Following River Routes and Artistic Transmissions in Medieval Central India

Notes

1. This account is glossed from translations of Ibn Baṭṭūṭa’s Riḥla by Samuel Lee, trans., The Travels of Ibn Batūta (London: Oriental Translation Committee, 1829), chap. 16–17; and by H. A. R. Gibb and Charles Beckingham, trans., The Travels of Ibn Baṭṭūṭa (A.D. 1325–1354), vols. 3–4 (Cambridge: London: The Hakluyt Society, 1971/1994); and H. A. R. Gibb, trans., Travels in Asia and Africa, 1325–1354, (London: Routledge and Kegan Paul, 1929; reprint, 1957).

2. Ibn Baṭṭūṭa notes that “the lieutenancy of Dawlatābād extends through a distance of three months.” See Lee, The Travels of Ibn Batūta, 162.

3. For a brief sampling, see Richard M Eaton, A Social History of the Deccan, 1300–1761: Eight Indian Lives (Cambridge University Press, 2005); Irfan Habib, Economic History of Medieval India, 1200–1500 (New Delhi: Pearson Education India, 2011); Mehrdad Shokoohy, Muslim Architecture of South India: The Sultanate of Ma’bar and the Traditions of the Maritime Settlers on the Malabar and Coromandel Coasts (Tamil Nadu, Kerala and Goa) (London: Routledge, 2003); Andre Wink, Al Hind: The Making of the Indo-Islamic World, vol. 2, The Slave Kings and Islamic Conquest in the 11th–13th Centuries (Leiden: Brill, 1997); Peter Jackson, The Delhi Sultanate: A Political and Military History (Cambridge: Cambridge University Press, 1999); Catherine B. Asher and Cynthia Talbot, India Before Europe (Cambridge: Cambridge University Press, 2006).

4. For studies on Gwalior and Chanderī, see Kalyan Kumar Chakravarty, Gwalior Fort: Art, Culture, and History (New Delhi: Arnold-Heinemann, 1984); Gérard Fussman et al., Naissance et déclin d’une qasba: Chanderī du Xe au XVIIIe siècle, vol. 2 (Paris: Boccard, 2003).

5. The term is transliterated as Kajarrā in Gibb and Beckingham, The Travels, 790, and as Kajwarā by Lee in The Travels, 162.

6. Although we do not have textual accounts confirming Kadwāhā’s official status as a ribāṭ, the archaeological context seems to fit such an assignation. I have treated this issue more extensively in a previous publication. See Tamara I. Sears, “Fortified Maṭhas and Fortress Mosques: The Transformation and Reuse of Hindu Monastic Sites in the Thirteenth and Fourteenth Centuries,” Archives of Asian Art 59 (2009), 25–26.

7. See Annette S. Bevridge, trans., The Babur-Nama in English (London: 1922), 590–91; W. M. Thackston, trans., The Baburnama: Memoirs of Babur, Prince and Emperor (Washington, DC: Freer Gallery of Art and Arthur M. Sackler Gallery, Smithsonian Institution, 1996). While Beveridge reads “Kachwa,” Thackston reads “Kachwaha.” Beveridge also noted the possible correspondence between Kachwa and Kadwāhā (p. 591).

8. Although not explicitly recorded in the extant corpus of earlier travelers’ reports, this particular central Indian route continued to be used during the Mughal era, as attested to by the writings of other foreign travelers, including Father Antonio Monserrate (1579–82), Niccolao Manucci (1655–56), and Peter Mundy (1632–33). See, for example, Michael Fisher, Visions of Mughal India: An Anthology of European Travel Writing (London and New York: I. B. Tauris, 2007), 42–46, 121–24; R. C. Temple, ed., The Travels of Peter Mundy, in Europe and Asia, 1608–1667, vol. 2, Travels in Asia, 1628–1634 (London: Hakluyt Society, 1914). See also Dilip Chakrabarti, The Archaeology of the Deccan Routes: The Ancient Routes from the Ganga Plain to the Deccan (New Delhi: Munshiram Manoharlal Publishers, 2005), 123–43.

9. While a full analysis has yet to be published, my preliminary field surveys have revealed many additional moments of building, extending from the fifteenth century through the colonial era and into the present day.

