Hostname: page-component-586b7cd67f-dlnhk Total loading time: 0 Render date: 2024-11-27T14:32:56.042Z Has data issue: false hasContentIssue false
  • English
  • Français

Divorce in Classical Athens

Published online by Cambridge University Press:  23 February 2012

Louis Cohn-Haft
Louis Cohn-Haft
Affiliation:
Castellina in Chianti, Siena, Italy
Get access Rights & Permissions [Opens in a new window]

Extract

The modern literature on divorce in Classical Athens is slight, the only detailed discussion that of W. Erdmann, Die Ehe im alten Griechenland (Munich 1934; repr. New York 1979) 384–403. A rare certainty in our knowledge is the ease with which a husband could terminate marriage. He had only to send his wife away, that is, back to her paternal family, and the marriage was at an end. From this it is tempting to infer that divorce in Athens was frequent, even casual. Not surprisingly that view has had a long tradition in works on marriage and family, law, society, and ancient Greece in general. It is a view almost surely incorrect, however, as the following examination of the evidence will show.

Type
Articles
Information
The Journal of Hellenic Studies , Volume 115 , November 1995 , pp. 1 - 14
Copyright
Copyright © The Society for the Promotion of Hellenic Studies 1995

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 See above all, Erdmann 388; also inter alios, Savage, C.A., The Athenian family: a sociological and legal study (Diss. Johns Hopkins U) (Baltimore 1907) 61Google Scholar, Paoli, U.E., La donna greca nell' antichità (Firenze 1953) 48Google Scholar, Harrison, A.R.W., The law of Athens: the family and property (Oxford 1968) 40Google Scholar (hereafter Harrison), the implications in Pomeroy, S.B., Goddesses, whores, wives, and slaves: women in classical antiquity (NY 1975) 64Google Scholar, Biscardi, A., Diritto greco antico (Milano 1982) 99–100Google Scholar, Cantarella, E., Pandora's daughters: the role and status of women in Greek and Roman antiquity, trans. Fant, M.B. (Baltimore 1987) 47Google Scholar; and such durable or recent general or popular works as Blümner, H., The home life of the ancient Greeks, trans. Zimmern, A. (1893) (repr. NY 1966) 149Google Scholar, Cohen, R., La Grèce au 5e siècle (1953 ed.), Vol. iiGoogle Scholar in Glotz, G., Histoire ancienne, 4 vols. (Paris 1925–1938) 255Google Scholar, Botsford, G.W. and Robinson, C.A. Jr, Hellenic history (5th edit., rev. Kagan, D.) (NY and London 1969) 358Google Scholar, Wilkinson, L.P., Classical attitudes to modern issues (London 1979) 53Google Scholar, JACT [Jones, P.V. et al. ], The world of Athens: an introduction to classical Athenian culture, (Cambridge 1984) 163.Google ScholarIsager, S. ‘The marriage pattern in classical Athens: men and women in Isaios’, C & M xxxiii (1981–1982) 85–87Google Scholar, dealing solely with the evidence from Isaeus, is a rare exception. She notes that divorce ‘is held to have been common’ in Athens, citing several recent authorities, then observes that although there are five cases in the works of Isaeus in which it is possible that a divorce occurred, only one is certain. The generalization quoted by Medick, H. and Sabean, D.W. ‘Interest and emotion in family and kinship studies: a critique of social history and anthropology’, in Medick, and Sabean, , eds. Interest and emotion: essays on the study of family and kinship (Cambridge 1984) 21Google Scholar provides an interesting comment on this point, but whether it applies to Athens is uncertain: ‘It is…almost a universal rule that when married life is insecure, the wife turns for support to her family of origin, so that a weak marriage tie produces a strong blood tie…’ (quoted from Young, M. and Willmott, P., Family and kinship in East London (London 1957, repr. 1972) 189.Google Scholar

2 Ar. Lys. 156 can be squeezed to appear to be a reference to divorce for adultery, but Menelaus is there depicted not as about to divorce Helen, but to kill her. We have no evidence that such a “Divorzio all'Italiano” was permissible under the law that required a cuckolded husband to divorce his erring wife; see below, n. 11. In the next line of the play Kalonike refers to the possibility of husbands simply walking out. Such action, given the freedom of movement of Athenian men, should not be construed as divorce; see, for example, the behavior of Euktemon in Isaeus vi. One may wonder too as to why Strepsiades does not divorce the wife he rants against in Clouds 41–74, and a guess is permissible that an inability to return her presumably large dowry might be the reason. (On dowry as a deterrent to divorce, see below, nn. 14, 48, and esp. 53.) The question, however, does not explicitly arise anywhere in the play.

