Talmud-Based Solutions to the Problem of the Agunah To my parents Agunah Research Unit, Volume 4 Talmud-Based Solutions to the Problem of the Agunah Avishalom Westreich Deborah Charles Publications 2012 Copyright © 2012 Deborah Charles Publications All rights reserved. No portion of this publication may be duplicated in any way without the expressed written consent of the publisher, except in the form of brief excerpts or quotations for the purpose of review. ISBN 978-1-906731-20-5 (hardback) Published and Distributed by: Deborah Charles Publications On behalf of The Agunah Research Unit, University of Manchester E-mail: dcp@legaltheory.demon.co.uk http://www.deborahcharles.co.uk Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne. This is the fourth volume of a five-volume series, by each of the members of the Agunah Research Unit: Vol.1: Bernard S. Jackson, Agunah: The Manchester Analysis (Agunah Research Unit, vol.1, based on the Working papers of Yehudah Abel, Nechama Hadari, Shoshana Knol, Bernard S. Jackson and Avishalom Westreich) Vol.2: Yehudah Abel, Confronting ‘Iggun Vol.3: Shoshana Knol, Agunah and Ideology Vol.4: Avishalom Westreich, Talmud-Based Solutions to the Problem of the Agunah Vol.5: Nechama Hadari, The Kosher Get: A Halakhic Story of Divorce Ordering details, for one or more volumes, may be found at: http://www.legaltheory.demon.co.uk/ARU.htm Acknowledgements The problem of agunot – the chained wives, whose husbands refuse to divorce and thus cannot remarry – has occupied a wide range of scholars from antiquity to modern days: from talmudic sages to modern day scholars; from rabbis to academics; from politicians to lay people: women, men and professional attorneys. All are concerned with, and seek to find a solution to, one of the major challenges to Jewish Law and Jewish society in both the past and the present. This book, devoted to discussing the modern agunah problem and its Talmud based solutions, is one link in this chain. The book is an outcome of several years of research, mainly at the Agunah Research Unit of the University of Manchester, where I spent nearly two years. Some periods of time at the Shalom Hartman Institute and in the Hebrew University Law Faculty were also devoted to this research. My current home institution, The Academic Center of Law and Business in Ramat Gan, enabled the completion of this work. Prof. Bernard Jackson, the founder of the Agunah Research Unit of the University of Manchester, has directed it in the past eight years with faith and vision. Without his guidance, support and constructive involvement, both in content and structure, this book could not have been written and completed. His contribution to the international agunah issue in general, and to this work in particular, cannot be overstated. The company of teachers, colleagues, friends and students has made this work possible. Their thoughtful observations and useful comments assisted me in the long quest for understanding the sugyot discussed in this book. In particular, I would like to express my gratitude to Mordechai Akiva Friedman, David Henshke, Berachyahu Lifshitz, Leib Moskovitz, Amihay Radzyner, Avi Sagi, Pinhas Shifman and the members of the Agunah Research Unit: Yehudah Abel, Nechama Hadari and Shoshana Knol. I would also like to thank my seminar students at the Academic Center of Law and Business in Ramat Gan and at the Hebrew University of Jerusalem, with whom I have studied and discussed the complicated issues arising from the agunah problem, and to Edward Levin for his excellent assistance in translations and linguistic editing. Special thanks to my family: my parents, Sara and Elimelech Westreich, my children, Tamar, Aryeh, Shmuel and Bacol, and to my wife, Miri: !"#$!%&#$'!(#$ )'". Earlier writings of mine were the basis of this book with some necessary updates and modifications: four working papers, published as part of the Agunah Research Unit publications (ARU Working Papers 9, 10, 11 and 15, available at http://www.manchesterjewishstudies.org/publications), and a few related articles, published in several periodicals (cited as Westreich, Divorce on Demand; Gatekeepers; Hafka‘at Kiddushin, Moredet, The Right to Divorce and Umdena: for full details see Bibliography, infra). References to these and other relevant writings are provided throughout this book. Tamuz 5772 July 2012 Contents Introduction 1 Chapter One: The Rebellious Wife (Moredet): A Talmudic Source for Unilateral Divorce? 1.1 Introduction 1.2 Tannaitic and Amoraic Sources Regarding the Moredet 1.2.1 Tannaitic Sources: Would the Moredet receive a Get? 1.2.2 The Babylonian Sugya of Moredet 1.3 Conclusions 4 4 7 7 13 23 Chapter Two: Ketubbah Stipulations and the Rebellious Wife in the Palestinian Tradition 2.1 The Rebellious Wife in the Palestinian Talmud 2.2 Ketubbah Stipulations in the Post-Talmudic Palestinian Tradition 25 25 29 Chapter Three: Palestinian Ketubbah Stipulations and the Geonic Moredet 32 Chapter Four: Marriage Annulment: From Mishnah to Talmud 4.1 Hafka‘at Kiddushin 4.2 Talmudic Cases of Hafka‘at Kiddushin 4.3 The Foundation of Hafka‘at Kiddushin 4.4 The Character of Hafka‘at Kiddushin 39 39 41 45 53 Chapter Five: Marriage Annulment in Post-Talmudic Times 5.1 Introduction 5.2 Mere Annulment or Coercion? 60 60 61 Chapter Six: Hafka‘at Kiddushin in Practice! 6.1 The Various Talmudic Approaches as Bases for the Later Disputes: Does Hafka‘at Kiddushin Require a Get? 6.2 Marriage Annulment in Practice: Can hafka‘ah be applied today? 70! Chapter Seven: Mistaken Marriage, Conditional Marriage: The Talmudic Basis of an Innovative Approach 7.1 Introduction 7.2 Mistaken Marriage; Conditional Marriage 7.3 “And the Sages Did Not Have the Power to Release Her without a Get” (BT Ketubbot 74a) 70 76 78 78 80 82 7.4 7.5 7.6 7.7 Ada‘ata dehakhi lo kidshah nafshah (‘Umdena) – Mistake or Condition? 85 The Integrated Approach 92 Iggrot Moshe: an Application of the Integrated Approach 98 Conclusion 104 Chapter Eight: From Theory to Practice: Will and Ability 8.1 The Pluralistic Nature of the (Normal) Halakhic Discourse 8.2 Legitimizing Solutions to the Problem of ‘Iggun 106 106 108 Chapter Nine: Who is an Agunah? 9.1 Introduction 9.2 No-Fault Divorce? 9.3 The Right to Impose Conditions on Divorce 9.4 The Motivating Force: Restricting the Right to Divorce 111 111 112 117 119 Epilogue 124 Bibliography 125 Abbreviations B.K.: BT: EH: Ket.: M: PT: T: Bava Kamma Babylonian Talmud Even Ha‘Ezer Ketubbot Mishnah Palestinian Talmud (Talmud Yerushalmi) Tosefta Introduction Is there a solution to the Agunah problem? Divorce in Jewish law is executed by a writ of divorce (henceforth: a get) delivered by the husband to his wife. Typically, divorce requires both spouses’ consent: the husband’s agreement is fundamentally mandated by talmudic law (a desideratum that has been unanimously accepted from the talmudic era onwards), while that of the wife is required according to medieval rabbinical authorities (“the ban of Rabbenu Gershom”, which was originally adopted by Ashkenazi 1 communities, but today is accepted by most, if not all, Jewish communities). 2 The Mishnah lists several cases in which the husband is coerced to divorce his 3 wife. Several others are added by the Talmud and post-talmudic authorities. Usually, these are cases of faults, blemishes, and so on, which support the wife’s 4 entitlement to the divorce. But even when a formal fault does exist, it is not always possible to obtain divorce, due to the husband’s recalcitrance or incapability. Does Jewish Law provide us with routes for solving this kind of cases? More than that. Not rarely, one spouse demands divorce, but cannot prove one or more of the classic faults. So, is this spouse – usually the wife – entitled to divorce when no formal fault exists? And if she (or he) is entitled to divorce, how can it be executed, if her or his spouse refuses to participate? This book discusses several halakhic routes for executing unilateral divorce, or, when divorce cannot be obtained, routes for establishing a formal halakhic separation between the couple by-passing divorce, as follows: get compulsion in no-fault-divorce cases, mainly on the basis of the law of the rebellious wife (moredet); constitutive marriage annulment on the basis of the talmudic maxim: 1 2 3 4 On the ban of Rabbenu Gershom and its spreading among Jewish communities see E. Westreich, Temurot. M Ketubbot 7:10. See, e.g., Rema (Rabbi Moses Isserles), gloss to Shul!an Arukh, EH 154:1. In those cases the wife receives her financial rights (the ketubbah payment, i.e., the husband’s obligation of payment written in the marriage document, including the dowry, the wife’s prenuptial property, etc.). Similarly, there are cases in which the husband is entitled to divorce. According to talmudic law, prior to the ban of Rabbenu Gershom, the husband usually had a right unilaterally to divorce his wife, with or without a basis in fault, and with or without her consent. The effect of a fault or blemish in such cases was therefore mainly financial (i.e., the wife would not receive her ketubbah payment). After Rabbenu Gershom’s ban, consent for divorce, or fault or blemish on the part of the wife, became needed to establish the husband’s right to divorce and not only to support his financial claims. 2 Westreich: Talmud-Based Solutions to the Problem of the Agunah “everyone who betroths does so subject to the consent of the Rabbis, and the Rabbis annul his betrothal” (!"#$%&#'(!")*+!%,)#-'./%!$&'0!")*+&!/1-&/!$&'0&!(2 ,#)#0); declarative marriage annulment on the basis of a mistaken transaction; and conditional marriage, whether explicitly stipulated by the couple, or implied retroactively by the halakhic decisor (posek). These routes are rooted in the talmudic literature, and were discussed and used (with more or less hesitation) by post-talmudic decisors. Some post-talmudic traditions adopted one solutions or another for actual practice, although it is not always clear what was the halakhic basis for them. Amongst them are the geonic tradition in a case of a rebellious wife (moredet) and the divorce clause in the Palestinian tradition, as reflected in Genizah ketubbot. Both are rooted in the two Talmudim – the Palestinian Talmud and the Babylonian Talmud; I shall discuss the halakhic construction on which they were based, and analyze the talmudic justification for their use. Accordingly, the structure of this book is as follows: Chapter One discusses the basis for get compulsion in the case of the rebellious wife (moredet) in the Mishnah and Babylonian Talmud. Chapter Two discusses the law of the moredet in the Palestinian tradition. Chapter Three discusses the relations between the posttalmudic Palestinian Tradition and the geonic law of the rebellious wife. Chapter Four turns to discuss marriage annulment. First, I analyze the Talmudic basis for constitutive marriage annulment. Then, in Chapter Five, I ask whether it is possible that marriage annulment was used in practice in the above post-talmudic traditions, the Palestinian and the geonic. Chapter Six turns to the way posttalmudic commentators viewed marriage annulment, and examines the possible use of marriage annulment in modern times. Chapter Seven analyzes a related route: a declarative marriage annulment on the basis of either mistaken betrothal (kiddushei ta‘ut) or terminative condition. It then reveals a unique and innovative approach, which integrates both, and thus enables retroactive cancellation of marriage due to an unexpected future event on the basis of the wife’s claim that “on this assumption she did not get married”. The object of the book is to present the main solutions to the Agunah problem, and indicate their (stable) basis in the Talmud and from it to later poskim. Finding a solution however is not enough. We need a prior willingness for its use, but this does not always exist. Chapters Eight therefore discusses the pluralistic nature of the halakhah, which in principle enables the use of the above solutions. Chapter Nine tries to define who is an agunah, a definition which may well prove crucial for the willingness of halakhic authorities to solve the problem. This involves discussion of the ideological dispute which stands behind the modern agunah problem: what is the halakhic basis for the spouses’ right to demand divorce? If this right does not exist, we don’t have a problem at all, and apparently no solution Introduction 3 is required and justified. But if the problem does exist, and the situation is defined as one of ‘iggun, the halakhic gates are opened for using one or more of the solutions to the agunah problem discussed in this book. Chapter One The Rebellious Wife (Moredet): A Talmudic Source for Unilateral Divorce? 1.1 Introduction Jewish law approves divorce by agreement. There are however several causes for which classic Jewish law sources approve unilateral coerced divorce. For 5 example, the Mishnah states: And these [men] are coerced [to divorce]: one who is affected with boils [a type of leprosy], one who has a polypus [a growth in the nose], one who collects [dog excrement], one who is a coppersmith or a tanner, whether these [conditions] were in them before they married or whether they arose after they married. But would Jewish law recognize a more general right, that of no-fault divorce? The answer, perhaps surprisingly, is Yes. A possible talmudic basis for this right is the law of the rebellious wife (moredet), which according to some post-talmudic views enables coercion of a get upon the wife’s demand. The most important and well known among these posttalmudic views is the geonic tradition regarding coercion of a get in a case of 6 moredet. It is true that this and other such views encounter severe criticism from other post-talmudic scholars. However, coercion of a get in the case of the moredet has deep roots in both tannaitic and amoraic sources. These roots are the object of this chapter. In what follows the relevant talmudic sources are analyzed, arguing that the roots of the well-known geonic and other post-talmudic traditions are to be found already in the tannaitic and amoraic sources. Several parts of our argument are based on Rashi’s interpretation of the Babylonian sugya (discursive unit), which is compatible with the geonic tradition but more far reaching in the way it legitimates coercion on the basis of talmudic interpretation. Although it is in the very nature of this kind of source that alternative interpretations are possible, which in our case differ inter alia as regards coercion, the interpretation here suggested has important advantages compared to the others, and may be considered as the simple reading of 5 6 M Ketubbot 7:10. Sometimes the right to divorce is based on post-talmudic sources: see, e.g., Moses Isserles (Rema), gloss to Shul!an Arukh, EH 154:1. See Friedman, Jewish Marriage, 324-325; Brody, HaGeonim, 298-300. Chapter One: The Rebellious Wife (Moredet) 5 7 the sources, as will be argued at length below. The argument that follows has both historical and dogmatic importance: historical insofar as it supports the claims (i) that coercion was implied, even where it was not explicit, already in tannaitic and amoraic sources, and (ii) that the geonic measures regarding divorce compulsion were based on talmudic interpretation rather than an independent enactment; dogmatic insofar as it points to a body of 8 9 opinion (here exemplified in the interpretation of Rashi, his predecessors and 10 successors ) amongst the Rishonim which supports the view (despite that 11 attributed to Rabbenu Tam, which was accepted by the main halakhic authorities and largely followed since) that coercion was authorized by the Talmud in the case of the moredet. Some background: Rabbenu Tam’s main argument is that coercion could not be found in the Babylonian Talmud, and the Geonim had no authority to introduce it 12 as an enactment (takkanah). Indeed, many Rishonim agree that coercion is a geonic enactment (takkanat haGeonim), while they debate whether the Geonim 13 had the authority for it. However, Friedman and Brody maintain that the Geonim themselves regarded it as a talmudic law, based on the conclusion of the sugya: “we make her wait twelve months for her divorce” (lit. on her get: ! "#! $%&"'() 7 8 9 10 11 12 13 This chapter focuses on the hermeneutic considerations which are the basis of the dispute regarding coercion in a case of moredet. Nevertheless, this dispute also reflects different conceptual models of marriage and divorce in Jewish Law; see Broyde, Marriage, 15-28. Ascribing this view to Rashi is accepted by many commentators, both Rishonim (Smag, Lavin 81; Ritva, 63b, s.v. *+,)(! -&(+! &.&"; Hagahot Maymoniot, Ishut, 14: 6), and A!aronim (Pene Yehoshu‘a, 63b, end of s.v #1-!"2+1!*)/0)*1), as well as by academic researchers (E. Westreich and A. Grossman, see note 9 infra). Recent researches show that early Ashkenazic Rishonim, mainly Rabbenu Gershom Me’or Hagolah and probably a few of his students, accepted the geonic tradition of moredet: see E. Westreich, Rise and Decline, 211; Grossman, Pious and Rebellious, 242. Rashi, who followed these scholars (see E. Westreich, ibid.; Grossman, !asidot Umordot, 443 n.137 [this note does not appear in the English edition]), tried to base their tradition on the talmudic sugya. Indeed, in many aspects Rashi continues Perushey Magenza, which largely continues the tradition of R. Gershom and his students: see TaShma, Hasifrut Haparshanit, 35-56. However, Ta-Shma (ibid., 43) claims that Rashi normally focuses on hermeneutic rather than halakhic considerations in his commentary (this claim is disputed by halakhic writers; see Talmudic Encyclopedia, vol. 9, 337, entry “halakhah”).! For example: Rashbam, see E. Westreich, Rise and Decline, 212; Grossman, !asidot Umordot, 435436. See E. Westreich, Rise and Decline, 212-218. See Sefer Hayashar LeRabbenu Tam, !elek haTeshuvot, 24. There is a wide range of opinions on this point, from accepting the takkanah (Rif), through accepting it with limitations (Ba‘al Hama’or’s opinion of hora’at sha‘ah) and totally rejecting it (R. Tam). See E. Westreich, Rise and Decline, 212 and further. 6 Westreich: Talmud-Based Solutions to the Problem of the Agunah 14 !"#$!% !&'% #()#% )*#)&). This view was adopted by some Rishonim, who treat coercion as a Talmud-based law rather than a takkanat haGeonim. Amongst them 15 16 are Rambam and Rashi. In many cases new manuscript discoveries shed light on the text of the talmudic 17 sugya. In this context, MS Leningrad-Firkovitch explicitly supports the geonic tradition in that Amemar’s opinion in a case of moredet who claims #+,%*#!- (he is repulsive to me) is stated as .#+%/0#1##2, i.e. “we forced him”, which could hardly be 18 understood other than coercion of a get. However, this surprising support for the geonic tradition is not without problems. A preliminary reading of the talmudic sugya reveals an intensive discussion on the financial aspects of moredet with no explicit mention of a get. Thus, if we accept MS LF, we must ask why was it only 19 in the era of Amemar that the get became an issue for the amoraim? Interestingly, Rashi, although having the traditional text of Amemar, integrates into his interpretation of the sugya the rule that the husband must give a get, and 14 15 16 17 18 19 The effect of the geonic takkanah was therefore to coerce the husband to give a get immediately and not only after 12 months. See, Friedman, Jewish Marriage, 324-325; Brody, HaGeonim, 298-300. See also Ramban, s.v. #-+'3)#4% 30#5-3, who ascribes the view that coercion is the Talmud’s final conclusion to some responsa of Rav Sherira Gaon, but rejects it, arguing that: “this coercion has never occurred to the Sages of the Talmud”: “in some of the responsa of Rav Sherira Gaon, of blessed memory, too, I saw that he interpreted that, by strict law, [divorce] is not coerced. When we [i.e., the final talmudic conclusion] said that we delay the get for twelve months, [this implies] an additional [Talmudic] enactment, that afterwards the husband is coerced to issue a get. This has no substantial basis; clearly, there is no new enactment here to coerce, and this coercion has never occurred to the Sages of the Talmud”. (%;/#132%/#!%!0#:-:%9#1'%#&#!)%#-0%+86%!)#)'%4)+%&343'&%&57-43 %=#)4:+%/#!3% <"$4%+,4.%&!%/#132%28(!:%)-3+%&3.%#&#)(!%!&07&%!"#$!%#()#%)*#)&%.+%/0#.'-%30)-!'23 %36%.##12%:3-+&.%#-2(%&,:%+,%.&+,%!+%=+3,-3%>32+%.':(%.07&%/!2%/#!'%!3.