Title The Identity of the Turkish Rulers to the South of Hindukushfrom the 7th to the 9th Centuries A.D.

Author(s) INABA, Minoru

Citation ZINBUN (2006), 38: 1-19

Issue Date 2006-03

URL https://doi.org/10.14989/48824

Right © Copyright March 2006, Institute for Research in HumanitiesKyoto University.

Type Departmental Bulletin Paper

Textversion publisher

Kyoto University

ZINBUN2005No.38

TheIdentityoftheTurkishRulerstotheSouthofHindukushfromthe7thtothe9thCenturiesA.D.*

MinoruINABA

1. TheTurkishRulersofEasternAfghanistan

Atthebeginningofthe1990 う民 many Bactriandocumentswerediscovered

inthenorthernareaofAfghanistanwhichwasstillinthemiddleofacivil

war.Thereisnodoubtthatthosedocumentswill ラtogether withtheBactrian

inscriptionfromRabataknear8urkhKotalandBuddhistdocumentsfromthe

cavesnorthofBamiyan , helpustoimproveourunderstandingofthehistory う

religionandthesocietyofancientAfghanistan.AnumberofthoseBactrian

documentshavebeenintroducedbyN.8ims-Williamswithtransliterations

andtranslations うwhich willsurelystimulatefurtherresearch(β倒S臼imsト-Will iam

2000)Justbeforethatdiscovery ぅbeing inspiredby8h.Kuwayamaうs workonthe

historyofK 瓦pi 釘and theGandh 瓦ra areapriortotheIslamicperiod(Kuwayama

1990) う I publishedapaperonthehistoryofeasternAfghanistan , speciallyfoｭ

cusi 時on Ghazni/Z 瓦bulistan fromthe7thtothe8thcenturyA.D・ ぅscrutiniz­

ingmaterialsavailableatthattilne(Inaba1991).Aframeworkofthehistory

ofthatareaduringtheperiodinquestionwasdeducedasfollows:

Inthelatterhalfofthe6thcentury , theHephtaliteempireinTokh 瓦rist 瓦n

collapsedduetoTurkishattackうi.e. WesternTuJue. Gandh 瓦ra ， whichhad

alsobeenunderthecontroloftheHephtalites うand K瓦pi 釘which wasadjacenttoGandh 瓦ra ， weregreatlyaffectedbythatevent. Thedynastylabeledthe

Khi 時alides byKuwayama(orNezakidesbysomescholarsbωed ontheleg-

*IwouldliketoexpressmygratitudetoProf. Y.YoshidaandDr. A.Palumbo , whoprovidedmewithinvaluableinstructionandsuggestions ぅand ぅas always , toDr. 8h.

Kuwayamawhohasneverfailedtokindlyandwithgreatgenerousitydiscusswithme

atgreatlength. AndIamalsogratefultoMr. AnthonyR.Blackforimprovingmy

English.

1

MINORDINABA

endofthecoins)hadbeenestablishedinK瓦pisl andwasexertingitsruleover

theK瓦bul valleyasfarasGandhara.Wheninthefirsthalfofthe7thcentury

Xuanzangvisitedthere うthe kingofK瓦pisl wasofthislineage.Inthe650 うh

ArabMuslimsreachedasfarasSlstanandintendedtomaketheirwayfurther

north うwhich resultedinconsiderablestrategicimportancebeingplacedonthe

southernborderoftheK瓦pisl kingdom うnamely K瓦bu l. Atthattime , K瓦bul

wasundertheruleoftheTurkswhohadbeenreportedbyXuanzangtohave

livedinthemountainareabetweenZ瓦bulist 瓦nand K瓦pi 釘うthat is ヲFulishisa­

tangna 弗栗侍薩慎那 . Graduallyhavingextendedtheirpower ぅthey cameto

beindependentand うeventually うdepri vedtheKhingalideKingofalmostthe

wholeofhisterritory.ItwastheywhowerereferredtoastheTurkSahinthe

Islamicsources.

Inthe680 うs ， aprincefromK瓦bul fledsouthwarddowntoZ瓦bulistan ， probｭ

ablyduetotheconflictsurroundingsuccessiontothethrone うand established

hisindependencethere(Cf.Kuwayama1999).Thenewr叫ers ofZabulist 瓦n

weredenotedintherecordsofArabandPersianhistoriansbythename ふん

rtbyl.Ithasbeenmaintainedthatthistitleshouldberead ムヂo znbylwhichmighthavesomerelationwiththecultsurroundingtheGodZhun(如 nii of

Xua 四ang). ConsideringtheTurkishsituationmentionedabove , howeverぅthistitleshouldbeunderstoodasacorruptedformoftheTurkishtitleiltiibiir.1

Thusうthe twokingdomsofKabulandZ瓦bulist 瓦n wereofthesameTurkish

originandbothactedasanimpregnablerampartagainsttheMuslimsfor

almosttwocenturiesfollowing.

Nevertheless うthere remainseveralproblemsunsolvedandunmentioned う

oneofwhichistheoriginofthoseTurks.TheBactriandocumentsmentioned

aboveandsomearchaeologicalfindingsoftheareainquestionfromthelastcenturyseemtoindicatetheTurkishtribenamedtheKhalajasapossible

candidatefortheoriginofthoseTurks.2 Theprimarypurposeofthispaperis

toverifythisidentification.

Asamatteroffact , theideaofidentifyingtheTurksinquestionwith

theKhalajtribeisnotveryrecent.L.Petech(1964)alreadysuggestedthis

identificationandreferredtotheexpressions“He-da-Iuo-zhi 詞達羅支" and“Ge­da-Iuo-zhi 葛達羅支Jうwhich appearedinChinesesources.A.Rehman(1988)

alsoconnectedtheTurksofK瓦bul andZabulist 瓦n withtheKhalajutilizingthe

materialscitedbyJ.Marquart(1

1 Cf.Bombaci1970:59;Rehman1988:180;Inaba1991:53-55;Sims-Williams2002:235.

2 Cf.Lee&Sims-Williams2003;Verardi&Paparatti2004;Inaba2004.

2

THEIDENTITYOFTHETURKISHRULERS

scholarsmaybe , theyseemnottohavesucceededinexaminingitsu 白ciently う

asPetechneverproducedacomprehensivecomparisonofinformationfromthe

ChinesesourcesandthatoftheArabicandPersiansources うnor didRehman

bringmaterialintheChinesesourcestolight. However ぅY. Yoshida(2000

[2003])recentlyhassuggestedtheassumptionthattheChineseHedaluozhiandGedaluozhicanbeconnectedwiththeword“kharalaじがう which appears

onthelegendsoftheNezakS瓦h coins うconsidering thelattertobesomekind

ofIndianizedformoftheword “Khalaj ." Iwilltakethissuggestionasmystartingpoint.

2. Hedaluozhi

Tobeginwith うlet usconsidertheChineseexpressionsHedaluozhiand

Gedaluozhi.However , sinceeachhasvariants うit isfirstnecessarytoexaminetheexpressionsthemselvestodecidewhichshouldbetakenasthecorrectform

orwhethertherearemultiplecorrectforms.

