Symbolic Jades of the Erlitou Period: A Xia Royal Tradition Author(s): Elizabeth Childs-Johnson Source: Archives of Asian Art, Vol. 48 (1995), pp. 64-92 Published by: University of Hawai'i Press for the Asia Society Stable URL: http://www.jstor.org/stable/20111255 Accessed: 12/07/2010 13:22

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=uhp.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

University of Hawai'i Press and Asia Society are collaborating with JSTOR to digitize, preserve and extend access to Archives of Asian Art.

http://www.jstor.org Symbolic Jades of the Erlitou Period: A Xia Royal Tradition

Elizabeth Childs-Johnson Hamilton College

are 1. here has been lively discussion recently about the cul Dawenkou jades marked by ornamental and refined tural identity of the Xia, allegedly China's earliest historical tool types. Longshan period jades of Shandong include for period, dating to ca. 2100?1700 bce. Due to the lack of any mally designed blades, some of which are directly related extant written documents from this period historians have to Erlitou period jade-working. The exploitation of jade been hesitant to confirm the period's historicity.1 None and the popularization of symbolic jade forms during the as a theless, due to C14 dates and the geographical distribution Erlitou period can be viewed generally climax to a of cultural remains throughout Henan, Shaanxi, and longstanding pre- and proto-historic tradition concen Shanxi provinces, archaeologists and art historians favor a trated originally not in central but in coastal China. Xia rather than Early Shang identification for the Erlitou ERLITOU PERIOD JADE TYPES AND THE culture.2 Excavated finds reveal plans of major religio-ad ERLITOU PERIOD STYLE ministrative centers and pubescent ritual bronzes such as jue and jia, and now ding and he, that mark the beginning Jades excavated from Erlitou period sites derive primarily of bronze casting for ritual purposes.3 Far more prominent, from the site of Erlitou of Erlitou period III, although a nonetheless, in the excavated material, and overlooked by few works date to periods II and IV.5 Almost all of the ex come or near or modern scholarship are the specialized, large in scale blades cavated jades from burials pits found in the nos. i 2 worked out of nephrite jade. Although these nephrite vicinity of the and "Palace" remains. The earliest were blades when published may be labeled ritual implements, jades discovered at Erlitou in 19676 and the latest thus are so there has been no discussion of why they are ritual or what far reported in 1987.7 Most of these jades artistically art on they signify artistically in terms of early Chinese cultural phisticated works of modeled utilitarian tools and to show no evidence of use and since the ma history. In order illustrate that these Erlitou jades iden weapons. They were tify amajor tradition artistically and historically, I shall be jority found in moderately well-to-do burials, these as gin by identifying stylistically representative types amidst jades may be generically described symbols of wealth, out and as archaeological remains and amidst collections in and perhaps proto-insignia. side China. Then I shall compare these representative jade Major types of these symbolic jades of the Erlitou cul works of art with excavated examples of earlier and over ture include the zhangc-blade, daod-knife, faceted yuee-ax, are lapping Longshan period sites, and of later provincial and and gef-dagger (Fig. ia). Other, less conspicuous forms to to a the handle and Bronze period sites further define what appears be guis-blade, zuh-arrowhead, attachments, major jade-working tradition unique to the Erlitou cultural various ornaments (Fig. ib). The names for certain of these are in some cases since are func period and theoretically Xia dynasty. jades self-explanatory they Due to recent excavations of the last ten years it has be tionally descriptive of the original weapon or tool from are come evident that during the Late Neolithic, ca. which they derive. Representative of this category the names 3500?2100 bce, ancient China witnessed a long period of ge-dagger, yue-ax, and dao-knife. Other for Erlitou as are jade-working. Various cultures aligning coastal China from period jades, such zhang and gui, drawn from northern Shandong to southern Zhejiang, such as the Eastern Zhou and later Han ritual texts like the Zhouli1 Hongshan of northernmost China, Dawenkou and Shan (Rites of Zhou) and Lij? (Records of Rites). are cases dong Longshan of northeastern China, and Liangzhu of Prototypes of Erlitou period jades in all trace to southeastern China, are individually representative of jade able Neolithic utilitarian implements originally fabri stone ex working on amajor scale.4 Each culture is characterized by cated in (Fig. 2). The jade weapon type yue, for idiosyncratic types of well-developed jade forms. Hong ample, is derived from the utilitarian stone ax, fuk (Fig. were shan jades, for example, comprise primarily amuletic 2d). Stone fu, which used to chop, split, and hew are dragon and cloud forms. Liangzhu jades, on the other wood, ubiquitous amidst archaeological finds of hand, are distinguished by specialized conga and bib shapes. Neolithic date.8 Early Neolithic examples from Hemudu

64 1

? ?

I\l

O O O pa ??\

i. A. of works of art from Henan: i. 2. Fig. Major types jade Erlitou, Yenshi, Zhang (Kaogu 1983.3, fig. 10:5-6, p. 204); Yue (1. KG 1984.1, fig. 5:2, 2. KG KG KG p. 38; 1976.4, fig- 6:4, p. 262; 3. 1983.3, fig. 70:1, p. 204; 4. 1978.4, fig. 1:1, p. 270); 3. Dao (1. KG 1985.12, fig. 8:1, p. 1092; 2. KG 4-10, KG Ge KG 2. 1975.5, fig- p. 305; 3. 1978.4, fig. 1:3, p. 270); 4. (1. 1976.4, fig. 6:6, p. 262; KG 1975.5, fig- 4-7> P- 305).

a Q. .

M fh

B. of works of art from Erlitou: 1. Handle 2. KG 6: 1 and KG 2. Secondary types jade (1, 1976.4, fig. 5, p. 262; 3. 1975.5, % 4-*4, P- 305); Gui (1. KG 1975-5, 4:6, 305; 2. KG KG % p. 1983.3, % 10:7, P- 204; 3. 1976.4, % 6:2, p. 262); 3. Arrowhead (KG 1983.3, fig. 7:4, p. 215); 4. Bead (1. KG 2. 1983.3, % 104, p. 204; KG 1984.1, fig- 5H-5, P- 38).

65 o o o m I

utilitarian A. From to nianhui Fig. 2. Five Erlitou period jade types and their prototypes: chan-spade zhang-blade (Zhongguo kaoguxuehui lunwenji c. From ben-adze to WW 1980.1:99); B. From lian-sickle to ge-dagger (Wen Wu 1984.2, fig. 3:9); gui-blade (Kaogu 1975.5, fig- 4^6, p. 305; 1984.2); e. From dao-knife to dao-blade D. From fu-ax to yue-ax (KG 1978.4, fig. 1:1, p. 270; KG 1981.3, fig. 6:1, p. 196); (KG 1975.5, fig- 4:10, P- 395)

a For the that the latest of in Zhejiang province,9 for example, show standard design example, yue appears during phase was ca. is characterized a of a broad and thick blade with stepped shoulder that Hemudu, 3200?2700, by slightly use a hole in fastened to a wooden or bone handle at a 90-degree angle. rounded blade shape and by the of perforated to at stead of a shoulder for attachment of handle to Examples from the earliest the latest phase Hemudu stepped site not ex evolve from crude to more refined shapes, from rough to blade.12 Although the author of the report does to his use of the term as to the more polished, and from small very large examples.10 plain yue opposed fu, a new hole for and The fu-ax is a tool whereas the yue-ax is weapon used refined shape, perforated attachment, slender are characteristics that to kill and slaughter theoretically humans and animals. proportions outstanding define later functional and bronze and Although the yue weapon evolved from the utilitarian fu symbolic jade yue.13 as a It is that about bce stone and tools tool, and although the tool may have been used apparent by 3500 jade to and and weapon and the initial weapon as a tool, their differences begin be consciously consistently polished, by in ca. bce tools and to appear to be practical and chronological. Hunting gear 3300 weapons begin appear commonly arrows are to been as Our means of utilitar cluding spears, bows, and known have specialized objects. distinguishing used ian and defensive axes are therefore and formal.14 A used throughout the Neolithic. Specialized weapons stylistic and for defense, on the other hand, begin to appear in archae thinner fabric, refinement through polish, perforated other forms hole for the ax as a ological assemblages at the time city walls and hafting qualify yue weapon. fu and is of defense appear, which is late in the Neolithic, during the The distinction in function between yue bce in and clarified in In the article "The Longshan era of ca. 3000?2000 Shandiong Shang script. Classification, and of Xia Nai at Henan provinces,11 but also earlier with the yue-ax that Nomenclature, Usage Shang Jades," In to revise traditional nomenclature for in emerges during the Late Hemudu and Liangzhu periods. tempted jades, a on ev common yue and fu ax types share a handle set at right cluding the yue, by initially drawing archaeological a an bowed idence.15 His discussion of in the section angle and rectangular blade with outwardly weapons "Weapons two are and clouded a lack of distinc edge of which one and sometimes sides beveled. Implements" is, however, by a tion between tool and between utilitarian Differences are marked primarily by refinement of form. weapon forms, 66 fu and weapon yue. At one point yue are described as flat fu and a weapon form of chan-spade, and at another point as large fu, as suggested by the Han Shuowen definition. Xia does not rely on archaeological data or written records from Shang times to amplify his discussion of axes but rather still depends on the traditional nomenclature of the Qing archivist, Wu Dacheng, when using qi1 (ch'i), for ex ample to refer to a yue ritual tool with flanking side dentil decor. In the Eastern Han Shuowen, yue is defined as a big fu ax or simply fu and iswritten both withe and without111 the metal radical.16 Earlier in Shang oracle bone inscriptions the yue graph is used as the verb "to hack or behead a sacrificial victim with the ritual ax."17 This verb of sacrifice11 in inscriptions frequently takes sacrificial animals or prisoners of war as its object. Although no bronze ver sions of yue have yet been discovered at Erlitou, later ex amples from Early and Late Shang sites of Zhengzhou and Anyang date mimick the shape of pre- and proto-historic versions from jade-working sites of Longshan and Liang cm cm zhu date. From a functional point of view the yue then was Fig- 3- Jade yue, 9.6 wide, 0.6 thick, Erlitou. Kaogu 1976.4, pi. 10:2. an ax used on ritual occasions during Shang times. Since the shape and form of Shang yue are stylistically based on Erlitou prototypes18 it is apparent that we are dealing with the specialized yue rather than with the utilitarian fu in elite burials of the Erlitou period. At Erlitou six yue have been excavated. The trapezoidal version with one or two hafting holes (Fig. IA2) was stan dardized earlier at Liangzhu and Longshan sites.19 One variation carries decorative serrations along the upper two sides of the blade and a blunt cutting edge with two-sided The excavated in like another exca cm cm cm beveling. latter, 1975, Fig. 4. Jade ge, 30.2 long, 6.6-6.9 wide, 0.5-0.7 thick, vated in 1982, is small, measuring 11.2 cm tall, 5.8-6.8 cm Erlitou. Kaogu 1978.4, pl-7:i xia.t wide, and 0.6 cm thick.20 A fourth example of this jade type of yue, from burial no. 3 at Erlitou, has no serrations but does have a turquoise stud that fills a small hole in the lower the yue. Ge is well known in oracle-bone26 and bronze part of the blade.21 The second type of yue, so far singular scripts, where it refers to the weapon used to attack and stab to Erlitou sites, is the round and square shape with large, to death an enemy.27 The ge graph is used as the signifie in central hole and blade that is faceted into four beveled a variety of other Shang words that are similar in meaning, as or or edges (Fig. IA2 left and upper right). This round, almost such "to cut down"0 "to attack,"P it is used in the disk-shaped version22 is known through two examples composition of the character for shields28 As demonstrated from K3 in area III at Erlitou (Fig. 3).23 The slightly larger long ago by James Menzies, the ge in design and function example measures 9.6 cm wide and 0.6 cm thick with a originated in the agricultural tool, the lianr-sickle (Fig. central hole that is 5.2 cm in diameter. The square and third 2b).29 As with certain other Erlitou jades, the ge are large, variation of yue is from burial no. 6; it is the largest known impressive, and sometimes decorated with abstract linear area exca example from Erlitou, measuring 21.0 cm tall by 23.0 cm motifs in the of the haft (Figs. IA4; 4). The ge wide, with a central hole that is 4.6 cm in diameter (Fig. vated from K3 measures 30.2 cm long, 6.6-6.9 cm wide, IA2 left).24 Serrated tooth decor, as on the trapezoidal ver and 0.5-0.7 cm thick; and the ge-blade from burial no. 37 sion, symmetrically aligns the two sides. The use of a char measures 43.0 cm long, 8.0 cm wide, and 0.5 cm thick.30 acteristically thin but tensile slice of jade material and ap The undecorated example unearthed from a pit within measures cm cm plication of decor through faceting or indentation clearly foundation F3 21.9 long and 3.8-4.7 identify these yue as Erlitou in style and date. wide.31 The Erlitou type of ge is distinguished from early The ge-dagger ax (Fig. 4) is another form of weapon Shang examples at Zhengzhou in jade and bronze by its new to this early historic period, when defense assumed an symmetrical shape and centered blade point.32 The tip of all-important role.25 As a weapon the ge appears later than most later Shang examples begins to point downward and