%)3)4%)4:3% ;=3+2%)7#,%3++. =+3,+). See Rambam, Ishut, 14:8-14. See the discussion infra. MS Saint Petersburg National Library 187 EVR I. I use hereafter the previous and more common name, MS Leningrad-Firkovitch or MS LF. It is possible to suggest another explanation, following Ritva, 63b, s.v. /#*)3$'% '#3: when the husband wants to divorce his wife he cannot do it immediately without paying her ketubbah. He is therefore coerced (.#+% /0#1##2) to divorce her only after the mishnaic process of decreasing the ketubbah (on this process see infra). Our interpretation of .#+%/0#1##2 is however more reasonable. It is also supported by Rashba, 64a, s.v. 3.#-3: Rashba deals with the traditional text of Amemar, and argues that its meaning cannot be coercion, since the Talmud doesn’t mention the words .#+%/0#1##2: “all that was said is ‘but if she said he is repulsive to me she is not coerced’, But they did not say ‘he is coerced’”. (3)-!% !+% ?.#+% /0#1##28% +4!% ?.+% /0#1##2% !+% #+,% *#!-% .)-!% +4!8% !+!% /!2% 3)-!% !+'). Accordingly, MS LF’s text, which did mention .#+%/0#1##2, must be interpreted as coercion of a get (but see Me’iri, 63b, s.v. =#)4(-.%#+3:$3, who rejects the possibility of a variant like MS LF). Rabbi Shlomo Riskin regards the stage of Amemar as a turning point, claiming that in earlier generations the moredet was not interested in divorce: see Riskin, Divorce, 40-42. Nevertheless, divorce was a consequence of the tannaitic laws of moredet (Riskin, ibid., 17-18). Below I suggest a different view of coercion in the earlier sources. Chapter One: The Rebellious Wife (Moredet) 7 probably understands this as authorizing coercion (where necessary). In fact, although the sugya deals with financial aspects, Rashi mentions the existence of a get four (!) times, sometimes requiring that it be given immediately, elsewhere 20 later. Accordingly, MS LF’s variant of Amemar appears not to be the only proof for coercion. There appear to have been broader interpretative advantages supporting this view throughout the whole sugya, beyond the specific dispute in the memra of Amemar. These will be described in the sections that follow. 1.2 Tannaitic and Amoraic Sources Regarding the Moredet 1.2.1 Tannaitic Sources: Would the Moredet receive a Get? Our Mishnah says (M Ketubbot 5:7): If a wife rebels against her husband, her ketubbah may be reduced by seven denarii a week. R. Judah said: Seven tropaics. For how long does he reduce it? Until the ketubbah is exhausted. Rabbi Yoseh says: he may reduce it for ever in case she inherits property, from which he may claim it. According to the Mishnah, in a case of moredet the court act against her by decreasing the value of her ketubbah (the sum due to her on termination of the marriage, guaranteed in the marriage document) in a gradual process (conversely, the Mishnah adds, in a case of a rebellious husband – a mored – the value of the ketubbah is increased). The Tannaim in the Mishnah argue about the exact weekly sum and the limits of this process. According to both opinions in the Mishnah it is a long process. Take for example a basic ketubbah with a value of 200 zuz (or: 200 denarii): according to the first view in the Mishnah (Tanna Kamma), decreasing the ketubbah can take more than half a year (200 / 7 denarii in a week = 28.57); 21 according to Rabbi Yehuda more than a year; and according to R. Yoseh it can 22 take forever. In fact, it is a much longer process even according to Tanna Kamma and Rabbi Yehuda, if we take into consideration other property which according to 20 21 22 Rashi’s commentators usually point to s.v. !"#$%&'&&(#)" as a source for coercion in his commentary (see for example Resp. Maharam, Prague Print, 946, 135a), but in fact Rashi repeats it several times: (1) 63a s.v. !*+,*(#&-(#-. (Mishnah): /0#!"#$*,%#1(#23),; (2) 63b s.v. $&("4%: !/&0#*)#$&!54; (3) s.v. &(&!: )/&0#$&!54-; (4) s.v. !"#$%&'&&(#)": !+,*(#)"+#!)6,&,#/0#!"#$*,%#)")#7!*,!5!". Accordingly, he views coercion as an integral part of every section of the sugya, and not only of one single part of it. See also infra, text at notes 43-44, for further discussion of Rashi’s references to a get. A tropaic (8&.'2/) is half a dinar and the amount of decrease per week is thus 3.5 dinars, half that of the Tanna Kamma. 9",.", i.e. the court continues to remove it from her property as an “overdraft”, in case she might get an inheritance and her husband would be able to collect from it. 8 Westreich: Talmud-Based Solutions to the Problem of the Agunah 23 the Mishnah’s commentators was also subject to the mishnaic decrease process. The tannaitic attitude of reducing the ketubbah in a gradual process is unanimously agreed in the Mishnah (but disputed in its specific details). It was however changed in a later tannaitic generation. The later opinion is found in the Tosefta (Ketubbot 5:7), which comments on the Mishnah thus: “If a wife rebels against her husband” – this is the first Mishnah. But our Rabbis enacted 24 25 26 that a court should warn her four or five consecutive weeks, twice a week. [If she persists] more than that, even if her ketubbah is a hundred maneh, she has lost it all. According to the Tosefta, there is no gradual process, but rather only a few weeks 27 28 of warning, and after that the moredet loses her entire ketubbah. This enactment introduces the prospect of a defined and rapid end to this process. The difference between these two tannaitic approaches is that between an almost 23 24 25 26 27 28 The additions to the ketubbah, husband’s gifts, dowry (!"#$%&'() and #*+,%-./%&'()) etc. might also be reduced according to the mishnaic rule of “until the ketubbah is exhausted” (01,"1(%&2(%23). The exact belongings that are subject to this process are disputed by the Geonim and Rishonim. See for example Ramban, 63b, s.v. #"2!0% ")&,+% ,1("; Rashba, 63a, s.v. &1$% 23": “ ‘Until the ketubbah is exhausted’ – this means, the main body of the ketubbah, addition, and dowry, except for her usufruct property, that is not included in the ketubbah [...] And there is an interpretation [...] that even according to rabanan he detracts from the usufruct property that she had when she rebelled”. (‘% 23 %&$% 7&"% <<<0,"1(% ##(,% -)&.7% 0#7% !"#$% &'()$% :";% 9.&)"2)"% 16'"1"% 0,"1(% +8&3% 57"+&6% 5401,"1(% 2!)( 02+$7% 037,% 0#% "&07% !"#$% &'()$% 1;"6% -),+#% "#&6.2% =<<<>7+&67). For the geonic view regarding the monetary aspects of the original law of moredet, comparing their own takkanah, see Brody, HaGeonim, 300-303. The words “a court” (-&2%1&,) do not appear in the Erfurt MS or in the print edition. In Vine MS they were marked for deletion: see Lieberman, Tosefta Kifshuta, 266-267. These words might be an addition influenced by the Bavli (see next section) or by misinterpretation of -&2% 1&, in the Yerushalmi (see infra, n.36), which was corrected later. In many cases MS Vine is closer to the tradition of the Yerushalmi, while MS Erfurt is closer to the Bavli: see Friedman, Tosefta Atikta, 7986; Westreich, Torts, 107 n.23. Here -&2%1&, appears in MS Vine and this fact supports the option of Yerushalmi’s influence. On the other hand, the word “twice” (?&$36) in MS Vine is similar to the Bavli’s tradition (see infra, note 26 and note 58), so the issue cannot be conclusively determined. Some variants and Rishonim do not include the word “five” (7$;"): see Lieberman, Tosefta Kifshuta, 267. “Twice” (?&$36) is according to MS Vine, the print edition and Talmidey Rabbenu Yonah. According to MS Erfurt and some Rishonim (Tosafot, Smag and more), the text here is: 1;.%?36. See Lieberman, Tosefta Kifshuta, 266-268, who concludes without any decision. “Warn her” (0,%-&+1$) in the Tosefta is without any public humiliating announcement, significantly different from 0&#3% -&*&+($, which is found in the parallel Babylonian baraita (see Lieberman, Tosefta Kifshuta, 267; infra, note 29). Warning in the Tosefta may thus reflect a private warning, by a messenger for instance. This is the most widespread explanation of the Tosefta. For an exceptional approach which integrates a gradual process into Rabbotenu’s rule (as in the Mishna) see Me’iri, 63a, end of s.v. %+$. &+&.$0, in the name of ?$7,% ?0&2&$#1% "2&307% 0$% &6#% 1"+"20% &#"2!. Me’iri rejects that view, following the Yerushalmi, which mentions explicitly an act of losing the ketubbah: % 1+,"7% .&0" 0./"&"%01,"1(, see infra. Chapter One: The Rebellious Wife (Moredet) 9 endless story (“the first Mishnah”) on the one hand and a story with a clear and sharp end (“our Rabbis”; “Rabbotenu”) on the other. The goal of the first opinion is to lead the wife to end her “rebellion”, and the process is designed to influence her finally (and voluntarily) to change her mind. But this was probably not enough for Rabbotenu, who apparently adopted a more coercive approach, with the 29 prospect of a more immediate loss of the ketubbah, and is thus thought to operate more efficiently than the Mishnaic process. The cause for this move is not explicit in the Tosefta. I assume that the character of the rebellious wives changed between the Mishnah and Rabbotenu. Maybe it was a move from a domestic moredet (!"#$%%) to a moredet from sexual relationships (!&%!' ()%(*%), which forced Rabbotenu to enact a process which would be effective almost immediately. Those two kinds of moredet are mentioned 30 in a later generation in an amoraic dispute, which may reflect an earlier situation. However reasonable this explanation, we cannot prove it from the texts of the Mishnah and the Tosefta, nor from their context: while the previous Mishnah and previous Halakhah of the Tosefta deal with sexual relations, what follows deals 31 with financial aspects. Explaining the very goal of Rabbotenu as inducing the wife to end her “rebellion” by providing a more drastic sanction is problematic. The takkanah of Rabbotenu is indeed more radical than the Mishnaic rule. But assuming its goal is to induce the wife to finish her rebellion, is it really more efficient than the rule of 32 the Mishnah? According to Rabbotenu, loss of the ketubbah occurs in a single action against the wife, and then, after only four weeks, there are no further possible sanctions, whereas the mishnaic process envisages a long period of time in which the required impact can be created. Thus, if the goal was to influence the wife to return to her husband, it would be more practical to use a lesser sanction over a longer period of time, rather than using the maximal sanction almost 33 immediately, where the moredet has been “in rebellion” for only four weeks. 29 30 31 32 33 In the parallel Babylonian baraita there is an additional element, a process of public announcement, which is actually a process of humiliation designed to end the rebellion. However, the public humiliation does not exist in the Tosefta (see supra, note 27, and compare Tosafot, 63b, s.v. #+),) nor in the Yerushalmi (see infra), and therefore I refer here only to the financial aspects of Rabbotenu’s ruling. See BT Ketubbot 63a, and implicitly in the Yerushalmi: PT Ketubbot, 30b, 5:8, see Shitah Mekubetset, 63a, s.v. $-.'!-#/!'01($'!.1. See also Riskin, Divorce, 4-9, discussing the definition of moredet, and 12-14, explaining the cause for Rabbotenu’s rule as “the increasing number of rebellious wives”. The humiliation does not exist in the Tosefta, see supra, n.29.' The term “marginal deterrence” is used by modern researchers to describe a legal system which imposes different measures of sanctions for different kind of offences, in order to create an efficient 10 Westreich: Talmud-Based Solutions to the Problem of the Agunah 34 On this analysis, the rationale of Rabbotenu may not have been deterrence but rather something different, namely to put an end to the conflict as quickly as possible, whether by bringing the couple back together or by leading to a complete separation between them. This might be a response to a social change, as suggested above, according to which the phenomenon of moredet from sexual relationships became common, and, for that reason, Rabbotenu were forced to enact a process which would be effective almost immediately. In their enactment the choice was put into the wife’s hand: preferably she may decide to withdraw her rebellion; however, if she insists, she is entitled to a divorce, but must forfeit her ketubbah. In this case the husband is compelled to give a get (otherwise the wife’s entitlement 35 has no meaning), by a physical coercion if required. This explanation is supported by the Yerushalmi. Another version of the baraita appears in the Yerushalmi and introduces Rabbotenu as follow (PT Ketubbot, 5:10, 30b): 36 The later court [enacted that we] warn her for four weeks (after which) she cancel her 37 ketubbah debt and leave. Comparing the Yerushalmi and the Tosefta, there is a significant addition in the Yerushalmi: “and leave” (!"#$%$), which means that after losing the ketubbah she ---deterrence for each offence. Accordingly, the sanction would be enhanced as a function of the severity of the offence, the number and extent of offences etc. (see George J. Stigler, “The Optimum Enforcement of Laws”, The Journal of Political Economy 78 (1970), 527-528: “If the offender will be executed for a minor assault and for a murder, there is no marginal deterrence to murder”; see also Steven Shavell, “A Note on Marginal Deterrence”, International Review of Law and Economics 12 (1992), 345; 351-352). On this view, it is more efficient to use a low sanction for a rebellion which is at its beginning and to enhance the sanction when the rebellion becomes more severe, rather than using the higher level sanction each time. & 34 35 36 37 It should be emphasized that this explanation is a suggested rationale of Rabbotenu’s view, rather than a proof, on the basis of the modern view regarding deterrence. The proof for this rationale is an internal proof, on the basis of Rabbotenu’s variant cited in the Palestinian Talmud. See Rambam, Ishut, 14: 8: $'()*&"%#$!*&$'$"&+%,$-. Rambam refers to moredet ma’is alay, but the halakhic implications of the different kinds of moredet are in regard to financial aspects and questions of timing, as discussed at length below, rather than regarding the character of coercion: see for example below, text at notes 89-93. The same applies to the mored, coercion of whom is compared to coercion in a case of moredet: see infra, notes 60-63. '$'.)& (./"& !.& +%/%'0& 1+!%/2")& +%3&'%.. The words +!%/2")& +%3&'%. does not refer to the warning (i.e. warning is done by the later bet din) since the word +!%/2") has no meaning according to this interpretation. It refers to the whole enactment, while the predicate of this sentence (“enacted”) is missing. Its meaning is the same as in the other two parallels of this baraita: $4%5'!&$4%'$./ (Tosefta) or $404$&$/62&$4%'$./ (Bavli).& !"#$%$&!'.$'-&'/.$)&"%!$. “Shoveret” means “writes a receipt” (shovar) for her ketubbah (see BT Sotah, 7a; Epstein, Mavo LeNusa! HaMishnah, 616. A parallel term in Tosefta, Ketubbot, 9:1, is clearer: !'.$'-& *(& $*& '/.$)), acknowledging that she received her ketubbah payments, or, more accurately, canceled her husband’s debt. Chapter One: The Rebellious Wife (Moredet) 11 38 receives a get. According to the first explanation of Rabbotenu above, which focuses on deterrence, this addition is not explained. But according to the second suggested interpretation it is meaningful. !"#$%$ is the completion of the whole process, representing its current goal: to lead to a separation between the couple. Get, according to this explanation, is a necessary condition for ending the story and 39 thus an integral part of Rabbotenu’s teaching. The version of Rabbotenu in the 40 Yerushalmi thus sheds light on their goal and rationale in the Tosefta. The view of Rabbotenu here suggested accords with Rashi’s interpretation of the Babylonian sugya. As argued below, Rashi creates a parallelism between all sections of the sugya. Accordingly, Amemar’s view that “we don’t force her” (&"' !'& ()%*%%+), or “we forced him” (!%'& ()%*%%+) as in MS Leningrad-Firkovitch, is parallel to Rabbotenu’s enactment. It is hard to understand it as a rule whose goal is to increase the deterrence on the moredet, since the words “we don’t force her” 41 are in her favour. We can conclude therefore that the goal of Rabbotenu in Rashi’s view is not to deter the moredet but rather to bring a quick end to the conflict – here by accepting the wife’s demand for divorce (after trying to convince her, even by public humiliation, as in the Bavli) and not forcing her to stay with her 42 husband. In the next section however, I discuss Rashi’s view further and point out its advantages as against other possible interpretations. The object of the law of the rebellious wife according to Rabbotenu is to put an end the conflict, either by bringing the couple back together or by a complete separation between them. According to the Mishnah, on the other hand, the object is to induce the wife to end her rebellion by decreasing the value of her ketubbah in a gradual process. What then shall we do after the moredet has lost her ketubbah? 38 39 40 41 42 See Friedman, Jewish Marriage, 322. There are also two other important differences between the Yerushalmi and the Tosefta: (i) Rabbotenu are replaced by “the later bet din” (see supra, n.24 and n.36); (ii) there is no mention of twice a week (see supra, note 26, and infra, text at note 58). This view differs from Riskin’s suggestion (Divorce, 17-18) that divorce in the tannaitic stratum is a final penalty for the wife, who really does not want it.& A quite different reading is possible: Rabbotenu of the Tosefta still had the object of coercing the wife back into the marriage, as in the Mishnah, but by a sharper financial sanction. Only in the Yerushalmi has the goal changed and become the quest for ending the conflict, “one way or the other”, and this is reflected by the addition: !"#$%$. However, I prefer the explanation suggested above. A revolution in our case is mentioned explicitly in the move from the Mishnah to Rabbotenu, while the alternative reading finds a more radical distinction between the Tosefta and the Yerushalmi. In fact, the Tosefta, Yerushalmi and the Babylonian baraita are three parallel versions of one turning point; these sources differ from one another in a few elements but not in essence. It is less likely that there was a significant but implicit change between the Yerushalmi and the Tosefta, while the sources explicitly point to Rabbotenu as the turning point of this sugya. The same argument is found in Ra’avad’s interpretation of !,&(%+'-): see infra, text at note 64. See Rashi, Ketubbot, 63b, s.v. !'&()%*%%+&"'. 12 Westreich: Talmud-Based Solutions to the Problem of the Agunah At this moment there are no more sanctions against the wife and so she is not likely to agree to go back to her husband. Would she receive a get? Rashi takes his view regarding compulsion of a get a step further. He suggests that after losing the ketubbah the wife receives a get both in the mishnaic rule and in that of Rabbotenu, albeit the get is not an essential part of the mishnaic rule of 43 moredet, contrary to Rabbotenu’s rule. As noted above, he repeats this point four times: (1) According to Tanna Kamma, the mishnaic gradual process of decreasing the ketubbah is until the ketubbah is exhausted. Rashi (63a, s.v. !"#$"%&'(%&()) adds here: !#$"%&*+#&!*,$'$&-.&!+&/"$0&1%&23*$, i.e. afterwards he gives her a get and she goes out (i.e., divorced) without receiving her ketubbah; (2) On 63b Rashi interprets !#&/'%+40 (she is to be consulted) as: &!-'.&"*&/'!54 !#&2$63"5&!'+)&/'2'634$ (we hold back her get and try to make her change her mind). (3) A similar interpretation is given for "(2$4& *'4(& '%'! (what is to be understood by “a rebellious woman”?): & /'"3$7$& !-'.& /'!54(& 8!"$*& /'7$%( !"#$"%, i.e. we force her [by] holding back her get and reducing the ketubbah. Quotations (2) and (3) are discussed at length below. For the moment, we may already draw two important conclusions: (a) the mishnaic gradual process of reducing the ketubbah does not deny a get but only postpones it, and after this process is ended she will receive a get, as mentioned explicitly in quotation (1); (b) Without the mishnaic gradual process (for example: according to Rabbotenu of the Tosefta), her get is not delayed, but she receives it immediately (or: after four weeks). This conclusion is explicit in the fourth appearance of get in Rashi’s commentary: (4) According to Amemar, moredet ma’is alay (see at length below) is not regarded as a moredet, but !