A Gold Coin of the Pāla king Dharmapāla
by PANKAJ TANDON1
THE PURPOSE of this brief paper is to present an important new coin which I believe to be the ﬁ rst known coin of the Pāla king, Dharmapāla. The Pāla dynasty was founded by Gopāla, who rose to power (some historians believe he was actually elected) in Bengal during the period of anarchy (matsyanyayam) that followed the decline of the Guptas and the invasions of Yasovarman of Kanauj and Lalitaditya of Kashmir. The dynasty lasted for almost four hundred years and raised Bengal to its highest level of power and achievement. Dharmapāla the son of Gopāla who ruled during the last quarter of the eighth and the early years of
the ninth centuries (c.AD 775–810) has been described as ‘one of the greatest kings that ever ruled in Bengal.’2 He expanded his inﬂ uence into the Deccan and up the Gangetic plain, installing his nominee on the throne at Kanauj. Both he and his son Devapāla engaged in an almost continuous struggle for supremacy with the Gurjara Pratiharas and the Rashtrakutas, and were able to maintain a strong position. After the death of Devapāla the Pāla realm started to crumble and, although it persisted in a reduced form for a long time, its former glory was never restored. Apart from their political power, the Pālas deserve an
important place in Indian history because of their cultural achievements. They were adherents of Mahayana Buddhism and prominent promoters of that faith. Indeed, the name Dharmapāla means ‘Protector of the Law,’ and is used as a name for eight ﬁ erce Buddhist deities whose role it is to defend Dharma. It was because of the efforts of Pāla missionaries that Buddhism came to be ﬁ rmly established in Tibet. The Pālas also patronized a ﬂ ourishing school of art, whose inﬂ uence is thought to be visible in Tibetan and South-east Asian art. Thus the Pālas occupy an important place in the political and cultural history of India. Despite their importance, and despite the fact that there are several copper plate grants and sculptural pieces attesting to their activities, the Pālas appear to have left little by way of coinage. Indeed, in
his study of the coinage of post-Gupta Bengal, B.N. Mukherjee asserted that ‘No metallic money was coined in the major portion of the Pāla … kingdom.’3 There is, however, one gold coin that has been put forward as a Pāla issue, but, because of its uniqueness, its authenticity has been called into question. I will consider this coin in detail later, after the new piece has been presented. This new coin is an important piece of evidence to help us understand the cultural and political climate of the times.
1 I would like to acknowledge the helpful comments of Joe Cribb and Wilfried Pieper. I am also indebted to Dr. T. Devendra Rao, who not only permitted me to examine and photograph some of his coins, but also loaned me some books from his library. 2 R.C. Majumdar, H.C. Raychaudhuri and Kalikinkar Datta, An Advanced History of India (London, 1963), p. 165. 3 B.N. Mukherjee, Coins and Currency Systems of Post-Gupta Bengal (New Delhi, 1993), p. 54.
The New Coin Of Dharmapāla The new coin is illustrated below by kind permission of the owner. It had been used as a pendant and still carries the loop attached to it for suspension purposes. Obverse: King ? riding on horseback left, with right hand holding up a spear, aimed at animal (lion or boar ?) at left, standard before the king. Four line Brāhmi legend at right: Sri / mā ndha / rma pā / lah, below horse’s forelegs, in two lines: kai / la, below horse’s hind legs: vo.
Reverse: Lakshmi seated facing on a single lotus, holding a lotus plant in each hand, ﬂ anked on each side by a sacred vessel, purna ghata. In top left ﬁ eld: Sri. Weight (including loop): 7.59 grams, 117.1 grains. Diameter: 21 mm. Die axis: 12 o’clock.
I believe the reading of the main part of the legend: Srimān Dharmapālah, is incontrovertible.4 All the letters are clearly legible. There could be some difference of opinion about individual letters taken on their own. For example, the ﬁ rst ma could conceivably be taken as pa or ya, the compound ndha might be seen as nva, rva, or rdha, and the pa in the third line could be read as ya or even ma. None of these alternative interpretations yield a sensible reading, while Srimān Dharmapālah is not only sensible, but also arguably the most plausible reading of the individual letters. The remainder of the obverse legend is less certain. One possible reading is kaila under the horse’s front legs, and vo under the back legs. The va seems to be sufﬁ ciently distinct from the dha in Dharmapālah to be read that way. The word kaila could be referring to the keela, or spear, being held by the king. Vo could have an as yet unclear meaning, or it could combine with kaila to yield kailavo. The vo could be an exclamatory sufﬁ x, and the phrase could serve
4 I am indebted to Shailendra Bhandare for his help with reading and understanding the legend. It was he who ﬁ rst suggested Srimān Dharmapālah as the reading.
