Volume - 1
Part - 1
Hindu temples result, in the early centuries A.D., from a felt need to give shelter to images that could make present for worship a divine force that otherwise remained invisible. "Seeing" the divinity - in this period increasingly presented in both aniconic and anthropomorphic form - becomes the central act of this developing form of worship, for which architects were called upon to provide a suitable environment. Vol. I of this Encyclopaedia has covered the Dravida form of architecture as it developed in southern India. Vol. Il, which begins with these volumes, traces the evolution of that form of temple architecture known as Nagara, found principally in northern India but with extensions also into the Deccan under the Calukya and Rastrakuta dynasties.
Building on earlier pan-Indian forms of urban and domestic architecture, architects of the Dravida mode of temple in South India had created, by the late sixth or early seventh century A.D., a palatial structure out of recognizable wooden forms to act as encasement for the inner sanctum in which the divine image was placed.
In North India, on the other hand, architects in the fourth to sixth centuries A.D. participated directly in the process of religious and symbolic experimentation that made visible forms of divine manifestation possible. A variety of solutions resulted, some tied directly to the ontology of manifesting divinity (as is the case with the created cave-cells at Udayagiri and the simple masonry cave-like cells that followed in Central India in the fifth century A.D.). Some solutions mixed sheltering fence-forms with that of the altar, as in later mandapika shrines. Kashmir, building partly on earlier Candharan forms, by the early seventh century A.D. had created a pragmatic pent-roof shed as shelter for divinity that survived there as a regional form for many centuries.
Some architects, however, began to play with symbolically subtle solutions to the need for an architectural shelter for divinity that led, by the sixth century A.D., to a new form of monument - the Nagara temple with its Latina sikhara - that spread widely across North India as a symbol for an emergent Hinduism. This form - potent in its architectural vocabulary - provided a symbolically vital integument for the interior sanctum; in which manifesting divinity was revealed.
This volume provides a background for the formation of temple architecture in North India, surveying the varieties of North Indian experimentation and their survivors as well as the emergent, dominant form of early Nagara structure in western, central, and eastern India and the Deccan. Further volumes will carry the evolution of this form forward and explore its expression and efflorescence in the high Hindu "medieval" period and after.
The style code used throughout this volume as reference for Chapters and Plates follows the Style Outline given below:
Vol. II, part 1: Foundations of North Indian style
I. Beginnings of North Indian Style, c. A.D. 350-650
A. Uttarapatha style, c. A.D. 360-575 Guptas and their feudatories
B. Early Vidararbha style , c, A.D. 350-500 Vakatakas (main branch)
C. Early Vidarbha (Vatsagulma) style, C, A.D. 450-5—Vakatakas of Vatsagulma
D. Apaanta style, C, A.D. 480-533 Traikutakas of Anirrudhapura
E. Kunkanadesa style, c, 450-610 Mauryas of Puri
F. Late Vidarbha style, c, A.D. 550-650 Kalacuris of Mahismati and Early Rastrakutas of Elapura
II. Varieties of North Indian style, c, A.D. 500-1100
A. Upper Indian
1. Magadha style, phase 1, c, A.D. 500-700 Later Guptas and minor dynasties
2. Madhyadesa style, c, A.D. 575-700 Maukharis and Puspabhutis of Kanyakubja
B. Central India
1. Dasarnadesa style, phase 1, c. late sixth- late seventh century A.D. Minor dynasties, mandapika and early nagara traditions
2. Malava style, c. early sixth – late seventh century A.D Aulikaras, Mauryas, and minor chieftains
3. Dasarnadesa style, phases 2 and 3, c. mid-eighth to early tenth century A.D. Pratihara period, mandapika shrines
4. Dahala style, phase 1, c. late eighth – early ninth century A.D. Kalacuris of Tripuri, mandapika shines
C. Western Indian
1. Surastra style, c. late sixth – mid-eighth century A.D.
a. Pre-Nagar ogase, c. late sixth – late seventh century A.D. Maitrakas of Valabhi and Garulakas of western Surastra
b. Early Nagar phase, c. late seventh – mid-eighth century A.D. Maitrakas of Valabhi
2. Maha-Gurjara style, phase 1, Arbuda school, c. seventh century A.D. Capotkatas of Bhillamala
D. Eastern Indian
1. Kamarupa style, c. seventh century A.D. Varmans of Kamarupa
2. Daksina Kosala style, c. late sixth-early eighth century A.D. Panduvamsis of Sripura and Nalas
3. Kalinga style, phase 1, c. late sixth-early eighth century A.D. Sailodbhavas
E. Southern Extension of North Indian style
1. Karnata style, Nagara phase 1, c. A.D. 62-0-750
a. Calukyas of Badami, Karnatadesa
b. Calukyas of Badami, Andhraadesa
2. Karnata style Nagara phase 2, c. A.D. 700-775 Rastrakutas of Elapura and Manyakheta
F. North-western India
G. Styloe of Kashmir and the Panjab, c. seventh – tenth century A.D. Karkotas and Upalas of Kasmira
Use of this outline for subsequent volumes in this series tentatively is as follows:
Vol. II, part 2: Period of Early Maturity, c. A.D. 700-900
III. Nagara style of Common lineages
IV. Nagara style of separate lineage
Vol. II, part 3
V. Beginnings of Medieval Idiom, c. A.D. 900-100
Vol. II, part 4
VI. High medieval period, c. A.D. 1000-1300
VII. Sultanate period, c. 14th-16th century A.D.
VIII. Mughal period, c. 16th-17th century A.D.
Vol. II, part 5: Annotated Glossary and Comprehensive Index.
Style and patronage are difficult masters, and sources of creativity in architectural matters are nearly impossible to attribute in ancient India. In these Volumes we follow style, while dividing chapters according to the likelihood of dynastic patronage. The realities of local guilds and master architects can only be suggested, though through them the greatness of this architecture was created and continually given expression.
As in volume I, the system of diacritics used in this volumes that used by Epigraphia Indian only by using c, ch, and s to suit international. As in Epigraphia Indica, corpus Inscriptionum, and most Archaeological Survey of India publications, e and o are used in order to make possible the distinction between these forms in Sanskrit and e and o in words of Dravidian origin.
Drawings made by the Institute have scales in feet or miles. Others retain those provided by their sources.
