Roman Coins from India

Paula J Turner

• Publisher: Routledge
• Year: 2016
• eBook ISBN: 9781315420677
• Edition: 1
• Topic: Sciences sociales
• Subtopic: Archéologie

This book is a detailed collation of the recorded finds of Roman coins on Indian soil, divided into Republican, Julio-Claudian and post-Julio-Claudian coins. It also includes chapters on the historical significance of the scarcity of Roman finds, the absence of base metal issues in the early empire, the predominance of early imperial denarii, and the difference in composition between the Julio-Claudian gold and silver hoards. There is considerable discussion on slashed gold coins and defaced silver coins and on imitation Roman coins found in India. Three exhaustive appendices include a catalogue of finds of Roman coins found in India, the present location of Roman coins found in India, and Roman coins in the Madras Central Government Museum. Copublished with the Royal Numismatic Society.

CHAPTER ONE

Historical Introduction

Roman coins have been found on Indian soil for almost two centuries; the earliest find was made in 1796 at Nellore,¹ where about forty aurei were unearthed in a pot. The find was communicated to the Society of Antiquaries in London, and much interest was aroused. This is the first record of a find, though slightly earlier Stavorinus (1798, vol. III, p.11) had claimed that finds of Roman coins were frequently made in the Surat region and a find was noted in Sri Lanka as early as 1574.² No doubt finds have been made but left unrecorded throughout the centuries.

The main factors governing our knowledge of coin hoards are three: first, the initial discovery is generally made by chance, since the position of hoards cannot be predicted accurately (unless a metal detector is available) without prior knowledge, though it may be known that finds have been made in that area before. For example, the discovery of Roman coins in the Karur district of southern India led to some speculative digging for buried treasure. Little (1883) hinted that this had occurred in the Karur area after two large hoards appeared there in 1856 and 1878. Secondly, a finder may be unwilling to admit to his windfall, especially if he fears he may lose it to state authorities without recompense. Finally, the official machinery which allows for the full study of a hoard by qualified personnel, usually in the form of Treasure Trove laws, may be primitive or dependent on the sympathy of particular officers to the interests of the antiquarian researcher.

These factors have all affected the quality of our knowledge of Roman coins found in India. The present study deals with finds containing coins up to the time of Constantine I, AD 307-337. There are over 75, and their distribution throughout India is shown on Map 1. The following discussion is based on the evidence provided by those finds, but it should be noted that further finds, and lost finds, may alter the general picture. For example, the Bamanghati hoard, recorded from a word-of-mouth report by Cunningham (1882), is sited in a part of India otherwise bereft of finds, and it is supposed to have contained third-century gold coins, which have not been found in any other Indian hoards. Unless and until similar finds are made in the same region, such a find can only be considered a freak occurrence; little significance can be attached to it because it is unique, and so it is difficult to establish how and why such a hoard came to be there. In addition, without a proper description of the hoard, it is impossible to know whether the coins were rescued intact from the earth, since they may have been in the hands of a villager for some time.

Once a find has been made, there are numerous possible fates for it. Interested and knowledgeable individuals like Sir Walter Elliot may be available to publish an accurate description of the coins; he was an authority on south Indian numismatics and wrote an exemplary report on the Vellalur hoard of 1841 (Elliot, W. 1886 and 1844). Less experienced observers, such as those who saw the Nellore find, may note a few salient points about the coins but leave us in doubt about the precise details. Our knowledge of a find of denarii at Karur in 1856 derives only from a brief note about it made by Henry Little in his description of a similar find made there in 1878; of the earlier find he noted that he ‘could not hear that a single denarius remained unmelted’.

In the princely territories, much depended on the attitude of a particular ruler. The ‘Ameer’ who received the Nellore hoard bountifully gave a coin each to his particular friends (but not to Colonel Sydenham who first reported it) and then he allowed Sydenham to study and draw them two or three at a time, but he ‘suddenly stopt and all… applications could not procure… another coin’ (Sydenham, Col., 1789, pp. 81-83). Davidson, the Governor in the area, himself got ‘an Adrian and a Faustina’, identifiable from his description as types RIC Hadrian 192-3 and RIC Antoninus Pius 394-5 (Davidson, A. 1790). Some of the coins from the massive Kottayam hoard went to the Rajah of Travancore, who allowed Bishop Caldwell, a numismatist, to study such coins as he had acquired; and the Rajah of Pudukottai was sufficiently enlightened to permit staff at the British Museum to examine the find made on his territory in 1898. In fact, he let them make selections from the hoard to fill the gaps in their own collection.³

