¹ The major edition of the Old Persian (OP) text of the Behistun inscription is that of Kent, R. G., Old Persian: grammar, texts, lexicon, second edition, New Haven, 1953, 116–35.Google Scholar The Elamite text has been published by King, L. W. and Thompson, R. C., The sculptures and inscriptions of Darius the Great on the rock of Behistun in Persia, London, 1907,Google Scholar and by Weissbach, S. H., Die Keilinschriften der Achämeniden, Leipzig, 1911Google Scholar (reprinted Leipzig, 1968). There are important comments on parts of the Elamite text by Cameron, G. G., “The Elamite version of the Behistun inscription”, Journal of Cuneiform Studies, XIV, 1960, 59–68.CrossRefGoogle Scholar The definitive version of the Babylonian text is now that of von Voigtlander, Elizabeth, The Bisitun inscription of Darius the Great: Babylonian version, London, 1978Google Scholar (Corpus Inscriptionum Iranicarum, Pt. I, Vol. II, Texts I).Google Scholar

In the notes below, the abbreviation DB + column and line-number refers to the OP version of the text, while DB/Bab + line-number refers to Von Voigtlander, and DB/E1 + section and line-number to King and Thompson. Following Kent, names from the OP texts are printed without the use of diacritics.

² Cameron, G. G., Persepolis Treasury Tablets, Chicago, 1948;Google ScholarHallock, R. T., Persepolis fortification tablets, Chicago, 1969;Google Scholaridem, “Selected fortification texts”, Cahiers de la Délégation Archéologique Française en Iran, VIII, 1978, 109–36.Google Scholar

³ DB I, 16–17; DB/Bab 6; DB/E1 1.14.Google Scholar

⁴ DB II, 7–8; DB/Bab 41; DB/E1 2.3.Google Scholar

⁵ Von Voigtlander, , 37 and 59.Google Scholar

⁶ DB, DNa, DSe, DPe and XPh, cf. Kent, , Old Persian.Google Scholar

⁷ Walser, G., Die Völkerschaften auf den Reliefs von Persepolis, Berlin, 1966. Falttafel 1 at end of the volume for the Naqsh-i Rustam tomb reliefs.Google Scholar

⁸ Roaf, M. D., “The subject peoples on the base of the statue of Darius”, Cahiers DAFI, IV, 73–160. For the Sattagydian see p. 116, carved figure 10.Google Scholar

⁹ Walser, , Völkerschaften, 89Google Scholar (Herzfeld having abandoned his identification with Delegation XIV); Schmidt, E. F., Persepolis I, Chicago, 1953.Google Scholar

¹⁰ Walser, , Völkerschaften, Falttafel 1.12.Google Scholar

¹¹ Walser, , Völkerschaften, Falttafel 1.13;Google ScholarRoaf, , 145, no. 24.Google Scholar

¹² Walser, , Völkerschaften, Falttafel 1.29;Google ScholarRoaf, , 144, no. 23.Google Scholar

¹³ Herodotus, , VII, 67–70.Google Scholar

¹⁴ Strabo, , XV, 2, 9.Google Scholar

¹⁵ SirWheeler, Mortimer, Charsada: a metropolis of the North-West Frontier, London, 1962;Google Scholar The opening chapter contains a review of the evidence identifying this site with the ancient Pushkalavati.

¹⁶ Other locations that might be proposed for Sattagydia, simply from a consideration of the map, are (i) Taxila-Bhir Mound; (ii) the region round Sibi in Baluchistan; or (iii) the Gardez region in eastern Afghanistan. There are objections to each of these. Taxila (for which the fullest account is that of the excavator, Marshall, J., Taxila, 3 vols., Cambridge, 1951)Google Scholar was far too large and rich, even in the late 6th and early 5th centuries B.C., to have been paying only a portion of the meagre annual tribute of 170 talents that Herodotus mentions for Sattagydia. Whether Taxila was part of the fabulously rich satrapy of Hiⁿduš is outside this enquiry, although it is noteworthy that whilst “Indians” fought against Alexander at Gaugamela as Achaemenid troops under the satrap Barsaentes, the various rulers of the Punjab and the middle Indus seem to have opposed Alexander more as independent chiefs, rather than as agents of any imperial state. I think that if Taxila ever was part of the Achaemenid empire, it was a loose subjection that did not take the form of the standard satrapy; in my opinion, the traditional location of Hiⁿduš in the region of Sind stands, and the enormous annual tribute of 360 talents of gold dust was the result of transit trade from the centre of India. The Sibi region is poor, but too far south of Gandara; it was, in a later century at least, under the control of the satrap of Arachosia (Arrian, , Anabasis, III, 8;Google ScholarCurtius, Quintus, IX, 7, 14). The Gardez region of eastern Afghanistan is too high to have allowed the scanty clothing shown in the reliefs to have been worn with any degree of comfort.Google Scholar

