Where Was Achaemenid India?

The Achaemenid Persians escape the historian.[1] Though they ruled the largest empire of the time for more than two centuries, much of what we know about them comes not from themselves but from other, often unsympathetic, observers. The Persians were perfectly capable of writing[2] but beyond bookkeeping and the anomalous Behistun (Bistun) inscription of Darius I, the Great[3] they appear to have been largely indifferent to the preservation of their version of the world.[4] Achaemenid rulers, satraps, and individuals emerge at irregular and widely separated intervals across three continents, but we know virtually nothing about them, particularly in comparison with their predecessors in Mesopotamia and Egypt and their fractious successors in Greece and Macedonia.[5] Achaemenid Persians did rule huge tracts of western Asia and neighboring regions, and their administrative arrangements in Central Asia lasted in one form or another at least until the coming of Islam.[6] However, the physical presence of the Achaemenid rulers over this vast extent remains subtle and difficult to distinguish from that of later groups.

Nowhere were the Achaemenids more unobstrusive than in India. In the territory of modern Afghanistan that was essentially an extension of the Iranian plateau lay the great satrapies of Aria, Arachosia, and Bactria,[7] with their capitals at Herat, Kandahar, and Balkh, respectively.[8] These large provinces were rich and important, and their locations are more or less established to everyone's satisfaction.

However, when one moves to the "Indian" satrapies at the eastern end of the empire, the picture becomes much less clear. There were four "Indian" satrapies, so classified on the basis of similar dress styles in the Persepolis reliefs.[9] The small, impoverished satrapies of Gandara and Sattagydia[10] nestled along the eastern side of Arachosia, on the plains and in the mountain ranges to the west of the middle Indus, in what is now Pakistan's North West Frontier Province.[11] The larger, but even poorer, satrapy of Maka straggled along the barren coast of Pakistani and Iranian Baluchistan and was avoided by any traveller who could take another route.[12] The little we know about these three satrapies leaves us with an impression of poverty and backwardness in a harsh environment.

The fourth "Indian" satrapy, India[13] itself, was totally different. Apparently conquered, annexed, or at least claimed by Darius the Great after 522 BCE[14] and after the voyage of Skylax of Caryanda to the mouth of the Indus from Kaspapayros,[15] India was described by Herodotus as fabulously rich, richer, in fact, than all the other satrapies together in terms of annual tribute, which was assessed in talents of gold dust rather than silver,[16] as was the tribute of all the others. The early imperial period Fortification Texts from Persepolis frequently mention India as a destination of parties of travellers.[17] Indian troops marched with Xerxes's army across the Hellespont and formed some of the contingents which tried to deflect Alexander's advance at Gaugamela.[18] Even after two hundred years there was considerable residual loyalty to the Achaemenid crown in its eastern provinces, although these troops may have been mercenaries. But we have no mention of India's capital city, virtually no idea of how its people looked,[19] no names for its satraps, and practically no notion of how the satrapy (if it ever was a satrapy) fitted into the imperial system.

So where was Achaemenid India? We must begin by deciding whether it existed at all as an independent entity, an issue debated hotly by Indian scholars. This is not an idle question. The basis of the debate has been the fact that (1) there was no historically attested "Achaemenid" presence in the northern Indus region at the time of Alexander's incursion in the late fourth century BCE, and (2) there is a complete lack of any identifiable mention of Persians in Indian texts from the fifth and fourth centuries BCE

In historical terms the most important single source is the Astadhyayi of the Sanskrit grammarian Panini, who was born near Taxila and whose main work was written in the fifth century BCE[20] In this work, Panini used geographical examples to illustrate his grammatical points. He seems not to have mentioned the Achaemenid empire or the Persians, and his description of governments revealed a preponderance of small, politically autonomous states in the northern Punjab and the northern Indus region which did not accord with any notions we might have of centralized Achaemenid control of the region. He does not mention the satrapal system, and there are no unmistakably Achaemenid names in his text. Mughal's work in the region of Multan on the central Indus confirmed that there was little, if any, archaeological contact in the mid-first millennium BCE between that part of the Indus and cultures further west, which emphasizes the role of the Indus River as a frontier.[21]

Indian archaeologists and historians have been divided on the question of Achaemenid control in India for at least fifty years.[22] Regrettably, much of the argumentation has stemmed from a priori assumptions about non-Indian control of the subcontinent. The results of these searches remain inconclusive, and there has been little help from archaeology.