+& /0'7''%& *+, i.e. we do not force her, or: no pressure is to be brought to bear upon her. This is interpreted by Rashi as follows: !#$"%&*+#&!*,$'$&-.&!+&/"$0&*+*&8!"$!5!+&!+&/0'7''%&*+, i.e. we do not hold her back, but he gives her a get and she is to be divorced without receiving her ketubbah. It is hard to understand these four repetitions as no more than a description of a 44 contingent event, which occurs only when the husband is willing to grant the get. It is much more plausible to understand it, following Rashi, as an integral part of 43 44 See supra, note 20.& Some Rishonim explained Rashi in this way. See Tosafot, 63b-64a, s.v. +#*; Rashba, 63b, s.v. &'%'! "(2$4&*'4( (Rashba quoted Rashi that the husband gives a get, but with an addition: “if he wants to”). But other Rishonim and A!aronim explained Rashi as here suggested: see supra, note 8. Chapter One: The Rebellious Wife (Moredet) 13 the halakhic rules of moredet. Rashi thus appears to endorse the view that the marital dispute must not remain static, without any movement towards a solution, and therefore that after loss of the entire ketubbah the husband is coerced to give a get. It is also possible that a broader consideration is involved here: a rule ascribed in the Bavli to Rabbi Meir 45 requires a ketubbah to be in existence. The result of accepting this rule is that a 46 get must be given after total loss of the ketubbah. Anyway, receiving a get is a required stage both according to Rabbotenu and according to the Mishnah, after the end of the process of losing the ketubbah debt. Supporting this view by Rashi’s interpretation has a dogmatic importance, by pointing to later opinion (i.e. Rashi’s view) which supports the view that coercion was authorized by the Talmud in the case of the moredet. But as an interpretation of the sources, can coercion be considered as their simple reading (the peshat)? I would argue that although it is not explicit, it is a reasonable explanation. Moreover, it has a significant advantage. It provides us with a harmonious view, which ties together systematically the tannaitic and amoraic sources. The talmudic sugya has a logical structure, in which two different options (already apparent in the tannaitic sources) are in tension in each of its sections, and where coercion is an integral issue throughout. Nevertheless, this systematic and logical structure, most clearly elaborated by Rashi, is opposed by competing interpretations of the sugya (notably that of Rabbenu Tam). We now turn to the talmudic sugya and its interpretation by the Rishonim. 1.2.2 The Babylonian Sugya of Moredet This section explores the Babylonian sugya of moredet, focusing on Rashi’s 47 interpretation compared to that of his opponents. Our analysis indicates Rashi’s advantages in every section of the sugya. However, its persuasiveness is attained by introducing a harmonious and systematic structure into the sugya as a whole. This fascinating structure, based on our previous conclusions (which found coercion already in the tannaitic sources), enables us to find coercion attributed in 48 amoraic sources as well as in the anonymous late talmudic stratum. 45 46 47 48 See BT, Bava Kamma, 89a (ascribing this view to R. Meir, on Mishnah, Ketubbot 5:1): “It is prohibited for any man to keep his wife without a ketubbah even for one hour. But what is the reason of this? So that it should not be an easy matter in his eyes to divorce her”. See Riskin, Divorce, 18. ! Palestinian Talmud (Yerushalmi) is discussed in the next chapter. I define here “amoraic” as attributed rather than anonymous sources. This distinction is significant here since the final development of the talmudic law of moredet is found in an anonymous stratum which belongs to the last generations of the Babylonian Amoraim or may even be a saboraic 14 Westreich: Talmud-Based Solutions to the Problem of the Agunah The first part of the sugya deals with the content of the rebellion, i.e. whether it is a rebellion regarding sexual relationships (!"#!$%&#%'#) or regarding domestic 49 duties (!()*##). The next part of the sugya is composed of a number of sections. At its beginning, the sugya cites a baraita (a tannaitic source other than the 50 Mishnah), which parallels the Tosefta with a few changes. The sugya continues as follows (BT Ketubbot, 63b): (a) [To turn to] the main text. If a wife rebels against her husband, her ketubbah may be reduced by seven denarii a week. R. Judah said: Seven tropaics. Our Rabbis (+,&'+-.), however, revised [their views] [and ordained] that an announcement regarding her shall be made on four consecutive Sabbaths and that then the court shall send her [the following warning]: ‘Be it known to you that even if your 51 ketubbah is for a hundred maneh you have forfeited it’… (b) Rami b. Hamma stated: The announcement concerning her is made only in the synagogues and the houses of study. Said Rava: This may be proved by a deduction, it having been taught, ‘Four Sabbaths consecutively’. This is decisive. Rami b. Hamma further stated: [The warning] is sent to her from the court twice, once before the announcement and once after the announcement. (c) R. Na!man b. R. !isda stated in his discourse: The law is in accordance with our Rabbis (+,&'+-.($!(*!). Rava remarked: This is senseless. Said R. Na!man b. Isaac to him, ‘Wherein lies its senselessness? I, in fact, told it to him, and it was in the name of a great man that I told it to him. And who is it? R. Jose the son of R. !anina!’ (d) Whose view then is he following? – [The first of the undermentioned:] For it was stated: Rava said in the name of R. Shesheth, ‘The law is that she is to be consulted’ (!-$/&(*#,), while R. Huna b. Judah stated in the name of R. Shesheth, ‘The law is that she is not to be consulted’ (!-$/&(*#,$/&)). The sugya introduces at (a) the dispute between the Mishnah and Rabbotenu. At (b) there are some clarifications about the procedure of announcing (hakhrazah). Then (c) the sugya cites R. Na!man b. R. !isda, who follows Rabbotenu in the ---passage; see infra, text at notes 75-77. The distinction between attributed sources and anonymous sources has a general importance in talmudic research: see Friedman, Ha’isha Rabbah. Friedman’s view is criticized by Brody, Stam HaTalmud, 220-224. However, Brody’s main criticism is chronological rather than in regard to the basic distinction between attributed and anonymous sources: see Westreich, Torts, 14-15. 49 50 51 See supra, text at nn.29-31. See supra, n.29. Additional changes are discussed below. The baraita states here that the same law is applicable to a woman betrothed or married, even to a menstruant, sick, or a woman “waiting for levirate”. The case of the menstruant is then discussed between R. !iyya bar Yosef and Shmuel. This discussion is a comment on the baraita and not part of the progression of the sugya. For that reason, I do not define it as a separate section. Chapter One: The Rebellious Wife (Moredet) 15 baraita. The phrase “the law is in accordance with” (%%%!"#!$#) which is used by R. Na!man bar Rav !isda usually means that the law is determined according to somebody and not according to his opponent, and here the main controversy is between Rabbotenu and the Mishnah. Thus Rav Na!man bar Rav !isda follows the rule of Rabbotenu, while Rava, who condemns this view as “senseless”, follows the Mishnah. Section (d) explains the authority for Rava’s decision. Accordingly, the dispute between “she is to be consulted” (#&" '(!$)*) and “she is not to be consulted” (#&" '(!$)*" '(+) is parallel to the earlier tannaitic dispute between the Mishnah and Rabbotenu. The verb '(!$)* means trying to convince the wife to change her mind. This is done by the mishnaic gradual process of reduction of the ketubbah. On the other hand, '(!$)*"'(+ means that the court does not use this process of convincing her, but the wife loses her ketubbah at once, and, as Rashi adds in his commentary, receives a get (similar to our conclusion above in regard to the Tosefta). 52 This explanation follows Rashi’s interpretation. Its simplicity and clarity are discernible, but it was not accepted by many Rishonim, including Rabbenu Tam. I assume that Rabbenu Tam’s objection to Rashi’s interpretation of these sections is not only as a result of local interpretative considerations. Rashi’s interpretation takes a harmonious view of the complete sugya, whose conclusion is the need for coercion, as will be shown hereafter. It was therefore necessary for Rabbenu Tam 53 to suggest different interpretations of almost every section of the sugya. 54 Rabbenu Tam suggests a different interpretation of the sugya. According to him, Rami bar Hamma in section (b) makes some additions to Rabbotenu. Rav Na!man bar Rav !isda’s determination of the law following Rabbotenu at section (c) does not refer to the tannaitic dispute. His statement opposes Rami bar Hamma by accepting the original law of Rabbotenu without any additions. This last view was condemned by Rava as “senseless”. This interpretation is problematic. Rami bar Hamma does not argue with Rabbotenu. I would even say that Rami bar Hamma does not even make additions 52 53 54 See Rashi, 63b, s.v. #&"'(!$)*: “we hold back her get and try to make her change her mind, and in the mean time we reduce her ketubbah by seven denarii a week” (" #($," '(-(./)0" #1(2" 3+" '(#4) 3&4&"'(-*(5"#,&4"#3&03!)"'(3/06"7!"703&0"8#&"-0./34). '(!$)* is thus exactly the mishnaic rule, and consequently this is Rava’s opinion. The dispute between Rav Na!man bar Rav !isda and Rava is therefore between determining the law in accordance with Rabbotenu and in accordance with the Mishnah (see Tosafot, 63b, s.v. 0#*(+0). Riskin, Divorce, 38-40, accepts Rabbenu Tam and rejects Rashi’s interpretation. Riskin’s main argument is that according to Rashi, Rava on section (c) rejects Rabbotenu, but on section (b) supports Rami b. Hamma’s interpretation of Rabbotenu’s teaching. This argument can easily be met by viewing Rava as interpreting Rabbotenu without determining the law according to them. See Tosafot, s.v. 0#*(+0. A few more interpretations will be mentioned below. 16 Westreich: Talmud-Based Solutions to the Problem of the Agunah to Rabbotenu, as described by Tosafot, but only interprets them: the baraita mentions two verbs (!"#"$%& and !")*'(), and Rami bar Hamma’s two statements refer respectively to these verbs, integrating them into one judicial process. The first statement describes how the public announcement is made, while the second exposes the timings of sending the messages for the wife, which are before and after the public announcement. Viewing Rami bar Hamma’s statements as an interpretation of Rabbotenu enables us to ascribe a more complicated object to him: to integrate the two 55 traditions of Rabbotenu, that of the Babylonian baraita and that of the Tosefta. Whereas his first statement, which mentions public announcement in synagogues and in batei midrashot, reflects merely the Babylonian baraita, his second statement, which deals with personal warning (!")*'(), reflects the concept of the 56 Tosefta (+,- !"$.& ) as well. Thus, presenting the public announcement and the personal warning as two parts of one process denies any possible disagreement 57 between these two sources. The variant .,(,-/""&01 of the Tosefta makes the link between Rami bar Hamma’s second statement and the Tosefta more stable and 58 explicit. Accepting this variant as the original text of the Tosefta increases the meaning of the integration between the Babylonian baraita and the Tosefta and makes it reciprocal: the baraita contributes the aspect of public announcement, 59 while the Tosefta contributes the number of personal warnings. In brief, interpreting “the law is in accordance with Rabbotenu” as rejecting Rami bar Hamma, who is understood as an opponent of the original meaning of Rabbotenu, while ignoring the Tosefta, as Tosafot suggest, is less likely. In fact, Rami bar Hamma does not oppose Rabbotenu; thus determining the law according 55 56 57 58 59 Compare Riskin, Divorce, 15-16, who sees the baraita as a result of a later redactor’s work, in order to make the Tosefta consonant with Rami bar Hamma’s rule of announcement.See supra, n.27. This is the variant of Vine MS and others: see supra, n.26. One could argue that this variant is a correction of the original text, influenced by the Bavli’s tradition of Rami bar Hamma and is not the source for his teaching: see Lieberman, Tosefta Kifshuta, 268. Nevertheless, if this were the case, we would expect to find it as MS Erfurt’s variant, which was more influenced by the Bavli than MS Vine (see supra, n.24). I prefer therefore to view this variant as the original basis for Rami bar Hamma and not as a consequence of his teaching (as to MS Erfurt: see Tosafot, 63b, s.v. 23"4). Accordingly, “[The warning] is sent to her from the court twice” in Rami bar Hamma’s second statement refers exclusively to the Tosefta. See Talmidey Rabbenu Yonah, in Shitah Mekubetset, 63b, s.v. ".(: “Rami bar Abba [variant reading: Rami bar Hama] came to interpret the Tosefta, saying, do not think that in these two times, both precede the public announcement, or that both follow the public announcement. Rather, one precedes the public announcement, and the other follows it” (-'2-+#$%+-/4'3-!+".(-/"&01-".(+-'*2(-,'().-2*(-$&'*'-.'15'.+-($1*-2,-2,2-$,-"&$' +#$%+-$)2-.)2'-+#$%+-/4'3-.)2-2*2-6+#$%+-$)2-!+".(). Chapter One: The Rebellious Wife (Moredet) 17 to Rabbotenu by Rav Na!man b. R. !isda has no implication for Rami bar Hamma’s own statements. Rashi’s interpretation of “the law is in accordance with” (%%%!"#!$#) as referring to the dispute between Rabbotenu and the Mishnah is much more probable. There is another advantage to Rashi’s interpretation. According to Tosafot, “she is to be consulted” (&'!$()) at (d) supports Rami bar Hamma’s teaching, and its meaning is that the court sends messages to the wife both before the announcement 60 and afterwards. But the verb (&'!$(); in a different inflection) appears earlier, in the first part of the sugya (on 63a) and its meaning there is totally different. At 63a the sugya deals with aspects of mored (a rebellious husband). One kind of mored is a financially rebellious husband, who refuses to support his wife and according to Rav is coerced to divorce her and pay the ketubbah ("*'+,'"-)./("')'*,"&0"')'*".(,*# 61 #2,1!"&1',). According to the mishnaic rule of mored we increase the value of the ketubbah. The Talmud then confronts this gradual process with Rav’s immediate rule of *'+,' and answers: “isn’t it required to consult him?!” (" #'2" '!,$(*$" ,*$, 62 '32?!), i.e. before the mored is coerced to give a get we try to convince him to reconsider his rebellion by increasing the ketubbah. '!,$(*$ here is far from Tosafot’s interpretation of &'!$() in the context of moredet. Rashi on the other hand follows the simple meaning of the sources, and interprets the verb 63 consistently (almost word by word) in its two appearances. Perhaps the similarity between those two led Rashi to interpret “she is to be consulted” in the sugya of moredet as including coercion, just as is in the sugya of mored. However, there are 60 61 62 63 Accordingly, &'!$() can also be accepted by Rabbotenu. Following this view, many Rishonim (Rif and others probably even earlier than Rif: see Halakhot Gedolot, 36, s.v. 14.,(#) determined the halakhah both according to Rabbotenu and &'!$(), while according to Rashi these are conflicting approaches (see Tosafot, ibid.; Rashba, s.v. 5', and more). Tosafot connect these different interpretations to different variants of the talmudic text at section (d): .2-"&*(!",#'*, according to Rashi (also in most MSS), which refers to Rava at section (c), who follows the Mishnah, or ",#)'*, #,.2-"&*(! (this variant appears in MS Munich 95) according to Tosafot, which refers to Rami bar Hamma and Rava, who according to Rabbenu Tam both follow Rabbotenu. The ruling *'+,' (he should divorce her) has the same meaning as *'+,#$" ,1,*" &'/,! ([the court] coerce him to divorce her): see Shmuel’s response to Rav (whose ruling is *'+,'), Ketubbot, 77a: "43 &,0$",#,/!'"*'+,#$" ,1,*"&'/,!5, i.e. rather than coerce him to divorce her (as Rav claims) let him be coerced to maintain her. See also Friedman, Divorce, 103-104. Some Rishonim (e.g. Rabbenu Hananel, and see also Tosafot, Yevamot, 64a, s.v. *'+,') dispute this, and argue that when &'/,! is not mentioned, physical coercion is not permitted and may lead to an invalid coerced divorce (get me‘useh). Their opinion led the later halakhah to distinguish between coercion of a get (#''/!) and obligation (2,'6) without physical coercion; see Shul!an Arukh and Rema, EH 154:21. But see different variants of this sentence, infra, at n.67. Rashi interprets '32"#'2"'!,$(*$",*$,?! as follows: ",'$3"&'.'06(,",2"&'!$()",)*5"&(0"8,12,"7,2".,06'"*(5 #12,1!"$3" &'/'-,(" ,2".,06'5. Rashi suggests the same interpretation, almost word by word, of "&'!$() #2 in a case of moredet: "&'.)'4"#325"#12,1!("&'16,/"8!"8,12,"#2".,0615"#'$3"&'.'06(,"#9':"1*"&'#5( 1252. 18 Westreich: Talmud-Based Solutions to the Problem of the Agunah broader considerations for Rashi, as discussed in this chapter. It should be mentioned that Rashi’s and Tosafot’s interpretations are not the only ones. A third meaning of “she is to be consulted” (!"#$%&) is suggested by 64 Ra’avad: the moredet has the option to choose either the mishnaic rule of 65 decreasing the ketubbah or the rule of Rabbotenu. It is difficult to explain, according to Ra’avad, what would cause the moredet to choose the rule of Rabbotenu and lose immediately all her ketubbah, especially if we follow those 66 opinions which deny coercion of a get in this case. The advantage of Ra’avad’s interpretation is that we can interpret consulting in 63a (!"#$%&) and in 63b 67 ("#'$%($) in a similar way, although it depends on questions of text. Yet, it is not explained why the choice either in mored or in moredet is given to the wife. Rashi on the other hand is systematic also on this point: in a case of mored he is to be consulted, while in the case of moredet she is to be consulted (and, as already mentioned, consulting has the same meaning in both cases). Rashi’s interpretation, 68 here again, is thus more simple and reasonable. We now turn back to the progress of the sugya of moredet. The sugya continues as follows: (e) What is to be understood by ‘a rebellious woman’? Amemar said: [One] who says, ‘I like him; but wish to torment him’ ()"$*(&+,-%'*)"$*(&",.). If she said, however, ‘He is repulsive to me’ ("$,*/"(%), no pressure is to be brought to bear upon her (lit. we do not coerce her). Mar Zutra ruled: Pressure is to be brought to bear upon her (lit. 64 65 66 67 68 Ra’avad agrees with Rashi that Rav Na!man, who determines the halakhah in accordance with Rabbotenu, means not according to the Mishnah (see Ritva, 63b, s.v. (.+*+%(). His argument is on the meaning of !"#$%& (see infra) and accordingly concerns the interpretation of Rava’s exact opinion. See Ramban, 63b, s.v. (). See Ritva’s explanation (63b, s.v. $01*20.(+)'). At 63a (the case of mored) the Rishonim (see for example Rashba, s.v. !&"3+4') introduce another variant: ",.*).*"#'$%($*'($' (compare the traditional text, supra, text at note 62: ",.*)".*"#'$%($*'($') which means that the wife is given a choice between two halakhic options, similar to the interpretation of this phrase in 63b (but on 63a, since it is a case of mored, her choice is between immediate divorce while receiving her current ketubbah and delaying the divorce but increasing the ketubbah, whereas on 63b her choice is between divorce without ketubbah and decreasing the ketubbah). It should be remarked that most MSS take the traditional text, despite MS Vatican 130, whose original text was ). but was corrected above the line to the traditional text. Interestingly, according to Shitah Mekubetset, ). was Rashi’s text in the first edition of his commentary (see 63a, s.v. '($'), and it is also the text of “Rosh and all of A!aronim” (ibid., end of s.v. .+*+%()'). I am therefore still doubtful whether Rashi in his last edition chose his text because of its advantages, or maybe this text is a result of his correction of the talmudic text, which was done in order to make his interpretation consistent with those two parts of the sugya. There is at least one more explanation of this passage: see Rashba’s explanation of Rif (63b, s.v. 5"'), which seems to integrate Rashi and Rabbenu Tam.