Adelmo compares images of the whore of Babylon and the Virgin Mary: ‘I was not so much struck by her form as by the thought that she, too, was a woman like the other, and yet this one was the vessel of every vice, whereas the other was the receptacle of every virtue. But the forms were womanly in both cases.’
(Umberto Eco, 1994: 241)
If we take the personifications seriously, we must take their predominant femaleness seriously too (Padel 1992: 160: cf. 158–9)
Personifications of abstract qualities abound in Greek art and literature, and in a number of cases they even acquired sufficient divine stature to merit worship. While the broad chronological span and the variety of contexts in which such figures appear makes any overview liable to oversimplification, one striking generalisation is inescapable: the great majority of personifications are female.1 The values represented can be thought of in various conceptual categories – social goods, ethical qualities, physical conditions – and may be undesirable, like Madness (Lyssa) or Strife (Eris), or ambivalent, like Nemesis (Retribution) or Peitho (Persuasion). Those that appear most frequently, however, are positive ‘good things’: terms for prosperity/happiness (Eutychia, Eudaimonia), Peace (Eirene), Victory (Nike) (Plate 9), Democracy (Demokratia), Justice (Dike), Good Order (Themis, Eunomia), Glory (Eukleia), Health (Hygieia). Given the subordinate status of real women in ancient Greek society, it seems ironic that the qualities deemed desirable by Greek men should be represented in female form.2 Andreia (Manliness) does not appear until the 30s BC, depicted on the monument from Aphrodisias which commemorates Zoilos, but that even ‘Manliness’ should be female rather underlines the paradox.3 The standard answer usually advanced to explain the paradox is linguistic: in Indo-European languages the abstract nouns that tend to be personified are usually feminine in gender. But this is only a start. Why should such nouns be feminine in the first place? And what about those personified nouns which happen to be masculine or neuter in grammatical gender? Further investigation is needed into the sociological and iconographic context in which these personified abstracts developed.
Plate 9 The Nike of Samothrace; marble original from the Sanctuary of the Great Gods at Samothrace, c.200 BC (?) (Louvre, Paris MA2369. Courtesy of the Louvre).
Even English, despite its lack of an explicit grammatical gender system, has a certain under-cover engenderedness. The only inanimate objects regularly given gender, usually by their male owners, are boats, cars, trains, and even church bells, but male-biased sexism can be found at many levels.4 An example with obvious application to the case of personified abstractions is the fact that only girls are given as names nouns which designate virtues: the characters Mercy and Charity in Dickens’ Martin Chuzzlewit may sound a little old-fashioned, but Prudence and Constance were still popular in my grandmother’s generation, and in my own Verity,
Felicity and Grace are not uncommon. Modern Greek girls’ names likewise include virtues, like Eirene, Elpida, Euphrosyne, Eugenia, Eleutheria, Nike and even Themis (Peace, Hope, Joy, Nobility, Freedom, Victory, Order).5 In ancient Greece real women tend not to be given abstract values as such, but compounds, like Eurykleia; where we do find a wide range of abstracts is in the context of prostitutes’ adopted names. In some cases these seem appropriate to their calling, e.g. Persuasion or Joy (Peitho, Euphrosyne), but others paradoxically signify virtues worthy of any good citizen wife or daughter: Fair Fame, Calm, Peace, Silence (Eukleia, Galene, Eirene, Sige).6
The gender of English personifications is neatly explained in an eighteenth-century dialogue by Joseph Addison: ‘It is a great compliment methinks to the sex, says Cynthio, that your Virtues are generally shown in petticoats. I can give no other reason for it, says Philander, but because they chanced to be of feminine gender in the learned languages’ (Dialogue upon the Usefulness of Ancient Medals, London 1726: 36). When faced with the question ‘Why are personifications predominantly female?’ most commentators have been happy to point to the congruence between the female sex of these figures and the feminine gender of the abstract nouns they embody: the phenomenon is an accident of grammatical gender. Why such abstracts should be feminine in the first place, though, has generally been left to students of linguistics, as part of the inconclusive debate on the origins of grammatical gender.7
