We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 19 November 2010
SGO I 01/01/07 is a highly refined 20-line elegiac epitaph from Cnidus which probably dates to the first century BC. Despite the poem's striking formal features and poetic qualities it has received almost no attention since its first publication by Sir Charles Newton in 1863. The first four sections of this article demonstrate that the inscription represents an important witness to certain epigraphic practices and epitaphic traditions, including those of inscribed dialogue and lament. The final section then presents a reading of the poem which centres on the competing poetics of consolation, memory and forgetfulness proposed by the epitaph's two speaking voices.
* jmh201@cam.ac.uk. I am very grateful to a number of people for helping this project along in a variety of ways. The staff of the British Museum were enormously helpful and supportive, and I am particularly thankful to Ian Jenkins for providing me with information about and access to the inscription. Christof Berns generously gave me unpublished information about the Cnidian necropolis and has allowed me to reproduce a personal picture of tombs from it here. Richard Hunter first pointed out this largely overlooked poem to me, and I owe many thanks to him as well as to James Diggle, Marco Fantuzzi, Chris Faraone, Regina Höschele, Donald Mastronarde and Helen Morales for reading and commenting upon earlier drafts. John Ma also helped me with some crucial lessons in epigraphy, and I am very grateful to him for his patience and enthusiasm. Finally I would like to thank the two JHS referees, who each provided me with extremely valuable comments and corrections.
1 The expedition report consists of two volumes (the first being of plates): Newton, C.T., A History of Discoveries at Halicarnassus, Cnidus, and Branchidae (London 1862–1863).Google Scholar In 1861 Newton became the first Keeper of Greek and Roman Antiquities at the British Museum.
2 Newton (n.1) vol. 2.2 474.
3 Newton (n.1) vol. 2.2 475. He prints the text (inscription no. 54) at vol. 2.2 768–69; there is a fairly accurate drawing of the stone in vol. 1 (pl. XCIV).
4 The inscription appears in the following editions: Kaibel, G., Epigrammata graeca ex lapidibus conlecta (Berlin 1878) 204Google Scholar; GIBM IV.1 829b; Geffcken, J., Griechische Epigramme (Heidelberg 1916) 208Google Scholar; Peek, W., Griechische Grabgedichte (GG) (Berlin 1960) 438Google Scholar; Griechische Versinschriften: Band 1: Grab Epigramme (GVI) (Vienna 1980) 1874; Blümel, W., Die Inschriften von Knidos Vol. 1 (Bonn 1992) 303.Google Scholar
5 See pages 73–74 of Zingerle, J., ‘Lexikalische Beiträge’, Glotta 19 (1931) 72–85.Google Scholar
6 Kalinka, E., Tituli Lyciae linguis Graeca et Latina conscripti Fasc. 1 (Vienna 1920) no. 205.Google Scholar
7 Blümel (n.4) 160.
8 Barth, M. and Stauber, J., Inschriften Mysia & Troas: Olympene (Packard Humanities Institute CD no. 7 1996) no. 2764.Google Scholar
9 McCabe, D., Didyma Inscriptions: Texts and List (Packard Humanities Institute CD no. 6 1991) no. 564 II.Google Scholar
10 For example, McCabe, D., Teos Inscriptions: Texts and List (Packard Humanities Institute CD no. 6 1991)Google Scholar no. 564 II (third century BC or later); IG IX 12 2:232a (second century BC).
11 Many thanks to a JHS referee for pointing out these Latin examples.
12 One of the most interesting aspects of this inscription that is regrettably beyond the scope of this initial study is its relationship with the epitaphic language and conceits found in Roman elegy. For example, one thinks immediately of the similarity of premise in Propertius 4.7 and 4.11, poems which also explore the possibility of a dead woman addressing her husband (or lover) from beyond the grave.
