¹ There is not even a laconic entry in the normally exhaustive Pauly-Wissowa, such as may be found by Kiechle, F., ‘Medismos’, in Lexikon der alten Welt (Stuttgart 1965) 1884Google Scholar, or Gugel, H., ‘Medismos’, in Kleine Pauly iii 1133Google Scholar. Gillis', D. recent study Collaboration with the Persians, Historia Einzels. xxxiv (Wiesbaden 1979)Google Scholar passes over the problem without comment.

Abbreviations for Iranian materials follow those utilized by Schmitt, R. in Kratylos xxv (1980) 1–16Google Scholar; cuneiform texts are cited according to The Chicago Assyrian Dictionary. Other abbreviations used in this essay are as follows: ANET = Pritchard, J., Ancient Near Eastern Texts³ (Princeton 1969)Google Scholar; CIS = Corpus inscriptionum semiticarum; Kent = Kent, R. G., Old Persian: Grammar, Texts, Lexicon (New Haven 1953)Google Scholar; RES = Répertoire d'épigraphie sémitique.

The present study represents a revised version of the initial chapter of my dissertation Medism: Greek Collaboration with Achaemenid Persia (Michigan 1979)Google Scholar. I am grateful to C. G. Starr, M. W. Stolper, G. L. Windfuhr, T. Cuyler Young, L. Koenen, J. H.Johnson and M. C. Root, who provided helpful advice on a number of specific matters. For the initial stimulus to explore Greek-Persian relations, I am indebted to the late G. G. Cameron.

² LSJ s.v. Ἀττικίζω, Ἀττικισμός, Λακωνίζω, Λακωνισμός and cf. Ἰωνίζω, Βοιωτίζω, Θεσσαλίζω, and Λυδίζω. These -ίζω verbs derived from nouns are called by Debrunner, A., Griechische Wortbildungslehre (Heidelberg 1917) 136–8Google Scholar, ‘Imitativa’, as they generally have the meaning ‘to be like x’. See also n. 10 below.

³ See also the traditions for the Medizing of Demaratus of Sparta and Themistocles (e.g., Plut., Them. 29Google Scholar; Ath. i 29f. Even after the conquest of Persia by Alexander the Great, the adoption of Persian language or dress created ill feelings among his Macedonian forces (Duris of Samos FGrH 76 F 14; Arr. An. vii 6.2–3, 8.2; and cf. Plut. De Alex. fort. i 8 = Mor. 329f–330a). For discussion see Badian, E., JHS lxxxv (1965) 160–1CrossRefGoogle Scholar and Bosworth, A. B., JHS c (1980) 1–21CrossRefGoogle Scholar. As C. G. Starr observes, such attitudes are not a late development as the Greek association of luxury with the Orient is at least as old as Archilochus (19 West) in the seventh century (IA ii [1975] 58–9).

⁴ Bonner, R. J. and Smith, G., The Administration of Justice from Homer to Aristotle i (Chicago 1930) 294–309Google Scholar, followed by Rhodes, P. J., The Athenian Boule (Oxford 1972) 162–71Google Scholar, argued that the only significant changes in the law before 411 were the inclusion of the boule and assembly in the judicial process by Cleisthenes and the exclusion of the Areopagus by Ephialtes. These changes are considered only a revision of the law from Solon's time. For a more recent discussion of the Attic impeachment procedure see Hansen, M. H., Eisangelia (Odense 1975)Google Scholar, with Rhodes, , JHS xcix (1979) 103–14CrossRefGoogle Scholar and Hansen, c (1980) 89–95. Although προδοσία is brought into conjunction with Μηδίζω or Μηδισμός only once in Herodotus (vii 30), it is connected with cooperation with Persia in about half of its 34 occurrences. Schreiner's, J. H. argument, C&M xxxi (1972) 84–97Google Scholar, that ostracism began as an attempt to combat Medism is not convincing: see Thomsen, R., The Origin of Ostracism (Copenhagen 1972)Google Scholar for other possible interpretations. For Themistocles cf. Thuc. i 138.6.

⁵ See Thalheim, T., ‘κακοῦργοι’, RE x (1919) 1529Google Scholar; Hansen, , Eisangelia 33–6Google Scholar.

⁶ Pritchett, W. K., The Greek State at War ii (Berkeley 1974) 27Google Scholar; cf. Roberts, J. T., Accountability in Athenian Government (Madison, Wisconsin 1982)Google Scholar.

⁷ ML 30 and 40, lines 32–8, which are badly worn and difficult to read. The dates are approximately 470 for the Teian fragments and perhaps 453–2 for that of the Erythrae stele. Any indication of a trial or legal procedure is lacking in the case of the Medizing Athenian councilor Lycidas (Hdt. ix 5), who was stoned to death with his family. This ‘lynching’ may be attributed to the threatening circumstances of the time, For discussion see How, H. and Wells, J., A Commentary on Herodotus ii 288Google Scholar and Hignett, C., Xerxes' Invasion of Greece (Oxford 1963) 281Google Scholar.

⁸ A query raised recently again by Momigliano, A., Alien Wisdom: The Limits of Hellenization (London 1975) 124CrossRefGoogle Scholar.

⁹ Myres, J. L., ‘Μηδίζειν: Μηδισμός’, Greek Poetry and Life: Essays Presented to Gilbert Murray (Oxford 1936) 97–105Google Scholar and PEQ lxxxv (1953) 8–9. His view is concisely stated in Bury, J. B. and Meiggs, R., A History of Greece⁴ (London 1975) 144CrossRefGoogle Scholar: ‘That the Greeks of Ionia had been long accustomed to regard Media as a resort against Lydia and to intrigue with the Median kings is shown by the word medism. For if such intriguing had first come into fashion after the rise of Persia and the fall of Lydia, the name chosen to designate it would naturally have been persism’.

