Published online by Cambridge University Press: 08 May 2015
In recent years the place (or plight) of women in ancient society has attracted much attention. Their legal position, the attitudes towards them on the part of men, the characterization of women in drama, the personality of known and spectacular individuals like Sappho and many other topics have been studied at considerable length from various points of view. It is on the surface, then, somewhat surprising that scarcely any attention has been paid to a topic which in another discipline, linguistics, has proved very popular of late, the question of women’s speech. Sixty years ago the idea that women exhibit a distinctive form of speech was already current; Jespersen devoted a chapter of his Language: its Nature, Development and Origin to ‘The Woman’. Since then with the growth of interest on the part of linguists in sociolinguistic questions an impressive body of literature relating to this topic has accumulated.
1 Note for example the following works (these are selected merely from the English-speaking world): Pomeroy, S.B.Goddesses, Whores, Wives and Slaves: Women in Classical Antiquity (N.Y. 1975),Google Scholar the issue of the periodical Arethusa entitled ‘Women in the Ancient World’ (11 ), Gould, J.P., ‘Law, custom and myth: aspects of the social position of women in classical Athens’, JHS 100 (1980), 38–59CrossRefGoogle Scholar and the source book compiled by Lefkowitz, M.R. and Fant, M.B.Women’s Life in Greek and Rome (London 1982).Google Scholar For women in Menander’s time see Fantham, E.Phoenix 29 (1975), 44 ff.CrossRefGoogle Scholar
2 One sociolinguistic sex-related topic has been well and profitably explored. This is the question of the naming conventions used by Athenian males in the fifth and fourth centuries B.C. when speaking of women. See Schaps, D. ‘The Woman Least Mentioned: Etiquette and Women’s Names’, CQ n.s. 27 (1977), 323–30CrossRefGoogle Scholar (dealing with the orators) and Sommerstein, A.H. ‘The Naming of Women in Greek and Roman Comedy’, Quaderni di Storia 2 (1980), 393–418Google Scholar (this contains interesting observations on Terence’s failure to observe the conventions of his models).
3 Jespersen, O.Language, its Nature, Development and Origins (London 1922), 237–54.Google Scholar For recent literature see the items mentioned by Adams below, p.43, nn. 1,3,5.
4 For the limitations imposed upon this kind of study see Adams, p. 43.
5 See Dover, K.J. ‘The Colloquial Stratum in Classical Attic Prose’, Classical Contributions (Studies in Honour of M.R. McGregor), (Locust Valley, N.Y. 1981), 15–25, especially 17.Google Scholar
6 Op. cit. (note 1) 38.
7 Women prose writers are not exactly thick on the ground. The erotic writer ‘Philainis’ may well be a man writing under & nom de guerre. For references see Lefkowitz-Fant, 160, n. 4.
8 See below, p. 39.
9 They are to be found in Austin’s, C.Comicorum Graecorum Fragmenta in Papyris Reperto (henceforth CGFPR) (Berlin and New York 1973).Google Scholar
10 See the notable essay by Sandbach, F.H. ‘Menander’s Manipulation of Language for Dramatic Purposes’ in Ménandre, Entretiens sur l’Antiquité Classique 14, Fondation Hardt (Vandoeuvres-Genève 1969), 113–36.Google Scholar (items from this colloquium will henceforth be cited as from Ménandre). Compare also Del Corno, D. ‘Alcuni aspetti del linguaggio di Menandro’, SCO 24 (1975), 13–48.Google Scholar
11 See Sandbach in Menander, A Commentary by Gomme, A.W. and Sandbach, F.H. (Oxford 1973) (henceforth Sandbach on/ad loc.), 4 ff.Google Scholar
13 See below, p. 30.
14 See Dover, K.J. ‘Linguaggio e caratteri aristofanei’, RCCM 18 (1976), 357–71, especially 368 ff. Google Scholar
15 See Dover, op. cit. for a telling example.
16 Cf. Macleod’s, C.W. edition of Homer, Iliad 24 (Cambridge 1982),Google Scholar index s.v. ‘colloquial expressions’; Stevens, P.T.Colloquial Expressions in Euripides, Hermes Einzelschriften x 38 (1976)Google Scholar and Moorhouse, A.C.The Syntax of Sophocles, Mnemosyne Supplement 75 (1982),Google Scholar index s.v. ‘Colloquialism’.
17 The notion was first propounded (apropos of Il. 14.331–36) by Gildersleeve, B.L.AJPh 28 (1907), 209.Google Scholar See also W.B. Stanford’s notes on Od. 4.681,6.262,19.351,23.1746.
