subshnnelike niches on the wall's central cardinal offsets ( 励川め are today called なイ e レ酊示 , that is, deities meant for an individual's personal devotion (see Fig. 48 ). ln contrast tO the icon1C lmage that makes divimty present 、 vithin the sanctum, 32 these figures act pnmanly as part of the temple's penpheral mediating ornament. Such ornament, however, IS essential to the temple's core symbolism. Little differentiates these deities 伝 01 れ the fabric of other ornamental motifs that mark and measure the temple's vanegated surface. There, a hlerarchy of images—from lnsects and plants tO humans, celestials, and dlvinitles—was meant tO matenalize and make visible a single reality. lt is in the sense of the temple as agent of human salvation that this exhibition s title, Gods, Guardians, and Lovers, seems appropnate. As the hlerarchy 0f lndia's cosmogony was glven visible form, the temple and its ornament became lts embodiment. lts surface represents the atom1C particles Of creatlon; its base, superstructure, and upper altar the three levels Of cosmic reality; and the multitude of deities enshnned by niches the particulate nature Of divme manifestation. Fneze From the bare cube of early stone temples (see Fig. 34 ) to the most hlghly ornamented shnne (see 日 g. 56 , p. 111 ) , the frieze of the temple represented the skln and fabric of cosrmc partuntion. As with a geode, the solidity of divine reality resided within the inner space Of the sanctum. Access for ritual was through the sanctum's door. Th1S inner reality could alSO 100n1 visibly on the extenor walls through blind "doorways" pro. 」 ected 伝 om the temple's cardinal faces as "niches" (see Fig. 39 ). Such niches framed divine lmages and, in many later temples, took the 応 rm of small subsidiary shrines. From Other offsets in the wall, celestial attendants emerge, muslclans hover, and an older order of deities stands guard. The temple branches into many forms, as had the trees Of ancient tree shnnes, its limbs entwined by the illusion of reality 33 ( 川アの . lt stands as an lnvisible cosm1C being, its ornament the created world. Elevation ln its high Hindu 応 rm , the temple often stood on a broad rectangular platform, ornamented by moldings and with niches containing divinities. Secondary shnnes often stood on the corners of this platform, as at Khajuräho (Fig. 49 ) , g1Ving matenal presence to the multiplicity of deities within this constructed cosmlc envlronment. BOth these and the central shnne Often had separate plinths 卍 ) ornamented in molded bands by rain-cloud elephants, sea creatures ( 川口同 , geese ( 〃研示の , dlvinized wamors ( ⅵめ′ ra ) , vines, and other motifs to suggest a celestial abode (see Fig. 50 ). The wall moldings 0 「 the temple ( イル a , "the altar's bonding") are usually placed above floor level, binding the sanctum. These include a 、 'water-pot ' molding ( 信 ) occasionally shown sprouting foliage and fruit, in reference both t0 the temple's rootmg in lndia's 34 water cosmology and tO lts ritual lustratlon. Niches placed on these moldings contaln lmages Of subsidiary mcarnatlons (for example, the Saptamätrkäs or Visnu S 4 レ a ー汀 ra め or sometlmes myth1C scenes 伝 0n1 the narratlve texts called 2 ra 卩 4 立 The frieze hierarchically places emphasis on the central divinity, with attending deities tO either side and guardians 0 「 the eight directlons on the corners. Celestial females, leogryphs (composlte animals) , muslclans, and Other ornaments flower 伝 0n1 recesses in the walls. The transltional cormce moldings above the frieze sometlmes carry narrative reliefs depictlng scenes 佇 0n1 the ePICS (the Ma 〃汀励汀 ra and R 汀川叫叩の or the 2 肝叩 . Few images are placed above ln the superstructure. ln its 応 rmal multiplicity, the temple's tower evokes the celestial realm 0 「 Fig. 42 Barrel-vaulted ( Ⅷ信励 / ) r00 「 of TeIi-kä-Mandir temple, Gwalior, Madhya Pradesh, about A. D. 75 ( ). 103