The terms ‘masculine’, ‘feminine’ and ‘neuter’ were first imposed by ancient grammarians on a language already gendered according to what they saw as the majority of biologically male or female creatures in each morphological category (cf. Foxhall, Chapter 5 in this volume). Aristotle cites Protagoras as his authority for distinguishing genders of nouns as arrena (masculine), thelea (feminine) and skeue (inanimate object). Aristotle uses the same terms, but sometimes substitutes metaxu (in between) for skeue, and notes that many inanimate objects are in fact designated by masculine or feminine nouns (Arist., Rhetoric, 1407 b 7). In the first century BC Dionysos of Thrace’s Art of Grammar standardised the term oudeteron (neither) for ‘neuter’, and added two further genders, the common, e.g. horse or dog (may be masculine or feminine), and the epicene, e.g. swallow (he chelidon) or eagle (ho aetos, always of same gender, whatever the sex referred to).8 The problem of mismatches between grammatical gender and sex comes under discussion as early as Aristophanes. In the Clouds Socrates is parodied as debating this along with other obscure topics in his ‘thinking-shop’. When Strepsiades begs to be taught the ‘unjust logic’, Socrates replies that first he must learn some basics, and proceeds to set him linguistic traps to fall into. First the fact that alektryon can mean either ‘cock’ or ‘hen’ is regretted, and a new word alektryaina is coined to distinguish the female. Much play is made with the gender of he kardopos, ‘kneading-trough’, one of those awkward feminine nouns of the second declension which looks as though it ought to be masculine. Then a problem arises with some men’s names which, being first-declension nouns, look distinctly feminine in the vocative, e.g. Amynias, Amynia (Ar., Nub. 657–93).
Despite such difficulties, the theory of ‘natural gender’, i.e. that grammatical gender reflects biological sex, was espoused in the nineteenth century by such scholars as Grimm; the great many inconsistencies and anomalies observable could be explained as confusions which have crept into this ‘logical order’ over time.9 However, the obvious problems with equating sex and gender, not least that no two languages have an identical gender division, have led many to question the ‘natural gender’ theory. Most late nineteenth- and early twentieth-century analysts tended to look for the origins of grammatical gender in the laws of congruence internal to language, rather than relating the question to the social context of language. Meillet, for example, asserts that the feminine gender is a grammatical category which, in most modern Indo-European languages, holds an important place in morphology, but on the whole is of little import: ‘Déjà en latin, le genre féminin n’avait plus de signification.’10
Adherents of such an approach would hold that grammatical gender has little connection with the way we see the world, but there is plenty of evidence to suggest otherwise.11 A sensible treatment of the question can be found in Marina Yaguello’s Les Mots et les Femmes, discussing sex and gender in the French language (‘Genre et sexe: la métaphore sexuelle’, 1978: 91–113). She summarises the problem succinctly:
The question with which we are faced, as with the more general problem of language-thought relationships, is this: Do we perceive death, the sea, the moon. etc., as feminine because the chance of a blind nominal classification has endowed them with the feminine gender? or. on the contrary, are they feminine because there are symbolic values attached to them which could be tied to mental and social structures and to cultural values? A problem of the chicken and the egg, you could say (Yaguello 1978: 98)
In the end, she concludes, the question of the origin of gender is a false problem: whichever way the arrow goes, gender > symbolism or symbolism > gender, the system as we have it certainly does convey an ideology linked to the social status of men and women.
It is interesting that extremes of both good and evil should be represented in female form, a point made by the Eco quotation with which I began. At the opposite end of the spectrum from our personified ‘good things’ we might think about the many female monsters of Greek mythology: Scylla and Charybdis, the Sirens and Harpies, Medusa, the Furies, the Chimaira, the Sphinx. Ruth Padel puts our female figures into this context: ‘Female personifications in classical Greek are a living part of a precise imaginative landscape. This landscape concentrated daemonic danger in female forms, such as Sirens and tragedy’s talismanic daemons, the Erinyes’ (Padel 1992: 161). Similarly, extremes of human behaviour often seem to find mythical expression in female form; those ‘monsters’ Klytamnestra and Medea spring to mind. At the ‘good’ extreme, apart from personified abstractions there are plenty of heroines representing the ultimate in virtue of one sort or another (‘personifying’ in a loose, non-technical sense): Penelope the chaste, Alkestis the selfless wife. In addition to monsters, the landscape, both real and imaginary, is peopled by more or less beneficent nymphs, all ultimately descended from Earth herself, and of course Olympos has just as many goddesses as gods. In a Judaeo-Christian culture we have become more used to deity being male but in Greek religion Olympian goddesses do not seem, as a rule, to be especially discriminated against. Given the inferiority of mortal women in Greek thought this might seem surprising, but denigration of female deities is rare. Aristophanes’ Peisthetairos and Euelpides touch on the subject when discussing a suitable patron deity for Cloud Cuckoo Land in Birds:
What’s wrong with Athena?