3 Hdt. i 59 ekpempein, v 39–41 exeō, apenta, tēs exesios, vi 61–63 apagesthai, apopempsamenos. Not unexpectedly Herodotus, whose interest is not in law but in literary effect, uses a variety of expressions. Exesis (v 40) is the only noun in this collection, and it is found nowhere else: L-S-J s.v.

4 Xen. Oik. 3.12 and the tendentious inference drawn by Erdmann 88.

5 Plato, Laws vi 784b–cGoogle Scholar, viii 841c–d, ix 868c–d, xi 926b–c, xi 929e. These are all cited by Erdmann 401–03, who concludes his treatment of divorce by observing that Plato' arrangements, being closer to modern law than to the contemporary laws of Athens, are therefore a notable advance.

6 See Humphreys, S.C., ‘The discourse of law in archaic and classical Greece’, Law and History Review vi (1988) esp. 455–456, 473–482.Google Scholar

7 Cox, C.A., ‘Sibling relationships in classical Athens: Brother-sister ties’, J. Fam. Hist, xiii (1988) 381Google Scholar finds another case by treating one feature of the tangled relationship of Plangon and Mantias in Dem. xxxix 24–25, 28 as a divorce ‘because she (sic) could not provide a dowry from her father's impoverished estate: she and her sons were cared for by her brothers’. To be sure the rejection by Mantias of his sons by Plangon and of Plangon herself can be called a divorce by the loose standard of Athenian procedure, but his later resumption of the marriage renders such a concept of divorce almost meaningless. (In the matter of dowry Cox is surely wrong on several counts: although the speaker asserts (xl 22–24) that Plangon's father was too poor to provide a dowry, her sons claimed there was a dowry (xl 14, 20); after living with a woman as his wife for a long enough time to have two sons, a man would not suddenly discover that he had received no dowry; if the brothers could support her and her sons, they could surely have supplied a dowry; and to split a technical hair, the woman would not herself have been responsible for provision of a dowry.)

8 I use the words regularly employed by scholarship, but it needs to be said that, as usual in Athenian law, these words lack the precision and exclusivity that we expect from legal terminology. Indeed one of them, as will be seen, is not found as a noun at all in the sources. It is also the case that there is no single word that corresponds to our “divorce”, meaning the legal termination of a marriage: Athenians, as the present discussion will show, were more concerned with how and why a divorce occurred than with the legality and the technicalities of the fact of a divorce. Marriage was fundamentally a private matter, defined by law but with no legal or public action required; so with divorce, which normally was a private action.

9 On divorce by mutual consent, see Harrison 39–40. Although mutual consent is humanly interesting, it was technically of no significance: the divorce was initiated by the husband. As for Plutarch's dependability for fifth century private life, note the difficulty of so much as identifying the lady in question: Stadter, P. A., A commentary on Plutarch's Pericles (Chapel Hill 1989) 238–39Google Scholar, where reference is made also to the very different version of the divorce in Athen. xii 533c–d.

10 While one must always preserve a skeptical stance with regard to any assertion of fact in a trial oration, there is no reason to doubt these allegations, and in any case they are important as indicating that a concern for the blameless wife's feelings would go over well with the ordinary Athenians on the dikastery.

11 The law is cited in Dem. lix 87. On adultery in Athens the fundamental treatment, that of Paoli, U.E., ‘Il reato d'adulterio (moicheia) in diritto attico’, Studia et documenta historiae et iuris xvi (1950) 123–82Google Scholar, was for decades accepted by legal scholars. (A good summary in Harrison 32–36.) A complete bibliography is provided in the notes to Cohen, D. ‘The Athenian law of adultery’, RIDA xxxi (1984) 147–65Google Scholar, who proposed a radically different understanding of the Athenian law(s). Cohen's views have been refined and expanded in Chapter 5 of his Law, sexuality and society (Cambridge 1991). They are discussed by Cantarella, E., ‘Moicheia: reconsidering a problem’, Symposion 1990 (Cologne 1991) 289–296Google Scholar, who had access to Cohen's new formulation. Cantarella focusses on legalistic issues and rejects the fundamental novelty of Cohen, namely that adultery was defined in Athens, as universally in other societies, as a crime against marriage (that is, in which at least one of the partners is a married person). On this issue she returns to the traditional view (Paoli's) that even for an unmarried man, sexual intercourse with women in certain other categories, including widows and unmarried women, was included under the rubric “moicheia” (which thus ought perhaps to be rendered otherwise than as “adultery”). L. Foxhall, ‘Response to Eva Cantarella’ 297–304 in the same volume, does not deal with legal questions, but with “the social construction of gender” (297) and the “ideology of sexual control” (299). The relation of this controversy to divorce is limited to conjectures as to the underlying attitudes of Athenians toward female sexuality.