¹⁰ For Ionian relations with Lydia see Graham, A. J., JHS xci (1971) 41–2Google Scholar and Harris, G., Ionia under Persia: 547–477 BC—A Political History (Diss. Northwestern 1971) 16–17Google Scholar. These encounters do not appear to have provoked a terminology with the same implications inherent in Medism. The term Λυδίζω appears in the sixth century (Hipponax 92.1 West, but apparently with the meaning ‘to speak Lydian’ (Masson, O., Les fragments du poete Hipponax [Paris 1962] 151Google Scholar, West, M. L., Studies in Greek Elegy and Iambus [Berlin/N.Y. 1974] 145)CrossRefGoogle Scholar. See Ar. Eq. 533 for a comedy by Magnes of this name. A possible political connotation is given by the Suda, s.v. Λυδίζω: τὰ τῶν Λυδῶν φρονῶ, but without any reference. As M. Stolper suggests to me, the late appearance of political connotations in these terms suggests they were modelled after the earlier meaning of ‘Medism’ and are to be distinguished in this respect from similar terms of the archaic period.

It has been argued that the Greeks must have known the Medes before the ninth century, based on the assumed chronology for the *ā>ē(η) sound change; see Mazzarino, S., Fra Oriente e Occidente (Florence 1947) 96–7Google Scholar and 341 n. 269–71, Benveniste, E. in La Persia e il mondo greco-romano (Rome 1966) 480Google Scholar and Laroche, E. in Mélanges de linguistique et de philologie grecques offerts à Pierre Chantraine (Paris 1972) 89–90Google Scholar. According to this hypothesis, the Ionic–Attic Μῆδοι would have been Μᾶδοι if the event had been later. For some severe criticisms of this argument see Szemerenyi, O. in Studien zur Sprachwissenschaft und Kulturkunde: Gedenkschrift für Wilhelm Brandenstein (Innsbruck 1968) 142–6Google Scholar and Lejeune, M., Phonetique historique du mycénien et du Grec ancien (Paris 1972) 235Google Scholar n. 2. As they observe, the view is filled with historical difficulties. The earliest reference to the Medes is in 837 as KUR a-ma-da-a-a in the annals of the Assyrian king Shalmaneser III: see Levine, L. D. in Iranian Civilization and Culture, ed. Adams, C. J. (Montreal 1972) 39–45Google Scholar and Iran xi (1973) 1–27Google Scholar, xii (1974) 99–124. Sennecherib (704–681 BC) even speaks of ‘the distant Medes, whose name no one among the kings, my fathers, had (ever) heard’ (Luckenbill, AR ii no. 238). Greek tradition is aware of the Median chief Δηιόκης, perhaps the (Akk.) Dayaukku, mentioned in the annals of Sargon II (721–705), but dependent on later sources. See Schmitt, R., AÖAW CX (1973) 137–47Google Scholar and Helm, P. R., Iran xix (1981) 85–90CrossRefGoogle Scholar. No object recognizable as an Iranian import for this period has been identified on the Greek mainland according to Muscarella, O. W., J. Anc. Near Eastern Soc. Columbia ix (1977) 31–57Google Scholar.

¹¹ Jonkers, E. J., ‘Μῆδοι, τὰ Μηδικά, Μηδισμός’, Studia varia Carolo Guilielmo Vollgraff (Amsterdam 1948) 78–83Google Scholar; cf. Macan, R. W., Herodotus IV–VI i 350Google Scholar.

¹² Herodotus (vi 28) describes him as an ἀνὴρ Πέρσης, but the similarity in name with the earlier Median Harpagus and the fact that a descendent of a Ἁρπάγος was a dynast in Lycia at the end of the fifth century (ML 93. 5) support such an identification. The transformation of Medes into Persians is illustrated by Datis, who becomes a ‘Persian’ in the later sources (schol. Ar. Pax 289), as does Mardonius (Nepos, Paus. 1.2). For Datis the Mede see now Lewis, D. M., JHS c (1980) 194–5CrossRefGoogle Scholar.

¹³ Bowra, C. M., Greek Lyric Poetry (Oxford 1961) 264Google Scholar, and Drews, R., The Greek Accounts of Eastern History (Cambridge, Mass. 1973) 7Google Scholar.

¹⁴ Bacchylides' (31.28) reference in 468 to the earlier capture of Sardis by the Περσᾶ[ν], probably reflects later terminology. On the other hand, apocryphal correspondence between Anaximenes of Miletus and Pythagoras (D. L. ii 5, viii 49) still speaks of ὁ Μήδων βασιλεύς and the Μήδων. For a discussion of the unreliability of Diogenes see Hope, R., The Book of Diogenes Laertius (New York 1930) 93–7Google Scholar. It is difficult to determine in such cases if the writers are modernizing or archaizing.

¹⁵ Highbarger, E. L., TAPA lxviii (1937) 98–111Google Scholar, preferred the period just before Marathon in 490, while Jacoby, F., Theognis (Berlin 1931)Google Scholar argued for the time just before Xerxes' campaign. A. R. Burn is almost alone in still contending for a date immediately after Cyrus' conquest of Ionia, (The Lyric Age of Greece [London 1960] 263)Google Scholar. See also West (n. 10) 65.

