a Ss AVY S> S, Ns \ tues Yi CEE: ae EZ ee Li eS i PAN: (ie tis LZ SAA Se ji ZZ Z ty ("te - og" ss foes js LL Yit3zj#7; 4 ee , Z SEE Lee : ep Cae A WY GY Gy tig Le y Lia Z LIST: SAL SZ tj Sipps ee, titjg se \ \\ AS SS NY RG MQ Qu OY \ \ \ \\ \ \ \ \\ NAY \ AA WY AX Ry A A MAY \ \ \ LY Iv WOM WS RAMN RN WV. KK WY \ \ ee Sui ada a A ft) 4 _ v4 fs 4 7 tae Vio. /4¢ 3 ira SMITHSONIAN INSTITUTION SOF BUREAU OF AMERICAN ETHNOLOGY BULLETIN 143 HANDBOOK OF SOUTH AMERICAN INDIANS JuLian H. Stewarp., Editor Volume 2 THE ANDEAN CIVILIZATIONS Prepared in Cooperation With the United States Department of State as a Project of the Interdepartmental Committee on Cultural and Scientific Cooperation. UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1946 For sale by the Superintendent of Documents, U. S. Government Printing Office, Washington 25, D.C. Price $4.25. N INST >= TaN ie JUN 19 1946 2 Ws SSEFICE LIBRE - LETTER OF TRANSMITTAL SMITHSONIAN INSTITUTION, Bureau oF AMERICAN ETHNOLOGY, Washington, D. C., June 1, 1944. Sir: I have the honor to transmit herewith a manuscript entitled ‘““Handbook of South American Indians. Volume 2. The Andean Civilizations,” edited by Julian H. Steward, and to recommend that it be published as a bulletin of the Bureau of American Ethnology. Very respectfully yours, M. W. Stiruine, Chief. Dr. C. G. Asso, Secretary of the Smithsonian Institution. i ici atc Acme elOY GAO ees » ! (romah. 2 vas, ‘ ; ; Xt e\ yNS o t "NX tA , --f CONTENTS PAGE 1 EES I es Rae Toa ee eae ee eee ee, See XXKV EES UNSUNG OS Be ea a ye, XXIX EKO WICGCMONLS -. 2-4. Ot ee ee a i XXX Peace USOTS (O.VOUIME 2 8 te keel ee aE ec we XXXII Part 1. The Andean Highlands: An introduction, by Wendell C. Bennett_- 1 LG SG (cae a eer ee a eae ere Tr eee 1 (Cie nn 3 [Egor by Saas 2210 5 ae Re a Se ee ee re eer Ss ee 3 LEAIC Tew oy 00Yef5 AUIS 10 |G ea eee me a CE oe ~ ee 4 LOO) a a een ee a ee ee ee 6 he Andean Highlands as a whole.......---.--252204. 228% ..=-=- 8 Enehlandenitunal divisions..<..25 222-3 eee oe ee 10 Mba © crit eal AMIGOS 2 ee rk i 13 LU eMcn gies Ns) ae a a ae a ee ae 13 (CORO TSG i a a eee 19 SHIDSISLENCE ACIVINICS....-22-22c2-n2ceeceee tae t oan = 20 WEAPONS AIG WALLA ROE a Pep eS ac IE ah TS = 23 TENTS 0) 471910 a ee nee 24 Architecture, and. building. «ry. 2B Soa ok I 25 (CLAN. ta ey oe hth roar en eo ee 27 Clothing and ornaments... 22 2 ep too = = 8 31 Se RS en a ee ae ear ae 32 of ar ree TU Th) ee ee a ee ee ees 32 IRC (ae rt i a ara? see ee ee 34 Kmowled ee. 6 a ee EEE ge 35 SLOT Ly: | a a ee eee es. ee 36 SET SOULMCTIWAMNMCS 22 ee a el ee eee we ee eI 37 ihe Diacuite-Atacameno.... 22.2. Face pete toe on = 38 TReEnWATONIMCN Gs 2 2 oP 38 PENG RF Rh it PT Ns i 39 The Afacameno =. foe obi oe I 41 IGE SORT czas 1 i a ee er ere ae 42 APD CenM Nabe oe ee PE es 42 Uden ae So a ae se. oS ee 42 five Norphern, Andes. cep: ten pee ee hs EE SS Eo 45 LEPC 6 Cy ee a ele IR IE nee) ee ame ee ee ee 46 ALE CNVITOMIMENG rs oe a ht A a 46 The Highland and Coastal cultures._._......____-__----- 47 DOLE Ta Ee eee pearre 50 fhe.environment and tribes 22.-4.2.2* at oe IS to 50 ANS 0) isi a re ee 55 BEB EE ESE UNC 2 i a i ge Oe ee et te OE eho I 57 SORT TES 5 2 0 ES a ee enn D ene een 59 VI CONTENTS PAGE Part 2. "The Central Aid esm eye. 2a oc ers Ay oe ae ieee pape ea 61 The archeology of the Central Andes, by Wendell C. Bennett_________ 61 Preface: 2 base hee als Meta CO UREA DIES Maun 9, PRT EN ee EME TUE PS 61 The region as a whole: archeological unity._____._.._.__---_--_-- 63 Problems sc 5 pee 2 eee es ie ee 64 Soulrees: 2. 245524255 ee eee ee es 66 Geogranhiye oi) se) Bap peer ae eae ee) ee 67 C@hronologyes 2a ale ee ied AT ee 70 Terminology. ester eet ain ne ee on noe or 74 The archeological periods.wn..ck eae ee 79 The ‘Chavin-Petiodsme4 24-414 sin wee wo 9 oo os 81 ‘The: Chavyimisitertst st etre ses 2S UY: Se ee 81 Coast*@hay in. Ue 2S ee aes sk eee ee 88 EM Salis = trot ee hank a ey enn om ee eo 4 92 The Harly* Periods:-South"Coast=<22e-.r ee ee eee 92 IN QZ CRRA Sn eRe A ie ee ee me ors = 4 92 PATA CAS tere ed ent on Oe omens eet em BE et 95 ‘The Harly*Periods:” Central Coast=<-2+22++ 22+ 2222": =e 96 Wihite=on=red 44% 2= 4422": eve ee Se eee 97 Interlocking ent 2s a. de ee) ee ee 98 arly viGimMater een eaeiae se eer nee on + ee 98 Phe Harly- Periods Nerth: Coasts:22.22420. 2 ee eee 99 BA TCG) O10 peg tl ee ll ha yk Ng id Par ef Mcp as 99 Galhnazoree tse eee har ee eee, Jenn ene 104 The Harly* Periods:* North Highlands’2]2" 0 22200 ae 104 VOC USIG ay Bt ttc e+ gr me ecplenth ntti eee ee 104 The Early Periods: South Highlands and Bolivia_____-_-_-_-_-_- 109 Harky aniahuana cont ces ease oot ee ehh een Ree 112 Classic’ Tiahuanacos 2. + 2 ee eee eee ee eee 114 CIPI Ba tt ee et He ee be eel in oe eileen ere 118 PUNC AT tt an tr on ene we Sear tee oe ne a ee al 120 Derived Miahuanaco: 2222 tt eo eee ae ee 121 Phe Middle Periods +--+ 24a +e =e ae tee as SP ee ee 122 The Middle Periods: Coast of Peri... =----2-22_-2__---2-_-_-_ 122 SPHEeSiteS2 Veg 7 + ne eee edie oo rc ee 122 Coast TiahuanacosA:=+2 2222 eee eee Sa 125 Coast Tiahuanaco2B es 1-22 ee ee eee ee eee 128 Coast-Highland Tiahuanaco relationships___-------------- 129 The Middle Periods: North Highlands_._-.__-_-------------- 130 Wilkawain-Tiahvuandco.. +++ a eee eee 131 Maranéne i227: tank eet: ser te eee See cee 132 Middle Huamaehuco-n:. sstennceeee ae ee 2 ieee)» = 802 YEE VM Ta Ge es ae a ste oe seeeates Ame 2 2 803 hie VENIaN CAV IGA Ra ee ee eee ln eet pli D = < 806 TRE tBUNA token toe wee se ea we lee mite BT. 806 PS GS cee RUA TRAE 208 5 See wep i nh oe 807 The Colorado and Cayapa______-2-.---2_--_------- 807 he WVMalabassenseccenc22.coeord bap aepellDl 2.2 oe 2 807 Phe SY wim bow OCA ee. ealqaae meriwamcrbeyo! = = 2 st 807 ‘The Inca/conquest of Heuadors.2.2 stearcieerm. beweemell 222s 808 Colonial and modern conditions____________________---___--- 812 Bibliography Soo (Serine Peels aemepban bee WE = 2 oo. 821 The archeology of Colombia, by Wendell C. Bennett__.....___._--_-- 823 Tien GN a ep a soe = a lt Eo 823 GeGerapliy Uh wae ION) a er nef oct amore SE oo oo 2 824 Archeological divisions______...______- Depo a, gia oe wx 826 Distribution andieamparicon ~~ -2. 0022-22 yeoladie Mis... <2 827 Chronology.___- I hth ol a no 8 8 oo pepe AES = 829 Bro blemseeo me. WOTSS. 5. 2-2» -tachan wk An wel. edadidl) os 830 SOLTCes Ore. a ee wee oe ee Wee ees babe ted . | 830 Archeological Zonésse)..- « --aaelgieth laalslne hag tetida ll . 832 TING OOD oo en ae ee os sn ee heh enee she». 832 Zee Pper Caneda=<-Harcncecececn- > sedhyerinnd alo: » - 834 Seem ODI TaaA A Fiennes. SIE SLOAN IRE IRNI 0) op) oo oo og enone AL » » 838 AAAS Tie feb ie oe le ee GD. ale re maw eiodd _ 841 BeA@ MI ChG Man -6 4 ose eee eee ee ee ey a eamepphi ) . = 842 GULP airona te SORE Oe NY Penida movemupuimetgee — - 844 (POAUDUACUSH Meee e eae oe ew ee ee oe ee EE eo ss 848 Sr MernAGeMEnOmee sore ay ee el 849 XIV CONTENTS Part 4. The Northern Andes—Continued. The archeology of Colombia, by Wendell C. Bennett—Continued. Archeological zones—Continued. PAGE Nhiscellaneos jute i ee ae 850 Babb liggrrarpiny eee ee ia cee I 850 The archeology of San Agustin and Tierradentro, Colombia, by Gregorio \Hermandez de Albass sot sees lta dea 851 San VAeUsHMiseees lee ec Oe ee ee 851 SLOne cul pire: 2. oon ou eee ae ee 851 ATehitectures 2.2. oe eee ee 853 Stone implements... . eee eee 854 Ceramics... oo cn en ee 854 Ornaments. vio. 422.2....-c.ce cn teh ee eee 856 Weapons... oo et dane thi ee 856 MierrademtTros 2s. o22)2-0.l ee eee 856 Bibliography.......o0lfauel.w! sebeuss. 6 eee eee 859 The archeology of the Popaydn region, Colombia, by Henri Lehmann. 861 Introduction{2 2-2 2 kee eee eee ee 861 Stoneisculptures on eee ee 861 Pictographs... teu iwe oh Je ott edt 4 een lo eee 861 Burial customs... 32 obec oe ee eee ee 862 Ceramics: oe cco eee eee eee 862 Stoneiaxkes foie sk hoe ee ee ee eee 864 Tribes of the Sierra Nevada de Santa Marta, Colombia, by Willard Z. Parkes 20232 eos eee ed oe nul 2 oe 865 Introductiones: o.oo ee ee ee 865 Soureesi2. 2h eee. pee ee 865 Geographyiie he 2 oe oe ne pe ae ee 867 Tribes 22 kee eo le oe es ae a ag Sl 868 Populatione e222 ous. ba ccd bn eas ee ee 870 Histone ee eae ce oh oll a 872 Gultarete cara Fie Seay 7a ee a 872 Subsistence: activitiesee +s.) luca ohio ea eee 872 Villaces and houses: 4.22005... eee ee ee 875 Engineering works... 22. oo a ee eee 878 Dress:and ornaments... .1ebedad ls Sue ee 879 diransportation.:......--2¢el Mazes sehoce eee eee 880 Manufactures: os220 2.20.20 Loh eer 880 Social organization: )/leiest, su! wales) be epee 882 Inifeveycle i eh oe el er 883 Recreational activities... 1... eee 884 Religioinesncectsccss.-k. 5. lee eee 885 Mythology: 22 cues2 2c. Aer bee See 886 Bibliography. 2b sees oul ee eee 886 The Chibeha; by vAs ln Wsroeber: 2 22 .2.2---2..L_. 2 eee 887 Tnatroditietiomi i See N ee tbe hl a 887 Habitat /and politieal divisions: —.......... site nt once Soe iphlore Lure iptgn 930 UCC Gat a i et ee a ee 931 Socialand political organization........... e229. ...- 931 Wife sey Cleter Sse Bet J hate je win 5 tet See pee Ref Ae 932 Vy E21 a ne a Sey ee eee em 933 Esthetic and recreational activities___._._._..._.__-_-_------- 934 FRCH eI OHUG WG MOR PIC: oi ean ee el Se ee ol oi lala 935 oreandtcarminess. oes . fee _ ok. Sew Le 936 Wievwulturesol, the Audaqul.ooo2 24.6 ot oesace eeu suseotseseke 936 The culture of the Moguex and Paez 220.22 o2le. 2 ele ceee-e 937 Sulbsisteneciactivabics 218200.) Aum moiiere eee 938 ouseszand ivillapesterr e502 ee eee ewe beeen 939 IPG ImeerIMesWOrKSe 2.22202 Bet eh SOA ee es 940 ‘Mesnspertationsuer yeaa tions Team. Wie ape opempenouy 941 Dreshandiormanientstrs (eesti Saline, Ca ee ee 941 ICUS WEE (GUDINES aS ily UE) l 2) ee a 942 Commicrce..sit Coston paltery And ier) swe’ by. 944 Socialvand political organization... 20002-2222 eee e- - 945 Pirerey Cle eaapinm Siiat Stier erat oe cee 947 Vices pre. Mo We Bee ok tN S 0a) 55 em 950 Esthetic and recreational activities_.............__._-------- 951 JEELNIG IE NES ge SNS Toll Fe ate een ee oe Eee ee ens 952 XVI CONTENTS Part 4. The Northern Andes—Continued. The Highland tribes of southern Colombia, by Gregorio Herndndez de Alba—Continued. PAGE ‘he culture OljpeeveaOsceec cess el ee eke ee SE ee 956 Subsistence activities... ¥elaeiine! bis comeing «2s 956 labile tlOns= 232 26.0)8 nae Us eels oa Oe 956 Dresssand, orsiaments ue fe 09 fol oe), eR 2 ees 956 Mianugacturess. 27 cse neces solo Joe ee 957 eocial and political organization. ...22.429" Det tay ee 957 WWANEARON se 52h es oe scat aes 957 Cannibaligmys so. .4 oe sees eee eas Oe 958 Esthetic and recreational activities.........._......--_--_- 958 Sicknessiand/deathyAii02 Jo wales. bes el ae 958 LNG Nite rT 959 BibNO STAN repays ow a 960 The modern Quillacinga, Pasto, and Coaiquer, by Sergio Elfas Ortiz. 961 The. Quillecinga and. Pastozs.... 4 22s6nee5 Jon eee 961 Mericul hureseed Uw oe) Oven bon io sedi bee 961 (IQUSGSeat eee eee emcee ceed ee 961 DWRessenc Goo ee eee scorer Oe 961 Mignufactures and tools..2.2-.2-.--2+...-.... 40a 964 Professions. and. trades... 2..2...+...2.3. Hie 965 Community jife. ooo. 3 ccc ce a seabowsssesse ee 965 Marriage s424s02seceseus donc td ydee ke | 965 Esthetic and recreational activities............_..._.------- 966 Medical practices and beliefs____.__.._.--_---.------------ 966 PhesCoaiquer’ ss. 2. 4sopeatesce cesses sons ee 967 Rresent-habitat. 0 2242224 .2.~S301Y Hos, SOUS eee 967 Culture .222)3 2 5 ih hr ROBBY DOD. ae 968 The Moguex-Coconuco, by Henri Lehmann______-____-_------.--- 969 Habitatiand trbes. =e ahee nessa a OL ee 969 Culttirés.5 4 ee ae ae RIO De See 973 Social and political organization.......27209 na 973 Cannibalism 395552254555 s5s60e5002 25-0 ee 973 Religion: 2.5525552.-T QR ARUib 90 (abr) DEER MAE, ae 974 Bibliography. sGe. Sek heels 2 974 CESS ET Ye echt cet eg a dS Sl a 3h Sta 2 2 975 Bibliography «2.x 2S ooo OE SR OL INGE Site. DEORE 4 ee 979 COONS oR ON ILLUSTRATIONS PLATES meine berivian igh Anges. oo. 2 ke ee ee eee eo ckedewnnan VU + Peruvian Coastal and Highland: deserts... s2sseucselvssel-seel ee Make Nitieaea. Bolivia... 2.2 eos one cece eo Mountain acriculture in. Perg__..... wagesg i sort slequrhslom. ccppal) o: Punanoleber and: Bolivia. 2. 3205-524 ceu ee eee milamas and yictinasiees( ) sete dian A od) Io eayudagetoe Laninalomele The upper. reachesioef the, Montaiia: 22.2ena5_ sie.) .ewed wloosel yeeow mubhewdesert. Central Coast of Peri... --«=—<2.—-.1.5Seteut S24ann2 4 Peruvian valley and: desertaterraine.! sc). 2abentioe sabseee esl . Goes 2.6.) or i atxtay mae eaters mee milean landscapes... red aienteldarer overs) ebremerna bya wlaos Ios . Chilean and Argentine landscapes_______.._._.-..------------------ .. The Northwest Argentine country_.-2.2-.22. 24 s2-2n22+-4ss4L2--- MEI ACLOT EAM LOS CRISS a ee pk oboe peer . Ecuadorean and Colombian landscapes. ____......_..-------------- . Colombian and Ecuadorean landscapes__.__.__.------------------ . Chavin de Hudntar architecture and stone sculpture__........._---- . Coast Chavin artifacts, structures, and sites...._...........-....--- . Gold objects of the Chavin and Paracas Necropolis styles_.__..._._-- felaracas CAvermas, Style DOLLLY.. = eqc.0—- eee nen ee ee . Paracas Necropolis and Cavernas style pottery and Necropolis mummy NGG ne ee ee ee a ang ee Sonmennd halve] yaa! maracas. Necropolis textiles... frre In dees’) fin Vi wcls po. oot Lads Peseta NOLS sviCinity Of PIScO, POM ne nn a SEE Ole eniazeaistyle DOLL CRY. eM ee mien marca sivle potbery<<.2.. 2. ...~meeoetidors baa moljortemen sign . White-on-red, Intermediate, and Interlocking style pottery from the RS nee ya UN ae fe eee An wo . Interlocking, White-on-red, and Intermediate style pottery from the Chancay. Valley 2 ees T here ole gt ate o Pachacamac Interlocking and coarser class of Early Lima style NE MET Ia) on ee Pi een . Finer class of Early Lima style pottery from Nieveria and Pachacamac. . Cultural representations in Mochica and Chimu pottery.__..---_--- / Mochiea and Karly Chimuspotteryiecs. bia =2eeeac eaete sulogtd de . Archeological structures of the North Peruvian Coast__.........---- . Gallinazo Plain and Two-color Negative from Viri Valley cemetery-_- . Recuay pottery and stone sculpture from the Callején de Huaylas---_- . Tiahuanaco, Bolivia, Gateway of the Sun_________._.--_-_-----_-- . Tiahuanaco, Bolivia, Classic Tiahuanaco pottery and carved monolith- miucara simlespottenye sed tneant vate deomtdale hue .dtoaetue . Pucara architectural features and stone sculpture______._...__-___-- . Coast Tiahuanaco-A style from Pachacamac___._._......._-___----- . Coast Tiahuanaco-A and -B style from the Central Coast__.___-_--- 595682—46——_2 XVII XVIII ILLUSTRATIONS PAGE 41. Black-white-red and black ware styles of the North Coast Middle PeTIOOS 5 Fee ee a keane 126 A2. Pottery, irom,a, Lambayeque site... .—....-2 2) 928 ee eee 126 43.. Textiles intheiCoast, dabuanaco style. 22202-5222 eee 126 44. Early Ica, Middle Ica, Late Ica, and Coastal Inca pottery styles___.. 126 45. Middle Ica II style pottery from Ica Valley___.___________-______- 126 46. Maddie and Jate-Coastal Periodttextilest. — 22-2. 2222 126 47. Late Chancay Black-on-white and White styles from the Chancay Walley ttitises ose ee oes anes ane eee 126 48. Great Late Period structures of the Central Coast_________________ 126 49. Late. Chimuw pottery ........1..s-42222.424.2._ See eee 126 50. Late.Chimu,goldwork from Huarmey_..._.J5!) 2 _2u eee eee 126 mt Chanchan,/PerG. 22.2 b ot oe ee ee ee 126 52. Archeological structures of the North Peruvian Coast_____________- 126 53. Mummy bundle from Late Period of the Coast of Peri_____________ 126 ne Chanapata, pottery... 4.02.2 oe ee) Eee ee 126 55. Peruvian weaving equipment of the Late Coastal Periods___________ 126 So. Peruvianewood carving... aoe ee 126 57. Metal tools and ornaments from prehistoric Peri________________-_- 126 58. Metal and stone objects from prehistoric Peri_____________________ 126 59. Miscellaneous Peruvian archeological objects. _________.__._.______- 126 60: Peruyian.whistling jarss......2 ee ee eee 126 ol, Cupisnique pottery ..._...-. 2 ._ eae ei Mele.) Dee ee 158 62. Cupisnique ornaments of shell and bone_.__..____--..-..._...-__-- 158 63.) Cupisnique.pottery..vu i908 atreds pak ahiviontijo1e teh ae 158 6£.14 Cupisnique pottery... 28GB Ae_ Sets _ sn ee 158 65. Carved stone mortars from Pacopampa_._________----_-----.----- 158 Go. Salinar, pottery... 3 de ee a Le Se eee 158 67. (Salinaripottetyao.22 4226 S92 iee ee ees rae.) bee alle 158 68. Salinar Period bonework and ey an Peer nee ee. 158 69. Burial types of the North Coast of Perfi______________.___--__-__-- 158 mot Mochica pottery... seo A _ ne 158 @ht Mochica‘eraftsmanship.._._. 2 3 a eo ee 158 72. Mochica construction and architecture___..._._._.._.._.________-_------ 158 73. Architectural:features at Sacsahuaman./2.02_ 20 SL .0 sete) Dose 182 fd Miews, of Sacsahuaman.. 05 se eo ee OE 182 75. Architectural features at Kenco and Ollantaytambo____.__-_--_---- 182 46.1 Ollantaytambo and. Limatambo_.....22s2ce5c005. ee ee 182 (i. Late tncaspottery iael . 34_ seal: _sampens hes. eoideohetnl saa 318 Wer INCA. DEONZe O08 Wainer Ss ee eS el beeen se sae 318 #o.t MiscellancousInca\artifacts2_ ore tet aide Aoki eee 318 80. Inca tunic and painted wooden goblets (queros).._.._-..---------- 318 81. Machu Picchu stone masonry and architecture__________-_-------- 318 $2.1 Two.views of Machu Picchuiuosc. cise 4_ ad) Jo eee eee 318 Se, [Inca,.architectures.2 /_ io! weet oven ology bee eee ee 318 $4. Inca.sitestand iterrainy su. wu) stole eoote noe eae eee 318 85. Pinchollo in the Colca River valley, Peri. Jove. 202222 2-22 382 86.1 AndeantHighland:scenes ci cd wut let Sines!) aioe, pee 382 87. Seventeenth- and eighteenth-century maps of Pert and Bolivia__-_--_- 382 88. Eighteenth-century maps of the Pert-Bolivia-North Chile region_-_--- 382 89. Maps of 1749 showing Audiencias of Lima and Charcas_____-------- 382 90. Highland industries in the Colonial Period____._._._...__-_--------.- 382 On; Colonialvart of Per@. ooo. 0522 occee leew ce kL 382 ILLUSTRATIONS XIX PAGE G2. Atgeciionvora Colonial textiles 220 9 OFO2 IS BY Bi sO BON oe 382 a5». Viodern.Qucchuaapriculires-sos2 2222222 IE Tt ee 8 446 94. Indians of the Peruvian Highlands threshing wheat_______________ 446 95, ‘The wheat: harvest in, the Cuzco areal li o22 22 Loe bees 446 Boe Quechua community -life and: houses<2 2.002209 _85 2UaEL ies een. 446 Sa Oucenuna weaving ay) Blo ONG BO? 20 GNIS OC) TT? Born i 446 eo Oucchuatiypesy Ais NAA BOON 10 BOTs Od JOT! SOIOG Poe 446 Boe rnohlandulestiviale~ x tals Se TIO 1) BOI BOTTING BGO, lo 446 Misa Modern Quechuatrelioion 2 20th: BONN. HOIO ss He? Bt ROL ows 2808 446 pee hullpasinwAymara country Baul 2b Bea Or! Or RTs GStnouass 574 Peed NaN ROT AC VIA UTC elt itis ope ase or tee er ne CT TO EEO Wet 574 fe Amora preparinaGuinoa-BEGG USING ONAL! WES MIATA woe 574 ae MAnmareWouUse: ty PSs S22 Cases fas free 8 ee A SNe 574 fe) Avimaradress.and manufactures js2242 5 tees a= OE OTL Ieee ov4 Pans siaisasion dake liticngat ae. SOT) POURS VOM) pie OTUs say 574 Reade n SAS abt ICA CHaEE DION wp)" be ono on et = OD OTA OTIS 574 Pens ae eA OT ANG ANCES iyi seat be peel 2 BY get i SLY OF 9. VEURL. MOT! VISIO 574 i ASA OR A AGIAN Sn 9f - Where. claos MP tne sot nn PAIS QI ONS 574 Hin?) Scenesamthe Aymara.countryao225. Jos2e e222 ee ES Oe 574 ene eA IA UA ped ag Ab pes en erin br stein OE! OO 574 ne a Asiniata Co gela ditGuns- 21s sale el tere or oe ce PI PATON 574 es Om anc Chipaya houses). 2 OOP UP) A PUR NO, PAOTOD AND. 582 114. Roofing a Chipaya house, Carangas, Bolivia__.__._._..._...._------- 582 if>. Urudishing and gathering totora reeds-. 2-2-2222 22 2. 5982 neo. Unoyand.Chipaya weaving and-spinning! 2 _ SloLeoe etd Seen. 582 ne SUmnangiChipaya Indians .~<<--=---..<=l28299_ Tesh Aoue F209. 582 ata Faye WV OTIA ET ws bial gem Do ge phe ee ni AW OOIUO DIOR BVAUGUE! 582 eee pe hinayansacrilcial- iteso=- 282028189 OLN DBO I DOS sic. BROOK) 582 20.) Chipaya country and ‘Chipaya Indians... 2 282228 L229 228 eee 582 121. Arica I and Arica II pottery from North Chile_-_..___.___._._-_-_- 590 122. Arica I and Arica II pottery from North Chile....____..___._---_-- 590 123. Fishing and hunting equipment, North Chilean Coast______-__-_-- 590 eae. UChild: burial: at»Playa-Miller,.Arieays 22-30 EU 20018 See Se 590 125. Toy replicas of artifacts, Playa Miller, Arica.......--.-.--_-_.--- 590 io.) Lextiles from Playa) Miller; Arica ioe roe eee SILO SOP 590 We7. sPrehistoric artifacts from.North: Chiles. 220 /00i008 DMM Sufi) Baye. 590 128. Prehistoric textiles from the vicinity of Arica__._._____._._._.___--- 590 ee a Hein CHOMOULEE Y= nn nd oe ee ee = eR OL DOR BONAR. 618 130. Miscellaneous Atacamefio manufactures________.__.------------- 618 131. Miscellaneous artifacts from North Chile__..___........_--..------ 618 132. Snuff tubes and tablets, North Chile._._... 2-2-2 eel lle 618 133. Containers, spatulas, and spoons, North Chile____________-_-__--- 618 134.) Mascellaneous artifacts, North, Chilesou- ose. ee SLL 618 135. The Pucaras (forts) of Humahuaca and Tileara___._________--__--- 632 136. Terraces and tombs of the Quebrada de Humahuaca and the Puna__ 632 iS/./ Treatment ofthe dead, Puna de Jujuy. 22222 Lelie lle lle 632 138. Manufactures from Humahuaca and the Puna____________-_---_-- 632 139. Diaguita pottery from the El Elqui Valley, Chile____.______-_-_-- 636 140. Diaguita pottery from the El Elqui Valley, Chile-____._______---- 636 celky LOTS ze TOlC 7 rj a la 638 142. Barreales ceramics and wooden mask from Atajo______---_-------- 638 Paar a mera arhineiS— 2. oo ee eee 638 PMMA SLOUCWOERK 42.2520 52205 525205022 0bo22245--eenen seen ee 638 xx ILLUSTRATIONS PAGE 145. The Pucara of Punta de Balasto, Fort Mendocino, Department of San- ta Maria’ Catamarca: 2... 22.2-...-<2s2eteee eee eee 638 146. Petroglyphs from the Diaguita region.........-....-.-2+-----._=- 638 147. Chaco-Santiaguefia painted funerary urns_...___.--.--..-_-__--_- 660 148. Chaco-Santigeuena: pottery... --. 2.2.24 1222220. .eeeee e 660 149. Ceramics from the Sierras of Cérdoba and San Luis_______________ 686 150. Projectile points from the Sierras of Cérdoba and San Luis________- 686 151. Cérdoba petroglyphs and pictographs___...._._._..-----_-_-_-_-- 686 152. Rock dwelling in the Piedras Blancas, Punilla Valley, Cérdoba____-_ 686 153. Araucanian graves, rewe, and dwelling___...___._-_._.._._---_---- 734 154. Araucanian pottery 2 o2 222. ..2-2.-.---~-...-..e eee 734 155. Araucanian Indians and Araucanian masks_____.___._____-_____-- 734 156, Araucanian Indians...2..2.2........-..+....-2s¢23) 43554 734 I5v. Pottery from ‘Carchi2: 2... ...a.sesseb see bee ee 782 158. Stone sculpture and clay figurines from Manab{f__________________- 782 159; ‘Cerro iNarrio pottery {22..22.....-.22.24-2225%se--2e0)s ee 782 160, Pottery from Azuay Provinee_...................03)5.=.)—eee 782 hol. Otevaloaericultures.2.2..2...2-.6.+.2- cess. = 814 162° Otovalowndiahis U0) oo oo ees eee ee 814 63. Otovalo Indians? 20.2 252.0222 enone eee 814 lea Otovalorscénes! 22 es_ 20 2 sk eet oie 814 165. Ecuadorean Indians and landscapes___._.___._-.-_._--------.--- 814 166. Modern and 18th-century Indians of Ecuador____.______-_-_._-_-- 814 67. OtovalovIndiaris: . 2.2 2. Je eb eee sent et ee tyes bye eee 168. Ecuadorean processions and markets____._.___..___-.-_-_-------- 814 G9, Wpper Cauca River pottery ...2..-......-sesdsch see eee 846 170. Quimbaya gold objects. =-._.-......--+-.--<4+2:seeee eee 846 171. Quimbaya, Sint, and Mosquito ceramics.____._...._._._.__---.----- 846 172. Chibcha and other artifacts... 222246 2-nes2t) Jone gate eee 846 173. A mound and temple at San Agustin_______._._.......-------_..-- 846 174. A carved stream bed at Lavapatas, San Agustin____________.___--_ 846 175. Tierradentro tomb,and. ceramics ...4- 222. 222esc eed eect eee 846 176. San Agustin stone carving .....- 22224. 2s0Shh eG. eee 846 177, Cagaba: villagess. 2) 22. Sect tes eee EL Lee le 878 178. Cagaba houses s/o he ee el A et eee ee 878 179. CAgaba trails and structures.___.__- -dawcidocssd hls 878 180. CAgaba industries at San) ae Ss — eS oo og ns} so [ a of oD Oo aE aS ene jae ) 1a, ty Museum, Philadelph Sl Puatse 4.—Mountain agriculture in Pera. used, in Colea Valley. (Courtesy James Sawders.) Bottom: Valley floor culti- Top: Ancient terraces, some still vation. (Courtesy Truman Bailey.) { Piate 5.—Puna of Pert and Bolivia. Top: Typical Peruvian high plateau. (Courtesy Truman Bailey.) Center: Alpacas feeding, Bolivia. (Courtesy Grace Line.) Bottom: Near Uncia, Bolivia. (Courtesy W. E. Rudolph.) Puate 6.—Llamas and vicufas. Top: A Highland scene. (Courtesy American Museum of Natural History.) Bottom (left): Llama with ear tassels. (Cour- tesy Grace Line.) Bottom (right): A young vicufia. (Courtesy American Museum of Natural History.) PLatEe 7.—The upper reaches of the Mentana. Top: Southeast corner of the Castillo at the famous archeological site of Chavin de Hudntar, Peri. (Courtesy Amer- ican Museum of Natural History.) Bottom: In the Urubamba drain- age, Junin, Peri. (Courtesy Tru- man Bailey.) Puate 8.—The desert Central Coast of Peri. Top: The bay and town of Ancén. Bottom: Rocky cliffs and sand dunes near Huarmey. (After Johnson, 1930.) Puate 9.—Peruvian valley and desert terrain. Top: Ancient site of Pacha- camac situated on arid hills at edge of fertile Lurin Valley. (Courtesy James Sawders.) Bottom: Majes Valley with desert hills in background. (After Johnson, 1930.) Puate 10.—Chilean landscapes. Bottom: An Araucaria pine and the Lanin Volcano, Central Chile. Grace Line.) Top: View along the Rio Itaba, Central Chile. (Courtesy Le, tif, Yes | | PuateE 11.—Chilean landscapes. Top: Osorno Volcano, Lake country. (Cour- ' tesy Julian H. Steward.) Bottom: The Juncal Valley leading over the Andes. | (Courtesy National Geographic Magazine.) if Puatse 12.—Chilean and Argentine landscapes. Top: Modern terraced agri- culture along Rio Caspana, North Chile. (Courtesy W. E. Rudolph.) Center: View from near the Cuesta Obisbo, Diaguita country. Bottom: Calchaqui Valley and La Poma. Irrigated valley below and dry upper slopes in foreground. (After Bowman, 1924.) ) (e) i=) fe) HH a aay Ce s 3 Se Ee oe Cr ay Qs SEA St i oS na) 2 c =| (=) 5 —" © iq os 1=O©) — a 2 ie ce — > oo ede a 6 § As o 7a." eS . | On a= oS TQ Puatse 13 Puate 14.—Ecuadorean landscapes. Top: Cafar Valley. Bottom: A 13,000-foot- high basin in South Ecuador. (Courtesy Chicago Natural History Museum.) Puate 15,—Ecuadorean and Colombian landscapes. Top: The Highlands near Mount Chimborazo. (Courtesy James Sawders.) Bottom: View of part of Bogota. (Courtesy Wilson Popenoe and the National Geographic Magazine.) Puiate 16. Colombian and Ecuadorean landscapes. Top: The approach to Cali, Colombia, in the Cauca Valley. (Courtesy Frank M. Chapman and the National Geographic Magazine.) Bottom (l The paramo of the Cordillera Central, Colombia. (After Rich, 1942, fig. 313.) Bottom (right): Modern mountain agriculture near Quito, Ecuador. (After Rich, 1942, fig. 298.) VoL. 2] ANDEAN HIGHLANDS—BENNETT 31 villages which specialize in making ceramics to be distributed via the system of local markets. A few Indians continue to work. with metals although the old skills are gone. Most of the objects once manufac- tured by the Indians have been replaced by superior ones of iron and steel. Weaving is still extensively done by the Indians. The girdle-back loom and the simple implements for spinning and weaving are still in use. In some places, an upright treadle loom is now employed by the Indians. As arule, women now weave on girdle-back and simple- belt looms, and men weave on the treadle looms. Most of the weaving done today is for the utilitarian garments used by the local Indians. However, there are individuals and localities which still specialize in finer weaving for the urban market. The products, however, are limited in techniques and rarely achieve a skill or fineness comparable with the pre-Spanish periods. | CLOTHING AND ORNAMENTS Although details of dress reflected regional and time differences, the standard clothing pattern remained the same throughout the archeo- logical periods of the Central Andes. The basic male costume con- sisted of a breechclout, a wrap-around skirt, a shirt with or without sleeves, and a shawl. The women wore a wrap-around single-piece dress, secured at the waist by a belt, and a mantle over the shoulders. Both men and women wore sandals and some kind of headgear. A woven bag with a shoulder strap was a standard part of the costume. Class distinctions in costume are clearly depicted in the Karly Mochica Period ceramics. Early Period graves on the South Coast differ in the elaborateness of clothing which accompanies the burials. Costume also distinguished certain occupational groups, such as war- riors, priests, and runners. In all periods, special costumes were made for dancers. Mortuary clothing was not only distinctive but also elaborate. A great variety of ornaments and methods of body decoration were employed in the Early Periods. Tattooing, scarification, and face painting were practiced. Skulls were artificially deformed. THar- plugs, nose plugs, and lip plugs were common, and ornamental facial masks were worn. ‘The artifacts discovered in excavating Early sites include many types of beads, necklaces, anklets, rings, collars, plumes, and metal or shell pincers for beard plucking. Although some of the scarification depicted on the modeled pottery vessels may have been ornamental, there is no indication of the leg or arm ligatures so typical of the Amazon. The Colonial Spanish Period introduced many changes in costume. As part of the process of reducing all Indians to one class level, dis- tinctions in clothing were eradicated. The Indians were forced to 595682—46——_5 4 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 adopt new types of clothing in an effort to make them forget their cul- tural unity. Likewise, many of the Indians voluntarily imitated the clothing of the Spaniards. The result was the creation of a new stand- ard costume for Indians, consisting of pants, jacket, shirt, ponchos (introduced in the post-Spanish time), and hats. At the present time, the Indians still use this 16th-century costume. The men wear sandals, short pants, shirts, vestlike jackets, and the sleeveless poncho. They still carry a woven or a leather bag. Head- gear varies enormously, and the Aymara wear both a knit stocking cap and a felt hat. Women wear several skirts, a belt, a blouse, a blanket, and a felt hat. Regional distinctions are still reflected in the details of the costume, and special dance costumes are still used. MUSIC Practically all the musical instruments used by the Jnca are archeo- logically old. From the Early Periods onward are found drums and tambourines; end flutes of clay, reed, or bone; gongs and clappers; seed rattles; straight and coiled trumpets of clay; shell trumpets; numerous pipes and whistles; and panpipes of reed or clay. In Colonial Spanish Periods, harps, violins, and guitars were added to the musical equipment. These, together with most of the ancient instruments, are still used by the mountain Indians. SOCIAL CULTURE Archeological evidence for aspects of social culture is meager and difficult to interpret. Although the /nca customs are comparatively well known from Spanish accounts, it is hard to determine whether these were characteristic of the Central Andean pattern as a whole. The few statements that can be made are based largely on inference from indirect evidence. Some sketchy information is available on the customs pertaining to the life cycle. Early Mochica Period ceramics depict mothers giving birth to children, assisted by a midwife and sometimes a second person as well. There is a suggestion that a medicine man was employed. Inca infants were placed in cradles almost immediately after birth, and cradleboards are depicted in the Middle Period ceramics on the coast. Again in Jnca times, there was a naming ceremony for a child about 2or3yearsold. At this ceremony the child’s hair was tied in bunches, and each participant in the ceremony cut off a lock and left a gift for the child. Although there is no direct evidence for the antiquity of this custom, it has a wide distribution from the Chibcha to the Calcha- qui and is probably old. The Jnca emphasized age grades, formal education at least for the upper classes, and puberty ceremonies. No convincing data are available on the antiquity of any of these. Had any great importance been placed on puberty ceremonies, the pictorial VoL. 2] ANDEAN HIGHLANDS—BENNETT 33 Mochica ceramics might be expected to furnish evidence thereof, but actually the ceramic design and modeling depict but few scenes that have anything to do with children. Information on marriage customs in pre-/nca times is likewise lacking. Burial practices are, of course, well known. In every period, the graves indicate an elaborate concern about the disposal of the dead. Graves are generally well made. The bodies were prepared especially for burial, and commonly they were elaborately wrapped in specially woven cloth. Grave goods were of good quality and quantity, al- though this depended in part on the importance of the deceased indi- vidual. Scenes on Early Period ceramics show funeral dances in which the dancers are represented as skeletons. The localized endogamous ayllu was a basic unit in Jnca social or- ganization. The patrilineal extended family and ayllu-like local group are undoubtedly old, but it is difficult to prove this archeolog- ically. However, standard types of headdress and face painting de- picted in the Mochica Periods might possibly represent symbols of ayllu affiliation, and the small village sites might suggest organized local groups. The /nca town of Machu Picchu had residence units for each ayllu, and in the remains of Late Period cities similar unit divi- sions are found. The /nca moiety is also of considerable antiquity in all probability, in spite of the lack of confirming evidence. Class distinctions apparently date back to the Early Periods, if one can so infer from ceramic designs. For example, one design shows an elaborately dressed figure carried in a litter by servants. Another is a banquet scene in which a prominent individual is being served. De- tails on many of the portrait jars can be interpreted as emblems of rank. It is not certain, however, that such class distinctions ever crystallized into a caste system in pre-/nca Periods. The elaborate mathematical political organization of Inca society was probably a recent development. In previous periods, chiefs, war chiefs, and others of high rank are represented, but there is little evi- dence of the extension of a political control over large areas. For ex- ample, Mochica ceramics of the Early Periods are found in abundance in the three valleys of Chicama, Moche, and Vird, but it is not certain that these valleys were united under a single ruler. Instead, the numerous battle scenes represent opposing warriors of about the same dress, and suggest considerable conflict within the Mochica culture. However, many scenes on Mochica vessels depict forms of punish- ment, implying an organized government and a legal system. The size of building units in all periods is proof that large numbers of people were effectively organized. In general, there seems to have been a gradual development of political systems, culminating in the Inca Empire. Previous to this, local groups were led by war chiefs. Then several groups might be united in a loose confederacy. The 34 SOUTH AMERICAN INDIANS [B. A. E. Buty, 143 Inca added their own caste to the political superstructure, as well as formalizing the government organization. 3 Many changes in the Jnca social culture took place during the Colo- nial Spanish Periods. Although the simple naming ceremony was not disturbed, baptism was added. Jnca age grades lost their signifi- cance, and the Inca schools for the upper caste were no longer main- tained. Marriage, insofar as possible, was controlled by the Church, which also took part in the burial services, although not to the total elimination of Indian customs. However, under the new economic regime, the Indians were unable to maintain the elaborateness of burial practiced before. Mummy worship was eliminated in the Early Colonial Quechua Period. The immediate effect of the Conquest Period was one of leveling the class distinctions of the Inca. Whether the Indians submitted to the Spaniards or joined the separatists, the leveling process was the same. The ayllu continued within the encomienda system, since it was a convenient unit to tax. Indian villages still had native chiefs, but since these chiefs also became the tax collectors, their interests often differed from the rest of the group. The majority of Indians were reduced to hatunrunus, or tribute payers, or yanaconas, the wandering peon class, who escaped tax payment. In effect the com- plicated Jnca class structure and political system broke down into a simple pattern of foreman and workers. In the Early Colonial Period, the corregimiento, or crown control, replaced the encomienda. Indian self-government was replaced by Spanish officials. The ayllu community continued as a convenient source of mita labor for the mines. Today the hacienda has replaced the corregimiento. The present- day Indians live either in independent communities subject to tax- ation, or on plantations as peon laborers. The ayllu, as a loosely organized village unit, continues. The simplified Indian customs associated with the life cycle continue, although the Church has for- malized the rituals of baptism, marriage, and burial. Indian chiefs of local groups are still recognized, and there are some who claim wider authority, although in reality the power of any chief is definitely limited. Wealth distinctions and differentiated age groups are still recognized, but class distinctions are not formalized nor sharp. RELIGION Formalized religious organization and elaborated practices are char- acteristic of the Central Andean cultures. The Inca established their sun-worship religion throughout the Empire, but local religious cults were not entirely eliminated. In pre-Jnca times, the number and variety of cults must have been enormous, and in all periods religious activities played a prominent role in the lives and energies of the VoL, 2] ANDEAN HIGHLANDS—BENNETT 35 people. The imposing ruins of Chavin and Tiahuanaco, as well as many of the large Coast pyramids, were probably built for religious purposes, and much time, manpower, and organization were required for their construction. Both priests and medicine men were prom- inent in religious activities. Early Mochica Period designs show medicine men massaging a patient or sucking out disease objects. In the same period, priests are represented, presumably as the for- malized leaders of religious cults. In /nca times, the priests were well organized and prominent in the ceremonies. | The elaborately prepared burials imply that ancestor worship was an important part of the religion at all times. For example, a con- siderable proportion of the time and energy of the Early Paracas peoples was devoted to the weaving of cloth for the mummy bundles. During Inca ceremonies, the decorated mummies of important indi- viduals were frequently paraded. Several stages of wrapping in a Paracas bundle, alternating decorated and plain fabrics, would sug- gest that this custom was old. Anthropomorphized and mythological beings are common in Early Period ceramic design. Everywhere, the feline, serpent, and condor divinities are prominent. Scenes of ceremonies are also depicted. In Inca times, ceremonies were associated with the agricultural calendar, and presumably this was also true in the more remote past. During the Conquest Period, the major changes in Indian society were economic and political rather than religious, in striking contrast to México. The pacified Quechua continued their religious practices with little interruption, and missionary activity was directed at the separatist group. During the Early Colonial Period, however, an active campaign to eradicate idolatry was started. The sacred mum- mies and idols were destroyed, and the remnants of the Inca priest eroup were restrained from practicing. The campaign was successful to the extent that by about 1650 the Church considered that the Que- chua had accepted Christianity. In the following periods, the Indian religious beliefs and practices that did not conflict with the Church doctrines were tolerated as superstitions. The Indians turned to the Church for prestige outlets. Today, Indian religion is a fusion of the old beliefs and practices with the Catholic. Ceremonies which once followed the agricultural cycles now merge with the Catholic ceremonial calendar. Small home ceremonies are still performed, and the number of minor magical be- liefs and practices is enormous. Sorcerers, diviners, herb specialists, and many other categories of medicine men are still popular. © KNOWLEDGE True writing and a recorded calendar were unknown in the Andean region. However, mathematical organization was characteristic of 36 SOUTH AMERICAN INDIANS [B. A. E. BULL, 143 the Inca Periods, and many records were kept by specialists on quipus, or knotted-string devices. Some writers believe that the quipus may also have been used for recording the days and the years. The antiq- uity of the quipu is not definitely known, but one Early Mochica Pe- riod vessel shows a runner carrying a comblike object which may rep- resent a string quipu. In any case, Larco Hoyle has pointed out that these running messenger figures on Mochica vessels carry bags which contained designed beans. These were seemingly used either as counters or as some kind of ideographic writing. ‘Today the quipu is exceedingly rare. Trephining and artificial skull deformation was practiced in Early and Late Periods. The use of herb medicines was prominent in Inca times and probably far earlier. Trephining and deformation are no longer practiced. Herb medicines are, however, prominent in every market of the Highland region of Perf and Bolivia. SUMMARY From this limited review, it is clear that the basic elements of the Central Andean culture pattern were largely established by the early archeological periods. Although it is not possible to assign any accu- rate dates, it would not seem unreasonable to say that by A. D. 500 the basis of this pattern was formulated. In about one thousand years of development up to the time of the Conquest, there is amazingly little evidence of technological advancement. Certain technological im- provements in metallurgy are noted, but the ceramics, weaving, archi- tecture, and carving illustrate a change in emphasis rather than in technology. The most significant changes occur in the fields of social and political organization. There is considerable evidence of a steady increase in population from the Chavin and Early Periods up to the Inca Period. With the advancing control of agricultural production and the increase in cultivable land by irrigation, a larger population could be supported even though less actual time was consumed on subsistence activities. In the Early Periods, such leisure time seems to have been devoted to ancestor worship and to the preparation of goods for burial. Later, this control of economy permitted the sup- port of a political superstructure. The Inca crystallized this development of political organization. They established themselves as a ruling class over a wide area in a series of successful military campaigns. The Inca system was not directed toward exploiting tribute from the conquered peoples, but rather was an attempt to build up a well-integrated economic organization. It is difficult to evaluate the total Inca achievement, but it was certainly successful in terms of organization of a large mass of population. Other types of trends might be studied, particularly as more detailed evidence becomes available for the various time periods. Kubler (see Vou, 2] ANDEAN HIGHLANDS—BENNETT 37 this volume, pp. 331-410) suggests many possibilities in his discussion of the historical periods, but many are not confirmable in the limited framework of archeological evidence. A careful study of settlement pattern and village type might be feasible. Likewise, one could analyze the dominance of various centers throughout the total history and consider the factors which have caused shifts in these centers. The over-all transportation pattern, land tenure and Jand use, leader- ship pattern and prestige, are other suggestions. Population shift likewise is of interest, in terms of colonization (mitimaes), forced withdrawal into the mountains or to the east, spread of archeological styles, and the mobility of vagrant groups following the Conquest. THE SOUTHERN ANDES The Southern Andes include those peoples who occupied the moun- tain region of Northwest Argentina, North Chile, and Central Chile south to the Island of Chiloé. The tribes farther to the south in Chile are included in volume 1 of the Handbook, as they are hunting and gathering peoples. The Central Andean region presents a com- paratively uniform pattern of culture throughout its history, but such uniformity is less evident in the Southern Andes. Cultures such as the Diaguita, the Atacameno, and others, which were directly adjacent to the Central Andes, reflect considerable influence from the Central Andean culture. The Araucanans, who were farther away, have a more independent pattern. The period of Jnca occupation undoubt- edly obscured many differences which had previously existed in the Southern Andes, but distinctive local orientation and cultural empha- sis is none the less recognizable. The tribes of Northwest Argentina and North Chile share many cultural similarities. Among such groups as the Diaguita, the Ataca- meno, the Humahuaca (Omaguaca), the Chicha, the Lule, and the Churajon of Arequipa, the same general cultural orientation is seen. All are agriculturists with about the same technical equipment, and all emphasize herding. The stone masonry, house types, village pat- tern, storage bins, and forts are closely similar throughout. Metal artifacts are almost identical in each group. Wooden artifacts, such as snuff tablets and tubes and many others, are shared by all. The pattern of warfare and the typical weapons are everywhere the same. Coiled basketry, weaving, and leatherwork do not differ greatly. Pottery, although differing in details of shape, is similar in general type of geometric design. The simple political and social organization, judging by limited historical knowledge, varied but little from one group to another. The inclusion of the Araucanians with the tribes of Northwest Argentina and North Chile is little more than a convenience based on 38 SOUTH AMERICAN INDIANS [B. A. B. Buty. 143 their southern geographical location. Culturally, the Araucanians are easily distinguished from other tribes of the Southern Andes, particularly in pre-Jnca times. All tribes of this general area, how- ever, show the effects of numerous contacts with the hunters of the Argentine Pampas, especially in their emphasis on hunting, their extensive use of the bow and arrow, and their loose political organi- zation. In the following description of the terrain and the cultural patterns, the Araucanians are kept distinct from the other tribes in the division devoted to the Southern Andes. See this volume, pp. 687-760. THE DIAGUITA-ATACAMENO THE ENVIRONMENT The mountain system of the Eastern Cordillera of Bolivia extends into the northwestern part of Argentina. The physical features are much the same, such as the high punas, dry intermont basins, and high peaks which rise above the snow line. Although much of the region is arid, some of the basins have streams which supply sufficient water for agriculture, at least with irrigation, and are surrounded by a puna suitable for grazing. The Andean pattern of intensive agri- culture was well established in these sections. Elsewhere in Argen- tina, the Andean peoples made little headway against the fierce nomadic tribes of the Chaco, the Pampas, and Patagonia. The grass- covered Pampas, today the richest agricultural and grazing district of South America, were not adaptable to cultivation with Indian techniques. In the Highlands, the principal centers of population concentration were, from north to south: Jujuy and Salta, Tucumd4n, Santiago del Estero, Catamarca, La Rioja, San Juan, Mendoza, and Cérdoba. Archeological finds demonstrate that all these centers were occu- pied by reasonably advanced cultures, typified by villages of rough stone construction, agriculture, ceramics of high quality, skillful met- allurgy, and political unification. These Indian cultures have long since disappeared, but historical references list the location of some of the tribes. The basins of Jujuy and Salta were occupied by the Atacameiio, who had also spread across the Puna of Atacama into North Chile. To the south, in the basins of