Oh, no. You can’t expect a well run city if you’ve got a female goddess standing
up there in full armour, while Kleisthenes gets on with his knitting (Ar., Av. 828–31, trans. Barrett 1978)
In the end they decide on the Persian cock, an unequivocally male bird.12
Although the majority of personifications are female, there are a number of male figures in cult, art and literature, and we might ask whether there is any significance in their masculinity. The male personifications most often depicted in Greek art are the youthful Eros (Plates 10 and 11) and his companions, Desire and Yearning (Himeros, Pothos), though even Eros seems to have had remarkably little place in cult.13 Arguably these occupy much the same position as our female figures, being beautiful young men of a suitable age, potential objects of men’s desire (Plate 10).14 In the Hellenistic period Eros becomes younger, increasingly represented as the mischievous child of Aphrodite, inspiring desire in others but not an object of desire himself (Plate 11).15 The only other personification regularly represented as a child is also male: Wealth (Ploutos) (Plate 12), who also appears in the arms of Kephisodotos’ ‘Peace holding the child Wealth’, erected in the Athenian agora between 375 and 360 BC.16 Wealth held a cornucopia, its connotations of agrarian plenty appropriate to a figure whom mythological tradition makes son of Demeter, and references to Ploutos as a deity are almost exclusively connected with Demeter and the Eleusinian mysteries.17 The eponymous Wealth of Aristophanes’ play of 388 BC seems unconvincing as a deity, despite the attempts of Chremylos to flatter him as ‘greatest of all daimones’; unlike the Ploutos of myth he is portrayed as an old man, his blindness stemming from Hipponax’s description.18 Wealth appears again as a child on a number of fourth-century vases, where he is always in the company of Demeter/Persephone, reinforcing his association with the Mysteries and the underworld.19
Plate 10 The ‘Palatine Eros’; Roman copy of an original in the style of Praxiteles, c.350 BC (Louvre, Paris MA2266. Courtesy of the Louvre).
Plate 11 Eros asleep; bronze original c.250–150 BC, possibly from Rhodes (The Metropolitan Museum of Art, Rogers Fund, 1943. 43.11.4. Courtesy of the Metropolitan Museum of Art)
Plate 12 Ploutos approaching a choregic monument; red-figure chous, c.400 BC (Antikensammlung, Staatliche Museen zu Berlin, Preussischer Kulturbesitz, F2661. Courtesy of the Berlin Museum).
Eros and Ploutos, then, though personifications of masculine nouns, are generally represented as youths or children, categories of ambivalent sexuality. The only other male personifications to appear with any frequency in the visual arts are Sleep and Death (Hypnos and Thanatos). On the Chest of Kypselos, as described by Pausanias, they were shown as children in the arms of their mother Night (Nyx), but thereafter they are almost invariably depicted in their Homeric capacity as pall-bearers to Sarpedon (Kypselos: Paus., 5.18.1; cf. Hes., Theog. 212. Sarpedon: Homer, Iliad 16.453–4, 671–2). On Euphronios’ much-discussed calyx-krater (c.515 BC) they are both mature, bearded men, with wings (Plate 13); vase-painters in the second half of the fifth century usually differentiate between the two by making Death older and bearded, often unkempt, Sleep a beardless youth. Such a differentiation surely reflects the idea that Death is a figure to be reckoned with, in contrast to his more obviously desirable younger brother.20 Evidence for their cult is sparse: according to Pausanias Sleep had an altar at Troizen; Plutarch mentions a shrine to Death at Sparta alongside those of two other male personifications, Fear (Phobos) and Laughter (Gelos).21 The kind of laughter respected in Sparta is clarified by Plutarch’s (Lykourgos 25) account of Lykourgos’ dedication of a statue of Gelos, indicating that humour could add light relief to a severe lifestyle, as well as having educational value in making reproof more palatable. Fear’s temple at Sparta is described by Pausanias (3.14.9) as outside the city, a place where adolescent boys had to sacrifice puppies to Enyalios the night before engaging in a vicious-sounding unarmed combat; Plutarch (Kleomenes 9) comments that the Spartans honour Fear as a positive force, ‘because in their opinion the state is held together above all by Fear’. Outside the Spartan context sacrifices to Fear on the eve of battle sound more apotropaic, meant to avert a powerfully disruptive force.22
Plate 13 Hypnos and Thanatos with the body of Sarpedon; red-figure calyx-krater by Euxitheos and Euphronios, c.515 BC (The Metropolitan Museum of Art, bequest of H. Durkee, gift of Darius Ogden Mills, and gift of C. Ruxton Love, by exchange, 1972, 1972.11.10. Courtesy of the Metropolitan Museum of Art).