Even if coins did receive notice on finding, many subsequently disappeared, particularly before the institution of proper Treasure Trove laws in India in 1878 (see Gupta, P.L. 1961c). But legislation alone cannot ensure the preservation or correct recording of finds. No area of India has a complete series of Treasure Trove reports, and those that do exist rarely provide a usable description. Usually the reference is limited to the number of coins found, the metal, and the fact that they are Roman. No mention is made of the issuer, type, weight or provenance. Nor does the registration of coins as Treasure Trove necessarily result in their safekeeping, since they were often either sold or distributed to museums without a proper record being kept; indeed the museums in India themselves were in the habit of passing duplicates to other institutions or even selling them. The Madras Museum thus disposed of the common issues of the Kathanganni hoard.⁴

These general problems concerning the recording of coin finds of course apply elsewhere, and not just in India. But the ‘efficacy of British Colonial administration in various areas’ (Raschke, M.G. 1978, p. 665) is an important factor governing our knowledge of Roman coins found in India. It is, however, difficult to assess this effect.

Table 1 shows the chronology of the finds. Records of the early finds, that is those found prior to 1856, are very variable; for example, very little is known of the Dharpul or Karur 1856 finds, and yet the Manikyala1830 find was quickly well-publicised because the coins were found in the famous tope during excavations. The discovery of a pot full of denarii at Vellalur in the rainy season of 1841 attracted the attention of Sir Walter Elliot, who published a full account of the hoard, which constitutes the first list of Roman coins found in India (Elliot, W. 1844). Between 1856 and 1878 no new finds were made. This is difficult to explain, because there was still great interest in Roman coins. George Bidie, part-time Superintendent of the Madras Museum from 1872-85, had initiated the collection of coins for display in the museum, which included what remained of the Kaliyampattur hoard. In 1874, he published a catalogue of the Roman gold coins prefacing his work with a general description of the coins, though in fact most of them can be satisfactorily ascribed (see Appendix III). He did, however, weigh the coins (Bidie, G. 1874). The increased incidence of finds after 1878 may be connected with the new Treasure Trove laws, though the Karur hoard found that year was only noticed by the Rev. Henry Little by good fortune, and not before two-thirds of the hoard (about 500 coins) had been melted down ‘to make bangles’ (Little, H. 1883), the rest passing to Little. He published his findings on the hoard, expanding on Bidie’s introduction to the catalogue.

The next work to deal at length with Roman coins found in India was a new catalogue of the Madras Museum collection. This was compiled by Edgar Thurston, the first full-time curator of the museum; he plagiarised his predecessor’s introductory comments, adding more information on recent finds culled from Little’s article. Thurston included silver and base issues, and marked out coins found in south India, without saying where the others had come from (Thurston, E. 1888).

The acquisition of the Vellalur 1891 hoard of denarii and the Vinukonda find of aurei by the Madras Museum led Thurston to revise his catalogue in 1894; he retained his earlier Introduction almost without change, but dispensed with the practice of including coin weights, and he omitted mention of those specimens not found in south India (Thurston, E. 1894).

Sewell’s paper, published some ten years later in 1904, was a new approach to the subject, for he supplied the first discussion of Roman coins found in India in list form, with references to each find. It was compiled according to emperor, which made it rather difficult to use, since hoards containing more than one issuer’s coins appeared in the list several times. His discussion of Roman trade with India, based on the coin finds, was pioneering, though he was prone to over-generalisation; in the main his conclusions are acceptable concerning the development of the trade, though his belief that social and moral conditions at Rome were the prime causes of the decline in trade would probably no longer be entertained seriously.

Rawlinson’s important book on the Roman trade with India, like Robertson’s famous account, scarcely mentioned coin finds (Rawlinson, H.G. 1916 and Robertson 1791). It was not until 1928, when E.H. Warmington published his classic work on the commerce between the Roman Empire and India, that an attempt was made to place the coin finds in an economic context. However, Warmington discussed in detail only eleven finds, and he marred his otherwise excellent and accurate work by making errors concerning the coin finds, such as including a coin of Commodus in the Kaliyampattur hoard, implying that the whole of the Pudukottai hoard was acquired by the British Museum, and so forth. His work was mainly based on the accounts of the classical authors, though he did try to take archaeological evidence into account as far as he could. The edition of 1928 was amended by the addition of an appendix in 1974.

When Sir Mortimer Wheeler arrived in India in 1944 to take over and revitalise the Archaeological Survey of India, one of his first projects was to provide a complete catalogue of Roman coins found in India, and to this end he employed two young students to search for information. The sensational finds at Arikamedu, including amphora sherds and a few pieces of Arretine ware, made a new appraisal of the coin evidence that much more important, and Wheeler’s first list appeared as an Appendix to the excavation report (Wheeler, R.E.M. et al. 1946). The finds were listed by provenance, and it represented the most important contribution to the study to date. Although it was a fairly complete list, it contained numerous errors, the worst being the duplication of the Karur finds under two names, Karur and Karuvur (though Warmington had already noted that they were one and the same), and the references were often incorrect.⁵ Happily, this was quickly emended, and a revised list appeared (Wheeler, R.E.M. 1951; the coin list is at pp. 374-81), with an essay on Roman contact with India which, for the first time, discussed the increasingly numerous archaeological materials. This new edition was not without some errors, but it has become the main source of reference on Indian finds of Roman coins for scholars in other fields.