Tucumdén, Catamarca, San Juan, and La Rioja, were the Diaguita (or Calchaqui), and the archeological evidence points to a much wider distribution of these tribes. The Lule lived around Santiago del Estero, and the Come- chigén occupied the basins of Cérdoba and San Luis. The Huarpe, an Araucanian-speaking group, migrated from Chile to occupy the Mendoza region or were Araucanized through influence from Chile. (See Handbook, vol. 1, p. 169.) In North Chile, the Atacama Desert extends roughly from the Peruvian border to the Rio Copiapé. The Atacama, the most arid desert in all of South America, is crossed only by the Rio Loa, and Vou, 2] ANDEAN HIGHLANDS—BENNETT 39 this river supports but one important oasis, around the town of Calama. The Atacama Desert was not only unsuitable for concen- trated settlement but also acted as an effective barrier, cutting off Central Chile from much contact with the Andean region to the north. Small groups of fishers, like the Chango, were able to exist along the rugged Coast, and the Atacamefo occupied the oasis of Calama and utilized the neeRberae punas for grazing. ’ In the States of Atacama and Coquimbo, between the Atacama, Desert and the Central Valley of Chile, larger streams cut their way to the Pacific. The valley flats that were large enough to support agricultural life were occupied by the Diaguita from Northwest Argentina. Northwest mains and North Chile both became part of the Inca Empire. However, influences from the Central Andes are reflected in the archeology long before the Jnca Period. The Tiahuanaco culture influenced and, in some places, temporarily replaced the local cultures. Western civilization has now Magely replaced or absorbed the Indian cultures in all of this region. New uses have been found for the land, such as cattle grazing, stock breeding, the growing of sugarcane and vineyards in Argentina, and the mining of copper and the extraction of nitrates in North Chile. THE DIAGUITA The Diaguta (or Calchaqui) occupied the States of Tucuma4n, Catamarca, La Rioja, Santiago del Estero, Salta, and San Juan in Northwest Argentina, and the States of Atacama and Coquimbo in Chile. Their language was called Kakan and differed from both Quechua and Aymara. The antiquity of Diaguita culture has not been finally determined, although it. was definitely pre-Jnca and con- temporaneous in part with Highland Tiahuanaco. Many of the characteristics of the Central Andean cultures are found. The sub- sistence was basically agricultural. The cultivated plants were similar to those of the Central Andes, and religious ceremonies were again associated with the agricultural cycle. Granaries, built like small houses, were numerous. However, terracing was not common, and it is doubtful that agricultural control was as complete as it was to the north. Hunting was common, and gathering also, particu- larly of algarrobo beans, from which a mildly intoxicating beer was made. Coca was chewed and tobacco and pariaca were used as snuff. Herding of domesticated llamas and alpacas was of great importance. | | The Diaguita lived in small irregular villages, marked by individual house units and streets. Some are large enough to be called cities. The houses were built of adobe or rough stones, sometimes set in mud, but no use was made of dressed stone. The walls were supplemented 40 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 by poles, and the roofing was of poles and thatch. Although some of the houses are of reasonably large size and may have been used as community centers, the large public edifices, so characteristic of the Central Andean pattern, are not found. Forts were built at strategic points, for the Diaguita were warlike. The spear thrower was used, but the bow and arrow was the prin- cipal weapon. The stone- and bone-pointed arrows were carried in a quiver. A bronze hand weapon with a point called a knuckle duster and the stone-headed club were other typical weapons. Bolas were also common, perhaps because of greater contact with the hunting tribes of Patagonia and the Pampas. Diaguita ceramics are of good quality and differ in shapes and designs from Peruvian and Bolivian types. ‘Two basic ceramic styles, the Santa Maria and the Draconian, have mutually exclusive distri- bution and suggest a basis for dividing Diaguita culture. Little use was made of gold and silver, but copper and bronze objects were numerous even in the pre-Jnca days. Many of the metal artifacts are of known Jnca types, such as axes, knives, chisels, hoes, and club heads. However, distinctive metal objects were also made: bronze bells, relief-decorated breastplates, scepters, socketed bronze axes, pestles, and knuckle dusters. Weaving was well developed, and the fabricated articles of clothing follow the standard Highland pattern. Coiled basketry and numerous wooden articles were also made. The little that is known of Diagwita social culture suggests a simpli- fied Central Andean pattern. The naming ceremony for infants was accompanied by cutting off locks of hair. Sororal polygamy and the levirate are reported. Burial was important and the graves contain offerings of many valuable objects. However, cemeteries which con- tain only child burials in urns are a distinctive characteristic. Some adult urn burials are also found, but direct burial in circular, stone- lined graves is more common. Urn burial may represent an influence from the Amazonian region, since it is not an Andean characteristic. Differences in richness of burial suggest that class distinctions existed, but there is no evidence for a rigid caste system. Permanent large- scale political organization was lacking. However, the chiefs of small units often united in alliances, some of which became quite powerful. Early Spanish accounts describe special rain ceremonies and others associated with the agricultural seasons in which ceremonial drunken- ness was a feature. These ceremonies were led by individuals desig- nated as priests, but the descriptions do not suggest a well-organized priesthood. The time of the Diaguita expansion into Chile is uncertain. In gen- eral, the Chilean Diaguita differed only in details of culture. The Inca mitima system eliminated most of this culture even before the Con- quest. VoL. 2] ANDEAN HIGHLANDS—BENNETT 4] THE ATACAMENO The Atacamefio occupied the desert of North Chile, the Puna of Atacama, and the Northwest Argentina region of Salta and Jujuy, although they had been displaced by the Draguita in this latter area in pre-Inca times. They spoke a distinct language, called Kunza. Little is known about Atacamefio culture except through archeology. The archeology of the Coastal section of North Chile (Bird, this volume, pp. 587-594) presents evidence of two Preagricultural Periods and one Agricultural Pottery Period, all of which antedated the typi- cal Atacamefio. The First Preagricultural Period, found in middens at Arica, Pisagua, and Taltal, is characterized by such fishing gear as shell hooks, composite hooks of weights and bone barbs, and harpoons with detachable forepieces. Percussion-flaked tools, stone side scrapers, and mortarlike lava bowls are also found. Burials are extended on and covered by reed mats, and particular attention was paid to the prep- aration of mummified infants. The lava bowls and percussion-flaked tools carry over into the Second Preagricultural Period, which is charac- terized by triangular points with concave bases, and some barbed and stemmed points. Fishhooks are of thorn, and the harpoons have bone forepieces. Bolas and spear throwers are found. The burials are flexed, folded in rush mats, and covered with bird- or guanaco-skin robes. In the First Agricultural Period, coiled basketry, plain pot- tery, weaving, and copper appear simultaneously. Maize, calabash, and cotton appear first, and beans may be later. The burials are flexed, rolled in rush mats, and the graves are marked by posts. Al- though certain features of this First Agricultural Period suggest the later Atacamefo, it seems more likely that the ultimate origin of this culture will be found elsewhere than the Coastal region. Isolated groups of Atacamefo survived in North Chile and on the Puna of Atacama into Spanish times, but, in spite of the evidence from the Chilean Coast, the culture has a respectable antiquity in the inte- rior. In Northwest Argentina, Afacamefio is as old, if not older, than Diaguita, and in the Calama region definite influence of Tiahuanaco culture is clear. The Atacameno appear to have been great traders, and are some- times called the middlemen of this region of the Andes. Their culture was eclectic, and at various time periods incorporated elements from the Coast, from the Tiahuanaco, from the Diaguita, from the so-called “Chincha” culture of southern Perf, and from the Inca. In general, the Atacamefio culture reflects the Central Andean pat- tern, although in many places the limitations of the terrain restricted its development. Agriculture was practiced where possible, and llama herding was important. Nowhere is there any great elaboration of culture. Since trade with the more advanced neighbors was com- mon, Atacamefio needs and crafts were modest. The ceramics are 42 SOUTH AMERICAN INDIANS [B. A. E. BULL. 143 simple and largely utilitarian. How much of their metalwork was manufactured locally is hard to say. Artifacts of wood and bone are the most characteristic. Among the typical wooden objects are tog- gles for llama harnesses, carved snuffing tubes and tablets, spoons, knives, and daggers which may have been used as agricultural imple- ments, weave swords, spindle shafts and whorls, carved idols, cups or goblets, boxes with covers, combs, tubes, drums, and flat slabs which are sometimes leather-covered and designated today as“ prayer books.” Bone pins and spatulas and small spoons were decorated with circle designs, and common bone counters were probably used in some game. The principal weapon was the bow and arrow. The arrows had a wooden foreshaft and an attached point. The bola and the spear thrower have not been found. There is no evidence of large public works or of strong political or social organization. In general, the position of the Atacamefio is intermediate between the Central Andean pattern and Diagutta. | THE ARAUCANIANS THE ENVIRONMENT The rift valley of Central Chile is the heart of the contemporary country and includes two environmental zones. Northern Middle Chile extends from the Rio Aconcagua, north of Valparaiso, to the Rio Bio-Bio. Here the valley is flanked on the east by the high Andes and separated from the Pacific Ocean by a low Coastal range. The climate is temperate and the soil fertile. The region was once occupied by tribes of Araucanian-speaking Indians. Later, the Inca invaded the area and extended their Empire as far south as the Rio Maule, but their political control was not very secure, and they were constantly struggling with the independent Araucanians to the south. However, Jnca influence was great enough to create a marked cultural difference between the Araucanians of northern Middle Chile and those of southern Middle Chile. The Spaniards were at- tracted to the temperate valley and soon replaced the Indian culture. Southern Middle Chile extends from the Rio Bio-Bio to the Island of Chiloé. The valley is of the same formation as in the north, but the climate is decidedly wet, resulting in a heavy rain-forest cover- age. This region was long occupied by the wilder tribes of Arau- canian Indians, who successfully blocked Jnca invasion and held off the Spaniards for several centuries. Today it is still the homeland of. the surviving Indians. Further south, the long Chilean Archipelago was occupied by primitive fishing and gathering groups which show little if any influence of the Andean pattern. THE CULTURE The numerous Araucanian-speaking peoples, who once occupied most of the Central Chile Valley and who live today south of the Rio Vow. 2] ANDEAN HIGHLANDS—BENNETT 43 Bio-Bfo, are united linguistically but apparently never formed a political, physical, or cultural unit. Instead, the Araucanians were composed of more or less independent tribes, such as the Picunche, Mapuche, Huilliche, Pehuenche, and Moluche. A summary of the salient features of the Araucanian pattern is difficult because much depends on the time period selected. The Araucanian groups. have been decidedly eclectic, acquirmg much from the Inca and perhaps pre-Inca peoples, changing radically after contact with the Spaniards, and coutinually absorbing new elements of Western civilization today. In Central Chile there are several passes through the Andes which permitted migration and contact with Argentina. The Araucanian tribes were once quite similar to the nomadic hunters of Patagonia. They depended largely on hunting, fishing, and gathering for subsist- ence; used skin clothing; lived in skin tents or shelters; and were organized on a hunting basis. Undoubtedly, some influences from the Central Andes reached them in pre-Jnca times. However, many of the cultural features by which the Araucanians are classified in the Highland pattern were firmly established only after the Inca conquest. The Jnca actually conquered some of the Araucanian tribes; others withdrew southward but still received considerable in- fluence. The resulting culture at the time of the Spanish Conquest was a mixture of the Archipelago or Patagonian with the Central Andean. In general, the groups in the north were closer to the An- dean pattern, those in the rain forests of southern Central Chile closer to the hunting-fishing pattern. Most of the Andean domesticated plants were known at the time of the Conquest. Although both men and women participated in agricultural activities, much of the work was left to the women. Planting and harvesting were, however, communal labor. Cultiva- tion was done with weighted digging sticks, stone shovels, and rakes. In the north some irrigation may have been practiced, while in the southern forests the slash-and-burn techniques were employed and there was no use of terraces or fertilizers. Llamas were introduced by the Inca. Fishing and the gathering of pifion nuts, numerous wild roots, berries, and greens were always important subsistence activi- ties. Coca was chewed, and tobacco was smoked in I-shaped pipes. The weapons followed the hunting pattern: bows and arrows, bolas, spears, flat clubs, slings, and bone daggers. The warriors used skin armor, shields, and helmets. The Araucanians were always fierce and organized fighters who put up stiff resistance first to the Inca and later to the Spaniards. Killing of captives and victory celebrations were standard practices. In contrast to the Central Andean pattern, water transportation was well developed, including dugout canoes, special boats built of planks, and reed balsas. After the horse was introduced, Arauca- nians became skilled riders. 44. SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 Pottery was everywhere simple, but proceeding from north to south it becomes cruder and has less ornamentation. At first only bark and llama wool were available for weaving. Only after the intro- duction of sheep did weaving become truly important. In Colonial and modern times the Araucanians have been noted silversmiths, but metallurgy probably does not predate the Inca conquest. Many artifacts, such as bowls and spoons, were carved of wood. Basketry was well developed, and skin tanning was a leading craft. The Central Andean pattern of dress replaced the earlier skin cloth- ing, and was itself replaced by European clothing. In general, dress was always simple. No footgear was used, and there were few dis- tinctions in class, occupational, or ceremonial garments. However, face and body painting was common. The thatched-roof houses were oval or rectangular in ground plan and built of stone, wattle-and-daub, adobe, or planks. Some were of large size with many occupants. Villages were composed of from three to eight houses arranged in a spread-out pattern. Cooperative labor was a feature in house building, but there were none of the large construction units which characterize the Central Andean pattern. Villages were composed of a number of extended families (kugas), each of which had a head man. The village recognized a chief. Position and inheritance followed patrilineal descent from father to son. Standard marriage was by purchase, but a mock ceremony of bride capture was a part of the ceremony. Important men were polygynous. The political organization was based on the village unit. Each village had its hereditary chief, who was controlled by a council of elders. The villages of a district were also ruled by a higher chief. Finally, the tribe, representing a number of districts, had a bead chief, advised by a council of five district chiefs. Although in times of war some chiefs attained considerable power and influence, class distinc- tions were not sharp, and there was no true caste system. Instead, the total political structure was basically democratic. The religion was strongly animistic. There were no formal priests, but medicine men were prominent. The ceremonies were of the occasional type rather than based on the agricultural cycle. In general, Araucanian crafts and material culture show a close resem- blance to the Central Andean pattern, while the social, political, and religious organization reflects the hunting pattern. Following the Spanish Conquest, many of the Araucanian Indians adopted the horse and became nomadic hunters and raiders. They spread to Argentina, where they caused considerable disturbance for a long period of time. Although pushed out of much of their territory by the Spaniards, they continued to resist up to the year 1887, when they were granted special reservation privileges by the Chilean Government. VoL, 2] ANDEAN HIGHLANDS—BENNETT 45 THE NORTHERN ANDES The Northern Andes include the Highlands and most of the west Coast of Ecuador, and the total Highland section of Colombia. The mountains of Colombia lay beyond the maximum expansion of the Inca Empire, and consequently the cultures were not subjected to the leveling process which obscured the picture of so many native cul- tures outside of the Central Andean area. Instead, many tribes of Colombia, notably the Chibcha, developed distinctive cultures, al- though still definitely part of the Andean continuum. Ecuador lies between the distinctive centers of Colombia and Peri. The period of intense Inca occupation obscured the previous Ecua- dorian patterns, but the archeological remains indicate that Highland Ecuador was as closely linked to Colombia as to Peri. For example, stone building is not a feature of Highland Ecuadorian culture, nor is llama herding. In fact, more significant links with Pert are found on the Ecuadorian Coast, where stone building and stone sculpture are outstanding. Throughout their total history the high civilizations of the Andes must have exercised a profound influence on much of the cultural activity in the South American continent. Many authors have stressed the possible contributions to Andean cultural development by the tribes of the Amazon. In fact, some authors attribute great im- portance to the Arawak in this respect. The botanical evidence that the marginal Amazon was a center of plant domestication gives weight to these considerations. The influence in the reverse is, however, often neglected since the Jnca political Empire did not penetrate deeply into the Tropical Forest nor the temperate plains regions. However, direct trade with Amazonian neighbors must have been an old estab- lished practice in the Central Andean region, and in the more marginal areas these contacts must have been even closer. (Further discussion of cultural relations between the Tropical Forest peoples and the Highlands will be found in the section dealing with the Montafia tribes, in volume 3 of the Handbook.) In the Northern Andes, the cultural links with the Tropical Forest cultures are particularly marked. The northern Ecuadorian Highlands and all of the mountains of Colombia are virtually surrounded by Tropical Forest regions, on the west, north, and east. Agricultural methods and house types in the Highlands are similar to those in the lowlands. In spite of the cultural exchange, however, one cannot speak of a typical Amazonian culture in the Highlands or the reverse. The differences between Ecuador and Colombia are of such magni- tude that each is treated separately in the following analysis. 46 SOUTH AMERICAN INDIANS [B. A. H, Buin. 143 ECUADOR THE ENVIRONMENT The country of Ecuador can be described in terms of three major topographical zones. The west Coast plain, broken by low hills, varies from tropical rain forests in the north to arid conditions in the south. The Highlands are composed of a series of intermont basins flanked on both sides by high mountain ranges. The eastern plains are in the upper Amazon drainage, and the Indian cultures of this section belong properly with the lowlands (the Montafia, Hand- book, vol. 3), rather than the Highland division. Ecuador lies strate- gically between the centers of high civilization in the Central Andes of Pert and the Northern Andes of Colombia. Through both archeo- logical and historical evidence, the Indian cultures of Ecuador show the effect of this intermediate position. Ultimately, large parts of Ecuador were actually incorporated in the Inca Empire. | The two parallel ranges of the Ecuadorian Andes are separated by a rift valley in which lie 10 intermont basins from 7,000 to 10,000 feet (2,100 to 3,000 m.) in altitude, arranged roughly in a row from north to south. In the mountains are some 30 active volcanos which have filled the valley and the basins with a volcanic ash which forms a porous soil and one subject to severe erosion. Although the forest limit is around 10,000 feet (about 3,000 m.) in Ecuador, the intermont basins were originally covered largely by brush. However, the origi- nal vegetation has been greatly changed by centuries of occupation, burning brush for charcoal, and clearing fields for planting. James (1942, pp. 127-29) gives a brief description of the principal intermont basins. The northernmost, Tulcdn, is on the Colombia- Ecuador border, and, although well drained, its altitude of 9,500 feet, (about 2,900 m.) is too high for many crops. The basin of Ibarra has an altitude of 7,000 to 8,000 feet (about 2,100 to 2,400 m.), but it is badly eroded. Some of the deep valley bottoms are but 2,500 feet (about 760 m.) in altitude and are utilized today for sugar- cane and cotton. The Quito Basin, in which the capital of Ecuador is situated, is one of the richest in agriculture, but its neighbor, Lata- cunga, is again dry and badly eroded. Riobamba is filled with a porous ash which makes cultivation difficult. Alausi Basin, in reality a river flat, lies largely below 8,000 feet (about 2,400 m.), and is in- tensively cultivated. The large basin of Cuenca is, like Quito, one of the most productive and most extensively occupied today. The three remaining basins in the south are Ofia, Loja, and Zaruma. All of these are widened flats of river valleys with rich soils. The archeological remains show the importance of many of these basins as centers of cultural development. At the time of the Jnca Empire, many tribal names were recognized in the Highlands of Vou. 2]... ANDEAN HIGHLANDS—BENNETT 47 Ecuador although all may not have been distinct cultural entities. Most authors agree on the five major groups in the Highlands before the Inca conquest: the Pasto, largely in- the Colombian State of Narifio and the basin of Tulc4n; the Cara, centered in the basin of Ibarra and once credited with organizing most of Ecuador into a single Empire; the Panzaleo, or Quito, around the basin of Quito; the Puruhdé, around Riobamba; the Cafari, in the Cuenca Basin; and the Palta, in Loja. The Palta seem linguistically related to the Jivaro and may have originally been part of that upper Amazonian group. The Highland basins are still densely populated, and the Indian percent- age of the population is high. The Coastal plain of Ecuador varies greatly in environment. In the north, like the Pacific Coast of Colombia, it is mostly covered with a tropical jungle, supported by a heavy double rainy season. As previously mentioned, the Indian groups which occupied this region are assigned to the Amazonian rather than to the Andean pattern, although their proximity to the Highlands was not without influence. South of the tropics in the Departments of Esmeraldas, Manabi, and Guayas, the Coastal plain is broken by low hills, and there is but a single rainy season from December to June. The vegetation is described as park land and scrub forest, and today this region is ideal for the raising of cacao. Farther south, are the desert condi- tions so typical of the Coast of Peri. Sparse settlement is possible on the few available oases. Among the tribes said to have occupied the Coastal region from south to north are the Pund, Huancavilca, Manta, and Esmeralda. Scattered remnants of these Indian groups still survive, although for the most part they have been replaced by a Mestizo population. THE HIGHLAND AND COASTAL. CULTURES A picture of the cultural patterns of Ecuador in pre-/nca times is difficult to reconstruct; first, because the period of Jnca occupation, although relatively brief, was intense; and second, because the Spanish chroniclers made little effort to distinguish between the imposed Inca customs already familiar to them and local variants that might have been earlier. Thus knowledge of pre-Jnca Ecuador depends largely on archeological evidence and the occasional statements which can be gleaned from the chroniclers. Large parts of the Highlands have not as yet been systematically explored by the archeologists, but on the basis of present knowledge it is clear that each major region of population concentration has a long and distinct cultural history. Collier (see pp. 769-780) sums up the evidence in terms of five major Highland sections, namely Carchi, Imbabura, Tungurahua-Chimborazo, Azuay-Cafiar, and Loja. At the present time, each of these has from three to seven chronological 595682466 48 SOUTH AMERICAN INDIANS [B. A. E. Bun. 143 periods established with certain authenticity. Although the earliest manifestations may not extend back before the Christian Era, as some have suggested, there is little doubt that future archeological work will establish a respectable antiquity in comparative Andean terms. Each of the archeological divisions presents distinct features, such as the deep-shaft, multiple-burial chambers of Carchi, the burial and dwelling mounds of Imbabura, the rough stone houses of Tungurahua- Chimborazo, and the excellent goldwork of Azuay-Cafiar. Although there were definite overlapping styles and influences from one region to another, such as the widespread Tuncahudn style of ceramics, there is no evidence of great cultural or political uniformity over a wide area. Still, every zone shows a respectable cultural inventory for each of its recognized periods. It is interesting to note the absence of strong Central Andean influence in the pre-/nca archeological periods. In- stead, the affiliations with Colombia seem better established. In general, however, the reconstruction of Highland Ecuador culture from the chroniclers and the archeological remains in the immediate pre-Inca period presents a pattern of the general Central Andean type. Subsistence was based essentially on the cultivation of maize, beans, squash, quinoa, and potatoes. From the deeper valleys many tropical fruits were obtained as well as the condiment aji. Women seem to have played an important part in the agricultural activities, and markets for exchange of food were prominent. The dog and the cuinea pig were the principal domesticated animals, although there is archeological evidence that the llama was known before the Inca conquest. Hunting was of greater importance than in the south, perhaps because of the greater abundance of deer, rabbits, and game birds. The principal weapons were still the spear, spear Jouve slings, and wooden clubs. Houses were arranged in fairly large villages. Most of them were of perishable materials until the Inca introduced stone masonry on a large scale. Pottery and weaving were well advanced and work in gold, copper, and gilded copper equaled the technical skill of the Cen- tral Andes. The clothing follows in general the Inca pattern, although the Cara men wore large cotton wrap-around blankets as a variant. The composition of the villages may have followed an ayllu pattern, although evidence is meager, and the cultivated land seems to have been individually owned and inherited. The local groups had chiefs, and in some cases this office was inherited in the male line. Chiefs of various units kept in touch with each other and frequently joined in loose confederacies. In fact, there is every indication that the High- land tribes were well on their way to state organization and sharper class distinctions. However, the historical account of the widespread Cara Empire is no longer acceptable. VoL, 2] ANDEAN HIGHLANDS—BENNETT 49 The common hair-cutting, naming ceremony for children was fol- lowed. Although the majority of the people were monogamous, the chiefs are said to have had several wives. Considerable attention was paid to death ritual and burial in mounds, deep-shaft tombs, coffins, and rarely urns. The religion emphasized the natural features of the mountain region. Idols of wood and stone are mentioned. Medicine men were prominent, but whether a priest group existed is not clear. On the Ecuadorian Coast, two archeological zones are outstanding. In the north, the Esmeraldas district is noted for its large circular mounds for burials and perhaps as bases of temples. Clay figurines of exceptional modeling are found, as well as a great variety of small metal objects made in gold, platinum, and gilded copper. Farther south, the Manabi region is characterized by an extensive use of stone. House enclosures are numerous, but more outstanding are the stone bas-reliefs, and the unique U-shaped stone seats which rest on carved animal and human figures. The pre-/nca cultures of the Coast are little known from the histori- cal sources. Fishing and hunting seem to have been of prominence in the subsistence, and some even suggest that the agriculture was left entirely to the women. Clothing was scanty in comparison to the Highlands. Houses were made of perishable materials. The Coastal groups were united because of the coastwise shipping, and some developed ocean travel and trade to a high degree. During the second half of the 15th century, the Jnca spread their control over both Coast and Highland Ecuador. The campaign was started by the Inca Emperor Tupac Yupanqui, but was not formally completed until the time of Huayna Capac, about A. D. 1495. The Inca Period ends with the Spanish Conquest shortly after A. D. 1534. However, in spite of this relatively short period of occupation, Inca influence was intense. Jnca pottery has been found in great quantity, and a vast amount of stone building is attributed to the Period, far more than is found in Argentina or North Chile. New food plants, such as the sweet potato, oca, sweet manioc, peanuts, and coca, are thought to have been introduced by the Jnca. Llamas, although they may have existed before, certainly increased in numbers and im- portance. (It is interesting to note, however, that the llamas and alpacas have not adapted well to the Ecuadorian péramos. Although they still exist today, the numbers are greatly reduced, and it is obvious that Ecuador is a poor country for breeding these animals.) The Inca made Quito a second capital of their Empire. Through their system of mitimaes, rebellious groups were moved out of the area and assimilated peoples moved in. Quechua, the Inca language, virtually replaced previously existing tongues spoken in the Highlands of Ecuador. 50 SOUTH AMERICAN INDIANS __[B.A.E. Buu. 143 The Spanish Conquest and subsequent historical periods repeat essentially the same sequence as noted for the Central Andes. The Coastal Indians were soon replaced or absorbed, especially as new commercial products and ocean shipping ports became important. The Indians of the Highlands continued to serve as labor supply on haciendas and in the mines. The effect of the historical periods has been to eliminate even more the cultural differences between the Keuadorian Indians and those of the Central Andes. COLOMBIA THE ENVIRONMENT AND TRIBES Of the present country of Colombia, only the western one-third is in the region of mountains and valleys. Here, according to James (1942, p. 79), “there are more different kinds of land than are to be found in any comparable area in South America.”’ The eastero two-thirds lies in the Llanos of the Orinoco or in the Tropical Forests of the Amazon. Not all the western third of Colombia is mountainous, and even in the Highland sections only a relatively small portion of the ter- rain offers what might be considered optimum conditions for devel- oping the Andean culture pattern. A review of the topography of western Colombia will clarify this statement. There are four major mountain ranges which run north to south in Colombia. Along the Pacific Coast is a low range called the Serrania de Baudé. In spite of peaks that rise to 6,000 feet (1,800 m.) in altitude, most of this mountain range, like the surrounding coastal plains, is covered by heavy tropical rain forest. In fact the Pacific Coast of Colombia is one of the wettest areas of South America, so that the flora and the fauna and the general environment are remarkably similar to the tropical Amazon forest. The scattered tribes which occupy this region, such as the Chocé in the north and the Colorado, Cayapa, and Barbacoa in the south, pertain to an Amazonian type of culture rather than an Andean. The next two mountain ranges, the Cordillera Oveidehiel and the Cordillera Central, reach altitudes of over 10,000 feet (3,000 m.) and over 18,000 feet (5,400 m.) respectively, and are separated by the rift valley of the Cauca. Large intermont basins are lacking but the mountain slopes are covered with forests, which, if the grade is not too steep, can be cleared away by cutting and burning for small- scale cultivation. However, no part of this region is suitable for large-scale cultivation in terms of Indian economy. In the Departments of Antioquia and Caldas, the Cauca River valley is flanked by narrow strips of flat land. Farther south, in the Departments of Valle and Cauca, the river has cut out a large flat VoL. 2]. ANDEAN HIGHLANDS—BENNETT 51 valley about 150 miles (240 km.) long and from 15 to 20 miles (24 to 32 km.) wide. The valley has an average altitude of about 3,000 feet (900 m.) above sea level, and is today an important region for sugarcane and for grazing. Jn pre-Spanish times, the same valley was apparently of little use. The flats were either swampy or cov- ered with a heavy grass resistant to indigenous agricultural techniques. Most of the Indian groups which inhabited the Cauca Valley region disappeared or were absorbed in the population long ago. The early Spanish accounts reported small populations with little sem- blance of organized political structure. Some of the tribes mentioned are the Coconuco around Popayan, the Lile around the present town of Cali, the Quimbaya around Cartago, the Nori around Antioquia, and the Sinw along the Sinfi River. Other tribes of less certain location were the Gorrén, Chanco, Ancerma, Pozo, Arma, Buritica, Dabeibe, Paucora, Caramenta, Carrapa, and Umbra. The Chibchan linguistic family was dominant. The general impression is that of numerous small tribes occupying limited areas of the mountain region on both sides of the Cauca Valley with little or no over-all political unification. The archeology of this region, although inadequately studied, presents an equally simple picture. Three major ceramic styles are distinguishable: the little-known Sin River style, the Quimbaya (archeological, not tribal) style found throughout most of the Depart- ments of Antioquia and Caldas, and the Upper Cauca style, which in- cludes a number of minor subdivisions. No remains of concentrated population centers have been discovered, which would confirm the pattern of isolated house sites and cultivated plots scattered about the mountain slopes. However, in spite of the lack of evidence of large- scale social organization or village pattern, the archeology reveals well- developed crafts. For example, the Quimbaya ceramics are varied and of good quality and the goldwork is outstanding. Although preservation of perishable objects is poor, numerous spindle whorls of clay show that weaving was common. Graves of the shaft-and- chamber type are well made and contain large quantities of mortuary furniture. The fourth major mountain range, the Cordillera Oriental, is sep- arated from the Cordillera Central by the Magdalena River Valley. The Cordillera runs north and then turns abruptly to the east, skirting the southern end of Lake Maracaibo and entering Venezuela. Ex- tensive intermont basins are located in the Departments of Cundina- marca and Boyacd, centered roughly around the towns of Bogota and Tunja. These Highland basins lie in the zone of tierra fria, between 6,500 and 10,000 feet (about 2,000 to 3,000 m.), high enough to temper the climate. Level stretches are extensive, the soil is fertile, and precipitation is ample. The natural vegetation coverage is of the 52 SOUTH AMERICAN INDIANS [B. A. E. BULL. 143 park-land and grassland type, which was easily cleared by the Indians. The péramos above these basins were of little use to the Indians al- though they are now utilized for cattle grazing. Below the basins, the mountain slopes are badly cut by erosion with little flat land. These high basins of the Cordillera Oriental possessed the most favorable environmental conditions in Colombia for the development of the Andean culture pattern and were, in fact, the homelands of the Chibcha civilization, frequently described as the third highest in the Americas. This region was not only the center of the greatest popu- lation density in Colombia’s past but still maintains that distinction today. Although the Chibcha culture has virtually died out, Spanish descriptions, confirmed in part by archeological work, testify to the extent of the population, the development of the crafts, the village pattern, and the advancements in large-scale political organization. In the northern and eastern extension of the Cordillera Oriental, large intermont basins are not found, and the pattern of scattered farms and habitations was similar to that of the Cauca Valley. Some of the tribes mentioned for this area are the Agata, the Lache, the Chitarero, Guané, Timote, and Cuica. Again, little is known about these now extinct tribes, although all were definitely influenced by the dominant Chibcha. (See Handbook, vol. 4.) The middle Magdalena River Valley and the swampy low flats which border the Caribbean Sea are covered with heavy tropical forests. The scattered tribes which occupy this area belong culturally with the Amazonian pattern, in spite of the fact that some of those in the Magdalena Valley, such as the Panche and the Muzo, were neigh- bors and enemies of the Chibcha and undoubtedly received cultural influences from. them. The tribes found in the tropical lowlands, along the Rio César and surrounding Lake Maracaibo, likewise belong in the Amazonian division, and will be described in the Hand- book, volume 3. The isolated mountains of the Sierra Nevada de Santa Marta, separated from the Cordillera Oriental by the Rio César, are covered with heavy rain forests. In terms of this analysis, this region would not seem particularly favorable for the development of the Andean pattern, although cultivation was possible on many of the lower mountain slopes. However, archeologically speaking, Santa Marta is the only region in Colombia outside of the Chibcha area with good evidence of fairly large villages, and is exceptional in its extensive use of stone as a building material. Village ruins of the now extinct Tairona culture are characterized by large, stone-lined, circular house platforms, stone-faced terraces, streets, reservoirs, bridges, stairways, and dressed-stone tombs. Ceramics, metallurgy, and the carving of bone, shell, and stone objects were technically advanced. The Indian groups in this area today, such as the Cégaba, Ica, and Chimila, speak VOL. 2] ANDEAN HIGHLANDS—BENNETT 53 the Arhuaco language, in turn related to the Chibcha. Although possibly descendants of the TJasrona, the contemporary Indian cultures do not assist much in explaining the more elaborate development in the past. Northeast of Santa Marta on the dry Goajiro Peninsula, the Indians of the same name who today maintain a culture highly influenced by Western contact, belong clearly in Tropical Forest pattern. Around the headwaters of the Magdalena River and in the Tierra- dentro region, between the upper Magdalena and the Cauca Rivers, the rolling, forest-covered mountains are continuous. Here again, as in so many parts of the Colombian Highlands, small agricultural plots and small isolated clusters of houses were located on the river flats and on the mountain slopes. The surviving Chibchan-speaking tribes, the Paniquita, Pdéez, Pantdgora, Pijao, and Andaki, still main- tain this pattern. ‘Two distinctive archeological periods are found in this region. The first, designated as San Agustin, is centered around the headwaters of the Magdalena River and is noted for its carved stone statues representing humans, anthropomorphic beings, animals, and birds. Small dwelling sites and mounds containing stone-lined temples are characteristic. This period is presumably quite old in time, but in spite of the size of some of the statues and the labor which must have been necessary to build some of the temple mounds, there is no evidence of great population density or elaborate social organi- zation. The other major archeological period is named after the region of Tierradentro. It is characterized by elaborate, subterra- nean chambers with a large central room surrounded by niches in which the walls are carved in relief and painted with geometric de- signs in black, white, red, and sometimes yellow. However, there is no evidence of large population or complex organization. The final region in the Colombian Highlands is the southern part in the Department of Narifio. Here the principal intermont basin is in reality one of a series which extends through Ecuador and, from a practical point of view, pertains geographically (as well as culturally) to the Ecuadorian Highlands rather than to Colombia. The Pasto and Quillacinga tribes were in this region. Kight major geographical areas with distinct styles of materials are recognized by archeologists in Colombia, namely, Tairona, Chibcha, San Agustin, Tierradentro, Narifio, Upper Cauca, Quimbaya, and Sint. All these fall within the Highland regions, since the archeolog- ical remains of the lowlands of Colombia. are virtually unknown. These eight major areas are represented by well-made ceramics, evi- dence of a competent weaving, excellent metallurgy, and some elabo- ration of burial methods. However, except for the Chibcha and, more dubiously, the Tairona, there is little evidence of large concentrated populations, advanced political organization, or many of the other 54 ‘SOUTH AMERICAN INDIANS [B. A. E. Bunn, 148 characteristics which typify the Andean pattern in Pert and Bolivia. For a brief time, the Jnca Empire skirted the soutbern border of Co- lombia, but its direct influence was probably negligible. However, archeological evidence shows that Colombia was not completely out- side of the sphere of Andean developments to the south. Quimbaya pottery clearly reflects certain Peruvian influences of the pre-Inca pe- riods. The stone statues of San Agustin were in all probability related to or influenced by Peruvian stone carving. Furthermore, Colombia was affected by some influence from the higher civilizations of Central America. Historically, Colombian cultures reflect the marginal position to the Peruvian Andes and to Central America. - In many ways, the growth of large, concentrated populations, urban patterns, and widespread political organization was handicapped by environmental limitations. For example, the domesticated llamas and alpacas, so important in southern Peri and Bolivia, were not herded in Colombia. In large part, this is due to the fact that the high paramos, unlike the punas, are not suitable for these animals. Even the wild forms, the vicufia and the guanaco, are not found in Colombia. Likewise, copper is rare and tin deposits are not found in Colombia, so that metalwork was limited to gold, some silver, and such alloys as could be made with the available copper. Limitations of terrain for cultivation have been mentioned previously. 