Two male personifications of definite vices are Sloth (Oknos) and Envy (Phthonos), though neither has an extensive iconography. The former appeared in Polygnotos’ painting of the underworld in the Lesche of the Knidians at Delphi; Pausanias refers to him as ‘a man the inscription says is Oknos’, suggesting that he was not a familiar figure; he was apparently represented as a seated man plaiting a rope, which a she-ass ate as he worked, but no further indication of his appearance is given.23 Envy, on the other hand, is explicitly of undesirable appearance in Lucian’s description of Apelles’ allegorical painting Slander. Envy stands before the throne of a king, who is flanked by Ignorance and Suspicion (Agnoia, Hypolepsis), while Slander (Diabole) drags her unfortunate victim forward, attended by Treachery and Deceit (Epiboule, Apate), and followed at a distance by Repentance (Metanoia), who turns to look at Truth (Aletheia). Envy, clothed in black, is the only male personification in the scene, ‘a pale, ugly man who has a piercing eye and looks as if he had wasted away in long illness’.24
Personifications in adult male form, then, are often of equivocal or even negative values, but the main point is their scarcity. For the most part literary and artistic tradition demands that abstract qualities, positive or negative, should be represented in female form. A good example of the strength of this can be seen in the story recorded by Kallisthenes, quoted by Athenaios, about Famine (Limos); the word is usually masculine in gender, though there are occasional instances of he limos (feminine) in Doric dialects.25 According to Kallisthenes, when the Arkadians were besieging the small town of Kromnos, one of the Spartans under siege managed to send a message back to Sparta via a riddle: the woman imprisoned in the temple of Apollo must be liberated within ten days, as she would no longer be capable of liberation after this time. The riddle is explained:
For this ‘woman’ is in the temple of Apollo beside Apollo’s throne, being a painted representation of Famine in the likeness of a woman. And so it became clear to all that the men in the besieged town were able to hold out only ten days on account of famine [dia ton limon].
(Kallisthenes, ap. Athen., 10.452b [ = FGrH 124 F 13])26
For Kallisthenes the masculine gender for limos was clearly the norm, but he makes no comment on the anomaly of Famine’s female form.