* Chapter One: The Rebellious Wife (Moredet) 19 we coerce her). Such a case once occurred, and Mar Zutra exercised pressure upon the woman and R. !anina of Sura was born from the re-union. This, however, was not [the right thing to do]. [The successful result] was due to the help of providence. Like earlier stages, the dispute in section (e) continues the basic tension of the sugya. Basically, both Amemar and Mar Zutra follow the Mishnah regarding moredet. As Rashi interprets it, the law of moredet involves forcing her by making her wait for her get (again, an interpretive addition of Rashi) and decreasing her 69 ketubbah, exactly the mishnaic rule. Amemar and Mar Zutra agree to apply this law when the wife claims “I like my husband but wish to torment him” (!"#$!%&#'( "#$!%&)'*+,), but disagree in applying it to a case of “he is repulsive to me” (! -#%+ #$'). According to Amemar, in this case we should not follow the mishnaic rule of moredet. Thus, the alternative option from earlier stages of the sugya arises: the rule of the Tosefta. Rashi therefore interprets the ruling as: “we don’t force her to remain under her husband, but he (must) give her a get while she loses her 70 ketubbah.” Coercing a get is not unique to Amemar’s teaching. It is part of an entire approach, whose roots are much earlier, in Rabbotenu of the Tosefta or even in the Mishnah. This approach is to be found, according to Rashi, at each section of the sugya, which consistently opposes the two approaches. However, at section (e), following Rashi’s interpretation, we have MS LF which raises explicitly the rule of 71 coercion. One comment should be made here. As we have seen, the dispute regarding “he is repulsive to me” at section (e) is equivalent to the dispute between the Mishnah and Rabbotenu, and to the dispute between “she is to be consulted” and “she is not to be consulted” at section (d). However, the rhetoric is quite different. “Consulting” at section (d) has a positive orientation, probably from the viewpoint of the husband or bet din. “Pressure” (or even coercion) at section (e), on the other hand, although having the same meaning, has a negative orientation, and probably 72 reflects the viewpoint of the wife. This fact may reflect diverse conceptions of different generations or of the sources of each part of the complete sugya. 69 70 71 72 See Rashi, 63b, s.v. #.#": "0(,0.!2#07,3,!"5#6!2#"4+/!"0,%!2#3,./!10/),+!%#+/!#.#". Rashi, s.v. %$: "(,0.! %$(! "%*,#,! 56! "$! 20,&! %$%! 8"0,"4"$! 1"$! 2&#3##.! %$. Rashi does not refer explicitly to the question whether it is done after announcement, as it is in the baraita, or immediately. Since the other elements of this rule are similar to the baraita, it is a reasonable understanding to apply the missing element (the announcement) here too.! See supra, text at notes 17-20. MS LF follows Rashi also in section (d): ")(-!2%+.!,"#%,: see supra, n.60, although there it is not unique – at that point the majority of MSS follow Rashi. See Riskin, Divorce, 41.! 20 Westreich: Talmud-Based Solutions to the Problem of the Agunah However, the redactor of the sugya integrated them into one complete sugya, 73 reflecting in fact the same halakhic approach. Following him, Rashi, as we have shown, interprets it in a clear and harmonious way. The Rishonim who rejected coercion opposed Rashi also in their interpretation of (e), sometimes in too complicated a way. Thus, for Rabbenu Tam, it is not completely clear whether the pressure upon the moredet who claims “I like him but wish to torment him” is according to the Mishnah or Rabbotenu. It probably can be according to both of them. However, it seems that Rabbenu Tam prefers to interpret it following Rabbotenu, according to whom the halakhah is fixed. On the other hand, “no pressure is to be brought to bear upon her”, in relation to the moredet who claims “he is repulsive to me”, is neither according to the Mishnah nor according to Rabbotenu. Its meaning is an immediate loss of the ketubbah and immediate divorce, depending of course on the husband’s will, without any four 74 weeks of warnings, announcements or other waiting periods. The next section of the sugya continues discussing aspects of moredet, mainly financial ones (whether she lost parts of her dowry or not). Finally, it reaches a highly important conclusion (63b-64a): (f) R. Zevid’s daughter-in-law rebelled [against her husband] and took possession of her silk [cloak] … Now that it has not been stated what the law is, [such clothing] is not to be taken away from her if she has already seized them, but if she has not yet seized them they are not to be given to her. And we make her wait twelve months for her divorce, and during these twelve months she receives no maintenance from her husband (( .,%( #-*#( *+#*,( $)( !"#$%&' )53&(#"'4&($)(,#)(.,%(#-*#(*+#*,(2"$3'(1./#0.). The conclusion, “we make her wait twelve months for her divorce”, belongs to a 75 late talmudic stratum, amoraic or even saboraic. It determines a waiting period of 12 months before receiving a get. The exact meaning of this passage is a matter of great dispute between talmudic interpreters, following the basic attitude of each commentator to the interpretation of previous stages of the sugya. The Geonim, according to both Friedman’s and Brody’s conclusions, referred to this passage as a late talmudic enactment, which determined coercion after 12 months of rebellion, 76 whereas the Geonim themselves applied coercion immediately. Rashi does not mention coercion explicitly. However, I assume that Rashi, as a continuation of his 73 74 75 76 In this sugya the Rishonim show awareness of the redactor’s work but in a different section. See Ritva, 63b, s.v. #%'*#6(#"$)', regarding Rav Huna bar Yehuda’s opinion in section (d).( See Tosafot, 63b-64a, s.v. )3.. See Friedman, Jewish Marriage, 323 n.37. See supra, n.14.( Chapter One: The Rebellious Wife (Moredet) 21 interpretation of the whole sugya, of which coercion is an integral part at every point, integrates coercion of a get at this stage as well. Rashi explicitly deals here 77 with the timing of the rule of moredet, which was earlier interpreted by Rashi as coercion of a get. Contrary to the Geonim and Rashi, Rabbenu Tam and his followers denied coercion here, in the same way as at every point in the sugya. There are a few versions of the interpretation of this passage according to Rabbenu Tam’s school. Their common denominator is that the Talmud teaches here that the husband is not allowed to divorce his wife, even if he wants to, without paying her financial rights, because she might reconcile during this period. However, after 12 months 78 the husband may divorce his wife, and she loses her ketubbah. An important question is which kind of moredet this passage refers to, “I like him, but wish to torment him” (ba’ena leh) or “he is repulsive to me” (ma’is alay). The Rishonim and A!aronim usually follow Amemar, who makes a distinction 79 between ma’is alay and ba’ena leh. Thus, to whom does this passage refer? Three 80 possible options are mentioned by the Rishonim: ma’is alay, ba’ena leh and both. Rashi at this point is not clear, and in what follows we will complete our previous discussion on his commentary of the sugya by explaining his view of this passage. The previous part of the sugya deals with a story about Rav Zevid’s daughter-in81 law, which Rashi interprets as a case of moredet ma’is alay. We may conclude that the present passage, which mentions the 12 months waiting period before receiving the get, continues that case, i.e. that of the moredet who claims “he is 82 repulsive to me” (ma’is alay). We have argued that Rashi interprets the whole sugya as a logical and systematic structure, which divides its parts between two basic concepts: the mishnaic on the one hand and that of Rabbotenu on the other. Interestingly, when describing the Mishnah and its followers, Rashi uses the 83 terminology of delaying, the same as is used by the Talmud in our passage: 77 78 79 80 81 82 83 See Rashi, 64a, s.v. !"#$%&!'() and s.v. *+&*,, and see infra. # See for example Ramban, 63b, s.v. !*(./#(-!), in the name of Rabbenu Hanan’el: “that is, if the husband said, ‘I will issue you a get immediately and I will take all the property’, we do not take heed of this, but we delay the get for twelve months, in order to consult her” (# "23!# *(.# 1.# *()"0 !3#4"(!"#.5&6.#.,'#&7*&#*+&*,#!"#$%&!'(#.".#8!&3#$%&76'(#."#!&+0%#"0#")5.)#*,"."#56#4"#$,.). In 63b the Talmud tries to support Mar Zutra, but finally rejects it in the words: .&!#.") (“This was not [the right thing to do]”). This makes possible and even preferable the determination of halakhah according to Amemar: see Rashba, 63b, s.v. ,/*)(#.&(/#&0&!. See Rashba, s.v. $&%2")#9$%&!'(); Ritva, 64a, s.v. $%&!'(). See also infra, text at note 90. See Rashi, s.v. !,"0. It is a matter of dispute amongst the Rishonim: see Rashba, 63b, s.v. 1-3(*!). See Ritva, 64a, s.v. $%&!'(). $&!'( (63b, s.v. $&0"(%); !,3),0# $&,7):)# .5&6# $&!'(/ (s.v. &0&!); or when introducing the opposite view: !,)!'!"#;!"#$%&:&&0#." (s.v. ."). 22 Westreich: Talmud-Based Solutions to the Problem of the Agunah !"#$%&!'(). If indeed Rashi deals here with ma’is alay, the fact that the terminology used here is the mishnaic one means that there is at this point a withdrawal from the earlier approach regarding ma’is alay. According to Amemar’s original teaching we do not force the moredet ma’is alay (!"#$%&*&&+#,"), i.e. in this case we would not use the mishnaic rule and delay her get, but her husband must divorce 84 her, probably according to Rabbotenu’s rule. Here, on the contrary, we impose a delay for giving the get, and according to Rashi this is in order to give her the 85 option to change her mind. It is a conceptual withdrawal from the strict approach of Rabbotenu, whose goal was the bringing to an end of the conflict as quickly as possible, towards an attempt to make it possible for the wife to change her mind, as 86 in the mishnaic rule. However, it is not a complete withdrawal, but a sort of combination of the two approaches. On the one hand !"#$%&*&&+#,", i.e. we would not use the decreasing process of the Mishnah. On the other hand, she would 87 receive her get not immediately (or after four weeks) but only after 12 months. In the meantime, according to Rashi, she has a chance to change her mind. Thus, if 88 she does change her mind, she probably would not lose her ketubbah. To sum up Rashi’s view, the final stage of the talmudic sugya imposes a waiting period of 12 months. The interpretation of this conclusion depends on the interpretative path of the sugya, especially as regards the legitimation of coercion. According to Rashi, here too coercion plays an important role: the final talmudic conclusion delays coercion for 12 months in a case of moredet ma’is alay. Yet, whether in ma’is alay or in ba’ena leh, at the end of the halakhic process the wife 89 can demand a get, and her husband is coerced to divorce her. The application of this late talmudic conclusion to moredet ma’is alay is a matter of dispute between Rishonim. Opposing Rashi, some Rishonim apply this 84 85 86 87 88 89 See Rashi, 63b, s.v. ,", and see also supra, n.70. Maybe she reconcile (!-#.)/01#&"),): see Rashi, 64a, s.v. !"#$%&!'(). See above, section 3.1. The current conclusion may therefore reflect a doubt, whether the halakhah in a case of ma’is alay is according to the Mishnah or according to the Tosefta (= Amemar’s #," !"# $%&*&&+ : see above, text at note 84): see Ba‘al Ha’itur, letter Mem, 68b: “the Savoraim [...] are undecided whether the law is as Rav Nahman bar Rav Hisda, who said the halakhah is in accordance with Rabbotenu; or as Rava who said in the name of Rav Shesheth, the halakhah is that she is to be consulted (#.(,2#1''#-.#.(,2#,-.+#&,#)%&1)-.+#!+"!#.(,2#,230#-.#.-#$(0%#-.+#&,#)!"#,4*3(2#5)'( !-#$&+"(%#!+"!). The ruling [therefore] is that we delay the get for twelve months”. See Riskin, Divorce, 44-46. See Ritva, 64a, s.v. $%&!'(). There are Rishonim who argue that the wife loses her ketubbah even if she changes her mind during the 12 months waiting period: see Ritva, ibid. As to Rashi, since this takkanah sounds in favour of the wife, I prefer the first explanation. The last takkanah of 12 months does not refer to ba’ena leh, as discussed above. Thus, the law here is the basic mishnaic law, agreed by both Amemar and Mar Zutra, and defined as: !"# $%&*&&+. According to Rashi, at the end of the mishnaic law, as already mentioned, she receives a get. Chapter One: The Rebellious Wife (Moredet) 23 90 rule to both kinds of moredet, while others apply it only to moredet ba’ena leh. 91 Amongst the latter are Rambam, Rashbam and Rabennu Tam, according to whom the law of moredet ma’is alay is as originally determined by Amemar. But at this point Rambam and Rashbam differ from Rabbenu Tam. While the latter rejects coercion, the former two both accept it, and according to their view the moredet 92 ma’is alay loses her ketubbah and receives a get immediately. In regard to coercion, therefore, there is an important group amongst the Rishonim which is in 93 favour of it and puts it within the core of the talmudic sugya. 1.3 Conclusions Coercion of a get in the case of a rebellious wife (moredet) is a matter of great dispute between talmudic commentators, whether Geonim, Rishonim or A!aronim. One most influential view was that of Rabbenu Tam, who strictly rejected coercion. Opposing the geonic view, Rabbenu Tam argued that coercion had no basis in talmudic sources. However, we have revealed a wide basis for coercion in tannaitic and amoraic sources, as well as in later anonymous talmudic discussions. Justifying coercion in a case of moredet is mainly a question of interpretation of talmudic sources. The interpretative option suggested here is a legitimate – we would even say: preferable – way of interpretation, with the significant advantages of clarity, simplicity and consistency. Furthermore, it creates a logical structure which holds together the Mishnah, the Tosefta and every stage of the discussion of the Babylonian Talmud. Not surprisingly, it was chosen by Rashi and some other commentators when interpreting the sugya. One methodological comment should be made here. Rabbenu Tam’s objection is based primarily on broader considerations, i.e. harmonizing all talmudic sources, and not on the simple meaning of this specific sugya. His main argument is 90 91 92 93 Supra, n.80. See Rambam, Ishut, 14:9-10; Shiltei Gibborim, 27a, A; Tosafot, 63b, end of s.v. !"#$%!. See the famous halakhah of Rambam, Ishut, 14:8: “([Bet din] asks her why she rebelled. If she says: ‘he is repulsive to me, and I cannot willingly have sexual relations with him’, [Bet din] coerce him to divorce her immediately, since she is not like a captive woman, who must have sexual relations with someone she hates, and she goes out (=she is divorced) without any ketubbah at all...” ()&$'%!(! )"#$%() $+') /!*3(') %$5!"') !*!%) &$+!2)/$*3-,) !')'34"') "'!2$) $#$%!) !"$*1%,) ".,%) 0%)/"-.,) ",) $#+,) "*!% ''2) "4!*2) %'4) %5*!) /"') $!#(') '34*() "$!4(2). Divorce in this case is without any delays, whereas Rabbotenu’s rule of four weeks of announcements and warnings (and then losing the ketubbah) are applied to the moredet ba’ena leh, together with the 12 months of waiting for her get: see Rambam, ibid., 9-10. ) The different types of moredet and the dispute between the Mishnah and Rabbotenu concerning the proper halakhic process do not relate to the character of coercion, which is the basic physical one: see supra, nn.35, 61. 24 Westreich: Talmud-Based Solutions to the Problem of the Agunah supported by some tannaitic sources, in which moredet is not mentioned amongst 94 the accepted cases of coercion. The Rishonim deal with those sources in accordance with Rashi’s approach, for example: solving the difficulty above by making a distinction between a case of coercion when the wife receives the amount of her ketubbah, as in the cited mishnayot, and coercion without receiving the 95 ketubbah, as in moredet. Explaining Rashi’s view in this way is based on a harmonising approach. Nevertheless, we may suggest a historical view: contradictory sources may be explained synchronically, as sources in a dispute (!"#"$%& !"'(")), or, as may be more accurate in our case, diachronically, as a developing law. That is to say, at an early tannaitic stage moredet was indeed not amongst the cases of coercion, but this changed during the generations, and the sources discussed in this chapter reflect this change in varying measures. Thus, coercion of a get in the Tosefta is an essential part of the halakhah and not only a possible outcome of losing the ketubbah as it is in the Mishnah, which may be the reason for not mentioning it amongst the mishnaic cases of coercion. Yet both sources focus on the monetary aspects, while coerced divorce is still not explicit. It becomes explicit only in late amoraic generations, Amemar according to MS Leningrad-Firkovitch or the final determination of late talmudic stratum: “we make her wait twelve months for her divorce” (01'(0&0!-&'%/'&/)'/!&,$&*+',-."). This halakhic process is influenced by sociological changes, which characterise the case of moredet. As briefly described 96 by Rabbi Ya‘akov Yehushu‘ah Falk (Pene Yehoshu‘ah): Even without that, we find a number of enactments regarding moredet, corresponding to changing circumstances: talmudic law, saboraic law, which was cited by Tosafot, and the law of metivta (i.e., the geonic law), which was cited by Rif and Rosh z”l... We have found a stable basis for get compulsion in the case of a rebellious wife. But the picture is not yet complete: was this a unique approach, developed merely in the Babylonian Talmud? Can we on the other hand point on branches of this law found in the Palestinian tradition? And what legal construction was built for justifying it? We turn now to examine these questions. 94 95 96 Mishnah Ketubbot, 7:10. See Tosafot, 63b, s.v. $20; Ramban, 63b, s.v. 0!3$,&,$"3". See Ritva, 63b, s.v. !4/".&0'.4&'3',. Actually, Ritva rejects this distinction: see ibid. See Pene Yehoshu‘ah, 63b, s.v. ",'0"& 0/.(2. The context of Pene Yehoshu‘ah’s statement is his question: 03/"2& 0,& /.0#"& 0!''/2& '0,0& 02/& ('$6& 50',?! (How could Rava be in a dispute with the baraita and describe their view so negatively?). His possible answer is that after Rabbotenu the Sages changed their mind again and moved back from Rabbotenu to the Mishnah as a result of changing circumstances (literally: “changing times”). Chapter Two Ketubbah Stipulations and the Rebellious Wife in the Palestinian Tradition 2.1 The Rebellious Wife in the Palestinian Talmud The previous chapter identified the basis in the Babylonian Talmud for get compulsion in the case of a rebellious wife. Is this conclusion compatible with the Palestinian Talmud (the Talmud Yerushalmi)? 97 The Yerushalmi discusses different aspects of moredet. As elsewhere, there are variations between the two Talmudim, the Babylonian and the Palestinian, both in citing tannaitic or amoraic sources and in the literary and conceptual development of the sugya. In our case, the Yerushalmi cites Rabbotenu differently and thus can shed light on their goal and rationale, as already discussed. In short, the baraita in the Yerushalmi varies on two significant points: (a) by mentioning divorce according to Rabbotenu, which is explicit in the Yerushalmi but not in the Tosefta 98 and the Bavli, and (b) in the absence of public humiliation, similar to the Tosefta 99 but contrary to the Bavli. We concluded, accordingly, that Rabbotenu’s goal is to lead to a separation between the couple and that divorce (where appropriate, coerced), being a necessary condition for such separation, is therefore an integral 100 part of Rabbotenu’s teaching. Another part of the Palestinian Talmud sugya is the question of the character of the moredet, whether her “rebellion” relates to her domestic or sexual role. These two options are raised implicitly in the Yerushalmi as two, not necessarily contradictory, alternatives, when explaining the differences between the rebellious wife and the rebellious husband. In the Bavli on the other hand these options are the core of an explicit dispute between two Amoraim, one of whom, Rabbi Yoseh 101 bar !anina, is mentioned also in the Yerushalmi. Interestingly, the anonymous conclusion of the Bavli limits the amoraic dispute to a domestic moredet, while both Amoraim agree to define moredet from sexual relationships as a moredet (!"#" 97 98 99 100 101 See PT Ketubbot 5:9-10, 30b. See supra, chapter 1, text at notes 36-40. The Yerushalmi uses the term '(!)*$%& while the Bavli uses '*#+!)