Tostartwith うlet usproceedwithapassagefromX·初 旬ngshu Vol.221b:

“Xieyu 謝周目 li es southwestofTokh 瓦ristan. ItwasformerlycalledCaojuzha 漕矩陀

orCaoju 漕矩but renamedHedaluozhiintheXianqingperiod.EmpressWuagain

changedthenametothepresentone.About400litoitseastliesJibin 厨賓and

toitsnortheastliesBamiyanatalmostthesamedistance.Indialiestothesouth う

PersialiestothewestandGuzganliesnorthwards.ThekingresidesinHexina

(鶴悉那 *yak-sit-na) whichis7ぅ000 liwide.Thekingalsosometimesresidesinthe

Aponi(阿婆体九 -bua-nia ) castle.Thelandproducesplentyofsaffronandwater

springsir 向ate thefields.PeoplefromTuJue , JibinandTokh 瓦rist 加 (吐火羅)

livethere.Jibinisrecruitingyoungsoldiersfromthiscountrytoprotectagainst

theArabs.Theysentanenvoyandpresentsinthe1styearoftheJingyunera

(710A.D.)andlatersubjugatedthemselvestoJibin.Inthe8thyearofKaiyuan(720) うthe EmperormandatedGedalωzhi xielifaShiquerandmadehimtheking

TheyhavesentseveralenvoysuntiltheTianbaoera. 門

HerethreenamesforZabulistanintheTangperiodarementioned , i.e. う

Caojuzha , Hedaluozhi(Gedaluozhi)andXieyu. TheChinesenamesapplied

toZ瓦bulist 瓦n havebeenanalyzedbyYoshida(Kuwayama1992:135-139) ぅac­

cordingtowhomCaojuzhaandXieyu , aswellasCaoli(Xugαoseng zhu αηう

Vol.4) ぅ Xieyue (Huilinyinyi う Vol. 100)andZhewutl 刷tana (Xuany 仇yinyi う

Vol.18)aretranscriptionsfor*dz 瓦ul くZ 瓦b 叫-istan) うwhile Hedaluozhiand

Gedaluozhi(andGelωdazhi intheJiutαngshu ; seebelow)musthavebeen

3

MINORDINABA

transcriptionsforawordcompletelydi 百erent from*dz 瓦ul くZ 瓦bu l. Asforthe

lattertranscriptions ,E.Chavannes(1969:160 うn. 4)presumesthatHedaluozhicouldbeanerroneoustranscriptionforDaluohezhi 達羅詞支which mightcorｭ

respondtoal-Rukhudh くal- RukhkhajappearinginIslamicsources ,especiallyingeographicalworkswrittenduringthe9thandthe10thcenturies うas aplace

nameofsouthernAfghanistanaroundthepresentQandahar.Asmentioned

above うPetech suggestedconnectingtheformtoKhalaj. WhileKuwayama

(1990:253-54) , distinguishingHedaluozhifromGedaluozhi うsupposes theforｭ

mertobeaplacenameandthelattertobeakindofTurkish0日cial title ,the*Qarata 己i.

HedaluozhialsoappearsintheZizhitongjia 凡 Vol. 200:

“On thedayguiweiofthe6thmonth[of661]う eight dudufuandseventy-sixzhou 州

hadbeenplacedinthesixteencountries ぅ i.e. Tl 山lOluo (Tok 瓦ra) うYanda (Hephtal) ヲ

Jibin ぅBosi (Persia)etc. 門

ThisaccountconcernsthereorganizationofAnxiduhufuandsixteencounｭ

triesintheWestinthe1styearofLongshuo(661) ラwhich isdiscussedbelow.

TheT:αng huiy αo Vol.73alsomakesreferencetothesameevent:

“Zhaozhi 僚枝 dudufu hasbeenplacedintheFubaoshidien 伏賓悲顛castle うthe

capitalofthekingofHedaluozhi. 刊

Thesamesourceliststhenamesofzhouandvillagesputunderthecontrol

ofthisdudufu. OnecanfindalmostthesameaccountintheXintαngshu

Vol.43 うas well.ItisobviousfromtheseaccountsthatHedaluozhiwasthenameappliedto

Z瓦bulist 瓦n duringtheXianqingera ,andaZhaozhidudufuwasstationedatthecastleofthekingofHedaluozhiwhensixteenwesterndudufuwerepositioned

inthe1styearofLo 時shuo (661)Recently ぅA. Palumbo(2001:122-24)hasdrawnattentiontothefactthat

HedaluozhiappearsintheDunhuangmanuscriptoftheLαozi hu αhujing 老

子化胡経 (British LibraryOr.8210/s.1857).Thename"d αoshi SuoDaoxuan道士索洞玄" iswrittenattheendofthismanuscript. ThisSuoDaoxuanis

knowntohavecopiedtwootherscriptures うone ofwhichisanon-titledDaoist

sutra(BritishLibraryOr.8210/s.2999)andtheotherthesecondvolumeoftheTaixu αηzhengyi benjijing 太玄真一本際経 (Brit ish LibraryOr.8210/s.3563).

Thedatethesetwodocumentswerecopiedisthe“25t h ofthe11thmonthin

the2吋year ofKaiyuan(5th うJanuary of715) ぺfrom whichonecaninferthatthemanuscriptoftheHuahujingalsobelongstothesameperiod.

4

THEIDENTITYOFTHETURKISHRULERS

3. Gedaluozhi , GeluodazhiandGeluozhi

TheformGedaluozhi うwhich isnearlyidenticaltoHedaluozhiexcluding

theinitialcharacter うappears intheCefuyu αηgui Vol.964aswellastheXin

tα ngshu. TheCefuyu αngui has:

“In the9thmonth[inthe8thyearofKaiyuan(720)], anenvoywassenttoapprove

theenthronementofGedaluozhixielifi α Shiquer 葛達羅支韻利護誓屈爾 as theking

ofXieyuandthatoftheGedaluozhitegin 特勤 as thekingofJibin."

However うthe Jiutαngshu Vol.198 うwhich containsalmostthesamecontent

thoughslightlyabridged うreads:

“[J ib in has]sentanenvoytoChinainthe7thyearofKa剖iYl

onescrollofastronomicalt舵ext 臼Sうa bookofsecretandforeignme吋dici 加ロ問e伺s. [The

Emperor]issuedanorderofi町estiture うsending anenvoyinreturntoapprove

theking[ofJibin]tobetheGeluodazhi 葛羅達支 tegin. 刊

Here うwe findtheform“G elu o d a zh i" . WhilethesimilaraccountintheTtαng

huiy αo Val. 99(Wudedi αn version)providestheform “Geluozhi 葛羅支Jう the

Sikuqu αnshu versionofthesamesourcehas“G elu o d a zh i ," whichmaymean

thatthecharacter “da 達" wasjustdroppedfrom “Geluodazhi円 to become

“Geluozhiうう in theWudedi αn version.3

3 Itshouldbenotedthatthereisanothervariantforthisword.TheSikuqu αnshu versionof1lαng huiy αo Vol.73has:

“T he ZhaozhidudufuhasbeenplacedintheFubaoshidiencastle うthe capitalofthekingofAluoluozhi 阿落羅支. 円