10. According to the 2011 Census of India, the population included 819 households with a total of 4,572 people. It is useful to note that the population has increased significantly over the past century. The population was recorded as 826 (433 males and 393 females) in 1901, in Charles Eckford Luard’s Central India state gazetteer series, vol. 1 (Calcutta: Superintendent Government Printing, 1908), 247.

11. Although not widely known among scholars, Bījāsan Devī occupies a position of regional prominence in central India and particularly in areas north of the Narmadā river. One of the more famous temples associated with this goddess is on a hilltop in Salkanpur, which is approximately 58 miles south of Bhopal. The temple at Kadwāhā appears, on the basis of style, to date to the late eighteenth or early nineteenth century.

12. I mapped a number of fragments while conducting fieldwork in the fall of 2012.

13. Madhusudan Dhaky and Michael W. Meister, eds., Encyclopaedia of Indian Temple Architecture (EITA) (New Delhi: American Institute of Indian Studies, 1998), vol. 2, pt. 3, 21–27.

14. The temple itself was likely buried at one point under a platform that once supported a small mosque built in the early fourteenth century, when the gaḍhī surrounding a Śaiva core was transformed into an Islamic ribāṭ. On the chronology of the fortress and ribāṭ, see Sears, “Fortified Maṭhas.” On the dating of the monastery, see Tamara I. Sears, Worldly Gurus and Spiritual Kings: Architecture and Asceticism in Medieval India (New Haven and London: Yale University Press, 2014), chap. 3.

15. R. N. Misra, “Religion in a Disorganized Milieu,” in Organizational and Institutional Aspects of Indian Religious Movements, ed. Joseph O’Connell (New Delhi and Shimla: Manohar and the Indian Institute of Advanced Study, 1999). Misra’s arguments are rooted in several decades of fieldwork, during which he noticed the high number of abandoned villages in the archaeological record and other evidence suggesting the relative itinerancy of the region’s population in antiquity (p. 72, n. 11). For an extended discussion of the relationship between jānapadas and the relationship between inhabited and wilderness spaces during this particular period in northern and central India, see Brajadulal Chattopadhyaya, “Historiography, History and Religious Centers: Early Medieval North India, circa AD 700–1200,” in Gods, Guardians, and Lovers: Temple Sculptures from North India, A.D. 700–1200, ed. Vishakha Desai and Darielle Mason (New York and Seattle: Asia Society Galleries in association with University of Washington Press, 1993), 36–46.

16. For more on geography, see Michael Willis, “An Introduction to the Historical Geography of Gopakṣetra, Daśārṇa, and Jejākadeśa,” Bulletin of the School of Oriental and African Studies, University of London 51, no. 2 (January 1, 1988), 271–78; Michael Willis, Temples of Gopakṣetra (London: The British Museum, 1997), 16–18; and Pranab Kumar Bhattacharyya, Historical Geography of Madhya Pradesh from Early Records (Delhi: Motilal Banarsidass, 1977).

17. At their point of junction with the larger Yamunā in the north, the two rivers remain separated by more than 100 kilometers, and their paths take them on a winding parallel journey that is separated most of the way by distances of anywhere between 50 and 150 kilometers.

18. The route along the Sindhu and its tributaries appear to have served as a major throughway centuries later for Ibn Baṭṭūṭa in his passage from Gwalior to Narwar and then onward to Chanderī, which is located directly along the Betwā River, just 25–30 kilometers southeast of Kadwāhā. While the route along the Betwā has been noted in the past, most recently by Dilip Chakrabarti, scholars seem to have missed the importance of the Sindhu, Mahuār, and Ahīrāvati for understanding routes of travel and the connection between places, in part because sites such as Kadwāhā remain largely understudied. Although Chakrabarti notes the routes along the Betwā and Chambal Rivers, he omits the Sindhu and its tributaries from his discussion of ways to and from Pawāyā. See Dilip Chakrabarti, The Archaeology of the Deccan Routes: The Ancient Routes from the Ganga Plain to the Deccan (New Delhi: Munshiram Manoharlal Publishers, 2005), 94–101, esp. 104–5.

19. For more on Padmāvatī and the nāgas, see H. V. Trivedi, Catalogue of the Coins of the Nāga Kings of Padmāvatī (Gwalior: Dept. of Archaeology & Museums, Govt. of Madhya Pradesh, 1957); Gwalior Museum, Madhya Pradesh (India), and Directorate of Archaeology & Museums, Terracottas of Pawaya: Central Archaeological Museum, Gujari Mahal, Gwalior. ([Bhopal]: Commissioner, Archaeology & Museums, Madhya Pradesh, 1990).

