¹ A good bibliography is contained in Pryce's, F. N.Catalogue of Sculpture in the Department of Greek and Roman Antiquities of the British Museum, vol. i, part I (1928)Google Scholar. To this may be added the early accounts given by Sir Charles Fellows in his Journal … in Asia Minor (1838–39), Account of Discoveries in Lycia (1840–41), Xanthian Marbles (1843), Lycia (1847), and Travels (1852); also the more recent references in Rodenwaldt, G., Griechische Reliefs Lykien (Sitzungsberichte der Berliner Akademie, 1933)Google Scholar, and in Picard, C., Manuel d'Archéologie Grecque (1939)Google Scholar.

² A similar view in regard to the Lycian pillar-tombs of the sixth century has recently been expressed by a Turkish archaeologist, Ekrem Akurgal, whose book Griech. Reliefs d. VI. Jhdts aus Lykien (1942) came into my hands while I was reading the proofs of this paper.

³ Here my thanks go to the Trustees of the British Museum for the extraordinary facilities which they gave me when photographing the reliefs, several years ago, when the present Director, Sir John Forsdyke, was Keeper of the Department of Greek and Roman Antiquities. To him I also wish to express my deep gratitude for the interest he has taken in this paper.

⁴ Fellows always speaks of ‘close by’ (Journal, I74) and ‘very near’ (Travels, 338), but he gives no accurate distance. Benndorf who has measured the distance, gives it in Reisen 1, 85 as ‘50 Schritte’.

⁵ T.A.M. i. 44Google Scholar, 6, 21: ‘..… ἀνέθηκεν

.’

⁶ Fellows calls it the burial-place of the kings and says, ‘And from finding the district to have been the burial place of the kings, it (sc. the Harpy Tomb) becomes the more interesting’ (Travels, p. 340)Google Scholar. Benndorf in his Reisen, vol. i, gives a view from the acropolis on plate 23 and describes this on p. 86 as follows: ‘Man erkennt hier in der linken unteren Ecke des Bildes die Harpagidenstele (i.e. Xanthian Stele) und übersieht rechts davon (i.e. to the south-west) die jetzt durch einige Saatfelder bezeichnete Agora, auf der sich einst das Sarpedoneion befand.’ For Kalinka, T.A.M., see infra, note 10.

⁷ The first map was made for Sir Charles Fellows by A. Hoskyn, Master of H.M.S. Beacon in 1840, and published in Spratt, , Travels (1847), vol. ii, plate 2Google Scholar. Benndorf, , Reisen, i, 85Google Scholar, rightly describes it as ‘nur dürftige Orientierung.’ Another map was given by Fellows in Xanthian Marbles, plate 2, but this is, again according to Benndorf, ‘eine Skizze nach verfehlten Schätzungen, daher mit Recht nicht wiederholt in den Travels and Researches’ The map in our Fig. 1 is from Benndorf's article in Oe. Jh. 3 (1900), p. 100, fig. 23, and was made by E. Krickl (Hauptmann im Genieregiment) in 1892. For Benndorf's description see above note 6. As for the Harpy Tomb, the monument still stands at its place, only the marble slabs with the reliefs have been removed to England. The sarcophagus between Harpy Tomb and theatre is of much later date.

⁸ E. Kalinka, Tituli Asiæ Minoris (here generally quoted as T.A.M.), vol. i (1901)Google Scholar, dealing with the early inscriptions in Lycian, vol. ii (1920), only with the late Greek and Roman inscriptions. The map is in vol. ii, fasc. 1, p. 95. It is marked ‘Form a Xanthi urbis. E. Krickl anno 1892 adumbravit,’ thus admittedly a copy of the map published 20 years before by Benndorf in Oe. Jh., which is marked ‘Planskizze von Xanthus, aufgenommen von E. Krickl 1892.’ Though it is clearly the same map, yet it is less carefully drawn. While in Benndorf's article it is quite obviously made by an architect, with explanation of figures in block letters, the drawing on Kalinka's map is not as accurate, the explanation of figures is in handwriting, and some of the figures (like S1, S2, S3, S4 in Oe. Jh. corresponding to (S)1, (S)2, S3, S4 in T.A.M.) have a slightly different explanation, and are sometimes not indicated in the right spot on Kalinka's map. One is led to the conclusion that this second map is not altogether reliable.

⁹ The term ‘Agora’ does not occur on Benndorf's map. It is only to be found on Kalinka's map where it applies to the remains of a square building surrounded by a stoa on the east and south. After examining both maps closely (see note 8), this proves to be a ‘late interpolation,’ inspired by the wish to adjust this map to the inscriptions of the Roman period, and by the interpolator's idea of an agora as surrounded by a stoa on each side of a rectangle. Yet Kalinka's map is fairly well known and because it is in a book dealing with the most important inscriptions from Lycia, it is frequently quoted by scholars, while Benndorf's article is almost forgotten and its map hardly known.