67 extremely thin isosceles trapezoids with usually three but seven also equidistant holes for hafting, respectively. The ones measure larger 65.0 (bottom)-6o.4 (top) cm long, 9.5 cm cm wide, and 0.4-0.1 thick; and 53.5 cm long and 8.8 cm wide (Fig. 5);36 a slightly smaller one measures 46.5-52.3 cm long and 9.8-10.3 cm wide; and the shortest one, 25.9 cm long and 11.5 cm wide.37 Two of the exca 5. cm cm cm Fig. A.Jade dao, 60.4-65.0 long, 9.5 wide, 0.1-0.4 thick, vated dao?the and that cm from b. largest measuring 52.3 Erlitou; Detail of geometric decor. Kaogu 1978.4, pi. 11:3; Wen decorated with the of Fong (ed.), The Great Bronze Age of China (New York, 1981), pl. 3. M57?are type geometric design found on both ge and zhang. As on the ge, these geomet ric motifs include sets of incised slanting lines here crossed a like X's to form diamond pattern that is framed by pairs an of vertical lines and by outer framing line mimicking the shape of the isosceles trapezoid blade (Fig. 5B). As on the yue, small-scale serrations decorate flanking sides; here they are systematized as evenly spaced nodules. The dao have one front face; the backs usually are left undecorated, al are though polished. Holes drilled from the front side only. are The zhang-blades the most surprising and exquisitely worked jades of the Erlitou period. As argued by Dai Yingxin,38 these eccentrically formed blades with con cavely bowed end originate in the tool type of similar shape called chans-spade, well known at the early southern Neo lithic site of Hemudu in Zhejiang (Fig. 2A) but also evi at dently Miaodigou period II in Henan. The peculiar shape of this spade is determined by the origin of this tool in the or shoulderblade of water buffalo water deer.39 The shape of this bone turned tool is mostly flat and narrow with a a median line runs the of strong longitudinally along length stepped shoulder used for hafting and a softly flaring broad the blade. of the Erlitou is the on Typical style emphasis blade theoretically used for digging. As a tool type it is but also decoration that is either and symmetry geometric known throughout the Taihu region of Ma Jiabang and linear or in the form of faceted such as decorate the edges, Liangzhu cultures.40 Although the lashing technique may two sides of the from the blade long ge K3. Proportionally, have differed slightly between tool and later Erlitou jade is and and the handle short. The middle is long elegant symbol, the design is similar. As Dai suggested, artistic li thicker than the sides and the are beveled. was edges evenly cense exercised in determining the arrangement of The Erlitou is identical in to excavated bronze jade ge type handle and length of the blade.41 In frontal outline both such as those in The blades of versions, published 1976.33 share the indented rectangular haft and long trapezoidally the bronze are that were examples sharp, suggesting they flaring biconcave blade with crescent-shaped edge. Al for use. The of the excavated designed workmanship ge though the secondary hafting holes perforated on the util from and also is in of Erlitou, yue, very high quality jade itarian bone spade are deleted on the symbolic jade blade, as out the team working; and, pointed by archaeological one perforation remains on the haft like the utilitarian at the refined of working Erlitou, exquisitely technique spade. Since there are few documented bone chan of sim and of match the best work grinding polishing produced ilar shape to those from the Taihu area at northern sites in in of Erlitou today jade workshops Beijing.34 Clearly, jades Henan, Shanxi, or Shandong, the inspiration for this pop are of artistic and in this masterpieces design respect repre ular Erlitou jade, on the basis of present evidence, seems to sent a artistic center for standards. major setting stylistic be primarily southern and Neolithic in origin. To date no The daod and the third and fourth forms zhang,c major zhang have yet been discovered in Liangzhu tombs or site of at are also for their symbolic jades Erlitou, noteworthy finds, although, according to very recent reports, zhang and with exquisite quality elegant style, emphatic symme have been "uncovered" amidst Shandong Longshan and and of The dao-knife is try strength design. self-explana Yueshi period remains.42 in name and for it is as a tool in name tory function, ubiquitous The zhangc derives from later ritual texts, primar excavations the Neolithic the throughout (Fig. 2e).35 By ily the Zhouli. There zhang and numerous variations based Erlitou this utilitarian saw has lost all functional ties period on descriptive prefatory qualifiers are cited. These varia with its in a formal as predecessor favoring strictly design tions, such yazhang,* dazhang,u and bianzhang,v are in of The four dao excavated at Erlitou are cases symbolic power. most simply descriptive of outstanding properties 68 Jadecm ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^H KaoguFong ^^^^^^^^^^^^^^^^^^^^^^^MiCj??JUkflH^Hi^^^^^^l^^^^^^^^^^^^lV^^^H 7//C Bronze ^^^^^^^^^^^^^^^^^^^^^^^^^Bfol^^^^^l^^^^^^^^^^^^^^^^^^^^^^^HI^^^I

sharp diagonal?this is the type represented in Han ritual cm 0.8 Fig. 7. Jade handle, 17.1 long, texts under the label zhang and defined "half a gui."47 This cm burial no. Erlitou. wide, 3, Kaogu "half a form of the is also very close to the 202. gui" zhang shape 1983.3, pl. 1:4, fig. 7, p. assumed by long dao blades that have been re?ut by slicing the blade in half (see the attempt to re?ut the upper edges of the dao in Fig. 12) and by trimming the ends into edges that are oblique.48 Although the name zhang does not ap pear in aXia or Shang literary context, that it describes the form represented at Erlitou is clear from the evolution of this jade as a type. Three zhang have been excavated at Erlitou. The exam ple discovered in 1975 is bone white in color, measuring cm cm and cm thick 46.0-48.0 long, 4.0 wide, 0.4?0.5 (Fig. 6).49 It is celebratory in design, elegantly long and flat with slightly forked head. In complement to this blade of classic proportions, small-scale, paper-thin thread ridges decorate the front side of the haft and leave a dentated sil houette of paired and more deeply cut dentils at the edges. This relievo design with staccato rhythm is comparable to those decorating the yue and dao blades but is by far the most complex and sophisticated version known amidst Erlitou cultural remains. A small perforated hole relieves the inner end of the handle. The other two zhang from Erlitou were found together in burial no. 3 from area V.50 are Although like the bone-white zhang they similarly thin cm and cm and long, measuring 48.1 54.0 long, respec cm tively, they are also considerably wider at 11.4 and 14.8 a cm. All three have in common an emphasis upon front face where and serrations are oriented. The perforations 54.0 cm-long zhang, like one of the Erlitou yue, has a turquoise in into the small hole at the of characterizing the zhang.43 Yaw of yazhang, for example, stud inserted, this case, edge the blade. The blade heads are worked to a refers to ya-teeth of yachi and to the dentils frequently methodically thin that forms a delicate half-moon in found decorating the haft of zhang, as pointed out by the very edge shape. of the burial does not the Han commentator Zheng Xuan.44 The zhang type, which The haft decoration zhang carry but it still is illustrated as one of six auspicious jades on several Han refined, raised relief of the 1975 excavated example, that is with of al stele,45 can be traced back to the early historic blade of the bears the elegant silhouette outlined pairs are as and more worked Erlitou period. In the Han Shuowen zhang described ternately deep lightly projections. thin and that is "half of a gui."46 This explanation is in part correct because Another type of jade, long, rectangular, amidst Erlitou remains is the so-called "handle" it refers to the degenerate form of zhang of Shang through prominent Not one has attached to Han date. This degeneration of the zhang blade is illus (Figs. iBi; 7). been found any its use.51 trated by the example from The Minneapolis Museum of other material that may signify Small, perforated holes at the and sometimes Arts (see Fig. 16). The blade is stripped of its usual haft and frequently appear handle's tip in from 10.0 to crescent-shaped mouth. By Shang times this form ismerely top edge (see Fig. 7). They vary length 17.0 cm in width from 1.6 to 2.2 and can be a shapeless long blade without handle and an end cut at a and cm, very thin,

69 averaging 1.0?2.0 cm thick.52 Their shape is distinctive coeval with the Longshan.60 On the basis of the excavated with an indented grip at one end and a blunt or broken jades from Erlitou, it is evident that certain forms, such as were as edge at the other. The longest example, at 17.0 cm, dis zhang, dao, ge, and yue favored symbolic markers covered in 1975, is decorated with a series of stacked semi and, in this respect, appear as a precedent for the Chinese human and tiger heads that alternate with plumelike motifs aesthetic favoring "politicized" art forms in jade. Stylis once imitative of a bird-feathered headdress (Fig. 7).53 tically, Erlitou jade works of art are classical in expression; Some fragmented examples show that this plume motif they are often characterized by a large scale, emphatic sym may degenerate into an independently repeated motif ver metry, elegant design, geometrically precise decorative de a or an ex tically aligning the handle (Fig. ibi). tail in the form of delicate relievo incised bands, Another symbolic type of jade form from Erlitou is the tremely refined quality of workmanship, a frontal name for one and a two-dimensional surface. guis-blade. The name^m is derived from the orientation, usually flat, were are numerous in of the six auspicious jades of Han times that made Comparative zhang-blades collections popular by reference in the Zhouli and its commentaries.54 outside China. Representative examples of Erlitou period on Like the zhang, which because of certain attributes took date belong, for example, to The Minneapolis Institute of other names, the gui too became various in label by Han Arts (see Fig. 16), The Field Museum in Chicago (Fig. 8), a times. The gui-blade as a generic type nonetheless has the Indianapolis Museum of Art (Fig. 9), and the Norton long history beginning during the Late Neolithic and last Gallery in Florida.61 The zhang from the Field Museum ap ing well into the Han and later dynastic periods. For this proximates in size the larger zhang with turquoise stud ex reason it, like the zhang, may be called by its generic name cavated from burial no. 3 at Erlitou in measuring 52.0 cm gui, first cited in ritual texts of Han date. Gui in the long and 10.0?11.8 cm wide. Standard features include the Shuowen is defined "as round at the top and square at the robust shape with crescentic tip worked to a thin beveled bottom."55 This description fits the jade of that shape edge, and a tang which is strong and rectangular. The tang known from Neolithic through Han times. As with the has a hole for hafting and a jagged silhouette referring to an zhang, gui too appear to have originated in agricultural the complex decor of miniature, multiplied thread-thin tool form, in this case, the benx-adze of Late Neolithic ridges that decorate the haft of the bone-white zhang times (Fig. 2c). Early versions of the gui from Shandong picked up at Erlitou (Fig. 6). Certain motifs on The Field show that the blade was heavily stepped on one edge, as is Musuem zhang underscore Erlitou aesthetics. Small-scale the adze tool (see Fig. 20).56 The gui, then, like the dao and ridges align the front of the haft (Fig. 8a) but similarly play zhang blades, appears to be agricultural in origin. ful pairs of small-scale serrations decorate the inward turn The three gui excavated at Erlitou57 seem insignificant ing arc of the mouth (Fig. 8b)?as if to emphasize the by comparison to their related but earlier counterparts from strictly aesthetic and symbolic function of this heraldic are Shandong. Gui are also small by comparison to zhang and blade type. The zhang from the other three collections 21.1 dao from Erlitou. The undecorated example measures robust with widely bowed, flaring mouths. The example measures cm cm long and 6.4 cm wide, and the other with a band fea from the Indianapolis Museum 31.2 long by cm cm turing a diamond motif at the haft measures 17.4 cm long 12.4 wide (Fig. 9), the Minneapolis example 40.6 and 7.4 cm wide (Fig. 1B2).58 The two have a double set of long by 8.4 cm wide, and the Norton example 39.37 cm cm perforated holes for hafting. Their tips are blunt and forms long by 12.7 wide. Each is typically biconcave in shape, cm simple, showing little connection with their utilitarian pro very thin and flat, averaging 0.4?0.8 thick. The three totype. A third, cruder example found at Erlitou, measur share with the two from M3 of Erlitou the geometric de re ing over 10 cm in length (Fig. IB2 right), may also be sign of the haft. Two sets of extended dentils frame exactly across lated to the gui as a type. two pairs of shorter ones, and all continue the haft in Ornamental forms, such as tubular beads and awl-shaped delicately raised ridges. The strength of design and the clas are are not numerous amidst ex sic balance in the of these four fit the spear points, known but proportions zhang a cavated finds (Fig. 1B4). Only six examples have been phase represented by examples from Erlitou, phase that are a found in excavations.59 Leaf-shaped arrowheads in jade evidently witnessed climax in the perfection of this type also represented (Fig. 1B3). of symbolic jade. Two ge-blades from the collections of Dr. Paul Singer CORROBORATION OF THE ERLITOU STYLE: (Fig. 10) and the Buffalo Museum of Science62 also reflect OTHER ERLITOU PERIOD JADES FROM COLLECTIONS a style and form comparable to the classic expression repre IN AND OUTSIDE CHINA sented by works at Erlitou. The ge from the Buffalo measures cm over 10 cm A variety of comparable jades from Western and Chinese Museum 43.8 long, which is collections corroborate the evidence that the group of sym longer than that from Erlitou. What is Erlitou in taste is the an rec bolic jades excavated at Erlitou served asmarkers of elite interest in symmetry and balance of the geometrically power that thrived during the Erlitou period and theoreti tangular handle and central blade point. By the Zhengzhou cally its preceding phase, tentatively labeled early Xia and period of early Shang times the point of the jade and

70 a. cm. VIHHHHMjMgMMMMM|^^^^^^^^^^^^^^^^^^^g^L^^^^^MC Jade zhang, long, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^Bface ofblade

a.

bronze ge bends downward and the middle is strongly marked by amedian Une.63 Small-scale dentils, as found on zhang, yue, dao, and now ge, align the end edge of the tang. In contradistinction to the Buffalo ge, the jade ge from the Singer Collection is hard in outline and design, suggesting a date later rather than contemporary with the Erlitou jades. Although the hafting hole is located at a dis tance from the haft, like the bronze ge from Erlitou,64 the blade point is centered and the tang fluted. The geometric decor of a diamond pattern framed by pairs of vertical lines also reflects a favored motif of Erlitou style. Comparable dao from Western collections, in particular the Arthur M. Sackler Collection (Fig. n) and The Art Institute of Chicago (Fig. 12), also document Erlitou pe riod styles. Both Sackler and Chicago dao are extremely long, extending beyond the longest Erlitou piece at 65.0 cm, inmeasuring 101.9 cm and 73.6 cm, respectively. They are 11.2 cm wide by 0.3 cm thick and 10.5 cm wide by 1.4 cm thick. The Art Institute example has been worked on both upper edges, indicating that there was an attempt to re?ut the blade (Fig. I2b), as was often done in post-Xia, now Shang through Han times, when refashioning dao as sions of what is a repeated bracket-shape on flanking zhang.65 Typical geometric designs of Erlitou inspiration short sides of the dao. are the diamond motif flanked by pairs of vertical lines at The version of yue with faceted blade ismore difficult to corners opposite of this long blade. The blade also appears locate in collections in and outside China. One example, to have been reworked on all four edges in forming a long almost identical in form to the rounded yue with blades slender piece of jade rather than a wider long trapezoid, faceted into four edges from Erlitou, is published as a rub are no which would explain why there symmetrically dis bing in the Qing catalogue Guyu tulu by Huang Jun (Fig. posed hafting holes. The Sackler blade, on the other hand, 13). The side serrations, which include two sets of three represents an artistic climax in the expression of the dao as dentils divided by a deeper cut, follow closely the Erlitou a classical work of art: the blade is perfectly symmetrical, interpretation. On other examples of this round-like ax, in formally balanced, an isosceles trapezoid with five equidis the Ashmolean Museum66 and the Bahr Collection in tant hafting holes bored from the front, thus with all the Chicago,67 for example, the serrations are hardened and the subtleties of Erlitou taste. Interest is paid to setting off the round blade is left unfaceted, suggesting that the two blades shape of the blade by creating the same shape in smaller size date either late in the Erlitou period or to the Shang, as rep either through incised outline, as on the excavated dao, or resented by Zhengzhou period remains.68 The Ashmolean through relief outline, as on the Sackler piece. Like the ex piece is quite small, measuring only 9.3 cm in diameter. cavated dao from Erlitou, the raised panel is in turn deco Shapes unknown through excavations at Erlitou but of rated with the symmetrical, geometric motif of diamonds Erlitou style also exist in collected works outside China. and pairs of slanting and vertical lines. The serrated motif, One type is the eccentrically long, gui-shaped blade, in other cases so carefully varied between sets of deep and known through examples published by C. T. Loo69 and in shallow troughs, is here varied as negative and positive ver the Indianapolis Museum of Art (Fig. 14).70 The former

71 cm Fig. 9. Jade zhang, 31.2 cm long, 0.7 thick, Erlitou culture. Indianapolis Museum of Art, Eli Lilly Collection.

cm Fig. 10. Jade ge, 47.0 long, Erlitou culture. Paul Singer Collection.

cm cm Fig. 11. Jade dao, 73.5 cm long, 10.5 wide, 1.4 thick, Erlitou culture. Courtesy of the Arthur M. Sackler Collection, Smithsonian Institution, Washington, D.C. S1987.450.

measures cm cm. 63.4 long and the latter, 57.15 Erlitou pe period aesthetics, such as perfect symmetry of trapezoidal riod features stand out in the affected length and handle design, equidistant hafting holes, and a flat shape that thins decor with serrations and paper-thin raised ridges, and an towards the once-functional cutting edge. Typically the end with further dentils. The heads of these long blades are backs are flat and the fronts highly polished. These and the spatula in shape, which suggests affinity with the strongly long dao, zhang, yue, ge, and certain eccentric forms that beveled edge of the gui and their tool prototype, the ben. exploit elegant length and flatness or finely worked deco Stylized length and small-scale decor through silhouette rative detail in silhouette stylistically identify Erlitou period typify Erlitou style that is here used lyrically in creating an jade-working. unorthodox shape of a symbolic jade. THE RELATIONSIP BETWEEN ERLITOU AND Another type of symbolic jade, not found in excavations LONGSHAN PERIOD JADES FROM HENAN, SHANXI, but of Erlitou style and date, are the short dao with double SHAANXI, ANHUI, AND SHANDONG perforations for hafting. Examples from the I. Wyman Drummond Collection in the American Museum of It may be observed that as a major art form jades repre Natural History,71 the Paul Singer Collection (Fig. 15), and senting the Erlitou style gradually disappear during the now are the A. W. Bahr Collection in The Field Museum72 Shang era, although the style survives in outlying, regional representative. All examples are characterized by Erlitou cultures so far represented by sites as far flung as Fujian and 72 A

cm 11.2 cm B. Fig. 12. A. Jade dao, ioi.o long, wide, 0.3 cm thick, Erlitou culture; Detail of short end of blade. The Art Institute of Chicago, Kate S. Buckingham Endowment. 1954.1197.