as an epithet for the king: ‘the illustrious Dharmapāla, wielder of the spear’. This reading of the legend is buttressed by the Devapāla coin which also may carry an epithet ending with an exclamatory va, and is discussed below. Another possible reading would take the letter kai and read it as the compound nka. The legend could then conceivably be restored as Vonkalah. Perhaps scholars of Brahmi and Sanskrit can elaborate on this, but it seems that Vonkalah might perhaps represent a reference to the Bangla homeland of the Pālas, and the coin could in this way be related to the coins with the legend Vangālamrigānka, which was an epithet of Anandadeva of Samatata.5 On the whole, I prefer the kailavo reading, as it seems more closely connected to the coin of Devapāla. Apart from the reading of the legend, there is the
question of the identity of the animal on the left being hunted by the king. At ﬁ rst, I naturally assumed it represented a lion, placing the coin squarely as a ‘mounted king lion-slayer’ type, similar in theme to the coins of Prakāsaditya. Upon closer examination it seemed that the animal might perhaps represent a boar. Normally, the Gupta style lions being hunted in the lion-slayer types are shown with mouth agape, thereby magnifying the ferocity of the hunted animal and, by extension, the prowess of the hunter. Here the animal’s mouth is only slightly open. Further, the face exhibits an elongated snout, more like a boar than a lion. Also, the expected lion’s mane is certainly not prominent, although one could argue that there is a small mane present. Thus the animal’s identity is somewhat ambiguous; it might be a lion or a boar.6
Next, it is worth considering the style of the coin to determine if there are any close parallels in the numismatic sphere. The mounted king of course has a well-known lineage in Indian coinage, with its frequent use in Indo-Greek, Indo-Scythian, Indo-Parthian, Kushan, and Gupta coinage. The mounted king as hunter is a less common motif, but we do see this in Gupta coinage. In his beautiful Rhinoceros-slayer type, Kumāragupta is shown mounted on a horse (see Fig. 2). The lion-slayer type of Prakāsaditya also depicts the king on horseback (see Fig 3). So the present coin seems to have clear antecedents in Gupta coinage.
5 See Mukherjee, Coins of Post-Gupta Bengal, p. 19. 6 It is conceivable that, if the animal is a boar, it represents the Pratihāras, who were enemies of Dharmapāla while, if it is a lion, it represents the Rāshtrakutas, also rivals (although later allies) of the Pāla king.
Another important coin closely connected to the present one is the so-called Bālamrigānka type published by B.N. Mukherjee.7 Another specimen was published by Rhodes and Bose, as their coin no. 27, although they did not provide any details. 8 As I have had the opportunity to examine and photograph this specimen, apparently not from the same dies as Mukherjee’s, I publish it here again for the beneﬁ t of scholars (see
This specimen weighs 7.32 grams and has a diameter of 23 millimeters, very much in conformity with Mukherjee’s specimen, although the die axis is approximately 2:30. There are two important parallels between this coin of Bālamrigānka and the Dharmapāla coin. First, the treatment of Lakshmi is very similar on both coins. The cross-legged posture, the simple lotus seat, the ﬂ ower in hand, the presence of the vase, and the use of the legend Sri, all point to a close afﬁ nity between the two coins. Second, the obverse legend arranged, vertically in horizontal lines, and in different locations on the ﬂ an, again suggests a close relationship. Joe Cribb has suggested that the Bālamrigānka coins could be Pāla issues.10 There are three known versions of the Bālamrigānka type, differing in the letters on the obverse. The illustrated coin has, in the left ﬁ eld, the letters Ja ya, and could represent an issue of Jayapāla. There are also coins with the letter Go, which could be issues of Gopāla, and coins with the letter De, which could be issues of Devapāla.
7 Mukherjee, Coins of Post-Gupta Bengal, pp. 58–61 8 N.G. Rhodes and S.K. Bose, The Coinage of Assam, The Pre-Ahom Period (Kolkata, 2003), vol. I, p. 77. 9 The coin is in the collection of Dr. T. Devendra Rao, whose kindness in letting me examine, photograph and publish it is gratefully acknowledged. The Kumāragupta rhinoceros-slayer coin published in Figure 2 is also in Dr. Rao’s collection, published here by kind permission. 10 Private communication.