12 The verb is ekballein, which L-S-J s.v., i 4, citing this passage, renders simply as ‘divorce’. In Dem. lix 59, in reference to the same incident, the more or less technical term, apopempein, is used. One may reasonably infer that the more graphic and forceful verb was employed for rhetorical effect, whether to be taken literally or not.

13 A law cited in Dem. lix 16 indicates that marriage of an Athenian to a non-Athenian was in the fourth century a criminal offense.

14 In the Athenian system dowry was a significant aspect of marriage. As regards divorce the requirement that dowry be returned with the woman must sometimes have served as a deterrent. On dowry, see n. 40.

15 Dem. xxvii–xxxi all relate to the orator's efforts to prove that his guardians had swindled him and thereby to recover his patrimony. The woman in the present case was the sister of the defendant, Onetor, and following the alleged divorce he supposedly gave her in marriage to Aphobos, the defendant in xxvii–xxix, to whom Demosthenes' father had willed his widow Kleoboule as wife, but who had taken the dowry and failed to marry her. Demosthenes' mother's fate is a significant part of that story (on which see Hunter, V., EMC 8 (1989) 39–48Google Scholar), but is unrelated to the divorce here under discussion.

16 Reported also by Plut. Alk. 8.4:

17

18 The four supporting sources are: Menander Epitrepontes, several passages, esp. 656–58, 714–15, 929–31, 1064, 1102–03 (Sandbach); P. Didot 1, especially 1–44; Plautus Stichus, opening scene through line 148; Rhetorica ad Herennium ii 38 (10 lines).

19 The right is understood as stemming from the nature of the agreement between the father of the bride and the bridegroom, which has the form of a conditional transfer of ‘ownership’, engyē (see also below, n. 26). It is this conditional nature which gives to the father the power to abrogate the agreement. See especially Wolff, H.J., ‘Marriage law and family organization in ancient Athens: a study on the interrelation of public and private law in the Greek city’, Traditio ii (1944) 53Google Scholar (repr. 1961 in Beiträge zur Rechtsgeschichte Altgriechenlands und des hellenistisch-römischen Ägypten (Weimar 1961), who emphasizes the limited nature of the exchange: ‘The aim of the engyē was to entrust rather than to alienate the object’. Pomeroy, S. B. ‘Greek marriage’, in Grant, M. and Kitzinger, R., eds., Civilization of the ancient Mediterranean (NY 1988) 1340Google Scholar puts it most strongly: ‘A married wman was actually only lent to a husband for the production of legitimate children for his oikos. Her family retained the right to reclaim her services to produce its own children’. The reclaiming appears in the legal literature as aphairesis: see especially Paoli, U.E., ‘La legittima afèresi dell'epikleros nel diritto attico’, Miscellanea G. Mercati, Vol. 5 (Biblioteca apostolica vaticana. Studi e testi cxxv [1946] 524–38)Google Scholar and Harrison 30–32. The noun is entirely appropriate, but not in fact attested in this context. Only verb forms appear in the texts. (Just, R., Women in Athenian law and life (London 1989) 74–75Google Scholar opines that ‘most probably’ a woman's brothers, if the father was no longer living, could exercise the right. There can, however, be no question but that sons inherited whatever rights and obligations their father had had.)

20 Rosivach, V.J. ‘Aphairesis and apoleipsis: a study of the sources’, RIDA xxxi (1984) 193–230Google Scholar (hereafter Rosivach) provides an exhaustive analysis, with references to the previous literature. The most detailed earlier treatment is that of Lewis, N. ‘Aphairesis in Athenian law and custom’, Symposion 1977Google Scholar (Actes du 3e colloque internationale d'histoire du droit grecque et hellénistique) 161–78, who concludes that ‘Attic law invested a father with an absolute right of aphaeresis…’. Sbordone, F. ‘Una tipica contesa familiare nella realtà giuridica e nel teatro antico’, Sileno xi (1985) (=Studi Barigazzi ii) 207–10Google Scholar deals only with literary questions arising from the texts.

