Published online by Cambridge University Press: 12 December 2019
Among the one-word fragments from unknown plays of Sophocles, fr. inc. 1111 R. (φίλανδρον) has been treated as one of the more straightforward. It derives from a passage in Hermogenes of Tarsos’ treatise Περὶ Ἰδεῶν (late second century c.e.), which includes the Sophoclean adjective, its referent and a brief gloss: … ὁ Σοφοκλῆς … φίλανδρόν που τὴν Ἀταλάντην εἶπε διὰ τὸ ἀσπάζεσθαι σὺν ἀνδράσιν εἶναι (‘… Sophocles called Atalante philandros somewhere because she enjoyed being with men’). Brunck assigned the fragment to Sophocles’ tragic Meleagros; most subsequent editors have edited the fragment as sedis incertae while commenting favourably on Brunck's ascription. This suggestion has also found support beyond Sophoclean scholarship, and, to my knowledge, no alternative has been brought forward. While the ascription of the fragment to the Meleagros is prima facie not implausible, I shall argue that a thorough analysis of the difficult passage in Hermogenes calls for a revision of the current lexicographical accounts of the word φίλανδρος—as well as φιλανδρία—and suggests that fr. 1111 may in fact originate in a satyr-play.
1 On the dates of Hermogenes and Περὶ Ἰδεῶν, see M. Patillon, Corpus Rhetoricum Tome IV: Prolégomènes au De ideis. Hermogène, Les catégories stylistiques du discours (De ideis). Synopses des exposés sur les Ideai (Paris, 2012), viii–ix.
2 Id. 2.5.5 [341]. Text of Hermogenes and page numbers follow Patillon's edition (n. 1) with the page numbers of Rabe's edition added in brackets. Translations of Hermogenes are adapted from Wooten, C.W., Hermogenes’ On Types of Style (Chapel Hill, NC / London, 1997)Google Scholar, here 82; all other translations are my own.
3 R.F.Ph. Brunck, Lexicon Sophocleum, in id., Sophoclis tragoediae septem cum scholiis veteribus … accedunt deperditorum dramatum fragmenta, vol. 4 (Strassburg, 1789), s.v. φίλανδρος.
4 Only Dindorf, W., Sophoclis tragoediae superstites et deperditarum fragmenta (Oxford, 1832), 370Google Scholar confidently edits it among the fragments of ‘the Meleagros’ (sc. fr. 356); it is also referred to as such in the various Lexica Sophoclea s.v. (Ellendt [Königsberg, 1835], Ellendt/Genthe [Berlin, 1872], Dindorf [Leipzig, 1870]). Subsequent editors list it among the incerta: Nauck in TrGF 1 (1856) has it as inc. fr. 1003, but notes ‘ad Meleagrum rettulerim cum Brunckio’, and this is repeated in TrGF 2 (1889) where the fragment features as inc. fr. 1006. Similarly, A.C. Pearson, The Fragments of Sophocles (Cambridge, 1917), 3.165 edits it as inc. fr. 1111, but notes that ‘Brunck was probably right in ascribing this fr. to the Meleager’. Radt (TrGF), who also edits it as inc. fr. 1111, refrains from explicit judgement, but see below, p. 10. H. Lloyd-Jones, Sophocles Fragments (Cambridge, MA / London, 20032), 213 briefly mentions it in his introduction to the Meleager without engaging with previous scholarship (‘Did Atalanta figure in Sophocles’ play? Fr. 1111, from which we learn that Sophocles called Atalanta φίλανδρος, might suggest it.’). The fragment is mentioned neither in the edition of Paduano, G., Tragedie e frammenti di Sofocle (Turin, 1982)Google Scholar nor in the detailed survey of Sophocles’ fragmentary plays in Jouanna, J., Sophocle (Paris, 2007), Annexe IIGoogle Scholar.
5 E.g. Arrigoni, G., ‘Atalanta e il cinghiale bianco’, Scripta Philologa 1 (1977), 9–47Google Scholar, at 47 and cf. below n. 38.