20. S. Sankaranarayanan and G. Bhattacharya, “Mahuā Inscription of Vatsaraja” Epigraphia Indica 37 (1967), 53–55.The inscription is affixed to the maṇḍapa (portico of Śiva temple 1, otherwise known as the Dhūrjaṭi Śiva temple, at Māhua, which is located approximately ten kilometers northeast of Kadwāhā. This Vatsarāja should not be confused with the later Vatsarāja (circa 777–808) of the Gurjara Pratīhāras.

21. This aspect of the inscription has led some scholars to conclude that Mahuā, at the very least, was “no provincial backwater” but rather fell solidly “on the main line of communication between Kānyakubja and Vidiśa” (Willis, Temples, 17).

22. V. V. Mirashi was the first to note that the Madhumatī River is mentioned in Bhavabhūti’s Mālatīmādhava; see his essay, “The Śaiva Ācāryas of the Mattamayūra Clan,” Indian Historical Quarterly 26 (1950), 6, n. 10. On Bhavabhūti’s place of origin, see V. V. Mirashi, Bhavabhutī (1974, rpt. Delhi: Motilal Banarsidass, 1996), 32–44.

23. In that play, Saudāminī travels to Padmāvatī from the sacred center of Śrisailam in modern-day Andhra Pradesh; in the process, she describes the flowing waters at the confluence of the Madhumatī and Sindhu Rivers. The description occurs at the opening of act 9. See Bhavabhūti, Bhavabhūti’s Mālatīmādhava, with the Commentary of Jagaddhara, ed. and trans. M. R. Kale (Bombay: Oriental Publishing, 1908), 79–80.

24. Michael Willis, Inscriptions of Gopakṣetra: Materials for the History of Central India (London: British Museum, 1996), 2.

25. R. N. Misra, “Śaiva Siddhānta,” in Religious Movements and Institutions in Medieval India, ed. J. S. Grewal (New Delhi: Oxford University Press, 2006), 54, 64, n. 77. For further connections to Gwalior, see Willis, Inscriptions, 2, n. 7, 109, and Arvind K. Singh, “Lintel Inscriptions from Gwalior,” Journal of History and Social Sciences 4, no 1 (July–December 2012).

26. F. Kielhorn, “Two Inscriptions from Terahi, [Vikrama-] Samvat 960,” Indian Antiquary 17 (1888), 201–2.

27. An edition of the inscription and a description of its contents can be found in F. Kielhorn, “Siyadoni Stone Inscription,” Epigraphia Indica 1 (1889), 162–79.

28. The term mahārājādhirāja is used in verse 22 of the Siyaḍoṇī inscription, and the term nr̥pacakravarttī is found in the fragmentary inscription from Kadwāhā. For more on imperial titles, see B. N. Puri, History of Indian Administration, vol. 2, Medieval Period (Bombay: Bharatiya Vidya Bhavan, 1975), pp. 2–4; D. C. Sircar Indian Epigraphy (Delhi: Motilal Banarasidass, 1965), 330–51.

29. V. V. Mirashi and A. M. Shastri, “A Fragmentary Stone Inscription from Kadwaha,” Epigraphia Indica 37 (1968), 117–24.

30. Unfortunately, the inscription is badly damaged on the right side. The translated text is largely from the surviving portions of lines 21, 22, and 23, which provide the second halves of verses 24 and 26 and the middle portion of verse 27. Unfortunately we have lost the entirety of verse 25, which may have contained more information about the origin of the dynasty. The Sanskrit text that can be deciphered was read by Mirashi and Shastri, “A Fragmentary Stone,” as: ... lajṭaḥ sa tasyānucarottamaṃ[m] / āyuṣmānnr̥patiḥ koyamiti papraccha saṃyamī // [verse 24] ... ndugauraṃ guṇināṃ pra[sū]tigottaṃ pratīhāramahīśvarāṇāṃ[m] / yasminn abhūd durbhaṭa i[ty upākhyaḥ] [verse 26] ... ppamā[?] ṇdiśikhaṇḍaratnavalayavyāsaktapādāmvu[mbu]jaḥ / Garjjadgūrjjarameghacaṇḍa ... [verse 27].

31. The best and most comprehensive discussion of royal titles and epithets is still found in D. C. Sircar’s classic study, Indian Epigraphy, 330–51.

