“The Bridge between Eastern and Western Cultures” Volume 4 Number 2 Winter 2006-2007

In This Issue From the Editor

About

The Silk Road is a semi-annual publi-cation of the Silkroad Foundation. TheSilk Road can also be viewed on-lineat <http://www. silkroadfoundation.org>. Please feel free to contact uswith any questions or contributions.Guidelines for contributors may befound in Vol. 2, No. 1 (June 2004) onthe website.

The Silkroad FoundationP.O. Box 2275Saratoga, CA. 95070

Editor: Daniel C. Waughdwaugh@u.washington.edu

© 2007 Silkroad Foundation© 2007 by authors of individual ar-ticles and holders of copyright, asspecified, to individual images.

Rather than attempt to commenthere on every article in this issueof our journal, let me share withyou some thoughts inspired byreading two important new bookswhich are closely related to certainof our contributions. In the firstvolume, Royal Nauruz inSamarkand, the eminent scholarProf. Frantz Grenet begins hisessay with the statement: ‘Apositive side to the so-called‘Ambassador’s painting’ atSamarkand is that we shall neverfully understand it…This meansthat research on this painting willnever stop and this is excellentnews, as this research had manyrepercussions on the knowledgeof Sogdian history and art’ (p. 43).Indeed, as most of our articles inThe Silk Road this time remind us,we live in a world of newdiscoveries, if one marked by sadreminders of what has been lostor is under the threat ofdestruction or theft. Were we tosecond-guess history, we mightask, for example, what if thebulldozer in 1965 had not torn offthe top of the hall with the‘Ambassador’s painting?’ Wouldit eventually have been properlyexcavated with enough moreintact to answer some of the nowunanswerable questions which areinviting such ingenious solutionsas the one proposed below byMatteo Compareti, who organizedthe conference in which Grenetand the other experts par-ticipated?

The conference reports in RoyalNauruz are fascinating reading inpart precisely because the authorsdo not always (and probably neverwill) agree on some of the mostimportant issues. Was the hallwith the mural part of a palace or

part of an elite but non-royalresidence? What is depicted? Isthe whole iconography connectedwith celebration of Nauruz? Is itabstract and symbolic or ratherrelated to a very specific politicalsituation? Is the Chinese scene onthe north wall a specific depictionof court culture in China or simplyemblematic of a Chinese prin-cess’s having been sent off as abride to Central Asia? It is certainlyinteresting that at least onecontributor (Markus Mode)explicitly disagrees with thepremise about Nauruz which isembodied in the volume’s title.While the consensus dating for thepaintings now has been narrowedto the 7th century, scholars cannotagree whether we might belooking at the late 640s or, say,the early 660s. Each choice hassome plausible arguments in itsfavor. Or, take the Sogdianinscriptions on the murals,published in a new reading in thisvolume by Vladimir Livshits. Whenwere they written? Do they recordmerely formulae and inter-pretation of the imagery by thosewho could not have witnessed thescene it depicts?

One can readily obtain a visualimpression of the scholarlydisagreements by comparing thevarious reconstructions of themurals on the west wall of the hall,the wall that arguably held thecentral imagery, since it faced theentrance to the room. Since onlythe lower part of the wall survivedthe ravages of time and thebulldozer, within certain limitsimposed by the rest of theimagery, the scholar is free toimagine what was in the upperregister. A number of key issuesare at play here. One basic and

• News from ancient Afghanistan.....5

• Prof. Tarzi’s 2006 excavations atBamiyan ..............................10

• A visit to the region of historicBalkh.................................. 27

• A new interpretation of the Afrasiabmurals.................................32

• Mapping Buddhist sites in WesternTibet ...................................43

• Han lacquerware in Xiongnugraves..................................48

• Ming-Timurid relations as recordedin Chinese sources ................54

• Hunting hounds along the SilkRoad....................................60

• An interview with Kyrgyz epicsingers.................................65

Next Issue

• A focus on food along the Silk Roads,with articles by Terrie Chrones, DarraGoldstein, Paul Buell, Debra McCown,Dru Gladney and Nancy Chen....

• The latest on Xiongnu archaeology

• Trade routes in Anatolia

.... and more

perhaps somewhat surprising oneis that most of the interpretationswhich have been advanced arebased on an imperfect publicationof the surviving evidence. As IrinaArzhantseva and Olga Inevatkinanote in their paper, the earliestdrawings of the paintingscontained many errors andomitted significant detail. Themore precise drawings undertakenduring restoration work in 1978have remained largely unknownand still need to be properly andfully published. The two scholarsdocument the value of those 1978drawings with a number ofexamples (and, one might add,the volume as a whole is valuablefor its extensive illustration).

As presumably any of thespecialists should admit, thechallenges of ‘reconstruction,’which perforce involves inter-pretation, and other kinds ofanalysis based on that which hasbeen reconstructed embody somereal dangers. For example,François Ory’s explanation for hisreconstruction drawings (see Figs.1-3, p. 91) should give everyonepause. A lot of his detail derivesfrom his reading of the other justlyfamous set of Sogdian paintings,those at Panjikent. Yet, as Étiennede la Vaissière pointedly notes (p.153; he is commenting on BorisMarshak’s interpretation of themurals, not on Ory’s recon-struction), ‘The painting ofSamarkand is not that ofPanjikent.’ Might one not respondto de la Vaissière, who seems toprefer for the west wall imagery avariant where the upper registeris based on a relief on a Sogdianfuneral couch from China (see hisfig. 3, p. 158), that ‘The paintingof Samarkand is not Sogdiancarved reliefs from China’? Is thekey and now lost image in theupper register the Goddess Nana,the Sogdian ruler Varkhuman, orthe Turk khaghan or somealternative configuration of thelast two? Choose one and a wholeskein of alternative reasoningunravels.

Part of the challenge here is tobalance the inventiveness ofmodern scholars in findinganalogies and sources for theimagery with the realities of thepreserved evidence and what tomoderns are the opaque worldviews and knowledge of thecreators of the paintings. I thinkno one nowadays would err on theside of underestimating thecomplexities of Sogdian culture,which embraced Iranian, CentralAsian, Indian and other elements.Indeed, what meticulous scho-larship is determining is how wide-spread certain motifs seem tohave been, even if they havesurvived to this day in scatteredlocations only in isolated andfragmentary examples. Mightthere not be a danger though ofimposing a rationality and systemon the Samarkand paintings whichthey do not merit? And mightthere not be a danger ofexaggerating the complexities?The late Boris Marshak’scontribution to this volume (whichhas been dedicated to hismemory) is noteworthy for hisopening admission that he hadbeen wrong in some of his earlieranalysis and for his insistence thattoo many of the other interpretiveschemes are at odds with thearchaeological evidence fromSogdiana, evidence which mostwould agree he knew better thanany other person. While of coursehe may not be right, Marshakshows what I would call a salutaryskepticism, for example, in regardto reading too much out of theSogdian inscriptions on thepaintings. It is worth rememberingone basic rule of argument, whichis that the chain of evidence is onlyas strong as each of its links. Ithink part of what Marshak isgetting at is that tendency to wantto construct an edifice of ‘proof’mainly out of unproven assum-ptions, at least some of which endup becoming ‘accepted fact.’

It is hard to know how we mayrespond to this volume when were-examine it in five or ten years,but I think there is a reasonable

bet that the articles in it which willhold up best with time are thosewhich are the narrowest in theirfocus on specific details: EtsukoKageyama’s careful comparison ofdetails of coiffure and garmentsof the Chinese women withrecently discovered depictions inearly Tang tombs; ValentinaRaspopova’s examination ofweaponry with her interestingobservations on the fact that to adegree the artists did not alwaysmatch weapons to the ethnicity ofthe individuals in the paintings;the previously mentioned articleby Arzhantseva and Inevatkina;and Alix Barbet’s technical studyof the painting technique. Thatsaid, Grenet surely is right aboutthe stimulation provided by theuncertainty concerning what weknow. That stimulation isabundantly evident in MatteoCompareti’s article which wepublish below: if his conclusionwould be accepted, it could indeedhelp to explain some of thepuzzling details of the‘Ambassador’s painting.’

While a volume such as RoyalNauruz is a landmark publicationpulling together both what isknown and what is hypothesizedand posing questions for futureresearch, the volume indirectlysuggests another kind ofdesideratum for the success ofthat research. Extensive as theliterature now is on theSamarkand paintings (and, ofcourse, many of the otherimportant bodies of materialunearthed along the Silk Road),much of it is not readily accessible.Markus Mode and the Franco-Uzbek Archaeological Missionheaded by Frantz Grenet are to becommended for their publicationof materials on the Internet.Imagine the benefit were they tocoordinate efforts and beginsystematically to post to one orthe other website digital copies ofthe scholarly literature that isoften published in books orjournals unavailable in mostlibraries. Surely it is in the interestof everyone that permissions be

2

obtained for copyrighted anduncopyrighted material to bereproduced in electronic form.That would truly be, to useGrenet’s words, ‘excellent news.’

If the Samarkand murals havein certain ways been ‘well known’and preserved starting with theirdiscovery over a quarter centuryago, the excavations at Bamiyanand surveys in southern Tibetreported on in this issue may benews to many and highlight theongoing threats to the pre-servation of cultural artificats. Thethreats come from variousdirections — illegal excavation,deliberate destruction for religiousor other reasons, indifference, thepressures of ‘modern develop-ment.’ In too many places — bethey in the Middle East, LatinAmerica, or Southeast Asia —there is no adequate security evenat sites which have long beenrecognized as having substantialhistorical value. It is hard to knowwhere the balance lies betweendiscovery and charting of newsites (as is being done in theproject reported on by Karl Ryavecin Tibet) and protection. As theSichuan archaeologist Huo Weinoted in a presentation I heard onthis Tibetan material a year ago,when the archaeologists returnedto at least one of the sites a yearafter the first discovery, the lootershad already been at work.

When we think back over thehistory of archaeology on the SilkRoad, there is, of course, the well-flogged issue of whether the‘foreign devils’ should becastigated for carting off to Europeso many objects, defacingBuddhist cave sites, etc. Therealso are the unintended con-sequences of excavations such asAurel Stein’s, where once the wordgot around locally about thepossible value of the artifacts, localentrepreneurs went to work onwhat had been covered up and leftbehind. By the time C. P. Skrine,the British Consul in Kashgar in1922-24 arrived in Khotan, hecould purchase sizeable chunks ofBuddhist murals from a shop that

was stocked with them. Skrinereported how one of these‘excavators,’ looking for objects tosell, entered a previously unknownsmall temple in the desert only tohave the walls and their muralsdisintegrate around him.1 It isimpossible to come up with abalance sheet here regardingwhether the world would havemore or less of the evidence ofhistoric remains had not many ofthem been removed by outsiderswhose motives may have beenaltruistic and scholarly orimperialistic. We cannot know, forexample, whether the remains atKizyl or Miran would have beenbetter preserved for posterity hadthe German expeditions in the firstinstance, and Stein in the secondnever been there. Even thoughnow, unlike then, proper legalframeworks are in place to protectcultural patrimony and prevent itfrom being illegally removed,enforcement is lax.

Such controversies plague theissue of the antiquities ofAfghanistan, where since 1922,with a hiatus in the Civil War andTaliban years, the DelegationArchéologique Française enAfghanistan (DAFA) has sofruitfully worked. Perhaps morewould have been left for futuregenerations to discover hadmaterial excavated prior to theend of the 20th century not endedup in the Kabul Museum for theshelling of the Afghan Civil Warand the iconoclastic Taliban todestroy. But who was to know?The world knows the grievous taleabout the wanton destruction ofthe Buddhas of Bamiyan. Theworld should know and supportbetter the resumption of seriousarchaeological investigations inAfghanistan. Would that therealso be some way to check theinroads made by ongoing illegalexcavations, whose pits pockmarkthe landscapes of many of thebetter known sites. Field work byDAFA at Bamiyan, now headed byZemaryalai Tarzi, is alreadyadvancing substantially what weknow about that former royal

residence and Buddhist monasticcenter, beyond what we knewwhile the Buddhas still stood.Read Professor Tarzi’s reportbelow: although phrased verycautiously, it encourages us tothink that even more significantresults may soon emerge.

This brings me to the secondbook about which I promised towrite here: Afghanistan: theRediscovered Treasures, thecatalogue for a remarkableexhibition at the Musée Guimet inParis until April 30 (which then willtravel to the United States). Thetreasures begin chronologicallywith the Fullol hoard, which datesfrom about four millennia ago. Thelargest portion of the objects comefrom the much later sites: AïKhanum on the Amu Darya, TilliaTepe, and Begram. Many yearsof the French-directed excavationsunder DAFA were devoted to AïKhanum and Begram. A Soviet-Afghan team headed by VictorSarianidi and Zemaryalai Tarzi,excavated the spectacular hoardof gold from Tillia Tepe in the late1970s just prior to the Sovietinvasion of Afghanistan. A well-illustrated catalogue of the TilliaTepe finds appeared in Russian in1983. Until the objects from thesevarious excavations resurfaced in2003, they had been assumed lostduring the Civil War and period ofTaliban rule. In fact they had beenlocked away safely in vaults of theCentral Bank. The exhibitionprovides a rare opportunity to seethese treasures and to support thereconstruction of the KabulMuseum, which will benefit fromthe proceeds of the tour.

Apart from its superbillustration in color of 228 objectsand numerous historic photo-graphs and drawings from theexcavations, the cataloguecontains a number of valuableessays, starting with that by OmarKhan Massoudy, the Director ofthe National Museum in Kabul,laying out the recent and tragichistory of his institution and itscollections. Several of the leadingFrench specialists — Pierre

3

Cambon, Jean-François Jarrige(who is also Director of the MuséeGuimet), Paul Bernard andVéronique Schiltz — bring readersup to date on the significance ofthe four major excavations andthe objects they uncovered. Thefinds at Fullol, discussed byJarrige, are important as evidenceconcerning the ‘Bactro-Margiana’cultural complex of ca. 2200-1800BCE, which connects to the better-known centers of culture inMesopotamia and the IndusValley. Bernard reviews theimportance of Aï Khanum as anoutpost of Hellenic culture. Visitorsto the exhibition will have theopportunity to see the well-knowninscription commemorating thevisit there by Clearchos, a discipleof Aristotle, and quoting theDelphic code of civic conduct.Véronique Schiltz’s essay on theTillia Tepe finds, discusses the yetunresolved issues of theirchronology and places the objectsin the context of other earlynomad art. A appendix by ThomasCalligaro on the technical analysisof the jewels set into the Tillia Tepegold demonstrates the broadgeographical sweep encompassedby the trade networks, fromnortheast Africa to Tibet andCeylon to the Baltic. And Chambonwaxes lyrical about the eclecticismand artistic imagination of thesteppe nomads (‘un mondenomade éclectique et ouvert, quijoue de la curiosité pour desmondes differents et cultive labeauté,’ p. 296). Not that we needto be reminded, but this isprecisely what we expect in the‘crossroads’ of civilizations whichlater would produce an analogoussyncretism of cultures depicted inthe paintings of Afrasiab.

In the most substantial of theessays, Chambon reviewscarefully the arguments about thechronology of the Begram finds.Here we seem to have aconundrum equivalent to thatregarding the paintings atAfrasiab, where there is muchscholarly disagreement andapparently little likelihood in the

near future of a definitive solution.Among his many interestingobservations are ones on thedegree to which some of theresults of the French excavationswere inadequately or incompletelypublished. He concludes thatthere is a cultural unity in the findsat Begram, Tillia Tepe and Sirkap(one of the most importantsettlements at Taxila in northernPakistan), with a strong indicationthat the Begram ‘treasure’ isthereby pre-Kushan in date, either‘Indo-Greek’ or Parthian. Whilethere are some early indicationsof Buddhism, this is not yet theera of the refined development ofGandharan Buddhist art under theKushans. The Begram ivories, withtheir stylistic diversity andenthusiastic evocations of thepleasures of life are well known.There are bronzes of strikingrefinement. The glass is nothingshort of miraculous — enameledbeakers with their scenes of thehunt, fish-shaped flagons, twocobalt blue blown glass vases…

I would like to think that in mylifetime I might visit a restoredKabul Museum and see thesetreasures there, under conditionswhich might ensure that they bepreserved safely to educate newgenerations of Afghans about thewonders of their heritage. As it is,the objects are arguably betterknown in the West than in theiroriginal home. The ‘News fromAncient Afghanistan,’ to quote thetitle of Nicholas Sims-Williams’article below, in fact can besurprisingly good. That there caneven be such an exhibition now inParis is, I suppose, some reasonfor hope in the face of the grimrealities which, alas, are regularfeatures in the news from modernAfghanistan.

Daniel Waughdwaugh@u.washington.edu

References

Afghanistan: les trésors retrouvés.Collections du musée national deKabul. Pierre Cambon and Jean-

François Jarrige, eds. Paris:Editions de la Réunion des muséesnationaux, 2006.

Matteo Compareti and Étienne dela Vaissière, eds. Royal Nauruz inSamarkand: Proceedings of theConference Held in Venice on thePre-Islamic Paintings at Afrasiab.Revista degli studi orientali, N. S.,Vol. LXXVIII, Suppl. no. 1. Pisa;Roma: Accademia editoriale,2006.

Note1. See Skrine’s field diary, BritishLibrary, Oriental and India OfficeCollections, MSS EurF 154/39, p.47, April 4, 1924, under theheading ‘New stupa found byAbbas Khan at Khadalik.’

4

A Hundred Years ofDunhuang, 1907-2007

An international conference co-sponsored by The BritishAcademy, The British Museum andThe British Library, will be held atall three venues, Thursday 17 toSaturday 19 May 2007.

In May 1907, the Daoist caretakerof the Dunhuang Buddhist cavesin northwest China revealed toAurel Stein FBA a hidden libraryin Cave 17. The library had beensealed a thousand years earlierand was packed with documents,manuscripts and paintings. Thisdiscovery revolutionised ‘orientalstudies’ throughout the world inthe early 20th century. In thiscentenary year we seek (1) toreflect on the discovery and (2)to review its impact on ‘orientalstudies,’ including the writing andre-writing of history and (3) todiscuss directions for the future.The rich finds from Dunhuanghave implications beyond ‘orientalstudies’ and need to beunderstood as part of worldculture. Details of the programmay be found at <http://www.britac.ac.uk/events/2007/dunhuang/prog.html>, wherethere is a link to a pre-registrationform.

News from Ancient AfghanistanNicholas Sims-WilliamsSOAS, University of London

Years of war and instability haveled to profound damage toAfghanistan’s cultural heritage,with the accidental or deliberatedestruction of ancient monumentsand works of art, the looting of theKabul museum and the pillagingof sites such as Aï Khanum bytreasure-seekers. At the sametime, a huge number of re-markable and importantantiquities have come to light,some no doubt as the result ofunofficial digging, others perhapschance finds resulting from thedisplacement of peoples to remotemountain refuges. These dis-coveries include documents fromall periods of Afghan history, ofwhich I would like to mention afew outstanding examples (inapproximately chronologicalorder):

• A group of nearly fifty Aramaicdocuments from the archive ofthe last Achaemenian satrap ofBactria, ca. 353-324 BCE. Thedocuments date from the reignsof Artaxerxes III, Darius III, andAlexander the Great — who isreferred to exactly as if he werea legitimate successor of theAchaemenids. This discovery wasmade public by Shaul Shaked inrecent lectures in Paris (Shaked2004); an edition by Shaked andJoseph Naveh is in press.

• An administrative document inGreek, a tax receipt dated in the4th year of the Graeco-Bactrianking Antimakhos I, perhapsequivalent to 170 BCE. Thisdocument, supposedly found atSangcharak in north centralAfghanistan and now in theAshmolean Museum, Oxford,was first published in 1994 andmost recently edited by ClaudeRapin (1996).1

• An acrostic inscription from theKandahar region in Greek verse

with Homeric allusions (Bernardet al. 2004). This inscription ofabout the 2nd century BCE tellsthe life-story of Sophytos son ofNaratos, who lost his fortune andthen made another as amerchant in distant lands beforereturning home to the delight ofhis friends. Perhaps the mostremarkable feature of thisinscription is that the author ofthese highly sophisticated Greekverses seems to have been ofIndian ancestry, as his name andpatronymic suggest.• Next in chronological order (15CE) is a Kharoshti reliquaryinscription of Vijayamitra, king ofApraca in NW India, whichcontains a synchronism betweenthe Azes era, usually identifiedwith the Vikrama era of 58/57BCE,2 and a previously unknown‘Indo-Greek era’ beginning 128years earlier (Salomon 2005).According to Richard Salomonmany well-known inscriptionswith dates in the 200s and 300scan plausibly be assigned to thisera of 186/5 BCE. On thisreckoning, for example, thetril ingual Dasht-i Nawurinscription of Vima I, thegrandfather of Kanishka I, maybe dated to about 93 CE.• Though the document itself isnot a new discovery, I mustmention here a Sanskritastrological text, theYavanajâtaka of Sphujiddhvaja,which also contains a syn-chronism, this time between theSaka era of 78 CE and theKushan era. According to theinterpretation recently proposedby Harry Falk (2001; 2004), thesynchronism establishes that thefamous ‘era of Kanishka’ beganin 127 CE. If this is accepted, itmust rank as one of the mostimportant of recent discoveries,since the date of Kanishka is a

fundamental point in thechronology of ancient Afghan-istan and India.• The Bactrian inscription ofRabatak, discovered in 1993 andfirst published by Joe Cribb andmyself in 1996,3 which describesevents of the first year ofKanishka and traces hisgenealogy back to his great-grandfather Kujula Kadphises.• A group of about 150 Bactriandocuments from northernAfghanistan, which date from the4th to 8th centuries CE (to bediscussed below).

• The Bactrian inscription ofTang-i Safedak (Lee and Sims-Williams 2003), which recountsthe foundation — or re-foundation — of a stupa inYakaolang district, Bamiyanprovince, probably in 714 CE,demonstrating the persistence ofBuddhism in Afghanistan wellinto the Islamic period.• Finally, I may mention a groupof Arabic administrative docu-ments, chiefly tax records, datedbetween 138 and 160 AH (= 755-777 CE). These documents,which will soon be published byGeoffrey Khan of CambridgeUniversity, mention some of thesame families and individuals asthe latest Bactrian documentsand seem to form part of thesame archive.

I do not have time today to talkabout all of these new documents,which cover almost all periods ofAfghan history from theAchaemenian period to earlyIslamic times. Instead, I shallrestrict myself to those that I knowbest, the Bactrian letters anddocuments which have come tolight during the past fifteen yearsand which appear to have beenwritten in northern Afghanistanduring the 4th to 8th centuries CE.I first reported on thesediscoveries in 1996, in a paperpublished in various forms inEnglish, French, Russian andJapanese.4 Since then, thenumber of Bactrian documentshas continued to rise and nowstands at more than 150. These

5

documents are written in a cursivescript derived from Greek, mostlyon parchment or leather,occasionally on cloth or wood.Apart from a couple of Buddhisttexts, the find consists of legaldocuments, letters, and economicdocuments. Many of them aredated according to an era whosestarting-point was almost certainlythe accession of Ardashir I, thefirst emperor of the Sasaniandynasty of Iran, in 223 CE.5

Although we have no reliableinformation on the discovery of theBactrian documents, it seemsfrom their contents that most ofthem derive from just one or twosources. In previous presentationsI have described in some detailthose which appear to come fromthe kingdom of Rôb (now Rui) inthe northern Hindukush, some ofwhich were written in places whichcan be identified: Rob itself;Samingan; Malr or Madr; Kah(modern Kahmard); and Warnu,which may correspond to modernQunduz, mediaeval Warwaliz [Fig.1].6 On this occasion I would liketo concentrate on a group of sevenor eight recently-discovereddocuments which originate in aslightly different area, theprovince of Gôzgân (Gûzgân,Jûzjân), to the northwest of Rob.7

These are all legal documents: adeed of sale, a loan contract, adeed of gift, a receipt, and severaldocuments regarding thesettlement of disputes. Both intheir physical form and in theirformulation they are very similarto documents of the same typefrom the kingdom of Rob. Almostall of them refer to the rulers ofGuzgan and to places in Guzgan;all but one are dated, and thedates fall within a period of 65years beginning in the year 436of the local era, i.e. 658/9 CE.

Luckily we possess a number ofother sources for the history andgeography of Guzgan during thisperiod. Amongst the geographicalworks one must give pride of placeto the Hudûd al-‘Âlam ‘TheRegions of the World,’ a Persianwork of the late 10th century,

which was written at the behestof a ruler of Guzgan. As EdmundBosworth has written (in Minorsky1970, p. viii), the anonymousauthor of this book ‘was essentiallyan armchair geographer,’ but ‘thedescription of Guzgan and itsdependencies ..., the one sectionof the book which must depend onpersonal observation andexperience’ is exceptionallydetailed and authoritative.Another important source is thehistory of Tabari, who recountsvarious episodes in the history ofGuzgan, some of them inconsiderable detail. Tabari’saccount of Guzgan begins with thefirst Arab invasion in the year 32of the hijra, i.e. 652/3 CE, just afew years before the earliest ofthis group of Bactrian documents.Chinese sources too preservesome relevant information, sinceGuzgan came under the nominalsuzerainty of the Tang dynastyafter the final defeat of theWestern Turks in the late 650s.Finally, we have at least one othersource which is actuallycontemporary with these

documents: a series of coinsbearing Bactrian legends whichinclude the name and titles of aruler named ‘Zhulad Gozgan, kingof Gar’; they also name the city ofAmber, presumably as a mint-town [Fig. 2, next page].

The city of Amber, nowadaysSar-i Pul, is well known fromIslamic sources. One of the newBactrian documents, a poorly-preserved loan-contract, waswritten in Amber [Fig. 3, nextpage]. Both the lender and theborrower are identified as ‘market-traders of Amber,’ which fits wellwith the description of Amber inthe Hudud as a ‘residence ofmerchants and the market ofBalkh.’

Another of our documentscontains the statement that it waswritten ‘in Sozargan, in Kalf, in thecity which they call Astof, in thepresence of the god Wakhsh, theking of gods, whom they worshipin Stof.’ I cannot identify thenames Sozargan and Astof, but itseems likely that Kalf is the Kalifor Kelif of the Islamic geographers.

Fig. 1. Northern Afghanistan

6

Kalif lay on the River Oxus, whichformed the northern border ofGuzgan; it was an importantcrossing-point, since it lay on theroad from Balkh to Bukhara andNasaf. Nowadays the principalsettlement named Kelif is that onthe north bank, in Turkmenistan,but there is also a village namedKilif or Kilift on the Afghan side.In mediaeval times Kalif wascounted as occupying both banksof the Oxus, being likened toBaghdad in this respect.

No less than five of ourdocuments were written either inGaz or in Lizg. These seem to be,in effect, two designations for thesame place.Probably Lizg,which means ‘thefortress,’ wasoriginally an epi-thet used todescribe a citywhose propername was Gaz. Acouple of docu-ments providesome detailswhich may help tolocate the fort-ress of Gaz. Oneis said to havebeen written ‘inGaz, at Wanin-dan, the court ofthe fortress,’which suggeststhat the ruler ofGuzgan may haveheld court at Gaz.

Another refers to ‘Gaz, in thedistrict of Andar.’ The Hudud givesthe name Dar-i Andara to amilitary camp where the ruler ofGuzgan had his residence; it isdescribed as being at the foot ofthe mountains, close to the townof Jahudhan (modern Maymana).If the name Dar-i Andara means‘the court of Andara,’ it may wellbe the place which once bore thename Gaz and which ourdocuments refer to as ‘the courtof the fortress’ ... ‘in the districtof Andar.’8

I turn now from the place-names mentioned in ourdocuments to the personal names

and titles, in particular, those ofthe rulers of Guzgan.

At least three rulers of Guzganare named in the documents:Kanag Gozgan, Skag Gozgan andYan Gozagan. I take it that Kanag,Skag and Yan are the personalnames of the ruler and that thename of the country, Goz(a)gan,is here used as a kind of family ordynastic name for its rulers. Asimilar usage is attested in Tabari,who (unfortunately for us) nevernames the ruler of Guzgan butrefers to him only as ‘al-Juzjani’or ‘al-Juzjan b. al-Juzjan.’

Kanag Gozgan is mentioned inthe earliest document, apurchase-contract dated in 658/9CE, in which both the vendors andthe purchasers are described as‘servants of Kanag Gozgan.’ Thename of Skag Gozgan occurs in adocument of 674/5, whichspecifies the payment of a fine forbreach of contract to ‘the treasuryof Skag Gozgan.’ Finally, the nameof Yan Gozagan is found in the verylast document of this group, datedin 722/3. This is a declarationintended to settle a disputebetween a number of persons whoare all described as ‘inhabitants ofLizag’ and ‘servants of YanGozagan, king of Gar.’

Fig. 2. Coin of Zhulad Gozgan, King of Gar (obverse and reverse).Photo copyright © 2007 Nicholas Sims-Williams.

Fig. 3. Bactrian loan-contract written in Amber (Sar-i Pul).Photograph copyright © 2007 Nicholas Sims-Williams.

7

The significance of the title ‘kingof Gar’ is obscure. I have not beenable to identify a specific placenamed Gar; possibly it is merelythe Bactrian word for ‘mountain.’The same title reappears on thecoins mentioned earlier, whichbear the name of yet anothermember of this dynasty: ‘ZhuladGozgan, king of Gar’ or ‘king of themountain country.’ These coinsbears dates in Pahlavi, but theyare practically illegible. Doubtsabout the era used add a furtherelement of uncertainty, so that wecannot at present be sure ofZhulad’s place in the sequence ofrulers. As it happens, the nameZhulad Gozgan also occurs in adocument of 705/6, a deed of giftissued by the ‘lord of Lizg’ in favourof three brothers, ‘Babay, Abgasand Zhulad Gozgan, sons ofKanag,’ whom he describes as ‘myown servants.’ It is not very likelythat the three brothers are thesons of Kanag Gozgan, the rulernamed 47 years earlier in thedocument of 658/9. Even if theywere, one would hardly expectthat Zhulad, the last-named andtherefore presumably theyoungest of the three, would havebecome king. It seems more likelythat this Zhulad and his fatherKanag belong to later generationsand perhaps to a junior branch ofthe royal family and that they werenamed after their i l lustriousforebears, the rulers KanagGozgan and Zhulad Gozgan.

To judge by their names, thekings of this dynasty were ofIranian stock. Although they seemto have ruled Guzgan withoutinterruption throughout the periodcovered by our documents, it isclear that they did not wieldabsolute power. During the earlypart of the 7th century, Guzgan wastributary to the khaghan of theWestern Turks. This must still havebeen the case at the time of theearliest document, a purchase-contract dated in 658/9, whichrefers specifically to the‘khaghan’s tax.’ No such tax ismentioned in the other docu-ments, all of which date from theperiod after the Western Turks

were finally defeated by the Tangat the very end of the 650s.However, China was too far awayfor the Tang to exercise effectivepower in this region, and ourdocuments do not contain anydirect reference to the Chinese. Sofar as I can see, the only trace ofChinese influence is to be found areceipt which ends with the phrase‘this is your tsak [ ],’ using aChinese word for ‘document’([ ]ce/chai [Early Middle Chinese— ]) in a context whereearlier documents of the sametype use the native Bactrian wordtsirg ‘(proof of) receipt.’

The real nature of politicalpower in this region is clarified bya letter to the Chinese emperorpreserved in an 11th centuryencyclopaedia. The author of theletter was a Turkish prince namedPuluo, a younger brother of thehereditary ruler of Tokharistan,who had his principal residence atQunduz, to the east of Balkh.Puluo was sent to live at the Tangcourt in 705. In this letter, whichwas written in 718, he mentionsthe king of Guzgan as one of the‘212 kings of various kingdoms,governors and prefects’ whorecognized the authority of hisbrother, the ruler of Tokharistan;at the same time he emphasizesthe loyalty of his brother to theTang. This state of affairs isdescribed as having existed sincethe time of his grandfather, thatis, one may presume, during thesixty years or so since the Chinesedestroyed the power of theWestern Turks and incorporatedtheir dominions into their ownadministrative framework: ‘Sincethe time of my grandfather andmy father up to that of the presentruler, the kings of Tokharistan havealways been the overlords of thesevarious countries ... The rulers ofTokharistan, from several gen-erations ago up to the present,have been sincerely devoted tothe great Tang dynasty; they havecome constantly to pay homageand to bring tribute.’