*,*$-&: see supra, n.29. Two more differences between the Yerushalmi and the other sources are mentioned supra, n.38. See supra, text at nn.34-40. See Riskin, Divorce, 21-23. This phenomenon is well known, and reflects the high level of Babylonian conceptualization: see Moscovitz, Talmudic Reasoning, 306-309. 26 Westreich: Talmud-Based Solutions to the Problem of the Agunah !"#$%&'($)"&(*(+,&'+&'%+-&(+$.&/(%/!%). Although this talmudic limitation cannot be literally derived from the amoraic dispute in the Bavli, it does not necessarily contradict their original teaching: according to the Yerushalmi these are alternative interpretations of the Mishnah, and not necessarily a conceptual dispute. 102 The following passage is of the greatest importance: 64((5&2((3!$&2$%%&((3!&!'3/&2('&'3/&2('&2(0!."&2(+('&)1$(&(0#&#%' R. Yoseh said: For those who write [a stipulation in the marriage contract]: ‘if he grows to hate her or she grows to hate him’, it is considered a condition of monetary payment, and their condition is valid. 103 In the Cairo Genizah ketubbot, dated to the 10th-11th centuries C.E., we find divorce clauses that are similar to the Palestinian Talmud in both syntax and content. The 5th century B.C.E. Jewish community of Elephantine also reflects a very similar tradition; thus all three may be considered part of a “long chain of 104 tradition in writing the Jewish marriage contract”. Historically, it is possible that this Jewish tradition was influenced by ancient Near Eastern traditions of stipulations in marriage documents. Another possible influence, Karaite (or Palestinian influence on the Karaites), is questionable: divorce clauses, as well as 105 provisions for slavery and burial, were absent from Karaite marriage documents. & The Genizah ketubbot and their relation to the Yerushalmi are discussed below, while here we focus solely on the divorce clauses of the Yerushalmi. R. Yoseh legitimates a condition in a case of hatred between the couple by referring to it as a monetary condition. But the exact content of the condition is not clear, and it is greatly disputed in both rabbinic and academic sources. The commentators usually deal with two main questions: First, what is the exact content of R. Yoseh’s condition – does it relate only to financial aspects, for example: rejecting the mishnaic process of decreasing the ketubbah in a case of moredet, or does it relate also to the marriage itself, enabling a coerced divorce in such a case? Second, suppose the condition refers to the marriage, how is it used in practice – by coercing the husband to give a get or by a judicial act of the bet din itself? The first 102 103 104 105 Yerushalmi, ibid. The ketubbot were discovered, researched, and thoroughly discussed by Mordechai A. Friedman (Friedman, Jewish Marriage). In the following discussion I rely on Friedman’s research in many aspects, as indicated in the footnotes below. & Friedman, Jewish Marriage, 319. See Friedman, ibid., 313–320; Olszowy-Schlanger, Karaite Marriage, 263-269. As to general influence, Friedman claims some influence of the Palestinian tradition on the Karaites (see Friedman, ibid., 46–49), while Olszowy-Schlanger disputes this. For an extensive analysis of the relationship between Karaite marriage documents and Babylonian, Palestinian, and Muslim documents, see Olszowy-Schlanger, ibid., 266-271. Chapter Two: Ketubbah Stipulations in the Palestinian Tradition 27 question is discussed here, the second question in the next chapters. 106 In a case described later in the Yerushalmi, a similar condition is mentioned. A man kissed a married woman (!"#$ %&$ '"#$ ()$ *('+$ '(')$ '),-$ ./)), and her entitlement to be paid the ketubbah fell to be decided. The amoraim did not regard her as a sotah (adulteress), which would mean that her husband should divorce her and she loses her ketubbah, but treated the case as one of hatred. Accordingly, they applied here the condition which was found in her ketubbah: 5*,#$('4%#$!02+$"'!($!"('#3,'-0$"01"($)%'$!%&0$"+'%#$*".!%$"02($("+%#$).!$*")$ 107 If this So-and-so (fem.) hates this So-and-so, her husband, and does not desire his 108 partnership, she will take half of ketubbah. Unilateral divorce is not explicitly mentioned in the Yerushalmi either in the sugya of the kiss story or in R. Yoseh’s condition. The amoraim discuss mainly the financial aspects of the condition: whether she is entitled to receive at least part of her ketubbah. However, these aspects were probably accompanied by divorce, and this presumes that it includes unilateral divorce on the part of the wife. This argument is based on the clause: !"('#('-0$"01"($)%', i.e. she would reject being in a partnership with him, which means that the wife has the right to demand and 109 obtain a unilateral divorce. One possible interpretation is that the wife’s entitlement to a coerced divorce is achieved by the quoted condition. If that is correct, the term “a condition of monetary payment” (*'66$ ""+() in R. Yoseh’s condition (which supplies its legitimization) includes stipulating the right to a unilateral divorce. *'66$ ""+( has therefore a wide meaning: “‘monetary stipulations’ include agreements to forfeit a 110 right or benefit assured one by law” (M.A. Friedman). Yet divorce is not the main legal consequence of the condition. Divorce is only part of the protasis (the “if” part of the condition) while the apodosis (the “then” 106 107 108 109 110 PT Ketubbot, 7:6, 31c. The word: "02( should be read as: "+2( or )+-(: see Lieberman, Hilkhot HaYerushalmi, 61, based on Maimonides, Or Zaru’a and Me’iri’s version. This reading was adopted by Friedman, Jewish Marriage, 317; Riskin, Divorce, 31 n.16. The two words: !"('#$,'-0 should be read as: !"('#('-0: see Lieberman, ibid. See Friedman, Jewish Marriage, 329: “Shutafut ‘partnership’ here clearly denotes ‘marriage’, as in Syriac. This felicitous term is particularly befitting in a stipulation which describes man and wife as equal partners in the business of marriage, each of whom can withdraw from the partnership at will.” Riskin, Divorce, 31-32. The right to unilateral divorce appears more explicitly later in conditions in Palestinians ketubbot from the gaonic period found in the Cairo Geniza (e.g., “and if this Maliha hates this Sa’id, her husband, and desires to leave his home, she shall lose her ketubbah money… and she shall go out by the authorization of the court”): see Friedman, Jewish Marriage, 327-346, and see infra, section 2.2. Friedman, Jewish Marriage, 319-320. $ 28 Westreich: Talmud-Based Solutions to the Problem of the Agunah part of the condition) is the financial aspect, which is also the core of the amoraic discussion that follows. Accepting the above explanation, that the entitlement to divorce is based on the condition, requires one to assume that “the text that is quoted omits... the wife’s exit from the marriage, the divorce itself which resulted 111 from her ‘hating’ her husband.” But why does divorce seem to be less significant in the conditions of the Yerushalmi? As we have argued, a get was an integral part of the law of moredet already in tannaitic sources, and in particular is part of Rabbotenu’s rule in the Tosefta, as is explicit in the Yerushalmi’s version of the baraita. Demanding a divorce therefore did not have to be based on any condition, but was based rather on the law of moredet itself. Accordingly, the reason why the amoraim do not discuss the right to demand divorce is that it was already known and accepted, rather than this being the “point of the innovation” of the condition. The same conclusion applies to the missing apodosis of Rabbi Yoseh’s condition: it might have mentioned the coerced divorce, but its core is monetary, i.e. to regulate the 112 financial terms of the tannaitic coerced divorce. R. Yoseh merely adds a financial aspect, which overrides the tannaitic rule of moredet. Namely, although a moredet loses her ketubbah, if the couple has stipulated that she would not lose it, the condition is valid since it is a condition of monetary payment. The justification for accepting the condition, “[it is] a condition of monetary payment and their condition is valid” (!""#$ %""&'($ %())$ "*&'), has its simple meaning: it is interpreted as referring to the monetary arrangements; the condition can be accepted precisely because it does relate to a monetary issue. The right of the moredet to receive a get may appear in this sort of condition, but its basis is not the condition but a more stable one: the basic tannaitic law of moredet. 113 Interestingly, some Geonim and Rishonim – Ramban and others – do explain the divorce clause of the Yerushalmi in the same way, i.e. as a clause which was 114 required for the financial agreements. Both are affected by their understanding of 111 112 113 114 Friedman, ibid., 318. This interpretation of R. Yoseh’s condition is briefly suggested by Friedman, Jewish Marriage, 320, as a second interpretative option. The present research supports this option and puts it in a wider context. See Rav Hai, infra, text at notes 144-146. See Ramban, BT Ketubbot 63b: when the couple explicitly stipulated that in a case of moredet the wife receives all her ketubbah, it is valid since it is %())$"*&'. As a support for this ruling Ramban quotes the Yerushalmi: “but if he wrote her that he accepts upon himself that even if she were to be rebellious she would take [all her ketubbah], as Rav [Yosef] ibn Migash, of blessed memory, said, that in that case it is considered to be a condition of monetary payment and is valid. And in the Yerushalmi: ‘for those who write: If he grows to hate her or she grows to hate him, it is considered to be a condition of monetary payment, and their condition is valid’ ” ($,-$,.#/$0,$.'+$%+$!*$*,* Chapter Two: Ketubbah Stipulations in the Palestinian Tradition 29 the Babylonian sugya of moredet but in opposite ways: the Geonim understood the sugya as a source for coercion, and therefore the Palestinian divorce clause was not required to be understood as legitimating coercion. The other Rishonim who cited the divorce clause understood the Bavli as excluding coercion, due to the adoption of Rabbenu Tam’s interpretation of the Talmud. Therefore they were motivated to 115 interpret the clause of the Yerushalmi as discussing only financial aspects. 2.2 Ketubbah Stipulations in the Post-Talmudic Palestinian Tradition Genizah Ketubbot indicate that the Palestinian custom of ketubbah stipulations, and in particular the Palestinian divorce clause, continued (maybe even expanded) in the post-talmudic era. In the Palestinian tradition the couple stipulated explicitly 116 in their ketubbah that the wife is entitled to a unilateral divorce, for example: $!"2$*,+)$.+#$324$!"$*54,2$666*+.24.2&#$+#7.$1!2$*!"#$%&#'$()*!$+,-.$*.!/$*0+0"$*)*$(+12 6(2*+.") And if this ‘Aziza, the bride, should hate this Mevasser, her husband, and not desire his partnership [...] and she will go out by the authorization of the court and with the consent of our masters, the sages. The Genizah ketubbot and the tradition found in the Palestinian Talmud are, in Friedman’s words as indicated above, part of a “long chain of tradition”. The Genizah tradition could thus issue unilateral divorce on the basis of the law of the rebellious wife. I support this conclusion in the next chapters, discussing the relations between the Palestinian and the geonic traditions and the place of 117 marriage annulment in both. This conclusion, however, raises an essential question. If, indeed, there was a basis in positive law for unilateral divorce, both for the Babylonian and for the Palestinian traditions, why was it necessary to write the divorce clause in the Palestinian ketubbot? One of the two citations of the divorce clause in the Yerushalmi, “the kiss story” cited above, suggests a unique version of it: “If this So-and-so (fem.) hates this So- ---$(+!+1=$ <+'!&2%+#2$ 63++52$ 12*$ (2''$ +1,.$ 3.*)$ 9!;0$ &:+'$ (#$ #%*$ %'1&$ 2'/$ 9!28.$ )2%'.$ 2!+41&$ 2'7" 3+,4!$?+%7$*0$(+12$9>3++52$12*$(2''$+1,.$.1,&$(+1$+1,&$(+1$(+#./)). 115 116 117 Z. Falk makes a similar argument, according to which the condition in the Yerushalmi deals only with the financial aspect, and suggests that coerced divorce might have not been part of this condition, as he claims to find in some of the Elephantine marriage documents: see Falk, Gerushin, 22. Here, however, we follow our previous conclusions, according to which the Yerushalmi did accept coerced divorce. Ketubbah no. 1, lines 23–24, in Friedman, Jewish Marriage, II, 9 (Heb.); 13 (translation). See infra, Chapters Three and Five. 30 Westreich: Talmud-Based Solutions to the Problem of the Agunah and-so, her husband … she will take half the ketubbah.” According to this clause, the wife is entitled to half of her ketubbah where she initiates unilateral divorce. Thus, we can suggest a reasonable explanation of the practical necessity for this clause: it was required in order to regulate the financial arrangements, which might vary from case to case. Following this argument, the divorce clause which stipulated a total loss of the ketubbah, normally found in the (later) Palestinian ketubbot, was written to exclude this different option, that of loss of only half of the ketubbah. In other words, the divorce clause was not required in order to legitimate unilateral divorce, since the latter had an independent basis, as argued above. It was required, rather, for the financial arrangements, which were subject to variation and therefore needed to be explicitly stated. As seen in the previous section, the structure of the divorce clause supports this interpretation: divorce is only part of the protasis (the “if” part of the condition) while the apodosis (the “then” part of the condition) is the financial 118 aspect, which is also the core of the amoraic discussion that follows. However, this explanation of the divorce clause as required for the financial arrangements is not completely satisfying, as regards the Genizah ketubbot. The equal distribution of the ketubbah which is mentioned in the Yerushalmi was unique to that case, and every other occurrence of the divorce clause – both in the early Elephantine marriage documents and in the later Palestinian ketubbot – has the standard financial arrangement, according to which if the wife unilaterally 119 demands divorce she completely loses her ketubbah. I doubt therefore if the half sharing of the ketubbah was practiced at all at the time of the Palestinian ketubbot from the Cairo Genizah. It may be the case that at some stage (the first centuries C.E., which are reflected in the Yerushalmi) this stipulation was required in order to exclude other possible financial arrangements. But in later times those alternatives were no longer in use and their exclusion was not necessary any more. The question of the necessity for this condition thus arises again: if only one arrangement was in practice, it did not need to be stipulated. And as regards the legitimization of coercion – we do have a rule of positive law for it. It appears therefore that the divorce clause in the Palestinian ketubbot was written as part of a general custom in the Land of Israel, according to which court 118 119 Supra, text at notes 110-111. ! According to Genizah ketubbot, the wife loses her ketubbah (mohar), but receives her dowry. Some ketubbot, however, distinguished between the delayed mohar payment, which was forfeited by the wife, and the advanced portion (the muqdam), which was considered as her personal property and therefore was not returned to her husband: see Friedman, Jewish Marriage, 333-335. Chapter Two: Ketubbah Stipulations in the Palestinian Tradition 31 120 stipulations were frequently written, even though they were not strictly required. This assumption is supported by the fact that some Palestinian ketubbot mention only the existence of the divorce stipulation without its details: “They agreed between themselves ‘concerning the matter of hate and love (!"#$%&'!"()'*)+',+, 121 i.e. the divorce stipulation) and life and death’ and all court stipulations”. We may conclude from this fact that the content of the divorce stipulation was known and common and there was no substantial need for it to be written. This might be also the reason why R. Yoseh in the Yerushalmi does not give any details of the divorce clause, but only rules that it is legitimate. Indeed, some scribes of the Genizah ketubbot were satisfied merely to mention its existence. Others, however, happily for us, preferred to write it out in detail. 120 121 See Friedman, Jewish Marriage, 15-18, 330. In Babylonia the opposite custom was observed: court stipulations were not written. See Friedman, ibid., 16. This might be also the basis for the custom in several places of not writing a ketubbah at all: see Rashi, Ketubbot 16b, s.v. -..'/%. Friedman, Jewish Marriage, 340. Chapter Three Palestinian Ketubbah Stipulations and the Geonic Moredet The Palestinian custom of ketubbah stipulations, and in particular the Palestinian divorce clause, continued (maybe even expanded) in post-talmudic era, as reflected in Palestinian ketubbot. At the same time, in Babylonia, the Geonim (approximately 7th-11th centuries C.E.) widely practiced the law of the rebellious 122 wife in order to enable unilateral divorce on the wife’s demand. These two posttalmudic traditions have developed two different institutions for a single object: enabling the wife to demand – and obtain – a unilateral divorce, even against her husband’s will. Was there an interaction between them? 123 In the Palestinian tradition the couple stipulated explicitly in their ketubbah that the wife is entitled to a unilateral divorce. According to the geonic tradition the wife’s right to a unilateral divorce was based on the law of the rebellious wife (moredet). The Geonim based their view on the Talmud, or more precisely: on a decree of the Saboraim, as cited at the final stage in the talmudic passage, which legitimated coercion of a get in the case of a rebellious wife: “we make her wait twelve months for her divorce” (!"#$!% !&'% #()#% )*#)&% +,% -.#+'/0, Ketubbot 124 64a). The effect of the geonic enactment was therefore (a) to coerce the husband to give a get immediately and not only after 12 months, and (b) to impose one or more of several possible monetary regulations in favour of the wife (such as not 125 losing her basic ketubbah). This is explicitly stated in some geonic responsa, as in 126 the following from Rav Sherira: 122 123 124 125 126 See Friedman, Jewish Marriage, 324-325; Brody, HaGeonim, 298-300. This is clearer in the Genizah ketubbot than in the Palestinian Talmud. Nevertheless, the two belong to one continuing tradition (see supra, text at note 104). See supra, text at notes 13-14. According to our analysis, the basis for coercion of a get is much earlier, in the Tosefta, or even in the Mishnah (see ibid., section 1.2.1). The exact monetary aspects of the geonic enactment(s) are not clear, and were probably disputed by the Geonim themselves. See Brody, HaGeonim, 300-304. Teshuvot HaGeonim, Sha‘arei Tsedek, Vol. 4, 4:15. Both Friedman and Brody assume that this view was largely accepted by the Geonim: see Friedman, Jewish Marriage, 324-325; Brody, HaGeonim, 298-299. Chapter Three: Palestinian Ketubbah Stipulations and the Geonic Moredet 33 This is our opinion [lit. we saw in the following way]: the original law was that [the bet 127 128 din ] does not oblige the husband to divorce his wife if she asks to divorce […] Later […] Nevertheless [the bet din] did not oblige the husband to write her a get […] [Later the Rabbis] enacted that when she demands divorce [the bet din] makes her wait twelve months [in case] perhaps they reconcile, but if they do not reconcile after twelve months 129 [the bet din] compels the husband and he writes her a get. After the Saboraim […] [the Geonim] enacted […] and [the bet din] coerces the husband and he writes her a get immediately [upon her demand] and she gets the hundred or two hundred [zuz, of her ketubbah]. This is the way that we have ruled for three hundred years and more. You should also act in this way. In his responsum, Rav Sherira indicates that the geonic enactment was a response to specific historical circumstances. A growth in the number of wives appealing to Muslim courts led to the fear of a coerced get (get me‘useh) and probably also to the fear of conversion to Islam. The Geonim responded by improving the wife’s 130 legal power in family matters deliberated in Jewish courts. In the basic ruling of compelling a divorce for the moredet, however, they followed the Talmud. The relationships between these traditions are not completely clear: are they independent, without any direct connection, i.e., contractual vs. normative, but having some similar characteristics as a result of their similar historical environment or common cultural background? Or are they connected, perhaps even reflecting similar legal constructions but only expressed differently, having the same normative basis and with some reciprocal influence between them? A link between the two traditions is made by Me’iri’s teachers’ teachers, who argue that the normative basis for the geonic compulsion of a get in the case of a 131 rebellious wife is R. Yoseh’s clause in the Palestinian Talmud: “and my teachers testified that their teachers explained that the geonic innovation in this matter is based on what is written in the Western Talmud in this sugya: !"