絲三第し窈を一 . こ氿イ第「物ン可野 ノノ ・ , こし三 finial / la をな汀 ) ThiS structure, then, as ln lnscnptlons that pralse a number of Fig. 46 ). Each of its rmmature r00 was crowned by its own ribbed bell-shaped 27 pyramidal r00 「 type made up 0 「 multiple staggered awnmgs and pinnacles; see led tO an elegant ne 、 archltectural structure known as a 〃れ観 ra ″ (a tiered A similar multiplication in the ⅲ〃な r00 「 over the temple's entry hall also on the wall's frieze. (See the essay by DaneIIe Mason. ) superstructure alSO had consequences for the increaslng sculptural representatlon high Hindu penod (see Fig. 47 , p. 107 ). 2 。 Such elaboration 0 「 ground plan and as た / ta that even ln lts own tlme was perceived as typifying northern lndia's Fig. 45 ) produced, by late in the tenth century A. D. , a multispired formula known the temple's plan. This evolution and multiplication 0 「 architectural elements (see each row Of nunlature rephcas in the tower was set above an eqtlllateral Offset in Ultimately, these distinct levels of subshnnes were compacted into one form ・ reference with its 0 , n potent S1gn value. 25 central mountain among many foothills (see Fig. 56 , p. 111 ) , an indexical 日 g. 39 の℃ / Visnu temple, Deogarh, Madhya Pradesh, about A. D. 500 ー 25. Fig. 40 叩 6 Ground plan with constructlng 墅 ld ofSiva temple no. 2 , Mahua, Madhya Pradesh, about A. D. 675. 101
pinched. Faces are either square or oval with somewhat receding chins and na 江 0 , slanting eyes. ln Gopädr1 to the southeast, the faces of images tend to be longer with pointed features and closely set eyes that Often taper downward at the outer corners. Chests and hips on both male and female figures incline toward narrowness. By the eleventh century, as the Kakanmadh temple at Suhänia, north of Gwalior, illustrates, the narrow chest was transformed into a distinct tube resting on wider yet relatively narrow hips (see Fig. 69 ). Early Madhyade'a images show んⅡ but flattened faces, usually having wide eyes that are pointed at the outer corners with defined lids, which sometlmes cover half of the bulgmg eye. FemaIe breasts 伝 om as early as the eighth century tend to resemble orbs posltioned high on the chest and pressed together above a tmy waist and wide, smoothly swelling hips (Figs. 67 , 70 ). lmages 伝 om the ninth century onward mamtain the short—waisted torso; breasts become even more globelike and limbs rounder. Lips tend to be んⅡ and often brows resemble cords arching upward at the outer corner Of the eye, while noses are straight and flanng, closely abutting the mouth (see Fig. 71 ). Typical images 伝 om Uparamäla around the K0ta area 伝 om the eighth through tenth centuries shOW broad shoulders, narrow hips, and long waists that bend sharply sideways (see Nos. 8 , 72 ). The gently rounded forms are npe yet restrained, With limbs tapenng tO thin ankles. The images have sweet, circular faces, the chin a small mound of flesh and long honzontal, but open, eyes. lmages 伝 om the more westerly area around modern Mandasor (see Fig. 72 ) seem to blend this Kota type with the image charactenstics of both Medapäta and Dafirnade'a—Avanti to the south. Figures 仕 om this latter reglon (see Fig. 73 ) , although little known, are charactenzed by 応Ⅱ three-dimensional forms, smooth, tapermg limbs, and rounded 」 oints. The swelling oval faces 0f the early images show a distinct single plane for the eyelid. Breasts are smooth and shallow, appendages gracefully weightless. Ornament is generally restrained and there appears tO be a focus on full, smooth surfaces. Farther east in Da'ärnade'a, a large, vaned area, bodies tend to be more weighty, often with wide h1PS, although the weight is in the midsection (see Fig. 61 ). Breasts are heavy but, unlike images 伝 om Madhyade'a and especially those 伝 om Jejekäde'a and Dähalade'a to the east, they merge somewhat with the body and often, particularly on earlier images, show gentle, fleshy rolls beneath them. Legs are well contoured, the musculature defined despite thick ankles. Faces tend to be 応Ⅱ and square with sensuously abundant lips, although the mouths are not wide. Eyes take a vanety of shapes but never the long, pointed forms of the lmages Of more easterly locatlons. lmages 伝 om Jejekäde'a (known pnmarily 伝 om Khajuräho) are closely related to those of neighbonng DähaIade<a, but there is more play with overall linear composition (see Fig. 74 ). Breasts are aggressively 応Ⅱ and round, and for both genders weight is concentrated in the chest. Eyes are long and pointed at the mner and outer ends, and chins are formed 伝 0n1 a round ball, usually without the dimple common to Dähalade'a images ・ Limbs are distinctly tubular, and there IS a preference for turnlng postures that remove figures 佇 om the relief plane, although this is somewhat a factor of chronolo , increaslng after the tenth century (see Fig. 75 