A Bactrian document dated in692/3 CE refers to two noblemen

or officials with Turkish names andtitles. One is named as ‘Ser theTurk, tudun of Gaz,’ the other as‘lord Magatur Bukla.’ It isinteresting to note that the titletudun is that given by the khaghanof the Western Turks to the officerwhom he sent to oversee the localrulers who had submitted to himand to control thepayment oftaxes. In the light of Puluo’s letterone might guess that the tudunhere carries out a similar functionon behalf of the ruler ofTokharistan, the immediateoverlord of the rulers of Guzgan.Even more interesting is the nameof ‘Lord Magatur Bukla.’ Magaturis no doubt a variant of the Turkishbaghatur, attested as a personalname and later as a term for a‘hero’ or ‘warrior,’ the source ofPersian bahâdur and Russianbogatyr’. As for Bukla [ ], Iam grateful to Yutaka Yoshida forthe very plausible suggestion thatthis name may be identified withthe the name transcribed intoChinese as Puluo ( [Early Mid-dle Chinese ]). As Prof.Yoshida points out, it is evenpossible that the Bukla in theBactrian document of 692/3 is thevery same person as the Turkishprince who arrived in Chinathirteen years later, the author ofthe letter to the Chinese emperor.

The historical informationpreserved in the Bactriandocuments is of course frag-mentary, but I hope that these fewexamples will be enough todemonstrate that they containfascinating and importantinformation, which can besupplemented and elucidated withthe help of historical andgeographical literature in Chinese,Persian and other languages.Together with the Aramaic, Greekand Arabic documents fromAfghanistan which I referred to atthe beginning of this paper, not tomention data from archaeology,numismatics, art history and otherdisciplines, they promise to makepossible a substantially newsynthesis of the early history ofAfghanistan.

8

[This article is part of a lecturegiven at the University ofWashington (Seattle) onNovember 10, 2005, in a seriesco-sponsored by the SilkroadFoundation and the SimpsonCenter for the Humanities, withadditional support from several ofthe university’s departments.References have been updated.]

About the Author

Nicholas Sims-Williams iscurrently Research Professor ofIranian and Central Asian Studiesat the School of Oriental andAfrican Studies, University ofLondon. His main interest is themedieval languages of EasternIran and Central Asia, especiallySogdian and Bactrian, and thehistory of their speakers. He hasjust completed the second volumeof his edition of the recently-discovered Bactrian documents(to appear in 2007), and is nowworking on a project on Bactrianchronology.

References

Bernard et al. 2004Paul Bernard, Georges-JeanPinault & Georges Rougemont.“Deux nouvelles inscriptionsgrecques de l’Asie centrale.”Journal des savants, 2004/2: 227-356.

de Blois forthcomingFrançois de Blois. “Du nouveau surla chronologie bactrienne post-hellénistique: l’ère de 223/224 ap.J.-C.” To appear in Comptesrendus de l’Académie desInscriptions et Belles-Lettres,séance du 2 juin 2006.

Cribb 2005Joe Cribb. “The Greek kingdom ofBactria, its coinage and itscollapse.” In: Osmund Bopear-achchi and Marie-FrançoiseBoussac, eds. Afghanistan: Anciencarrefour entre l’est et l’ouest.Turnhout (Belgium), 2005: 207-25.

Falk 2001Harry Falk. “The yuga ofSphujiddhvaja and the era of theKusânas.” Silk Road Art andArchaeology, 7 (2001): 121-36.

Falk 2004Harry Falk. “The Kaniska era inGupta records.” Silk Road Art andArchaeology, 10 (2004): 167-76.

Lee and Sims-Williams 2003Jonathan Lee and Nicholas Sims-Williams. “The antiquities andinscription of Tang-i Safedak.” SilkRoad Art & Archaeology, 9 (2003):159-84.

Minorsky 1970Vladimir Minorsky. Hudûd al-‘Âlam, 2nd ed. Clifford E.Bosworth, ed. London, 1970.

Rapin 1996Claude Rapin. “Nouvellesobservations sur le parchemingréco-bactrien d’Asangôrna.”Topoi 6 (1996): 458-69.

Salomon 2005Richard Salomon. “The Indo-Greek era of 186/5 B.C. in aBuddhist reliquary inscription.” In:Osmund Bopearachchi and Marie-Françoise Boussac, eds.Afghanistan: Ancien carrefourentre l’est et l’ouest. Turnhout(Belgium), 2005: 359-401.

Sims-Williams 1997aNicholas Sims-Williams. New lighton ancient Afghanistan: thedecipherment of Bactrian. London,1997.

Sims-Williams 1997bNicholas Sims-Williams. “Nou-veaux documents sur l’histoire etla langue de la Bactriane.”Comptes rendus de l’Académiedes Inscriptions et Belles-lettres,1996 [1997]: 633-54.

Sims-Williams 1997cNicholas Sims-Williams. “Novyebaktriiskie dokumenty.” VestnikDrevnei Istorii, 1997/3: 3-10.

Sims-Williams 1997dNicholas Sims-Williams. “Newfinds from ancient Afghanistan” (in

Japanese). Oriente, 16 (1997): 3-17.

Sims-Williams 1998Nicholas Sims-Williams. “Furthernotes on the Bactrian inscriptionof Rabatak.” In: Proceedings of theThird European Conference ofIranian Studies, Part 1, Old andMiddle Iranian Studies.Wiesbaden, 1998: 79-92.

Sims-Williams 2001Nicholas Sims-Williams. Bactriandocuments from northernAfghanistan I: Legal and economicdocuments. Oxford, 2000 [2001].(Vol. II: Letters and BuddhistTexts, forthcoming 2007.)

Sims-Williams 2004Nicholas Sims-Williams. “Nou-veaux documents bactriens duGuzgan.” Comptes rendus del’Académie des Inscriptions etBelles-Lettres, 2002 [2004]:1047-58.

Sims-Williams 2005Nicholas Sims-Williams. “Bactrianlegal documents from 7th- and8th-century Guzgan.” Bulletin ofthe Asia Institute, 15 (2001[2005]): 9-29.

Sims-Williams and Cribb 1996Nicholas Sims-Williams and JoeCribb. “A new Bactrian inscriptionof Kanishka the Great.” Silk RoadArt and Archaeology, 4 ([1995-]1996): 75-142.

Shaked 2004Shaul Shaked. Le satrape deBactriane et son gouverneur:Documents araméens du IVe s.avant notre ère provenant deBactriane. Paris, 2004.

Notes

1. For references to the earliereditions see Rapin 1996, 458 n.1.

2. For a different view see Cribb2005, 213-14.

3. Sims-Williams and Cribb 1996;revised edition in Sims-Williams1998. In 2003 I finally had theopportunity to visit Kabul and to

9

8. Some other indications mightlead one to seek Gaz further tothe east, in the mountains to thesouth of Balkh. Sources from the14th to the 19th centuries mentionhere a valley and a village calledDarra Gaz, which may beidentical with the Gazzah orJazzah of early Islamic sources.Jazzah is frequently mentioned inTabari’s account of the events ofthe year 119 of the hijra (i.e. 737CE), when the Turgesh khaghanwas defeated in battle by Asadb. ‘Abdallâh. From Tabari itappears that Jazzah was afortified place with a garrison; itwas counted as belonging toJuzjan, but it is clear from theitineraries of the combatants thatit lay in the extreme east ofJuzjan, close to Balkh. It seemsunlikely to me that such anoutlying fortress would have beena major administrative centre, orthat the ruler of Guzgan wouldhave held court there; but I haveto admit that this is the only placewithin the confines of Guzganwhich is definitely known to haveborne the name of Gaz.

Fig. 1. Bamiyan VR1= Royal City site no.1, overall plan of excavations. Copyright © 2006 DAFA; Zemaryalai Tarzi.

examine the inscription with myown eyes; as a result, a thirdedition is now in preparation.

4. Sims-Williams 1997a-d; seealso http://www.gengo.l.u-t o k y o . a c . j p / ~ h k u m /bactrian.html. Sims-Williams2001 is the first volume of acomprehensive edition of thedocuments.

5. This important point has beenestablished by François de Blois(see de Blois forthcoming).

6. I previously accepted thealternative identification ofWarnu (Greek Aornos) asKhulm, but the newly-discovered Aramaic documentsname both Khulm and Warnu,almost certainly implying thatthey are different cities.

7. This group of documents isnow edited and translated inSims-Williams 2005. For a moredetailed version of the followingdiscussion see Sims-Williams2004 (in French), but the datesof the documents given therenow need to be corrected.

8. Some other indications mightlead one to seek Gaz further tothe east, in the mountains to thesouth of Balkh. Sources from the14th to the 19th centuries mentionhere a valley and a village calledDarra Gaz, which may be identicalwith the Gazzah or Jazzah of earlyIslamic sources. Jazzah isfrequently mentioned in Tabari’saccount of the events of the year119 of the hijra (i.e. 737 CE),when the Turgesh khaghan wasdefeated in battle by Asad b.‘Abdallâh. From Tabari it appearsthat Jazzah was a fortified placewith a garrison; it was countedas belonging to Juzjan, but it isclear from the itineraries of thecombatants that it lay in theextreme east of Juzjan, close toBalkh. It seems unlikely to methat such an outlying fortresswould have been a majoradministrative centre, or that theruler of Guzgan would have heldcourt there; but I have to admitthat this is the only place withinthe confines of Guzgan which isdefinitely known to have bornethe name of Gaz.

Bamiyan 2006: The Fifth ExcavationCampaign of Prof. Tarzi’s Mission

Zemaryalai TarziThe Marc Bloch II University, Strasbourg

This was the fifth consecutive yearthat I led the French survey andexcavation mission at Bamiyan.We have to excavate in unusualand difficult conditions due to thefact that we work on privateproperty rented for the purpose ofexcavation but restored to itsoriginal cultivated state at the endof each archaeological season.Another significant reason why wemay not be able to complete ourplans is the necessity of oftenhaving to circumvent houses,walls, gardens and, most of thetime, irrigation systems. Recog-nizing the incoherence shown inthe drawings of the site, I decidedin 2006 to excavate themonuments separately, eventhough I may later attempt to linktheir relationships to one anotherin a larger context. Amongst themonuments I chose are the chapel(sanctuary) found around Stupano. 2 on the ‘Eastern Monastery’site (MO in the text), and Stupano. 1, which is the largest inBamiyan and measures approx-imately 30 m on each side. We arejust in the beginning phases ofexploration of the other sites suchas MR (Monastery of the ancientor former king), VR1 (Royal Citysite no. 1) and VR2 (Royal City siteno. 2). The results will producesignificant information on thehistory and archaeology ofBamiyan.

Choice of Sites

We have proceeded on thefollowing sites, from West to Eastin the Bamiyan Valley, as part of aplan established in 2001:

I. The Royal City site no. 1=VR1 (1st campaign of excava-tions) [Fig. 1, facing page]

The site is located on the rightside (north) of the Kabul–Band-e-Amir road where the Valley ofTchehelsotun emerges and moreprecisely at the westernextremity of the ancientBamiyan bazaar, today in ruins.As we anticipated, we could notehow the cave architecture of theancient city progressively gaveway to built architecture. In2006 we opened the site. In2007 we will build on the successof the first season and continueour work there.

II. The Royal City site no. 2 =VR2 (1st campaign ofexcavations) [Fig. 2]

The site is located to the westof the niche of the 55 m Buddhastatue, almost at the foot of thecliff itself. There we uneartheda remarkable ensemble of glass-blowing or glazing workshops.Next year we will investigate thissite further as well as sectorswithin the royal city or itsneighboring sectors.

We owe to Alfred Foucher(1923, 1925, 1942-1947) theinitial identification of theBamiyan royal city, notwith-standing subsequent andunfounded contrary propositionsmade by M. L. Carter (1985).Furthermore, as indicated by theChinese pilgrim Xuanzang(Julien 1853, Beal 1983, Watters1904 and Pelliot in Godard et al.1928), the royal city, Bamiyan’scapital, was built up against thecliff and crossed the valley. Itswall was 6-7 li long. Today theruins indeed still abut the GreatCliff and spread east to westfrom Sorkh Qol (west of the

Fig. 2. Bamiyan VR2 = Royal City site no.2, excavation plan for thenorthern part of the site. Copyright © 2006 DAFA; Zemaryalai Tarzi.

11

large 55 m Buddha) to the westof Sang Tchaspan. In this waythe city was built on both slopesof the Tchehelsotun valley. Ourexcavations, although as yetlimited in scope, have addedsignificantly to our knowledge ofthe city.

III. The Monastery of thepreceding king = MR (The 2nd

campaign of excavations didnot take place because oflandmines.)

We ran a series of surveys in2005 (1st campaign) whilemoving around without appar-ent danger. However in 2006this site, located between thetwo dynamited Buddha statues,had been entirely anddeliberately l ittered withlandmines, as indicated byseveral hundred white and redrocks. Where did the minessuddenly come from? Thequestion remains curiouslyunanswered. In 2007, we willexcavate there (2nd campaign)according to our pre-establishedplan.

IV. The Eastern Monastery =MO (5th campaign) [Figs. 3,10]

We have been excavating on thissite since 2002 (Tarzi 2003,2004 a-c, 2005). It is located tothe south of the eastern part ofthe Great Cliff and to thesoutheast of the niche of the 38

m Buddha. In 2006 we openedfour sites, extended ourexcavations around the chapelof Stupa no. 2, freed Stupa no.4, unearthed the northern sideof the Great Stupa no. 1, andbegan excavations in the ‘cloverfield’ to the northwest of Stupano. 1, practically at the foot ofthe Great Cliff.

Fig. 3. Bamiyan, MO = site of Eastern Monastery, excavation site 2005-2006.Copyright © 2006 DAFA; Zemaryalai Tarzi.

Fig. 4. Bamiyan Royal City site no. 1 (VR 1), excavations on the southern sideof the site, view from the northwest towards the southeast. Photograph copyright© 2006 DAFA; Zemaryalai Tarzi.

12

Site I. The Excavations of VR1[Fig. 4, facing page]

Our goal was to excavate a pre-Islamic part of the Bamiyan royalcity at the limit of the cave andopen air built habitat. We had avery complicated negotiation withthe owner of the chosenemplacement, an ancient serail(this word designates a caravanserail, not a Turkish palacereserved to women), located tothe west of the ancient and nowdestroyed bazaar. Thanks to thesupport of the BamiyanGovernment and of the nationalsecurity we were able to come toan agreement. We excavated adozen squares 4m x 4m. Webegan with square T15 by grottoA, to the north, and squares T16-T19 to the south. The site wasenlarged to the west with squaresR18, S18, R19, S19 and thebeginning of squares S20 and T20.

From north to south, it was veryinteresting to observe in the ruinsthe transformation of cavehabitats into built habitats. Onecan note that the serail, probablyestablished around the first partof the 20th century, revealed amore or less level terrain of whichthe archaeological layers T15, T16and T17 were in chaos. We founda broad mix of Turk pre-Islamic(6th-9th centuries) and Ghurid(12th-13th centuries) ceramics, andsome Ghaznavid, Khwarazmshahiand modern ceramics. Further tothe south the stratigraphybecomes more logical.

Without going into too muchdetail one can observe that thereare three distinct periods. The firstis Turk pre-Islamic, just prior tothe Ghaznevid; the second period— of the Ghaznevids and theKhwarazmshahs — is short, andthe site for a time was abandoned.The most flourishing Muslimperiod for this site followed, thatof the Ghurids, who establishedtheir second capital at Bamiyan.Following the Mongol hiatus — thebrutal passage of Chingis Khanand the near total destruction ofall habitats, as seen in the layers

of ashes in the excavations —Bamiyan became a township of noimportance and was repeatedlyoccupied by invaders from all over.In squares S19, T18 and T19 wefound the corner section of thebase of a fortification. There isplenty of documentation for thestudy of this Muslim period ofBamiyan (Baker and Allchin 1991;Barthold 1913, 1977; Bosworth1961, 1965, 1973, 1996; Elliot1867-1877; Gardin 1957, etc).However more research in depthshould be done. While ourcontribution has produced somesatisfying results, the excavationof this site is not finished.

The excavated area may wellencompass the Bamiyan ceramicstudios, judging from the finds ofa significant number of small clayrods used by potters in their kilns.Our knowledge of ceramics atBamiyan had previously beenlimited to a series of ceramicshards, most found on the surface,with the exception of some foundthrough the surveys on the sitesof Zohak and Gholghola cities. Theresearch undertaken prior to ourswas the work of famousspecialists, Gardin (1957), Gardinand Lyonnet (1987, and in LeBerre et al. 1987b) and finallyBaker and Allchin (1991), butshould, however, be confirmednow through excavations such asours. We hope that our

excavations will shed moredefinitive light on the subject, inview of our discovery of abeautiful and significant series ofceramics. In addition, we found inthe Muslim layers a large numberof funerary pits (containing bonesfrom bodies which had been leftexposed to the air) that had beeninserted in the constructions oragainst the walls. As CharlesMasson had already noticed in theregion of Jalalabad and as we hadalso noted in several of ourexcavations in Hadda, the practiceof inhumation of bodies in-creasingly becomes the norm inAfghanistan following the Mongolperiod (15th and 16th centuries).

Site II. The Excavations of VR2[Fig. 5]

The location of this site is to thewest of the niche of the 55 mBuddha by Sorkh Qol, practicallyat the foot of the cliff. We chosethis location in the hope of findingthe fortified walls of the pre-Islamic city of Bamiyan. We wereclose. Instead of finding thesouthern face of the fortified wall,we discovered constructions builtagainst it. A careful study of our1967 photographs shows that thisarea was used as a cultivated field,irrigated by a canal which is drytoday. A few years later it becamea pond or reservoir. The layer of

Fig. 5. Bamiyan Royal City site no. 2 (VR 2), view of the northern part of theexcavations from the Great Cliff. Copyright © 2006 DAFA; Zemaryalai Tarzi.

13

loess (pure clay) at the bottom ofthe pond is visible in our trenches.We excavated in two places: A, thefirst sector, to the north and at theimmediate south of the fortifiedwall; B, the second sector to thesouthwest of the first em-placement of a tower also visiblein our 1967 photograph.

A - Glass workshops (?) [Figs. 6,7]

In this sector we excavated eight4m x4m squares, C5-C7, D5-D7and E5-E7. The chronologyremains the same as for VR1.However the positioning of thestructures does not change: theysuperimpose one another from theend of the Turk period until the endof the Ghurid period. Two largeensembles were excavated.Because all of the constructionshave not yet been fully unearthedit is difficult to establish therelationship of one to the other.Nevertheless we are certain thathere is an ensemble of workshopswhere the kilns (vertical), such assmall size tandoors, most likelyserved as smithies or hearths

rather than ceramic kilns. Wefound a grinding device in situ,with a fixed (inferior) circularmillstone and a mobile superiormillstone. The floor with itsremains of glass paste (pâte deverre) and glass shardsoriginating from the making ofvases suggest that these wereworkshops for glass manu-facture. There were as well twoovens or smithies and twostorage pits.

Until now no serious studyhas been done regarding theindustry of glassmaking duringthe Muslim period in Afghanistanand specifically in Bamiyan.Readers may wish to refer to myresearch (Tarzi 2001, 97-99)concerning this craft innorthwestern India and itsrelationship with China and theMuslim world.

B – The Sorkh Qol tower

Based on my own recollectionsand 1967 photographs, the twoSorkh Qol towers reminded me ofsites in Chinese Turkestan. Ithought then they might belongto the fortifications of the royalcity, the pre-Islamic capital ofBamiyan. Since these two towersnow have been destroyed, Iwanted to find their locations inorder to excavate their found-ations and establish theirrelationship to the wall thatconnected them. Nothing is left ofthe first tower located to thenorthwest. Itsmaterials weretotally repro-cessed for theconstruction of ahouse which isnow inhabited.The site of thesecond tower,larger than thefirst and furtherto the southeast,r e m a i n sunoccupied. Wewere told by theinhabitants that,prior to its finaldestruction, all

that was left were the west andsouth walls and the southwestcorner of this tower, which hadobviously been built on a squareplan. In addition, during its longperiod of abandonment, it hadbeen used to store hay and foodfor cattle; its plan had undergonesuch degradation that it hadalmost became circular. A furthercomplication for this investigationis the fact Sorkh Qol is located inthe path of streams which flow intothe valley on the weather’s whim.Thus, as we observed, a majorstorm of 2005 had devastated theirrigation channels in the area.

The raw bricks fallen from thetwo walls had the followingdimensions: 38x20x6 cm,33x20x6 cm, 31x19x6 cm,27x20x6 cm. We discovered thatthe foundations of the tower wereplaced on a sandy ground in whichwe found many glazed shardsdating from the Muslim period.Near the eastern part of the towerdeeper excavations (1.40 m)unearthed only the foundationseat of a wall, aligned northeastto southwest, 0.70 m in width, cutin its center by the foundation seatof a second wall, itself positionedeast-west and 0.40 m in width.The two walls formed angles of 25o

and 65o. At the same level of thesewalls but further to the northwestof our excavation, we discovereda more or less rectangular tiledgutter. It is probable that we arein the courtyard of a largeconstruction. Unfortunately, given

Fig. 6. Bamiyan Royal City site no. 2(VR 2), general view of the glass orceramics workshop. Photograph copy-right © 2006 DAFA; Zemaryalai Tarzi.

Fig. 7. Bamiyan Royal City site no. 2 (VR 2), glass work-shop, detail, showing lower part of a millstone. Photo-graph copyright © 2006 DAFA; Zemaryalai Tarzi.

14

rapid erosion due to the streams,excavations in 2007 may bringonly minor results.

Site IV. The Excavations at MO

Since 2002 we have beenexcavating the site of the EasternMonastery referred to in our textas MO. At the end of eachexcavation campaign we re-burythe remains on site under severalmeters of soil. The first two weeksof excavations at MO were devotedto the costly and difficult task ofreopening squares previouslyexcavated. We enlarged GalleryA9 in which we had previouslyfound many sculptures, largelycomposed of clay moldings withpolychrome intact. Our discoveriesthis year however were moremodest than in previous years: aclay statue representing a sittingBuddha in meditation [Fig. 8], thebottom part of a standing claystatue [Fig. 9] in situ, and manymore fragments. Our efforts herewere concentrated on therestoration of previously un-earthed moldings too fragile to bemoved. We asked the DAFA

( D e l e g a t i o nd’Archéologie Françaiseen Afghanistan) to sendus their restorationexpert to demonstratethe use of newpolymerized glass-based products for theconsolidation of clay. Inthe same central sectorof Massif A, weexpanded our excava-tion squares aroundStupa no. 2 in order toacquire a good and easyread of the chapel’s planthat surrounds it. Firstwe had to unearth thewestern side of thestupa and pursueexcavations on thenorthwestern sides. During thisseries of excavations wediscovered a clay votive stupa tothe west (Stupa no. 5), two otherclay votive stupas (Stupas nos. 6-7), and a series of benches onwhich presumably had stood 22large clay Buddha statues. Of thisnumber, 14 pairs of feet and 5

isolated feet are preserved in situ;some are preserved to the knees.

Chapel A (= Caitya A), aroundStupa no. 2 [Figs. 10; 11, 12 nextpage]

In order for us to acquire a betterunderstanding of the architecture

Fig. 8. Bamiyan Eastern Monastery(MO), Gallery A9. Clay molding withpolychrome, representing a sittingBuddha in meditation. H 43 cm, W29 cm at shoulder. Inv. no. BAM.V.06.1. Photograph copyright ©2006 DAFA; Zemaryalai Tarzi

Fig. 9. Bamiyan Eastern Monastery (MO). Bottompart of clay Buddha statue along W wall of GalleryA9. Photograph copyright © 2006 DAFA;Zemaryalai Tarzi.

Fig. 10. Bamiyan Eastern Monastery (MO). Detail of excavation plan, showingCaitya A. Copyright © 2006 DAFA; Zemaryalai Tarzi.

15

of the site and ensembles in theBamiyan Valley, we first studiedthis monastic complex — a caitya,a sort of a large cruciform chapelplaced in a square plan —separately from its surroundingcontext of the other constructionsof the MO. This large caitya wascommunicating with Gallery A9 tothe east, which in turn gaveaccess to a similar yet probablysmaller complex that had in itscenter Stupa no. 4 (see below).To the north, it communicatedwith other galleries and corridors.To the south, it was delimited byan attached portico where wefound significant traces of fire (cf.2005 report). Excavation of theMO1 square gave us more preciseinformation on the western wallof this ensemble.

Caitya A is surrounded by fourwalls, 2.15 m thick. At each of the

four corners is a cell 3,50m x3,50m square (cells N, NE, SW,SE). A corridor parallel to the‘enclosing’ wall joined each cell tothe next one. These four cornercells and their communicatingcorridors had their inner sidesdefined byenormous blocksof masonry simi-lar to the doubletowers of thecorners of theTall-i Takht forti-f i c a t i o n s(Stronach 1978)and the NeeRoyal city( G h i r s h m a n1946). Thesedouble massifswere also rein-forced by thecorners of thecaitya’s centralcourtyard. The

width of the corridor was reducedat the level of the masonry blocks,thus creating large pilastersagainst the enclosing wall. We didnot find any traces of fire or asignificant amount of compactedclay on the floor of the corridor(s)or of the four corner cells, whichbrings us to conclude that theircovering was not made out ofwood. In this case one cansuppose that each of the fourcorner cells was roofed by a cupolaresting on squinches and thecorridors by barrel vaults.

The nearly square centralcourtyard measured 9.50 m on aside. In its center is Stupa no. 2,also with a square plan andmeasuring 5.25 m. Therefore thecircumambulation path (pradak-sinaptha) did not exceed 2.20 min width.

In the middle of each side ofthe courtyards each arm of the‘cross’ is a structure like an ‘iwan’opening on Stupa no. 2 (iwan N,S, E and W). Logic would suggestthat each of these iwans becovered with a barrel vault.However signs of fire at the levelwhere there were niches tosupport beams and joists provethat each iwan was covered witha wooden roof and had anattached portico. The Pradak-sinapatha around Stupa no. 2initially had a schist or slatedlimestone ti led floor, whichsubsequently was covered with a

Fig. 11. Bamiyan Eastern Monastery (MO), Gallery A9. View of chapel (caitya)from E with Stupa no.2 in its center. Photograph copyright © 2006 DAFA;Zemaryalai Tarzi.

Fig. 12. Bamiyan East-ern Monastery (MO), NEcorner of courtyard ofStupa no.2, with pairs offeet on benches. Photo-graph copyright © 2006DAFA; Zemaryalai Tarzi.

Fig. 13. Bamiyan Eastern Monastery (MO), Stupa no. 2from SW. Photograph copyright © 2006 DAFA; ZemaryalaiTarzi.

16

layer of lime stucco, as also wasStupa no. 2. Finally, probablybetween the 6th and 9th centuriesCE, the floor was completelycovered by a layer of clay whichhas retained the traces of fire.

Stupa no. 2 [Fig. 13, facing page]

In 2005 we excavated three sidesof this stupa. Two (east and south)were unearthed completely andthe north partially. In 2006 wecompleted excavation of themonument. Its form and its stuccocoating links it to the stupas ofGandhara and more specifically ofHadda, the location of the firststupa to be found to the north ofthe Hindu Kush. It consists of aplatform on a plinth, a torus, scotiaor cavetto of the ‘classical’ type ofHadda, and on each side, sevenpilasters, two of which are locatedat the corners. Each pilaster is ona molded base composed of aplinth, a torus, and cavetto orhollow molding. The shafts of thepilasters are without anydecoration. The capitals are of theCorinthian type with a row ofpolylobed acanthus leaveshanging on the abacus [Fig. 14].

In the lower levels then, there isno difference from the Gandhara

stupas. At the upper level though,there is something new: on eachside of the stupa are threerelatively deep niches on top ofmolded bases, flanked by pilastersor columns whose circular, moldedbases are like ionic bases. The fewarchitectonic elements still left onthis row are very interestingindeed, but will require furtherinvestigation in order to give acomplete description.

Despite the deterioration of itsexterior, Stupa no. 2 had not beendisturbed by illegal excavation inthe past. So we excavated itscenter vertically, in the hope ofdiscovering a reliquary and itsrelics. We were absolutelysurprised at the finds. In thesoutheastern corner of the stupa,within the compact masonry,about 0.50 cm from the top of theedifice, we found a bronze bell, abronze earring with screw andchiseled rings and a thin goldcircular plaque in the shape of aring [Fig. 15]. At the same depthtowards the northwest, wediscovered ivory beads, finestones and gems, and further tothe west we found 5 more bronzerings, one of whichseems to be a ringwith a bulge for abezel, a smallbronze coin, etc. Ourdiscoveries ceasedat about 1.60 cm,almost at the groundlevel of Stupa no. 2.At that level we alsofound severalfragments of verysmall gold petalsand a small and verythin bronze coinbroken into several

u n i d e n t i f i a b l efragments. We might interpretthese findings in the following way.It may be that the persons incharge of the restoration of thisstupa, between the 6th and 9thcentury CE, left the relics indisarray. Such has been notedpreviously, for example in Butkara(Facenna 1980-1981), whererelics were divided and placedbehind the niches of a monumentinstead of inside a reliquary. Thiswas exactly the case with ourStupa no. 4 to which we shallreturn later. In conclusion I shouldnote that, while we attempted toreach a deeper floor such as thatattained in A6 in 2003, weexcavated Stupa no. 2 only to adepth of 550 cm. As our workersbecame increasingly endangeredin the narrowing trench of theexcavation, we had to cease ouractivities.

Stupas nos. 3, 5-7, around Stupano. 2

Stupa no. 3 [Fig. 16]

Stupa no. 3 was partiallyunearthed last year in the eastern

Fig. 14. Bamiyan Eastern Monastery(MO), one of the pilasters of stupano. 2. Photograph copyright © 2006DAFA; Zemaryalai Tarzi.

Fig. 15. BamiyanEastern Monas-tery (MO), a partof the excavatedrelics from Stupano. 2. Photographcopyright © 2006DAFA; ZemaryalaiTarzi.

Fig. 16. Bamiyan Eastern Monastery (MO), NW cor-ner of votive Stupa no. 3, detail. Photograph copy-right © 2006 DAFA; Zemaryalai Tarzi.

17

‘iwan’ or arm of the cross of ChapelA. In 2006 we unearthed itseastern side and its eastern andnorthern stairs. It is a miniaturizedclay votive stupa, preserved ontwo levels only, since it waspartially crushed due to thecollapse of the walls and the roof.Its plan is square, but with its fourstairs and their landings it has acruciform shape. In some placesit preserves its entire painted reliefdecoration, l ittle columns,modillons etc. Its north-southlength measures 196 cm and itseast-west length measures 199cm. Our stupa resembles the clayvotive stupas of Tape Sardar atGhazni (Taddei 1985). There is adifference, however, in that thestupa in Bamiyan is not placed ina lotus flower as is the stupa atGhazni. Stupa no. 3 of Bamiyan issimilar to the stupas of the kindrepresented on the baked clayplaques of Kashmir (Kak 1933)and the painted murals of theBamiyan grottos. Prof. Taddei haddated the second period of TapeSardar from the 6th to the 7th

century CE (Taddei 1968). Thedestruction of Bamiyan in the 9th

century CE is historically attested.It is therefore reasonable to datethe construction of Stupa no. 3 ofthe MO of Bamiyan between the6th and 9th century CE.

Stupa no. 5 [Fig. 17]

This stupa is the twin of Stupa no.3, almost of the same dimensions,better preserved in its form butless so in its miniaturized reliefdecoration. It is to be noted thatsome panels on the façade wereadorned with images: seatedfigures in a nimbus at the lowerlevel and standing figures in animbus at the upper level, possiblyimages of the Buddha in miniatureand made in the pure tradition ofthe Bamiyan artistic school.

Stupas nos. 6 and 7 [Fig. 18]

These two clay votive stupas werebuilt on an irregular square plan.One side of the square measuresapproximately 115 cm. Two levels

were preserved; the higher levels,as with Stupas nos. 3 and 5, werecompletely crushedby the collapse ofthe walls and roofs.Their lower levelsconsist of a platformon a plinth and anunadorned scotia.Their upper levelshave the beginningof a square bodyadorned by pilasterswhich have dis-integrated. This typeof stupa is oftenfound in Haddawhere it is usuallystuccoed.

Stupa no. 4 [Fig. 19]

We anticipated we would find astupa in this location (squares NEB20 and NE C20). In order toreach it we had to demolish a wallseparating two fields. With theconsent of the fields’ owners, MM.Bahauddin and Khan Mohammadwe proceeded without damaginga very old poplar tree. We weredisappointed with the results ofthe excavation because the stupahad been extensively plundered inthe past. Was this act ofvandalism, dating probably fromthe Ghurid period, due to thesalvaging of the construction stoneor the limestone coating, or wasthis systematic destruction aimedat recovering the reliquary hidden

Fig. 17. Bamiyan Eastern Monastery (MO), clay votive Stupa No. 5, located Wof Stupa no. 2. Photograph copyright © 2006 DAFA; Zemaryalai Tarzi.

Fig. 18. Bamiyan Eastern Monastery (MO),clay votive Stupas nos. 6 and 7, located Nof Stupa no. 2. Photograph copyright ©2006 DAFA; Zemaryalai Tarzi.

Fig. 19. Bamiyan Eastern Monastery (MO), Stupano. 4. Photograph copyright © 2006 DAFA;Zemaryalai Tarzi.

