P E U S E P O L I S 559 difficult alpine region was far from convenient. The practical capitals were Susa, Babylon, and Ecbatana. This, at the same time, accounts for the fact that the Greeks were not really acquainted with the city until it was taken by Alexander. 1 Ctesias must certainly have known of it, and it is possible that he may have named it simply Hepo-cu, 2 after the people, as is undoubtedly done by certain writers of a somewhat later date. :! But whether the city really bore the name of the people and the country is another question. And it is extremely hazardous to assume, with Sir H. Rawlinson and Oppert, that the words and Pdrsd, "in this Persia," which occur in an inscription on the gateway built by Xerxes (D. lin. 14), signify "in this city of Parsa," and consequently prove that the name of the city is identical with the name of the country. The name Persepolis appears to have been first used by Clitarchus, one of the earliest, but unfortunately one of the most imaginative annalists of the exploits of Alexander. The word was no doubt meant to allude to the " Persians," but apparently he pref erred this extraordinary form 4 to the regular " Persopolis " 5 for the sake of a play on the destruction (Tre/oo-i?) which he relates. Later writers have followed him in the use of the name Persepolis. 6 For information about the capture and treatment of the city by Alexander we are almost entirely dependent on narratives which are based on Clitarchus, since Arrian unfortunately disposes of this episode in a very summary fashion. The course of events may be traced somewhat as follows. Alexander, having crushed the resistance of the Persian army under Ariobarzanes at the " Persian Gates," 7 marched rapidly on the capital. Ariobarzanes had made his way thither with a few followers, but was refused admission by Tiridates, the commandant of the citadel, who had already commenced negotiations with Alexander, and at last surrendered the place with its immense treasures to the conqueror. In a subsequent battle Ariobarzanes was killed. 8 Alexander then ordered a general massacre, and gave up the city to be plundered. In the citadel he placed a garrison of 3000 men under Nicarchides, 9 and then caused 1 ^Eschylus, whose knowledge of the world is certainly not very extensive, takes the " city of the Persians " to be Susa. Of. especially Pers., v. 15 with v. 761 (r6S darv Zoucruv). Herodotus does not mention the capital of Persis at all. " The only expression that could be interpreted in this sense is ^s Ilc pcras, "to the Persians." But perhaps es Il^pcras, with him, means only "to the land of Persis." No doubt, when he says that the body of Cyrus was conveyed es Il^pcra?, this might be explained on the supposition that he wrongly imagined that Cyrus was buried in Perse polis. Xenophon, who knew of Pasargadas from Ctesias, calls it llepcrai (C l/r., viii. 5, 21) ; but, as he was not acquainted with the country, this goes for nothing. Of more importance is the fact that Plutarch, Artax., iii. (probably after Dinon), places Pasargadse ev Hfyxrcus, where the expression applies to the country and not to the city. 3 So undoubtedly Arrian (iii. 18, 1, 10), or rather his best authority, King Ptolemy. So, again, the Babylonian Berosus, shortly after Alexander. See Clemens Alex., Admon. ad gentes, c. 5, where, with Cleorg Hoffmann (Pers. Miirtyrer, 137), /ecu is to be inserted before Ilepo-ais, and this to be understood as the name of the metropolis. 4 Ilepo-f TroXtj means strictly " city-destroying." TlepcraliroXis, a well- authenticated reading in Strabo and ^Elian (I.e.), is no improvement. 5 This form is actually restored by later scholars, and seems to have been used by the geographer Ptolemy (vi. 4). 6 Besides the historians who draw upon Clitarchus (Diodorus, Curtius, Justin, Plutarch in Alexander), Strabo (79 sq., 727 sq.), Pliny (vi. 115, 213), and several others. Justin (i. 6, 3) introduces the name Persepolis in an account which is based on Ctesias, just as Arrian (vii. 1, 1) once employs it, although he can scarcely have got it from his excellent sources. 7 On this locality, see the paper of Fr. Stolze in the Verhand- lungen der Gesellschaft fur Erdkunde in Berlin, 1883, Nos. 3 and 6. 