² Goetze, A., Die Annalen des Mursilis, MVAG 38 (1933)Google Scholar (referred to as AM).

³ A.M. 41.

⁴ Garstang, J. and Gurney, O. R., The Geography of the Hittite Empire, Occasional Publications of the British Institute of Archaeology in Ankara No.5, London, 1959Google Scholar (referred to as G.G.) 84 and map 1. The term “centre of gravity” is used by Goetze in his review of G.G., JCS 14 (1960), 43–8Google Scholar.

⁵ Orientalia 27 (1958), 395Google Scholar, and RHA 62 (1956), 10Google Scholar.

⁶ Achaeans and Hittites (1960), 33Google Scholar.

⁷ AS 18 (1968), 175Google Scholar.

⁸ Ibid. n. 59.

⁹ Kleinasien, in Kulturgeschichte des Alten Orients, 2. Neubearb., Munich (1957)Google Scholar (referred to as Kleinasien ²) 49, 84, 102, and map. Cf. also Goetze, A., Madduwattas, MVAG 32 (1927)Google Scholar (referred to as Madd), 148, and Kizzuwatna, in Yale Oriental Series, Researches,(referred to as YOSR) XXII (1940), 23Google Scholar. For similar views, see Barnett, R. D., AS 3 (1953), 78Google Scholar and Neufeld, E., The Hittite Laws, Luzac & Co., London (1951), 2Google Scholar.

¹⁰ See Mellaart, J., AJA 62 (1958), 26CrossRefGoogle Scholar, and Macqueen op. cit., 175.

¹¹ Goetze, , JCS 14 (1960), 47Google Scholar; Houwink ten Cate, Ph. H. J., The Luwian Population Groups of Lycia and Clicia Aspera during the Hellenistic Period, Leiden (Brill), 1961, 196Google Scholar; Professor Mellink in a personal letter states: “My inclination is to believe in habitation all along the coast and in the mountains even where we have no traces on the surface. An exploration of the archaeology of the areas involved would have to be organised in quite different ways from the usual survey and digging methods. So I see no reason why geographical identifications could not be postulated forthe coastal regions in the second millennium B.C.”.

¹² KBo. VI. 28Google Scholar, Obv. 6-9. See Cornelius, , Orientalia 27, 237–9Google Scholar, G.G., 64, Goetze, , YOSR XXII, 22Google Scholar. Goetze dates this crisis to the earlier part of the reign of Tudhaliyas III, pointing out that by 1. 16 of the text “we are referred to a period before Suppiluliumas ‘emerged’ (šara išparzašta) ‘and ascended the throne’”. Cf. Gurney, The Hittites, Penguin, 1952, 27Google Scholar.

¹³ P. 95, and see map 1.

¹⁴ See, for example, the maps of Gurney, , The Hittites, Penguin, 1952, xviGoogle Scholar, Cornelius, op. cit., 390, and Macqueen, op. cit., 176. Goetze favours a more north-easterly location for the Lower Land (see his map in Kleinasien ²). A full discussion of his location occurs in YOSR XXII, 22–4Google Scholar.

¹⁵ See, e.g. Garstang, , AJA 47 (1943), 37Google Scholar n.5, G.G., 64, Goetze, Kleinasien,² map, Houwink ten Cate, , The Luwian Population Groups etc., 194Google Scholar, Cornelius, op. cit., 388. On the advantages of Tyana's position, see Magie, D., Roman Rule in Asia Minor, Princeton University Press, 1950, 201Google Scholar.

¹⁶ Goetze concludes that the Lower Land “is that province of the Hittite empire which the Arzawean invasion hits” (YOSR XXII 23)Google Scholar, and the Lower Land also appears in Muwatallis' Prayer (KUB VI, 45 + 46, II, 38-40 – translated into English in G.G., 116-19) which lists all the deities of the Hittite kingdom.

¹⁷ Cf. G.G., 64.

¹⁸ If one accepts Goetze's location for the Lower Land (see n.14 above), then Tuwanuwa must have lain to the south of this territory.

