Acknowledgements. Earlier versions of this paper were read by S.M. Wall; for discussions of a proximate draft and specific questions I am indebted to Prof. G. Huxley. A.A.D. Peatfield (Modhi and Petsophas) and J. Driessen (Dikta in Linear A and B). My greatest obligation, however, is to Sandy MacGillivray, who suggested that I write about Dikta, and encouraged me to complete this study. Writing, reading and reflection were carried out during my tenure of the Macmillan-Rodewald Studentship at the British School at Athens.

¹ ‘The Ancient Name of Zakro’ GRBS 8 (1967) 85–7.

² The first season of the resumed excavation took place in October-November 1986 (Macgillivray, J.A., Sackett, L.H.et al., BSA 82 (1987) 135–54Google Scholar), following a survey of the Roussolakkos area in spring 1983 (BSA 79 (1984) 129–59). A second season of explorations was carried out in March-May 1987 in BSA 83 259–82) below.

³ Line 3: ἐς ἐνιαυτόν; cf. Harrison, J.E., BSA 15 (1908–1909) 329–37Google Scholar, followed by Bowra, C.M., ‘A Cretan Hymn’ in On Greek Margins (Oxford 1970) 187Google Scholar, for Zeus as a year-god, coming ‘for the year’ in spring. Verbruggen, H., Le Zeus Crétois (Paris 1981) 105Google Scholar, prefers to understand ἐς ἐνιαυτόν as an annual invocation of a god who was always present in the sanctuary.

⁴ Ed. pr. of the hymn by Bosanquet, R.C., BSA 15 (1908–1909) 339–56Google Scholar, with the commentary of G. Murray, ibid. 357–65. Republished by M. Guarducci. Inscriptions Creticae (hereafter IC III.ii.2. The discovery of the hymn, on 24 and 28 May 1904, is described by Bosanquet in letters to his sisters, reprinted in Richard Can Bosanquet. Letters and Light Verse (hereafter RCB) (Gloucester 1938) 147–149.

⁵ Diodoros also seems to refer to the visibility of ancient ruins at Knossos, belonging to the ‘House of Hera’ (v.66.1: ἔτι ϰαὶ νυ̑ν δειϰνύται ϑεμέλιἁ Πέας οἰϰόπεδα ϰαὶ ϰυπαρίττων ἄλσος ἐϰ παλαιου̑ χρόνου ); cf. PM I 344, II 7, 334 n. 1, for the relationship of the later sanctuary of Rhea to the Minoan palace.

⁶ So Guarducci, M. in IC III page 9.Google Scholar

⁷ For the discovery of the site of the temple, which had also been sought at Kouremenos, (BSA 9 (1902–1903) 329–35)Google Scholar and inland at Aghios Nikolaos (ibid. 336–43; cf. BSA 8 (1901–2) 288 for Bosanquet's initial doubts over the location of the sanctuary at Roussolakkos), see BSA II (1904–5) 298–300.

⁸ IC III.ii. I.4—5; cf. BSA 15 (1908–9) 340 for the findspot of the inscription, whose provenience from Roussolakkos was assured by ‘the characteristic red earth’ with which it was encrusted.

⁹ IC III.iv.8.4–5; cf. the parallel formula used of the sanctuary of Athena Polias in 4–6: Ἀϑαν[α]ίαν Πολιάδα ϰαὶ ϑεοὺς ὅσσο[ις] ἐν᾿ Αϑαναίαι ϑύεται πάντας.

¹⁰ The story of Talos is told at Argonautika IV. 1635–1693.

¹¹ So Guarducci, , IC III page 9.Google Scholar

¹² Apollonios places Zeus' childhood in both a Diktaian (i.509) and an Idaian cave (ii.1234; iii.134), and, in a further passage, appears to assimilate Διϰται̑ος to Ἰδαι̑ος: ὅσσοι ἔασιν/ Δάϰτυλοι Ἰδαι̑οι Κρηταιέες,οὕς ποτε νύμφη/ Ἀγχιάλη Διϰται̑ον ἀνὰ σπέος ἀμφοτέρη̢σιν/ δραξαμένη γαίης Οἰαξίδος ἐβλάστησεν (i.1129–1132).