13 Heberdev, R., Tituli Pisidiae linguis Graeca et Latina conscripti. Tituli Termessi et agri Termessensis (Vienna 1941).Google Scholar
14 Newton (n.1) vol. 2.2 no. 53 (768) (vol. 1 pl. XCIV) = Blümel (n.4) 409.
15 Christof Berns per e-litteras (4 May 2009). A drawing of the same monument (Tomb 1 in the south necropolis) appears as fig. 37 on page 155 of Love, I.C., ‘Knidos-Excavations in 1967’, Türk Arkeoloji Dergisi 16/2 (1967) 133–59Google Scholar; on the necropolis, see also page 153 of Love, I.C., ‘A preliminary report of the excavations at Knidos, 1969’, AJA 74/2 (April 1970) 149–55.CrossRefGoogle Scholar
16 The importance of considering sepulchral inscriptions in their ‘visual, archaeological, or topographical’ contexts is discussed on pages 4–9 of Oliver, G.J., ‘An introduction to the epigraphy of death: funerary inscriptions as evidence’, in Oliver, G.J. (ed.), The Epigraphy of Death (Liverpool 2000).Google Scholar
17 Newton (n.1) vol. 2.2 475.
18 Kaibel (n.4) 75: ‘I a. Chr. n. saeculo non antiquior titulus rara quadam verborum compositionisque arte conspicuus’.
19 Noteworthy features among the letter forms (each about 1.75cm tall) also include: epsilon with a shorter second horizontal hasta; non-lunate sigma and epsilon; cursive zeta (Z); theta with a small horizontal hasta that does not touch the inside of the circle; mu with oblique hastae which are straight and do not splay outward at the bottom; pi with a slightly shorter second vertical hasta; xsi with no vertical hasta (Ξ); full-size omicron; upsilon with no horizontal hasta at the bifurcation; phi with a reduced circle-size; and omega with slightly extended horizontal hastae.
20 For example, Blümel (n.4) 171 = Newton (n.1) vol. 2.2 745, no. 28 (vol. 1 pl. XC), first half of the second century BC (extremely similar but for the non oblique hastae of the mu); Blümel (n.4) 185 = Newton (n.1) vol. 2.2 no. 43, second to first century BC (with less-pronounced apices).
21 Peek (n.4) GVI 1873 = Bernand, E., Inscriptions métriques de l'Égypte gréco-romaine (Paris 1968) no. 33Google Scholar; for a discussion of the inscription, see Zucker, F., ‘Grabgedicht aus Herakleopolis’, Journal of Egyptian Archaeology 40 (1954) 118–23.CrossRefGoogle Scholar
22 An early example (431 BC) of a stone containing Konkurrenzgedichte is Peek GVI 20: two epigrams from a Polyandrion tomb in Athens' Kerameikos: cf. Peek, W., ‘Ein milesisches Polyandrion’, Wiener Studien 79 (1966) 218–30.Google Scholar
23 Gutzwiller, K.J., Poetic Garlands: Hellenistic Epigrams in Context (Berkeley 1998) 22.Google Scholar
24 ‘The meanings do not exclude each other, and in AP ἅλλo may sometimes cover both’: Gow, A.S.F., The Greek Anthology: Sources and Ascriptions (London 1958) 29.Google Scholar On ἅλλo, see also Fraser, P.M., Ptolemaic Alexandria (Oxford 1972) vol 2. 863Google Scholar; pages 453–54 of Wilcken, U., ‘Nachtrag zu den Kairener Zenon-Papyri’, Archiv für Papyrusforschung 6 (1920) 447–54.Google Scholar
25 Gutzwiller (n.23) 22 (with bibliography).
26 Cf. Gutzwiller (n.23) 22; see also pages 105–08 of Obbink, D., ‘New old Posidippus and old new Posidippus: occasion and edition in the epigrams’, in Gutzwiller, K.J. (ed.), The New Posidippus: A Hellenistic Poetry Book (Oxford 2005) 97–118.Google Scholar
27 ‘Die drei Teile der Inschrift, durch zweimaliges ἅλλo getrennt, bieten nicht, wie meist in solchen Fällen, Variationen desselben Gedankens, sondern zuerst fortlaufende Mitteilungen und dann ein Gespräch’: Zucker (n.21) 119.