#$%!"#!&"'()*!&"+", 132 -""./! #/,! &/00! "#$(! ("#$%! "#. What is the exact meaning of this link? Me’iri 127 128 129 130 131 132 I have added “the bet din” when the Gaon refers to the judicial act. When he refers to the enactment I have added “Rabbis” for the first two enactments and “Geonim” for the last one. Verb in plural (&"'""10). Similarly, all the judicial acts infra are formulated in the plural (!2&"'""10 &"5/)!2&"3"4)0), in contrast with the actual writing of the get: 76!,+!'(/)/. This enactment is the final section of the talmudic passage (#7"6#! #(%! "14"! 48"4(! ,+! &$",%0/), which Rav Sherira ascribes to the Saboraim. See Rav Sherira’s responsum, ibid., and see E. Westreich, Rise and Decline, 217-218; Brody, HaGeonim, 295. Supra, text at note 102. Bet haBe!irah, Ketubbot, 63a, s.v. (*4/0'!*/0+(,!&"*!/,3. Me’iri’s teachers’ teachers explicitly link the geonic rebellious wife to the divorce clause of the Palestinian Talmud. However, the Palestinian Talmud and the later Genizah ketubbot are part of a single tradition. M.A. Friedman even suggests Westreich: Talmud-Based Solutions to the Problem of the Agunah 34 133 himself opposed coercion in cases of moredet. His discussion of the geonic measures relates to their financial enactments, according to which the wife would 134 not lose her basic ketubbah (and other monetary components). Me’iri rejects these enactments (“it is not correct to rule like them”), but then cites his teachers’ teachers who find some support for the Geonim in the customary Palestinian divorce clause. Accordingly, the link between the two traditions does not relate to the coerced divorce but rather to the financial aspects of the rebellious wife. Nevertheless, taking the words of Me’iri’s teachers’ teachers (as cited in Me’iri’s commentary) out of their context in Me’iri’s text reveals a different intention: it appears that Me’iri’s teachers’ teachers tried to legitimate the coerced divorce itself and not (only) the financial aspects. Thus, they interpret “if she grows to hate him” in R. Yoseh’s condition as: “if she grows to hate him, so that he is required to divorce her (!"#$%& '(!& ))*+") whether while [receiving] all the 135 ketubbah or with a small reduction”. In the same way, Me’iri’s teachers’ teachers refer to the fear that she may (unjustifiably) “take herself out of her husband’s control” (!%,-&.+/&!/0,&,+)1!%) as the reason for their seeking to find support for the geonic ruling, which means that the wife had the option of unilateral divorce and this needed justification. The divorce clause accordingly gives the wife the right to initiate unilateral divorce, and the geonic enactments were based on this custom. Me’iri’s teachers’ teachers’ argument is as follows: this condition was practiced not only in the Land of Israel, but was also known and used in Babylonia. Thus, the divorce clause was at first a widespread practice. Then the decree of the Geonim made it an obligatory norm, even when it was not written, thus authorizing them to compel a divorce in all such cases (or require different financial arrangements, according to Me’iri). This is similar to other cases defined in the Babylonian Talmud as “court stipulation” (tnai bet din), i.e., a clause in the ketubbah (for example: benin dikhrin [a clause that gives preferential inheritance rights to the sons of this wife]), which became a binding practice, so that the spouses are obliged to follow it even if it is not written explicitly in their ---that Me’iri’s teachers’ teachers based themselves also on an actual ketubbah, and not only on the Palestinian Talmud; see below.& 133 134 135 See Me’iri, ibid., s.v. 2+#-3/!& +%(.$(: according to our [i.e., Me’iri’s] opinion the husband is not coerced [to give a get] (4+1(5&4+'&(67,.%(). Ibid., s.v. 7.#(/-&.(/%7!&4+.&(!*: “in the financial [lit. collection] issue the Geonim innovated”. Similarly, they mention (ibid.): “if she hates him she shall take her ketubbah or part of it and she shall leave ('07()”. The addition '07( to the divorce clause in the Palestinian Talmud shows as well that they understood this clause as legitimating unilateral divorce. Chapter Three: Palestinian Ketubbah Stipulations and the Geonic Moredet 35 136 ketubbah. 137 Some scholars have accepted this view as historically correct. Amongst them, 138 an interesting compromise view is suggested by Moshe Shapira. Shapira bases the geonic tradition on the Palestinian divorce clause, following Me’iri’s teachers’ teachers, but in a unique way: as a cause for the cancellation of the talmudic 12 months’ waiting period and not as a basis for compulsion of a get (or for other 139 financial aspects). Therefore he argues as follows: (a) at first, there was a practice of writing the divorce clause, which became more and more widespread, to the extent that it became possible to coerce a divorce even if the divorce clause was 140 not explicitly included. The divorce clause included, in addition to unilateral divorce, the right of the wife to receive her ketubbah or part of it. (b) Thus, according to Shapira, the 12 months’ waiting period became otiose, since (based on the divorce clause) no sanctions were left during that period against the wife: she got alimony, and when divorced received her full ketubbah. (c) The Geonim ruled, therefore, that the coerced divorce should be effected immediately upon the wife’s demand, canceling the 12 months’ waiting period. However interesting this argument is, it is historically unconvincing. Shapira bases his argument on the claim that according to the divorce clause the wife receives her ketubbah (and thus that the 12 months’ waiting period lost its function). This claim is based on another citation of the divorce clause in the 141 Palestinian Talmud, which gives the wife half of the ketubbah, and on Me’iri, who adds the option of receiving all of the ketubbah: “[the divorce clause stipulates that] if she hates him she will receive her ketubbah or part of it and leave”. Historically, however, this is inaccurate: we do not find in the divorce clauses any precedents for receiving the ketubbah in full. Even receiving half the ketubbah was not the practice written into the Genizah ketubbot at the time of the Geonim, which 142 always mention the wife’s total loss of the ketubbah. Thus, Shapira’s description 136 137 138 139 140 141 142 See M Ketubbot 4:7-12. In the Palestinian ketubbot, however, the divorce clause was written. This may be explained as part of a general custom in the Land of Israel, according to which court stipulations were frequently written, even though they were not required: see supra, text at notes 120-121. See Lieberman, Hilkhot haYerushalmi, 61 n.!. M.A. Friedman doubts whether this description is historically possible: see Friedman, Jewish Marriage, 325-327, and see also below. Shapira, Gerushin, 124-130." Shapira, Gerushin, mainly at 129." Since this argument explains the geonic decree, we must assume (according to Shapira’s reasoning) that the process here described existed in Babylonia as well. The “kiss story”; PT Ketubbot 7:6, 31c." See supra, text at note 119. 36 Westreich: Talmud-Based Solutions to the Problem of the Agunah of stage (a) above is doubtful as to the wife’s receiving the ketubbah, and therefore his whole historical reconstruction becomes problematic. According to Shapira’s reasoning the 12 months’ waiting period was still relevant, since the wife could lose at least part of her ketubbah, and there the geonic decree had no reason for canceling this waiting period. Beyond these arguments, it is hard to accept Me’iri’s teachers’ teachers’ view, following either Shapira’s explanation or the classic interpretation of it, as a support for coerced divorce or for the financial arrangements. It is correct that the divorce clause was a common practice. Nevertheless, the Geonim do not refer to the Palestinian tradition of making such a condition as their normative basis. They refer rather to the Talmud as the source for coercion, and explain their decree as relating to the timing of coercion and to the monetary aspects. And even as regarding these latter details, the Geonim did not mention any contractual aspect 143 (!"##$%&'() as their basis but rather the needs of their time. Indeed, it is possible that the Geonim were familiar with the Palestinian tradition (but not as a basis of their enactments). According to the following responsum, they interpret it as relating to the financial aspects of the law of the rebellious wife 144 rather than to the basic right to demand divorce. This familiarity may be deduced from Rav Hai Gaon, who legitimates some kinds of financial arrangements in cases 145 of the rebellious wife on the basis of: “since it is a condition of monetary payment, and it is valid” ()%%*"$&"+$!"##$%%&'(,). This is almost word for word the 146 Palestinian justification of the ketubbah clause, and it is cited here as a support for monetary aspects rather than for coercion. Yet, even with regard to the financial aspect of the geonic decree on the rebellious wife, the Geonim did not refer to the Palestinian Talmud as their basis. Thus, a distinction should be made between the positive law aspects of the rebellious wife, which were regulated by geonic enactments (either financial or the timing of coercion), and the contractual aspects, which were left to the spouses’ agreement. The geonic enactment on the rebellious wife was a piece of independent legislation, not based on the Palestinian divorce clause. In other words, it appears that there is some interaction between the Geonim and the Palestinian divorce clause with regard to financial aspects, but not with regard to 143 144 145 146 See supra, text at notes 125-130. The structure here suggested is similar to Me’iri himself, as discussed above, and to some other talmudic commentators, but with a significant distinction: those Rishonim rejected coercion, while the Geonim supported it, but found its basis in the Talmud: see supra, text at notes 113-115. See Teshuvot HaGeonim (Harkavi edition), 523. In the Babylonian Talmud we find )%%*$"&'($!"##-,$.-/ (Ketubbot 56a), and similarly in the Tosefta (Kiddushin 3:8). The formula &"+$!"##$%&'( is unique to the Palestinian Talmud. Chapter Three: Palestinian Ketubbah Stipulations and the Geonic Moredet 37 the right to demand divorce unilaterally, and not as a support for the enactments. Me’iri’s teachers’ teachers’ explanation of the geonic decree is a result of 147 dogmatic acceptance of Rabbenu Tam. Since, according to Rabbenu Tam, the Talmud does not mention coercion, we need a different basis for the Geonim, and this suggestion finds its basis in the Palestinian tradition. As Me’iri mentions, his teachers’ teachers were aware of the anachronistic character of their interpretation: And they (i.e., his teachers’ teachers) wrote at the end of their writings that it is better for us to take pains to interpret their teachings (i.e., the teaching of the Geonim) than to say 148 that they explicitly uprooted the whole talmudic passage without any reason. Perhaps Me’iri’s teachers’ teachers were faced with a real situation, which proved 149 the catalyst for their assumption. Mordechai Akiva Friedman assumes that Me’iri’s teachers’ teachers were not only aware of Rabbi Yoseh’s condition in the Palestinian Talmud, but were also familiar with the real practice in the Land of Israel at their time, i.e., they saw a “real” Palestinian ketubbah which included a similar clause. According to Friedman, the teachers’ teachers are likely to have been Ra’avyah (R. Eliezer b. Joel Halevi), who examined a ketubbah that was brought from the Land of Israel and apparently contained the divorce stipulation, 150 similar to the divorce clause in the Palestinian Talmud. This actual finding “could have led him to conclude that there was a direct connection between the 151 (Palestinian) clause and the (Babylonian) geonic enactment.” Me’iri’s teachers’ teachers thus base the geonic tradition on the Erets Israel custom. As argued here, the actual interaction between the two traditions might be limited from an historical perspective. But the very fact of making such a link has a dogmatic significance. For Me’iri’s teachers’ teachers, the Palestinian tradition is 152 sufficient to legitimate the problematic geonic tradition, probably even in relation to what they (following Rabbenu Tam’s view) regarded as non-legitimate coercion. This attitude towards the Palestinian tradition gives it an enormous dogmatic weight: it can justify customs, norms etc., even if they lack a normative basis in the Babylonian Talmud. The core question now is what exactly can be supported by the Palestinian precedent. Some scholars argue that the Palestinian tradition is based on a variation 147 148 149 150 151 152 See supra, text at n.11. Me’iri, Bet haBe!irah, Ketubbot, 63b, s.v. !"#$%&' ()!' "&*. Me’iri himself needs this anachronistic support for the financial aspects, as argued above. Friedman, Jewish Marriage, 327.' See Sefer Ra’avyah, Vol. 4, §919 (Mishpete haKetubbah), 309. ' Friedman, ibid.' At least according to the teachers themselves; see above.' 38 Westreich: Talmud-Based Solutions to the Problem of the Agunah of marriage annulment, and according to their view no get was required in order to 153 154 perform divorce. The results are (dogmatically, rather than historically) far reaching: a preliminary agreement between the spouses can be a basis for marriage annulment, and the fact that it was done in Erets Israel in the past gives it its 155 legitimization. An alternative explanation of the Palestinian tradition is that the husband was coerced to grant his wife a get on the basis of the preliminary agreement. Here too, there is an important dogmatic implication: according to the view of Me’iri’s teachers’ teachers, a preliminary agreement can dissolve later 156 problems of get me‘useh, when divorce is initiated solely by the wife. As mentioned, some parts of the Palestinian ketubbot as well as later interpretations of the geonic enactment raise another possibility: that of explaining these two traditions on the basis of marriage annulment. We now turn to discuss this issue: the next chapter discusses the talmudic sources of marriage annulment, after which, in chapter 5, we return to the Palestinian and the geonic traditions, examining their possible use of marriage annulment. 153 154 155 156 I, however, reject this view; see infra, text at nn.260-272. See Westreich, Divorce on Demand, 360-363. Below, I accept this explanation; see ibid. . We find precedents for this kind of condition, as in the monogamy condition, according to which the husband committed himself to divorce his wife if he takes a second wife: see E. Westreich, Temurot, 26-29. These cases are beyond the scope of the current discussion.! Chapter Four Marriage Annulment: From Mishnah to Talmud 4.1 Hafka‘at Kiddushin Annulment of marriage (Heb. Hafka‘at kiddushin) is mentioned in various contexts in the Babylonian Talmud. A number of famous talmudic passages discuss the concept of marriage annulment: “the Sages annulled his betrothal” (!"#"$% &"'()"*+% &#,-% (!#".*/0). In a similar way the Palestinian Talmud, when discussing a case where a get was formally void but validated by the Sages, mentions the notion that: “their [i.e., the Sages’] words uproot the words of the Torah” (!-(1%"-,)% &"-*(.%&!"-,)), according to which the Sages have the authority 157 to annul the marriage in certain circumstances. From early classic commentators to modern Jewish Law scholars, the character of marriage annulment in Jewish Law has been much debated. In particular, what is the legal construction of marriage annulment and what are the conditions for its application: does it always entail retroactive annulment of the marriage or may it be only prospective, and based on what authority? And does a writ of divorce, which is mentioned in several talmudic passages in the context of annulment, have 158 a significant role in this process? These debates revolve around the appropriate reading of talmudic sources. Nevertheless, textual analysis of the main passages reveals support for almost all the competing opinions. Typically for layered talmudic text, there is no homogeneous meaning; each reading exposes one or more possible aspects of the text. Indeed, some scholars have pointed in the past to the contribution to the issue 157 158 See BT Yevamot 90b; 110a; Ketubbot 3a; Gittin 33a; 73a; Bava Batra 48b; PT Gittin 4:2, 45c. The exact context of these passages will be discussed below. Hafka‘at kiddushin (annulment) literally means cessation of the betrothal. The legal result, however, is annulment of marriage, and it is commonly used in this context. In the following I therefore refer to annulment of marriage and annulment of betrothal as synonymous, unless otherwise explicitly specified. For the moment, see Freimann, Seder Kiddushin, and the classic literature he cites (e.g., R. Solomon ben Abraham Adret [Rashba] and R. Asher ben Ye!iel [Rosh], 66-72); Berkovits, Tnai, ch.4; Sho!etman, Hafka‘at Kiddushin. Additional sources and references to modern debates are cited below. On the (more general issue of the) authority of the Sages to uproot the words of the Torah, see Gilat, Perakim, 191-204; Elon, Jewish Law, chs. 14-16. Both Elon and Gilat discuss also the particular case of marriage annulment; see below. On the authority to uproot the words of the Torah as a halakhic tool used in cases of conflict between Jewish law and morality see Sagi, Yahadut, 230256. 40 Westreich: Talmud-Based Solutions to the Problem of the Agunah of the ultimate talmudic redactor, especially in interpreting annulment as 159 retroactive. But the picture which has been drawn is still incomplete, as regards both the development of the concept and the question of the authority of the Sages. A re-reading of the sources is therefore required. The advantage of revealing the talmudic strata is not merely for the purposes of historical research of the talmudic text, but also for analysis of the legal status of marriage annulment. This kind of tension between talmudic layers is a classic ground for creating contradictory interpretations amongst talmudic 160 commentators. In what follows attention is directed to several post-talmudic interpretations of the concept of marriage annulment, each of which was influenced by a different talmudic stage. Revealing the various approaches throughout the talmudic sources and post-talmudic literature is thus essential for establishing the actual basis for modern proposals for practical implementation of marriage annulment as a solution to the problem of agunot. Jewish Law is normally characterized by a pluralist discourse and, despite acrimonious controversies, the merits of competing arguments are recognized, 161 receiving some legitimacy – at least on a post factum level. Nevertheless, Jewish family law, and especially marriage annulment, is characterized quite differently. In the last few decades some proposals of marriage annulment were raised as a solution to the problem of agunot. On the basis of the analysis of talmudic and post-talmudic sources we might expect some acceptance of these proposals. Yet those solutions have met with severe objections, frequently resulting in total 162 rejection, accompanied by strong emotional reactions. These phenomena patently diverge from the pluralist hermeneutic discourse normally characterizing Jewish Law. The discussion that follows thus makes a significant contribution: it reveals the deep and stable basis of marriage annulment in talmudic and post-talmudic 163 literature. 