¹⁰ He says so quite plainly in his commentary, and again refers to the older agora round the spot where the Xanthian Stele and the Harpy Tomb are standing: T.A.M. ii. 1, 96Google Scholar ‘infra arcem ad meridiem situm est forum saxis stratum, ubi praeter cetera aedificia exstructae sunt duae illae columnae quarum una monumentum Harpyiarum nominatur, altera insignis est longitudine tituli Lycia lingua inscripti et epigrammate Graeco.’

¹¹ Benndorf, , ‘Zur Stele Xanthia,’ in Oe. Jh. 1900, iii., 98 ff.Google Scholar; König, F. W., ‘Die Stele von Xanthos’, Klotho, 1936Google Scholar; Meriggi, P., ‘Zur Xanthosstele,’ in Acta Jutlandica (Aarskrift for Aarhus Universitet) 1937, ix. 504 ffGoogle Scholar.

¹² This, in Lycia, usually marks the beginning of an inscription and shows the way it was set up. It always faces the direction from which worshippers or visitors are expected to come.

¹³ C.I.G. iii, 4269 b, commenting on the Xanthian Stele, says: ‘Praeterhanc stelam Xanthi in foro etiam Σαρπηδόνειον collocatum fuisse novimus ex Appiano bell. civ. iv. 78.’ But Kalinka in T.A.M. ii. 96Google Scholar ‘Tota hac regione (sc. prope theatrum) multae parietinae inveniuntur, inter quas illud quoque Σαρπηδόνειον fuisse puto cuius Appianus b. civ. iv. 78 mentionem facit. Confer M. 313 sq. ubi Sarpedo Glaucum appellat:

An inscription (T.A.M. ii. 265Google Scholar) has been found to the southeast of the theatre, erected by Aichmon after a victory, and its last line runs: Σαρπηδόνι καὶ Γλαύκωι ἥρωσι. As this inscription obviously presupposes a heroön of Sarpedon and Glaucus, the C.I.G. iii. 4269 b add. comments: ‘Titulus fortasse positus fuit in Sarpedonio.’ And Benndorf, (Historische Inschrift vom Stadttore zu Xanthus, Festschrift für Otto Hirschfeld, 1903, 29)Google Scholar concludes: ‘Da s Sarpedoneion lag wahrscheinlich auf dem Hügel über dem Theater, innerhalb der Ringmauer.’ The Sarpedoneion was also mentioned in Aristot., pepl. 53Google Scholar; Athen. i. 13 sq; and Plin., N.H. 13, 88Google Scholar.

¹⁴ For collected evidence and general literature on this subject see Oe. Jh. 1931, xxvii. 82 ffGoogle Scholar.

¹⁵ Plut., consol. ad Apoll., 21Google Scholar; Val. Max., ii, 6, 13. Both writers state that, among the Lycians, the male members of a family in mourning had to wear women's clothes. As a reason they give the belief in Lycia that mourning was something unworthy of a man, and so he had to put on a woman's dress to make it less conspicuous. But this is clearly a belated and rationalistic attempt at an explanation of this ancient custom (cf. Hauser, in Philologus 54 (N.F.8) 389 ff.)Google Scholar. It is proved that, in earlier times, only the female members represented the family in Lycia. The tradition that men had to put on female garments on certain religious occasions where the family as a whole was involved is in itself only one of the many survivals of such ancient customs in Lycia.

Furthermore, the wearing of long dresses by priests and singers or musicians on religious occasions in early archaic Greece as well as in Minoan Crete points to an interesting parallel. And in Persia, Assyria, and Babylonia the king's attendants wore a similar dress for certain other reasons. It will be proved that, as far as the east side of the Harpy Tomb is concerned, the Persian tradition had some importance and coincided with Lycian customs.

¹⁶ For the Satrap sarcophagus: Mendel, , Cat. Mus. Ottom. i. 33 ffGoogle Scholar. (where Mendel has proved that all these attendants were male); for the Payava Tomb: Smith, B.M. Cat. ii. 47Google Scholar, pl. 11 (the prince represented here is the satrap Autophradates, 375–362 B.C.); for the Nereid Monument: B.M.Cat. ii (fourth frieze), the prince seems to be a Persian satrap but cannot be identified); for the Heroön from Gyeulbashi: Benndorf's monograph (the scene has been thought to depict the Ilioupersis with Priam and Hecuba enthroned above the besieged city; it is more probable, however, that it refers to some event in Lycian history or legend).

