Coins, Art, and Chronology: Essays on the pre-Islamic History of the Indo-Iranian Borderlands
Edited by Michael Alram and Deborah E. Klimburg-Salter
Verlag Der Osterreichischen Akademie Der Wissenschaften, Wien 1999

Evidence from the Rabatak Inscription of Kanishka I

Joe Cribb

| page 1 | page 2 | page 3 | page 4 |


[1] During the last twenty years I have been addressing a large part of my research to the question of Kushan chronology. This is not an exercise in isolation or an intellectual game, but an attempt to provide a firm framework for the study of the history, culture, religion, and art of the Kushan period.

[2] The relationships between the Kushans and their neighbours and the more distant powers of Rome, China and Iran can only be understood if the date of these contacts is known. The mixture of classical, Indian and Iranian influences in Kushan cultural and religious remains has been extensively documented, but the meaning of these influences also needs a temporal context. Our understanding of the religious attitudes prevalent in Iran and India during the Kushan period will also be enhanced by the establishment of a chronological framework. It is perhaps in the field of art history that the importance of the Kushan question is most dramatically felt, where sculptures, dated in uncertain eras, and archaeological contexts, no matter how well stratified, remain undatable until an answer which can gain wide acceptance is found.

[3] Johanna van Lohuizen-de Leeuw has aptly expressed the necessity of solving the chronological problems: "The uncertainty regarding the eras, and particularly the one which was customary in the time of Emperor Kaniska and his successors, has been a stumbling block to every discussion that went somewhat deeper into the problems of the history of North India" (van Lohuizen-de Leeuw 1949: 1).

[4] The problems inherent in this question, which currently allow a range of up to 200 years of uncertainty, arc obvious. Compare for example the dates for year 1 of the Kanishka Era used in two recent exhibitions with a special focus on the Kushans: AD 78 used for the 1985 exhibition at Cleveland, Ohio, Kushan Sculpture, Images from Early India, and AD 232 used for the 1996 exhibition at Vienna, Weihrauch und Seide, Alte Kulturen an der Seidenstrasse. The two dating systems do not reflect the development of' evidence and analysis during the ten years which separate the two exhibitions, but two different approaches to the question. An answer cannot be reached by averaging out the answers or by holding a vote among scholars as to which view they most favour. Neither an average, nor a consensus provide the solution. Just as in the natural sciences, where there can only be one correct description of a given phenomenon, so there must also be one accurate answer to the dates of the Kushan kings which renders all other solutions incorrect.

[5] Having read extensively in the literature on the Kushan problem, I came to an understanding that the difficulties so widely experienced were because many of the answers were based on too narrow an approach to the problem. Enquiries have too often been focused on the justification of the choice of a particular date for the beginning of the reign of the most famous Kushan king Kanishka 1, or on the overinterpretation of one particular piece of evidence without an analysis of its context and the broader implications for the chronology of his predecessors and successors. There has also been a deep-seated reluctance to re-examine the evidence, often numismatic, underpinning many of the theories presented by the pioneers of the topic more than a hundred years ago.

[6] In beginning to look at this question I decided to avoid the assertion that the exact date of the beginning of Kanishka I's reign could be precisely fixed to a particular year until appropriately precise evidence emerged. As a practicing museum curator I often have to express an opinion on the dates of Kushan kings, both in response to enquiries and in the expression of general information in popular exhibitions and publications. When making such statements I always take care to couch them in terms which retain the level of uncertainty inherent in our present state of knowledge.

[7] In my research I have attempted to adopt similar caution and to focus on the question of the level of uncertainty by examining, one after another, the bodies of evidence relevant to chronology in as much detail as possible, and in this way to continue broadening the basis on which the level of uncertainty can be gradually reduced. I have learned the extent to which numismatic evidence has underpinned the whole study of the Kushans, right from the time of the first scholars like Masson and Prinsep who studied their coins without any clear understanding of who the Kushans were. A key issue stressed by many looking at the numismatic evidence has been the question of establishing sequences in the numismatic material and this has been at the centre of each piece of research undertaken. An additional support to the use of numismatic sequences can be their relationship to dated inscriptions, provided the eras involved are correctly identified. In some cases the dated inscriptions add quantitive measurement to the reigns revealed by the sequences.