and No. 46 ). FIesh, even on earlier images, is relatively hardened compared tO images 伝 0n1 farther west. Drapery IS smooth, almost lnvisible, sometlmes with incised patterning but never with the nppling effect common t0 groups ofimages in Dähalade'a. Dähalade<a sculptors 伝 om early on exhibit a preference for figures carved in the round; their backs, although unseen, are cursorlly but fully articulated (see No. 62 ). Poses move away 佇 om the plane of the relief, often fully twisted like a corkscrew (see No. 49 ). MuItipIe subsidiary figures and decorative motifs crowd the relief plane. Drapery in the tenth century and later 仕 om around both JabaIpur in the south and the easterly ShähadoI Distnct often takes a distinctlve nppling 応 rm (see No. 50 ). Fig. 69 月韲示 , Käkanmadh temple, Suhänia, Madhya Pradesh, about A. D. 1015 ー 35. 125
伝 om areas to the east, linking Märude'a in this way with Medapäta and Gurjarade'a. Carving, however, is lush but impreclse, unlike the cnsply cut contemporaneous works 伝 0n1 the area tO the south. Ninth—century lmages 伝 0n1 the eastern area Of Märude'a, ancient Sapädalaksa, show petite bodies with doll—like rounded limbs (see NO. 55 ). Female breasts are small, round, and high, while male images feature wide pectorals mergmg with narrow waists. Faces have rounded, bulgmg cheeks, and closely set eyes. Somewhat later, these features transform intO , i Ⅱ 0 , y bOdies with gently swelling musculature, long, thin torsos, and sensuously na 0 嶬 , waists (see No. 12 ). There is a softness of the flesh but not the んⅡ fleshiness of the images 佇 om Medapäta. Works 伝 0n1 neighbonng Sürasena consistently tend toward heaviness (see No. 28 ). Bodies are short with heavy and tapermg limbs. The flesh congeals mtO an almost cementllke surface. Figures shOW little movement either laterally or away 伝 om the rehef plane and images 応 el solid. Waists tend to be high and Fig. 68 Mahisäsuramardini, Pipälädevi temple, Osian, Rajasthan, 9th century. 124
伝 om plan to frieze, making them proportionally wider in relation to the 励 ra on the wall frieze than they are on the lower moldings. The architects, therefore, were faced with the problem Of restonng a visual hierarchy concomltant with the symbohc importance of the 励 ra images and their location, especially in relation to figures that came to be placed on the side walls (see Fig. 78 ). One solutlon, evident in the case Of the 市た 2 , was tO reduce the sigmficance of the images on the side offsets by substituting lesser beings for major deities. A second solution was to shnnk the width 0f the side walls by altenng the 22 relationship bet 、 veen plan and elevation. Yet a third solution, evident even ln early temples beanng lmages in places other than the 励 ra , was to develop a hierarchy of frammg to glve order to the wall's offsets. Thus the 励記 ra niche was articulated as a subshnne—sunken more deeply intO the 、 plane—or made more elaborate or larger than the niches Of images on the side Ⅱ planes (see tFig. 4 ). A fourth solution that became popular 伝 0n1 the ninth century ln western lndia and substantially altered the 100k Of the wall was tO indent a na な 0 , mner sectlon Of the corner and intermediary offsets, restormg pnmacy tO the 房イ ra , which was to retain its 応Ⅱ width (see Fig. 80 ). These vertical recesses soon became a crucial, although optional, ingredlent 0 「 the developed wall in all styles of north lndian temple architecture (see Fig. 79 ). Ninth—century structures in bOth western and central lndia resolved the architect's search for appropnate occupants Of the intermediary OffsetS (pratirathas) by treating the iconogr 叩 hy 0f this offset as attendant t0 the 励 ra い hierarchy implicit in the plan (see Fig. 40 ) ]. As with the corner offsets of early eighth—century monuments, these intermediary OffsetS on many eighth— and early mnth—century temples display a m1X Of deities, some m or in and Of themselves but usually subsidlary in relation to the monument's dedication (see Fig. 80 ). ln ninth-century temples, however, a different order of beings appears. After initial expenments With rmnor deities, female attendants became the fixed occupants 0f this 0 et. At first these women often held flowers or fly whisks, 23 ObJects denoting their attendant functlon, or they danced in celebration Of the deity. CeIestiaI females who have no specific divine