¹⁶ Thomsen (n. 4) 97. Although Lewis, D. M., ZPE xiv (1974) 1–4Google Scholar, has argued for a date in the 470s, this still places the ostraca in proximity to the Persian Wars. For Kallias see Shapiro, H. A., Hesp. li (1982) 69–73CrossRefGoogle Scholar.

¹⁷ According to Jacoby, F., Hesp. xiv (1945) 185Google Scholar n. 207, this is an ‘old superstition’ which can be traced back to Wilamowitz in 1889. The theory is not unfounded. Of the epigrams attributed to Simonides, Μῆδοι appear in those assigned to Marathon (21 Page), Thermopylae (23 Page; cf. Hdt. vii 228), Artemisium (24 Page), Salamis (13 Page; cf. Plut., Hdt. Mal. 36Google Scholar), Plataea (17 Page = Thuc. i 132) and Cimon's Eurymedon campaign (46 Page). In another, the foreign invaders are called the ‘barbarous-tongued nations of the Medes’ (14 Page). Exceptions to this principle were regarded by Jacoby as later literary revisions, as in the Athenian epigram celebrating Marathon where Περσốν occurs (ML 26). This ethnic is generally regarded as a later addition to a Salamis monument, either directly after the event (Hammond, N. G. L., JHS lxxxviii [1968] 27Google Scholar) or a decade or so later (ML pp. 54–7). Other epigraphic exceptions are either regarded as later revisions (ML 12), or dependent on restorations derived from subsequent literary authorities (ML 24). The enemy at Marathon is also designated ‘Medes’ by Aeschylus, epigr. 2 Page. For discussion of epigrams cited above see Page, D. L., Further Greek Epigrams (Cambridge 1981) 186–302Google Scholar.

¹⁸ A. Pers. 236 mentions Μῆδοι, 765 a Μῆδος, who is either a mythical eponymous ancestor of the royal house (Podlecki, A. J., The Persians by Aeschylus [Englewood Cliffs, N. J. 1970] 92Google Scholar) or perhaps Cyaxares (Broadhead, H. D., The Persae by Aeschylus [Cambridge 1960] 192, 279Google Scholar), and 791 refers to ‘the Median (expedition)’—τὸ Μηδικόν.

¹⁹ The little that is known of this figure is almost totally contained in the Suda, s.v. Διονύσιος Μιλήσιος. Drews' attempt to resurrect him, (n. 13) 20–2, should be balanced with the more skeptical Pearson, L., Early Ionian Historians (Oxford 1939) 27, 110Google Scholar, as the tradition appears to be conflated with testimonies of other authors named Dionysius. For Greek literature concerning Persia see the listing of Cantarella, R. in La Persia e il mondogreco-romano (Rome 1966) 489–504Google Scholar, esp. 498 n. 49 for the Persica.

²⁰ See How and Wells (n. 7) ii 189 and Pearson (n. 19) 203–5. In the second century, Agatharchides of Cnidus rejected the mythological explanation on the basis of the accent, which was Πέρσας not Περσᾶς (On the Red Sea 6 = GGM i 113Google Scholar).

²¹ Macan attributed ‘Mede’ as a general designation for the Persian forces in Herodotus' account to one of his ‘sources’ (Hdt. IV–VI i 285 and Hdt. VII–IX i.2 429).

²² Drews' attempt to demonstrate that the early Persica were primarily ‘histories of the Persian Wars’ ([n. 13] 31 and 159 n. 46) is not very convincing. Although some of the surviving fragments include references to the Persian Wars, the total number of the remaining fragments is too few and their contents too diverse for such a surmise. Choerilus' late-fifth-century epic Persica cannot be used as a paradigm for the earlier Persica as it was an entirely different literary genre.

²³ D.L. ii 12 cites as his authority the third-century Lives of Satyrus. Different charges are listed by other writers apparently for two trials, a prosecution by Thucydides son of Melesias and another later by Cleon. J. A. Davison, CQ iii (1953) 39–45, who accepts both accounts, dates the earlier trial to 456/5. Meiggs, R., The Athenian Empire (Oxford 1972) 435–6Google Scholar, suggests 450. See, however, Dover, K. J., Talanta vii (1975)Google Scholar, Mansfeld, J., Mnem.⁴ xxxii (1979)Google Scholar, xxxiii (1980).

²⁴ Note the comment of Cary, M., CAH vi (1957) 56Google Scholar: ‘In the fourth century, the crime of “medism” became respectable in Greece, and it remained in honour so long as the Mede remained to medize with.’ See also Lewis, D. M., Sparta and Persia (Leiden 1978)Google Scholar.

²⁵ Starr, Chester G., Political Intelligence in Classical Greece, Mnemos. supp. xxxi (Leiden 1974)Google Scholar provides a concise discussion of this neglected topic.

²⁶ Media makes an earlier appearance in the OT as one of the regions where the Assyrians resettled the captive Israelites after the fall of Samaria in 721 (II Ki. xvii 6, xviii 11). The sons ofJavan are listed in Gen. x 4Google Scholar (cf. I Chron. i 7Google Scholar) as Elishah, Tarshish, the Kittim, and the Dodanim (LXX, ῾Ρόδιοι), generally identified with locations in Western Anatolia and Cyprus, while the last seems to clearly refer to Rhodes. von Rad, G., Genesis (London 1963) 140Google Scholar, interestingly suggests that the Table may be contemporaneous with Anaximander's famous map of the world.

²⁷ G. Cameron at one time attempted to demonstrate that the language of these prophecies indicated a date between 561 and 550, but he later indicated to me that he thought the matter was best kept open. See the summary of his early paper, ‘Media in the Old Testament’, JAOS li (1931) 370Google Scholar. In contrast Torrey, C. C., JAOS lxvi (1946) 7Google Scholar, argued that the passages were interpolations of the third century, but I prefer a period closer to the events.

