but for the present purpose it is essential to consider the archaic readings. On Karlgren's widely accepted scheme of reconstructed phonology ('Grammata Serica Recensa; BMFEA. 29. 1957, 1–332) the text would originally have been read *dzəg dzīwan kīər ngīuk 'ăng ziəg dz'iər ts'ieng. Karlgren prefers the reading *dzewən/hsϋn to *dziwan/hsϋan for , and is followed by Hansford and others in this. With due respect for his authority, I suggest that he places insufficeint weight on the W. Han use of variants with the latter reading. which he dismisses as caused by reliance on erroneous interpretations; he also discounts the reading in his usual Sui and T'ang sources. In support of the former reading he states that 'in Han times the w. was written with the phonetic ziuĕn/sϋn. see Shuowen'. I can find no support for this in the Shuo wen under the entires for under the entries for (la, 4a, SPTEK); but see n. 33. There is also pre-Ch'in support for the more usual reading (n. 18). If Karlgren's view is adopted it is much less easy to make sense of the passage.

¹² Legge's translation simply follows the E. Han theory that Shun used an armillary sphere (see below); Karlgren chooses his interpretation of ch'i cheng as ‘sun, moon and five planets’ from the Han commentators.

¹³ The basic meaning of tsai is one of location, as in the example from the Canon of Yao ‘ he was on the throne for thirty years', l, 12a, SPTK; elsewhere in the Canon we find ‘regulate and attend to the transformation in the north ’, l. 2b. The present context suggests the more abstract second sencse as appropriate. The (?thired century B.C.) glossary Erh ya supports this interpretation by giving ch'a ‘examine’ as a possible meaning of tsai

¹⁴ In another chapter of the Book of Documents (Hung fan ‘The Great Plan’) roughly contemporary with the Canon, We find the pa cheng ‘eight (concerns of) government’ listed as food, wealth, sacrifices, works, population. crime, embassies and the army, 7, 2b, SPTK. An interesting parallel to the usage of the canon is found in the (possibly pre-Ch'in) Wang chih chapter of the Li chi , where one official is charged with the function ch'pa cheng ℈regulating the eight (concerns of ) government’, 4, 9b, SPTK; the list given (4, 17a) differs form the Hung fan but includes simila pracitcal matters. Could the ancient character for ‘eight’)have been corrupted to ‘seven’ + in the canon? The list of the chi' cheng given by the earliest interpreter of the Canon, Fu Sheng (. 200 B.C. see below is ‘spring, autumn, winter, summer the signs of heaven, the pattern of earth and the way of man’, quoted in Shih chi , 1, 24 and 27, 1292 (both in T'ang comm.), Peking, 1962. This is less directly concerned with state affairs than the pa cheng, but still represents the overall cosmic concerns of the ruler. A further ‘cosmic order’ scheme is given by Ssu-ma Ch'īen (writin c. 90 B.C.) who lists ‘heaven, earth, the twenty-eight divisions of the sky, the ten mothers and twelve sons (i.e the tow sets of cyclical singns), the (standard) bells and pitchpipes’, Shih chi, 25, 1253. Karlgren quotes the defective version of this list in 25, 1243; his interpretation of the Ch'i cheng as ‘the seven directors (i.e sun, moon and planets)’ (1948,79) discounts the early glosses and the paralled of the pa cheng. As support he offers a debatable rendering of a passage form the shih chi (n. 30), a probably spurious fragment ascribed to Fu Sheng (n. 32) and the a priori assumption that the ch'i cheng must be connected with calendrical astronomy.

¹⁵ ef. the pharse y؛ shih in the Hung fan chapter. 7. 4b, SPTK, meaning ‘precious (i.e. royal) food’.

¹⁶ In the phrase tu liang heng ‘the (standrad measures of) lenght, capacity and weight’, 1, 6b, SPTK.

¹⁷ Chou li, 2, 36a, SPTK. A cognate term is heng/*g'ang , referring to the upper crossmember of a set of girdle pendants, Shuo wen, la, 4b, SPTK; see the diagram of an excavated set of these in Hansford (1968, 104).