The ﬁ rst coin of the series would then be the Bālamrigānka coin with the letter Go on the obverse, representing a coin of Gopāla, the father of Dharmapāla. Mitchiner has published such a coin, number 79 in his catalogue of Bengal coins.11 Indeed, an examination of this piece shows that the style is the crudest of the three Bālamrigānka coins, indicating perhaps that the style evolved and developed over time. We will see a similar evolution in style between our coin of Dharmapāla and a later coin of his son Devapāla, which is in a much more reﬁ ned style. The second coin among the known Bālamrigānka coins is marked by the letter De on the obverse,12 and this could possibly be an issue of Devapāla. Finally, the coin illustrated in Fig. 4, with the letters Jaya could be an issue of Devapāla’s cousin Jayapāla, the son of
Dharmapāla’s brother Vākpāla. Jayapāla was the commander of the army and led a Pāla campaign in Assam.13 Finally, it would be appropriate to say a word about the denomination of this coin. As it carries a suspension loop, it has not been possible to obtain an accurate measurement of the weight of the coin itself. But the weight, including the loop, of 7.59 grams suggests the coin originally weighed approximately 7.25 to 7.30 grams. This is very similar to the recorded weights of the Bālamrigānka coins, which Mukherjee has argued conform to a tripāda-suvarna, or three-quarter suvarna, standard. The fact that the Bālamrigānka coins conform to the weight standard of the Pāla coins rather than the earlier Samatata issues is a further argument in favour of assigning these coins to the Pālas. Almost all of the Samatata
issues weigh between ﬁ ve and six grams. For example, the 31 Samatata coins in my own collection average 5.56 grams in weight and, apart from one low-weight outlier at 4.39 grams the weights of these coins fall within the range of 5.04 to 5.85 grams. These coins are thus signiﬁ cantly lighter than the Pāla-Bālamrigānka coins. Further, the quality of gold of these coins also seems better than the Samatata coins, although this statement is based only on visual inspection rather than a proper metallic analysis. 11 M. Mitchiner, The Land of Water: Coinage and History of Bangladesh and Later Arakan (London, 2000), p. 59. Mukherjee, Coins of Post-Gupta Bengal, also publishes a specimen of this type on plate VI, no. 2. 12 See Mukherjee, Coins of Post-Gupta Bengal, plate VI, no. 1. 13 See R.C. Majumdar, ed., The Age of Imperial Kanauj (Bombay, 1963), p. 50.
Relationship to the coin of Devapāla I turn now to consider the relationship of the Dharmapāla coin to one previously claimed to be an issue of his son Devapāla. It was originally published by Ajit Ghosh in 1951 and assigned by him to Devapāla. 14 This attribution has been called into question, and indeed the authenticity itself of the coin has been doubted.15 I have been able to examine the coin myself and am providing a photograph of it by kind permission of the owner
The most important point to note about this coin in relation to the Dharmapāla coin is that the two coins show marked similarities of style and the forms of letters, so much so that I would assert that the Dharmapāla coin now provides strong evidence to suggest that the Devapāla coin is indeed genuine. The two greatest areas of similarity are precisely the same areas in which the Dharmapāla and Bālamrigānka coins also showed an afﬁ nity for one another: in the treatment of Lakshmi and in the arrangement and paleography of the legends. The treatment of Lakshmi on the two coins is virtually identical. On both coins, the goddess is seated in a full padmasana, with her wrists resting on her knees. Each hand holds a lotus plant with long, wavy leaves and a central ﬂ ower stalk that ends with a bell-shaped bloom and with petals or leaves on either side. The goddess is not nimbate and is ﬂ anked on each side by a sacred vessel. The word Sri is at the top left on both coins. Finally, the lotus seat, while not identical (the Dharmapāla coin
has a single lotus, the Devapāla a double), has a curious feature not known on other coins: the presence of a central stem below, ﬂ anked by leaves. The similarity of the reverses is so great that, I would argue, the two coins can be conﬁ dently identiﬁ ed as the products of the same atelier. Turning now to the obverse legend, the letter forms are clearly very similar. The letter Sri, for example, is virtually identical on the two coins. Other letters also show the same kind of marked similarity, although there do appear to be some differences, perhaps representing the idiosyncrasies of individual celators. The arrangement of the legends is also very much alike, with letters arranged as a vertical legend in short horizontal rows. I would argue that the legend arrangement also suggests that the Dharmapāla coin increases the Devapāla coin’s 14 A. Ghosh, ‘A unique gold coin of the Pala emperor Devapala Deva,’ JNSI XIII, Part II, (December 1951, publ. June 1952) pp. 123–5. 15 See Rhodes and Bose, Coinage of Assam, chapter 6.
