Minerals and Metals in Ancient India v1.c28.s8.t1: Metallic Art-Traditions in Gujarat
Minerals and Metals in Ancient India v1.c28.s8.t1
- Language
- English
Metallic Art-Traditions in Gujarat
v1.c28.s8.t1.1
In his otherwise excellent work Swarnakamal (1980) has surveyed the traditions of Gujarat in producing metallic art objects and icons. A summary of the analytical data obtained by him is presented in Table 20.4 which clearly depicts the long, almost unbroken tradition across the centuries of casting copper, bronze and brass icons in Western India. Apart from chemical estimation of the percentages of major elements, the spectroscopic data show the presence of trace elements hinting at the kinds of ore or raw material from which the metals were extracted.
v1.c28.s8.t1.2
Rosdi in Central Gujarat is a Harappan site which produced mostly copper artefacts significantly contaminated with lead and zinc as impurities suggesting that the ore came from the Zn/Pb bearing Southern Aravalli area. Tin was deliberately added specially in the ring to assist casting at lower temperature. The Harappan artefacts from Rangpur, Gujarat were also mostly made of copper and bronze, the latter containing 2-11 per cent tin. Bangles contained higher percentages of tin. Several samples contained nickel varying in percentage from 0.2 to as much as 5.9. Nickel was derived from the ore. (Lal, 1962-63) A Somnath axe (29.8 cm long) was made of bronze: copper alloyed with 12.8 per cent tin, 1.21 per cent lead and containing 2.57 per cent iron. Metallographic analysis of the Rosdi ring (Swarnakamal, 1980: 42) showed that this was carefully cast and subjected to mild working when the dendritic structure was broken.
v1.c28.s8.t1.3
The four pre-Gupta Period icons found in Gujarat and described by Swarnakamal (vide items 3-6 in Table 20.4) illustrate that during second-fourth century AD the province was a melting pot for intermingling of different art traditions — Gāndhāra, Mathurā, Amarāvatī and Kuṣāṇa or Indo-Pārthian styles. The Fasting Śākyamuni (no. 3) was made of pure soft copper. Absence of sulphur and core markings or dendrite and presence of traces of silicon and iron suggest that it was made of native copper like the Zanskar Valley material in Taxila as reported by Sana Ulla. (Swarnakamal, 1980: 48-51) The rough finish Analysis of Metallic Items and Icons from Ancient and Medieval Gujarat
v1.c28.s8.t1.4
| Sl. No. | Site | Item | Elemental Percentages | | | | Impurities |
| | | | Cu | Sn | Zn | Pb | Fe | |
| 1. | Rosdi, Central Gujarat, Harappan Site | Chisel, Celt, rod, bangle | 95.5 - 98.5 | up to 1.23 | up to 1.54 | up to 1.20 | — | Fe, Ni, As, Bi, Ag, Si, Mg |
| 2. | Rangpur | Ring Bangle | 88.6 86.4 | 11.0 11.1 | — | — | — | Fe, Bi, Ag, Si, Mg, Ni 1.8 %, Pb, As |
| **Pre-Classical Gupta Age** | | | | | | | | |
| 3. | — | Fasting Śākyamuni Gāndhāra Style 2nd-4th century AD | 99.0 | — | — | — | — | Si, Mg, Fe |
| 4. | — | Meditative Buddha Mathurā Style 2nd-3rd century AD | 81.9 | 15.8 | — | 0.8 | — | Fe, S |