¹⁸ The Shuo wen (la, 4a) defines as ‘a fine jade’ and quotes the phrase ‘a headpiece of hsϋan and tassels of jade’ form the (? fifth century B.C.) Tso chuan.. The current text characters were similar enough to be easily confused. A homophone for found as a variant is (see n. 19). which the Shuo wen defines as ‘a variant for ’. This apparent interchangeability of and is further confirmed by the following gem-names in texts of Warreing States date or earlier: Shan hai ching , 5, 9n, SPTK; Mu t'ien tzu chuan , 16b, SPTK: Shih ching , ode 134/ For the shup wem has ‘a pearl that is not roung’, la, 6a. The gem-name in two texts c. 300 B.C. vouches for this connexion: Ch'u ts'u 13, 20a, SPTK, and I Chou sh , 7, 3a, in Han-wei ts'ung-shu (HWTS) alone occurs in the Yu kung chapter of the Book of Documents in the context of a tirbute-list, 3, 5a. The term in a list lf royal treasure (Mu t'ien tzu chuan, 3b, HWTS) is suggestively close to .

¹⁹ Thus om Shih chi, 28, 1355, and in Fu Sheng's Shang shu ta chuan , ap. T'ai p'ing yϋ lan , 29, 3a, SPTK.

²⁰ Shih chi, 25, 1253 and 27, 1291 have .

²¹ Karlgren's view of these variants (1948, 78) is that they were introduced in Han times as the result of a quite baseless theory that meant something rotating. He does not makeit at all clear why anyone should have made such a conjecture if was, as he claims, read hsϋn/*dziwən rather than hsϋan/*dziwan.

²² See Fu sheng's commentary, P. 57 below. Han shu , 21a. 969, peking, 1962, has , combining both variants of interest. Also Cheng Hsuan in Shih chi, 27, 1292 (comm.).

²³ See the quotation from Han text of the Tso chuan, n. 18.

²⁴ Shih chi, 121, 3124. When classical studies began to be encouraged under Emperor Wen (179–155 B.C.) Fu Sheng was recognized as an expert on the Book of Documents.

²⁵ T'ai p'ing yϋ lan, 29, 3a, SPTK.

²⁶ The use of chi with the ‘wood’ radical here rather than with ‘jade’ as in the opening quotation suggests that one of them may be the result of a copyist's error.

²⁷ Fu Sheng's use of these two words is highly reminiscent of the (? third century B.C.) appendix to the Book of Change, I ching which has 8, 5b, SPTK, 'chi means the subtlety of movement, the first visible signs of fortune '. As his next sentence makes plain, for Fu the word chi/*kiər (however written) refers to the infinitesimal movements with which new events begin; he obviously feels that Shun was involved in an act of divination at the beginning of his reign.

²⁸ 'At the still point of the turning world. Neither flesh nor fleshless; / Neither from nor towards; at the still point, there the dance is, / But neither arrest nor movement. And do not call it fixity, / Where past and future are gathered…', Burnt Norton, Collected poems, ondon, 1963, 191. In his comments on the Book of Change, Wang Pi (A.D. 226–249) took a similar view of the hsϋan chi to Fu Sheng: 'If we look at things from their roots, although their principles may be broad the intellect can cover them by a single name. Thus, if one places oneself at the hsϋan chi ij₣ to observe the great rotation, the movements of heaven and earth will not cause any surprise', Chou i lueh lieh 2b, SPPY.

²⁹ See Harper, D. J. ‘The Han cosmic board’, Early China, 4, 1979, 1–10.CrossRefGoogle Scholar.

