¹ It is impossible to express here my indebtedness to all the works which have been written on Hittite geography. But I must acknowledge the help which I have received from “The Geography of the Hittite Empire” by John Garstang and O. R. Gurney (referred to as G.G.) an invaluable collection of source-material as well as a bold reconstruction of the map of second-millennium Anatolia. I have also to thank Prof. Gurney and Mr. James Mellaart for their help in discussing the evidence and for assistance with references. They must not, of course, be held responsible for the conclusions which I have drawn from it.

² Annals of Mursilis II, years 3–4. Text in Götze, , Die Annalen des Muršiliš, M.V.A.G. 38 (1933), (referred to as A.M.)Google Scholar

³ Letter, Tawagalawas, K.U.B. XIV, 3Google Scholar. Sommer, , Die Ahhijavā-Urkunden (1932) (referred to as A.U.)Google Scholar, ch. I. Partial translation in G.G. 111 sqq.

⁴ Friedrich, , Staatsverträge des Hatti-Reiches in Hethitischer Sprache (referred to as Verträge) 1 (1926)Google Scholar. Partial translation in G.G. 89–90.

⁵ ibid., sect. 9.

⁶ ibid.

⁷ ibid., sect. 2.

⁸ A.U. 1, 1 sqq.

⁹ ibid. 1, 16 sqq.

¹⁰ Apasas G.G. 88; Millawanda G.G. 80 and especially note 3.

¹¹ A.M. 39.

¹² ibid. p. 36.

¹³ Treaty with Mursilis; Friedrich, , Verträge II (1930)Google Scholar. Partial translation in G.G. 93–4.

¹⁴ ibid, §4. In §5 Manapa-Dattas is told not to covet a frontier-place of Hatti. This need not mean that the Seha-River-land bordered on the land of Hatti in its strictest sense, but merely that it bordered on a dependency of the land of Hatti (e.g. the Lukka-lands). Cf. page 169 above, where the dependent land of Mira is regarded as part of Arzawa.

¹⁵ A.M. 75.

¹⁶ Translated G.G. 95.

¹⁷ It also suggests a connexion with the sea. See G.G. 96.

¹⁸ Treaty of Muwatallis with Alaksandus (Verträge II) §11. Partial translation in G.G. 102–3.

¹⁹ Among other countries. K.U.B. XXIII, 11–12, lines 3–8Google Scholar. Translation G.G. 121.

²⁰ See G.G. 79.

²¹ §§1–4

²² §§5 sqq.

²³ See above p. 169 and note 3.

²⁴ A.U. 4 and 74.

²⁵ G G. 101.

²⁶ G.G. 103.

²⁷ Alaksandus-treaty, §14; G.G. 102.

²⁸ G.G. 104.

²⁹ Karkisa and Masa pp. 173–4; Warsiyalla p. 178.

³⁰ Alaksandus-treaty §11.

³¹ Verträge, I, 51 sqqGoogle Scholar.

³² K.U.B. XXIII, 11–12, line 6; G.G. 121Google Scholar.

³³ K.U.B. XIV, 1Google Scholar. Götze, , Madduwattaš (MVAG 32 [1927])Google Scholar. For Hapalla see pp. 24 sqq. (Rs. 21 sqq.)

³⁴ K.U.B. XIX, 22. G.G. 83 and 99Google Scholar.

³⁵ Götze, , Madduwattaš 38Google Scholar.

³⁶ E.g. G.G. 74, Götze, , Kleinasien² (1957), mapGoogle Scholar.

³⁷ P. 172 above.

³⁸ Manapa-Dattas treaty, §1.

³⁹ Tawagalawas-letter IV, 6.

⁴⁰ Kupanta-Inaras treaty, §§5–6. In fact he fled from somewhere near Pitassa, rather than from his homeland.

⁴¹ K.U.B. XV, 34Google Scholar and duplicates 33a, 33b and 38. See ANET. 352, Goetze, , J.C.S. 14 (1960) p. 48Google Scholar.

⁴² See p. 170.

⁴³ Cf. Otten, , J.C.S. 15 (1961) p. 112Google Scholar.

⁴⁴ See note 14.

⁴⁵ Deeds of Suppiluliumas, fragment 13, E i 7 sqq. See Güterbock, in J.C.S. 10 (1956) p. 65Google Scholar.

⁴⁶ Translation in Sturtevant, and Bechtel, , Hittite Chrestomathy (1935) pp. 42 sqqGoogle Scholar. See Goetze, , J.C.S. 14 (1960) p. 46Google Scholar.