18

behind the niches of themonument? Indeed, in con-firmation of our secondhypothesis, we found at 200 cm areliquary [Fig. 20]. This littlereliquary — a small globular pot,7 cm high with a 5.5 cm maximum

diameter — contained 5 beads, 2of which were totally decomposedmarine pearls and one of whichwas an ochre-orange soft stone ina disk shape with a hole in itscenter [Fig. 21]. The lid sealing the

relics was madeof a type ofd e c o m p o s e dmastic; we willknow more about

its origin when the results comeback from the laboratory.

Stupa no. 1 (western survey)

This survey was done differentlyfrom our usual square pattern.The goal was to find the north-west corner of the Great Stupa no.1 in order to get the exactmeasurement of one side of thesquare plan for this monument.We established that the northernside measures 29-30 m. This verylarge survey of more than 10 m inlength improves our under-standing of this monument andthe site of MO [Fig. 22]. In additionthe excavation revealed thewestern landing of the northernstairs of the stupa [Fig. 23]. Twoother discoveries should be noted:

• It looks as though in its firststate of decoration the stupa hada stucco coating.

• It is also important to explainwhy around the base of thestupa there is a considerablethickness in the layer left by thefire. This layer is composed ofburnt clay fragments, burntwood and many iron or bronze

Fig. 20. Bamiyan Eastern Monastery(MO), Stupa no. 4. Reliquary in theshape of a small globular pot. H 5.7cm, Dia. 5.5 cm. Inv. no. BAM, V.06-187. Photograph copyright © 2006DAFA; Zemaryalai Tarzi.

Fig. 21. Contents of the reliquary BAM,V.06-187: beads, agate, ivory and twosmall pearls. Photograph copyright ©2006 DAFA; Zemaryalai Tarzi.

Fig. 22. Bamiyan Eastern Monastery (MO), Stupa no. 1, largesurvey to the northwest of the monument. Photograph copy-right © 2006 DAFA; Zemaryalai Tarzi.

Fig. 23. Bamiyan Eastern Monastery (MO), Stupano.1, the west landing of the northern stairs of thestupa. Photograph copyright © 2006 DAFA;Zemaryalai Tarzi.

19

fragments [Fig. 24]. It is quitepossible that Stupa no. 1 wasencircled by a wooden railing(vedika or pado-vedika) sup-ported in some places by claymasonry.

Northwest Sector (excavations inthe clover field)

For several years I wanted to opena site that would extend thenorthern limits of the MO to thenorth and at the foot of the GreatCliff. Reaching an agreement withthe owners took a long time, butwe eventually were able to openthe site on August 13th. Our guestof honor, Roland Bezenval,Director of the DAFA, broke theground with the first shovelful ofdirt. It became a very largesurvey, 25m x 2.5m, that crossedthe field south to north. Due to thecollapse of the cliff in the northernpart of the survey, there one hasto go at least 7 m deep to obtainresults. Given the size of the taskwe will have to pursue theoperation in 2007.

The Chronology

The pre-Islamic period

In 2003 and 2004 we discoveredthe two great periods of Bamiyanaround Stupa no. 1 (BN, BS, BWand A6 and A8). I had initiallyestimated the first period to be ofthe 3rd century CE. However,thanks to the discovery in 2004 ofceramics and of an Indian coin

with a hill andthree arches,the date canbe pushedback to the 2nd

century CE(Rapson 1908;Allan 1936;M i t c h - i n e r1976). I datedthe secondperiod to the6th century CEin my firstthesis onBamiyan (Tarzi1977). This

date corresponded to that of theconstruction of the 55 m Buddhastatue and with the restoration ofthe ensemble of the 38 m Buddhastatue. It is to this second periodthat I attribute the realization andextension of the ‘EasternMonastery’ (MO), in which wehope to find the 1000-foot-longreclining Buddha statuementioned by Chinese pilgrimXuanzang. The carbon 14 analysisof vegetal and animal fibers fromthe coating of the 38 m Buddhastatue done by the German teamin charge of the protection of thefragments of the two large statuessupports my proposed dating,viz.: the second great period ofBamiyan begins in the middle ofthe 6th century CE. When we reachthe level of the fire of the 9th

century CE, I will ask the Germansfor more analysis. What is certainis that the moldings unearthed inGallery A9 date from before the6th century CE. I have carried outa careful comparison with the claymoldings of Hadda on the site ofTape Shotor (Tarzi 1976) and ofTape Sardar in Ghazni —excavated by my colleagues andfriends Profs. Maurizio Taddei(1968, 1981, 1985) and GiovanniVerardi (1981, 1989). Based onstylistic and technical criteria inthe analysis of the pre-moldedclay curls, I am able to date theseto the 3rd - 5th century CE.

Because the history of CentralAsia is yet inadequately knownand no dates are absolutely

certain we have to proceed withcaution when attempting toexplain a chronological hypo-thesis. The only writteninformation available to us is in theChinese and Muslim sources thatshed some light on the secondgreat period of Bamiyan. It is ourexcavation data which enable usto determine the first great periodand place it between the 3rd andthe 5th century CE, with thepossibility that we can extend thisperiod back to the 2nd century CE.Of particular interest in thischronology is to learn what causeda general degradation involvingthe profanation of many statues.One could attribute thisdestruction to the Sasanians in thetime of Shapur or to theHephthalites.

We know for certain that thehiatus between the first andsecond periods is rather large andtranslates into an accumulation ofsoil and embankment 1 m – 1.5m thick. Also certain is that thesecond great period of Bamiyancorresponds to its seizure by thewestern Turks. Indeed whetherthe central power at Bamiyan isHephthalite or local (Tajik), politicsof Central Asia at the time weregoverned by the Western Turks,the same ones who askedXuanzang to go through Balkh andBamiyan, two cities that were noton the initial itinerary of theChinese pilgrim.

The ceramics and chronologicalprecision

The study of the ceramics weexcavated remains to be done.Yet the discovery of shards foundin the site of MO along its long wallV, in survey N (north), has alreadyproduced valuable information(Tarzi 2005). These pre-Islamicceramics, which we will detail later,consist of two lots: one from the2nd – 5th century CE and anotherfrom the 6th – 9th century CE.Indeed it is to Ya’qub bin Laith as-Saffar in 871 CE that I attributethe end of life at the MO, thedestruction of Bamiyan in general

Fig. 24. Bamiyan Eastern Monastery (MO), Stupa no.1: tracesof fire, 9th century CE. Photograph copyright © 2006 DAFA;Zemaryalai Tarzi.

20

and of the statues and theirtemples. Further study of the firstgroup of ceramics needs to bedone. The second lot is composedof shards generally similar to theones found by Le Berre in thefortified castles of the Hindukushand published by Jean-ClaudeGardin and Bertille Lyonnet (in LeBerre 1987a, b) and Piers Baker(1991).

It is with too much caution thatthese ceramics are termed ‘turco-hephthalite’ and attributed both tothe Hepthalites and the Turks.Hephthalite ceramics, generallymade in the Kushan tradition, aredifficult to identify. In someregions they seem to be inspiredby the Kushano-Sasanids, and insome cases can be included in theBegram III group (Kuwayama1991). In Hadda, for example,nothing allows us to distinguishbetween the Kushan andHephthalite ceramics. On theother hand, there the Turkceramics are obvious and specific.These hand-made ceramics havebeen very well analyzed by Gardinand Lyonnet, both from thestandpoint of techniques ofshaping and polishing and the useof under-glazes, and from thestandpoint of the symbols andmotifs in the painted designs,which reflect a connection with thepeoples of the steppes. Theseceramics appear at a later date(second half of the 6th century CE)at a time when the Hephthalites

had been politically underminedby the fleeting coalition of theSasanians and Western Turks [Fig.25]. However, in our researchregarding Bamiyan in particularand more generally Central Asiato the northwest of India, we finda discrepancy between the hand-made Turk ceramics and the dozenor so surviving issues ofHephthalite coinage.

We do not yet have definitiveanswers regarding the Heph-thalites and their art. However,there is no reason to doubt thewell founded hypotheses of mycolleagues Roman Ghirshman(1948), Robert Göbl (1967) andShoshin Kuwayama (1989, 2002).We should continue to studyclosely the Hephthalites takinginto consideration the Tape Sardar(Ghazni) Italian excavations, the2004 and 2005 Afghan ex-cavations at Khwadja Safa andTape Narendj in Kabul, and myown excavations in Hadda. Tothese we should add the DAFAexcavations at Tape Alghata nearthe village of Dadal (Myadan-Wardak province) and myexcavations in Bamiyan. Sadlythere has also been a series ofillegal excavation in Khawar,Messe Aynak of Logar or Khord inKabul, and many other sites. Ourduty is to provide documentationso that specialists will be able tostudy and interpret our databaseobjectively. Until our futureexcavations uncover Hepthalitecoins we will continue to attributethe pre-Islamic art of Bamiyan tothe post-Kushan period, includingthe Kidarites, Hephthalites andTurks.

The Islamic Period

The ceramic and numismaticevidence

A. Ceramics

The transition between the pre-Islamic and Islamic period isrepresentative of the history of theBamiyan Valley. One should takeinto account the writings of

historians such as Yaqubi andTabari, who describe the gradualshift from one religion to anotheramong the local population. Suchstudy is necessary to understandwhy we find in our excavationsTurk ceramics of the early Islamicperiod. Despite attacks by theSamanids and the Ghaznevids,several Sardars (Sar = lords,wrongly labeled by the Arabs asSher) managed to stand up tothem and remain loyal to theirreligion, Buddhism, until 987 CE.This is probably why the ceramicsof Bamiyan during the Samanidand Ghaznevid periods occupy amodest place [Fig. 26] among ourfinds. On the other hand theGhurid period is amplyrepresented in its ceramic forms

and their new decorations, all ofwhich are very importantregarding the history of the region[Figs. 27-29, next page]. Oncemore thanks to historical texts,especially the Tabakat-i Nasiri ofDzuzdjani (Dzuzdjani 1963-1964;Raverty 1881), and studies byBosworth (1961) and Barthold(1977), we learn that Bamiyanserved as a capital for theShansabani kings of Ghur, startingwith Fakhr al-Din Mascud (1163CE) and ending with Djalal al-DincAli (1213 CE) and ‘Ala’ al-DinMuhammad. We know that the cityof Gholghola also served ascapital. The royal Ghurid cityextended over a kilometer fromthe south where today’s Bamiyan

Fig. 25. Bamiyan Royal City site no. 2(VR 2),Turk ceramic (BAM.V.06.”tower”—140 cm). Photograph copy-right © 2006 DAFA; Zemaryalai Tarzi.

Fig. 26. Bamiyan Royal City site no. 1(VR 1), ceramic with painted geomet-ric and vegetal decoration in relief inthe Muslim layers (12th century CE?)(BAM.V.06.MS18). Photograph copy-right © 2006 DAFA; Zemaryalai Tarzi.

21

airport is to the edge of theplateau on the north and wascalled Tape Almas. Gholghola wasits citadel. This city and its citadelwere taken first by theKhwarazmshahs (1215 CE) andlater by Chingis Khan.

According to Gardin (1957), theKhwarazmshahi and Ghuridceramics originating fromGholghola city were the productof an imported Iranian workshop,whose brief duration lasted fromapproximately 1175 into the 13thcentury CE. As Gardin also noted,these ceramics were not limited

to the city ofGholghola. Wehave to investi-gate further whythere is such anabundance ofMuslim ceramics

in Bamiyan onseveral levels and inmany places in thevalley. Once wefinish our photoprocessing for theexcavations of 2006,we will be able totackle this questionmore seriously.

B . N u m i s m a t i cevidence

To date we have notcleaned the coins wefound. In 2006 wefound no Greek orKushan coins. Wefound several thinMuslim bronze coins

probably from the Ghaznevidperiod, a few Ghurid coins, andeven a few from the modernperiod such as the reign of theformer king of AfghanistanMohammad Zaher Shah alMotawakelellalah.

Conclusion

It is a challenge to draw aconclusion from this scientificpatchwork without being able toinclude carbon 14 analysis or otherlaboratory test results. Thanks tothe discovery of ceramics and

coins in our excavations of 2003-2005, we know that the beginningof Bamiyan was around the 2nd or3rd century CE. The discovery of abronze coin depicting the imageof a Greco-Bactrian king (first halfof the 2nd century CE) is veryimportant despite the fact that itwas not found in the earliest layersof Bamiyan. Further investigationswill be necessary. Our exca-vations, especially of the Turkceramics, have shed light on theend of the Buddhist period of thevalley around the 9th century andon the relationship between thispre-Islamic period and thebeginning of the Islamic period.One can say that this phase ofarchaeological uncertaintyconcerning the transition betweenthe two great periods is areflection of the history ofBamiyan, when its Buddhistsovereigns converted to Islam andits monuments devoted to the cultof Sakyamuni were brutallydestroyed by Ya’qub bin Laith as-Saffar in the 9th century CE.Nonetheless, Buddhism would laststill until the Ghaznevid period.

Further study of the Muslimceramics and the study of thenumismatic evidence will enableus to establish more precise dates.We have found so far Ghaznevid,Khwarazmshahi and Ghurid coinsdating from the 10th-13th centuriesCE right up to the abandonmentand destruction of the site byChingis Khan. Once we havecleaned these coins it will bepossible to date with more

Fig. 27. BAM.V.06.MS20-160 cm.

Bamiyan Royal City site no. 1 (VR 1), ceramics withpolychrome of the Muslim period. Photographs copy-right © 2006 DAFA; Zemaryalai Tarzi.

Fig. 29. BAM.V.06.MS20-160 cm (12th century CE?).

Fig. 28. BAM.V.06.MS20-160 cm (12th century CE?).

22

precision the stratigraphy, whichwill then make possible the use ofthe glazed ceramics as anindispensable tool in identifyingthe archaeological layers weencounter in Bamiyan.

About the Author

Born in 1939 in Kabul, ProfessorZemaryalai Tarzi completed hisstudies under the supervision ofProfessor Daniel Schlumberger, inthe process obtaining three PhDs.From 1973 to 1979, he wasDirector General of Archaeologyand Preservation of HistoricalMonuments of Afghanistan. Helater directed the excavations inBamiyan and Hadda on the sitesof Tape Shotor and Tape TopeKalan. Exiled to France in 1979,he is currently Professor of EasternArchaeology at the Marc Bloch IIUniversity of Strasbourg, France,and will go to Kabul University inearly March 2007 to teach theirarchaeology Master’s students inFarsi. He is Director for the FrenchArchaeological Missions for theSurveys and Excavations ofBamiyan, the mission having beenco-funded by the NationalGeographic Society since 2004.Prof. Tarzi is the author of someseventy-five articles and books.He is also President for theAssociation for the Protection ofAfghan Archaeology, Inc. (SanFrancisco, California), whosemanaging director and founder ishis daughter Nadia Tarzi. For moreinformation please contactinfo@apaa.info and/or visit http://www.apaa.info.

References

Allan 1936

John Allan. Catalogue of the Coinsof Ancient India. London: BritishMuseum, 1936 (reprinted 1967).

Allchin 1972

F. Raymond Allchin. “A cruciformreliquary from Shaikhan Dheri.”In: Pratapaditya Pal, ed. Aspectsof Indian Art. Papers Presented ina Symposium at the Los Angeles

County Museum of Art, October,1970. Leiden: Brill, 1972: 1-26.

Allchin and Hammond 1978

F. Raymond Allchin and NormanHammond, eds. The Archaeologyof Afghanistan from Earliest Timesto the Timurid period. London;New York: Academic Press, 1978.

Abu al-Fazl 1939

Abu al-Fazl ibn Mubarak. The A’in-i Akbari [by] Abu’l-Fazl ‘Allami.Henry Blochmann, tr. Douglas C.Phil lott, ed. 2nd ed. revised.Bibliotheca Indica, N.S., Work no.61, Vol. 1. Calcutta: Royal AsiaticSociety of Bengal, 1939.

Ball 1982

Warwick Ball, avec la collaborationde Jean-Claude Gardin. Archae-ological Gazetteer of Afghanistan.Catalogue des sites archéolo-giques d’Afghanistan. 2 tt. Paris:Editions Recherche sur lescivilisations, 1982.

Baker and Allchin 1991

Piers H. B. Baker and F. RaymondAllchin. Shar-i Zohak and theHistory of the Bamiyan Valley,Afghanistan. Ancient India andIran Trust Series, No. 1. Oxford:BAR, 1991.

Barthold 1913

Wilhelm Barthold (V. V. Bartol’d).“Bamiyan.” The Encyclopaedia ofIslam, Vol. 1 (Leiden: Brill;London: Luzac, 1913): 643-4.

Barthold 1977

Wilhelm Barthold (V. V. Bartol’d).Turkestan down to the MongolInvasion, 4th ed. E. J. W. GibbMemorial Series, N. S., 5 (1e

ed.1928), London: E. J. W. GibbMemorial Trust, 1977 (reprint ofrevised and expanded 3rd ed.,1968; 1st ed. 1928).

Barthold and Allchin 1960

Wilhelm Barthold and F. RaymondAllchin. “Bamiyan.” The Encyclo-paedia of Islam, New ed., Vol. 1(Leiden: Brill; London: Luzac,1960): 1009-1010.

Barthoux 1930

Jules J. Barthoux. Les fouilles deHadda. T. 3. Figures et figurines,

album photographique. Mémoiresde la Délélegation archéologiquefrançaise en Afghanistan [Mém.DAFA], 4. Paris: G. van Oest,1930.

Beal 1983

Samuel Beal, tr. Si-yu-ki. BuddhistRecords of the Western World.Translated from the Chinese ofHiuen Tsiang AD 629. 2nd ed. NewDelhi: Oriental Books ReprintCorp., 1983 (1st ed., 1884).

Beal 1888

Samuel Beal, tr. Life of Hiuen-tsiang by the Shamans Hwui li andYen-tsung, with a prefacecontaining an account of the worksof I-Tsing. London: Trübner, 1888.

Breshna 1972

A. G. Breshna. “A Glance at theHistory of Fine Arts inAfghanistan.” Afghanistan 25/3(1972): 11-21.

Bivar 1971

A. D. H. Bivar. “Hayatila.” TheEncyclopaedia of Islam, New ed.,Vol. 3 (Leiden: Brill; London:Luzac, 1971): 303-304

Bivar 1998

A. D. H. Bivar. “The SasanianPrinces at Bamiyan.” in: Vesta S.Curtis, et al. eds. The Art andArchaeology of Ancient Persia:New Light on the Parthians andSasanian Empires. London; NewYork: I. B. Tauris, 1998: 103-110.

Bosworth 1961

Clifford E. Bosworth. “The earlyIslamic history of Ghur.” CentralAsiatic Journal, 6 (1961): 116-133.

Bosworth 1965

Clifford E. Bosworth. “Ghurids.”The Encyclopaedia of Islam, Newed., Vol. 2 (Leiden: Brill; London:Luzac, 1965): 1099-1104.

Bosworth 1973

Clifford E. Bosworth. TheGhaznavids, Their Empire inAfghanistan and Eastern Iran,994-1040. 2nd ed. Beirut, 1973(1st ed., 1963).

Bosworth 1996

Clifford E. Bosworth. Les dynasties

23

musulmanes, trad. de YvesThoraval. Paris, 1996 (original ed.:The Islamic Dynasties: AChronological and GenealogicalHandbook. Edinburgh, 1967)

Carter 1985

M. L. Carter. “Hsüan-tsang and theColossal Buddhas at Bamiyan.” In:A. K. Narain, ed. Studies inBuddhist Art of South Asia. NewDelhi: Kanak, 1985: 117-125

Chavannes 1903

Edouard Chavannes. Documentssur les Tou-Kiue (Turcs) Occi-dentaux. Paris: Libraire d’Amer-ique et d’Orient, 1903

Dupree 1958

Louis Dupree. Shamshir Ghar:Historic Cave Site in KandaharProvince, Afghanistan. Anthropo-logical Papers of The AmericanMuseum of Natural History, Vol.46, pt. 2. New York, 1958.

Dupree and Kolb 1972

Louis Dupree and Charles C. Kolb.“Ceramics from Aq Kupruk, Darra-i-Kur and Hazar Gusfand.”Transactions of the AmericanPhilosophical Society, N.S., 62/4(1972): 33-42 (reprinted inAfghanistan, 27/4 (1974): 45-69).

Dupree 1967

Nancy Hatch Dupree. The Valleyof Bamiyan, 2nd ed. Kabul: AfghanTourist Organization, 1967.

Dzuzdjani 1963-1964

Dzuzdjani. Tabakat-i Nasiri. Newed. by Abd al-Hayy Habibi. 2 vols.Kabul, 1342-43/1963-64.

Elliot 1869

Henry M. Elliot. The History ofIndia as Told by Its OwnHistorians: The MuhammadanPeriod. John Dowson, ed. 8 vols.London: Trübner, 1867-1877.

Faccenna 1980-1981

Domenico Faccenna. Butkara I(Swat, Pakistan) 1956-1962.Instituto italiano per il medio edestremo Oriento, Centro studi escavi archaeologici in Asia, Reportsand memoirs, Vol. 3, pts. 1-3.Rome: IsMEO, 1980. Vol. 3, pts.4-5/2. Rome: IsMEO, 1981.

Foucher 1923

Alfred Foucher. “Rapport A.Foucher” [A letter addressed to E.Senant about the antiquities ofBamiyan]. Journal Asiatique, avril-juin 1923: 354-368.

Foucher 1925

Alfred Foucher. “Notes surl’itinéraire de Hiuan-tsang enAfghanistan.” In: Etudes Asia-tiques publiées à l’occasion duvingt-cinquième anniversaire del’Ecole Française d’Extrême-Orientpar ses membres et sescollaborateurs. T. I. Publicationsde l’École française d’ExtrèmeOrient, t. 19. Paris: G. Van Oest,1925: 257-284.

Foucher 1942-1947

Alfred Foucher. La vieille Route del’Inde de Bactres à Taxila. 2 tt.Mém. DAFA, 1. Paris: Les Éditionsd’art et d’histoire, 1942-1947.

Fussmann 1974

Gérard Fussman. “Nouvelledécouverte à Bamiyan.”Afghanistan, 27/2 (1974): 57-73.

Gardin 1957

Jean-Claude Gardin. “La poteriede Bamiyan.” Ras Orientalis, 2(1957): 227-245.

Gardin and Lyonnet 1987

Jean-Claude Gardin and BertilleLyonnet. “Céramique et Chrono-logie.” In: Le Berre 1987a: 96-120.

Hambis 1967

Louis Hambis, ed. Mission PaulPelliot. T. III. Douldour-Aqour etSoubachi, Planches. Paris, 1967.

Ghirshman 1946

Roman Ghirshman. Begram.Recherches archéologiques ethistoriques sur les Kouchans.Mém. DAFA, 12. Cairo: Institutfrançais d’archéologie orientale,1946.

Ghirshman 1948

Roman Ghirshman. Les Chionites-Hephtalites. Mém. DAFA, 13.Cairo: Institut français d’arché-ologie orientale, 1948.

Göbl 1967

Robert Göbl. Dokumente zurGeschichte der Iranichen Hunnenin Baktrien und Indien, 4 vols.,Wiesbaden: Harrassowitz, 1967.

Godard et al. 1928

André Godard, Y. Godard, andJoseph Hackin. Les antiquitésbouddhiques de Bamiyan. Mém.DAFA, 2. Paris; Bruxelles: G. vanOest, 1928

Hackin and Carl 1933

Joseph Hackin and Jean Carl.Nouvelles recherches archéolo-giques à Bamiyan. Mém. DAFA,3. Paris: G. van Oest, 1933.

Hackin et al. 1959

Joseph Hackin, Jean Carl andJacques Meunié. Diversesrecherches archéologiques enAfghanistan (1933-1940). Mém.DAFA, 8. Paris: Presses uni-versitaires de France, 1959.

Higuchi 1983

Takayasu Higuchi. Bamiyan:Afuganisutan ni okeru Bukkyosekkutsu ji in no bijutsukokogakuteki chosa, 1970—1978-nen (in Japanese; Bamiyan: Artand Archaeological Researches onthe Buddhist Cave Temples inAfghanistan 1970-1978). 4 vols.Kyoto: Dohosha, 1983-1984; 2nd

ed., 2001.

IsMEO activities 1972

[IsMEO activities]. “ArchaeologicalMission in Afghanistan.” East andWest (Rome), 22/3-4 (1972):379-384.

Julien 1853

Stanislas Julien. Histoire de la viede Hiouen-thsang et de sesvoyages dans l’Inde depuis l’an629 jusqu’en 645, par Hoei-li etYen-thsong, suivie de documentset d’éclaircissements géograph-iques tirés de la relation originalede Hiouen-thsang. Paris:L’Imprimerie impériale, 1853.

Julien 1857-1858

Stanislas Julien. Mémoires sur lescontrées occidentales, traduit dusanscrit en chinois, en l’an 648par Hiouen-thsang, et du chinois

24

en français par M. Stanislas Julien,2 tt. Voyages des pèlerinsbouddhistes, II-III. Paris:L’Imprimerie impériale, 1857-1858.

Kak 1933

Ram Chandra Kak. AncientMonuments of Kashmir. London:India Society, 1933.

Klimburg-Salter 1989

Deborah E. Klimburg-Salter. TheKingdom of Bamiyan: Buddhist Artand Culture of the Hindu Kush.Naples: Istituto universitarioorientale, Dipartimento di studiasiatici; Rome: Istituto italiano peril medio ed estremo oriente, 1989.

Kohzad 1953

Ahmed Ali Kohzad. “The MostAncient Temple in Bamiyan.”Afghanistan, 7/2 (1953): 12-17;translation of “Qadim tarinmacbade Bamiyan.” Aryana, 2/6(1944-1945): 20-26.

Kuwayama 1987

Shoshin Kuwayama. “LiteraryEvidence for Dating the Colossi inBamiyan.” In: Gherardo Gnoli andLionello Lanciotti, eds. Orientalia:Iosephi Tucci Memoriae Dicata. 3vols. Serie oriental Roma, 56.Rome: Istituto italiano per il Medioed Estremo Oriente, 1985, Vol. 2:703-727.

Kuwayama 1989

Shoshin Kuwayama. “TheHephthalites in Tokharistan andNordwest India.” Zinbun: Annalsof the Institute of Research inHumanities (Kyoto University), 24(1989): 83-134.

Kuwayama 1991

Shoshin Kuwayama. “The Horizonof Begram III and Beyond: AChronological Interpretation of theEvidence for Monuments in theKapisi-Kabul-Ghazni Region.” Eastand West (Rome), 41/1-4 (1991):79-120.

Kuwayama 2002

Shoshin Kuwayama. Across theHindukush of First Millennium: ACollection of the Papers. Kyoto,2002.

Le Berre 1987a

Marc Le Berre, M. Les monumentspré-islamiques de l’Hindukushcentral. Avec la collaboration deHenri Marchal et une contributionde Jean-Claude Gardin et BertilleLyonnet. Mém. DAFA, 24. Paris:Editions Recherche sur lescivilisations, 1987.

Le Berre et al. 1987b

Marc Le Berre, Jean Claude Gardinet Bertille Lyonnet. “Donnéesarchéologiques inédites surl’histoire de Bamiyan (Afghani-stan).” In: Gherardo Gnoli andLionello Lanciotti, eds. Orientalia:Iosephi Tucci Memoriae Dicata. 3vols. Serie oriental Roma, 56.Rome: Istituto italiano per il Medioed Estremo Oriente, 1985, Vol. 2:775-785.

Litvinskii and Zeimal’ 1971

Boris A. Litvinskii and Tamara I.Zeimal’. Adzhina-Tepa. Arkhitek-tura, zhivopis’, skul’ptura.Moscow: Izd-vo. Iskusstvo, 1971.

Mitchiner 1976

Michael Mitchiner. Indo-Greek andIndo-Scythian Coinage. Vol. 9.Greeks, Sakas and Their Con-temporaries in Central andSouthern India: Indo-Parthian,Western Satraps, Chutus, Abhiras,Satavahanas. London: Hawkins,1976.

Miyaji 1973

Akira Miyaji. “La scène deParinirvana dans la grotte F àBamiyan.” Journal of the NagoyaUniversity Faculty of Letters, 9(1973): 51-73.

Miyaji 1978

Akira Miyaji. “The Parinirvanascenes of Bamiyan: an icon-ographical analysis.” In: TakayasuHiguchi, ed. Japan-AfghanistanJoint Archaeological Survey in1976. Kyoto: Kyoto University,1978: 17-31.

Mustamandi 1969

Shahibye Mustamandi. “Nouvellesfouilles à Hadda (1966-7) parl’Institut Afghan d’Archéologie.”Arts Asiatiques, 20 (1969): 15-36.

Rapson 1908

Edward James Rapson. Catalogueof the Coins of the AndhraDynasty, the Western Ksatrapas,the Traikutaka Dynasty and the“Bodhi” Dynasty. London: TheBritish Museum, 1908.

Raverty 1881

Henry G. Raverty, tr. Tabakat-i-Nasiri: A general history of theMuhammadan dynasties of Asia,including Hindustan; from A. H.194 (810 A.D.) to A.H. 658 (1260A.D.) and the irruption of theinfidel Mughals into Islam, byMinhaj-ud-din, Abu-’Umar-i-’Usman. 2 vols. Calcutta: AsiaticSociety of Bengal, 1881.

Rowland 1938

Benjamin Rowland, Jr. The Wall-Paintings of India, Central Asiaand Ceylon: A Comparative Study.Boston: Merrymount Press, 1938.

Rowland 1947a

Benjamin Rowland, Jr. “TheColossal Buddhas at Bamiyan.”Journal of the Indian Society ofOriental Art, 15 (1974): 62-73.

Rowland 1947b

Benjamin Rowland, Jr. “Studies inthe Buddhist Art of Bamiyan, theBodhisattva of group E.” In: K.Baratha Iyer, ed. Art and Thought,issued in honour of Dr. Ananda K.Coomaraswamy on the occasionof his 70th birthday. London:Luzac, 1947: 46-52.

Scerrato 1960

Umberto Scerrato. “A Short noteon Some Recently DiscoveredBuddhist Grottoes near Bamiyan,Afghanistan.” East and West(Rome), 11/2-3 (1960): 94-120.

Schlumberger et al. 1964

Daniel Schlumberger, BrunoDagens, and Marc Le Berre.Monuments préislamiquesd’Afghanistan. Mém. DAFA, 19.Paris: G. Klincksieck, 1964.

Spuler 1952

Bertold Spuler. Iran in früh-islamischer Zeit: Politik, Kultur,Verwaltung und öffentliches Lebenzwischen der arabischen und derseldschukischen Eroberung, 633

25

bis 1055. Wiesbaden: F. Steiner,1952.

Stronach 1978

David Stronach. Pasargadae: AReport on the ExcavationsConducted by the British Instituteof Persian Studies from 1961 to1963. Oxford; New York:Clarendon press, 1978.

Taddei 1968

Maurizio Taddei. “Tapa Sardar:First Preliminary Report.” East andWest (Rome), 18/1-2 (1968):109-124.

Taddei 1985

Maurizio Taddei. “Clay Stupas andThrones at Tapa Sardar, Ghazni(Afghanistan).” Zinbun: Memoirsof the Research Institute forHumanistic Studies (KyotoUniversity), 20 (1985): 17-32.

Taddei and Verardi 1978

Maurizio Taddei, and GiovanniVerardi. “Tapa Sardar: SecondPreliminary Report.” East andWest (Rome), 28/1-4 (1978): 33-135.

Taddei and Verardi 1981

Maurizio Taddei, and GiovanniVerardi. “Buddhist Sculptures fromTapa Sardar, Ghazni(Afghanistan).” La parola delpassato. Rivista di studi antichi(Naples), fasc. CXCIX (1981):251-266.

Tarzi 1976a

Zemaryalai Tarzi. “Hadda à lalumière des trois dernièrescampagnes de fouilles de Tapa-é-Shotor (1974-1976).” ComptesRendus de l’Academie desInscriptions et Belles Lettres,Séances de l’année 1976, juillet-octobre (Paris): 381-410.

Tarzi 1976b

Zemaryalai Tarzi, ed. Zaryab.“Bamiyan, and Where is theSleeping Buddha ?” Aryana(Afghanistan Republic), Summer1976: 30-33.

Tarzi 1977

Zemaryalai Tarzi. L’architecture etle décor rupestre des grottes deBamiyan, 2 tt. Bibliothèque du

Centre de recherches sur l’Asiecentrale et la Haute-Asie, 1. Paris:Imprimerie nationale, 1977.

Tarzi 1988a

Zemaryalai Tarzi. “Bamiyan.”Encyclopaedia Universalis, Vol. 3(Paris, 1988): 225-26.

Tarzi 1988b

Zemaryalai Tarzi. “Bamian. ii.History and Monuments.”Encyclopaedia Iranica, Vol. 3, fasc.6 (London; New York: Routledge,Kegan Paul, 1988; 1989 ed. ascomplete volume): 658-660.

Tarzi 1997

Zemaryalai Tarzi. (1997),“Préservation des sites de Hadda:monastères bouddhiques de TapeShotor et de Tape Tope Kalan.”Mari, Annales de RechercheInterdisciplinaires, 8 (1997): 207-222.