Recently another source of the first importance became available to scholars. These, the Elamite administrative records known as the "Persepolis Fortification Texts," are the single largest source of information about the early phase of the Persian empire to have emerged in years. The vital point about these tablets is that they were not intended for public consumption but for bureaucratic convenience: lies, bombast, and posturing were all very well in public proclamations, but there would be no conceivable reason to preserve them in bookkeepers' files recording ration issues and kept, as they were, in a language and script accessible to few and stored in a capital city rarely visited by the Achaemenid rulers.

Among the many types of texts identified by Hallock were those he classed as Q Texts, dealing with the issue of rations to parties of travellers.[23] Hallock recorded 303 separate travel ration texts (PF 1285-1579 and PF 2049-2057), of which 32 were issues to travellers to the eastern part of the empire, including India.[24] In addition, another 14 texts[25] mention men from India.

The crucial point of this is that by the later part of Darius the Great' reign (522-485 BCE), the central bureaucracy treated "India" as just another destination for imperial travellers. We are not told where it was: obviously, the bookkeepers knew, or at least dealt with people who knew. But travel to India was on exactly the same footing in terms of supplies and support as travel to any other region known to have been in the empire. As far as the writers of the Fortification Texts were concerned (and they, if anyone, would have known the facts) India was Achaemenid.

With Hallock's work in hand, the debate must shift from whether Achaemenid India existed at all, to where it was. To be a candidate for the Achaemenid Persian administrative center in India, an archaeological site must have had certain specific features. It must obviously have had an occupation dating to the late sixth or early fifth century BCE It must have been large and complex, and must have been reasonably centrally located in its district and on major communication routes. It must also have been the only site of its type in the vicinity because, by analogy with known Achaemenid administrative centers, it is highly unlikely that the central administration in Persia would have chosen a city which was simply one of a large number of similar sites.[26] Our eyes are drawn irresistibly to the enormous and long-lived site of Taxila, in the northern Punjab east of the Indus River. This is not an original idea. The site's principal excavator, John Marshall, proposed this identification in his final report.[27] However, Marshall's identification was based on Classical sources and a frankly pro-Western bias, distressing to many of his Indian readers. There was little in his discussion of archaeological material that would support such an identification, and the Elamite evidence was not available to him. Marshall proposed that the Bhir Mound at Taxilacould have been founded by Darius the Great after about 518 BCE, although he admitted that he had found no tangible evidence to support this view. The principal excavations on the Bhir Mound were not conducted with much regard for stratigraphic recording, and the collected pottery was in any case published in such a manner as to prevent much in the way of detailed analysis. The subsequent work of Sharif[28] was intended to correct this imbalance and to check the stratigraphic sequence of the earliest large mound in the Taxila area. These broadly succeeded in their objectives but were of such limited extent that they yielded no information for the architecture of the Bhir Mound.

Marshall's treatment of the pottery from the Taxila sites was extraordinarily misleading because his presentation of the ceramics took the form of a series of pot types from all the Taxila sites with no regard for stratigraphical and, therefore, chronological considerations. However, Sharif's more careful excavations revealed a well-defined collection of grey wares in his two lowest strata,[29] ceramics that were derived in some degree from the Painted Grey Ware Culture first studied at Hastinapura and now examined more generally by Tripathi.[30] The link is important because it confirmed that the earlier material from Taxila's Bhir Mound was, on the whole, within the northern Indian tradition of the first millennium BCE It also broadly confirmed the later sixth century BCE date for the foundation of the Bhir Mound proposed by Marshall, although the site appears not to have been founded by Achaemenids.

In contrast to general north Indian links at the Bhir Mound, there were certain "Western" pottery styles common to Taxila/Bhir Mound and contemporaneous sites further north and west, in Swat, across the Indus, and on to the Iranian plateau. One such pottery style was a highnecked water jar common throughout the occupation of the Bhir Mound[31] and found in numerous variants at the northwestern Pakistani site of Balambat,[32] which possibly had an Achaemenid level, as well as in a slightly later context at Charsada.[33] The same shape was found in more secure Achaemenid contexts at Kandahar in southern Afghanistan, Tepe Yahya Period II in Kerman, and possibly in the Persepolis region and Nad-i Ali in southeastern Iran.[34] A second shape found on several northwestern Pakistani sites was a wide shallow bowl with convex sides.[35] The excavators in the Swat valley have dated this shape to their Swat Period VI, or the first half of the first millennium BCE[36] The shape was present throughout the main occupation of the Bala Hisar at Charsada and occurred in all levels of the Bhir Mound at Taxila. The shape was also found at Kandahar but not to the west or north: in both Iran and Central Asia potters preferred vessels with incurved sides.[37]