Sichuan provinces and now Hong Kong and Vietnam.73 The climax of jade-working, as witnessed in the refined, large-scale symbolic blades from Erlitou, in turn seems to be anticipated at several sites during the Longshan period. Dates calibrated from the northern Longshan and southern Liangzhu cultures are ca. 3300-2100 bce, with the Shandong Longshan covering ca. 2400-2000 bce, and the Erlitou ca. 2100-1800/1700 bce.74 Archaeological data il lustrating the interrelationshp of Erlitou and Longshan jade-working traditions come from several stray finds and excavations in Anhui, Shaanxi, Shanxi, and Henan

provinces. During the Longshan period, beginning at some time during the mid-third millennium bce, a major cultural change ismanifest in the appearance of nonutilitarian, sym bolic jade forms. In southern and coastal Liangzhu period tombs, refined jade yue appear in large numbers alongside "ritual" cong and bi.75 Dao also occasionally appear in ri ;fv~ Liangzhu period remains, such as at Changxun in Zhejiang province and Fuquanshan in Shanghai,76 where standard Liangzhu jades, like bi, cong, and yue, are otherwise prominent. Thus, gradually, during the Liangzhu and also forms and Fig. 13. Rubbing of jade yue, n.d., Erlitou culture. Huang Jun, Guyutulu Longshan periods symbolic jade concomitantly (Beijing, 1939), I:i. weapons increase in production, whereas "ritual" jades,

73 Short cm cm 1 cm Fig. 15. jade dao, 11.9 long, 5.5 wide, under thick, Erlitou culture. Paul Singer Collection.

such as the cong and bi, well known in the south and coastal China, decrease. The most dramatic find corroborating the new emphasis to be played by the symbolic jade is represented by the nu merous stone dao and yue (called fu in the site report) ex cavated at Xuejiagang in Qianshan county, Anhui pro vince.77 Remains from Xuejiagang site derive mostly from cemetery burials with periods II?III representing Liangzhu and two final short periods of occupation representing probably Shandong Longshan (PIV) and Shang (PV). Period II is rich but period III still richer in number of bur ial artifacts. During period III ceramics are made mostly by hand, although rims of vessels are turned on the wheel. are They gray to black in color with a percentage showing cm Fig. 14. Long jade blade, 63.4 a black skin on an otherwise gray vessel, a treatment that cm long, 8.8 wide, 0.4 cm thick. typifies later classical Longshan and Liangzhu period ce Indianapolis Museum of Art, Eli ramics. Amidst IV remains there is a Lilly Collection. period large percent age of eggshell-thin, highly polished blackware dingy tripods, guiz-pitchers, and bei^-cups which point to an occupation level of classical Shandong Longshan date.78 Although during period III few jades appear, refined stone versions of dao and yue, both of which become popular as formal jade types during the Shandong Longshan and Erlitou periods, abound. Forty-nine yue and 36 dao were excavated from 80 tombs of PHI.79 The latter dao and yue from Xuejiagang are flat and thin, are and described by the excavators as showing evidence of use (Fig. 17).80 Certain of the dao apparently were re?ut and reused,81 possibly for the purpose of amassing in a tomb symbols of agricultural wealth and military might. There is no use mention about or lack of use with regard to what appears to be yue rather than fu axes at this site. The spe cial position of these two types of implements, the yue 74 cm Fig. 16. Jade zhang, 38.64 long, 10.79 cm wide, 0.8 cm thick, Late Erlitou/Early Shang period. The Minneapolis Insitute of Arts, bequest of Alfred Pillsbury.

. OO OOOOO

A. Anhui. Fig. 17. Drawing of stone vue, Songze/Liangzhu culture, Xuejiagang, Kaogu B. stone Xuebao 1982.3, fig. 24, p. 309); Drawing of dao, Songze/Liangzhu culture, Xuejiagang, Anhui. KGXB 1982.3, fig. 25, p. 310.

not before the of the and Erlitou weapon and dao-agricultural tool, is only illustrated by height Shandong Longshan are as stone of ver their large numbers but by the artistic interest taken in periods, they significant precursors jade an sions that become the hallmark of classic Erlitou painting certain of their upper surfaces with abstract period are also as one chan bracket motif that approximates the bird-feathered head styles. They significant geographically which Late Neolithic influences dress decorating Liangzhu and also later Shandong Long nel through southern, may to have reached the north. shan jades (see Fig. 20).82 By comparison other tool and the creation of tools and weapon types, yue and dao are greatest in number. The dao It becomes apparent that as the range in size from 15.0 to 51.6 cm long (Fig. 17B) and haft weapons symbolic forms occurred during Liangzhu two are and and climaxed ing holes range in number from to thirteen. All dao Longshan periods (ca. 3300?2100 bce) to in most the classic and trapezoidal in shape, and, according the report, during Shandong Longshan (2400?2000 bce) cases both sides of the blade are characterized as sharp and Erlitou (2100-1700 bce) periods. Although the archaeolog use. is still certain and beveled, thus, they appear designed for The hafting ical data piecemeal, symbolic jade types one or stones from sites in Shaanxi and Shanxi holes vary between being drilled from two sides. related Longshan to The sandstone material varies in color from granitic to can be used document this trend of symbolic jade-mak III at the and Erlitou amethyst purple. Since the stone tools from period ing during Longshan subsequent periods. from in Xuejiagang are dated by C14 to around 3000 bce,83 well Collected jades Cangshanmao Yenanshi, 75 Shaanxi, include bi-disks, degenerate cong, stray imple were collected and out of 400 or 500 only a part have been a cm is no as to were ments, plus yellow-green dao that is 54.6 long.84 The recovered. There accounting which jades latter blade is beveled on both sides, is slightly bowTed in found in tombs and which were simply collected. The re shape, and according to the excavators shows use along its covered part of the jades associated with this site are repre middle edge. Four small holes for hafting pierce the upper sented by a description of 28 zhang, 9 gui, one fu, 5 yue, half of the blade. Three additional holes aligning the upper one qi, 3 ge, and close to 40 dao.90 Outside of one fu and edge indicate earlier use of the jade. Decor in the form of qi, representative specimens selected for description in symmetrically positioned dentils appears along the two clude those that are familiar in Erlitou period tombs and for are short ends of this blade. This feature is typical of Erlitou this reason particularly useful in corroborating the dao and thus can aid in dating these pieces to a coterminous Erlitou tradition of jade-working. According to the au or The thor's and more none of Longshan Erlitou cultural period of expression. present convincing argument,91 cong from Cangshanmao show a total loss of vigor in form these tool and weapon jades show evidence of use. Outside and decor by comparison with their predecessors amidst of the few ornamental huang and bi shapes, dao, yue, and Liangzhu cultural finds, which is another tendency point zhang have blunt cutting edges, indicating that "they were ing to a date in Longshan or the Erlitou period, when these not used implements but rather were ritual implements."92 ritual cong become obsolete as major works of art. Figurai Short and long dao, oblique and arc-shaped zhang imagery is reduced to linear motifs of circles and lines as on blades, yue, and now ge from Shimao are illustrated.93 It is similar cong from Taipingchang in Sichuan85 and Taosi in stated that most of the nephrite jades from this site are black Shanxi,86 also dateable to this Longshan/ Erlitou cultural or dark green in color, are finely worked, and are very thin. are phase. Gray and blackware ceramics picked up at Cang The four longest dao extremely thin, in one case mea shanmao further confirm dating these site finds and jades to suring 0.115 cm thick, thinner than comparative versions the late Longshan/Erlitou cultural phase. from Erlitou. The dao are in general long, varying 14?26 Finds from Taosi in Xiangfen, Shanxi province are only and 49?55 cm and are trapezoidal in shape with a slightly partially published and promise to be rewarding in the fu bowed blade. Shorter, thicker dao vary 13-16 and 19-28 ture. Over 230 burials have been discovered and 109 re cm in length. The two published zhang were in the first re as ported.87 The site is identified a new type-site for the port called chan-spades for reason that they originated in Longshan phase in the Middle Yellow River Valley. Finds the spade type of tool.94 As discussed earlier, like the dao, are are generally comparable to those from the Sanliqiao cul the zhang very thin and long with sharp blades and are ture, a type-site of the Henan Longshan. As at Cangshan mostly black in color of jade. According to the more recent mao, jade cong and bi are found in burials. Cong are ex publication of finds at Shimao, the longest zhang measures cm tremely small (1.3 and 2.6 tall) and show formal 49.0 cm long and 7.8 cm wide at the tip of the blade, and degeneration and lack of decor: the usual four vertical the majority of others average above 30 cm long. As illus a channels separating prisms are described as forming an trated in drawing, the haft of one of the zhang is charac a as eight-pointed star and the decor is limited to three hori terized by serrated silhouette is popular on Erlitou an zontal ducts. Certain of the cutting edges of the yue (de zhang (see Fig. 25A).95 Other zhang are simplified to scribed as chan?) are said to be blunt and thin, indicating oblique instead of half moon-shaped blade edge and are that yue here are symbolic and not used objects. The one without a dentate silhouette; this shape is generally similar published dao is comparable to those from Cangshamao in to the bronze version of zhang from Erlitou96 and the de Shaanxi. It measures 25.0 cm long, is asymmetrical with generate jade version represented by the blade from The two perforated holes for hafting, and the blade is beveled Minneapolis Museum of Arts (Fig. 16). The three ge de are on both edges. scribed in the second report typologically and stylisti Jade and stone implements from Shimao in Shenmuxian, cally comparable to the Erlitou jade ge. Their dimensions, Shaanxi are the most rewarding in documenting the cul which range from 36.5, 29.4, and 21.0 cm in length and 0.6 tural interp?n?tration of Erlitou and earlier Longshan tradi to 1.0 cm in thickness,97 are comparable to Erlitou ge. The cm tions in the Yellow River Valley. Ceramics from Shimao black ge measuring 29.4 long from Shimao is compara are characteristic of period II at Kexingzhuang, which is ble to Erlitou versions in favoring a symmetrical blade with a on equivalent to the Longshan Neolithic in Shaanxi but could central point that is sharp the upper and lower edges overlap into the subsequent Erlitou phase.88 A variety of and with facets extending along the length of upper and tombs have been cleared but only a selection of the finds, lower parts of the blade. Neither has a central median line a including jades, have been published. It is reported that but both have hafts differentiated from the blade and per are a one finely worked jades rather than ceramic vessels promi forated hole for attachment of handle. Only of the nent in tombs of this date; this phenomenon reflects elite jade yue is illustrated with the exception of another labeled are burial interests typical of both Shandong Longshan and qi. All yue characteristic of Erlitou period types. The a Erlitou. According to the 1988 report of Dai Yingxin,89 the yue labeled qi has rectangular shape decorated with den on as one majority of the jades from Shimao were not excavated but tils flanking sides at the haft, typifies version of

76 B

a cm cm Fig. 18. a. Drawing of stone/jade gui, 18.0 long, 4.5 wide, 0.85 cm thick, Shandong Longshan, Liangchengzhen, Rizhao, B. a Shandong. Kaogu 1972.4, fig. 2, p. 57; Revised drawing of cm cm. stone/jade dao, 48.7 cm long, 12.0-15.0 wide, 1.5 thick, Shandong Longshan culture, Liangchengzhen, Rizhao, Shandong. HIP KG 1972.4, % 3,P- 57

THE SYMBIOTIC RELATIONSHIP BETWEEN JADES WORKED IN SHANDONG DURING THE CLASSICAL

LONGSHAN PERIOD AND THOSE WORKED IN CENTRAL CHINA DURING THE ERLITOU PERIOD

The evidence thus far suggests that symbolic jades of Erlitou period date witnessed an earlier phase of develop ment presently identifiable with the Liangzhu and Longshan phases at Xuejiagang in Anhui province and later Longshan/Erlitou phases in the north at Taosi in Shanxi, and Cangshanmao and Shimao in Shaanxi. Other evidence corroborating that there existed a thriving jade-working are the Erlitou yue. All cutting edges of yue are half-moon in tradition of Longshan/Erlitou period date the exquisite shape, and blunt. Their backs are flat and their bodies thin. examples of jade gui and dao, and now possibly zhang (see a The yue have either one or two perforations for binding below) that can be associated with the culture of the handle. Although the quality of jade-working at Shimao Shandong Longshan. Ceramic finds from the Dawenkou does not match the high standards of Erlitou examples, ty through Shandong Longshan cultural phases indicate that pologically and stylistically the two jade-working centers, Shandong underwent a long period of development com Shimao and Erlitou, are comparable. If Shimao is dated to parable to the Songze and Liangzhu cultural phases of south the Longshan then the site finds are suggestive that the coastal China.98 The Dawenkou culture is estimated as ex Longshan Neolithic and Erlitou periods are overlapping tending from ca. 4500 to 2300 bce and the succeeding and related in cultural interest. The evidence from Shimao, phase of the Shandong Longshan from ca. 2400 to 2000 in Taosi in bce." The latter and chrono and also Cangshanmao Shaanxi and Shanxi, is phase synchronizes overlaps particularly rewarding in indicating that Erlitou period pre logically with the Erlitou culture. Although unlike Erlitou no decessors and contemporaries produced specialized sym early style bronzes have been found at Shandong bolic jade forms concentrating on zhang, yue, dao, and ge, Longshan sites, this northeastern culture was thoroughly fa just like Erlitou. Ge are not represented in lesser finds from miliar with the sophisticated art of working jades of formal or stone Cangshanmao yet from Taosi. Earlier, mostly but design. also jade prototypes, particularly dao and yue from Two site finds are pivotal for identifying the interrela Xuejiagang in Anhui and earlier Liangzhu period tombs, tionship of northeast and north inland cultures, of show early southern precedent for specialized tool and Shandong Longshan and Erlitou. One is the well-known weapon jades. Although the archaeological evidence is still discovery of symbolic jades at Liangchengzhen, Rizhao in piecemeal, on the basis of the above-mentioned data, the 1963100 and the other is the more recent discovery at Linqu, transformation from utilitarian to pure symbolic forms?as Zhufeng, also in Shandong.101 Two symbolic jades, imita they appear in select burials at Erlitou?exists during the tive of the tool shapes of ben and dao, reportedly made of Longshan, and, as will be discussed below, was a fait ac stone but possibly black jade (Fig. 18), were found along compli by the Shandong Longshan period that begins ca. with black pottery fragments at Liangchengzhen.102 The 2400 BCE. ben, properly called gui in its transformation as symbolic