32. These are all cataloged in Willis, Inscriptions, 112–13. See also Mirashi and Shastri, “A Fragmentary,” and Arvind Kumar Singh, “Inscriptions from Kadwaha, Dist. Guna (M.P.),” Prāgdhārā, Journal of the U.P. State Archaeological Organisation 10 (2000), 229–39.

33. Willis, Temples, 17. Although the temples at Baktar are unpublished, the rest are catalogued in EITA 2, no. 3, 3–61; R. Trivedi, Temples of the Pratīhāra Period in Central India (New Delhi: Archaeological Survey of India), 67–70, 95–96, 121–25; and Willis, Temples, 36–41, 59–60, 68, 73, 79–81 (pls. 8–12), 93–96, 100–103, 118–19, 141–42.

34. The same may have been also true of Thubon.

35. Here I deviate from Krishna Deva’s general assessment that Kadwaha’s surviving temples from the tenth century “pertain to a cognate style” (EITA 2, no. 3, 17). His brief survey of the site as a whole contains a number of inconsistencies. He identified first ten surviving temples from the tenth century, and then later eleven temples (EITA 2, no. 3, 17), and he missed at least two temples that should have been included in the set.

36. This point is discussed further in Sears, Worldly Gurus, 158–67. See also F. Kielhorn, “A Stone Inscription from Ranod (Narod),” Epigraphia Indica 1 (1892), 351–61.

37. The active involvement of local individuals in the construction of temples in the vicinity of Kadwaha can be seen again in the inscription from Siyaḍoṇī noted above (Kielhorn, “Siyadoni Stone Inscription”). According to the inscription, more than two dozen individuals collaborated in the construction and sustenance of a local Viṣṇu temple during the tenth century. Although the local chief Undabhaṭṭa is named as one of the donors, the vast majority of the temple’s foundation and maintenance was provided for by local craftsmen and merchants. From the inscription, we learn that the temple stood on a field donated to Viṣṇu by the entire town; that it was set up initially by a merchant named Canduka, who also donated property to the temple as an endowment. Other merchants gave endowments; some donated houses, and others gave goods procured from potters, sweet makers, liquor distillers, oil millers, stone carvers, and betel-nut sellers.

38. Adam Hardy, The Temple Architecture of India (Chichester, UK: Wiley, 2007); Hardy, “Indian Temple Typologies,” Glimpses of Indian History and Art: Reflections on the Past, Perspectives for the Future, ed. Tiziana Lorenzetti and Fabio Scialpi (Sapienza Universita Editrice, 2012), 101–25 (esp. 104–6).

39. See, for example, Phillip B. Wagoner, “Modal Marking of Temple Types in Kakatiya Andhra: Towards a Theory of Decorum for Indian Temple Architecture,” in Syllables of Sky: Studies in South Indian Civilization in Honor of Velcheru Narayana Rao, ed. David Shulman (New Delhi: Oxford University Press, 1996), 431–72. See also Michael W. Meister, “Style and Idiom in the Art of Uparāmala,” Muqarnas 10 (1993), 346; Ajay Sinha, Imagining Architects: Creativity in the Religious Monuments of India (Newark, DE: University of Delaware Press, 2000).

40. On style and idiom in Indian architecture, see Meister, “Style and Idiom;” Meister, “Art-Regions and Modern Rajasthan,” in The Idea of Rajasthan: Explorations in Regional Identity, ed. Karine Schomer, Deryck Lodrick, Joan Erdman, and Loyd Rudolph, vol. 1, 143–76 (New Delhi: Manohar, 1994; Darielle Mason, “A Sense of Time and Place: Style and Architectural Disposition of Images on the North Indian Temple,” in Gods, Guardians, and Lovers,” 116–38.

41. On bhūmija, see Krishna Deva, “Bhūmija Temples,” in Studies in Indian Temple Architecture, ed. Pramod Chandra, 90–113 (New Delhi: American Institute of Indian Studies, 1975).

42. On phāṁsanā, see Michael W. Meister, “Phāṁsanā in Western India,” Artibus Asiae 38, nos. 2/3 (January 1, 1976), 167–88. On śekharī, see Adam Hardy, “Śekharī Temples,” Artibus Asiae 62, no. 1 (January 1, 2002), 81–137.