7 The habitation pattern ia Colombia has changed markedly in the 400 years following the Spanish Conquest. Today, James (1942, p. 94) lists 14 centers of population concentration in Colombia. The only correspondence with the pre-Columbian Indian pattern is in the high basin region in the Cordillera Oriental in the Depart- ments of Cundinamarca and Boyacd, the old Chibcha territory. Otherwise, the growth of new centers has been made possible by fac- tors outside of the range of Indian economy and technology. Wheat and barley are cereals which grow at higher altitudes than maize, and thus expanded the cultivable terrain. The ox-drawn plow could cope with grasslands resistant to Indian digging-stick methods. Domesticated cattle, sheep, and horses flourished on the high p4ramos worthless to Indian economy. The commercial market for cinchona aad later for cacao opened up new terrain. Many of the mountain slopes, cultivated in desultory fashion or not all by the Indians, have become the prize coffee lands. Discovery of oil created new centers in the Cordillera Oriental. Shipping ports developed on both the Caribbean and the Pacific. Irrigation methods and planting of sugarcane now make the Upper Cauca Valley a rich region. Plati- num and gold mines have formed new communities. Bananas have turned the tropical lowlands into a profitable region. Most of the new centers have not been dependent on the limited Indian population. The Spaniards soon discovered that, outside of VoL. 2] ANDEAN HIGHLANDS—BENNETT 55 the Chibcha region, the Indians were either too few in number or too resistant to be an important source of labor. Consequently, Negroes were imported at an early date and have since been brought in greater numbers to work in the tropical regions. As a result, the Indian element in the Colombian population is small. Although many in- dividuals in the old Chibcha area are of mixed descent, the Indian language and culture have largely disappeared. A few scattered Indian groups can still be found in the Santa Marta region, in the higher basins of the mountains along both sides of the Cauca Valley, and in the rolling forest country of Tierradentro and the upper Mag- dalena River. Although the Colombian Government in its official census recognizes, perhaps too conservatively, a total of 60,000 In- dians today, most of these occupy the eastern two-thirds of the coun- try, which lies outside the Highlands. THE CHIBCHA In Colombia, the Chibcha culture of Cundinamarca and Boyacdé achieved an intensity comparable to that of the Central Andean pattern. The Chibcha are frequently compared to the Inca. How- ever, the Jnca Empire represents the ultimate expression of a pattern whose roots lie deep in the archeological past. The Chibcha culture has no such demonstrable antiquity. No long-term archeological sequences have yet been established in Colombia, although certain local styles, like San Agustin in southern Colombia, may be as ancient as the Early Periods in Pert. San Agustin is famous for its varied stone sculpture, since other aspects of the culture are little known, but what preceded or followed this spectacular development is still to be discovered, and the stone- carving art seems to have had little influence on cultures elsewhere in Colombia. Other regions of the country present similar local styles, little related to each other. Only the Tairona sites in Santa Marta have been studied in detail, and these are apparently as recent as the Chibcha in time. The Chibcha territory has long been a rich agricul- tural section, but archeologists have uncovered no pre-Chibcha materials there, and in any case there is still no good evidence of centuries of occupation. Undoubtedly, a large part of this situation is due to the lack of sufficient archeological work, but there is some doubt that any picture will ever be presented comparable to the Central Andean region in complexity and antiquity of advanced civilization. Nonetheless, the Chibcha culture had advanced so far that it is frequently classed as the third highest civilization of the Americas. Chibcha culture was based on intensive sedentary agriculture, and as such is unique in Colombia. The cultivated crops were the same as those in the Central Andean region with the exception of 56 SOUTH AMERICAN INDIANS [B. A. E. Bun. 143 a few of the root plants. The principal agricultural implements were a digging stick and a wooden spade. It is doubtful if irrigation was known in pre-Spanish times, and probably the Chibcha were less versed than the Jnca in the use of fertilizers and in soil conservation. This region is beyond the natural range of llamas and alpacas, so that the dog and the guinea pig were the only domesticated animals. The absence of suitable domesticated animals made the high péramos virtually worthless to the Chibcha. The common weapons were the spear and spear thrower, darts, slings, and shields. The bow and arrow was known, but was not of great importance. A unique weapon was a two-bladed wooden sword. Transportation was limited to foot travel because of the lack of animals and of streams suitable for navigation. Stones, set up at intervals in circles, mark some of the house sites, but in general, building walls were of poles, in some cases coated with clay plaster. Stone masonry and construction with adobe brick were unknown. The chiefs lived in large compounds surrounded by pole palisades. However, there were no large public works, such as large religious constructions, pyramids, mounds, or fortified sites, which would have required the organization of mass labor. The houses were arranged in villages which never grew to the size of true cities. The craftsmanship was in many senses inferior in quality to other areas of Colombia. Ceramics were competently made, but there was no great variety of forms nor outstanding artistry in decoration. Weaving fibers were limited to the domesticated cotton. The most elaborate decoration of cloth was in painting, either with roller stamps or freehand. Copper was rare and bronze was unknown. Most metallurgy was in gold or in a combination of gold, silver, and copper, called tumbaga. Although goldworking techniques were numerous, workmanship was less competent than that of the Quimbaya. There was no stone carving, although wooden statues have been reported. The clothing followed the Andean pattern: breechclout, shirt, and shawl. Footgear was not worn. Noseplugs and earplugs were common, and the body was painted with roller stamps. The hair-cutting, naming ceremony for children, already mentioned for Northwest Argentina, the Central Andes, and Ecuador, was also practiced by the Chibcha. Puberty ceremonies for both sexes were of considerable importance. Burial was elaborate, especially for chiefs, although the pattern of ancestor worship was not overemphasized. Positions of chiefs and priests descended through the matrilineal line, and there is some evidence for matrilineal descent in other aspects of the culture. Clans, however, were not very important, and property was owned individually rather than collectively. The three major classes in Chibcha society were the nobles, the commoners, and the slaves, that is, prisoners of war. This class system was gradually VoL, 2] ANDEAN HIGHLANDS—BENNETT 57 assuming caste proportions. At least, the chief or ruler was of almost divine importance, and his whole life was surrounded by complex protocol. There was also a specially trained and privileged class of warriors with their respective war chiefs. At the time of the Spanish Conquest, the Chibcha had begun the organization of states and were on their way to unite the whole ter- ritory under one leader. Two loosely organized states were promi- nent at the time of the Conquest, and both were expanding. The conquered peoples were not incorporated with the thoroughness of the Inca political machine, but rather subjected to the payment of tribute. The religious practices and beliefs were complex, but formalized religious organization was weak. A special group of priests partici- pated in the ceremonies, but their principal function was that of an intermediary between the people and the gods. Although a long period of training was necessary in order to become a priest and al- though the position was inherited, the priests themselves were not formally organized. Religious centers, or temples, were but little different from the common dwellings. They did contain idols, and sacrifices, including human, were performed at them. The gods fol- lowed the Central American pattern of being departmentalized, that is, having specific functions, such as commerce or weaving, assigned to them. Chibcha culture offered little resistance to the Spanish invaders, and since the Conquest, the language and much of the culture have dis- appeared. This can be explained in part by the new introductions from the Old World which made radical changes in the economy. In comparison, it is dificult to explain why the Aymara held on to their language and customs with such tenacity while the Chibcha let both slip away. An important contributing factor to the elimination of the Chibcha culture is the fact that the Colonial Spaniards found the Chibcha territory ideal for their own settlements. It was the only region in Colombia where agricultural labor was both abundant and tractable. Spanish cattle could utilize the pd&ramo country; Spanish oats and barley grew well in the high plateau region. Unfor- tunately, there was no place for the Chibcha to retire. FUTURE STUDIES This brief survey of the Highland region of South America indicates clearly the need for new studies. No list of detailed studies would be practical, but a few general suggestions can be selected. (1) Contemporary Indian cultures.—One of the most outstanding gaps in our knowledge of the indigenous cultures of South America is the lack of competent studies of the contemporary Indians of High- 58 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 148 land Ecuador, Perd, and Bolivia. Today, this is the region of the largest and most concentrated Indian population in all of South Amer- ica. Furthermore, these Indians are a functioning part of the con- temporary economic systems of the three countries. Still, the mono- eraphic literature is limited, and field studies rare. Recent investi- gations, when published, will remedy this situation somewhat, but, considering the size and importance of the area, much still remains to be done. These Indians are essentially of the old physical types, speak their aboriginal languages, and maintain much of their old culture, but they have also been deeply influenced by 400 years of European contacts. Contemporary studies face the difficult problem of dealing with a dual system, one basically Indian and one basically Kuropean in pattern. The conflicts and the integration of these two systems have never been properly studied. (2) Basic Andean pattern.—In this Introduction, a cultural pattern found throughout much of the Highland region of western South America has been sketched. More detailed studies are needed to determine the significant composition of this basic pattern. ‘The en- vironmental setting for the development of this pattern has been emphasized, but it is obvious that a favorable environment is merely a setting, and not the final determinant. (3) Central Andean pattern.—The Central Andean pattern, which culminated in the Jnca Empire, had its roots in the archeological past. More detailed studies of the composition of the Central pattern are needed. From an archeological point of view, earlier periods in the different subdivisions, such as Chavin, Mochica, Nazca, Paracas, Recuay, Tiahuanaco, and Chiripa, should be examined in detail to determine the elements which they possess in common. Archeolo- gists concerned with the complex problem of chronology emphasize differences in these early periods. A study of the similarities would be equally stimulating and of value in discussing the cultural history of wider areas. (4) Highland Andean horizons.— Papdwuco ft ) a 2 pe RLY ‘ } Ge ay nS KR QLake Junin Fad = Hiaets so ofp Oo ROAM Ra H & i | S °*HUANCAYO oe n Ss fe | © CENTRAL wih iw eAYA CUCHO ‘ ; he \ > HIGHLANOS ass me) a 2 eOLLANTAYTAMBO + %. 1. a °CUzZCO = H J °PUCARA_} ger> BOLIVIA } SOUTH ( HIGHLANDS Ley | . @COCHABAMBA ° a ) Rivers, lakes © CITIES, TOWNS, SITES ---- NATIONAL BOUNDARIES 100 200 300 400 KILOMETERS Mar 2.—Archeological sites and regions of the Central Andes (Pert-Bolivia area). (Revised from Kroeber, 1944.) Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 63 quently, knowledge of the archeological past, the unwritten history of these Indians, is of importance. THE REGION AS A WHOLE: ARCHEOLOGICAL UNITY The Perti—Bolivia archeological region forms a unit which can be compared with other archeological areas in South, Central, and North America, not to mention the Old World. The unity is more than geographical. A trained person can generally recognize a speci- men from this region regardless of its time period. Throughout the archeological past, the Peri—Bolivia area has lacked many things which are important elsewhere, such as any knowledge of the wheel or the arch, writing or dating, urn burial, and tetrapod ware. Such negative factors could be multiplied, but positive points are needed to explain the unity of the region as a whole. Some of these are listed below. (1) The basic subsistence of agriculture and herding was essentially the same in all periods, although the details vary in terms of the en- vironment. The cultivated food plants that were known in all periods are maize, beans, squash, potatoes, manioc, quinoa, and oca. Coca chewed with lime was the standard narcotic, rather than tobacco. Domesticated llamas and alpacas were utilized for meat, wool, and transportation in all periods, although their natural habitat is in the Highlands. (2) All periods knew the same building materials, although the environment influenced the selection somewhat. Adobes were used for building on the dry Coast, stone more commonly in the High- lands. Advanced stone masonry is not a common achievement for all periods, but has a wide distribution. _ (8) The same major crafts were developed in all periods. These include pottery, weaving, metallurgy, and architecture, as well as such minor crafts as basketry, shell inlay, featherwork, and wood carving. (4) Individual distinctions in social rank, specialized priesthood, complex religion, good military organization, and perhaps the clan were always present, insofar as can be judged from the archeological materials. | (5) Certain general trends follow throughout the time periods, ir- respective of geographic divisions. Early Periods everywhere empha- sized realistic or naturalistic design and individual skill in craftsman- ship. The Middle Periods are marked by conventionalized designs. Late Periods prefer geometric patterns, culminating in the Inca Period, which is frankly geometric. A general reduction in color range is noticeable from Early to Late Periods. Likewise, a gradual growth from the Early Period villages to the Late Period cities is seen in most areas. 595682—46——7 64 SOUTH AMERICAN INDIANS [B. A. E. BuLL, 143 In résumé, the differences in archeological periods throughout Pert and Bolivia are essentially those of selection and emphasis rather than startling contrasts in basic techniques or materials, Weaving furnishes an excellent illustration. The basic fibers in all periods are wool and cotton, in spite of the fact that wool was essentially a High- land product and cotton was limited to the Coast. Furthermore, the basic weaving techniques were known throughout all the periods, and no improvements were made on the standard girdle-back loom. The differences in textiles from region to region, or from period to period, are essentially due to selection or preference for certain techniques and designs, rather than radical changes in technology. This unity of achievement in Peri and Bolivia, regardless of geo- graphical and time differences, implies that the total development was relatively independent of the rest of the Americas. This does not deny the possibility of external influences from Colombia, Central America, and the Amazon, but rather suggests that the region was sufficiently advanced and organized to resist any wholesale migration or influence from the outside. PROBLEMS In a region as large as the one treated here and one which maintained a comparatively complex civilization for many centuries, the number and variety of problems involved in the study of the materials are myriad. A few samples are discussed below: (1) Chronology.—The major emphasis of archeologists working in the Andean region is and has been chronology. The absence of any written records or calendrical dates has made it imperative to arrange the mass of materials in this region in at least a relative chronological order. The techniques which have so far been employed in the chrono- logical studies are standard practices in archeological work, such as stratigraphy in refuse sites, building stratigraphy, grave overlapping, combination of building and grave stratigraphy, grave isolation, unit- site isolation, developmental sequences in art styles, the use of historical materials, and distribution reconstruction. In spite of all this, more refinements are necessary. No attempt has yet been made to apply the science of tree-ring dating to the Andean region. Practically all the chronological studies have so far been based on ceramics, which are the best preserved and the easiest to handle. It is equally im- portant, however, to include other materials such as textiles, metals, stone carving, and masonry. A sound Andean chronology is important for cross-dating all South American archeology, and it is equally basic for numerous other types of study, such as the tracing of art sequences. (2) Trends.—The Pert-Bolivia region presents a long and complex archeological past, good historical source material on the Inca at the time of the Conquest, and, finally, numerous living Indians who still VOL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 65 preserve their language and much of their old culture. Thus there are excellent opportunities for tracing long-term cultural trends. There are, unfortunately, some limitations since the archeology presents an incomplete picture. The material culture of the past civilization is the best represented, and even here preservation factors are unequal in different regions. Archeological interpretations of the social culture must always be treated with caution. Certain evidence is, however, available. Painted and modeled ceramics may portray the bird and animal life of the time, the type of dress, and general scenes of social activities. Occupation and subsistence can be determined in part from direct evidence, such as remains of plants, irrigation projects, and cultivated fields. Burials furnish considerable information about practices pertaining to the treatment of the dead. The ruins them- selves furnish good evidence about the size of villages, the types of houses, the religious structures, and fortifications. Considerable validity can be given to such interpretations if they are checked by studies of contemporary Indian ethnology and by the description of the Inca at the time of the Conquest. The problem, then, of tracing long-term trends is one of major importance. (3) Cultural development and geography.—An interesting and important problem concerns the relationship of topography and en- vironment to cultural development in different archeological periods. Many of the factors which determine contemporary settlement in the Peruvian Highlands were also effective in the past. However, the mountains and deserts which are such difficult barriers to contem- porary transportation were apparently much less so when all trans- portation was by foot or by llamas. Many studies could grow out of this general problem. (4) Cultural contacts.—A study of the influence of the Andean civilization on other areas of South America is important, and the reverse is equally so. Also, studies of contacts within the Pert- Bolivia area should be made, emphasizing such factors as trade, mi- eration, diffusion, and conquest. (5) Technical studies.—Such materials as textiles and metals can be subjected to technical studies by themselves. Some such studies have already been made, principally on the textiles. The fineness of the thread and the complexity and variety of weaving techniques have been carefully analyzed, and the Peruvian achievement has been measured in terms of hand weaving throughout the world. In metallurgy, the techniques and processes can be determined from the specimens. Other types of technical studies could be made, including architecture, city planning, and farming methods. (6) Art studies—Much attention has already been given to Peruvian archeological material by art students. Studies can be made of design or modeling or sculpture, and also, in connection with 66 SOUTH AMERICAN INDIANS [B. A. E. BULL. 143 chronological evidence, of the art developmental sequences. Only a few studies have been initiated of the influence of one medium, such as textile pattern, on another, such as ceramic painting. SOURCES The archeological bibliography for the Pert-Bolivia region is very extensive. It is impractical to mention any but a few of the numerous titles for this area. Fortunately, some excellent bibliographies exist: Dorsey (1898), Means (1928), Richardson and Kidder (1940), Schwab (1936). Likewise certain journals devote sections to bibliography in this field, such as the Journal de la Société des Américanistes de Paris; the Boletin Bibliografico de la Universidad Mayor de San Marcos, Lima; the Boletin Bibliografico de Antropologia Americana, México; and the Handbook of Latin American Studies, Cambridge, Mass. Finally, specialized bibliographies are found in many of the monographs cited here. Pre-nineteenth century sources of value for the archeologist are few. The early chroniclers are of some use, but their principal interest is for the study of Inca ethnology. Consequently, biblio- graphic references to them will be found in that section of the Hand- book. A few, like Cieza de Leén, give descriptions of ruins and statues; but the majority are of value in portraying life at the time, which serves as an aid in interpreting the archeological remains. The travelers of the 19th century, although not trained archeologists, were good observers, and consequently their writings are valuable source materials. Some of the outstanding ones are: Bastian (1878), Castelnau (1852), Middendorf (1894), D’Orbigny (1867), Rivero and Tschudi (1851), Squier (1877), Wiener (1880). A number of authors treat the Pert-Bolivia field as a whole, either from the point of view of archeological résumés, special topics, or art studies: Baessler (1902-1903), Doering (1936), Fuhrmann (1922 a, 1922 b), Raoul and Marie d’Harcourt (1924), Hewett (1939), Joyce (1912), Kelemen (1943), Kroeber (1944), Langlois (1935-36), Leh- mann and Doering (1924), Mead (1924), Means (1931), Muelle and Blas (1938), Schmidt (1929), Seler (1893), Tello (1929, 1942), Thomp- son (1936), Wasserman-San Blas (1938). The materials in this section have been presented from the point of view of time period and geographical location in one of six major areas. Bibliographical sources are best cited in terms of the geograph- ical divisions, since most accounts deal with more than one time period. Some of the outstanding sources for the six regions are listed below. For the North Coast of Peri.—Bennett (1939), Holstein (1927), Kroeber (1925 a, 1926 a, 1930 a, 19380 b, 1944), Larco Hoyle (1938-39, 1941), Muelle (1936), Tello (1938), Uhle (19138 a). . VoL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 67 For the Central Coast of Peré.— Gayton (1927), Kroeber (1925 b, 1926 b), Muelle (1935), Newman (n. d.), Reiss and Stiibel (1880-87), Strong (1925), Strong, Willey, and Corbett (1948), Uhle (1903, 1908), Villar Cérdova (1935). For the South Coast of Peré.—Berthon (1911), Doering (1927), Gayton and Kroe- ber (1927), Kroeber (1937), Kroeber and Strong (1924 a, 1924 b), Putnam (1914), Tello (n. d.), Uhle (1918 b), Yacovleff and Muelle (1932). _ For the North Highlands of Perti.— Bennett (1942, 1944 b), Kinz] (1985), Me- Cown (n. d.), Tello (1923, 1930). For the Central Highlands of Peri.—Bingham (1913, 1915 a, 1915 b, 1916, 1922, 1930), Fejos (1944), Franco Inojosa (1935, 1937), Franco Inojosa and Gonzalez (1936), Markham (1910), Means (1938), Pardo (1987), Rowe (1945), Tello (1937), Valedrcel (1934 a, 1934 b, 1935 a, 1935 b). For the South Highlands of Peri and Highland Bolivia.—Bandelier (1910), Ben- nett (1934, 1936), Casanova (1942 c), Franco Inojosa and Gonzalez (1936), Kidder (1948), Posnansky (1914), Stiibel and Uhle (1892), Tschopik (n. d.), Valcdrcel (1935 ¢). Archeological materials in Perfi and Bolivia have been the subject of special technical studies. Although many of the sources already cited deal with architecture, ceramics, metallurgy, and weaving, a few specialized studies are cited below. For special studies of metallurgy.—Baessler (1906 a), Lothrop (1987 a), Mead (1915), Nordenskidld (1921). For special studies of the Peruvian quipu and calculation.—Locke (1923), Nor- denskidld (1925 a, 1925 b), Wassén (1931). For special studies of textiles and weaving techniques.—Baessler (1906 b), Craw- ford (1915, 1916), d’Harcourt (1934), Levillier (1928), Means (19380), Montell (1929), O’Neale (1937, 1942), O’Neale and Kroeber (1930), Stafford (1941), Yacovleff (1933). GEOGRAPHY This brief review of the general geography of Pera and Bolivia is intended to relate certain geographic features and divisions to the archeological past. Since there are no indications of important cli- matic changes during the time period considered, the environmental characteristics of contemporary Peri and Bolivia are also applicable to the archeological past. The excellent preservation of delicate textiles in the dry sand graves, and the remarkable condition of many early adobe buildings, testify that the Coastal desert conditions have long prevailed. Geographically, Peri is commonly described in terms of three vertical strips: the desert Coast, the high Andes, and the Montafia or eastern slope of the Andes, which merges into the Amazon jungle. An analysis of the archeological materials shows that these three strips did not always produce the sharp cultural divisions that the contrast- ing environments would suggest. On the other hand, such environ- mental factors are always of importance in the over-all interpretation. The most extensive archeological work has been carried out on the Coast and in the Andes, and consequently the history of these two strips is reasonably well known. The Montajfia region has received so 68 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 little attention archeologically that its ultimate importance can only be guessed at. It is generally conceded that the Andean civilizations did not penetrate deeply into the true Amazon tropics. On the other hand, it may ultimately be shown that the Andean cultures drew heavily on the varied resources of the Tropical Forests and the Mon- tafia. A number of archeologists have stressed the probable im- portance of the Amazon region, particularly in the initial develop- mental stages of the Andean civilizations. The Montafia will be described in the Handbook, volume 3. The Pacific Coast of Peri and North Chile is the most desert region of South America. 'Topographically, the Coast is composed of pied- mont plains; mountain spurs which run transversely to the Andes; some traces, particularly in North Chile and southern Pert, of an old Coastal mountain range; and a series of rivers which originate in the Andes and flow into the Pacific. There are some 25 principal rivers on the Peruvian Coast, as well as a number of unimportant ones. Al- though the flood plains of some of the rivers merge, most of them are separated by a good stretch of desert as well as by mountain spurs. The rivers are classified as permanent, semipermanent, or sporadic, depending on whether they head in the continental watershed or not. Most habitation on the Coast of Peri depends on the river valleys and irrigated agriculture. Farm land is limited but exceptionally fertile. In preagricultural times, the Coast of Peri must not have been very attractive, Fishing and collecting shellfish would have supported a certain population, and it is possible that wild-plant collecting and hunting were once profitable in the valleys. Evidence for such suppo- sitions has been obscured or obliterated by many centuries of intensive cultivation with irrigation, so that it is difficult to discover what the original wild-plant and animal life was like. It would also follow that agriculturists could make little use of the Coastal valleys without a good knowledge of irrigation since most cultivation would be limited to the flood plains of the rivers. On the other hand, once the valleys were settled by agriculturalists with a good knowledge of irrigation, the development of high civilization would be possible. The climate is not severe. The land is fertile. The mountain spurs and desert stretches which separate valleys make them relatively easy to defend. Materials for building, such as adobe and stone, are plentiful, and fishing and some hunting would supplement the agricultural produce. Each of the major valleys on the Coast of Pera is in itself a unit with an independent cultural history. However, during some pe- riods adjacent valleys might share the same development or be domi- nated by the same culture. Consequently, for descriptive purposes, several valleys can be grouped together. The valleys included in such a grouping may not be the same in every time period, however, and throughout the archeological history of Perd, the cultures always VoL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 69 show a tendency to revert to valley independence. In general, a division into North Coast, Central Coast, and South Coast valleys is satisfactory, but in any particular study a division into smaller units is necessary. The dryness of the Coast of Perti preserves materials, particularly when buried in cemeteries in the desert sands. Such perishable ob- jects as textiles, calabashes, featherwork, wooden artifacts, and plants are all well preserved. This is apt to give the false impression that the Coast cultures were more advanced than those in the rainy Highlands, where preservation is bad. The Highland region of Peri and Bolivia presents many contrasts in topography and environment. Much of the terrain is uninhabit- able because of the high altitudes and the barrenness. The habit- able parts are largely intermont plateaus, of which the altiplano in southern Pert and Bolivia is the largest. Life is also possible in the deep river valleys which cut through the mountain chain. Most of the Highland region is unforested, and great stretches of the terri- tory are classed by geographers as grassland, bushland, or even des- ert. Because of the altitude, the climate is cold throughout the year but freezing temperatures are rare in the inhabited sections. The range in temperature throughout the 24 hours of the day is great. Most of the region receives ample rainfall in the season from Decem- ber to March. The wild-animal life today is limited, although vicufia, viscacha, fox, chinchilla, deer, and game birds are available. However, both the llama and the alpaca were domesticated from some wild form like the vicufia or guanaco, and there are other indications that game was far more abundant in the past. The same can be said for wild plants, since this is one of the regions in which potatoes, quinoa, oca, and arracacha may have been first domesticated. In contrast to the Coast, the Highland region was probably quite attractive to early hunting and gathering groups, although traces of these have not yet been found by the archeologists. For agriculturists, there are also numerous advantages in the high mountains. Stone and adobe are available for building materials, although wood, in general, is absent. There is clay for pottery, wool for weaving, grass for plaiting, and reeds in the lakes for making mats and boats. Likewise, the Andes are rich in minerals; and copper, silver, gold, tin, and mercury were all utilized in pre-Spanish times. The present population clusters in the Highlands correspond closely to those of the past with the exception of some of the modern mining centers. These population clusters are found in natural areas, such as plateaus or mountain river valleys, where land is available for sub- sistence agriculture, and where grasslands are available for pasturage. The clusters are separated from each other by stretches of virtually 70 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 uninhabitable mountain wastes. Each of these clusters represents an archeological unit with distinct cultural features. Some of the principal clusters are: (1) The Cajamarca region of the far North Highlands. (2) The Huaraz region in the Callejé6n de Huaylas. (3) The Hudnuco region. (4) The Mantaro River region. (5) The Cuzco region in Central Peri. (6) The Arequipa region. (7) The Puno region in southern Pert. (8) The Altiplano region of Bolivia. (9) The Eastern Cordillera region of Bolivia. In contrast to the Coast, preservation is poor in the Highlands. Objects of ceramics, stone, some bone, and some metal are preserved, but other materials usually disintegrate. Consequently, much of the evidence for the arts and crafts is based on analogy and on indirect evidence. Furthermore, conditions of life in the Highlands seem quite severe. Warm clothing and good housing are necessary. The scarcity of fuel is a major problem. Available land for agriculture is limited, and many crops will not grow in the high altitude. In spite of the fact that the Highlands seem far more limited in potentialities than the Coast, the archeological history shows that they have al- ways played a dominant role. For the purposes of this section, the cultural-geographical areas have been grouped into six major regions (map 2). This is dictated in part by a desire for simplification, but also because of the avail- ability of archeological evidence. Insufficient work has been done in many of the areas to justify separate treatment. The six major regions discussed in this section are: (1) The North Coast.—Theoretically all the Coastal valleys between Casma and Piura are included, but practically the discussion is limited to the three adjacent valleys of Viri, Moche, and Chicama. (2) The Central Coast.—The valleys from Lurin to mune are included with special attention to the Rimac, Ancén, Chancay, and Supe Valleys. (3) The South Coast.—The valleys from Rio Grande (Nazca) to Cafiete, with special attention to the Nazca, Ica, and Pisco (Paracas) Valleys. (4) The North Highlands.—The region of the Callej6n de Huaylas and the Chavin section of the Marafion River are included here. (5) The Central Highlands.—The region around the town of Cuzco is covered. (6) The South Highlands and Highland Bolivia.—The Puno region of Pert and the altiplano of Bolivia are grouped together in this category. CHRONOLOGY At the present time, there is no absolute chronology in Andean archeology. The only concrete date is 1532, the year of the arrival of the Spaniards in their historic conquest of the area. Otherwise, calendrical dates are guesses and, furthermore, prospects of obtaining “Vou. 2] - ARCHEOLOGY OF CENTRAL ANDES—BENNETT 41 accurate dates on the basis of internal evidence are meager. There was no writing in pre-Spanish Pert, no dated stones, no evidence of a recorded calendar. So far no one has attempted tree-ring dating, and the material present does not seem to be well adapted to this type of work. Consequently, archeological dating in Pert and Bolivia must be in terms of relative chronologies. ‘These involve two major factors: one, geographic location; and two, time sequence. On the Coast of Perti each valley can be considered as a separate unit, as can each of the inhabited areas in the Highlands. Within a valley or an isolated Highland area, a relative sequence can be set up showing the succession of different styles and periods. Once this has been accomplished, adjacent valleys or areas can be compared. In many cases, it can be demonstrated that a cluster of valleys on one part of the Coast has comparable sequences or overlapping styles. The problem of linking local time sequences in valleys or areas which are widely separated is more difficult, but can be approached by examining styles which have a wide distribution, trade pieces, and the like. Geographic distance is, however, always difficult to inter- pret in terms of time. For example, a distinctive style in a Central Coast valley appears also in a North Coast valley, but it is exceedingly difficult to estimate the length of time necessary for such a wide distri- bution. If the style spread because of military conquest, only a short time factor would be involved; but if it spread by gradual infiltration, then the time might be exceedingly lengthy. To some extent the time factor can be judged by detailed comparison of the styles in the two valleys. Unfortunately, few of the Coastal valleys or the Highland areas in Andean region have local sequences satisfactorily established. Much more work is needed, but there is no doubt that, as the gaps are filled, a trustworthy relative chronology for the whole Pert-Bolivia region can be established. The broad outlines have already been laid down although many points are still unsatisfactory, particularly those involving the relationship between Coastal and Highland regions. A résumé of the generally accepted sequences so far established is given in the “‘Chart of Perti-Bolivia archeological periods’, on page 80. All the six major geographical divisions are represented, namely, the South, Central, and North Coasts, and the North, Cen- tral, and South Highlands. The Central Highlands division appears incomplete because, in spite of extensive investigation, so far only one style has been discovered which is demonstrably older than the classic Inca Period. The valleys and areas chosen to represent the major geographical divisions are those in which archeologists have established sound local sequences. Major periods are indicated by capitals; styles, not iso- lated as periods, by small letters. ‘The period names correspond to Te SOUTH AMERICAN INDIANS [B. A. E. Bunn, 143 those used in this report, although a few alternative names are indicated by parentheses. The relative time of a style or period is suggested by its horizontal position as indicated by the general divi- sions in the column at the extreme left. The suggestion of dates by centuries is a gross approximation, but on the conservative side accord- ing to most authorities. Some would extend the whole chart back several centuries before the Christian Era. The Nazca and Ica Valleys are selected for the South Coast because between them a good sequence can be established. The nearby site of Paracas is included because it is of special importance. The inter- mediate South Coast valleys of Chincha and Cafiete do not have complete sequences and so are not included in the chart. By combin- ing the four valleys of Lurin, Rimac, Ancon, and Chancay, a fairly good sequence for the Central Coast can be established. Although every period has not been found in each of these valleys, it is highly probable that future excavation will remedy this. On the North Coast, a fairly complete sequence has been outlined for the valleys of Viri, Moche, and Chicama. To the south, the valleys of Casma, Nepefia, and Santa are not included, although they present very interesting material. Of the valleys north of Chicama, only Lamba- yeque has been investigated with any intensity, and even there the sequence is far from final. The North Highlands are here represented basically by Chavin and the Callején de Huaylas, but reference to Huamachuco is also included. The Central Highlands cover only the Cuzco region. The South Highlands are combined with Bolivia for a sequence represented by Puno in Pert and Tiahuanaco in Bolivia. Other Highland regions have been too little studied for inclusion in this chart. Andean archeologists have long used the terms Early, Middle, and Late Periods to refer to the gross time divisions preceding the Inca. Today, it is well established that a Chavin horizon precedes what was formerly called the Early Periods. Consequently, the terminology for the major time divisions will have to be revised, since terms like pre-Early are confusing. For the time being, the chronological divi- sions can be designated as the Chavin Periods, the Early Periods, the Middle Periods, the Late Periods, and the Jnca Periods. Likewise, it is possible to speak of three pan-Peruvian horizons, namely, Chavin, Tiahuanaco, and Inca. Each of these three widespread styles is sep- arated by a respectable time interval in which local styles are devel- oped in the different geographic areas. The Chavin horizon is the earliest yet isolated in the Peruvian re- gion. Chavin style has been found in early sites unmixed with other known styles in the North Highlands of Pert, on the Central Coast, on the North Coast, and as a minor design influence in the Cavernas Period of Paracas on the South Coast. The North Highlands site of VoL, 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 73 Chavin de Hudntar seems the most logical center of distribution al- though there is no final evidence for this. In the interval between the pan-Peruvian Chavin horizon and the Tiahuanaco horizon, local cultures dominate the major geographic areas. The distinctive Nazca style appears on the South Coast and the Mochica style develops on the North Coast. On the Central Coast, the Early Period interval is occupied by the White-on-red and the Interlocking styles, and in the North Highlands, Chavin is followed by the Recuay style. The second pan-Peruvian horizon is Tiahuanaco. Its center of origin and distribution seems to be the altiplano of Bolivia around Lake Titicaca. In this region, the Tiahuanaco style dominates the archeological scene almost up to the Jnca time. Tiahuanaco-in- fluenced styles have a wide distribution in Bolivia, Pert, Chile, and Argentina. In Perf, the style is found in quantity on the Central Coast, on the North Coast, and on the South Coast, where it mixes with the last phases of the Nazca style. In the North Highlands, a Tiahuan- aco style follows the Recuay Period. On the Coast, the Tiahuanaco Period soon breaks up into a number of local styles. Somewhat later, a Coastal reformulation of Tiahuanaco, called the Black-white-red geo- metric style, has a limited distribution. The basic Tiahuanaco plus the local break-down followed by the Black-white-red geometric forms the Coastal Middle Period. In reality, much of this development is independent of the Bolivian Highlands. Following the Tiahuanaco-influenced Middle Periods, new local styles were developed, such as the Ica style on the South Coast, the Chancay Black-on-white style on the Central Coast, and the Chimu style on the North Coast. These Late Period styles inherit a great deal from the Middle Periods, but are still quite independent. The Chimu style also shows a revival of many elements from the Early Period Mochica style. The final pan-Peruvian horizon is the Inca. The center of origin and distribution is the Cuzco region in the Central Highlands, where Inca material is found in great abundance and where the Jnca main- tained their capital. At the time of the Spanish Conquest, the Inca were in political control of all of Peri and Bolivia as well as parts of Ecuador, Northwest Argentina, and Chile. The archeological remains confirm this wide distribution. In the Pert-Bolivia region, no trace has yet been found of early nomadic hunters nor of prehorticultural or preceramic material. _ There is likewise no good evidence for the early stages of agricultural or ceramic development. However, it still seems probable that early nomadic hunters passed through the region at one time, and recent botanical evidence points to the Andean region as one of the potential centers for the original domestication of such plants as manioc, pota- 74 SOUTH AMERICAN INDIANS [B. A.B. Bunn. 143 toes, and even corn. There are large shell heaps on the Coast of Pert which may ultimately produce evidence of these early stages, but so far little excavation had been effected. A number of reasons can be suggested why evidence of these early stages has not yet been discov- ered. One, the region is a large one in which relatively little scientific work has been done. ‘Two, the remains of such early stages may well have been obscured or obliterated by some 2,000 years of intensive agricultural occupation. Three, archeologists have been too occupied in the study of the advanced cultures to concentrate on the search for early ones. Four, early remains are nowhere elaborate and therefore are difficult to identify. In fact, they are usually first discovered by accident. Someday it may be possible to cross-date Peruvian and Central American materials. That the high civilizations of the Andean region and those of Central America had the same basic culture is evident in the common food plants, such as maize, beans, and squash; in the ter- raced and irrigated agriculture; in the development of such crafts as ceramics, metallurgy, and weaving; in the types of stone and adobe masonry; in the similarity of construction techniques; in the patterns of grouping temples, pyramids, sunken courts, and inclosures; in the use of the corbeled arch; and in numerous other generalized points. At the same time, detailed similarities in styles, ceramic shapes, tech- niques, and the like have not been satisfactorily demonstrated between the Andean region and Central America. Until valid relative chro- nological sequences have been established for Ecuador and Colombia, there is little possibility of cross-dating the Peruvian region with Central America. TERMINOLOGY There is considerable variation in terminology for styles and periods among the Andean archeologists. The accompanying chart (p. 80) follows the terminology used in this section of the Handbook and is based, for the most part, on published accounts. For convenience, a list of standard and alternative terms follows, utilizing the chart as a basic outline. 1. Chavin Periods: Chavin. A standard term for the site of Chavin de Huéntar, the style, and the period wherever found. Some authors limit the usage so that the term refers only to the Highland style. Coast Chavin. The Chavin coastal style and sites. Cupisnique. A variant term for the Coast Chavin, Chongoyape. Coast Chavin style from a site in Lambayeque Valley. Nepefa. Coast Chavin style from a site in Nepefia Valley. - Early Ancon. Coast Chavin style from shell heaps at Ancon. Puerto de Supe, Primitive Ware. Coast Chavin style from shell heaps at the pert of aig Valley. VoL, 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 79 Cerro Sechin style. From a site in Casma Valley with carved stone pillars and blocks which seem distantly related to Chavin. Moxeke style. From a site in Casma Valley with relief clay carvings related to the Coast Chavin style. El Salinar. Newly discovered style, and perhaps period, in Chicama Valley, combining certain Coast Chavin and White-on-red style features. Thought to be intermediate between Coast Chavin and Mochica. 2A. Early Periods, South Coast: Nazca. Standard term for polychrome style of South Coast. Proto-Nazca. Alternative term for Nazca. Early Nazca. Alternative term for Nazca. Nazca-A. Early stylistic division of Nazca. Nazca-B. Stylistic division of Nazca. Nazca-X. Stylistic division of Nazca, transitional between Nazca-A and Nazca-B. Pre-Nazca. Alternative term for the Nazca-B style, which is thought by some writers to be earlier in time than Nazca-A. Chanca. Generic term for a Highland style related to Nazca-B. Paracas. The name of a site on the Pisco Peninsula, but frequently used as an over-all term for the two related periods at that site. Paracas Cavernas. One period and style at the site of Paracas. Paracas Necropolis. Another period and style at the’site of Paracas, usually con- sidered to be slightly more recent than Cavernas, and distantly related to - Nazca. Cerro Colorado style. Alternative term for Paracas Cavernas, and sometimes for Paracas in general. Ocucaje. Ceramic style in Ica Valley related to one or both of the Paracas styles. Early Chincha. An early incised ware in Chincha Valley. Proto-Chincha. An alternative term for Early Chincha. 2B. Early Periods, Central Coast: White-on-red (W-on-R). Standard term for style and period found in Chancay Valley and elsewhere. Interlocking. Standard term for style and period on the Central Coast, at Chancay and Pachacamac and elsewhere. Once thought to antedate the White-on-red. Intermediate. A transitional period between White-on-red and Interlocking at Chancay. ‘Wari. Term for a generic style in the Highlands along the pee? River, with some relationship to the Interlocking style. Wanka. Alternative term for Wari. Negative. A style, not a period, found mixed with the ifemeeiae at Pacha- camac. Early Lima. A composite style with a number of strains which may or may not be contemporaneous. Found at Nieverfa and elsewhere in the Rimac Valley, and thought by some to antedate the Coast Tiahuanaco Middle Periods, Proto-Lima. Alternative term for Early Lima. Nieveria, Alternative term for Early Lima. Aramburti. Alternative term for Early Lima. 76 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 2C. Early Periods, North Coast: Mochica. Generally accepted term for the white and red, modeled ceramics and period of the North Coast valleys. Muchik. Alternative term for Mochica. Early Chimu. Alternative term for Mochica. Proto-Chimu. Alternative term for Mochica. Early Moche. Alternative term for Mochica, in the Vird, Moche, and Chicama Valleys. Mochica-I. Stylistic subdivision of Mochica. Mochica-II. Stylistic subdivision of Mochica. Chimbote style. Alternative term for Mochica-II. Gallinazo. A style and period isolated at Gallinazo ruins in Viri Valley. Thought by some to follow Mochica, but other authorities would like to place it earlier. Negativo. Alternative term for Gallinazo. Tipo Nepefia. Unusual alternative term for Gallinazo. Middle Moche I. Alternative term for Gallinazo. 2D. Early Periods, North Highlands: White-on-red. Style found near Huaraz, probably related to Coast Period of same name. Recuay. Style and period named after sites in the vicinity of the town of Recuay. Recuay—A. Stylistic strain in Recuay Period. Recuay—B. Stylistic strain in Recuay Period. Andean Archaic. Stylistic strain in Recuay Period, thought by some authorities to be a distinct period. Huaylas-I, Alternative term for Andean Archaic. Huaylas-II, Alternative term for Recuay. Lagenaria. Alternative term for Andean Archaic. 2E. Early Periods, South Highlands and Bolivia: Tiahuanaco (Tihuanacu, Tiawanako). Name of a site, a style, and a period in the Highlands of Bolivia. Also used as a general term for the style wherever found. | Early Tiahuanaco. A pre-Tiahuanaco style and period found at the site of Tia- huanaco in Bolivia. Classic Tiahuanaco. A term for the developed Tiahuanaco style at the site of Tiahuanaco and elsewhere in Bolivia. Tiahuanaco-I. Stylistic division of Tiahuanaco. Tiahuanaco-II, Alternative term for Classic Tiahuanaco. Derived Tiahuanaco. A style closely related to the Classic Tiahuanaco but with a@ wider distribution in Bolivia and perhaps elsewhere. Chirtpa. Lia 5 - 3? Bee Baa eo an iD >> . > Ht ii ii Hee oe eat AN 4 FH se oh Te pe sch ea = a Ze WN I SY S se ACIS Figure 10.—Classic Tiahuanaco pottery shapes and designs. a—a’, vase shape with typical step design; b—b’, open-bowl shape with design on inner rim, of Classie pumas; c—c’, constricted-neck bowl with spout and serpent design; d—d’, this bowl and design are both unusual Classic style. (After Bennett, 1934, fig. 16.) Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 117 axes. Bone needles, disks, and incised tubes are found, and shell wa- cut into disks and beads. Nothing is known about woodwork, cals abashes, basketry, and other of the more perishable materials, although they doubtless existed. The Classic Tiahuanaco Period places great emphasis on masonry. No other Peruvian period matches the exactness and precision of the stonework in its careful cutting, dressing, squaring, and notching. The nearest rivals are Chavin and Jnca but even in those cultures the masonry is inferior. Classic Tiahuanaco also emphasizes the stone-carving design style, in contrast to stone carving as such. For people so adept in the cut- ting and working of stone, thesculpture is remarkably simple and plain. The intricate designs, however, which are applied to statues, slabs, and gateways are surprising to find on stonework. For example, the ““Gateway of the Sun” has a central, upright figure with an elaborate crown and a staff in each hand. On both sides of this central figure are three rows of running, human figures, some with condor masks. Along the base of the gateway is a frieze of front-view faces, each sur- rounded by tab appendages. The incised designs on the statues add other elements, such as condor heads, puma heads, fish heads, and geometric units. The style as a whole is angular, rigid, and highly conventionalized, and few concessions are made to the difficult medium on which it is applied. The similarity to textile designs has already been mentioned. The ceramic design, in contrast to that found on stonework, is quite simple. Thus two design styles can be recognized, one for ceramics, one for stone carving and textiles. These two groups are not alto- gether mutually exclusive, but the more elaborate designs of the stone carving rarely if ever occur on ceramics; and, likewise, the winged king condor, the profile head, the full-front-bodied warrior, and even the curl-tailed puma are not common on carved stone. Tiahuanaco style has generally been defined in terms of the stone-carving design, but the ceramic style is clearly recognizable. Little can be reconstructed about the daily life of the builders of Tiahuanaco. The art and the existing remains presen‘ only a limited picture of social organization, political organization, or the economic life of the times. Tiahuanaco art is highly impersonal. In all prob- ability, the designs themselves had symbolic meaning, but sound inter- pretation is difficult. The central figure on the ‘‘Gateway of the Sun” has been called an important religious personage, perhaps Viracocha; the orientation of Calasasaya with the movements of the sun has been used to imply a sun cult; the ubiquitous step design has been called symbolic of the mountains which flank the altiplano; but, in reality, there is slight factual basis for such interpretation. 118 SOUTH AMERICAN INDIANS [B. A. EB. BuLL. 143 The harsh environment of the altiplano makes it unlikely that the economic life in the past was very different from that of today. Pota- toes, oca, and quinoa have long been the basic foods. The domesti- cated llamas and alpacas are common today and their presence in the past is indicated by bones, modeled pottery, and painted designs. It is doubtful if the available farm land was much greater in the past, and consequently the population has probably always been distributed in clusters around the general region of Lake Titicaca. The interpreta- tion of the site of Tiahuanaco as a type of Mecca to which pilgrims came to pay tribute in labor and stone carving is feasible, but an alter- native explanation that Tiahuanaco was the center of a large and con- centrated population is not consistent with existing environmental conditions. CHIRIPA The Chiripa style has been found at its type site on the south shore of Lake Titicaca and in a refuse deposit on the Island of Pariti. In spite of the limited discoveries, Chiripa is represented by a distinctive house type, a village pattern, characteristic ceramics, and numerous bone tools, stone artifacts, and metals. No Tiahuanaco influence is seen in these materials. Stratigraphic excavations have placed the Chiripa Period as definitely pre-Decadent Tiahuanaco, but whether it comes before or after the Classic Tiahuanaco is still not established. In any case, Chiripa is of considerable interest, since if it is shown to be post-Classic, it would demonstrate that foreign styles could exist in a region dominated by Tiahuanaco; and if it is considered to be pre- Classic, it represents an important Early Period. The Chiripa village consists of a number of rectangular houses, built corner to corner around a circle. At the type site, 3 houses were un- covered and an estimate was made that a total of 14 formed the village circle. All the doorways face toward the central plaza. ‘The house walls have 3 feet (1 m.) of foundation made of small stones packed in clay. The walls were built up higher with rectangular adobes and roofed with sticks and grass. All walls are double, and the hollow space between served as storage bins entered from the inside of the house by means of rectangular windows with adobe lintels. The door- steps are paved with flagstones, and the entrance itself has a double thick wall with a slot for a sliding, panel-type door. The circular vil- lage plan and the double-wall house type have not been found else- where in Pert and Bolivia. This is surprising, since house walls with an air space between the two facings would seem exceedingly practical for warmth in the cold altiplano. The houses rested on a low, raised ridge below which was about 644 feet (2 m.) of stratified deposit contaiming wash dirt, rough stone walls, a few flexed burials, potsherds, and other artifacts. Under the 119 ARCHEOLOGY OF CENTRAL ANDES—BENNETT VOL. 2] ips str thin but no pottery. some p) These contained flexed burials packed clay floor of one house were 13 graves, some in stone-lined and covered boxes P] and remnants of plain cloth tes that the house was not abandoned after the g indica burials had been put beneath the floor. of gold, a few shell beads, Everyt (After Bennett, 1936, figs. 27, 28.) Figure 11.—Chiripa modeled and painted ware. 120 SOUTH AMERICAN INDIANS [B. A. E. BULL. 143 Ceramics (fig. 11) are represented by fragments, although a few vessels can be partially restored. Similar fragments are found along the house floors and in the strata underneath the houses, which sug- gests that the site was occupied for a relatively long time period. Many of the fragments are from plain brown ware cooking vessels. Most of the painted pieces are from a flat-bottom open bowl with perpendicular sides and a slightly out-flaring thick rim. One restored vessel of this type has a diameter of 12 inches (30 cm.) and a height of 5 inches (14 cm.). Other vessel shapes are rare. The character- istic painting is with a thick yellow pigment on a red-base slip. Black or brown are infrequent additions, and on some pieces the color areas are separated by incised lines. The painted designs are mostly variations on a simple step pattern and triangles. Many fragments have relief decoration in the form of human faces or of pumas with the body in profile and the head in front view. Included with the ceramics are long tubes, burnt black on the inside and decorated on the outside with incised and relief design. These tubes may have been used as fire blowers. Finally, painted clay bricks are found near the houses, with their sides colored green, white, and red. Bone artifacts are numerous, and consist of polished points, needles, buttons, beads, scrapers, chisels, knives, and the like. Some of the scrapers and knives have serrated edges. Stone artifacts include hand grindstones and metates; chipped hoes or scrapers; bolas; and one T-shaped, polished ax. Both shell and stone beads are common. Although metal artifacts are rare, both copper beads and gold strips are associated. At the type site, the houses had burnt down and the debris of the circular village formed a mound with a slight depression in the center. This mound was selected in the Decadent Tiahuanaco Period for the building of a semisubterranean, stone-faced temple. Likewise, Deca- dent Tiahuanaco intrusive graves cut into the old Chiripa house foundations. Thus, the Chiripa Period is clearly established as earlier. ‘The Chiripa refuse layer on the Island of Pariti seemed to cover a grave of the Classic Tiahuanaco Period. Unfortunately, this stratigraphy is not clearly established. The Chiripa style is still limited to these two sites, and consequently its position in the total Peruvian framework remains obscure. At present, the only suggested relationship is with the Pucara style in the South Highlands of Perd. PUCARA In the Department of Puno in the South Highlands of PerG, north of Lake Titicaca, is an important site known as Pucara (pl. 38). Pucara is represented by a characteristic type of stone building, a type of statuary, and a distinctive ceramic style. Although it has as VoL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 121 yet to be thoroughly described, the implication at the moment is that it represents a basic style like Chiripa plus a strong Tiahuanaco influence. At the same time, it cannot be dismissed as another branch of Highland Tiahuanaco since there are too many distinctive and independent features. The masonry employs dressed stone with- out the Tiahuanaco jointing and notching. Stones are arranged in a pattern of large, concentric enclosures with small roomlike divisions. The stone carving includes some statues which resemble those at Tiahuanaco but without the strict conventionalization nor the use of the fine-line incised design. Perhaps all the stone carving does not belong to the Pucara Period since the Tiahuanaco Group Four, or geometric style, is also found. The ceramics (pl. 37) are characterized by the typical Chiripa shape, that is, a flat-bottom open bowl with perpendicular sides and a slightly out-flaring thick rim. However, this is but one of a number of Pucara shapes. Likewise, decorated clay tubes are asso- ciated. ‘The vessels are painted in black and yellow on a red base, and the color areas are commonly separated by incised lines. Simple geometric elements are common, but the designs are dominated by the feline figure, either painted or in low relief. The style is far more developed than Chiripa and seems to have considerable Tiahuanaco influence in it. The distribution and ultimate importance of this site and period must await further exploration and publication. DERIVED TIAHUANACO The Tiahuanaco style materials outside the altiplano of Bolivia do not represent an expansion of the total Classic Tiahuanaco complex. The characteristic architecture, stone sculpture, and even ceramics have not been found outside of the home region. At the same time, the materials from the Eastern Cordillera of Bolivia, from North Chile, and from the Coast of Peri bear an undeniable relation- ship to the Classic Tiahuanaco style. The term “Derived Tiabua- naco”’ is used to express this relationship, and there is a possibility that a period representing such a derived style may be isolated. For example, at the site of Arani near Cochabamba, one group of pottery is well made and utilizes three- and four-color designs on a red base. This group cannot be classified as Decadent Tiahuanaco, but at the same time it lacks many of the typical Classic Tiahuanaco charac- teristics. Likewise, ceramics from the Island of Pariti in Lake Titi- caca vary from the typical Classic Tiahuanaco, in spite of the tech- nical excellence of manufacture. However, more evidence is needed to establish the Derived Tiahuanaco as a definite period. If ever so isolated, it should be placed chronologically as transitional from the Karly to the Middle Periods. 122 SOUTH AMERICAN INDIANS [B. A. E. BULL. 143 THE MIDDLE PERIODS The Middle Periods in the Peré-Bolivia region correspond roughly to the pan-Peruvian Tiahuanaco horizon. Subdivisions and local styles are numerous, as can be judged from the chart (p. 80) and from the glossary of period terminology (p. 74). The principal uniting factor is the influence of the Tiahuanaco style. It has already been mentioned that no known period of Highland Tiahuanaco spread in its entirety to the Coast of Peri. Consequently, it is advisable to distinguish the Coastal style with the designation Coast Tiahuanaco. In this section, the Middle Periods on the Coast of Peri are de- scribed as a whole, although two major subdivisions are recognized, and numerous local variants are mentioned. Following the discussion of the Coast, the Middle Periods of the North Highlands and of the South Highlands and Bolivia are described. THE MIDDLE PERIODS: COAST OF PERU THE SITES In the literature, Coast Tiahuanaco styles are also called Epigone or Epigonal and, in Peri, by the name Andino. Local terms are fre- quently used to indicate materials excavated at a particular site in a valley. The glossary of period terminology explains the various terms. The following list reviews some of the coastal valleys and sites in which Middle Period materials have been encountered. It is reasonable to assume that similar finds will be made in every valley between Nazca and Chicama, but although the local collections confirm this, excava- tion information is not always available. The list starts on the Cen- tral Coast at the site of Pachacamac, not because it is considered to be the center of distribution but because it was one of the first places in which scientific excavations revealed the presence of the Coast Tiahuan- aco materials. (1) Pachacamac.—The ruins of Pachacamac are situated near the mouth of the Lurin Valley. Uhle’s (1903) excavations at the unit called Pachacamac revealed an old temple, made of rough stone, which had later been covered by debris and reused as the foundation for a step-pyramid construction which can still be seen. Graves were found in the soil at the base of the older temple, intrusive in the fill of the covering debris, and at the base of the more recent pyramid. These three grave levels contained materials of three distinct styles, each of which was con- firmed by finds in the cemeteries around the other parts of theruins. Uhle divided the materials from the oldest graves into two periods on stylistic grounds; namely, a style closely related to Highland Tiahuanaco and a style, apparently derived from the first, which he called by the term ‘‘ Epigone.’’ The graves from the middle level contained materials which Uhle designated as post-Epigone. ‘The upper graves contained a coastal variant of the Inca style. The graves of the Tiahu- anaco, Epigone, and post-Epigone styles, all belong to the Middle Periods under discussion here. Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 123 (2) Nieveria.—Local collections in the Rimac Valley contain specimens of the Coast Tiahuanaco style. The collection excavated at Nieveria is no exception, but is, unfortunately, badly mixed. In the published analysis of this material (Gayton, 1927), the strains called Tiahuanaco Epigonal and Nazca-Y definitely pertain to the Middle Periods. (3) Ancén.—The excavated grave materials from the large cemetery at Ancén were divided by Strong (1925) into several periods. Of these, Middle Ancén-I, Middle Ancén-II, and Late Ancén-I belong to the Coast Tiahuanaco Periods. The stylistic division of these three groups is confirmed in part by grave depths and grave stratigraphy. Middle Ancén-I has an Epigonal style associated with a local incised red ware. Late Ancén-I is a Black-white-red Geometric style. In gereral, Uhle’s sequence at Pachacamac is confirmed by the Ancén analysis. (4) Chancay.—At the site of La Calera de Jecuan, 35 graves were excavated which contained various Middle Period styles, all mixed together. Probably more excavation would allow some of these to be distinguished as periods. (5) Supe.—A mixed collection was found at the site of San Nicolas. Careful analysis (Kroeber, 1926 b) allowed this mixture to be divided into two dominant Middle Period styles, although further excavation is needed to confirm this. (6) Viri.—Local collections in the Viri Valley contain quite a variety of Coast Tiahuanaco materials. The excavated sites of Huaca de la Cruz, Huaca Larga, and Taitacantin contained the Black-white-red Geometric style. At Huaca de la Cruz the graves which contained this style of material were intrusive in the Mochica levels. (7) Moche.—On a platform of the Huaca del Sol at Moche, both Coast Tiahuanaco style fragments and graves have been found. Other graves contain the Black-white-red Geometric style. This in part suggests the triple division found at Ancén and Pachacamac. The Huaca del Sol site contained a pressed- relief ware and a style called Queneto associated with the Coast Tiahuanaco. Coast Tiahuanaco pieces have also been found at Chanchan, near Trujillo, but without excavation notes. Local collections, presumably from the ruins of Chanchan, contain several other styles which are probably contemporaneous and related to the Middle Periods. These include Cursive Modeled, Cursive Tripod, Black-white-and-red Recuoid, and Red-and-white Chanchan styles. (8) Chicama.—Coast Tiahuanaco styles are common in Chicama Valley, but the only recorded excavation is at a site called Pata de Burro. The Chicama Valley seems to represent the northernmost limit of distribution of the Coast Tiahuanaco style in its pure form. Influences have been noted in the valleys to the north, but no clear-cut period has been isolated. (9) Chincha and Cajfiete.—South of Pachacamac, the evidence for the Coast Tiahuanaco is more confused. In the Chincha and Cajfiete Valleys excavations have still not uncovered any clear-cut Coast Tiahuanaco styles. (Middle Cafiete, presumably of the Middle Periods, does not show Coast Tiahuanaco influences. See fig. 12.) (10) Ica.— Middle Periods in the Ica Valley have been designated (Kroeber and Strong, 1924 b) as Epigonal, Middle Ica-I, and Middle Ica-II. The ceramics of all three of these are vaguely related to the Coast Tiahuanaco. ‘The textiles, however, show a much closer relationship. (11) Nazca.—The Middle Ica-I and Middle Ica-II styles are also found in Nazca Valley in the sites excavated by Uhle. The Ica Epigonal is replaced by the Nazca-Y style, which seems to be a vague transition from the earlier Nazca Period to the Coast Tiahuanaco. Other sites in the Nazca Valley, however, contain excellent Coast Tiahuanaco material. The site of Pacheco, although not adequately described, is known to have large urns and other vessels with complex painted designs clearly of Coast Tiahuanaco style. [B. A. E. BULL. 143 SOUTH AMERICAN INDIANS 124 SO QO OL rater, O O OSS . (Re- te Valley ane fig fiete pottery from Cerro del Oro, C iddle Ca drawn from Kroeber FicurE 12.—M 3 (3, 3 ; 74, fig. a ; 73, i 75, fig. e ? , pls. 71, fig. 2 1937 b VoL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 125 The Coastal Middle Periods are divisible into two major subperiods, as well as numerous minor styles and local variations. The first of these major subperiods, called here the Coast Tiahuanaco-A, bears a definite relationship to Highland Tiahuanaco and has a fairly consist- ent spread up and down the entire Coast. The second major sub- period, here called the Coast Tiahuanaco-B, might equally well be named the Black-white-red Geometric style. Unlike the first sub- period, the Coast Tiahuanaco-B seems to be a coastal development with no direct relationship to Highland Tiahuanaco. In a number of places, such*as*Moche, Ancén,'and Ica, the Coast Tiahuanaco-B is separated from the Coast Tiahuanaco-A by intervening styles or periods. Nonetheless, ;recent excavations by Willey (Strong, Willey, and Corbett, 1943) reveal an unbroken sequence of development. The over-all picture is of a dominant style, definitely related to Highland Tiahuanaco, which spread up and down the Coast of Pert. From the beginning, it mixed with local styles and tended to lose its unity. Somewhat later, a local formulation of the earlier style, characterized by Black-white-red Geometric design, in turn spread along the Coast of Peri. Local Middle Period styles developed or intruded in many places on the Coast. Some are contemporaneous with the early, or Coast Tiahuanaco-A, some are contemporaneous with Coast Tiahuanaco-B, and some fall in between the two. Some of these local styles can be satisfactorily placed. Nazca-Y (pls. 24, bottom; 25, d, e, g,) and Ica Epigonal (pl. 44, a, 6) are contemporaneous with the Coast Tiahuanaco-A. The post-Epigone style at Pachaca- mac and Middle Ancén-II fill the interval between A and B (pl. 40, a-c). The local styles found at Chanchan and the Huaca del Sol seem to be in part contemporaneous with Coast Tiahuanaco-A and in part in the intermediate period. Everywhere, the Coast Tiahuanaco represents a mixture, both with earlier Coastal styles and with outside influences from such areas as the North Highlands. This makes the description of the characteristics of the major subperiods exceedingly difficult, as the following résumés indicate. COAST TIAHUANACO-A The Coast Tiahuanaco-A is represented by the Tiahuanaco and the Epigone styles at Pachacamac (pl. 39), by Middle Ancén-I, and by the Pacheco site in Nazca Valley (figs. 13, 14). Although not as clearly isolated, the Epigonal styles at Ica, Nieveria, Supe, and Moche can also be included. In general, Coast Tiahuanaco-A is characterized by the following features: (1) Ceramics with the basic colors of black and white on a red-slip base. (2) Some vessels or fragments with four or more colors on ared-slip base. (8) Some vessels or fragments with polychrome 126 SOUTH AMERICAN INDIANS. [B. A. E. Bun. 148 (Drawn from a, Pacheco bowl; b, Pacheco (?). American Museum of Natural History specimens.) \ | 2m, - i 2 mt AN NaS ee Oz TAN MAN Ce SRE AS i =) ONE At a Cy ‘ = (3) i } Gn | Se a) iilsh Wis alsSy| Figure 13.—Coast Tiahuanaco-A style pottery in the south. (Moa di i ie «| Ve a Pw = \ eS santihlt TI ell fl \ ° \\\" wD YZ To A NT , iff sul . L 6 (ea (i yy ZZ} iol tu ica XK S : shih TONZ om ESS cuate S — Sess SS = =< See ike > SS; ae CS ‘ SSs Se SS ey CS apie SSS FR 5 ile Sic ~ 0 a9 2 2 eae) aS) ®o # SeS n > 3 Ze CS nN 5 o fet fy =} {99} oF co) Sates s&s ae SS m8 Ins} =: we 2 Ei op) fa} » OV Ss =| 1s @) a eta © > | @ = PLATE 17 Top ) polly Ae 1943 b] (After Tello PLATE 18.—Coast Chavin artifacts, structures, and sites. Top: a, c, Coastal Chavin or Cupisnique style vessels from Chongoyape, Lambayeque Valley; b, Coastal Chavin stone bowl, same provenience. (Ves- sels are 814 inches (21.5 em.) high.) (Courtesy Museum of the American Indian, Heye Foundation.) Center: Structure and altar(?) at Aspero, Puerto Supe. Bottom: Aspero site looking toward the valley, ruins of above structure at center of picture. (Courtesy Department of Anthropology, Columbia Uni- versity.) PLATE 19.—Gold objects of the Chavin and Paracas Necropolis styles. Top: Paracas Necropolis ham - (Courtesy American Museum of Natural History.) Bottom (left): Gold wrist mered gold ornaments. ornament, Chongoyape Chavin style, Lambayeque Valley, Pert. Bottom (right): Ear ornaments, Chon- (Cylinder is 1034 inches (27 em.) high; ear ornament goyape Chavin style, Lambayeque Valley, Pert. é is 544 inches (13 cm.) in diameter.) (Courtesy Museum of the American Indian, Heye Foundation.) G Puare 20.—Paracas Cavernas style pottery. (a, b, d, e, Courtesy Museo Na- cional, Lima; c, photo of cast, Courtesy American Museum of Natural History.) 6 PuaTE 21.—Paracas Necropolis and Cavernas style pottery and Necropolis mummy bundle. a, b, c, Necropolis style; d, typical Cavernas style; e, two- color negative style of Cavernas; f, partially unwrapped Necropolis mummy bundle. (a—-e, Courtesy Museo Nacional, Lima; f, Courtesy American Museum of Natural History.) PLatTE 22.—Paracas Necropolis textiles. Top: Section of a shawl with poly- chrome embroidered warriors carrying trophy heads. Bottom: Shirt with over- all embroidery design. (Courtesy American Museum of Natural History.) PuiatEe 23.—Aerial photos, vicinity of Pisco, Pera. Top: Strange belt of pock- markings on ridge. Bottom: Ancient irrigation system in Pisco Valley, cultural affiliation unknown. (Courtesy James Sawders.) PuatE 24.—Nazca siyle pottery. Top row, Nazca A style; second row, Nazca X style; third row, Nazca B style (this includes tall jar) ; panpipes, probably Nazca B; two bottom vessels, Nazca Y style. All from Nazca Valley except panpipes, which is from Ica Valley. (After Gayton and Kroeber, 1927, pls. 1, 6, 11, 8, 7, 14, 13; panpipes after Kroeber and Strong, 1924, pl. 29.) PLatTE 25.—Nazca style pottery. a-—c, f, Nazca A style; d, e, g, Nazca Y or Epigonal (Tiahuanaco influenced) style. (After Schmidt, 1929, pp. 335, 331, 337, 330, 330, 342, 330.) 7. PLATE 26.—White-on-red, Intermediate and Interlocking style pottery from the Chancay Valley. a, 6, White-decorated type of White-on-red style; c, d, White-zoned type of White-on-red style; e, Three-color Intermediate; f, g, Interlocking style. (After Willey, 1948, pls. 1, 6.) Puate 27.—Interlocking, White-on-red, and Intermediate style pottery from the Chancay Valley. a, Interlocking; b-e, g, White-on-red; f, three-color Intermediate. (After Kroeber, 1926, pls. 88, 90, 86, 89.) PuatrEe 28.—Pachacamac Interlocking and coarser class of Early Lima style (Nieveria). a-—c, Interlocking vessels. (After Strong and Corbett, 1943, pl. 4.) d-g, Early Lima vessels. (After Gayton, 1927, pl. 93.) PuatTe 29.—Finer class of Early Lima style pottery from Nieveria and Pacha- camac. 6, g, Most suggestive of Interlocking; h, 7, Coast Tiahuanaco in- fluenced. (After Gayton, 1927, pls. 91, 94, 95; h, 7, after Schmidt, 1929, pp. 273, 272.) e | | Soo see nae Puate 30.—Cultural representations in Mochica and Chimu pottery. a, c, Fishing from balsas; }, d, f, g, house types; e, torture to death. Proveniences as follows: a, Unknown; 6, c, Chimbote; d—f, Trujillo; g, Chicama Valley. d is Middle or Late Chimu; others are Mochica. (After Schmidt, 1929, pp. 155, 157, 156, 158.) Puate 31.—Mochica and Early Chimu pottery. a, Warrior; b, nude prisoner; c pottery figurine of woman with infant; d, sleeping man; e, portrait jar of man with mutilated nose and lips; f, Mochica jar retaining Coast Chavin influence,, g, man shooting birds with blowgun. (After Wasserman-San Blas, 1938, figs. 367, 329, 360, 372, 264, 473; f after Schmidt, 1929, p. 211.) Se Puate 32.—Archeological structures of the North Peruvian Coast. Top: The Temple of the Sun at Moche, a Mochica pyramid. Bottom: Great wall of Santa Valley, cultural or period association uncertain. (Courtesy James Sawders,) PuatEe 33.—Gallinazo Plain and Two-color Negative from Virtii Valley Cemetery a—c, e, Negative; d, f, Plain. (After Bennett, 1939, fig. 13.) ST ‘cannon RIRRERIA EOI OORASROEON OST SSERSCESOSS TSEC SSS ESOT SSESSSNGNAIO TN oe "0 PuatEe 34.—Recuay pottery and stone sculpture from the Callejoén de Huaylas. a-c, Recuay Three-color negative style. (After Schmidt, 1929, pp. 241, 235, 232.) d-g, Stone sculptures from the vicinity of Aija, Ancash. (Courtesy Donald Collier.) Cece “4 “6261 “Jprayog 1OJJV) “Aoywod oovuBNnyeLy, OIsse[D :7ybr waddy) (‘StopMeg sourvp ASoJANOD) “UNG oY} Jo ABVMOZRYH ‘RIATOG ‘oovuRNYyRIE—'ce aALVIg . i= SS ae a \ \ MDHOA AK SI PLATE 36.—Tiahuan- aco, Bolivia. Top: Classic Tiahuanaco pottery. (Courtesy American Museum of Natural History.) Right: Carved Clas- sic Period mono- lith. (Courtesy Grace Line.) g Puate 37.—Pucara style pottery. a, Hollow-based bowl; b, d, sherds with cats’ heads; c, f, fragments of pottery tubes; e, g, anthropomorphic head vessels. (Courtesy Peabody Museum, Harvard University; c, f, Museo Nacional, Lima.) PuatTE 38.—Pucara architectural features and stone sculpture. Top (left): De- tailed view of Temple. Top (right): General view of the site. Center and bottom: Pucara style stone sculpture. (Courtesy Peabody Museum, Harvard University.) PuLatTE 39.—Coast Tiahuanaco-A style from Pachacamac. (After Schmidt, 1929, pp. 295, 285, 283, 276, 271.) h Puiate 40.—Coast Tiahuanaco-A and -B styles from the Central Coast. a—c, Post-Epigone beakers; d, Middle Period pressed ware from Huacho; e, h, Black, white, red geometric styles from Ancén; f, g, Epigonal and Black- white-red Geometric styles from Huacho. (After Schmidt, 1929, pp. 296, 228, 229; b, e, h, after Willey, 19438, pl. 1.) Puate 41.—Black-white-and-red and black ware styles of the North Coast Middle Periods. From Huaca de la Cruz and Taitacantin. a-—d, Black-white- and-red; e-g, Black ware. (After Bennett, 1939, figs. 9, 10.) Puatr 42.—Pottery from a Lambayeque site. A style intermediate between the Middle Periods and Late Chimu. a, d, e, Black ware; b, c, Orange and Black-and-orange. (After Bennett, 1939, figs. 19, 20.) PuatTe 43.—Textiles in the Coast Tiahuanaco style. Top: Fragment of an em- broidered garment. Bottom: A hat in velvet technique. (Courtesy John Wise, New York.) Sk g Puate 44.—Early Ica, Middle Ica, Late Ica, and Coastal Inca pottery styles. a, b, lea Epigonal or Early Ica; c, Middle Ica 1; d, e, Late Ica 1; f, Inca; g, Late Ica II. (After Kroeber and Strong, 1924, pls. 30, 31, 36, 40; g, after Schmidt, 1929, p. 305.) Puate 45.—Middle Ica II style pottery from Ica Valley. (After Schmidt, 1929, pp. 304, 307, 316, 313.) PiLatE 46.—Middle and Late Coastal Period textiles. Top: Late Period textile detail, double cloth. (Courtesy American Museum of Natural History.) Bottom: Middle Ica shirt. (Courtesy John Wise.) y Black-on-white and White styles from the (After Schmidt, 1929, pp. 251, 245, 250.) Puate 47.—Late Chanca Chancay Valley PLATE 48.—Great Late Period structures of the Central Coast. To p: View of Pachacamac, looking from the Temple of the Sun. (After Strong and Willey, 1948, pl. la.) Center: Cajamarquilla, city of the Rimac Valley. (Courtesy Grace Line.) Bottom: Aerial photo of Fortress of Paramonga or La Fortaleza at Pativilea. (Courtesy National Geographie Magazine; photos by Capt. H. B. Grow and Peruvian Naval Air Force.) | w~ ie) —_, S . m— AN else figs ’ a—c, From El Brujo, Chicama Valley (After Bennett, 1939 ttery. Black ware from a Lambayeque site. PuatE 49.—Late Chimu po PuateE 50.—Late Chimu goldwork from Huarmey. Tov: Gold collars and plumes. Bottom: Stirrup-spout vessel, goblet, and plate. (Courtesy American Museum of Natural History.) OAT j eee ee J appHe tile Date ey Zs % ‘et tle WOW? My Z oo “ Ye WY imi Yb | Puate 51.—Chanchan, Pera. Top: Aerial view showing extent of a part of the | great Late Chimu city. Bottom: Aerial close-up of one enclosure at Chanchan. (Courtesy James Sawders.) PuatTE 52.—Archeological structures of the North Peruvian Coast. Top: A ‘temple in the Chicama Valley with remnants of an ancient highway crossing the desert at the left. Bottom: Wall arabesque at Chanchan. (Courtesy James Sawders.) NeOR BRU Baga (Courtesy 4 st of Pert. tory.) 1s RNS ARN RT eye setopinennars Nor nee iod of the Coa American Museum of Natural H PuatE 53.—Mummy bundle from Late Per Tos 14.) b ? b) Crd 12 ) b b ) figs. 10, 11 ae) d and modeled type d 1se , punctate type. (After Rowe, n ame, h d type; -1ncise Puate 54.—Chanapata pottery. a, e, Black-inc Black PuatTE 55.—Peruvian weaving equipment of the Late Coastal Periods. Top: Workbasket containing small weaving implements. Bottom: Double-cloth loom. (Courtesy American Museum of Natural History.) PuaTe 56.—Peruvian wood carving. a, b, Digging-stick handles from Trujillo (Late Chimu); c, d, Digging-stick handles from Huacho (Mid- dle or Late Periods) : e, Idolinlaid with shell, from Pachacamac (period?) ; f, Oar or rudder handle, Ica (Middle or Late Periods). (Height of eis 10% inches (27.5 cm.), other objects not to same scale, but range in height from 3) inches (8.8 cm.) to 6 inches (15.3 cm).) (After Schmidt, 1929, pp. 428, 414, 429.) qd PuatEe 57.—Metal tools and ornaments from prehistoric Perf. a, Mace head of copper or bronze; b, knife of copper or bronze; c, gold tweezers; d, gold tweezers with handle; e, silver crown or headpiece. Specimens a, b, Inca style, from vicinity of Tiahuanaco; c, d, from Ica (period?); e, from Chimbote (period?). (Respective heights: a, 5 inches (12.5 cm.); 6, 6 inches (15 cm.); c, 2 inches (5 em.); d, 4% inches (11.2 em.); e, 13% inches (84 cm.)). (After Schmidt, 1929, pp. 388, 369.) | te | | Cc PLATE 58.—Metal and stone objects from prehistoric Peri. a, Copper mask with shell and pyrites eye and shell teeth; 6, c, gold beakers; d, e, human figures of bronze; /, bronze cat; g, stone receptacle with serpent relief. Specimen a, from Chicama Valley (probably Early Chimu); b, from Ica (probably Middle . Period); c, provenience unknown; d-g, vicinity of Cuzco (probably Inca). (Respective heights: a, 744 inches (18 em.); 6, 234 inches (7 em.); c, 6% inches (17.5 em.); d, 54 inches (13 em.); e, 434 inches (12 em.); f, 238 inches (6 cm.); g, 238 inches (6cm.).) (After Schmidt, 1929, figs. 375, 372, 373, 378, 459.) PuatE 59.—Miscellaneous Peruvian archeological objects. Top: Musical in- struments. Bottom: Ty of sandals. (Courtesy American Museum of Natural History.) Airenters PuatE 60.—Peruvian whistling jars. Top: Late Chimu style. Bottom: Actual cross section and diagrammatic sketch of a whistling jar. (Courtesy American Museum of Natural History.) Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 127 designs clearly related to the Highland Tiahuanaco style. (4) Some textiles, usually tapestries, with designs which resemble the stone- carving style at the site of Tiahuanaco. (5) Flaring-sided goblets and flaring-sided cups, or variants of these. (6) Double-spout vessels and face-collar jars. (7) Ceramic design elements, such as puma and condor figures, tridents, steps, scrolls and the like. =(hi p =i Y/R Ze Fe ig) a IW AeA oA ‘ ae Sein ; FicurE 14.—Coast Tiahuanaco-A style vessels. (Drawn from George Hewitt Myers Collection, Washington, D. C.) Other characteristics can be added to the above list for specific local areas. For example, large, thick urns are a Coast Tiahuanaco-A type on the South Coast. Tapering-spout jars with rounded handles, skull bowls, and yarn bowls with two flat, horizontal handles are asso- ciated on the Central Coast. For the North Coast, pressed-reliet design and black ware can be added. The greatest unity of Coast Tiahuanaco-A is found in the textiles 595682—46——11 128 SOUTH AMERICAN INDIANS [B. A. E. Bui. 148 (pl. 43). Weaving art is well developed and includes most of the known techniques. Intricate and ambitious tapestries are the most characteristic. Other frequent techniques are: brocade; warp-pat- tern weaving; double-cloth; painted cloth; both square and flat braid- ing; velvet or pile-knot, particularly in the form of squared hats; interlocking warp; knitting; some embroidery; and tie-dye patch- work. Red and yellow are the typical yarn colors, plus tan, brown, orange, and blue. The textile designs show a close relationship to Highland Tiahuanaco. Pumas, condors, human heads, and human figures are remarkably similar to the stone-carving designs on the “Gateway of the Sun” and onstatues. In fact, the textile patterns furnish one of the best keys for the relationship of Coast and Highland Tiahuanaco. No buildings have been associated with the Coast Tiahuanaco-A subperiod except a rough stone wall at Pachacamac. Presumably, other buildings at that site, as well as parts of Chanchan and others were built during the period. So far, Coast Tiahuanaco-A materials have been found mostly in cemeteries and rarely in habitation sites or in positive association with buildings. At Pachacamac, the burials are in the form of bale-shaped mummies with false heads of metal, wood, clay, or painted cloth. The bundles are found in conical or cylindrical chambers lined with stones or adobes and covered with stone, cane, matting, or wood. On the South Coast, the large urns might have served for burial. Also underground cham- bers, covered with logs and hung with tapestries, have been reported to contain bundle burials of the Coast Tiahuanaco-A style. Little attention has been paid to other artifacts associated with the Coast Tiahuanaco-A style. At Pachacamac are spindle whorls of clay, bone, and stone with incised designs; wooden staffs; bone spatulas with carved heads; shell, bone, and teeth amulets; and small wooden scales. There is no mention of metals, although presumably copper, bronze, gold, and silver were utilized. At Ancén, solid clay figurines are associated. COAST TIAHUANACO-B Coast Tiahuanaco-B materials have been found at Pachacamac, although not clearly isolated; in Late Ancén-I; at Chancay; at Huaca de la Cruz, Taitacantin, and Huaca Larga in Virt Valley; in one iso- lated grave at the Huaca del Sol; and as a stylistic influence in Middle Ica-IT in the Ica and Nazca Valleys. The style is characterized by geometric designs painted in black, white, and red. The typical vessel shapes are: round-bottom cups, annular-base bowls; squared stirrup-spout jars; face- or plain-collar jars with two small handles at the base of the collars; and flaring-collar bowls with two flat rim-to-body handles. VOL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 129 Other characteristics are difficult to discover, since most of the ma- terials are from graves. Inspite of the wide spread of the design style, little political unity is implied. In general, the Coast Tiahuanaco-B is transitional to the true Late Periods and shares characteristics with them. COAST-HIGHLAND TIAHUANACO RELATIONSHIPS In Highland Tiahuanaco four periods are recognized: Early Tiahua- naco, Classical Tiahuanaco, the hypothetical Derived Tiahuanaco, and Decadent Tiahuanaco. Coast Tiahuanaco has been divided into two subperiods. The discussion which follows is limited to the Classic Tiahuanaco and the Coast Tiahuanaco-A, since only these two show any marked resemblances. Only two of the typical Highland Tiahuanaco ceramic shapes are found on the Coast: the goblet and the cup. Even in these shapes the Highland forms have flaring sides, the Coast forms generally have straight sides. Other Classic Tiahuanaco ceramic shapes do not occur on the Coast, nor do other shapes associated with the Coast Tiahua- naco-A occur in the Highlands. The ceramic design colors, on the other hand, are closely related, siace both Coast and Highland Tiahua- naco employ black, white, yellow, and gray on a red-slip base. Like- wise, fine finish and polish are characteristic of both. The textile designs of the Coast Tiahuanaco-A Period are in many cases almost identical with the stone-carving designs of Highland Tiahuanaco. The ceatral front-view figure of the “Gateway of the Sun,’ the running figures on the side, and the small sun faces all occur on Coast textiles. Were textiles preserved in the Highlands, they would probably be even more similar to the Coast fabrics. In Highland Tiahuanaco, a distinction can be made between ceramic design and stone carving-textile design. This distinction is by no means maintained on the Coast, where the characteristic ceramic designs of the Coast Tiahuanaco-A are closely related to the textile designs. For example, a large urn from Pacheco is decorated with a front-view figure with both hands outstretched and holding staffs, like the central figure on the “Gateway of the Sun.”” Examples could be multiplied to show that the textile-stone carving design is an important factor in the relationship of Highland and Coast Tiahuanaco. Al- though this helps explain many of the similarities, it does not account for the two corresponding ceramic shapes, or for the ceramic design colors which are typical of both. Black and white, plus other colors, on an over-all red base are not colors derived from the textiles. Many attempts have been made to explain the relationship of High- land and Coast Tiahuanaco, but none is completely satisfactory. One is that another site must exist somewhere between the Highlands and the Coast with materials which would account for the similarity. . This 130 SOUTH AMERICAN INDIANS [B. A. E. Buty. 143 is a good suggestion, but the hypothetical site has yet to be found.2?, A second is that such well-known sites as Pachacamac and Ancén are not the real centers of contact of Highland and Coast Tiahuanaco. Again this hypothetical coastal center which would explain all has not been discovered. A third, already mentioned above, attempts to explain the relationship in terms of the transfer of textiles and their designs. As has previously been mentioned, this does not explain the total ce- ramic situation. A fourth reverses the process and claims that Coast Tiahuanaco is a local outgrowth of Nazca, which then spreads to the Highlands. According to this explanation the Highland Tiahuanaco plain-stone masonry was later influenced by Coast textile design. It is true that the intricate, fine-line incision on Highland Tiahuanaco stonework is amazingly incongruous. On the other hand, the stages of development from Nazca to Coast Tiahuanaco are still hypothetical. The pan-Peruvian Tiahuanaco horizon does not seem due to a mass migration from the Bolivian Highlands to the Coast. At least, the differences are greater than one would expect under such circumstan- ces. Furthermore, Highland masonry and stone carving are not found on the Coast. Archeological evidence does not even indicate that the Coast Tiahuanaco-A was a politically united empire, such as the Later Inca Period. An alternative to political unity might be unity in religion. The Highland Tiahuanaco type site has frequently been interpreted as a religious center. The spread of such a domi- nant religion with its conventionalized design symbols would account for much of the Highland-Coast Tiahuanaco relationship. It would explain in part the spread of the design style without the accompany- ing evidence of political unity. However, these questions still remain unanswered. THE MIDDLE PERIODS: NORTH HIGHLANDS The Peruvian Andes have furnished little evidence for the distri- bution of Highland Tiahuanaco culture. Dressed-stone masonry and stone carving are found in many places, but the relationship with Tiahuanaco is not specific. In spite of a considerable amount of exploration and excavation in the Central Highlands near Cuzco, no Tiahuanaco Period has yet been isolated. In the North Highlands, Chavin is frequently compared with Tiahuanaco, but whatever rela- tionships exist are certainly not the result of a direct spread of a Bolivian culture. Recuay too has stone buildings and stone carv- ings which bear some resemblance to Highland Tiahuanaco, but do not suggest direct contacts. In the Callején de Huaylas in the North Highlands, some ceramics are clearly related to the Coast Tiahuanaco in shape and design. These include flaring- or straight-sided goblets, round-bottom cups, ? Recently, Dr. Tello has uncovered a site near Ayacucho which may well be the key to the problem. Vow. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 131 double-spout jars, and face-collar jars. All are decorated with Coast Tiahuanaco designs, painted in three or four colors on a red-slip base. Local collectors claim that these ceramics are found in dressed-stone tombs built up above ground. WILKAWAIN-TIAHUANACO Near the town of Huaraz at the ruins of Wilkawain, Coast Tia- huanaco-influenced materials have been found in stone-lined box graves under large boulders; in house-site refuse; and in stone-lined and covered tombs, 8 feet (2.5 m.) deep. Although a number of styles are represented, the materials clearly belong to the same period, which has been named Wilkawain-Tiahuanaco (fig. 15). FicgureE 15.—Two vessels of the Wilkawain-Tiahuanaco style. (Drawn form - specimens in the National Museum of Lima.) The different ceramic styles which are found together in the deep tombs are: (1) Polished red ware collar jars, open bowls, plates, and modeled monkeys; (2) black ware straight-sided goblets, double bowls, and double-spout vessels, all without pressed-relief design; (3) two-color negative ware high-collar jars and shallow plates; (4) orange ware two-handled ollas, goblets, plates, and sieves; (5) painted ware with from two to four colors, always on a red or orange base. The designs of this painted ware are somewhat similar to the Coast Tiahuanaco of Middle Ancén-I and Nazca-Y. The Wilkawain-Tiahuanaco is associated with stone building, not only in the stone-lined graves but also in one-room, above-ground house sites. Three-story temples with projecting cornices and carved- stone puma heads set in the walls may also be part of this period. Walls are made of split stones laid in horizontal rows, alternating thick and thin. No dressed stone is employed, but the masonry technique is reminiscent of Chavin. However, this technique is com- 132 SOUTH AMERICAN INDIANS [B. A. E. Bun. 143 mon in the North Highlands and does not necessarily imply direct relationship to the earlier period. Copper pins, a wooden point with a gold-leaf overlay, stone beads, hammerstones, and chipped flint and obsidian points are all associated with the Wilkawain-Tiahuanaco culture. Perhaps the carved stone puma heads on tenons should also be included. This North Highland extension of Coast Tiahuanaco influence sug- gests that contacts were frequent. It is also known that the North Highland influence was affecting the Coast at about the same time. Such exchanges aid greatly in the cross-dating. MARANON The little-known Marafién style in the North Highlands is generally considered as pertaining to the Middle Periods. It is characterized by shallow bowls and tripod plates, painted on the inside in a cursive style. The colors are variants of brown, red, and tan, and the prin- cipal designs include a dragonlike animal and elaborate geometric elements such as steps, scrolls, frets, and diamonds. Similarly deco- rated fragments have been found around Cajamarca, and the style is sometimes named after that town. Occasional pieces have been found on the North Coast of Peri (fig. 16) and in the vicinity of Huaraz. MIDDLE HUAMACHUCO In the far North Highlands, near the towns of Huamachuco and Cajabamba, three cultural periods. are distinguished. The earliest is found at the ruins of Marca Huamachuco and designated as a Mid- dle Period style. Massive stone walls of pirca construction, rising as much as three stories in height, are identified with this period. Stone sculpture is also included. There are crested animal heads with tenons, and likewise human heads with earplugs and ornamented headdresses. Blocks are carved with relief geometric and animal patterns. The ceramics may be plain red slip, two-color negative, or white paste with red and black cursive designs. The latter are similar to the Marajién style of ceramics. THE MIDDLE PERIODS: SOUTH HIGHLANDS AND BOLIVIA DECADENT TIAHUANACO Following the spread of Tiahuanaco, there is a return to local, re- gional styles. In Bolivia, a Decadent Tiahuanaco Period can be isolated which, although related to the Classic Tiahuanaco, is technically in- ferior. Whether Decadent Tiahuanaco should be classified as a Middle Period or a Late Period is uncertain. It is clearly pre-Inca and post-Classic in the stratigraphic excavations. Likewise, it has no definite connections with the Coast Tiahuanaco Periods. Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 133 Decadent Tiahuanaco represents a break-down of the Classic style. Classic designs are usually complete; Decadent designs either leave out parts of the figure or use such elements as a head, face, or tail feathers for the total design. Classic-design figures, in spite of their rigid conventionalization, seem to have meaning; Decadent are used as pure design and may be placed upside down or combined. In ese ger (es ES NL MOERT ELIZ noe Taal te ve e i 0 \" ' 1 Eg MO Figure 16.—Interior design of tripod bowl, Marafién style. (Drawn from specimen in Rafael Larco Herrera Museum, Chiclin.) the Classic style, geometric elements are usually combined with figures; the Decadent style uses many geometric elements by themselves, such as steps, perpendicular and wavy lines, double S’s, horizontal wavy lines, angular and curved scrolls. Decadent designs are carelessly executed, and the colors are dull and limited to black, white, and red. One or two colors on an orange, red, or brown slip are typical. 134 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 FIGURE 17.—Shapes and designs of Decadent Tiahuanaco style pottery. a-a’, Cochabamba style kero shape with line and dot design; b-b’, rounded-base cup with interlocking design with circles; c-c’, two- handled jug with step design; d-d’, open, rounded base bow] with black-on-red scroll design; e-e’—-e’”’, cup, with two samples of design, one a depleted condor-puma combination, the other an elongated human face in profile; f, kero-shape goblet with Decadent human-face design. (After Bennett, 1934, fig. 17.) Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 135 Some of the Classic Tiahuanaco vessel shapes carry over, such as the goblet, the two-handled libation bowl, and the the flat-bottom cup. Even in these, new variations are found, such as straight rather than flaring sides, ornamental raised bands, and libation bowls without the modeled puma and llama heads. Several new shapes are introduced, including collar jars, angular-body vessels, spouted ves- sels, and one- or two-handled jars. In total, the Decadent Tiahuan- Figure 18.—Tiahuanaco stone statues at Mocachi, Bolivia. (Redrawn from Casanova, 1942 ¢, fig. 3.) aco vessel shapes are more varied than those of the Classic Period (fig. 17). None of the important buildings at the Tiahuanaco site has been assigned to the Decadent Period, although the small temple east of Calasasaya probably belongs to it. This temple is a semisubter- ranean structure with facing walls made of upright stones and small stones, and has a variety of carved stone heads as wall decorations. Some of the stones are dressed, some rough, and all of them, including the carved heads, were probably collected from older buildings and reassembled in this crude temple. Similar semisubterranean temples have been associated with the Decadent Tiahuanaco at Chiripa and Lucurmata. Nostone carving is definitely associated with this period, although it is possible that some of the technically inferior carvings might be assigned to it (fig. 18). 