¹⁹ Garstang's remark, AJA 47, 40Google Scholar, that there is no direct route westward into the territory of Arzawa proper from the vicinity of Konya, refers to the impassable barrier presented by the Sultan Daǧ, since at that time he was placing Arzawa in Pamphylia.

²⁰ Whether or not the Lower Land had been included in the Arzawan conquests, it was certainly once more under Hittite control during the reign of Suppiluliumas (see KUB. XIX, 22, 4–7Google Scholar, translated by Houwink ten Cate, , JNES. 25 (1966), 28Google Scholar). The reassertion of Hittite control in the general area may have been the prelude to further military operations by the Hittites (perhaps Suppiluliumas' reconquest of Arzawa mentioned in § 3 of the “Alaksandus Treaty” – Friedrich, , Staatsverträge des Hatti-Reiches in Hethitischer Sprache II, (1930)Google Scholar (referred to as Verträge II), 53, is to be assigned to this period). And Hanuttis' attack on Hapalla perhaps represented the first stage of a drive by Suppiluliumas against other hostile states.

²¹ AM.. 39.

²² Partial translation in G.G., 89-90.

²³ P.86.

²⁴ P.91.

²⁵ Quoted by Macqueen, op. cit., 177, n. 68.

²⁶ Ibid.

²⁷ Orientalia, 27, 294–5Google Scholar.

²⁸ Cornelius' proposal incorporates a series of superficially attractive linguistic equations. However, the use of such linguistic equations is a highly unreliable and often dangerously misleading method of determining geographical locations. One might mention in passing that Mayer and Garstang, using the same data as Cornelius, once made out an equally plausible case for identifying Mira-Kuwaliya with the area marked out by the Lycia Xanthos River in the west and the Cestros River in the east (JEA. 11 (1925), 28–9Google Scholar). They equated Kuwaliya of the Hittite text with the Lycian Kabalia (incorporating Balboura, Bubon and Oenoanda) as defined by Ptolemy (see Ramsay, W. M., The Cities and Bishoprics of Phrygia, Oxford, Clarendon Press, 1895, 266Google Scholar). Mira was equated with the Milyas district (Garstang also suggested that the name survives in the Lycian town Myra, and Tritsch in fact suggested that the Hittite Mira should be identified with the Myra, Lycian, Archiv Orientální 18, 1-2 (1950), 497Google Scholar) and Wiyanawanda with Oenoanda (cf. Laroche, RHA. 69 (1961), 61 no. 17)Google Scholar.

²⁹ Goetze, Madd. rev. §§ 38-39. Cf. G.G. 76.

³⁰ The frequent association of Mira–Kuwaliya with Pitassa and Sallapa is mentioned by G.G. 91 although the locations which they assign to these places is open to question.

³¹ See, e.g. G.G. 74, and map 1, Goetze, Kleinasien ² map, and JCS., 14 (1960), 48Google Scholar, Macqueen, op. cit., 176. Cf. also Barnett, R. D., CAH. II, new ed., fasc. 68 (1969), 4Google Scholar.

³² Orientalia 27, 393 and 396.

³³ e.g. AJA. 47, 40–41Google Scholar, and G.G. 75 f, and 85.

³⁴ AJA. 47, 40Google Scholar.

³⁵ Op. cit., 177.

³⁶ Discussed above. See notes 12 and 20.

³⁷ See, e.g. Macqueen, op. cit., 169. The question of Lukka's and Millawanda's locations is still a matter of some dispute. However, convincing arguments have been put forward for identifying Millawanda with Miletus (see G.G. 80-81, and cf. Albright, , AJA. 54 (1950), 168CrossRefGoogle Scholar, Huxley, , Achaeans and Hittites, 11 ff.Google Scholar, and Stubbings, , CAH. II, new ed., fasc. 26 (1964), 22Google Scholar). If this proposal is correct, it is very likely that places like Attarimma and Iyalanda (which appear in the Tawagalawas document, KUB. XIV, 3Google Scholar, and were almost certainly Lukka settlements) should be located in Caria (note G.G.'s very plausible identification of Iyalanda with Alinda, 78). On the assumption that Sallapa lay to the east or south of the Salt Lake, this would mean that Tawagalawas' envoy travelled several hundred miles before reaching the king at Sallapa (KUB. XIV, 3, I, 6 ff.Google Scholar). However the distance involved is not inconceivably greater than would be the case with G.G.'s and Macqueen's locations for Sallapa.