¹³ Bosanquet, , RCB 138–9, 147Google Scholar, in letters written in May 1903 and May 1904 notices the presence of sponge-boats from Kalymnos (9—10) and Kasos in Palaikastro bay; the harbour was again filled with boats from Kalymnos in October 1986.

¹⁴ Argonautika iv. 1690–1693. Guarducci, M., IG III.vii ‘Samonium Fanum’ p. 157Google Scholar, identifies the sanctuary of Athena founded by the Argonauts as that of Athena Samonia on Cape Sidheros-Salmonion. Athena Samonia is invoked together with Zeus Diktaios in the oath taken by the Hierapytnian settlers at Arkades (IC III.iii.5.11–13). Athena (Polias) was also worshipped at Itanos, however, (see n. 9 above) and will have been among the ‘gods in Dikta’ (see IC III.ii.1.7).

¹⁵ At v.80.4 Diodoros lists the sources on which his Cretan excursus (v.64–80) is based (τοι̑ς τὰ πιϑανώτερα λέγουσι ϰαὶ μαὶ μάλιστα πιστευομένοις): Doriades, Sosikrates, Laosthenidas and Epimenides ὁ ϑεόλογος. Diodoros' working methods elsewhere do not encourage confidence that he examined the tradition independently, and he is more likely to have consulted earlier sources through an intermediary; Jacoby, , FGH IIIb 468Google Scholar Anhang F1, Kommentar 342, suggests that Laosthenidas may have been ‘der exzerpierte Autor’ Nevertheless, Diodoros’ account is consistent, in placing Zeus' birth at Dikta and his upbringing on Ida, and seems to preserve a less contaminated tradition than other allusions.

¹⁶ Cf., e.g., the scepticism of Wilson, A.L., Minos 16 (1977) 102–3.Google Scholar In general, modern discussions of the location of Dikta have been concerned less with the site of the sanctuary than with the identification of a cave and a mountain. See further below pages 39–42.

¹⁷ The three suggested elements of an identification have all been admitted, but separately; Et. M.'s unbearded statue as a cult-image in the sanctuary: Guarducci, , IC III page 9Google Scholar; cf. Willetts, , Cretan Cults and Festivals (London 1962) 211Google Scholar; Δίϰταν ἐς ἐνιαυτὸν ἓρπε in the hymn to Zeus: Guarducci, , IC III p. 16Google Scholar: ‘Juppiter ut annuls festis in suo fano Dictaeo adsit invocatur.’ The identification of the Minoan town at Palaikastro with Diodoros' ruined city was suggested, tentatively, by Bosanquet, , BSA 15 (1908–1909) 351Google Scholar (cf. BSA 40 (1939–40), followed by Guarducci, , IC III page 5Google Scholar, and Faure, P., Fonctions des cavernes Crétoises (hereafter Fonctions)(Paris 1964) 97.Google Scholar

¹⁸ Strabo X.4.12 (478). Strabo's distances are given in round numbers (100 and 1000 stades), but, although the latter is inaccurate (Ida is 130 rather than 185 km. from the eastern coast of Crete), the former meets the location of Palaikastro almost precisely; cf. the longitude given by Ptolemy iii.17·9 for ‘Mt. Dikte’; 55° 30′, compared with Itanos 55° 40′ and Salmonion 55° 50′.

¹⁹ Both Apollonios (n. 11 above) and Kallimachos (Hymn to Zeus 4–7) appear to interchange Dikta with Ida, and their confusion was transmitted to Roman writers: cf. Martial iv. 1.1–2.