Muchisobscureconcerningtheformationofthepresenttextofthe1lαng huiy αo . T.F旧uhata (1989;1998)pointsoutthatthemanuscriptswhichwereusedasthebaseforconstructingtheSikuqu αnshu versioncouldpreserveanolderform.Furuhatamentionsthreemanuscriptsbelongingtothislineage ぅtwo ofwhicharenowintheNationalCentralLibraryofChineseTaipeiandtheotherbeingintheSeikadoBunkoLibrary ヲJapan. Inthesethreemanuscripts うthe wordinquestioniswrittenAluoluozhi. AllthreecouldbedatedtotheQingperiod , whileうaccording toFuruhata , thislineageofmanuscriptscoulddatebacktotheSouthernSongdynasty.Ifthatisthecase , theformAluoluozhicouldalsobefairlyold. Comparingwithotherforms ぅthe firstcharacter 阿could beinterpretedasanerrorfor 詞(or うpossibly 阿which hasthesamesoundas 詞) . Thoughthesecondcharacter 落is completelydi 旺'erent fromotherforms ぅit ispossiblethatthischaracterreflectstheexistenceof1-/r-asasecondconsonantintheoriginalword うwhich

isquiteinteresting(Seebelow).YetthisformisfoundonlyintheSikuq切附hu versionofthe 1lαng huiy αo andisnotveryeasytoconnectwithotherformsviaasatisfyingexplanation.Thatiswhyitisomittedfromthediscussionhere.

5

MINORDINABA

Table1.iltiibiirinvariouslanguages

Language Source

本 ilt ii b ii r Turkic

韻利設 xiel iflα Chinese Jiutαηgshu うXin tαngshu etc.

υL 入L てo~Y] po Bactrian BactrianDocuments

hitivin α MiddleIndic(7) NezakS瓦h Coins

ふん Arabic/Persian Baladhurl,Tabarletc.

Fromthesourcescitedabove うit isobviousthatGedaluozhiandGeluodazhi

aretranscriptionsofthesameword.Ontheotherhand うcom paringGedaluozhi(whoseMiddleChinesesoundis*kat-dat-Ia-t~re) withHedaluozhi(whoseMC

soundis> 出a-dat-Ia- 印刷ぅonly theinitialsoundisdifferent.Asdiscussedbelow う

thedifferencecanbeexplainedbythedifferenceinthedateoftranscription

andhowtheinformationwastransmitted.Therefore ウit iswarrantedthatwe

tentativelytakethesetwoformstobeslightlydifferenttranscriptionsofthe

sameword.

4. hitiviTI α kharal忌.ca

Asarest此ofthediscussionabove ぅHedalωzhi/Gedaluozhi aretho 時ht to

bethemostappropriateforms.Nextletusconsiderwhatwordthesetermsmusthavetranscibed. Thereisaclueintheso-callednyckyMLK タ=Nezak

S瓦h coinsfou 日d chieflytothesouthofHindukush.Thecoinswhichbearthe

PahlavilegendnyckyMLK'havebeenscrutinizedbyR.GobI(1967).Among

them ぅone groupofcoins ぅcategorized byhimasEmission208(referredasNumH208hereafter) うhas aBrahmilegendassociatedwithitasfollows:

“srThitivirakharal 瓦yap 瓦ramesvara srTvahitiginadevak 瓦rita rp. " (G 凸bl 1 967- i : 142

145)

Thesecondwordhitivirais ぅaccording toH.Humbach(1966:60) ぅthe Midｭ

dIeIndicformoftheTurkishtitleiltiibiir ・ This seemstobehighlyprobabletakingintoaccountthatSims 羽Tilliams (2000:254)identifiesthewordυ山ー

てO~Tj PO hilitberappearingintheBactriandocumentswithiltiibiir.4

AccordingtoM.Mori(1967:427) う ilt ii bii r wasthetitlegiventothechiefs

ofclansotherthantheAshinainEasternTuJue.AsforWesternTuJu札the

4 Asfortheformhilitber ,seeSims-Williams2002:235.

6

THEIDENTITYOFTHETURKISHRULERS

Jiutαngshu Vol.194reads:

“Tong YabguQaganwasbraveandsmart ぅvery goodatwarfare.Inthenorth うhe

conqueredTieLu 鉄勤うin thewest うhe repelledthePersians ぅand inthesouth うhis

territorystretchedasfarastheborderofJibin.Allofthesecountriessubmitted

tohim.Hepossessedhundredsofthousandsofarchers うbeing dominantoverthe

WesternRegions 西域 . First うhe residedattheoldhomeofWuSun 烏孫うthen

movedhispalacetoQianQuannorthofShiguo(C 託) . Hegavethetitlexielif α to

allthekingsofWesternRegionsandalsodispatchedatutuη 吐屯 to eachofthese

regionstocollecttaxesandtributes.HeheldunparalleledpowerintheWest. 門

ThisrecordinformsusthatnotonlyweretheTurkishchiefsbutalsothe

r叫ers ofthecity-statesinTransoxianaandBactriagiventhistitle(Cf.Kuwayama

1990:234).ThelocalrulerofthecityofRob(kl

thetitleυ仏L1" O~Y] PO inthreeoftheBactriandocumentsdatedtothe7thcenｭ

tury(Doc.N うDoc.P andDoc.QofSims-Williams2000). Accordingtothe

Xintαngshu Vol.221andtheCefuyu αngui Vol.964quotedabove うthe King

ofXieyu=Z 瓦bulist 瓦n wasgiventhetitlexielij α. Assumingthatthisxielij α is

identicalwithiltiibiir(Cf.Mori1967) うalong with叫Lてo~Y] po andhitivi 同 we

canconsiderthistitletobethe“ iltiibiir ofGedaluozhi. 刊

Ontheotherhand うon thewords “hitivira kharal瓦va" appearingonthe

legendofNumH208うthere seemstobenodiscussionoftheword“kharal瓦va"

besidesthatofHumbach(1996[1998]:251)whosuggeststhepossibilityof

thiswordbeingakindoferroneoustranscriptionforthenameofsometribe う

possiblytheKharlukh=Qarluq. However ぅthe firsthalfofthe8thcenturyto

whichthesecoinshavebeenattributedis うto thebestofourknowledge うtoo

earlytobethedateofQarluq うs appearanceeventothesouthofAmuDarya.5

Moreoverうno informationぅprior orsubsequent うhas surfacedimplyingtheir

5 V.Minorsky(1982:287-88)saysthatagroupofQarl 問were reportedbyal-TabarltobelocatedtothesouthofAmuDarya.Whatismeanthereisthe“Jabguyaうう who helpedSuluQaghanofTiirgi 話who werefightingwiththeArabsinTokh 瓦rist 瓦n inthe730 冶

(Cf.Tabarl-ii:1547 う1590-91 う1604 ぅ1609). MinorskythoughtthatthisJabguyawasaYabguoftheQarluq うwhich leadhimtoassumethatinthebeginningofthe8thcentury う

therealreadywasagroupofQarluqleadbytheirYabgulivinginTokharist 瓦n. Actually う

al-Taba r! うmentions thisYabguas “Jabghuyaal-Kharlukhl刊on oneoccasion.However う

asinotherinstances うsince heonlyusedtheterm"Jabghuya"or“Jabghuya al-Tukh 瓦rl ， "

itseemsnottobeafirmenoughbasistoinfertheexistenceofQarluqsinTokharistanatsuchanearlydate.O.Pritsak(1951:274& n.2う275) criticizesthissuppositionandpointsoutthatitwasafter766thattheQarluqbecameanheirofWesternTuJue.