A more confused case is that of the masculine Eleos, ‘Mercy’, whose deification is most often heard of in the context of the ‘altar of Pity’ in the Athenian Agora. A number of our later Greek sources seem to be unhappy with the masculine, and use Philanthropia instead, a word obviously associated conceptually, though not exactly a synonym. In a sample exhortation for an Athenian audience, Apsines (Ars Rhetorica, Spengel i. 39, third century AD) obscures the issue with the statement: ‘You have an altar of Eleos, and universal philanthropia for all is thought to be a god.’ Sopatros (fourth century AD) follows his example, in a speech put into the mouth of Demosthenes: ‘You see how the Athenians worship Eleos along with Athena Polias, you see how they build an altar of Philanthropia on the Agora’ (Sopatros, Diairesis Zetematon, Walz, C. (1835) Rhetores Graeci, Stuttgart: vii. 210). Eleos’ masculinity becomes a serious problem when the altar is referred to by Latin sources: an accurate translation of eleos is, of course, the feminine misericordia, with dementia as a possible alternative; there is no suitable equivalent which would keep the masculine. Seneca the Elder mentions the Athenian altar of Misericordia, and Quintilian maintains ‘that the Athenians, the wisest people, understand misericordia not as a state of mind but as a god’.27 In the Thebaid Statius gives a lengthy description of the altar in the Agora, at which Adrastos and the Argive widows make supplication after the disastrous campaign of the Seven against Thebes. He calls the altar’s deity Clementia, which raises a potential problem in describing Mercy’s cult statue, but, fortunately for Statius, there is no such thing: ‘There is no image, the goddess’ form is entrusted to no metal.’28
Problems of representation obviously also arise with personifications of nouns which are neuter in gender. Old Age (Geras) appears only on five vases from the first half of the fifth century, all depicting Herakles’ confrontation with him, a story for which no literary account survives (Plate 14). Geras is represented as an old man in every case, being particularly wizened and deformed, leaning on a walking-stick, on pelikai in the Louvre and Villa Giulia (Louvre G234, ARV2 286,16; Villa Giulia 48238, ARV2 284,1; see Shapiro 1993: 89–94, nos. 34–8). Grief (Penthos) has human form implied by the story which makes it/him ask Zeus for a share in the honours being handed out to the gods, a fable Plutarch attributes to Aisop. The Greek is equivocal, Grief being designated only once by a personal pronoun in the dative, autoi, which could of course be masculine or neuter, but English translations are obliged to designate Grief as ‘he’ in order to maintain the personification (Consolation to his Wife, 609; e.g. trans. Russell 1993: 299–300). It would be interesting to know how Kratos was represented in the original production of the Prometheus Bound. Its/his companion Bia was presumably female, though masculinity might be expected of ‘Force’ acting as guard of the captive Prometheus; the character is silent, however, in contrast to Kratos, who has a lengthy exchange with Hephaistos in the opening scene of the play. The pair were apparently depicted in the context of the punishment of Ixion on a fragmentary late fifth-century skyphos in a private collection in Basel, but unfortunately all that survives is Bia’s hand and the two inscriptions.29
Plate 14 Herakles and Geras; red-figure pelike by the Geras Painter, c.480 BC (Louvre, Paris G234. Courtesy of the Louvre).
In Monuments and Maidens, Marina Warner (1985: 63–87) discusses the ‘allegory of the female form’ in the classical tradition. She argues that while at first the female form of personified abstractions was an accident of grammatical gender, iconography took hold, perpetuating the tradition. In addition to this, a further rationale for depicting concepts in female form is provided by the female figures of Greek myth, especially Athena; this affixing of meaning to the female form, she suggests, can be seen today in advertising (Warner 1985: 85–6, on ‘engendered images’). Her account is persuasive, but the argument needs to be taken further back.
Literary and artistic tradition did indeed perpetuate a female iconography for personified abstractions, but this iconography has its roots in attitudes towards the feminine. In a male-dominated society extremes of both good and evil tend to be represented in female form, as ‘the other’; further, it is noticeable that all the personified ‘good things’ we have seen are either handsome youths or beautiful young women of marriageable age.30 Is it too fanciful to suggest that they are so represented because both abstract and image are indeed objects of men’s desire? In Prodikos’ ‘Choice of Herakles’ the superficially desirable Vice is described in explicitly sexual terms, appealing to the hero to take the option of a life of decadence and indulgence.31 Aristophanes makes play with just this connection with the silent female characters in several plays who are leered at by the men on stage: the ‘Peace-treaties’ (Spondai) of the Knights, ‘Reconciliation’ (Diallage) in the Lysistrata, ‘Vintage’ and ‘Festival’ (Opora and Theoria) in the Peace (Spondai, Ar., Eq. 1389–95; Diallage, Lys. 1114–36; Opora and Theoria, Pax 523–6, 706–14, 871–6, 1329–57). While the very fact that women have a low profile makes the female form a practically suitable vehicle for abstract ideas in search of an incarnation, psychologically their desirable form conveys the desirability of the abstract values they embody.
Thanks to Gideon Nisbet for discussing ideas for this chapter at an early stage, and to the anonymous readers for their detailed and helpful comments. Thanks are also due to Martin Seeker & Warburg for permission to reproduce the extract from Umberto Eco, The Name of the Rose, 1994.