21 Others have also favored this interpretation of the Demosthenic text, e.g., A.T. Murray p. 7 n. c in Vol. v of the LCL Demosthenes (1939), Wilkinson (n. 1) 53.

22 In the Epitrepontes and the Stichus the women retain their husbands and all ends happily. If the other two texts followed the conventions of New Comedy they too ended happily with divorce avoided.

23 Reliance on Roman Comedy dies hard: see Bickerman, E.J., ‘La conception du manage à Athènes’, BIDR xvii (1975) 21 n. 102Google Scholar, who on the authority of Paoli accepts that Plautus and Terence ‘generally reproduce faithfully the customs and legal rules presented in the original’; also Lewis (n. 20) 171 n. 34, who holds to ‘Fraenkel's Law’, according to which ‘legal elements indispensable to the plot…are surely Greek, while those only superficially connected with the action of the play are likely to be Roman’; McDonnell, M. ‘Divorce initiated by women in Rome’, AJAH viii (1983) 54–57 and nn. 7–24Google Scholar; and Konstan, D. ‘Between courtesan and wife: Menander's Perikeiromene’, Phoenix xli (1987) 122 n. 6CrossRefGoogle Scholar who strains to show that Plautus and Terence knew Greek law and strove to cite it accurately, but fails to explain why cornic writers would subject a Roman audience to scholarly precision or what such pedantry would accomplish, even if we allow them to have been students of Classical Attic law. On the other hand Harrison 19 is appropriately cautious and sparing: see his index of sources.

24 In line 930 Charisios says his wife has not left (divorced?) him (the verb used is apoleipein), and there is the possible implication that her failure to initiate a divorce settles the matter. The text, however, is fragmentary and does not demonstrate definitively that Smikrines could not on his own have removed his daughter from the marriage. It does suggest that the father would prefer to persuade her to initiate the divorce herself, on which point see below, Grounds for Divorce (p. 9).

25 Given the fundamentally private and informal nature of the various procedures, an outsider might have been unable to distinguish which one had been used, with the one important exception noted above: apoleipsis was a matter of public record.

26 Wolff (n. 19) 46–53 argues that the terms engyē (‘solemn promise’) and ekdosis (giving of the bride), imply a transfer of property that is for a limited purpose, procreation, and allow the original owner to retain his interest in the ‘property’. Modrzejewski, J., ‘La structure juridique du mariage grec’, Scrirti in onore di Orsolina Montevecchi (Bologna 1981) 258–60Google Scholar takes the same position, emphasizing the critical importance of offspring: that is, the existence of a legitimate child nullified the right of aphairesis. That view has logic on its side, but no concrete evidence, and is appropriately treated by MacDowell, D.M., The law in classical Athens (Ithaca, NY 1977) 88Google Scholar (hereafter MacDowell) as no more than a conjecture. The sentiments of jurors might, of course, have overridden the letter of the law in such an instance; see below, n. 32. A fresh treatment of the significance and meaning of engyē, emphasizing social reality rather than legalistic detail is found in a valuable article by Patterson, C.B., ‘Marriage and the married woman in Athenian law’, in Pomeroy, S.B., ed. Women's history and ancient history (New York 1991) 49–54.Google Scholar

27 One may conjecture that in such a situation, with the facts not publicly known, a cloud might permanently hover over the reputation of a woman peremptorily taken back by her father, who would therefore prefer she take the step herself rather than leave the initiative to him. Such an hypothesis may explain the apparent contradictions among the passages cited from the comedy of Menander, in which the father tries to persuade his daughter, but says he will remove her anyway.