[8] This paper presents the evidence provided by a landmark discovery in Kushan studies, so remarkable that I have already heard rumours that some scholars would prefer it to be a forgery rather than believe its evidence. I am pleased to acknowledge that it undermines several of the conclusions I have in print on the interpretation of Kushan numismatic evidence. I am pleased because the levels of uncertainty which I see as my prime concern are dramatically reduced by its discovery. The most important revelation of the document, from the point of view of chronology, are its solution of the Soter Megas question and its confirmation of the early Chinese account of the origins of Kushan rule under the leadership of Kujula Kadphises and the Kushan expansion into India in the reign of his son. Uncertainty remains, but for the first time we are able to make a list in correct order of the Kushan kings from their rise to power during the first century AD and match this list to the sequence suggested by their coins. We are still lacking the lengths of the reigns of the early and late Kushan kings, but we can make some approximate estimates of the chronological contexts of their reigns.

[9] I will now present a brief account of this new discovery, and discuss its significance for reducing our levels of uncertainty about the absolute chronology of the early Kushan kings. In the first appendix I have also attempted to throw light, on the basis of numismatic and inscriptional evidence, upon the spread of early Kushan rule and in the second I have reviewed the evidence of the various eras, both actual and imagined, which help create the chronological context within which the early Kushans and their contemporaries existed. This review concludes with a reassessment of the Kushan Era used in the Dasht-e Nawur inscription which points to an even tighter reduction of the levels of uncertainty regarding the date of the first year of Kanishka I.


[10] I would like to acknowledge the importance to my work on Kushan coins of the generous support of the Neil Kreitman Foundation, and to express my gratitude to Neil Kreitman and the Trustees of the Foundation for making this possible.

[11] My thanks are also due to Professor Ikuo Hirayama and Mrs. Michiko Hirayama, Bill and Katie Barrett, Katsumi Tanabe, Nicholas Sims-Williams, Osmund Bopearachchi, Frantz Grenet, Yevgeny Zeymal, Michael Alram, Bob Senior, Chantal Fabregues and Aman ur-Rahman and my colleagues Liz Errington, Helen Wang, Vesta Curtis and Andrew Burnett for their encouragement, advice and support during this study.

[12] I would also like to express my deep debt to my sadly missed colleagues Nicholas Lowick and Martin Price who both taught me to understand the importance of numismatics.

A new Bactrian inscription of Kanishka the Great

[13] In 1993, a photograph arrived on my desk which has demanded a fundamental reassessment of our understanding of the history of the early Kushan kings. Sent by a British charity worker, Tim Porter at the request of Sayed Jaffa, Governor of Baghlan in northern Afghanistan, the photograph showed a rectangular piece of stone 90 cm wide, 50 cm high and 25 cm thick. All over one face of the stone was an inscription written in Greek letters. A preliminary examination of the photograph showed the inscription was written in Bactrian, and mentioned the names of Kushan kings.

[14] The stone bearing the inscription was found in March 1993, together with fragments of a sculpture of a lion and architectural elements, in a hill known locally as the Kafirs' Castle, in the region called Rabatak forty kilometers north of Pul-i Khumri (Ball 1982: 1226, no. 944).

[15] I invited Professor Nicholas Sims-Williams of the School of Oriental and African Studies, London, to work on the stone with me, and the following discussion is based on his reading and translation of the text. Our joint paper on the Rabatak inscription has now been published (N. Sims-Williams and J. Cribb 1996).

[16] The Rabatak Inscription

(Translation by Nicholas Sims-Williams)

[1] . . . of the great salvation, Kanishka the Kushan, the righteous, the just, the autocrat, the god [2]

*worthy of worship, who has obtained the kingship from Nana and from all the gods, who has

*inaugurated the year one [3] as the gods pleased. And he *issued a Greek *edict (and) then he put it into Aryan.