²⁸ Eichrodt, W., Ezekiel (Philadelphia 1970) 380–1Google Scholar; cf. Zimmerli, W., Ezechiel (Neukirchen/Vluyn 1969) 644Google Scholar. Lud is problematic; it may refer to Lydia, which supplied mercenaries to Egypt during the Saite dynasty (Mellink, M. J., ‘Lud, Ludim’, IDB iii [1962] 1978–9)Google Scholar, but it appears elsewhere with Put (Isa. 66. 19, Heb. pwl is normally emended to pwṭ as the LXX reads phoud), so it may be a misreading for lubim, ‘Libya’ (Nah. iii 9Google Scholar). In the Table of Nations, Lud is a descendant of both Ham (Gen. x 13Google Scholar) and Shem (x 22), and the Ludim appear in the prophets with other African peoples (Ezek. xxx 5Google Scholar) as archers (Jer. xlvi 9Google Scholar), so it is possible that there were two peoples known as Lud who must be distinguished from each other. The older identification of Put (Old Persian Putaya) with Punt should be laid to rest; see Cameron, G., JNES ii (1943) 308Google Scholar. Jud. ii 23Google Scholar appears to locate Put and Lud in Asia Minor as among the peoples conquered by Nebuchadnezzar, but this Hellenistic treatise is filled with anachronisms and difficulties, so need not be given serious consideration.

²⁹ Rowley, H. H., Darius the Mede and the Four World Empires of the Book of Daniel (Cardiff 1935)Google Scholar persuasively contended that Darius the Mede was a product of confused traditions, unrelated to any historical personage. Berger, P.-R. in ZfA lxiv (1975) 192–234Google Scholar argues for some genuine sixth-century elements in the Book of Daniel, but see Hartman, L. F. and Di Lella, A. A., The Book of Daniel (Garden City, N.Y. 1978) 35–6, 50Google Scholar, and Koch, K., ‘Dareios, der Meder’, in The Word of the Lord Shall Go Forth: Essays in Honor of David Noel Freedman, ed. Meyers, C. L. and O'Connor, M. (Winnona Lake, Indiana 1983) 287–99Google Scholar for more recent discussion.

³⁰ Bickerman, E. J., JBL lxv (1946) 254–6Google Scholar, argued that the Persian titulary for the west was ‘king of Persia’, and defended the reliability of Ezra's account. In his opinion, both versions are genuine: the Hebrew is the oral proclamation of the herald; the Aramaic the official written decree. In contrast Torrey (n. 27) 11 viewed these passages as third century traditions and assigned the transition to ‘Persian kings’ to the reign of Artaxerxes I.

³¹ The lists of the subject peoples of the empire place Persia first and Media second in the Canal stelae inscriptions from Maskhūta, Shallūfa, and Kubri (Suez) = nos 8–10 in Posener, G., La première domination perse en Égypte, Inst. français d'arch. or., bibl. d'étude xi (Cairo 1936)Google Scholar; cf. the Apis stelae (nos 3–5). In the sculptures of Cambyses and Darius from the Serapeum stelae and the temple at Hibis, the Iranian rulers are depicted as Egyptian pharaohs (no. 31) performing the traditional religious rites. See Root, M. C., The King and Kingship in Achaemenid Art, Acta Iranica xix (Leiden 1979) 123–8Google Scholar. The Wadi Hammamat hieroglyphic inscriptions of Atiyawahy (Posener nos 24–30) and Ariyawrata (31–4), the governors of Coptos, from the end of Darius' reign to the 17th year of Artaxerxes I, designate them ‘Persian’ officials.

³² Vallat, F., CDAFI iv (1974) 161–70Google Scholar. The phrase is absent in the hieroglyphic inscriptions of the statue (J. Yoyotte, JA cclx [1972] 253–66) that was discovered on the Apadana mound at Susa, perhaps brought there by Xerxes after his punitive Egyptian expedition. See Hinz, W., AMI viii (1975) 120–1Google Scholar. Roaf, M., CDAFI iv (1974) 73–8Google Scholar, points out the peculiar Persian character of the statue and differences with New Kingdom reliefs, although it clearly reflects Egyptian workmanship.

³³ As on the statue of Udjaḥorresnet, the chief physician of Cambyses, and on the stele of Ahmose, general for the Egyptian satrap in the time of Darius (Posener [n. 31] no. 1b, lines 11–12, 18–19; cf. no. 6, line 5; see Lloyd, A. B., JEA lxviii [1982] 166–80Google Scholar). Aramaic papyri from the Elephantine military colony include a number of bearers of Babylonian and Persian names (e.g., Kraeling, E., The Brooklyn Museum Aramaic Papyri [New Haven 1953]Google Scholar nos 3, 4.24; Cowley, A. E., Aramaic Papyri of the Fifth Century BC [Oxford 1923]Google Scholar no. 5). Ethnic designations appear infrequently, as in Kraeling no. 5.17, which mentions a witness named ‘Atarparan b. Nisai, the Mede’, whose home must have been Nisaya, a district in Media (DB I 58). See also Driver, G. R., Aramaic Documents (Oxford 1957)Google Scholar for the important administrative documents of the Egyptian satrap in the late fifth century and Spiegelberg, W., SPAW (1928) 604–22Google Scholar for the demotic correspondence of the Khnum priests of Elephantine with the satrap Pherendates in the time of Darius I. For discussion of the papyri see Porten, B., Archives from Elephantine (Berkeley 1968)Google Scholar and Bresciani, E., SCO vii (1958) 132–88Google Scholar. The Asiatic/African bifurcation was traditional in Egypt, as is noted by Posener, G., ‘Sur l'orientation et l'ordre des points cardinaux chcz des Egyptiens’, in Göttinger Vorträge vom Ägyptologischen Kolloquium der Akademie (Göttingen 1965) 69–78Google Scholar.