28 See R.D. Anderson, P.J. Parsons and Nisbet, R.G.M., ‘Elegiacs by Gallus from Qasr Ibrim’, JRS 69 (1979) 125–55Google Scholar (the pentameters on the Gallus papyrus are indented): ‘But Greek papyri, so far as I know, do not indent; nor do Greek inscriptions before the imperial age' (130). In Latin elegy indentation of the pentameter occurs as early as 101 BC (an elegiac inscription from Corinth): see pages 86–87 of S., Dow, ‘Corinthiaca VI: the Latin elegiacs of ca. 101 B.C. VII. Greek elegiacs of the Roman Empire’, HSPh 60 (1951) 81–100.Google Scholar
29 Fantuzzi, M. and Sens, A., ‘The hexameter of inscribed Hellenistic epigram’, in Harder, M.A., Regtuit, R.F. and Wakker, G.C. (eds.), Beyond the Canon (Hellenistica Groningana 11) (Groningen 2006).Google Scholar Fantuzzi and Sens demonstrate that there were poets outside the Alexandrian literary élite (such as ‘professional epigrammatists who worked on commission and had no real connection to the inner centers of cultural privilege or authority’) who nevertheless abided by ‘“Callimachean” metrical practice' (106).
30 Zucker (n.21) 121.
31 There is already evidence from the Classical period for professional poets writing epitaphs (for example, Eur. Tro. 1188–91).
32 See pages 48–50 of Allen, F., ‘On Greek versification in inscriptions’, Papers of the American School of Classical Studies 4/3 (1888) 35–204.Google Scholar
33 Cf. Zucker (n.21) 119, 121.
34 Herodes' epitaph for a woman named Aphrodisia is signed Ἡρώδηζ ἔγραψενthe one for Apollonius is signed Ἡρώδυ: Bernand (n.21) nos 5, 35. On the poems, see von Wilamowitz-Moellendorf, U., ‘Zwei Gedichte aus der Zeit Euergetes' II’, Archiv für Papyrusforschung 1 (1901) 219–25.Google Scholar I thank an anonymous referee for pointing out that IG 12.5.30 (Paros, second century BC) represents another example of a signed epitaph: the last line tells us that διOνύσιOζ Mάγνηζ ΠOιηΤήζ ἔγραψεν.
35 Cnidus was a Doric foundation. On the Doric koine see Buck, C.D., The Greek Dialects (Chicago 1928)Google Scholar: ‘a type of modified Doric […] prevails in the inscriptions of the last three centuries B.C.’ (it retained ‘a majority of the general West Greek characteristics, but with a tendency to eliminate local peculiarities’) (176).
36 The form Λάθα occurs seven times in Sacco's repertory of Lethe references in funeral epigram (six of these belong to the second/first centuries BC): Sacco, G., ‘Lethe negli epigrammi funerari’, Epigraphica 40 (1978) 40–52.Google Scholar
37 Λάθαζ Tr. Adesp. 372 Snell-Kannicht (line 4); AP 7.711 line 6; Peek (n.4) GVI1585 (Cyzicus, second to first century BC). δύσΤανε: Soph. OC 542; Eur. Med. 442, 995; Theoc. 15.31.
38 Kaibel (n.4) 204: ‘Utraque enim binorum distichorum stropha ionice, utraque ternorum dorice scripta est […] sententiarum tamen quam dicunt strophicam responsionem frustra quaeres’.
39 See Sens, A., ‘Doricisms in the new (and old) Posidippus’, in Acosta-Hughes, B., Kosmetatou, E. and Baumbach, M. (eds), Labored in Papyrus Leaves (Washington DC 2004) 65–83Google Scholar on the potential socio-political significances of Doric forms in epigrams by Posidippus of Pella.