18 On Gildersleeve’s definition, Hephaistos’ utterance at Il. 1. 580 ff. would qualify as feminine syntax.
19 Wackernagel, J.Kleine Schriften 2 (Göttingen 1953), 992.Google Scholar One word of address is found used only by women (4 times) to men. This is κάμμορε. One hesitates, however, before classifying this as female speech since it is always the same man who is so addressed, Odysseus, and Telemakhos at Od. 5. 351 refers to Eurykleia as expecting εκείνον τον κάμμορον. At Ap. Rhod. 4. 1318 κάμμορε is used by a man.
20 Notable is Lysias 32. 11–18.
21 Herodas uses μã as a marker of female speech (1. 85; ?4. 20; 5. 13, 56, 59; 6. 4, 21): cf. Theokr. 15. 89, Aiskhylos, Hik. 870, 960. See Meister, R.Abh. kôn.sàchs. Ges.der Wiss. 13 (1893), 683f.Google Scholar and V. Schmidt (cited below in note 40), 12 ff.
23 For remarks on some of these (and other poetesses) see West, M.L. ‘Die griechischen Dichterinnen der Kaiserzeit’, Kyklos (Griechisches und Byzantinisches: Rudolf Keydell zum neunzigsten Geburtstag) (Berlin and New York 1978), 101 ff.Google Scholar
24 I am thinking of the quite remarkable (and very late) letter in P. Grenfell 1. 53.
25 Much of this material is collected by Gilleland, M.E. ‘Female Speech in Greek and Latin’, AJPh 101 (1980), 180–3Google Scholar (his Anecdota Bekkeri 2. 855 is actually a scholion to Dionysios Thrax = GG. 1.3.5 39.24 Hilgard: it gives a ‘fragment’ of Alkman for which see Bergk, PLG 5 3.78 — the lemma is lost). This article and that of Adams add some further relevant items. There are no doubt more.
26 Plat. Phaidon 60 a 4 is a generalization about behaviour, not language.
27 27 Cic. de or. 3. 45 (Fordyce in his Catullus-commentary is wrong to use this passage to illustrate Cat. 84. 5 — see Nisbet, R.G.M.PCPAS 24 , 110;Google Scholar that has nothing to do with female speech, but is an attack, traditional in invective, on the man’s ancestry on his mother’s side — see Nisbet’s commentary on Cic. in Pis. p. 194).
28 See Jespersen, op. cit. (note 3 above), 242 f.; Trudgill, P.Sociolinguistics: an Introduction (Harmondsworth 1974), 90Google Scholar and Adams, p. 44.
29 Cf.Dover, K.J.Greek Popular Morality (Oxford 1974), 81.Google Scholar On the term θείος άνήρ see M.L. West on Hesiod, Op. 731 and the monograph of Bieler referred to there.
30 (Or the male belief in this tendency.)
31 Add Phot, α 1756 Theodorides to Carlini’s testimonia here.
32 See the entry Effeminatus (by Herter, H.) RAC 636Google Scholar for passages where the speech of effeminate men or eunuchs is mentioned. Note [D.H.] composit. verb. 128 ΰπό γυναικών ή κατεαγότων άνθρώπων λέγοιτ’ ν … The young prologue speaker of Menander’s Hydria is described by Quint. Inst.Or. 11. 3. 91 as speaking uoce effeminata when imitating a woman (for this habit in comedy see below pp. 33,36). The interpretation of this passage offered by Gaiser, K.ZPE 47 (1982), 31,Google Scholar defending the arguments set out in his Menander Hydria against Hunter, R.L.CR 29 (1979), 210Google Scholar seems to me strained.
33 For these terms see Adams, p. 44.
34 Opinions about which lines are usable in this play are likely to show considerable variation: the figure in the left-hand column is quite arbitrary. If one adds the newly discovered parts of the opening scene (P. Oxy. 3 368–71, first published by Turner, E.G. in PBA 63 , 315 ff.Google Scholar and available separately) the numbers arrived at (by me) are 190 and 30.
35 YCS 22 (1972), 142.
36 This may also hold good for the type of female characters that have survived. Perhaps the hetairai (in the case of Samia, retired hetairai) are slightly over-represented at the expense of respectable matronae such as Sostrata in Terence’s Adelphoe (= x in Men. Ad. b.). In fact one character dominates our view of Menandrean female characters. This is Habrotonon; ‘la figura di donna che possiamo con più agio esaminare in tutti suoi caratteri psicologica è quella di Abrotono’: Zini, op. cit (above, p. 26) 57.