43. Although one might be tempted to see such a distinction as modal, I would argue that it falls under the category of “style,” since it represents a single, primarily surface-level intervention that is applied to temples built in different modes. Unlike the formation of a new mode, such as bhūmija or śekharī, it does not constitute a systematic transformation of the temple’s frame as a whole.

44. In both cases, a recess was inserted between the kapilīs (threshold projection) and karṇas to create a clear distinction between the wall and the vestibule.

45. The ninth-century practice of inserting recesses between each projection can be seen at sites such as Terahi, located less than ten kilometers to the northeast, and farther afield at Gwalior, as seen in the Chaturbhuj temple (circa 875 CE). Trivedi, Temples of the Pratīhāra Period, 14–16, 121–125, 129–132; EITA 2, no. 2, 37–38, 55–58.

46. Such may be the case, for example, with the Śiva temple that accompanies the monastery. It is unfortunate that a proper relative chronology of Kadwāhā’s temples has yet to be developed, and only a few temples at Kadwāhā have been dated with any degree of precision. The Caṇḍāl Maṭha has been placed in the late ninth century (EITA 2, no. 2). In his entries for EITA (2, no. 3, 21–27), Krishna Deva proposed a “late tenth century” date for temple 1 (Śaiva) of the Pachalīvālā group and for both of the temples of the Khirnīvālā group. A more general date of “tenth century” was given to all three temples of the Akhatiavāla group, the two temples of the Nahalvār group, the temple accompanying the monastery, and temple 2 (Vaiṣṇava) of the Pachalīvālā group. Three temples—the Eklāvālā temple and both temples of the Morāyat group—were excluded from EITA, which, for northern India, currently run only up to 1000 CE. Presumably, their absence indicates that Krishna Deva may have envisioned them as belonging to the eleventh century. By contrast, Michael Willis has suggested a date of the first quarter of the tenth century for the smaller temple next to the Toteśvara Mahādeva in the Morāyat group (Willis, Temples of Gopakṣetra, 80). In addition, Michael Meister, in passing, has placed the Viṣṇu temple in the Pachalīvālā group around circa 1000–1050, in contrast to Krishna Deva’s later date of the tenth century. See Michael W. Meister, “Regional Variations in Mātṛkā Conventions,” Artibus Asiae 47, nos. 3/4 (January 1, 1986), 233–62; fig. 9.

47. EITA 2, no. 2, p. 124.

48. See above.

49. My assessment dovetails with M. A. Dhaky’s analysis of the dissolution of older “Daśārṇa” elements from the temples of Khajurāho and further east in Ḍāhala-deśa during the tenth century. See Madhusudan Dhaky, “Genesis and Development of Māru-Gurjara Architecture,” in Studies in Indian Temple Architecture, ed. Pramod Chandra, 114–165 (New Delhi: American Institute of Indian Studies, 1975), 151. Other examples can be located in the Kacchapaghāta territories further north, in the Gwalior region, as seen in the Kakanmaṭh temple at Sihoniya and the Kacchapaghāta-era temples at Batesar and Naresar. Dhaky’s suggestion, in passing, that the link may have been through the Cāhamānas is probably correct; another factor may have been the proximity to Uparāmala. For a few sample comparisons in western India, see Ambikā temple at Jagat, the Śiveśvara and Takṣakeśvara temples at Ēkaliṅgji, the Mahāvīra and Sachiyāmātā temple complexes at Osian, and the Ghaṭeśvara temple at Barolī. EITA 2, no. 3, pls. 392, 421–22, 426, 709–10, 722–23.

50. On the Mahā-Gurjara temple style, see Madhusudan Dhaky, “Genesis and Development,” 146–51; M. A. Dhaky, “The Chronology of the Solanki Temples of Gujarat,” Journal of the Madhya Pradesh Itihasa Parishad 3 (1961), 1–83. The molding sequence here also corresponds roughly to what is seen at the Udayesvara temple at Udayapur.

51. Hardy, “Śekharī Temples,” 94 n. 38.

52. On Kacchapaghāta temples, see Michael Willis, “Architecture in Central India under the Kacchapaghāta Rulers.” South Asian Studies 12, no. 1 (1996), 13–32.

53. For an extremely cogent discussion of the relationship between sculpted representations of different modalities and the knowledge to realize them in built form, see Madhusudan A. Dhaky, The Indian Temple Forms in KarnÚātÚa Inscriptions and Architecture (New Delhi: Abhinav Publications, 1977).