Obviously, these are tentative suggestions rather than solid facts. However, it is noticeable that some of the early Bhir Mound pottery from Taxila, whose mid-to-late sixth century BCE date is not in dispute, bears a surprising resemblance to material from contemporaneous Kandahar and southern Iran, where an Achaemenid presence is certain. This would indicate that contacts of various sorts were maintained between the Kandahar region of southern Afghanistan (and hence to the Iranian plateau proper) and the upper Punjab by way of the Gumal Pass area. In geographical terms such a direct route between the fertile Punjab and the important nexus of trade routes meeting at Kandahar eliminated the need to negotiate the mountainous terrain between modern Ghazni, Gardez, Kabul, and the Peshawar vale.[38] It also passed through solidly Achaemenid territory in Sattagydia[39] and reiterates the importance to the empire of Sattagydia as a way station on the route to India.[40] Furthermore, it put the Achaemenid rulers in at least nominal control of a belt of contiguous territory including the whole of the Kabul River valley, from its source to its junction with the Indus, which would have allowed movement down the Kabul River and eventually to the Indian Ocean, as was done by Skylax of Caryanda.

So we come back to our original question: Where was Achaemenid India? We know now that it existed, although we do not know when it ceased to be a part of the empire. There are apparently no other sites in the region with Taxila's potential. But where are the Achaemenid remains? And why do we not have more historical evidence from Indian sources?

The question of Achaemenid remains is settled quickly. We simply do not know enough about eastern Achaemenid architecture to say whether the buildings at Taxila/Bhir Mound's earliest levels were Achaemenid Persian or not. The only substantial excavations of an eastern Achaemenid site (Kandahar) remain largely unpublished, except in preliminary form.[41]

As for the silence of Indian historical records, it may be that the Achaemenid presence east of the Indus was so ephemeral, or so short-lived, that it was never really noticed, which would tie in with Panini's failure to mention it, as we noted above. After all, just as Taxila was on the eastern edge of the western world, it was simultaneously on the western edge of the Indian world, well away from the central region of the upper Ganges and open to many influences from the north and west. More excavation at Taxila would help solve the question of whether the Persian presence in northwestern India was based there, as would more work with the existing material in the Taxila museum. Until and unless another contender is found, Taxila's Bhir Mound remains the most plausible candidate for the capital of Achaemenid India.[42]

Professor David Bivar first introduced me to Achaemenid Iran at the School of Oriental and African Studies in London and remained both a friendly critic and inspiring teacher when I went on to Corpus Christi College, his alma mater at Oxford. It is to him that I owe my continuing interest in the history and culture of the borderlands between Afghanistan, Pakistan, India, and China. I therefore offer this essay to him to thank him, in a small way, for the guidance and good example he has provided to me and to many others over his entire career. I also thank my wife, Monica Barnes, for reading numerous drafts of this paper.

[1] There is no conventional shorthand way to refer to the rulers of the first Persian empire. To themselves they were "Achaemenids" (Old Persian [OP] Haxdmanisiyd), based on their clan affiliation and descent from their eponymous ancestor Haxdmanis (OP) or Achaemenes (Greek), while to the Greeks and people of the Near East, they were Persians, which merely meant they came from Persis in southwestern Iran and which was not the whole truth because many Persians were not Achaemenids. For the purposes of this essay, the terms "Achaemenid," "Persian," and "Achaemenid Persian" will be used interchangeably to represent the ruling dynasty of the Persian empire from the mid-sixth century to 332 B.C.

[2] See G. Windfuhr, "Notes on the Old Persian Signs," Indo-Iranian Journal 12 (1969-1970), pp. 12125, on the creation of the Persian cuneiform syllabary and its independence from other cuneiform systems. The Achaemenid administrators used whatever other languages were appropriate to their surroundings: for example, Aramaic (R. A. Bowman, Aramaic Ritual Texts from Persepolis, Oriental Institute Publication 91 [Chicago, 1970]) and Elamite (R. T. Hallock, Persepolis Fortification Texts, Oriental Institute Publication 92 [Chicago, 1969]).

[3] R. G. Kent, Old Persian Grammar, Texts, Lexicon, 2d ed. (New Haven, 1953[, text DB.

[4] The editio princeps of OP texts (Kent, Persian) contains 96 attested and 8 spurious texts. Subsequent work has revealed very few additional texts, most of which are laconic and stereotyped building inscriptions.