Tarzi 2001

Zemaryalai Tarzi. Relation entreles villes et les monastères aunord-ouest de l’Inde: le cas deKâbol. Thèse d’habilitation, 2 tt.Strasbourg, 2001.

Tarzi 2003

Zemaryalai Tarzi. “Bamiyan:Professor Tarzi’s Survey andExcavation Archaeological Mission,2003.” The Silk Road, 1/2 (2003):37-39.

Tarzi 2004a

Zemaryalai Tarzi. “RecentArchaeological Investigation of theFrench team at the Bamiyan Site.”In: Protecting the World HeritageSite of Bamiyan, Messages fromthe Field: International Sym-posium. Tokyo, 2004: 16-17.

Tarzi 2004b

Zemaryalai Tarzi. “‘Kaweshhayetazae bastanshenasi dar Bamiyan’et sa traduction en allemand:‘Bamians Buddhistische Denk-mäler und neue Ausgrabungen.’”Andescha, 16, Jänner 2004: 3-8,60-68.

Tarzi 2004c

Zemaryalai Tarzi. “La missionarchéologique à Bamiyan en2003.” Les Nouvelles

d’Afghanistan 104, mars 2004:20-21.

Tarzi 2005

Zemaryalai Tarzi. “Les fouilles dela mission archéologique françaiseà Bamiyan sous la direction de Z.Tarzi.” In: L’art d’Afghanistan dela préhistoire à nos jours: nouellesdonnées, actes d’une journéed’étude, UNESCO, 11 mars 2005.V. and A. Marigo, eds. Paris:Centre de recherches et d’étudesdocumentaires sur l’Afghanistan,2005: 95-124.

Verardi 1983

Giovanni Verardi. “Osservazionisulla coroplastica di epoca kusananel nord-ovest e in Afghanistan inrelazione al materiale di TapaSardar, seguite da una pre-cisazione sulla natura e la datadelle sculture di Ushkur.” Annalidell’Istituto Universitario Orientale(Naples), 43 (1983): 479-504.

Watters 1904

Thomas Watters. On YuanChwang’s Travels in India, 629-645 A.D. Thomas W. Rhys Davidsand Stephen W. Bushell, eds. 2vols. Oriental Translation Fund, N.S., 14-15. London: Royal AsiaticSociety, 1904-1905; 3rd ed., NewDelhi, 1996.

26

Balkh and the Plains ofTurkestan

Frank HaroldUniversity of Washington, SeattlePhotographs by Ruth Harold

If I had to choose a single placeto represent all the glories andcalamities of Central Asia, I wouldpick the oasis of Balkh in northernAfghanistan. Balkh was old longbefore Alexander captured it, andover the course of 2500 years hasseen more than a score ofconquerors come and pass on.The Arabs, impressed by Balkh’santiquity and wealth, called itUmm-al-belad, the mother ofcities. When the Silk Road was thechief artery of commerce betweenEast and West, Balkh was secondto none. But then came ChingisKhan, who wreaked upon it theutter devastation that has madethe Mongols’ name a byword forbarbarism. Balkh never fullyrecovered, and eventually fadedinto a vil lage; the seat ofgovernment shifted to scruffy butvigorous Mazar-i Sharif, site of arevered shrine. What the visitorcomes to see in Balkh is chieflythe melting walls of the old city,enclosing a vast field of rubble andwreckage; it is a place ofmemories rather than monu-ments. But for those who savorthe melancholy pleasure of ruins,there is no more evocative sitebetween Xian and Trebizond.

Why here, on the drab plainsof Turkestan between the HinduKush mountains and the river AmuDarya (Oxus)? At one level,geography holds the key. Balkhsits on an alluvial fan built up bythe Balkab River, well suited toirrigation. The region calledBactria in ancient times wasrenowned for its grapes, oranges,water lilies and later sugar cane,and an excellent breed of camelstoo. To this day, some of the

world’s most luscious melonscome from nearby Kunduz. Mostsignificantly, several natural traderoutes intersect at Balkh. Fromthere, caravans could follow thewell-watered foot of themountains westward towardsHerat and Iran, or across the Oxusto Samarkand and China [Fig. 1].The valley of the Balkab still givespassage to Bamiyan and thenceto Kabul; of all the routes acrossthe Hindu Kush, this is the mostwesterly and the easiest. Butgeography is at most opportunity,not destiny; and the greatness ofBalkh owes even more to those

inventive Iranian peoples whopromoted craftsmanship andtrade, built cities and wrote poetryall across ancient Central Asia. Onthe down side, Balkh was usuallyrich rather than powerful, andbecame the envy and the prize ofmore warlike neighbors.

Always a place of importance,the province of Bactria and itscapital city flash into view in thefragmentary annals of historiansand travelers. Bactria first appearson the list of the conquests ofDarius, who incorporated it intothe Achaemenid empire.Traditionclaims that Zoroaster taught hereand died here, in the sixth centuryBCE, or even earlier; the Zoro-astrian faith became the statereligion of the Achaemenids, andlater of the Sasanians. Alexandertook Bactria in 329 BCE, marriedthe princess Roxane, and madethe region his base for furtherconquest and for the amalga-mation of the Greek and Iraniancivilizations. That vision, far in

Fig. 1. Balkh and its neighbors. Copyright © 2007 Frank Harold.

27

advance of the times, survived foranother three centuries in thesmall Graeco-Bactrian kingdomsthat thrived and quarreled on bothsides of the Hindu Kush. The ruinsof one of their cities have beenfound at Aï Khanum, close by theRiver Oxus. They wrote no historybut contributed greatly to thedevelopment of Gandharanarchitecture and sculpture, andthey minted the most gorgeoussilver coins of the ancient world.

Bactria reappears with itsannexation by the Kushans (129BCE), whose large and powerfulempire stretched from CentralAsia deep into India. This was afortunate era, when the landsthrough which the caravan routespassed were divided among a fewstable states which submergedtheir differences in the interestsof trade; and Balkh flourished atthe crossroads, as a depot andtrans-shipment point for theworld’s luxuries. ‘From the RomanEmpire the caravans brought goldand silver vessels and wine; fromCentral Asia and China rubies,furs, aromatic gums, drugs, rawsilk and embroidered silks; fromIndia spices, cosmetics, ivory andprecious gems of infinite variety’(Dupree 1967, 71). With themerchants came monks preachingthe new religion of Buddhism, andBalkh became a center of worshipand learning, famous for itstemples and monasteries.

By the time the Chinese pilgrimXuanzang passed through Balkhon his way to the fountainhead ofBuddhism in India (630 CE), thecity had become part of theSasanian empire. Sasanianviceroys ruled from Balkh, and asplendid fire-temple had beenerected in the suburbs. Thebazaars were still humming withtrade, the countryside fertile, anda hundred Buddhist temples andmonasteries testif ied to thecontinued vitality of Buddhism.Xuanzang was particularlyimpressed by the wealth of thechief monastery and its associatedstupa. But he noted also that the

building had recently been looted,and that the monks had becomelax in the performance of theirduties; perhaps he sensed that theglory days of Buddhism were over.There was frequent strife with theTurkic nomads across the Oxus,and the first Arab incursions werejust fifteen years ahead.

The times that followed wereturbulent ones in Central Asia.Balkh changed hands repeatedlyamong Arab, Persian and Turkicrulers, and was sacked more thanonce, yet it continued to prosper.The Arab geographers Yaqubi andMoqaddasi (9th and 10th cen-turies) depict Balkh as it wasunder Samanid rule, whenBukhara was the center of power.A large and bustling city of mudbrick some three square miles inarea, it held perhaps 200,000persons. Mud-brick wallssurrounded the city, pierced byseven gates. Buddhist andZoroastrian temples had beendestroyed; instead, a splendidFriday mosque occupied thecenter, and many more mosqueswere scattered among thedwellings. The city was home,not only to Persians and Turks butalso to communities of Jews andIndian traders; a Nestorianmetropolitan had his see in Balkh.It nourished poets and scholars,lawyers and even geographersand astronomers. But peace wasa sometime thing; even whenBalkh came under Seljuk rule for

over a century, the nomads werenever far away.

Catastrophe struck in 1220,when Chingis Khan chose to makean example of Balkh, perhaps aspunishment for an uprising. Onehundred thousand Mongolhorsemen embarked on an orgyof slaughter and destruction thatleft nothing standing; a few weekslater they returned to pick off thehandful of wretched survivors.Balkh remained in ruins for acentury and was so described byMarco Polo (1275) and by IbnBatuta (1333); and yet revivalmust have been under way, forTimur (Tamerlane) chose Balkh toproclaim his accession to thethrone (1359). Timur and hissuccessors favored Balkh; theyrestored the walls [Fig. 2] andendowed the city with quitesplendid buildings, some of whichsurvive.

Balkh remained worth fightingover, by Uzbeks, Safavids,Mughals and eventually the risingpower of Afghanistan under theDurrani Shahs. But the city slowlydeclined as its surroundings grewswampy and malarial, theirrigation canals fell into disrepair,and cholera struck again andagain. Uzbeks of nomad originsbecame the dominant element.By the beginning of the 20th

century the population was downto 500 households, and theadministrative center of Afghan

Fig. 2. Balkh. The Timurid walls of the Bala Hissar. Photograph copyright© 1970 Ruth Harold.

28

Turkestan had migrated (1866) tonearby Mazar-i Sharif. A newchapter had begun, the one inwhich we are still living.

Of all this eventful history, littleenough remained on the groundat the time of our visit in 1970.One arrives in the center of anagricultural market town, neatlyplanted with trees and grass, thatshow off two Timurid edifices. Oneis the mausoleum of Khwaja AbuNasr Parsa [Fig. 3], erected in1462/63 in honor of adistinguished theologian; it isconsidered one of the finestexamples of late Timuridarchitecture and often features ontourist posters. The shrine consistsof a tall octagonal brick chambersurmounted by a fluted dome;entry is through a high portalflanked by a pair of corkscrewcolumns. The entire exterior isclad in brilliant blue tile mosaic,much of which has been slowlypeeling off the walls. The interior

is cool anda u s t e r e ,d e c o r a t e dwith stuccohoneycomband paintedf l o r a ld e s i g n s .Across thepark stands atall gatewaywith somedeco ra t i vetile work;this is all thatremains of amadrassahbuilt in the17th centuryin Timuridstyle. As faras I cana s c e r t a i n ,b o t hbuildings stillstand.

The earliermonumentstake somesearching. Acouple ofnondescript

mounds probably mark the sitesof that Buddhist temple and stupa,whose statuary Xuanzangdescribed as being ‘lustrous withprecious gems.’ A small brickmosque decorated with carvedstucco survives from early Islamictimes, but we failed to find it. Whatwe had come to see was the walls

[Fig. 4], battered and weather-beaten but still sixty feet high inplaces, that enclose the BalaHissar, the High Fort. Theramparts were built in Timuridtimes (14/15th centuries) uponfoundations that likely go back tothe Kushans and possibly further.They enclose a roughly circularfield half a mile across, thatprobably corresponds to thecentral city of medieval Balkh. Nowthere is only dry scrub and lowmounds of debris; here and therepotsherds and broken bricks callmutely for attention. There isnothing much to see; but I havenever forgotten what it felt like,up there on those worn stumps ofwall, gazing out over nothingness.

We stayed in Mazar-i Sharif(‘Tomb of the Exalted’), whichboasted a decent hotel and evenoccasional electricity.There isnothing ancient or traditionalabout Mazar, which only rose toprominence in the 19th century.We found it a bustling ‘third-world-modern’ town of straight widestreets, motor traffic, governmentoffices and shops. Uzbeks,Tadjiks, Hazara and Pashtunsmeet and chaffer in the seat ofpower, which is also a center fortrade in Karakol lambskins andcarpets. More recently, Mazar hasall too often been in the news. Itwas largely spared Soviet shelling,but saw much fighting when theTaliban seized most of northernAfghanistan, and again when that

Fig. 3. Mausoleum of the theologian Abu Nasr Parsa (1462-1463 CE). Photograph copyright © 1970 Ruth Harold.

Fig. 4. Balkh. The Timurid walls of the Bala Hissar. Photograph copyright© 1970 Ruth Harold.

29

regime was toppled by theNorthern Alliance with the help ofAmerican airpower. Miraculously,the great shrine of the Sharif Ali,which lent the city its name, hassurvived the turmoil and hasrecently been restored.

Hazrat (the noble) Ali is one ofthe central figures of Shia Islam,and almost as much revered asthe Prophet Muhammad himself.Ali ibn Abi Talib was Muhammad’scousin and son-in-law, andeventually became the fourthCaliph. But his reign was marredby discord; the Caliph wasassassinated in 658 CE, andaccording to orthodox traditionwas buried in Najaf, Iraq. Afghansbelieve otherwise: the body of the

slain Caliph, tied onto a camel’sback, was carried out to Turkestanand buried in a secret location.Five hundred years later, thanksto dreams and visions, the gravecame to light and a shrine wasbuilt over it. Chingis Khan leveledit, but Ali ’s sepulcher wasrediscovered during the reign ofHusain Baikara, the last TimuridSultan of Herat, who erected agrand mausoleum on the site(1481 CE) [Figs. 5, 6, 7]. This isthe building, many times restoredand re-decorated, that one seestoday. With its two domes,impressive courtyard and portals,excellent blue tile-work and a flockof white pigeons, the shrine ofHazrat Ali is one of the mostspectacular buildings in

Afghanistan. Pilgrims flock to thetomb, which has a reputation formiraculous cures; and manythousands come here each springto celebrate Nauruz, the PersianNew Year.

Neither Balkh nor Mazar lookanything like a caravan city of themiddle ages, but nearbyTashkurgan does (or did in 1970)[Figs. 8, 9, 10, facing page]. Thetown officially goes by the nameof Khulm. A ruined mud-brickcastle looms over the town; it isonly a couple of centuries old, butthe weathered walls give it an

antique air. The covered bazaarwas fascinating, a place oftraditional crafts, small open-fronted shops and inviting chai-khanas (tea houses). You sat on atakht, a throne, by the side of thestreet, sipping your tea whilepuffing on a hookah, and watchedthe parade of Central Asia pass by.That was in a time of peace, whichseemed like innocence. I wonderhow Tashkurgan has fared.

About the Authors

Frank and Ruth Harold arescientists by profession andtravelers by avocation. Frank wasborn in Germany, grew up in the

Fig. 5. Mazar-i Sharif. Shrine of Hazrat Ali. Photograph copyright© 1970 Ruth Harold.

Fig. 5. Mazar-i Sharif. Shrine of Hazrat Ali. Photograph copyright© 1970 Ruth Harold.

Fig. 5. Mazar-i Sharif. Shrine of HazratAli, view in courtyard. Photograph copy-right © 1970 Ruth Harold.

30

Middle East and studied at the CityCollege, New York, and theUniversity of California atBerkeley. Now retired from fortyyears of research and teaching, heis Professor Emeritus ofbiochemistry at Colorado StateUniversity and a member of thevolunteer faculty at the Universityof Washington. Ruth is amicrobiologist, now retired, and anaspiring painter. The Harold familylived in Iran in 1969/70, whileFrank served as Fulbright lecturerat the University of Tehran. Thisexperience kindled a passion forAsian travel which has since takenthem to Afghanistan and back toIran, into the Himalayas, up anddown the Indian sub-continent

and along the Silk Road betweenChina and Turkey. They make theirhome in Edmonds, Washington.

Sources

The history of Balkh, and thereports of ancient travelers, arecovered in some detail by Guy Le

Fig. 8. Tashkurgan. Photograph copy-right © 1970 Ruth Harold.

Strange, The Lands of the EasternCaliphate (Cambridge, 1905;reprint Lahore, 1977); and in

articles on Bac-tria and Balkh byFrantz Grenet inthe Encyclo-paedia Iranica,Vol. 3 (London,1988). A veryreadable accountof the land, itspast and itspresent, is inNancy HatchDupree’s TheRoad to Balkh(Kabul, 1967);unfor tunate ly,this little gem isnow a collector’sitem. The mostc e l e b r a t e dtraveler’s tale issurely The Roadto Oxiana, byRobert Byron(London, 1937;reprint Oxford,1982). Discerning

readers will also enjoy The LightGarden of the Angel King:Journeys in Afghanistan, by PeterLevi (London, 1972). Afghani-stan’s recent travails have beenextensively covered, for instance,in Larry P. Goodson, Afghanistan’sEndless War (Seattle, 2001).

Fig. 9. Street scene, Tashkurgan. Photograph copyright© 1970 Ruth Harold.

Fig. 10. Chaikhane (tea house) in thebazaar, Tashkurgan. Photographcopyright © 1970 Ruth Harold.

31

Further Evidence for the Interpretation of the‘Indian Scene’ in the Pre-Islamic Paintings atAfrasiab (Samarkand)Matteo ComparetiVenice

The Sogdian paintings at Afrasiabwere discovered accidentally morethan forty years ago during roadconstruction near Samarkand.However, only in 1975 was the firstbook concerning them published(Al’baum 1975). Archaeologicalexcavations continued at theAfrasiab site for some time,leading to the discovery of otherfragments of schematic paintingsbetween 1978 and 1985(Akhunbabaev 1987). Since 1989French archaeologists have beenexcavating at the ancient site incollaboration with Russian and,after the collapse of the SovietUnion, Uzbek colleagues, butwithout a specific connection tothe paintings themselves.

The room where the paintingswere discovered came to be calledthe ‘Hall of the Ambassadors,’since its western wall, facing theentrance, depicts several repre-sentatives of non-Sogdian landsbearing gifts. While from the startthere has been general agreementon the identification of thatimagery, the interpretation ofwhat exactly the ritual is that thescene depicts and some of itsdetails have been a matter ofdebate. This article is a con-tribution to that discussion.

General Interpretation

At the end of the 1980s Chiara SilviAntonini offered persuasiveevidence to identify the scene ofthe western wall in the so-called‘Hall of the Ambassadors’ as arepresentation of the Iranian NewYear Festival (Nawruz) inconjunction with the coronation ofthe local king, Varkhuman (SilviAntonini 1989). In this author’s

opinion, Silvi Antonini’s idearemains the key for a correctinterpretation of the entire cycleof the paintings at Afrasiab. Whilea detailed study by Markus Modea few years later disputed herinterpretation (Mode 1993), thegreat specialist of Sogdian studies,the late Boris Marshak, not onlyaccepted it but also added otherimportant elements to the generalinterpretation of the whole cycleand, especially, of the southernwall paintings (Marshak 1994).Soviet scholars continued to studythe Afrasiab paintings, althoughtheir interesting results did notbecome widely known because oftheir publication in Russian in rarejournals or books (e.g., Maitdinova1984; Akhunbabaev 1990; Motov1999).

In recent years other authorssuch as Etsuko Kageyama, FrantzGrenet, Sergei Iatsenko, andSimone Cristoforetti together withthe present writer reopened theproblem of the interpretation ofthe Afrasiab paintings consideringsingle scenes or details(Kageyama 2002; Grenet 2003;Yatsenko 2004; Grenet 2005;Grenet 2006; Compareti andCristoforetti 2005). Finally, aconference was organized by thepresent writer in March 2005 inVenice in order to collect in onevolume the most recent results ofthe investigations by thosescholars who had already studiedthe paintings together with a newtranslation of the Sogdianinscriptions at Afrasiab by VladimirLivshits (Compareti and de laVaissière 2006).

The inscriptions are extremelyimportant in order to attempt achronology for the paintings.

Fragmentary inscriptions on thewestern wall mention the name ofa sovereign, Varkhuman, whocorresponds to the local kingrecognized as governor ofSamarkand and Sogdiana by theChinese Emperor Gaozong (649-683) in the period between 650-655 (Chavannes 1903, 135).1 In658 Gaozong even sent an envoyto the court of Varkhuman for anofficial investiture (Anazawa andManome 1976, 21ff., cited byKageyama 2002, 320). However,according to Islamic sources,when Sa‘id ibn Othman conqueredSamarkand in 676 he did not findany king there. So it is possible tosuppose that Varkhuman was nolonger regent of Samarkand at thetime of the Arab conquest (Frye,1954, 40; Smirnova 1970, 275;Fedorov 2006, 222-223). All of theevidence suggests that theAfrasiab paintings were executedaround this period.

Other literary sources (Chinese,Persian and even Sogdian) couldbe useful for the comprehensionand chronology of the muralpaintings. They have already beenextensively considered in a recentstudy by Frantz Grenet, who alsodemonstrated that the spatialorganization of the Afrasiabpaintings corresponds to anestablished scheme well known inSasanian Persia (Grenet 2005,124-130) and in India too whereit possibly originated (de laVaissière 2006, 148).2 Inparticular, as already noted byMode (Mode 1993), a passage ofthe Tangshu (History of the TangDynasty, composed between 1043and 1060) referring to 7th-centurycentral Sogdiana mentions a royalpavilion where representations of

32

Persians and Byzantines appearedon the western wall, Chinese onthe northern wall, and the Turksand Indians on the eastern one(Chavannes, 1903, 145).3 As willbe shown in the present paper aswell, this is exactly the partitionof the walls at Afrasiab, the onlyexception being the western one.

At present it is almostunanimously accepted that thescenes depicted in the ‘Hall of theAmbassadors’ are part of a uniquecycle. The western wall most likelyrepresents the initial stage of theSogdian Nawruz celebrated in 7th-century Samarkand. According tothe mobile Sogdian (and, moregenerally, Iranian) calendar,during the 7th century Nawruz fellin summer (Compareti andCristoforetti 2005).4 Thefragmentary paintings enable usto recognize foreign delegatesshown during the presentation ofsome gifts to an upper centralfigure that is irretrievably lost [Fig.1]. The presentation of gifts (or,better, tributes) on the occasionof Nawruz is a very ancient Iraniancustom, and, according to theinterpretation by RomanGhirshman and other scholarswhich was accepted by SilviAntonini, it can be clearlyobserved among the very ancientreliefs of the Apadana at

Persepolis (Silvi Antonini 1989,125-126).

It seems correct to consider theNew Year Festival as a goodoccasion to celebrate thecoronation of a king amongancient Iranians. The hypothesisis supported by some literarysources on ancient Persian habits,transmitted, however, onlythrough Muslim authors (Ibid.,118-126). So, the lost figure in theupper part of the western wall wasmost likely that of a king, possiblyVarkhuman himself. It is notexcluded that he was probablyrepresented together with his wifeaccording to a scheme muchappreciated on 6th-centuryfunerary monuments belonging topowerful Sogdians settled in Chinaand recently excavated in the areaof Xi’an (China).5 On thosemonuments the couple isrepresented under a canopy bothin realistic and paradisiacal scenes(Marshak 2001), but it is notexcluded that the two royal figurescould have been positioned at acertain distance from each other.In this case, a possible alternativeexplanation is Mode’s hypothesisthat the two rows of tributes-bearers were proceeding towardstwo different sovereigns(Varkhuman and the TurkishKhaghan) (Mode 2006). Also of

interest is Sergei Iatsenko’s veryinteresting observation about thenecklace and the reconstructedtorque in the hands of the envoyfrom Chaghanyan on the westernwall. The envoy can be recognizedin the lower left part of the scenesince he is wearing a robeembellished with so-calledsenmurv-like creatures.Considering that all the otherforeign envoys are bringingtributes to be presented to thefigure that was depicted in theupper part of the scene, then mostlikely the torque was intended fora man and the necklace for awoman. On the western wall theycan only be the Samarkand royalcouple (Yatsenko 2004). In thisway, as will be discussed moreextensively below, the King andQueen of Samarkand could havecounterbalanced the Chineseemperor and empress on thenorthern wall.

Boris Marshak has identified thescene on the southern wall as thefinal stage of the Nawruz festival,when the Sogdian sovereigncelebrated a funerary ritual inhonour of his ancestors in a templewhich — according to the Tangshu— was built on the eastern fringesof the town (Marshak 1994, 11-15).6 While the paintings are veryfragmentary, the shape of an

Fig. 1. Western wall, Afrasiab (after F. Ory and Grenet reconstruction).

Fig. 2. Southern wall, Afrasiab (after F. Ory and Grenet reconstruction).

33

enormous horse ridden by aperson who was depicted largerthan the other people around himcan be discerned among thepaintings of the southern wall [Fig.2]. It is highly probable that thisperson was king Varkhumanmoving in the direction of thetemple of his ancestors. Therepresentation of this buildingsurvived only partially in the leftpart of the scene depicted on thesouthern wall, almost on thecorner with the eastern wall (thedirection of the royal paradeaccording to the Chinese sources).Two people wearing ritualZoroastrian masks (padam) in themiddle of the procession could beconsidered in charge of thesacrifice for the royal ancestors.In fact, they accompany aharnessed horse not mounted andfour geese probably destined to besacrificed in honour of Iraniangods (Grenet 2005, 125).

Chinese subjects appear on thenorthern wall exactly as reportedin the Tangshu [Fig. 3]. It is anextremely complicated represen-tation which is divided in twoparts, both of them connectedwith astronomical-astrologicalmatters. It is beyond the aim ofthe present article to discuss indetail all the elements which ledto its general interpretation(Compareti and Cristoforetti 2005;Compareti 2006; Comparetiforthcoming; Compareti andCristoforetti forthcoming). Here itwill be enough to say that the twoparts of the scene constitute anexact parallel with the western andthe southern walls. In fact, mostlikely, on the northern wall thereis a representation of the ChineseNew Year Festival. Some Chinesepoems report the custom of very

ancient times when the emperorwas expected to perform a ritualhunt in a special part of theimperial park and kill evil animalssuch as felines. After the hunt,members of the Chinesearistocracy went at a pond in thesame park to listen to musiciansand singers on boats shaped likebirds’ heads, while some peoplehad to dive in the water in searchof animals such as fishes, turtlesand ducks. This is exactly what canbe discerned in the hunting andaquatic scenes in the left part ofthe northern wall. Also theappearance of a dragon isrecorded in Chinese poems incoincidence with a festivitypossibly to be identified with theNew Year Festival and, in fact, acomposite winged monster can bediscerned below the boat in theaquatic scene (Comparetiforthcoming). A person leadingtwo horses in the water is probablyperforming a Chinese ritual whichhas not yet been identified. As willbe observed below, the presenceof this last element is quiteimportant for the comprehensionof the whole cycle.

There is something more to beadded as regards the aquaticscene. One lady on the boat isrepresented larger than theothers, because she is possibly theChinese empress. She is feedingthe fishes just below the boat and,in fact, her left hand is opened asif to drop something into the water.Such an attitude calls to mind atypical Chinese festivity, theDuanwu Festival, still performedtoday around the summer solsticeto remember a poet whocommitted suicide in ancienttimes, during the Warring Statesperiod (453-221 BCE). Every year

on thato c c a s i o np e o p l efeed fishesin order top r e v e n ta q u a t i ca n i m a l sf r o meating the

body of the poet and race in riversand lakes on dragon-shaped boatssearching, ideally, for his corpse.Other people dive in the water tolook for his body which, needlessto say, was never recovered. It isquite clear that the second part ofthe ancient Chinese New YearFestival, as reported in poems,and the Duanwu Festival verymuch resemble each other.

As is well-known, in Gregoriancalendrical terms, the ChineseNew Year Festival falls in January-February, while the Duanwu fallsaround mid-June. The latter dateconstitutes an exact parallel withthe Sogdian Nawruz which, duringthe 7th century, was celebrated insummer. The artists in charge ofthe execution of the paintings atAfrasiab confused the twofestivities because of theirsimilarities. Possibly the artistseven confused them deliberatelyin order to have a calendricalcoincidence between a localSogdian festivity and an importantChinese one (namely, theDuanwujie) which could balancethe Chinese New Year celebration.

The Eastern Wall

At this point we turn our attentionto the eastern wall [Fig. 4, facingpage]. Many scholars agree thatthis scene represents India mainlyon the basis of the passage in theTangshu, even though there is noTurk representative depicted here.Unfortunately, once more, theChinese source is not specificabout the subject of the scene,and the fragmentary state of thepainting makes an identificationextremely difficult.

The surviving fragments areconcentrated just in the lower part

Fig. 3. Northern wall, Afrasiab (after F. Ory and Grenet reconstruction).

34

of the wall. Starting from the leftpart, one can observe two sittingpeople facing each other in atypical attitude of teacher andstudent. Between them is a roundobject which was probablyintended to be an armilla used forastronomical purposes. In a recentpaper, Frantz Grenet arguedconvincingly that this imagerepresents the transmission ofastrology from the Greeks to theIndians according to aniconography rooted in Classical art(Grenet 2003). Next to them is arider on a horse but so far there isno satisfactory explanation for itsmeaning. After the entrance, onthe right portion of the wall, asecond enigmatic scene can bediscerned: a person wearing along robe is carrying a child whiletwo big birds spread their wings.The background is aquatic. Grenetproposed to identify this scenewith a representation of Krishnaand his foster-mother, Yashoda,just before the attack of the crane-demon Putana (Grenet 2005,128). Above them, fragments ofan image of a horse and the legsof a person could be intended asanother episode of Krishna’syouth, when he fought the horse-demon Keshin (Ibid.; Grenet2006, 45) [Fig. 5]. On the rightpart of the painting, thebackground is still aquatic butinfant-archers and a man grabbingthe tail of a bull together with a

kneeling person appear amongfishes, turtles, water-birds andflowers. Grenet identified theinfant-archers as a multiplerepresentation of Kama, theIndian god of love who wasreproduced according to aniconography borrowed fromClassical art. His presence couldbe justified because of the ‘allusionto Krishna’s romances with thecowgirls’ (Ibid.). However, asGrenet himself noted, MarkusMode had already observed thatthose archers should be betteridentified with pygmies fightingcranes exactly as described(although differently represented)in Roman sources (Mode 1993,98; Grenet 2006, 46-47). Mode’sintepretation would fit better giventhe fact that a source in Classicalart was proposed for the firstscene on the left. Moreover, thedemon-crane Putana is one, whilein the painting there are parts ofat least three birds. Unless thereare new discoveries, theinterpretation of this scene willremain open.

The long-haired man grabbingthe tail of a buffalo in the watermay recall an enigmatic sceneengraved on a panel of a 6th-century funerary bed from Chinawhich possibly belonged to aSogdian [Fig. 6].7 It is possible torecognize an archer shooting abird while sitting on an ox amongwaves, a scene which has not yetbeen identified. However, theIndian (or, better, Vishnuite)context of the panels of thisfunerary bed, recently pointed outby Penelope Riboud (Riboud 2004,46, fig. 24), may mean that itprovides a good comparison withthe detail in the Afrasiab paintingeven if rendered differently. Otherelements such as the pygmy

archers shooting cranes couldbe part of the same schemebut, unfortunately, thefragmentary state of theeastern wall does not allowone to discern more.

Among the different scenes onthe eastern wall, thefragments of ‘teacher and

student’ with armilla and theinferior parts of the horse close tothe legs of a person areproportionately larger than theother figures in the samecomposition. Possibly, this fact isnot to be neglected, especially ininterpreting the scene identifiedby Grenet as Krishna fightingKeshin, since on the other wallslarger characters have veryimportant roles.

In his mostrecent paperdevoted to theinterpretation ofthe Indian scenesat Afrasiab,Grenet produceda 5th-centuryGupta relief fromM a n d o rrepresentat ingKrishna fightingagainst Keshin(Grenet 2006, fig.2), although two

other specimens of Indiansculpture — one from a privatecollection and the other fromMathura — provide good parallelswhich support his arguments(Harle 1985, figs. 7-8). Lastly, ablack wax inscribed seal (possiblyKushano-Sasanian) kept in theAshmolean Museum deserves tobe considered [Fig. 7, next page].The figure, which appearstogether with a Bactrianinscription, could be interpeted asHeracles fighting the monstruoushorse of Diomedes or Krishnafighting Keshin according to a

Fig. 4. Eastern wall, Afrasiab (after F. Ory and Grenet reconstruction).

Fig. 5. Eastern wall, Afrasiab, detail(after Al’baum 1975, fig. 26, detail).

Fig. 6. Panel of theVahid Kooros Col-lection funerary bed(after Riboud 2004,fig. 25).

35

common iconography which waspossibly rooted in Classical art(Ibid.).8 The interpretation is againcomplicated, typical for objects ofCentral Asian provenance in whichGreek, Iranian and Indianelements co-exist, exactly as inthe painting at Afrasiab underexamination.