⁴⁷ G.G. pp. 30–1. But cf. Goetze, , J.C.S. 14 (1960), p. 45Google Scholar; Güterbock, , J.N.E.S. 20 (1961), 95Google Scholar.

⁴⁸ Manapa-Dattas treaty, §1.

⁴⁹ Kupanta-Inaras treaty, §§5–6.

⁵⁰ Mś and Ḳrḳś. See Breasted, , Ancient records of Egypt III, §309Google Scholar; Wainwright, , J.E.A. XXV, 149Google Scholar.

⁵¹ A.U. I, IV, 6Google Scholar.

⁵² Alaksandus-treaty, §14.

⁵³ As far as can be seen from the Annals of Mursilis.

⁵⁴ e.g. G.G. 81; Huxley, G. L., Achaeans and Hittites (1960) p. 33Google Scholar.

⁵⁵ e.g. G.G. 80; Schachermeyr, , Hethiter u. Achäer (1935)Google Scholar.

⁵⁶ A.S. 4 (1954) 177–8, 5 (1955) 82Google Scholar.

⁵⁷ J.C.S. 14 (1960) p. 47Google Scholar.

⁵⁸ A.J.A. 68 (1964) p. 269Google Scholar.

⁵⁹ G.G. 84. It need not of course be confined to the Hermus valley, but probably included the Maeander and Cayster valleys as well.

⁶⁰ G.G. 96.

⁶¹ e.g. there would be no question of the control of “such widely separated districts as Miletus and the Caicus valley.”

⁶² Lines 105 and 173.

⁶³ In the Catalogue; Iliad II, 824–7Google Scholar.

⁶⁴ Leaf, , Troy, a Study in Homeric Geography (1912) pp. 180 sqqGoogle Scholar.

⁶⁵ For the suggestion that the Lycians migrated from the north-west to their later home in the south-west cf. Adams, Pythian, Bulletin of the British School of Archaeology in Jerusalem, No. 1 (1921), p. 4Google Scholar; Wainwright, in J.E.A. XXV (1939) p. 153Google Scholar.

⁶⁶ Jones, A. H. M., The Cities of the Eastern Roman Provinces (1937) p. 36Google Scholar. Ramsay, , Historical Geography, p. 156 sqqGoogle Scholar.

⁶⁷ For these areas see Burney, C. A. in AS. VI (1956) 179–203Google Scholar.

⁶⁸ G.G. 86. J. Mellaart points out that the Akar Çay is a very poor boundary and can be crossed by anyone with the help of a long pole. His suggestion is “to identify the R. Astarpa with that tributary which runs from Emirdağ north-east towards the Sakarya. As newly defined the Astarpa would form the eastern boundary with Hatti, whereas the Siyanti (upper Sakarya in the Turkmen Dağ) would form the northern boundary with Hittite territory, in this area the Land of Wilusa.” This is slightly different from my own suggestion, and Mr. Mellaart's knowledge of the ground lends weight to his proposal. But although the actual boundary of Mira is uncertain, the area in which the country lay is clear. My own identification of the Siyanti with the upper Porsuk Çay perhaps finds some confirmation in the tale of Madduwattas, who was given the land of the River Siyanti by the Hittite king. This name perhaps indicates that Madduwattas received rather more of the river-valley than was included in the Land of Mira. His fief may in fact have included the plain of Eskişehir, through which the Porsuk Çay flows. This rich plain on the edge of Hittite territory would give him an excellent base from which to expand.

⁶⁹ G.G. 75 sqq.

⁷⁰ J.C.S. 14 (1960) p. 47Google Scholar.

⁷¹ K.Bo IV, 13, i, 39 ffGoogle Scholar. K.U.B. VI, 45, ii, 34 ffGoogle Scholar.

⁷² In a letter.

⁷³ Ramsay, , Historical Geography, p. 360Google Scholar suggests there was a route across it in the classical period.