54. On the relationships among mode, style, and idiom, see particularly Wagoner, “Modal Marking,” and Meister, “Style and Idiom.”

55. I recorded at least two such fragments, excavated by the Archaeological Survey of India from the vicinity of the Toteśvara Mahādeva temple, while conducting fieldwork in 2001–2 and again in 2012.

56. For example, compare the lower half of the arch with a diagram provided by Parul Pandya Dhar in The Toraṇa in Indian and Southeast Asian Architecture (New Delhi: D. K. Printworld, 2010), p. 32, fig. D.7a.

57. Cynthia Packert Atherton, The Sculpture of Early Medieval Rajasthan (Leiden: Brill, 1997), 56–59, pl. 59; EITA 2, no. 2, 254–59, pls. 569–77. The version seen at Khed is more ornate; it contains decorative bands between the illikā-toraṇa base and the flaming upper course.

58. A number of good illustrations of this type of niche can be found in Dhār, The Toraṇa, pp. 158–71.

59. On the evolution of the gavākṣa, see Hardy, The Temple Architecture of India, 63–82. Although Hardy does not discuss this form specifically in this chapter, the flaming niche seems to emerge from similar patterns of combination, abstraction, and evolution.

60. References to tapas can be found in verses 7, 10, and 14 of the Kadwāhā inscription.

61. Kadwāhā inscription, verse 16. What can be read of the Sanskrit is as follows: tasyāvagamya sa kathāṃ karuṇāvimuktav[b]āḥpaḥ kṣa[ṇaṃ] tadanu kopavipāṭa[lākṣaḥ /] – – ∪ – ∪∪∪ – ∪∪ – ∪ – – – ∪ – ∪∪∪ – [14] jalada[s sa]lakṣmī[ḥ] //

62. Paraphrased from verses 17–18. Unfortunately these verses fall in a fragmented section of the inscription and cannot be read in full. What can be read of the Sanskrit is as follows: atha prabhāvāgatakārmuken[ṇ]a v[b]āṇaiś ca dīptaḥ sa dharāvr̥ṣāṃkaḥ / ātta[sva]līlas tripurāṃtakasya ∪/– – ∪ – – ∪∪ – ∪ – – [// 17] ∪∪∪ ∪∪∪ – – – ∪ – – ∪ [15] kaṃraḥ sakalam api sa jitvā śātravaṃ śarvvakalpaḥ / surapatiramaṇīnāṃ puḥpa(ṣpa)-vr̥ṣṭyāvakīrṇaḥ puram anupama – – – ∪ – – ∪ – – [//18]

63. The problem of determining affiliation is particularly pronounced even at a site such as Khajurāho, which Devangana Desai, after years of careful research, conjectures may have been variously Śaiva Siddhānta and Vaiṣṇava Pāñcarātra; see Devangana Desai, The Religious Imagery of Khajurāho (Mumbai: Franco-Indian Research, 1996.

64. It is found in the context of an early tenth-century Jain temple in Āmvāñ. See EITA 3, no. 2, pl. 739.

65. On vulnerability of corners in medieval Indian temple architecture, see Michael Meister, “Śiva’s Forts in Central India: Temples in Dakṣiṇa Kośala and Their ‘Daemonic’ Plans,” in Discourses on Śiva, ed., Michael Meister, 119–43 (Philadelphia: University of Pennsylvania Press, 1984). See also Darielle Mason, “A Sense of Time and Place,” 116–38; Stella Kramrisch, The Hindu Temple (Calcutta: University of Calcutta, 1946), 38.

66. Helene Brunner, “Maṇḍalas and Yantra in the Siddhānta,” in Maṇḍala and Yantras in the Hindu Traditions, ed. Gudrun Bühnemann, 152–77 (Leiden: Brill, 2003), 159, 168.

67. On the iconography and placement of mātr̥kā sets, see Michael W. Meister, “Regional Variations in Mātṛkā Conventions,” Artibus Asiae 47, nos. 3/4 (January 1, 1986), 233–62; and Katherine Harper, The Iconography of the Saptamātr̥kās (Lewiston NY: E. Mellen Press, 1989). For a specific discussion of mātr̥kās around the exterior of a temple, see Desai, Religious Imagery, 165–68.

68. Unfortunately the main temple next to the monastery at Surwāyā is too damaged to determine its original program. For an overview of the temple, see EITA 3, no. 2, pl. 49.