136 SOUTH AMERICAN INDIANS [B. A. E. BuLL. 143 Decadent Tiahuanaco does not suggest the strong social or religious organization of the Classic Period. It is found at many sites in the altiplano of Bolivia and southern Pert, but the influence probably carried no farther afield. The characteristic Decadent Tiahuanaco had disappeared before the arrival of the Inca. THE LATE PERIODS Following the pan-Peruvian Tiahuanaco horizon, local styles emerge in the major geographical regions of Peri and Bolivia. Although most of these are distinctive styles, the break with the Middle Periods is not a sharp one. In many areas, the transition from Middle to Late is easily traced. Some of the Late Periods have rather wide distribution, but none can be called pan-Peruvian. Chronologically, the Late Periods fall between the Tiahuanaco and the Inca horizons. The Jnca ultimately incorporated the whole region of Peri and Bo- livia in their political Empire, but even before that was achieved Inca influence can be seen in the closing phases of the Late Periods. Description of the Late Periods again follows the geographic se- quence of South, Central, and North Coast, and North and South Highlands. The Central Highlands were apparently dominated by the developing Jnca civilization during the time of the Late Periods. THE LATE PERIODS: SOUTH COAST ICA The valleys of the South Coast had reverted to local cultural styles even during the Middle Periods. Although such styles as the Middle Ica-I (pl. 44, c) and Middle Ica-ITI (pl. 45) in the Nazca Valleys can be correlated with Middle Periods elsewhere, their local character is outstanding. The Late Period on the South Coast, called Ica or Late Nazca, is in reality a development and continu- ation of the trend started in the Middle Periods. The sequence established for Ica Valley includes Middle Ica-I, Middle Ica-II, Late Ica-I (pl. 44, d, e), Late Ica-ITI (pl. 44, g), and Inca (pl. 44, f). These periods are not sharply distinguished, but change gradually from one to another. The Ica style is found in the Cafiete, Chincha, Ica, and Nazca Val- leys, and there are some suggestions of its extension much farther to the south. In fact, some authorities speak of a dominant Chincha (or Ica) culture which mixed with southern elements to form a Chin- cha-Atacamefio. 'This combined culture presumably had a wide spread in southern Peri and Northern Chile. Most authorities, however, await further evidence before accepting the Chincha- Atacamefio concept. Vou. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 137 Most surface ruins in the South Coast valleys are assigned to the Ica or Jnca Periods. Although this suggests more people and better building organization than in the Early Periods, at no time could these valleys have supported truly large populations. Tambo Colo- rado in the Ica Valley and La Centinela and Tambo de Mora in Chin- cha Valley are typical of Ica Period ruins. These consist of com- binations of pyramids, stairways, terraces, and courts. True towns are rare, and the houses associated with the outstanding ruins seem to have been either the residences of important people or ceremonial rooms. Squared, hard-baked adobes are common in the construction, and many walls are made of puddled clay or tapia. Some of the buildings have typical Inca wall niches and narrow-topped doorways. Some walls are decorated with colored plaster, with frets made by arrangements of adobe bricks, and rarely with clay arabesques. Cemeteries are located near the ruins and in the hot sands. Two grave types are reported: a shaft grave with a chamber opening to one side and a roomlike chamber grave with a roof of leaves supported on poles. The burials were wrapped in leaves, skins, and cloth and bound with rope. Ceramics and other artifacts were placed in the graves. Some skulls show artificial occipital flattening. Ica ceramics can be subdivided on the basis of the presence or ab- sence of Inca influence. Late Ica-I Period is more or less a continua- tion of Middle Ica, while Late Ica-II shows the Jnca influence. Late Ica-I ceramics are characterized by the use of red, white, and black colors, and the predominance of textile pattern designs. Some design units are frankly geometric, such as series of diamonds or squares, and others are highly conventionalized birds or fish, arranged inrows. Most designs are found in the woven fabrics as well as on the ceramics. Open bowls with angular sides, rounded bottoms, and beveled rims are typical. Other common shapes are constricted, flaring-collar jars, and a long tubular-collar vessel with a flaring rim and flat handles. In Late Ica-II ceramics, Jnca influence is noted in the vessel shapes, and in such features as horizontal flat loop handles and pointed bases. The true Inca aryballoid shape occurs with decoration in the local Ica style. A typical vessel has a barrel-shaped container with a short flaring collar and two looped handles. Other shapes carry over from the Late Ica-I, and in general the colors of the ware and the design tendencies are the same as before. Although much of Ica pottery is decorative and well executed, it lacks the originality and the freedom of expression found in the Nazca Period. In fact, the Ica ceramics show virtually no Nazca influence. There is, however, some Chimu influence from the North Coast. Weaving was well developed in the Ica Period (pl. 46, bottom). Tapestry, embroidery, and weft-pattern weaves are all common. 138 SOUTH AMERICAN INDIANS [B. A. BE. Bunn. 148 There is frequent use of applied fringes and tassels, and borders of small insets of tapestry. Slings are found in great numbers, made by a combination of braiding and warp-weave tapestry. Textile designs are essentially geometric in spite of the use of small conventionalized bird and animal figures. Small design units are either arranged in horizontal bands and slanting rows or framed in squares, diamonds, and lozenges. The textiles show some unusual parallels with Chimu pieces from the North Coast, implying widespread contacts. In general, the Late Period on the South Coast is hard to character- ize. Weaving and ceramics were equally developed, and utilitarian objects and ornaments were made of gold, silver, copper, and bronze (pls. 57, c, d; 58, 6). The architecture is competent but not unusual. Little can be implied about the social and political organization. The whole area fell to the Inca with little resistance, and their social organization dominates the scene. THE LATE PERIODS: CENTRAL COAST Most of the Late Periods on the Central Coast following the Middle Periods are heavily influenced either by the Chimu style from the North Coast or by the Jnca style from the Central Highlands. At the ruins at Pachacamac, a pure Jnca Period is found preceded by a style which shows this combined influence. Black ware double jars, aryballoid shapes, collar jars with flat handles, and stirrup-spout vessels all demonstrate this. Face-collar jars with handles painted on the body of the vessel are a local element in the mixture. Elsewhere on the Central Coast, the same situation exists. In some sites, the predominance of the Chimu style is great enough to suggest an actual extension of the Chimu culture itself. CHANCAY BLACK-ON-WHITE An exception to the general picture of Late Periods on the Central Coast is a Black-on-white style found at several places in the Chan- cay, Ancén, and Rimac Valleys. The ceramics (pl. 47) are composed of thin porous red or orange clay, covered by a scaly white slip over which are painted black or dark-brown designs. Stripes, both wide and narrow, single and grouped, are typical, and other common ele- ments are rows of dots, wavy lines, cross-hatch, toothed diagonals, and small birds or animals. For the most part the designs are skill- fully applied. Bowls and dishes are found, but two ceramic shapes are dominant: a goblet and an elongated globular vessel with a bulging collar which is frequently decorated with a face and two loop handles. All evidence indicates that the Chancay Black-on-white style is late, and probably contemporaneous in part with Inca. The absence of definite Jnca or Chimu influence is interesting. This is the only distinctive style in the Late Periods of the Central Coast, and it dem- VoL. 2] ARCHEOLOGY OF CENTRAL ANDES—BENNETT 139 onstrates that local groups were able to carry on their own cultural traditions in spite of the political dominance of the Inca. THE LATE PERIODS: NORTH COAST CHIMU On the North Coast the dominant Late Period is the Chimu, or Late Chimu. It follows the Middle Periods without a sharp break and ultimately merges with the Jnca Period. The traditional Inca history tells of the conquest of the Chimu realm, but certainly strong Inca stylistic influences occurred before. The chronological posi- tion of Chimu is well established by stratigraphic grave series at the ruins of Moche. The style has been well isolated in cemeteries in Virt Valley, at Chanchan, and in Chicama, Lambayeque, and Piura Valleys. Chimu culture has a dominant, outstanding style, in spite of the fact that many of its elements can be traced to previous periods. Many of the pottery shapes and textile techniques carry over from the Middle Periods. The earlier Mochica tradition is revived in modified form. North Highland influences are numerous, and later, Inca elements are introduced. These varied influences from different sources are, however, blended with some original elements into a characteristic Chimu style. Subdivisions of the Chimu Period have been suggested. All of these have not been confirmed, but a basic dual division is apparent, namely, a Chimu subperiod without Jnca influence (pl. 42) and a Chimu subperiod with Jnca influence (pl. 49, d, e). Although this division is based on the presence or absence of known Inca elements, it can be confirmed even in the detailed differences of common cooking ollas. The Chimu Period was a time of great population increase and expansion. ‘This is marked by the first appearance of true cities; and by the numerous and ambitious irrigation projects, which opened up much new terrain. Almost every quebrada in a valley was utilized in Chimu times. Finally, the Chimu Period has a wide distribution from Piura in the north to Casma in the south. Outside of this area, Chimu trade pieces or evidence of Chimu influence are noted from the Coast of Ecuador to the Nazca Valley on the South Coast and even at Machu Picchu, near Cuzco. The Chimu Period is marked by cities of considerable size, such as Chanchan (pls. 51; 52, bottom) near Trujillo, Pacatnamt or Bar- ranca in Pacasmayo Valley, and El Purgatorio in Lambayeque Val- ley. Of these, the city of Chanchan is the largest and one of the best known. The building of Chanchan was probably started in the Mid- dle Periods, but it was greatly enlarged during the Chimu and Inca Periods. Today, the remains cover about 11 square miles (18 km.”) 140 SOUTH AMERICAN INDIANS [B. A. B. Burn, 143 and consist of at least 10 large walled-in units, arranged roughly within a rectangular area with the narrow end toward the sea. Be- tween these units, frequently called ‘‘palaces,” are irrigated areas, reservoirs, and other ruins. Some cemeteries are found within the city, but others are on the sandy section which borders the ocean. Also near the shore are deep cuts down to the water level in which totora reeds grow. The units within the city quite possibly represent subdivisions of Chimu society, such as clans. Each unit differs from every other in detail, but all have certain features in common. They are all walled-in, and some have two or three enclosure walls. The Tschudi group covers an area 1,600 by 1,100 feet (480 by 355 m.) and its en- closure walls still stand over 30 feet (9 m.) high. Almost every unit contains a stone-lined reservoir, irrigated gardens, streets, houses of which some had gabled roofs, pyramids, cemeteries, and rows of cut-out cells which have been called prisons. The building materials are large and small rectangular adobes with rounded tops, puddled clay, or tapia, and a composition material of clay, shell, and stone. In some places, walls are built on rough stone foundations and reinforced with algarrobo logs. The occasional rains on the North Coast have destroyed much of Chanchan. There are still visible, however, traces of wall paintings, walls decorated by niches or by special arrangements of adobe blocks, and walls with relief clay arabesques cut out of a thick clay plaster (pl. 52, bottom). The small design units represent birds, fish, small figures, and some geometric elements, all apparently derived from textile patterns. The arrangement and the magnitude of the Chimu cities suggest a developed social and political organization, and this is partly con- firmed by the traditional Jnca histories. Individual distinctions in rank are portrayed in the ceramic designs and by the degree of elab- orateness in burial. Unfortunately, the Chimu ceramics do not depict the wealth of detail found in the Mochica Period. Typical Chimu graves are unlined shafts which may be marked by long sticks or paddles. Burials are commonly in a seated flexed position, wrapped in cloth. Extended burials or burials flexed and lying on their sides are variants. Burials are accompanied by ceramic and other grave furniture such as metal, cloth, bone, shell, wood, and calabash artifacts. Preservation in the Late Period is much better than in the Early. Chimu ceramics (pl. 49) are fairly difficult to characterize, since, in spite of a certain basic unity, they vary greatly in shapes and designs. Black ware and monochrome red ware are characteristic, although some painted ware is found, particularly in the Jnca phase. Many of the vessels are mold-made and are decorated by pressed Vow. 2] \ ARCHEOLOGY OF CENTRAL ANDES—BENNETT 141 relief, stipple, modeling, incision, appliqué, champlevé, and paddle marking. Ollas may have crude white designs painted on the orange clay base. The stirrup-spout vessel is about as characteristic as it was in the earlier Mochica Period. However, Chimu stirrup-spouts are gener- ally rectangular in cross section and have a small modeled monkey or other figure at the base of the spout. Other typical shapes, many of which are derived from the Middle Periods, are: double whistling jars (pl. 60); double-spout vessels with flat connecting bridges; tapering-spout vessels with a flat bridge which connects the spout and a modeled figure; globular bowls with tapering spouts, with and without single flat handles; and globular ollas with flaring or angular rims. Under /nca influence other shapes are added, such as: ary- balloid jars, both of black ware and with painted designs; vessels with modeled animals at the collar base; collar jars with flat handles; and shallow plates. The modeling of complete vessels and of small figures is generically related to the Mochica Period, but with considerable decline in the realistic technique. In Chimu ceramics, the figures are stylized and none would ever be called a portrait jar. The animals and birds are difficult to identify. Some scenes are still portrayed in the relief work, but they lack the photographic quality of Mochica. Textiles of the Chimu Period are well preserved. The development of weaving is also represented by looms, weave swords, weave daggers, and spindle whorls. A characteristic decorative technique is large- area embroidery but not of the over-all type. Tapestries are still common, particularly as belts and borders, and the warp-locked technique is typical. Gauzes, single-faced brocades, and warp- pattern weaves are plentiful. Plain cloth is decorated by painting, tie-dye, and ikat. The fabricated pieces include breechclouts, head bands or turbans, large mantles, and shirts with and without short sleeves. Copper, bronze, gold, and silver were all utilized in the Chimu Period. Copper and bronze were used for ornaments, but utilitarian artifacts are more frequent, such as points for digging sticks, needles, clubs, and knives. Gold and silver were hammered into goblets, featherlike pieces, masks, plates, elaborate earplugs, beads, and other ornaments. In fact, metalwork is one of the outstanding achieve- ments of this period (pl. 50). THE LATE PERIODS: NORTH HIGHLANDS Little is known about the Late Periods in the North Highlands- Middle Period material is, however, abundant and might, in part; have continued into the Late Period times. At the site of Chavin, much unidentified Late remains have been found, including con- 142 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 struction work of rough stone facing walls, stone-covered shallow canals, and stone-lined pits. The associated ceramics are mostly plain wares from cooking vessels and simple painted wares. None of the material is particularly distinctive, and all may be contempo- raneous with the Jnca occupation. 8 ] €) ; ri Ty: 5 iv; 4 : ¢ ‘ £, I ft k > aa a i wh ¥ ¥ * hh Ml i ‘wy ) th ia bit Fr ; ; (a thy Pahnad hy r r ' i , , ¥ 7) Put a oF 1 "7 : : * : 4 me A ” i ; gs . , , 4, ’ 4 ; . : ,' Hl + doll " a" ' nF i ; nee nd 3 / 4 a4 ; any rik -¥ j rs " rae) ij tape ‘+ i ‘| o 4 , i woth oe satu: Bevis iy bine patil iy : é ae... ee) ted bAS py Pe i, i as | > in ida Ria ind @ | 7 LS Baas ‘ } , i _? : r i, cs we. oie F e's - bs fl a , a. § ous oy j i i) 84: dt y ie Pal ei i vi P 7 . ae ng i th wae ol boll ; j > ton . ah ah p Vai Ra wi haha u i fouiti! wre i ais ; enone aed di ayoan ; F | vt maw ne ih si . ny ts, ' ot ehoctic tier ; Se mer | iit eae wire: Inbavic cai | a, poi iota, oxida endian nian ‘eat A CULTURE SEQUENCE FOR THE NORTH COAST OF PERU By Raragt Larco Hoye THE CUPISNIQUE CULTURE GEOGRAPHY Cupisnique pottery was first discovered by the author, who found it only in sherds, which were very numerous, in the Cupisnique Valley (map 1, No. 2), located between the Jequetepeque and Chicama Valleys on the North Coast of Peri. The name Cupisnique was selected to designate the pottery style and the culture. Cupisnique culture graves in the Chicama Valley are located at the following sites: Palenque, Barbacoa, Salinar, and the haciendas of Sausal, Gasnape, Roma, Santa Clara, Casa Grande, and Salamanca y Mocollope. Outside of the Chicama Valley isolated Cupisnique finds have been made in Pacopampa, Pacasmayo, Piura, Lambayeque, Chao, and Santa. The specimens from the various non-Chicama sites, although similar to the Cupisnique style, differ in distinct and peculiar char- acteristics.! CULTURAL AND CHRONOLOGICAL FOSITION Previous to the discovery of the first Cupisnique cemetery at Barbacoa, in 1939, the occasional specimens of the style, in pottery or stonework, which were found in the Chicama region, were classified as ‘“‘Chavin” after the Highland temple site of Chavin de Hudntar. The stone sculpture at this site bears a stylistic relationship to the Coastal finds. It is the opinion of the author, however, that the Cupisnique culture had its origin on the Coast rather than in the Andes. He believes that the important site in the Nepefia Valley was the principal religious center in primitive Pera from which the feline motif in decorative art radiated throughout all Perd. It should be emphasized that, in considering these particular forms of the feline motif, we are probably dealing with the symbols of a religious cult which was embraced by various peoples. The char- acteristic local expressions of the feline motif as seen in the art of Paracas, Pacopampa, and Chongoyape are an evidence of this. 1 A type of pottery discovered in the Virfi Valley, resembling Cupisnique, has been called ‘‘Vira-Cupis- nicoid’”’ by the author. It appears likely that this type is a fusion of the Salinar culture and Cupisnique influence. 149 150 SOUTH AMERICAN INDIANS [B. A. HE. BuLu, 148 It should further be noted that, while most Cupisnique cemeteries yield grave artifacts which are characterized by the many forms of the cat representation, the feline theme is entirely absent in the ceme- tery of Santa Ana, which belongs to the Cupisnique culture but is a variant in some respects as the pottery is orange rather than dark in color. Thus it would seem that some of the tribal groups of the North Coast did not accept the feline idea from Nepefia during the Cupisnique times. To conclude, the Cupisnique culture can be placed in the North Coast sequence as antecedent to Salinar, which was in turn followed by Mochica (Proto- or Early Chimu). Sculpture, ceramics, and the carving of small objects in stone and bone were advanced in development. The sculptural art is of particular merit. Metals were unknown, and the presence of small plates or disks of gold in some of the Salinar-Cupisnicoid tombs can probably be attributed to Salinar influence.’ Settlements at this time were small groups of houses which may represent some sort of joint family arrangements. The Cupisnique culture contributes strongly in culture elements to the formation of the Mochica culture. CULTURE SUBSISTENCE ACTIVITIES There is no evidence concerning actual agricultural practices or techniques, but several domesticated food plants were known at this early period. Ceramic representations of plants, as well as the original foodstuffs, have been found buried with the dead. These include peanuts, yuca, a cultivated species of caigua, and gourds. It is surprising that maize,’ potatoes, and other products common to the region in later archeological periods have not been discovered. In addition to. horticultural produce the Cupisnique people added llama and deer meat to their diet. The remains of shellfish and edible land snails testify to the importance of these as a food. Fish were, undoubtedly, another important food item. It seems reasonable to believe that they domesticated the llama and the dog, as skeletons of both of these animals have been recovered from human graves. ARCHITECTURE _ The few known ruins in the vicinity of Cupisnique and elsewhere in the Valley of Chicama show three types of construction: (1) those _of stone that the author considers most ancient; (2) those of conical adobes that pertain to a probable later period; and (38) a subtype of combined stone and adobe. 2 Gold objects have been found in association with Chavin-like cultures, or cultures related to the Cupis nique, at Puerto de Supe and Chongoyape; although they have not, as the author states, been found with Cupisnique proper.— EDITOR. 3 Maize and beans have been found with the related early Ancén-Supe culture to the south.— EDITOR. VoL. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 151 The sole pottery vessel of Cupisnique style depicting a house prob- ably corresponds to the stone-adobe subtype. This vessel (pl. 61, b) shows a house of simple rectangular plan with a gabled roof. The rectilinear figure seen on one side seems to be a doorway. On the Pampa de los Fésiles and at Hacienda Sausal (Barbacoa and Palenque) there are remains of walls constructed of fairly large, irregular stones set in mud and fitted with stone spalls. Adobe structures of the period are built up of conical or odonti- form adobes. Walls of adobe were built up by placing rows of coni- cal adobes point to point (fig. 19, a) and then by filling the angles and interstices with clay mortar. In the construction of walls of great thickness, a double row of adobes was laid point to point. On these another layer of adobes was alined in reverse in such a manner that their pointed ends touched the bases of the lower row, and likewise fitted base to base with each other. The conical adobes always had their flat bases toward the outside, in order to form a flat surface for the wall. The adobes were laid on a slight incline, so that a transverse cut shows the rows as a series of undulating faces (fig. 19, b). The walls were given a uniform surface by coating them with clay plaster. Both solid circular and pyramidal constructions were initiated in this cultural period. $ = LO KS f Figure 19.—Arrangement of conical adobes in Cupisnique structures. (Redrawn from Larco Hoyle, 1941, fig. 191.) 152 - SOUTH AMERICAN INDIANS [B. A. E. Buu, 143 CLOTHING AND ORNAMENTS The Cupisnique people wore a simple loincloth, leaving most of the body uncovered. The headdress consisted of a cap with an attached covering for the back of the neck. The bone ear orna- ments mentioned below, necklaces of stone and bone beads, and bone rings were common adornment. Necklaces of beads usually had a central piece with magnificent engravings. Some of the dead were found with rings on two, three, and even five fingers. ARTS AND MANUFACTURES Ceramics. —Notwithstanding the technical skill displayed in the finished product, the pottery does not attain perfection. A coarse- grained clay and temper were used in making even those vessels which have highest quality. Precise formulas apparently were not followed in the mixing of clays and tempers, so that results were not uniform. Open-kiln firing, a method which appears to be associated only with red or orange-colored pottery, was not employed. Their closed kilns resulted in an imperfect (or reduced) baking of the clay. Only in the last cultural stage of the Cupisnique do red, maroon, dull purplish red, and cream-colored ceramics appear. Impressions of molds on Cupisnique vessels show that pottery of this period was mold-made much as that of the later Mochica Period. Ceramic representations (pls. 61, 63, 64) include: anthropomorphs, zoomorphs, and phytomorphs. Vessel forms most common are stirrup-mouthed jars, bottles, globular vessels, and truncated cones. Even houses are depicted. Designs were incised on the anfired, dehydrated vessel surfaces much as in the carving of bone or stone. Afterward, color was sometimes applied (red, maroon, cream, and black) in the undeco- rated zones between the incised lines. Contrasts between smooth and rough surfaces were also utilized in pottery decoration. Rough- ening was accomplished by punctating, scraping, simple hatching or combing, and cross-combing the surface. Small bumps or pro- tuberances were occasionally placed on both smooth and rough sur- faces of the vessels. Geometric motifs are simple, and include and combine straight paral- lel lines, angles, simple or double chevrons, rectangles, combined and divided diagonally, herringbones, triangles, and rhomboids. Star figures and leaves are also a common design element. Much of the Cupisnique decorative style suggests a religious sym- bolism similar to the art forms of the Nepefia Valley. In fact, to eliminate from Cupisnique these Nepefia-like elements would se- riously deplete the total range of decorative motifs. The Cupisnique representations of the Nepefia type are, however, noticeably modified. Vou. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 153 In modeled pottery, the Cupisnique sculptor reproduced animals, god or demon forms, plants, and dwellings. In his work can be ob- served the first evidences of naturalistic representation of North Pe- ruvian prehistory. Notable examples are the earliest known facial portrait jar of Peri and a scene composed of a mother nursing an infant (pl. 64, ¢). Much of the elaborate, sculptured funerary ware of the later cultures of Northern Peri undoubtedly were of Cupis- nique origin. Carving in stone and bone.—The Cupisnique craftsman excelled in the carving, sculpturing, and polishing of small objects of stone, shell, and bone. He developed techniques to control very hard rock, such as porphyry, granite, turquoise, lapis lazuli, and quartz. Out of these materials he made beads, amulets, small figures, rings, earplugs and ear pendants (pl. 62, top), various receptacles, and mortars. Working with turquoise, which was a scarce medium, he flaked the stone and set the flakes into a thick layer of plaster. Softer media, such as slate and anthracite, were also used; the latter was made into amulets and mirrors, which were ground and polished to a smooth surface, which was sometimes convex to reduce an image. Bonework includes rings (pl. 62, bottom), which were skillfully carved with religious or demonic motifs, beautifully worked beads, earplugs and pendants, combs, spatulas, and needles. Animal bones, especially llama bones, and bird bones were used to make these artifacts. The techniques developed by the Cupisnique artisans for this exacting and elaborate carving in such durable materials as the very hard varieties of rock are not known. Textiles.—Balls of cotton yarn and fragments of carbonized cloth have been found in the tombs. The textiles are coarse and open, imperfectly woven, and done in a simple but uneven interlacing technique. Stamps made of pottery are a Cupisnique artifact, but there is no evidence to show that these were used to decorate textiles. It seems more likely that the stamps were used to decorate the human body. Several of the stamps present traces of red paint adhering to them. The bone needles were undoubtedly used to sew the textiles in making clothing. Metals.—In no Cupisnique Period excavations has metal of any sort been discovered. BURIAL AND RELIGION Disposal of the dead was clearly associated with the religious ideas centering around the feline deity. In Cupisnique, a large percentage of the offerings to the dead are sculptured, modeled, or incised repre- sentations of the gods, particularly the feline. This is in contrast to the later cultures, especially Salinar, of the North Coast, in which 154 SOUTH AMERICAN INDIANS [B. A. E. Bunn, 143 the seemingly highly religious background of the grave offerings was replaced by other themes. The feline deity or a feline cult is widespread in Perf. When the cat demon appears in the Cupisnique culture it is conceived of an- thropomorphically and is represented walking erect on its hind feet as aman. In some cases, cat and human representations appear on the same modeled face: half of the countenance feline and the other half humanized-feline, highly stylized. The condor and the serpent appear to be other minor Cupisnique divinities. Burial procedure was more or less uniform, with some variation in the arrangement of the body. Graves.—The graves were irregular holes excavated directly into the earth. The shape of the grave depended upon the position and the degree of flexure or extension of the corpse. Some graves were circular, others elongated or elliptical. In some graves, the dead had been covered with angular rocks. Depth of the graves varied from 32 inches (80 cm.) to 6 feet (2 m.). Disposition of the body.—The flexed position was the most common (pl. 69, top), the head being turned either to the left or right or doubled forward on the chest; the skeleton was found either on its side, on its back, or face downward; the legs were tightly flexed and usually drawn up together near the chest, but in the case of those individuals placed on their backs, the legs were sometimes spread apart and flexed on each side of the body. Some skeletons were only slightly flexed. No definite orientation to the cardinal points was observed for either graves or the bodies found within them. Coloring of the bones.—The bones were commonly colored with a red powder composed of clay with traces of lead and mercuric oxides. The coloring of the skeleton is not complete, although in some instances the bones are so nearly covered with the red pigment as to suggest that some of the burials were secondary. In some cases, the skull was covered with a double cloth beneath which was a thick layer of the red powder. Double burials in a single grave have also been found. Red seems to have been the funeral color of Cupisnique as it was for the North Coast until the rise of the Chimu Empire. In one Cupis- nique grave, a small quantity of green powder was found. Grave offerings.—It is not known whether the dead were clothed for burial, but in the majority of cases artifacts and jewelry were placed with or on the body. Many grave objects, including pottery vessels, are of a purely ceremonial or religious nature, although some graves contained pottery marked by fire as though in cooking. Vessels were placed at the sides of the body, and sometimes in the hand. VoL. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 155 Remains of foods were found in the graves, and shells had often been placed in the dead man’s hand. The various objects and ornaments of stone and bone described above were all recovered from graves. Dog skeletons are found in graves. THE SALINAR CULTURE INTRODUCTION The first cemetery of the Salinar culture was discovered by the author at the site of the same name in 1941. Previously, four pot- tery specimens, acquired by the Museo Arqueologico ‘‘Rafael Larco Herrera”’ at Chiclin as a part of a much larger collection, had been set aside as being sufficiently distinctive to be a separate style. More pottery vessels of this new style were obtained by the museum from an ‘‘huaquero,” and this led to the discovery of the Salinar cemetery in the upper Chicama Valley. The name of the type-site was applied to this heretofore unrecognized prehistoric culture. GEOGRAPHY Salinar is located on the Pampas de Jaguey Hacienda, only a few kilometers from the Cupisnique site of Barbacoa. It is a vast ceme- tery, 228 graves having been encountered within its limits to date. Within the cemetery, graves of the Mochica culture were found super- imposed over Salinar burials, and Salinar graves were found which intruded through earlier Cupisnique interments. As with burial grounds restricted to Cupisnique, Mochica, or Viri cultures, Salinar is situated out of the cultivated land on the sterile slopes of the hills bordering the valley. Other Salinar type cemeteries have since been found to the north- west, east, and southeast of Hacienda Sausal. A few isolated graves of the culture have also been located between Sausal and the origi- nal Salinar cemetery. In the Vira Valley, east of the port of Guafiape, a typical Salinar cemetery was isolated, and in the same valley, adjacent to Hacienda San Ildefonso, some curious “‘hybrid”’ vessels were recovered which have resemblances to both the Cupisnique and Salinar styles. Three vessels from the Santa Valley are known which appear to be variants of the Salinar style; however, the provenience of these particular specimens is open to some doubt. At present, the area in which Salinar-type finds have been made centers between Hacienda Pampas de Jaguey on the north and Ha- cienda Sausal on the south, a sector on the right side of the upper Chicama Valley. The only other important location for Salinar is Guafiape in the Vira Valley. 156 SOUTH AMERICAN INDIANS [B. A. E. Buty, 143 CULTURAL AND CHRONOLOGICAL POSITION The Salinar culture of the North Coast of Peri constitutes an im- portant chronological and evolutionary step between the Cupis- nique and Mochica cultures. In it we can observe the course of de- velopment of the ceramic and pictorial arts from one stage to the other. Except for the carved bone spatulas, which were done in a manner reminiscent of Cupisnique, only two feline representations were found among all the Salinar artifacts. These were pottery vessels. For some reason, as is the case with the Cupisnique culture site of Santa Ana, the Salinar peoples did not incorporate the feline motif into their decorative art. Its absence makes it likely that the feline symbolism was not a part of their religious beliefs. In the evolution of social aggregations of the North Coast of Peri, Salinar takes its place among the cultures of limited geographical spread and political dominance which existed before the formation of the great organized regimes. But the people of Salinar contributed many valuable cultural elements to the formation of the first great civilization of the north, the Mochica. CULTURE SUBSISTENCE ACTIVITIES Maize, which was not found in the Cupisnique culture, has been recovered from Salinar graves. Squash and gourds are also recorded for Salinar. Pottery representations of the licuma and the pepino indicate their former use as foods. There are, however, no life representations or any other evidences that give a clue to types of agricultural practices that might have been employed by these people. There are various sea shells, including clams and choros, in the eraves, and there are shells of land snails. Skeletons of birds found with the dead suggest that they also served as food. Artifacts of llama bone are known for this period, and it is likely that this animal was eaten. CLOTHING AND ORNAMENTS There is considerable emphasis on headgear in Salinar clothing. Judging from pottery representations, there was a cap that was adapted to the form of the head. The cap was belted with a coiled strip of cloth or fibers. This coil or belt for the cap crossed, in some cases, in the front, with the two ends fastened upon the head. Some- times the ends hung down over the back. There were also conical caps similar to the present-day ‘‘chullo”’; and still others of a rectangu- lar form, like a mitre, with a visor attached. The hair was usually combed in bangs, down to the eyes in front and cut back into a step in front of the ears. On the sides and back VoL. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 157 it was usually trimmed off at the level of the neck but sometimes hung down to the shoulders. A variation was a headdress where the hair was tied at the top, giving the head a conical appearance. There is no evidence for face painting, but three vessels show individuals with incisions beside the eyes, nose, and mouth, which suggest that the Salinar people adorned the face with skin incisions, or which may indicate a beard, mustache, etc., etc. Clothing covering the body is never well depicted. Some figures which do not show the genitalia may be the individuals covered with a long shirt of some sort. However, from fragments found on actual bodies of the dead, it is certain that clothing was worn. They wore both circular and tubular ear pendants, finger rings, nose pendants, and necklaces of stone, shell, or pottery beads. These beads were cylindrical, spheroid, or truncated-conical in form. They also wore necklaces composed of bell-like pieces of pottery and pottery phalli. Bracelets were composed of small, cut snail shells. ARCHITECTURE No actual buildings have been identified as belonging to the Salinar culture, but two pottery vessels give considerable detail upon house types. One of these shows a round tower supported by step-designed pillars. A decorative frieze of continuous loops, which are perforated in the center, surrounds the tower. The roof is flat. The second house is four-sided and pent-roofed with an open front (pl. 66, c). The stairsteps, or stepped symbols, are the vertical supports on each side, and they are fastened together with beams. A cross- piece, circular in cross section rests upon these beams, and in turn, holds up the roof. The roof slopes toward the rear of the house. A great central upright in the front-center of the house helps support the crosspiece. From these two ceramic representations, the development of a quite definite architectural style is observed. It is, unfortunately, impossible to know what materials were employed. ARTS AND MANUFACTURES Ceramics.—On first view, Salinar ceramics (pls. 66, 67) seem rather crude and simple; but closer inspection shows a marked technological improvement over Cupisnique pottery. The paste was carefully prepared with sifted ingredients. Firing usually resulted in an even red color throughout. This indicates the employment of pottery ovens permitting abundant oxidization during baking. Ninety-three percent of the ware studied has a dull, natural red surface resulting from the firing; the remaining 7 percent is black, dark red, purple, or dark brown, and has a petrous appearance as does the Cupisnique pottery. The darker vesssels have been bur- 158 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 nished with a smoothing implement, and some of the red vessels were apparently coated with a film which upon firing resulted in a surface finish similar to that of the transitional Cupisnique or Cupisnique- Salinar pottery. Molds were used in manufacture, and handles and spouts were at- tached to the vessels later. Bottle forms are the most common, although there are others. The principal shapes and their variations are given below: I, Stirrup-mouthed vessels. A. Human representations. B. Zoomorphic representations upon truncated-conical or semi- globular forms. C. Globular pitchers, truncated-conical or conical. D. Stirrup-mouthed vessels with triple conduits. II. Bottle-shaped jars with cylindrical spouts and flat, semicircular handles. A. Anthropomorphic, entire body. B. Zoomorphic, entire body and animal-head representations. C. Globular forms. D. Globular forms, elongated. E. Truncated-conical, double truncated-conical, and conical forms. Ill. Jars with spout and bridge handle, with a sculptured figure, upon globular or truncated-conical pitchers. (In this type appear the same variations as listed for the two types above.) Among the bottle forms there are some with a short, broad spout which may be classifled as “porongos.’”’ They have a globular or truncated-conical form and cylindrical, semicircular handles. There are also little or miniature globular ollas, truncated-conical vessels, and some with human heads and forms. Surface decoration of the pottery was by both incising and paint- ing. The Salinar potter effected the incising while the clay was wet and soft. The old feline motifs, so prominent in Cupisnique, have disappeared in Salinar decoration. The geometric motifs, however, continue, and are a little more developed and perfected than in the earlier period. Incisions are used to aid in anthropomorphic and zoomorphic delineations; and the roughening of the vessel surface, and the addition of small nodes or protuberances are additional tech- niques of nonpainted decoration. Both red and white pigments were used to decorate the Salinar pot- tery, and were applied after partial dehydration and before firing. As noted above, most of the vessels do not have aslip ora base. The paint, usually white, was applied with a thick brush. The same geo- metric and other simple designs executed in incising were also made with paint. The stepped symbol, loops, stars, wavy lines, parallel arched lines, and volutes constituted the principal motifs of the Sali- nar artist. Paint was frequently used to bring out details of model- ing as well as to draw independent designs. It is interesting to note that in this period are, seemingly, the beginnings of the ceramic paint- Cc PuaTE 61.—Cupisnique pottery. a, Feline representation with slight anthropo- morphic resemblance; b, house with roof sloped in two directions; c, bottle with anthropomorphic (?) head at top. (Courtesy Rafael Larco Hoyle.) PuatE 62.—Cupisnique ornaments of shell and bone. Top: Ear pendant of shell which had been inlaid with turquoise. Bottom: Bone rings. (Courtesy Rafael Larco Hoyle.) PLATE 63.—Cupisnique pottery. a, Fox; b, mollusk representation, probably Spondylus pictorum: c, roughened surface with protuberances; d, red ware jar, representing feline serpent, minor divinity of Cupisnique. (Courtesy Rafael Larco Hoyle.) PuatEe 64.—Cupisnique pottery. a, The feline serpent; b, geometric bottle form; c, woman nursing child (a unique example); d, feline god. (Courtesy Rafael Larco Hoyle.) Vers tottifie Vipygy Wipe OUI[O} OU J, yb o¢ | ( oO |AOY “onbrustdt NS . <= SS eX OoIB'y [oBVJeyy ASoJINOD) jo ¥BY} OF ABTIWMITIS ‘onbiustdny ut }S B UL AjLop iegodde jou soo UI[O} OY ABR. L i 2s 7 yorym “eduredoowg jo AYUTATp -eduredooeg Wodj Sivj1OWl 9U0}sS SRK Uy Wij x 84844 Sc RRR AGE N [MO Poa BOT G9 LV Td g Puate 66.—Salinar pottery. a, Dog or feline; 6, monkey; c, four-sided, pent- roofed house; d, anthropomorphic head-and-spout with flat bridge-handle; e, woman; f, unidentified bird; g, stylized owl head; h, dove;7, owl. (Courtesy Rafael Larco Hoyle.) g A PuatTe 67.—Salinar pottery. a, Single-spout jar with flat handle and incised and painted stepped design; 6b, incised and painted decoration; c, single-spout jar with flat handle and white design; d, fantasy shaman attending a paticnt; e, incised vessel; f, roughened, noded, and incised surface; g, feline; h, noded vessel; 7, monkey. (Courtesy Rafael Larco Hoyle.) PuatTe 68.—Salinar Period bonework and metalwork. Top: Three bone spatulas incised in a style suggestive of Chavin or Cupisnique. Bottom: Nose ornament, ring, and sheets of beaten gold. (Courtesy Rafael Larco Hoyle.) PuatTEe 69.—Burial types of the North Coast of Peri. Top: Cupisnique. Center: Salinar. Bottom: Mochica. (Courtesy Rafael Larco Hoyle.) h PLATE 70.—Mochica pottery. a, Excellent and unusual relief sculpture; 6, Ai apaec, the Supreme Divinity, as an agriculturist, shelling corn; c, the Supreme Divinity conquers a demon vampire; d, death’s head of Ai apaec; e, removing a tumor; f, representation of a face pockmarked by disease; g, the earthly divinity sails the seas; h, portrait study. (Courtesy Rafael Larco Hoyle.) PLATE 71.—Mochica craftsmanship. a, Pottery model, used in making a pottery mold; }, c, two sides of a mold made from a model; d, warrior carved in bone, red with mother-of-pearl inlays; e, copper casting, a warrior attacked by a dog; f, wood sculpture with mother-of-pearl! inlays; g, spear thrower wrapped with gold; h, head ornament of beaten gold; i, stone box with engraved figures. (Courtesy Rafael Larco Hoyle.) PuatE 72.—Mochica construc- tion and architecture. Top: Aerial view of aqueduct of Ascope. Bottom: Adobe col- umn at Tambo Real. (Cour- tesy Rafael Larco Hoyle.) Vou.2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 159 ing art that was later to crystallize in the Mochica culture with its expertly drawn red and white designs.‘ Although anthropomorphic sculpturing and modeling of Salinar pottery have not attained to the Mochica level of style, the bird fig- ures, animals, plants, and houses show a continued advancement in technique. In the human figures, as in most of the representative pottery of the region, the form of the body is subordinate to the vessel shape. Because of this, heads are out of proportion, so that greater capacity can be gained for the vessel. In making the features, the hair is done with incised lines; the eyes made by placing a dot in the center of a simple incised circle, by an incised diamond, or by deeply indenting a small sphere of clay; and the nose is usually a small pro- tuberance with holes in the sides. The mouth is a long incision, or double incision, with small pieces of clay inserted or with indenta- tions to serve as teeth. The ears are represented as very large. Breasts and genitalia are indicated in a disproportionate size. Some attempt was made to give anatomical proportions to the lower ex- tremities; but the arms are thin and poorly made. Fingers are very crudely executed with incisions. In the sculpturing there is a tendency to depart from naturalism and stylize certain features; yet, in the animals and birds the artist undoubtedly tried to capture natural qualities and attitudes. Some of these are: cats, monkeys, rats, owls, parrots, doves, hummingbirds, and other birds, unidentified. Plants include the cactus, lucuma, and pepino. The appearance of pornography in modeled ceramics, because of its important position in Mochica art, should be especially noted. The scenes depict natural coitus, and there is no perversion.. Carving in stone and bone.—A few bone spatulas (pl. 68, top), with incised designs somewhat similar to those of Cupisnique carved upon them, appear to be the only tie to the outstanding carving of the earlier period. These artifacts do not have the same perfection, nor do they treat the design motifs as in Cupisnique.. Sculptured stone has not been found in connection with Salinar. Beads of turquoise, and other stones, are of the most simple shapes and unadorned. Textiles.—The fragments of cloth from the graves which have not disintegrated are of a tight warp and of a textile type commonly called “‘lawn.’’ ‘The weave is simple interlacing of one or two strands, which are well twisted and uniform. The material has not been identified, but is probably wool or cotton. Needles, both of bone and of wood, were presumably used in sew- ing garments. 4 See Bennett, this volume, for reference to the ‘‘White-on-red”’ style, pp. 92, 97. 595682—46 13 160 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143. Metals.