¹⁷ The conquest of Lycia by Harpagos, the general of Cyrus, is to be dated not later than 538 B.C. This campaign consisted mainly in the siege and capture of Xanthus, described in detail by Herodotus i. 176. This Harpagos was a Mede and ἀνὴρ οἰκήιος of Deiokes (Hdt. i. 108) and συγγενής of Astyages (Hdt. i. 109), thus of royal blood himself. As the ruler of Xanthus who erected the Xanthian Stele on the agora calls himself son of Harpagos (T.A.M. i. 44Google Scholar; the stele dates from the beginning of the fourth century), it is very probable that members of the house of Harpagos were in some sort of command in Lycia ever since the conquest.

¹⁸ E. Herzfeld, Iran in the Ancient East, pl. 67a and b; Pope, A. U., Survey of Persian Art, iv, pl. 88Google Scholar; Schmidt, E., The Treasury of Persepolis (Oriental Institute of Chicago Communications 21), 1939, 21 ff.Google Scholar, figs. 14, 16. These are the most recent publications dealing with the subject.

¹⁹ It is to be noted that these two court officials take precedence over the carrier of the royal weapons and over the officers of the king's bodyguard. They are also the only persons to accompany the king on several other occasions (as shown on other reliefs from Persepolis, e.g. the portals). And every time the sole attribute of their office is a towel or napkin, neatly folded, or a fly-whisk, or a scent-bottle. And their attire is always the same. Neither of them can be the famous Hazarapatis, the Major-domo and Grand-Vizier of the empire, who was the commander of the king's bodyguard (Xenophon translates Hazarapatis by Chiliarch, in Cyrop. viii. 6)Google Scholar. One of them may be the ‘Eye of the King’ who was still more prominent than the Hazarapatis, and to whom was entrusted the control of the empire (Meyer, E., Gesch. d. Alt. iii. 43)Google Scholar. And it seems very likely, as Schmidt, E. F. has shown (Treasury of Persepolis, 26 ff.)Google Scholar that the other was the ‘Cupbearer,’ who held the rank of a priest in Xerxes' times and was also responsible for the king's safety. The office of the Cupbearer was, at least in later Achaemenian times, just as that of the Hazarapatis himself, in the hands of eunuchs, as several literary sources indicate. And this information seems born out by the reliefs where the person is depicted without beard or moustache (which would be visible above the muffler). Cf. Marquart, J., Untersuchungen zur Gesch. von Eran, i. 57 ff., 224 ff., ii, 158 ff.Google Scholar; König, F. W., Altpersische Adelsgeschlechter, in Wiener Zeitschr.f. d. Kunst des Morgenlandes, 1924, xxxi, 289 ff.Google Scholar; 1926, xxxiii, 23 ff., 37 ff.; 1928, xxxv, 1ff.; König, F. W., Der falsche Bardija, in Klotho 4, 1938, passimGoogle Scholar; Schmidt, E. F., Treasury of Persepolis, 26 ffGoogle Scholar.

²⁰ This court ceremonial was by no means a short-lived institution but a long established religious ritual, as is proved by an Assyrian fresco painting, almost identical in contents with the Persian reliefs (Fig. 6), Syria ix, pl. xxiii ff. Fragments of a similar painting from the palace of Niniveh are in the British Museum. For the king with two attendants accompanying him, many more examples of Assyrian art could be mentioned, chiefly reliefs, e.g. Assurbanipal's Hunt, the Banquet of Assurbanipal, Sanherib's Sacrifice (Meissner, Babylonien und Assyrien, figs. 46, 48, 117), etc. For the description of the Assyrian ceremonial in contemporary literature, see Peiser, F. E., Studien zur oriental. Altertumskunde, in M.V.A.G. 1898, 253, 1. 16 ffGoogle Scholar. On the other hand, this same court ceremonial was continued by the Seleucids, after them by the Arsacids (Philostrat., , Vita Apollon. Tyan. i, 27 ffGoogle Scholar. describes such an audience at the Arsacid court in the first century A.D.), and after them by the Sasanids (Arabic and Byzantine writers give ample information about this; cf. Nöldeke, , Tabari, 113, 221Google Scholar) and by the Khalifs all through the Middle Ages.

²¹ The date of the Treasury reliefs has recently been stated as between 490 and 486 B.C. (Schmidt, E. F., Treasury of Persepolis, 33Google Scholar), and the king and crown prince may be taken to represent Darius and Xerxes, as on the corresponding reliefs of the Tripylon. The Apadana was completed by Xerxes himself. But as for the Hundred-Column Hall, E. Herzfeld discovered in the south-west corner a stone slab stating in Babylonian that Artaxerxes I erected this structure on the foundations prepared by his father Xerxes (Herzfeld, , Altpers. Inschr., in Arch. Mitt. Iran, 1. Erganzungsband, 1938, p. 45Google Scholar), and thus the date of these reliefs cannot be before 465 B.C. It was possibly somewhat later in the reign of Artaxerxes I.