⁷⁴ E. I. Gordon has observed (in Gurney, , C.A.H. II,² XV(a), p. 20Google Scholar) that the mention of the “king of the Hurrians” in the “Annals of Tudhaliyas IV” makes it impossible to date this section of the text to so late a period. The Assuwa campaign must consequently be referred to another, earlier Tudhaliyas, presumably in the period before the kingship of Suppiluliumas. The order, and even the number, of kings in this period is uncertain, but campaigning in the west must be relegated to a date before the collapse from which the Empire recovered under Suppiluliumas—perhaps to the reign of “Tudhaliyas II” about 1450–1430, when “Tudhaliyas came to Arzawa” as mentioned in the Alaksandus treaty. Another text which may well be moved to this period is the Madduwattas text (Götze, , Madduwattaš, M.V.A.G. 32 [1927]Google Scholar) in which the “father of my Sun” may well be the Tudhaliyas who campaigned in the west, and the author, probably an Arnuwandas, a monarch of the period of the collapse. It was at just this time that western Anatolia became temporarily important in Near Eastern affairs (cf. the Arzawa letters in the Amarna correspondence and the operations of the Lu-uk-ki in Cyprus, (E.A. 38, 10)Google Scholar. These are surely Lukka-people from the north-west rather than Lycians from the south-west, and their actions are remarkably similar to those of Madduwattas and Attarsiyas the Ahhiyawan, who also descend on Cyprus from north-west Anatolia. These operations seem to fit this period of Hittite weakness just as well as the period of the final Hittite collapse.

Needless to say, the revised dating of these texts, if accepted, makes it impossible to use them as background-material for the Trojan War.

⁷⁵ The list is given in G.G. p. 105 and pp. 121–2.

⁷⁶ Page, D. L., History and the Homeric Iliad, (H.H.I.) (1959) pp. 102 sqqGoogle Scholar.

⁷⁷ The suggestion, though less likely, is by no means impossible. There seems to be a very decided move of north-western names (e.g. Lukka/Lycia, Karkisa/Caria, Masa/Mysia) towards the west and south-west coasts at the end of the second millennium, and Assuwa/Asia may well have moved with them. It is perhaps worth while to observe that the Trojan Catalogue in Homer (in all probability a document with its origins in the second millennium) looks much more interesting—and convincing—if we move the Lycians, Carians etc. back to their second millennium positions and assume a later misunderstanding after the movements have taken place.

⁷⁸ Cf. A.U. IX. This text has normally been dated to Tudhaliyas IV, because of the mention of Assuwa. But if Assuwa belongs to an earlier period, so presumably does A.U. IX. The suggestion of Forrer (Reall. d. Ass. 1, 157Google Scholar) that the text belongs to an earlier Tudhaliyas, is not then as “out of place” as e.g. Huxley, G. L. (Achaeans and Hittites, p. 5)Google Scholar supposes.

⁷⁹ By Forrer, , MDOG 63 (1924) 1–22Google Scholar; OLZ (1924) 113–18Google Scholar; KƖF 1 (1929), 252–72Google Scholar; RLA I (1928)Google Scholar s.v. Ahhiyava.

⁸⁰ Most recently Page, H.H.I.; Huxley, Achaeans and Hittites.

⁸¹ op. cit. pp. 10–14.

⁸² op. cit. p. 11.

⁸³ G.G. 80–1.

⁸⁴ Schachermeyr, F., Hethiter u. Achäer (1935)Google Scholar.

⁸⁵ For the texts see Sommer, A.U.; Güterbock, , Z.A. N.F. 9 (1936), 321 sqq.Google Scholar; Schachermeyr, , Hethiter u. Achäer, 41–2Google Scholar. For a summary of their contents see Huxley, , Achaeans and Hittites 1–10Google Scholar. The evidence for a connexion with Mira, (A.U. XVIII and V)Google Scholar is not sufficient to give any certainty that the two are in contact.

⁸⁶ A.U. I passim.

⁸⁷ K.U.B. XIX, 5. G.G. 95Google Scholar.

⁸⁸ Page, , H.H.I. 24, n. 2Google Scholar; Huxley, op. cit. p. 12.

⁸⁹ Sommer is, I am sure, right in saying that if the country had had an obviously Anatolian name there would never have been any suggestion that it lay overseas. See Sommer, , “Aḫḫijavā und kein Ende?”, (1937) p. 286 sq.Google Scholar, Page, , H.H.I. p. 8Google Scholar.

⁹⁰ The identification of Ahhiyawā with the Troad is not new. It appears for instance in the maps of Götze (Kleinasien ²) and Lloyd, (Early Highland Peoples of Anatolia (1967), p. 17)Google Scholar. What I hope I have done is to justify this position with reference to the rest of the geography of western Anatolia in a fuller and more formal way than is to be found in these works. Apart from Ahhiyawā, my map differs in several important respects from that of Götze.

It may be as well to say here that I do not believe in the historicity of the “Trojan War” as it is portrayed in the Homeric epics. The reasons for my unbelief are too many and too complicated for inclusion in this article, but I hope to give them elsewhere.

⁹¹ Accepting the usual reconstruction of the Hittite king-list as seen e.g. in G.G. p. ix. But see notes 74 and 78 above.