28 In Lys. xxxii 4–6 a certain Diodotos, a man of considerable wealth, departing on a military campaign (in which he in fact lost his life), left a will providing for the settlement of his estate and the remarriage of his wife, who was his niece, in case of his death. All these matters he put into the hands of his brother, who as her father would anyway have automatically become her kyrios upon her husband's death. This is the only evidence we have directly on this score. In two other instances, Dem. xxvii 4ff. and xxxvi 7–8 men who were ill and anticipating death made wills in which they provided for the guardianship and remarriage of their wives. One is tempted, in the face of such substantial evidence, to take these cases as normative, but the cautionary note needs to be sounded that written wills were always looked upon with suspicion in the fourth century as being untraditional and also easily forged. And apparently with some reason: in two of the cited cases, the appointed guardians swindled their wards, and in the third, Dem. xxxvi, a grown son, whose normal receipt of his father's estate had been pointedly and insultingly bypassed by the will, claimed that there was no such document at all. That he lost the case does not lessen the force of his argument that a man with a mature son would not normally make a will. One may doubt that men leaving home on military service or a lengthy voyage left wills. It is altogether more likely that a verbal transfer of authority to a close relative was customary. For example, Apollodoros (the loser in Dem. xxxvi) describes in another case, Dem. 124 ff., how while he was on military service, an acquaintance back in Athens, attempting to assist him in a financial matter, brought to the meeting Apollodoros' father-in-law who, although no statement is made to this effect, was the person most likely to have had charge of his affairs and of his family during his absence.

29 A welcome confirmation of the historicity of at least part of the tale is found in a dedication from the agora, Hesperia vi (1937) 341, mentioning ‘Kleokrateia, daughter of Polyeuktos of Teithras, and wife of Spoudias’.

30 He was Polyeuktos' son, the disinheritance of a son required legal action, see Rudhardt, J., ‘La reconnaissance de la paternité: sa nature et sa portée dans la société athénienne (Sur un discours de Démosthène)’, MH xix (1962) 50–52.Google Scholar In Dem. xli 5 Leokrates’ ouster from the family is discreetly glided over:

31 The speaker's circumlocutory claim that the quarrel between Polyeuktos and Leokrates is irrelevant to the present case perhaps suggests a ground other than money for their enmity.

32 The distance between law and social sanction is put neatly, in the context of divorce initiated by a wife's father, by Lewis (n. 20) 178: ‘[the father] was under no legal constraint to justify such action…; …he…was under strong emotional and social pressures to seek an accommodation’.

33 There is a huge literature on this subject, in which controversy over technical detail abounds but is not germane to the present problem. The most recent comprehensive treatment is that of Karnezis, J.E., The epikleros (Athens 1972)Google Scholar (in Greek with detailed English summaries), containing thorough references to the previous literature; the ancient sources are listed 236–38. Karnezis' work is not universally admired; see the harsh review by MacDowell, , JHS (1976) xcvi 228.Google Scholar Some of the nuances in interpretation may be observed especially in the work of Balogh, E., ‘Some notes on adultery and the epikleros according to ancient Athenian law’, Studi in memoria di Emilio Albertario, 2 vols. (Milano 1953) ii 697–719Google Scholar, Harrison 10–12, 13–38, Lacey, W.K., The family in classical Greece (Ithaca, NY 1968) 139–45Google Scholar, Schaps, D. M., ‘Women in Greek inheritance law’, CQ xxv (1975) 53–57CrossRefGoogle Scholar, and Economic rights of women in ancient Greece (Edinburgh 1979) Chap. 3, MacDowell 95–97.

34 The question of whether claiming an heiress was a right or an obligation is readily—and cynically—answered by reference to her fortune. If a substantial property came with her, marriage to her was a right which the eligible relative was probably glad to exercise, and if he failed to do so, the next in line could be counted on to supply the necessary eagerness; if she was poor or still worse destitute, it was a duty that could be avoided only by supplying a dowry of a value specified by law. Andok. i 117 ff. is refreshingly explicit and unambiguous on this distinction.

35 If the existence of legitimate offspring was decisive in sealing a marriage, as some have argued (see n. 26), the situation here under examination illustrates either the apparently still more powerful element of a husband's freedom to divorce at will or the overriding importance of keeping property within the family. Possibly if the child had been a son (the text is messy on this point: in 40 she had a daughter, in 43 “children”, of whom only the daughter is specified), it would not have been permissible to remove the mother from her marriage; see Harrison 11–12, 308–11. Perhaps then the mention of the divorced wife's having been given in the new marriage by her brother in the presence of other relatives is an indication of her family's having accepted a situation they might legally have opposed.

36 The absence of concrete evidence has not prevented the notion that, as expressed by Ehrenberg, V., The people of Aristophanes: a sociology of old Attic comedy (NY 1962) (orig. Oxford 1943) 146Google Scholar, ‘childlessness often led to divorce’. The famous instance in Hdt. v 39–41 is not germane: it occurred in an earlier period, in Sparta not in Athens, had to do with royalty not ordinary citizens, and the king in that anecdote, although acknowledging his duty to produce an heir, refused to divorce his childless wife.