159 160 161 162 163 See Atlas, Netivim, 206-24 (= Atlas, Kol Demekadesh); idem, Addition; Sho!etman, Hafka‘at Kiddushin, 352-355. See Friedman, Tosefta Atikta, 149. See Sagi, The Open Canon. Specific references are supplied below. See Goldberg, Hafka‘at Kiddushin ‘Enah Pitron; idem, ’Ein Hafka‘at Kiddushin, as opposed to Rabbi Shlomo Riskin’s proposal: Riskin, Hafka‘at Kiddushin. For an English version, see Riskin, Hafka‘at Kiddushin (English Version), criticized by Rabbi Jeremy Wieder, Rebuttal. See also Riskin, Response; Wieder, Rejoinder. Another debate is between Rabbi Uri’el Lavi and Prof. Berakhyahu Lifshitz: see Lavi, Ha’im, as opposed to Lifshitz’s proposal: Lifshitz, Afke‘inhu. See also Lifshitz’s response: Lifshitz, Al Masoret, and Lavi’s reply: Lavi, Hafka‘at Kiddushin Einah Ma’aneh. Several aspects of these debates are discussed below.! The debate over marriage annulment, however, requires an extensive discussion, taking into consideration various aspects, e.g. hermeneutical, political (i.e. the struggle over authority which characterizes the Israeli family law) and sociological: see Westreich, Gatekeepers. Chapter Four: Marriage Annulment: From Mishnah to Talmud 4.2 41 Talmudic Cases of Hafka‘at Kiddushin Two prototypes of constitutive annulments are found in the Babylonian Talmud. The first is annulment granted directly after the betrothal and taking effect from the moment of the betrothal, due to some fault in the betrothal procedure (hereinafter: immediate annulment). Although divorce in Jewish Law is executed by a writ of divorce willingly given by the husband to his wife, here no such writ would be required. The second is annulment issued quite a while after a valid betrothal (and marriage) took place (hereinafter: delayed annulment). All cases in the latter group deal with a defect in the divorce procedure. They include a writ of divorce which was written, given to the wife, or sometimes delivered to an agent, but which for some reason external to the writ itself was invalidated (hereinafter: an externally flawed get). Annulment, applied in these cases due to a variety of reasons, makes the couple practically divorced, despite the formal defect in the get. Two cases are included in the first group: 164 (a) the case of Naresh – a minor orphan girl was married (in a ceremony valid only by Rabbinic law, and not by Torah law) to a man who sought to marry her after she became an adult, but a second person kidnapped her and 165 married her (by Torah law); and 166 (b) the case in which the woman was forced to get betrothed and then 167 168 willingly (from a formal perspective ) gave her consent (!"#$%&'%"()). In both (a) and (b), the Talmud records that the marriage was annulled due to the misconduct of the “husband” when betrothing his “wife”: 6'"*"+&'"!%#"$(&,*-.&%'*"/$01%&,2%'3&1(!&%-&%!/&43"0(&5,2%'3&1(!&'!/&1%'& 164 165 166 167 168 BT Yevamot, 110a. Her agreement is not mentioned, but she probably gave it, at least after being kidnapped (otherwise the marriage was not valid, and annulment was not required): see Rabbenu Nissin ben Reuben Gerondi (Ran), Yevamot, 38a in Alfasi (Rif) (in the Vilna edition); R. Yom Tov ben Abraham Ishbili (Ritva), BT Yevamot 110a, s.v. 1%'; cf. Ramban, BT Yevamot 110a, s.v. "!1&-.. BT Bava Batra, 48b.& The formal validity of the marriage is based on an expansion of R. Huna’s statement: “If someone were forced to sell, the sale is valid” which was made by Amemar: “If the wife were forced to accept the betrothal, the betrothal is valid” (BT Bava Batra, 47b). R. Huna’s statement is discussed by Benny Porat, Ha!oze Hakafuy, 102-106. The moral problem with the man’s act is obvious, and is therefore a reason for take action against it, even in contradiction to the formal laws of marriage and divorce. This explanation rejects the assumed equivalency of formal halakhic rules and moral principles (for further discussion: see Sagi and Statman, Religion and Morality, 5-8). Accordingly queries 1, 3, and mainly 4 in Porat, ibid., 103, are easily resolved. 42 Westreich: Talmud-Based Solutions to the Problem of the Agunah He acted improperly; they, therefore, also treated him improperly, and the rabbis 169 annulled his betrothal. Three cases are included in the second group. In all of these cases a valid writ of divorce was written and given but was invalidated by external events: 170 (c) The first is a case of conditional divorce. The husband initially made a condition whose fulfillment would invalidate the writ of divorce, and then tried to fulfill the condition (that is, to invalidate the get), but an unexpected accident prevented him from doing so. In principle the claim of an unforeseen event (!"#$) is acceptable, and in this case it means that the condition is considered as fulfilled, the get is annulled, and the wife is not divorced. Rava, however, according to one tradition in the Babylonian Talmud, argues that the claim of an unforeseen event cannot be accepted here, and the wife is divorced. The Talmud explains that Rava’s reasoning is that in order to prevent extreme results the Sages enacted that the marriage is annulled. The results which the Sages feared are (1) the wife’s remarrying when she was not properly divorced, if indeed it was an unexpected accident and the get was invalidated; or (2) the woman becoming an agunah, if the event was really not unforeseen and the get was valid, but a chaste observant 171 woman would fear that it is invalid and therefore would not remarry. (d) A dying person who gave his wife a get (in order, for example, to exempt her from being bound to a levirate marriage) but later recovered from his 172 illness. According to R. Huna, the writ of divorce is annulled, since it was given under the assumption that he would die but he did not. Both Rabbah and Rava disagree with R. Huna in cases where he has not explicitly made such a stipulation, due to a fear of a mistake: people would erroneously think 169 170 171 172 BT Yevamot, 110a; Bava Batra, 48b. This reasoning is mentioned in the case of Naresh by Rav Ashi, and his sources are discussed below. In the second case it is mentioned by Mar bar Rav Ashi according to the following textual witnesses: Mss Oxford, Florence, Munich, Vatican 115 and print editions, while according to others (Mss Hamburg, Paris and Escorial) it is Rav Ashi here as well (see also infra, n.241). Following the version of “Mar bar Rav Ashi”, we may consider it as a “transferred” statement, but there is no reason to ascribe the transmission to a later editor (compare Ben Menahem, Hu ‘Asa; Sho!etman, Ones, 118-120; Porat, Ha!oze Hakafuy, 106 n.148). In my opinion it is reasonable to assume that Mar bar Rav Ashi used his own father’s statement, which fitted properly his case: according to Rav Huna and Amemar’s reasoning, the betrothal is formally valid though immoral. Therefore the response is “improperly”, i.e. beyond the formal borders of the halakhah. In fact, by contrast with H. Ben Menahem’s view (ibid.), +%#&'($)* is a modification of a common expression (see infra, n.203), used for Hafka‘ah by both Rav Ashi and Mar bar Rav Ashi, so it is hard to derive any proof from its literal meaning. BT Ketubbot, 2b-3a.( “On account of the chaste women and on account of the loose women” (BT Ketubbot, 2b-3a). BT Gittin, 72b-73a. Chapter Four: Marriage Annulment: From Mishnah to Talmud 43 that in the above case the get becomes valid only after the husband’s death and this is the reason for its annulment when the husband recovered. Because of that fear, explains the Talmud, although the get is invalid by Torah law (since he recovered), the wife is (regarded as) divorced. Here too, the Sages enacted marriage annulment. (e) The husband sends the get to his wife by a messenger, but cancels the get (as 173 he is entitled to do) before the messenger delivers it. In order to prevent 174 extreme results, such as the wife’s remarrying unaware of the cancellation, Rabban Gamliel the Elder enacted that no one may cancel a writ of divorce before the wife receives it, unless in the presence of the messenger or his wife. His descendants, Rabban Shimon ben Gamliel and R. Judah haNasi (Rabbi), disputed the status of the get when the husband ignores Rabban Gamliel’s decree and cancels it. According to R. Judah haNasi, the get is void and the wife is not divorced, while Rabban Shimon Ben Gamliel does not void the get, and the wife is divorced. The reasoning behind Rabban Shimon ben Gamliel’s view is the authority and validity attributed to the Sages’ decrees, since otherwise: “how is the power of the Court [i.e., the 175 court of Rabban Gamliel, who enacted the regulation) [left] unimpaired!” But, the Talmud asks, if the writ of divorce is annulled according to Torah law, how can the Sages regard a married woman as a divorcee? The authority for that, explains the Talmud, is based on the concept of annulment of marriage. The Babylonian Talmud introduces the annulment in the last three cases in the following way: “Everyone who betroths [a woman], does so subject to the consent of the Rabbis, and in this case the Rabbis annul [his] betrothal” (kol demekadesh ada‘ata derabanan mekadesh ve’Afke‘inhu rabanan leKiddushin mineh; !"#$%#!&' ()*)%!+)",#)$&!+*-.!,(*)/$01,!"#$%!+*-.#!12/#1). The authority for the annulment is derived from a kind of preliminary consent: the husband betroths subject to the 176 willingness of the Sages, and such a stipulation gives the Sages the authority to 173 174 175 176 BT Gittin, 33a; Yevamot, 90b. See further BT Gittin, 33a, the various explanations of R. Yo!anan and Resh Lakish of this enactment, and cf. PT Gittin 4:2, 45c. Interestingly, Resh Lakish explains it as forestalling the problem of agunot, and according to Rashi’s commentary agunot here has the modern meaning: a married woman, whose husband (after cancelling the first get) refuses to divorce her: see Rashba, Gittin 33a, s.v. 1(,. This reasoning is found in the Babylonian Talmud, but is missing in the parallel texts in the Tosefta (Gittin 3:5) and the Palestinian Talmud (Gittin 4:2, 45c); see Weiss, Le!eker HaTalmud, 389 n.366. The wife probably does the same, otherwise it might be considered as a marriage made in error: see Shitah Mekubetset, Ketubbot 3a, s.v. "#$%(!&'. 44 Westreich: Talmud-Based Solutions to the Problem of the Agunah annul the marriage. This is derived from his saying: “according to the laws of Moses and Israel” (!"#$%&'($)'*+,), viewed as a form of condition, according to which the betrothal depends on the Sages’ consent. Another possible understanding of the authority to annul marriage is that it derives from the unique character of marriage as a legal and social institution, which was subject to the consent of the 177 Sages. The meaning of the concept of annulment of marriage is the subject to fundamental disputes among both traditional commentators and modern scholars. Hafka‘ah in cases (a) and (b) takes effect at the time of the betrothal and annuls the betrothal ab initio. It appears that in these cases the Sages invalidate the act of betrothal (by making the money ownerless or declaring the cohabitation to be 178 promiscuity) and thus the hafka‘ah prevents the betrothal from becoming valid. Annulment in cases (c), (d) and (e) takes place some period of time later, after valid betrothal (and probably also marriage) took place. So we may ask whether a similar legal construction is to be applied to these cases, i.e. is the betrothal annulled ab initio, resulting in retroactive annulment of the marriage? Or is hafka‘ah here prospective, taking effect only from that time on? According to the second possibility, annulment would refer to the status of marriage and not to the act of betrothal, by contrast with the previous reasoning. A related question is the role of the get in this process. If annulment is indeed prospective, a possible understanding of the ruling is that it validates the externally flawed get, which was not valid according to Torah Law. The get on this analysis is a substantive element in the process of hafka‘ah. If annulment is retroactive, a get is not necessarily required. As mentioned above, however, all the talmudic cases of delayed annulment do involve a get. Many Rishonim (but not all) regard this as supporting the view that demands a get in the process of hafka‘ah. However, this does not necessarily mean that a get is an essential component of the process of annulment. We may argue that though a get is indeed necessary, and hafka‘ah is thus limited to cases in which a get was given, this is due to various external reasons, such as preventing a “slippery slope” in the use of hafka‘ah, which will damage the stability of Jewish marriage, while conceptually the hafka‘ah remains a retroactive annulment of the marriage. 177 178 Ritva explains the statement “according to the laws of Moses and Israel” as a sort of condition ('&!%", 1%),-'&.#%$'*/)'!0'()0'(/*(). It might also be the view of Rashi: see Riskin, Hafka‘at Kiddushin (English version), 12-14. Others dispute this: see Berkovits, Tnai, 120-121, 134-135. In the definition of marriage as a social institution, I follow Atlas, Netivim, 207-209. The common denominator of the various explanations is that the authority to annul marriage is unique to marriage and divorce, and not part of a wider authority of the Sages.' See Lifshitz, Afke‘inhu, 318-319. In the talmudic sugyot however we find different approaches: see infra.' Chapter Four: Marriage Annulment: From Mishnah to Talmud 45 We may summarize the issue as follows: when a get is given, does hafka‘ah still annul the marriage retroactively or does it operate only prospectively, from the time of the giving of the (faulty) get? And is the get an essential element in the process? This issue in particular involves questions of both history and dogmatics, and is discussed later in this chapter. An additional issue is the authority of the Sages to enact marriage annulment. Do the Sages have the authority to annul marriages by virtue simply of their jurisdiction? Or may the power to annul be not the a priori authority of the Sages, but rather the agreement of the spouses? The latter view, although conceptually less radical than the former, is significant for both historical analysis and for its legal implication, in that it expands the normative basis for any suggested terminative condition as a possible solution for the present problem of agunot. Thus both the normative basis of the talmudic concept of hafka‘ah and the manner of its application in the various cases are critical. Analysis of the historical development of the talmudic concept of hafka‘ah may assist us in answering these questions. 4.3 The Foundation of Hafka‘at Kiddushin It seems that all the possible approaches mentioned above regarding both the character of hafka‘ah and the authority of the Sages to enact it are already found in the talmudic sources. As a starting point, however, we should indicate the earliest 179 talmudic source which discusses hafka‘ah. In cases (a) and (b) annulment is mentioned by Rav Ashi or Mar bar Rav Ashi, 6th-7th generation Babylonian 180 Amoraim. In cases (c) and (d) annulment is mentioned by the anonymous stratum of the Talmud as an explanation of several amoraic statements and appears to 181 belong to a late chronological stage of the Talmud. Case (e) is a discussion rd between two 3 generation Babylonian Amoraim of a tannaitic source: the dispute between Rabbi and Rabban Shimon ben Gamliel regarding cancellation of a get which was sent by a messenger. It seems therefore that this case is the source of the concept of marriage annulment, and in what follows I shall support this argument. 179 180 181 See Weiss, Le!eker HaTalmud, 393. See n.169 above. See Friedman, Ha’Isha Rabbah, 283-321; Halivni, Mevo’ot. Robert Brody has recently criticized Friedman and Halivni’s approaches: see Brody, Stam HaTalmud. However, his list of early anonymous passages in tractate Ketubbot (ibid., 228-232) does not include case (c) above, although it belongs to this tractate. I assume therefore that Brody agrees that this case, and similarly case (d), belong to a later talmudic stratum. This leaves us with case (e) as the earlier source of marriage annulment, as discussed further in this section. 46 Westreich: Talmud-Based Solutions to the Problem of the Agunah In the case of the messenger (e), Rabban Shimon ben Gamliel rules that the wife is divorced despite the cancellation of the get. This is interesting from a conceptual point of view, and both Talmudim discuss it. In the Palestinian Talmud we find as 182 follows: !0'+"/!#/%#+4!1&!1"(+3!1&'!5%&'!%'&*!#$"&+!,-!%',!"#$%&!./!0/*,!1+!,2%(+3%2!)!"#$%&"!'&( :"/26!#(!7%8",#!/#"!"#$&#!#"9%!"2%/ The Yerushalmi first cites the dispute between Rabbi and Rabban Shimon Ben Gamliel when the husband ignored Rabban Gamliel the Elder’s enactment and cancelled the get. According to Rabbi, despite the breach of the enactment, the get is void and the wife is not divorced. According to Rabban Shimon Ben Gamliel, on the other hand, the get is not void, and the wife is divorced. !5#$%&!/#3!"'+/!1,"!#$&%3!/",!,'"6!'&*!;%&'*!/+($!%/+!5#/%#+4!1&!1"(+3!1&'!'+/!6"/% =;,'"6!%'&*!1%'<"(!1,%'&*" Rabban Shimon Ben Gamliel’s view is reasonable. So what is the reasoning behind Rabbi’s ruling? Rabbi, according to the Yerushalmi, claims: how can you, Rabban Shimon ben Gamliel, say that the get is valid: “the Torah said that [the get] is void [when the husband cancels it], while they [= the Sages] said that it is not void [i.e., the husband’s cancellation is invalid] – can their words uproot the words of the Torah?!” !/#"!5."'6%!/#3!"'+/!1,"!$&3,!#-4!%2>+!."'6%3!/",!,'"6!/#!1%%,!#(!.%&2("!.%6%-!#(!1+3!%9" =,+"'6!"6+"'6!1%/!.'6"!'&(!"'+/3!/#/!*"( This passage is Rabban Shimon ben Gamliel’s answer. He doesn’t state explicitly that the Sages do have that authority. He rather proves it from a different case in which what was regarded according to the Torah as terumah (a ritual giving of a certain percentage of the agricultural produce to a priest) could be cancelled (and 183 defined as non-terumah, i.e. !ullin) by the Sages. Surely, in marriage, as in the case of terumah, the Sages do have the authority to uproot the words of the Torah. :1#!<%%*!1/+!@&8*!,*4/&!0/%%32!1*"%!%&'#!'+/!/&/!'&!,%(3"/!?' In the final passage Rabbi Osha‘aya bar Abba rules in favour of Raban Shimon ben 182 183 PT Gittin 4:2, 45c. The following passages are short and difficult to explain. Therefore I do not translate it literally, but rather cite each Hebrew passage and explain it in a more detailed manner.! The consequence of this act is far reaching: after the ruling of the Sages there is a permission for a regular person (a stranger, '-, i.e. not a priest) to eat the fruits (assuming that another terumah was made), while according to Torah law they are considered as a terumah, forbidden to any person who is not a priest, and their eating results in the severe punishment of death performed by heaven (mitah biyde shamayim). Chapter Four: Marriage Annulment: From Mishnah to Talmud 47 Gamliel. Rabbi Osha‘aya bar Abba claims against Rabbi Yehuda Nesi’a, Rabbi’s grandson, that [nearly] no one can understand (!"#$%%&#!'() his grandfather’s ruling ()*+&#,&-'). The core of the dispute between Rabbi and Rabban Shimon ben Gamliel according to the Yerushalmi is whether the Sages have the authority to rule against the Torah, including declaring a married woman as a divorcee. According to Rabban Shimon ben Gamliel, this authority does exist, here as well as in other cases, and does not depend on any preliminary consent of the spouses (which is irrelevant to the case he compares: terumah). 184 The Palestinian Talmud here deploys a concept similar to hafka‘ah: the wife is considered as a divorcee despite the defect in the get since “the Sages uproot the words of the Torah”. This very explanation is found in the Babylonian Talmud in 185 the name of Rav !isda, in the following sugya. 186 In a different context, Rav !isda and Rabbah, two third generation Babylonian 187 Amoraim, discussed whether the Sages have the authority to uproot the laws of the Torah (,./0,#!(#.*&#./$1"#!%20(#!%&#0%*). According to Rav !isda, the Sages do 188 have such an authority, while Rabbah challenges his view. One of Rav !isda’s proofs is Rabban Shimon ben Gamliel’s view in the case of a cancelled get. Rabbah then replies: “Everyone who betroths [a woman], does so subject to the consent of the Rabbis, and the Rabbis annul his betrothal” (#!2*.&#'01&'#3&$(&#!'( !%3/&%$"# !2*.