Texts from ancient India provide us an insight into the worship of divinities in India and the shelters devised for them. In the Astddhydyl (c. fourth century B.C.), Panini mentions a number of Vedic deities (Agni, Indra, Varuna, Bhava, Sarva, Rudra, Mrda, Vrsakapi, Pusa, Aryama, Tvasta, Surya, Soma, Vastospati, Mahendra, Apamnaptr, Nasatya) who received oblations. Female deities include Indranl, Varunani, Agnayi, Usa, Vrsakapayi, and Prthivl, the last always referred, to as a pair with Dyaus. Post-Vedic female divinities named include Bhavani and Sarvani (popular in the Vahika and Pracya regions), Rudrani, and Mrdani. Theistic devotion (bhakti) had its beginning in Panini's time, a fact made clear by his reference to devotion to Vasudeva and Arjuna, as from names like Varunadatta and Aryarnadatta that indicate that the sons so named were born through the grace, respectively, of Varuna and Aryama. Such devotion extended also to the Lokapalas, to yaksas, and Panini mentions paired deities such as Sivavaisravanau, Sankarsanavasudevau, and Skandavisakhau. Panini knew of images under worship (areas), the mention of which might presuppose the existence of shrines.
Patanjali's Muhdbhdsyn (c. second century B.C.), which is a detailed commentary 'on Panini, mentions the worship of Vasudeva-Krsna as both hero and deity; his identity as one of the four Vyuhas is well established and that with Visnu is suggested. The performance of Visnu's Balibandhana and Krsna's Kamsavadha exploits are popular. Patanjali mentions Siva-bhagavatas, the devotees of Siva, and discusses their unsocial practices.
The Muhdbhdsyu specifically refers to the temples of Dhanapati (Kubera), Rama (Balararna), and Kesava (Vasudeva), with worship attended by dance, music, and elaborate rituals. Contemporary representations of Kupiro yakho (Kubera yaksa) are known from Bharhut and of Balarama from Mathura, An inscribed image of fourarmed Vasudeva- Visnu carrying gada and cakra in his upper hands and clasping a mutilated sankha in the lower hands, held against his chest, from Malhar (Bilaspur District, Madhya Pradesh), is assignable to the close of the second century B.C. Worship of Gauri, Sarasvati, Laksml, and Yami also had become popular.
Kautilya's Arthusdstru, a compilation completed as late as the third century A.D., refers to' the placement within a fortified city of temples that enshrine Siva, Vaisravana, the Asvinikumaras, Sri(Laksml), and Madira (perhaps a fertility goddess associated with the cult of the Great Mother). The ArthaSdstra prescribes that images of Aparajita (Durga), Apratihata (Visnu), Jayanta (Kumara), and Vaijayanta (lndra) should be set in niches as well as ones of Vastudevatas. Most deities in the Arthasastra are common to Panini and Patanjali as well, and pertain to the earliest strata of the manuscripts not much distant from the age of Patanjali.
Early Buddhist and Jaina literature, as well as Kautilya's Arthnsdstru, refer to various types of structures and their embellishments prevailing in the early centuries B.C. and in the Saka-Kusana and transitional periods. Bas-reliefs from Bharhut, Saficl, Bodhgaya, Mathura, and Amaravati (c. second century B.C. to third century A.D.) corroborate this literary testimony. Such evidence can conjure up a picture of a contemporary Indian city, with moat (parikha), rampart (prakara), bastioned and turreted gate houses (dvarattalakas or gopurattalakas), corner-bastions (karnattalakas), ornamental gates (toranas), and busy streets lined with private and public buildings, such as the royal palace (raja-prasada or raja-nivesana), shops and emporia, punyasalas, caityas, and an assortment of small, medium, and large residential houses (including multi-storeyed mansions).
The mansions and the royal palace had various types of pavilions or chambers (known as ktagara, kutagarasala, candrasala, simhapanjara, or harmya). A kutaara or kutagarasala was a roofed pavilion on any upper storey; the former normally was square on plan with a conical roof, the latter rectangular, with a vaulted roof with gabled ends crowned by small stupis or kalasas. A candrasala was an open type of pillared pavilion, normally on the sky-storey. A simhapanjara usually was a bay-window projecting from an upper storey enclosed by a parapet (vedika), lattice (jala), or bars (salakas). Harmya was a rectangular kutagara topped by valabhi or sabha-kara sikhara situated on the uppermost storey.
Shrines were modelled after prevailing domestic structures and the forms of kutagara, kutagarasala, candrasala, etc. were freely borrowed from civil architecture. An independent shrine with a small chamber and peaked roof came to be designated kutagara or kutagarasala (the former square with a domical roof, the latter rectangular with a vault). An example from Amaravati (Fig. 1) is labelled the "kutagarasala of the Mahavana at Vaisali."
A basic form for a shrine was a modest platform with a top slab frequently depicted in Hinayana Buddhist reliefs. According to the Samyutto Nikaya (Yakkha- suttas), the Buddha once relaxed on the "tankitamanca" in the bhavana of Yakkha Suciloma at Caya. The commentary explains tankitamanca to be a stone slab resting on four other stones, obviously referring to a four-legged 'stone dais or altar. The term might alternatively have meant an altar carved with designs, as in the case of the Asokan period vajrasana at Bodhgaya.
Often such altar-platforms were placed under trees, which had been taken as objects of worship (caityas) in India from great antiquity. This combination of platform and tree occurs abundantly on bas-reliefs at Sancl (Plate 3), Bharhut, Bodhgaya, and Amaravatr. A tree enclosed by a railing is designated "cetiya" on an Amaravati relief from the second century B.C. The provision of an umbrella (chatra), a mark of royal dignity, gave a similar significance.
Frequently the dais was enclosed by a railing (vedika) demarcating a sacred area (sthana), This became in due course the symbol or cognizance of a shrine. While describing the caitya (shrine) of Purnabhadra, the Aupapdtika-satra uses the expression "kia-veyaddi," which Coornaraswamy interpreted as "having an earthen or stone slab altar." The term is equivalent to Sanskrit "vitardi" and means a railing or enclosure. In the Aupnpdtikn-sutru, the railing obviously was regarded as an integral part of a shrine. Made originally of bamboo or timber, this vedika subsequently was constructed of brick or masonry, and ultimately of stone.