³⁴ See Spiegelberg, W., Die sogenannte demotische Chronik des Pap. 215 (Leipzig 1914)Google Scholar. For discussion of the Ptolemaic date and nature of this document see Meyer, E., SPAW (1915) 296–9Google Scholar and Johnson, J. H., Enchoria iv (1974) 1–17Google Scholar.

³⁵ As in the funerary inscriptions of Petosiris at Tuna el-Gebel and those of Smatawyntefnakht; see Lloyd (n. 33) 177–8.

³⁶ RES 3022 = Garbini, G., Iscrizioni Minee = Seminario di Semitistica Richerche xGoogle Scholar (Ist. Or. Napoli 1974) no. 247. The Egyptian connections of the Minaeans also are reflected in the designation of their colony at Dedan (al-˓Ulā) as Ma˓īn of Muṣrān. For further discussion of the chronology of the Minaean kingdom and its contacts with the Levant see my discussion in ‘Dedanite and Minaean (South Arabian) Inscriptions from the Hismā’, in Annual of the Department of Antiquities, Jordan (1983). The pre-Islamic Safaitic texts from North Arabia even utilize ‘Mede’ to designate the Iranian successors to the Achaemenids, the Parthians and Sassanids, in their conflicts with Rome. See CIS v 4448Google Scholar and Winnett, F. V., Safaitic Inscriptions from Jordan (Toronto 1957)Google Scholar nos 78, 88. The attempt of Smith, S. in Archaeologica Orientalia in Memoriam Ernst Herzfeld, (Locust Valley, N.Y. 1952) 206Google Scholar n. 1, to find ‘Persian’ in the occurrences of frs in pre-Islamic inscriptions is puzzling since it is frequently attested in Old Aramaic and Hebrew texts as ‘horse(man)’. See Jean, C. F. and Hoftijzer, J., Dictionnaire des inscriptions sémitiques de l'Ouest (Leiden 1960) 237Google Scholar. For the pre-Islamic texts see Harding, G. L., An Index and Concordance of Pre-Islamic Names and Inscriptions (Toronto 1971) 465Google Scholar and Ryckmans, G., AOF xiv (1941) 54–6Google Scholar. The only occurrence of frs as ‘Persia’ in pre-Islamic texts that I am aware of is in a third-century AD inscription from Yemen. See Ryckmans, J., Le Muséon lxxx (1967) 508–12Google Scholar.

³⁷ The origins of the expression ‘Persian, born in Egypt’, are greatly disputed. During Ptolemaic and Roman times the meaning of the term evolved from a description of social position to a designation of secondary legal status. Oates, J. F., YCS xviii (1963) 1–129Google Scholar, argues that the term was selected because of its odious connotations. Boswinkel, E. and Pestman, P. W., Les archives privées de Dionysios, fils de Kephales (P. L. Bat. 22) (Leiden 1982) 56–63Google Scholar, suggest it arose in a military milieu and originally meant ‘son of a Persian (soldier)’. The demotic expression of Wynn ms n Kmj, ‘Ionian born in Egypt’ (e.g., P. dem. Ryl. 21) has been cited as the comparable Egyptian phrase, but note the demotic expression Mtj ms n Kmj in P. dem. Lille i no. 1 (243 BC). In contrast, the Greek expression Μῆδος τῆς ἐπιγονῆς occurs only once (P. Tebt. 815, fr. 2, R iii 53–4 [228/221 BC] and in the context of numerous Πέρσης τῆς ἐπιγονῆς and other epithets. See E. Bresciani, PP xxvii (1972) 123–8. The only occurrences of Prs in demotic are in official Ptolemaic documents or in the context of Ptolemaic propaganda: see Lorton, D., JEA lvii (1971) 160–4Google Scholar and the demotic ostracon recently published by Bresciani, E. in Das ptolemäische Ägypten, ed. Maehler, H. and Strocka, V. M. (Mainz 1978) 31–7Google Scholar. Native Egyptian documents utilize a different terminology: see P. Vindob. D 10000 and Zauzich, K.-T., Papyrus Erzherzog Rainer (Vienna 1983) 165–74Google Scholar for the use of ‘Mede’ in a third-century fragment of the prophetic book of the Lamb to Bocchoris (a reference I owe to L. Koenen) and the second-century derogatory reference to the Medes in the hieroglyphic texts of the Horus myth from Edfu: Chassinat, E., Le temple d'Edfou viGoogle Scholar = Mém. inst. français d'arch. or. xxxi (Cairo 1931), 214–15Google Scholar with Kees, H., NAWG (1930) 346–7Google Scholar. Greek references to Medes during the Ptolemaic period are rare (e.g. P. Lon. vii 2052Google Scholar), whereas Persians are frequently mentioned. See Launey, M., Recherches sur les armées hellenistiques i (Paris 1949) 563–80Google Scholar for discussion. In sum, Greek ethnics are normally precise for this period, while Egyptian ethnics are general and ambiguous.