³⁸ JCS. 14 (1960), 48Google Scholar.

³⁹ The text of Muwatallis' prayer in KUB. VI, 45Google Scholar (translated in G.G., 118), where a reference to Sallapa is juxtaposed with references to Ussa, Parshunta, Huwatnuwanda (?) the River Hulaya, and the Lower Land, is highly compatible with Goetze's proposal.

⁴⁰ KUB. XIV, 15 ii 7 ff.Google Scholar, AM., 49.

⁴¹ A.M., 47.

⁴² e.g. by G.G., 76–7, and Macqueen op. cit., 177.

⁴³ Chiefly because of its association with Kaneš. See G.G. 22-3. Cornelius, , RHA. 62, 3Google Scholar, is more tentative in accepting the identification. Cf. Laroche, RHA. 69 (1961), 67, no. 45Google Scholar.

⁴⁴ Orientalia, 27, 383Google Scholar.

⁴⁵ Goetze, A.M. 51Google Scholar. There is no general consensus of opinion on the location of Walma. G.G. (p. 86) place it near Ipsus (suggesting that the town of Holmi indicates a survival of the name Walma), while Goetze places it between the lakes Trogitis and Caralis (see Kleinasien ² map). Houwink ten Cate (op. cit., 194 n. 3) expresses a preference for Goetze's proposal. Cornelius, while proposing a location for Walma at Holmi near the mouth of the Calycadnos suggests that a different Walma should be located at Holmi on the Maeander, (RHA. 62, 9)Google Scholar. The latter Walma is the one he associates with Mursilis' expedition to Arzawa.

⁴⁶ KBo. IV, 10 Obv. 32Google Scholar, translated in G.G. 67. See also G.G. 74.

⁴⁷ There is general agreement on this by Goetze, G.G., and Cornelius, as Houwink ten Cate points out (op. cit., 194).

⁴⁸ 69-72. See also Garstang, , “The Hulaya River Land and Daddassas”, JNES. 3 (1944)Google Scholar. Cf. Cornelius, , Orientalia 27, 390Google Scholar, map, and Houwink ten Cate, op. cit., 194 n. 4.

⁴⁹ Laroche's discovery was briefly reported in “The Times”, 25/9/73, 7Google Scholar.

⁵⁰ This was advocated by Güterbock, , JNES. 20 (1961), 86 n. 3Google Scholar.

⁵¹ See above n. 46 and Mursilis' treaty with Kupanta-Inaras § 9 (Friedrich, , Verträge I (1926)Google Scholar, G.G. 90).

⁵² Cf. Forrer, , Forschungen I/1, 17, and G.G. 91Google Scholar.

⁵³ See G.G. 91 par 4.

⁵⁴ See below.

⁵⁵ G.G. 90 § 10.

⁵⁶ Ibid.

⁵⁷ KBo. 1, 24Google Scholar. See G.G. 90 and n. 4 for references.

⁵⁸ Cf. Houwink ten Cate, op. cit., 195.

⁵⁹ Cf. Forrer, , Forschungen I/1, 16–17Google Scholar.

⁶⁰ See above.

⁶¹ For the difficulties of the route between the Pamphylian plain and the mouth of the Calcycadnos River, see D. Magie, op. cit., 266-7, and 1144 n. 24. Note, however, that in Roman times at least there was a coastal route which connected Pamphylia with eastern Cilicia (Magie, 270).