6 At Id. 2.3.18 (328) Hermogenes claims that δριμύτης and ὀξύτης are interchangeable (and suggests a near synonymity of ὀξύτης and τὸ ὀξέως λέγειν). ‘L'image dans les deux cas est celle d'un esprit perçant et pénétrant’, as Patillon (n. 1), c puts it. For the systematic place of δριμύτης within Hermogenes’ theory of ἰδέαι, see Hagedorn, D., Zur Ideenlehre des Hermogenes (Göttingen, 1964), 70–6CrossRefGoogle Scholar; Patillon, M., La théorie du discours chez Hermogène le Rhéteur. Essai sur les structures linguistiques de la rhétorique ancienne (Paris, 1988), 112–13, 257–9Google Scholar.
7 Cf. Id. 2.5.3 (340): σαφέστερον δὲ ἔσται τοῦτο διὰ τῶν παραδειγμάτων (‘examples will make this clearer’).
8 Wooten (n. 2), 82.
9 Lempp, U., Hermogenes (Stuttgart, 2012), 96Google Scholar: ‘Das Wort ‘Mannesliebe’ bedeutet hier fast das Gegenteil von dem, was wir üblicherweise darunter verstehen; denn heute bedeutet es Lüsternheit und Ehebruch’ [my emphasis].
10 Montanari, F. (ed.), GI – Vocabolario della lingua greca (Turin, 1995)Google Scholar, s.v.: ‘poster. attrazione per gli uomini Hermog. Id. 341.15.’
11 The particle seems to give special emphasis to the repeated νῦν, and thus underlines that the pointed meaning φιλανδρία is at issue. Significantly, both Wooten and Lempp leave δήπου untranslated; contrast the translations quoted in the next note.
12 This understanding is clearly supported by the anonymous scholiast ad loc. (Walz, Rhet. 7.2.889). Accordingly, Ruiz-Montero, C., Hermógenes: Sobre las formas de estilo, Introducción, traducción y notas (Madrid, 1993), 240Google Scholar translates: ‘Aquí esa expresión de philandría indica casi lo contrario del sentido en que nosotros solemos utilizarla, pues en ese pasaje claramente quiere significar la incontinencia y el hecho de ser adúltera’ [my emphasis]. Similarly, Patillon (n. 1), 157 renders it as ‘peut-être en effet l'expression “amour de l'homme” signifie-t-elle ici le contraire de son sens de l'usage habituel: car il veut assurément signifier ici la licence et l'infidélité conjugale’ [my emphasis].
13 φιλανδρία is used in a variety of authors in the Common Era, invariably in the sense of ‘love of one's husband’: e.g. Ph. De Abrah. 245, 253; Joseph. AJ 18.159; App. B Civ. 4.4.23 bis, 4.6.39; Dio Cass. in Zonar. 9.17 (1.281 Boissevain); Clem. Al. Paed. 3.8.41.5, 3.11.57.3; Strom. 4.19.121.2; Ps.-Lucian, Halc. 2.
14 E.g. Attica: IG 2 4249 (first/second cent.), 12644 (second cent.); Sparta: IG V.1 483 (early second cent.?); 581 (c.100–150), 600, 605 (Roman period); 601 (reign of Caracalla); Delphi: CID 4:160 (c.150–200); Macedonia: EAM 54 (c.150–200); 201 (c.200–250); Perinthos-Heracleia 283 (third cent.); Lydia: SEG 33: 1031 (159/160); Bithynia: IK Apameia (Bith.) / Pylai 13 (later Roman Imperial period); Paphlagonia: Marek, Kat. Amastris 45 (203); Phrygia: SEG 37: 1099 (second/early third cent.), 1099 [2] (132–211, c.160?); Pisidia: IK Selge 15 (c.222–235).
15 See, for example, Ap. Ty. Ep. 55.1, 58.7; Plut. Mor. 57D; 243A, 257F; 499C, etc.; Lucian, Dial. Meret. 7.3 (297); Fronto, Ep. 12.2; Ptol. Tetr. 4.5.3; Max.Tyr. Or. 40.3; Iambl. Bab. fr. 35.60; Clem. Al. Strom. 4.20.128.1, Paed. 3.7.39.1. Inscriptions: Sparta: IG V.1 1187 (second cent.); Arcadia: IG V.2 182 (third cent.), 492 (superlative, second/third cent.); Macedonia (Thessaloniki): IG X.2 1 903 (second/third cent.); Crete: SEG 45: 1251 (first/second cent.); Lydia: SEG 29: 1199 (after a.d. 212); TAM V.2 954 (222–235); Bithynia: IK Iznik 1352 (superlative; second/third cent.).