In any case, in our opinion, it isnot so important to figure out theexact interpretation for the detailof the man with the horse. Thevery presence of this animalrepresents an interesting element.A reasonable hypothesis can beoffered to explain its inclusion inthe cycle of paintings at Afrasiabwhere, as already observed, afunerary sacrifice in the presenceof royal characters is depicted onthe southern wall as a continuationof the Nawruz on the western wall.On the northern wall the Chineseemperor and empress too areperforming local rituals connectedwith calendrical matters whichcorrespond quite precisely to theSogdian events. In both thescenes of Sogdian and Chineserituals the presence of the horseseems to be very important,although on the northern wall theexplanation is not yet fully clear.Persian literary sources of theIslamic period explicitly indicate aconnection between the horse andwater (Cristoforetti 2006).Moreover, in the painting on thenorthern wall, two horses areswimming, since the connectionwith the aquatic element shouldhave been extremely important forthe Chinese too (Riboud 2003;Mao forthcoming). So, if the

hypothesis of the Indian characterof the eastern wall is correct andif that fragmentary paintingactually represents the lower partsof a man with a horse, then oneshould expect to find also therethe depiction of an Indian festivityor celebration connected withroyalty when a kind of horsesacrifice occurred. Such a ritual isreported in Indian literary sourcesexplicitly and is described as themost important royal sacrifice thatonly important sovereigns couldhave afforded to celebrate sinceVedic period: the ashvamedha(Dumont 1927).9

The aim of the horse sacrifice,or ashvamedha, in ancient Indiawas the recognition of a king asan universal sovereign. It was alsoa magical ritual with very ancientsolar reminiscences celebrated inorder to ensure fecundity to thekingdom. In fact, the king wasexpected to accomplish it at theend of his reign, when it wasalmost time for the succession ofthe designated new king (Ibid.,x). The preparation for theashvamedha took normally oneyear but, sometimes, even two.The sacrificial animal — which waschosen because of his color, speedand other characteristics (Albrightand Dumont 1934, 110-111) —had to be ritually immersed in apond and later was set freetogether with one hundredcastrated horses and obliged tomove in a northeastern direction.The land crossed by the horse wasconsidered to be under that king’sdominion. Thus this was a warningto the neighboring kingdoms.Young nobles, and among themthe designated successor, had tofollow the horse in order to allowit to pass wherever it wanted andprevent it from coupling with anymare for one entire year. At thetime of its return to the startingpoint, a three-day celebration tookplace during which encomiasticsinging, music and offerings hadto be performed at court. On thesecond day the sacrificial horsewas attached to the royal war-chariot and led by the king himself

to a sacred pond. When hereturned, the queen andconcubines had to embellish themane and the tail of the animalwhile praying. Then the horse waschoked in the northern part of thearea designated for the sacrifice,and the queen had to lie next toits corpse while the king and hiscompanions performed anenigmas-competition. Only at thispoint, the dead horse, togetherwith other sacrificial animals,could be quartered in the presenceof priests, nobles and commonpeople.

Ritual celebrations and othersacrifices in honor of Indiandivinities such as Agni and Somawere performed as well. Inparticular, the ritual liquid Somawas solemnly offered on the thirdday of the ashvamedha, and, later,all the objects touched by it hadto be thrown in the water(Dumont, 1927, 227). On thefourth day a purification bathoccurred for all the participants inthe ashvamedha, together withthe sacrifice of twenty-one sterilecows. Although we do not knowexactly the representation of thatIndian ritual by the Sogdians (norby the Indians themselves), atleast this part of the ashvamedhacelebration resembles very muchthe Duanwujie.

The ashvamedha took placearound February-March (but alsoin spring or summer; see Ibid., 9-10) and had very clear calendricalconnections: the wandering of thehorse around and outside thekingdom had the symbolic valueof the annual movement of the sunwhich conferred to the king thelegitimacy to govern on the landduring the period of the year.Furthermore, of considerableinterest is the fact that, accordingto some Indian literary sources,among the characteristics of thedesignated sacrificial horse was aspot on its body or a tuft of hairon the forehead with a particularshape representing the Pleiadesconstellation. There is an exactparallel between such acharacteristic of the horse of the

Fig. 7. Kushano-Sasanian inscribedblack wax seal, Ashmolean Museum(after Harle, 1985, fig. 1).

36

ashvamedha and a similar spot ortuft required for sacrificial horsesin ancient Mesopotamia. Possiblythis was due to the association ofa certain period of the year, whenthe Pleiades were particularlyvisible in the sky, with phenomenasuch as inundations by largerivers. In this way, the Pleiadescould have been seen as a link tothe ferti l ization of the land(Albright and Dumont 1934, 124-127). So here is another linkbetween horse and water inconnection with calendricalmatters. As is well-known, manyelements coming from Meso-potamian culture were acceptedin Central Asia such as, forexample, the presence of thegoddess called by Kushans andSogdians Nana who, in the Iranianmilieu, should be possiblyidentified with Anahita and,consequently, with Venus (Tanabe1995; Grenet and Marshak 1998).Astral symbols seem to haveplayed a very important role inancient Sogdiana too, and for thisreason the artists at Afrasiab wereprobably attracted by theirpresence also in other culturalspheres. Specimens of Sogdian artobjects produced both in themotherland and in the coloniesalong the so-called ‘Silk Road’present quite often images of ahorse or a pegasus with a pole onits head embellished on its topwith a half moon containing a staror a flower (Compareti 2003, 34).Since much of the horserepresented on the eastern wallat Afrasiab disappeared, anyattempt to find such astralsymbols is completely useless.However, as now seems likely, thehorse was associated with theaquatic element and had astralrelationships almost in everyancient culture; for this reason itcould have been chosen to berepresented on the walls atAfrasiab.

Other details of the eastern wallcould be considered to have aprecise parallel with elements inthe other three painted scenes atAfrasiab. Just as in the scenes

representing Sogdiana and China,where the sovereign was depictedtogether with his queen, for thecelebration of the ashvamedha thepresence of the royal couple wasrequired. The character of theIndian sacrifice concerned thelegitimacy of a king, and atAfrasiab the representation ofsuch a concept can be observedon two opposite walls: the eastern(or Indian) one and the western(or Sogdian) one where theNawruz is celebrated in con-junction with the coronation ofVarkhuman. Finally, all the sceneson the four walls present clearconnections with astrological-astronomical matters which couldbe considered reciprocally(although not exactly) balanced:when a king decided to performthe ashvamedha, this had tohappen around February-March, aperiod of the year quite close tothe celebration of the Chinese NewYear Festival (January-February).The Sogdian artists and theirpatrons were certainly aware ofthe astronomical-astrologicalcommon notions of ancient‘Westerners’ such as Indians,Greeks and Mesopotamians, butthey had to find also commonfeatures with China that, in thosedays, still followed a differentsystem and only later — mostlikely through the Sogdians —would have adopted ‘western’features such as the division of theweek into seven days, each onelinked to a planetary deity(Chavannes and Pelliot 1913, 158-177). For this reason they cleverlyopted for representations on thenorthern wall of the Chinese NewYear Festival, which fell

approximately during theashvamedha, and the Duanwu,which fell around the SogdianNawruz.

It is not possible to find an exactcontemporary parallel for thepainting of the eastern wall since,unfortunately, the few elementsreferring to the ashvamedha inIndian art are limited to therepresentation of the sacrificialhorse itself — for example, in atleast one sculpture and coins ofthe Gupta period (Huntington1985, 187-188; Lindquist 2003).Nevertheless, hints of a horsesacrifice exist in Indian art,although they date to a periodmuch later than the Afrasiabpaintings. These are the reliefcarvings of the great platform ofthe early 16th century HazaraRama temple at Hampi (ancientVijayanagara) (Dellapiccola andVerghese, 1998).10 Almost fortyyears ago, the great indologistHermann Goetz had alreadyrecognized among these reliefssome images of foreignersrepresented in processional rowstogether with animals such aselephants and horses (Goetz1967, 195-196, fig. 11). Inparticular, one relief is extremelyinteresting because of the pose ofone foreigner (possibly a Persian?)holding the bridle of a recalcitranthorse with one hand while theother seems to be beating theanimal with a kind of dagger [Fig.8]. Why would a person of possiblyIranian appearance have beenrepresented in this way? Could hispose be considered the final partof a horse sacrifice? In the pre-Islamic Iranian sphere there was

Fig. 8. Relief carving, Hazara Rama temple, Hampi (after Goetz1967, fig. 11).

37

an important funerary horsesacrifice, the chakharom, which isjust alluded to, for example, on aSogdian ossuary from Shahr-iSabz (Uzbekistan) by the solepresence of a harnessed horse(Grenet 1993, 61, n. 44).11

However, if the people in the reliefat Hampi are Iranians then theyshould be considered in allprobabil ity Muslims and notMazdeans. What is needed is anexample from Indian art dated toa much earlier period.

As noted above, in Indian artthere are very few specimenswhich can be identified asrepresentations of theashvamedha, and they are all dateto the Gupta period. Nevertheless,an interesting horn cylinder sealfrom the Greco-Bactrian site ofTakht-i Sangin [Fig. 9] which has

been recentlyattributed toIndia (Bernard,1994, 112-113) presents a scenevery similar to the relief of Hampiand, possibly, to the recon-struction of that detail of theAfrasiab painting reconsideredhere. It is not easy to give an exactinterpetation to the scene carvedon the seal, which was termed byClaude Rapin ‘an Indian myth’(Rapin 1996, 50, fig. 29b; Rapin1995, 275-281); however,what is really important toconsider here is the pose ofthe person standing in frontof the horse, which atAfrasiab as well might havefollowed an establishediconographic formula. It isnot possible to be too specificabout the pose of the manwith the horse on the hornseal since the scene isdepicted too sketchily and itcould represent just a riderwith his steed or even aprocession. Processions of

people and animals can beobserved already in very ancientPersian art (for example, atPersepolis12), and it is well knownthat the Achaemenids extendedtheir control on some parts ofnorthwestern India, where theyinfluenced local cultures. However,the processions of Persepolis donot look exactly like the scene onthe Indian horn seal, where theman seems to stand in front of theanimal without walking and hisright hand seems to grab the hiltof the sword while with the left,possibly, he holds the bridle.

Since the temple at Takht-iSangin is much older than anyother specimen of Indian artalready considered in the presentstudy, then it seems to be likelythat a pure Indian iconography isinvolved at Afrasiab. In any case,independently from the origin ofthe iconography, it is clear that theIranians were aware of it. In fact,it is possible to observe verysimilar scenes in a problematic 7th-century (?) textile fragmentacquired on the antiquarianmarket embellished with ninehorses on three parallel lines withpeople wearing caftans holding inone hand the bridles and with theother one a weapon or a stick. Theposition of those men indicates aclearly hostile attitude towards thehorses; for this reason the sceneof the textile could be intended asa representation of a sacrifice(Kitzinger 1946, fig. 46; Trilling1982, pl. 17) [Fig. 10]. As thepresent writer has already noted,

notwithstanding the long period oftime which separates the creationof the textile and the relief atHampi, they resemble aston-ishingly each other (Compareti2005, 37-40). The same could besaid for the earlier horn seal fromTakht-i Sangin. That fragmentaryPersian textile offers a good, ifenigmatic, parallel to thespecimens in Indian art. If ancientIranians knew the Indianiconography for the horse sacrificeand had even adopted it, then itcould be considered likely that itis exactly something similar thatthe Sogdians copied for therepresentation of India on theeastern wall at Afrasiab. Thehypothesis is hardly demon-strable, but at least it can offer apossible alternative identificationto the battle between Krishna andKeshin. In any case, the depictionof Indian subjects at Afrasiabaccording to that iconographycould be simply explained becausethe context there was obviouslyIranian.

An Indian King in the Paintingsat Afrasiab?

We might, finally, propose areconstruction for the upper partof the eastern wall which wascompletely lost: probably also inthis scene there was a kingtogether with his spouse; it is notunlikely that the royal couple hadlarger dimensions exactly as onthe other walls. As in the case ofVarkhuman and Gaozong on the

other walls, could this Indianking be identified? It wouldseem that Harsha (606-647)is the most probablecandidate, since he was avery powerful sovereign ofnorthern India in that period.True, the Chinese pilgrimXuanzang, who visited Indiain the first half of the 7th

century and was a respectedguest at Harsha’s court,recorded important Buddhistfestivities celebrated in thatkingdom but gave no hintsabout the ashvamedha (Beal

Fig. 9. Indianhorn cylinderseal, Takht-iSangin (afterRapin 1996,fig. 29b).

Fig. 10. 7th century fragmentary textile,Egypt? (after Trilling 1982, pl. 17).

38

1884 [1983], 206-224). Xuanzangwas a devoted follower of theDharma, which could explain hisscant interest in his memoir inreligious customs extraneous toBuddhism. However, thisexplanation would not explain whyIndian sources too, such as theHarshacharita, are silent about theashvamedha during Harsha’sreign, even though animalsacrifices should have existed in7th-century India (Kane 1918,236).

In the second part of the 7th

century, Adityasena, one of thelast representatives of the GuptaDynasty (or Later Gupta) inMagadha performed theashvamedha (Auboyer 1965, 387;Asher 1983), a fact that definitelycreated great sensation andwhose echoes, possibly, reachedother regions outside India. Forthis reason he could have beenconsidered an appropriatecandidate to appear together withVarkhuman and the ChineseEmperor Gaozong at Afrasiab.

This last hypothesis — althoughhighly evocative — is hardlydemonstrable. Despite the effortsof some students of Sogdian artto reconstruct the otherfragmentary parts of the samepainting with reference India,proof of the proposed iden-tifications is still lacking (cf. Mode2002).

About the Author

Matteo Compareti graduatedfrom The University of Venice ‘Ca’Foscari’ in oriental languages andliteratures and in 2005 defendedhis PhD in Iranian studies at TheUniversity of Naples ‘L’Orientale.’He is a specialist in art history ofSogdiana, especially pre-IslamicSamarkand, although his interestsinclude also Sasanian Persia. Atpresent his research focusesmainly on the iconography ofMazdean divinities.

References

Akhunbabaev 1987

Khasan G. Akhunbabaev.“Domashnie khramy rannesredne-vekovogo Samarkanda.” In:Gorodskaia kul’tura Baktrii-Tokharistana i Sogda, Tashkent,1987: 10-21.

Akhunbabaev 1990

Khasan G. Akhunbabaev.Sotsial ’no-planigraficheskaiastruktura kvartala znati naAfrasiabe i problema izucheniiarannesrednevekovogo Samar-kanda. Samarkand, 1990.

Al’baum 1975

Lazar I. Al’baum. Zhivopis’Afrasiaba. Tashkent, 1975.

Albright and Dumont 1934

William F. Albright and Paul-ÉmileDumont, “A Parallel between Indicand Babylonian Sacrificial Ritual.”Journal of the Royal AsiaticSociety, 54/ 2 (1934): 107-128.

Anazawa and Manome 1976

W. Anazawa and J. Manome.“Korean Envoys Depicted in theMural Painting from AncientSamarkand.” Chosen Gakuho, 80(1976): 1-36 (in Japanese).

Asher 1983

Frederick M. Asher. “Historical andPolitical Allegory in Gupta Art.” In:Bardwell Smith, ed. Essays onGupta Culture. Delhi; Varanasi;Patna, 1983: 53-66.

Auboyer 1965 (1961)

Jeannine Auboyer. L’India fino aiGupta. Milano, 1965 (original title:La vie quotidienne dans l’Indeancienne (environ IIe s. avant J.-C.-VIIe s.). Paris, 1961; in Englishas: Daily Life in Ancient India, fromApproximately 200 B.C. to 700A.D. New York, 1965).

Beal 1884 (1983)

Samuel Beal, tr. Si-Yu-Ki. BuddhistRecords of the Western World.Chinese Accounts of India.Translated from the Chinese ofHiuen Tsiang AD 629. London:Trubner, 1884 (Reprint New Delhi,1983).

Belenitskii 1961

Aleksandr M. Belenitskii. “Rabotypendzhikentskogo otriada v 1961g.” Arkheologicheskie raboty vTadzhikistane, IX (1961): 53-75.

Bernard 1994

Paul Bernard. “Le temple de Takht-i Sangin en Bactriane: temple dufeu ou pas?” Studia Iranica, 23(1994): 81-121.

Chandra 1983

Lokesh Chandra, “HellenisticEchoes in the Legend of Krishna.”In: Joachim Deppert, ed. Indiaand the West. Proceedings of aSeminar Dedicated to the Memoryof Hermann Goetz. New Delhi:Manohar, 1983: 149-154.

Chavannes 1903

Edouard Chavannes. Documentssur les Tou-Kiue (Turks) Occi-dentaux. Paris, 1903.

Chavannes and Pelliot 1913

Edouard Chavannes and PaulPelliot. “Un traité manichéenretrouvé en Chine.” JournalAsiatique, 11 sér., t. 1 (1913): 99-199; 261-394.

Compareti 2003

Matteo Compareti. “Notesull’iconografia del pegaso e delcavallo bardato nell’arte iranica.”In: Matteo Compareti and G.Scarcia, eds. Il falcone di Bistam.Intorno all’iranica Fenice/Samand:un progetto di sintesi per il volodel Pegaso iranico tra Ponto,Alessandretta e Insulindia.Venezia, 2003: 27-37.

Compareti 2004

Matteo Compareti. “Remarks onthe Sogdian Religious Iconographyin 7th Century Samarkand.” 2004:http://www.eurasianhistory.com/data/articles/a02/422.html.

Compareti 2005

Matteo Compareti. I Popoli Iranicie la navigazione nell’OceanoIndiano. Venezia, 2005.

Compareti 2006

Matteo Compareti, “A Reading ofThe Royal Hunt at Afrasyab Basedon Chinese Sources.” In:

39

Compareti and de la Vaissière2006, pp. 173-184.

Compareti forthcoming

Matteo Compareti. “On theMeaning of the Dragon in thePaintings at Afrasyab (AncientSamarkand).” Eurasian Studies, 6(2006): 64-76 (in press).

Compareti and Cristoforetti 2005

Matteo Compareti and SimoneCristoforetti. “Proposal for a NewInterpretation of the Northern Wallof the ‘Hall of the Ambassadors’at Afrasyab.” In: Central Asia fromthe Achaemenids to the Timurids:Archaeology, History, Ethnology,Culture. Materials of an Interna-tional Scientific ConferenceDedicated to the Centenary ofAleksandr Markovich Belenitsky,ed. Valerii P. Nikonorov, St.Petersburg, 2005: 215-220.

Compareti and Cristoforettiforthcoming

Matteo Compareti and SimoneCristoforetti. “New Elements onthe Chinese Scene in the ‘Hall ofthe Ambassadors’ at Afrasyabalong with a Reconsideration of‘Zoroastrian’ Calendar.” StudiaIranica, forthcoming.

Compareti and de la Vaissière2006

Matteo Compareti and Étienne dela Vaissière, eds. Royal Nauruz inSamarkand. Proceedings of theConference held in Venice on thePre-Islamic Paintings at Afrasyab.Rivista degli Studi Orientali, N.S.,LXXVIII, Suppl. n. 1. Pisa; Roma:Accademia editoriale, 2006.

Cristoforetti 2006

Simone Cristoforetti, “Afrasyabtoponimo e Afrasyab eponimo:considerazioni sulla riemergenteplausibilità di una lectio facilior.”In: Compareti and de la Vaissière2006, pp. 163-171.

Dallapiccola and Verghese 1998

Anna L. Dallapiccola, AnilaVerghese, Sculpture at Vija-yanagara: Iconography and Style.New Delhi: Manohar, 1998.

de la Vaissière 2006

Étienne de la Vaissière. “Les Turcs,rois du monde à Samarcande.” In:Compareti and de la Vaissière2006: 147-162.

Dumont 1927

Paul-Émile Dumont, L’Asvamedha.Description du sacrifice solenneldu cheval dans le culte védiqued’après les textes du Yajurvedablanc. Paris: P. Geuthner, 1927.

Fedorov 2006

Michael Fedorov. “Returning to theSogdian Incense-Burner of theLate VII-Early VIII c. AD: A Por-trait of Ikhshid Varkhuman?”Iranica Antiqua, 41 (2006): 221-231.

Fowden 2004

Garth Fowden, “The Six Kings atQusayr ‘Amra.” In: La Persia eBisanzio. Roma, 2004: 275-290.

Frye 1954

Richard N. Frye, ed. and tr. TheHistory of Bukhara. Translatedfrom a Persian Abridgement of theArabic Original by Narshakhî.Cambridge, Mass.: MediaevalAcademy of America, 1954.

Goetz 1967

Hermann Goetz. “Frühe Dar-stellungen von Moslems in derHindu-Kunst.” Oriens, 18-19(1967): 193-199.

Goossens 1993

Roger Goossens. “Un texte grecrelatif à l’ashvamedha.” JournalAsiatique, t. 217 (1930): 280-285.

Grenet 1993

Frantz Grenet. “Trois nouveauxdocuments d’iconographiereligieuse sogdienne.” StudiaIranica, 22/1 (1993): 49-67.

Grenet 2003

Frantz Grenet. “L’Inde desastrologues sur une peinturesogdienne du VIIe siècle.” In:Carlo Cereti, Mauro Maggi, and E.Provasi, eds. Religious Themesand Texts in Pre-Islamic Iran andCentral Asia. Studies in Honour ofProfessor Gherardo Gnoli on the

Occasion of His 65th Birthday.Wiesbaden, 2003: 123-129.

Grenet 2004

Frantz Grenet. “Maracanda/Samarkand, une métropole pré-mongole. Sources écrites etarchéologie.” Annales. Histoire,sciences sociales, 5-6 (2004):1043-1067.

Grenet 2005

Frantz Grenet. “The Self-Image ofthe Sogdians.” In: Étienne de laVaissière and Éric Trombert, eds.Les Sogdiens en Chine. EFEO,Etudes thématiques 17. Paris(2005): 123-140.

Grenet 2006

Frantz Grenet. “What was theAfrasyab Painting About?” In:Compareti and de la Vaissière2006, pp. 43-58.

Grenet and Marshak 1998

Frantz Grenet and Boris I.Marshak. “Le mythe de Nana dansl’art de la Sogdiane.” ArtsAsiatiques, 53 (1998): 5-18.

Harle 1985

James C. Harle. “HeraklesSubduing the Horse(s) ofDiomedes and Krishna Slaying theDemon-Horse Keshin: a CommonIconographic Formula.” In: JanineSchotsmans and Maurizio Taddei,eds. South Asian Archaeology1983: Papers from the SeventhInternational Conference of theAssociation of South AsianArchaeologists in Western Europe,Held in the Musées Royaux d’Artet d’Histoire, Brussels. Naples:Instituto universitario orientale,1985: 641-652.

Huntington 1985

Susan L. Huntington, The Art ofAncient India. Buddhist, Hindu,Jain, with contributions by John C.Huntington. New York; Tokyo:Weatherhill, 1985.

Kageyama 2002

Etsuko Kageyama. “A Chinese Wayof Depicting Foreign DelegatesDiscerned in the Paintings ofAfrasiab.” In: Iran, questions etconnaissances. Vol. I: La période

40

ancienne, ed. Ph. Huyse. Paris,2002: 313-327.

Kane 1918

P. V. Kane, ed. The Harshacaritaof Ba’abhatta (Text ofUchchhvasas I-VIII). Bombay,1918.

Kitzinger 1946

Ernst Kitzinger. “The Horse andLion Tapestry at Dumbarton Oaks:A Study in Coptic and SassanianTextile Design.” Dumbarton OaksPapers, 3 (1946): 1-72.

Lindquist 2003

S. E. Lindquist. “EnigmaticNumismatics. Kings, Horses, andthe Ashvamedha Coin-Type.”South Asian Studies, 19 (2003):105-112.

Maitdinova 1984

Guzel M. Maitdinova. “Kinterpretatsii zhivopisi AfrasiabaVII v. n. e. (stsena v lodke).”Izvestiia Otdeleniia obshchest-vennykh nauk AN TadzhikskoiSSR, 2, Dushanbe, 1984: 20-27.

Malville 2000

John McKim Malville, “The CosmicGeometries of Vijayanagara.” In:John McKim Malville and Lalit M.Gujral, eds. Ancient Cities, SacredSkies: Cosmic Geometries andCity Planning in Ancient India. NewDelhi: Indira Gandhi NationalCentre for the Arts; Aryan BooksInternational, 2000: 100-118.

Mao forthcoming

Mao Min. “The Scene of theBathing Horses at the «Hall of theAmbassadors» in Afrasyab(Samarkand): Its IconographyInfluence from China and ItsCultural Link with the NawruzFestival.” Paper Presented to theSecond Conference on TurfanStudies, Turfan, August 25th-30th

2005, forthcoming.

Marshak 1994

Boris I. Marshak. “Le programmeiconographique des peintures dela ‘Salle des ambassadeurs’ àAfrasiab (Samarkand).” ArtsAsiatiques, 49 (1994): 5-20.

Marshak 2001

Boris I. Marshak. “La thématiquesogdienne dans l’art de la Chinede la seconde moitié du VIe siècle.”Comptes Rendus de l’Académiedes Inscriptions et Belles Lettres,2001, No. 1: 227-264.

Marshak 2006

Boris I. Marshak, “ Remarks on theMurals of the Ambassadors Hall.”In: Compareti and de la Vaissière2006, pp. 75-85.

Mode 1993

Markus Mode. Sogdien und dieHerrscher der Welt. Türken,Sasaniden und Chinesen inHistoriengemälden des 7.Jahrhunderts n. Chr. aus Alt-Samarqand. Frankfurt/Main etc.:Peter Lang, 1993 (= EuropäischeHochschulschriften. Reihe XXVIII,Kunstgeschichte. Bd. 162).

Mode 2002

Markus Mode. “Court Art ofSogdian Samarqand in the 7thcentury AD: Some remarks to anold problem.” 2002: http://www.orientarch.uni-halle.de/ca/afras/index.htm, accessedJanuary 15, 2007.

Mode 2006

Markus Mode. “Reading theAfrasiab Murals: Some Commentson Reconstructions and Details.”In: Compareti and de la Vaissière2006, pp. 113-128.

Motov 1999

Iurii A. Motov, “Izobrazheniemisterii prazdnika Mikhragan vnastennykh rospisiakh afrasiab-skogo dvortsa.” In: Istoriia iarkheologiia Semirech’ia. Sbornikstatei i publikatsii. Almaty, 1999:57-81.

Rapin 1995

Claude Rapin. “Hinduism in theIndo-Greek Area. Notes on SomeIndian Finds from Bactria and TwoTemples in Taxila.” In: AntonioInvernizzi, ed. In the Land of theGryphons. Papers onCentral AsianArchaeology in Antiquity.Monografie di Mesopotamia, 5.Firenze: Le lettere, 1995: 275-291.

Rapin 1996

Claude Rapin. Indian Art fromAfghanistan. The Legend ofSakuntala and the Indian Treasureof Eucratides at Ai Khanum. NewDelhi: Manohar, 1996.

Riboud 2003

Penelope Riboud. “Le cheval sanscavalier dans l’art funérairesogdien en Chine: à la recherchedes sources d’un thèmecomposite.” Arts Asiatiques, 58(2003): 148-161.

Riboud 2004

Penelope Riboud. “Description dumonument”, and “Quelquesremarques sur l’iconographiereligieuse du monument.” In:Catherine Delacour and PenelopeRiboud, eds. Lit de pierre, sommeilbarbare. Présentation, aprèsrestauration et remontage, d’unebanquette funéraire ayantappartenu à un aristocrate d’AsieCentrale venu s’établir en Chineau VIe siècle. Paris, 2004: 15-31and 43-47.

Schiltz 1994

Véronique Schiltz. Gli Sciti. VIIIsecolo a.C.-I secolo d.C., Parigi,1994.

Silvi Antonini 1989

Chiara Silvi Antonini. “ThePaintings in the Palace of Afrasiab(Samarkand).” Rivista degli StudiOrientali, 43 (1989): 109-144.

Sinor 1990

Denis Sinor. “The Establishmentand Dissolution of the TürkEmpire.” In: Denis Sinor, ed. TheCambridge History of Early InnerAsia. Cambridge: CambridgeUniversity Press, 1990: 285-316.

Stein 1983

Burton Stein. “Mahanavami:Medieval and Modern Kingly Ritualin South India.” In: BardwellSmith, ed. Essays on GuptaCulture. Delhi; Varanasi; Patna,1983: 67-92.

Smirnova 1970

Ol’ga I. Smirnova. Ocherki iz istoriiSogda. Moskva: “Nauka,” 1970.

41

Tanabe 1995

Katsumi Tanabe. “Nana on Lion.East and West in Sogdian Art.”Orient, 30-31 (1995): 309-334.

Trilling 1982

James Tril l ing. The RomanHeritage. Textiles from Egypt andthe Eastern Mediterranean 300 to600 A.D. Washington D.C., 1982.

Yatsenko 2004

Sergey A. Yatsenko. “The Costumeof Foreign Embassies andInhabitants of Samarkand on WallPainting of the 7th c. in the ‘Hall ofAmbassadors’ from Afrasiab as aHistorical Source.” Transoxiana, 8(June 2004): http://www.transoxiana.org/0108/yatsenko-afrasiab_costume.html.

Notes

1. In 630, during the reign ofTaizong (630-649), the Chinesedefeated the Eastern Turk Empirein Mongolia and used Turkishcontingents in order to subjugatethe Western Turk Empire in theTarim Basin and Transoxiana. Thistask was completed during thereign of Gaozong between 657-659 (Sinor 1990, p. 310). Turkguards can be observedeverywhere in the paintings of thewestern wall at Afrasiab: they canbe recognized by their hairstyleand other facial traits such as theabsence of a beard. In fact,Chinese literary sources describethem as having long hair andplaits. They do not carry gifts inthe paintings, and for this reasonthey have been considered to beguards in service in the territoryof their former empire after thesubmission to the Tang.

2. A possible similar scheme is,mutatis mutandis, in the famousearly 8th century Omayyadpainting at Qusayr ‘Amra (Jordan)(Fowden 2004).

3. The king of that Sogdian region(identified with Kushanya) wentinto that pavilion to pray; so itwould not be incorrect to identify

the building as a temple too. Thisdoes not seem to have been thecase in Afrasiab where the ‘Hall ofthe Ambassadors’ has beenidentified as the private space ofa prominent person, possiblyVarkhuman himself (Marshak2006, p. 75).

4. Chinese sources also report thetime of year when the SogdianNawruz was celebrated(Compareti and Cristoforettiforthcoming). Grenet noted thatone of the Turk guards wrappedhis robe around his hips, probablysince the season was too warm(Grenet 2004, pl. B).

5. Recently Mikhail Fedorovsuggested that Varkhuman is theperson dressed like a warrior inthe left corner of the southernwall, who is, however, depicted onthe same scale as the other peoplearound him (Fedorov 2006). For adifferent interpretation of thatdetail, see Compareti 2004.

6. It is not clear if this part of thecelebration should follow orprecede the Nauwruz (Comparetiand Cristoforetti forthcoming).

7. Unfortunately, the funerarymonument was not excavatedaccording to scientific criteria, andit is now part of a private collection(the Vahid Kooros Collection,Houston).

8. The same figure of Krishnapresents some traits borrowedfrom Hellenistic culture (Chandra1983).

9. Also classical authors wereaware of the ashvamedha. In hisbook on Apollonius of Tyana,Philostratus (1st century CE) givesa description of an Indian horsesacrifice which could be onlyinterpreted as the ashvamedha(Vit. Apoll. Ty., 2, 19, 15, cited inGoossens 1930).

10. This is in all probability arepresentation of the Mahan-avami, a nine-days festival whichcomprised also horse sacrificesexactly as in the ashvamedha tobe held in March-April orSeptember-October: (Stein 1983,75-88). It is worth remembering

that the city of Hampi presents anurban plan which denotes acomplex system of astralrelationships (Malville 2000).

11. Other harnessed horsespossibly ready for sacrifice can beobserved in at least one paintingat Panjikent and on a 6th-centurySogdian funerary bed from China(Belenitskii 1961, p. 72, fig. 15;Marshak 1994, pp. 11-15).

12. The motif was wide-spreadamong many ancient culturesconnected with ancient Persiafrom very ancient times to theIslamization of Iran and CentralAsia. Processions of this kind canbe observed, in fact, at Pazyryk(3rd century BCE; see Schiltz 1994,p. 284, fig. 215) but also in someunpublished fragmentaryterracottas from 10th-11th centuryDvin (Armenia), directly observedby this writer in the State Museumof Armenian History.

42

Mapping Early Buddhist Sites in Western Tibet:Recent Findings from Tsamda County, ChinaKarl E. RyavecUniversity of Wisconsin, Stevens Point, WI, USA

In this article, I present recentfield survey findings of earlyBuddhist sites in Ngari or westernTibet during August 2006. Thesites are in the core region of theformer Guge kingdom [Fig. 1],and date from the 10th to 13th

century Second Diffusion ofBuddhism period, known inTibetan as the ‘Chidar’ (phyi dar).This survey of abandoned templesand cave murals at several locales

in Tsamda county of Ali prefecturein the Tibet Autonomous Regionis designed to study the spatialpatterns of newly discovered sitesin relation to the better knownsites of historical importance. Inthis way, it may be possible todeduce approximately how manysites probably lie in the canyonsof Ngari awaiting detaileddocumentation. Historical Tibetanrecords do not contain maps, and

so it is often difficult to locate earlytemples listed. Also, many smallershrines and cave murals are notrecorded in the surviving historicalrecords, and it is thus necessaryto survey areas to document sites.