Shrines of yaksas, nagas, and other divinities worshipped in the early centuries B.C. (copiously referred to in early Buddhist and Jaina literature) were of this sort. Yaksa-shrines are called jakkhayatna or often simply cetiya, bhavana, or ayatana. Tree-worship is far older than worship of stupas, supported by the fact that "caitya" originally meant "vrksa-caitya" while "stupa" denoted a funerary monument embodying the concept of memorial. Both existed before the time of Gautama Buddha or Vardhamana Mahavira: the Buddhists (and, to a lesser extent, the Jainas) adopted both forms and, to an extent, conceptually and formally amalgamated them. The stupa came to symbolise the parinirvana of Gautama Buddha and of past Buddhas, and the tree their enlightenment. By Asoka's time, the worship of stupas was well established, taking on the features and attributes of the older tree-worship. The stupa became "caitya" and a stupa-shrine came to be designated "caitya-grha," "cetiya-ghara," or "grha-stupa. "
Shrines represented on early reliefs show the following varieties: (1) a platform (with or without some symbol); (2) a platform under an umbrella; (3) a platform under a tree; (4) a platform enclosed by a railing; (5) a platform within a simple pillared pavilion; and (6) permutations and combinations of the above. These shrines were mostly modest roofless structures, some had large compounds, a few were provided with one or more toranas, and some are represented having roofs.
Bodhigharas (not illustrated)
Bodhigharas were Buddhist shrines meant for worshipping the Bodhi tree under which Gautama received enlightenment. The spot is represented as a dais (Bodhimanda) under a peepul tree. Coomaraswamy has discussed literary references to Bodhigharas and the early representations on has-reliefs. He shows that these were hypaethral shrines, in some cases with two or more storeys (timber-built galleries that the worshipper could conveniently ascend for lustrating and honouring the Bodhitree).
Bharhut has yielded two reliefs of double-storeyed Bodhigharas, one of which shows three doorways on the ground floor and two ornate windows; the upper floor represents a modest shrine, probably with an apsidal end. The other relief, labelled the "Bodho of Sakamuni," shows a large, complex structure with a circular plan and multiple ornate windows on the upper storey.
Sanci has four representations of Bodhigharas, two of which-appear to be octagonal, one circular, and one apsidal. Three are two-storeyed; the apsidal one is four-storeyed. The top storey of the latter has two ornate windows at lateral ends, as on the apsidal Bodhighara from Bharhut.
Of the two reliefs of Bodhigharas from Amaravati, one is circular, the other apparently rectangular, with a conspicuously tall upper storey, a sala on each side, and a pair of projecting simhapanjaras supported on stilt-like pillars.
A relief of a Bodhighara from Mathura seems to depict a two-storeyed square structure with polygonal projections at the four corners. The spreading branches of the Bodhi tree jut out of numerous windows on the ground floor and upper gallery.
Other shrines (Figs. 2-10; Plates 1-3, 7)
Other religious structures depicted on bas-reliefs at Bharhut, Sancl, Bodhgaya, and Mathura are largely of kuta, sala, and cap a types. A small apse-ended shrine (capakara on plan and gajaprsthakrti in elevation) with three finials on the sikhara is shown at Mathura (Fig. 2). The Jetavana scene at Sanci shows three shrines, of which one is a small circular structure enclosed by a railing with an octagonal ridged roof crowned by a stupl (Plate 1; Fig. 3); two are larger shrines having sala-sikharas crowned by four stupis (Figs. 4-5) and with ornate entrances marked by gavaksa arches that terminate in finials.
Two of the Jetavana shrines figured at Bharhut (labelled "Gandhakuti" and "Kosamba-kuti") closely resemble the salakara shrines at Saficl. The ensemble of a small domed structure and two larger rectangular structures in an Amaravati panel labelled "Savathi" (Fig. 7) confirms the tradition of Jetavana-shrines associated with the Buddha.
A naga shrine shown at Sanci is a square pillared pavilion with an octagonal sikhara crowned by a stupi (Plate 2). The domed sikhara is pierced by gavaksa windows and enclosed by a vedika.
An example of a domed pillared pavilion with a kapota-cornice is seen in the depiction of the Sudhamma-devasabha at Bharhut (Fig. 6). A more complex shrine, with a praggriva-Iike projection at the cornice level and a front window through the domical roof, is shown at Sand (Plate 3). Larger pillared pavilions, roofed by a sabha-kara sikhara and with cornice, are also represented at Sanci. One has a single front window (Plate 7); the other, with two front windows, appears to be a two-storeyed structure, with vedika as balustrade on each storey. Two-storeyed shrines are frequent- ly depicted at Bharhut, One, with an imposing sabhakara sikhara and crowned by ten stupis. (Fig. 8), is enclosed by a vedika railing on each storey. The lower tala shows five stunted pillars with ornate capitals; the upper storey shows three gavaksa-arches.
A shrine suggesting a double-level thatched hut (Fig. 9), engraved on a bronze plaque from Sohgaura inscribed in the Mauryan Brahmi script, shows a form continued at Saficl, Bodhgaya, and Mathura.
Similar pavilions appear on roughly contemporary Audumbara copper coins; these depict three varieties of shrines (all Saiva, as indicated by the presence of trisula-cum-parasu), Variety "a" (Fig. 10a) is an elaboration of the Sohgaura type, with two kapota cornices surmounted by a domical sikhara. Variety "b" has a square sikhara (Fig. 10b); Variety "c" is similar, but lacks pillarets at the grlva level (Fig. 10c).
Shrines in early epigraphy (not illustrated)
There is considerable epigraphic evidence, in fact, for the early worship of divinities in shrines. Two inscribed Garuda-dhvajas at Vidisa, one set up in honour of the "supreme god Vasudeva" by the Bhagavata Heliodorus (a Yavana ambassador from the court of the Indo-Bactrian king, Antialkidas, of Taksasila) in the 14th year of the reign of Bhagabhadra (c. 131 B.C.), the fifth Sunga king, the other erected by GautamIputra Bhagavata, the ninth Sunga ruler, in his 12th year, attest to the existence of Vaisnava shrines. The first referred to must be the elliptical shrine for which foundations have been excavated near to Heliodorus's pillar.
Three inscriptions from Nagari, District Chittorgarh, Rajasthan, refer to the con- struction of a stone wall that encloses a place for worship of Sankarsana and Vasudeva by the Bhagavata king, Sarvatata, who probably belonged to the Kanva dynasty. The site preserves a massive stone enclosure and the plinth of an elliptical brick temple.