identlty ( 叩“ , Ⅷ ns ) thus seem tO have first appeared on exterlor walls Of north lndian temples as attendant figures on intermediary projectlons. Certain figures, especially those 伝 01n the ninth and tenth centunes, are set in poses that visually relate them tO the deity on the 励記 ra and their subsidlary posltion in the hierarchy is expressed by reduced scale. An alternatlve tO shnnking such lesser figures was tO remove the niche frame altogether. The tenth-century temples called 、 'Mahä—Gurjara," which appear pnmanly in Gujarat and southern Rajasthan, exhibit a preference for this solution. With the exception of figures on the 励記 ras and sometimes the corner 市 , all wall figures are unframed in this architectural configuration and directly addorsed tO the wall surface where they frequently stand on single open lotuses whose stems sprout 伝 om the masonry body of the structure (see Fig. 81 ). This unframed configuration began t0 appear with increasing frequency 伝 om the mld-tenth century onward in central lndia as well (see 日 g. 79 ). A fantastic, reanng, lionlike animal ( ) finds a fixed location in the vertical indentations, which, as mentioned, allowed adjustment Of the wall rhythm and hierarchy. The 応 rm of the ロ may have nugrated to these recesses 伝 om an earlier location as a prancing bracket tO the sides 0 「 the colonettes that framed the bhadra's niche (Fig. 43 ). By the tenth century it appeared in vertical indentations on monuments 0 「 varymg wall articulation 伝 om Kutch t0 KhaJuräh0 (see Figs. 4 , 79 ). AIthough 叩 can also be found here, the is never placed on any pro. 」 ection. This is alSO the case with the ascetics ( 川Ⅲ 1 な ) whO replace り , as in the recesses in the eleventh—century western lndian temple, at tlmes leenng at the 4254r 示 , linkmg the indentatlons and projections 1ntO a playful narratlve. Where flanking offsets have figures, females always appear ln some relation to the intermediary 0 爪 et. This is also true of the walls of enclosed halls, where Fig. 82 娵 Extenor wall and enclosed hall 7 曲ロⅢ叩叩の , Ambikämätä (Devi) temple, Jagat, Rajasthan, about A. D. 961. 日 g. 83 畩 Exter10r wall, Kandanyä Mahädeva temple, KhaJuräho, Madhya Pradesh, about A. D. 1025 ー 50. Fig. 84 Exter10r wall, NiIakantha Mahädeva temple, Sünak, Gujarat, about A. D. 1075. 134
eighth century onward, Gane'a often appears in the 励記川 (central offset) niche of a 2 southern wall ( No. 1 ). SchoIars who have pubhshed this 3 sculpture have called it eighth century, 5 mnth century, and ninth tO tenth century. exammatlon Of the carvmg as well as certain details, partlcularly Of ornament, seems tO warrant an early date—probably no later than the end of the eighth century. Ornament IS simple, as are the round lotus pedestal with stem, layout of figures and pierced rehef ground. This Gane'a is really carved in the round, the features of h1S back roughly but fully chiseled. The naturalism of the lotus pedestal as well as the so 仕 undulating flesh and weight of the bodies also indicate an early date. The particular 応 rm of the たル川たん 4 in the center Of the crown, more monkey than lion, as well as the dimensionality of the foliate scrollwork connect this image Of Gane'a with monuments Of the eighth century. These forms retaln a taste Of earher structures, such as the Munde'vari temple at Ramgadh, Madhya pradesh, datable to A. D. 636 7 that are Just emerglng 伝 om Gupta conventlons—convenuons Where the vegetal and animal forms that become stylized ornament by the ninth century stlll retaln a clear VISIOn Of their prototypes ln nature. R ( 肥れ化 : Lee 1975 : 17 , 20 , no. 8 ; The Asia Society 1981 : 11 , 1982 : 3 ; Sutton 1983 : 364 , fig. 9 ; Newman 1984 : 44 , 75 , 84 ; Bird et 1985 : 28 , fig. 6 ; Chandra 1985 : 112 ー 13 , no. 46 ; Courtnght 1985 : pl. 1 ; Pal 1986 : 71 , no. 16. EITA, V01. 2 , pt. 1 , pp. 115 ー 23. 7. Ramgadh is located southeast 0f Varanasi. See ( 石ー T 月 , vol. 2 , pt. 2 , fig. 168 ). Archeolog1cal Museum, Lucknow Umvers1ty fragment 伝 om Ahar, Madhya Pradesh, in the See, for example, the probably late eighth-century Robert Brown. Pramod Chandra. 3. Richard Newman, Paul Courtnght. See the essay by Danelle Mason. vol. 2 , pt. 1 , pl. 220. 1. Doorway 0fAurangabad Cave NO. 6. See EITA, 6. 5. 4. 2. 