²⁰ The story of the concealment of Zeus by Rhea is told at Theogonj 453–491; see 481–4. Bosanquet, (BSA 15 348)Google Scholar seems to have assumed that Hesiod referred ‘expressly’ to the Diktaian Cave in this passage, but the variant reading Δίϰτον in 477 has no authority.

²¹ Diodoros v.70.6, quoted on page 1 above; cf. v.70.2; Apollodros i.1.6; Agathokles fr. 2 ap. Athenaios 375F: τὴν Διὸς τέϰνωσιν ἐπὶ τη̨̃ς Δίϰτης,ἐν ἡ̨̑ ϰαὶ ἀπόρρητος γίνεται ϑυσία.

²² Apollonios Rh. i.508ff.: Aratos Phain. 32ff., Dionysios Hal. Ant. Rom. ii.61.

²³ Schol. to Aratos Phain. 33: ἐγεννήϑη μὲν ἐν τῇ Δίχτῃ Παρέχονται ενδείξεις που οδηγούν στο συμπέρασμα ότι το τοπωνύμιο Δίχτη χαι ο επιθετιχός τύπος του Διχταίος τα οποία φέρουν ποιχιλία τοποθεσιών Κεντριχή χαι Ανατολιχή Κρήτη, ανήχουν χυριολεχτιχά στην αρχαία τοποθεσία του Παλαιοχάστρου που ερευνάται εχ νέου αυτή την περίοδο από την Βρεταννιχή Αρχαιολογιχή Σχολή. Εναλλαχτιχές αναγνωρίσεις της ταυτότητος της Δίχτης αποδίδονται στην σύγχυση των διαφόρων παραδόσεων για τη γέννηση χαι την ανατροφή του Κρητός Διός η οποία παρουσιάζεται σε αρχαίες πηγές της Ελληνιστιχής περιόδου. Η δημιουργία του μύθου του Διχταίου σπηλαίου αποδειχνύεται αποτέλεσμα παρόμοιας διαδιχασίας συγχώνευσης . Αλλο ένα στοιχείο σύγχισης εισάγεται με την παράθεση ενός Ελληνισ-τιχού οριαχού χειμένου. Η σωστή αναγνώριση της Δίχτης παρέχει μία τοπωνυμιχή συνέχεια που συνδέει μεταξύ τους τις φάσεις οίχησης της αρχαιολογιχής τοποθεσίας του Παλαιοχάστρου. Diod. v.70.2–4.

²⁴ Idaian Cave: cf. Diod. v.70.4: Diktaian Cave: cf. the references collected by Guarducci, , IC III 7–8Google Scholar, and see the discussion below, pages 000–00.

²⁵ Arkalochori, : excavation reports in BSA 19 (1912–1913) 35–47Google Scholar (J. Hazzidakis and J. Bambakas), and brief notices of Marinatos’ unpublished excavations in 1934–5 in AA 1934 252–4; 1935 248–56; cf. AJA 39 (1939) 134–6; cf. Faure, Fonctions 160–162, Nilsson, M.P., Minoan-Mycenaean Religion ² (Lund 1950) 60–1Google Scholar; Psychro: British excavations: Hogarth, D.G., BSA 6 (1899–1900) 94–116Google Scholar; cf. Faure, Fonctions 151–9, Nilsson, loc. cit. 61–4; for the identification of the Psychro as the Diktaian Cave, see Boardman, J.. The Cretan Collection in Oxford. The Dictaian Cave (Oxford 1961) 2–3Google Scholar: ‘more from regard for the opinion of its excavators and general usage in the last half-century than from any conviction about the accuracy of the identification.’

²⁶ Idaian Cave: Cook, A.B., Zeus II (CUP 1925) 935–939Google Scholar; Faure, Fonctions 99–131; Verbruggen, op. cit., n. 3 above, 71–99; for the recent resumption of excavations in the Cave, see Sakellarakis, J.A., PAE 1983 415–500.Google Scholar

²⁷ Lines 9–10. The Kouretes received the infant Zeus from Rhea at Dikta [and concealed him]. There is no reference to a birth-cave in the hymn.