7

MINORDINABA

Table2.Khalajinvariouslanguages

Language Source

Khalac ?

詞達羅支/葛達羅支 Chinese Jiutαηgshuう Xin tαngshu etc.

xα 入ασo Bactrian BactrianDocuments

Kharal 託a MiddleIndic(7) NezakS瓦h Coins

εi ム Arabic/Persian I~takhrl ， Mas'lidletc.

activitytothesouthofHindukush.

Thereisadifferentviewonthereadingofthefinalconsonantofthis

word.Thecharacterstranscribingthesoundv-andc-areverysimilarinthe

Brahmiscriptandsometimesnoteasytodiscern.Thecoinsofthistypewere

alreadyreportedinthe19thcenturyandA.Cunningham う J . PrincepandE.

Thomashaveshowntheirreadings(Cunningham1962:269) ぅwhich are “Ai 廿re1加n

cha払?" “Ain1n cha"and“Kharala cha"respectively.Allofthemreadthefinal

consonantasch-=c-.Recently ヲ J . Harrr 凶ta (1996:378-9)6andYoshida(seeabove)alsoagreeintheirreading“kharaHica ."

KhalajiswritteninArabicscriptas 九i ムAs isknown ぅ r intheArabicscripttranscribesthesoundj-andc-.IntneBactriandocuments うthe nameis

writtenasXαλασO ぅthe Bactrianσtranscribing thesounds-andc-.Withthis

inmindonecansafelyassumetheoriginalnameofKhalajasKhalac.This

tribewaslivinginthesouthwesternregionofAfghanistaninthe10thcentury

(Cf.Minorsky1940). MeanwhileotherpassagesintheBactriandocuments

indicatetheexistenceofthattribeinthenorthernskirtofHindukushinthe7thandthe8thcenturies(Sims-Williams2000:82 ヲ98-104; Cf.Sims ・机Tilliams

2002:234). Therefore ぅif "kharal 瓦va/kharalaca" ofNumH208istakenasthenameofsometribe ぅKhalaj isamoreappropriatecandidatethanQarluq.

Theseconsiderationslendsupporttothereading“hitivira kharal泌がうfrom

thelegendofNumH208asthe"iltiibiirofKhalaj

5. KhalajandtheTang

Ofthetwophrasesdiscussedabove , i.e. “Gedaluozhixielifiα竹 and “hitiviγα

kharal従ピ う xielijαand hitiviraaredifferenttranscriptionsofthesameword.

6However, Harmattaidentifiesthiswith 本Qarg i1a c"i 昌撤支which isthenameofthekingofK 瓦pi 釘mentioned intheJiuTIαngshu andtakesitasanamefortheroyalfamily.

8

THEIDENTITYOFTHETURKISHRULERS

Hereafter うthe plausibilityofidentifyingGedaluozhi うwhich maybeidentical

withHedaluozhi うwith kharal 必a shouldbediscussed.Forthatpurpose ,takingalookatthepoliticalanddiplomaticrelationsbetweentheTangandTuJue

wouldbeuseful うas itisimportanttoconsiderunderwhatcircumstancetheseChinesetranscriptionsappearedinthesources.

AfterthebreakupofEasternTuJuebytheTang , thelattergradually

expandedtheirinfluencefurtherwest うwhile WesternTuJuereestablisheditspowerunderTongYabguQaganatthebeginningofthe7thcentury.Afterthe

deathofTongYabgu ぅYipi DuoluQagan 乙毘日出陸可汗established hisruleover

theWesternTuJue. However , in642 うthe TangbroughtYipiShekuiQagan

乙毘射置可汗 into powerandcrushedDuoluQagan. AshinaHeluhadbeen

ayαbgu forDuoluQaganbutsurrenderedtotheTangafterDuoluQagan 冶

defeat. Nevertheless うHelu rebelledafterthedeathofEmperorTaiZong 太

宗. HeinvadedtheterritoryofShekuiQaganandtriedtoconquerGaochang

高昌 and Tingzhou 庭州(650-651). TheTangdispatchedarmiesunderthe

commandofSuDingfang 蘇定方ヲAshina Mishe 阿史那禰射and AshinaBuzhen

阿史那歩真. In657 うSu DingfangwonasweepingvictoryoverHeluatSuy 瓦b.

ThelatterfledtothewestbuteventuallywascapturedatCaebyXiaoSiye着嗣業.

SubjugatingHelu うs rebellion うthe Tangimmediatelysetintopostwarmanｭ

agement うincluding therelocationoftheAnxiduhufutoKueaandtherearｭ

rangementofHell どs formerterritoryunderthenewduhufu.TheX'初 旬ngshu

Vol.215breads:

“After Heluうs downfallうhis formerterritorywasdividedintosomezhouandxu αη

勝、 to whichmanyclansweremadetoimmigrate. TheMukun 木昆clan wasrefashionedintotheFuyan 旬延 dudufu， alongwiththeTuqishiSllogemohe 突騎施

索葛莫賀 clan beingrefashionedintotheWalu 日毘鹿 dudufuうthe TuqishiAlishi 突

騎施阿利施 clan beingrefashionedintotheJieshan 繋山 dudufuうthe HuluwuQue胡株屋闘 cl an beingrefashionedintotheYanbo 盤泊 dudufuうthe SheshetiTun 揖

舎提轍 cl an beingrefashionedintotheShuanghe 讐河 dudufuうand theShunishiChuban 鼠尼施慮半clan beingrefashionedintotheYingsuo 鷹裟 dudufu. Then う

thetwoduhufu う Kunli ng 昆陵 and Mengchi 蒙池うwere placedincontrolofthese

dudufus.Ineverycountry ぅwhich wassubjugatedbytheTang うa zhouwassetup ぅ

beingputunderthecontroloftheAnxiduhufuuptotheborderofPersiaintheWest. 門

ThereorganizingprojectofWesternRegionsbeganintheyearfollowing

Helu'ssubjugation うi.e. 658andwasbroughtaboutinthe1styearofLongshuo

9

MINORDINABA

(661)accordingtotheT:αng h'l均αo Vol.73:

“On the2凶day ofthe5thmonthinthe3rdyear[ofXianqing](7t h う June of658) う

theAI 凶 d'l山ifu wasmovedtoKuca.TheformerAnxi[dudu]wastransformed

intotheXizhou 西川 dudu， towhichQuZhizhan 麹智湛wasappointedtogovernthe

oldplaceofGaochang.AsWesternRegionswerebeingsubjugated うenvoys were

dispatchedtoSama 訂rqa 凶(Ka 時gω) andTokh 瓦訂ris 抗t 瓦加n (Tl 山

thes討it印ua抗ti ぬon凡うthe productsofthosec∞oun 凶ltrie 白s (訪其地風俗物産) ，う and thehistory

Oぱf thecreationandabolition0ぱf zhωau orxu αη there aswell. Theysubmitted

drawingsandplatesasreportsoftheirmissions うwhich wereutilizedbythe0血cers

tocompilethesixtyvolumesoftheXiyutuzhi 西域圃志 "

"Onthe17thdayofthe6thmonthinthe1styearofLongshuo(7t h う July of

661) ぅ 幼au andxu αn wereplacedinTokharist 加. TheenvoyWangMi 時yuan 王

名遠 submitted theXiyutuji 西域園記 . Herequestedplacingdudufuinthesixteen

countrieswestofKhotanandeastofPersia うlaying 80zhau う 1 10 xu αη and 126

militarystations 軍府 . HealsoerectedastonemonumentinTokh 瓦ristan torecord

theEmperor うs achievementsandvirtues. 門

ThemissionofWangMingyuanwasalsorecordedintheXin tαngshu

Vol.58:

"Xiyutuzhi う sixty volumes. Gaozong 高宗dispatched envoystoSamarqandand

Tokh 瓦ristan toinvestigatethesituationandtheproductsofthosecountries うafter

whichheordereddrawingstobesubmitted.TheEmperoralsoorderedthe0伍cers

tocompile[thebookonWesternRegions].XuJingzong 許敬宗received thisorder

andcompleteditinthe3rdyearofXianqing."