—Thin, hammered gold objects and jewelry are found in Salinar graves. This is the first appearance of metal in the culture sequence of the North Coast of Perf. Most of the gold was beaten into thin sheets, and rudimentary openwork forms were made. How- ever, the technique of soldering gold was known.5 A ring, a nose ornament, and circular, oval, and trapezoidal frag- ments of gold were recovered from the graves (pl. 68, bottom). All objects weighed less than 2 gm. each. In one tomb a small ovoid sheet of metal was found, covered with a greenish oxide. This suggests that the Salinar artisans had mixed copper with gold, or that they had used an impure gold. Unfortu- nately, this fragment is so small that it would have to be destroyed to make a quantitative chemical analysis. BURIAL A very definite mode of burial was practiced. Bodies were interred at full length with the legs extended and slightly crossed (pl. 69, cen- ter). Usually, the individual Jay on his right side and was supported in this position by pottery vessels and rough stones. Arms were ex- tended at the sides, or, in some instances, the left hand was placed over the pelvis. Sometimes the legs were slightly flexed. The de- gree to which this pattern of arranging the body was followed is in- dicated by the fact that only 2 percent of all burials recorded had the bodies placed on the left side. The outline of the grave was an dohoartd ellipsoid. Often the bodies were placed along the side wall of the grave and were then covered with great slabs of stone leaned against the wall, forming a sort of crude sarcophagus. Or, as a variant of this, the wall of the grave was undercut and the body and accompanying funerary offerings placed in the niche and sealed off with stone slabs. Occasionally, two individuals were placed in asingle grave. In these cases they were extended side by side usually back to back, with one at a slightly higher level (6 inches or 15 cm.) than the other. From one to three pots were placed in the grave with each body. They were put either at the head, thorax, shoulders, legs, arms, or feet. Red powder was found in most of the graves but not as small bundles of powder as in Cupisnique. It was scattered in chunks or lumps within the grave or was placed in special receptacles. In color, it is not as vivid a red as that found in the Cupisnique graves, being more of a dark red or sometimes a purplish red. The dead were covered, or partially covered, with cloth, and adorned with necklaces and bracelets. In the mouths of some of the 5 This is one of the earliest evidences of soldering from Peri. See Root, volume 5, Handbook of South American Indians.—EDITOR. Vou.2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 161 skeletons were found the small oval or circular sheets of gold, possibly placed there for magical purposes. It is to be noted that this custom was accentuated in the Mochica culture. In addition to pottery, other funerary offerings include: gourds that had been filled with meat, pumpkin seeds, maize, mollusks, land snails, birds, dogs placed at the feet of the dead man, conical pieces of white chalk, mortars, fragments of rough quartz, and round stones, usually white in color. The burial form of Salinar antedates the Mochica burial type in which the dead person is placed in the grave at full length but on the back. It is likely that the stone-slab tombs of Salinar are the precur- sors of the Mochica stone-lined box tombs. It should be mentioned that the burials of Salinar-Cupisnique type, found in the Viré Valley, are not of the Salinar type but the flexed type of burial which char- acterizes Cupisnique proper. Salinar bodies follow no very definite orientation in the graves with reference to the cardinal directions; however, the majority of them are oriented within the arc of 35° to 65° west of magnetic north. SHAMANISM The first representation of shamanism or medical practices for the North Coast of Peri are recorded in Salinar. One vessel shows a seated individual with another person before him who is reclining on his back. The seated individual has placed his hands upon the patient in much the manner of the Mochica shamans who are shown in the pottery of that period (pl. 67, d). THE MOCHICA CULTURE GEOGRAPHY The Mochica territory includes that part of the Coastal belt of North Pera lying between lat. 7° 36’ and 9° 20’ S., and between long. 78° 51’ and 79° 28’ W. It embraces the rich valleys of Chicama, Santa Catalina (Moche), Virié, Chao, Huamanzafia, Santa Ana (Lacramarca), and Nepefia, an area of approximately 6,585 sq. km. (about 1,600 sq. miles). Mochica pottery has been found north of Chicama, in the Valley of Jequetepeque, only in small quantity. In the region of Pallasca, in the Sierra of the Department of Ancash, there are vessels and stone objects showing Mochica influence. These would indicate that Mochica influences, carried by conquerors or traders, reached well into the interior. CULTURAL AND CHRONOLOGICAL POSITION The cultural elements of three periods—Cupisnique, Salinar, and Viré Negative—all contributed to the make-up of the Mochica culture. 162 SOUTH AMERICAN INDIANS [B. A. E. Buun. 143 In analyzing the Mochica ceramics, the author sees evidence of four progressive periods. The first two are represented only in the Chicama and Santa Catalina Valleys; and the last two are found in all of the North Coast valleys of the Mochica territory. It is the author’s opinion that the Mochica style originated in the Chicama Valley. Throughout the four Mochica ceramic periods are seen the rise, evolution, and decadence of the arts. Beginning in naturalism, the trend is toward stylization, followed by a rejuvenation, and then a degeneration. During the final stage of decadence, there appeared a new culture and people in North Coastal Peri. These people, of whose origins we are not sure, left behind a pottery of slightly modified Tiahuanaco style. Their influence appears impressed upon the various institu- tions and ways of life of the previous Mochica. A fusion of this Tiahuanaco style with the Mochica left a hybrid pottery (Kroeber’s Cursive), which is the true Middle Chimu style because it links the Mochica with the Late Chimu. CULTURE SOURCES OF INFORMATION This summary is based principally on observation of over 30,000 ceramic vessels, many of which are realistically modeled and painted to represent a wide variety of cultural activities, and on a study of many other Mochica artifacts of metal, stone, wood, and bone. These objects are in the Rafael Larco Herrera Museum of Chiclin, Perd. The summary also includes information obtained by the author during many archeological excavation and reconnaissance trips in the North Coast area. The ethnology of present-day groups, both of the Coast and Highlands of this region, supplements the archeological data. SUBSISTENCE ACTIVITIES Agriculture.—The Mochica people were advanced agriculturalists. Their agricultural activity is evident today not only in the rich valleys that they dominated but also in the marginal areas that they brought under cultivation through great irrigation works. In all the valleys there are canals and aqueducts, some of which are still utilized today. The canal of La Cumbre, 113 km. (about 75 miles) long, is fed from the headwaters of the Chicama River and irrigates the fields adjacent to the site of Chanchan. In order to cross ravines which interfered, numerous aqueducts were built, following the plans of canals that were traced for the purpose of irrigating new lands. The most im- portant are those of Ascope (pl. 72, top) in the Chicama Valley, Mampuesto in the Santa Catalina Valley, and Tambo Real in the Santa Valley. The aqueduct of Ascope, which was constructed by sedimentation, has a length of 1,400 m. (about 4,500 feet), a cubic VoL. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 163 content of about 785,000 cu. m. of earth, and a weight of more than two million metric tons. This Mochica aqueduct is one of the master engineering works of ancient Perd. The Mochicas employed a system of straight furrows and small leveled plots with retaining earth banks for irrigating crops on the flat lands, and curvilinear furrows for irrigating sloping terrain. For fertilizer they used bird guano from the islands. The principal plants cultivated were maize, beans, peanuts, pota- toes, yuca (manioc), sweet potatoes, ulluco, aji, maize, pumpkins, gourds, chirimoyas, custard apples, pacae, granadillas, licumas, pepinos, coca, and cotton. In addition, other plants were used for medicial purposes, including varieties of cactus, habillas, ashango, maicheles, and many other curative herbs. Hunting.—The Mochicas hunted to obtain meats to supplement the plant foods. Nets and the estélica (spear thrower) and dart were used to kill deer; the blowgun and spear thrower for birds, such as doves and wild ducks; and maces or clubs for sea lions. The domesti- cated llama and guinea pig were eaten. The Mochicas gathered land snails and hunted iguanas as additional foods. Fishing.—The Mochicas fished in the ocean from large balsas, much like those now seen on Lake Titicaca. They also had smaller balsas similar to the small totora balsas, or ‘‘caballitos,” still used today by the native fishermen of the Coast. The construction of these little craft appears to have remained unmodified through many centuries. The principal fishing equipment included nets with gourd floats, un- barbed fishhooks of many sizes, and wooden harpoons used for large fish. Pottery representations indicate that the Mochicas caught fish ranging in size from the shark to the anchovy. From along the rocky beaches, they collected shellfish, many species of which have. either since disappeared locally or were brought in from farther north in Mochica times. Spondylus pictorum and Strombus galeatus are examples of the latter. Food preparation.—The Mochicas prepared various dishes of meats mixed with vegetable foods. In a container consisting of two gourds fastened together with a cord they placed doves or guinea pigs over maize. ‘They cooked guinea pigs on small spits over hot coals. They ate from gourd, pottery, and silver food containers with sticks sharp- ened at both ends or with pottery spoons. Large, bell-shaped vessels were receptacles for kitchen refuse. Chicha was a fermented maize drink and was served in gourd, pottery, or silver containers. Domestication of animals.—The llama and the dog were the principal domesticated animals. In addition, the Mochicas captured young deer, pumas, monkeys, and parrots in order to have them as household pets. 164 SOUTH AMERICAN INDIANS [B. A. E. Bun. 1438 ARCHITECTURE General architectural knowledge also evolved from Cupisnique into Mochica. The Mochicas did not build great cities of the order of Chanchan, but their structures are handsome and reveal consid- erable knowledge of architectural principles. They attained esthetic harmony by studying the strength and nature of their building materials. Construction was principally with rectangular, mold-made adobes. These were developed from the odontiform adobes of the earlier periods. Rough stone was used for wall foundations. In large buildings, or in walls built along the sides of roads, semicircular, bread-shaped adobes were used. Roofs were made of straw, cane, matting, and totora, supported with algarroba beams. Algarroba timbers were also used as a framework for the great masses of adobe in the solid pyramids. Dwellings were usually small and consisted of rooms connected by rectangular doors, some of which were arched at the top. Houses frequently had patios and terraces, and some had small entrance rooms. Roofs were gabled, with an open space between the two slopes at the peak, so that the houses not only provided shelter against the rains but were also well ventilated. Palaces were erected on outstanding sites or on the tops of pyra- mids. They were approached by broad stairways. Both exteriors and interiors of the palaces were decorated with symbolical figures executed as frescoes and polychrome murals in relief. Adobes with decorative geometric stucco motifs were used in the building of palaces and temples. Forts, located at strategic points in the Mochica territory, are solid structures. Often they are surrounded by massive high walls. The stairs, in contrast to those of the palaces, are narrow and steep, to prevent free access. ‘The smooth walls are inclined, sacrificing artistic harmony to obtain strategic efficiency. Temples were built upon the great solid pyramids and were deco- rated with complicated polychrome friezes of a religious character. The principle of the column was known and frequently employed. On the Tambo Real Hacienda (pl. 72, bottom), in the Santa Valley, a monumental column 26 feet (8 m.) high stands on the center of a great pyramid. The Mochicas knew and controlled the techniques of clay covering, plastering, and stuccoing for wall finish. ROADS Sections of roads are still in existence on the pampa of Chicama and in the valley of Santa. These roads were 9.80 m. (about 33 ft.) wide 6 Special attention should be called to the arch form mentioned above. There are examples of it in Mochica temple and tomb construction. Although known at this time, the arch was rarely used. vou.2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 165 and ran the length of the territory, even through the most precipitous regions. Minor branches extended back from the main roads. Small rectangular platforms at intervals along the roads are presumably foundations for buildings which housed messengers. The road width of 9.80 m. (about 33 ft.) should be noted. A pyramid of this period is 98 m. (about 330 ft.) square at the base. This consistent factor, discovered in many other measurements, suggests that the Mochica unit of measure corresponded to 98 cm. (about 3.3 ft.). DRESS AND ORNAMENTS Clothing and jewelry of men were far more elaborate than that of women. Women, as a rule, wore only a large shirt, or camisa, and, rarely, simple ear pendants. Men wore a breechclout, an underskirt, a sleeveless undershirt, a decorated overskirt hanging nearly to the knees, and a very showy, short-sleeved outer shirt that came down a little above the fringe of the underskirt. Both imner and outer clothing were tied around the waist with a belt decorated with artistic motifs. The Mochicas went barefoot but painted their feet and lower legs to resemble boots. Headdresses are sumptuous, but varied. Some are relatively simple, being a circular band or turban. Others are more complicated and are harmoniously combined in three parts: the cap with neck covering; the chin strap; and the turban proper, that surrounds the head like a crown and is adjusted over the other two parts. The turban was made of the finest cloth, of various colors and decorated with figures, or of treated feline or monkey skins. Stuffed birds, and pieces of gold, silver, or copper wrought in the shape of animals, ceremonial knives, and disks adorned the headdress. ‘The final result as seen on digni- taries such as war leaders, rulers, or priests was very impressive. These turbans were distinctive of rank, office, or profession, and serve to identify the interpreters of messages, the messengers, fisher- men, agriculturists, priests, governors, and military chiefs. As jewelry they wore great ear ornaments of gold, silver, copper, or of bone or wood with inlays of turquoise, mother-of-pearl, and lapis lazuli. The ear ornaments are simple circular and tubular forms, tubes with ridges or rosettes at the end, and varieties of pendants. Discoidal and other varied forms of nose ornaments were worn by nobles through a perforation in the nasal septum. Necklaces are of geometrically shaped beads, series of little idols, fish, stylized felines, fruits, and seeds. They were made of semiprecious stones (turquoise, lapis lazuli), quartz, rock crystal, gold, silver, copper, bone, shell, wood, and pottery. The central pieces of the necklaces were large beads representing human or mythological beings. ‘The Mochicas also wore finger rings and covered their fingernails with thin sheets of gold, 166 SOUTH AMERICAN INDIANS [B. A. E. Bu. 143 Although most of the actual textiles have decayed, it is known that the Mochicas knew and used llama or vicufia wool and brown cotton. Zoological, botanical, and geometric motifs were used in the decoration of fabrics; in addition, they attached gold and silver disks to their garments. Feathers were also used to adorn clothing and turbans. They painted the face and body with geometric figures, and they made incisions in the skin of the lips and cheeks to represent foxes, iguanas, felines, serpents, birds, and simple linear arrangements. No evidence of tattooing has been found. TRANSPORTATION The llama, the only domesticated beast of burden, carried cargo in bags, saddle-bags, and large baskets. There is also ceramic repre- sentation of llamas carrying mutilated persons. Occan travel along the coast was in_ the great balsas and in the smaller ‘‘caballitos’” of totora. These craft were propelled with long paddles of wood or planed-off sections of Guayaquil cane. MANUFACTURES Ceramics.—The Mochica potter carefully selected his clays to obtain a uniform paste. First, the potter modeled a thick-walled clay model (pl. 71, a), which was fired. Over this he made the molds, which were cut vertically into two parts, following the.groove down the sides of the model. From these negative casts (pl. 71, 6, c) were obtained two positives, which were joined to form the vessel. The spout, handle, and base were made separately and added to the main body of the vessel. The junctures were obliterated; and the com- pleted vessel was polished, painted, and exposed in the open air to dry before firing. Pottery was fired in open ovens, producing com- plete oxidization. For paints, they used colored clay mixed with silicious materials that when fired produced a brilliant surface. Polishing was done with spatulas of bone. In special cases they inlaid the pots with turquoise, slate, and bits of gold and silver. Textiles.—The Mochicas prepared thread by hand on a spindle having a whorl. Textiles were woven on hand looms. The few existing specimens of cloth and the pottery representations show the great variety of fabrics used for clothing and their rich decoration. Both open-mesh and close-weft weaving was employed. The thinner cloth served as underclothing and the more tightly woven for the outer garments. The textile industry was in charge of the women. Preparation of skins.—Hides were prepared both depilated and with the hair. Judging from the appearance of the treated skins, they used lime and alum to treat and bleach them, obtaining a smooth, Vor. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 167 consistent, and flexible finish. Complete small animals and birds were stuffed for turban ornaments. Work in gourds.—The gourd, an important article for domestic use, was decorated with incisions and inlays. Metals.—The Mochicas obtained gold, silver, copper, and lead (pl. 71, e, g, h). The quantities of these metals which are found sug- gest that they not only got native metals but may have known tech- niques of extraction of ores.. Pure lead, perhaps from silver-lead ore, has been found in tombs.’ They alloyed gold with silver and with copper. They also gilded silver and copper with a gold amalgam that was put on by the means of fire. A sheet of copper is gilded with a very fine layer of gold hammered over it. In soldering, they used alloys of silver and gold to solder silver, and alloys of copper and gold to solder gold. Sheets of gold as fine and delicate as note paper have been found in Mochica graves. Cascabels were made on stone molds, and metals were polished with stone burnishers.. GOVERNMENT The remains of urban constructions, the expansion of agriculture: through the great irrigation works, the outstanding architectural mon- uments, and the network of roads throughout the Mochica country attest to a life organized by mature governmental methods. In the high degree of artistic and technical attainments, it is suggested that governmental influence was directed toward great material achieve- ments and the diffusion of cultural knowledge. The Grand Sefior, or Supreme Ruler (Cie. quich),* considered of divine origin, was dominant throughout the Mochica territory. His visage is encountered on funerary ceramics in archeological sites throughout the Mochica valleys. Sometimes these vessels represent him in full youth; at other times he appears in the sober majesty of the adult ruler. The Cacique (Alaec) was the regional governor, and his effigy is encountered only within the valley or sector of his particular administration. The Grand Sefior, Cacique, was the military chief as well as civil ruler. Representations of this man with great feline teeth indicated his divine origin and religious functions. Commoners approached the rulers with attitudes of great reverence, their hands placed together and head inclined toward the ground. When invited to a banquet by one of the chiefs, the guest always sat on a lower level while the host ate upon a throne covered by a sun. shelter. Women never appeared at such social functions.. 7 The presence of cinnabar (mercuric ore) in: Mochica graves shows that this mineral was known. The author considers it not unlikely that the Mochica controlled techniques by which they freed mercury from the cinnabar and used this metal in the extraction of gold ore.—AUTHOR. See Root’s discussion of metal- lurgy in prehistoric Pert, volume 5, Handbook of South American Indians, in connection with this.—EDITOR. 8 This name and that of the divinities, mentioned subsequently, are taken from a vocabulary of the historic Mochica collected by Father Carrera 168 SOUTH AMERICAN INDIANS [B. A. E. Bun. 143 The rulers presided at festivals, hunted and fished as a diversion, and were conducted on magnificent litters, attended by a great retinue, to different parts of their realm. During such visits or journeys they maintained contact with all parts of their domain by means of mes- sengers. The rulers were severe and unmerciful in the administration of justice. Delinquents were punished with mutilation: cutting off the upper and lower lips, the nose, and the feet. In serious cases, they stripped the guilty of his clothing, skinned off his face, and stoned him to death. Afterward the body was abandoned to the birds of prey. Such sanctions were apparently carried out in public with great display. WARFARE The war chiefs were both protectively and gorgeously attired. The helmet or headdress was amply quilted in order to deaden the blows of the mace. The great ear ornaments were probably not only a mark of rank but also a protection for the side of the head and face. Warriors wore a breechclout, skirt, shirt, and bracelets of metal that protected the wrists. The common warrior usually wore a semicircular or coni- cal helmet of great thickness, and carried a mace that had a sharp metal point on the handle end. He also hurled darts with the estdlica, or spear thrower (pl. 71, g). A semicircular knife was used in close com- bat. Blows of the mace were warded off with a small circular or quadrangular shield fastened to the wrist of the left arm. Armies made use of scouts, who moved ahead of the vanguard of the troops. These soldiers are represented on the pottery as keeping watch from the peaks of hills. War trumpets were made of the Strom- bus galeatus or of pottery. Dogs were carried with a special harness into battle in order to distract the enemy in the clamor of the battle (fig. 20, a). Prisoners were conducted naked from the field of battle (fig. 20, 5), and were sacrificed to the gods by being thrown from the mountain tops (fig. 20, c). Their bodies were afterward quartered and taken home as trophies, LIFE CYCLE During childbirth the mother, in a sitting position, was sustained by a man with the midwife assisting. A mother carried her small baby on her back in a manta or in a small cradle made of wood or wild cane, in which the baby was held by a net which allowed it freedom of the legs. Children were given pottery toys representing animals, humans, and utensils. They were also amused with rattles and whis- tles, As children, they were schooled in a specific art or industry. It is observed that there are very few ceramic representations of children; they appear only with their mothers or other adults. Tombs of children are rarely encountered. The children were buried with their Vou. 2] CULTURE OF N. COAST OF PERU—LARCO HOYLE 169 toys some of which are excellent miniatures of regular pottery vessels. In general the children ran naked, although some of them wore a very simple shirt. Women had full care of the children and conducted other domestic work. They also helped the men in more arduous labors. The men engaged in construction work, irrigation, tilling the fields, working in the mines, and in the opening of roads. They fought wars, hunted, iv. MON LY BP x AES a@ LS? GX So ose Soe = UPPER LIMIT OF POTATO CULTIVATION -<— —- - ------—-----—-------- Se STA rg AT sane ses] BES WITHIN THE LIMITS OF AGRICULTURE 4 Figure 30.—Climatic cross section of a typical intermontane basin in the Peru- vian Andes. The cultivation and pasture zones are shown, and the thickness of the dark symbols on the right is proportional to the amount of each staple that is produced at the corresponding elevation. (After Bowman, 1916, fig. 35.) In the Colonial era, the Spaniards first preempted the key basin formations, such as the Valleys of Cuzco (11,000 feet (3,000 m.) above sea level), Jauja, Huamanga, Yucay, and Anta, where, depend- ing upon altitude, the possible crops were maize, barley, wheat, and potatoes, or sugarcane, alfalfa, and fruit, as in Abancay Valley (6,000 to 8,000 feet (2,000 to 2,500 m.) above sea level). The most lucra- tive encomiendas were established in the deeply canyoned environ- ments, that is, in the quechua proper, while the puna regions of the Collao were long left untouched by Europeans, until well after the middle of the 16th century (Levillier, 1921-26, 1:19, 127; Montesinos, 1906, 1:193). Here, then, we shall use an empirical definition of the term “Quechua.” It refers specifically to certain Colonial Indian groups, to distinguish between pre-Conquest and post-Conquest versions of Peruvian culture. Those human agglomerations are regarded as Quechua which speak the Quechua language, or inhabit an environ- 334 SOUTH AMERICAN INDIANS [B. A. E. BULL. 143 ment of deep Highland valleys known as quechuas. Quechua- speaking peoples inhabiting the puna or the Coastal plains may, of course, be regarded as Quechua. Quechua also, however, may be certain non-Quechua-speaking Indians of the deep Highland valleys, from Northwest Argentina, North Chile, and eastern Bolivia to the Ecuadorean Highlands.! POPULATION The Viceroy of Peri was concerned mainly with the government of the two great Audiencias of Lima and Charcas (pls. 87, 88, 89). The Audiencia of Lima included most of what is known today as the Republic of Peri, and the Audiencia of Charcas included southern Peri and Bolivia. Here, therefore, any reference to Colonial Pert may be taken to concern only these areas which are today defined as Peri and Bolivia, and in which the Quechua are most thickly dis- tributed. Modern Ecuador, on the other hand, corresponds to the Colonial Audiencia of Quito, and its dense Quechua populations were subject to that government rather than to the Viceroyalty of Pert. Many differences may be pointed out between the Quechua of Ecuador and those of Pera or Bolivia; the discussion of the Colonial Indian populations of Ecuador should, therefore, be conducted separately. (See Murra, this volume, pp. 785-821.) As for the Audiencias of Lima and Charcas, their territorial defini- tions remained fairly stable throughout the Colonial era (fig. 31), and the populations may be studied by means of various counts taken at intervals during the 16th, 17th, and 18th centuries (table 1). TaBLE 1.—Total Indian populations in the Audiencias of Lima and Charcas Number of | Number of Date Indian Indian Source persons | tributaries LUT) bet ee ee oe ee 2 ae 1, 490, 137 315, 335 | Zavala, 1935, pp. 323-26. TD io ee mare ne Bien TILE, 2 ene ere el eRe LSE Tee 287,395 | Vazquez de Espinosa, 1942, pp. 703, 720. NORGE SEARS Ee _ Peas a FA 1 £230) 798 (saris) At Diez de la Calle, 1648, p. 18a. PSUS SL Re ee ce ee anew ee a Se Se oe So 275,078 | Morales, 1871, p. 333. G28 Soe oe a SE Mee Pee eS Pe Se ee EES 231,008 | Vazquez de Espinosa, 1942, pp. 703, 720. Ges Se ee Se ee eee 612, 780 143, 363 | Manso, 1859, App. p. 7. {SUSE Tee soe ears SEEe eek ek ee 610, 190 141, 248 | Escobedo, in Hesaike, 1901, p. 92. OG is stem Oe. eee eS eS kee | G08)894 jose ee ee 2 Memorias, 1859, vol. 6, App. pp. 6-9. | The accompanying graph (fig. 32) reveals a movement of popula- tion density characterized by unrelieved loss. At no moment during the Colonial era do the Indian populations appear to have undergone phases of recovery, such as those determined for the 16th century in México. It is also to be noted, however, that not until 1720 did any great losses through pestilence occur in Per. According to Dr. 1 During the historic period, Quechua has spread widely also into the Montafia of eastern Peri and Ecua- dor and even somewhat into the Amazon Basin. These areas will be described in the Handbook, volume 3.—EDITOR. Vou. 2] COLONIAL QUECHUA—KUBLER 335 7, *Saguaden Gactilsibe, wef Bat an Os . Ly. ae ‘ogi = ca i) Set ne =— BOUNDARIES AND TERRITORIES Lines Names 1665 .xx2 AUD DE LIMA 1775... AUDH DE LIMA 1859+—+ Figure 31.—Map showing boundary changes in the Central Andes, 1665-1859 (After Ogilvie, 1922, fig. 41.) 995682—46—_—24 336 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 Cosme Bueno, the plague of 1720 reduced the Sierra Indians by two- thirds (1763-78, n.p.). Although such losses in México during the 16th century threatened the extermination of the Indian race, no events of this class are known for the early Colonial history of Pert. (Kubler, 1942 a, pp. 606-43; Varinas, 1899, pp. 208-16; J. T. Polo, 1913.) Hence, the great decline in the Indian population of Pera before 1720 must be assigned to other sources than disease alone. 1,490,317 INDIANS 300 1586: 1,230,798 INDIANS 250 : 150 456/ 1572 ‘Sg! 1628 1720 1754 /78t FIGURE 32.—Indian population decline in Colonial Peri. Census for the Audien- cias of Lima and Charcas, between 1561 and 1781. Figures on vertical axis represent hundreds of thousands of Indian tributaries. Varinas, for instance, attributes loss to the greed of those respon- sible for the administration of justice among the Indians (Varinas, 1899, p. 216), after noting that otherwise such losses were inexplicable when the great fertility of the Indians, the long absence of wars, and the superior moral climate of Christianity were taken into account. As a very general explanation, this may be allowed to stand. (See also Romero, 1923 d). Certainly no other historical factor may be invoked which will account for the startling rate of decay. The Civil Wars and the schismatic state of Indian society in the 16th century, the inhuman service of the mines in the 17th century, and the oper- ations of debt servitude in the 18th century were all causes of a state of Indian disaffection which culminated in the great rebellions of the Late Colonial Period. But the charge of corrupt administration Vou. 2] | COLONIAL QUECHUA—KUBLER 337 voiced by Varinas tells nothing of the efforts at reform, or of the ways in which the population declined. A remarkable deceleration occurred in the rate of decline between the years 1572 and 1591. Both before and after this period the rate of loss appears to have been far more rapid. It is not unreasonable to associate this change of rate with the administrative reforms of the viceregal government of Francisco de Toledo. The systematic re- ducciones, the urban concentrations and regroupings, the administra- tive reorganization of the yanaconate, and the many other measures enforced by Toledo are surely reflected in this section of our curve. In other words, the achievement of the Toledan government may be interpreted as the regrouping of the Indians. The process of dis- persal to inaccessible areas was not only halted, but reversed. That Toledo’s measures, however, were merely remedial and palliative is demonstrated by the rapid acceleration in the rate of loss before 1628. Precisely how may this rate of loss be described? In less than two centuries the Indian population was more than halved. It is a de- cline with which a greatly accelerated death rate apparently has little to do, as we have inferred from the relative absence of epidemics. Rather does it seem necessary to account for the great losses by move- ment of populations away from the Viceroyalty. Im effect, such a hypothesis is supported by a comparison between the regional figures compiled in 1628 and again in 1754 (table 2). Of a total of 73 prov- inces, 62 are listed with figures in both counts (table 2). Of these 62 provinces in turn, all but 11 show loss between 1628 and 1754. These 11 provinces, however, gained by 1754. Three of them, the Cercado Province of Lima, and Cuzco and Cajamarca Provinces, were urban areas. The remaining eight, with exception of the Province of Piura, where Coastal agriculture was practiced, are all situated along the eastern Andean frontier, in the upper ranges of the Montajia zone. Paucartambo Province was a coca-growing district in the upper Montafia; Carabaya Province was noted for its resources in placer gold, and may also be designated as a Montafia province; in Porco Province important silver mines were situated; Tomina had the Montafia at its eastern border and was a great agricultural province growing the food for the laboring populations of the mines; Cocha- bamba had the same character, as a food-growing area for the service of the mining industry; the high plateaus and quechuas of Sicasica Province also provided food for the miners; and Larecajes Province edged well down into the Montafia. The suggestion lies very close that these increases at the south- eastern frontiers of the Audiencia are merely the official reflections, within administrative boundaries, of a far more extensive process of population displacement. Reports are frequently encountered, in connection with draft labor (mita) at the mines, which tell of the 338 SOUTH AMERICAN INDIANS [B. A. B. Bunn, 148 TABLE 2.—Movement by provinces between 1628 and 1754 in the Audiencias of Lima and Charcas : Amount i : Persons } Persons | Move- Economic type of 4 NOES Province in 1628 | in 1754 | ment province cree Tima 2.5 o see El Cercado..-..___- 1,921 2,078 | Elus_...| Urban: 2222422 eee 157 DO... -xscee Catietes: 2s cesk a: 2 3, 984 2,861 |, Minus, |.2--o22-242--6--24$524-_ 8 eee ‘Do. Se Tents ee 4, 222 3, 162 |...d0...02|---2-22--oeencloe ee Dos tvgeek wiauyosstyy. 7? 15, 241 6/885 422 -dOv. . 2\s.-Jaasscee teed ee ee ae paler cecil Feecoutir Sy Sr ds 15, 835 Ge | 22200. 3.) 22 - one ene na ee ee Os eae i re 15, 518 8,162) |22.dos..2| 22 -cel S ne Bete eee euencay aaa a 8, He 5,184 |...00_.._|) 22-00 -n eee Oey eee NAG ale 4 , 16 Cf on an | ones Sere. Dore sees Cajatambo-_-_-____- 25, 577 6,648 |... 00". |_.--- a eee DO ig siete inaylas: «2.5 - 28, 751 12,541. | 3do..2.|2s0s-232s2.-3 ee eee Oe eee Conchucos.__--_-- 25, 512 10) 709 |_-d0___ _ |. 8. 252. eee DOs wee Fass Huamalies_______- 8,,140)| =. cdo. s2.| cLewc.42-14-. 2 ee eee Dos oes Huanuco...._____- MOPASDIN? TaNAg7 do... |... hn Dome ates Manmade POY. wy 30, 266 8 454 | d0secz|scsese-. tt 2. eee us ne Eppa A aule = Speen a 32, 021 21, 062.| @do |S... 2... eee BOUT Oe Sse nuplosS 739 DOe2 ee 1d bs alia oA 8, 700 155108) |" Blus2-_—| (Coast farmings== == -se== 6, 406 Doss tet .. Cajamarca_-_____-- 36, 817 42,187 |__.do____| Urban mercantile_-____- 5, 370 DOM eter: Huamachuco._-__- DOP. cccatoe Cajamarquilla_-__ 6, 453 3986! |. Minus: ;|2s22-2.2-2 222) fe ae eee DOE Le fener Chachapoyas_--_-- 3, 750 DO he Luya-Chillaos____ 9, 433 2,014) 2. Oe. oi ele. cose see eee ee ID OUSOVEN Es WAM AS22 eee ae | owe Oe ee Cn ee lenea cease Hupmanga a ee Humana eae eee sen Wi Q6 Qiu) aed subenea ce. sechebe se bk ee One ee ee UTC ae eee 12, 337 6; 935" |---00_..-|--2) 2-2-2 coe eee IDO. 2228-34 Angaraes_._______- 10, 595 9,953) | 2. VA if YZ} fe a ee carrer ) S ee OE ety pele secre comme guile <== DMF ——_ SS — I< - ——— ——<—— —=—==_ -— —— = —— YS xX <7 —_, SaaS 7 = a = —— = FZ DPIZEZEZYIIZD 27 ZEISS w Figure 36.—Colonial Peri. a, Don Melchor Carlos Inga; b, a municipal scribe; c, an Indian domestic interior; d, the July harvest, showing dogs and chickens. (After Guaman Poma, 1936, pp. 739, 814, 871, 1049.) : Vou, 2] COLONIAL QUECHUA—KUBLER 363 Among the women, Indian dress persisted intact at high rank. A great lady (coya) is shown in the lliclla fastened by a jeweled tupu (ibid., p. 757). Vazquez de Espinosa notes a few types of regional women’s headgear. In Los Canas a coiled black scarf was worn, but in the Collao, women affected high black woolen bonnets sur- mounted by crescent-shaped ornaments (Vazquez de Espinosa, 1942, pp. 602, 606). In 1793, Haenke observed that the process had been reversed; among ordinary Indians the women’s costume was most thoroughly Hispanicized, and the men retained old-fashioned dress (Haenke, 1901, p. 100). It may be that in Haenke’s time the traditional Indian female costume stood much closer to European style than in the 17th century, and that the men’s costume looked Indian because it was archaic, retaining many 17th-century elements (pl. 86). Such, at least, has been the history of peasant costume in Europe. The dress of the great hereditary Indian curacas in the 18th cen- tury continued to approximate European court styles (fig. 43, p. 393) An example is given by Tupac Amaru’s costume in 1781. He wore blue velvet suiting with gold braid, a cloak of the same material in scarlet, and a three-corned hat over the Inca fringe upon his forehead. But over the velvet suit was a richly embroidered unco; and from a gold chain at his neck hung a golden figure of the Sun (Angelis, 1836— ay pe 11): Among ordinary Indians, the poncho (pl. 91), which differs from the unco in that the side seams are left unsewn, was introduced, probably from Chile, in the 17th century (Montell, 1928, pp. 238-42). The uses of the poncho are closely associated with horseback riding. Skinner (1805, p. 349) points out that it was ultimately adopted by Spanish cavalry in Europe, for its protection to horsemen against the weather (pls. 91, top; 92). MANUFACTURES No aspect of Colonial life is so poorly studied as its material cul- ture. It is impossible at present effectively to distinguish among Indian, Mestizo, and Creole productions; the regional manufactures are badly confused, and the periods of workmanship show great lacunae, especially in the 16th and 17th centuries. Of extraordinary importance is the fact that the Quechua communes often established and owned community workshops. Mita laborers were assigned to these establishments, from which the income was reserved for tribute payments. Such workshops were usually given to textile production (Pefia, 1698, p. 332). Nearly every Highland province supported textile industries; obrajes are recorded for Cuzco, Huaylas, Cajatambo, Conchucos, Hudnuco, and Cajamarca Prov- inces. (Manso, 1859, p. 149; Bueno, 1763-78, n. p.; Vazquez de Espinoza, 1942, passim; Zimmern, 1944.) 364 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 Certain districts at the eastern frontier made a specialty of lumber industries, and in Angaraes Province, many Indians were carpenters and joiners, producing furniture for sale in neighboring districts. At Racche in Tinta Province, ceramic containers were manufactured, and glass factories existed in Cochabamba Province. In La Plata Prov- ince, the cylindrical wooden vessels for chicha, called keros, were made and painted. Hu4nuco supported a local industry of manu- facturing roof tiles from the rinds of gourds (Vazquez de Espinosa, 1942, pp. 448, 655). Silversmithing was an important industry, of which the main Colo- nial centers were Cuzco and Jauja, where the Indian craftsmen con- tinued to use the cubical bronze hammers of antiquity (VA4zquez de Espinosa, 1942, pp. 475, 599). Specifically Peruvian are the stirrups, braziers, tupus, and alms dishes of Colonial manufacture (Harcourt, n. d.). SOCIAL AND POLITICAL ORGANIZATION Encomienda.—The conversion of the individual Indian from the status of an Inca subject into a Colonial Quechua was achieved by means of the Spanish grants known as encomiendas. The reorgan- ization of Indian society, the orientation of the economy, and the adulteration of native religion were made possible by this systematic insertion of Spaniards as colonists at all points where a dense Indian population existed. The term encomienda designates the fiduciary commission of the labor of Indians to a White colonist who is their trustee. 'The Indians providing the labor constituted the repartimiento; the colonist is the encomendero (fig. 37, 6) (Torres Saldamando, 1879-80, vol. 3). The encomendero owned no title to the land on which the Indians of his repartimiento worked and lived, although he was entitled to a fixed share of the fruits of their labor. In the exercise of his rights, the encomendero destroyed the agrarian economy of the region by demanding tribute in precious metals and stones, or in kinds of produce which were not locally grown. To enforce his demands, the encomendero was obliged to terrorize the curaca, who in turn terrorized the tributaries in his charge (Santillan, 1879, p.55ff.). This situation arose in part because the encomendero was forbidden to visit the Indians of his repartimiento or to demand any hospitality from them. The law was more often broken than observed, but in theory all supervision over the Indians was meant to be exer- cised by the curacas and the resident clergy, but not by civilian col- onists (Anonymous, 1889, p. 166). Matinzo describes in detail the internal, or “minor,” government of each repartimiento. As in pre-Conquest society, the moieties, called Hanansaya and Hurinsaya, were maintained. Each had its chief Vou, 2] COLONIAL QUECHUA—KUBLER 365 f(A Brey Cs DAR Tiiovi p.. ey traci mij nie (adi aft W tt SPO ATO ykXd penn sA u u u Figure 37.—Colonial Peri. a, The Viceroy Cafiete and Sayri Tupac; 6, an encomendero with Indian servants; c, a curate and curaca; d, the mayordomo of a cofradia. (After Guaman Poma, 1936, pp. 440, 551, 588, 688.) 366 SOUTH AMERICAN INDIANS [B. A. E. But. 148 (primera persona, segunda persona), who refrained from all inter- ference in the affairs of the opposite moiety, with the exception that the Hanansaya chief (primera persona) enjoyed precedence in matters of protocol. Each moiety consisted of several ayllus, and at public functions the representatives (curacas) of the ayllus of the Hanansayas were seated on the right, on slightly more elevated positions than the Hurinsaya curacas. The primera persona held the right to call meetings and direct gen- eral business in the sense of the chairman of a committee. He was also the public accountant, possessing greater skill in arithmetic than many Spaniards. The curacas of each ayllu were entrusted with the collection of tribute. The total tribute for the community was depos- ited by the primera persona with the corregidor, who placed it in the community cashbox (caja de comunidad), to which he and the primera persona held the keys. The primera persona also selected and assigned the Indians to the work at the mines, in the Spanish settlements, in the tambos, and at all the various forms of draft labor. The living of all these various officials was supplied out of the tribute collected from the community. Although the curacaships and higher offices were traditionally hereditary, in the Colonial era the rule was that the curacas and other officials sanctioned by the Spaniards had arisen to their posts by cunning and by connivance with the encomen- dero or with the visiting judge (Loaysa, 1889, pp. 586-89). In general, then, most rich areas of Peri were caught during the 16th century in the network of the encomiendas. Few provinces escaped colonization, and even there, as in the case of the Neo-/nca in Vilcabamba, the flight from European culture produced alienating effects in Indian society. Colonial Indian life as a whole was brought into being by the encomienda; as such it was a creative institution, mediating European culture for large native populations. At the same time, however, the proliferation of repartimientos brought about certain deep disorders in Colonial life. For example, by royal decree of October 28, 1541, the pastures and streams were reserved for communal use by all members of Colonial society (Montesinos, 1906, 1:119). In practice, however, the vast holdings of the Europeans in livestock rapidly crowded available pas- turage to a saturation point at which the Indian was excluded from large stock-raising activities. Thus, although the encomendero often was granted no title to any land (Falcén, 1867, p. 459), he had access to such a reservoir of free labor that he could accumulate huge herds, 2,000 to 20,000 head in number. In some areas, not only the com- munal pastures but also the food-growing areas were preempted for the nourishment of these herds. As the land was withdrawn from Indian use, so also were the irrigation waters brought under European control, so that the Indians could neither raise adequate crops nor Vow. 2] COLONIAL QUECHUA—KUBLER 367 water their own few cattle. A contemporary observer notes that the Spanish stock-breeding industry, made possible through encomienda, was partly to be blamed for the great losses of population suffered dur- ing the 16th century, since the Indians were forced to move their fields higher and higher, in barren and rocky terrain, where the yield was capable of sustaining but a fraction of the former population (Molina of Santiago, 1916, p. 149). The same general effect was produced by the foundation of towns and monastic establishments. In Highland Peri, arable valley bot- tomland is the most valuable possession in Indian society. An exqui- site balance between cultivated land and population density had been achieved under the /nca regime. With the advent of the Spaniards, however, the foundation of numerous towns on bottomland incapa- citated the best soil for agricultural production. The Indians previ- ously resident upon the site of the town were forced to emigrate or to enter Spanish service through encomienda (Lorente, 1867-72, 1:22; Anonymous, 1889, p. 183). The process whereby the Indian was expelled from the regions of arable soil was further accelerated by direct purchase. The situation was such that the Indians, to escape the heavy labor drafts attendant upon encomienda service, sold their communal lands at low prices (Montesinos, 1906, 1:254, 282). At all times the sale of land by Indians to encomenderos was subject to regulation, but by 1562 cheap sales had become so numerous that further transactions were forbidden excepting by viceregal license. 3 Another pernicious effect of encomienda was that the native ani- mal resources of the Highland were systematically destroyed to make room for European stock. All the colonists participated in the pro- cess, even Negroes and Mestizos, until, in 1556, the herds of llamas and deer were so few in the once rich valley of Huamanga that serious food shortages resulted (Anonymous, 1889, p. 187; Montesinos, 1906, 1:243). Corregimiento.—After the 16th century, the encomienda was dis- placed by Crown government (corregimiento); the administration of Indians by private beneficiaries was supplanted with the institution of Crown officials called corregidores (fig. 38). Before the constel- lation of Colonial customs associated with Crown government can be discussed, however, it is important to examine the main events in the decline of the encomienda system. The tenure of encomienda was never made permanent, in spite of the prolonged agitation by encomienda holders and missionaries. (See Zavala, 1935, cap. 6, pp. 183-223.) Yet the number of encomi- endas rose to a peak at the end of the century. The following table 3 suggests the spread of the institution for the Viceroyalty of Perd. 595682—46-——26 368 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 TABLE 3.—Rise of the encomienda Corregi- Number | Number | Vacant | mientos Date jof reparti-| of enco- | encomi- | (Crown Source mientos |menderos| endas | reparti- mientos) 1648 ue Se oe aa a S00R Ee ee Zarate, 1870, pp. 181-82; Montesinos, 1906, 1:192-93, 196. 1556; [2 _2. oe | be ere el 2 Big ee sek bu Zavala, 1935, p. 326. 1561 477 427 24 18 id. 1574 614 970 6 21 | Lopez de Velasco, 1894. Lifian, 1859, pp. 303-04. 1591 775 (Goa) | ee ee 80 | ZAvala, 1935, p. 325; Torres Saldamando, 1879-80, vol. 3. 