²⁸ For the identification of Modhi, which appears to be current, see the discussion on page 41 below.

²⁹ IC III 7–8.

³⁰ See n. 11 above.

³¹ n. 22 above.

³² Ephoros ap. Strabo x.4.8 (476).

³³ Dion. Hal, ii.61.2.

³⁴ Plato. Laws 624–625.

³⁵ Laws 625B.

³⁶ See n. 26 above for discussions of the Cave's history: the great finds of bronzes belong to the chronological range in which the composition of the Odyssey can be placed.

³⁷ The reference is to the tradition, discussed below, that the infant Zeus was fed and protected by bees in a sacred cave on Crete; cf. Columella ix.2.3.

³⁸ IC III page 8, discussing CVA British Museum III H plate 32a and c, for which see Cook, A.B., JHS 15 (1895) 2CrossRefGoogle Scholar; id., Zeus ii, 929 with plate 42.

³⁹ Μεταμορφώσεων Συναγωγή xix, based on the Ornithogonia of Boios, which seems to belong to the late Hellenistic period (Jacoby, FGH IIIb Suppl. I 582Google Scholar; II 478 n.4).

⁴⁰ IC III page 8: ‘quod si probaveris, fani Dictaei famam iam VI a. Chr. n. late patuisse colligendum erit;’ cf. Cook, , Zeus ii 929.Google Scholar

⁴¹ Georgics iv. 152, quoted above; cf. Columella ix.2.3 with n. 37 above.

⁴² Antoninus' date is probably Antonine, and although he is drawing on an earlier writer (Boios, n. 39 above), the extent to which he himself interprets the tradition is uncertain; it is clear, however, that he is reporting a conflation, of the cave in which Zeus was brought up, and the place where he was born.

⁴³ Aelian, , Πεϱὶ Ζῶων xvii. 35.Google Scholar For Antenor, see Jacoby, , FGH IIIb 463.Google Scholar

⁴⁴ Cf. Apollodoros i.1.6: a similar process of rationa lisation, by which traditions concerning birth place and upbringing were combined, is evident in Antoninus Liber alis’ references to the cave in which Zeus was born.

⁴⁵ BCH 84 (1960) 188–96; cf. Fonctions 97: ‘Les seuls sanctuaires que l'on connaisse chez les Etéocrétois se situent ou bien sur les sommets, ou bien auprès des sources.’

⁴⁶ Strabo x.4.12; Et. M. s.v. Δίϰτη; schol, to Apollonios Rh. i.1130; Arrian FGH 156 fr. 65.

⁴⁷ Hoeck, , Kreta I (Göttingen 1823) 406–408.Google Scholar

⁴⁸ Hogarth op. cit., n. 25 above, 94–95.

⁴⁹ In addition to the old name, Καϰό Κεφάλι which Faure, Fonctions 98, records, Greger, S., Village on the Plateau (Manchester Ph.D. 1985)Google Scholar I reports that in the Lasithi village of Magoula the mountain peak is known as Madara.

⁵⁰ Bosanquet, , BSA 15 (1908–1909) 351.Google Scholar

⁵¹ Cf. Beloch, K., Klio II (1911) 433–5Google Scholar, and the exemplary discussion of the tradition by Toutain, J., RHR 64 (1911) 277–91.Google Scholar

⁵² Beloch, op. cit. 434.

⁵³ Bosanquet, op. cit., n. 50 above, 351 (‘possibly the cone of Modhi’); Faure, P., BCH 84 (1960) 189–96CrossRefGoogle Scholar; ibid. 86 (1962) 36–8: ‘un important sanctuaire sur le sommet du cône du Modhi qui domine de ses 539 mètres toute la région’; cf. Fonctions 94–9.