17. Twenty-armed Gaqe'a ProbabIy extenor wall, central offset Provenance unknown, probably central-eastern Madhya Pradesh (D ähalade'a) About early 1 Oth century Sandstone; 68.3 x 23 cm JamesW. and Marilynn AlsdorfCoIIection, Ch1cago His t 、 venty arms fanned out around him like a halo, th1S dancing Gape'a, elephant-headed Remover 0f Obstacles, combines extraordlnary exuberance With a gentle, inward hesitancy. Left leg steppmg forward in the dance, heel slightly raised to stamp, and nght knee bent, he swings h1S h1P and bulglng belly backward, counter- balancing with h1S head and opposite shoulder. ln his many hands Gane'a holds an array of attnbutes, includmg h1S own broken tusk ( No. 16 ) as cornucopia bursting with fohage (upper le 仕 hand) , a reference to his early role as Lord 0fAgnculture and thus れⅲ呼 As his name implies, he is lord of Siva's host of rotund dwarfs (g 叩 ) who h1S own corpulence, probably indlcatlve 0f his as a 4 or early lndian 10Ca1 nature s int. Son of Siva and Pärvati, Gane'a holds iva' s snuling snake (upp er nght hand) , crescent n100n (center le ) , and skull cup (center nght) as well as the conch (center le and mace (center nght) that may indlcate an attempt tO mcorporate Vaisnava afflliation. His supple, curhng trunk plucks a sweet ball 伝 om the ubiquitous bowl in his forward le 丘 hand. Other hands dlsplay 川 of the dance and musical instruments. dances on a lOtus pedestal similar in f0n11 tO the earlier Gape'a ( NO. 16 ). TO his le 仕 a long, llthe rat, his レ〃 4 〃 4 , twists upward tO gaze appreciauvely at the god, while a second rodent crouches nearby. Directly above the god's head, a heavenly garland-bearer pr0Jects e 〃 - 化伝 0n1 an ornamental, multilayered cloud. Flankmg clouds hold celestial couples playing muSIC t0 accompany the dance. AdditionaI muslclans at the god's feet beat percusslon as they sway vlvaciously to the divine rhythm. S corkscre 、 pose With one shoulder forward to balance an outthrust hip, seen even 1 れ ore clearly in the dancing drummer by the god's nght leg, is a typical 1 167
25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Michael 嶬を Meister, Jain Temples in Central lndla," in 月平“な 9 リ月なイ 月は町ら ed. U. P. Shah and M. A. Dhaky (Ahmedabad: L. D. lnstitute, 1976 ) , pp. 223 ー 41. I here choose to avoid the term 川レ . The word れ畩ロ ra in th1S penod seems to have referred to the muluspired 立韲 i 応 rm. Meister, "Phärhsanä in Western lndia. For example, see Michael 嶬を Meister, 'Histonography ofTemples on the Chandrabhägä, Reconsidered," in ん市 石叩んア , な B ⅲ 6 H ⅳ 0 研月れ , G. S. Ga1 and Frederick S. Asher, eds. (New Delhi: Oxford & IBH Publishing Co. , 1985 ) , pp. 121 ー 24. Michael W. Meister, "Präsäda as Palace: Kütina Ongms of the Nägara Temple, 月市ら月豆 , vol. 49 ( 1989 ) , pp. 25480. See also, idem "Siva's Forts in Central lndia"; idem, "On the Development ofa Mo 甲 h010 for a Symbolic Architecture. Knshna Deva, "Bhümija Temples," in & 祠沁ⅲん市 TempIe 月″“ tu 肥 , ed. Pramod Chandra (DeIh1: Amencan lnstltute oflndian Studies, 1975 ) , pp. 9g113. SteIIa Kramrisch, M 明立 4 研 S Ⅷ (Philadelphia : PhiIadeIphia M useum of Art, 1981 ) , provides the rich religous and mythic setung for Saiva sculpture. See also MichaeIW. Meister, "Display as Structure and Revelation: On Seeing the Shiva Exhibition, ” S 市ⅲ防 C 明ⅧⅢ 4 れ , vol. 7.4 ( 1981 ) , pp. 84 ー 89. Oleg Grabar, 月肥 M ど市ロ朝れ研 Om 明肥 (Princeton: Pnnceton Un1versity press) ()n press). StelIa Kramnsch, "The Temple as Purusa," in S ルイ沁ⅲん市 TI 明甲 月″“ tu 肥 , pp. 40 ー 46. Ananda K. Coomaraswamy , Ya 越 , 2 vols. (Washington, D. C. : Sm1thsoman lnstitution Press, 1928 ) , p. 31. Ananda K. Coomaraswamy, "An lndian Temple: The Kandanyä Mahadeo" and Svayamätrnvä: Janua Coeli," in Tra 市れ月灯イ町川川 , C00 川韲ルロ川ァ , な & [ P 叩にパ , Roger Lipsey, ed. (Pnnceton: Pnnceton Umversity Press, 1977 ) , pp. 3 ー 10 , 465 ー 520. Dhaky, "Präsäda as Cosmos," 77 肥月韲 石わ ra げ Bulletin, vol. 35 ( 1971 ) , pp. 215 ー 17. These moldings are called the " 市厖〃渤心 or the "bindmg" ofthe "altar" ( 市 ) ; see Meister, "Readmg Monuments and Seeing Texts. Michael 嶬′ . Meister, "Symbol and Surface: Masonic and PiIlared Wall-Structures in North lndia, ”月れ面バ 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 月朝 , V01. 