The Nanaghat (District Poona) inscription of Naganika of the first century B.C. refers to the performance of Vedic sacrifices by the Satavahana royal family and opens with obeisance to such divinities as Dharma, Indra, Sankarsana- Vasudeva, Candra-Surya, and the Lokapalas (Yama, Varuna, Kubera, and Vasava).
An inscription from Mora (Mathura) of the reign of Mahaksatrapa Sodasa (c. A.D. 10-25) records the installation of images of the five Vrsni heroes in a stone shrine. Another Mathura inscription of the same reign, engraved on a doorjamb, records construction of a shrine, torana, and vedika at the mahasthana of Bhagavan Vasudeva. Numerous Kusana inscriptions also refer to the setting up of images of the Buddha and of Jaina Tirthankaras and to the foundation of shrines for them.
An inscription from Nandsa, District Udaipur, Rajasthan, dated A.D. 226, records performance of Vedic sacrifices following construction of shrines to Brahma, Indra, Prajapati, and Visnu.
Maurya and Post Maurya Periods: Structural Remains (Fig. 11)
From the time of Asoka Maurya (c. 272-232 B.C.) to the early Kusana period, evidences from rock-cut shrines and from surviving foundations of constructed shrines suggest that temples existed in circular (vrtta), elliptical (vrttayata), and apsidal (capakara) forms. The Ajivika caves at Barabar, District Gaya, Bihar, which contain inscriptions of Asoka and his grandson, Dasaratha, preserve both circular and elliptical hut-forms with domical or vaulted roofs. The facade of the Lomas Rsi cave replicates a large timber gavaksa-arch, supported by curved rafters (gopanasi) within an ogee-shaped frame of laminated planks, crowned by stupi-finials. The architrave of the entrance represents jalavatayana (latticed wickerwork) for the first time.
Also assignable to the Maurya period is the plinth of a circular brick-and-timber stupa-shrine that survives at Bairat, District Jaipur, Rajasthan. Enclosed by a pradaksina, the shrine was preceded by small praggriva. Temple no. 40 at Safici was originally an apsidal stone temple of the Maurya period raised on a rectangular plinth, the superstructure built of timber. Another unusual complex of four, elliptical, stone halls was unearthed at Rajgir (Rajagrha), the ancient capital of Magadha. This complex has been identified with the Buddhist Ilvakamravana-vihara, but only the foundations survive. An apsidal brick temple was also excavated at Sarnath and an elliptical brick hall formed part of the Chositararna at Kausambi, the ancient capital of Vatsa.
Structural forms prevalent during the Maurya period continued in subsequent centuries, as recorded in numerous bas-reliefs from Bharhut, Safici, Bodhgaya, Mathura, and Amaravati. The apsidal plan in this period perhaps was more popular than either the circular or elliptical plan. Buddhist cave-shrines begin to replicate complex wooden structures with apsidal ends, barrel-vaulted naves, and side aisles, but the type was not restricted solely to Buddhist use. Three astylar apsidal shrines in stone from c. the first century B.C./A.D. are known from Taxila (Taksasila). Temple no. 18 at Sanci was an apsidal shrine, and apsidal structures are shown on reliefs from Mathura (Fig. 2) and Amaravati (Fig. 1) as well. At Ramatirtharn, Sankaram, and Nagarjunakonda in Andhradesa, also, numerous apsidal temples of brick were constructed under the patronage of the later Satavahanas and Iksvakus. At Nagarjunakonda, these apsidal chapels begin to house image of the Buddha as well as stupas. An apsidal temple enshrining Siva is known from Nagarjunakonda and there are also examples of an apsidal Jaina shrine at Udayagiri, Bhuvanesvara (c. 25 B.C.), and an apsidal Naga shrine at Sonkh near Mathura (c. second century A.D.).
Nagarjunakonda, by the third or fourth century A.D., also preserves rectilinear shrines with square cellas, a type that ultimately eclipsed all other forms.
Pillar-types in this period include plain octagonal pillars with that follow a pristine timber tradition and are seen often in rock-cut shrines during the second and first centuries B.C. Later pillars develop a ghata base and a capital-type that has affinity to the Maurya bell-capital but loses the ridges of the lotus petals and becomes simply an inverted ghata. At Bharhut, pillars usually lack ghata base but show inverted lotus-capital, cable design, and large, flaring Taranga-bracket adorned adorned with criss-cross pattern (Fig. 11d.) This pillar-type may later show an amalaka set in box above the ghata, a stepped abacus, and various kinds of animal brackets. Variant forms with ghata base and plain bell-capital appear on reliefs at Sanci, occasionally surmounted by ghatapallava crowned by animals (Fig11a-b). A Bodhgaya relief shows a pillar with a ghata base and capital surmounted by a standing bull (Fig.11c).
Volume - 2
By A.D. 700, architects in Northern India had already created a unique architectural form for the temple, condensing a range of past cosmogonic and cosmological symbolisms into the rich decorative morphology of the temple itself.' This "Latina-Nagara" type of temple (see frontispiece, p. ix) combined the axis of a world-pillar, the cube of a sacrificial altar, and the body of a palace to house an image of divinity that was presented in visible form within its sanctum. Its distinctive high curved superstructure quickly dominated the North Indian skyline, from west to east and from the Himalayan foothills to the upper Deccan. Preliminary stages in the evolution of this form, as well as a variety of the alternative types and traditions of temple structures that survived in Northern India throughout its history, have been presented in the first set of volumes in this series (Vol. Il, pt. 1, "Foundations of North Indian Style").
The range of eighth- and ninth-century temples covered by this present set represents what we have designated the "Period of Early Maturity," in part following a phrase used by Stella Kramrisch for classical Indian sculpture. The Latina temple does reach maturity in this phase, its form understood and expressed in a variety of regional idioms by architects patronised by political powers intent on incorporating and marking both territory and populations within their growing hegemony. Building these temples gave merit to their individual patrons, provided a powerful tool for communities of priests, and helped both validate and perpetuate a growing "State" order, in part by helping to incorporate lineage clans within a broader social order.