1628 Naan | ye 82 | Vazquez de Espinosa, 1942, p. 772 ff. 1In but 58 districts listed. As always with 16th-century statistics, arithmetical inconsistencies appear in the tabulations, but the general sense of the figures is reli- able. It will be noticed first that the number of encomenderos rose far more rapidly than the repartimientos. This was possible because many single repartimientos were fractionally assigned to several holders, and because individuals and institutions were frequently granted pensions. ‘These pensions, often called encomiendas, some- times carried no assignment of Indians. The increase in the number of repartimientos, on the other hand, is explained by new coloniza- tions and by the progressive fragmentation of already extant popu- lations into more numerous encomiendas. Most striking of all is the increase in the number of Crown repartimientos, called corregimi- entos. By 1628, their number had been stabilized, and thereafter remained constant. The process whereby this was achieved depended upon the fact that repartimientos normally escheated to the Crown at the end of their assignment to private individuals. If the encomi- enda were granted, as was the custom, for two or three lives or gener- ations, after that period, it reverted to the Crown, to be administered by the corregidor of its district. The number of encomiendas thus incorporated to the Crown was very high at the beginning of the 17th century. In New Spain, for example, there had been 721 encomi- endas in 1574 (Lépez de Velasco, 1894); by 1602, only 140 survived (Zavala, 1935, pp. 173, 314). The appointment of corregidores in private encomiendas was in- augurated by Viceroy Toledo in the 1570’s (Lorente, 1867-72, 1:20; Poma de Ayala, 1936, p. 489). Prior to Toledo’s government, the Indians suffered heavily from the unnecessary litigation into which they were drawn. Toledo’s primary purpose in establishing the corregidores was to reform this situation. ‘The Indians were hence- forth required to seek justice from their corregidor, who was also charged with supervising the collection of tribute and protecting the Indians of the repartimiento from unwarranted abuse. In legal affairs, be it noted, the Indians retained the right of appeal from their corregidor to the Audiencia. The living of the corregidor was paid by the Colonial government out of tribute, and in no case was the Vor. 2] COLONIAL QUECHUA—KUBLER 369 official to accept remuneration for his services from his Indian clients. Immediately after its institution, this reform interrupted the extor- tions practiced upon the Indians by the solicitors at the Audiencias, by the clergy, the encomenderos, and the curacas, with the result that COREGIMIEU'TO. i ELOREC a BLDA Tp me Dee ay? amg H 30 »lato Leg ty ‘ a | 1 2 . a D SSTSS eoigeoousys yrs bu Pro briwas Figure 38 —The Corregidor’s table. (After Guaman Poma, 1936, p. 505.) 370 SOUTH AMERICAN INDIANS [B. A. E. Bunn, 148 Viceroy Toledo felt it essential that the corregidores be continued after his administration (Lorente, 1867-72, 1:20). It soon became apparent, however, that from the point of view of humane administration, the corregidores were a worse evil than the situation they were intended to remedy. ‘Their salaries were so small that they found it necessary to supplement their incomes with pre- cisely the kinds of extortion they were meant to eradicate. As Whites their residence among the Indians was an exile which they alleviated by illegally surrounding themselves with superfluous deputies and servants (Aponte, 1867, p. 544). It should be noted that Viceroy Toledo also achieved the elimination of the large numbers of supernumerary Indian officials in each commu- nity. The mandones supervising groups of 10 and 20 Indians were sup- pressed and only the mandones of larger groups, such as the pachaca (100 tributaries) and the pisca pachaca (500 tributaries) survived. Thus, the cost of Indian government was reduced, but in place of the swarm of Indian officials, there appeared the Mestizos, Negroes, and mulat- toes with whom the corregidores surrounded themselves (Montesclaros 1859, p. 20). Indian participation in local government was thereby reduced to a minimum that characterized all the subsequent genera- tions of the Colonial regime. Tribute.—The fundamental quantum of Colonial government was Indian tribute. Tribute maintained the caste and status of the Kuropeans to whom encomiendas had been awarded. From the various forms of tribute, all public works, religious instruction, monastic foundations, institutions of learning, hospitals, and civil salaries were provided. In the individual Indian existence, the pay- ment of tribute occupied a central position, setting the tone for virtually all forms of activity. The great differences between pre-Conquest taxation and Colonial tribute must be explained in terms of the differences between the two economies. The Jnca economy was self-sufficient. Under Colonial administration, however, the delicate equilibrium between population and resources was upset by the introduction of a mercantile econ- omy based upon exports and imports. Vast quantities of local products were traded abroad for necessities and luxury goods, such as wines and textiles, armor and tools. The absolute demand upon Indian productive capacity was raised to a multiple of the capacity itself (Santillan, 1879, pp. 74~—75). The Jnca system of compulsion to labor had provided an elaborate and efficient scale of rewards for work achieved. The absence of such rewards in the Colonial exploitation of labor seriously affected the social morality of Quechua life. Since the curacas of huarangas and pachacas, instead of receiving exemptions for their labors of super- vision and administration, were reckoned as tributaries, they found it Vou. 2] COLONIAL QUECHUA—KUBLER 371 necessary to exploit their subordinates in order to meet their own tribute payments. Thus the cost of Colonial government devolved upon the members of the society least able to meet the charges, and the moral responsibilities of government were disregarded by the minor authorities. Mita.—During the protracted disorders following the Conquest. extravagant demands were made upon Indian labor, endangering the basic agricultural and stock-raising economy. A modification of the pre-Conquest mita was accordingly introduced as a measure for sta- bilizing food production, in which only a limited number of Indians was expected to appear for hire at stated intervals (Montesclaros, 1859, p. 23). The mitayo was paid for his labors partly in cash, in the hope that his sense of monetary values would be affirmed (Lifian, 1859, pp. 303-04). The proportion of draft labor drawn into service varied both according to region and time. Viceroy Toledo had legislated that only one-seventh of the dwellers in any settlement should be liable. In the 17th century, this ratio grew slowly (Mon- tesclaros, 1859, p. 23). By 1620, one-seventh prevailed only for the Sierra, excepting Potosi; one-sixth was the ratio on the Coast; and in the Audiencia of Quito, the most populous area of the Viceroyalty of Pera, one-fifth was drafted (Esquilache, 1620, in Memorias, 1859, 1:89). In the 16th century, the mita had been restricted to public works, but as early as 1615 the legal use of draft labor had been extended to textile manufactures and to mining enterprises. It remained for- bidden, however, to utilize mitayos in the search for treasure among the huacas, in the coca industry (cédula of 1609), in vineyards, olive plantations, sugarcane mills, and in lumber industries (Montesclaros, 1859, pp. 25-27). In any case, the nature of the work done in mita varied according to the resources of the province. In mining areas, all mitayos served the mines, as described below. In agricultural and manufacturing areas, the mita was dedicated to those activities. The exploitation of the great mines of Peri was achieved at all times during the Colonial era by Indian draft labor. An elaborate mechanism for the administration of this mita was first set up by Viceroy Toledo in the decade of the 1570’s. Later on, in the 17th and 18th centuries, the service of the mines enforced many cultural adjustments upon the thousands of Quechua families dedicated to it. The entire Colonial epoch in Perf has not incorrectly been designated as a vast religious and political organization for the exploitation of the mines (Belatinde, 1932, p. 11). On the basis of the census taken during his visita general, Viceroy Toledo assigned the necessary Indians to the mita of a given mining center from among the provinces neighboring the mines. For example, 17 provinces surrounding Potosi were known to contain a population 572 SOUTH AMERICAN INDIANS [B. A. E. Bunn, 148 numbering about 80,000 tributaries. Of these, one-seventh were annually allotted to the mita, numbering 11,199. Each mitayo, however, was not expected to serve more than 18 weeks annually, with the result that a man’s corvée fell due but once every 7 years for one-third of a year. In theory, then, no single Indian could be called to the mita of Potosi more than four or five times during his life as a tribute payer, between the ages of 18 and 50. It should be noted, however, that if the population of a given prov- ince dwindled, the mita assignments remained unchanged. In Chu- cuito Province, early in the 17th century, the mita actually called each individual every 5 or 6 years (Messia, 1603, 2n Lorente, 1867-72, 2:352). If the draft in a given province could not be mustered, the burden fell upon the curacas. When death, pestilence, or wholesale truancy depleted a community, its curacas were compelled by law to make up the missing man-days by hiring laborers at exorbitant rates (Messia, 1603, 7n Lorente, 1867-72, 2:369 ff.). This substitute labor was re- cruited from among the indios mingados, who offered themselves of their own volition for hire in the mines at a rate higher than that paid to the mitayos (Montesclaros, 1859, p. 42). In Potosi alone, there were said to have been about 40,000 resident mingados, whose wages in 1601 were twice the rate paid the mitayos (Fernandez de San- till4n, 1868, p. 451 ff.). The precise operation of the mita may be reconstructed in some detail from viceregal reports, which dedicated more and more space to the problems of the mines as their resources dwindled during the 17th century. Such impoverishment was, of course, related to the static condition of Colonial mining technology. The proclamation of the coming mita period was sent out 2 months ahead of schedule to the curacas in each community. The curacas in the minor government of each repartimiento thus had opportunity to muster their men and to receive payments from those Indians wishing to evade the mita (Salvatierra, 1899, p. 30 ff.). Upon being assembled all the mitayos of a given province set out for the mines in a great pro- cession. Thus the mita setting out for Potosi from Chucuito annu- ally consisted of 2,200 Indians with their families, amounting to more than 7,000 persons. Each mitayo took 8 to 10 llamas and some al- pacas for meat; the wealthier ones were accompanied by as many as 30 or 40 llamas to carry their corn and potatoes and their sleeping mats. The procession encamped each night of the journey, sleep- ing under the open skies. Each trip from Chucuito to Potosi was 300 miles (480 km.), lasting 2 months, and involving the movement of 30,000 to 50,000 animals. After 1650, the mitayos were paid a daily rate for traveling expenses. Vou. 2] COLONIAL QUECHUA—KUBLER 373 Of the whole crew, only a fraction returned to Chucuito at the con- clusion of their turn. Some had died, and others, having no herds or food left for the return trip, stayed in Potosi, coming under the high charges levied for residence there. Many others preferred not to return home to the local mita of Chucuito, which was additional to that of the mines, and escaped eastward to the relative freedom of the lowland valleys. If, however, the mitayo returned to his home, he had spent 4 months in travel, 4 months at labor in the mines, and 2 months in compulsory service in Potosi. As the 17th century progressed, not only were the mines themselves becoming exhausted but also their exploitation had depopulated the surrounding provinces. A substantial portion of each mita sent to the mines failed to return to the home province, and the greater part of the losses may be accounted for in terms of fugitives, either to other provinces or to the unconverted areas of the eastern Andean valleys. (Vazquez de Espinosa, 1942, p. 632; Navarra y Rocaful, 1689, 7n Memorias, 1859, 2:239-45.) About 1603 it was proposed that large, permanent Indian towns be founded near the mines. Actually, the Indians resisted these attempts at industrial urbanization, and the cost of the program was prohibitive because of the fact that most of the land near the mines had long since passed into private Spanish hands. The proposal to establish a permanent settlement for 37,800 Indians near Potosi was never realized. (Mes- sia, 1603, an Lorente, 1867-72, 2:357-58; Esquilache, 1620, in Mem- orias, 1859, vol. 1.) In 1720, moreover, the Crown actually com- manded the cessation of the mita of Potosi, and its replacement by voluntary labor. The decree was never executed (see Whitaker, 1941, pp. 21, 90), and the mita at the mines continued nearly to the end of the Colonial era (Armendariz, 1736, in Memorias, 1859, 3:152). Money.—-Before 1557, the money economy of Pertii was conducted in the absence of coinage (Levillier, 1921-26, 1:260-61; Montesinos, 1906, 1:245-46). Later on, when the dies for Peruvian coinage had been widely distributed, they were forged to such an extent by civilians and clergy that the value of Colonial money remained uncer- tain until Viceroy Toledo reformed and enforced the mint laws (Anony- mous, 1889, p. 192). Thus, the use of a monetary medium of exchange was attempted in the absence of valid forms of metal currency. The entire early Colonial system was founded, as regarded economic life, upon the assumption that all values, goods, and services could be assessed in terms of a money that had no physical reality. The pesos, ducados, castellanos, and tomines of Spanish currency were fictional in Perf. The Indian who was entitled to such and such a pay never saw it, but if he were remunerated, it was in bullion or in goods. The confusions arising from this situation were multiple. The powerful 3l4 ‘SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 debtor could assign arbitrary values of goods to the fixed scales of mon- etary pay and tribute with impunity. Among the Indians, further- more, whose notion of property was in terms of tangible objects, such as lands, crops, or herds, all shared communally, the idea of a money- value presented serious difficulties. Property.—The Indian sense of property can best be understood by reference to pre-Conquest customs regarding inheritance. It was a loose arrangement, in which the appointed successor to an estate provided from its revenues for the support of the family of the deceased. The heirs possessed the estate in common without dividing it, and the property was administered by a representative of the ayllu concerned. Yet no heir could claim a share of the harvest unless he had taken part at its sowing (Santillan, 1879, p. 2-709; Polo de Ondegardo, 1873, pp. 162-63). These customs persisted long after the Conquest, although Europeans often interfered, imposing a written testament designating heirs of their own choosing. What was true for inheritance was true for property in general. Land belonged to the people who sowed the crops in it; property was a function of participation in the labor of production (Polo de Onde- gardo, 1873, p. 164). The Spanish concept of property, however, as derived from the Roman laws of absolute individual ownership, was only partly intelligible to the Indians. The Spaniards, while encour- aging the differentiation between rich Indians and poor Indians, frus- trated the incipient sense of private ownership by the administrative compromise of imposing tribute exactions upon whole communities rather than upon individuals. When a curaca, for instance, had attained the wealth necessary to acquire the standard symbols of prestige, such as saddle horses and breeding stock and firearms, he was forbidden to possess them by sumptuary laws. ts Be Berccccke CTA & Beprages Moree OS REC EK. ‘ aS : SRN HON i’ SN ooh : Cac Ataars ‘ 4 Beara RA : 3 HI Naocas Sos ; PROG su & . sequel x — — CS wy BE PACAG SES - FEN SRN AO . ~ ASKA aS ARISEN So — SMe REO Vee Be ARES — BO oe Soren Aate : | tk, 5 SF PA RTE DE Gnanlgas @. AR? ~~. “Er OBISPADO PLA | PAZ Diuidids en todas sus Proumcias nuebarnente emmenva a << PARTE RE Yimana Q de aegun £2u Fes Meacias NeRCNE PLATE 87.—Seventeenth- and eighteenth-century maps of Pertti and Bolivia. Top: Audiencia divisions of 1657. of La Paz, 18th century. (After Sanson, 1657.) Bottom: The Bishopric (From Juicio de Limites entre Pertti y Bolivia, 1906.) RAEN sae NERA RI, AA AAAS SSALAAAI AA SLL A = omc Al CAA LLL LLL LLL CL A A i | | manatee erro (9061 ‘BIATIog A Nisg o1}U0 SoPUNT op OLOME WOT) “T8L{ ‘oozn~) Jo oltdoysig oy, -7yhey (CELA1 “BI[Ipowmlg A ouRD znID JaIJW) ‘eCLLT ‘SvoIBYyQ pus BUILT JO SBLNUVIPly oY} JO SBIOUTAOIG pue SOpIyIVd oy], -7faT ‘“WOIsed o[IYO YPION-eIATOg-nieg 94) Jo sdew AINjU9I-YUIE}YSI— "88 ALVIg SEHR 1h GEIACIYRR Wy 99 Tis uray wep oes se Penna RVINVOM Cry oe coae & Equatear Is dos. Belen Ce Bave dle Guavagin Bs TL G3) Jizan bpd nuit? RAO S | PARTIE SEPTENTRIONMEDE PEROU. ie an lon toe la pun oeculerdgle de on nn on Amazone breve ha (arke de Mire 1s COS DAMINE ne Le S pog gs toner Sodan aa ORBLE BOUT, # phe. Lacties NU hs de 20 ah Pyare ELLE JF leplenirion Bag acted ia ~nnt eponnngnopmninnannnannanthnannsseneetngeenene eC to, Ay ipa men oF 22 ees oo Tioga Yin le CoM is We Ge Certere- We Cee rg! a 4 1h Lan oe a BD 3 Ly ar i é ue ny tie tees tg Cig ar OD ae 3 Lig Z b; *Platae A ay j y ae j Vis chi An Oo 2544 Fong Los s Gehicas PRI MERIDIONALE he : 3 y pe VEROU tar le S Robert deVaugondy Bile de A Ronenr Googe. wd du Bon Avec Prankeye A7¢9 tad tacucs niarines on dune heures Se nee hiriariade 5 a ey ee Puate 89.—Maps of 1749 showing Audiencias of Lima and Charcas. (After Robert, 1748, pp. 204, 205.) if ye BURR: ee PERSIE RE ER OO Has PAE Te ke her: Bes . ville é Soaks Puate 90.—Highland industries in the Colonial Period. Top: Working the silver mines of Potosi. (After Vanderaa, n. d., vol. 32, pl. 42 b.) Bottom: A sugar mill, a mine shaft, and windmills. (After Chatelain, 1719, vol. 6, map preced- Inf pelle) A icaca ge) Lop Colonial art of Per 1 poncho from the Lake T PuaTE 91 iti Colonia et ey See g oc AS ee 5 Se m & Sao eS etch 1) Seis (= qg a0 S . “= Ss ey Ss She AQ ¢ fo) ae @m » . ats ee) iy a prose Sur I SF eee — fa coer gus S52 O aes pinden, 1941.) (After 8 1750. c1réa, BASS? ce i ag gees 4 ee Reet en er 3 Ti ‘ Se ee Oe gga np « MOD “> SS Gis oa Ne a mat EF 4 a a Ki Sx PuiatsE 92.—A section of a Colonial textile. From the Lake Titicaca region of Pert. The designs are combination of Jnca and European motifs. (Courtesy American Museum of Natural History.) Vou. 2] COLONIAL QUECHUA—KUBLER 383 nearly successful that the Franciscan friars in the city asked leave to depart, and Pizarro’s Indian wife, Dofia Inés, the daughter of Huayna Capac, attempted to cross the lines to join the besieging Indians. Had the campaign been successful, it was the alleged plan of the Indians at the siege of Lima to reconstitute the Inca state and, through the Spanish women remaining in Pert, to create a new generation combining the virtues of the two races (Valverde, 1879, p. 80). The significance of the rebellion of 1536-37 lies in the fact that it was the final Indian attempt to smother concentrated force with over- whelming mass. Manco conceived the rebellion, it is true, as a total effort in southern Perf, but he employed tactics analogous to those of game-beaters. If he succeeded in taking the lives of some 800 Span- iards during the 16 months of the rebellion, Cuzco held firm, occupied by only 80 horsemen among 190 Spaniards supported by some hun- dreds of Indian warriors, who withstood the pressure of Manco’s effectives, estimated at 180,000 men (Levillier, 1921-26, 2:391—93). Yet the siege of Cuzco remains a remarkable operation in an exten- sive campaign, during which Indian strategy, tactics, and material equipment were greatly improved. There was unity of command, there were efficient communications with other forces, and improved weapons were used, such as the weighted slings (aylyo), strung with llama tendons which could not be severed, and designed to entangle the legs of the horses. There were also staked pits, retarding forti- fications, and ambitious flanking movements. Manco had at his dis- posal a certain number of horses and firearms captured from the relieving parties sent out of Lima, as well as seven or eight captive Europeans who served as grooms, armorers, and powdermakers (Herrera, 1726-27, vol. 5, pp. 191, 193). These improvements, nevertheless, were far outbalanced by the traditional weaknesses of Inca warfare. For example, the campaign at Cuzco displays certain remnants of the ritual behavior of cere- monial war (Lépez de Gémara, 1749, p. 120). Manco’s great mass attacks were launched periodically at the full moon, whea his men were at the mercy of the cavalry couater attacks. At all times his troops were massed so densely that their pikes were of no avail against the horses. ‘Then, at new moon, when the horses would have been at disadvantage in the dark night, the Indians ceased attacking to perform sacrifices to the lunar deity, thus forfeiting the advantage of sustained pressure upon the enemy. This 20-day rhythm of battle was known to the Spaniards, and they exploited it accordingly (Kub- ler, 1944). It also became necessary for the Indians periodically to disband large forces which were sent in search of food. Indian needs were vastly greater than those of the small Spanish force, and after the fall of Sacsahuaman it was an easy matter for the Spaniards to gather 595682—46——27 384. SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 enough corn at nearby fields (Valverde, 1879, pp. 34, 43). The Indians, on the contrary, had to demobilize huge numbers of men for agricultural labor, and this cumulative drain upon their strength finally contributed to the failure of the rebellion. Manco commanded a popular army at Cuzco, levied from all avail- able sources. It certainly lacked the professional training of the pre-Conquest Inca armies, and in battle its behavior was that of a surging rabble, while in defeat the defection of large contingents was a daily phenomenon. The Spaniards systematically exploited all the weaknesses of such an agrarian militia. They harvested or burned the crops sown with great difficulty by the Indian armies. Prisoners of battle were mutilated, usually by cutting off their right hand (Valverde, 1879, p. 44). In the fourth month of the siege, the Span- iards adopted a devastating practice. The besiegers composed a@ mass of humanity in which only a fraction was tactically effective. The great remainder consisted of women and children and families of the fighting men. The food and care of the attackers were provided by these camp-followers, and Hernando Pizarro ordered that all women caught in battle be killed (Valverde, 1879, p. 52). The progressive demoralization caused by this practice contributed largely to Manco’s withdrawal from Cuzco in February 1537, in the 10th month of the siege. Finally, a major cause of the failure of the rebellion was Manco’s lack of control over the proletariat of the yanaconas distributed throughout Perf. These servants functioned as spies, domestics, pro- viders, and nurses for the Spaniards. Above all, they could mingle freely among the rebels to learn their plans of attack (Pizarro, P., 1921, 1:330). Against such large-scale espionage, Manco had no protection. In Manco’s defense, it should be pointed out that he had no choice but to attack in mass. He had had some experience with Spanish weapons and tactics during the campaign against Quizquiz, in 1535, where he served as an Indian auxiliary. When planning his own campaign against the Spaniards, however, Manco faced the dilemma of choosing either to use untrained masses of men or to approximate European tactics without the proper equipment. That he decided upon the mass attack was reasonable and fitting. The rebellions of the 18th century.—The last century of the Colonial era was punctuated by Indian rebellions, occurring at fre- quent intervals throughout Per4. The normal form was that of local riots, occasionally involving whole provinces, in which the animosity of the Indians was directed against their corregidores. Indian rebel- lion was usually conducted within a framework of loyalty to Church and Crown; these institutions were rarely questioned, and the causes of social oppression were identified by the Indians with the minor Vou. 2] COLONIAL QUECHUA—KUBLER 385 resident officials. None of the rebellions achieved pan-Peruvian pro- portions, not even the great revolt of the 1780’s, under the brief lead- ership of José Gabriel Tupac Amaru and his associates. The earliest riot of which we have any record occurred in Lima during the government of Viceroy Castellar (1674-78). It differed considerably from the rural and provincial revolts of the following century, in that it developed among the atypical Indians of the metropolitan area. In 1730, in the Province of Cochabamba, where the greatest number of Mestizos in the Viceroyalty of Pert were resident (Manso, 1859, p. 194), another tumult broke forth when the regional inspection attempted to take the census (Armendariz, 1736, 1n Memorias, 1859, 3:280-87). At about the same time, in Asdingaro, Carabaya, Cota- bamba, and Castrovirreina, the exasperated Indians murdered several corregidores. In Cotabamba, the disturbance was precipitated when the corregidor made the mistake of attempting to collect tribute pay- ments during a Church festival. In 1734, at Andahuailas, another revolt against a corregidor was recorded; in this affair, the clergy evi- dently opposed the official (Garcia y Sanz, 1876, pp. 88-89). The most ambitious and extensive outbreak prior to the great rebellion of 1780 was the so-called Tarma Revolt, which erupted in 1742. Its history will be recapitulated in the Handbook, volume 3. Juan Santos himself disappeared from Spanish view after 1750, but it is to be noted that with him, as with the later leader, José Gabriel Tupac Amaru, a steadfast adherence to the Catholic Church governed his actions. Also like the rebellion of 1780 is the fact that the revolt of 1742 developed eastward from the Montajfia, rather than west, toward the administrative center of life in Pert. In the west, nevertheless, may be noted the so-called Revolt of Huarochiri. It began in 1750 in Lima with a conspiracy among urban Indians. The conspirators were executed, but the revolt soon broke out again in Huarochiri Province, where the corregidor’s lieutenant and his family were murdered. ‘The responsible Indians then wrote letters inciting the rest of the province to rebellion. An Indian militia was formed and commanders selected. ‘The roads were blocked and bridges broken. Soon, however, Indian resistance was crushed by a Spanish force of 400 men. It was discovered that one of the ring- leaders had made a plan of Lima and laid out the strategy for invading the capital; another of the leaders had been to the Montajia to confer with Juan Santos, and had traveled Pert in disguise as a merchant, arousing the curacas (Manso, 1859, pp. 94-99). The revolt of Tupac Amaru.—The history of the revolt of Tupac Amaru is intricate and obscure; to clarify its course, the reader may refer to the following list. of its episodes and their general calendar (pls. 87, 88): . 386 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 148 — . The Chayanta Rebellion (August 1779 to May 1781). 2. José Gabriel Tupac Amaru and the Siege of Cuzco (November 1780 to May 1781). 3. Julidn Apasa and the First Siege of La Paz (March to June 1781). 4, Andrés Tupac Amaru and the Second Siege of La Paz (August to October 1781). 5. Final ome Miguel Bastidas, Felipe Velasco Tupac Amaru. These various episodes, it should be emphasized, were confined to the southern Highlands of Pert. Their relation to one another is not that of an organized rebellion conducted simultaneously in various theaters under unified direction. On the contrary, as one episode waned, another matured to take its place, and the rebel groups migrated ac- cordingly. Leadership was discrete and confused, but it is likely that fighting Indians from certain provinces, such as Pacajes or Chayanta, participated in nearly all the major episodes. The Chayanta rebellion unfolded in the Viceroyalty of Buenos Aires, but its participants were Quechua and Aymara Indians resisting the extortions of corregidores. La Paz was briefly besieged in Feb- ruary 1781, until the arrival of troops from Buenos Aires (Segurola, 1881, p. 375). By this time, the revolt of Tupac Amaru was well underway; Cuzco had been besieged in January 1781, and there is no doubt that the rebels of Chayanta were in communication with their colleagues to the northwest. (Angelis, 1836-37, p. 223; see also Segurola, 1881, pp. 375, 379.) In general, the revolt of Tupac Amaru, insofar as it was directly under his control, suffered from inadequate force and organization. At no time did he command the full resources of more than five provinces: Lampa, Aséngaro, Carabaya, Chucuito, and Paucarcolla. Neighboring Aymara provinces, such as Larecaxa, Omasuyos, and Pacajes, never unconditionally supported his insur- rection. In public declarations, Tupac Amaru phrased the aims of the rebellion in the language of administrative reform and did not formally offer his followers the promise of an independent Peruvian Indian state. Others did so in his name after his death, such as the illiterate Julidn Apasa; they, however, were Aymara-speaking, and far less Hispanicized than the descendant of the Inca rulers. Tupac Amaru’s communications were poor. He was unable to make effective use both of the Chayanta Rebellion and of the Aymara insurrection about La Paz. His rebellion was constructed only within the weak pattern of the hereditary local curacaships; it was sustained by no other Indian administrative network, for with the institution of the corregidores, the Indians themselves had long since been oc- cluded from the higher processes of Colonial government. (Cf. p. 347.) Although the rebellion of the 1780’s takes its name from the episode staged by the idealizing and romantic Tupac Amaru, its most sustained and substantial achievement occurred in the long siege of La Paz after Vou. 2] COLONIAL QUECHUA—KUBLER 387 his execution. (Fig. 40.) Thus, in 1781, three distinct and separate insurrections were in progress, connected only by the most tenuous relations: the Chayanta rebellion, the rebellion of Tupac Amaru, and the siege of La Paz under the command of Julidn Apasa. That none of them succeeded may be attributed in large part to the fact that they remained separate and unrelated. The tactical conduct of the siege of La Paz reveals remarkable va- riety and inventiveness, vitiated, however, by the lack of discipline and training among the peasant levies. Early in the campaign, the Indians were armed only with stones, and they attacked most heavily during rainy weather, in the justifiable hope that the European fire- arms would be made ineffective. At this stage, Indian morale was good; it is reported that the attackers were most solicitous to conceal and bury their dead; the few prisoners preferred suicide to captivity (Segurola, 1881, pp. 387-88). Soon the Indians managed to spread the siege by burning surrounding communities, impounding all live- stock, and by investing the roads over which reinforcements might come. Firearms were procured and aimed by snipers sheltered within the ruins of the burned houses at the city’s edges. Captive Spaniards were made to serve the newly acquired artillery, and the rebel effec- tives were augmented by Mestizo deserters from the city. By April the attackers could circulate unseen all about the city, using the burnt- out shells of houses. Attacks in force were delivered by files of foot soldiers moving behind the cover of horses and pack animals (Segu- rola, 1881, pp. 390-96). Their armament then included cannon, mortars, and muskets. The cannon were used to fire hand grenades; explosive sling missiles, rockets, and incendiary arrows wrought much damage. The artillery, however, did little harm, for the fire of the four mortars was laid by Mariano Murillo, a captive and naturally uncooperative loyalist, whose arms Julian Apasa later had struck off (Segurola, 1881, pp. 405, 412) in punishment for attempting to com- municate with the city. Meanwhile, the Indian host had encamped upon the plateau above La Paz, in what has been described as “‘another city,’’ with a church, many dwellings, a prison, and other buildings, in a manner suggesting the practice of the great Inca sieges (Bram, 1941, p. 85; Segurola, 1881, p. 421). The camp had 24 cabildos, each with its gallows and whipping post. A great tent was called the Palacio; in it lived Julian Apasa with his young chola wife, a government of four ‘‘oidores,” various ambassadors, two captive priests, and Bonifacio Chuqui- mamani, the cholo secretary (Borda, 1881, p. 440). The ‘‘oidores” administered the sale of coca and the estates of the deceased; they were also the treasurers and the supply officers for the army (Borda, 1881, p 441). The ritual life of the camp was punctuated by frequent performances of great wheel dances (bailes de rueda). It is striking 388 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 to note that, as in the revolt of Manco Capac and in the siege of Cuzco in 1536-37, the dances were performed every 20 days. During them, military activities were suspended. The collapse of the first siege of La Paz late in June 1781 happened 1A pveso Congu CO gran do yo fa ~ SOR [f I j ink a4 co a a Fiaure 40.—The capture of Tupac Amaru I. (After Guaman Poma, 1936, p. 449 Vou. 2] COLONIAL QUECHUA—KUBLER 389 less because of local conditions at the site of the siege than because of Spanish depredations among the supporting provinces. The crushing of the revolt of Tupac Amaru did not become generally known about La Paz until September (Segurola, 1881, p. 481), but the consequent demoralization of the rebel provinces bore results far earlier. The second unsuccessful siege of La Paz began on August 4, under the direction of Andrés Tupac Amaru, who styled himself the son and heir of José Gabriel Tupac Amaru. Precisely who he was and where he came from have not been determined. The new leader, in order to legitimize his position, invented and circulated a strange account of the rebellion. According to Andrés, in a forged letter purporting to be from José Gabriel and dated July 1 at Tinta, he had been be- queathed all his “‘father’s’”’ powers. These powers, it was claimed, derived immediately and legitimately from Charles III. Andrés had little to do with the actual conduct of the second siege, but its military leaders acknowledged his authority. As in the first siege, the Indians attempted the ruse of impersonating Spanish soldiers. When this failed, an effort was made to flood the city. By October 12, the headwaters of the river had been dammed; a great head of water col- lected, and when the dam was suddenly removed, a torrent rolled down upon the Spaniards. The device was common in mining enter- prises; at La Paz, however, it was unsuccessful (Segurola, 1881, pp. 485, 494). Another Indian stratagem was for venders to offer food to the starved Spaniards at the edge of the city, and to take captive those who came forth (Segurola, 1881, p. 488). Final episode.—Over the last episodes of the great rebellion, the spirit of José Gabriel seemed constantly to hover. His relatives, real and fictitious, dominated the scene. Finally, in May 1783, the flames of the revolt flickered once again in Huarochiri Province. An individual styling himself Felipe Velasco Tupac Inca Yupanqui, in- voked his “‘cousin,’”’ José Gabriel, in calling the Indians of the cor- regimiento of Parinacochas to his banner. They were assured that José Gabriel was still alive, seated upon the imperial throne in the realm of the Gran Paititi. During the month of June Spaniards were imprisoned, the roads and bridges leading into the province from Lima were cut, and a general Peruvian rebellion was planned for August 29. The conspirators, however, were apprehended and executed before its eruption (Angelis, 1836-37, p.183). For present purposes, the most significant aspect of the Indian re- bellion is its striking lack of formal indigenous cultural content. Had the rebellion been successful, and had it resulted in the creation of an independent Peruvian Indian state, that commonwealth would have assumed and continued the institutional culture of the Colonial era. Such a prospect would have been inconceivable or repugnant to the Neo-Jnca in Vilcabamba; conversely, the cultural autonomy of the 390 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 148 Quechua in Vilcabamba in the 16th century would have been unsatis- factory to the Christianized and Hispanicized Indians of the late 18th century. At all points the late Colonial Indian philosophy of re- bellion found its limits within the horizons of Spanish institutional culture. In essence, the Indians wished to capture Spanish insti- tutions, not destroy or displace them by others. A quantitative measure for the processes of Colonial acculturation is suggested by the similarities and differences between the rebellions of Manco Inca and José Gabrial Tupac Amaru. To both leaders a relatively stable Andean community pattern was available for manipulation. For Tupac Amaru, however, formal government DOMINIOS DEL PORTUGAL Figure 41.—Map of Pert, showing Intendencies in the 18th century. (After Wilgus, 1932. Reproduced with permission from Breve Historia de América, by Carlos Pereyra, 1930.) Vou. 2] COLONIAL QUECHUA—KUBLER 391 and established religion were to remain Hispanic and Catholic. Hence, most changes within the basic Andean community pattern must be treated as functions of the spread of formal Hispanic cul- ture; the constant and unchanging traits, accordingly, are those which never came into direct conflict with the Church or with Colo- nial government. PERU eli ‘ 2 64.0% Moses * ° ° “Fee, : a re) t ° J; es eeo, °° *ee, 8 wR e 5.0 A8an BA°% Seen ° - ¢ at pS PROVINCIA ®, coofe, Van°e, iar o : *ee, fe 2 OE Shins © wo were, aes : *, Re EMQuitos? F 3 “y ao veases, [log Silas coo @, CS Sires ? é @ e o & o&s is mad e 5 42° ¥ - 4 2 aes ©8380 ..° fog saura SES: » 2 qUCUMAN - tg? rocoto Fe, : Saecey. ; 4 ° "Fee. *e, 14: 6 : \~: Ps ROB, ° & is le ae se. waa Tt ae % mays 9 ve) oe & he 3 J: o ‘x f eo} % . 3 0} “a Miers 8: CS Figure 42.—Map of La Plata, showing Intendencies in the 18th century. (After Wilgus, 1932, Reproduced with permission from Breve Historia de América, by Carlos Pereyra, 1930.) 392 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 148 ESTHETIC AND RECREATIONAL ACTIVITIES Music.—Harcourt’s incontrovertible observation that Andean musical folklore is the richest and most original in America (Harcourt, 1925, p. 207) postulates an intricate Colonial development from indigenous sources. Its character derives from the blending of 16th-century European polyphony with a presumptive native penta- tonic scale (Harcourt, 1925, pp. 1382-53). Stringed instruments, unknown in antiquity, were introduced in the 16th century; the harp, with a diatonic scale through five octaves, and the charango, a mandolin, are used by the Quechua (Harcourt, 1925, pp. 79, 85). Among percussion instruments, Indian use of the small drum called tinya as a tambourine is perhaps the result of Catalan influence (Harcourt, 1925, p. 18). The rich repertory of Quechua songs contains religious chants, funeral lamentations, love songs, vocal accompaniments for choreo- graphic and instrumental performance, farewell songs, and pastoral songs. Noteworthy is the absence of cradle and nursing songs (Har- court, 1925, p. 168). Dances.—An intimate union between Christian liturgy and Quechua dancing may be noted in a drawing by Guaman Poma, executed early in the 17th century, showing masked Indian boys performing a, duelling dance before the church altar (Poma de Ayala, 1936, p. 783). About 1650, Church festivals were the occasion for native dances (Cobo, 1890-95, vol. 4). According to Frézier (1716, p. 249), the death of Atahuallpa was enacted in dancing on the day of the Nativity of the Virgin (September 8) by Indians in bird costumes, bearing figures of the Sun and Moon. In 1905 an “Inca” dance was still performed in Tarma, with Indians impersonating Huascar and Juan Pizarro, as well as princesses and a sorcerer (Harcourt, 1925, pp. 105-07). An important Colonial development was the perform- ance of dances choreographically related to industrial activities; Harcourt reports a wool-cutter’s dance at Ayacucho in which the workmen brandished the shears of their trade (Harcourt, 1925, p. 120). Many other dances of this class have been noted in the Are- quipa region (Mejia Xesspe, 1923). Leisure.—The inability and the reluctance of the Indians to perform the required amounts of work incited the colonists to puni- tive measures, which further increased Indian unwillingness, bringing in turn even sterner punishments. The Spaniards do not seem to have comprehended that, for the Indian, no work was worth doing which was not infused by ceremonial symbolism. Under the Inca, the tributary gave labor for the fields of the Sun and to the public works of the Emperor, and in working for his own sustenance, he worked for his community (Polo de Ondegardo, 1873, p. 161). All Vou, 2] COLONIAL QUECHUA—KUBLER 393 work was punctuated by ritual and festival occasions; work itself was ceremonially performed. In Christian life, work and worship were separate concepts. The day of rest evoked no response from the Indian whose understanding of leisure was in terms of ceremonial exercises. Under Christian planche.XXXI . | PLIC24TZ. | dapres Ficure 43.—An early 18th-century drawing of the Incas. A, The Emperor; 8B, the Coya or Queen; C, Indian; D, Indian carrying a cape or mantilla; E, houses; F, G, plan and cross section of furnace for cooking herbs; H, vessels found in ancient tombs. (After Frézier, 1716, opp. p. 247.) 394. SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 direction, the tributary was expected to do unadorned work for 6 days, divorced from all forms of ritual behavior. His daily devo- tions were a separate category, and labor, from being a form of piety, was degraded into physical toil, without spiritual compensation. For the great reformer, Viceroy Toledo, this problem of ‘‘ psycholog- ical unemployment”’ never presented itself. To the administrator, the real problem was quite different, and it had a double aspect: to educate the Indians to systematic habits of work, and to regulate the extravagant demands made upon their labor by private parties. Alcoholism.—One result of these solutions to the problem of leisure was vastly to increase the Indian consumption of alcohol. Drunken- ness, induced by consuming large quantities of chicha or sora, was more common than in antiquity, when it had been restricted to cere- monial occasions (Santillan, 1879, pp. 77-79). In the 19th century, Indian alcoholism presented a serious social problem, since govern- ment revenue had in large part to be raised from taxes on alcohol. As Belatinde points out (1932, p. 98 ff.), one-seventh of all government income was derived between 1869 and 1903 from this source. But Belatinde, in moralizing upon Indian alcoholic degradation, takes no account of the increase in population, or of the gradual exploitation of alcoholic consumption by tax collectors. Coca.—In his daily life the Quechua had access to a luxury which had been denied him under the Jnca rule: coca. In antiquity, the use of coca was subject to strict regulations and was commonly enjoyed only by the ruling caste. In the Colonial era its cultivation was commercially encouraged, and the narcotic was used by Indians throughout Venezuela, New Granada, Quito, and Perdé. It was grown chiefly in the Andean area between Huamanga and La Plata. The mastication of the leaves allayed thirst and hunger; within certain limits the herb supplied dietary deficiencies; its cultivation was arduous only where the first breaking of the ground was con- cerned; the industry was a business for some 2,000 Spaniards in the 16th century; and the leaves conveniently were circulated as currency among the Indians themselves. Matienzo declared that without coca there could be no PerG. When abused, however, the use of coca induced pellagralike infections of the mucous membranes. It was also blamed for making women barren, and for increasing the in- dustrial death rate. In 1555-61 Viceroy Cafiete attempted to sup- press its cultivation because of the association of coca with idolatry,* and because the expanding industry was drawing labor away from the mines. These legislative efforts soon had to be abandoned, because of the economic role played by coca, and Indian resistance to its return 4 The herb was frequently used medicinally by Europeans against colds and dental decay; in the Bishopric of Quito, however, the association of coca with idolatry led to its proscription. Europeans using it were liable to excommunication (Pefia, 1698, pp. 570-71). ‘ Vou. 2] COLONIAL QUECHUA—KUBLER 395 to the status of a restricted commodity, as under Inca rule. (Mati- enzo, 1910, chs. 44-45; Santill4n, 1879, pp. 115-16; Anonymous, 1889, pp. 214-16; Falcén, 1867, p. 484; Quiroga, 1922, p. 100 ff.) RELIGION The 16th-century conversions.—The Jesuit Blas Valera, writing about 1591, left a succinct account of the nature of 16th-century efforts to Christianize the Quechua populations of Perfi (1879, pp. 207-15). In the first months after the Conquest the Indians were forcefully herded into the Christian family, being given baptism without cate- chism or doctrinal instruction of any kind. Such mass baptisms were practiced in northern Perf, at Cajamarca, and among the Coastal tribes as far south as Pachacamac. Not only were these early con- versions impermanent, but the bad example of the first generation of colonists brought about the social and moral degradation of Indians living in proximity to the Europeans. On the encomiendas, the grantee was obliged to maintain a resident clergy for religious instruction of the repartimiento Indians. Since the living of the curate or friar in question was collected in part from the encomendero himself, the religious instruction often served the encomendero’s wishes, favoring his techniques of exploitation (Toledo, 1867, p. 4). In certain cases, the encomendero even appointed the curate to the position of major-domo, putting him in the position of extortioner (Anonymous, 1889, p. 201). The priest enforced his exactions by appointing officers of justice, dispensing judgment in civil matters, and maintaining prisons (Lorente, 1867-72, 1:3-4). For such services the resident curate was paid not only a substantial cash salary, amounting to as much as 300 or 400 pesos annually, but his extensive food needs were provided by the Indians of the reparti- miento (Vega, 1896, pp. 109-10). In general, then, the Indians were reluctant to enter Christian life during the 16th century. Their behavior with regard to conversion impressed the Spaniards of the generation of the 1560’s as obstinate, secretive, and hypocritical. The Indians retorted that Christian doctrine was imposed by force rather than by love, that the Spaniards had never seriously attempted to remove linguistic obstacles, and that evil Christian works had corrupted the Christian faith in Perf. In the eyes of the Indians, as we may gather from the remarkable “‘Colo- quios de la Verdad,”’ by Pedro de Quiroga, written about 1563, the Christian ministers were corrupt and greedy. The doctrine itself had often been degraded through being taught by Negroes and yanaconas. From the Indian point of view, the missionaries had failed in the following objectives: (a) To convince the Indian of the existence of the true God; (6) to demonstrate that the universe was created by the one and only God; (c) to explain the mystery of the 396 SOUTH AMERICAN INDIANS [B. A. E. Bun. 143 Redemption; and (d) to induce the Indians by good example, and thus to lead them to understand and participate in the Sacraments (Quiroga, 1922, p. 115 ff.). Sixty years later, Father Arriaga, writing about 1620, still faced the same difficulties as had his 16th-century predecessors. The catechism which he proposed as essential for the instruction of the Indians yields an index to the persistence and vitality of Indian religion at the time he was writing. For Arriaga, the catechism should insist upon the following cardinal points: Only one God exists; God created the universe; the Devil is a fallen angel taking vengeance on God through man and the idols of Indian worship. Man originated in the Cre- ation of Adam and Eve, contrary to the Indian doctrine of multiple human origins. Christ instituted the Sacraments for redemption from original sin. Among the Sacraments, that of confession must be clearly defined, to distinguish it from Indian confessional rites. The doctrine of the intercession of the saints needed special clarification, to avoid the great confusion between the saints and Indian cult-objects (Arriaga, 1920, pp. 127-28). Finally, a radical conflict was present between the Indian and the Christian concepts of afterlife. The traditional Indian view regarded the poor, the old, and the infirm as despised by the supernaturals, in contrast to the ethical affirmations of humility and the love of the poor contained in Christianity (Polo de Ondegardo, 1916 a, p. 8; Arriaga, 1920, p. 53). In brief, Quechua pre-Conquest religion subsisted as an integral system of beliefs about divine essence all through the 16th century. As such, it was dominant over Christianity, which was not accepted by the Quechua in its essential beliefs until an advanced date in the 17th century (cf. p. 400). Survivals of Quechua religion: Objects of worship.