| 5. | — | Two Dancing Girls Amarāvatī style. 2nd century AD | 76.6 | 15.8 | — | 6.4 | — | Al, Fe, Zn |
| 6. | — | Indo-Pārthian Figure carrying Flower-container, Kuṣāṇa style. 2nd-3rd century AD | 88.6 | 0.13 | 7.6 | 2.49 | — | Fe, Sb |
| **6th century AD** | | | | | | | | |
| 7. | Akota, Central Gujarat | Jīvantasvāmī 30 x 7.8cm | 68.6 | 8.7 | 3.5 | 13.2 | 1.9 | Al, Mo, Ag, Bi |
v1.c28.s8.t1.5
| 8. | Akota | Ṛṣabhanāth or Ādināth* | 78.5 | 0.9 | 10.4 | 9.0 | — | Fe, Ag, Bi, Si, Mg |
| 9. | Akota | Jīvantasvāmi 40.5 x 11.2 cm | 73.4 | 5.53 | 3.56 | 9.82 | 4.91 | Ag, Cd, Au, Sb, Bi, Mg |
| 10. | Akota | Ambikā installed in the Vidyādhara Kuḷa* | 76.7 | 1.61 | 16.32 | 4.04 | — | Fe, Ag, Bi |
| **6th-7th century AD** | | | | | | | | |
| 11. | Mahudi | Pārśvanātha | 89.3 | 0.95 | — | 9.6 | — | Zn, Fe, Ag, Bi |
| 12. | Akota | North Gujarat Pedestal of a Jain image | 77.4 | 8.1 | — | 10.5 | — | Zn, Fe, Al, Ag, Cd, Au, Sb |
| **7th century AD** | | | | | | | | |
| 13. | Mahudi | Ṛṣabhanātha | 66.0 | 5.9 | 12.8 | 2.5 | 1.00 | Ni 3.8 Al 6.7? |
| 14. | Akota | Standing Sarasvati with halo* | 80.0 | 8.1 | 6.5 | 3.4 | 2.00 | Ni, Al, Cd |
| 15. | Akota | Tritirthaka of Pārśvanātha 22 x 11 cm | 80.8 | 5.3 | 7.4 | 2.2 | — | Fe, Ni, Al |
| 16. | Akota | Standing Pārśvanātha* | 73.2 | 6.6 | 12.8 | 4.5 | 1.7 | Al, Ag, Cd, Mg |
| 17. | Akota | Unidentified Jina* 26.7 x 16.4 cm | 78.3 | 1.9 | 10.1 | 3.8 | 3.2 | Al 1.5. Ni 0.3., Cd |
| **7th-8th century AD** | | | | | | | | |
| 18. | Mahudi | Tirthankara with Yakṣa-Yakṣi | 82.0 | 11.2 | 3.5 | 2.0 | — | Fe, Ni, Al, Bi |
| 19. | Akota | Tritirthaka of Pārśva-nāth 20.8 x 12.7 cm | 88.8 | 3.9 | — | 3.6 | 1.9 | Ni, Cd, Mg |
| 20. | Akota | Ambikā on Lion | 79.7 | 12.7 | 3.9 | 2.6 | — | Fe, Ni, Al |
| | | | | | | | | (Contd.) |
v1.c28.s8.t1.6
| Sl. No. | Site | Item | Elemental Percentages | | | | Impurities |
| | | | Cu | Sn | Zn | Pb | Fe | |
| **8th century AD** | | | | | | | | |
| 21. | Akota | Sarasvati without Halo* | 85.5 | 3.2 | 6.7 | 2.6 | 1.9 | Al |
| 22. | Akota | Cāmaradharīṇī* | 73.2 | 2.2 | 10.7 | 11.3 | 1.4 | Fe, Al, Ag, Cd, Au, Sb, P, Mg |
| **8th-9th century AD** | | | | | | | | |
| 23. | Lilvadeva in Zalod, N.E. Gujarat | Caturviṃśati-Paṭṭa of Ādināth | 88.1 | 2.1 | — | 7.6 | — | Fe, Al |
| 24. | ” | Unidentified Jina | 81.6 | 5.9 | — | 5.9 | 2.2 | Al 2.1, Zn, Mg |
| 25. | ” | Pārśvanātha | 73.4 | 7.5 | — | 11.3 | 2.5 | Al 2.6 |
| **9th century AD** | | | | | | | | |
| 26. | Akota | Unidentified Jina* 34 x 10 cm | 72.7 | 5.7 | 6.8 | 11.2 | 1.95 | Ag, Bi, Si |
| **9th-10th century AD** | | | | | | | | |
| 27. | Akota | Mutilated Back Stela | 78.0 | 3.7 | 2.6 | 15.7 | — | Fe, Al, P |
| **11th century AD** | | | | | | | | |
| 28. | Lilvadeva | Pārśvanātha | 80.1 | 0.4 | 6.8 | 4.5 | 2.5 | Ni 4.2 |
| 29. | ” | Tirtirthaka of Pārśvanātha (AD 1037) | 69.9 | — | 16.9 | 9.9 | — | Fe, Sn, Al, Mg, Sb, Bi, P |
| 30. | Lilvadeva | Tirtirthaka of Ādinātha | 63.4 | — | 20.6 | 13.6 | — | Fe, Ni, Al, Sn, Ag, Mg, Sb, Bi |