77 a cm no. Fig. 19. Openwork upper part of jade pin, 23.0 long, burial 202, Shandong Longshan culture, Linqu, Zhufeng, Shandong. Kaogu 1990.7, pi. 1:1.

measures 18.0 cm cm and to blade, long, 4.5-4.9 wide, date the Shandong Longshan period. Ceramic types and 0.85-0.6 cm thick. The cutting edge is heavily beveled on style indicate that the burial finds date to the classical phase one side and is only slightly so on the reverse side. This when highly polished eggshell-thin, black vessels were per form of blade is the predecessor for the generalized jade gui fected in form and design.104 The several jades that were from Erlitou (Fig. IB2). The Liangchengzhen blade is in excavated from burial no. 202 approximate those from cised with semi-human masks on both ends of the handle. Liangchengzhen and also Erlitou. The jades include two The dao from Liangchengzhen, mislabeled chan in the yue, an elaborate openwork jade ornament (Fig. 19), a jade report, is much longer than the gui, measuring 48.7 cm stickpin decorated with miniscule images of the human long, 12.0-15.0 cm wide, and 0.5 cm thick (Fig. i8b). face, and a three-holed dao. The Linqu dao has not been Three equidistant holes bored from the front side align the published in photograph but is described as measuring a cm 6.0 cm cm as upper edge, where wooden handle may originally have 21.7-23.7 long, wide, and 0.7-0.8 thick, been hafted to the blade. An extra, smaller hole is out of black-green in color, and as trapezoidal in shape. From the alignment with the others, suggesting that this blade had an drawing of its form in the tomb plan, the dao, like the earlier phase of use or additional means of support. The Liangchengzhen blade, has three equidistant perforated piece was broken in several parts. In size and trapezoidal hafting holes and another hole along one short side. The shape the type is close to the undecorated versions exca yue are generally comparable to others excavated in vated at Erlitou, although it is not as rigorously symmetri Shandong, for example, from Dantucun, Wulianxian,105 cal in its proportions, which would indicate an earlier date. and to the plain rectangular type known at Erlitou (Fig. According to Liu Dunyuan, the black pottery fragments 1A2). The Longshan period yue from Dantucun curiously found with the gui, some of which are now in the Shanghai has serrated decor along its upper edges, as is familiar on 103 Museum, once belonged to the edge of a pan.ab The Erlitou and related Shimao jades. decorative motifs are similar in style to the incised lines of The dao, yue, and gui from Liangchengzhen and Linqu hooks and brackets framing the two semi-human masks on succeed an already well-developed tradition of jade-work the jade gui. This stylistic comparability justifies dating the ing represented in Dawenkou period finds.106 As symbolic jades to the classic phase of the Longshan period in jades, they represent a peak of sophistication and refine ca. bce. ment in is in most re Shandong, 2400-2000 both style and form that comparable Jade remains from Linqu, Zhufeng in Shandong also spects to the slightly later, or overlapping in date, jades

78 reverse Fig. 20. Jade gui, obverse and faces, n.d., Shandong Longshan cul ture. Palace Museum, Beijing.

21.0 cm Fig. 2i. A. Jade gui, tall, Shandong Longshan cul ture; B. Detail of decoration at the base. Hotung Collection.

79 Preference for certain details of decor and varied type of symbolic jades suggest that the two cultures, Erlitou and were Shandong Longshan, close and in regular contact. Both cultures, for example, favored the dao blade as a sign of wealth. The Liangchengzhen and Linqu types retain the trapezoidal shape with ends of uneven height that is typical of the earlier functional dao, as represented atXuejiagang and of dao from other Longshan and Erlitou cultural contexts in Shaanxi and Shanxi provinces. The dao blades of Erlitou 22. dao, 41.0 cm 11.7 cm on Fig. A.Jade long, wide, Shandong cultural date, the other hand, are more advanced stylis culture; b. Sketch of the mask with extended Longshan remaining tically in a formal that is limb on obverse and reverse ends of the blade. Paul favoring shape strictly symmetrical. Singer The from Erlitou also show a formal evolution be Collection. gui yond Shandong Longshan styles. On the basis of current ar chaeological data, gui do not appear to be a primary type of symbolic jade favored at Erlitou cultural sites. The few gui from Erlitou finds seem secondary and subordinate, sterile versions of a type already well established within the cultural tradition of Shandong Longshan. The gui from on Liangchengzhen, the contrary, show their origin in the adze and therefore a date stylistically earlier than that of the Erlitou examples. Recently reported but only partially documented by ar chaeological data are three zhang said to be from Shandong, illustrated and briefly discussed in recent articles by Liu Dunyuan, Wang Yongbo, and Yang Boda.108 The three zhang are said to come (i) from a stone tomb in Shangwanjia inWulian county, Shandong; (2) from Long shan remains at Dafanzhuang, Linyi county, Shandong; and (3) from Yueshi remains at Simatai, Haiyang county, was Shandong. None of these finds excavated. The zhang from Dafanzhuang and Simatai are said to derive from Longshan and Yueshi period remains. Presuming that these will eventually be published with proper provenance, their presence amidst theoretically Shandong Longshan and Yueshi contexts, nonetheless, would reinforce the symbi otic relationship between Shandong Longshan and Erlitou cultural periods and the evidence that the sophisticated zhang of Erlitou provenance and date grew out of a north ern Longshan period tradition of jade-working and still earlier agricultural tool of southern Hemudu and Liangzhu our affiliation. A large gap in knowledge about the trans mission and interaction of mid- through late third millen nium bce Liangzhu and Longshan traditions still remains. from the Erlitou period. The Shandong dao and gui are Several extant jade blades of the gui, dao, and yue type not clearly symbolic, tools and not utilitarian in function. from collections in and outside China further solidify the use Their thinness alone renders impractical. The sophisti interaction between the Shandong Longshan culture and cation of jade-working during the Shandong Longshan is jade-working traditions at Erlitou. Representative gui marked not only by the taste for formal jade blade shapes blades of Shandong Longshan style are in the Palace but by complex imagery calligraphically worked into the Museum in Beijing (Fig. 20) and in the Hotung Collection art as jade's surface. This sophisticated of working jade, (Fig. 21). Two representative dao belong to the Paul Singer noted, is matched by the quality of ceramics well known Collection in New York (Fig. 22) and the Arthur M. through the highly burnished blackware vessels of classic Sackler Collection inWashington, D.C. (Fig. 23).109 The were shape that designed for ritual use.107What is the rela gui blades are characterized by a long rectangular shape that tionship between the two cultures of Erlitou and Shandong is usually pierced by two hafting holes, unlike the un Longshan and their sophisticated jade-working traditions? earthed example from Liangchengzhen, and by two 80 ^^^^^^^^HHMf^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^HBS^-'^^^^Hcm ShandongLongshan ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^HK&l^^^l decor onone ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^H^HfcJ^^^Hof the Courtesyof theArthur ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^H^^^HSacklerGallery,

A

extensions surrounding the semi-human mask on the un earthed blade. Further geometric display of the type found on the Liangchengzhen sherds is typified in the symmetri cal display of hooks and curls above the mask motif. Significant to this interpretation is the pattern of framing vertical Unes and horizontal bands filled with a linked dia mond or brioche pattern. The layout of these geometric motifs interestingly correspond with those on the gui blades from the Beijing Palace Museum (Fig. 20) and Erlitou (Fig. IB2). The framing patterns of verticals and horizontals on the Beijing Palace Museum blade ismatched on the reverse by an incised image of a geometric version of the semi-human mask as found on the Liangchengzhen 1 gui (Fig. 8a). Here the eyes are framed by what has the ap pearance of openwork cloudscroll motifs, the same type that figure on the Linqu stickpin (Fig. 19), suggesting by its comparability that this gui is a classic jade of Shandong Longshan cultural expression. The significance of this im age, although worn and barely decipherable, is underscored by the crowning motif of bird with outstretched wings above the semi-human mask, a motif that recalls the painted images on Xuejiagang yue and dao. This combina tion of images, semi-human mask and bird, decorating gui is at the basis of a series of related, openwork jades and gui blades in collections mostly outside China.110 Like the imagery on the gui blades, that on the two dao from the Singer and Sackler collections further illustrates that the Shandong Longshan culture had a rich religious tradition of symbolic imagery. The Sackler dao is 47.8 cm long, 17.0 cm tall, and 0.9 cm thick (Fig. 23). The Singer dao is comparable in thickness but is otherwise completely symmetrical, measuring 48.0 cm long (Fig. 22). If exam ined closely, the incised image located midway on the short sides of the Sackler dao can be read as a semi-human mask B in profile (Fig. 23B). Figurally, this semi-human image is envisioned frontally, split into profile images overlapping beveled edges, one heavily and one slightly in imitation of the two short sides. If opened up, it would represent a the adze tool. The imagery on the Hotung Collection gui frontal image comparable to that represented on the Erlitou (Fig. 21) can be read as a worn version of the more repre jade handle (see Fig. 7) and Beijing Palace Museum gui. sentational motif depicted on the Liangchengzhen blade. Brackets framing the eye are composed of the same hooks The incised semi-human is reduced to a central set of eyes and curls as frame the gui blade masks. An added attribute surrounded by hook-and-curl motifs once related to the is the limb ending in a claw that can be pursued below the 81 head of this motif. The image of semi-human mask with vessel shapes of Shandong origin, such as the so-called gui, in on can be to similar and versions at Li limbs ending claws earlier Liangzhu jades immedi compared later Erlitou. ately comes to mind.111 This image of the semi-human also makes the suggestion that Erlitou may represent the mask with limb extension is repeated on the Singer dao occupation of the usurper from the east, Yi, of the Eastern (Fig. 22B) but with a total loss in meaning since the once Yi who conquered the Xia king, Tai Kang, and then ruled representational content has dissipated into an abstract, as king of Xia. Other historical data for the interchange be geometric pattern of extended hooks and curls. The shape tween the Xia and Eastern Yi of Shandong solidifies this of the dao has also lost all vibrancy in its completely, rigidly possibility. Fu Sinian with keen insight postulated long ago symmetrical, rectangular shape. The question arises, does that in early historic and proto-historic China two major the rigid symmetry of this piece qualify it as an Erlitou cultural groups, called Xia and Yi, existed in northeast and piece with Longshan style imagery? Given the comparabil central north China.114 In the Bamboo Annals there is fre ity of the semi-human mask and limb pattern with images quent reference to Yi tribes who come from the east to on other jades of attributed Shandong Longshan date, this make their submission to Xia kings, who are enfiefed by or rectangular blade belongs within the northeastern, Xia kings, who revolt against Xia kings; and of Xia ex Shandong Longshan tradition. peditions to the Yi in the east.115 The documentation of The Singer and Sackler dao blades and related Beijing historic records for the frequent social and political contact Palace Museum gui, however, can be brought to bear in between Xia and Eastern Yi complements and corroborates explaining the figurai imagery on the jade handle and also the interdependent relationship of Erlitou and Longshan semi-human mask decorating the recently published cultural and jade-working traditions. bronze plaque from burial no. 57 at Erlitou.112 As discussed PROVINCIAL ERLITOU CULTURAL JADES IN earlier, a facial image of the tiger alternates with semi-hu FUJIAN AND SICHUAN man facial masks on the Erlitou jade handle (see Figs, ibi; 7). Although neither the tiger motif, the totemic-like rep The appearance of numerous and large-scale ritual jades etition of motifs, nor handle type has been discovered asso amidst certain regional finds from outside the Yellow River ciated with jade works of art at Shandong Longshan sites, Valley in southwestern and southeastern China is tantaliz the image of the semi-human at Erlitou directly reflects the ing evidence for an extensive Erlitou cultural influence. subject portrayed in the jade art of the Shandong Longshan The richest finds derive from excavations and reports made culture. The splayed nostril, almond-shaped eye, and grin since 1931 in Guanghanxian in Sichuan province. Other ning mouth are all features generically similar to subjects of finds come from the far southern provinces of Fujian, and Shandong Longshan style. The same semi-human image, just recently reported, from Yuenan, modern-day Viet on highly abstracted, appears the recently excavated bronze nam,116 and Dawan, Hong Kong.117 plaque with turquoise inlay uncovered from burial no. 57. The stone and jade types from Tanshishan in Fujian be Circular eyes are framed by an upper display of geometrically long to two phases of the Tanshishan culture.118 The first abstract bird feathers and by a lower extension representing phase levels (xia and zhong) is stated to be prehistoric in abstractly simplified limbs with flanking claws. Since jade date and comparable to the Liangzhu culture, and the sec working during the Shandong Longshan period was the ond or latest phase (level shang) is defined as Bronze Age in predominant art, with developed forms of imagery and date. All tools and weapons were fabricated out of stone standardized types of symbolic jades, there were evidently and are comparable to southern cultural types from Jiangxi, various ways that Shandong Longshan influences could Guangdong, and the Taihu Valley of Bronze Age date. The affect Erlitou cultural expression. The impact of Shandong exception cited amidst these tools and weapons, however, Longshan upon Erlitou is documented by the presence of is the blade with sharp edge called ge from Meili in gui but also dao at Erlitou. Although these gui are not a Zhangpu. The shape of this blade is generically comparable major jade of Erlitou cultural production, the dao is shared to the bronze version of zhang with rounded blade head and perfected by both cultures. Religious imagery in the from Erlitou. The diagnostic trait identifying this shape as form of the semi-human mask with feathered headdress a zhang and not ge of Erlitou cultural type is the jagged sil and limb with claw extension also now appears to be trans houette of the haft and the flared outline of the blade. mitted from earlier Liangzhu art and is a preferred form of Neither dimensions nor a detailed description of this find is representation in both Shandong Longshan and Erlitou art. provided. Other tool and weapon finds, including a large The suggested symbiotic relationship between the two number of ge, are also generically comparable to northern cultures in Shandong and Henan appears in other ways. types of Bronze Age date. The appearance of well-estab The independent yet shared interest in symbolic jade lished symbolic jades of Erlitou stylistic inspiration in the far a an blades, such as dao and yue, and probably zhang, as form south?possibly import from the north?may not be sur of art is one. As illustrated by Li Boqian, plentiful ceramic prising given the longstanding interaction that is suggested data also demonstrate that the direction of influence was of elsewhere in the south, such as at Guanghanxian in central, earlier Shandong Longshan styles on Erlitou.113 Certain north Sichuan. 82 Fig. 24. Jade fu, zhang, and collared disk, . VS - ^^Hl*?jHSw^%? f ' Taipingchangculture/Late ^ * Bronze ^A|^'| " mE$ti^?S^lt^K??tlh?' ?L ' & Neolithic/Early Age,Hanzhou, ^P>^L' * ?J^B^^?l^^?^BQKv^^f ????k Sichuan. D. S. Dye, "Some Ancient m??/r JMBSM^S^k.-!*^^^9 *&7?hHL? ^KwK and Curves in Circles, Squares, Angles Mt '** Earthand inStone inSzechwan, Mr I <^t\W* ' ^EtU^^K?*^ ^Bnl"4 China," -^ Uf*\ %&f ^Bfta^^l^?B^*^B^J?:

The earliest reports of jade finds from Sichuan appear in basis of external comparisons of ritual bronzes.123 The dat in in English 1931 and 1933 the Journal of the West China ing of these three pits, the earlier one from Taipingchang Border Region Research Society (Xieda xuebao), published by and later two from Sanxingdui, covering the eras of West Union in the China University Chengdu, Sichuan. In Longshan to Erlitou and Yinxu period I, is reflected stylis D. a 1931 S. Dye, Professor of Geology, gave brief descrip tically in the excavated stone and jade implements. tion of six different types of Neolithic stone and jade im According to Cheng's assessments, the "ceremonial" im an plements recovered from the bottom of irrigation ditch plements found in the pit included 3 cong, several yuanac a In over of "progressive farmer."119 1934 David C. Graham up rings with collar, 20 bi-disks, more than 2 zhang, sev more dated Dye's report by describing in detail the prove eral long fu/yue, plus ornaments in the form of turquoise nance a of the find.120 He provided map and illustrated the flakes, presumably for inlay, and 15 jade beads.124 The out were rectangular pit of which the finds allegedly taken mixed cultural character of these finds points to a date at on a ex the farm of Mr. Yen. He also carried out informal the end of Liangzhu/Longshan on the one hand and to the at as near cavations this site, known Taipingchang, Han dominance of Erlitou period influences on the other. The zhou and carefully documented his excavations in a pre bi, mostly made out of local gray sandstone but with some notes are liminary report, adding illustrations and about made out of jade, extraordinary for their unprecedented numerous stratigraphy and pottery shards. In 1949 Cheng size, compared to standard Liangzhu period bi.125 The Te-k'un (Zheng Dekun) carefully reviewed Graham's largest one has a diameter of 70.0 cm, a hole diameter of work and distinguished artifacts found in the pit from those 18.0 cm, and a thickness of 7.0 cm.126 The three cong are, recovered from the cultural strata. In updating this earlier on the other hand, very small: one is 5.5 cm tall and 7.5 cm research at Taipingchang, Cheng was able to conclude that, wide with a hole 6.5 cm in diameter; and the another is on the basis of polished stone tools and gray-slipped pot even smaller, measuring 3.0 cm tall and 5.7 cm wide. The tery, the ceremonial pit and dwelling site date to the last third cong, recently published,127 measures 11.0 cm tall and cm phase of prehistoric Sichuan, which is equivalent to the 9.0 in diameter. The latter two are plain and the former or "Black Pottery Culture in North and East China,"121 has widely spaced grooves and a simplified circle on each Shandong Longshan. prism face. These motifs are feeble simplifications of the on This pit find the property of Mr. Yen has been chal standardized mask that was masterfully worked in intricate two on lenged recently by extremely rich finds from other detail and levels of carving cong of the classic Liangzhu large pits opened in 1982 at Sanxingdui, Guanghanxian.122 period.128 The degenerate decor and small size of these two are excavators to at The latter pits dated by the PHI cong testify to the end of the classic Liangzhu, probably ca. on to Sanxingdui the basis of internal evidence and the pre 2400 BCE, when symbolic jades begin to dominate as the or or beginning Late Shang phase Yinxu period I on the preferred burial implement.

83 The mixed cultural character of the Taipingchang finds type blades and measure 24.8 and 26.4 cm, respectively. is also marked by the presence of long fu/yue129 represent The third type continues to reflect Erlitou styles in its pa ing late Liangzhu period influences130 and zhang represent per-thinness, trumpet-shaped blade point, and haft heavily ing classic Erlitou period influences (see Fig. 24). One decorated with raised lines and profile serrations with two zhang measures 39.4 cm long and 15.5 cm wide, and the pairs of notched crossbars (Fig. 25 a). The latter form, how other 36.5 cm long and 12.4 cm wide. They are extremely ever, has lost considerable balance in design and decor. The thin, varying from 0.46 to 0.49 cm thick. As with Erlitou fourth type is distinguished by a half-moon-shaped blade excavated zhang, these examples have an exquisite silhou point and unemphatic haft protrusions so that in shape it is ette comparable to a trumpet and a haft carefully outlined long and slightly bowed without a prominent hilt (Fig. in notches and paired protrusions also worked into a rhyth 25B). Degeneration of form and aesthetics of classic Erlitou mically, jagged outline. Three other zhang-blades, presum style is marked in the taste for disruption of standardized ably from the same or a related find at Taipingchang, are form and design. generically similar in style. One is striking in length, one of Manneristic tendencies are especially marked on zhang mea 2 the longest so far documented, at 56.1 cm. Another blades from pit at Sanxingdui. Fifteen examples, some of sures 41.4 cm long; and still one other, broken at the haft, which are jade and some stone, were excavated and sepa is estimated to measure 45 cm long.131 All three are simi rated into two groups called yazhang and bianzhang by the larly thin, averaging 0.4 to 0.5 cm thick. Artistically speak excavators. Their differences derive from the shape of the ing, the sophisticated working of these jades is evident in blade point: the former is forked and the latter is drawn at the care for form and small-scale geometric detail. The an oblique angle;137 both can be defined as zhang. The lat geometric decor of parallel lines and jagged silhouette with ter is a simplification of the Erlitou zhang that became stan pairs of protruding paper-thin crossbars on the haft of the dard during later Shang and Zhou eras (see Fig. 16). The 2 longest example is somewhat hardened, suggesting that fifteen examples from pit show eccentric diversion from a these jades come at the end rather than the beginning of classic Erlitou period style in variety of other ways. Erlitou period influences. Variations include the taste for dramatic length without Other artifacts showing influence of Erlitou period jade balance, eccentric forking of the blade point, disinterest in working found in Taipingchang excavations include a re serrated handle designs and, most importantly, the interest cut dao-blade with geometric designs. Although the frag in portraying figurai motifs. The three green-to-gray jade ment is described as a "borer,"132 its decor of lozenges and blades (called bianzhang in the report) are surprising for as groups of parallel lines characterizes it a cut-down dao of their incised imagery and emblematic signs.138 Four bands Erlitou inspiration. The medium exploited is identified as of imagery decorate the upper and lower parts of the sim hard sandstone,133 which indicates it was locally worked. plified obliquely shaped blade. The rows of human figures Another curious and so far unpublished dao from San who are dressed in local costume alternate with bands fea as sun xingdui is characterized by the typical Erlitou style geo turing what the excavators describe mountain and as metric decor and design. What marks the latter jade typ motifs. Small-scale, profile motifs of zhang with forked em ical of southwestern taste, however, is the preference for blades are positioned to the left and right sides of the radically large size, double or triple that of classical Erlitou blematic mountain motif. Clearly the zhang-blade was re to users styles.134 Other evidence for the influence of Erlitou style ligiously significant of these blades, who have been are the remains of numerous turquoise flakes that were de identified as those of the early Shu culture.139 No such im at at or context. signed for inlay.135 The taste for turquoise inlay Erlitou is agery is known Erlitou in any early Shang 2 at same represented by several finds, and in the case of jades, by the The ge-blades from pit Sanxingdui show the on popular stud inlay zhang and yue. local stylistic tendencies that characterize the zhang. They The jades from the pits at Sanxingdui, in contrast to are eccentrically thin, flat, and long, with the longest mea cm as those represented in finds at Taipingchang, show stylistic suring 59.4 and the thinnest, 0.3 cm,140 if in mockery tendencies that have more to do with local manufacture of the Erlitou prototype that was large in scale and power than with current dynastic or cultural styles of central north ful in silhouette. The original Erlitou period interest in China. They go far beyond stylistic tendencies of Erlitou in faceting the two long edges of the blade and decorating the are reflecting idiosyncracies typifying local taste and manufac handle with geometric decor survive but manneristi an 2 a ture, with pit 1 jades representing earlier and pit later cally transformed. The ultimate transformation of proto 1 at can in two phase of production. Amidst the extant jades from pit typical Erlitou period ideals be read the pub as two a Sanxingdui, zhang and ge blades stand out.136 It is not lished ge described zhang. The ge have handle were a specified how many zhang excavated from pit 1, al decor including pairs of raised lines and notched silhou on are though the excavators distinguish four types based the ette that mimick those of the zhang. Decor and shape a form of the blade head. The first two types are ge in shape manipulated for the purpose of creating uniquely local are and need not be classified as zhang despite their syncretic expression. These regional eccentricities made promi a eccen use of the zhang-style haft. The third and fourth are zhang nent by the small-scale image of zhang-blade with

84 ' Di?, M

1 H

A" Il bad LJ B LiN f l?E3ET

cm no. Fig. 25. A. Drawing of two jade zhang, 38.0 and 24.8 long, Sanxingdui culture/Early Bronze Age, pit 1, b. cm Sanxingdui. Wen Wu 1987.10, fig. 12:3-4, P- 8; Drawing of two stone zhang, 50.0 and 68.2 long, Sanxingdui no. culture/Early Bronze Age, pit 2, Sanxingdui. WW 1989.5, fig. 34:8-9, p. 16.

trie notching at regular intervals rendered on the face of the THE DISAPPEARANCE OF ERLITOU STYLE JADES ge-blade. The latter ge, furthermore, has a blade tip perfo DURING THE SHANG rated with the design of a profile bird. The other blade tip The Erlitou period was the era par excellence for creating has been compared to the gaping mouth of a fish.141 large-scale, classically designed symbolic jades. Typical The diversion from classic forms and decor of the north types, such as the zhang, long and short dao, and faceted ern Erlitou style at Sanxingdui in Sichuan emphasizes the yue gradually wane asmajor art forms in favor of primarily an independence of this southern tradition during era bronze vessels during the successive Shang dynasty. If which, although extending beyond Xia into Early Shang, Erlitou type jades appear in later Shang contexts they show once is clearly indebted to mainstream Xia cultural stim a loss of creativity and a significant diminution in size. The a uli.142 What appears to be widespread influence of the latter are key attributes in distinguishing Erlitou from later Erlitou culture is now well documented not only in Shang-style jade blades. Sichuan, where remains have been adequately published, An extremely long zhang-blade was published in 1966 as a but by piecemeal finds throughout the south, in Fujian, work from Yangzhuangcun, Nanxiao, Zhengzhou.143 In a Hong Kong, and now Vietnam. What this evidence sug 1958 farmer while cultivating the ground uncovered the gests is that Erlitou influence was clearly more than a loose blade about 1.5 meter below the surface. The blade mea was sures 66.0 cm cm cm a stan tribal expression. Rather, Erlitou, like the later Shang, long, 13.0 wide, and 0.4 thick, more a likely strong administrative center with centralized dard large scale for Erlitou period zhang. The shape of the as control ismeasured through the high standard in artistic blade also appears to be completely Erlitou in style. The an production both locally and provincially throughout head is crescentic and the haft is characterized by sets of cient China. longer ridges that frame shorter sets and theoretically con tinue as raised ridges across the width of the haft. Despite