46 ( 1985 ) , pp. 129 ー 48. My par 叩 hrase 伝 om Dhaky, "Präsäda as Cosmos,' p. 220. For the development ofthis doorway imagery in the early Gupta penod, see Michael 嶬を Meister, "Därra and the Early Gupta Tradit10n," in Chhavi 乢 Rai Krishna D ロん羸〃ん川 6 , Anand Knshna, ed. (Banaras: Bharat KaIa Bhavan, 1981 ) , pp. 192 ー 205. For the concept 0f "meditational constructs" see T. S. Maxwell, "Nänd, parel, Kalyänpur ・ Saiva lmages as Med1tational Constructs, ” in D な ( 0 パ 0 〃 & Ⅷ , pp. 62 ー 81. For "limnality" in a different religlous context, see Linda Seidel, S 研 G ん : 77 肥 Ro 川Ⅲ肥 んイ研月ⅲ e (Chicago: University of Chicago Press, 1981 ). ThiS sect10n, ln part, is based on a presentation for the "Pafica—mahäbhüta' Semmar held at the lndira Gandhi National Centre for the Arts, New Delhi 1992. MichaeI 嶬を Meister, "The Hindu Temple: Axis and Access," in C 明 ( な研 韲ら月〃厩明イ Mo 川 , Kapila Vatsyayan, ed. (New Delhi: lndira Gandh1 National Centre for the Arts & Abhinav Publications), pp. 269 ー 80. Meister, '&leasurement and propo 10n idem, "Symbol and Surface. ' Michael 嶬を Meister, "The Hero as Yog1n: A Virastambha 伝 om Chittor," ln D ⅲ旧ババ研ん市月な , Lokesh Chandra and Jyotindra Jain, eds. (Delhi: Agam Kala Prakashan, 1986 ) , pp. 283 ー 86. Michael 嶬仁 Meister, "Temples, Tirthas, and Pilgnmage: The Case of Osiäfi, in R 酊 - C れ市 D. Handa & A. Agrawal , eds. (New Delhi : Harman Publishing House, 1989 ) , pp. 275 ー 82. See Joanna G. Wilhams, "Siva and the Cult ofJagannätha: に 0n0 部叩 hy and Ambiguity, ' in D ⅳパ & Ⅷ , pp. 298 ー 311. Meister, "On the DeveIopment 0 「 a Mo 甲 h010 for a Symbolic Architecture"; idem, "Altars and Shelters in lndia," ロロ甲 (Art and Archaeology Research Papers), vol. 16 ( 1979 ) , p. 39. Meister, "Mupde'vari: Ambiguity and Certainty. Richard Lannoy, 7 方に S i 行 : 月 & " 研ん市 CuIture and So (London: Oxford University Press, 1971 ). Roger Eliot Fry, "lndian Art," in I 立 レ ctl 催 5 (Cambndge: Cambndge University Press, 1939 ). 115
25. 14. 15. 16. 17. 18. 19. 21. 22. 23. 24. external acts ofworship and its emphasis on personal expenence. The text IS translated intO French with an introduction by Charlotte Vaudeville, L' ん ca ) 広レ H ロ叩わ D ″ ) ′ ( ⅲⅧ , (Par1s: Publications de l'Ecole Franca1se d' Extreme Onent, V01. 73 , 1969 ). Rocher, op. cit, p. 252. For further informatlon on royal patronage of temples in north lndia, please refer to the essay by MichaeI - 引 lis. The E たⅣ叩町卸ロ , which describes the temples ofBhuvanesvara in onssa, IS explicit about temple building as a 応 rm Of イⅲ as are medieval digests on ntual. See ch 叩 ter 49 , verses 17 and 29 0 「 the E た叩 " 瘤 , edited by Dr. U. N. Dhal (DeIhi: Nag PubIishers, 1986 ). The 石たⅲ叩肝叩 a belongs to a subgenre 0 「 2 r 衂〃 c wnting, the 立 / za 信′ la 立 For details on these texts see note 7. Like the レ ' ⅳⅢ襯 la ( ) ″韲叩町ⅲ la , the 7 ) レ汀 ra たな川な〃第川 ) shows Päfcarätra influence, although it is probably considerably later in date. For some comment on the text see Franqoise MaIIison, 、 'DeveIopment of Early Knshnaism in Gu. 」 arat: Visnu— Ranchod-Krsna," in Monika ThieI-Horstmann, ed. , Bhakti ⅲ C 肝肥 R い夜翫〃 , 797 屮 7982 (BerIin: Dietnch Reiner Verlag, 1983 ) , pp. 245-57. For bo 市 the D ra 川励川 and the レ 'iST Ⅲ d 〃 at ・川 ra 〃卩 a , I am using the editions 佇 om Nag Publishers in Delhi. ー am using the Anandä<rama edition, 1894. This interpretation is discussed in my p 叩 er cited earlier on the 7 劬 / K / れ川 0 「 the S たイ叩肝叩 a. I have taken the liberty in my translation ofreducing the number ofdifferent names and substituting for an alternate name the more familiar "Siva" when I suspected that the different names might be confusing. See Rocher, op. cit. , p. 233. For further discussion 0f genealogrs 0f different subcastes in the medieval per10d, see the essay by B. D. Chattopadhyaya. This is my argument in the p 叩 er I have cited in note 8. The P ⅲ - ゾ回リ a 川川黻川 0 「 the twelfth century, S. D. Belvakar, ed. (Calcutta: The Asiatic Society, 1914 ) , chapter 1 , p. 28. A good source ( ) 「 informatlon about traditional Jain accounts Ofthe life and converslon 0 「 Kumärapäla iS Georg Bühler's 77k , なん日 e 川 a 夜川市汀り , translated from German into English by ManiIal PateI ⅲ the Singhi Jain Senes, no. 1 1 (Santiniketan: Smghi Jaina Jnanap1tha, 1936 ). The text is published 斤 om the Jaina Atmänanda Sabhäⅲ Bhavnagar, 19 ( ) 9. 93