| **Medieval Period** | | | | | | | | |
| 31. | Gujarat AD 1350 | Ambika | 68.4 | 1.6 | 18.5 | 9.5 | — | Fe, Ag, Bi |
| 32. | Gujarat AD 1480 | Model Temple with four doors 10 x 24. 5 cm | 68.6 | 0.2 | 28.9 | 1.6 | — | Fe, Mg, Bi |
| 33. | Gujarat AD 1485 | Viṣṇu-Nārāyaṇa | 58.9 | 0.5 | 36.8 | 2.1 | — | Fe, Al, Ag, Si, Mg, Bi |
v1.c28.s8.t1.7
| 34. | Rajasthan 15th-16th century AD | Rajput Prince on Horse | 72.9 | 0.4 | 21.8 | 2.7 | 1.0 | Al 0.3, Ag, Mg |
| 35. | Gujarat AD 1554 | Kāl Bhairava | 76.7 | 1.2 | 13.8 | 6.3 | 1.5 | Al, Mg |
| 36. | Gujarat 17th century AD | Cauri-Bearer | 58.3 | 1.5 | 35.5 | 2.3 | — | Fe, Ni, Al, Ag, Cd, Si, Mg |
| 37. | Rajasthan 17th century AD | Dīpalakṣmī Rajput Girl | 58.8 | 0.9 | 33.2 | 4.5 | 0.8 | Al 1.6, Mg |
| 38. | Gujarat 18th century AD | Dīpalakṣmī | 52.8 | 1.1 | 39.9 | 2.9 | — | Fe, Al |
| 39. | Gujarat AD 1752 | Tirthankara-Seated | 62.3 | — | 36.0 | 0.5 | — | Fe, Ag, Bi |
| 40. | Nepal | Ṣaḍakṣarī Lokeśvara form of Avalokiteśvara | 60.5 | 2.75 | 35.3 | 2.37 | — | Fe, Ni, As, Au |
v1.c28.s8.t1.8
∗ Note: Swarnakamal proposed somewhat later dates for these Akota icons; we however accept the earlier dates (in the table) as suggested by Shah.
v1.c28.s8.t1.9
suggests that the image was probably cast in a sand mould like many other Gāndhāra metallic objects. The 18 cm high 13 cm wide pure copper image is of solid cast and reflects the strong Buddhist tradition. Reedy (1983) has quoted the Sāriputra text disapproving hollow casting: “If a statue is made with a hollow body, there will be strife and quarelling”.
v1.c28.s8.t1.10
Both the Meditative Buddha of Mathurā style (No. 4) and the Amarāvatī style dancer pair (No. 5) belong to the same age and significantly contain the same percentage of tin. The latter however contains 6.4 per cent lead also which was a deliberate addition. We may recall that the Andhra Sātavāhanas and Kṣtrapas introduced lead coinage in the country. Swarnakamal (1980: 60-61) has drawn our attention to the object no. 6, an Indo-Pārthian figure assaying 7.6 per cent zinc, a Roman brass jar-handle found at Akota and a brass Atlas statuette recovered from Samalaji, and then wrongly concluded that these items, like the Taxilian brass, ‘came from outside India’. Such a conclusion must be rejected in view of our earlier discussions on zinc ore mining, zinc metal and brass artefacts in ancient India. (vide chapters 17 and 18)
v1.c28.s8.t1.11
Swarnakamal has studied the bronze brass icons from different parts of Gujarat: North (Mahudi on the bank of Sabarmati), North-East (Lilvadeva in the Panchmahal district) and Central (Akota near Baroda). His results of chemical and spectrographic analysis of the principal icons, together with the respective dates as suggested by U.P. Shah (1956 and 1959) and other art scholars, have been cited in Table 20.4. We have discussed why the dates suggested by Swarnkamal are not acceptable by us. Banerjee (1982) has also doubted whether Swarnakamal’s proposition of dates would be universally acceptable.