85 the location of the find in Zhengzhou municipality the straight, rectangular shapes with diagonal blade and no han as or zhang is completely Erlitou in style and form. Since dle, is commonly represented in stone pottery in Zhengzhou is not outside the realm of Erlitou cultural Shang and Western Zhou tombs. Dao blades in the Erlitou influences during the Erlitou period, this zhang likely dates tradition, on the other hand, seem to disappear altogether are as to a time within the Erlitou cultural phase. unless certain of the crude form of zhang identified Large-scale jades of Erlitou period design, like the zhang reworked dao.151 and dao, are rarely represented amidst burial goods or resi Gui-blades, although not common amidst Erlitou finds, dential remains at Erligang and later Shang period sites. The are popular, although uninteresting and at small scale after jade ge is the exception. Several large-scale, long ge have the Erlitou period. By Shang times their points take a vari been excavated at Zhengzhou sites. One example, dark ety of shapes, sometimes round and sometimes in the form green in color, measuring 57.0 cm long and 8.0 cm wide, of an equilateral triangle. Numbers of jade gui with round comes from burial no. 7 at Baijiazhuang.144 Another exam blades were excavated from Shang burial M5 at Anyang.152 no. ple from burial 3 at Panlongcheng in Huangpi county, The latter type is represented in burials from the Western measures a cm In terms of at Hubei, record 93.0 long.145 Sector cemetery Xiaotun and the former in laterWestern exca as length Erligang period jade ge are longer than those and Eastern Zhou burials, such at Fengxixiang, Changan vated from Erlitou. Their prototype nonetheless is clearly in Shaanxi, and Tongshan in Gansu.153 By Eastern Zhou a Erlitou. Stylistically Erligang ge differ in favoring either times the gui blade with symmetrical point is standardized, strong median line down the longitudinal axis of the blade, as illustrated by additional examples from the Chunqiu site or a blade head that points slightly downward, four blade in Wuxian, Jiangsu.154 It is also significant that earlier facets (rather than the two of Erlitou design) along their Liangzhu-type cong and bi occasionally appear in Shang as length.146 Since the ge appeared later than the yue in Late and Western Zhou burials,155 but, with certain of the are Neolithic jade assemblages (e.g., Shimao), the continued Erlitou symbolic jades, they very small and crude, with as a art or production of ge in large scale and major jade form out decor, polish, any of their former classic expression. in early Shang times is understandable. CONCLUSIONS Yue-ax types simulating those of Erlitou design also con none tinue to be produced during the Shang period,147 but Zhang, dao, yue, and ge?specialized jade types and signi possesses the classical balance and sophistication of design fies of elitist control?stand out amidst Erlitou-associated that characterize their predecessor. The so-called handle Longshan remains and collections in and outside China. are shaped jade, initially seen at Erlitou, on the other hand, is These symbols of formalized tool and weapon shapes area very common in Shang and Western Zhou burials,148 al distributed in the immediate of Erlitou cultural finds in as though it too is invariably undecorated except for the rigid Shaanxi, Shanxi, and Henan, but well in distant repetition of the plume motif that was standardized during Guanghanxian in Sichuan and elsewhere in south China the Erlitou period. and outside China in today's Vietnam. The Erlitou culture as elite art was a whose influence was Large-scale zhang and dao drop off markedly evidently major power perva works during the Shang. The only type of zhang that ap sive and culturally dominant. As suggested by shared jade in in or are de works art and pears excavated Shang remains, jade stone, (and ceramic) of cross-cultural influences of as generate examples of their prototype. Apparently 183 stone Shandong Longshan upon Erlitou, which if argued a zhang were unearthed from 41 of 900 small burials in the chronologically fits Xia identification, it is evident that this an Western Sector cemetery of Yinxu. The only examples Xia settlement could be culturally challenged by equally tun are center Yi. the evi published there and from Xiao burials very crude, sophisticated of power, the Eastern Given all less than 15 cm long, and all with blades that have one dence for the symbiotic relationship between jade-working centers blunt and one oblique end.149 They look as if they had been of Erlitou and Shandong Longshan, for the wide on cut down the middle of an original long dao blade. spread influence of Erlitou certain southern sites, and Zhang reported from Western and Eastern Zhou sites, for the classic forms of developed jade symbols of wealth, a such as Shangkangcun in Fufeng, Shaanxi; Dasikongcun in the Erlitou culture produced major artistic tradition that are a con Anyang; and Dongjiao in Luoyang, Henan, similarly strongly suggests power different from the Shang yet we are at crude and shapeless150 and bear no comparison to their tiguous with the Longshan. If to accept the finds nu as as original creative and classic form of Erlitou date. The Erlitou and related sites Xia in date, which noted is merous zhang from collections, as represented by Figure corroborated by stratigraphical and chronological data, or as 16, probably date to Late Xia Early Shang times, when then the tradition of working large-scale jade blades sym art can as classic standards were relinquished. The latter type, repre bolic forms be recognized uniquely and charac sented by the example in the Hotung Collection (Fig. 21) teristically Xia. exhibits degeneration of Erlitou style attributes in its lack of proportion and balance of design. The handle and blade head lose clarity of shape and ultimately dissipate into 86 was an or Notes suggests that jade, like the cowrie and turquoise, import ob ject acquired through trade (p. 223). 7. shehui Erlitou on Zhongguo kexueyuan kaogu yanjiusuo gongzuodui, This article was first presented at the International Symposium Xia "1987-nian Yanshi Erlitou yizhi mucang fajue jianbao," KG Culture, University of California, Los Angeles, May 23-25, 1990. 1992.4:194-303. were said to have been collected 1. on Fragmented jade(s?) See for example the article by David Nivison the "Astronomical with a bronze and from area V at Erlitou in 1987 re along ding jia (KG Evidence for the Bamboo Annals' Chronicle of Early Xia," and 1991.12:1138). sponses to it assembled in Early China 15 (1990) 187?196. Recent publica 8. For utilitarian stone fu see Yuan Te-hsin (ed.), Zhongguo lishi wenwu tions collating past articles concerning the Xia include Xu Zhongxu (Taipei, 1977), pi. 38: Cheng Te-k'un, Archaeology in China, I: Prehistoric (ed.), Xia wenhua lunwen xuanji (Zhongzhou Guji Publishers, 1985); Tian China (Cambridge, 1966), pis. X:i, 3; XL3; XIL2-3; Chang, Archaeology Changwu (ed.), Xua Xia wenming (Beijing University Publishers, 1987); in Ancient China, 174, p. 212; 141, p. 178; 132, p. 167. For Li Xia tansuo Renmin and fig. fig. fig. Min, Shangshi (Henan Publishers, 1985); see examples from Liangzhu tombs Nanjing bowuyuan, "Jiangsu Chou Hung-hsiang, Xiashi Xia wenhua yanjiu shumu (Hong Kong, 1990). Sidun de KG 1981.3, 6:1?2, 196; 2. a Wujing yizhi shijue," fig. p. Jingzhou For review of the evidence favoring a Xia identification rather dichu bowuguan, "Zhongxiang Liuhe yizhi," Jianghan kaogu 1987.2, than Early Shang for Erlitou cultural remains see K.-C. Chang, The 12:1-8, p. 13; 17:1-3, p. 20. For utilitarian stone fu from Dawenkou Ancient ed. figs. Archaeology of China, 4th (Yale Uniersity Press, 1986), pp. see Shandong Provincial Cultural Bureau and Jinan Municipal Museum 307-316; and Louisa Huber, "The Bo Capital and Questions Concer (eds.), Dawenkou (Beijing, 1974), fig. 28, p. 37. ning Xia and Early Shang," Early China 13(1988)146?77. In identifying 9. Lin Huadong, Hemudu wenhua chutan (Hangzhou: Zhejiang renmin Erlitou as Xia, Huber comprehensively considers data from archaeolog chubanshe, 1992), pp. 119-126. ical, cultural, and historical viewpoints. In addition to relevant data from 10. Lin, Hemudu wenhua, 5-6; Ancient China, source ma fig. Chang, Archaeology of stratigraphy and available C14 dates, she assembles historical fig. 176, p. 213. terial alongside archaeological data to demonstrate that Xia and early 11. See note 25. were Shang not only compatible but interconnected culturally, what she 12. Lin, Hemudu wenhua, 5-6:3-4, 123; 2-6:2-3, 47; 17 as see fig. p. fig. P- P- describes "comparatively homogeneous" (p. 55, also pp. 55-62). 13. For refined of what to be made out of same examples appear yue-axes Du Zhengsheng brings together the sort of convincing interdisci jade or stone, however labeled fu, from tombs see data with the forceful conclusion that Erlitou is the Late Xia Liangzhu Shanghaishi plinary dy wenwu baoguan weiyuanhui, "Shanghai Qingpu Fuquanshan Liangzhu nastic capital of Jie at Zhenxun, and Yanshi Shangcheng is the Early wenhua mudi," Wen Wu 1986.10, 5-12, 8; at (hereafter WW) figs. p. Shang dynastic capital of Tang Bo ("Xiadai kaogu ji qi guojia fazhan 31:1-4, p. 10; 50:1?3, p. 16. All these yue are distinguished from utili de tansuo," Kaogu (hereafter cited KG) 1991.1:43-56). At present this tarian prototypes by their thinner fabric and refined, polished surface. historical-cultural continuity and the evidence from stratigraphie and For other refined, similar called of stone and a implements chan-spades C14 dating strongly argue that Erlitou is Late Xia culture and that from another Late Neolithic culture, see Dawenkou, 27, 36, see jade fig. p. Yanshi Shangcheng is Early Shang. For C14 dates from Erlitou Qiu pis. 23-24. The confusion as to whether an ax should be labeled yue or Shihua et al., "You guan suowei 'Xia wenhua' de tan shisi(Ci4)-niandai fu is evident in the tendency to use both when describing the same im ceding de chubu gaogao," KG 1983.10:923-928. from tombs. The axes that are described in note 7, for See for KG bronze Wen plement Liangzhu 3. example 1976.4, fig. 4 (PHI jue), p. 260; are example, labeled fu. On the other hand, in the report of finds from Fong (ed.), The Great Bronze Age of China (New York: Metropolitan Fanshan, all axes are labeled 1 Yuhang, Zhejiang yue (WW 1988.1, pp. Museum of Art, 1980), color pi. (jue). The chance recovery of a 14-16, area pi. 1:2). bronze ding-tripod and jia from V at Erlitou indicates that the ding 14. K.-C. used similar criteria in the in his as a Chang describing yue cast bronze vessel type was also known during the Erlitou period Archaeology of Ancient China, pp. 165?167. Here Late Neolithic axes, 1991.12, 1 (KG fig. left, p. 1138). are as called yue, defined "ritualized," due to their specialized qualities 4. For a discussion in on from these Late Neolithic English jades identifiable either their refined matrix and or and Erlitou cultures see the through design through (Hongshan, Liangzhu, Shandong-Longshan) use an or the of inscription incised clan symbol. exhibition brochure by E. Childs-Johnson, Ritual and Power: Jades of 15. See Xia Nai, "Shanghai yuqi de fenlei, dingming he yongtu," KG Ancient China (New York: China Institute in America, 1988) and also 1983.5:455-467 and later English version, "The Classificiation, "Dragons, Masks, Axes, and Blades from Four Newly Documented Nomenclature and Usage of Shang Dynasty Jades," in K.-C. Chang Jade-Working Cultures of Ancient China," Orientations (April (ed.), Studies of Shang Archaeology (New Haven: Yale University Press, I988):30-4I. 1986), pp. 185?207 and esp. pp. 223?229. Xia stated that his examina 5. For excavated from Erlitou see shehui jades Zhongguo kexueyuan tion is based on an "to treat the excavated Erlitou "Yenshi Erlitou sanbachu fa archaeological methodology, kaogu yanjiusuo gongzuodui, yizhi artifacts as the basis of research and then seek to relate this material to jue jianbao," KG 1975.5:302-309, 294 (where jades date to PHI); written records" There is, however, no of "Yenshi Erlitou xin faxian de he KG (p. 207). analysis archaeolog yizhi tongqi yuqi," 1976.4:259?263 nor one ical data of how blade type is distinguished from another in "Erlitou chutu de he KG 1978.4:270 (PHI); yizhi tongqi yuqi," (no terms of name or function of the in a context. Xia still re Henan Yenzhi Erlitou KG object Shang date); "1980-nian qiu yizhi fajue jianbao," on lies the traditional "old approach" by seeking explanation in later lit 1983.3:199-205, 219 (PHI); "1981-nian Henan Yanshi Erlitou mucang erary ritual names, as in his erratic treatment of various forms of the yue fajuejianbao," KG 1984.1:37-40 (PII, PIV); "Henan Erlitou erhao ax. For on p. 213 he calls the ax with hole and KG Yanshi example, large perforated gongdian yizhi," 1983.3:206-216 (PHI); "1982-nian qiu a as faceted cutting edge from Erlitou biqi instead of a yue used in the Erlitou yizhi jiuchu fajue jian-bao," KG 1985.12:1085-1094, 1108 site report (p. 213). Xia Nai states: "The preliminary report terms them (PPII?III); "1987-nian Yanshi Erlitou yizhi mucang fajuejianbao," KG 'yueh' [Xia fig. 40:2]. They should be called 'pi-chi' [biqiac*] instead" (p. 1992.4:294-303 (PHI). no 213). There is explanation for this statement. Xia is following Wu 6. KG 1975.5. From the earliest reports on discoveries and excava the author of (An Illustrated Ancient tions at Erlitou there is reference to of Dacheng, Qing Guyutukao Study of piecemeal fragments jade (e.g., as Jades, p. 59), who in 1889 distinguished yue a perforated flat fu and the CASS, Excavation Team, "1959-nian Henan Yanshi Erlitou Luoyang flat ax with side serrations as as we learn later on 226. is not a qi, p. Qi shijie jianbao," KG 1961.2, fig. 3:26, p. 84, white jade piece). Due to known in Shang oracle bone inscriptions and is known in only one the scarcity of jade amidst the earliest excavated finds of 1957 at Erlitou Zhou bronze inscriptional example, where if the graph indeed does re ("Henan Yanshi Erlitou KG 1965.5:215-224, yizhi fajuejianbao," pi. fer to then it is a reference not to a but to a name 10 a qi weapon (see e.g. 5:15 [ornament], [fragment of cong-tube],) author Fang Yousheng Zhou Fugao, Jinwengulin, Hong Kong, 1974, no. 1616). Xia classifies the