³⁰ According to Ssu-ma Ch'ien ‘The seven stars of the Northern Dipper (a to Ursae majoris) are what (the Canon of Yao) calls “the hsϋan chi yϋ heng for setting in order the seven (concerns of) government”', Shih chi, 27, 1291. This does not (pace Karlgren) involve any suggestion that Ssu-ma used the words of the Canon as names for these stars, which he does not; he is simply noting (after Fu Sheng ?) an important ancient reference to the Dipper. Near the end of the first century B.C. Liu Hsiang wrote that hsϋan chi referred to the ‘angular array’ kou ch'en ( to Ursae minoris) and the ‘pivot star’ shu hsing Σ1694 Cam.), Shuo yuan , 18, lb, SPTK. In the Chou pi (first century B.C.) the hsϋan chi is similarly a circumpolar star; on this point see NŊda Ch؛ryŊ ShŪhi sankei nokenky؛ , Kyoto, 1933, 65 ff. The apocryphal books wei shu of the early first century A.D. begin to use hsϋan, chi, and yϋ heng as actual names for the stars of the Northern Dipper; see the examples quoted in K'ai yϋan chan ching chapter 67.

³¹ See Yabuuchi Kiyoshi . Ch؛goku no temmonrekihŊ , Tokyo. 1969, 64–70, also the discussion in Needham (1959,284–390).

³² Shih chi, 27, 1292 (commentary). Now that the hsϋan chi yϋ heng has become a terrestrial object. Ma makes the stars of the Northern Dipper into the ch'i cheng; why these should have needed ‘setting in order’ is unclear. His pupil Cheng Hsϋan suggested that the ch'i cheng were the sun, moon and five planets, presumably because these are obvious candidates for observation with an armillary sphere, Shih chi, 1, 24 (comm.). This was the view adopted by Karlgren. There is, however, no evidence of this theory being current before the E. Han; a fragment of Fu Sheng's commentary mentioned by Karlgren, in which he adopts Cheng Hsiian's theory, is clearly spurious. Its first and only appearance is in a note in the thirteenth century A.D. encyclopaedia Yϋhai , 2, 50a (repr. Taipei, 1964), which is decisively outweighed by the two T'ang quotations of the version given above (n. 14), also noted by the Yϋ hai. In his rendering of as ‘the sϋn-stone apparatus’ karlgren essentially follows Ma and Cheng, but x012B;s unable to give us any notion of what sort of apparatus might have been involved or why it should have been made of stone.

³³ Ku yϋ t'u k'ao , 1889, 50a and b. On 52a Wu illustrates a similar but less regular object, which he suggests may be identical with the I yϋ ‘Jade of the I(tribes)’ mentioned as part oi a funerary display in tne Book of Documents 11, 9a, SPTK He notes, however, that it resembles the specimen he designates haϋan chi. Under its entry for hsϋn the Shuo wen says this is another name for the I yϋ (la, 4a). Could an association along thes lines explain Karlgren's statement (n. 11) that was written under the Han?

³⁴ ibid., 51b.

³⁵ Jade: a study in Chinese archaeology and religion, Field Museum of Natural Histor: Chicago, 1912, 104.

³⁶ Loehr (1975, 11) claims that Laufer disavowed his earlier views in Archaic jades, New York, 1927, 28 f. However, in the passage cited Laufer simply revises his description of the linear markings on the discs (see n. 67) and does not discuss their purpose.

³⁷ For a list of these see Needham (1959, 777), to which should be added 'Le plus ancien instrument d'astronomie: le Pi', Ciel et Terre, LXXV, 5–6, 1959. In this Michel reverses his previous archaeologically impossible position that the pi is a degenerate form of the three-lobed discs, and claims with equal lack of justification that the pi is intended for astronomical purposes.

³⁸ 'Les jades astronomiques chinois', Communications de l'academic de Marine (Brussels), IV, 1949, 111 ff. I have translated Michel's French. A convenient summary of Miche's position is given in Needham (1959, 333–9).

³⁹ One of the most obvious faults of Michel's theory is the pointlessness of using a costly and elaborate device for this purpose. Logically, since one has to know the location of the pole to construct the apparatus properly the resulting device yields no new information at all.