Interest in western Tibet hastended to focus on the region’shistorical importance inspearheading the SecondDiffusion of Buddhism, and the

i c o n o g r a p h i corigins of TibetanBuddhist art.W e s t e r ns c h o l a r s h i plargely startedwith the expe-ditions andfindings of thenoted Tibet-ologist GiuseppeTucci (1933). Hisinterests weresimilar to thoseof Sir Aurel Steinwho earlier ex-plored ancientBuddhist sitesalong the SilkRoad during thefirst decades ofthe twentiethcentury. In bothcases, theseearly archae-ological and art

Fig. 1. Religious and cultural sites in the core region of the GugeKingdom, ca. 10th-14th centuries CE. Copyright © 2007 Karl E.Ryavec

43

history surveys revealed a regiondessicated by climate change overthe past two millennia withnumerous settlement areasabandoned due to dried-upirrigation sources. Recently,archaeological investigations ofwestern Tibetan pre-Buddhist andBuddhist sites have commencedunder collaborations between theAmerican archeologist MarkAldenderfer (2001, 2004), and theChinese archaeologists Hou Weiand Li Yongxian (Hou and Li 2001)of Sichuan University. And I havestudied the aerial extents ofancient abandoned farmland inseveral canyons of the formerGuge kingdom based on 2-footresolution satellite imagery, andmade some preliminary findingson likely historical populationlevels (Ryavec 2005).

Tsamda county lies in the heartof the canyon country ofsouthwestern Tibet carved by theSutlej river of the Induswatershed, and bordered on thesouth by the Himalayan frontierwith India. This region containsmany ancient acropolis fort sites,some probably dating from thefirst millennium BCE Iron Age.Also, the terrain abounds inhabitable caves. Traditionally,many people l ived in cavesespecially during the cold wintermonths. Under the Tibetan empireca. 7th to 9th centuries, westernTibet was annexed and referred toas Zhangzhung. The indigenousBonpo religion remained strongwhile Buddhism seems not to havemade any inroads, unlike inCentral Tibet where it was fosteredas a court religion of theemperors. Little is known of thispast period, with studies by asmall number of specialistsreferring to Zhangzhung in termsranging from a coalition ofterritorially based lineages togroup of kingdoms, and an empire(Bellezza 2003). However, mostscholars agree that theZhangzhung capital was located atKyunglung on the eastern borderof Tsamda county where the Sutlejriver originates on the Tibetan

Plateau west of sacred MountKailash. After the collapse of theTibetan empire in the 9th century,the new Guge kingdom that arosein the 10th century was sometimesstill referred to as Zhangzhung.

It is not clear what sort ofbroader cultural influences mayhave interacted with the historicaldevelopment of Zhangzhungduring the pre-Buddhist period.The nearest major cultural hearthin this part of the world was theHarrapan, or Indus valleycivilization that flourished c. 5000– 2000 BCE, long before availablearchaeological evidence pertainingto Zhangzhung appeared. Whatimpact the urban Harrapan culturemay have had on historicaldevelopments in western Tibet isspeculative at best, as outlined byGeoffrey Samuel (2000).Furthermore, George van Driem(1998) postulates that thenorthern Neolithic culture ofKashmir (2500 – 1700 BCE), muchcloser to western Tibet, representsa colonial exponent of the northChina Majiayao Neolithic culture(ca. 3900 – 1800 BCE). Van Driemaptly points out that this provides

an explanation for the fact thatTibeto-Burmans inhabit both sidesof the Himalayas, the greatestnatural land barrier on earth.Certainly it is reasonable toassume these settlements in theupper Sutlej River valley had anagrarian base long beforeBuddhism arrived, but some timeafter the historical domesticationof the important grain crop plantsof wheat and barley in the MiddleEast about 10,000 years ago.According to J. P. Mallory andVictor Mair (2000), the weight ofcircumstantial evidence for theearly settlement of eastern CentralAsia during the Mesolithic periodindicates western origins. Inaddition to the cultivation ofcereals, archaeological evidencefor the domestication of sheep alsofavors this western originhypothesis. The extent to whichwestern Tibet was geographicallyconnected with these earlydiffusions and later tradingnetworks requires further detailedinterdisciplinary research.

Most of the Buddhist monas-teries in western Tibet were builtunder patronage of the western

Fig. 2. The acropolis fort complex above the present-day village andmonastery of Mangnang. Note the numerous caves of the former resi-dents. Photograph copyright © 2006 Karl E. Ryavec.

44

Tibetan Guge dynasty (ca. 10th to17th centuries CE), which becameinstrumental in supporting a

resurgence of Buddhist art andliterature. During this period manyacropolis fort sites were re-developed to accommodatetemples. In fact, the former Gugecapital at the fortress of Tsaparangis the main destination of mosttourists going to Tsamda countytoday, and is described in detailin guide books to western Tibet.

Yet almost every canyon in thecounty has an ancient acropolisfort complex that saw the addition

of Buddhist temples during the‘Chidar’ period. Buddhist monksalso painted cave murals as theydid throughout the Silk Roadregion.

New Sites

The acropolis fort complex [Fig. 2,facing page] above the village ofMangnang [Fig. 3] contains aruined temple with an approx-imately 4 m 12th-century Buddhiststatue still largely intact [Fig. 4].Mangnang is well known as thesite of an 11th century templewhere the famed Bengali Buddhistmaster Atisha was invited to stayone year ca. 1043 before travelingto Central Tibet where he passedaway. Available guidebooks toNgari make no mention of thisimportant surviving statue in thefort above the Mangnang villagetemple complex.

Bedongpo village lies in a smalltributary of the Sutlej atapproximately 4,200 m. Ruins ofa fort/monastery complex towerabove the vil lage [Fig. 5].Historical records show that themonastery of Bedongpo was abranch of the main monastery ofToling constructed in 996 CEduring the beginning of theBuddhist period in western Tibet.Various other ruined temples arealso found in the Bedongpo valley,

Fig. 3. Mangnang village, and ca. 11th century temple complex. Himalaya fron-tier between Tibet and India in background. Photograph copyright © 2006 KarlE. Ryavec.

Fig. 4. Ca. 12th century ruined statueof a Buddhist deity, approximately 4m tall, on the top of the Mangnang fortcomplex. Photograph copyright © 2006Karl E. Ryavec

Fig. 5. Ruined ca. 12th century fort/temple complex above the village ofBedongpo. Note small new Buddhisttemple at foot of site. Photographcopyright © 2006 Karl E. Ryavec

45

but only near the village was asmall temple rebuilt after China’spolitical and economic reforms inTibet during the 1980s. About 2km downstreamfrom Bedongpois a relativelyunknown sitecalled Drisa[Fig. 6]. Num-erous chortendating fromcirca the twelfthcentury lie onthe valley floorbelow an exten-sive complex ofcaves and anacropolis fort.Traces of redpigment on thewalls of at leastone of theruined fortb u i l d i n g si n d i c a t ep o s s i b l etemples. No traces of houses,however, are noticeable, indicatingthat the former farmers probablylived year-round in the caves. Theentire valley here below Bedongpovillage consists of large tracts ofabandoned farmland, alsoindicated by a surviving chortennear the edge of an ancient field[Fig. 7].

The Khyunglang cave [Fig. 8]contains 3 walls with surviving

circa twelfth-century murals [Figs.9, 10, facing page]. There aresome other caves in the valley, butladders are required to access

them. The Khyung-lang cave lies about5 km upstream fromShangtse fort, thereputed summercapital of the Gugekings.

Conclusion

Tsamda county con-tains many surviving

early Tibetan Buddhist forms of artand architecture relativelyunknown except to a handful ofspecialists. The area’s importanceas one of the core agrarian basesof the kingdoms of Guge andPurang starting in the tenthcentury led to royal patronage forsubstantial temple construction.

There are several problems,however, that make it difficult ifnot impossible for scholars andtourists to visit most of these sites.In recent years, tourists to Tsamdacounty have generally only beengranted access to the maintemples at Toling in the countyseat, the nearby fort and templesat Tsaparang, and the Dungkarcaves and fort easily accessedfrom the main road into thecounty. Chinese border securityconcerns place most of the other

sites off limits,ostensibly dueto sensitivityabout the loca-tions of militarybases andpatrols. Chinamight do wellto learn fromits neighborIndia, whichnow allowsscholars totravel within 1km of itsclaimed border

Fig. 6. Drisa fort complex with ca. 12th century chorten. Photograph copyright© 2006 Karl E. Ryavec.

Fig. 7. Ca. 12th century chorten and aban-doned farmland in the Bedongpo valley.Note the numerous caves of the formerresidents. Photograph copyright © 2006Karl E. Ryavec.

Fig. 8. Entrance to the Khyunglang cave.Photograph copyright © 2006 Karl E. Ryavec.

46

with China. But China, unlikeIndia, has to deal with its citizensattempting to flee the country.Thus the wider off-limit areas onthe Chinese side reflect, in part, amethod of preventing people fromgetting close enough to the borderto walk into India. A second

concern relates to efforts byChinese cultural offices preventingphotography at the sites. Thisproblem could be easily solved ifthose scholars who have alreadytaken high-resolution photos (andofficial Chinese-sponsored surveyshave already done this) made

them freely available. This wouldthen make it possible for morepeople to visit the sites without theneed to take photographs of themurals, as complete images wouldalready be available.

Acknowledgements

I would like to thank the TibetanAcademy of Social Sciences andProf. David Germano of theUniversity of Virginia for facilitatingmy field research in the TibetAutonomous Region of Chinaunder the general auspices of theTibetan and Himalayan DigitalLibrary project.

About the Author

Dr. Karl E. Ryavec is an AssistantProfessor in the Geography/Geology Department at theUniversity of Wisconsin, StevensPoint. His Ph.D. dissertation(University of Minnesota, 2002)was entitled “Land Use Change inCentral Tibet, c. 1830-2000.”Starting in 2004, Dr. Ryavecconducted field research inTsamda county in China tounderstand better the growth anddecline of the western TibetanGuge kingdom. E-mail:<kryavec@uwsp.edu>.

References

Aldenderfer 2001

Mark Aldenderfer. “Piyang: aTibetan Buddhist Temple andMonastic Complex of the 10th -11th

Centuries AD in Far WesternTibet.” Archaeology, Ethnology, &Anthropology of Eurasia, 2001,4(8):138-146.

Aldenderfer and Moyes 2004

Mark Aldenderfer and HolleyMoyes. “Excavations at Dindun, apre-Buddhist village site in farwestern Tibet.” In: Huo Wei andLi Yongxian, eds. Essays of theInternational Conference onTibetan Archaeology and Art.Chengdu, China: Center forTibetan Studies, Sichuan UnionUniversity, 2004: 47-69.

Fig. 9. One of the three c. 12th century mandala murals in the Khyunglang cave.Photograph copyright © 2006 Karl E. Ryavec.

Fig. 10. Detail of a Guge warrior in one of the three c. 12th century mandalamurals in the Khyunglang cave. Photograph copyright © 2006 Karl E. Ryavec.

47

Bellezza 2003

John V. Bellezza. “Ancient Tibet:Bringing to Light the Forgotten. Acomprehensive survey: of Pre-Buddhist sites in Upper Tibet(1992-2002).” Athena Review, 3/4 (2003), online at <http://w w w . a t h e n a p u b . c o m /12tibet.htm>.

Huo and Li 2001

Huo Wei and Li Yongxian. “XizangZhada Xian Piyang-Dongga Yi Zhi1997 Nian Diaocha Yufa Jue”[Survey and Excavation of thePiyang-Dongga Site in ZandaCounty, Tibet in 1997]. ActaArchaeologica Sinica, 3 (2001):397-426.

Mallory and Mair 2000

J. P. Mallory and Victor H. Mair. TheTarim Mummies: Ancient China

and the Mystery of the EarliestPeoples from the West. London:Thames & Hudson, 2000.

Ryavec 2005

Karl E. Ryavec. “Aerial Survey ofAbandoned Agricultural Fields inthe Ancient Tibetan Kingdom ofGuge: Recent Findings from 2-footResolution Quick Bird Imagery ofBedongpo and Environs.” AerialArchaeology Research GroupNewsletter, 30 (2005): 18-25.

Samuel 2000

Geoffrey Samuel. “The IndusValley Civilization and Early Tibet.”In: Samten G. Karmay andYasuhiko Nagano, eds. NewHorizons in Bon Studies. BonStudies, 2. Osaka: NationalMuseum of Ethnology, 2000: 651-670.

Tucci 1933

Giuseppe Tucci. Secrets of Tibet:Being the Chronicle of the TucciScientific Expedition to WesternTibet. Translated from the Italianedition by Mary A. Johnstone.London; Glasgow: Blackie & Son,1933.

Van Driem 1998

George Van Driem. “NeolithicCorrelates of Ancient Tibeto-Burman Migrations.” In: RogerBlench and Matthew Spriggs, eds.Archaeology and Language. Vol.2. London; New York: Routledge,1998: 67-102.

48

Han Lacquerware and the WineCups of Noin UlaFrançois LouisBard Graduate Center, New York

Lacquer work is today recognizedas one of the centrally distinctivecomponents of Han materialculture (206 BCE–220 CE). What’smore, the Former Han period (206BCE–8 CE) has come to becelebrated as the apogee ofChinese lacquer art (see Wang1982, 80–99; Prüch 1997; Fuzhou1998; Barbieri-Low 2001; and Li2004 for further reading on Hanlacquer). These insights arerelatively recent and entirely theresult of archaeological dis-coveries. Precisely a century haspassed since the firstarchaeological discovery andidentification of a lacquer vesselfrom Han China. In the spring of1907, while surveying the Hanborder fortifications north ofDunhuang — and just weeksbefore coming upon thesensational medieval library at theMogao Caves — Aurel Stein dryly

recorded his historic find, awooden ear cup with scrollornament from the ruins of a Hancommand center (Stein 1921, Vol.2, 645; Vol. 4, pl. LII). SinceStein’s discovery, and especiallyover the past forty years,archaeologists have unearthedthousands of Han and even pre-Han lacquer artifacts, severalhundred of which are fortunatelystill in fine condition.

We now know that the use oflacquer as a protective, water-proof coating made from the sapof the lacquer tree (rhusverniciflua) goes back to Neolithictimes in China. But as anartistically emancipated craft,lacquering came into its own onlyin the late fifth century BCE in thestate of Chu in southern China.From that time on it was thepreferred means of decoration for

all types of wood-based artifacts,whether vessels, boxes, furniture,musical instruments, arms,chariots, or coffins. By the Qin(220–206 BCE) and early Haneras, lacquering had become soprominent a craft that certainvessels were even produced as‘pure’ lacquer artifacts without awood substrate, using insteadlacquer-drenched ramie fabric tobuild a core.

As a commodity, lacquer workwas in many respects akin towoven silk during the Han era.Both had relatively little intrinsicmaterial value. Made fromrenewable resources, silk andlacquer products, unlike artifactsmade of jade and gold, werevalued primarily on the basis oftheir design and manufacture.This meant that they could bemade to cater to a relatively broadspectrum of the population. Plainsilk fabric and utensils simplyvarnished in raw brown lacquerwere widely available andessential commodities. Butpatterned silks with complexweave structures and glossy,colored lacquers with artfullypainted red and black decoration

or even gold and silver inlays couldbe very expensive and functionedabove all as means of socialdistinction. The quality of lacquerwork found in archaeologicalcontexts can thus explain muchabout the wealth and socialposition of its last owner; it caneven il luminate his or herrelationship to the Han imperialcourt.

While the discovery of Hanlacquer ware in a military stationfrom the ancient Han frontier maynot be as spectacular as the findsfrom aristocratic tombs near bigtowns, it is by no means unusualor surprising. Lacquered artifactswere available everywhere in theHan Empire, even though themajority was made in thelacquering workshops of centraland southern China where lacquertrees grew abundantly. Hanlacquers have even been found inareas far beyond the ancient Hanfrontier, as far north as Lake Baikaland as far west as Begram inAfghanistan (for the Begram findssee Hackin 1954, 295-297, figs.243–249; Mehendale 2005,1.4.3).

The lacquer artifacts from suchdistant sites are sti l l poorlyunderstood, despite the fact thatmany of them were already foundin the 1920s and 1930s. Althoughwidely discussed early on, theyhave received little attention sincethe major discoveries in thePeople’s Republic of China tookcenter stage. Now, however, inlight of recent insights on Hanlacquer and in view of new

discoveries in Mongolia andBuryatia, it is worth taking a freshlook at some of the early finds.

Most recently, Han lacquerartifacts have been reported froma number of Xiongnu cemeteries.While some evidence wasdiscovered in tombs in the TamirRiver Valley in eastern Arkhangai,some 300 kilometers west ofUlaanbaatar (Waugh 2006), themajority of finds come from themountains between Ulaanbaatarand Lake Bai-kal north of it( M i n i a e v1998; Torbatet al. 2003).Of these re-cent finds alacquered Hanchariot iscertainly themost extra-o r d i n a r y(Miniaev andSakharovskaia2006). Themost sig-nificant earlydiscoveries,h o w e v e r ,remain thewine cupsdiscovered inthe mid-1920s in theNoin Ula

Mountains, about 100 kilometersnorth of Ulaanbaatar.

Four relatively well-preservedlacquer cups from Noin Ula havebeen adequately published [Figs.1–5]; a few more have beenreported though not illustrated inthe main surveys. Unfortunately,no proper excavation report of thetombs was ever prepared, as itspotential authors had fallen victimto Stalinist terror (Maenchen-Helfen 1965). And the Japanese

Fig. 1. Wine cup, dated 2 BCE, fromNoin Ula tomb 6. Seen from the side.State Hermitage Museum, St. Peters-burg (photograph © 2005 DanielC.Waugh).

Fig. 2. Wine cup, dated 2 BCE, from Noin Ula tomb 6. Seen from the bottomwithout handles. State Hermitage Museum, St. Petersburg (photograph afterUmehara 1960, pl. 59).

Fig. 3. Wine cup from Noin Ula tomb 23. Mid-first centuryCE. National Museum of Mongolian History, Ulaanbaatar(photograph courtesy of Thierry Ollivier).

49

archaeologist Umehara Sueji(1893–1983), who was able tostudy the lacquers extensivelysoon after their discovery andplanned on publishing a majoranalysis of the site, lost most ofhis research materials during the1945 napalm bombing of Tokyoand had to reconstruct his bookmanuscript after the war. Finally,there has been some confusiondue to the dividing of the Noin Ulafinds between the StateHermitage in St. Petersburg andthe Museum of Mongolian Historyin Ulaanbaatar.

Umehara explains that the fourHan lacquer cups come from three

different tombs, a fact thatappears to have escaped severallater authors (Umehara 1960, 28–32, pls. 59–62). The Russianexpedition of 1924–25 led by PetrK. Kozlov, according to Umehara,found an inscribed and dated cupof 2 BCE in the large kurgan 6[Figs. 1, 2] and two un-inscribedcups in kurgan 23, about 100meters west of kurgan 6 (Figs. 3,4). Sergei I. Rudenko latermentions in his inventories of theKozlov expedition, published in1962, that tomb 23 actuallycontained four lacquer cups, oneof which [Fig. 3] had beenreturned to Mongolia (Rudenko

1969, 112). The fourth cup wasnot discovered by Kozlov, but byMongolian scholars who inves-tigated the tombs in summer1927. This cup, found in fourfragments, is also inscribed anddated to the year 2 BCE [Fig. 5],but was discovered in tomb 5,which lay in the vicinity of tomb6. It has always been kept inUlaanbaatar (Umehara 1944, 16;Umehara 1960, 29). The two cupsnow in Ulaanbaatar have recentlybeen shown in two travelingexhibitions in Europe (Paris 2000,147; Bonn 2005, 51), where theywere both assigned to tomb 6,without any mention of Umehara’saccount of the 1927 investigationsor the reports that the uninscribedcup [Fig. 3] was found in the innerburial chamber of tomb 23, northof the coffin (cf. Trever 1932, 47,pl. 29, 1; Rudenko 1969, pl. 48).These discrepancies are likely theresult of oversights by thecatalogue authors. The cata-logues, especially the one fromParis, do, however, have the virtueof providing outstanding colorillustrations of Noin Ula lacquers.

The two cups from tombs 5 and6 [figs. 1, 2 and 5] carry importantinscriptions that identify them asofficial products manufactured ingovernment workshops for theimperial court. Both share thesame basic design of facing birdsand spirals, yet they show verydifferent styles: one bold the othertender and fragile. These stylisticdifferences have been recognizedas typical for two distinct regionalstyles — those from Sichuan andShaanxi, respectively. As AnthonyBarbieri-Low has illuminated in hisexcellent recent dissertation, thecup from tomb 5 [Fig. 5] is atypical example of the thousandsof mass-produced vessels fromthe two imperial luxury workshopsin Shu and Guanghan; fifty lacquervessels from these Sichuanworkshops are still known today(Barbieri-Low 2001, 421–422).The cup is of beautiful quality,despite being mass-produced, andis explicitly designated in itsinscription as ‘fit for use by theemperor (chengyu).’

Fig. 4. Wine cup from Noin Ula tomb 23. Mid-first century CE. State HermitageMuseum, St. Petersburg (photograph after Umehara 1960, pl. 63).

Fig. 5. Wine cup, dated 2 BCE, from Noin Ula tomb 5. National Museum ofMongolian History, Ulaanbaatar (photo courtesy of Thierry Ollivier).

50

In contrast, the cup painted withfine lines from kurgan 6 [Figs. 1,2] is unique in the archaeologicalrecord. It was produced in apalace workshop in the capitalChang’an in Shaanxi by themaster artisan Wang Tanjing anddesign painter Hu. On the bottom,separate from the incisedinscription made by themanufacturing office, the cupbears the two additional largecharacters ‘Shanglin,’ which referto the imperial park in Chang’an.This reference led Barbieri-Low tospeculate that the lacquerworkshop might actually havebeen located in the imperial park.It is, however, much more likelythat the inscription was simplypart of the palace inventorysystem that designated the cup foruse in the one of the imperialpalaces in Shanglin Park.

There is indeed good reason tothink that this cup was used inShanglin Park in the year 1 BCEto host the chief Xiongnu leader,Shanyu Wuzhuliuruoti (r. 8 BCE–13 CE). According to Ban Gu (32–92 CE), this shanyu had requestedan audience at the capital as earlyas the year 3 BCE. But the Hancourt extend the formal invitationonly after months of deliberationscentering on the vast expense andthe bad luck occasioned by visitsfrom the Xiongnu leaders. Asubsequent illness of the shanyufurther delayed the visit. When theshanyu finally arrived, he came inthe company of five hundred men,more than in any delegationbefore, and all of them eager toexperience the Han court’s wealthand generosity. The emperor, whofor astrological reasons haddecided it best to ‘reside in theGrapevine Palace (Putao Gong) inShanglin Park and to treat hisguest with additional honors,’invited the shanyu to stay atShanglin Park as well — a privilegeregular courtiers could only dreamof. ‘The shanyu, appreciative ofthis favor, was also regaled with370 robes, thirty thousand boltsof various patterned silk fabrics,and thirty thousand pounds of raw

silk, in addition to the same giftsas had been given in the year 25BCE [to his predecessor]’ (Hanshu94B.3817).

Ever since the Xiongnu and theHan court had reached a peaceaccord in 53 BCE, diplomaticexchange between the two hadintensified. By the end of the firstcentury BCE there was a wellestablished system of tributarytrade between the two rulers,according to which the emperorprovided huge gifts to the shanyu,who in turn acted nominallysubmissive, promising to keep thepeace and enable mutual trade.Before Wuzhuliuruoti’s visit in theyear 1 BCE, there were fourinstances of a Xiongnu shanyuattending an audience at the Hancourt, each more generouslyrewarded than the one before(Barfield 1989, 63–67). Back in hisown country, as Thomas Barfieldhas pointed out, the shanyu wasobliged to distribute among hisnobil ity the wealth he hadobtained through this tributarytrade (Barfield 1981). Such giftdistribution among the Xiongnunobility offers one explanation asto why the two imperial lacquercups were reportedly found indifferent tombs. It is of course alsopossible that the occupants oftombs 5 and 6 were both at theChinese court in 1 BCE. Indeed ithas often been suggested thatkurgan 6 is the tomb ofWuzhuliuruoti Shanyu himself(Paris 2000, 146).

The two cups made in the year2 BCE, undoubtedly in anticipationof the expected Xiongnu visit, arethe earliest lacquer cups from NoinUla. The ear cups found in tomb23 belong to an altogetherdifferent category. They are ofnoticeably lower quality than thevessels made in the governmentworkshops, have no officialinscriptions, differ in style, andmay be as much as seven decadesyounger than the cups dated to 2BCE (cf. a vessel from the tombof Wang Xu [d. after 69 CE] inPyongyang; Harada 1930, pl. 61).The symmetric and yet organic

cloud and scroll ornament withinterspersed animals that wasomnipresent during the FormerHan period [Figs. 1, 2 and 4],faded away over the course of thefirst century CE. In the 40s, aftertwo decades of civil war followingthe murder of Wang Meng (r. 9-23 CE), the aristocracy of the LaterHan dynasty (25-220) aban-doned the old luxuriantornamental style in favor of afrugal one, and the imperiallacquer workshops henceforthproduced only undecoratedvessels, red on the inside, blackon the outside, before finallyshutting down for lack of fundsearly in the second century. Withthe trendsetting imperial eliteforfeiting luxurious display, theHan lacquer industry becomesfiercely commercial, and the oldornamental style survives only inan increasingly simplified,downgraded form. The two cupsfrom tomb 23 represent thiscommercial type of ware typical oflacquer production in the firstcentury CE. Most likely these cupsreached the Xiongnu not as officialimperial gifts but through trade.We can view them as evidence forfree forms of private tradebetween the Han and Xiongnustates, conducted both at Hanfrontier markets, by the countlessmembers of the embassies thatwent to and fro, and by themilitary stationed along theborders. Yü Ying-shih has adeptlydescribed this kind of frontiertrade in his classic study (Yü 1967,93–132).

Because lacquerwares weremade in only some regions inChina, they were among thefrequently traded goods in the HanEmpire. And because lacquerswere unique Chinese productswith excellent qualities, it seemsreasonable to assume that theyfigured prominently in interna-tional trade — like bronze mirrorsor silks (Maenchen-Helfen 1973for an overview of bronze mirrorsin Xiongnu contexts). EarlyChinese sources are full of proudreferences to the infatuation ofHan’s neighbors with Chinese

51

goods, yet the sources virtuallynever mention lacquer specificallyas an export good. Lacquerevidently did not fit the Hanrhetoric of wealth in the same waygold, jade, and silk did. When theHan court provided wine cups totheir tributary delegations, theyhanded them out as party favorsafter the main banquet rather thanas serious gifts worth entering intothe national records. Never-theless, some recipients of suchtokens of imperial grandeur —Chinese officials perhaps more sothan others — treasured them,sometimes over generations, andrarely ever used them.

Not surprisingly, imitations ofthe imperial wares were alsoavailable on the market. Some, asBarbieri-Low has uncovered, weredeceptively similar to authenticofficial wares, complete with fakeofficial inscriptions (Barbieri-Low2001). The majority, however,were of lesser quality andemphasized either the bold redline perceived as typical of theimperial Sichuan style or the finered lines typical of themetropolitan style of the FormerHan capital, Chang’an. Thefragments of lacquer vesselsrecently discovered in Xiongnutombs in the Tamir Valleyrepresent such commercialcategories from around the mid-first century CE (Waugh 2006).The bronze-mounted handle of anear cup found in Feature 201 [Fig.6], for instance, continues thecharacteristic design of diagonal

lines seen on ear cup handlesmade for the court earlier on [cf.Fig. 5]. The bowl found in Feature97 [Fig. 7], on the other hand,combines on its black exterior astriking pattern of sketchy redlines and dots — a fadedcontinuation of the old imperialSichuan-style — with a design ofdelicate scrolls painted in red,gray, and yellow. Such multi-colored painted ornament wastypical for the commercialproducts of the early Eastern Hanera (25-220 CE) and hinted at theearlier tradition of expensive goldand silver inlays. Like manyEastern Han artifacts for personaluse, such as bronze mirrors, silks,or jewelry, this lacquer bowl wasmagically charged to enhance

fertility through its auspiciousinscription yi zi sun, ‘may it bringyou sons and grandsons’ [Fig. 8].

The regularity with whichremains of Han lacquers are foundin Xiongnu tombs of the late firstcentury BCE and the first centuryCE suggests that the Xiongnu eliterecognized fine Han lacquers asprestigious and useful possess-ions, if not for their associationwith the Chinese court, then fortheir appeal as exotic commoditiesand their connection to theXiongnu ruling family, whosepolicies resulted in the availabilityof Chinese artifacts. A moreprecise picture of Xiongnuperceptions and uses of fineChinese commodities such as theperishable lacquers, however,awaits further research.

Fig. 7. Han lacquer bowl fragment, first century CE, found at Tamiryn UlaanKhoshuu, Feature 97, Arkhangai aimag, in 2005 (photograph © 2005 DanielC. Waugh).

Fig. 6. Bronze-mounted handle of ear cup found at Tamiryn Ulaan Khoshuu,Arkhangai aimag, Feature 201, in 2005 (photograph © 2005 Daniel C. Waugh).

Fig. 8. Inscription on Han lacquer bowl,first century CE, found at TamirynUlaan Khoshuu, Feature 97, Arkhangaiaimag, in 2005 (photograph © 2005Daniel C. Waugh).

52

About the Author

François Louis is an AssociateProfessor of Chinese Art andDesign History at the BardGraduate Center for Studies in theDecorative Arts, Design, andCulture in New York. He haspublished widely on early andmedieval Chinese metalwork andis currently working on a book onKhitan material culture.

References

Barbieri-Low 2001

Anthony Barbieri-Low. “TheOrganization of ImperialWorkshops during the HanDynasty.” Unpublished Ph.D.dissertation, Princeton University,2001.

Barfield 1981

Thomas J. Barfield. “The Hsiung-nu Imperial Confederacy: Organ-ization and Foreign Policy.” Journalof Asian Studies, 41/1 (1981): 45–61.

Barfield 1989

Thomas J. Barfield. The PerilousFrontier: Nomadic Empires andChina, 221 B.C. to A.D. 1757.Cambridge, Mass.; Oxford:Blackwell, 1989.

Bonn 2005

Kunst und Ausstellungshalle derBundesrepublik Deutschland,Bonn, ed. Dschingis Khan undseine Erben: Das Weltreich derMongolen. Munich: Hirmer, 2005.

Fuzhou 1998

Zhongguo qiqi quanji bianjiweiyuan hui. Zhonggo qiqi quanji:3 Han. Fuzhou: Fuzhou meishuchubanshe, 1998.

Hackin 1954

Joseph Hackin, J.-R. Hackin, JeanCarl, and P. Hamelin, eds.Nouvelles recherches archéol-ogiques à Begram (ancienneKâpicî), 1939–1940. Mémoires dela Délégation archéologiquefrançaise en Afghanistan, t. 11.Paris: Imprimerie Nationale,1954.

HanshuBan Gu. Hanshu [History of theFormer Han Dynasty]. Beijing:Zhonghua shuju, 1962.

Harada 1930Harada Yoshito. Lo-lang: A Reporton the Excavation of Wang Hsü’sTomb. Tokyo: Tôkô Shoin, 1930.

Li 2004

Li Zebin, ed. Han Guangling guoqiqi [Lacquer from the Han Stateof Guangling]. Beijing: Wenwuchubanshe, 2004.

Maenchen-Helfen 1965Maenchen-Helfen, Otto J. “Reviewof Umehara, Moko Noin Urahakken no ibutsu. (Studies ofNoin-Ula Finds in North Mongolia)and Rudenko, Kul’tura khunnov iNoinulinskie kurgany.” ArtibusAsiae, 27/4 (1964 - 1965): 365–369

Maenchen-Helfen 1973Otto J. Maenchen-Helfen. TheWorld of the Huns: Studies in theirHistory and Culture. Berkeley etc.:University of California Press,1973.

Mehendale 2005Sanjyot Mehendale. Begram Ivoryand Bone Carvings. <http://ecai.org/begramweb>. Berkeley:University of California, 2005.

Miniaev 1998

Sergei S. Miniaev. Dyrestuiskiimogil’nik. Arkheologicheskiepamiatniki Siunnu, vyp. 3. St.Petersburg: Evropeiskii dom,1998.

Miniaev and Sakharovskaia 2006

Sergei S. Miniaev and L. M.Sakharovskaia. “Khan’skaiakolesnitsa iz mogil’nika Tsaram.”Arkheologicheskie vesti, 13(2006).

Paris 2000

L’Asie des steppes: d’Alexandre leGrand à Gengis Khan. Paris:Réunion des Musées Nationaux,2000.

Prüch 1997

Margarete Prüch. Die Lacke derWestlichen Han-Zeit. Frankfurta.M.: Peter Lang, 1997.

Rudenko 1969Sergei I. Rudenko. Die Kultur derHsiung-nu und die Hügelgräbervon Noin Ula. Bonn: Habelt Verlag,1969. [Translation of Kul’turakhunnov i Noinulinskie kurgany,Moscow-Leningrad, 1962.]