claim to be genuine. In their careful analysis arguing that the Devapāla coin is not genuine, Rhodes and Bose point out that the coin is so different from other known coins that its very uniqueness renders it suspect. Now, with the emergence of this coin of Dharmapāla, that unique status has gone and, accordingly, the argument of Rhodes and Bose loses much of its force. Finally, let me turn to the reading and interpretation of the legend. Ghosh had originally read the legend as Sri / Mā / Ha rsha / Vā / nde va / Pā la and attributed the coin to the Pāla emperor Devapāla. D.C. Sircar reinterpreted this to mean Srimān Harshavat Devapāla, and opined that, ‘It is difﬁ cult to say whether the coin may be attributed to king Devapāla (c.AD 810–50.) of Bengal and Bihar.’16 J.P. Singh rejected the attribution to the Pāla king, largely on the basis of the fact that the ﬁ rst part of the name of the issuing king was Harsha and therefore this must have been the main part of his name.17 Largely on this logic he attributed the coin to Harshapāla Varmana Deva of Assam. P.L. Gupta implicitly supports this attribution, despite acknowledging that the coin shows evidence of Pāla artistry.18 Rhodes and Bose argued against this attribution, saying
that they did not feel a coin of this quality was within the capability of Assamese celators of the time.19 As already mentioned since the coin was unique, they concluded that it was probably not authentic but that, even if it was, it was unlikely to be of Assamese origin. I feel the Dharmapāla coin allows us to look afresh at this coin in two important ways. First, we know from the Dharmapāla coin that the Pāla celators were using the title Srimān, and that they were conjoining the ﬁ nal na of this word with the ﬁ rst letter of the name.20 Thus we see the compound letter ndha on the Dharmapāla coin. This would suggest that the reading of the legend on the Devapāla coin proceeds from Sri to ma, to be followed not by Harsha, but moving to the lower right panel, with nde. The lower left panel would then be read last. The second
piece of evidence from the Dharmapāla coin relates to the possible epithet for the king being followed by an exclamatory vo, or, in the case of the Devapāla coin, va. Thus the king’s name on the Dharmapāla coin is followed by kailavo, while the king’s name in the Devapāla coin is followed by Harshahva If we accept this argument, the legend on the Devapāla coin could be read as Srimān Devapālah Harshahva. Reading it in this way negates Singh’s argument that the name Harsha occupies the most important place. Rather, the name becomes clearly Devapālah, followed by what is possibly an epithet, Harshahva. I leave this for the Sanskrit scholars to analyse, but suggest that perhaps Devapāla was known as the embodiment of happiness or joy. Conclusion In this brief note, I have presented an important new gold coin, attributing it to the Pāla emperor Dharmapāla. If this attribution is correct, this would be the ﬁ rst known coin of that illustrious king. The coin serves as a link between some other post-Gupta coins of Bengal, and has a particular similarity with a previously published coin of Dharmapāla’s son, Devapāla, thereby supporting that coin’s claim for authenticity.
16 D.C. Sircar, Studies in Indian Coins (Delhi, 1968), p. 383. 17 J.P. Singh, Monetary Development in Early Assam (Jorhat, 1989), reviewed and discussed in Rhodes and Bose, Coinage of Assam, pp. 52–3. 18 P.L. Gupta, Coins (India – The Land and the People) fourth ed. (New Delhi, 1996), pp. 83–4. 19 Rhodes and Bose, Coinage of Assam, p. 53. 20 Shailendra Bhandare pointed out to me that this way of presenting Srimān Dharmapālah does not really conform to the rules of sandhi in Sanskrit, which would have required the conjoined words to become Srimat Dharmapālah. He argued, however, that, as this was about the time that ‘Apabhramsa’ languages were emerging, perhaps the rules were more likely to be bent. This may explain the sufﬁ x va seen at the end of the legend.