16 The pronoun ἡμεῖς thus comprises not just Hermogenes and his contemporaries, as Wooten, Lempp and others seem to assume, but also speakers of Greek in the fifth century b.c.
17 The two aspects first appear in neat opposition (κυρίως … οὐ κυρίως — οὐ κυρίως … κυρίως) before they collapse in the comparative κυριωτέρα. Arguably, πως qualifies this shift and could be rendered as ‘paradoxically’ vel sim.
18 Thus the rendering in LSJ s.v., paralleled by the entries in GI (‘nemico della caccia’) and Bailly (‘qui hait la chasse’). These lexicon entries only give Xen. Cyn. in support; Poll. Onom. 5.63 and 6.172 (the former, however, clearly derived from Xen.) are not mentioned.
19 To credit the hound who gives up the chase—and hence spares his prey—with φιλανθρωπία in the ‘common’ meaning of ‘pity’ would conform to a general tendency in Cyn. to anthropomorphize animals. Hermogenes discusses this in his treatment of the ἰδέα of γλυκύτης (2.4.17–18 [335]), to which δριμύτης is related or subordinated; cf. Patillon (n. 6), 112; Rutherford, I., Canons of Style in the Antonine Age. Idea-Theory in its Literary Context (Oxford, 1998), 70Google Scholar with n. 19.
20 Thus Delebecque, E., Xénophon: L'art de la chasse (Paris, 1970), 59Google Scholar translates: ‘parce qu'ils n'aiment pas l'animal sauvage’. On other compounds in -θηρος (from θήρα or θήρ respectively), see the list in Hoffmann, O., ‘Ἀλέξανδρος’, Glotta 28 (1939), 21–77Google Scholar, at 25.
21 Hermog. Id. 2.5.3 (340): ἡ δὲ τῆς δριμύτητος λέξις […] μὲν καθ’ ἑαυτὴν οὐδὲν ἔχει τοιοῦτον, ἔννοιαν δέ τινα σημαίνουσα, ἧς οὐκ ἔστι κυρία, ἤ τισιν ἄλλοις ἐφεπομένη κατά τινα οἱονεὶ χαριεντισμὸν γίνεται δριμεῖα καὶ ποιεῖ τὴν δριμύτητα.
22 Hagedorn (n. 6), 72–3.
23 Hermog. Id. 2.5.8 (341): ὅλως τε πολὺς ὁ κίνδυνος ἐν ταῖς τοιαύταις δριμύτησιν ἐκπεσεῖν εἰς ψυχρότητα· ἐπεὶ καὶ οἱ τὰ σπουδαῖα γελοῖα ταυτὶ συντιθέντες τοῦ γελοίου γ’ ἕνεκα οὐδὲν ἄλλ’ ἢ ταῖς τοιαύταις χρώμενοι δριμύτησιν εὐδοκιμοῦσιν· εὐλαβεῖσθαι οὖν χρή (‘In general, however, in using words subtly one must be careful not to lapse into frigidity (psychrotēs). For those who combine the serious and the comic for the sake of a laugh are in fact praised because they use such clever expressions. Thus one must be careful’).
24 Eur. Andr. 177–80: ἃ μὴ παρ’ ἡμᾶς εἴσφερ’· οὐδὲ γὰρ καλὸν | δυοῖν γυναικοῖν ἄνδρ’ ἕν’ ἡνίας ἔχειν, | ἀλλ’ εἰς μίαν βλέποντες εὐναίαν Κύπριν | στέργουσιν, ὅστις μὴ κακῶς οἰκεῖν θέλει (‘Do not bring such customs to us: for it is not right for one man to hold the reins of two women. Whoever does not want to live miserably is content to seek just one love for his marriage-bed’).