87 ax weapon according to its blade type, in the category he labels "tao 38. Dai Yingxin, "Shaanxi Shenmuxian shimao Longshan wenhua shaped end-edge implements, frequently occurring in large sizes" (p. yizhi diaocha," KG 1977.3:155, 154-712; "Shenmu Shimao Longshan 224). wenhua yuqi," Kaogu yu wenwu (hereafter KGWW) 1988.6:244-249; 16. Ting Fu-pao, Shuo-wen chieh-tzu ku-lin (Taipei: Shang-wu yin-shu Hayashi Minao, "Chug?k? kodai ch?kei gyokkotsu hakukei gyokki," kuan reprint, 1959), chin-pu (jinbuae) pp. 6332?6334; kepu (gebua*), pp. Tohogakuho 54 (March, 1982)17?9, 1-45. 5698-5699; Paul L.-M. "On the of the Pu Shou 39. Lin, Hemudu wenhua, 155. Serruys, " System p. was [bushoua?] in the Shou-wen chieh-tzu, Chong-yang yen-chiu yuan li-shih 40. Ibid., p. 159. Lin states that this chan-spade probably used to no. sow yu-yen yen-chiu-so chi-kan 56 (1985), 452, p. 734. In the Shuowen yue crops. is also written with the metal radical6 (see pp. 719.3). In his commentary 41. Dai, "Shenmu Shimao," p. 245. on the latter graph Zheng Xuan notes that yue is also written without 42. See Wang Hongming, "Shandong sheng Haiyangxian shiqian a the metal radical,"1 which helps clarify that the identification of yue here yizhi diaocha," KG 1985.12 fig. 6:3, p. 1062 for zhang found amidst ax is the ax as used in oracle bone inscriptions, where it refers to "to with Yueshi period remains at Simatai, Haiyangxian, Shandong. Another to on at the yue11" in ritual slaughters. zhang is reported by Wang Yongbo be exhibit the Linyi to use 17. In oracle bone incriptions yue is used verbally11 and refers of Municipal Museum ("Shilun Guanghan Sanxingdui faxian de yurui," the yue for sacrifices; see Sun Yirang in Qiwen zhulie, xia vol., p. 19 and Jinian Sanxingdui kaogu faxian liushi zhounian, Ji BaShu wenhua guoji Li Xiaoding, Jiagu wenzijishi (Taipei, 1965), pp. 0459, 0479 for examples xueshu taolun hui lunwen, Guanghan, Sichuan, PRC, April 1?6, 1992, or of the graph. Sun refers to the Shuowen definition in identifying the manuscriptp. 121, fig. 4:6-7; "Yazhangxinjie," KGWW1988.1, fig. 1:6, out two were acle bone term of sacrifice and the graph of the bronze ritual implement p. 38). Liu Dunyuan points that the above picked up see illustrated in Zhou, Jinwengulin, pp. 214-215, 545. Also Chang amidst Longshan and Yueshi remains ("Yazhang yu Shangdai tongge," cover Tsung-tung, Der Kult der Shang Dynastie in Spiegel der Orakelin-Schriften Wenwu Tiandi 1994.3:11?12, inside for illustrations). Color pho (Wiesbaden: Otto Harrassowitz, 1970), p. 135. tographs of the latter two zhang picked up amidst Longshan and Yueshi a are 18. Comparative and representative bronze versions of the yue of period remains, plus third from Shandong illustrated without cita Shang date include those from M5 at Anyang (Yinxu Fuhao mu, 1980, pi. tion by Yang Boda in "Jade Zhang in the Collection of the Palace 13) and the royal burial at Sufutun, Yiduxian, Shandong (WW 1972.8, Museum, Beijing," Orientations (February 1995):53?60. fig. 28, p. 29). 43. Zhouli, "Dianrui," "Kaogongji" chapters, Sibu Congkan ed. axes are 19. Representative from the Liangzhu and Longshan cultures (Shanghai: Shangwu publishers, 1989), 5/37a, 35b, 36a, and i2/4b, 3b, are an illustrated in Childs-Johnson, Ritual and Power: Jades of Ancient China, where yazhang, zhang, dazhang, zhongzhang cited. For analysis to figs. 25-26, 53, pp. 18, 20; "Four Jade-Working Cultures," figs. 32-34, of the Zhouli and other Warring States and Han literary references see p. 40. Also see Mou Yongkang et al., Liangzhu wenhua yuqi (Beijing, zhang Zhou Nanquan, "Zhongguo gudai yu, shi zhang yanjiu," 1989), pis. 232-239. KGWW 1993.5:58-65; Feng Hanyi and Tong Enzheng, "Ji Guanghan 20. cm KG 1978.4:270. For the jade yue with blade 7.7 long and 7.6 chutu de yushiqi," KG 1979.2:32-33; Xia Nai, "Shang Dynasty Jades," cm tall, excavated in 1982, see KG 1985.12, fig. 8:4, p. 1092. pp. 217-220; Dai, "Shenmu Shimao," pp. 244-246; Wang, "Shilun 21. KG 1983.3:203. The yue measures 9.2 cm tall, 8.0-9.24 cm wide. Guanghan Sanxingdui," pp. 3-14; Zheng Guang, "Cong Erlitou yizhi 22. a a Xia Nai calls this weapon biqi, bi-shaped qi ("Shang Dynasty de yazhang tanqi," Wenwu Tiandi 1994.3:15. uses cata in Jades," p. 227). He qi instead of yue in following the Qing 44. See Zhou's discussion of this point "Gudai yushi zhang," p. 60; note loguer of jades, Wu Dacheng (see 15). In the excavation report yue Feng and Tong, "Ji Guanghan," pp. 32-33; and Zheng, "Cong Erlitou," is correctly used (KG 1976.4:262; fig. 6:3-4). P- 15 as as 23. KG 1976.4, fig. 6:3-4, p. 262. 45. Xia Nai relates that, early the Han, ritual jades included the 24. KG 1984.1, fig. 5:2, p. 38. so-called "Six Auspicious Jades" ("Shang Dynasty Jades," figs. 36-37, are on a 25. One of the major characteristics of Shandong Longshan (e.g. pp. 208-209). These so-called auspicious jades incised Han stele in Chengziyai) and Xia sites (e.g. Wangchengang and Dong Xiafeng) is the that is, for example, recorded the Qing compendium by Hong Shi, construction of a city wall, presumbably for defensive purposes; KG Li Xu, Hongshi Huimuchai Congshu ed. (1972), vol. 5, pp. 3-6. Wang are 1980.2:97?107; Chang, Archaeology of Ancient China, p. 250; Anne Yongbo also illustrates drawings of the so-called auspicous jades that as on Underhill, "Variation in Settlements during the Longshan Period of illustrated engravings three Han stele (see his "Shilun Guanghan Northern China," Asian Perspectives 33:2 (Fall 1994): 197?228, esp. table Sanxingdui," fig. 5:4, 8?9). 1, p. 203. It is in connection with this desire to defend that weapons such 46. Shuowen, yu-pu,?l pp. 141?142. In this ex as the ge originated. 47. Zhou Nanquan's article type of degenerate zhang from 26. Shima Kunio, Inkyo bokuji kenkyu (Hirosaki, 1958), p. 327.1-2. cavated sites is illustrated in fig. 5, p. 61. Figure 5:10 (a jade from the 27. Li Hsiao-ting, Chia-ku-wen-tz'u chi-shih Chung-yang yen-chiu Liangzhu site of Zhanglingshan inWuxian, Zhejiang) should be deleted is to the of yuan lishi-yu-yen yen-chiu-so chuan-k'an chih wu-shih (Nankang, from his chart since it formally unrelated evolution this jade seems to J965), p. 3753- For the Han definition see Serruys, "On the System of type. At the end of his article Zhou misunderstand the stylistic the Pu Shou," no. 451, pp. 733-734. evolution of the zhang from Neolithic through Han times by deciding as 28. Li, Chia-ku-wen, pp. 3777, 3771, 3765. For other bone cognates to identify all oblique-edged zhang zhang and those from the Erlitou as and related terms see pp. 3753-3804. and Longshan periods with dentiled haft qi (p. 63). For the post cut 29. James M. Menzies, The Shang Ko (Toronto, 1965), pp. 1-9. Erlitou type of zhang with rectangular haft and obliquely blade that see 30. KG 1975.5:305-306 (K3); KG 1992.4, fig. 3:3, p. 296, pi. 1:4 is Late Shang in date, Zhongguo kexueyuan kaogu yanjiusuo (M57). Anyang fajuedui, "1975-nian Anyang Yinxu de xinfaxian," KG 1976.4, 12 31. KG 1976.4:261-262. fig. far right, p. 271. are usu 32. Ibid., p. 260. 48. If Longshan and Erlitou period dao re?ut, the recutting is so a 33. Ibid., p. 263. ally executed by halving the blade that it takes the shape of "half of as is in the Eastern Han Shuowen notes This 34. See for comparison, the ge from the Erligang period site of gui" described (see 47, 65). is illustrated amidst the remains in Sichuan the Wangjinglou in Xinzheng county, Henan; KG 1981.6, fig. 1, p. 556. type of recutting jade (see see 35. For utlitiarian stone dao of Neolithic date see Cheng, Prehistoric drawn example in Feng and Tong, "Ji Guanghan," pp. 32-33). Also 220. China, pi. XIV: 1-4. Xia, "Shang Dynasty Jades," p. 36. KG 1975.5:306 (K3); KG 1992.4:296 (M57). 49. KG 1978.4:270. 37. KG 1978.4:270. 50. KG 1983.3, fig- 7, PP. 202-203.

88 in 51. In Carved Jades of Ancient China (Berkeley, 1938), pi. IX: 1, Alfred Bahr Now the Field Museum of Natural History, Chicago (New York, a a Salmony publishes Shang bronze ge which is inserted into jade han 1927), pi. XIII:i. are at dle of Erlitou inspiration. 68. Similarly hardened examples known from the Fu Hao tomb no. 52. See examples in KG 1978.4, fig. 2:2, p. 270; KG 1976.4, fig. 6:5, Anyang (Yinxu Fuhao mu, pi. 115:2) and from tomb 18 at Xiaotun, p. 262, pi. 6:2 right; KG 1984.1, fig. 5:3; KG 1975.5, fig. 4:4-5, 12-24, Anyang (KG 1981.4, pi. XVIL5). p. 305; KG 1985.12:1092-1033; KG 1994.3:296. 69. C. T. Loo, Inc., An Exhibition of Chinese Jades, pi. VI: 5. 53. KG 1975.5, fig. 44, P- 305 70. This piece has been published in Yutaka Mino and James 54. Zhouli 5/35b, 36a, 37a; 12/ib, 3a, "Dianrui" and "Kaogongji" Robinson, Beauty and Tranquility: The Eli Lilly Collection of Chinese Art com chapters. Representative of the nineteenth and twentieth century (Indianapolis, 1983), pi. 8, pp. 64-65. are mentators on the function and typology of the jade gui the Qing 71. The short dao in the American Museum of Natural History in ace. no. scholar Nie Chongyi, Xinding Sanlitu, reprinted by Guji Chubanshe New York, 70.3.3022, is published here for the first time; it measures cm (Shanghai,. 1984), vol. 10, p. 3; the Qing scholar Wu Dacheng, 17.0 long by 7.7 cm wide. Guyutukao, 1898, p. 21; the early twentieth century scholar Ling 72. For the dao from The Field Museum in Chicago see Laufer, 1. Shunsheng, "Zhongguo gudai ruigui de yanjiu," Min-tsu-hsueh yen-chiu Archaic Chinese Jades, pi. Ill: so notes chi-kan (Bulletin of the Institute of Ethnology):ij5-203; and Berthold 73. See 116, 117. Laufer, Jade: A Study in Chinese Archaeology and Religion (Chicago, 1912, 74. Erlitou C14 dates are cited in KG 1977.4, PP- 217?232 and in republished by Dover, 1974), chap. 2, pp. 80-103. Chang, Ancient Civilization of China, pp. 315, 318. For Longshan C14 55. Shuowen, tu-pu,^ pp. 6167-6169; quoted in Xia Nai, "Shang dates also see Chang, fig. 203, p. 247. Dynasty Jades," pp. 216-219. 75. In most elite burials of Liangzhu cultural date, yue-axes are the to 56. It is also possible, but unlikely due length, that the gui origi only weapons found in addition to ritual bi-disks and cong-tubes. The as at axes are cases nated in the chan-spade, such found Dawenkou period sites (see in all extremely refined, thin, and highly polished, de as use note 13). signed artistic works of art and not apparently for (see Liangzhu 57. KG 1983.3, fig- 10:7, PP- 202-203; KG 1975-5, fig- 4:6, p. 305. burials cited in notes 8, 13). ver One of the gui has remains of hemp wrapping and three lines in 76. WW 1984.2, fig. 3:8, p. 14; fig. 5, pp. 15-16 (Changxun); WW on 10 milion its haft mimicking where the blade would have been bound 1986.10, figs. 29-30, p. 10; fig. 31:8-9, p. (Fuquanshan). to a wenwu handle. 77. Anhuisheng gongzuodui, "Qianshan Xuejiagang Xinshiqi 58. KG 1983.3, fig. 10:2-4, p. 204; KG 1984.1, fig. 5:4-5, p. 38. shidai de yizhi," KGXB 1982.3:283-324. 59. KG 1983.3, fig- 7-4, P- 215; KG 1985.12, fig. 8:3, pp. 1092-1093. 78. Ibid., p. 322. 60. For the evidence that the Erlitou culture derived in part from the 79. In the report certain axes are described as chan and others as yue. see on no local Longshan culture of Henan reports Meishan and Dong-xi There is explanation for the differentiation (ibid., pp. 308?311). In in are afeng sites in KG 1975.5:282-294, Kaogu Xuebao (hereafter KGXB) troducing the stone tools from these burials the yue, chan, and fu cited, no are 1982.4:427-475 (Dongxiafeng); KG 1980.2:97-107, KGXB 1983. but fu described in the catalogue of finds. This lack of distinction 1:55?91 (Meishan); Yin Weizhang, "Erlitou wenhua tantao," KG and consistency of nomenclature characterizes the difficulty in distin at 1978.1:1-4; Huber, "The Bo Capital," pp. 56-60. A predecessor of guishing weapon and tool types this time when quality and refinement a a are as a Erlitou may be Wangchenggang, site with stamped earthen wall that all-important means of signifying wealth. See notes 7-8. has been identified with the Xia founder's, Yu's, capital Yangcheng; 80. KGXB 1982.3:297. Henan wenbo tongxun 1978.1:30?31; Chang, Archaeology of Ancient China, 81. Ibid., figs. 24:3-4, 25, p. 310. pp. 315-316; Zou Heng, "Shilun Xia wenhua," Xia Shang Zhou 82. See ibid., fig. 25:3, 20, p. 310. kaoguxue lunwenji (Beijing, 1980), pp. 95-182; An Jinhuai, "Jinnianlai 83. The excavators compare ceramics of PHI at Xuejiagang to those near Henan Xia Shang wenhua kaogu de xinshouhuo," WW 1983.3:1?7; Li from the Songze culture Shanghai (KGXB 1980.1) and to those near Boqian, "Erlitou leixing de wenhua xingzhi yu zushu wenti," WW from Beiyinyangying Nanjing (KGXB 1958.1). This stage precedes ca. 1986.6:41-47. the classic Shandong Longshan of 2400-2000 bce. to 61. Four Erlitou period zhang belong the Minneapolis Institute of 84. Ji Naijun, "Yenanshi faxian de gudai yuqi," WW 1984.2:84-87. Art. One (no. 50.46.312) is illustrated in color in Childs-Johnson, "Four 85. Cheng, "The T'ai-p'ing-ch'ang Culture," pi. 1:3. are Jade-Working Cultures," fig. 36, p. 41; and three published in Na 86. Zhongguo shehui kexueyuan kaogu yanjiusuo Shanxi gongzuo Chih-liang, Chinese Jades from The Minneapolis Institute of Fine Arts dui, "Shanxi Xiangfen Taosi mudi fajue jianbao," KG 1980:18-31; (Rutland, Vt.), 1977, pis. 9-11, p. 49. See also Horace F. Jayne, The "1978-1980-nian Shanxi Xiangfen Taosi mudi fajue jianbao," KG Chinese Collections in the Norton Gallery and School of Art (West Palm 1983.1:30-42. cat. no. are as Beach, Fla., 1972), 46; C. T. Loo, Inc., An Exhibition of Chinese 87. KG 1980.1:30. Yue here described chan-spades. Archaic Jades (New York, 1950), pi. IV3. 88. Dai, "Shaanxi Shenmuxian," pp. 157, 154?172. 62. Joan Hartman, Ancient Chinese Jades from the Buffalo Museum of 89. Dai, "Shenmu Shimao," p. 239. Science, exhibition catalog (New York, 1975), pi. 55 (mislabeled Shang 90. See chart illustrated in ibid., pp. 246?247. or Western Zhou). 91. Ibid., pp. 244, 239. In his earlier analyses Dai Yingxin argued that a were 63. One of the bronze versions of ge from Erlitou also has strong the jades functional. For example, he separated the dao into vari a ous median line but is balanced in proportions and without tip that bends categories, including shan-dao, lian-dao, and dao. Shan means to cut (see KG 1976.4, fig. 3:1, p. 260). down and lian refers to sickle; thus, the function of the two longest forms 64. KG 1976.4, fig- 3:3, P- 260. of dao seem similar rather than dissimilar. That described as lian-dao was are common see use 65. Re?ut dao in collections; Huang Jun, Guyutulu differentiated, seemingly, by the of rectangular rather than round 1:11-12 to re?ut as as to a (Beijing, 1939), and appear be early Shang times. holes for hafting the blade wooden handle and by its smaller size A re?ut dao-blade, for example, is known from the Late Xia/Early ("Shaanxi Shenmuxian," pp. 154-172). see Shang site of Taipingchang, Sichuan; drawn figure 76 in David C. 92. Bushi shiyongqi ershi ligi, Dai, "Shenmu Shimao," p. 244. "A Graham, Preliminary Report of the Hanchow Excavation," Journal 93. Ibid., fig. 1, p. 240; color pis. 1, 4. of theWest China Border Research Society VI (1933-1934). 94. Shima, Inkyo bokuji kenkyu, p. 327.1?2. 66. John Ayers and Jessica Rawson, Chinese Jade throughout the Ages, 95. KG 1977.3, fig- 2:2, p. 155. Victoria and Albert Museum (London, 1975), fig. 5, p. 24. 96. The bronze zhang from Erlitou is published in KG 1976.4, pi. 5:4, 67. Berthold Laufer, Archaic Chinese Jades Collected in China by A. W. fig. 3, p. 260.