⁶² One might note in passing that Houwink ten Cate has shown that Cilicia Aspera was one of two major Luwian centres on the southern Asia Minor coast in the first millennium B.C. This of course does not mean that it was a major Arzawan centre in the second millennium B.C. However, in his account of various Luwian onomastic elements, Houwink ten Cate discusses the element Kuṷa which occurs in a few names in Hittite texts and which calls to mind Kuwaliya (op. cit., 152). He points out that in the town of Sinekkalessi the name Μιρασητας (Mira-zita/i) appears three times and κουαλις (Kuṷa+li) many times. The occurrence of these two names together leads him to make the suggestion – though very tentatively – that this area may have been the centre of the kingdom of Mira-Kuwaliya – i.e. on the boundary between Pamphylia and Cilicia Aspera (op. cit., 194).

⁶³ See A.M. 51.

⁶⁴ Published by Knudtzon, J. A., Die Zwei Arzawabriefe, 1902Google Scholar. See also Meyer, E., Geschichte des Altertums, I. 2, 474Google Scholar, and Barnett, R. D., AS. 3 (1953), 64 n. 5Google Scholar.

⁶⁵ A.M. 69, Cf. Friedrich, , Verträge II, 7Google Scholar.

⁶⁶ See, e.g., G.G. 93 ff., Macqueen, op. cit., 176 (map), and Goetze, Kleinasien,² map.

⁶⁷ The inference that the Astarpa formed the boundary of Arzawa as well as that of Mira-Kuwaliya need not automatically follow from Mursilis' text, as Forrer once asserted on the grounds that immediately after defeating Piyama-Inaras, Mursilis apparently crossed over into Arzawan territory (Forschungen I/1, 18Google Scholar. Cf. Macqueen, op. cit., 169.). The text may give a somewhat misleading impression here, and one should note the similar expressions used by Mursilis in describing his advance first to the Sehiriya River, and then to Aura, (A.M. 44 and 48Google Scholar). Note in particular obv. II 9-11, 48 … “and I joined up with him in Sallapa. Then I marched into the Land of Arzawa, and when I reached Aura …” There is of course no suggestion that the Sehiriya River or Aura lay in Arzawan territory.

⁶⁸ Belleten, V (1941), 42–3Google Scholar. For the identification of Habesos with Antiphellus, see Pliny, , N. H., V, 28Google Scholar. Cf. Bilabel, , Gesch., 66Google Scholar, Garstang, , AJA. 47, 47.Google Scholar, and Tritsch, , ILN Mar. 21, 1953, 448Google Scholar.

⁶⁹ The Apasas-Ephesus equation is strongly supported by Cornelius, who stresses the possible linguistic relationship between Apaasa, and Ephesus, (Orientalia 27, 395Google Scholar n. 3 and n.4), infers from the early cult of Artemis that Ephesus was an important pre-Greek state, and suggests that Parijana which is named amongst the Assuwan states (see KUB. XXIII, 11 and 12Google Scholar. Cf. Bossert, H. Th., Asia, 24Google Scholar, Forrer, , Klio XXX, 171Google Scholar) is to be equated with Priene. Note also Hanfmann's, G. M. A. article “A Hittite Priest from Ephesus”, AJA. 66 (1962), 1–4CrossRefGoogle Scholar. Hanfmann states (p. 4): “the appearance at Ephesus of a work such as the bronze group published here would be more readily explained if Bronze Age Ephesus was a royal residence and an important centre of a major kingdom”. It is an interesting speculation, but the isolated discovery referred to certainly does not allow one to draw any confident conclusions as to the significance of the site in the Late Bronze Age period, or its possible identification with Apasas.

⁷⁰ E.g. Huxley, , Achaeans and Hittites, 33Google Scholar.

⁷¹ Extract from a personal letter.

⁷² Garstang, , Belleten V, 43Google Scholar, rejected this possibility on the grounds that the text of Mursilis' Annals implies that Ukha-zitis fled before the Hittites to a neighbouring island. There is no such island in the vicinity of Attaleia/Antalya, while on the other hand, “a suitable island is found opposite Habesos in Castelorizo (Meis)”. However, the text is far from conclusive in this respect.