25 Cf. Lloyd, M., Euripides Andromache (Warminster, 1994)Google Scholar, on vv. 215–21.
26 Cf. Patillon (n. 1), 158 n. 694: ‘Chez Euripide le mot est employé avec une syllepse: s'agissant d'Hélène […] il désigne, en mauvaise part, l'amour licencieux pour les hommes; s'agissant d'Hermione, il désigne l'amour conjugal.’
27 A passage such as Sophocles’ Trachiniai (550–1) ταῦτ’ οὖν φοβοῦμαι μὴ πόσις μὲν Ἡρακλῆς | ἐμὸς καλῆται, τῆς νεωτέρας δ’ ἀνήρ (‘It is this that I fear: that Herakles may be called my husband but the man of this young girl’), uttered by Deianeira, brings this out in a δριμύς way of its own.
28 According to Hoffmann's historico-linguistic study (n. 20) on names and appellatives in -ανωρ and -ανδρος, -ανωρ (restricted to Homer and choral lyric) had the primary function of referring to a single ἀνήρ or the male gender, whereas -ανδρος was mainly used to refer to a multitude of ἄνδρες (at 38). Only when -ανωρ was no longer in use, thus Hoffmann's claim, would some forms in -ανδρος—among these φίλανδρος—be used to cover both functions (at 44–6). In his reading, the ‘bad sense’ would therefore be the older or primary one. It must be said, however, against Hoffmann's claims to the contrary (at 39), that neither Helen's στίβοι φιλάνορες (‘man-loving traces’) in Aesch. Ag. 411 nor Clytaemestra's claim at Ag. 856 οὐκ αἰσχυνοῦμαι τοὺς φιλάνορας τρόπους | λέξαι … (‘I am not ashamed to speak out about my man-loving ways …’) are particularly convincing examples for φιλάνωρ ‘loving one man, sc. the husband’; rather, both passages are purposefully ambiguous (pace Fraenkel, E., Aeschylus Agamemnon [Oxford, 1950]Google Scholar, on v. 856, cf. on v. 411). Contrast, however, Aeschylus’ use of φιλάνωρ in Pers. 136. As for φίλανδρος in Aeschylus, this is attested once, metaphorically in Aesch. Sept. 902 (lyr.) πέδον. Only few commentators discuss the adjective, and the comment by Tucker, T.G., Septem contra Thebas (Cambridge, 1908)Google Scholar, ad loc. perhaps says more about the time in which it was written than about Aeschylus’ use of the word: ‘Destitute of any prince the land is a widow who has just lost her beloved spouse.’ Hoffmann (n. 20), 46 gives the same explanation. Cf. M.M. Vock, ‘Bedeutung und Verwendung von ἀνήρ und ἄνθρωπος und der stammverwandten Derivativa und Komposita in der älteren griechischen Literatur (bis nach 350 v.Chr.)’ (Diss., Freiburg, CH, 1928), 61, 65, 68. The entries s.v. in the various Lexica Sophoclea mentioned above (nn. 3 and 4) are revealing of the ambiguity of the adjective. Ellendt (1835) gives no gloss on the word, but whereas Dindorf (1870) translates amans viri, Genthe's revised edition of Ellendt (1872) has virorum amans.
29 As with φιλανδρία, the entry in GI corresponds almost exactly to LSJ: φίλανδρος in Soph. fr. 1111 R. and Plut. Thes. 26.2 is rendered as ‘che ama i comportamenti maschili’; Bailly gives alternative translations for the Sophoclean use (sc. ‘qui aime les hommes ou les habitudes viriles’, emphasis in the original).
30 On the myth of Atalante, see e.g. Escher-Bürkli, J., ‘Atalante [4]’, RE 2 (1896), 1889–95Google Scholar, Arrigoni (n. 5), Grossardt, P., Die Erzählung von Meleagros. Zur literarischen Entwicklung der kalydonischen Kultlegende (Leiden, 2001), 270–5CrossRefGoogle Scholar.