⁵⁴ See, e.g., Kerenyi, C., Zeus und Hera (Leiden 1972) 22Google Scholar; Lasserre, F., Strabon VII (Budé, Paris 1971) 156Google Scholar, in addition to Bosanquet, Guarducci and Faure, cited above.

⁵⁵ Sherds of the later periods were not evident on a visit to the peak sanctuary in April 1987.

⁵⁶ ADelt Chronika 1972 (appearing 1976) 652; cf. id., Kadmos 11 (1972) 101–112.

⁵⁷ See Peatfield, A.A.D., BSA 78 (1983) 277Google Scholar: ‘Modhi and Traostalos (a peak sanctuary S–E of Modhi, associated with Zakro) have very crude two-roomed structures delineated only by lines of stones, even though from their topographic position they should be more important;’ cf. Rutkowski, B., The Cult Places of the Aegean (Yale UP 1986) 80.Google Scholar

⁵⁸ Excavation by Myres, J.L., BSA 9 (1902–1903) 356–87Google Scholar; C. Davaras, op. cit., n. 56 above; cf. id.Kadmos 11 (1972) 102; ‘The new finds … make one of the richest series ever found in a peak sanctuary.’

⁵⁹ I owe this observation to Dr. J.A. Macgillivray. The evidence of cult practices found in Building I in the first season of the resumed excavation (BSA 82 (1987) 143–8) has been supplemented by the discovery, during the second season, of the torso of a substantial chryselephantine cult statue, discussed by the excavators on pages 267 below.

⁶⁰ Raison, J., Pope, M.. Corpus Transnumère du Linéaire A (Louvain 1981)Google ScholarPK Z 8a; Z 15; Z 11a; Z 12a.

⁶¹ See Karetsou, A., Godart, L., Olivier, J.-P., Kadmos 24 (1985) 128–31.CrossRefGoogle Scholar

⁶² Cf. Godart, L., Aux origines de l'Hellénisme. Hommages à H. van Effenterre (Paris 1984) 121–8Google Scholar, accepting only 13 correspondences between Linear A and B signs as established.

⁶³ Karetsou, Godard, Olivier, op. cit., n. 61 above, 130–1, Za 2.1.

⁶⁴ For Dikta in the Knossos tablets, see the references collected by Mcarthur, J.K., The Place Names of the Knossos Tablets, Minos 19.2 (1985) 29–31Google Scholar; di-ka-ta-de (allative form, Δίϰτανδε): Kn. Fp 7.2; F 866; Fh 5467.a; G 7509.1; di-ka-ta-jo (ethnic, Διϰται̑ος) Kn. Fp 1.2 (Ventris, M. and Chadwick, J., Documents in Mycenaean Greek ¹ (Cambridge 1973) no. 200).Google Scholar It should be added that much careful work that has been devoted to sorting the groups in which place names occur on the Knossos tablets has not so far suggested an eastern context for any of them; cf. Wilson op. cit., n. 16 above; Bennet, J., AJA 99 (1985) 231–49CrossRefGoogle Scholar; Killen, J.T., ‘The Knossos Tablets and the Geography of Mycenaean Crete’ in Mycenaean Geography, ed. Bintliff, J. (Cambridge 1977) 45Google Scholar: ‘The only place on the tablets which seems certain to have an easterly location is di-ka-ta(-de), which is pretty certainly Dikte.’

⁶⁵ So Karetsou, Godard, Olivier, op. cit., 131–2.

⁶⁶ See the references collected in n. 21 above.

⁶⁷ IC III.iv.9.77–78.

⁶⁸ IC III page 5: ‘quod nomen haud sane ineptum videtur, cum plana valus, floridis nunc olivetis magna ex parte obducta, hic illic palustrem adhuc speciem exhibeat.’

⁶⁹ Bosanquet, , BSA 9 (1904–1905) 298Google Scholar; BSA 15 (1908–9) 339; followed by Cook, A.B., Zeus II 930Google Scholar; Nilsson, op. cit., n. 25 above, 464–5; Faure, Fonctions 97; G. Huxley, op. cit., n. 1 above, 86–7.