Likewise うthe T:αng huiy αo Vol.36reads:

“On the9thdayofthe5thmonthinthe3rdyearofXianqi 時(14th うJune of658) ,asWesternRegionswerebeingsubjugated うenvoys weresenttoSamarqandand

Tokh 瓦rist 瓦n toinvestigatethesituation うproducts andthehistoryofcreationand

abolitionofzhauorxu αn there.Asaresultぅmaps anddrawingsweresubmitted う

whichwereutilizedbythe0白cers incompilingthesixtyvolumesoftheXiyu

tuzhiunderthedirectionofXuJingzong.Whenthebookwascompleted , allthe

scholarshighlyadmireditsextensiveness."

AccordingtotheJiutαngshu Vol.198 うWang Mingyuanhadbeenagoverｭ

norofNanyouxu αn 南由牒 う Longzhou 臨州 . HewasdispatchedtoTokh 瓦rist 瓦n

atthecommencementofthereorganizationprojectin658andreturnedto

10

THEIDENTITYOFTHETURKISHRULERS

Chinain661submittingtheXiyutujiasareportofhismission.7

TheXiyutuzhicompiledbyXuJingzongandotherofficerswasalsoreferred

toastheXiguozhi 西園志 . Itwascompletedassixtyvolumes ぅwhich later ,withtheadditionoffortyvolumesofplatesanddrawings うwas enlargedtoonehundredvolumesin666(the3計rd yl戸ea訂r ofLi加nde 二the Is剖t Y戸ea紅r 0ぱf Ganfl 島engω )(F1αe仰η Zh'l凶T匂uli

AccordingtotheXi仇7ηl t的α7η19ρ8hz 仏1" う the edicttocompilethebookwaωs issuedin

the3r吋dy戸ea訂r ofXi同anqing andthesixty-volumeversionwascompletedwithin

thatveryyear うwhich maymeanthatthe 五rst versionoftheXiyutuzhiwas

theverbatimcompilationofthevarioussourcesconcerningtotheWestern

Regions。百 ered toandaccumulatedbytheCentralGovernment(Cf.Uchida1965:143).TheF1αe印η zhz川uli

theXi旬νt匂1， t加uz}幼hi waωs compiledbasedonthetravelr問epo ぽrt 臼s ofXua 佃nzang うWan

Xuansuo 壬玄策and others.Togetherwiththeseworks ぅprior workslikePei

Ju'sXiyutuji8 andthebiographicaldatafromtheGαoseng zhu αη might also

havebeenutilized , withinformationpassedbetweentheembassiesandthe

Tang. Thepassage “Whenthebookwascompleted , allthescholarshighlyadmireditsextensiveness ううin theX如 何ηgshu seemstoindicatethat うunder

thedirectionofXuJingzongwhohadmuchexperienceinthecompilationof

theHistorieslikeSuishu , Jinshuandetc. うthis bookwascompiledfromvarious

sourcesavailableatthattime.TheXiyutujisubmittedbyWangMingyuanmusthavebeenapartof

theforty-volumesupplement.ThisreportbyWangMingyuan ぅthough titled

tuji う had tocontainnotonlyplatesanddrawingsbutalsootherinformation

collectedduringhisthreeyearmission.Forinstance ぅthe Jiutαng8hu Vol.198

reads:

“In the3rdyearofXianqing うwhen investigatingthesituationofthatcountry

(=Jibin) うthey notethatthefounderoftheroyalfamilywasXinnie 馨草ぅand

thatdowntothepresentkingHejiezhi 昌顕支うthe thronewaspassedfromfather

tosonfortwelvegenerations."

TheoriginalChineseofthesentenceunderlinedis 訪其園俗うwhich cor-

7 G.UchidathinksthatWangMingyuanwassenttoTokh 瓦rist 瓦n twice,i.e.inthe3rdyearofXianqi 時and the1styearofLo 時shuo (Uchida1965:142).Nevertheless うconsidering

thepurposeandresultsofhismission うit ismoreappropriatetothinkofitashavingbeencarriedoutforthreeyears.

8 PeiJu うsX 句u tujiissupposedtohavebeenreferencedinthedescriptionoftheWesterncountriesintheSuishu.Thebookwascompletedby606(Kuwayama1990:125-126).

11

MINORDINABA

rectlycorrespondstowhatismentionedintheXintαngshu Vol.5andthe

Tα ng huiy αo Vol.73 うnamely ヲthat theenvoysdispatchedtoSamarqandand

Tokh 瓦ristan intheXianqingera “investigatedthesituationandtheproducts

ofthosecountries(訪其地風俗物産) ." Therefore ぅthe informationconcerning

JibinintheJiutαηgshu isthoughttobederivedfromWangMingyuan 冶re­

port.Moreover , fromtheaccountoftheTIαng huiy α0 ， itcanbededucedthatthereportprovidedinvaluabledataforrefashioningthesixteendudufuinthe

WesternRegions.

TheaccountfoundintheXintαngshu Vol.221bonZ瓦buli stan saysthatthe

nameHedaluozhiappearedintheXianqingera.Consideringtheaboveleads

ustoassumethatthealternatenamewastheresultofWangMingyuan 冶

missioninTokh 瓦ristan. Unfortunately ぅno informationrevealshowhisinvesｭ

tigationwascarriedout.Itisalsounknownwhetherhehimselfexploredfrom

Tokh 瓦rist 瓦n tothesouthofHindukushorhecollectedthedatathroughinforｭ

mants. Nevertheless , itisclearthathismissionwasalmostconcurrentwith

theemergenceoftheKhalajpowertothesouthofHindukushandherecorded

thenameasHedaluozhi.

6. HedaluDzhi/Gedaluozhi , kharal託a andKhalaj

Thoughitisnotevidentwhythenamewasrecordedinthe ぅso tospeak う

IndianizedformHedaluozhi うthere aretwopossibilities.ThefirstisthatWangMingyuanorhisinformantrecordedthenameintheIndianizedform. The

secondisthattheKhalaj=Khalacthemselvesreferredtotheirownnamein

Indianizedform.ThelattermayberatherprobableastheKhalajthemselves

usedtheIndianizedform"kharal 説ゲ intheirowncoinsaboutahalfcenturylater.