⁷³ Magie, op. cit., 266 refers to a “difficult route which led up from Side across the mountains to Lake Beyşehir” and then to Iconium (see also 1140 n. 18). One might compare with this route the modern Turkish highway which connects Konya with the Pamphylian plain by way of Beyşehir.

⁷⁴ G.G. 89, §§ 2-3 of Mursilis' treaty with Kupanta-Inaras.

⁷⁵ A.M. 73.

⁷⁶ A.M. 51.

⁷⁷ A.M. 53-7.

⁷⁸ A.M. 59. See also the references to Arinnanda, and Puranda, in Laroche RHA. 69 (1961), 64 no. 27 and 69 no. 58Google Scholar.

⁷⁹ A.M. 61.

⁸⁰ Prior to Mursilis' expedition, Mashuiluwas, who represented Hittite interests, could count on only limited support from his own people (A.M. 39), and even after the Hittite campaign had been brought to a successful conclusion, Mursilis still had considerable reservations about the loyalty of the people of Mira (A.M. 75).

⁸¹ A.M. 63-5.

⁸² Friedrich, Verträge II, 5Google Scholar.

⁸³ E.g. by Sidney Smith in a letter to “The Times” (referred to by G.G. 108) and Goetze (see Kleinasien,² map). Cf. Huxley, , Achaeans and Hittites, 33Google Scholar, and Houwink ten Cate, The Luwian Population Groups etc., 196Google Scholar, and contrast Cornelius, , Orientalia 27, 396Google Scholar. See also Eilers, W., OLZ (1935), 201–13Google Scholar, Sommer, , Die Ahhiyavā-Urkunden, Bayerische Akademie der Wissenschaften, Munich (1932), 157Google Scholar, and Albright, , AJA. 54, 168Google Scholar. Huxley (p. 20) draws attention to the fact that the Persians probably called Caria Krk, and refers to Herzfeld, E., Altpersische Inschriften, Berlin (1938) No. 14 p. 27 line 28Google Scholar.

⁸⁴ Friedrich, , Verträge II, 7Google Scholar.

⁸⁵ Cf. Magie, op. cit., 39–40, who suggests that the Maeander route was probably later in origin than the so-called Persian “Royal Road” (the Hermus route).

⁸⁶ From the western end of the Pamphylian plain, there was an important route of communication with the interior, perhaps leading north from the vicinity of Attaleia (see De Planhol, X., De la Plaine Pamphylienne aux Lacs Pisidiens, Paris (1958), 23–4Google Scholar). Cf. the modern Turkish highway which links Antaly a with Isparta, Burdur, etc. As described by Magie (p. 265), the route in Roman times “climbed the steep ascent to the mountains of Milyas past the entrance to the valley in which lay Termessus, and over a pass 3,000 feet above sea level into the plain of Isinda, near the headwaters of the Istanos Çay. From here it lay over the mountains to the plain north of Cibyra and thence to Laodiceia on the southern Highway. This was presumably the route used by the army of Mithradates when it attacked Termessus and perhaps invaded Pamphylia”. There is no reason to doubt that a similar route was used in the second millennium B.C. – a route providing direct contact between the Pamphylian plain and the western end of the Maeander valley.

⁸⁷ A.M. 75.

⁸⁸ G.G. 94.

⁸⁹ As indicated by § 9 of Mursilis' treaty with Kupanta-Inaras (G.G. 90).

⁹⁰ A.M. 75 (Sec. 50).

⁹¹ Macqueen (op. cit., 169) claims that the parting of the ways occurred at Sallapa since there are no names after Sallapa which are common to the lists of places mentioned in the two “western” campaigns undertaken by the Hittites. However, it is not the Hittite king's intention to give an exhaustive list of the places through which he passed. He mentions only those places where something of significance occurred which had a direct bearing on the campaign in question. It is quite possible that he passed through Aura on his way to Millawanda without mentioning this fact. However he does refer to Aura in his march to Arzawa, for it was at Aura that he received word of Piyama-Inaras' preparations to meet him at Walma (A.M. 49).

⁹² Cf. Goetze, Kleinasien,² 102Google Scholar.