[4] In the year one it has been proclaimed unto India, unto the *whole of the realm of the *kshatriyas, that (as for) [5] them - both the (city of) . . . and the (city of) Saketa, and the (city of) Kausambi, and the (city of) Pataliputra, as far as the (city of) Sri-Campa [6] - whatever rulers and other *important persons (they might have) he had submitted to (his) will, and he had submitted all [7] India to (his) will.

Then King Kanishka gave orders to Shafar the karalrang [8] *at this . . . to make the sanctuary which is called B . . . ab, in the *plain of Ka . . ., for these [9] gods, (of) whom the . . . *glorious Umma leads the *service here, (namely:) the *lady Nana and the [10] lady Umma, Aurmuzd, the gracious one, Sroshard, Narasa, (and) Mihr. [interlinear text: . . . and he is called Maaseno, and he is called Bizago]

And he likewise [11] gave orders to make images of these gods who are written above, and [12] he gave orders to make (them) for these kings: for King Kujula Kadphises (his) great [13] grandfather, and for King Vima Taktu, (his) grandfather, and for King Vima Kadphises [14] (his) father, and *also for himself, King Kanishka.

Then, as the king of kings, the devaputra [15] . . . had given orders to do, Shafar the karalrang made this sanctuary.

[16] [Then . . .] the karalrang, and Shafar the karalrang, and Nukunzuk [led] the worship [17] [according to] the (king's) command.

(As for) *these gods who are written here - may they [keep] the [18] king of kings, Kanishka the Kushan, for ever healthy, *secure, (and) victorious.

[19] And [when] the devaputra, the *ruler of all India from the year one to the year *one *thousand, [20] had *founded the sanctuary in the year one, then *also to the . . . year. . . [21] according to the king's command . . . (and) it was given also to the . . ., (and) it was given also to the . . ., (and) also to [22] . . . the king gave an *endowment to the gods, and . . .

The historical implications and numismatic context of the Rabatak inscription

[17] The most startling revelation of the Rabatak inscription is the previously unrecognised Kushan king, Vima I Tak[to] (line 13), whose position among the early Kushan kings is clearly indicated. Now that we know his name we can recognize his portrait sculpture, and his name in two other stone inscriptions and on several coins which were previously associated with the king identified in this inscription as his son Vima II Kadphises. He can also be recognized as the anonymous issuer of the Kushan "Soter Megas" coins.

[18] His portrait sculpture is the massive stone figure of a seated king found during the excavations of the Mat sanctuary, near Mathura. It can be identified as his portrait on the basis of its inscription, which calls him Vima Tak . . . (in Brahmi). He is also the Vima of the Dasht-e Nawur inscription, where he is named Oohmo Tak... (in Bactrian).

[19] His name appears on three different groups of coins. The first are the bull and camel coins with the name Vema (in Kharoshthi), which I mistakenly attributed to Vima II Kadphises when I first read them (Cribb 1981). I have since found further specimens reading Vema Tak . . . (Gobl 1993: no. 45). The second consists of two coins of the "Soter Megas" type, with both Greek and Kharoshthi inscriptions. The Greek has the usual anonymous inscription, but on an example in the Lahore Museum and on another in the British Museum the king's name Vema has been added to the end of the Kharoshthi inscription. The third series is represented by a single coin, now in the British Museum, but published almost a hundred years ago (Smith 1898: no. VII). This coin is a copper drachma, with a seated image, with the same posture as the Mat sculpture, accompanied by the Soter Megas symbol and the Bactrian inscription Oohmo Tak . . . (for further details and illustrations of these coins see Sims-Williams and Cribb 1996).

[20] As exciting as the revelation of a new king might be, in many ways it is the context in which this name appears that is the most important contribution to Kushan history. Vima I Tak[to] is named as the second of three ancestors of Kanishka I. Their family relationships are also described: great grandfather Kujula Kadphises, grandfather Vima Takto, father Vima Kadphises.