40 For ‘the intense privacy’ (and especially ‘domestic privacy’) of some Hellenestic grave monuments, see Bruss, J., Hidden Presences: Monuments, Gravesites, and Corpses in Greek Funerary Epigram (Leuven 2005) 56–57.Google Scholar
41 See page 31 of Alexiou, M., The Ritual Lament in Greek Tradition (Lanham MD and Oxford 2002)Google Scholar: ‘the ancient ritual custom of passionate invocation at the tomb is known to have persisted, alongside the literary convention of epigrams in dialogue form’.
42 For example, κOύφα σOι| χθών πάνωθε πέσOι, γύναι, ‘May the earth lie light above you, lady’ (462–63).
43 For a Hellenistic sepulchral epigram that plays with the idea of such a dialogue, see Philetas of Samos 2 Gow-Page = AP 7.481. In the first two lines of this epitaph a young girl's grave stele speaks, but in the second distich the narrative voice informs us that the girl's words are a response to her father (who is imagined as mourning at her tomb): χ μικκ Τάδε παΤρ λγει πάλιν· ῎|σχεO λύφαζ (‘and the little girl says these things back to her father: “Leave off your grief”’): cf. Bettenworth, A., ‘The mutual influence of inscribed and literary epigram’, in Bing, P. and Bruss, J.S. (eds), Brill's Companion to Hellenistic Epigram (Leiden 2007) 69–93CrossRefGoogle Scholar, page 77.
44 Peek (n.4) GVI 1873–77, 1880. A striking Roman example of a dialogue-epitaph is CIL 6.1252, a three-part epitaph for the wife of a freedman who lived in the reign of Tiberius. On the front of the stone is a six-line elegiac epigram in Greek; each side of the stone also contains another poem (of 12 lines on the left; of 14 on the right). These three ‘parts’ can easily be read as Konkurrenzgedichte: see Courtney, E., Musa Lapidaria (Atlanta 1995)Google Scholar no. 180 with commentary; cf. also Hutchinson, G., Propertius: Elegies Book IV (Cambridge 2006) 231Google Scholar (on the same page Hutchinson mentions Atthis' epitaph in passing). I am grateful to a JHS referee for pointing out this inscription.
45 In Peek (n.4) 1875, the wife (Thermis) speaks first to her husband (Simalos); in 1877 the inscription (second century AD?) is from a sarcophagus that the husband has built for both himself and his wife.
46 Tsagalis, C., Inscribing Sorrow: Fourth-Century Attic Funerary Epigrams (Berlin and New York 2008) 260CrossRefGoogle Scholar: ‘the composer of the epigram invites his fictive reader to assume that the speaker in the “second person” verbs of the [first] three verses is Melite's husband’.
47 On γνμαι in epitaph (with special reference to fourth-century Attic cases) see Tsagalis (n.46) 17–61.
48 Cf. also in Roman elegy Propertius 2.13.57–58: sed frustra mutos revocabis, Cynthia, Manis, | nam mea qui poterunt ossa minuta loqui?
49 The use in stonecutters' workshops of epitaphic ‘handbooks’ (which would have contained thesauri of appropriate sepulchral phrases and motifs) has long been postulated: for the Latin-language evidence see Cagnat, R., ‘Sur les manuels des graveurs d'inscriptions Romaines’, Revue de Philologie 13 (1889) 51–65Google Scholar; see also pages 10–11 of Mclean, B., An Introduction to Greek Epigraphy of the Hellenistic and Roman Periods (Ann Arbor 2002).CrossRefGoogle Scholar