Stein 1921Aurel Stein. Serindia: DetailedReport of Explorations in CentralAsia and Westernmost China. 5vols. Oxford 1921. Available onlineat: <http://dsr.nii.ac.jp/toyobunko/VIII-5-B2-9/>.

Torbat et al. 2003T. S. Torbat, U. Erdenebat, and D.Seveendorzh. Egiin golyn savnutag dakh’ arkheologiindursgaluud: khurli in ueesMongolyn ue. Ulaanbaatar, 2003.

Trever 1932Camilla Trever. Excavations inNorthern Mongolia (1924–1925).Leningrad: J. Fedorov, 1932.Available on-line at: <http://depts.washington.edu/silkroad/museums/shm/trever.pdf>, withthe photographs displayedseparately at: <http://depts.wa s h i n g t o n . e d u / s i l k r o a d /museums/shm/shmno inu la .html>.

Umehara 1944

Umehara Sueji. Shina Kandaikinenmei shikki zusetsu [Cata-logue of inscribed and datedlacquers from Han China]. Kyoto:Kuwana Bunseido, 1944.

Umehara 1960Umehara Sueji. Môko Noin Urahakken no ibutsu [Studies of NoinUla Finds in North Mongolia].Tokyo: Toyo Bunko, 1960.

Waugh 2006

Daniel C. Waugh. “The Challengesof Preserving Evidence of ChineseLacquerware in Xongnu Graves.”The Silk Road, 4/1 (2006): 32–36.

Wang 1982

Wang Zhongshu. Han Civilization.New Haven: Yale Univ. Pr., 1982.

Yü 1967

Yü Ying-shih. Trade and Expansionin Han China. Berkeley: Universityof California Press, 1967.

53

Trade and Commerce on the Silk Road after theEnd of Mongol Rule in China, Seen from ChineseTextsRalph KauzInstitute of Iranian Studies/Austrian Academy of Sciences, Vienna

Traffic and exchange on the SilkRoad are generally perceived tohave taken place more before andduring Mongol rule in Asia than inlater periods. That is, in this viewthe ‘Silk Road’ is a historicalphenomenon which came to a haltsometime during the late‘European’ Middle Ages. Fewwould think about merchants andenvoys crossing Asia well after theMongols, before the Europeans‘re-discovered’ Central Asia. Thispaper will discuss the traffic alongthe route after the Mongols inorder to demonstrate that the SilkRoad did not break off completelyin the middle of the 14th century,but continued to function for arather long time subsequently.

The historical documentation oftrans-Asian contacts in Chinesetexts started during the Handynasty, when Emperor Han Wudisent Zhang Qian, the first well-known traveller along the SilkRoad, to the Yuezhi to propose apolitical and military coalitionagainst the Xiongnu. Around thebeginning of the Common Era, theHan conquered vast areas of the‘Western Regions’ (Xiyu), andconsequently the Han Empirestretched far into Central Asia.After the downfall of the Han,interactions between Western,Central and Eastern Asiapersisted, though on a lessinstitutionalized level. They roseagain under the rule of thecosmopolitan Tang dynasty, whenSogdian and other merchantsbrought various exotics to Chinaand exported Chinese silks andother products in exchange. ForPeroz, son of the last Sasanianshah, China was the only haven

from the onslaught of the ArabMuslims. Already during the Tang,but much more during thefollowing dynasties, directoverland contacts betweenEastern and Western Asia declinedto a certain degree; however,maritime trade increasedenormously instead.

These circumstances changedunder the Mongols, when travelerslike John of Plano Carpini andWilliam of Rubruck, not to mentionMarco Polo and others, broughtreports and narratives of Centraland Eastern Asia under Mongolrule back to Europe and implantedthe idea of (relatively) free traveland trade into the heads of theircountrymen. Nowadays images ofjourneys on the Silk Road aremostly connected with Marco Poloand the Mongol period. Few peoplewould probably think of theaforementioned Han ventures,Sogdian traders, and even less ofpersons like the Chinese officialChen Cheng, the Timurid painterGhiyas ad-Din and the PortugueseJesuit Benedict Goës, who crossedCentral Asia decades and evencenturies after the fall of theMongols.1 What were the motivesof these latter travelers and whatwas the historical and politicalbackground of their endeavors?

The famous Great Wall of China,in the shape we can appreciatetoday, was built during the Mingperiod (1368-1644), and in thepopular view it is still a symbol ofChinese seclusion (cf., however, DiCosmo 2006). In fact though,according to a major historicalsource of this dynasty, the‘Veritable Records’ (Ming shilu) aswell as other texts, numerous

embassies arrived in China fromCentral and Western Asia duringthis period. They number 89embassies coming from theCentral Asian center Samarkandalone — the first arrived on 27October 1387 and the last on 3May 1618.2 If we take the average,about one embassy fromSamarkand arrived in either of thetwo Chinese capitals under theMing (Nanjing and later Beijing)every three years, though wemust admit that this calculation ispartly incorrect because manymore embassies came at thebeginning of the dynasty than atits end.

The embassies arriving fromSamarkand during the Mingdynasty seemingly reveal theclose continuous contacts withCentral Asia. However, Samarkandstill lies in Central and not inWestern Asia. If we turn furtherwest (or better southwest), wefind that 21 embassies arrivedfrom Herat (in 14 cases writtenHalie and in 7 Heilou) between theyears 1409 and 1484. Herat wasthe second capital of the Timuridsafter Shahrokh ascended thethrone in 1405 and from this basehe re-conquered parts of hisfather’s empire from various rivals.Herat was the contemporaneouscapital of Khorasan — andKhorasan can absolutely beregarded as a province of Iran.3

But Herat was not the furthestwestern city mentioned in theMing records: three embassiescame from Kerman, the importantcenter of southeastern Iran,written in the ‘Veritable Records’with different spellings (1415,1424, 1425), and we also find five

54

embassies from Isfahan (1419-1483) and seven from Shiraz(1413-1484). The last three citieswere (and are) definitely on whatis considered Iranian soil and atleast at the beginning of the 15th

century they belonged to theTimurid Empire. Thus it seemsthat we may affirm commercialrelations between Iran and Chinataking place well into the 15th

century.

Relations between China andWestern Asia apparently wenteven beyond the region of Iran:we learn of 23 embassies arrivingfrom Arabia (or Mecca), which theChinese called Tianfang4 (1433-1618), and of five coming from acountry called Roumi (1423-1445)and finally eleven others from onepronounced similarly, namely Lumi(1524-1618). Roumi and Lumishould be both the transcriptionof ‘Rum,’ the popular term forAnatolia and pars pro toto for theOttoman Empire.

According to Chinese texts,therefore, embassies from placesas far as Samarkand, Arabia andthe Ottoman Empire arrived inChina during the whole Mingdynasty, though the numbers atthe beginning of the dynastysurpass those in the second halfby far. As a result continuousexchanges between Western andEastern Asia seem to have takenplace even centuries after theMongols’ conquests.

However, the real basis of thedata gathered from the ‘VeritableRecords’ can be doubted, and onemay consider all or at least anumber of these embassies to be‘fake embassies.’ First, one mayquestion whether embassies fromArabia and the Ottoman Empirestill arrived in China in the late 16th

and even early 17th centuries,because Asia was separated bypolitical and religious obstaclesduring that period. Another reasonfor these doubts is the existenceof various Persian credentialletters to the Chinese Emperorwhich were found as attachmentsto Persian-Chinese glossaries of

the period. These glossaries werecompiled by the ‘Muslim Office’(Huihui guan), a sub-office of the‘Office of the Four Barbarians’ (Siyiguan) which was in charge oftranslations from variouslanguages and thus essential fordiplomatic and commercialexchange. Linguistic inconsis-tencies and errors allude to theprobability that the texts werewritten by corrupt Chinese officialsfor ‘fake embassies.’5 These lettersindicate that the embassies camefrom places much closer to Chinathan they pretended in order toprofit from the lucrative tributesystem. Finally, the Chinese textsshow that the Chinese officialsthemselves were often not sure ofthe veracity of many embassies.One early example recorded bythe critics of the SupervisingSecretary (jishizhong) Huang Ji on26 December 1424 may be givenhere: ‘Many of the envoys aretrading barbarians who conducttheir private businesses under afalse pretext to bring tribute andunder false authorization in orderto attain an official position...’(Ming shilu 1966: Renzong shilu,j. 5, p. 160).

The reality of intense trans-Asian exchanges well after the fallof the Mongols in Iran and Chinamay be doubted after theseremarks. To il luminate thispreliminary reasoning we willconsider briefly the history of theregion before we turn again to theembassies and put them in theframework of the broader politicalcontext of the period. Finally, wewill address the question, ‘Why didthe Silk Road decline?’

When Zhu Yuanzhang defeatedthe Mongols and founded the Mingdynasty in 1368, he was forced todeal with areas across the bordersof China, because the rest of theMongol armies fled with the lastYuan emperor Toghan Temür intothe steppes.6 Serious Mongolpretenders survived until 1388,some of them recognized even bythe Chinese as having legitimateclaims. Thus Zhu Yuanzhang hadto include the regions north of

China into his strategic schemes.Furthermore, many people ofCentral Asia (the so-calledsemuren) had settled in Chinaduring the Yuan dynasty, andsome of them were obviouslybrought back to ‘Samarkand’ as itis written in Chinese texts, but thistoponym probably includes otherplaces of Central Asia as well.

Timur came to power at aboutthe same time as the first MingEmperor; he founded a CentralAsian empire with its capital atSamarkand. Though of Turkicorigin, Timur stood in the traditionof the Mongols and legitimized hisrule by marrying a woman withChingisid lineage. His pretensionsclearly aimed far beyond the areain which he grew up, the core ofthe Ulus Chaghatay dominion. Heconquered Iran, led his army intoIndia and far into Asia Minor wherehe defeated the Ottoman SultanBayazid, but the east, Chinaincluded, was to be spared untilthe end of his rule. Finally, Timurgathered his army to attack China,but he died in the early stages ofthis campaign at Utrar, on 18February 1405.

Both rulers saw themselves aspossible successors of theMongols, and their respectiveaspirations reached beyond thearea they actually ruled. Thus it isnot surprising to find them spyingout the other’s ambitions. In factwe cannot prove this statement,yet the texts handed down maylead one to assume it. Timur sentseveral embassies to the Chineseemperor, bringing the greatlydesired horses as tribute withthem and Zhu Yuanzhang alsosent embassies to the ruler ofCentral Asia and Iran. We maysuppose that Timur presentedhimself as a tributary to theChinese in order to gathernecessary information on thestrategic conditions in early MingChina. The first of these embassiesarrived in late 1387 and was sentby the ‘son-in-law’7 Timur — thusstate the Chinese texts. TheseTimurid embassies (some elevencan be counted) continued until

55

1396, and all of them brought withthem relatively high numbers ofhorses which we will rememberwere the most sought after tributeitem by the Chinese. Why theywere brought to a temporary endis not quite clear. Fortunately,however, we can reconstruct theprobable development of therelationship between the Timuridsand the Ming of those years froma number of sources: first theaforementioned Chinese texts;second, Timurid historiographerswho also disclose an amount ofimportant information; and thirdeven Europeans, namely theCastilian Ruy Gonzáles de Clavijoand the Bavarian JohannSchiltberger, who both sojournedin Samarkand at the turn of thecentury. They did not leave much,but it is all the more valuableinformation because of this.

A letter, allegedly sent by Timurhimself, seemed to have beencrucial for the interruption ofrelations between both empires(Kauz 2005, 64-67). In this letter,which was received in late 1394at the Chinese court, Timurunconditionally acknowledged thesuzerainty of China in mostobedient terms. As a result theChinese government sent anembassy under the leadership ofthe official Fu An and others,among them the eunuch Liu Wei,8

to Samarkand to thank Timur forhis submission. Chinese andTimurid texts differ completely intelling the story. According to thefirst, Fu An refused to kowtow toTimur and was ordered to travelaround the Timurid Empire torealize its size. Strangely enough,the Timurid Sharaf ad-din Yazdigives a much more favorableimpression of the reception: Fu Anwas allowed ‘to kiss the carpet’ —a high favor — and was graciouslydismissed. Whatever happened,the embassy must have been afailure, because out of 1,500accompanying soldiers only 17returned in 1409.

Zhu Yuanzhang died in 1398,and his death was followed byyears of struggle for the

succession. The victor of thesestruggles was Zhu Di, son of thefirst Ming emperor and governorof Beijing. His rule of 22 years(1402-1424) as the YongleEmperor may be considered oneof the most remarkable reigns ofChinese history. Zhu Di pursuedan expansionist foreign policy. Themovement of the capital fromNanjing to Beijing, close to theMongols who were stil l athreatening enemy, can be seenas the symbol of this policy. Themost important objectives of thispolicy were:

- Large-scale campaignsagainst the Mongols withparticipation of the emperor;- Occupation of Vietnam;- Maritime expeditions to theWestern Indian Ocean underthe leadership of admiralZheng He;- Active foreign policy towardsCentral Asia.

It is certainly the last objectivewhich is of interest here, but theCentral (and Western) Asia policycan only be seen in the context ofthe overall foreign policy. Oneexample of this is the presence ofTimurid envoys travelling overlandand Hormuzian envoys travellingoverseas to Beijing at the sametime in early 1421! Zhu Di hadwoven a close net connecting themajor parts of Asia with China atits center. This is certainly anextraordinary feature in thehistory of Chinese foreignrelations.

Before we turn to the peak ofthese relations, their ratherdisastrous beginning must berevealed. Only a short time afterhis inauguration, Zhu Di sent anembassy ‘to Samarkand, Heratand other places to present theemirs there with brocade’ (Mingshilu 1966: Taizong shilu, j. 15,p. 270). Thus goes the officialreason; the factual motive,however, was probably to inquireafter the remainders of the formerChinese embassies and to resumetributary relations. According tothe accounts of the Europeansalready mentioned, Clavijo and

Schiltberger, this mission probablyhad a disastrous outcome. Clavijotells us that the Chinese shouldbe hanged (Clavijo 1928, 223-225), and we read nothing of thereturn of these envoys in Chinesetexts. Timur wanted to deliver thetribute in person and he startedto organize a campaign againstChina in late 1404. The motivesbehind his plans remain obscure;his self-exculpating words aboutleading his army against non-believers should be doubted. Hismain objective was more likelydirected against the Moguls, whopossessed the eastern part of theformer Ulus Chaghatay; againstChina the campaign might havebeen a raid similar to the one heled against India some yearsbefore (Kauz 2005, 75-78). Luckilyfor the Chinese, the campaigncame to no avail, because, as wehave seen, Timur died in its earlystages.

Timur’s successors did notcontinue his aggressive politics,but pursued instead peacefulexchanges with China: Fu An andalso Chen Dewen, anotherChinese envoy, were allowed toreturn home with the rest of theirrespective entourages. Thefollowing years witnessed themost intensive exchange betweenthe two empires. This exchangewas documented in both Chineseand Timurid sources, which allowus to give a fairly exact picture ofthe interactions. It was theChinese Emperor Zhu Di who wasthe driving force behind them; hisprobable aims were to propagatehis rule and legitimize it by thenumerous audiences given toforeign envoys. Possibly a militaryalliance against the continuingMongol attacks was also part ofhis scheme. The Timurids on theother hand were primarilyinterested in the various Chineseexports such as silks andporcelain, though we know at leastone occasion when Shahrokh, theTimurid ruler until 1447, wantedto impress Chinese envoys byordering the extensive decorationof his capital Herat.

56

Some letters which wereexchanged between the two rulershave survived in Chinese andTimurid sources and have becomethe object of various translationsand research (e.g., Fletcher 1968;Kauz 2005, 93-129). They showthat diplomatic misunder-standings were overcome byrather pragmatic responses: inparticular the Chinese Emperor didnot insist on his alleged superiority— which was of paramountsignificance for the Chinesepolitical system — andacknowledged the Timurid ruler as(nearly) his equal. An almostmodern system and network ofdiplomatic and political exchangesbetween Western, Central andEastern Asia developed for the fewdecades of the Yongle reign. Thisregional network was certainlyconnected with the internationalnetwork under Ming guidance.

However, further developmentsshow that these interactionsdepended almost exclusively onthe foreign policy of Emperor ZhuDi. After the death of this emperoron his return from a campaignagainst the Mongols in 1424, theChinese embassies to the Westmet with increasing difficulties. Itseems likely that they neverreached their destinations in theWest, although embassies fromthe West continued to arrive inrather large numbers. Thus,mutual contacts continued, albeitin a rather one-sided manner. Oneexample may be given: the‘Veritable Records’ tell of envoysarriving at the Chinese court fromQi’erman (Kerman), Roumi (Rum/Ottoman Empire) and Kuncheng(Kun-City = Qom?) on 14September 1425. They broughthorses and the ubiquitous ‘localproducts’ as tribute with them andreceived silks, brocades and clothin return (Ming shilu 1966:Xuanzong shilu, j. 7, p. 184, j. 8,pp. 205, 216). Here we maychallenge again the credibility ofthis embassy: it is possible thatKerman and Qom sent envoys toChina because both cities werewithin the sphere of the Timurid

Empire, but envoys from theOttoman Empire seem much lesslikely, since it was rather moreorientated towards Europe inthose years. However, it is notaltogether impossible thatOttoman merchants/envoystravelled through the whole ofWestern Asia and joined theirTimurid colleagues somewhere inCentral Asia. We may recall thatSchiltberger returned in thesesame years on a similar route backhome to Bavaria.

Chinese texts record embassiesfrom Western and Central Asiauntil the very end of the dynasty,whereas the number (andsuccess) of Chinese embassiesdeclined rapidly after the death ofZhu Di. We may thus turn to theseefforts from the Chinese sidebefore we consider again theWestern. During the Xuanzong era(1425-35), the foreign policy ofEmperor Yongle was not yetcompletely abandoned: one lastmaritime expedition was sent tothe Western Indian Ocean (1431-33), and three embassies weresent to Central Asia. However, thefirst succeeded, whereas the latterthree in all probability did not.Their orders were all given in thesecond half of Xuanzong’s rule, butthey met with difficulties at theChinese borders where Tatars andTibetans maltreated the localpopulation and no doubt alsoforeign missions, though it seemsthey might have been less severeon the Central Asians than on theChinese (Kauz 2005, 162-172). Itis not clear at exactly which partof the route the Chinese envoysdecided to return, because allincidents are badly recorded.These records do not support adefinite conclusion — they justallow a most probable assumptionthat the embassies did not crossthe borders of China. Theinsecurity at the borders and theChinese military incapabilityobviously hindered their passagebeyond China.

However, the tribes in China’swest were not the major menacefor China; the attacks of the

Mongols had much greater impact.The Oirats or Western Mongolsbecame the most importantsource of danger for the Ming inthe 1440s, even managing tocapture the Chinese emperor ZhuQizhen in the battle of Tumu in1449. Nevertheless, the Chinesegovernment was still capableenough to react quickly andinstalled his half brother on thethrone. After their captive lost hispolitical importance, the Mongolsreleased the former emperor, andthe Ming faced the awkwardsituation of housing anotherpossible emperor inside the wallsof the capital. A number of militaryand civil officials finally overthrewthe Jingtai Emperor in 1457 andinstalled the former emperor totake up his reign again under thetitle ‘Heavenly Harmony’(Tianshun).

Zhu Qizhen had obviouslylearnt his lesson from captivity bythe Mongols and consideredpreventive measures again,possibly even further assaults onthe Mongols. Two further intendedMing embassies (1457 and 1463)to the Timurids must probably beregarded as parts of these politicalschemes, especially because bothof them had to be conductedunder military guidance. However,the course of these missionsshows the almost completestructural military and admin-istrative incompetence whichmade sending any more envoysto the West out of the question. Ithad proved to be difficult to obtaineven the necessary horses for theundertaking. The envoys and theirentourages struggled hard toreach Hami or, in the case of thesecond mission, did not evenarrive at this oasis close to China.Maybe the latter did not evenleave the capital (Kauz 2005, 211-219). Thus the last Chineseattempts to exchange embassieswith the West came to an end.

This did not mean that thetraffic collapsed completely. TheCentral Asian side had obviouslyfar less difficulty sending envoysand merchants to China. One large

57

embassy which arrived in early1453 deserves special mentionbecause the Chinese texts relatethat no fewer than 121 towns sentenvoys — it seems that the wholeMiddle East and Central Asia senttheir men to China. The reasoncould have been the inaugurationof the aforementioned JingtaiEmperor. Their great numberseems to nourish our suspicionthat we are dealing with fakeembassies just adopting names offoreign places. Fortunately, oneimperial Chinese edict written in1453 in Mongolian and Chinesehas survived in Turkish archives.The recipient was a ‘small vassalkingdom’ named La’er, probablythe transcription of Lar in southernIran, a trading center of mediumimportance in the Western IndianOcean region. This documentproves that contacts betweenChina and rather smallprincipalities in the Middle Eastactually existed.

However, the Chineseadministration gradually lostcontrol of the various tributeembassies arriving in the capital.The number of envoys increased(sometimes dramatically), but thetribute they brought was lackingboth in quality and quantity. Alsoembassies came more often thanthey were entitled to do andstayed much longer in China thanbefore. Central Asians obviouslylived more comfortably in Chinathan they did back home; thisassumption is also confirmed bynumerous requests of ‘people ofSamarkand’9 to settle down inChina. But it must beacknowledged that far fewerembassies arrived in China fromthe middle of the 15th century untilthe end of the dynasty than in thepreceding hundred years or so,the ratio being approximately 35to 75. For the last period theannals of the Ming dynasty madethe following comments: ‘In thereign of Wan li (1573-1620) theintercourse with Samarkand wasstill animated, for those foreignersliked to carry on trade with theChinese people. Besides this, it

was the custom that when theyhad entered China, the Chinesegovernment took charge of theirmaintenance’ (Bretschneider1910, Vol. 2, 267).

Thus the history of theinterchanges between Western,Central and Eastern Asia under theMing dynasty can be roughlydivided into two major periods:the time until the mid-15thcentury and the period afterwards.In the time around 1500 theMiddle East and Central Asia sawa number of major changes: theShi’i Safavids came to power inIran and partially blocked thetraffic between the Sunni rulers inthe East and in the West of theirdominion. The Timurids perished,though they found a successor inBabur who conquered northernIndia in 1526, and were replacedby the Uzbeks, who, however,could not establish an empire asstrong as Timur’s. It can generallybe said that the political andeconomic importance of CentralAsia declined rapidly after 1500.The discovery of the maritimeroute to India and China by Vascoda Gama and the followingEuropean expansion towards theIndian Ocean did not have muchinfluence over the Silk Roadtraffic.10

To conclude, the embassiesfrom Timur to the new dynasty inChina received a favorableresponse, mainly because theybrought much desired horses withthem. However, when Chinareciprocated and sent Chineseenvoys to Samarkand, these weremade much less welcome,probably because of the Chineseattitude to degrade Timur to amere subordinate of the MingEmperor. The relations betweenthe Timurids and the Mingdeveloped very well only afterTimur’s death, and the Minginitiatives proved paramount forthe development of intercourse onthe Silk Road. The Chinese sidewas rather more interested in thepropaganda of its superiority,whereas the Timurid side favoredcommercial aspects. China

stopped sending embassies to thewest after several failuresdisclosed its military andadministrative incompetence. TheTimurids and their successorscontinued to send embassies,though these were exclusivelycommercial in character. It is mostprobable that a number of theseembassies did not come fromCentral Asia at all, but startedsomewhere near the borders ofChina and just disguisedthemselves as coming from placesfurther away. However, it may beconfirmed that the traffic towardsChina carried on almost until thevery end of the Ming dynasty,although on a rather lower scale.

[Portions of this article werepresented in a lecture, co-sponsored by the SilkroadFoundation and ColumbiaUniversity, at the Asia Society inNew York on November 11, 2005.]

About the Author

Doz. Dr. Ralph Kauz is a researchscholar at the Institute of IranianStudies/Austrian Academy ofSciences, Vienna. His thesis for hishabilitation on Ming-Timuridpolitical and economic relationswas published in 2005 (seecitation below). Among his otherrecent publications are “PostalStations in Ming China (1368-1644),” in: Angela Schotten-hammer, ed., Trade and TransferAcross the East Asian“Mediterranean” (East AsianMaritime History 1) (Wiesbaden:Harrasowitz, 2005): 75-89, and“Bengali Textiles as Tribute Itemsto Ming China,” in: Rosemary Crill,ed., Textiles from India: TheGlobal Trade (Calcutta: SeagullBooks, 2006): 39-55. He wrote inThe Silk Road, 3/1 (2005) abouthis project to publish a translationof the “Khataynameh” (Book ofChina), written by `Ali AkbarKhata’i in 1516.

58

References

Amiot 1789

Jean Joseph Marie Amiot. “Recueilde suppliques, lettres de créance.”Mémoires concernant l’histoire, lessciences, les arts, les mœurs, lesusages & c. des Chinois. 15 vols.Paris, 1776-1791. Vol. 14 (1789):239-308.

Bretschneider 1910

Emil Bretschneider. MediaevalResearches from Eastern AsiaticSources. Fragments towards theKnowledge of the Geography andHistory of Central and WesternAsia from the 13th to the 17th

Century. 2 vols. London: Rout-ledge & Kegan Paul, Trench,Trübner & Co., 1910.

Clavijo 1928

Ruy González de Clavijo. Clavijo:Embassy to Tamerlane, 1403-1406. Guy Le Strange, tr. and ed.London etc.: Harper & Brothers,1928. Extended excerpts alsoavailable on-line at <http://depts.washington.edu/silkroad/texts/clavijo/cltxt1.html>.

Di Cosmo 2006Nicola Di Cosmo. “The Origins ofthe Great Wall.” The Silk Road, 4/1 (2006): 14-19.

Fletcher 1968Joseph F. Fletcher. “China andCentral Asia.” In: John KingFairbank, ed. The Chinese WorldOrder: Traditional China’s ForeignRelations. Cambridge, Mass.:Harvard University Press, 1968:206-24, 337-368

Kauz 2005Ralph Kauz. Politik und Handelzwischen Ming und Timuriden:China, Iran und Zentralasien imSpätmittelalter. Iran – Turan 7.Wiesbaden: Reichert Verlag,2005.

Ming shilu 1966Ming shilu, 133 vols. Taibei:Zhongyang yanjiuyuan lishi yuyanyanjiusuo, 1966.

Rossabi 1963Morris Rossabi. “A Translation ofCh’en Ch’eng’s Hsi-Yü Fan-Kuo

Chih.” Ming Studies, 17 (Fall1963): 49-59.

Rossabi 1975

Morris Rossabi. China and InnerAsia: From 1368 to the PresentDay. London: Thames andHudson, 1975.

Rossabi 1990

Morris Rossabi. “The ‘decline’ ofthe central Asian caravan trade.”In: James D. Tracy, ed. The Riseof Merchant Empires: Long-Distance Trade in the Early ModernWorld, 1350-1750. Cambridgeetc.: Cambridge University Press,1990: 351-370.

Thackston 1989

Wheeler M. Thackston, tr. and ed.A Century of Princes. Sources onTimurid History and Art.Cambridge, Mass.: The Aga KhanProgram for Islamic Architecture,1989.

Watanabe 1975

Hiroshi Watanabe. “An Index ofEmbassies and Tribute Missionsfrom Islamic Countries to MingChina (1368-1644) as recorded inthe Ming Shi-lu, classifiedaccording to Geographic Area.”The Memoirs of the Toyo Bunko,33 (1975): 285-326

Yule and Cordier 1916Henry Yule and Henri Cordier, tr.and ed. Cathay and the WayThither: Being a Collection ofMedieval Notices of China. Vol. 4.London: Hakluyt Society, 1916.

Notes

1. For an English translation ofChen Cheng’s account, seeRossabi 1963; for Ghiyas-ad-din,see Thackston 1989, 279-297; forGoës, Yule and Cordier 1916, 198-254, and on-line at <http://depts.washington.edu/silkroad/texts/goes.html>.

2. I used Watanable 1975 to countthe number of embassies fromCentral and Western Asia to Chinain the period mentioned. Thenumbers might not always becorrect to the last digit, butdifferences are minor.

3. It is always difficult to applytoponyms to different historicalperiods. When I speak of Iran inthis article, I mean the regionwhose borders roughly encompassthe modern country.

4. According to the hithertounpublished Chinese glossary‘Zengxu zazi’ Tianfang is thecountry of the ‘Kingdom of theKaaba’ (mamlakat-e ka’beh).

5. Some of these letters werealready published in 1789; seeAmiot 1789.

6. For an overview of Chineserelations with Inner Asia beginningunder the Ming, see Rossabi 1975.

7. Chinese fuma, a translation ofthe Mongolian word kürgän, asTimur called himself. Son-in-law ofChingis Khan is meant, thuslegitimating his rule.

8. Eunuchs played a crucial rolein diplomatic and tribute relationsduring the entire Ming dynasty.

9. ‘Samarkand’ was here also usedas a collective term for CentralAsia as a whole.

10. On the decline of the overlandtrade, cf. Rossabi 1990.

59

Hunting Hounds along the SilkRoad – Which Way Did They Go?Sir Terence ClarkLondon

In 2001 I journeyed for the firsttime along part of the Silk Roadfrom Khiva to Bukhara,Samarkand, Tashkent, Bishkek,Lake Issyk-kul and Almaty torealise an ambition I had nurturedsince studying Russian atCambridge years before. Thefollowing year I journeyed alonganother part from Bishkek overthe Tien Shan to Kashgar, thenthrough the oasis towns on thesouthern side of the Taklimakanbefore crossing the desert to thenorthern oases and on toDunhuang, finally leaving viaUrumqi to Almaty. During thesejourneys I came across varioushunting hounds peculiar to CentralAsia but similar in many aspectsto the Saluki, the hunting houndof the Middle East, where I hadspent much of my professionalcareer. I was curious to knowwhether there was indeed arelationship between them and, ifso, whether these hounds owedtheir origins to Western or CentralAsia.

Hunting hounds of the Salukifamily [Fig. 1], characterised by

their long limbs, deep chest,tucked up waist, wedge-shapedhead on a long neck, pendulousears and whippy tail, have beenknown to exist in Western Asiasince at least the fourthmillennium BCE. Archaeologicalevidence from this period in theshape of seal impressions fromTepe Gawra innorthern Iraqshows repre-sentations ofsuch hounds inpursuit ofcervidae (Clark1995, 132).Similar hounds,though with dis-tinctive prickedears and a tailcurled over theback and gener-ally known asTesem also occurin Ancient Egyptfrom ca. 3,750-3,400 BCE(Brewer 2001, 32). In the absenceof evidence elsewhere to thecontrary, it would seem likely thatthe Saluki type of hunting hound

emerged first on the greatplains of Mesopotamia,where they were used forhunting mainly by sight thewhole range of the abundantgame then to be foundthere, as well as predatorssuch as fox, jackal and wolf.However it cannot beexcluded that among thenomadic tribes of CentralAsia, who may have left fewtangible clues, similarrequirements for a huntinghound on the steppes andsemi-deserts there mayhave shaped a similar kindof hound, possibly with a

denser coat to protect it from thecolder climate [Fig. 2].

It has been shown recently thatthe dog was first domesticatedfrom the wolf in Eastern Asia andspread outwards from thereacross the world probably about15,000 years ago or possibly40,000 years ago (Savolainen etal.). According to the Russiancynologist V. A. Gorodtsov huntingand other types of specialiseddogs emerged in Central Asia inthe Neolithic period (8-10,000BCE) (Plakhov and Shelestovaforthcoming). This accords witharchaeological evidence in

Western Asia and leaves aconsiderable period of millenniafor distinctive breeds of huntinghounds to have developed beforethe advent of the Arabian Salukito Central Asia, which Russiancynologists, such L. P. Sabaneev,A. A. Sludskii and E. I.Shereshevskii, generally agreecame in the wake of the Muslimconquests in the 7th and 8th

centuries and was crossed withlocal breeds to produce the CentralAsian Tazy (Ibid.). Sabaneev saysthat the admixture of blood fromthese local breeds brought abouta change in the smooth-hairedSaluki’s appearance to a longercoat, fringes, pendulous ears anda generally coarser build. He doesnot specify what these localbreeds were but only that theyFig. 1. Smooth-haired Iraqi Saluki. Photo-

graph copyright © 2007 Terence Clark.

Fig. 2. Central Asian Tazy in Kazakhstan. Photographcopyright © 2007 Terence Clark.

60

were longhaired mountain dogswith pendulous ears (Sabaneev1993, 13).

However this attribution of theorigin of the Tazy to the eastwardmovement of Salukis with theconquering Arabs seems tooverlook other evidence thatsuggests the presence of verysimilar smooth-haired huntinghounds in the region at muchearlier dates, which might equallyhave contributed to thedevelopment of the Tazy. Thereare for example petroglyphs fromthe 1st and 2nd millennium inKazakhstan and Kyrgyzstan thatshow stylised dogs in huntingscenes, though in general theyappear to have pricked ears andan upward curling tail more likethe Tesem of Ancient Egypt [Fig.3].