⁴⁰ It is surprising that Hansford (1968. 67) continues ot entertain Michel's theory after stating this point explicitly.

⁴¹ Michel's principal sources are the scholarly and accurate study of Maspero, H., 'Les instruments astronomiques des Chinois au temps des Han', Mélanges Chinois et Bouddhiques, 6, 1939, 183 ff., and a translation of the Chou pi suan ching by Biot, E., Journal Asiatique, 1841, 3me ser., 11. 593 ff. He treats neither with particular respect. Thus, in support of his identification of the yϋ heng and the ts'ung he quotes (via Maspero) from a description of an armillary sphere by Ts'ai Yung , c. A.D. 180, Shih chi, 1, 24 (comm.). Following the fashion introduced by Ma Jung (p. 59), Ts'ai refers to its sighting tube as a yϋ heng; Michel quotes the statement that its bore is one inch in diameter, but suppresses the inconvenient statement that it is eight feet long, rather large for a ts'ung. Again in Michel (1949) there is a quotation from the Chou pi (via Biot), in which observations of a circumpolar star (called the hsϋan chi: n. 30) are made using a gnomon and sighting string, 2, 2b-3b, SPPY. Michel omits passages and reorders others to suppress the gnomon and string, and make it seem that the hsϋlan chi is an instrument. He gives his readers no warning of this apart from a few dots between passages; the mutilated text is adopted in Needham (1959, 337).

⁴² See Neugebauer, P. V.,Sterntafeln von 4000 vor Chr. bis zur Gegenwart, Leipzig, 1912;Google Scholar a simple polar plot is accurate for the area of the celestial sphere involved. Even supposing the fit to be exact for a given specimen, extant trilobate discs are far too variable for this to be more than a chance effect.

⁴³ On the subject of the linear markings found on one face of Wu Ta-chx0027;engx0027;s disc, see the discussion in n. 67. Miche's interpretation of these involves the precession of the equinoxes, a feet of which the Chinese were unaware until the fourth century A.D., see Yabuuchi (1969, 81) and Needham (1959, 356).

⁴⁴ See Harper (1979) and Loewe, M. A. N., Ways to paradise, London, 1979, 75 ff.,Google Scholar which connects the shih with the well-known TLV mirrors. The arguments sketched here have been set out at length in Cullen, C., ‘Some further points on the shih', Early China, 6. 1981, 31–46. See also Harper, D. J., ' The Han cosmic board: a response to Christopher Cullen', ibid., 47–56, and further discussion pending in Early Chirna, 7. A similar hypothesis has also been advanced in the collective work Chung kuo t'ien wen hsϋeh shih , Peking, 1981. 185 f. I owe the suggestion of following up the possible connexions of the hsϋan chi and the shih to a conversation with Dr. Loewe in early 1977.

⁴⁵ K'ao-ku, 1978, 5, p. 340. Modern forms of characters have been used for clarity.

⁴⁶ Teh shih is a model of the Kait'ien ‘umbrella-(like) heaven’ cosmography current form Warring States times on wards. The clearest account of this at present availbale is Nakayama Shigeru, A history of Japanese astronomy, Cambridge, Mass, 1969, 24–35.

⁴⁷ See for instance, the design on the lid of a lacquer funerary chest c. 433 B.c., recently excavated: Wen-wu, 1979, 7, p. 40. On this object the year of the deceased's death is marked in a manner highly reminiscent of th Shih.

⁴⁸ See shih chi, 128, 3229; the date given corresponds to 532 B.C. Loewe (1979,77) discusses this story.

⁴⁹ See n. 27.

⁵⁰ Perhaps the most likely possibility is that the yϋ heng is the crosswise representation of the Northern Dipper on the heaven-disc of the shih; the rotation of this constellation round the celestial pole played an important role in early Chinese astronomical thinking. Remembering the evidence that hsϋan chi was a name for the pole star, it is interesting that there are signs that yϋ heng was also a name for the Dipper itself (see reference of n. 44).

⁵¹ See Part I, p. 60.