8 This is mentioned by Curtius only, but it has great intrinsic prob ability. The massacre at the taking of the city appears to be confirmed by Plutarch (Alex., 37) from the letters of the king. 9 This again is only found in Curtius. Alexander was in the heart of a country which he had laid waste, but by no means thoroughly subdued, which hated him bitterly, and which was the native land of the dynasty ; he was amongst a people who still felt themselves to be the royal palaces to be set on fire, certainly not in a drunken freak, but apparently with deliberate calculation on the effect it would produce on the minds of the Asiatics. 10 Now it has hitherto been universally admitted that " the palaces " or " the palace " (TO, /^uo-t Aeta) burned down by Alexander are those now in ruins at Takhti Jamshid, as already described. From Stolze s investigations it appears that at least one of these, the castle built by Xerxes, bears evident traces of having been destroyed by fire. 11 The locality described by Diodorus after Clitarchus corresponds in important particulars with Takhti Jamshid, for example, in being supported by the mountain on the east. 12 And, if there are other details, such as the triple wall, which it is difficult to reconcile with the existing state of things, we must bear in mind on the one hand the great destruction that must have been wrought in the course of thousands of years, and on the other that small inaccuracies are not to be wondered at in a writer like Clitarchus, who is constantly straining after effect. Tli ere is, however, one formidable difficulty. Diodorus says that the rock at the back of the palace containing the royal sepulchres rises so steep that the bodies could be raised to their last resting-place only by mechanical appliances. This is not true of the graves behind Takhti Jamshid, to which, as Stolze expressly observes, one can easily ride up ; on the other hand, it is strictly true of the graves at Nakshi Rustam. Stolze has accordingly started the theory that the royal castle of Persepolis stood close by Nakshi Rustam, and has sunk in course of time to shapeless heaps of earth, under which the remains may be concealed. He and Andreas, our highest authorities on the topography of this district, 13 consider this spot peculiarly adapted for the site of a citadel, while the water-supply would suffice for a numerous court-retinue and garrison, and for a royal residence with its palaces and gardens. Nevertheless we are unable to adopt this suggestion. The vast ruins of Takhti Jamshid, and the terrace constructed with so much labour, appear to us of more importance than any number of doubts and conjectures. These remains can hardly be anything else than the ruins of palaces and the other belongings of a kingly residence ; as for temples, the Persians had no such thing, at least in the time of Darius and Xerxes. And it can hardly be supposed that such solid structures were much more numerous in former times, and that these alone have survived owing to their peculiar situation on the terrace. For, in the first place, it is evident at a glance that the situation itself is of an excep tional kind. Moreover, Persian tradition at a very remote period knew of only three architectural wonders in that region, which it attributed to the fabulous queen Humai (Khumai) the grave of Cyrus at Murgab, the building at Haji abad, and those on the great terrace. 14 It is safest therefore to identify these last with the royal palaces destroyed by Alexander. Clitarchus, who can scarcely have visited the place himself, has simply, with his usual the dominant race, and knew that their king was still alive. That in these circumstances he should have a strong garrison under a trust worthy Macedonian was simply a matter of course. Nicarchides after wards commanded a trireme in the fleet that sailed from the Indus to the Tigris (Arrian, Indica, xix. 5 ; after Nearchus). 10 See art. PKRSIA (p. 582 below). 11 Dr Stolze has kindly explained to the writer of this article that the layer of charcoal in the " hall of a hundred pillars " is apparently the result not of a conflagration but of gradual decomposition. 12 The name of this mountain too, fiaaiKiKov &pos, is identical with SlMikuh, which is at least tolerably well established by Ouseley (ii. 417) as a synonym of Kuhi rahmet. 13 We are here again indebted to private communications from Stolze, as well as to his published papers. 14 See especially Hamza Isp., 38 ; Tabari, i. 690, 816 (cf. Nuldeke, Geschichte der Perser . . . aus . . . Tabari, p. 8). The ruins at Takhti Jamshid are alluded to as the work of Humai, in connexion with an event which occurred shortly after 200 A.D.