StillitmustnotbeforgottenthatHedaluozhiandGedaluozhiarenotexact

transcriptionsfor“kharal託a. ' う The secondcharacter 達usually transcribesthe

sounddiir-(Cf. 達磨くdharma) andisnotexpectedtoappearhere.Thoughthe

detailedlinguisticcircumstancetothesouthofHindukushduringthisperiod

isstillambiguous うit maysafelybesupposedthatMiddleIndicandBactrian

werethetwomainlanguagesthere うbecause theyaretwoofthethreelanguagesusedintheNezakSahcoins うlegends (thethirdbeingPahlavi).Accordingto

Sims-Williams(1997:23) うwhen anl-andanr-comeintodirectcontact うthe lｭ

tendstochangetod-intheBactriandocumentsbelongingtothe7thcentury

andlater.Inferringfromthis うit canbeassumedthatwhenthenameKhalac

wasIndianizedto"kharal 託a" ,ortoaformsimilartoit うand wastransmitted

12

THEIDENTITYOFTHETURKISHRULERS

toWangMingyuan , therebybeingaffectedbytheBactrianpronunciation ,thesecondconsonantwasheardasd-. Ontheotherhand ヲconcerning the

ChinesetranscriptionofMiddlelndic うwe findinstanceswhereMiddlelndic

d-istranscribedbyMiddleChinesel-/r-andMiddlelndicl-istranscribed

byMiddleChinesed-.AccordingtoS.Karashima(1994:19 う27 ， 67) うthis may

reflectthefactthatintheMiddlelndicofnorthwesternIndia ,pronunciationof

d-andl-hadmovedcloser.ThecaseofHedaluozhi うGedaluozhi andkharal 説a

couldbeexplainedinthiswayaswell.

Nevertheless うas noticedinbothoftheexplanations うit isnotatallaneasytasktoascertaininwhichstageofcommunicationthesoundd-orthe

correspondingChinesecharacter 達appeared うas thislargelydependsonthe

conditionofactualcommunicationorconversations.9

ConcerningthereasonwhythenameofKhalajwaslndianized うwe may

recallthepredominanceofIndiancultureintheareastretchingovertheHinｭdukushduringthisperiod. AbouthalfoftheNezakS瓦h coins うlegends are

writteninlndicwithBrahmiscript.lO BesidesうIndian culturespreadevento

BactriaandSogdianaintheprecedingperiod , i.e. うthe KidariteandHephｭ

thaliteperiod. Thisphenomenawas ぅaccording toKuwayama(2002:154) , aconsequenceofthedispersalofmerchantsawayfromGandh 瓦ra tothoseareas

afterthefalloftheHephthalites. Ontheotherhand , F.Grenet(2002:213)regardsthistobeduetothecloserelationshipofBactriaandSogdianawith

NorthwesternIndiaundertheruleoftheKidaritesandHephthalites. Harｭ

matta(1996:370)referstothecoinsissuedinKhuttalaroundthisperiodonwhichthenamesoftherulers(Hephthaliteprinces , accordingtoHarmatta)

arescribedinlndianizedforms.lndianizationofthenameofKhalajmaywell

9 Marq凶rt (Marquart& deGroot1915:258 ぅn.1) andPetech(1964:294)referstothe

word 丸山> whichappearsintheitineraryofAbuDulafMis'arb.Muhalhil , quotedby

al-Ya 伊t inhisMu 'j・αm al-Buldiin(iii:443). Thisiscalled “the FirstLetter"ofAbu

DulafandthewordinquestionappearsalongwiththeQarluqinthedescriptionofthe

peoplewholivedalongtheroadfromWesternTurkestantothewesternpartofChina.

Marquartsuggeststhepossibleconnectionbetweentheletter..kwith 達supposing that

r.and んare confusedhere. However うthe “F irst Letter"ofAb 己Dulaf isthoughtto

oebaseaonhearsaycollectedinBukhara(Minorsky1955:11-18)andtheform がムふ

hasnotprovedtobefoundanywherebuthere. Therefore ぅthough thewordwhiChto

someextentresemblesHedaluozhi/GedaluozhiandappearstogetherwiththeQarluqis

interesting , itisnotpossibletomakeanykindofconclusiveargumentb邸付on this

word.

lO Gるbl hasclassifiedtheNezakS忌h coinsintoabouteightygroups ぅmore thanahalfof

whichutilizeBrahmiinscriptionsorletters(Cf.GぬI 1967-i:132-186).

13

MINORDINABA

followthesametendency. Theymighthavethoughttheirnameshouldbe

IndianizedinordertoberecognizedasoneoftherulersoftheIndianworld.

Thiscouldbereflectedinthefactthatinthe8thyearofKaiyuan うabout sevｭ

entyyearsaftertheXianqingera うKhalaj sentenvoystotheTangcourtwho

referredtothemselvesasGedaluozhi. Forthem うreferring tothemselvesby

suchanamewasprobablyproperandbefittingarulerwhocouldsendenvoys

tothegreatTangdynasty.

Itisnotclear うhowever ， whythistimeadifferentうthough slight ,transcripｭtionwasadopted. Toascribethischangetothedifferenceinperiodorthe

whimofatranscriberisasimpleremedy. Still うthe factthatsomeArabic

andPersiansourcesapplytheformQalajinsteadofKhalajshouldbenoted.

D~ ωan Lugat αt-Turk う a Turkishdictionaryofthe11thcenturybyMaqm 己d

al-K 説garI explainstheetymologyofthenameKhalajthroughtheTurkishwords “qal aどうIn Jami(al-Ttαωarikhう a worldhistoryofthe14thcentury

compiledbyRa 訂d aI-DIn ぅthe tribeiscalled“Qalaj ." Thefactthatsources

writtentothewestofPamirdi 旺'er intheirtranscriptionsofthefirstconsonant

seemstocorrespondtothedualChinesetranscriptions.Maqmudal-K 説garI

alsostates:

“The OyuzandQif ，ぬq sometimeschangeqiijtokhii'.TheyareasectionoftheKhalaj.Theysay:KhIYZIMkhlzim'Mydaughter(bint!) うwhile theTurkssay:

QYZIMqlzim. 刊(Dankoff &Kelly1982-ii:263)

Meanwhile うin theMiddleChinesetranscription ， 言IT and 葛usually tranｭ

scribethesoundkhaandqα respectively. Therefore うit mightbeassumed

thatHedaluozhirecordedintheXianqingeratranscribedthenameKhal 泌

whichisattestedintheBactrian うArabic andPersiansources うwhile Gedaluoｭzhirecordedinthe8thyearofKaiyuantranscribedthenameQalacwhichis

witnessedintheworksofMahmudal-K 泌さarI andRa 訂d aI-DIn.Itmayalso

besupposedthattheoriginalnameofthetribewasQala 己うwhich theKhalaj

peoplethemselvespronouncedKhalacbychangingqajtokha'(Cf.Doerfer

1971:173)

Bethatasitrna ぁmy conjecturehereconcerningtheexistenceoftwoChinesetranscriptionsisthattheymightbepartlycausedbythedifferencein

thetimeoftranscription.ThosewhoactedasintermediatersbetweenKhalaj

andtheTangeachtimemusthavebeendi 旺'erent aswell うwhich mayimply

thatthepronunciationofthosemediatorscouldhavedi 旺'ered slightly.