⁵² Laufer (1912. flgs. 36 and 37): Nott (1936. pl. XVIII): Michel H., ‘Les jades astronomiques chinois: une hypothèse sur leur usage’ Bulletin des Musées Royaux d'art ed d'Histoire 1947. also reproduced in Needham (1959. fig. 151): Jenys, Soame, Chinese archuic jades in the British Museum, British Museum, London, 1951, pl. XGoogle Scholar (see pl. VII(a) here); Willetts 1965, black-and-white, pl. 37 (VI(b) here); Hansford 1968. black-and-white, pl. 15s (VI(a)); Dohrenwend 1971, 54(VII(b)); Rawson and Ayres (1975, cat. no. 42). Not all these specimens have the linear markings that appear on Wu Ta-ch‘eng’ specimen (see pl. I(b) and n. 67). Other identieal specimens not la belled hsüan chi: a specimen in the collection of the Kokugawa Kobunka Kenkyūsho , Ashiya , illustrated in Hayashi Minao . Chūgoku kodai no saigyoku zuigyoku (Ceremonia jade in Ancient China). Tōhō Gakuhō, 40, March 1969, pl. 24. 1; specimen in the collection of the Metropolitan Museum of Art, New York, illustrated in Na Chih-liang (pl. VIII(c) here), Yü-ch'i t'ung-shih , Taipei (2nd ed.). I, 1974, pl. 99; Lochr (1975, black-and-white, pl. 106(pl. VII(c) here)).

⁵³ Trilobate dises lacking flanges: Cheng, Te-k'un, Archaeology in China, III, Chou China, Cambridge, 1963, 186Google Scholar, refers to the specimen illustrated in Hsün-hsien hsin-ts'un ku-ts'an-mu chih ch'ing-li, , K'ao-ku hsüeh-pao (KKHP), 1936, 1, pl. 9, 5; Loehr (1975, black-and-white, pl. 232); Shan0hsi shen-mu-hsien shih-mao lung-shan wen-hua i-chih tiao-ch' , Kao-ku (KK), 1977, 3, pls. 4, 8 and 9 (pl. VIII(b) and (a) here). dise with four flanges: Jenys (1951, pl. XI). Other forms: salmony (1952, pl. 31, 2); Na Chih-liang, Chih-liang, Ch'i-pi , Ku-kung chi-k'an , I, 4, April 1967, pl. 7: Pieces from the Yamato Bunkakan Collection, Illustrated Catalogue Series, 5, 1976, pl. 137; (2) dises with four flanges and variant flange forms: na Chih-liang (1967. pl. 6); Loehr (1975, pls. 105 and 107); (3) disc with three flanges and three sets of triple notches: Salmony (1952, pl. 31, 1).

⁵⁴ Ho-mu-tu i-chih ti-i-ch'i fa-chüeh pao-kao , KKHP, 1978, 1. pp.51–4. Radiocarbon tests of acorn and wood fragments from the fourth layer have given the dates 6727 ± 140 B.P. and 6960 ± 100 B.P. (ibid., 93).

⁵⁵ This is referred to in the text as ying-shih , a term used to describe a stone with a jade-like appearance.

⁵⁶ KKHP (ibid., 53, T17 (4): 73).

⁵⁷ Willetts puts forward the idea that the Neolithic prototype of the jade disc pi was a practical implement, possibly the fly-wheel of a drill or a mace-head, on the basis that many other ritual jades are clearly derived from practical prototypes (ibid., 56–7). It by no means follows that all ritual objects had their origin in practical implements. In the case of the toroids from Ho-mu-tu, both their size and the fact that they were made of jade makes them unlikely to have been eīther fiy-wheels or mace-heads. It would, of course, be dangerous to assume that the Ho-mu-tu toroids were directly ancestral to the rings from Yangshao and Lungshan sites. Apart from the great gap of time involved, there is evidence that jade rings may have been imported into China form Baikalia around the second millennium B.C.(Watson, W., Cultural frontiers in ancient East Asia, Edinburgh, 1971, 36).