¹⁷ Marquart, J., Untersuchungen zur Geschichte von Eran, Heft II, Leipzig, 1905, 175.Google Scholar

¹⁸ Gershevitch, I., The Avestan hymn to Mithra, London, 1959, 174–5.Google Scholar Although Gershevitch was concerned with linguistic, rather than geographical, arguments, he does identify a particular geographical locality for Avestan Iškata = OP Sattagydia, which was that proposed by Marquart. Even if Iškata was Sattagydia, the location in the Hindu Kush is invalidated by the evidence of the clothing in the sculpted reliefs, and the extreme difficulty of reconciling a location in the central part of the Hindu Kush with any plausible reconstruction of Vivana's campaign.

¹⁹ Marquart, , 73–4.Google Scholar

²⁰ The Avestan hymn to Mithra, 174–5.Google Scholar

²¹ Frye, R. N., The heritage of Persia, London, 1962, 47.Google Scholar

²² MacDowall, D. W. and Taddei, M., “The early historic period: Achaemenids and Greeks”, in Hammond, N. and Allchin, F. R. (ed.) The archaeology of Afghanistan from the earliest times to the Timurid period, London, 1978, 187 and figure 4.1.Google Scholar

²³ Above, n. 7 and 8.

²⁴ Walser, , Völkerschaften, Falttafel 1.5; 1.6; 1.10.Google Scholar

²⁵ The heritage of Persia, 47.Google Scholar

²⁶ DB/Bab 6, in listing the lands of the empire, included both KUR pa-ar-ú-pa-ra-e-sa-an-na and KUR sa-at-ta-gu-ú in the same line, Von Voigtlander, 12 and 54.

²⁷ In DB/Bab 6 the only difference from the OP text DB I, 16–17Google Scholar is that the Gaⁿdāra of the OP version is represented by Parauparasaena of the Babylonian. Since all the other names of the two lists match directly one may assume that these do as well. Support for this reconstruction comes from the parallelism of OP Armina (DB I, 15)Google Scholar and Babylonian Urartu (DB/Bab 6), an illustration of the same region having two different names.

²⁸ Herzfeld, E. E., The Persian empire, Wiesbaden, 1968, 341–3.Google Scholar

²⁹ Schmidt, , Persepolis I, Figure 2.Google Scholar

³⁰ Mughal, M. R., “Excavations at Tulamba, West Pakistan”, Pakistan Archaeology, IV, 1967, 13–152.Google Scholar

³¹ Mughal, , 55.Google Scholar

³² As first stated by Herodotus, , III, 97.Google Scholar

³³ Dr. F. R. Allchin, personal communication.

³⁴ Raverty, H. G., Notes on Afghanistan and Baluchistan, London, 1888, 425.Google Scholar

³⁵ Raverty, , 312.Google Scholar

³⁶ Punjab Government, Gazetteer of the Bannu District (1883–4), Calcutta, 28.Google Scholar An early historical appreciation of the importance of that whole region of the middle Indus for links with Afghanistan is that of Burnes, A., Cabool: a personal narrative of a journey to, and residence in that city in the years 1836, 7, and 8, London, 1843, 287.Google Scholar

³⁷ Whitehouse, D., “Excavations at Kandahar, 1974: first interim report”, Afghan Studies, I, 1978, 9–39;Google ScholarMcNicoll, A., “Excavations at Kandahar, 1975: second interim report”, Afghan Studies, I, 1978, 41–66;Google ScholarHelms, S. W., “Old Kandahar excavations 1976: preliminary report”, Afghan Studies, II, 1979, 1–8.Google Scholar

³⁸ Stein, M. A., Report on the archaeological survey work in the North-West Frontier Province and Baluchistan for the period from January 2nd, 1904, to March 31st, 1905, Peshawar, 1905, 7–10.Google Scholar

³⁹ Toynbee, A. J.. A study of history, VII, London, 1954, 639–40.Google Scholar

⁴⁰ See n. 3, above.