37 The same holds good for Latin comedy. In one play we have a 42% female contribution; in others, like Pseudolus, nothing.
38 Cisteilaria - Synaristosai, 42%: Eun. 32%.
39 This, although a connected utterance, is actually interrupted by a line and a half from Onesimos which Habrotonon, self-absorbed, does not hear: see Bain, D.Actors and Audience (Oxford 1977), 138 ff.Google Scholar
40 This account supplements and updates the excellent discussion by Dedoussi, C.Hellenika 18(1964), 1 ff.Google Scholar(See also Dodone 6,2ff., where she suggests restoring ώ τάλαν in Donatus’ note on Ter. Eun. 899: ‘au interiectio est perturbatae mulieris, ut apud Graecos †tautay’: Boeckh had conjectured ai; ai τάλαν ?). Important additional comments were made by Schmidt, V.Sprachliche Untersuchungen zu Herondas (Berlin 1968), 36 ff.Google Scholar
41 See Sandbach on Men. Dysk. 500 and Austin, C.Menandri Aspis et Samia II Subsidia interpretationis (Berlin 1970), 73.Google Scholar
42 For this use of τâλαç see V. Schmidt, op. cit. note 42, and Wendel, T.Die Gespráchsanrede im griechischen Epos und Drama der Blütezeit (Stuttgart 1929), 22.Google Scholar
43 Scholl. Plat. Theait. 178e (cf. Souda ω 84 Adler), Ap. 25c. Liddell and Scott erroneously state that this notice says that μέλε was originally used by women only (so also Ussher on Ar. Ekkl. 120).
44 The restoration of the part of Sostratos’ mother in this scene suggested by Ritchie is certain: the actual allocation of parts is not; it is just possible that here we have Getas using a wheedling tone. For discussion see Sandbach ad loc.
45 Not included in the Oxford text, but included by Arnott, W.G. in the first volume of his Loeb edition (London 1979), p. 366.Google Scholar
47 Browne, G.M. plausibly suggests ώ τάλαν as a supplement here (BICS 21 , 50).Google Scholar Cf. as well as Dysk. 591, Makhon 400.
48 For τάλαν in poetry see Wendel, op. cit. (above, note 42).
49 Cf. Schmidt (above, note 40), 37.
51 In these two examples it could be argued that the genitive depended on όίμοι.
52 See Handley, Jacques and Sandbach ad loc. and also Marcovich, M.ICS 2 (1977), 206 f.Google Scholar who argues forcefully that Knemon should not complain about his ερημιά (his suggested restoration is unconvincing).
53 See Sandbach in Gomme-Sandbach, 722.
54 Schiedsgericht, 77.
55 They are collected by Sandbach on Sam. 255.
56 Webster, Studies in Menander, 102 claims that Menander makes Habrotonon use ■γλυκύτατε as a kind of nursery endearment; cf. also Williams, T.Hermes 91 (1963), 311Google Scholar who is wrong to adduce Aiüos Aristeides 2.475 ff. Dindorf (= 36.96 Keil) for there ■γλυκείαν means ‘freshwater’.
57 Cf. Ar. Lys. 872 where γλυκύτατον is said man to wife and note the gravestone from Ptolemaic Egypt: Αωρητηϊ γλυκÌ γλοκυτάτηι Εύφημος άνήρ … (see Crisaiolo, L.Scritti in onore di Orsolina Montevecchi [Bologna 1981], 113).Google Scholar
58 Note how Silenos speaks to the baby Perseus at Aiskhylos,Diktyoulkoi 812 f. (LI-J) ό πάπα[ς S]ï παρέζει τψ μίκκψ ra γ«λ[οΐ]α. Since we have reached the nursery, this seems an appropriate place to mention that Greek mothers tended to call their sons ανδριάντες (Phot, α 1768 Theodorides).
59 At Theokr. 15.16 πάππα has been conjectured as a supposed form of address by wife to husband (cf. ‘daddy’ and ‘momma’ used by North American Blacks, but in this context ad hoc ‘baby talk’ is more appropriate than a regularly used vocative, as Daniel points out): see Gow and Dover ad loc. The MSS. and the Antinoe papyrus’s πάντα has, however, been successfully defended by Daniel, R.W.ZPE 27 (1977), 79 ff.Google Scholar
60 See Shipp, G.P.Modern Greek Evidence for the Ancient Greek Vocabulary (Sydney 1979), 530.Google Scholar
61 See Sandbach ad loc. and in Ménandre (above, note 10), 131. Ιχω σε is practically a formula of tragedy.
62 Note also Helen’s address of Telemakhos at Od. 15. 125 and what Eurynome says to Penelope at Od. 18.170 (τécoç). Williams, ibid, (above, note 56) notes that Phaidra’s nurse in Eur. Hipp. 9 times addresses Phaidra as (ώ) τίκνον.