⁵⁸ Min-peï chien-ou ho chien-yang hsin-shih-ch'i shih-tai i-chih tiao-ch'a , KKHP, 1973, 1.

⁵⁹ Andersson, J. Gunnar, Children of the yellow earth, London, 1934, 272Google Scholar f. and fig. 121.

⁶⁰ Ta-wen-k'ou: hsin-shih-ch'i shih-tai mu-tsang fa-chüeh pao-kao , Peking, 1974, 98.

⁶¹ Shan-tung an-ch'iu ching-chih-chen hsin-shih-ch'i shih-tai mu-tsany fa-chü , 1959, 4, p. 27, fig. 10, 1.

⁶² See n. 9. Hayashi (1969, 208) thinks that the CFT may be the jade form to which the scholastic ritualists of the late Warring States gave the term kuei-pi (Chou li, 5, 36a, SPTK). His reason for this theory is that the CFT might have been taken by them as a pi disc with three kuei sceptres lying around its circumference. He makes it clear that this is not intended as an explanation of the original of the form in Shang times, and that it is, in any case, an identification that must be regarded with caution.

⁶³ Che-chiting chia-hsing ma-chia-pin hsin-shih-ch'i shih-tai i-chih ti fa-chìeh KK, 1961, 7, pp. 349 f. Two jade specimens were excavated; in each case they were found next to the head of the skeleton.

⁶⁴ See n. 61.

⁶⁵ Hsiao-t'un , vol. 1, I-chih ti fa-hsien yü fa-chiien , part 2, Nanknng. 1959. pi. 62b; Kuo Pao-chūn Hsün-hsien hsin-ts'un , Peking, 1964, pl. 50, 1.

⁶⁶ Apart from the astronomial theories of Wu Ta-eh'eng and Michel, various attempts have been made to link the shape of the CFT to some practical function. Salmony (1952, 69 f.) feels that these objects were cord-holders on silk-bales; Na Chih-liang (1967, 13–15; 1974, 157–9) proposes an elaborate theory in which the CiT is derived from a spinning-whorl (fang-lun . He suggests that a whorl that had been made unbalanced through loss of part of its circumference was restored to balance by the deliberate breaking of the circumference at a further two points, thereby giving a three-lobed form. He feels that this accidentally produced shape would have been adopted for its attractive suggestion of rapid swirling, and that the flanges were added later for purposes of decoration. Kuo Pao-chün (Ku-yü hsin-ch'üan Li-shih yü-yen yen-chiu-so chi-k' , vol. 20, 2, 1949,30 f.) thinks they originated from part of a loom. Theories such as these have arisen because the shape of the CFT has so impressed scholars that they have felt it required some special practical explanation, thereby ignoring the possibility that its origin might be understood through an analysis of its stylistic make-up.

⁶⁷ A further feature of the CFT which has been claimed as significant is the presence of linear markings as seen on one face of Wu Ta-ch'eng's specimen (see pi. 1(6)). These occur on four (Ku-yü t'u-k'ao, 1889 (see pi. I); Nott (1936, pi. XVIII); Needham (1959, fig. 151); Dohrenwend (1971, 54)) out of the ten specimens of the CFT-type known to me, not, as Hansford (1968, 66) claims ‘on nearly every specimen’. Whatever the origin of the markings on Wu's specimen, there seems a strong possibility that those on other specimens are the result of deliberate forgery, copying from the line drawing given by Wu (pi. I(fc)). As Loehr (1975, 10 f.) has pointed out, the parallel lines in Wu's drawing correspond to a single deep groove on the original specimen (Laufer, 1927, 28 f.). On the specimens described by Nott, Michel (see Needham, 1959, fig. 151) and Dohrenwend, two single lines replace the deep groove, suggesting that a forger copied these from Wu's drawing without actually having seen his specimen. It is possible that the lines were added later to a genuine ancient specimen of a CFT; however, judging from photographs alone, the crudity of workmanship shown by Nott's and Michel's specimens is such as to suggest that these two pieces may be modern forgeries.