⁹² A.U. XIII.

⁹³ A.U. XV.

⁹⁴ A.U. X.

⁹⁵ Tawagalawas Letter, II, 58 sqq.

⁹⁶ A.U. I.

⁹⁷ A.U. IV.

⁹⁸ K.U.B. XXIII, 13Google Scholar. See G.G. 120–1; Page, , H.H.I. p. 28, n. 25Google Scholar; Huxley op. cit. pp. 7–8. This portion of the “Tudhaliyas Annals” is unaffected by the redating mentioned in note 74.

⁹⁹ See note 74.

¹⁰⁰ A.U. XVII.

¹⁰¹ See below p. 182 sq.

¹⁰² A.U. I.

¹⁰³ Page, , H.H.I, pp. 10–14Google Scholar.

¹⁰⁴ The author is an old man, apologising for the actions of his youth (Taw. IV, 52 sqq). Atpas and Piyamaradus, who appear in the text, appear also in K.U.B. XIX, 5Google Scholar, a letter from Manapa-Dattas of the Seha-River-land. This presumably means that Atpas, Piyamaradus, Manapa-Dattas and the author of the Tawagalawas letter are contemporaries. Since Manapa-Dattas was installed as king of the Seha-River-land by Mursilis II at the beginning of his reign, he is unlikely to have survived until Muwatallis, the successor of Mursilis, was an old man. The latter part of the reign of Mursilis is much more probable.

¹⁰⁵ Cf. Goetze, , C.A.H. II ², chapter XXI, p. 35Google Scholar.

¹⁰⁶ The attempted reconquest of the Syrian provinces by the rulers of the Nineteenth Dynasty was begun by Sethos I (1318–04). Even in his first year his armies reached the Lebanon, if not further.

¹⁰⁷ H.H.I, p. 14.

¹⁰⁸ p. 171 above.

¹⁰⁹ Manapa-Dattas treaty, §3, translated by O. R. Gurney in G.G. 93.

¹¹⁰ A.M. pp. 67–73. The translation is taken from Gurney, O. R., The Hittites (1952) pp. 115–16Google Scholar.

¹¹¹ He is also shrewd, for by frightening Manapa-Dattas before forgiving him he ensures the return of escaped prisoners and a constancy which another vassal might not have shown.

¹¹² e.g. Gurney, , The Hittites, p. 49Google Scholar.

¹¹³ The destruction of Troy VI by earthquake is placed by Blegen “some time within only a few years of 1300 B.C.” (Blegen, Troy, selections from C.A.H.² (1961), p. 11Google Scholar. Cf. Troy III, I, 18Google Scholar).

¹¹⁴ Huxley, , Achaeans and Hittites, p. 19Google Scholar.

¹¹⁵ At least two sites on the European shore of the Sea of Marmara, Selimpaşa and Toptepe, have yielded wheelmade Grey Minyan sherds of Troy VI–VII type. See French, D. H. in A.S. XV, (1965), p. 34 and fig. 13:1Google Scholar; and 16 (1966), p. 49 and fig. 6, nos. 32–3. There is also a site of Troy VI type near the tip of the Gallipoli peninsula (Blegen, , Troy III, 1, p. 18Google Scholar). The whole area has yet to be properly surveyed.

¹¹⁶ pp. 137–202. I am grateful to Mr. Mellaart for allowing me to see this article in manuscript.

¹¹⁷ If the Land of the River Siyanti is to be situated there—see note 68.

¹¹⁸ K.U.B. XXIII, 1Google Scholar = A.U. XVII.

¹¹⁹ Gurney, , The Hittites, p. 50Google Scholar.

¹²⁰ Page, H.H.I, p. 6Google Scholar.

¹²¹ Huxley, , Achaeans and Hittites, p. 16Google Scholar.

¹²² The dating of the destruction of Troy VIIa is a matter of some doubt. In the official publication Blegen says “about 1240 B.C.” (Troy IV, 1, p. 12Google Scholar). Archaeologically speaking, Mycenaean IIIB pottery was in common use, while the style of IIIA had not been wholly abandoned. There was no IIIC pottery. The following level, VIIb 1, built without an interval after the destruction, shows some IIIB and some IIIC pottery. Troy VIIA lasted “one, or at the most two generations”, and the duration of Troy VIIb 1 “can hardly have been less than a generation, and may have been more”. If we believe with Blegen that IIIB pottery came to its end shortly before 1200 B.C., then a date about 1250–1240 B.C. is indicated for the destruction of Troy VIIa. Desborough, (The Last Mycenaeans and Their Successors 1964)Google Scholar dates the end of IIIB to c. 1200 (p. 241) and the fall of Troy VIIa (p. 164) to “between 1250 and 1230”. If however we believe that IIIB pottery lasted until about 1180 B.C., (as for instance Stubbings in C.A.H. I², VI, p. 75)Google Scholar then a destruction of Troy VIIa about 1200 B.C. can be contemplated. The latter figure, it seems to me, unnaturally stretches the “one, or at the most two generations” since the earthquake which ended Troy VI, and I would support the earlier date. Tudhaliyas IV is to be dated to ca. 1265–1240, so that the fall of Troy VIIa is reasonably likely to have taken place during his reign.