society, dommance 0 「レロ〃 1 ロ ideology, heterogeneity 0 「 social groups, and many-stranded religlous beliefs and practices. For these implicatlons see B. Chattopadhyaya, 月平“な ofRural & なん川群市イ RuraI & り , ⅲ石祠 ) , M 司ん市ロ ; also 、、 The Making of Early MedievaI lndia" (forthcoming). References t0 forest reg10ns ( 可襯の are found in different verslons 0 「 Ma. 」 or D. 16. 17. 18. 19. 20. 21. 22. 46 Rock Edict 13 of A'oka in a context that suggests the forest people did not reconcile tO the penetratlon 0 「ル 12u , an lmpenal power intO their reg10n; for the text Of the edict see D. C. Sircar, op. cit. , pp. 34 ー 37. D. C. Sircar, op. cit. , p. 395. For example, the Nalas who were certainly associated with Koraput and Bastar 伝 om the first half of the sixth century; R. C. Majumdar, 77 肥 C 信 / 月 , pp. 188 ー 90. The spate in the emergence 0 日 ocal ruling lineages ln Gupta and post-Gupta tlmes is underlined inJ. G. De Caspans, 、、 lnscnptions and South ASian Dynastic Tradition,' ln 7 元市朝〃イ P 司ⅲⅲ S 月朝ロ , R. J. Moore, ed. (New Delhi: Vikas Publishing House Pvt. Ltd. , 1979 ) ; B. D. Chattopadhyaya, "Political Processes and Structure of Polity in Early Medieval lndia" ・ R. S. Sharma, U Decay ⅲ ん市ロ , chapter 10. F. K1eIhorn, "Two ChandeIla lnscnptions 佇 om AJaygadh," 石 rap ロん市朝 , V01. 1 (Repnnt, New Delhi: Archaeolog1caI Survey oflndia, 1971 ) , p. 337 ; Munshi Deviprasad, "GhaCayälä lnscnptlon 0 「 the Pratihära Kakkuka,' 肝い戸 Ro / 月豆ロ & り′ ( 1895 ) , pp. 513 ー 21 ; D. R. Bhandarkar, 。、 Gha!ayälä lnscnptions of Kakkuka: Samvat 918 , ' ' 石 2 叩 / ん市 ca , V01. 9. Repnnt (New Delhi: Archaeolog1cal Survey oflndia, 1981 ) , pp. 277 ー 81. For a disct1SS1()n 0 「 different theones the ongm of the Candellas, see chapter 17 ( 、、 O ロ n ofthe Chandellas") inJ. N. Asopa, 〇畩ⅲ〃肥 R 卩 " な (DeIhi-Varanasi-Calcutta: Bharatiya Publishing House, 1976 ) , pp. 208 ー 17. Asopa, 0 「 course, traces their descent 斤 om an old Brähmana ruling lineage. See S. K. Mitra, op. cit. , chapter 2. For an analysis ofdifferent dimenslons 0 「 the process see B. D. Chattopadhyaya, 、 ongm ofthe Ra. 」 puts: The Political, Econom1C and SOCial Processes in EarIy Medieval Rajasthan," 77 肥ん市聞 Historical R れ塒 vol. 3 , no. 1 ( 1976 ) , pp. 59 ー 83. For the unusual orlgln myth that traces the ancestry of some 10Ca1 kingdoms through the protectlon gwen to 2 child by mice, see Rornila Thapar, 、 'The Mouse in the Ancestry, ' in , 4 川月イ / lara (Professor R. N. Dandekar FeIicitation Volume), S. D. Joshi, ed. (New Delhi: 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. A. 」 anta Books lnternational, 1984 ) , pp. 429 ー 30. H. V. Tnvedi, ん“′甲〃肥 ra 川な ras , C わ〃 , K 韲〃 c/ 叩〃第 イ Two Minor D 川口ⅵい ( Co 甲町 ん“Ⅷ川ん市 ca Ⅲ川 , V01. 7 , pt. 2 ) (New DeIhi: ArchaeoIog1cal Survey of lndia, 1978 ) , p. 76. lbid, p. 79. 月 Co ⅲ 0 〃可、 P た加 and S ⅲ ん“叩朝 (Bhavnagar Archaeolog1cal Department, n. d. ), p. 89. F. K1eIhorn, "lnscnptions 丘 om Kha. 」 uräho," 石 2 rap / ん市 ca , V01. 1 , p. 130. See Chitrarekha Gupta, 、、 TheWnters' CIass ofAncientIndia—A Case Study ⅲ SociaI MobiIity, ' 77 肥ん市 0 加川な and So / H ⅳ 0 り′ Review, V01. 2 , no. 2 ( 1983 ) , pp. 191 ー 204. Ancestnes ofmodern Käyastha castes like the Nigamas, srivästavas, and . Mäthuras may be traced to early medieval records 伝 om Madhya Pradesh, Uttar Pradesh, and Rajasthan in the penod W1th which the present essay concerned. B. D. Chattopadhyaya, 、 Markets and Merchants in Early Medieval Ra. 」 asthan, & / & 1 化 Probings, V01. 2 , no. 4 ( 1985 ) , pp. 413 ー 40 ; V. K. Jain, Trade イ 7 元 ⅲ月Ⅲん市 a い . D. 7 00 卩 0 の (New Delhi: Munshiram Manoharlal Publishers Pvt. Ltd. , 1990 ) , chapter 9. Devendra Handa, 〇豆 : H ⅳ 0 り′ , 月“ 0 / ) ′ , , 4 灯イ月ⅲ“ tu 尾 (Delhi: Sundeep Prakashan, 1984 ) , chapter 6. V. K. Jain, op. cit., pp. 245 ー 49. D. R. Sahni, 、、 Ahar Stone lnscnption, 石 2 叩 / ⅲん市 ca , V01. 19 ( 1927 ー 28 ) , Repnnt (New Delhi: Archaeolog1caI Survey oflndia, 1983 ) , pp. 52 ー 62. For further information on this sub. 」 ect, see the essay by Phyllis G ロ no 圧 S. B. Dasgupta, 月〃んな 0 市にロ ( ) T 汀 B 渤ⅳれ , 3d ed. (Calcutta: University of Calcutta, 1974 ). For Täntnc elements in Jainism, see R. N. Nandi, Religious ん立ⅲⅲ ( ) の C なⅲ〃肥 Deccan (Delhi: Motilal Banarsidass, 1973 ) , chapters 9 and 10 ; R. B. P. Prasad , 山ⅲな川ⅲ石韲 M . 司 K ロ川酊ロを .. 4. D. 50 720 の (New DeIhi: MotiIaI Banarsidass, 1975 ) , chapter 3. Pra 伝 ) 渤韲盟市 ( ) ( ゾ K ta 川ロ , Sansknt text with English translation by S. K. Nambiar (Delhi: MotilaI Banarsidass, 1971 ). The allegoncal play Pra 伝 ) 渤韲イ ro , wntten by Kc»a N'li'ra and staged at the court 0 「 a Candella ruler, is evidence 0 「 acnmony that certainly exrsted among followers 0 「 differen い ects in early medieval lndia. Serg10 Meliton Carrasco Alvarez, 、 Brähmanical Mona.stic lnstitutlons ln EarIy Medieval North lndia: Studies in