31 The Amazons are ἀνδροκτόνοι (as a translation of Scyth. Οἰόρπατα) in Hdt. 4.110.1; Aesch. PV 723–4 speaks of the Ἀμαζόνων στρατὸς στυγάνωρ (‘man-loathing army of Amazons’)—already the Iliad has the tantalizing noun-epithet formula Ἀμαζόνες ἀντιάνειραι (3.189; cf. 6.186). According to Hoffmann (n. 20), 37–8 ἀντιάνειρα is not ‘man-like’—sc. in analogy to ἀντίθεος—but ‘opposed, hostile to, fighting men’ vel sim. (from ἀντίος); more nuanced is the thorough discussion in Blok, J., The Early Amazons. Modern and Ancient Perspectives on a Persistent Myth (Leiden – Boston – Cologne, 1995), 155–93Google Scholar, who traces a shift in meaning of ἀντιάνειρα from ‘equivalent to men’ to ‘hostile to men’, and 175–7, discussing the various ancient interpretations of the epithet; cf. already Detienne, M., Dionysos mis à mort (Paris, 1977)Google Scholar, 86, 123 n. 85 (earlier literature). ἀντιάνειρα is also (if much later) an epithet for Atalante: Nonnus, Dion. 35.82 (where an Amazon is mentioned only a few lines later, 91). On Atalante, see further below.
32 Atalante and the Amazons are frequently paralleled, compared or juxtaposed in modern scholarship (see e.g. Detienne [n. 31], 33; Carlier-Detienne, J., ‘Les Amazones font la guerre et l'amour’, L'Ethnographie 76 [1980–1], 21, 29Google Scholar; Just, R., Women in Athenian Law and Life [London and New York, 1989], 241CrossRefGoogle Scholar; Barringer, J., ‘Atalanta as model: the hunter and the hunted’, CA 15 [1996], 48–76Google Scholar, at 49, 59–66, with further literature); in ancient sources, explicit comparisons of Atalante and the Amazons are attested very late, e.g. Eust. Il. 4.240.20 van der Valk, listing those who might be considered a γυνὴ ἀνδρεία: the Sarmatian women, the Amazons, Rhodogyne and Atalante (cf. Eust. Or. 16 p. 279.9–11 Wirth). According to Barringer's study on visual representations of Atalante in vase painting, however, the Amazons (and Maenads) provided an important model for Atalante—this, as well as the shared epithets φίλανδρος and ἀντιάνειρα (cf. n. 31), suggests that an implicit parallel was drawn significantly earlier.
33 Tzetz. Schol. ad Antehomerica 22 lines 7–14 (Papathomopoulos, M., Nouveaux fragments d'auteurs anciens [Ioannina, 1980], 59Google Scholar): τῶν ἱστορικῶν Ἑλλάνικος μὲν λέγει· | τοῦ Κιμμερικοῦ συμπαγέντος Βοσπόρου | πάλαι περάσας ἦλθε τῶν Ἀμαζόνων | πλεῖστος στρατὸς χρύσασπις ἀργυραξίνης | θῆλυς φίλανδρος ἀρρενοβρεφοκτόνος· | μερίζεται δὲ πρὸς πολλὰ τῆς γῆς τὰ πλάτη | καὶ τοὺς ἀραγμοὺς τῶν μαχῶν συνεκρότει. | ταυτὶ μὲν ῾Ελλάνικος οὗ Λέσβος πάτρα (‘Of the historians, Hellanicus says: long ago when the Kimmerian Bosporos was frozen, an immense army of Amazons set across, bearing golden shields and silver axes, womanfolk, man-lovers, killers of male offspring. The army dispersed over many stretches of the land and made the clashing sounds of battle. This is the account of Hellanicus, whose fatherland is Lesbos’).
34 Οn this discourse in general, see Carlier-Detienne (n. 32), 11–33, who does not, however, deal with this or other such potentially early sources.
35 Cf. e.g. Strabo 11.5.1.
36 See further below, pp. 11–12.
37 In Lucian, Symp. 25. For an attempt at reconstructing the Meleagros, see Grossardt (n. 30), 83–7.
38 In addition to the editors of Soph. fr. 1111 R. listed above (nn. 3–4), see e.g. Grossardt (n. 30), 85 and n. 44 (cf. 243 n. 64, 270 n. 5, 279), who refers to fr. 1111 as belonging to Meleager and, on this basis, assumes that Atalante was among the personae of the play, even though he elsewhere admits that the participation of Atalante in the Calydonian boar hunt is securely attested only for Euripides and later texts (at 270).