50 Cf. Peek (n.4) GVI 1833–72.
51 Important examples of this include Callimachus 31 Gow-Page = AP 7.524; Dioscorides 22 Gow-Page = AP 7.37; Meleager 121 Gow-Page = AP 7.79; Leonidas 70 Gow-Page = AP 163; Antipater of Sidon 22 Gow-Page = AP 164; also Posidippus 19 Gow-Page = 142 Austin Bastianini = APl. 16.275. On dialogues in epitaphs and epigram (with a number of examples), see pages 306–28 in Fantuzzi, M., ‘The epigram’, in Fantuzzi, M. and Hunter, R. (eds), Tradition and Innovation in Hellenistic Poetry (Cambridge 2004) 283–349.Google Scholar
52 Newton (n.1) vol. 2.2 769.
53 Page 24 of Day, J., ‘Rituals in stone: early Greek grave epigrams and monuments’, JHS 109 (1989) 16–28.CrossRefGoogle Scholar
54 Rossi, L., ‘Lamentazioni su pietra e letteratura trenodica: motivi topici dei canti funerari’ ZPE 126 (1999): 29–42.Google Scholar
55 Derderian, K., Leaving Words to Remember: Greek Mourning and the Advent of Literacy (Leiden 2001) 31Google Scholar, n.64. Alexiou (n.41) distinguished θρνOζ (performed by professional mourners) and γOζ as the two principal forms of ritual lament (102–03). On Homeric lament see also Tsagalis, C., Epic Grief: Personal Laments in Homer's Iliad (Berlin 2004).CrossRefGoogle Scholar
56 Derderian (n.55) 34. Achilles is explicit on this point at Il. 23.9: ‘let us lament Patroclus, for this is the due of the dead’ (ΠάτρOκλOν κλαίωμεν γρ γέραζ σΤ θαννΤων).
57 Derderian (n.55) 36; cf. Alexiou (n.41) 166.
58 Alexiou (n.41) identified the contrast of past and present as a theme that recurs ‘in the lament throughout the Greek tradition’ (171), though with some diachronic variation in the details: 165–71.
59 Alexiou (n.41) 171.
60 Alexiou (n.41) 178–81.
61 Alexiou (n.41) 182–84.
62 For commentary on this section, see Russell, D.A. and Wilson, N.G., Menander Rhetor (Oxford 1981) 346–50.Google Scholar
63 See Heath, M., Menander: a Rhetor in Context (Oxford 2004).CrossRefGoogle Scholar A less substantial and detailed treatment of rhetoric appropriate to epikedion speeches is to be found in [Dionysius of Halicarnassus] Ars Rhetorica (Section 6, μέθOδOζ. πιΤαφων).
64 On Homer in Menander Rhetor and the view of Homer as the inventor of ancient rhetoric, see Cairns, F., Generic Composition in Greek and Roman Poetry (Edinburgh 1972) 34–40.Google Scholar
65 Men. Rh. 434 line 19 Russell-Wilson.
66 Men. Rh. 434 lines 11–12.
67 Τζ φOρμάζ Men. Rh. 434 line 16.
68 Poets of the later Hellenistic period also seem to have experimented with poetic forms of male lament, of which Moschus' Lament for Bion is a clear example; largely modelled on that poem is Ovid Am. 3.9 (the ‘Lament for Tibullus’): see Reed, J.D., ‘Ovid's elegy on Tibullus and its models’, CPh 92/3 (1997) 260–69.Google Scholar On the lament for Moschus (as well as Bion's ‘Epitaph for Adonis’) and Menander Rhetor's precepts for funereal ‘monody’, see Manakidou, F., ‘Epitaphios Adonidos and Epitaphios Bionis: remarks on their generic form and their content’, MD 37 (1997) 27–58.Google Scholar
69 Men. Rh. 436 line 26, 437 line 1.
70 Suter, A., ‘Male Lament in Greek Tragedy’, in Suter, A. (ed.), Lament: Studies in the Ancient Mediterranean and Beyond (Oxford 2008) 156–80.Google Scholar
71 Wright, E., The Forms of Lament in Greek Tragedy (Unpublished PhD Dissertation, University of Pennsylvania 1986).Google Scholar On lament in Alcestis, see now also Swift, L., The Hidden Chorus: Echoes of Genre in Tragic Lyric (Oxford 2010) 350–64.CrossRefGoogle Scholar
72 See Suter (n.70) Appendix at 171–72.
73 The ‘full’ male laments (according to Wright's (n.71) criteria) are Teucer's in Soph. Ajax 992–1039; Creon's in Soph. Ant. 1261–1346; Oedipus' in Soph. OT 1307–366.