Ripe and self-fulfilling in its symbolic structure, the temple in this period still required architects to work out a number of practical issues concerning how it could be used. Interior spaces along a longitudinal "axis of access" evolved from a simple sheltered portico in front of the sanctum door to a variety of sheltering halls and covered entries; circumambulatory spaces were provided first by open platforms or as paths incorporated within the fabric of the temple itself. Balconies allowed light into these ambulatory paths and halls; and screened windows sometimes were used to filter light and provide ventilation. A proliferation of niches, on the interior as well as on exterior walls, also allowed priests to elaborate increasingly complex and particularized iconographic programs for use in rituals. While something is known, however, of the histories of the hegemonic powers -Pratiharas, Cahamanas, etc. - within whose reigns these temples were built, we know very little about the priests or communities that they were built to serve.
While sharing always a symbolic potency and form, temples in the period separate into major stylistic groupings in different regions as well as showing significant and continual idiomatic variation. Central and western of North India seem to have extended to large degree the decorative and aesthetic traditions developed in territories ruled previously by Gupta dynasts; temples in Malwa, Gujarat, and southern Rajasthan instead a number of decorative conventions found previously in Satavahana, Vakataka, and Kalacuri territories in the Deccan. While little could be said in earlier Chapters about such differences in architectural terms, temples form these regions in this period of early maturity share significantly separate canons, as reflected in the following Style Outline used to Outline used to organise this volume:
Vol. II, part 1: Foundations of North Indian
I. Beginnings of North Indian Style, c. A.D. 350-660
II. Varieties of North Indian Style, c. A.D. 500-1100
Vol. II, part 2: North India, Period of Early Maturity
III. Nagara Style of Common Lineage, c. early eight-late ninth century A.D.
A. Central India
1. Gopagiri style, c. eight century A.D. Mauryas of Gopagiri and Kanyakubja
2. Dasarnadesa style,. C. late eighth-ninth century A.D. Gurjara-Pratiharas of Kanyakubja
3. Dahala style, phase 1, c. late eighth-late ninth century A.D. Kalacuris of Tripuri
4. Madhyadesa style, phase 2, c. eithth-ninth centuries A.D. Gurjara-Pratiharas of Kanyakubja
5. Jejakabhukti style, phase 1, c. late ninth-early tenth century A.D. Candellas of Kalanjara and Kharjuravahaka
Himacala style, phase 1, c. eighth-ninth centuries A.D. Hill Dynasties
C. Western India
Maha-Maru Style, phases 1 & 2, c. A.D. 700-925
1. Marudesa: Pratiharas of Mandavyapura
2. Marudesa: Pratiharas of Jabalipura and Kanyakubja, phase 1
3. Surasena janapada: Surasenas of Sripatha
4. Sapadalaksa: cahamanas of Sakambhari
5. Pratiharas of Kanyakubja and their feudatories, phase 2
Malava style, Uparamala, phases 1 & 2, c. A.D. 650-900
1. Mauryas of Medapata and Uparamala
2. Gurjara-Pratiharas and their Maurya feudatories in Malava
IV. Nagara Style of Separate Lineage, c. early eithth-late ninth century A.D.
A. Western India
1. Surastra style, c. early eighth-late ninth century A.D. Saindhavas of Bhutambilika
2. Maha-Gurjara style, phase 2, c. A.D. 700-900
a. Minor dynasties of northern Gujarat
b. Pratiharas of Jabalipura and Bhilamala
c. Capotkatas of Anahillapataka
d. Caps of Vardhamanapura
e. Samas of Kaccha
B. Eastern India
1. Vanga style, phase 1, c. A.D. 700-900: Palas
2. Kalinga style, phase 2, c. A.D. 700-900: Bhauma-Karas
Use of this outline for subsequent volume in this series can tentatively be projected as follows:
Vol. II, part 3
V. Beginnings of Medieval Idiom c. A.D. 900-1000
Vol. II, part 4
VI. High medieval period, c. A.D. 1000-1300
VII. Sultanate period, c. 14th -16th centuries A.D.
VIII. Mughal period, c. 16th-17th centuries A.D.
Volume - 3
Closely following the publication of Part III of Volume I of the Encyclopaedia of Indian Temple Architecture, I am happy to see that Part III of Volume II has been successfully completed. The American Institute of Indian Studies and its editor Shri M.A. Dhaky deserve the greatest appreciation of scholars and researchers for their perseverance in this painstaking gigantic task.
With each succeeding part of the two volumes the complexity increases. The developments of the earlier period were comparatively easier to describe and delineate. The vision, idea, and form were discernible: the symbolism, form, and function could be clearly identified.
The ancient Indian temple was a shrine situated at the confluence of rivers, mountains, and near oceans. It was a place of pilgrimage and in turn became a sacred place and provided space for a symbolic pilgrimage.
With the passage of time the Indian mind and its intellectual and artistic discourse became more complex and multilayered. The seeds of a system of establishing correspondences between the abstract and the concrete, the physical and the psychical, the temporal and the spiritual lay in early speculative though methodologies of ritual, philosophic schools, seminal notions in mathematics, and schools of medicine. However, it was early and late medieval India, in the north as well as south, east, and west, which endeavoured to give shape and substance to these notions through two different but related strategies. One was through the elaboration of mythic, the Puranic and others, through concretisation of the mythical into a structured language of narrative. It would appear that the authors and compilers of the Puranas found a language of myth and narrative to restate and interpret the revelation of the Vedas and other similar texts. The dynamics of the ecological concerns of the Vedas and the Upanisads was now couched in a descriptive language of another order. This is as evident from a re-reading of the myth of the descent of the Ganga as it is from the myth of Mahisasura-mardini or the Visnu as Varaha rescuing Prthvi or riding the Garuda. Countless myths and legends related to Siva, Visnu, and the Devi can be so interpreted at a primary level. Other layers of meaning were superimposed. Also, the Puranas endeavoured to translate the abstruse philosophic notions into simple but sometimes involved narrative. A theological layer can also be discerned. The method of the Puranas provided the tools of making multilayered statements which could be read concurrently at the purely socio-cultural, even political, level or at deeper levels of philosophic thought or even mathematical equations. An elaborate code developed which had significance for the lay and the initiated at his or her level of comprehension. The emphasis here as elsewhere was on interconnections, mutual interpenetration, and multilayered meaning which could be retrieved in different configurations. Multidimensionality was its principal attribute: simultaneity and concurrency, and reoccurrence its chief instrument.