37 Athenians were as unlikely as any others to acknowledge sterility, which was comfortably denied by men before the advent of modern science, as in the case of another Spartan king, Ariston: Hdt. vi 61.2.

38 He is quoted (7) as saying her goodness should not be rewarded by letting her grow old childless with him. The assertion is clear, the sentiment subject to a variety of interpretation, ranging from the prudential concern to be looked after in old age to the emotional satisfaction of realizing her female destiny by bearing children.

39 In 18, and twice in 33, in the active.

40 Since the divorce was officially inscribed (15), it must have been technically an apoleipsis, initiated by the wife. Still, in describing the actions, Demosthenes asserts (17) that it was the men who had the divorce registered, and nowhere is there mention of the wife appearing before the archon. One supposes that these confusing details occur because the language of a speaker before an Athenian court was not the precise technical jargon of a lawyer and because the generalities were sufficiently well known so that no confusion resulted. There is, of course, always the possibility of a deliberate effort to befuddle the hearers. Amid a large scholarly literature on dowry, a detailed and balanced account may be found in Harrison 8, 45–60, 297–301, to which may be added the treatment in Schaps (n. 33) Chap. 6, with full bibliography. Divorce provided a real test of the control over dowry. In principle and in law dowry went with the woman, to be administered by whoever was her legal guardian. A divorcee had no place to go but back to her father's family, which would thereby gain control of the dowry. In practice, however, the guardian of a divorced wife might have found it difficult to recover the dowry from her ex-husband, since the breakup of the marriage would strain if not sever the relations between the two families. In the case of divorce instituted by the wife there would almost surely be hostility between the husband and the wife's blood relations, and that could have served to encourage the husband to be defiant about returning the dowry.

41 For the seriousness of acts envisaged as causing corruption of family and home, see Paoli (n. 11) especially 123, 126, 140–41. The point is powerfully made by Euphiletos in his speech to his wife's lover, Lys. i 26. A slave woman who was part of the household was a different matter, and a husband's trifling with her was a trifle indeed, as indicated in Euphiletos' rather coy reference to his own dalliance with his wife's maid, Lys. i 12.

42 Harrison 41 disagrees, but on the basis of the quite inadequate evidence of Dem. xxx.

43 Surprisingly, even this ostensibly unexceptionable assertion has been contested. Foxhall, L., ‘Household, gender and property in classical Athens’, CQ xxxix (1989) 38Google Scholar finds, with no evidence cited but perhaps following Wolff, that if a woman did not like her husband's management of her dowry ‘or anything else he did’ she could easily walk out, ‘taking her dowry with her’. Although it is possible, even likely, that a wife could walk out without ceremony, there would be no place for her to walk to unless she had persuaded her male kin that her move was justified. The problem of how she was to get hold of the dowry to take with her is discreetly ignored. Wolff, H.J., ‘Die Grundlagen des griechischen Eherechts’, Tijdschrift voor rechtsgeschiedenis (= Rev. d'histoire du droir) (1952) 11Google Scholar (repr. in Zur griechische Rechtsgeschichte, ed. E. Berneker, Wege der Forschung xlv [Darmstadt 1968]), perhaps bemused by his effort to show that an Athenian wife was less strictly tied than a Roman manus-wife, flatly states that an Athenian wife was free to divorce at will, but he does observe that the actual practice of this right was bound to be restricted by social pressure. Keuls, E.C., The reign of the phallus (NY 1985) 101Google Scholar on the other hand sees the woman's position as very bad indeed: ‘…the wife's technical right to divorce in case of maltreatment was largely illusory….’ because ‘her guardian…would normally be loath to have her back to marry off a second time’.

44 The only way the direct testimony of a woman (as distinct from simply quoting her, for what that would be worth) could be brought into court was by quoting a formal oath, taken in an appropriate temple before witnesses. This procedure is referred to occasionally in the trial orations, but there is no single instance in which it was actually introduced in the proceedings. In Dem. Iv 27 and Is. xii 9 reference is to the offer of such oaths, not to their having been taken. The unique case in which the oath was actually taken—with decisive effect—is referred to, but not introduced in court, in Dem. xxxix 25 and xl 10–11. On the subject of women's testimony in court, see Humphreys, S.C. ‘Kinship patterns in the Athenian Courts’, GRBS xxvii (1986) passim, esp. 72.Google Scholar

45 The following effort at a reconstruction of the rationale and procedure of this type of divorce is entirely conjectural. Others have imagined the generalities and specific details differently; see e.g., Mossé, C., La femme dans la Grèce antique (Paris 1983) 54–55Google Scholar, Wolff (n. 43) 11.