[5] For recent analyses of the role of the satrap and the nature of the satrapy in Achaemenid Iran, see C. Herrenschmidt, "Designation de Empire et concepts politiques de Darius IeI d'apres les inscriptions en vieux-perse," Stlr (1976), pp. 33-65; W. Hinz, Altiranisches Sprachgut der Nebeniiberlieferiingen: Gottinger Orientforschungen, Reihe 3, Iranica, vol. 3 (Wiesbaden, 1975[; and D. M. Lewis, Sparta and Persia, Cincinnati Classical Studies, n.s. 1 (Leiden, 1977).

[6] R. N. Frye, The Golden Age of Persia (London, 1975[, pp. 35-37.

[7] OP Haraiva, Harauvatis, and Bdxtris (Kent, Persian, p. 137, DNa 22-4).

[8] Pre-Islamic Herat remains to be explored. For Kandahar see S. W. Helms, "Excavations at 'The City and Famous Fortress of Kandahar, the Foremost Place in All of Asia,"' Afghan Studies 3-4 (1982), pp. 1-24. For Balkh see J.-Cl. Gardin, Ceramiques de Bactres, MDAFA 15 (Paris, 1957).

[9] G. Walser, Die Volkerschaften auf den Reliefs von Persepolis, Teheraner Forschungen, vol. 2 (Berlin, 1966), pp. 89-95.

[10] OP Gaddra and Oatagus (Kent, Persian, p. 137, DNa 24-5).

[11] For Sattagydia, see D. Fleming, "Achaemenid Sattagydia and the Geography of Vivana's Campaign (DB III, 54-75)," JRAS (1982.2), pp. 102-12. For Gandara, G. Tucci, "On Swat: The Dards and Connected Problems," EW, n.s. 27 (1977), pp. 9-103.

[12] The area of Maka is now known as the Makran coast. Apparently, it has always been poor, and there is little evidence, ecological or otherwise, to suggest that it was once more prosperous. See G. Le Strange, Lands of the Eastern Caliphate (London, 1905), p. 329; D. W. Engels, Alexander the Great and the Logistics of the Macedonian Army (Berkeley, 1978), pp. 114-17.

[13] OP Hindus (Kent, Persian, p. 137, DNa 25.

[14] It is not mentioned in DB but is in the later DNa.

[15] Herodotus 4.44.

[16] What exactly was intended by Herodotus' tribute list is far from clear. The argument that the Persepolis reliefs show vassals in the act of paying tribute at the New Year festival (A. T. Olmstead, History of the Persian Empire [Chicago, 1948], p. 238) remains at most an interesting theory because we have no written mention of this event taking place. The Indian tribute was 360 Euboean talents of gold dust per year (Herodotus 3.95). The whole question of whether anything remotely like this fabulous amount of precious metal was ever brought from any part of the empire in one year is too complex to consider here.

[17] Hallock, Fortifications Texts.

[18] Herodotus 7.65; Arrian Anabasis 3.8.

[19] M. D. Roaf, "The Subject Peoples on the Base of the Statue of Darius," CDAF14 (1974), pp. 73-160.

[20] J. E. Schwartzberg, ed., A Historical Atlas of South Asia, chap. 3, "From the Vedic through the Classical Age," section IIIB, "The Pre-Mauryan and Mauryan Periods" (Chicago, 1978), pp. 165-72.

[21] M. R. Mughal, "Excavations at Tulamba, West Pakistan," Pakistan Archaeology 4 (1967), pp. 41-66.

[22] S. Asthana, "The Period of the Early Historical Culture," chap. 8 of History and Archaeology of India's Contacts with Other Countries from Earliest Times to 300 B.C. (Delhi, 1976); S. Chattopadhyaya, "The Achaemenids and India," Indian Historical Quarterly 26.2 ( 1950), pp. 100-17; P. H. L. Eggermont, "Alexander's Campaigns in Sind and Baluchistan and the Siege of the Brahmin Town of Harmatelia," Orientalia Lovaniensia Analecta 3 (1975); R. C. Majumdar, "Achaemenian Rule in India," Indian Historical Quarterly 25 (1949), pp. 153-65; A. V. W. Jackson, "The Persian Dominions in Northern India Down to the Time of Alexander's Invasion," in Cambridge History of India, vol. 1, Ancient India, ed. E. J. Rapson (Cambridge, 1935).

[23] Hallock, Fortification Texts, p. 6, points out that most rations were issued for one day, which meant that there had to have been elaborate supply arrangements along the entire route, along with reliable and acceptable means of payment or credit.

[24] Arachosia 8, Aria 4, Bactria 2, India 8, Kerman 6, Gandara 3, and Sagarta 1. The Indian texts are PF 1318, 1383, 1397, 1524, 1552, 1556, 1572, 2057.