v1.c28.s8.t1.12
On the basis of the dates of the icons affixed by him, and which are at variance from those given by the art scholars, Swarnakamal reached certain general conclusions which may be quoted (1980: 165):
v1.c28.s8.t1.13
The earliest metals used in making of artifacts, objects of self-adornment and worship were more or less pure copper, and rarely copper-tin alloy. The metal generally used during the Gupta Period was a kind of lead-bronze alloy, and in the post-Gupta periods (7th-8th centuries AD) it was copper-tin alloy. In the metal objects of the period ninth-fourteenth centuries AD, we find a transition from the use of copper-tin alloy to copper-tin-zinc alloy, which are known as zinc bronze, and thereafter to copper-zinc-lead alloy known as lead-brass. In the post-medieval period, particularly after fourteenth century AD, we find a marked transition from lead brass to real brass technology.
v1.c28.s8.t1.14
We agree with him only partially. It is true that in the pre-Gupta and Gupta Period there was an aesthetic tendency to use pure copper for making icons (Fasting Śākyamuni, No. 3, Sultanganj Buddha etc.), and after fourteenth century the use of tin dwindled giving preference to zinc. But the situation between sixth-fourteenth century AD is not as simple as Swarnakamal has proposed. If he failed to find the use of zinc in the ancient Gujarat icons before ninth century AD, it is because he chose to ignore the earlier dates suggested by the epigraphers and art scholars. Regarding lead we do not find any intermission of its use across the centuries second-fourteenth century AD as suggested by the model given by Swarnakamal; the proportions of lead in the icons merely fluctuated.
v1.c28.s8.t1.15
There might have been regional scarcities such as of lead in Akota of seventh century AD, of zinc in Lilvadeva during the period eighth-ninth century AD and of tin in Lilvadeva during eleventh century AD. Particular compositions could reflect comtemporary and local art styles apart from the availability/scarcity of the constituent metals. We do not find much use of tin in the Gujarat icons after tenth century AD, whereas this metal continued to be popular in South India as attested in the composition of Cola bronzes.
v1.c28.s8.t1.16
In Gujarat there were tendencies for producing (i) zinc-free or low Zn lead bronze (Nos. 5, 12,19, 23-25) and (ii) tin-free or low Sn lead brass (Nos. 6, 8,10, 17, 28-31) but these tendencies were feeble and not restricted to one place or any sharply defined time period. Only a few samples (Nos. 9, 16 and 26) contained reasonable amounts (more than 3.5 %) of all the three metals. The percentage of lead never fell below 2 per cent. Low proportions of tin or lead might have been used to improve fluidity and the ease in deoxidation of copper melt. Thus Swarnakamal's claim that Gujarat introduced specific alloys such as zinc bronze, lead bronze, lead brass and actual brass in specific periods and locations cannot be supported on the basis of the presently available data; more data are essential before any firm conclusion can be reached on this point.
v1.c28.s8.t1.17
The highlights of Swarnakamal's work has been the chemical, spectrographic and metallographic analyses of the Gujarat icons. The key metallographic observations made by him are summarised in Table 20. 5. In most of the samples lead globules remained as insoluble discrete phases, and where there was very little Sn or Zn, no dendrite-producing alloy phase was noticed. In many instances, slip and strain lines, flattened grains and broken dendrites provided evidences for hot or cold working by techniques such as light or heavy hammering. There were cases of cold working preceded or followed by annealing.