⁷⁰ IC III.iv.9.67–71.

⁷¹ In Bosanquet's, translation, BSA 40 (1939–1940) 75Google Scholar: ‘The aforesaid boundaries clearly separate the land of the Itanians from that which formerly belonged to the Drag-mians and Praisians, and is now possessed by the Hierapytnians; and the sanctuary of Zeus lies outside the disputed land and is encompassed by enclosures and by various other landmarks and signs as was easily seen from the plans submitted to us.’

⁷² IC III.iv.9.81–82. Similar prohibitions are a common feature of Greek sacred laws; see, e.g., Sokolowski, F., Lois sacrées des cités grecques (Paris 1969)Google Scholar no. 65.78–81 (mystery cult of Andania: prohibition of wood cutting); 79 (Delphi: pasturage of ἱερά χωρα : 116 (Chios: prohibition of pasturage); 177 (Kos) 80–82: μὴ ἐξ·έστω δὲ τοὶς ϰοινωνου̑σι τω̑ν ὶερω̑γ [γε]ωργει̑ν τὰ τεμένη. Prohibitions of this kind are specific, and never merely formal. A prohibition of wood-cutting (μήτε ξυλεύηι), for example, implies the existence of a sacred grove; cf. Robert, L., Le sanctuaire de Sinuri près de Mylasa I (Paris 1945) 17Google Scholar; while the regulations concerning pasturage and cultivation suggest that the sanctuary enclosed areas of open ground.

⁷³ For the Hierapytnian claims, see ibid, lines 71–84. The stretch of walling exposed in 1905, which was identified as a περιβολος, extended for a distance of 36 m., and enclosed Blocks Χ and Π on the north and east. Since the wall reaches the main street at Χ 51–66, however, although its continuation was not traced, it seems that some parts of the Minoan town lay outside it, for example, Blocks B and M.

⁷⁴ The fertile ground of Kouremenos is separated from the remainder of the Palaikastro plain by the hill of Rizoviglo and is defined by a boundary wall of antique, but uncertain, date: ‘a wall, probably ancient’ Dawkins, R.M., Tod, M.N., BSA 9 (1902–1903) 329Google Scholar; ‘in view of its rubble and mortar construction … possibly contemporary with the Venetian occupation’ Peatfield, A.A.D., BSA 79 (1984) 156.Google Scholar It should be noticed, however, that the coast line has changed somewhat since antiquity, both through the accumulation of silt and as a result of the increase in the sea level in eastern Crete; so that an exact site for Heleia is hard to place.

⁷⁵ Bosanquet and Huxley cited in n. 69 above.

⁷⁶ Huxley ibid. 86–7: ‘Bosanquet once proposed to identify the ruined city mentioned by Diodoros with the remains at Palaikastro. Against the proposal is the fact that in Hellenistic times at least, the neighbourhood of Palaikastro was known as Heleia; it is so named in the Magnesian arbitration of ca. 130 B.C. in a dispute between Itanos and Hierapytna.’

⁷⁷ Huxley ibid. 87. It is possible that the city of Dragmos mentioned in the same inscription (IC III.iv.9.58, 68), whose former frontier with Itanos ran along the Karymes valley (59: [ὡς ὁ Σέδαμνος] ἐς Καρύμας), should be located in the area of Zakro, although an appropriate site remains to be identified.

⁷⁸ Bosanquet suggested the identification in BSA 15 (1908–9) 351; cf. n. 15 above.

⁷⁹ See, e.g., a letter dated 29 May 1904 (RCB 149): ‘Well, the inscription gives us the name of the place: for it proves that the temple of Dictaean Zeus was here, at P.K., and we know that the temple was at a place called Heleia, from the Itanos boundary inscription …’

⁸⁰ See n. 64 above.