The other equally potent instrumentality closely related to the first was the arts, specially those of architecture, sculpture, and music. Through the arts, the creator-visualiser could concretise the fundamental principles at the universal level and yet be grounded simultaneously in the reality of specific time and space. He could coalesce into a single artistic creation the spiritual and the temporal, the abstract and the concrete, the geometrical and the figurative, the micro and the macro, the unmanifest and the manifest.
It took time for these developments to attain fruition. Contours of these transitions which appear at first sight as total transformations can be traced, although difficult. However, by the tenth and the 11th centuries, the consequences are clear in architecture and sculpture. In the Puranas they can be discerned from the comparatively simpler Puranas like the Visnu, the Vayu, and the Agni to the Visnudharmottara to the later Puranas. The movement is from the simple to the complex, from one to many dimensions, from the direct to the oblique. This is perhaps the unseen but the real raison d' etre of medieval Indian temple architecture as of medieval Indian sculpture. The multiplicity of heads and arms in iconic concepts, the gradual expansion of the temple horizontally, the ascension of vertical sikhara and the rhythm of its repetitive cognate turrets (anga-sikharas, sikharikas), peaked mandapas, and sub-shrines are clear indicators of parallel articulations of the singular concern with the notion of the one and the many, the formless, and many forms, the relationship of the physical, the psychical, the temporal, and the spiritual. The beginnings of these elements can be seen in the late Gupta and early medieval architecture and its pinnacle is reached in the 11th and the 12th centuries through a vigorous formative process in the tenth century.
Ironically, and paradoxically, the king/royalty appropriate the notions and the system to make his or their royal presence. It is an interpenetration into an existing and evolving complex system. An outstanding example is the employment of the Daksinamurti myth of Siva by Rajaraja Cola. Siva's mythical abodes, forts, and temples also become palaces or like palaces as has been said. In north India, a shrine also begins to be equated with 'prasada' that is a palace, never rejecting it as a house of God. Whether the idea of treating a deity as the king or king as the divine is Indian or foreign is not so seminal as is the fact that the artistic structure was so designed that it could represent these two levels and many more simultaneously. Understandably, these temples have been interpreted by different scholars in diverse ways and viewpoints.
Concurrent and basic was the acceptance of the paradigm of the Purusa as a design concept. Although the "Purusasukta" of the Rgveda provided the seeds of a total concept, it is in the arts that it developed into a sophisticated term of reference for actual structure. Architecture evolved the Vastupurusa-mandala as a basic grid for ground plan and elevational format; embodying the sculpture, the silpa texts evolved an elaborate system of sutras, bhangas, and asanas which served as an armature. Musical theory conceived the Sangita-purusa as a basic framework of structure and drama is the Natya-purusa itself. Many articles in this volume give details of the system of correspondence between the limbs of the human body and the different architectural members of the temple structure. It is clear from a reading of these articles that the basic grid or model had the inbuilt potential of multiple expressions.
Given the two unifying unseen threads of the employment of myth as narrative and the paradigm of Purusa as a geometrical grid, there was the possibility of countless permutations and combinations-geometrical, algebraical, arithmetical, horizontal, or vertical. The model was flexible enough to allow for containing the regional, sub-regional, local, and even individual predilections. It was broad enough to have the potential of subsuming the purely ephemeral, temporal royalty, feudatory lordship, and the rest. After a lapse of time, it is the perennial symbolism, the signification which can convey multilayered meaning that sustains. Herein lies the power and the efficacy of these structures as artistic expressions. While the motivations and intentions of the human instrumentality no doubt were determinants, today it is the perennial and universal which bespeak beyond history and temporality. This is as true of the great temples in south India, Brhadisvara (Tanjavur), Sriranganathasvami (Srirangam), as it is for Kandariya Mahadeva (Khajuraho), Lingaraja (Bhuvanesvara), and other such temples at Dilwara (Mt. Abu) in north India. Understandably, Sri Aurobindo, A.K. Coomaraswamy, and Stella Kramrisch focussed attention on the symbolism, significance, and meaning not eschewing the relation of form, function, and meaning. Paul Mus did the same in respect of the 'stupa', specially Borobudur in Java, also a medieval monument in terms of chronology and history.
In contradistinction, others have endeavoured to read only political, social, and economic messages through these temples. Some amongst them have attributed school, style, and technique to purely socio-economic and political conditions, and patron. While the latter were undoubtedly primary factors for creativity and specially, in the case of temples where surplus funding was essential, they cannot be in the very nature of creativity and the flow of a living tradition be the sole determinants of the final artistic product and characteristic style. The poet's, the painter's, the architect's, the sculptor's, the musician's, the dancer's, and the dramatist's skill lies in his ability to be contemporary without being ephemeral or purely local or time bound. In the case of medieval Indian architecture and its magnificent monuments, the sthapatis and the craftsmen appear to have taken the challenge of meeting dynastic desires and yet fulfilling a higher and more lasting purpose of creating a cosmos on earth with its multiplicity of the vegetative, animal, human, and divine all fusing into the one unknowable presence of the One. Even the particularity of the icon loses meaning. What remains as residual experience is fullness of the empty space of the garbhagrha where dualities are merged and lost.
Each one of the temples-through their regional schools and sub-schools, and clearly distinguishable styles and differences in superstructure-are Meru or Mahameru, They interconnect earth and heaven, move from the outer to the inner, the gross to the subtle, the mundane to the sacred, the physical to the spiritual. It is a journey from multiplicity and complexity to oneness and unity.
The chapters in the Encyclopaedia unfold the many rainbow colours, almost like the sancari bhavas (transitory states) in a grand medieval drama of Indian temple through the differences in schools and styles conditioned by dynasty and patronage. Material, forms, and techniques differ but, at the level of vision and essence, there is unity. As in the field of music and dance, where there are distinctive sampradayas and gharanas, and the ragas and compositional pattern are diverse, so also in the medieval temples: but the ultimate relish of rasa, a state of wonder (adbhuta) they evoke, is identical. Evoking the experience of being elevated to a higher level of consciousness through the monument and the ritual is the final goal. The temple structure itself, like a grand Indian musical composition, adopts a basic grid like an octave with its ascending and descending notes (its sthayi, antara, pallavi, anupallavi, and caranam) and then enlarges the built structure vertically and horizontally through repetitive motifs to create the effect of a single cascade or mood.