¹²³ e.g. Cook, J. M., The Greeks in Ionia and the East (1962) p. 19Google Scholar.

¹²⁴ If “the king of Ahhiyawā retreated” from the king of the Seha River-land in A.U. XVI The sense of the verb is of course uncertain. See Huxley, , Achaeans and Hittites, p. 7Google Scholar.

¹²⁵ As the majority of Hellenists would have it.

¹²⁶ Chiefly Troy VIh. See Blegen, , Troy III, 1, p. 17Google Scholar.

¹²⁷ ibid.

¹²⁸ Troy IV, 1, 9.

¹²⁹ ibid.

¹³⁰ Troy III, 1, 18Google Scholar.

¹³¹ , Levanto-Helladic style, Troy III, I, 340, 347Google Scholar. Levanto-Mycenaean, shapes Troy III, 1, 44Google Scholar.

¹³² Troy IV, 1, 8Google Scholar.

¹³³ Troy III, 1, 17Google Scholar.

¹³⁴ Cf. Page, , H.H.I. 69–70Google Scholar.

¹³⁵ Troy IV, 1, 10Google Scholar.

¹³⁶ For contact with Cyprus and the Levant during the Amarna period see note 74.

¹³⁷ Ugaritica, II, p. 156, fig. 60Google Scholar: 11, Lachish, II, Pl. LXIII, 8.

¹³⁸ e.g. C.A.H. ² p. 12; Troy IV, 1, 6Google Scholar.

¹³⁹ Mellaart, , A.J.A. 62 (1958) 15–18Google Scholar.

¹⁴⁰ It was first suggested that Minyan ware was brought to Greece at the beginning of the Middle Bronze Age ca. 1900. (Mellaart, op. cit. p. 20). But excavations at Lerna in the Argolid have shown “gray ware of Minyan character” there in the EH III period. (Caskey, , Hesperia 29 (1960), pp. 296 sqq.Google Scholar). In fact many towns including Lerna, Tiryns, Asine and Zygouries, were destroyed at the end of EH II ca. 2100, and this seems to be the most likely time for the beginning of “Minyan” penetration.

Even if further research on the Early and Middle Helladic periods in Greece suggests that the origins of Greek Minyan ware are not after all to be found in north-west Anatolia, we are still left with the similarity of culture between Troy VI and the Greek mainland, and the older theory of a two-pronged movement from south-east Europe into Greece and north-west Anatolia. The possibility of speakers of an early form of Greek in and around the Troad remains.

¹⁴¹ Mellaart, op. cit. p. 19; Chadwick, , C.A.H. II², XXXIX (1963), p. 13Google Scholar; Piggott, , Ancient Europe (1965), p. 121Google Scholar, but cf. Palmer, , Mycenaeans and Minoans (1961), pp. 225 sqqGoogle Scholar.

¹⁴² See note 74.

¹⁴³ Troy VIIb 2Google Scholar. Blegen, , Troy IV, 1, 142 sqqGoogle Scholar.

¹⁴⁴ Blegen, , Troy IV, 1, 252–3Google Scholar.

¹⁴⁵ Dunbabin, T. J., The Greeks and their Eastern Neighbours (1957) pp. 65 sqqGoogle Scholar.

¹⁴⁶ Troy IV, 1, 147 and 250Google Scholar.

¹⁴⁷ Some Trojans may well have spent the intervening period on defensible heights such as the Balli Daǧ, within the Troad itself. (Troy IV, 1, 147Google Scholar).

¹⁴⁸ Carruba, in Compte Rendu de l'onzième Rencontre Assyriologique Internationale (1964) p. 42Google Scholar.

¹⁴⁹ Cook, J. M., C.A.H. II², XXXVIII (1961) pp. 4 sqqGoogle Scholar.

¹⁵⁰ Strabo XIII, 1, 3. See Leaf, , Strabo on the Troad (1923) 43 sqqGoogle Scholar.