⁴¹ DB/Bab 80 (Von Voigtlander, , 35 and 59).Google Scholar

⁴² DB/Bab 83 (Von Voigtlander, , 36 and 59).Google Scholar

⁴³ DB III, 66.Google Scholar

⁴⁴ Not Irdumaka as proposed by Cameron, in JCS, XIV, 1960, 66.Google Scholar He has himself corrected his former opinion to Kandumaka in Von Voigtlander, 36, notes to 1. 81.

⁴⁵ DB/Bab 81 (Von Voigtlander, 35–6 and 59).Google Scholar

⁴⁶ Cameron, in JCS, XIV, 1960, 66, note to DB/E1 3.27–8.Google Scholar

⁴⁷ The Persian empire, 334.Google Scholar

⁴⁸ p. 36, notes to DB/Bab 81.

⁴⁹ OP, 13 Anamaka, (DB III, 63);Google Scholar Babylonian, 13 Tebet (DB/Bab 81); Elamite, 13 Anamaka (DB/E1 3.24—5).

⁵⁰ DB/Bab 81; DB/E1 3.24–5.

⁵¹ Tarn, W. W., The Greeks in Bactria and India, second edition, London, 1951, 460–2.Google Scholar

⁵² Natural History VI, 92.Google Scholar

⁵³ For example, Ptolemy, VI, 18, 4Google Scholar put both Kapisa and Ga[n]zaka in the Paropamisus, although it is virtually certain that Ga[n]zaka (the later Ghazni) has at no time been counted as of the Hindu Kush/Paropamisus. Herzfeld's suggestion (The Persian empire, 337) that the Paropamisus was the highland part of the area known as Gaⁿdāra is borne out by a comparison of the different parts of the Behistun inscription; see n. 27, above.Google Scholar

⁵⁴ Bernard, P., “Une problème de toponymie antique dans l'Asie centrale: les noms anciens de Qandahar”, Studio Iranica, III, 1974, 171–85; see p. 180Google Scholar.

⁵⁵ Fraser, P. M., “The son of Aristonax at Kandahar”, Afghan Studies, II, 1979, 9–21;Google Scholar see p. 17 n. 26.

⁵⁶ Hackin, J., Recherches archéologiques à Begram, Paris, 1939,Google ScholarNouvelles recherches archéologiques à Begram, Paris, 1954.Google Scholar

⁵⁷ OP 7 Viyakhna, , DB III, 68;Google Scholar Babylonian 7 Adar, DB/Bab 82; Elamite 7 Viyakhna, DB/E1 3.29.

⁵⁸ Raverty, , Notes on Afghanistan and Baluchistan, 72.Google Scholar

⁵⁹ p. 36.

⁶⁰ If Gadutava was in Arachosia and not in Sattagydia, as is stated in DB/E1 3.27–28, Gandamak could to some observers have been the site of the battle, as neither the extent of Achaemenid Gaⁿdāra into the upper Kabul river valley, nor the northward boundary of Achaemenid Arachosia, has been defined with precision. It is certainly conceivable that the two rival armies manoeuvred around each other in the snow for two months between battles, but given the apparent ferocity and urgency of the struggle it is unlikely that much time would have been wasted on careful movements. In any case, such a reconstruction faces the same criticism as that of Herzfeld: in order to be in Arachosia for the third battle of the series it would have been necessary for Vahyazdata's lieutenant to have fought back up the Kabul valley and past Ghazni, and Vivana would no more have left his rear unprotected here than in the Bolan pass.

⁶¹ Which probably happened in some way, as Darius recorded that Sattagydia has submitted, DB/Bab 84; see note 5, above.

⁶² Herzfeld, , The Persian empire, 300,Google Scholar and Toynbee, , 611, both located the Yautiya/Outioi in south-central Iran, in the region later known as Lāristān.Google Scholar

⁶³ As proposed by Hinz, W., Neue Wege im Altpersischen, Wiesbaden, 1973, 61.Google Scholar

⁶⁴ Bernard, , Studia Iranica, III, 1974, 178 n. 23.Google Scholar

⁶⁵ The evidence for the identification of the Old City of Kandahar with the City of the Arachosians has been summarized by Fraser, , Afghan Studies, II, 1979, 17, n. 26.Google Scholar The location of Aršada in Arachosia is mentioned in DB III, 72,Google Scholar DB/Bab 83, DB/E1 3.31.

⁶⁶ Even if OP cuneiform did exist before Darius I, it does not seem to have been used for record-keeping or administrative work, and so may be discounted in this context.

⁶⁷ See most recently Stronach, D., Pasargadae, Oxford, 1978, 283–95.Google Scholar

⁶⁸ Von Voigtlander, , 7.Google Scholar