⁶⁸ There is in addition a single example of a ritual stone axe of reported Neolithic date that bears a flange in very low relief on one side; the flange on the other side is apparently broken. The form of the flange appears to be a variant of the six toothed type described above. The axe is 16–7 cm. long by 13 cm. wide and has two perforations. It was excavated from a site at Tan-t'u-ts'un in Wu-lien county, Shantung province, with a number of other fine ritual stone axes having one and two perforations. (Shan-lung hsien wen-vm kuan-U-ch'u , SHan-tung hsien po-wu-kuan shan-tung-lilcli'u, P'u-ch'a pu-fen , Peking, pl- 2, n08) In the absence of any excavation details through which to date and identify this site more exactly, this single flanged specimen is not enough evidence to be able to argue for the Neolithic origin of the flange on ritual implements.

⁶⁹ Length 65–2 cm.; width 9–6 cm.; maximum thickness 0–4 c m.; found in 1975, KK, 1978,4. p. 270, pi. 12, 1;Wen, Fong (ed.). The Great Bronze Age of China, New York, 1980,Google Scholar eat. no. 3. There is an increasing tendency for t he Erh-li-t'ou culture to be identified with the Hsia dynasty (Wen Pong, 1980, 2 f., 70 f.). So far, however, no written evidence has been discovered which would confirm this theory in the way that the oracle bones did in the case of the Shang.

⁷⁰ See the remarks by Robert W. Bagley in Wen Fong (1980, 73).

⁷¹ Cheng, Te-k'un. Archaeology in China, I. Prehistoric China,Cambridge, 1959, pl. XIV. 4.Google Scholar

⁷² Andersson, J. Gunnar, The site of Chu Chia Chni, Hsi Ning Hsien, Kansu, BMFEA, 17, 1945, 52,Google Scholar pi. 21 (2); Cheng Te-k'un (1959, pi. XIV, 3); Kan-su yung-ching ta-ho-chuang i-chihfa-chiieh pao-kao KKHP, 1974, 2, pp. 43 f., pi. 16, 5, 6, 7, 10.

⁷³ A further two specimens of the ritual jade tao have symmetrically arranged teeth in thecentre of one end of the blade (Willetts, 1965, colour pi. 7, Shang or Chou dynasty, length 14 in., Freer Gallery of Art, Washington; Loehr, 1975, black-and-white, pi. 210, Shang or WesternChou, length 61 cm., Grenville L. Winthrop collection in the Fogg Art Museum, HarvardUniversity). In both cases, the overall asymmetry of the specimen suggests that the other end of theblade has been broken off and the fracture repaired by repolishing. The result is a blade flared atone end but relatively narrow at the other. In the case of the Freer specimen, the fractured endhas been considerably recut and teeth have been added of a type different from those at theunbroken end, whereas on the Winthrop specimen, the repolished end has been left undecorated.

⁷⁴ Yen-shih erh-li-t'ou i-chik hsin-fa-hsien ti t‘ung-ch'i ho yϋ-ch'i , KK. 1976, 4, p. 262, pi. 6, 1 (see pi. IV(c) and (d)).

⁷⁵ Huai-an ch'ing-lien-kang hsin-shih-ch'i shih-tai i-chih tiao-ch'a pao-kao KKHP, 1955, 9, pi. 1, 1; 14–4 era. in length; Nanching-shih pei-yin-yang-ying ti-i-erh-tz'u ti fa-chϋeh KKHP. 1958, 1, pi. 4. 9; 13 cm. in length. These pieces are also discussed andillustrated by Hayashi (1969, 218, pi. 29. 3 and 4).

⁷⁶ Na Chih-liang (1967, 11) maintains, however, that the flange was an integral part of thestone axe, and served as a means for tightening thongs in the atachment of a haft. In putting forward this suggestion, he ignores the archaeological evidence for the development of boththe flange and axe forms.