³⁸ Gardiner, A. H., Ancient Egyptian Onomasticon (London 1947) i 81–2Google Scholar; Erichsen, W., Demotisches Glossar (Copenhagen 1954) 185Google Scholar; Crum, W. E., A Coptic Dictionary (Oxford 1939) 190Google Scholar. However, J. H. Johnson prefers to derive the Coptic word MATOI (‘soldier’) from Egyptian md3yw, an ethnic term for the Nubian people employed during the New Kingdom in the army and as ‘policemen’. Afterwards, the term was used to designate even native Egyptians as ‘police’. See Gardiner i 74 for the texts.

³⁹ Bengtson, H., ‘Die ‘Ionier’ in der Überlieferung des Alten Orients’, Philol. xcii (1937) 148–55Google Scholar = Kl. Schr. (Munich 1974) 76–82Google Scholar, and Sethe, , NAGW (1916) 131–3Google Scholar. The archaic hieroglyphic ḥ3w-nbwt (roughly meaning ‘around the baskets’, a metaphor for the Mediterranean islanders) was also applied to the Greeks during the Ptolemaic period (Gardiner [n. 38] i 206–7). The attempt of P. Montet, RA xxviii (1947) 129–44, to derive Hau-Nebwet from Ἑλλοί and vaῦs and trace Greek presence in Egypt back to the early third millennium is filled with multiple linguistic and historical improbabilities, as is observed by J. Vercoutter, RA xlvi (1947) 125–58 and xlviii (1949) 107–209.

⁴⁰ B.M. 21901. Text and translation in Wiseman, D. J., Chronicles of Chaldaean Kings (626–556 BC) in the British Museum (London 1956) 55Google Scholar. The same explanation for the Greek use of ‘Mede’ is offered by Duchesne-Guillemin, J. in ‘Media’, Kleine Pauly iii 1128Google Scholar: ‘Da die Griechen zuerst mit Medern unter Iranier in Berührung kamen nannten sie oft die Iranier Meder’. Cf. Bengtson, , Kl. Schr. (n. 39) 82Google Scholar n. 41. What this view fails to explain is the supplanting of ‘Mede’ as a general term for Iranians in Greece.

⁴¹ Harmatta, J., AAntHung xix (1971) 3Google Scholar, notes that the internal Iranian politics of Cyrus' era are generally neglected in standard treatments of the period. See Helm, P. R., Iran xix (1981) 85–90CrossRefGoogle Scholar for a recent statement concerning Median history.

⁴² B.M. 35382 = Nabonidus Chronicle ii 15. For text and translation see Grayson, A. K., Assyrian and Babylonian Chronicles (Locust Valley, N.Y. 1975) 107Google Scholar. Kent has suggested that the geographical designation Pārsa ‘seems to have been imposed by an outside source’ (p. 9). More recently, Windfuhr, G., Acta Iranica vGoogle Scholar = Hommages et opera minora, monumentum H. S. Nyberg ii (Leiden 1975) 466–8Google Scholar, has associated Parsua with a westward movement from the Iranian province of Parthawa (Parsu/awa).

⁴³ Weidner, E., AOF vii (1931–1932) 1–7Google Scholar; Hallock, R. T., Persepolis Fortification Tablets (Chicago 1969)Google Scholar nos 692–5, 2033, which he discusses in Seals and Sealing in the Ancient Near East, ed. Gibson, M. and Biggs, R. D. (Malibu 1977) 127Google Scholar. Although Herodotus provides the genealogy of Cyrus (i 107; vii 11), Cyrus is not mentioned as a member of the old royal dynasty of Anshan.

⁴⁴ Cyrus Cylinder = ANET 316, where Cyrus claims to be from ‘a family (which) always (exercised) kingship’.

⁴⁵ At Murghab: ‘I am Cyrus the (Great) King, an Achaemenian’ (Kent, CMa, CMb, CMc); Ur: ‘Cyrus, Great King, … King of the universe, king of Anshan’ (UET i 194); Babylon: ‘Cyrus, king of Anshan’ (v R 35). For Anshan see n. 43; the use of the archaic term in Mesopotamia must have helped to establish his legitimacy as the new ruler. Hansman, J., Iran x (1972) 101–25CrossRefGoogle Scholar, has recently identified the area near Maliyūn in southwest Iran as the city of Anshan.

⁴⁶ Gadd, C. J., ‘The Harran Inscriptions of Nabonidus’, AS viii (1958) 35–92Google Scholar = ANET 562. The text of Gadd for H2 i 42 reads šar (Māt?) mi-ṣir (ālu) ma-da-a-a (māt) a-ra-bi, ‘the king(s?) of the land(?) of Egypt, the city of the Medes, the land of the Arabs’. On the possibility that ‘king’ (šar) refers to all three names, Gadd notes, ‘To extend the idea of “king” over the two following descriptions would be contrary to Babylonian usage’ (76 n. 3). For further discussion see Ephcal, I., The Ancient Arabs (Leiden 1982) 180–91Google Scholar.

⁴⁷ Gadd (n. 46) 77.

⁴⁸ Tadmor, H., Assyriological Studies xvi (Chicago 1965) 351–64Google Scholar. For another chronological interpretation see Lambert, W. G., Proceedings of the 5th Seminar for Arabian Studies (1972) 53–64Google Scholar.

⁴⁹ Contra Röllig, W., ZfA lviGoogle Scholar = n.f. xxii (1964) 229.

⁵⁰ Nabonidus Chronicle ii 3=Grayson (n. 42) 106. Cyrus was in Ecbatana in 537 (Olmstead, , HPE 57–8Google Scholar). According to Herodotus, Ecbatana was also the capital city of Cambyses (iii 64).