46 Erdmann 395–96 sees the support of a male relative as crucial. This view is not attested in any source but seems certain anyway. Nevertheless Gould, J., ‘Law, custom and myth: aspects of the social position of women in classical Athens’, JHS c (1980) 44 n. 40Google Scholar believes on the authority of Plutarch that she ‘could not be represented by others’. A question arises to which no one has found an answer and that is the relation of the procedure under examination to that of divorce by peremptory action of the wife's father (pace Rosivach, see above, who solves the problem by reducing aphairesis to a kakophemism for apoleipsis, but one may doubt that the undoubted amateurism of Athenian law was quite that casual). If the residual right of a married woman's family to take her back was an unrestricted right, why not just take her back and not bother with the divorce before the magistrate? Perhaps a desire to make the husband's culpability a matter of public record? Or perhaps the publicity of a divorce before the archon made recovery of the dowry easier? Or …?

47 Plut. Alkibiades 8.5 interprets the episode differently, slightly closer to the conjecture above. Neither the testimony of [Andokides] nor that of Plutarch is altogether reliable. On the less than dependable and perhaps not contemporary text of Pseudo-Andokides, see Maidment 534–39 in the LCL Andokides (1941). Cox (n. 7) 381 believes that the husband's right prevailed.

48 It is perhaps at this point that the woman's blood relations, if they still wanted the divorce, could resort to aphairesis, a power presumably greater than that of the husband's claim, if the argument above, at n.19, is correct; unless, of course, the existence of a child served to nullify the family's power (see n. 26), which may have been the situation in the case of Hipparete's unsuccessful effort. Alkibiades' determination to keep his unwilling wife may perhaps be credited to the fabulous dowry ([Andok.] iv.13–14) he would have had to return had she succeeded in divorcing him.

49 Dem. xxx 15: ‘the wife-initiated divorce was registered.’ MacDowell 48 notes that a central record office in Athens was an innovation of the period 403–399.

50 Plut. Per. 24.5, Is. ii 8–9, Dem. lvii 41. So also in Dem. xli, the single case of aphairesis, the father had a new husband ready, thereby nullifying the possibility of his daughter's return being understood as dismissal by her husband. The situation of a divorcee was somewhat similar to that of a widow of child-bearing age, except that a widow was not as it were automatically under suspicion as somehow unworthy. On widows see the detailed study of Hunter, V., ‘The Athenian widow and her kin’, J. Fam. Hist, xiv (1989) 291–311.CrossRefGoogle Scholar

51 See esp. MacDowell 235–37. It is a commonplace that in general Athenian officials had extremely little initiatory or discretionary power. Divorce for adultery was required by law (above, n. 11), but although some cuckolded husbands may have—for whatever reason—evaded the requirement, there is absolutely no evidence for any public agency ever having taken the initiative to correct such a situation. Even in the hypothetical instance of a husband bringing an adulterer before the Eleven for punishment, Cohen (n. 11) (1991) 116–122 does not suggest that this or any other public body would take steps to determine whether the required divorce had in fact occurred. Presumably the mere fact of the public airing of the crime would be sufficient to guarantee the husband's conforming to the law.

52 Just as laws themselves were enforced through suits brought by individuals. One needs to remember that Athens maintained no public prosecutor nor any other agency to initiate legal action even in crimes against the state.

53 The requirement that the dowry be returned with a rejected wife may well have contributed as a deterrent to divorce. It is emphasized as primary by MacDowell 88, Flacelière, R., ‘Crete and Greece’, in Grimal, P. (ed.), Histoire mondiale de la femme i (Paris 1965) 314Google Scholar, Murray (1936) 136–37 in Vol. iv of the LCL Demosthenes. While it no doubt played a part, if one wishes to argue on such purely monetary grounds one must accept that the prospect of greater fortune was a simple arithmetical calculation, but the element of enmity with a closely tied family would need to be balanced against the new tie with a richer but not necessarily more powerful family.