In the 4th century BCEAlexander the Great and hisSeleucid successors hadestablished an empire from theMediterranean to Afghanistan andthe Indus. It is known that theGreeks used Saluki-like hounds forhunting. Indeed there are groundsfor believing that the very nameSaluki comes from the Arabic wordSaluqi for Seleucid (Smith 1980).The Greeks traded with westernCentral Asia and their influencecan be seen in some of the

S c y t h i a n s ’s p l e n d i dexamples of thegoldsmith’s art,including thefamous pectoralfrom the TolstaiaMogila, showingsmooth-ha i redhounds witheither pricked orcropped ears inpursuit of hare( P r z e z d z i e c k i2001, 67). TheSeleucids’ capitalat Seleucia inc o n t e m p o ra r yIraq was retainedby their suc-cessors — theParthians, though

the following Sasanians built a newtown at Ctesiphon on the oppositeside of the Tigris in 226 CE — andwas the great hub controllinginternational trade from Rome toChina that became known as theSilk Road (Valtz Fino 2005, 149).Much of this trade for some 500years was in the hands of theSogdians, based around Bukhara,Samarkand and the Ferghanavalley, who were for long exposedto Hellenistic influence, not leastsince Alexander’s wife Roxannewas a Sogdian (Sinor 1990, 175).So it is entirely possible that Salukihounds were being conveyedeastwards along that route wellbefore the Arab conquests. Indeedif we look further east in China inthe Qin Period (221-207 BCE) wefind funerary bricks from nobletombs with graphic examples ofsmooth-haired hunting houndswith cropped or pricked ears inhunting scenes (Przezdziecki2001, 98). Similar hounds appearon stamped bricks (Ibid.) and instone reliefs (Schafer 1985, 77)later in the Han Period (206-220CE).

Be that as it may, it is clear thatby the 7th century hounds lookingremarkably like contemporarySalukis were being represented inart across Central Asia and China.In a remarkable exhibition at the

British Library on the Silk Road in2004 there were two 7th centuryTang dynasty terra cotta figurinesof such hounds in unmistakableposes: one sitting upright and theother crouching on the crupper ofa horse behind its huntsmanmaster [Fig. 4]. A magnificentmural in an imperial 7th centuryTang tomb near Chang’an showsa falconer with a Sparrowhawk onhis arm accompanied by abeautifully represented featheredSaluki (Whitfield 1999, 89). A 9th-century painted scroll in the BritishMuseum of the Paradise ofBhasajyaguru from Dunhuang inthe time of the Tang dynastyshows a smooth-haired houndSaluki reaching for a piece ofmeat. Another Tang paintingshows two Salukis in a butcher’s

shop, while a Song dynasty colourand ink on silk painting from the10th century shows three mountedhunters carrying unmistakableSalukis on their horses (Watersand Waters 1984, 92, 40). Muchlater a Jesuit painter resident inBeijing in the mid-18th centurypainted several Salukis for apresentation album to theQianlong Emperor, though it issuggested that foreign dignitariesmay have given them as tribute

Fig. 4. Tang dynasty figurine. Photo-graph copyright © 2007 Victoria andAlbert Museum. Reproduced with per-mission.

Fig. 3. Petroglyph from Kazakhstan. Photograph copy-right © 2006 Renato Sala of the International ScientificProjects of the Laboratory of Geo-archaeology, Almaty.Reproduced with permission.

61

(Rawski and Rawson 2005, 188-191, 410-411).

Against this academic evidenceI set off to explore for facts on theground today. The start in thewalled oasis town of Khiva inwestern Uzbekistan appearedauspicious as I spotted a Berkutor Golden Eagle sitting on a stand.The Berkut is traditionally usedfor hunting right across CentralAsia and is often used intandem with Tazys [Fig. 5]. Myhopes were soon dashedhowever as it turned out to bea mere tourist attraction. Wecrossed the Oxus or Amu Daryaas it is now called and droveparallel with it along thesouthern side of the Kyzyl Kumin ideal hunting country butwithout success, apart fromsome information about anoasis town far from our roadthat was described as thecentre for hunting with Tazys.However I noticed on the mapan area outside Bukhara that wasdesignated as a gazelle naturereserve. I reasoned that if gazellewere indigenous there huntinghounds might also be found there.We stopped at the entrance to thereserve, and I went in with ourguide to speak to the Director incharge. Bounding to meet uscame the familiar form of what Isupposed to be a smooth fawnSaluki. On closer inspection itproved to have turned back ears,which are more commonlyassociated with Greyhounds,though it definitely was not a

Greyhound; yet itwas not a Salukieither. All theirkeeper could tellus was that thishound and all theothers we sawthere had comeoriginally fromRussia and thathe used them forseeking out fallengazelle in thereserve. During avisit to Russia in2004 I saw many

similar hounds, which were calledHortaya (khortaia) [Fig. 6]. SomeRussians believe that this breedmay have descended from thosesmooth-haired hunting houndsshown in the early Scythianrepresentations mentionedabove.

After taking in the manydelights of Bukhara, Shakhrisabz

and Samarkand,we drove on viaTashkent andBishkek deep intoKyrgyzstan. Ourdestination was ayurt in the TienShan Mountainsnear the formerSilk Road cara-vanserai at TashRabat. All alongour route I hadbeen told aboutthe decline inhunting with

hounds, particularly since thedeparture of many of the ethnicRussian population, and I hadgiven up all thought of seeing anymore hunting hounds. It wastherefore a pleasant surprise to begreeted on arrival in Tash Rabatby a hound with a dense black coatthat the local Kyrgyz called aTaigan [Fig. 7]. This breed isparticular to the high mountainswhere it has developed the abilityto hunt all manner of animals inthe rare atmosphere above 2,500m even in the depths of winter. Itappears to be related to the Salukibut with a broader head, a stockierbuild and a dense coat.

It proved to be the first of anumber of such hounds that wewere to see. We took a walk upinto the mountains hoping to seeinto China and on the way cameacross a yurt that was protectedby a red and white Taigan. Awoman emerged from the yurt

who turned out to be someoneto whom we had given a liftearlier. She welcomed us intoher yurt and showed us a tinyblack Taigan puppy. Herhusband appeared and saidthat he hunted marmot andmountain goat for the pot, andthe hounds were very affectiveeven when the temperature fellto –30 degrees C. As weexplored further in the area wepassed near another yurt fromwhich a man hailed us, offeringus hospitality. We declined butasked if there were any Taigansnearby. He indicated a valley

Fig. 5. Mounted Berkutchis and Tazys in Kyrgyzstan. Pho-tograph copyright © 2007 Terence Clark, from originalon display in Cholpon Ata Museum, Kyrgyzstan.

Fig. 6. Russian Hortaya in Uzbekistan. Pho-tograph copyright © 2007 Terence Clark.

Fig. 7. Taigan at Tash Rabat. Photograph copyright ©2007 Terence Clark.

62

where we would find some. Wesplashed through an icy mountainstream up the valley and arounda bend came on another yurt anda mud-brick house underconstruction. As we approachedthree Taigans rushed out to greetus. They were very friendly andlooked in good condition: one of

them was heavily pregnant andwas due to give birth within a fewdays. All, we were told, wereexcellent hunters. A little furtheron I saw another Taigan on theother side of a stream by a smallhouse. As soon as we stopped toget a better look, the door of thehouse flew open and some younglads rushed out, jumped ontohorses and raced across thestream towards us, followed by avery lively Taigan puppy. It wasonly with difficulty that wemanaged to extricate ourselvesfrom their pressing invitations totheir house. We did relent furtheron where a woman appeared froma small house with a plate ofdifferent dairy products: kaimak(a sweet thick cream), curds andcheese with delicious fresh bread.

As we descended from themountains the next day we passedmany yurts and great herds ofhorses, sheep and cattle — andeven a few yaks. We stopped atone where a woman was makinglittle round cheeses (qurut) thatshe was setting out to dry andharden in the sun. She showed us

round her beautifully decoratedyurt, from the roof of which hunga fur from a fox caught by herTaigan. Further on we came acrossa young lad walking by theroadside with a beautiful black andtan hound that had something ofthe old-fashioned Bell-Murray typeof Afghan Hound about it [Fig. 8].

Nearby I spotted outof the corner of myeye a familiar shapegliding along thebase of a farmhousewall. We made alittle detour andwere warmlywelcomed by thefarmer’s wife andher small bright-eyed daughter, whoproudly showed usthe black and whiteTaigan bitch, ratherstrangely calledTarzan, and her fivetiny puppies. Thebitch gave a warninggrowl when I

stepped too near to take a pictureof the puppies and the next thingI knew was that she had bitten mein the ankle, to the mortificationof her owner! No serious harm wasdone and I was assured that thebitch had been vaccinated againstrabies, but it was a warning notto mess with a Taigan withpuppies.

It was clear from all theseencounters that the Taigan, likethe Saluki in the Middle East, isheld in a position of high esteemboth as a pot-filler and as acompanion. However, as in otherparts of the region, the pressuresof modern life on the habitat ofthe hunting hound’s prey isleading to a decline in the numbersof the Taigan and, according tolocal sources, the carelessness ofsome of the hunters has resultedin some crossing with otherbreeds. Nevertheless in my shortexploration I had seen enough ofthe breed to form the impressionthat it is sti l l hanging onsuccessfully in this area ofKyrgyzstan.

By the time of my next tripalong the Silk Road a year later Ihad established via the Internetcontact with people in Kyrgyzstanand Kazakhstan so that my searchfor Central Asian hunting houndswas less haphazard. This time myearlier impression of the situationof these hounds was reinforcedboth factually and visually. Ourguide, Almaz Kurmankulov, was afounder of the Kyrgyz TaiganSociety, which had recorded about200 purebred Taigans and wasencouraging breeders to preservethe breed as part of the nation’sheritage. Starting in Bishkek hetook us first to meet a well-knownhorse and Taigan breeder on hisfarm. Here we saw a number ofhis breeding stock, all of themblack and white, with thedistinctive ring at the end of thetail where the last two vertebraeare fused together. One bitch wassurrounded by a litter of livelypuppies, two of which were dueto be presented later to King JuanCarlos of Spain. We had intendedto see one of the kennels on LakeIssyk-kul, but all the hounds wereaway in the mountains where theyspend the summer months withthe flocks and herds. As we wereto see later this is the bountifultime for Taigans when marmotsare abundant, which they catch tofeed their puppies. Instead wemade a diversion high into themountains to visit ZarnaiSagenbai, one of the fewremaining Berkutchis inKyrgyzstan. We found him and hiswife at their yurt, outside of whichsat a screaming Berkut. Zarnai puton a demonstration of his skill,flying the Berkut from his horseto a lure in the valley below [Fig.9, next page]. He and his sons stillflew their eagles to hunt mainlyhare for food and fox for fur butalso for wolf to protect theirlivestock. In the past they usedto fly them in tandem with Tazys,as an old photograph in theCholpan-Ata Museum illustrated[Fig. 5, facing page].

As on our last trip we made forTash Rabat whence we planned todrive over the Torugart Pass into

Fig. 8. Afghan Hound? Photograph copyright © 2007Terence Clark.

63

the Xinjiang-Uighur AutonomousRegion of China. As we arrived atour yurts we were greeted withthe news that there was a Kyrgyzcommunity some 30 km away withTaigans that Almaz was keen toregister. So he and I and a localRussian driver set off to find them.Before long we found ourselves ata military checkpoint before theborder with China. The prospectsof passing it did not look too goodas neither of my companions hadpassports with them. However afriendly officer pointed to a whitehouse a few hundred metres awaywhere he said we would find manyTaigans and allowed us to pass.Sure enough we found severalmature hounds there, some withyoung puppies that kept poppingout of deep holes in the groundwhere they lived, protected fromthe biting wind and predators.Almaz duly measured them andrecorded their details. A young ladcame up and asked if we wouldlike to see some more. So underhis guidance we set off in ourancient car across country wherethere was not so much as a dirttrack. After some while westopped at the top of a bluff fromwhere we could see below a yurtbelching smoke from its chimney:we had arrived. Scrambling downwe were met by several Taigansof different ages and sizes and inno time Almaz was submerged ina heap of playful puppies, grownfat on regurgitated marmot meatfrom their parents. The lady of the

yurt dragged out from herunderground den a veryreluctant bitch to show usher recently born puppies.There could be no doubtthat in these remote partsTaigans still formed anessential part of the Kyrgyznomads’ way of life.

Fascinating though thejourney along the SilkRoad through the UighurRegion proved to be, interms of hunting hounds itwas totally unrewarding.According to borderofficials and our guide, wewere not allowed to depart

from the main road to visit Kyrgyzvillages in which I had expressedan interest; so I could not makeenquiries there. Among the Uighurand Han Chinese I met only blanklooks when I showed themphotographs of the hounds. Yet acontemporary mural in the KhotanMuseum illustrating medievaltravellers on the Silk Road showedin the central foreground anunmistakable Tazy.

How different was the receptionat our final destination inKazakhstan! In Almaty KonstantinPlakhov, a biologist at the Instituteof Zoology and a champion ofKazakhstan’s native breeds, cameto the hotel and tookme to call on AskarRaibaev, Presidentof the Dog BreedingFederation ofKazakhstan. As Ientered his house Ialmost fell overRoshan, a beautifulTazy that occupiedmuch of the centreof the room with hereight puppies. Askardescribed the con-siderable effortsbeing made to preserve not onlythe Tazy but also the Tobet [Fig.10], a huge shepherd guard dog,of which I was shortly to see somespecimens at my next stop — theSunkar Breeding Centre in abeautiful setting outside the town.The Centre contains a large

number of raptors of differentkinds, some of which are there forbreeding and some forrehabilitation, after being foundinjured, and release into the wild.Running loose were severalmature Tazys and puppies, whilein well-designed kennels someTobets padded massively up anddown. Both the Tazy and the Tobethave suffered from a decline innumbers and in quality, and theCentre is endeavouring topreserve the breeds and toencourage their wider distribution.It is hard to judge its success.Certainly I heard of some Kazakhswho maintained quite largekennels of Tazys for hunting ontheir farms and on the steppe, andI met some Tazy breeders inAlmaty. However, the problem isalso one of changing lifestyles:there is no longer the same needto hunt for food or for fur, andyounger people are moreinterested in computer gamesthan hunting in the often harshconditions of the steppe.

Throughout my tours I wascollecting here and there from thevarious hunting hounds bothmtDNA and DNA samples to sendto Dr Peter Savolainen in Sweden,who is undertaking research intothe origins of dogs. The research,

which has of course importantimplications for the history ofmankind in this region, continues,and in due course it may lead toan answer to my initial questionabout the western or central Asianorigins of these hunting houndsalong the Silk Road.

Fig. 9. Kyrgyz Berkutchi. Photograph copy-right© 2007 Terence Clark.

Fig. 10. Kazakh Tobet. Photograph copyright © 2007Terence Clark.

64

About the Author

The author studied Russian atCambridge University while in theRoyal Air Force but on joining theBritish Foreign Service was sentto study Arabic at the School ofOriental and African Studies atLondon University and at theMiddle East Centre for ArabStudies in Lebanon. He spentmuch of his career in the MiddleEast, latterly as Ambassador toIraq and to Oman. In retirementhe retains close links with theMiddle East through a numberof academic bodies and haswritten extensively in books andjournals on the history and politicsof the region and on the Saluki andhunting. E-mail <Sirterenceclark@aol.com>.

References

Brewer 2001

Douglas J. Brewer. “The Dogs ofAncient Egypt.” In: Douglas J.Brewer, Terence Clark and A. A,Phillips. Dogs in Antiquity. Anubisto Cerberus: The Origins of theDomestic Dog. Warminster, UK:Aris & Phillips, 2001.

Clark 1995

Terence Clark. “Saluqis in Iraq.”Ch. 5 in: Gail Goodman, ed. TheSaluqi: Coursing Hound of the East

/ Al-Saluqi: kalb al-sayd al-sharqi.Apache Junction, Az.: Midbar, Inc.,1995: 131-134.

Plakhov and Shelestovaforthcoming

Konstantin N. Plakhov and A. S.Shelestova. Borzye Tazy i okhotas nimi. Almaty: forthcoming.

Przezdziecki 2001

Xavier Przezdziecki. Our Levriers.Nice: Les Amis de XavierPrzezdziecki, 2001. (Tr. of LeDestin des Levriers. Paris, 1984.)

Rawski and Rawson 2005

Evelyn S. Rawski and JessicaRawson. China: The ThreeEmperors, 1662-1795. London:Royal Academy of Arts, 2005.

Sabaneev 1993Leonid P. Sabaneev. Sobakiokhotnich’i: Borzye i gonchie. Ed.E. A. Kalganov. Moscow: Terra,1993 (first published in 1895-1898in journal Priroda i okhota).

Savolainen et al. 2002Peter Savolainen, Ya-Ping Zhang,Jing Luo, Joakim Lundeberg andThomas Leitner. “Genetic Evidencefor an East Asian Origin ofDomestic Dogs.” Science, 298, 22November 2002: 1610-1613.

Schafer 1985

Edward H. Schafer. The GoldenPeaches of Samarkand: A Study

of T’ang Exotics. Berkeley:University of California Press, 1985(1st ed., 1963).

Sinor 1990

Denis Sinor, ed. The CambridgeHistory of Early Inner Asia.Cambridge: Cambridge UniversityPress, 1990, p.175.

Smith 1980

G. Rex Smith. “The ArabianHound, the Saluqi — FurtherConsideration of the Word andOther Observations on the Breed.”Bulletin of the School of Orientaland African Studies (University ofLondon), 43/3 (1980): 459-465.

Valtz Fino 2005

Elisabetta Valtz Fino. “In the Wakeof Alexander.” Ch. VIII in: MilbryPolk and Angela M. H. Schuster,eds. The Looting of the IraqMuseum, Baghdad: The LostLegacy of Ancient Mesopotamia.New York: Abrams, 2005.

Waters and Waters 1984

Hope Waters and David W.Waters. The Saluki in History, Artand Sport. 2nd ed. Wheat Ridge,Co.: Hoflin, 1984.

Whitfield 1999

Susan Whitfield. Life along the SilkRoad. London: John Murray, 1999.

65

A Thousand Years on the Silk Road:Epic Poetry and Music from theKyrgyz RepublicRysbai Isakov, Epic SingerAkylbek Kasabolotov, MusicianDr. Helen Faller, Anthropologist

In February-March 2006, withgenerous support from theSilkroad Foundation, anthro-pologist Helen Faller toured theUnited States with two youngartists from Kyrgyzstan in a seriesof university residencies called theKyrgyz Cultural PerformancesProject. The artists were RysbaiIsakov, a laureate epic singer, who

performed episodes from theKyrgyz national epic Manas, thelongest in the world at over half amill ion lines, and AkylbekKasabolotov, a member ofKyrgyzstan’s Tengir TooEnsemble, who shared hiscountry’s unique nomadic musicaltraditions. The three of them werein residence at eight universities

in six states and performed forover 3000 people. The purpose ofthe tour was to provideopportunities for Americans tolearn about Central Asianperformance culture from twotalented cultural ambassadors andto provide the artists withexperiences that would help themto develop their art in new ways.Video footage of the KyrgyzCultural Performances concert atthe University of Texas at Austinis on view at www.silkroadfoundation.org andrealaudio.cc.utexas.edu:8080/ramgen/cola/centers/creees/i m a g e s / m e d i a /kyrgyz_022206E.rm. Informationabout the project can be found on

the Central Asian CulturalExchange website — www.cace.us.

In addition to being a manaschy— reciter of the Manas epic —Rysbai is the Executive Director ofthe Manas Miras Foundation, themission of which is to preserve theManas. Rysbai explained toAmerican audiences that hesuffered from terrible migraineheadaches as a child and that onlyafter he received a vision andbegan to recite the Manas did theydissipate. In fact, if he doesn’trecite regularly, he becomes ill.Akylbek is in his first year ofgraduate school at Bishkek’sConservatory, where he studiesunder composer and director ofthe Tengir Too Ensemble,Nurlanbek Nishanov. UnlikeRysbai, he considers himself anartist who needs to practice inorder to perform. In addition tostudying and playing profession-ally, Akylbek teaches music at theuniversity level.

This interview, here somewhatshortened for publication, tookplace, mostly in Russian, over teain Helen Faller’s Philadelphia homein early mid-March 2006.

Helen: I wanted to start by asking howlong you have each been pursuing yourart, although (she nods to Rysbai) youdon’t consider it art, right? If you couldtalk a little about the place of theManas in your life, in your childhood,for example, that would be interesting.

Rysbai: I had a grandmother whotold me all those folk tales. Epics,legends, and all that. And ofcourse she told me about theManas. I don’t know at what age,but in my childhood I was familiarwith the Manas and when I alreadyhad learned how to read, mygrandmother gave me a book. Andshe wanted me to read from it notas it is read silently, but as it isrecited. And when I read it, shebroke into tears.

H: How old were you at the time?

R: I was six, no, not six. Older.And that’s how I grew up. I grewup in the spirit of the Manas. AndI began to recite at the age of 12.

How I started to recite I’ve toldyou, right? Before [I began torecite] I was very ill. I was very,very ill. It’s impossible to describeit in words — it wasn’t a physicaldisease. And then I had a vision.After that I began to get betterslowly. At the beginning I didn’teven understand what it was, itwas incomprehensible. But,afterwards, after I started to recitea little, I noticed that my headwasn’t aching like it used to.

H: You were already older than 12 atthat time?

R: That was when I was around15 or 16. Only then did I start tonotice, but up till then I didn’tquite…Only at the age of 12 did Istart to understand…notunderstand but rather to feel it…

H: Did people start inviting you torecite the Manas at that age?

R: No. I recited at school onspecial occasions. I was very shyabout reciting the Manas. Myupbringing made me shy. But,there were occasions.

H: They called upon you.

R: Yes, people who wanted tohear. And when they asked, Ibegan…I closed my eyes and Idon’t know how I performed. Theepic began.…I probably started torecite the Manas properly from theage of 20. That’s when I startedtruly to understand the Manas.Until then, I was somehow blind.

H: And how did that happen? I’mcurious.

R: I became accustomedsomehow and began no longer tosuffer from shyness when Irecited. Step by step…

H: But by the age of 20 you hadalready lived in Bishkek for four yearsafter graduating from high school.Right?

R: I became a student at theuniversity. In Bishkek Iparticipated in competitions andfestivals. We all competed amongourselves. From every departmentin the university.

H: That was 10 years ago. Didn’tstudying interfere?

R: No. … I took ten years off fromschool and then went backafterwards.

H: When did you start working for theManas Miras Foundation? At first youworked in television, right?

R: After I got my universitydegree, I worked in televisionprogramming. But before that weestablished a society amongourselves of creative youngpeople. I worked there as theexecutive director. We developedoral folk projects in the form ofcontemporary performance. Wesubmitted a proposal to thetelevision company and startedworking there. We collaboratedthere with writers from the Houseof Poets. At that time I receivedan invitation from Baitkochernezov(Sponsor of the Manas MirasFoundation) which I accepted, andwe started the Foundation.

H (turning to Akylbek): I would like tolearn more about how you came towork in music.

Akylbek: Well, my father’syounger brother is a musician. Heplays the komuz (Kyrgyz lute).Even now he works at the KarakulMusic School. And he was the onewho showed me the musical pathin the road [of life]. He had mestart as a student at the musicschool in 1996. And there I learnedhow to play the balalaika, theRussian ‘lute.’

H: And for the komuz there are nowritten notes. Is that how they taughtthe instrument?

A: No.

H: They taught students to play byusing musical notation.

A (nods): And around then theyopened a department of mountainregion instruments. At the time,there was no such thing.

H: But I thought your father’s brotherwas a komuz player?

A: Yes he is. But he was the onlykomuz player, and he took meunder his wing. So that’s how Iended up learning how to play thebalalaika. But, I also have agrandmother who plays the

66

komuz. After I finished musicschool, I wanted to study at theConservatory in the capital city ofKyrgyzstan, Bishkek.

H: And that was five years ago?

A: Yes. In 1996 I entered theKarakul Music School and Igraduated in 2000. And in thesame year I immediately began atthe Conservatory. At that time theConservatory offered onlytraditional singing and the komuz.But in 2000 they opened adepartment of national music thatincluded wind instruments and thekylkiyak (Kyrgyz cello).

H: It was Nurlanbek. [NurlanbekNishanov is a Kyrgyz composer anddirector of Tengir Too, who has workedwith the Aga Khan Trust for Culture forsome five years. Akylbek is hisprotégé.]

A: Both Nurlanbek and Chytyrbaevopened the new department. Soafter entering the Conservatory, Igave it a lot of thought anddecided that I wanted to play myown national instruments. And Iimmediately started playing windinstruments. I gave up thebalalaika and began from nothing.Nurlanbek started me out fromnothing. And when I graduatedfrom the Conservatory, heimmediately gave me work inTengir Too.

H: But, before joining Tengir Too, youalso played in a group, right?

A: Yes, yes. I played when I wasin music school in an ensemblecalled Aidigil. I played a littlesybyzgy (Kyrgyz flute), but onlythe sybyzgy. But, in Karakul thereweren’t really any musicians ofprofessional stature. And thenNurlan aga [an honorific meaningolder brother or uncle] invited meto work with Tengir Too. I workedwith Tengir Too for two or threemonths, but then they closed dueto lack of finances. And then in2003 we started to work again.

H: That was because of the Aga KhanFoundation? You started getting moneyfrom them or…

A: Yes. And in addition to that wegot support from the Kyrgyz State

Technical University. So we startedto work there, and we’re therenow.

H: I have a question. You both had tomove to Bishkek in order to seriouslypursue your callings?

A: Yes.

R: I didn’t know that I was goingto become a manaschy. I justdecided to go to university.

A: I wanted to continue studyingmusic.

H: Hm. I ask because it’s not anabsolute contradiction but it seemssomewhat strange that you as ruralpeople who have very deep folk rootsin your blood, in order to develop thoseroots to reach a higher level had tomove to Bishkek, which is a much lessKyrgyz place. In the sense thatRussians live there, Russian speakers.In Bishkek, things aren’t really so muchculturally Kyrgyz as they are post-Soviet, right?

A: If I had stayed in Karakul, Iwouldn’t have learned to playnational wind instruments. Therearen’t any professionals likeNurlanbek there. And I want to saythat Nurlanbek aga is the onlyKyrgyz folk musician who works ata professional level. Now he hasbegun to revive all our musicaltraditions and has started to makeinstruments. And he composesworks to be played on folkinstruments. For that reason I hadto leave home to study. Becausethere weren’t any professionals,any teachers.

R: For me, as I said, the situationwas different. I never dreamed ofbecoming a manaschy. I simplythought about getting intouniversity. Just like everyone else.I got in and suddenly that waswhat happened. I’ve already saidthat I started truly to understandthe Manas when I was studying inBishkek.

H: It’s interesting to me that in orderto continue to do your work, you haveto be in the city.

R: Well, we aren’t always there.We very often leave the city forthe villages to tour and perform.

H: I would like to know from you whyyou decided to participate in thisproject, to take part in a residenceacross what seems like almost all ofAmerica?

R: For me it was an honor. Andnot everyone is given such anopportunity, right? To travel toAmerica.

H: Akylbek, why did you decide toparticipate in this project?

A: We have a saying in Kyrgyz thatstates, if they invite you, it isnecessary to go. And the secondreason is that I wanted to showmy culture to Americans. It’s asRysbai aga said, for us it is anhonor. Athletes and culturalactivists have a special role. It isour obligation…

R: To bring to light the splendorof the people, the people’s culture.

H: And what did you want to get outof this experience?

R: We wanted to perform as weshould, so that our peoplewouldn’t be ashamed of us afterour performance. We hoped forthat.

H: Akylbek, did you want somethingdifferent or the same thing?

A: The same thing. For the publicto understand and listen to us. Weare very pleased. It has beengood. Our dream has beenrealized.

H: How do you feel you’re beingreceived by American audiences. Howdo you feel when you perform?

A: I feel comfortable, somehow.

R: Yes, and it’s not only our musicthat they understand. If a personhas music in his or her soul, thenit is possible to understand withoutlanguage. But, the astonishingthing is that they understand theManas as well. You saw thatyourself in Texas. The audiencetold us what was happening in theepic. They asked the questionabout the woman speaking [thestudents heard Rysbai vocalizingKanykei’s distress and wanted toknow more]. That shows that theyunderstood the words. And onewoman said that she began to cry

67

when I was reciting. Moments likethat make it clear to us thatAmerican audiences comprehendus with their souls.

H: Did you have any other additionalgoals for your trip to America?

R: We didn’t come here only toacquaint America with our culture.At the same time we wanted tolearn, to see how Americans livewith our own eyes. What life isreally like for Americans. To hearabout it is one thing, but to see itwith your own eyes is another. Wedidn’t even know that Americanswould be so hospitable…but ofcourse their traditions arecompletely different. And that wasa surprise for us. I personallythought that Americans wouldn’tinvite us to their houses as guests.

H: And what else surprised you?

R: Wyoming was surprising. Theheavenly nature in Wyoming, thebeautiful mountains, and of coursethe antelope. How the sheep roamfree.

H (addressing Akylbek): And didanything surprise you other than that?

A: Well, seeing the cities and thecivilization itself, yes. And the cars,by comparison with ours… Thereare differences… The people hereare good and very open.

H: You didn’t expect that?

A: No. I didn’t expect that. Nomatter where you go, people areready to answer your questions.And the cities, the cities are verybeautiful.

H: What haven’t you liked in America?I have asked you several times. Itprobably seems like a strange questionto you, since you haven’t answered.

R: Aha, what haven’t we liked?That of course is a matter for eachperson. That’s personal… Well, Inotice that very few people are ahealthy weight.

H: Yes, Americans eat too much.

R: I think it’s not just a matter ofovereating. I eat a great deal tooand I don’t put on weight.

H: Yes, but you walk 11 km every day[to and from work in Bishkek].

R: There you go. If you’re goingto eat a lot, you need to be active.But, that’s my personal opinion.There’s a lot more, of course. Ifyou compare Texas with Wyomingand New York, it’s magnificent. Wesaw Texas first, and thenWyoming. In those placeseverything was clean. Like glass,right? But when we arrived in NewYork, it was unexpected. New Yorkwas dirty. It turns out that a bigcity means big trash. (Laughs)

H: Akylbek, you didn’t answer thequestion about what you foundsurprising in America.

A: Well, what has been surprisingis to see such big cities, howthey’re built, and also I’msurprised by how much work isinvolved…and how well thebuildings are looked after. Insideeverything is in order. So, all thathuman labor, that surprises me.How everything is kept up. Andanother thing — the relationshipbetween man and nature is veryclose. In my opinion, that’s verygood.

H: What do you mean exactly?

A: Yesterday we learned [whileguests at the house of somePhiladelphia Kyrgyz] that sheeparen’t slaughtered the same waywe do it in Kyrgyzstan. Here theyuse a special razor. The attitudeof people to nature here isrespectful, not only towardsanimals but towards all nature.Also, it was good to meet localmusicians, who played differentkinds of music. I played with them.They taught me music and Itaught them music. It was veryinteresting.

R: I would like to collaboratecreatively with Americans again.

H: Although you haven’t yet seen allof America, you’ve spent some timehere and I wonder if your opinion of ithas changed.

R: Yes. Yes, now we’ve seen it withour own eyes, right? Of course ourknowledge isn’t complete.Everywhere we go, we are treatedas guests. We live in hotels. Inshort, everything is done for us —

every detail is looked after. Withthis experience we can’t say thatwe really know America, but wedo have some understanding ofthe place now. And ourimpressions are completelydifferent from what we have beentold about America.

H: Interesting.

R: In Kyrgyz we say: it’s better tosee something once than to hearabout it a thousand times. Iwanted to make one more point.Here it seems easier for people torealize their dreams. I would liketo extablish a creative connection,a scholarly and creative con-nection, with Americans with thegoal of preserving the Manas epic.To create that connection andrealize it… In the Manas there aresacred things — not for anyparticular nations or tribes. Butthings that should be sacred forall people. Like the Bible, like theKoran! Like Buddhism andKrishnaism and so on. The Manasis also a part of those worlds. Oneshouldn’t divide them. And it’s notright to love only one and deny theothers, the way that religiousfanatics do. For example, onefanatic loves only the Bible anddenies all other religions. Anotherloves only Islam and deniesanother religion. If a person trulybelieves in the sacredness of hisreligion, then he ought to valueother sacred things.

H: Certainly. Akylbek?

A: Well, for me this experience hasgiven the inspiration when I goback to Kyrgyzstan to practice andplay music even more. To makeeven more instruments. To learnhow to play new instruments. Andeventually to compose a piece ofmusic. That’s what I feel inspiredto do.

H: And the piece of music will be forKyrgyz folk instruments or for…

A: For folk instruments.

H: Good. Thank you for taking the timeto do this interview. As bolsyn.

R and A: As bolsyn.

68