63 Herodas is of course writing a kind of Ionic. He uses the non-Attic plural at 5. 71
65 Cf. Ziebarth, RE s.v. Eid (v. 2. 2076 f.). At 254 the effect of putting on Agathon’s κροκωτός makes him swear by Aphrodite. At 268 before the intrigue is under way he swears by Apollo.
66 What follows is not intended as a general account of oaths in Greek or even of oaths in Menander. On the latter see Wright, F.W.Studies in Menander (diss. Princeton, 1910) (Baltimore 1911 ), 1–55Google Scholar which, although obviously out of date, is still of value. Webster, T.B.L.Studies in Menander, 99 ff.Google Scholar discusses oaths from the point of view of their suitability for individual characters and classes of person. See also Feneron, J.BICS 21 (1974), 88 ff.Google Scholar It is a pity that the relevant RAC article will appear under ‘Schwur’ rather than ‘Eid’.
67 The allegation in Hesykhios (μ 413 Latte) that men also use this oath presumably rests on a misunderstanding of a scene in comedy where a man mimicked a woman. Other such notices are quite explicit in stating that men do not swear by the two goddesses: e.g. Photios (s.v. μα τώ θεώ): άνδράσι δ'ε ου πρέπει τούτον όμνύναι and Phrynikhos, 166 Fischer: νη τώ θεώ• όρκος γυναικών ου μεν άνηρ όμεϊται, ει μη γυναικίζοιτο. Cf. also the scholion to Ar. Ekkl. 155.
68 See Thierfelder, A.Studi Urbinati n.s. B35 (1961), 113 ff.Google Scholar Sandbach misses this in his note on Dysk. 874. The oath is also found in Makhon 297 used by a woman of the same profession as the speaker of the first line of Synaristosai.
69 This oath was, for reasons quite unknown (cf. Wilamowitz, Der Glaube der Rellenen 2. 141),Google Scholar used by men in Socratic literature. Socrates himself employs it seven times in Plato and we find him using it in Xenophon as well Mem. 1.55). Why it should be given to Lakhes (Plat. Lakh. 181 a 4) is extremely puzzling.
70 Sostratos in Dis ex. 20 ff. by saying άρνήσεται … εις μέσον τε πάντες οι θεοί / ηξουσι implies that his Samian girl-friend will make a denial μα πάντας θεούς (unless we are to imagine a string of oaths of the kind put forth by Parmenon at Samia 309 f.). For this oath see the Ashmolean Papyrus edited by Mahaffy, J.P. (TRIA 6 , 202).Google Scholar A recently published fragment of New Comedy has a further oath by all the gods (P.Oxy. 3432. 9).
71 See Sandbach ad loc.
72 In S’indi’ classici in onore di Q. Cataudella 2 (Catania 1972), 306 and An Introduction to Menander, 130.
73 For which see Bader, B.RM 113(1970), 304 ff.Google Scholar Reconstruction of the lost opening of Bacchides and Dis ex. must now take account of the new reading in line three of fr. 8 Lindsay extracted by M. de Nonno from the Priscian ms., Vat. Lat. 3313 (RIFC 105  391 ff.).
74 PIFAO 337, first published by Boyaval, B.ZPE 6 (1970), 5Google Scholar and now fr. 1 in Sandbach’s text.
75 See Handley and Sandbach on Dysk 202.
76 See Sandbach on Epitr. 484 and in Mënandre, 131.
77 τάλ«ινα appears two lines later. The vocative of φίλος, φίλτατος may be favoured by women in invocations of the gods.
78 Also Ήρακλείδαι καί θεοί fr. 730 which some have suspected.
79 This study has not concentrated on expressions avoided by women although it was made clear enough that several oaths were exclusively male. The characteristically Attic form of address ώ râγ is not attested for women in Menander (three certain occurrences in the mouths of men, the latest Mis. A82) although it has been suggested as a supplement in the mouth of the slave girl Doris at Men. Pk. 1003. Lysistrate twice uses ώ râγ (Ar. Lys. 501, 1162).
80 Feneron, op. cit. (note 66) 90 remarks on the lack of variety in women’s oaths.
81 See Adams, J.N.The Latin Sexual Vocabulary (London 1982), 218.Google Scholar Women are known to have used obscene language for ritual purposes and when suffering from delirium (for the latter cf. Hippokrates, Epid. 3. 17. 11).