1 For general accounts of Greek personification see Gombrich 1971 and Webster 1954. On personifications in Greek art 60CM00 BC see Shapiro 1993; for a collection of references to personification cults in the same period see Hamdorf 1964. I discuss the cults of six personified abstract nouns at length in my Ph.D. thesis (Stafford, forthcoming); five are of feminine nouns (Themis, Nemesis, Peitho, Hygieia, Eirene), one of a masculine noun (see below on Eleos).
2 Blundell (1995: 17) comments on the paradox of female deities’ prominence in a patriarchal society.
3 Zoilos, a freedman of Augustus, himself stands next to Time, ‘Honour’, while other personifications on the frieze include Mneme (Memory), along with the less abstract Polis (City-state) and Roma: Erim 1979: Pis 21–9.
4 That bells should be ‘she’ is particularly striking, given such proper names as ‘Great Tom’ (thanks to Lin Foxhall and Hamish Forbes for a bell-ringing lesson). The inherent androcentrism of English is discussed at length by Spender 1990 and Key 1975. The attribution of gender to the world around us can be seen as part of a more general human penchant for anthropomorphism (Yaguello 1978: 11).
5 Thanks to the Greek students in my Greek Religion seminar group at Royal Holloway (1994/5) for compiling a list, which also included: Agape, Athanasia, Arete, Eudokia, Eudoxia, Euthymia, Eumorphia, Eutychia, Zoe, Parthenia, Pistis, Sophia.
6 Athenaios, 13.577a, 583e, 587f, 593b, 4.157a. A ‘mythical’ poetess Phantasia, wife of Memphes, daughter of Nikarchos is mentioned by Eustathios (Commentary on the Odyssey 1379.62) and Photios (Bibliotheke 3.151ab). Thanks to Richard Hawley for references.
7 For surveys of nineteenth- and twentieth-century theories on grammatical gender see Fodor 1959: 7–30, and Ibrahim 1973: 14–50.
8 Ars Grammatica, 634b (ed. G. Uhlig, 1883, Grammatici Graeci, Leipzig: i. 24–5). Latin grammar early fell under Dionysios’ influence, and via Latin most of the modern grammars of Europe are indebted to him. See Robins 1951: 1–68 on Greek and Latin grammatical theory.
9 Cf. Varro, De Lingua Latina 9.55–62, for a defence of natural gender.
10 Meillet 1921/38: ii. 24–5: the feminine is a subdivision of the ‘genre animé’, which not only covers living beings, but is also extended to everything considered ‘animated’, e.g. earth, tree (opposed to the fruit it bears, which is ‘inanimée’, neuter gender), hand; once such grammatical categories have been created, all animate substantives have to be assigned masculine or feminine gender, the division often seeming entirely arbitrary. But even Meillet suspects that in the prehistoric languages which Latin, etc., developed from, the feminine gender did have some kind of value, though he does not enlarge on this. Commenting on Meillet (in the same volume), Mauss points out the necessity of taking into account social and psychological factors when studying this question of categorisation.
11 See the linguistic and sociological studies cited by Key 1975 and Spender 1990. Many of their points are relevant to our enquiry, especially in so far as they relate questions of linguistic gender to the place of women and the feminine in society.
12 See Loraux 1992 for a discussion of gender difference and the divine. Cf. Blundell 1995: 191 for the suggestion that Athena’s rather ‘masculine’ iconography reflects her problematic status as a powerful female deity.
13 On Eros, see Hermary et al. 1986: esp. 851, on his cult. See Shapiro 1993: 110–24 on the ‘ubiquitous triumvirate’. A philosophical genealogy makes Eros the result of an unlikely union between Resource (Poros) and Poverty (Penia): Plato, Symposium 203b-c; cf. Plut., On Isis and Osiris 57: 574d.
14 Cf. the relationship between the active/passive opposition and the masculine/feminine distinction in grammatical gender: e.g. the active agent phylax is masculine, while the more passive abstract phylake, ‘guarding’, is feminine. Lysippos’ famous statue of the Opportune Moment (Kairos) at Sikyon was also a desirable youth, described in an epigram by Poseidippos (AP 16.275) and represented in later reliefs and gems; see Pollitt 1986: 53—4, fig. 47, and Kershaw 1986: ch. 3. Pausanias records an altar to Kairos at Olympia (5.14.9), and cf. Alexander Rhetor’s analysis of the use of a personified Kairos in Demosthenes’ Olynthiac 1.2 (De figuris, Spengel, L., ed. 1854–6, Rhetores Graeci, Leipzig: 3. 19).