⁷⁷ The fact that there are flanges on jades as early as the Erh-li-t'ou period indicates thattheir formation was entirely independent of bronze flanges which do not occur until the Anyang period.

⁷⁸ An-yang yin-hsϋ wu-hao-mu ti fa-chueh ,. KKHP,1977, 2, p. 77, pi. 22, 2. Scholars have agreed that the date of this tomb lines with the MiddleAnyang period (Wen Fong, 1980, 8). Another similar specimen of a rectangular axe, made ofstone, is illustrated inHou-chia-chuang Taipei, 1962, pi. 105, 8; 17'1 em. in length.

⁷⁹ An-yang-sMhhsi-chiaotiyin-taiwen-huai-chih Wen-wu, 1958, 12, plate inside the back cover, 31. Despite the poor quality of the photograph, the flange design exemplified by the jade in plate IV(ft) is clearly identifiable.

⁸⁰ See also Salmony (1952, pi. 31, 2) and Hayashi (1969, pi. 24, 1).

⁸¹ See note 54.

⁸² Specimens from Lung-shan related sites: Miao-ti-kou yii san-li-ch'iao Chung-kuo k'o-hsϋeh-yϋan kϋao-ku yen-chiu-so Peking, 1959, pi. 70, 8, made of stone, 4 cm. in length;Li-hsi fa-chlieh pao-kao Chung-kuo k'o-hsϋueh-yϋan k'ao-ku yen-chiu-so, Peking, 1962, pi. 34, 13, made of jade,8'2 cm. in length.

⁸³ Ch'ang-liai-lisien kuang-lu-tao ta-ch'ang-slmn-tao pei-ch'iu i-chih KKHP, 1981, 1, fig. 8, 17; 86.

⁸⁴ These three trilobate jade discs have been excavated from a site at Ssu-p'ing-shan,. Ying-ch'eng-tzu in Liaoning province (Sekai bijutsu zenshu I, Tokyo, 1953, colour plate 8; Hayashi Minao, 1969, 210).

⁸⁵ KK, 1977, 3, p. 157 and 172. A further trilobate of Shang date is a button-like shell object 2 cm. in diameter (fig. 11; Te-k'un, Cheng, Archaeology in China, II, Shang China, Cambridge, 1960Google Scholar, pi. XXVe).

⁸⁶ Hou-chia-chuang, Taipei, 1962, pl. 125. 10 and 11Google Scholar. The specimen illustrated here in fig. 10 appears to have been damaged in the drilling of a second perforation. The presence of a fifthlobe can, however, be inferred from the form of the rest of the disc.

⁸⁷ See n. 53: KKHP, 1936, 1.

⁸⁸ See nn. 57 and 66.

⁸⁹ Hayashi, Minao. In Shū jidai no kikagakuteki nn monyō ichi, ni ni tsuile , Tōhōgaku 26, 1963, 1–16).Google Scholar

⁹⁰ Similar theories have been put forward by Ma Ch' eng-yiian (Wen Fong, 1980, 8 and 30)and T'ang Lan (Ts'ung ho-nan cheng-chou ch'u-t'u ti shang-tai ch'ien-ch'i ch'ing-t'ung-ch'i t'an-ch'i , Wen-wu, 1973, PP. 5–7).

⁹¹ This form bears a striking similarity to the trilobate disc.

⁹² 1969, 208–11. Hayashi feels that the whorl motif may have a symbolic connexion withthe radiance of celestial bodies. Such a connexion would, in his view, strengthen his conjecturethat the CFT should be identified with thekuei-pi referred to by the late Warring Statesritualists (see n. 62) who stated that this object was particularly associated with the worship ofthe sun, moon and stars. If this is so, it would turn out that the CFT might after all have anastronomical connexion, if only a symbolic one. However, this view relies on two very tentativehypotheses; it is particularly hard to believe that the scholastic ritualists could have had detailed knowledge of the religious circumstances under which trilobate discs appeared nearlya thousand years before.