89 97-Dai, "Shenmu Shimao," chart p. 248, fig. 1:7, p. 240 (no. Dongson culture of Iron Age date. Four zhang were discovered, two at SSY118). the type-site Phung Nguyen and two in the Xom Ren remains, both in 98. See Chang's review, Archaeology of Ancient China, pp. 244-247. Vinh Phu province. ca. 99. Other Shandong Longshan cultural dates of 2035 +/- 115 to 117. One very small zhang was discovered, broken, in a tomb at are on see 2405 +/- 170 cited in Yan Wenming, "Longshan wenhua he Dawan Lamma Island; Li Xueqin, "Xianggang Dawan chutu wenwu Longshan shidai," W/W 1981.6:41-48, 46 (Sanlihe); Li Boqian, "Erlitou yazhang yu wenti," Nanfang 1992.1:25?29; Yang, "Jade Zhang," notes leixing de wenhua xingzhi yu zushu wenti," KG 1986.1, 17?18, p. 53; Shang Zhixiang, "Xianggang Dawan yizhi chutu yazhang zhuiji," p. 47; KG 1977.4:225. Wenwu tiandi 1994.3:14-15. 100. Liu Dunyuan, "Ji Liangchengzhen yizhi faxian de liangjian 118. Zeng Fan, "Guanyu Fujian shiqian wenhua yicun de tantao," shiqi," KG 1972.4:56-57. KGXB 1980.3:264-284, pi. 1:8; "Fujian Zhangpu Xinshiqi shidai yizhi 101. The Shandong Archaeological Team iA, CASS, "Shandong diaocha," KG 1959.6:273-275. Linqu Zhufeng Longshan wenhua mucang," KG 1990.7:587-594. 119. D. S. Dye, "Some Ancient Circles, Squares, Angles and Curves 102. Liu Dunyuan, "Lun (Shandong) Longshan wenhua taoqi de jishu in Earth and in Stone in Szechwan, China," Journal of the West Chna he yishu," Shandong daxue xuebao 1959.3. Border Region Research Society (Xieda xuebao) ^(1930-1931), Chengdu, 103. Liu Dunyuan, "Shandong Rizhao Liangchengzhen kancha Sichuan, pp. 102-105. jiyao," KG 1960.9. 120. David C. Graham, "A Preliminary Repeort of the Hanchow 104. The Shandong Archaeological Team iA, "Shandong Linqu Excavation," The Journal of the West China Border Research Society Zhufeng," KG 1990.7:594. VI(i933-i934):ii4-i3i wenwu n.m. 105. Shandong xuanji (Beijing, 1977), figs. 5,8, 121. Cheng Te-k'un, "The T'ai-p'ing-ch'ang Culture," Xieda xuebao axes as 106. See the jade published chan-spades from Dawenkou, pi. i(i949):i-i5, 67-81. 122. wenwu 24; Childs-Johnson, "The Four Jade-Working Cultures," pis. 11-12, p. Sichuansheng guanli weiyuanhui et al., "Guanghan 33 Sanxingdui yizhi yihao jisikang fajue jianbao," WW 1987.10:1-15; 107. See Du Zaizhong, "Shilun Longshan wenhu de 'danketao,'" KG "Guanghan Sanxingdui yizhi erhao jisikang fajue jianbao," WW 11-12. 1982.2:181, pis. 1989.5:1-20; "Guanghan Sanxingdui yizhi," KGXB 1987.2:227-254; 108. See note 42; Yang, "Jade Zhang," pp. 53-54. "Lun Guanghan Sanxingdui yizhi de xingzhi," Sichuan wenwu 109. Another representative jade from the Fogg Museum in 1988.4:9-12; "Sichuan Guanghan chutu Shangdai yuqi," WW a an on Cambridge is yue with incised mask both faces; Max Loehr, 1980.9:76-77; Robert W. Bagley, "Sacrificial Pits of the Shang Period at Ancient Chinese Jades from the Grenville L. Winthrop Collection in the Fogg Sanxingdui in Guanghan County, Sichuan Province," Arts Asiatiques on Art Museum, Harvard University (1975), fig. 192, p. 151. XLIII(i988):78-86; Noel Barnard, "Some Preliminary Thoughts the no. See the examples illustrated in Childs-Johnson, "Four Jade Significance of the Kuang-han Pit-Burial Bronzes and Other Artifacts," an orna at Working Cultures," figs. 23-24, which include openwork jade paper presented the Conference of Ancient Chinese and Southeast a ment with eagle and cloud motifs from the Mus?e Cernuschi and jade Asian Bronze Culture, Kioloa, Feb. 8-12, 1988. huang with similar theme from the Art Institute of Chicago. Gui blades 123. Pit I is dated Late Neolithic through Late Shang (WW a are to with comparable bird motifs and semi-human mask in part illus 1987.10:14) and pit II Late Shang (WW 1989.5:19). was trated in Doris Dohrenwend, "Demonic Images from Early China," Ars 124. The pit at Taipingchang said to be 7 feet long, 3 feet wide, a Orientalise (1975); Na Chih-liang, Yu-chi tung-shih (Taipei, 1964), pi. 3; and about 3 feet deep and burial pit (Cheng, "The T'ai-p'ing-ch'ang one to was Huang Jun, Guyutulu, ce I, 7. On of these related works in the Freer Culture," pp. 1-2, 68-69). According Graham, the top of the pit a con 20 stone on Gallery of Art (Dohrenwend, "Demonic Images," fig. 37a-b), is originally covered with about disks lying their flat sides with set siderably hardened and simplified representation of this semi-human others up vertically aligning the sides ("Hanchow Excavation," pp. mask and winged bird. On other gui of Shandong Longshan style the 115, 116-118). Dye argues that the original configuration of these large a cone bird is eliminated and two facial images, one more animal-like and the to small sandstone disks formed shape with the larger disks lying other more human, alternate on front and reverse sides. at the bottom ("Some Ancient Circles," p. 102). were in. See Mou Yongkang, Liangzhu wenhua yuqi, cover. A gui blade 125. Stone and jade bi excavated from burials at Fuquanshan to stone from the Art Institute of Chicago, published originally by Alfred small by comparison the examples from Sichuan. They vary be tween cm Salmony in Archaic Chinese Jades from the Edward and Louise B. 13.0 and 23.0 in diameter (WW 1986.10:16). The largest bi was cm Sonnenschein Collection (Chicago: Art Institute of Chicago, 1952), pi. excavated from Sidun, Jiangsu only 18.2 in diameter (KG 1?2 a LIV:2; Carved Jade of Ancient China, pi. XXVII: is transitional piece 1981.3:197). Two of the largest bi known outside China, from The Field on measure cm documenting the influence of Liangzhu styles the Shandong Museum in Chicago, 39.0 and 41.0 in diameter (Childs on Longshan tradition. The image of the semi-human mask the gui Johnson, Ritual and Power, fig. 12, p. 17; "The Four Jade-Working reflects Liangzhu style, although the subject and type of jade blade doc Cultures," fig. 12, p. 36). 100 were ument Shandong Longshan taste. 126. Over fragments of bi preserved (Cheng, "The T'ai 1. 112. KG 1992.4, color pi. p'ing-ch'ang Culture," pp. 4?5, 70?71). 113. Li, "Erlitou leixing," p. 45 illustration. In discussing Shandong 127. Feng and Tong, "Ji Guanghan chutu," fig. 1, p. 33; Zhongguo on Longshan cultural influences the Xia, Li expands upon earlier ideas meishu quanji, gongyi meishu bian, 9 (yuqi) (Beijing, 1986), pi. 79, p. 29 of Zou Heng ("Shilun Xia wenhua," Xia Shang Zhou) and Zhen text. a in Zhengxiang ("Erlitou wenhua shangjue," Henan wenbo tongxun 1978.4). 128. See typical mask of the Liangzhu culture Childs-Johnson, to cover 114. Fu Sinian, "Yi Xia Dong Yi shuo," in Papers Presented Ts'ai "Four Jade-Working Cultures," and figs. 18-19, P- 35 on as as Yuan P'ei His Sixty-fifth Birthday (Nankang: Institute of History and 129. Cheng labels the zhang zhengui and the long fu/yue yangui Philology, Academia Sinica, 1935), pp. 1093-1134. ("The T'ai-p'ing-ch'ang Culture," pp. 2, 68). The jade labeled collared 115. James Legge, "The Annals of the Bamboo Books," chap. IV in disk from Taipingchang (Dye, "The T'ai-p'ing-ch'ang Culture," pi. 1:5) The Chinese Classics, vol. Ill: The Shoo King or The Book of Historical is well known in Shang and Western Zhou times (see Yinxu Fuhao mu, Documents (Hong Kong, 1865), p. 119 (Di Xiang), p. 125 (Di Fa, Di pis. 91-94); the form's earlier history is not yet clear. Early examples in to Gui), p. 122 (Di Shi). ivory from the late Neolithic site of Dawenkou may relate the on or 116. Diep Dinh Hoa, "New Findings Zhang in the Phung Nguyen Neolithic early historic prototype of this form (Dawenkou, fig. 87:4, Culture," Archaeology Enters the Twenty-first Century Symposium Papers, p. 102). axes more Beijing University, May 23-25, 1993. This culture precedes the 130. Long fu/yue commonly appear in southern, Liangzhu

90 burials than do in or period they northern, Dawenkou Shandong KG 1965.10:502; and elsewhere in Zhengzhou, KG 1957.1:72. see WW Longshan period burials; 1984.2, fig. 112, p. 4 (Fuquanshan, 147. See, for example, Yinxu Fuhao mu, pis. 115-116; Yeung Kin Wenwu ziliao in Shanghai); congkan 1980.3, fig. 61, p. 23 (Caoxieshan, fong, Jade Carving Chinese Archaeology, vol. I (Hong Kong: The and WW Wuxian); 1978.7, fig. 37, p. 15 (Shixia, Guangdong). Chinese University Press, 1987), pi. XXXIV8-10; also see note 51. and WW or are 131. Feng Tong, "Ji Guanghan yuqi," 1979.2, figs. 6?8, p. 148. These handles with "petal" plain design common in later The cm is illustrated in in 32. zhang measuring 56.1 long color Zhongguo Shang and Zhou tombs and show little stylistic difference from their meishu Erlitou see quanji, 9 (yuqi), fig. 78. prototypes; examples excavated from Zhengzhou, KG "The Hanchow "The 132. Graham, Excavations," fig. 76; Cheng, 1977.1, pi. 5:5; Dasikongcun, Anyang, S. H. Hansford, Chinese Carved T'ai-p'ing-ch'ang Culture," p. 77. Jades (London, 1968), pi. 11:4-6; Fuhao tomb (M5), Anyang, Yinxu "The 133. Cheng, T'ai-p'ing-ch'ang Culture," p. 76. Fuhao mu, pis. 156-160; Lingtai, Gansu of Western Zhou date, KG 134. This dao-blade is stored in the Sichuan see also unpublished presently 1977.2, pi. XVL4-5; Yeung Kin-Fong, Jade Carving, pis. Provincial Museum in at Chengdu. It is least three feet long, dark green XLVL9-16, LXXVL9-16. to black in decorated on one side with the standard de color, geometric 149. Zhongguo shehui kexueyuan kaogu yanjiusuo Anyang gong of Erlitou and takes a isoscelic also of sign style, perfect trapezoid shape, zuodui, "1969-1977-nian Yinxu Xichu mucang fajue baogao," KGXB standard Erlitou on style. 1979.1, fig. 79:1 left, described p. 105. One similar example was 135. in the form of remains of flakes in burial finds found in Turquoise inlay 1975 within the excavated foundations Fio at Anyang, KG are common to Erlitou and see KG 12 Longshan period sites; 1976.4, fig. I976.4, fig. right, p. 271. Broken examples, totaling 8, of similar type, 3:2, 260 and KG Shen cm were in tun p. (Erlitou) 1977.3, fig. 2;i, p. 155 (Shimao, 15.8 long, found Xiao burials; Yinxu fajue baogao . muxian) IQ58-1961 (Beijing, 1988 reprint), pi. 67:2-3, p. 255. See WW 8. KG Ma Dezhi et 136. 1987.10, fig. 12:1-9, pp. 5-6, 150. 1969.8:9; al., "1953-nian Anyang Dasikongcun and are terms drawn from the Zhouli and 137. Yazhang bianzhang used faxian baogao," KGXB 1955.9, pi- ? (M23, 32); Guo Baozhun and Lin Nie of the and the excavators to describe by Chongyi Qing (Sanlitu), by Shouchin, "1952-nian qiuqi Luoyang Dongjiao fajue baogao" KGXB the with of head are zhang differing shapes (WW 1987.10, fig. 12:3-4, P- 8; ?955-9, pi- 5 lower right. Single examples also known from Eastern 13:7, In the refers to the Zhou tombs of fig. p. 9). archaeological report bianzhang zhang Springs and Autumns and Warring States dates; see whose blade-head takes the of an shehui shape oblique angle. Zhongguo kexue yuan kaogu yanjiu suo, Luoyang Zhongzhoulu 138. Two of these are 68.2 and examples long, measuring 50.0 cm; (Beijing, 1959), fig. 53:3, p. 115; Shangcunling Guoguo mudi (Beijing, and a third measures cm Guo 27.3 long (WW 1987.10:8-9). I959), pis. 10, 21; Baozhun, Shanpiaozhen yu Liulige (Beijing, 1959), For a discussion of the of see 139. significance Shu Dong Qixiang, pis. 35:6; 118:13. "BaShu shehui BaShu lishi It is to account xingzhi chutan," (Chengdu, 1991), pp. 151. difficult for this disappearance since the dao was Gudai de Ba Shu and more re so an 23-43; Tong Enzheng, (Chengdu, 1979); monumental artistic tradition during the Erlitou period. As sym Li "Shu rende cently, Shaoming, laiyuan yu zushu," Jinian Sanxingdui bolic blades, dao do not appear in Shang or Western Zhou tombs. For liushi zhounian Shu wenhua lishi kaogu faxian ji (Ba yu guoji xueshu taolun reworked dao in Shang times see note 65. Sichuan. hu?), April 1-6, 1992, Guanghan, 152. Yinxu Fuhao mu, pi. 84. WW For 140. 1989-5, fig- 34:3-5, PP- 16, 12-13. 153. eccentrically long examples in pottery and stone from 141. WW Gansu of Eastern see 1989.10:5. Tongshan, Zhou date, KG 1976.1, figs. 1-2, pp. 142. Zhen makes the same in her discussion of the and in stone Zhengxiang point 46?47; and shell from Fengxixiang, Fengxi, Shaanxi, see dates of from wenwu jades Sanxingdui (Zhongyuan 1993.1:9?11). Fengxifajue baogao (Beijing, 1962), pis. LXXXVIL6, XC:6). 143. WW There is no reason to assume that because this WW 1966.1:58. 154. 1988.11, fig. 8:105, p. 19. For additional examples see blade was found in that it dates to the or CII:n Zhengzhou Erligang Early Fengxi fajue baogao, pi. (Kexingzhuang); Luoyang Zhongzhoulu, There was no excavated context. Both Xia Nai Shang period. ("Shang pis. 82:9, 48:3, 11; and WW 1985.2, fig. 6, p. 44 (Qishan, Licun, note Dynasty Jades," p. 219, 20) and Yin Weizhang ("Erlitou wenhua Shaanxi). zai KG make this at tantao," 1984.4:353) assumption. 155. See M5 Anyang, Yinxu Fuhao mu, pis. 81-82; the Western 144. KG 1957.1:72, V:i2. Sector at pi. Cemetery Anyang, "Xichu mucang," KGXB 1979.1, fig. 79:3, 145. renmin chutu wenwu xuan Cultural at Zhongguo gongheguo (Beijing: p. 104; Dasikongcun Anyang, Hansford, Chinese Carved Jades, fig. Relics no. Press, 1976), 15. 11:2; and Luoyang Zhongzhoulu, pi. 51:8 (Western Zhou). for 146. See, example, ge-blades from Minggonglu in Zhengzhou,

91 Chinese Characters

aa. a. % # b. n ab. & c. it ac. d. 77 ad. e. ^ ae. f. 31 af.

ag

ah.

ai.

aj k. * i. m. A n. o. ^(*) p. &

s.

t. ^4 u. *.# V. w. #

x. 5^r

y. ^ff z.

92