15 I discuss the Hellenistic iconography of Eros briefly in Stafford 1991/3.
16 Pausanias 1.8.2; Julio-Claudian copy, Munich Glyptothek 219. The group is represented on a set of six Panathenaic amphorai of 360/59 BC from Eretria (e.g. Eretria Museum 14815; Simon 1988: PI. 5) and on a second-century AD Athenian bronze coin (Simon 1986: no. 4). On Kephisodotos and the Athenian cult of Eirene see Stafford, forthcoming: ch. 6.
17 This character of agricultural wealth is already suggested in the birth-place Hesiod assigns Ploutos in ‘the thrice-ploughed field, in rich Crete’ (Theogony 969–74). Homeric Hymn to Demeter 486–9; cf. Drinking song, Attic skolia 885 (Campbell 1988). Ploutos is amongst the otherwise unexceptional list of Demeter’s associates in the call to prayer in Aristophanes’ Thesmophoriazousai of 411 BC (295–300), though the idea that Wealth should be a god seems to be ridiculed in Euripides’ Cyclops just a few years later (316–17).
18 Ar., Plout. 230. MacDowell 1995: 329–31 describes Wealth as ‘one of the most complex allegories in any of Aristophanes’ plays’, and points out the ambiguity of his status as god/man/abstract, especially noticeable in his speech at 234–14.
19 Cf. Plato’s derivation of Pluto from ploutos, as the king of the underworld is ‘giver of wealth’ (Kratylos 403a). See Clinton 1994 for representations of Ploutos in vase-painting, and on ancient and modern confusion with Pluto; he concludes that ‘P. was a mere personification and never given formal worship as a god; no sacrifices are attested’ (416). The only other appearance of Wealth in a ritual context is less obviously anthropomorphic, in Plutarch’s description of ‘the driving out of ox-hunger’ at Chaironeia, a ceremony which he relates as being practised in his own time (Plut., Table Talk 6.8 = Moralia 693 e-f); on the ritual, see MacDowell 1995: 270–1, nn. 17–18.
20 See Shapiro 1993: 132–65 on vase-paintings. Sleep undergoes a similar process of ‘juvenation’ to Eros in the Hellenistic period, often being virtually indistinguishable from the latter: Stafford 1991/3.
21 Paus., 2.31.3; Plutarch, Kleomenes 9; cf. statues of Sleep and Death at Sparta (Paus., 3.18.1).
22 Plut., Theseus 27 and Alexander 31; Appian, Punica 21; cf. Fear at Selinous (IG xiv. 268).
23 Paus. 10.29.1. See Shapiro 1993: 178–9 on this and another possible representation of Oknos (black-figure lekythos, Palermo 996).
24 On not being quick to believe slander 6–8. Botticelli’s Calumny of Apelles, in the Uffizi, follows Lucian’s description closely, except for the figure of Truth; see Warner 1985: 316–17.
25 Used e.g. by the Megarian in Ar., Acharnians 743.
26 See West 1966: 231 (ad Theog. 227) for further references to Limos personified: ‘the gender of the noun and the sex of the god are variable’.
27 Seneca, Controversiae 10.5.10; Quintilian, 5.11.38. Cf. kairos/occasio for Greek/Latin gender-crossing.
28 Thebaid, 12.492–3. Wycherley 1954: 148 comments that the lack of a statue is lucky for Statius ‘otherwise the slight difficulty of the gender of the deity … might have become acute’. On the altar of Eleos see Stafford, forthcoming: ch. 7.
29 Aesch., PV 1–87; on the disputed date and authorship of the play see Griffith 1977.
30 Homer’s Litai are the only female personifications to be characterised as elderly, their age in relation to Ate dictated by the logic of the allegory (Horn., II. 9.502–12).
31 Prodikos fr. 2 DK (= Xen., Mem., 2.1. 21–34); on Herakles’ encounter, cf. Fox, Chapter 2 in this volume.