Itisnowappropriatetosumupthediscussion:

First うfrom thevariantsinChinesetranscription , HedaluozhiandGedalu-

14

THEIDENTITYOFTHETURKISHRULERS

ozhishouldbetakenasthecorrectforms うboth ofwhichweretranscriptions

ofthesameword.

Second うthere aretwowordswhichseemtobeindicativeofthetitleofsomeruler. OneisGedaluozhixielif α recorded intheChinesesourcesand

theotheris“hitivira kharal瓦とがうwhich appearedonthelegendsoftheNezak

Sahcoins ウboth ofwhichareconcernedwithalmostthesameperiodandarea.

Byassumingthetitle“khalac iltiibiir， ' う it becomespossibletoidentifybothas

beingdifferenttranscriptionsofthesametitleusingdifferentlanguagesand

scripts.

Third う if theseweretranscriptionsoftheIndianizedformofKhalaj うthe

ChineseHedaluozhiandGedaluozhicanbeunderstoodtocorrespondtothe

phonicvariantsoftheoriginaltribename うi.e. KhalacandQala じ .

Thetitlementionedabovewa剖s bornbytherulerswhomweknowfrom

theIslamicsωour 陀ce侃s tωo havebeen

Khala 吋j i詰s themostc∞on凶siおstωen 凶t hypothesistoexplainthepoliticalsituationof

theareasouth0ぱf Hi加nduku 凶1凶sh aroundthattime.

7. KhalajinCentralAsia

AsforthequestionofwhenandfromwheretheseKhalajcametoAfghanistan う

regrettablywedonothavesu 茄cient information.StillMinorsky'spioneerwork

(1940)andotherindirectinformationleadustothefollowingdescription.

It hω been pointedoutthattheoriginoftheKhalajmightbeassociｭ

atedwiththeHephthalites.Thisassumptionisbasedontheaccountinthe

Mα !ii tz!:tα l- ' Ulii m (119-120)byal-Khwarizmlwhichsays"TheKhalajandKanjlnaTurksaretheremnantsoftheHephthalites. 竹It willnot うhowever うbe

verypromisingtoexpectexactinformationconcerningethnicityinthisstateｭ

mentbecausethisworklumpstheKhalajandtheKanjlna(=Kumedh)inthe

samecategory(Cf.Clauson&Bosworth1965:8-9).Ontheotherhand ぅcon­

sideringthatthesourcesbelongingtotheIslamicperiodgenerallydistinguish

theKhalajfromotherTurksorTurkmans ぅand Huicao(Kuwayama1992:38 う

122-123)alsodistinguishesTurksinKめul andZabulistanfromtheTurksto

thenorthofHindukush うit isnotveryprobabletothinkoftheKhalajasa

partofWesternTuJue.IwillconfinemyselfheretosaythattheHephthaliteempiremightwellhavebeenanamalgamofvariousnomadicandnon-nomadic

people11andthattheKhalajpeople うwho livedsomewhereinCentralAsia ,had

11 Cf.Maroth1990;Clauson&Bosworth1965 ぅloco sit.

15

MINORDINABA

beeninvolvedin(ortookpartin)therapidexpansionoftheempire ぅmeaning

onesectionofthemmovedintoAfghanistan.AsMinorskypointedout , IbnKhudadhbih(α l-MIα s ii likωαα al-M，αmaωli批t依k : 28 う

31)r陀epoωrt 臼s thatagroup0ぱf Khala 吋j andQa紅r It叫waωs livingtotheeast0ぱf Sy戸r

Darya.Othersourcescanbecitedaswell.TheXintαngshu Vol.217mentions

atribecalledtheBoma 駁馬adj acenttoJiegu 結骨(=Kirgiz).Theformerwas

alsoreferredtoastheElazhi 遁刺支 (υαγ. Helazhi 昌刺支) orBila 弊刺 . The

Tongdi αn Vol.200callsthesametribeHela 局刺(Cf. Shiratori1970:615~620).

Moreover ぅal-Id6sI (OpusGeographicum:714~715) describesthataboutacouｭ

pIeofdayseastofTar 瓦z residedthewinterquarteroftheQarluqandthe

winterquarteroftheKhalajadjacenttoit. Thegeographerfurtherrecords

thatonthewayfromTaraztotheterritoryofKIm 瓦k therewasacastlecalled

xyxmwhichwastheresidenceoftheKhalajking.AlthoughtheexactlocationoftheKIm 瓦k territoryishardtospecify うal- IdrIslandotherArabandPersian

geographerstendtoconnectittotheIrtishbasin.T.Moriyasu(1977:28-30&

ns.113-115)suggeststhepossibilitythatthetribementionedintheChinese

sourcesmightbethesameastheKhalajbasedonthearrangementofthegeoｭ

graphicallocations.Asamatteroffact うIslamic sourcesdescribetheKIm 瓦k as

awesternornorthernneighborofKi 培iz (Cf.Minorsky1982:305).Thelatter

was ぅaccording totheXintαTゅhuぅ the neighboroftheBoma/Elazhi/Helazhi.Thisevidenceseemstoindicatethat うfrom the9thtothe10thcenturies う

asectionoftheKhalajstilllivedtotheeastofT紅白and westoftheIrtish

basin ぅpossibly aroundIssik-Kulandeastward.TheJαhiin-n iim仏 a 13thcenｭ

turyPersianmanualofgeographybyMul}.ammadNajIbBakran ,saysthatthe

“Khalaj areakindofTurkswho , inancienttimes うhad cometoZ瓦bulistan

fromtheterritoryofQarl 問・刊 ( Jα hii n-n ii mα: 73). ThisalsorevealsthattheoriginalhomeoftheKhalajwasinthatarea.

Unfortunately うno informationisavailableconcerningthelaterhistoryof

theCentralAsianKhalaj.Fromthe10thcenturyon , theKhalajweremenｭ

tionedintheareasouthofAmuDarya うespecially innorthernIndiaandthe

easternsectionoftheIranianplateau うwhich hasalreadybeendescribedby

Minorsky(1940)

16

THEIDENTITYOFTHETURKISHRULERS

Bibliography

Bombaci うA. (1970)OntheAncientTurkicTitleEltabar.In:ProceedingsoftheIXth

MeetingofthePerm αnent Intern αtion αl AltaisticConference , Naplesぅpp.1-66.

Cunni時ha瓜A. (1962)LaterIndo-Scythians.(reprint) ぅDelhi.

Cα}

Ta剖ipe i.

Clauson ぅG. & C.E.Bosworth(1965) ヲAI-Xw 訂izml onthePeopleofCe 山古I Asia.

JRAS1965 ぅ pp . 2-1 2 .

Da 此off うR. &J.Kelly(tr.)(1982)CompendiumoftheTl 川ic Dialects.3Vols ぅH紅白

vardUniversityPrinting0血ce o

Doerfer うG. (1971)Khal α'J Materials.IndianaUniversity.

Furuhata うT. (1989)OnthevariousversionsoftheT' αng hui-y αo . Ti5hi5gaku78 ぅ pp . 82

95.(Japanese)

Furul 同九T. (1998)AStudyonCirculationoftheTtαng hui-y αo . TheTi5yi5shi-Kenkyu

57(1) , pp.96-124.(Japanese)

G凸bl ぅR. (1967)DokumentezurGeschichtederiranischenHunneninBαkt 巾n und

Indien.4Vals. うWiesbaden.