[21] For the first time we have a firm structure for the history of the early Kushan kings, affirming Kanishka's direct connection to the Kushan kings before him. The Rabatak inscription, written on the orders of Kanishka I probably at the end of his first year, makes it clear that the Kushan kings traced back the origin of their dynasty to Kujula Kadphises and that they considered there to be a direct succession from him to Kanishka I through his son the new king Vima and his grandson Vima II Kadphises. All four members of the dynasty were designated shao, king, and therefore were thought to have had their own distinct reigns.

Kushan Chronology

[22] This sequence of four Kushan kings can be matched to the record of the early Kushans presented by both coins and other inscriptions.

The Coin Sequence

[23] The coins attributable to the early Kushans suggest the sequence of the first four kings: Kujula Kadphises, the anonymous Soter Megas ruler now identifiable as Vima I Takto, Vima II Kadphises and Kanishka I.

[24] There are several groups of coins of Kujula Kadphises. Their position at the beginning of the Kushan series is clear from their derivative types which copy Indo-Greek, Indo-Parthian, Indo-Scythian and Roman coin designs. Overstrikes and the copied designs show that Kujula Kadphises issued coins south of the Hindu Kush as a contemporary and successor to the Indo-Parthian rulers Gondophares and Abdagases in Kabul, to Abdagases and Sasan in Gandhara/Taxila, and to Sasan in the Sind, and conqueror of their Indo-Scythian neighbour the satrap Zeionises (Jihonika) in Kashmir (Cribb 1993).

[25] The next Kushan coins are the Soter Megas series, issued after Kujula Kadphises' coinages, as demonstrated by their use of designs copied from and overstruck on the coins of Sasan, the last Indo-Parthian ruler in Taxila and Gandhara and a contemporary of Kujula Kadphises. Alongside the Soter Megas coins are those issued in the name of Vima I Takto already identified above, some of which (bull and came] types) follow directly issues of Kujula Kadphises.

[26] After the coins of the Soter Megas types the next Kushan issue consists of the coins in the name of Vima II Kadphises. They copy the script styles and part of the inscriptions of the Soter Megas coins; they also continue the use of the 'Attic' weight standard reintroduced for the Soter Megas coinage, but with new denominations, including gold.

[27] The use of gold is continued by Kanishka I. The script style and use of Greek language on Kanishka's first issue copy exactly those of Vima II Kadphises' last issues.

[28] The reign after Kanishka I is confirmed as that of Huvishka by a die link between a late coin of Kanishka I (Gobl 1984: no. 80) and an early coin of Huvishka (Gobl 1984: no. 314).

The Inscription Sequence

[29] The sequence of kings named in inscriptions other than the Rabatak inscription also features a series of four early Kushan kings, but the first king is called Kushan rather than Kujula Kadphises. The sequence of Vima I Tak[to] and Vima II Kadphises is in agreement with their dated inscriptions in the same unknown era: the year 279 Dasht-e Nawur inscription of Vima I Tak[to] preceding the year 284/7 Khalatse inscription of Vima II Kadphises (previously misread as 184 or 187, corrected in Cribb 'Numismatic Perspectives on Chronology in the Crossroads of Asia', in press).

[30] The inscriptions of the king called simply Kushan are dated in the years 122 and 136 (Azes Era), i.e. 19 and 33 years after the year 103 inscription of Gondophares (see below for a further discussion of these inscriptions). This king Kushan occupied the same chronological slot, as a contemporary of Gondophares' successors, as Kujula Kadphises.

[31] The inscription on the Mat sculpture of Vima I Tak[tol naming him as Kushanaputro, i.e. son of Kushan, points to the same identification of king Kushan as Kujula Kadphises, the father of Vima I Tak[to]. The so-called Heraus coins are also issued in the name of king Kushan, but identified on a bilingual issue as coins of Kujula Kadphises (Cribb 1993).

[32] The inscriptions of Kanishka I start a new era, based on his regnal dates, a practice already apparent in his first year, according to the Rabatak inscription. He can be placed after the other three early Kushan kings because the names of his successors for the next 100 years are clearly indicated in later inscriptions.

Kushan king list, based on coin and inscription sequences: Table 1

g o  t o  p a g e  2