74 Suter (n.70) 165.
75 Suter (n.70) 166.
76 Suter (n.70) 164–65.
77 Peek (n.4) GVI 2005.
78 Wood, S., ‘Alcestis on Roman sarcophagi’, AJA 82 (1978) 499–510CrossRefGoogle Scholar; LIMC 1 s.v. Alkestis. For an ancient meditation on Alcestis' exceptional virtue, see Pl. Symp. 179b–c. On ‘tragic vocabulary’ in fourth-century Attic funerary epigrams, see Tsagalis (n.46) 269–71 (Tsagalis finds four parallels with lines from Alcestis in his corpus epitaphs).
79 So Lattimore, R., Themes in Greek and Roman Epitaphs (Urbana 1962) 203.Google Scholar
80 See Lattimore (n.79) 164 and, for example, Callimachus' epitaph on Saon of Acanthus (41 Gow-Page = AP 7.451).
81 A Homeric idea: see Sourvinou-Inwood, C., ‘Reading’ Greek Death: To the End of the Classical Period (Oxford 1995) 139–40.Google Scholar
82 On the theme, see Lattimore (n.79) 217–18.
83 Hes. Theog. 227 calls Lethe the daughter of Eris (not of Hades, as Theios calls her in our poem), and the ‘plain of Lethe’ is mentioned in Arist. Ran. 185, though in neither of these texts does the image of Lethe water or any ‘water of oblivion’ appear.
84 On ‘The Myth of Er’ and reincarnation, see Halliwell, S., Plato Republic 10 (Warminster 1988) 21–23Google Scholar; on the River 'Aμελήζ, memory and death, see Vernant, J.-P., Mythe et Pensée chez les Grecs (Paris 1969) 79–94.Google Scholar
85 Albinus, L., The House of Hades: Studies in Greek Eschatology (Aarhus 2000) 143.Google Scholar See West, M.L., Orphic Poems (Oxford 1983) 1–3Google Scholar for the view that there was no ‘Orphism’, only Orphic texts.
86 On the Orphic leaves, see generally Pugliese Carratelli, G., Le lamine d'oro orfiche. Istruzioni per il viaggio oltremondano degli iniziati greci (Milan 2001)Google Scholar; Graf, F. and Johnston, S.I., Ritual Texts for the Afterlife: Orpheus and the Bacchic Gold Tablets (London and New York 2007).Google Scholar On the fountain they mention (which is generally thought to be a fountain of the Lethe), see especially Bernabé, A. and San Cristóbal, A.I.J., Instructions for the Netherworld: The Orphic Gold Tablets (Leiden 2008) 29–35.Google Scholar Bernabé and San Cristóbal mention the epitaph for Atthis in their survey of the history of the Lethe, but only in passing: ‘The “water of forgetfulness” is also mentioned by an inscription of the 1st century B.C., in which a dead spouse consoles her widower (IKnidos 303, 11): “I did not drink the last water of the forgetfulness of Hades”, which means that she still preserves her identity and still remembers her husband' (30). As Sacco's study (n.36) demonstrates, however, the mere fact that the Lethe is mentioned in Atthis' epitaph is not itself remarkable.
87 PEG II.2 474 = Graf-Johnston 1. For bibliography, see PEG II.2; especially Foti, G. and Carratelli, P., ‘Un sepolcro di Hipponion e un nuovo testo orfico’, PP 29 (1974) 91–126.Google Scholar See also Ferrari, F., La fonte del cipresso bianco (Turin 2007) 115–40.Google Scholar
88 Cf. the reconstructed 
in line 12 of the leaf from Petelia (PEG II.2 476, also fourth century BC, from southern Italy); cf. also 
in line 3 of the leaf from Rome (PEG II.2 491, 260 BC).