69. EITA 2, no. 3, 98, pls. 272–74. M. A. Dhaky, who wrote the EITA entry on this temple, also notes the relative rarity of this feature.

70. Krishna Deva, in his entry on this temple for EITA accidentally mistakes Brahmā and Savitr̥i for Umā-Maheśvara (EITA 2, no. 3, 24).

71. For comparisons, see the goddess temple at Mahuā, the Śiva temple near Bhīmagajā at Pañhārī, the Jarai-kā-maṭh at Barwāsāgar, and the Śiva temple accompanying the monastery at Terāhī (EITA 2, no. 2, pls. 118, 132; and Trivedi, Temples of the Pratīhāra Period, pl. 116). At earlier sites, such as Naresar and Amrol, Garuḍa appears alone in the central position (lalāṭabimba) (EITA 2, no. 2, pls. 7, 8, 13).

72. See Meister, “Regional Variations,” 245.

73. I documented a fragment of a door lintel containing Naṭeśa at its center in storage in the vicinity of the Mohaja Mātā temple in Terahi in 2001 or 2002; it has since be moved to a more centralized site for storage by the ASI. Examples of doorframes featuring Naṭeśa in the lalātabimba from Buḍhi Chanderī can be seen in the outdoor sculpture garden at the Chanderī Museum today.

74. The inscriptions from Gurgī and Chandrehe are edited and translated by V. V. Mirashi in Inscriptions of the Kalachuri-Chedi Era, vol. 4, pt. 1, Corpus Inscriptionum Indicarum (Ootacamund: Gov. Epigraphist for India, 1955), 198–204, 228–233. They describe him as following in the Mattamayūra lineage of the sages Purandara and Śikhāśiva, also described as “lord of Madhumatī” (verses 4–5). Purandara is the first sage mentioned in the fragmentary inscription from Kadwāhā (verse 4), edited by Mirashi and Shastri (1968), and Madhumatī is the river that flows through the general region. Purandara is also named in an inscription connection with a second Mattamayūra monastery not far from Kadwāhā at the site of Ranod, published by Kielhorn, “A Stone Inscription,” 351–61.

75. I discussed this issue in “Constructing the Guru,” Art Bulletin 40, no. 1 (March 2008), 14–16.

76. Ranod inscription, verse 3. For a Sanskrit edition, see Kielhorn, “A Stone Inscription,” 351–61.

77. Translated by Mirashi, verse 3. Here the pairing of Naṭeśa and Caṇḍī, also known as Cāmuṇḍā and Caṇḍikā, is of particular interest because of an iconographic turn traceable at Kadwāhā. In nearly all of the temples at which Naṭeśa is prominently featured, Cāmuṇḍā appears as his counterpart in bhadra niches, effectively replacing Pārvatī as the god’s consort.

78. The power of SÏiva’s dance has been eloquently assessed and reassessed on numerous occasions. In South Indian contexts, dancing SÏiva came to signify the act of granting grace or liberation through the destruction of ignorance, personified as a demon, in the famous Naṭarāja images from the Chola period. See Ananda Coomaraswamy, The Dance of SÏiva: Fourteen Indian Essays (New York: Sunwise Turn, 1918), 56–66; Stella Kramrisch, The Presence of Śiva (Princeton, N.J.: Princeton University Press, 1981), 439–42; and Padma Kaimal, “Shiva Nataraja: Shifting Meanings of an Icon,” Art Bulletin 81, no. 3 (1999), 390–419.

79. For example, see R. N. Misra, Sculptures of DÚāhala and DakshinÚa Kosala (Delhi, 1987); Atherton, 18–20, 94, 103–6.

80. George Kubler, The Shape of Time: Remarks on the History of Things (New Haven: Yale University Press, 1962).

81. R. N. Misra, Ancient Artists and Art-Activity (Simla: Indian Institute of Advanced Study, 1975), 24.

82. Misra, Ancient Artists, 33.

83. Misra, Ancient Artists, 34-35.

84. Iron slag still visible in the foundation markings of abandoned medieval settlements in the vicinity of Kadwāhā, such as a place once known as Anghora Cukhru, suggest that Kadwāhā too may have been linked to metal production. I noted the presence of iron slag among the foundations of Anghora Cukhru in December 2012. It is also worth noting that Chanderī has long known for its textile production.