The only new list to appear since Wheeler’s day was compiled by P.L. Gupta, in his catalogue of two hoards in the Andhra Pradesh Government Museum at Hyderabad (Gupta, P.L. 1965a). While an up-dated list was welcome, Gupta’s review of the coin finds was hardly any improvement on Wheeler’s; numerous errors limit the value of his work severely. Not only did he fail to include 16 finds for which information was available at that time, some of which had actually featured in Wheeler’s lists, but he clearly made little or no attempt to see bibliographical material for himself, nor did he approach any Indian museums to check on the location of the finds.⁶

No other lists have been published since 1965; a paper by Raschke (1978) includes a discussion of some aspects of the finds and the promise of a full paper in Roman coins found in India which has not so far appeared. The coins have received important attention in two books on Roman economics; Sture Bolin discussed them in his controversial work on the role of the denarius in the Roman economy, and Rodewald considered the finds in the context of the monetary policy of Tiberius (Bolin, S. 1958 and Rodewald, C. 1976). Both relied on Wheeler’s second list for information; Rodewald, although writing some eleven years after Gupta’s publication, appears not to have known about it. Mention of the Indian hoards has, of course, been made in many other numismatic papers, but the above represents a summary of all the main works of reference.

Notes

1.    Davidson, A. (1790) is the most often quoted reference to this hoard. Note that unless a specific reference is made, hoards of Roman coins from India will not receive footnotes in the text, as a full bibliography to each individual find is provided in Appendix I below.

2.    References to finds from Sri Lanka are listed at the end of Appendix I.

3.    Twenty coins were selected; see the section on the British Museum in Appendix II below, and the article by Hill, G.F. (1898).

4.    Various institutions received coins from this hoard, such as the Fitzwilliam Museum, Cambridge, and the British Museum, as well as museums in India. It has not proved possible to trace every coin, and none can certainly be identified in the Madras Museum; see Appendices II and III below.

5.    Wheeler, apparently, did not examine the coins, even in the Madras Museum, where the then curator, T.G. Aravamuthan, provided him with information. Some of the references in his list proved very hard to find; correct information is given at Appendix I.

6.    Except, obviously, in the Andhra Pradesh Museum. Gupta made repeated references in this work (1965a) to coins which ‘might exist in the Madras Museum’ - for example, Salihundam, Ongole, Mallayapalem and others. (cf. pp. 59-60 in Gupta’s catalogue).

CHAPTER TWO

Analysis of the coin finds

1. Geographical distribution

Map 1 shows all the known finds of Roman coins in India. There are two main concentrations, in the Coimbatore district of south India, comprising almost exclusively coins of the Julio-Claudian period (Map 2), and along the Krishna river in Andhra Pradesh (Map 3). The grouping in the Coimbatore district has long been noted, and its proximity to the famous beryl mines has been suggested as the reason for it.¹ Pliny specifically referred to the superiority of south Indian emeralds, though now only the beryl varieties can be obtained.² The Coimbatore area is also conveniently situated for the Palghat gap, a pass through the ranges of hills running from the Nilgiris in the north, through to the Annamalai and Cardamum hills in the south. This pass is the only way through from the Kerala coast into central southern India, an important route if regular travel around Cape Comorin was considered too hazardous. The finds are not on the hilly areas, the exception being the aureus (perhaps of Claudius) from Bishopsdown in Ootacamund. They are not found significantly close to the Cauvery River, which supports the theory that the Cape was rarely if ever rounded in the early Roman Imperial period. If ports on the Coromandel coast had been regularly used at this time, one would have expected more finds from the south-east of Tamil Nadu and along the rivers.³

Warmington (1974, p.292) thought that the Pudukottai coins were proof that the Cape was rounded, but the coins in this hoard were all worn, which may be an indication that the coins were circulated for some time before burial, in which case the location would have little relevance. M.P. Charlesworth (1951, p. 136) pointed out that only shallow draught native vessels could have passed through Adam’s Bridge, and that large western ships would have had to make the treacherous circumnavigation of Sri Lanka. A comparison between Maps 2 and 3 shows that the rounding of the Cape was probably undertaken in the later period, for the finds of coins dating to after the Julio-Claudian years (which are mainly second-century) have a wider distribution and the group of finds from the mouths of the K...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Acknowledgements
• Chapter 1 Historical introduction
• Chapter 2 Analysis of the coin finds
• Chapter 3 The historical significance of Roman coins found in India
• Chapter 4 Slashed coins and imitations
• Chapter 5 Conclusions
• Appendices:
• Maps
• Tables
• Plates
• References