v1.c28.s8.t1.18
Annealing was done below or at recrystallisation temperature. Twinned grains are pairs of identical grains, whose lattice structures are mirror images Metallography of Some Gujarat Icons
v1.c28.s8.t1.19
| Sample No. in Table 20.4 | Discrete Phases | Evidence of Annealing or Heat-treatment | Evidence of Cold or Hot working |
| 7. | Lead (L) and Inter-metallic compounds (IC) | Annealing after casting | Slip and strain lines (SLSTL) and Flattened Grains (FG) suggest severe hammering or cold working |
| 8. | Lead (L) and Oxides (O) | Twin Grains (TG); annealing was done at Recrystallisation Temp. (RT) after cold working | Cold working |
| 9. | L at Grain Boundary (GB) | Annealing below RT | No evidence |
| 10. | — do — | — do — | — |
| 11. | L and O; otherwise no alloy phase; no dendrite; polyhedral grains | — | — |
| 12. | Same as 7 | Heat-treated or annealed below RT; then . . . . . | mild working |
| 13. | Distorted polygonal grains | Annealing below RT | No evidence |
| 14. | L at GB; large grains | A few twinning crystals suggest annealing at RT before cold or hot working | SLSTL and FG suggest cold or hot working |
| 15. | Equiaxial grains | — | — |
| 16. | Same as 14 | Same as 14 | — |
| 17. | L and IC at grain boundary | Annealing and Heat-treatment below RT suggested by etched structure | No evidence |
| 18. | Dendritic Structure | No evidence | Mild hot or cold working |
| 19. | O at grain boundary | No evidence | No evidence |
v1.c28.s8.t1.20
| 20. | — | No evidence | Mild hot or cold working suggested by broken dendrites (BD) |
| 21. | As in 14 | As in 14 | — |
| 22. | Lead precipitation; distorted polygonal grains without twinning | Annealing below RT | Extensive parallel SLSTL and FG indicate heavy cold working causing permanent deformation |
| 23. | O and L as globules; no dendrite, and no alloy phase | — | — |
| 24. | IC visible in internal structure | No evidence | BD observed in etched structure suggests cold working |
| 25. | Same as 24 | — | Same as 24 |
| 26. | L and IC | Annealing below RT | Parallel SLSTL and FG indicate heavy cold working (after annealing) such as severe hammering |
| 27. | — | Annealing after casting at intermediate temperature | No evidence |
| 28. | — | Annealing below RT | — |
| 29. | L precipitated; distorted polygonal grains without twinning | As in 26 | — |
| 30. | — | As in 26 and 29 | As in 26 and 29 |
v1.c28.s8.t1.21
of each other, formed on each side of a common plane. Fine grain structure and the presence of twinned grains indicate that the object might have been first cold worked and then annealed at the recrystallisation temperature. Such an object (no. 8) is the beautifully modelled Ṛṣabhanāth of the Gupta Period (5th-6th century AD) from Akota; Swarnakamal pushes its date to a much later period probably because it contains 10.4 per cent zinc! Another object (No. 15), also from Akota-Tirīrthaka of Pārśvanātha (22 x 11 cm) — showed equiaxial grains along with a few twinning crystals, and also strain lines. Swarnakamal suggests (1980: 112) that this object was probably annealed to recrystallisation temperature, and then worked in cold or hot condition. This icon also contains appreciable amount of zinc 7.4 per cent and certified by the art scholars to belong to seventh century AD. Thus sophisticated casting, cold/hot working and annealing of bronze/brass icons used to be done in Akota and other places of Gujarat during seventh century AD; this is well before the Cālukya or Solaṅkī rule (AD 961-1222) which is claimed by Swarnakamal (1980: 172) to be the era introducing brass icons.
v1.c28.s8.t1.22
The chronological list of icons in Gujarat attested by art scholars, epigraphers and palaeographers, reproduced by Swarnakamal (1980: 11-15) show very few noteworthy icons after tenth century AD and before the fourteenth century AD. The Somanātha temple was sacked by Mahmud of Ghazni in AD 1026 and things did not settle down in the western part of India till the invading Muslims consolidated their hold in this part of the country which took about three centuries. In the thirteenth century AD mining of zinc ore and smelting the metal were resumed at Zawar and thus the brass technology was boosted. For four centuries (AD 1350-1752) Gujarat produced brass icons with very high precentages of zinc and very little lead and tin (Table 20.4). The eighteenth century AD brass icon Dīpalakṣmī (No. 38) assays nearly 40 per cent zinc. The seated Tīrthaṅkara image (No. 39) bears an inscription of the date AD 1752 which is five years before the war of Plassey and eight years before Zawar Mine had to slow down and eventually close its production on account of the repeated Marāṭhā invasions in Rajasthan.