39 The studies of Ley, A., ‘Atalante – von der Athletin zur Liebhaberin. Ein Beitrag zum Rezeptionswandel eines mythologischen Themas auf Vasen des 6.–4. Jhs. v.Chr.’, Nikephoros 3 (1990), 31–72Google Scholar and of Bergamasco, A., ‘Atalanta e il ruolo della donna nell'immaginario figurativo Greco’, in Colpo, I., Favaretto, I., Ghedini, F. (edd.), Iconografia 2005. Immagini e immaginari dall'antichità classica al mondo moderno (Padua, 2006), 397–402Google Scholar on Atalante's iconography trace a development from an initial emphasis of her manlike qualities (in the hunt or as an athlete) to an erotically attractive (and interested) female who still pursues male activities but is in some sense tamed. Both studies date the turning point in this development to the mid fifth century b.c. Cf. Most, G., ‘Of motifemes and megatexts: comment on Rubin/Sale and Segal’, Arethusa 16 (1983), 199–218Google Scholar; the role of Atalante in the myth of Meleager can be gleaned from the useful appendix in Grossardt (n. 30), 285–90.
40 Phillips, A.A. and Willcock, M.M., Xenophon & Arrian: On Hunting with Hounds (Warminster, 1999)Google Scholar, ad loc.
41 Cephalus is introduced as he who ‘was snatched by the goddess’ (ὑπὸ θεᾶς ἡρπάσθη, Cyn. 1.6).
42 E.g. Thgn. 1287–94; Eur. Phoen. 151–2.
43 An early and remarkably detailed version of the courtship competition is preserved in the fragmentary Hesiodic Atalante-Ehoie: see Hes. frr. 72–76 M./W. ≅ frr. 47–51 Most ≅ *frr. 2–*4 Hirschberger (cf. Hirschberger, M., Gynaikōn Katalogos und Megalai Ēhoiai. Ein Kommentar zu den Fragmenten zweier hesiodeischer Epen [Munich and Leipzig, 2004]Google Scholar, ad locc.).
44 In Ar. Lys. 785–96 the problem of Atalante's misogamy is compounded by the fact that her spouse, Milanion, had equally resisted the female sex before he met and eventually married her. Already the scholiast, however, describes this as a self-serving distortion of the traditional story by the male chorus; schol. Ar. Lys. 785a, b (Hangard): ἐπίτηδες δὲ τοῦτο ὁ τῶν ἀνδρῶν χορὸς παριστορεῖ.
45 Cf. Aesch. Diktyoulkoi fr. 47a.824–32 with Laemmle, R., Poetik des Satyrspiels (Heidelberg, 2013), 299–305, 399–400Google Scholar; and see 399–402 Studien II s.v. ‘Inversion der Begierde’ for further examples.
46 Cf. already Hunt, A.S., ‘1083. Satyric drama’, The Oxyrhynchus Papyri, Part VIII (1911), 60–71Google Scholar (ed. princ.); Laemmle, R., ‘Zur Autor- und Stückzuweisung von P.Oxy. 1083 fr. 1 (S. **F 1130 R.)’, ZPE 208 (2018), 44–66Google Scholar offers doxography and new arguments for Sophoclean authorship.
47 Cf. Maas, P., rev. ‘The Oxyrhynchus Papyri Part VIII’, PhW 39 (1911), 1214–19Google Scholar, at 1217 and id., ‘Zu dem Satyrspiel Oxyrh.-Pap. VIII 1083’, in id., Kleine Schriften, ed. W. Buchwald (Munich, 1973 = repr. of BPW [1912], 1426–9), 50–3, at 53 (tentatively attributing the fragment to Aristias’ Atalante).
48 My warmest thanks to Patrick James and Simon Westripp, who generously let me read the drafts of the Cambridge Greek Lexicon s.v. φιλανδρία and φίλανδρος, and to Richard Hunter and Cédric Scheidegger Laemmle, who generously read drafts of this article.