²² It is very probable that the detailed account of Harpagos' campaign by Herodotus, and his stories about the miraculous preservation and the rise of Cyrus (in which Harpagos plays a predominant rôle) were partly derived from some member or members of the house of Harpagos, and later corrected by some Persian friend of Herodotus (Zopyros ?). This Harpagid family claimed descent from Deiokes the Mede (cf. note 18), and as they seem to have stood in close connexion with Lycia, Herodotus may well have come across them there. For this ‘Harpagid tradition’ in Herodotus, see Schubert, R., Herodots Darstellung der Kyrossage, 1900, 76Google Scholar; Justi, F., Grundriss der iranischen Philologie, ii, 410Google Scholar; Prasek, J. V. in Klio, 1904, iv, 199 ff.Google Scholar; and How and Wells, Commentary on Herodotus, 1936, note to book i.

²³ On the short side (north). Cf. Mendel, , Cat. Mus. Ottom., i, 189 (Fig. to the left)Google Scholar; Winter, , Sarkorphage von Sidon, 14 f.Google Scholar, pl. 7, 18. The sarcophagus dates from the end of the fourth century B.C. Yet it is hardly likely that it could have been made by an artist who actually saw the Persepolis reliefs. Persepolis was sacked and burnt down immediately after its capture by Alexander. It is agreed that the sarcophagus is of Attic workmanship, and the many allusions to the Persian court ceremonial and customs in Aeschylus' Persae (in language, expressions, ideas, and even in the metre of the dialogues) show that this ceremonial was quite well known in contemporary Athens. Otherwise, how could all these allusions have been understood by the listeners? Writers like Herodotus and Xenophon had also their share in making the people of Athens well acquainted with Persian customs and ritual. Surely the Greeks of the mainland, and even more so the Asiatic Greeks had not to rely on hearsay to describe or depict Eastern ceremonial in a work of art.

²⁴ Babelon, , Traité, ii, 8 ff.Google Scholar; Rev. Num. 1908; Six, , Num. Chron. 1898, 199 ff.Google Scholar; Hill, B.M. Cat. Coins, Lycia, pl. vi,ff.,; cf. also head from Ephesus, Pryce, , B.M. Cat. B 215Google Scholar, Fig. 132.

²⁵ This was a sacred law, and had been a rule in Persian art from its very beginning. Xenophon, , Cyrop. viii, 3, 14Google Scholar even goes a little further: when describing the splendid procession of Cyrus, he states that the very tall charioteer of Cyrus was yet much smaller than Cyrus himself. Does this mean that the rule of emphasising the difference in size between the king, his immediate followers, and the othe people, was also applied to simple narratives in an oral tradition? I cannot help feeling that Xenophon was simply describing a picture or relief, though he does not say so this time. It is to be noted, however, that Xenophon usually does mention reliefs and pictures if he describes them, e.g. Cyrop. i, 2, 13Google Scholar.

²⁶ A small boy or girl bringing offerings to their dead parents was frequently depicted on Greek vases and reliefs. The best examples to be compared with this scene on the Harpy Tomb, are the well-known archaic Laconian reliefs (Ath. Mitt. iv, pl. 8, 1–2), where also the offerings are the same as on the Harpy Tomb.

²⁷ Among the earlier tombstones with this group of man and dog, cf. the Anaxandros stele in Sophia (from Apollonia; Jahrb. 1902, pl. 1), the Naples stele (Rayet 19; B.-B. 416; it can be traced back to Sardis), the Alxenor stele from Orchomenos (Naxian; B.-B. 41), and the Agathocles stele (Athens, , Nat. Mus. 724)Google Scholar and Aegina stele (A.D. i, 33)Google Scholar. The tradition survives in the fourth century as shown on the Delphi stele (Bulle, pl. 265), the Thespiae stele in Athens (Collignon, Stat. funér, Fig. 68), and on the Ilissos stele (Conze ii, pl. 211). But vases prove that this was a favourite subject also in the sixth century, cf. the Timonidas pinax from Corinth, the amphora 2303 in Munich (Richter, , JHS—VOL. LXIIGoogle Scholar. Ancient Furniture, Fig. 163), and that it also spread to Italy (South Italian amphora in Rome, Vatican; Collignon, Stat. funér., Fig. 67).

²⁸ See the inscription on a late Lycian rock tomb from Bel near Sidyma (T.A.M. ii, 1, 245Google Scholar; J.H.S. 1914, xxxiv, 5 ffGoogle Scholar. n. 10):

.…

²⁹ All these minor details could not be dealt with in this paper. Also the question of dating the monument and of analysing its style must be left for a later occasion, The attempt at an interpretation of the other three sides of the Harpy Tomb was briefly outlined in a paper which I read to the Hellenic Society at Cambridge on May 4th, 1943. I trust I shall be forgiven for not compressing it into a few pages for the sake of immediate publication.