[25] PF 785, 1317, 1397, 1410, 1425, 1437, 1511, 1525, 1529, 1548, 1552, 1558, 1572, 1601.

[26] Balkh, Kandahar, and Herat are all relatively isolated, although huge. Charsada, the ancient Pushkalavati and probable capital of Gandara (M. Wheeler, Charsada: A Metropolis on the North-West Frontier. Being an Account of the Excavations of 1958 [Oxford, 1962]), is some distance from the later Islamic centers of the Peshawar vale.

[27] J. Marshall, Taxila (Cambridge, 1951), pp. 1213.

[28] M. Sharif, "Excavations at Bhir Mound, Taxila," Pakistan Archaeology 6 (1969).

[29] Marshall, Taxila, pp. 398-438 and pls. 121-31; Sharif, "Excavations at Bhir Mound," p. 77.

[30] B. K. Thapar, "The Pottery," in B. B. Lal, ed., "Explorations at Hastinapura and Other Explorations in the Upper Ganga and Sutlej Basins 1950-52," Ancient India 10-11 (1954-1955), pp. 5-151; V. Tripathi, The Painted Grey Ware: An Iron Age Culture of North India (Delhi, 1976).

[31] Marshall, Taxila, pl. 122-54; Sharif, "Excavations at Bhir Mound," fig. 10-8, 9.

[32] A. H. Dani, "Timargarha and Gandhara Grave Culture," Ancient Pakistan 3 (1967), pp. 1-407, esp. figs. 57-8 and 60-2.

[33] Wheeler, Charsada, p. 54; layer 32 was attributed by Wheeler to the late fourth to early third centuries B.C.

[34] A. McNicoll, "Excavations at Kandahar, 1975: Second Interim Report," Afghan Studies 1 (1978), pp. 41-66, esp. fig. 6-3 and possibly 6-4; C. LambergKarlovsky, Excavations at Tepe Yahya, 1967-1969: Progress Report 1, American School of Prehistoric Research Bulletin 27 (Cambridge, Mass., 1970), p. 26, fig. 8-K, O; W. A. Sumner, "Cultural Development in the Kur River Basin, Iran," Ph.D. diss., University of Pennsylvania, 1972, pl. XLII-T[?] [sic]; G. Dales, New Excavations at Nad-i Ali (Sorkh Dagh), Afghanistan, Research Monograph Series, no. 16, Center for South and South-East Asian Studies (Berkeley, 1977), pl. 132, Type A1.

[35] . Balambat: Dani, "Timargarha and Gandhara Grave Culture," fig. 58-13, 14, 15; Charsada: Wheeler, Charsada, fig. 11-10; Taxila/Bhir Mound: Marshall, Taxila, class XV; Sharif, "Excavations at Bhir Mound," fig. 11-8; Swat valley: G. Stacul, "Excavation near Ghaligai (1968) and Chronological Sequence of Protohistorical Cultures in the Swat Valley," EW, n.s. 19 /1969), pp. 44-91, esp. fig. 13-A.

[36] Stacul, "Excavation near Ghaligai," p. 85.

[37] D. Whitehouse, "Excavations at Kandahar, 1974: First Interim Report," Afghan Studies 1 /1978), pp. 9-39, fig. 6-64, 65, 66, 67, 68; McNicoll, "Excavations at Kandahar, 1975," fig. 5-9; J.-Cl. Gardin and B. Lyonnet, "La prospection archeologique de la Bactriane orientale /1974-1978): Premiers resultats," Mesopotamia 13-14 (1978-1979), pp. 99-154 and fig. 10; Dales, "New Excavations at Nad-i Ali (Sorkh Dagh)," pl. 19.

[38] The basic route through the Gumal area is laid out in H. G. Raverty, Notes on Afghanistan and Parts of Baluchistan (London, 1888), p. 5.

[39] Fleming, "Achaemenid Sattagydia."

[40] A point forcibly made by A. Toynbee, "The Administrative Geography of the Achaemenid Empire," in A Study of History (London, 1954), vol. 7, annex VI C(ii)(c)3, p. 640.

[41] An edition of the 1974-1975 seasons is in the final stages of preparation and will be published by British Archaeological Reports, International Series, in 1994. The work of the 1976-1978 seasons of Kandahar is available only in summary form (Helms, "Excavations at 'The City and Famous Fortress of Kandahar'," and references therein).

[42] This contradicts the conclusion in Fleming, "Achaemenid Sattagydia," n. 16.

By: David Fleming

Copyright © 1998-2008 The Circle of Ancient Iranian Studies (CAIS)