⁵¹ The standard treatment of the relationship between Median and OP is Mayrhofer, M., ‘Die Rekonstruktion des Medischen’, AÖAW cv (1968) 1–23Google Scholar. Not all Iranian scholars accept his hypothesis. Gershevitch, I., TPhS (1964) 1–29Google Scholar, prefers to explain the differences on the basis of internal dialects and points out a number of ordinary non-technical ‘Median’ words in OP. Lecoq, P., Acta Iranica ii (1974) 55–62Google Scholar also argues in favor of pronounciation variants rather than Median loanwords, viewing the language of the OP inscriptions as a sort of unspoken koine utilized by western Iranians. However, he accepts ‘great king’ as Median in origin (58). The title ‘great king’ appears first with the Hittites, as it is completely absent in Sumerian titulature. It first occurs in Akkadian with Shamshi-Adad I, and continues in use to Assurbanipal. In Babylon, both Kurigalzu and Nabonidus used the title šarru rabū. For discussion and references see Seux, M. J., Epithètes royales akkadiennes et sumériennes (Paris 1967) 298–300Google Scholar. Harmatta (n. 41) 12 believes the traditional Achaemenid royal title was dropped by Cyrus after his subjugation of the Medes in 550 and M. A. Dandamaev suggests that Cyrus assumed the royal title of the Median kings in Persien unter den ersten Achämeniden (6. Jahrhundert v. Chr.) (Wiesbaden 1976) 94Google Scholar. For the propaganda aspects of this terminology see Cameron, G. in The Idea of History in the Ancient Near East, ed. Dentan, R. C. (New Haven 1955) 82–4Google Scholar, and Nylander's, C. essay, ‘Achaemenid Imperial Art’, in Power and Propaganda: A Symposium on Ancient Empires, ed. Larsen, M. T. (Uppsala 1980) 345–59Google Scholar.

⁵² Harmatta (n. 41) 11 provides a convenient list of Median terms borrowed by the Persians for state organization and administration. For the Median derivation of ‘satrapy’ (*xšaθrapā-, rather than OP xšaçapāvan) see Mayrhofer, M., Donum Indogermanicum Festgabe für Anton Scherer (Heidelberg 1971) 48Google Scholar and Fouilles de Xanthos vi (Paris 1979) 181–5Google Scholar; cf. Schmitt, R. in Studies in Greek, Italic, and Indo–European Linguistics offered to L. R. Palmer, ed. Morpurgo, A. D. and Reid, W. (Innsbruck 1976) 373–90Google Scholar.

⁵³ Iranian and Greek sources suggest the prominence of Medes in the Achaemenid court. See Mayrhofer, M., Onomastica Persepolitana (Vienna 1973)Google Scholar for Median names, but cf. Zadok, R., Israel Or. Stud. vii (1977) 111–12Google Scholar, who notes that Iranian names in Babylon have only ‘Median’ forms and that Herodotus designates as Persian individuals with names that have ‘Median’ features; this leads him to express reservation about Mayrhofer's linguistic criteria for ‘Median’. The attempt to distinguish Persians and Medes in Achaemenid art on the basis of different costume by Hinz, W., Altiranische Funde und Forschungen (Berlin 1969) 63–93Google Scholar, and von Gall, H., AMI v (1972) 261–83Google Scholar, has received some harsh criticism from Roaf (n. 32) 94–103 and Cook, J. M., The Persian Empire (London 1983) 230Google Scholar, who point out that Persians frequently adopted the Median dress (Hdt. i 135; vii 62). But Root (n. 31) 282 observes that the alternating costume perhaps suggests ‘the harmonious interaction of the two functional aspects—rather than the interaction of Medes and Persians’, i.e. ethnicity is not the theme in the artistic portrayal of the officialdom.

⁵⁴ See the listing of Achaemenid royal titles by Nylander, C., Orientalia Suecana xvi (1967) 157–66Google Scholar; cf. Lecoq, P., Acta Iranica iii (1979) 55–6Google Scholar. The Babylonian title varies for Xerxes including ‘King of the city of Persia, city of Media, Babylon and the Lands’. Babylon was dropped from the titulary in 481; see Cameron, G., AJSL lviii (1941) 323–4Google Scholar and Schmitt, R., AAntHung xxv (1977) 91–9Google Scholar for discussion.

⁵⁵ Ctesias (ap. Nicolaus of Damascus, , FGrH 324Google Scholar F 66) departs from this tradition, giving Cyrus a humble origin as the son of a shepherd and outlaw, but it is generally agreed that this is the product of the propaganda circulated at the court of Artaxerxes II for the purpose of discrediting Cyrus the Younger, a descendant of another branch of the Achaemenid family. See A. Cizek, AC xliv (1975) 547, but cf. Mallowan, M., Iran x (1972) 3Google Scholar, who prefers to trace the dissension between the two Achaemenid branches back to the court of Darius I.

⁵⁶ Nabonidus Chronicle ii 2–4 = Grayson (n. 42) 106. Harmatta (n. 41) 14–15, suggests there was a clash between the military aristocracy and the state bureaucracy, the former supported by the Deiocid dynasty and the latter siding with Cyrus, which conflicts with this reference.

⁵⁷ Since Astyages was without male issue (Hdt. i 109), Cyrus could have been viewed popularly in Media as the heir apparent to the throne and may have taken the hand of a Median princess after the defeat of Astyages to add to his legitimacy in the eyes of the Median populace (cf. X. Cyr. iii 5.19). Frye, R., The Heritage of Persia (New York 1963) 110Google Scholar suggests the continuation between the Median and Achaemenid state was virtually complete, ‘with only Cyrus replacing Astyages’.