GrenetぅF. (2002)RegionalInteractioninCentralAsiaandNorthwestIndiainthe

KidariteandHephthalitePeriod.In:Sims-Williams うN. (ed.)Indo-IranianL仰­

gu αges αnd Peoples.Oxford うpp.203-224.

Harn 凶ta ぅJ. (1996)TokharistanandGandh 瓦ra underWesternTurkRule(650-750).

In:Litvinsky ぅB.A. うZha 時Guang-da &R.Sh.Samghabadi(eds.)Histo γνof Oi り­

iliz αtions ofCent γαl Asi α II I. Paris うpp.367-383.

Humbach うH. (1966)BaetrischeSpracl 時nkmiiler I.Wiesbaden.

Humbach うH. (1996[1998])Pangul ぅa Turco-BactrianRuler.BulletinoftheAsi αInsti­

tute10 う pp . 247-25 1.

IbnKhurd 瓦dhbih ，α l-Mα salikωαα l-Mα malik. ed.deGoeje ，M. ょうLeiden う1967.

al-Idrlsl う Opus Geogn αphicum (KitabNuzhatal-MustaqfiIkhtiriiqal-Afaq').ed.Cerulli う

E. うA. Bombacietal. うNaples &Romeう1982-1984.

Inaba , M.(1991)ThreeKingsofZabulisUininthe7th-8thCenturies.Sein αηAzi α

Kenky 匂 35 ぅ pp . 39-60 . (Japanese)

17

MINORDINABA

Inaba うM. (2004)N瓦y Qal'a , Wujlrist瓦n andtheKhalaj.AnAppendixforVerardi&

Paparatti2004.

Karashima ぅS. (1994)ChouAgonKyonoGengonoKer 向u. Tokyo.(Japanese)

al-Khwaxizmi うMα fati:!}α l- 'Ulum.ed.vanVloten , G. うLeidenう1968.

KuwayamaぅSh. (1990)Kapis i:-G αndhara ShiKenkyu.InstituteforResearchinHuｭ

manities うKyoto University.(Japanese)

Kuwayama うSh. (ed.)(1992)Hi 仰の Wαng Wu-Tianzhuguoz}川町Record ofTravels

inFiveIndicRegio ηs ， Translation αηd Comment α旬. InstituteforResearchin

Humanities , KyotoUniversity.(Japanese)

Kuwayama うSh. (1999)HistoricalNotesonK瓦pisi andKめul intheSixth-Eighth

Centuries.Zinbun34 う pp . 25-77 .

Kuwayama うSh. (2002)AcrosstheHindukushofth εFirst Millennium.Institl 府for

ResearchinHumanities うKyoto University.

Lee うJ. &N.Sims-Williams(2003)TheAntiquitiesandInscriptionofTang-eSafedak.

SilkRo αdArt αnd Archaeology9 ぅ pp . 159-184.

Mar6th うM. (1990)DiepolitischeGeographieAfgl 間山tans im7.-8.Jahrhundert.In:

Har 江川ta ， J.(ed.)FromAlexar 時r theGreattoKillTegin ぅ B吋apest ぅ pp . 1 33-137 .

Marquart うJ. (1901)Erans αhr nachderGeogn αphie desPs.MosesXorenac'i.Berlin.

Marquart うJ. &M.J.deGroot(1915)DasReichZabulundderGotZunvom6.-9.

Jahrhundert.In:Weil うG. (ed.)Festschr ザt 印刷 γd Sαch肌 Berli nぅ pp . 248-292 .

Minorsky ぅV. (1940)TheTurkishDialectoftheKhalaj.BSOS10(2) ぅ pp . 4 1 7-437 .

Minorsky ぅV. (tr.)(1955)AbuDulafMis'aribnMuhalhil'sTrl αvels inIran(ci 陀αA.D.

950).Cairo.

Minorsky ぅV. (tr.)(1982)Ifududal-'Alam:TheRegionsoftheWo 似. (2nded.) うCam­

bridge.

Mori うM. (1967)HistoricalStudiesoftheAncientT1川ic People1.Tokyo.(Japanese)

Moriyasu ぅT. (1977)Aproposdestribusdel'AsieCentralequisetrouventdansles

sourcestibetaines:DRU-GUetHOR.Journ αl ofAsi αn αnd Afric αn Studies14 う

pp.1-48.(Japanese)

Muhammadb.NajibBakr 瓦n ， Jαhan-nam α. ed.Riyahi ラM.A. うTehran ぅ1963.

Palumbo , A.(2001)LaｫScritturadiLaozicheconverteibarbariｻ.Sincretismoe

conflittoideoligicoinunciclodiletteraturareligiosadellaCinamedievale.Ph.D.

18

THEIDENTITYOFTHETURKISHRULERS

Dissertation(IstitutoUniversitarioOrientale うNapoli).

Petech うL. (1964)NotessuK瓦pisi eZ瓦bul. Rivistαdegli Studi0巾T山li 39(4) ヲpp.287-

294.

Pritsak うO. (1951)VondenKarlukzudenKarachaniden.ZDMG101 ぅ pp . 270~300.

Rehma 凡A. (1988)TheLαst TwoDyn αsti ωof theSiihis.(reprint) ぅNew Delhi.

Shiratori うK. (1970)ShinaHondoShuuiShominzoku.ShiratoriKurakichiZensl 凶 4 う

Tokyo ぅpp.549-739. (Japanese)

Sims-Williams ぅN. (1997)NωLight onAncientAfghanistan.SchoolofOrientaland

AfricanStudies うLondon University.

Sims-Williams ぅN. (2000)BactrianDocumentsfromNorthernAfghanistan1.Oxford.

Sims-Williams ヲN. (2002)AncientAfghanistanandItsInvaders:LinguisticEvidence

fromtheBactrianDocumentsandInscriptions.In:Sim トWilliams うN. (ed.)Indoｭ

In αηtαη Lαngu αges αnd Peoples う Oxford ぅ pp . 225-242 .

al-Tabarl うTα 'rzkhα l-Rusulωαα l-Muluk. ed.deGoeje , M.J. うLeiden ぅ1879-190 1.

Uchida , G.(1965)TouKousouChokusenSaiikishiKouroku.Kenkyu(KobeDα句αku

Bung αkukai) 35 ぅpp.140-149. (Japanese)

Verardi うG. &E.Paparatti(2004)BuddhistCω es ofJiighurz αηdQ αrabiigh-e Gh αzn i，

Afgh αnist αn. IsIAO-Rome.

al-Y 瓦qut う Mu 'jαmα l-Buldii n. 5Vols. うBeirutう1984.

Yoshida , Y.(2000[2003])Review:NICHOLASSIMS-WILLIAMS.Bαctrian Docuｭ

mentsfromNorthernAfgh αnist αη I : LegalandEconomicDocument.Studiesin

theKhaliliCollection うVo l. 3.CIlr.Pt.2 う Inscriptions oftheSeleucid αηdP αrthian

Periods αηd ofE αstern In αη αnd CentralAsi α. Vol.4うBactrian. BulletinoftheAsia

Institute14 ヲpp. 154~ 159.

19