For the researcher and the scholar it is also an essential prerequisite to understand the 'what' and the 'how' before asking the 'why'. The volumes lay bare the structure, design, and the nuances of differentiation. Here is a mine of information and knowledge assiduously compiled by the senior scholars Krishna Deva, M.A. Dhaky, and Michael Meister. As a reference work this Volume along with others will serve, metaphorically speaking, the mountains of encyclopaedic knowledge which must be traversed before reaching the pinnacle of experience. I would like to record my, and of others like me, admiration for this diligent scholarship. I would again like to congratulate Dr. (Prof.) Frederick Asher, President of the American Institute of Indian Studies (AIIS) and its Vice-President Dr. Pradeep Mehendiratta for their direction and guidance, and to Prof. M.A. Dhaky for his exemplary scholarship and deftness as editor.
I am happy that Indira Gandhi National Centre for the Arts (IGNCA) could join the AIIS as a co-publisher in bringing out this volume. Through our combined efforts as co-publishers (and otherwise), I hope Indian art history will slowly but surely take a new turn of eschewing the 'either or' approach, and, instead, embrace both.
Some twenty years ago, a skeptical officer at the agency that has provided by far more financial support for the American Institute of Indian Studies' Center for Art and Archaeology than any other agency, pronounced with certainty that not a single volume of the Encyclopaedia of Indian Temple Architecture ever would be published. At the time, he may have had reason for doubt. But, with this volume, the project is very nearly complete. Just two more volumes and perhaps the most invaluable resource of all, the glossary volumes, remain, and they soon will be ready for press as well.
What has happened during these twenty years to change the outlook for a resource that is now widely acclaimed in many countries? Several things, mostly a series of highly effective partnerships. First, the Center became part of the American Institute of Indian Studies, where the administrative work could be shared with an extraordinarily capable and experienced new set of colleagues under the direction of Pradeep Mehendiratta, the Institute's Director-General. This left scholars to deal with scholarship. Second, the scholars themselves gave untold hours-no, even years-to the project. At the head of the list, we need to acknowledge M.A. Dhaky, who has devoted what must be described as a lifetime to the Encyclopaedia, all the while maintaining his personal scholarship, to say nothing of friendships around the world. Michael Meister provided a close editor's hand, substantial original writing, and a wise voice in shaping Pramod Chandra's original vision. Many others, too, have participated in this project. The numbers go well beyond the scholars responsible for each chapter. Four in particular merit special mention V.K. Venkatavaradhan, whose devotion to the project has insured letter-perfect text, and three draftsmen- S. Dorai, S. Pandian, and N. Ravi who have set the standard for architectural drawing in India: And Dharmapal Nanda' s complementary share at the same level of excellence on the score of photography.
Two other partnerships need acknowledgment. Following publication of the first volume, the doomsayers were proved wrong, and the Smithsonian Institution became a major sustainer of the project. Without their financial support, the project would have failed long ago. We remember fondly the late Kennedy Schmertz, whose initial support provided the much-needed expression of confidence. Gretchen Gayle Ellsworth followed him with similar support for the project. And finally we recognize with gratitude Francine Berkowitz's support for the Center for Art and Archaeology and its leading project, this Encyclopaedia. Rarely has there been a representative of a funding agency who as often appears to represent the Institute as the agency that actually employs her. Finally, the Indira Gandhi National Centre for the Arts has provided generous support for the publication of several of these volumes. We acknowledge with gratitude Kapila Vatsyayan's confidence in the Institute and this project.
The Part 3 in the sequence takes further the publication programme for the Volume II of the Encyclopaedia of Indian Temple Architecture series of this Institute and in size is comparable to the preceding two parts. The buildings discussed here were founded in the tenth century, a period which represents the beginning of the medieval architectural styles in north India (and in parts of Pakistan), that came on the scene before the advent of the High Medieval which began from the 11th century. The building activity in the tenth century was underway in the domains of several different regional and imperial dynasties. Buildings, in a few cases, were erected by the rulers themselves, some by their vassals, provincial governors, wealthy and powerful generals, and other officers such as those on ministerial posts, also treasury officers, next the opulent merchants, and, no less, a few were founded by the heads of different religious sects, particularly the Saivaite pontiffs and Jaina abbots.
The Sanskrit terms used here have been extracted from the various medieval vastu-works of central and western India as well as Orissa and its bordering tracts in eastern India. This Part has 207 drawings and 20 site-maps in the first (text) bind, and 913 black-and-white illustrations in the second bind.
The history of this Project has not been touched upon in any of the previous parts; hence is briefly outlined here. A personal project which was destined to be a seed for this Project, was initiated late in the year 1966 at the American Academy of Benares, Varanasi (after 1970, American Institute of Indian Studies), with a modest title and limited scope, as The Dictionary of Indian Architectural Terms. Its tentative scheme I then had worked out for approval and had discussed with Dr. Pramod Chandra, the initiator and the first Director of the American Academy of Benares. (Dr. Chandra had already explored the idea with the Indian Advisory Committee of the American Academy of Benares whose members had fully supported it.) In the following year, after my discussions with K.R. Srinivasan in Madras (Chennai) and Krishna Deva in Delhi, its scope was enlarged and the Project then was promoted as the Encyclopaedia of Indian Temple Architecture with essay volumes discussing regional and period styles where the appropriate technical terms extracted from the relevant vastu-texts in Sanskrit were envisaged to be employed in the descriptions of the buildings, reflecting as they supposedly do the perception and language of the ancient/medieval architects and sculptors involved in their construction. The annotated and illustrated glossary of terms was envisaged to follow the essay Parts.
While promoting this modified and enlarged version of the Project, Prof. Chandra enlisted the support of the Smithsonian Institution in the U.S., as also of the academics connected there with the studies in the field of Indian art and architecture. The schema of the volumes then contemplated and next rigorously followed was based on three co-ordinates-region's political/historical background, socio-religious context, and the temporal bracket involved, in short within the perspective of space-time-cultural continuum. The classificatory organization of styles thus resulted is somewhat analogous to that involving phyla, genera, and species in biology which are set within the broad geographic areas and geological times. Likewise, in this instance, the architectural styles here are viewed in a continuous process of formal and stylistic evolution as they perceptibly reveal. The scheme was sufficiently flexible to accommodate new entries of the fresh stylistic groups and it can, wherever and whenever felt necessary, be modified when fresh discoveries/researches in certain regions and temporal areas so demanded or warranted. (In point of fact, the schema of each Part had been more than once revised, enlarged, and remodelled in response to the exigencies of new archaeological findings including the fresh epigraphical discoveries necessitating the modification of history and the chronology of events as well as of buildings etc., etc.) Based on the premises of the strongly marked stylistic differences, the series of publication was next designed to have two volumes, Volume I to cover the South Indian buildings and Volume II to discuss those of North India.