—The theogony did not survive intact, but underwent certain processes of deformation and reemphasis. As the Inca state was decapitated by the Conquest, so also was its religious symbolism decapitated. Garcilaso presented Pachacamac and the Sun as the supreme binary deities of Inca religion, the one invisible, and the other visible (Garcilaso, 1722, vol. 1, bk. 2, ch. 2). Pacha- camac, it is said, had been incorporated with Jnca religion as a politi- cal stratagem, to secure the coordination of the Coastal area by recog- nition of its supreme deity (Calancha, 1638, pp. 365-66). By 1566, however, Pachacamac was no longer taken into account by the official ecclesiastical inquiries into idolatry (Calancha, ibid.). In general, however, the adoration of celestial bodies was continued by the Quechua tribesmen. The rites of the sun, called Punchao or Inti (Valera, 1879, pp. 138-40), took precedence over those of other major planets. However zealously the Christians destroyed the idols and the instruments of idolatry, there remained for Indian worship Vou. 2] COLONIAL QUECHUA—KUBLER 397 the mountains and hills, the streams, springs, and lakes, the ocean and many other natural forms whose adoration never ceased (Polo de Ondegardo, 1916 a, p. 43). Devotion to the pacarinas sometimes determined the place of res- idence of the tribe, and often proved a serious obstacle to the efforts of the Colonial government to move the Indians to more productive sites, The theory of multiple human origins, as represented by the belief in the pacarinas, with its complicating ancillary proposition that the progenitor of any species may unaccountably occur in another realm of nature, was the most objectionable heresy the Christians felt it necessary to combat (Arriaga, 1920, pp. 69-127). Special Colonial forms of static huaca worship were evolved at the mines. The mountains and the mines themselves were implored, as huacas, for a plentiful yield, with nocturnal rites of dancing and drinking. The gold mines were called Coya; the silver mines and their metals, Mama. Mercury and its ores were adored as Ichma or Linpi, while ores in general were addressed as Corpa (Calancha, 1638, vol. 2, bk. 10, pp. 371-72). The mining city of Potosi was adored as a sacred object (Loaysa, 1889, p. 593; Calancha, 1638, pp. 371-72). The spiritual force of the huaca was both contagious and infinitely divisible, so that even a burned fragment, or the place where a huaca had been destroyed, was venerable, or huaca itself. The Indians of Huaylas worshiped in the 17th century at the Rimac bridge in Lima, because their huacas had been jettisoned there in the 1570’s (Arriaga, 1920, p. 22). The mallquis, or mummies of the dead members of the ayllus, were also huacas, and the bodies of the dead lords had their own priests and rites. Throughout Pert, the Indians were in the habit of re- moving the bodies of the deceased from the churchyards to transfer them to the Machay or mountain caves and niches in which such huacas were kept in the Colonial era. When apprehended in the act, the Indians gave as their reason Cuyaspa, i. e., love for the dead. The custom was prevalent as late as the second quarter of the 17th century (Arriaga, 1920, pp. 14, 25, 61; Calancha, 1638, p. 377). A social difference was present between the cult of the mallquis and that of the other portable huacas. In any community there were two social strata based upon priority of residence and designated as the Huari (old settlers) and the Llacuaz (immigrants within the mem- ory of recent generations). The number of artifact huacas was great- est among the old settlers, and mallqui worship was more common among the Llacuaz (Arriaga, 1920, p. 1388; Calancha, 1638, p. 372). Artifacts of European manufacture frequently attained status as conopas, such as fragments of glass, silk, or sealing wax. In general, they were regarded as luck pieces or talismans. 398 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 Sacrifices.—The sacrifices regarded as appropriate offerings to the deities and cult objects were characterized by their status as materials of value upon which some labor had been expended. It was essential that animal sacrifices be made of domestic animals on which time and labor had been spent, rather than wild animals. Yet the sacrifices of animals were no longer publicly but secretly performed by the Quechua. The rites no longer involved valuable livestock, such as llamas, but only guinea pigs, as offerings, and also for divinatory and therapeutic purposes. Practitioners.—In Quechua religion, a sacerdotal caste must be dis- tinguished from the immense number of common “‘sorcerers.”’ The proliferation of the sorcerers practicing an infrasocial or antisocial magic was a Colonial phenomenon. Christian doctrine contains a distinction between idolatry and sorcery. ‘The former consists either of superfluous worship or false worship; the latter always involves a pact with the demon, regardless of formal cult (Pefia, 1698, p. 236). Hence, the priests of Quechua religion were always true idolaters, unlike the sorcerers, who might or might not be idolaters. Both in antiquity and after the Conquest, the sorcerers were men and women of low caste. It was exceptional that nobles or wealthy men engaged in the profession. In the Colonial era, the animal sacri- fices which were provided by the client to furnish the sustenance of the spellbinder were commuted to offerings of money, clothing, or food. The number of sorcerers had increased enormously, and Polo de Ondegardo assigns the increase to the general spread of indigence in the Colonial communities. Their services were therapeutic, divi- natory, and magical. Their rites were often practiced with Christian formulas, especially in the treatment of disease. The customary rites of securing confes- sion of sins, and of sucking or extracting the disease from the body of the patient, were alternated with kneeling Christian prayer, sprink- lings of holy water, invocations of God and Jesus, and signs of the Cross. The sorcerers also attempted to explain and adapt Chris- tian doctrine, by pronouncing that God’s goodness was finite, that the Christian remission of sins was not accorded to great sinners, that natural events were actually governed by the huacas, that the Chris- tian saints were huacas, and that Jesus and the Devil were related as brothers. Hence, the sorcerers played a role in the formation of Quechua Catholicism, while bringing support to Quechua idolatry. In essence, however, their crafts were contrary to religion whether Quechua or Christian, in that their magic was illicit, infrasocial, and proliferant without relation to doctrine. (Arriaga, 1920, p. 41; Polo de Ondegardo, 1916 a, pp. 26-30; Calancha, 1638, pp. 377-79.) Among the White population of Pert, the Indian sorcerers not infre- Vou. 2] COLONIAL QUECHUA—KUBLER 399 quently enjoyed some prestige, chiefly as healers and diviners; for Europeans to have recourse to Indian sorcery was regarded by the Church as only a venial sin, of no great consequence (Pefia, 1698, p. 242). Ceremonials.—The chief calendrical occasions for huaca festivals were three: (1) at the time of the Corpus Christi festival (Oncoy- mitta and Intiraymi), when the huacas were adored to prevent the destruction of the crops by drought or frost; (2) about Christmas- time, at the beginning of the rainy season, dedicated to lightning and thunder as rain-bringers; (3) the harvest festivals, called Aymoray or Ayrihuaimita (Arriaga, 1920, p. 52). A variant of the harvest festival was described in detail by Cristébal de Molina of Santiago, as occurring in April 1535. All the mummies of the Jnca Emperors, with their attendants, were exposed before dawn along an avenue laid out in a plain near Cuzco. Each mummy was equipped with furnishings relating to its importance. The mem- bers of the Inca royal caste alone were present, garbed in their richest clothes. As the sun appeared, their chant slowly grew in volume until noon, then diminished until sunset. Manco Inca, richly installed in a litter apart from the others, gave the cues for the performance (Molina of Cuzco, 1916, pp. 160-62). In certain areas, the sun feast, called Intiraymi, fell on the solstice in June, coinciding with Corpus Christi, and was celebrated simul- taneously with the Christian ritual, under its protection (Polo de Ondegardo, 1916 a, pp. 21-22). June was also the month during which the festival dedicated to the Pleiades was celebrated. In the district of Lima these rites were celebrated publicly in 1617. The moment chosen was at the time of the greatest visibility of the con- stellation, when the crops were in danger of frost. The rites were intended to influence the activity of the constellation with regard to weather (Rel. geogr. Indias, 1881-97, 1:205-06; Arriaga, 1920, p. 76). In September, after the sowing, the feast called Citua or Situay was secretly celebrated, with many local variants. It was mainly a rite of lavatory purification, attended by blood sacrifices in token of loyalty to the Sun and the Inca Emperor (Polo de Ondegardo, 1916 a, p. 23; Molina of Cuzco, 1916, pp. 35-41). As late as December 1571, during the growing season, the old festival of Raymi was still observed in Cuzco itself. During it, puberty rites were performed. In antiquity, the festival had been intimately associated with the idol called Huana- cauri. In the 1550’s, the /nca Paullu housed this idol by his house in Cuzco, and the Raymi festival was celebrated there until the destruc- tion of the figure by the Spaniards. Elsewhere than in Cuzco, Raymi was celebrated in different ways and at different times of year. For example, in Potosi, Raymi was held during the sowing period and 595682—46——_28 400 SOUTH AMERICAN INDIANS [B. A. B. Bunn. 143 again at Corpus Christi, at which times the Indians feasted and drank in costume, aod performed various ancient sacrifices (Polo de Onde- gardo, 1916 a, p. 19 ff.; Cobo, 1890-95, 4:36). The Christianization of the Quechua: Extirpation of idolatry — The failure of the men of the 16th century to achieve the Christianization of the Quechua may be traced chiefly to the sparse and undermanned organization of the Church in Per4d, to the profane disturbances of the Civil Wars, and to tbe powers of resistance or survival implicit in native Quechua religion. Early in the 17th century, however, the Church threw its forces into an ambitious, systematic, and successful campaign to displace idolatry. By 1660, more or less, the Catholicization of the Quechua had been achieved. On the one hand, genuine conversions were sought. The priests, friars, and missionaries made it their aim to secure the good will of the secular leaders of Indian society. An intensive program of education was conducted among the children of the various communi- ties. The civil and economic status of the Indians was improved by teaching them new crafts. Among ordinary Indians, the priests attempted a gradual eradication of the most salient vices. Sodomy and bestiality, for instance, drew reproof or punishment more imme- diately than the simple huaca worship practiced by most Indians. Huaca worship, in turn, was treated more severely than drunkenness and trial marriage (Pefia, 1698, pp. 1388-438, 149-51). On the other hand, very stern repressive measures were taken against the professional practitioners of idolatry. The native priests and sorcerers were arrested and isolated from their communities. The corregidores were entrusted with the task of seeking out the eremitic professionals (Anonymous, 1923). All cult sites, idols, and instru- ments were to be destroyed. ‘The ritual dances and songs were not to be performed, and the priests undertook wherever possible to de- stroy the drums, feathered garments, animal masks, and panpipes associated with these rituals. In the churches continuous predication against idolatry was employed, and local informers were encouraged with the promise of absolution for their past participation (Pefia, 1698, pp. 246-47). Thus, under Viceroy Montesclaros (1607-15), 600 idols from Huaro- chiri alone were burned in the public square in Lima, and an Indian idolater named Hernando Paucar was publicly flogged in the presence of the Viceroy. Thereafter, the campaign to extirpate idolatry assumed extravagant proportions. In 1617-18, systematic idol-and- witch-huaots were conducted. Nearly 6,000 persons confessed to idolatry in the corregimiento of Chancay alone. In that same district, 679 “‘sorcerers’”’ were discovered. The priests confiscated 603 chief huacas, 3,418 conopas, 617 mummies (mallquis), etc., in some 31 settle- ments in the Archbishopric of Lima alone (Arriaga, 1920, pp. 13-16). Vou, 2] COLONIAL QUECHUA—KUBLER 401 The great campaign of extirpation was accompanied by asystematic effort to extend the peripheral conversions. Yet, in 1626, the feeling existed among the clergy that little or nothing had been achieved. An archepiscopal report of that year expresses horror at the discovery of numberless communities in which no Indian did not engage in idolatry, and the fault was blamed upon the small numbers of the clergy in Pert. It was even recommended that the enforcement of Christianity among the Quechua be confided to the Inquisition (Tovar, 1873, p. 327 ff., 338-39). As late as 1646, archepiscopal inspectors were appointed to travel throughout the provinces in search of idolaters (Tovar, 1873, p. 413). It is somewhat surprising, therefore, to read in 1667 that genuine idolatry no longer existed in Perfi (Pefia, 1698, pp. 231-40). The campaign of extirpation had ceased; the Indians were regarded as practicing Christians, and the Church expressed satisfaction with the spiritual condition of the Indian communities. Of course, a double process is present: true idolatry was on the wane, but also the clergy became more tolerant of ‘‘superstitions’’ which formerly had been regarded as pure idolatry. From the churchmen’s point of view, an old scholastic distinction between idolatry and superstition made the change of attitude pos- sible. The distinction turned upon the question of whether the subject apprehended divinity in the objects of his worship. If, in the view of the clergy, the Indians practiced huaca worship without attributing any divine essence to the huacas, they were the victims only of superstition, and defined as those who “seek good or flee evil by useless and disproportionate methods” (Pefia, 1698, p. 240). Thus, it became possible for many baptized Indians, while sincerely believing in Christianity, still to revere their huacas, certain stones and moun- tains, without becoming idolaters. The churchmen understood that the object of this veneration was merely to propitiate the irrational forces of nature, and that as such it could be tolerated by curates as being nonheretical, i.e., without error or infidelity, although in effect it contaminated the worship of the True God and was to be dis- couraged. Gradually, the conditions under which the ritual was performed came to be taken into account by the clergy. For instance, when Indians participated in the great huaca festivals, it was granted that the participation might be undertaken without interior faith. If the Indians were moved by the compulsions of Indian society, the act was external and not to be regarded as true or perfect idolatry. In any case, it was urged after 1667 that mildness be shown in the pun- ishment of actual cases of idolatry because of the scanty indoctri- nation and the limited understanding of the Indians (Pefia, 1698, pp. 232-33). When an Indian threw stones upon an apacheta, the 402 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 act would constitute heretical idolatry only if performed in the belief that divinity resided in the heap of stones. If, on the other hand, the offering were made to no assumed divinity, the act was regarded as nopheretical idolatry. Finally, if the offering were made in the mere belief that fatigue was thereby lessened, the act was regarded as superstitious and as only a venial sin (Pefia, 1698, pp. 239-40). After 1650, many Indian rites which had earlier been the objects of the campaign to extirpate idolatry suddenly entered the class of rela- tively harmless superstitions. Among them were such customs as deforming the heads of infants, various puberty rites (as in the wara- chikoy festival, when the hair of the boys was cut for the first time, and in the rites of ablution performed when girls first menstruated), the chuquiragua ceremonies, in which engaged couples were driven through a bonfire of green branches, and certain rites of therapeutic magic. Among the last named were such customs as placing the gar- ments of the sick person out of doors, with portions of herbs or corn, so that they might be stolen by casual passers. ‘Therapeutic baths of ritual character, as well as the custom of spitting into ichu bushes were also tolerated by the clergy. The household ritual of placating the fires on the hearth with offerings of corn or chicha, public lamen- tations during eclipses as well as the beating of the dogs, the placing of guardian stones in the fields, and the many rites connected with house building also came to be tolerated (Pefia, 1698, pp. 239-40). In these instances, the clergy attempted to determine whether the work of the Demon was present by the following standard of measure: Ii the effect of the action was measurably greater than its cause, diabolic intervention was assumed to be present; if, however, the effect was less than the cause, the observance was regarded as unaf- fected by demonic influences and therefore to be classified among the harmless instances of vain observance. Clear instances of this latter were the innumerable beliefs in bad omens, such as snakes crossing one’s path, the augury of a sterile year from the birth of twins, the reading of evil omens in the cries of birds and animals, prognosti- cations from the twitchings of muscles, and the various forms of divi- nation in which coca was used (Pefia, 1698, p. 241 ff.). An example of the latter was the custom of spitting into the hand with fingers point- ing downward; the course of the coca spittle was taken to predict the outcome of the event. It must not too readily be taken for granted that all the concessions came from the clergy. The 17th-century distinctions between heret- ical and nonheretical idolatry, between superstition and vain observ- ance, were far from being purely doctrinal solutions to the problem of Christianization. On the contrary, the religious education imposed upon the parishes insisted vigorously upon the fundamental articles of faith. A community continued in its course of heretical idolatry Vou. 2] COLONIAL QUECHUA—KUBLER 403 unless it displayed knowledge of the canonical books of Holy Scripture as revealed by the Holy Ghost; knowledge of the traditions of the Church as codified by the Council of Trent; knowledge of the articles of faith contained in the Credo; knowledge of the Pater Noster and the Commandments. Among the Indians, these teachings were fre- quently confused and perverted. or instance, the Indians commonly held the Trinity as three divisible and separate persons. The Father was not infrequently regarded as older than the Son. A natural con- fusion arose when the Indians, to the horror of the clergy, identified the material fabric of the Church with its mystic substance, or when God was identified with some temporal monarch, and confused with the King of Spain. Yet the clergy regarded these errors as anything but heretical, because of the ignorance of the spirits in which they arose (Pefia, 1698, pp. 268-85) and because of the imperfections of instruction. Catholicism became a functioning part of Quechua life when the Indians accepted Christianity’s system of divine essence. The sur- vivals of Quechua ritual and the doctrinal errors or confusions men- tioned just above may not be taken as proofs that the Indians rejected Catholicism. On the contrary, the cardinal question is always that of divine essence. As long as the huacas partook of divinity, their worshipers were not Christians. In the moment, however, when divinity forsook the components of Quechua religion, that religion ceased to exist, however vigorously the veneration of the huacas and other cult objects may have continued. Ever since the mid-17th century, Quechua religion has been Catholicism, although its ritual texture still preserves innumerable intact constellations of pre- Conquest observances. The Catholic clergy—The Christianization of the Quechua was, in many respects, the most profound change wrought in Quechua ‘lif during the Colonial era. At no point can it have been said to have occurred without the direct interference of European clergy, whether regular or secular. In 1569, the total number of priests and friars resident in the Vice- royalty of Peri was about 350, distributed among 477 repartimientos (Anonymous, 1889, p. 172). The estimated need at that moment was for some 1,500 curates, and in Mexico at this time a smaller total population was served by many more spiritual ministers. In 1583, accordingly, it was resolved in the Ecclesiastical Council held in Lima that each curate should serve no more than 200 to 300 families (Tovar, 1873, pp. 445-46; Levillier, R., 1919 a, pp. 273-415). This optimum ratio was never uniformly achieved, but by 1638, the total number of active curates had been increased ‘to a point at which the average parish contained about 358 families (V4zquez de Espinosa, 1942, pp. 22-58). 404 SOUTH AMERICAN INDIANS [B. A. BE. Bun, 148 Hence, if Christianization requires an ample working staff of curates and missionaries, the epoch of the most rapid Christianization may be assigned to the period in which the number of European workers was most rapidly expanded. The maintenance of the curates in the parishes was, of course, provided by the parishioners. The stipends were fixed by a synodal constitution, whence they took the name of sinodos. Payment to the curate was made out of the tribute collected by the corregidor. The sinodos, however, were inadequate for the needs of the ministers, who increased their income by various methods (Montesclaros, 1859, pp. 13-14). For example, the priest collected funds from the various lay organizations (cofradias) in the parish, and he benefited from the many gifts that were customary. Called camaricos, these consisted mainly of foods. The curates, whether regular or secular, were always men of Euro- pean caste. The intention of the Church itself had long been to en- courage the creation of a clergy recruited at least in part, from the native populations. For instance, a papal brief from Clement XIII in 1766 confirmed and repeated earlier resolves, taken in 1697 and 1725, that Indians should be admitted to the religious orders, educated in the colegios, and promoted according to their capacities to suitable positions and dignities (Garcia y Sanz, 1876, p. 225). The frustration of these good intentions on the side of the hierarchy was achieved by the civil government through the Spanish institution known as royal patronage. Not long after the Contact, the Papacy had con- ceded the Spanish Crown the rights of appointment to ecclesiastical positions in America. (See Mecham, 1936.) In Pera, accordingly, the practice was for the Bishop to form panels of three names for each benefice or curacy; from such panels the Viceroy or the President of the Audiencia, as vice-patrons representing the Crown, selected the effective appointment (Juan and Ulloa, 1826, pp. 335-37). Hence, as the institution of corregimiento had deprived the Indians of par- ticipation in Colonial government (see pp. 345-347), so did royal patronage deprive them of participation in religious government. During the great Indian rebellion of the 1780’s, when the Indian rebels sought to capture Spanish institutions, it was characteristic that ecclesiastical offices also were appropriated by Indians. For example, in 1782, an Indian from Paucartambo, named Nicolas Villca, conse- crated himself Bishop of Cuzco. He was tonsured, and accepted the hand-kisses and genuflexions of his followers, while dispensing pastoral benedictions (Angelis, 1836-37, p. 170). The lay associations.—The prestige system of the modern Quechua community (see Mishkin, this volume, pp. 443,459), designated as the institution of cargos, has its origin in the intricately organized lay associations of the Catholic parish, called cofradias in the Spanish- Vou, 2] COLONIAL QUECHUA—KUBLER 405 speaking world, or sodalities in English. In canon law, the cofradfa, or sodality, is regarded as an association of the faithful with episcopal approbation, existing to promote the maintenance of the cult. The cofradias are among the most characteristic institutions of Spanish Catholicism; they flourished most amply in Spain during the 16th and 17th centuries, and served as an instrument of the Catholic Reform (Pfandl, 1924, p. 83). In Pert, the existence of Indian cofradias (fig. 37, d) is reported as early as 1575 by Father Arriaga. One dedicated to the relief of the souls in Purgatory was established in that year in the district of Hu- aylas, immediately after the destruction of the huacas by the priests. About 1610, several cofradias are reported in Cuzco (Romero, 1923 a; Arriaga, 1920, pp. 85, 99). After 1650, more or less, cofradias were established wherever parish conditions permitted them. Their membership was usually restricted to 24, including two major-domos annually elected under the curate’s supervision. The members were selected from within a single parish. Ideally, all the adult males of a parish should belong to the cofradias; the parish con- tained as many cofradias as were necessary for the maintenance of the various altars, the seasonal and annual Church ceremonies, and the charitable enterprises of the community. Certain cofradias, moreover, offered the privileges of mutual benefit associations. The membership was responsible for the funeral expenses of its members, ‘and occasionally undertook to provide for the families of deceased members. But the main function of the cofradia was always to sub- sidize the festival occasions of the cult, especially on certain saints’ days and during Holy Week. Its members might endow the associ- ation by testamentary disposition, but whether endowed or not, the association was held responsible by the curate for providing cult expenses (Juan and Ulloa, 1826, pp. 647-49; Haenke, 1901, p. 113). The wide diffusion of the cofradia system cannot be assigned to any date earlier than the general Christianization of the Quechua and must, therefore, be placed after 1650. Almost no record of the cof- radia structure within a single Colonial community has been preserved, but its integral survival and amplification in many modern Quechua communities suggest the importance of studying it in great detail. The ritual life of the Catholicized Quechua very likely unfolded almost entirely within the sodalities, with their costumes, insignia, privileges, responsibilities, and interlocking memberships. The Catholic calendar.—The Church regarded the non-observance of certain festivals by the Indians as a mortal sin. Every Sunday was, of course, a compulsory observance, as well as Good Friday and Sat- urday, Easter Sunday, and the Nativity. Other compulsory festivals were the days in the calendar dedicated to the Circumcision, the Ascension, Corpus Christi, the Nativity of the Virgin, the Purification, " 406 SOUTH AMERICAN INDIANS [B. A. E. Bun. 148 the Annunciation, the Assumption, and the days dedicated to Saint Peter and Saint Paul. In addition to these festivals, each month con- tained other lesser holidays, usually celebrated in services and pro- cessions by some part of the community: In January, the feast of the Kings; February, St. Mathias’ day; March, St. Joseph’s day; May, St. Philip and St. James, and the Invention of the Cross; June, the Nativity of St. John the Baptist; July, St. James the Greater and St. Anne’s days; August, St. Lawrence and St. Bartholomew; Sep- tember, St. Matthew and St. Michael; October, St. Simon and St. Judas; November, All Saints’ Day and St. Andrew; December, St. Thomas, St. Stephen, St. John the Evangelist, the Day of the Inno- cents, and St. Sylvester (Pefia, 1698, p. 672). It cannot be empha- sized too strongly that religious education was achieved by this intri- cate ritual calendar. Nearly one-fourth of each year, it is true, was consumed in its celebration, but it will be noted that the feasts invoke all the major events of the genealogy and Life of Christ, and that among them each and every one of the apostles is commemorated. By this means, the ritual calendar itself became a lengthy catecheti- cal exercise; the parishioners learned the contents of the New Testa- ment through direct participation in its festivals. Among the cofra- dias, arduous preparations were undertaken for the appropriate celebration of the particular day. Individual prestige was accumu- lated by the discharge of the major-domo’s responsibilities, and at many points the prestige system and the experience of Christian doctrine coincided within the demands of the ritual calendar. MYTHOLOGY AND LITERATURE Mythology.—The gradual historical transformation of Quechua mythology is sparsely documented. A detailed account of the my- thology and ritual of the Province of Huamachuco, written about 1561, deserves special mention (Augustinians, 1865, pp. 5-58). A creator named Ataguju was given a triple aspect suggesting an influence from Christian Trinitarian teachings. The Christians figured in the myth- ological narrative at a time prior to the creation of the Indians by Ataguju. The Conquest also assumed mythological proportions, and among its consequences was the multiplication of the huacas by fragmentation (cf. Arriaga, 1920, pp. 23-25). Messianic cult.—The hostile fusion of Christian symbols with Quechua mythology assumed a violent expression in a religious out- break in 1565 (Molina of Cuzco, 1916, pp. 93-103). As a phenomenon of revived huaca worship, it spread throughout the Provinces of Chuquisaca, La Paz, Cuzco, Huamanga, Lima, and Arequipa, both among the hatunrunas (encomienda dwellers) and the yanaconas (urban Indians). It was reputed to emanate from Vilcabamba, Vow. 2] COLONIAL QUECHUA—KUBLER 407 whence it was disseminated by Titu Cusi’s agents in order to alienate the pacified Indians from their Spanish masters. It was quelled only in 1572 with the suppression of the separatist government of Vilca- bamba and the execution of the young Inca ruler, Tupac Amaru I. The cult centered about a belief in the renaissance of the huacas, and it was associated with a ritual chant. It was believed that all the huacas had returned to life, some being allied with the ritual of Pachacamac, and others with the cult of Tiahuanaco. These two factions of huacas were united in giving battle to the Christian god. The battles took place in the skies, and it was predestined that, as God had defeated the huacas during the Conquest, so were the huacas now to destroy God and the Spaniards. The huacas had planted crops of worms with which to destroy the hearts of the Spaniards and their horses and livestock, as well as those Indians who had embraced Christianity. All adherents of the new huaca cult were to renounce baptism, and to cleanse themselves by fasting, and abstinence from Spanish customs and tools. The huacas were no longer to reside in objects and places as before, but were to incorporate themselves in the bodies of the faithful fol- lowers of the cult. Hysterical behavior marked the reception of the huaca by an Indian, and the spirit then was enshrined in a small oratory, where sacrifices were offered and dances performed. It has been suggested that this last aspect of the cult was influenced by the Christian theory of demoniac possession (Molina of Cuzco, 1916, p. 99, n. 265); certainly the whole phenomenon bears the closest affinities to North American messianic, or Ghost Dance, religions. Literature.—The existence of a literature in Quechua written by Indians is difficult to prove. The corpus of Quechua texts is incom- plete, but there is no evidence that its monuments were composed other than by Europeans and Mestizos. An Indian theater flourished, it is true, to such an extent that its performances were forbidden in 1781, in connection with the execution of Tupac Amaru (Rojas, 1939, p. 108), and Harcourt asserts that the play “Ollanta”’ was instrumental in fomenting the rebellion itself (Harcourt, 1925, p. 180). Its texts, however, are probably of European composition, such as the one writ- ten early in the 19th century by Pedro Zegarra, who had also trans- lated Racine’s Phédre into Quechua (Middendorf, 1890-92, 3:116). Older Quechua literary texts, the religious plays or autos sacramen- tales, were composed by European curates in the 17th century. Such was “The Prodigal Son,’ in three acts and 23 scenes, written in the mid-17th century by Juan de Espinoza-Medrano, a Creole archdeacon of the cathedral chapter in Cuzco (Middendorf, 1890-92, 4: 2). The text of ‘“Usca Paukar,’’ a play in which the central figure is a beggar of Inca descent, also may be attributed to a European in the 408 SOUTH AMERICAN INDIANS [B. A. E. Bunn, 143 19th century. Middendorf points out its romantic tendencies, and certain influences from the Faust cycle of northern European literature (Middendorf, 1890-92, 3: 94-95). In poetry, many chants, lamentations, and songs (yarahuis) pub- lished by Harcourt (1925) were recorded by him for the first time. He notes that Spanish forms (the redondilla and the copla) saturate modern Quechua verse. The insistence upon allusions to the dove is greater than in European folklore, and characterizes Quechua verse (Middendorf, 1890-92, 4: 220). Since pigeons were introduced from Europe, this diction either is Colonial or an adaptation of pre-Conquest bird-symbolism. The failure of the Indians to record their own compositions deprives their life of a literary component, which, on the other hand, is a capital constituent of Colonial culture proper. EDUCATION One consequence of the campaign against idolatry was the renewed foundation of collegiate establishments for the education of the sons of the curacas. The Colegio del Principe was founded by Arriaga in 1619 in the building which housed the Jesuit novitiate in the Indian ward of Lima, called El Cercado. At the invitation of Viceroy Esquilache, 14 sons of curacas from various provinces foregathered to receive Christian instruction. At the Viceroy’s expense, they were clothed in special garments of green and red, and obliged to wear shoes, stockings, and hats. By 1621, the Colegio housed 30 students, living in dormitories, and receiving instruction from a Jesuit in reading, writing, singing, and counting. The day was spent mainly in doc- trinal and ecclesiastical exercises, as in a monastic establishment. (Arriaga, 1920, pp. 118-19, 167; Tovar, 1873, p. 237; Colegio, 1923.) In the 18th century, the curriculum consisted chiefly of Latin and Spanish, grammar and rhetoric, mathematics and music. A similar establishment was founded in Cuzco in 1628 as a Jesuit boarding school called the Colegio del Sol or Colegio de San Francisco de Borja. Both in Cuzco and Lima the boys matriculated at the age of 10, remaining interned until 18 (MacLean, 1943). A counterpart of the Lima establishment was founded at the same time as a house of‘correction. It was likewise situated in the Indian ward of Lima, and bore the name Casa de Santa Cruz. In 1621, some 40 native priests and sorcerers were incarcerated there and earned their keep by spinning wool. (Arriaga, 1920, pp. 6-7, 168; Tovar, 1873, pp. 236-37.) On the various repartimientos, the officials representing the ayllus (primeras personas, segundas personas, mandones) were the immedi- ate objects of clerical or civilian instruction in matters of doctrine, language, and trades. The education to a trade, such as that of Vow. 2] COLONIAL QUECHUA—KUBLER 409 painter or metalworker, was felt necessary to prevent the officials of the minor government from becoming financial and moral burdens to the community. (Matienzo, 1910, ch. 6, pp. 16-17; Rel. geogr. Indias, 1881-97, 1:180; Means, 1932, p. 154.) Popular education was conducted entirely by catechetical methods in the parishes, and it was restricted to the knowledge necessary for the discharge of the simple tasks of daily devotions (MacLean, 1943, p. 57), such as incipient literacy, rudimentary arithmetic, and the musical culture necessary for liturgical purposes. (Cf. Vargas, 1940; Mongrovejo, 1673, p. 18.) THE PROBLEM OF THE QUECHUA COMMUNE In the last analysis, the Quechua commune is the repository and the unit-cell of Quechua ‘‘culture.’”? Recent estimates yielded about 3,000 communes in Pert, sustaining 1,500,000 dwellers (Poblete, 1938, pp. 54-55, 60). Discussion of the commune, therefore, occupies a central position in recent writings upon social problems; its survival means the survival of an Indian society in Pert; its extinction will bring the disappearance of any recognizable Indian cultural component of Peruvian nationality. The correct solution to the problems of the Quechua commune is today a matter of grave political interest (Valdez de la Torre, 1921). The retention of a communal regimen of property by Indian villagers is the main issue. The present paper should dem- onstrate the seemingly great survival value possessed by such com- munes. They have withstood many generations of attack and undermining, from encomienda to latifundismo, from the yanaconate to the colonos and the industrial proletariat of modern Peruvian society. The explanation may be that no mode of exploitation can survive in Perti without having the reservoir of labor in the communes to draw upon. It is difficult to imagine the disappearance of the Quechua commune otherwise than under conditions in which the institutions or groups both exploiting it and yet vitally interested in its preservation should also disappear. The age-long survival of the commune is, therefore, an attainment of the successive governments, whether Inca, Colonial, or Republican, which have exploited it. As long as it survives in its habitually depressed condition, new exploiters will be attracted to encouraging its continuation, and they will be favored by its singular tenacity and stability. BIBLIOGRAPHY Acosta, 1940; Almansa, 1899; Andagoya, 1865; Angelis, 1836-37; Anonymous, 1889, 1923, 1943; Aponte Figueroa, 1867; Arciniegas, 1939; Arriaga, 1900, 1920; Augustinians, 1865; Avila, 1939; Baudin, 1928; Becker, 1932; Belatinde, 1932, 1938; Betanzos, 1880; Benzoni, 1857; Bingham, 1912, 1922; Borda, 1881; Bowman, 1916; Bram, 1941; Bueno, 1763-78; Calancha, 1638; Calvete de Estrella, 1889; Castafieda, 1929; Catholic Ency., 1907-12; Cieza de Leén, 1864, 410 SOUTH AMERICAN INDIANS [B. A. E. Buby. 143 1877, 1880, 1913, 1918, 1923 a; Cobo, 1890-95; Colegio, 1923; Conquest of Perit, 1929; Cuneo-Vidal, 1925; Cusi Yupanqui, 1916; Dictionnaire pratique, 1925-28; Diez de la Calle, 1648; Encyclopedia Univ., 1905-30; Enriquez de Guzm4n, 1862; Estete, 1918; Falcén, 1867; Ferndndez, 1876; Ferndndez de Santill4n, 1868; Fisher, 1929; Frézier, 1716; Frisancho, 1923; Garcia y Sanz, 1876; Garcilaso, 1722; Garro, 1942; Gillin, 1936; Gonzdlez de Holguin, 1901; Gutiérrez de Santa Clara, 1904-10; Haenke, 1901; Hanke, 1935; Harcourt, 1925, n. d.; Herrera y Tordesillas, 1726-27; Hoyo, 1917; Hussey, 1936, 1942; Jij6n y Caamajio, 1919 ¢; Juan and Ulloa, 1826; Kaerger, 1901; Kubler, 1942 a, 1942 b, 1944; La Casas, 1892; La Farge, 1940; Levillier, 1919 a, 1921-26; Lexikon fiir Theologie, 1930-38; Library of Congress, 1936; Lifian y Cisneros, 1859; Lizdrraga, 1909 b; Llano y Zapata, 1904; Loaysa, 1889; Lépez de Gémara, 1749; Lépez de Velasco, 1894; Lorente, 1867-72; Mackehennie, 1909-18; MacLean y Estenés, 1943; Manso de Velasco, 1859; Maridtegui, 1928; Maridtegui Oliva, 1942; Markham, 1871, 1892; Matienzo, 1910; McBride, 1921; Means, 1919, 1928, 1932; Mecham, 1934, 1936; Medina, 1930; Mejfa Xesspe, 1923; Meléndez, 1681-82; Memorias, 1859; Mendiburu, 1874-90; Métraux, 1942 a; Middendorf, 1890-92; Molina of Cuzco, 1916; Molina of Santiago, 1916; Mongrovejo, 1673; Monsalve, 1604; Montes- claros, 1859; Montesinos, 1906; Morales, 1871; Morta, 1922-25; Ocampo, 1907; Oliva, 1857; Oviedo y Valdés, 1851-55; Pachacuti Yamqui Salcamaygua, 1879; Pefia, 1698; Pert, 1906; Pfandl, 1924; Pizarro, Hernando, 1872; Pizarro, Pedro, 1921; Poblete, 1938; Polo, J. T., 1877, 1913; Polo de Ondegardo, 1873, 1916 a, 1916 b; Poma de Ayala, 1986; Quiroga, 1922; Recopilacién, 1681; Relaciones geogrdficas de Indias, 1881-97; Ricard, 1933; Rio, 1812; Rodriguez de Figueroa, 1913; Rojas, 19389; Roman y Zamora, 1897; Romero, 1923 c, 1923 d; Rowe and Escobar, 1948; Sdenz, 1933; Salvatierra, 1899; Sdnchez, 1867; Sancho, 1917; Santillan, 1879; Sapper, 1938; Sarmiento de Gamboa, 1906; Segurola, 1881; Skinner, 1805; Spinden, 1941; Steck, 1938; Toledo, 1867; Toledo and Almansa, 1899; Torres Saldamando, 1879-80; Tovar, 1873; Trimborn, 1936; Tupac Amaru, 1941; Ulloa, 1826; Urteaga, 1929; Vaca de Castro, 1909; Valdez de la Torre, 1921; Valera, 1879; Valverde, 1865, 1879; Vargas Ugarte, 1939, 1940; Varinas, 1873, 1899; Vazquez de Espinosa, 1942; Vega, 1896; Velasco, 1941-44; Whitaker, 1935, 1938, 1941, 1942; Willcox, 1921; Xérez, 1872; Zarate, 1853, 1870; Zavala, 1935; Zimmerman, 1938; Zimmern, 1944. THE CONTEMPORARY QUECHUA! By Brernarp MiIsHKIN INTRODUCTION POPULATION AND DISTRIBUTION (SEE MAP 1, NO. 4) Of Pert’s total population of 7,023,111 (Census of 1940), 2,847,196, or 40 percent of the population, were classified as Indians, and 3,283- 360 as Mestizos. In the 1876 Census, the Indian component of the population was put at 57.6 percent of the total. These figures re- veal a noteworthy trend. Pert, like other of its neighbors in the “Tndo-American”’ bloc, is on the road to becoming a Mestizo country (Censo Nacional de 1940, Resultados Generales, Primer Informe Oficial, Lima, 1941). It should be pointed out that the results of the 1940 Census do not pretend to describe the racial composition of the population in exact terms. No elaborate biological criteria were used to set off one group from the other, but rather an effort was made to indicate the larger divisions in the population on the basis of rough common-sense judgments. It was ascertained that 13 percent of the population per- sonally filled in the blank on race; 87 percent were entered in one or the other racial categories by the census taker. Generally speaking, sub- jective factors such as personal preference, or arbitrary choice, seem to have played a large part in determining race designation. Cul- tural criteria doubtless were utilized to some extent. But the cultural criteria that might have been considered in separating the Indian from the White-Mestizo part of the population were left to the ana- lytical ability of each individual census taker. The difficulties in the way of drawing up a usable list of such criteria are clear enough, so that no objective formula for classifying the population in terms 1 This summary of Quechua ethnology is based mainly on field data collected by Dr. Bernard Mishkin in the course of study of a Quechua village in the Department of Cuzco in 1937-38, and on briefer personal sur- veys in 1938 and 1941-42. The village of Kauri is situated in the high Andes of Quispicanchis, between the valley of Urcos and the Marcapata on the eastern side of the mountains. It mayor may not be a typical Quechua community; there is little published material with which to compare it. Among the few sources which have been drawn upon for comparative data is the work, ‘‘Nuestra Comunidad Indigena,”’ by Hilde- brando Castro Pozo (1924). Three little-known papers of great value which are quoted frequently in the following pages are: ‘Los Varayocc,’’ by Pastor Ordofiez; ‘‘Sistemas de Arrendamiento de Terrenos de Cultivo en el Departamento del Cuzco y el Problema de la Distribucién,”’ by Francisco Ponce de Ledén; and “E] Ayllu de Qqueros,” by Luis Yabar Palacios. All three works may be found in the Revista Universi- taria of Cuzco. 411 412 SOUTH AMERICAN INDIANS [B. A. E. Bunn. 143 of cultural divisions was at hand. What criteria the census taker selected and to what extent they influenced his judgments cannot be guessed. An extensive list of the cultural criteria that might fit the needs of the census taker in the field would include language spoken, type of settlement, utilization of land, amount of land owned, clothing worn, nature of religious worship, etc. Of all these, language is certainly the least complicated to deal with statistically and is of primary importance. The 1940 Census offers the following pertinent data: TaBLE 1.—Language groups of Peruvians over 5 years of age} Number of | Percentage speakers | of speakers SS SS Se | SS Language Spanish ere: es Leone hee Peal AN Be ea ee a ee 2, 443, 390 46. 7 \ONGTs( a1) (sae Oe otk ee eee a en eee a ae a eee ene Yo 1, 625, 156 31.0 Spanishvandi@uechwa s/t. 5. 7225 22a aes SS ea eee ee ee 816, 966 15.8 PA IM ATs eee PE 2 aan I 2 a em rt alr 184, 743 3. 5 SpanishlandiAvyimaraleer Grose 2 eee ee SSO. TRE A 8 47, 022 0.9 Winclassified ea: seers aie 2S ae em ee a a 111, 075 2.1 Mota sees Seek Soe ok ee ee ee ee ee Se eae 5, 228, 352 100. 0 1 Adapted from table, facing page 24, of Estado de la Instruccién en el Perfi Segun el Censo Nacional de 1940 (Informe Especial), Lima, 1942. TABLE 2.—Language groups, exclusive of Spanish, among Peruvians over 5 years of age, arranged by Departments } Indians speaking— Total Department Quechua population Quechua and Aymara Spanish TANMAZONAS= Eee ae ee eee ne re ed peas oe oe we ees 1, 715 9,970: |-2- essen 53, 982 JATICRS Nt Se pas ee Se et Sas setae em SIS 194, 888 105; 1355 |taee eee 354, 892 PASOUI ELIT AC Mee ae eae eater a er We SL ee a 186, 448 28: 819" |_ 22. eee ee 216, 243 WAT Q UIP Ae Pet sas es AP ae ek oh tee 37, 817 38, 347 817 225, 244 PAST ACU CLL OME e ete ee Re ee ee en ete 246, 947 49: QAO: os see 299, 769 Cajamarca. $2 oat oe eo le eee he 3, 368 21, 80802. aaa eee 405, 975 COPE oh cl Sa aa, Eee ls pa ON pa sale ep pie. | roti 8 al 3) 900: so eee ee 72, 295 CUZ COR Se ee eR oe ees a ee ae ie 326, 238 77, 378 61 411, 298 MMIANcaveligaee et tee ese ene eee een ee nee 160, 153 40, 434 Wc oo ee 203, 128 PAU AMO COM ee Se Rh Re Le oe ot es 97, 557 68,1125 eet. 6 eae 193, 235 Ce eee ne eee ee ne cle ee ania eee ee 2, 157 14 982022 seoseeaee 118, 371 UU Tn a ee te eee | Ee pee Bel oe 112, 397 169, O2ilel: == - 2 ee 361, 878 Mapliihertad sees -eeentmens rene eens oP esen ere ees 534 Be (23 pena e 319, 514 Lambayedtess< 188 ee Eek 4, 562 6856) | 324 162, 967 Og Uaa Gy ihe al OM lly ch mn NM ci kale, Sa WRebaapl 17, 072 SCSI ee 721, 818 Loretoct< sts on AE oe ATEN oe oe ke eo ee 8, 881 34;,02%, (22: se asses 137, 932 Wiad reer iOse see ne ee eae eee eee e eee e ene 92 AG0 | 2 oci See 4, 098 IMfOQue2U8i2 a. 588 a 8 ee ob Rs Se 4, 414 3, 856 2, 651 29, 034 IBA Fae re! eee eae BL ome ee. Pane meine eas ee 87 SUG ee eee 343, 357 TEU ee Se a lek Led La rae Bg 210, 236 40, 736 176, 129 463, 080 Dan Wartinwsee coerce soe ae eee eo. eee 9, 580 55662) |. 2 eon eee 77, 197 PUN ACTA eee ae earn pt one try ee Ie aS eee 13 391 5, 085 31, 684 Mumbes ek Besta ale at beta ate. Died se. vealed eee 961/62 16 Sea 21, 212 UMreported 2 aie cet ose Seek Se ee Se eee ee ae a I eS ee 149 ZING Get SE ed SE PL Dee EO ES 1, 625, 156 816, 966 184, 743 5, 228, 532 1 Adapted from table facing page 24, op. cit. Study of the foregoing tables shows that the Quechua-speaking pop- ulation, including bilingual persons, are exactly equal in number to the Spanish-speaking inhabitants of the country. Those speaking VoL. 2] CONTEMPORARY QUECHUA—MISHKIN 413 Spanish only are a minority of the whole population, and are concen- trated mainly along the coast. The Quechua speakers, as the table of political divisions demonstrates, make up the majority of the popula- tion in the central and southern departments. Cuzco, with 326,238 using Quechua exclusively and 77,378 bilingual residents of a total of 411,298, is 98 percent Quechua speaking. In Ayacucho, all but 1 per- cent of the population can and do speak Quechua. The center of Aymara in Pert is Puno, while some Aymara islands exist in Tacna and Moquegua. It is apparent that bilingualism is a common phenomenon in Pert. Approximately 16.5 percent of the population over 5 years of age are classified as bilingual (Spanish-Quechua and Spanish-Aymara).