the space flanklng the central window projections m1m1CS the intermediary offset of the sanctum (see Fig. 82 ). Poses and attributes of ロ 2 ′示 and レア show great vanety. However, the vanety IS probably a function of the multiplication of such images and seems largely aesthetically motlvated, although at times types bear literary significance. The placement of type withm any designated location does not appear to be canonically dictated, although specific cases exlst in which figures exhibit a clear relationship to the niches they flank by either pose or attribute. Thus the utilization and placement, including relative placement, of m or deities, 市た 2 , ロ′示 , and レア on はⅡ S seems tO have resulted 伝 0n1 a search for a hierarchy of significance to reflect the symbolic hierarchy 0 「 the floor plan. The differentiatlon made salient by framlng and scale allows the wall's ornamental rhythm to visually present this hierarchy. Axiality in relation to the sanctum and door, compass direction, sequential arrangement, clrcumambulatory order, hierarchy within the wall, and visual coherence are all at play on the "medieval' temple exterlor. The foregoing reduces the iconographic program of the wall frieze to its essentlals. After the ninth century the wall was seldom so simple, however, and vaned greatly 伝 om reglon to reglon. ln many parts of central lndia, for example, the frieze came to be divided by honzontal moldings into two or three repeatlng levels of images (see Fig. 83 ). Contemporaneous western lndia retained a single level of images but filled the larger walls with small figure groups or side offsets articulated as P111ars 、 vith elaborate raised niches supenmposed on them (see Fig. 4 ). Figure sculpture on the temple exterlor is by no means limited tO the wall frieze (see Fig. 38 ). lmages were placed in a vanety of locations on the superstructure Of bOth sanctums and halls, in niches on the binding moldings ( レルの , and in the moldmgs of the basement の (Fig. 84 ) and platform ( / ロの and covered the hall extenors as well (Fig. 82 ). The types of images placed ln each Of these locatlons were never arbitranly chosen, however. Their selection resulted 伝 0n1 reasoning that, if not always evident, can often be discerned by exammatlon 0f the histoncal lineage of each type. ln addition, there were Other forms Of contemporaneous stone archltecture that could include a retlnue Of images. These include the freestanding arch 00m 卩の and pillar, the monastery, and the sometimes elaborate step wells popular in western lndia (see Fig. 85 ). ScuIptures 伝 om these types of structures, divested of their onglnal context, are frequently iconographically and compositionally indis- tinguishable 伝 0n1 images that were once part 0 「 standing temples. The often multiple and multipurpose halls that came to front the temple were less stnctly related tO the cosmogonic function of the temple than were the sanctum walls. Thus architects were freer to be innovatlve with plan and decoratlon as lt suited use and patron. ln general, the interlors of these halls reutilize forms for bOth images and ornament found on the temple extenor. However, thiS space, sometlmes with an overarching 10tus ceiling (see Fig. 57 ) , multiple pillars ornamented with bursting and winding vegetatlon, and repeated depictions Of dancers and muslclans has a function beyond the representatlon 0 「 the germlnative nature Of divinity. lt is also the forecourt Of the palace Of heaven (the inner sanctum) in which the divine regent is celebrated in stone and li 応 . The door, while containing features and configuratlons charactenstic of specific reglons at particular points in tlme, also bears certaln lmmutable propertles. These include the rlver goddesses on the lower 」 ambs, flankmg the entry, along With Other aquat1C and protectlve motifs predominatlng in the 10 ℃ r portlon. Celebratory and apotropalc lmages were located in the side 」 ambs, and celestial lmages were featured in the lintel and overdoor. Vegetation entwined the whOle. Like the wall images, the elements of the configuration of doorways became Fig. 85 well established early in the tradition—in fact substantially earlier than was the Queen's step well (RäVi ki Väv) , case with wall imagery. Doors of the late Gupta penod make this evident (see Fig. Patan, Gu. 」 arat, about A. D. 1060. 135