As in the life of an individual, so in the history of an institution, visitations of ups and downs were inevitable; the Project, as a result, sometimes received serious jolts, temporary stagnation, and even at times was threatened to be totally abandoned. However, the Varanasi Center, and along with it the Project, thanks to the efforts of the past Presidents, particularly Prof. Edward Dimock Jr., and the Vice-President of AIIS and the Director-General of its Indian operations in Delhi, Dr. Pradeep Mehendiratta, has survived and for the past two decades has been progressing at a fairly consistent pace; its productions-despite a few hostile, partisan, venomous, and motivated reviews (as against a bulk of appreciative ones) they received-have been welcomed. In point of fact, the wide use made of the volumes by several researchers in the field the world over, bears testimony to their usefulness and hence the validity of the Project. (The earlier Parts of the volumes are now largely out of stock.)
Strong reservations on the usage of Sanskrit terms had been voiced in some quarters for describing the buildings in the first Parts when they became available in print. However, we have parallels for the usage of the characteristic terminology of a given land in other fields, for instance the employment of the Arabic and Persian terms for Islamic architecture, Greek and Latin jargon for the Hellenic and Roman architecture, French, German, and Spanish terms in the domain of Romanesque and Gothic architecture, Chinese terms in understanding the Chinese paintings and philosophy: And, in India too, in the realms of Indian philosophy, literature, and Indology in general, several Sanskrit (and Pali-Prakrit) terms are used. Why, then, the opposition in the field of Indian temple architecture? Though there is some reason for such resistence in regard to the EITA volumes, this was due more to the excessive use by a few contributors not only of the genuine Sanskrit terms but also of Sanskrit words for which common English words were readily available. For instance pillar or column (and not 'stambha'), doorsill (and not 'udumbara') etc., should have been used in lieu of the Sanskrit words/terms shown here in parenthesis. (In India, of course, such Sanskrit words are widely known and readily understood.) Keeping the afore noted criticism in view, I have, for this volume, as far as was possible within the intentions of the contributing scholars, converted such terms into their equivalent English words. But the total negation of the usage of Sanskrit terms is uncalled for, because, for many architectural details, the only terms available are in Sanskrit. Some of these in origin denote their functional/locational aspects, some are metaphorical and hint to their remoter origins as well as meanings inherent in the formal shapes implied, and the associated decorative details in the total cultural context of ancient and medieval India.
As I glance through the scholarly writings of the new generation in the U.S., U.K., and Germany, researchers like Thomas E. Donaldson, Michael D. Rabe, Phillip B. Wagoner, Walter Smith, Adam Hardy, Falk Reitz (besides my friend of long standing Michael Meister), there is encountered a sizeable sprinkling of Sanskrit terms. These scholars have mastered the Sanskrit jargon to an appreciable degree. Of course, discretion has to be maintained on not unduly mixing up the jargon specific to the ancient and medieval periods, or of northern and southern India; and even in north India that of the eastern provinces and of the rest of the northern regions. (Earlier writers on Indian temple architecture have introduced confusions and hence erred exactly on this score.) Also to be avoided, as far as possible, is the usage of terms of the modern Middle Indo-Aryan or Dravidian languages-Oriya, Gujarati/Rajasthani, Tamil-e-for it is advisable strictly to employ the Sanskrit terms alone for the national as well as international usage. But these are the problems which will be tackled and discussed at some length in Part 5 of each Volume.
The present Part deals with the dawn of the medieval architecture whose onset took place precisely at the beginning of the tenth century. Medievalism in north India simultaneously manifested itself in all regions and in almost all fields of cultural expressions-literature in its manifold modes, elaborate ritualistic forms of worship, costumes and ornaments, music and dance, and above all art and architecture, converging as they all did to the same ideals, and exhibited identical inward spirit and external unity in formal appearance. Architecture, by an assured, as though pre-determined, course of evolution, by then had left behind the relative simplicity, undue massiveness, ponderousness of appearance met with in the earlier ages. The medieval era ushered in the first definite stage of creating fully organic, highly integrated, and convincingly articulate appearance of a "temple" as it will look, or should look, as Gothic era contemporaneously was to do in the medieval Europe for the "church".
The temple's lower structure and superstructure, their highly organized, stratified, and moulded surfaces in vertical rise, and their structural spread along the horizontal plane for the first time attained a well-balanced, logical, wholesome and handsomely manifest, and truly architectonic image. The figural presence and decorative art of the temples (both on the prasada and the adjoined hall) now blended harmoniously with the moulded surfaces, so much so that they cannot be separated from the total rhythm of form and features. The supple torsos, the elegant body bends, the nicely formed faces, and the fine jewellary of the associated divine and semi-divine figures, and no less the humans, wherever appearing, were depicted as engaged in the manifold activities of life including worship as well as love: They added worldly colour together with otherworldly intentions. And their specific placement on the exterior of the temple-body clarified their functions as well as the iconological import of their association even when they were progressively losing the contemplative and serene looks of the classical, post-classical, and even pre-medieval times. For the first time the prasada looked as Purusa, embodiment of Eternal Man. as well as the configuration of the total Cosmos, a perfected concrete symbol of the total existence.
Some of the greatest masterpieces of temple architecture, a few among these also possessing pretension of scale, were created in this age. True, there was still not that outburst of prolific building activity, as it will be from the beginning of the 11th century onwards till the early phase of late Medieval centuries, in the latter times it particularly happened in western India. However, much of this period's material contribution has been destroyed by the ravages of time as well as the vagaries of man. And yet, selecting from the surviving buildings, we may mention, in passing, the Laksmana and Visvanatha temples in Khajuraho, the ruined temples at Kotai and Kerakot in Kaccha, and the Muktesvara and Gauri temples at Bhuvanesvara as representing the finest gems of this early phase of medieval architecture. There will, in the next phase, be built grander edifices, notable for their more evolved form and hence even more cogent appearance. But, qualitatively, the level of excellence of form and decor and their happiest marriage visibly manifest and joyously celebrated here, in the tenth century, was not to be duplicated afterwards.
Send as free online greeting card
Email a Friend