⁵⁸ From its singular occurrence in the Babylonian Chronicle, Olmstead (HPE 38) suggested that Cyrus ‘had just revived the title’ of King of Parsa, but it is ‘king of Anshan’ that is regularly employed in the Babylonian context (see n. 45). There is also no evidence to support Bickerman's suggestion that this was the western titulary of the Achaemenid kings (n. 30), as its absence in Phoenician inscriptions (n. 59) and Thracian OP texts makes clear (DGh = Mayrhofer, M., Supplement zur Sammlung der altpersischen Inschriften, SÔAWWien. cccviii [1978] 16Google Scholar; cf. Hdt. iv 87, 91). The chronology for the period is vexed. The conventional date of 550 for the defeat of Astyages has been challenged by Drews, R., Historia xviii (1969) 1–11Google Scholar, who prefers a date of 554/3 for the event. The date of 547 for the fall of Sardis is also not secure; see the discussion in Grayson (n. 42) 107 and 282. Mallowan (n. 55) 6, prefers a date of ‘545 BC or possibly a year or two later’.

⁵⁹ Myres (n. 9) 97 minimized this fact by stressing that Cyrus' title appeared only once (Hdt. i 206), ignoring the other passages cited in the text above.

⁶⁰ The inscription (Byblos 13) was found in 1955 and has been frequently discussed. See Starcky, J., MUB xlv (1969) 259–73Google Scholar; Röllig, W., Neue Ephemeris f. Semit. Epigr. ii (Wiesbaden 1974) 1–15Google Scholar; Schiffman, I., Riv. studi fenici iv (1976) 171–7Google Scholar; Cross, F. M., IEJ xxix (1979) 40–4Google Scholar. Starcky dated it to 400, but most would now date it a century earlier. All but Rölling restore the broken line as [mlk]prs wmdy, ‘king of the Persians and the Medes’. Röllig observes that the letter before mdy can be read as m or k, the latter of which seems more likely. The other reading is dependent on the biblical phrase (Est. i 14, 18; x 2; Dan. viii 20Google Scholar) appearing in literature from the fourth century or later and not contemporaneous with the inscription. The phrase 'don mclakim, based on the Akkadian bēl šarrāni (Seux [n. 51] 318–19), is the Aramaic equivalent of OP ‘great king, king of kings’ (see Huss, W., ZDPV xciii [1977] 139Google Scholar). The Median/Achaemenid royal titles were probably all imitations of those of Assyrian kings, as was suggested by von Wesenkonk, O. G., Oriental Studies in Honor of C. E. Pavry (London 1933) 488–90Google Scholar.

⁶¹ The ‘seizure’ of Egypt by a ‘Persian man’ is also emphasized in the cuneiform inscriptions from the recently discovered statue of Darius at Suza (DSab and n. 32 above).

⁶² As argued by Mallowan (n. 55) 3. Darius' marriages to Atossa and Artystone, the daughters of Cyrus (Hdt. iii 88; vii 2), do not suggest any bitterness between the two ‘rival’ branches. The old view that the two Achaemenid families of Cyrus II and Darius I were located at Anshan and Parsa respectively is now discredited by evidence that they are different names for the same site (see n. 45). Far more attractive is T. Cuyler Young's suggestion to me that there was a signifıcant anti-Persian/pro-Median element in the revolt of Gaumāta. It should be noted that the Old Babylonian Bisitun text describes the rebel Gaumāta as ‘a certain Mede’: i 15 = von Voigtlander, E. N., The Bisitun Inscription of Darius the Great, Babylonian Version (London 1978)Google Scholar. Evans, J. A. S., Herodotus (Boston 1982) 57Google Scholar, presents a similar view of the revolt. The propaganda aspects of Bisitun have been emphasized by Dandamaev, (n. 51) 1–90, who views Darius as the usurper and Gaumāta (Bardiya) as the legitimate heir of Cambyses; cf. Bickerman, E. and Tadmor, H., Athenaeum lvi (1978) 239–61Google Scholar, Herrenschmidt, C., Annales (ESC) xxxvii (1982) 813–23CrossRefGoogle Scholar and Cook (n. 53) 44–57. In contrast, Wiesehöfer, J., Der Aufstand Gaumāta und die Anfänge Dareios I (Bonn 1978)Google Scholar, characterizes Gaumāta as a pretender and emphasizes the religious and social dimensions of the rebellion. On the problem of succession in Iranian kingship see Frye, R., AAntHung xxv (1977) 75–82Google Scholar. Another expression of Median unrest is the revolt of Fravartish, who claimed to be a descendent of Cyaxares and was supported by the Median palace guard of Darius; the largest number of enemy casualties and prisoners of any in the Bisitun inscription are recorded for this rebellion: perhaps as many as 34,425 dead and 18,000 captured (but the Aramaic Elephantine copy has 108,010, as Cook notes in his discussion of the battle; cf. DB II, 13–17 Kent). These outbreaks among the Medes help explain the new tone and emphasis in Darius' inscriptions.

⁶³ The development of the OP script is now generally assigned to the reign of Darius: see Schmitt (n. 1) 17–20 for a summary of recent discussion. In addition, Darius seems to have established the custom of adopting a throne name, perhaps to gloss over the illegitimate nature of his rule: ‘there is an apparent break in the tradition of naming between the older line of the Achaemenids on the one hand and Darius I and his successors on the other’ (Schmitt, R., AION xlii [1982] 93Google Scholar). For a more general treatment of the period see Hinz, W., Darius und die Perser: Eine Kulturgeschichte der Achämeniden (Baden-Baden 1976)Google Scholar and n. 52 above.