89 PEG II.2 475 = Graf-Johnston 8.
90 PEG II.2 476 = Graf-Johnston 2.
91 PEG II.2.477 = Graf-Johnston 25.
92 One of these is from a Roman cemetery near Rhethymnon, which shows that the leaves were still used even in the period (first century BC) to which Atthis' epitaph probably dates.
93 The grouping (of eight leaves) was first made by Zuntz, G. (Persephone: Three Essays on Religion and Thought in Magna Graecia (Oxford 1971) 355–93).Google Scholar See Janko, R., ‘Forgetfulness in the golden tablets of memory’, CQ 34 (1984) 89–100CrossRefGoogle Scholar on the inclusion of the leaves from Hipponium (publ. 1974) and Thessaly (publ. 1977) in Group B.
94 For a summary of attempts (and objections raised to the endeavour) see Janko (n.93) 90.
95 The last three lines of Janko's (n.93) reconstruction of Group B's archetype (199) synthesize these verses. On the idea of an archetype for the leaves, see also Ferrari (n.87) 118–21.
96 Bernabé has amassed much evidence for the Orphic mysteries' connections with Osiris and Egypt: PEG II.2 40–63.
97 On other epitaphs in which ‘we catch an echo of Orphic phrase’, see Gardner, P., The Sculptured Tombs of Hellas (New York 1896) 205–06.Google Scholar
98 On the problem of dating the Orphic hymns, see Athanassakis, A.N., The Orphic Hymns (Missoula 1977) vii–viii.Google Scholar
99 See bibliography in Wood (n.78) especially 504.
100 Cole, S.G., ‘Voices from beyond the grave: Dionysus and the dead’, in Carpenter, T.H. and Faraone, C.A. (eds), Masks of Dionysus (Ithaca 1993) 276–95.Google Scholar
101 Some of the evidence is collected on pages 198–99 of Quandt, W., De Baccho ab Alexandri Aetate in Asia Minore Culto (Halle 1912).Google Scholar
102 Following a discussion of eschatological themes in Pindar's threnoi, Rossi (n.54) observes that in the fourth century BC epitaphs began to exhibit ‘questo nuovo modo di vedere l'oltretomba’ and to contemplate the possibility that the soul could be separated from the body and attain a sort of divine status in death. It is not until the beginning of the second century that inscriptions exhibit these sorts of themes with an ‘intento consolatorio nei confronti dei vivi’ (34).
103 On sleep and death in Homer, see Albinus (n.85) 90–97.
104 Sacco (n.36) 49.
105 For the view of death as sleep as diametrically opposed to death as migration of the soul, see notably Plato Apol. 32d, where Socrates says that 
it is either to pass into a state of no longer existing, perceiving or feeling (like ὕπνος) or a μεταβολή […] καὶ μετοίκησις 
106 Il. 16.457=675. See Sourvinou-Inwood (n.81): in Homer, ‘it is not only the burial involving the erection of a tymbos-and-stele that is said to be the due of the dead, but also mourning and weeping for them’ (129–30, n.59); see also Garland, R., ‘
’, BICS 29 (1982) 69–80.Google Scholar
107 Sacco (n.36) 51. See also the ‘anti-echo’ in a first- to second-century AD Chersonnese epitaph (GVI 1684), all 17 lines of which are a husband's apostrophe of his dead wife: 
(11–12).
108 Though perhaps coincidental, it is nevertheless striking that in both Atthis' and Sabinus' epitaph the phrase 
appears in the same context.
109 Gow, A.S.F. and Page, D., The Greek Anthology: Hellenistic Epigrams 2 (Cambridge 1965)Google Scholar; see also pages 116–17 of Hunter, R., ‘Callimachus and Heraclitus’, MD 28 (1992) 113–23.Google Scholar
110 Newton (n.1) vol. 2.2 475.