# /,2%1$4'/# 53&$(). According to Rabbah, the Sages do not have the authority to uproot the words of the Torah. Rather, their authority to rule that the wife is divorced is derived from the unique character of betrothal, which was made 189 subject to the consent of the Sages. 184 185 186 187 188 189 The precise meaning of the hafka‘ah in the Yerushalmi (which we may also define as: “quasihafka‘ah”) will be discussed below. BT Yevamot, 89b-90b. The whole sugya is extensively analyzed in Friedman, Ha’Isha Rabbah, 346-357. For further discussion on the literary aspects of this sugya see Westreich, Hafka‘at Kiddushin. The context is the laws of terumah: a case in which according to the Torah the act of terumah was valid, but the Sages invalidated it. The similarity between the Bavli and the Yerushalmi is apparent; see further below. See Albeck, Mavo, 289-290; 307-308. The generation is significant: teachings of third generation Babylonian Amoraim are still found in the Palestinian Talmud, while those of later generations rarely exist; see Zussman, Veshuv LiRushalmi Nezikin, 98-99, and notes 178a, 179. In the specific context in which Rav !isda initially expressed his view it was Rav Natan bar Rabbi Osha‘aya who was in dispute with him. However, the general discussion regarding the authority of the Sages to uproot the words of the Torah was between Rav !isda and Rabbah. Rabbah could have replied that his view is according to Rabbi, who rejected the authority of the Sages to uproot the words of the Torah and disputed Rabban Shimon ben Gamliel in this point. Nevertheless, Babylonian Amoraim usually preferred to minimize the scope of tannaitic disputes Westreich: Talmud-Based Solutions to the Problem of the Agunah 48 One may argue that the discussion between Rabbah and Rav !isda is a later expansion of the basic amoraic dispute, and was actually inserted by later editors. If this is correct, the concept of marriage annulment cannot be ascribed to Rav !isda and Rabbah. However, here this is not the case. The discussion between Rav !isda and Rabbah was indeed wide and complex in its origin and included several arguments for each side. It didn’t happen on just one occasion but was a continuing 190 debate which included several participating Amoraim. Therefore it is most reasonable to see the “messenger passage” as part of the actual discussion between these two scholars, both in regard to Rav !isda’s argument and in regard to 191 Rabbah’s response. This conclusion follows S.Y. Friedman’s analysis of the sugya. Friedman points to the literary character of the debate, which included seven arguments (a typological number). We find corpora of seven arguments in other debates between Rav !isda and Rabbah (see Eruvin, 43a). Therefore, concludes Friedman, the impression is that we have here “a full and defined corpus”, rather than “a continuing accumulation of proofs, some of whom were 192 added after the time of Rav !isda”. The similarities between the Bavli and the Yerushalmi cannot be overstressed. Both discuss “uprooting the words of the Torah”, almost in the same words. And in both the sugya has a similar structure, which includes the precedents for both 193 invalidating terumah and validating a cancelled get. Apparently there was an interaction between the two Talmudim and either the basic Palestinian sugya was transmitted to Babylonia and adopted by Rav !isda (while expanding its spectrum 194 to a few other cases) but rejected by Rabbah, or Rav !isda’s view was 195 transmitted to Erets Israel. According to the latter alternative, the process was as ---and not to create parallelism between amoraic disputes and tannaitic ones: see Goldberg, Tsimtsum Ma!lokot, 139-142. Rabbah preferred therefore to explain Rabban Shimon ben Gamliel in accordance with Rabbah’s own view. 190 191 192 193 194 195 See BT Yevamot, ibid.: “Rav Hisda sent to Rabbah through Rav Aha son of Rav Huna [...] He said to him: it was my intention to raise objections against your view from [the Rabbinical laws which relate to] the uncircumcised, sprinkling, the knife [of circumcision], the linen cloak with tzitzit, the Shavuot lambs, the shofar, and the lulav.” It is unlikely to assume that Rabbah’s original response was omitted from an unknown reason, with no real indication for it, and the current response was added by a later talmudic stratum. See Friedman, Ha’Isha Rabbah, 350-351. Friedman argues that only the second half of the sentence is Rabbah’s original statement, but there is no reason, neither literary nor conceptually, to make such a distinction. This parallelism is an additional indication that the “messenger case” is an integral component of the general sugya of uprooting the words of the Torah. This is a common phenomenon: see for example Dor, Torat Erets Israel. See Zussman, Veshuv LiRushalmi Nezikin, 98-99, and ibid., notes 178a, 179; Florsheim, Rav ! isda, 126-131, indicating a few of Rav !isda’s students who transmitted his teachings. Chapter Four: Marriage Annulment: From Mishnah to Talmud 49 follow: Rav !isda and Rabbah argued; both supported their views; Rav !isda supported his argument from Rabban Shimon ben Gamliel’s opinion; Rabbah rejected that support. This debate was partially transmitted to the Yerushalmi, which discusses the main argument citing some of the sources, but retains the simple meaning of those sources without any mention of Rabbah’s final conceptual 196 development. We can now describe more precisely the process by which the concept of hafka‘ah was constructed. First, a tannaitic source – Rabban Shimon ben Gamliel’s view – validated an invalid get based on a decree of the Sages. This source describes an act of marriage annulment, or quasi-annulment, but yet in a casuistic 197 formulation rather than conceptualized. Then early Palestinian and Babylonian Amoraim (Rav !isda) based this on the general principle of the Sages’ authority to uproot the words of the Torah. Rabbah rejected this view. He did agree with Rabban Shimon ben Gamliel that the Sages have the authority to validate the divorce, but suggested an innovatory reasoning: the Sages have a specific authority to annul the marriage, since “Everyone who betroths [a woman], does so subject to 198 the consent of the Rabbis, and the Rabbis annul his betrothal.” Rabbah formulates a new and innovative concept: the concept of marriage annulment. According to the Palestinian Talmud and Rav !isda, on the other hand, no such an innovation is required. The Sages indeed have a wide authority to uproot Torah Laws, thus they may declare the get as valid, despite its cancellation by the husband. Though this view was rejected by Rabbah, it was revived a few generations later by Rav Ashi. 199 In the case of Naresh Rav Ashi explains that the annulment of marriage is a result of the misconduct of the “kidnapper”: 5!"#"$%!"&'(")*%+#,-%'!#".)/0'%+1'!2%0*&%',%'&.%32"/*%4+1'!2%0*&%!&.%0'!% 196 197 198 199 Rabbah is hardly mentioned in the Yerushalmi (if at all; see Zussman, ibid., 131-132 n.179). In our case as well, only Rav !isda’s view is reflected in the Yerushalmi (maybe as a result of Rav !isda’s students’ transmission: see previous note) while there is no reflection of Rabbah’s view. See Moscovitz, Talmudic Reasoning, 1-6. It is more evident in the Yerushalmi and the Tosefta, where even the general reasoning of “since if so, what becomes of the authority of the bet din” is missing; see infra, n.217. According to the present analysis, the source of the concept of marriage annulment is the talmudic sugya in Yevamot 90b. Accordingly, the parallel sugya in Gittin 33a is a shortening of the original one. The basic discussion between Rabbah and Rav !isda is presented there anonymously, and Rabbah’s approach is the final conclusion of that passage (compare Gilat, Perakim, 201). Yevamot, 110a; see case (a), supra, text at nn.164-165. According to some talmudic variants the source of this statement might be case (b): see n.169 above (for current purposes this issue doesn’t have any further implications). 50 Westreich: Talmud-Based Solutions to the Problem of the Agunah He acted improperly; they, therefore, treated him also improperly, and the Rabbis annulled his betrothal. Rav Ashi’s explanation is composed of two different parts: one completely in Hebrew (“he acted improperly; they, therefore, treated him also improperly”, !"#$ %&#$'! "()! #*! #)+! ,'-.(! /%&#$'! "()! $)+) and one in Aramaic (“and the rabbis annulled his betrothal”, $-0-1!$-)#2-3(!%0*4!#$0-+3."#). The shift from one language 200 to another indicates that his teaching might be based on two different sources. Obviously, the second part – the Aramaic – is a quotation of Rabbah’s explanation 201 of the authority of the Sages to annul marriage in the case of the cancelled get, which, as we have argued, is an original statement of Rabbah. Rav Ashi uses the core of Rabbah’s teaching – the marriage annulment concept (afke‘inhu) – as a stock phrase but omits its first part, which bases the authority to annul marriage on the previous consent of the husband (“everyone who betroths does so subject to the consent of the Rabbis”, )231!%0*42!"5+2"!)2312!('). Instead, he cites a different reasoning whose sources are found in the teachings of the Amoraim of earlier 202 generations, such as Rav Hamma’s regarding the improper act of a debtor, or 203 even in tannaitic sources. What stands behind Rav Ashi’s new formulation?! Some talmudic commentators restored the omitted part of Rav Ashi’s teaching, the principle that “Everyone who betroths does so subject to the consent of the 204 Rabbis”, as Rashi writes: ! [The Sages] annulled his betrothal: since every one who betroths does so subject to the consent of the Sages ("#$! 6-1'7! 5+2*! $(#5! )2312! ('2), as we say: “according to the laws of Moshe and Israel” (("4)-#!$)1!52'). And the Sages said that where one kidnaps a wife from her (intended) husband the betrothal is not valid. Rashi’s view was challenged by Tosafot. Tosafot argue that since the mere act of betrothal was against the will of the Sages, how can we say that he betroths subject 200 201 202 203 204 The use of different languages in an amoraic statement is usually allows us to distinguish between the basic amoraic statements and later anonymous additions: see Friedman, Ha‘Isha Rabbah, 301 (but not always: see Friedman, Tosefta Atikta, 434 n.52). Nevertheless, this method can be expanded to make a distinction between the various sources of the amoraic statement itself: see Westreich, Torts, 52 n.60. See supra, text at notes 188-189. This concept is used by Rav Hamma in order to explain a verdict of Rava to Rav Papa; see BT Ketubbot, 86a. The general idea that an improper act (%&#$'!"()) prompts an improper response is found in several sources. See for example BT Yoma, 75a. See also Ben Menahem, Hu ‘Asa, 157, who suggests that a dispute between Bet Shammai and Bet Hillel regarding mi’un (BT Yevamot 107a) is the source for that concept. His suggestion however is based on a general substantive similarity rather than on literary parallelism: Bet Shammai does not in fact use there the language of %&#$'!"(). BT Yevamot, 110a, s.v. "3#. Chapter Four: Marriage Annulment: From Mishnah to Talmud 51 205 to the consent of the Rabbis (!"#$% &'()"% *+,"*). Some commentators reply to Tosafot’s query. Maharam of Rothenburg for example explains (following Rashi), that although he acted here in a rude way, he didn’t mean to act against the will of the Sages and therefore we can say “everyone who betroths does so subject to the 206 consent of the Rabbis”. Nevertheless, according to Maharam, in cases where he did intend to act against the will of the Sages, we cannot say “everyone who betroths does so subject to the consent of the Rabbis” and we cannot annul the marriage. 207 Maharam’s explanation, as well as those of others, do not deal with one central argument against Rashi’s view. If Rav Ashi had only shortened Rabbah’s statement, Rashi’s explanation would have been preferred. But Rav Ashi replaced Rabbah’s reasoning by a different one. This fact is significant. Its meaning is that according to Rav Ashi, we do not need Rabbah’s explanation since we have the alternative reason: “he acted improperly; they, therefore, treated him also improperly”. Rav Ashi thus rejects Rabbah, and opposes his view, that the authority of the Sages depends on the unique character of marriage (“everyone who betroths does so subject to the consent of the Rabbis”). Rather, Rav Ashi is close to Rav !isda: the Sages have by definition the authority to annul marriage, since, we now may add, the Sages have the authority to uproot the words of the Torah when required. In the case of Naresh they decided to use that authority due to the misconduct of the “kidnapper”. 208 This explanation follows Tosafot, in that Rav Ashi’s view is based on the authority of the Sages rather than on “everyone who betroths does so subject to the consent of the Rabbis”. It is reasonable to conclude accordingly that Rav Ashi opposes Rabbah. Tosafot however do not say that Rav Ashi accepts Rav !isda, and consequently the sugyot that mention kol demekadesh are in dispute and follow 205 206 207 208 See Tosafot, Bava Batra, 48b, s.v. -'.+, and see also Ri in Tosafot, Yevamot 110a, s.v. /0.12, who leaves this issue without decision. Cited in Mordekhai, Kiddushin, §522. Maharam defines the act of the kidnapper as mere insolence (*$2,(%*134-). For a different explanation of Rashi’s view see Atlas, Netivim, 207-209, according to whom the authority to annul marriage derives from the unique character of marriage as a legal and social institution, which was subject to the consent of the Sages, and does not depend on the intent of the husband: “the primary validity of marriage comes from the laws of Moses and Israel, while the Sages established the form and conditions of the institution of marriage. If one acts improperly [...] even though he does not act in accordance with the opinion of the Sages, nor does he heed their will, he cannot extricate himself from the authority of the Sages by changing the form and the conditions for validating the institution of marriage”. See also Edrei, Koa! Bet Din, 34; Wieder, Rejoinder, 76 n.21. Supra n.205. This view was adopted by Weiss, Le!eker HaTalmud, 391-392. 52 Westreich: Talmud-Based Solutions to the Problem of the Agunah Rabbah. Rather Tosafot harmonize Rav Ashi with the other sugyot of kol demekadesh by arguing that even according to Rav Ashi when the betrothal was valid, and the annulment is applied only later, the concept of the authority of the Sages to uproot the words of the Torah is not sufficient and we need kol demekadesh as a support. Here however we follow the simple meaning of Rav Ashi, according to which he revives Rav !isda’s expanded view of the authority of 209 the Sages. One comment should be made here. Following Rav Ashi’s ruling, Ravina agrees that when the betrothal was effected by money (kiddushei kesef) the Sages could annul it. Nevertheless he wonders how the Sages could annul the betrothal when it 210 was effected by cohabitation (kiddushei bi’ah). According to the present analysis, Ravina did not challenge the authority of the Sages to annul cohabitation by betrothal. Rather, he discusses the procedure, and thus the legal construction, by which the Sages annul the betrothal. Since we are dealing here with annulment of 211 the validity of an act of betrothal, the act itself should be defined as an act which does not effect betrothal. There was less difficulty in the case of betrothal by money, since here the annulment could be understood as due to the authority of the 212 Sages to declare money as ownerless. But in case of betrothal by cohabitation, Ravina challenged Rav Ashi: how (and not by what authority) could the Sages give 213 the act a new meaning which would affect its legal validity? Rav Ashi then answers that this is possible by declaring his cohabitation to be an act of mere 209 210 211 212 213 As regards the expanded authority of the Sages see also Tosafot, Ketubbot 11a, s.v. !"#"$%&, who ascribe this view to Rav Huna as well: in some circumstances a bet din may supply the consent in conversion of a minor, but according to Ri such a conversion is only by rabbinic law (miderabanan) and not valid according to Torah law (mi-de’orayta). The converted man may now marry a Jewish woman even though he is a gentile mi-de’orayta, since, explain Tosafot, the Sages have the authority to uproot the words of the Torah. The expanded authority of the Sages is suggested also by Rabbi Akiva Eiger (Gittin, 33a, s.v. '()"*+,-') as an explanation of the view of Rabbi (!): According to R. A. Eiger, Rabbi agrees that when the husband cancelled the get in the absence of any bet din, the marriage is annulled. The annulment however is not based on “kol demekadesh” but rather on “uprooting the words of the Torah”: “for in an instance in which there is an important reason, the Sages have the power to uproot the words of the Torah” (.-&*%#./"01.!"(0'2(.!&.0$3.0'+*#.4"&56.3"$.65.7".$'76.4*%.4'+&$3.-#-.8'()"*+,-3). For additional sources who support an expanded authority of the Sages see Gilat, Perakim, 201-204; Elon, Jewish Law, 521533. For full citation and analysis of Ravina and Rav Ashi’s discussion see below. See supra, text at note 178. See Rashi, Yevamot, 110a, s.v. 6)"2 (and in all the other occurrences of Ravina and Rav Ashi’s discussion). See for example Rashi, Yevamot 90b, s.v. 7"3+ (and in slightly different words in the other occurrences): “when he betroths by cohabitation, what [kind of] annulment can you apply here [lit. can be said], how did they define the cohabitation?” (.(-"$."-(.80&"&#.-5"-.-2*+,-."-&.9(-"$$.7"3+ :('"'7."-&). Chapter Four: Marriage Annulment: From Mishnah to Talmud 53 promiscuity (see below).! The conclusion of this section is of the greatest importance. Rav Ashi is a later generation Amora, and his decisions are generally accepted. What makes it more 214 decisive in our case is the possibility that it was accepted also by his son, Mar bar Rav Ashi. The implication of the above analysis is therefore that the final talmudic stage accepts a significantly expanded authority of the sages as initially suggested by Rav !isda and by the Yerushalmi in its interpretation of the view of Raban 215 Shimon ben Gamliel. The Sages, hence, have the wider authority, the authority to uproot the words of the Torah. 4.4 The Character of Hafka‘at Kiddushin How does hafka‘ah work – is it a retroactive annulment of marriage or is it prospective, i.e. an act terminating the marriage only from now on? And what if at all is the role of the get in this process? In order to examine this issue we must return to the case of the get messenger, which provides us with the earliest source 216 for annulment. According to Rabban Shimon ben Gamliel, the husband cannot cancel a get which was already given to an agent (messenger) to deliver to his wife in the absence of the agent or the wife. In his words: 6"#$!%$&!'$(!)*+!",!%+!-./!*.0'!12!3$4*"1!.1*!*15(1!.1!1*+$!*0$. ! He (the husband) can neither cancel it nor add any additional conditions, since if so, what 217 becomes of the authority of the bet din ("#$!%$&!'$(!)*+!",)?! This is quite explicit: the husband cannot cancel the get, so the get is valid. The Sages act here by validating the get (or, more precisely, by preventing the husband 218 from invalidating the get), rather than by actively annulling the marriage. This view seems to be shared also by the Palestinian Talmud, which merely discusses 219 the cancellation of the get and its validation by the Sages. 214 215 216 217 218 219 Depending on the exact version of case (b); see supra, n.169, and infra, n.241. Tosafot try to harmonize Rav Ashi with the view of kol demekadesh ada‘ata derabanan mekadesh by contrast with the conclusion of this section. However, even according to their view the result is some expansion of the authority of the Sages to uproot the words of the Torah. See further supra, text at notes 208-209. See supra, text at notes 179-182.! BT Gittin, 33a; Yevamot 90b. In the Yerushalmi, Gittin 4:1, 45c (cited above, text at note 182) and in the Tosefta, Gittin 3:5, the reasoning of Rabban Shimon ben Gamliel’s ruling (“since if so, what becomes of the authority of the bet din”) is missing. This can be done by removing the legal power of the husband to cancel the agency of the messenger. See supra, text at notes 182-184. Edrei, Koa! Bet Din, 34 n.121, identifies this view as the view of the Yerushalmi, but argues for a different view in the Bavli. See also Gilat, Perakim, who explains 54 Westreich: Talmud-Based Solutions to the Problem of the Agunah But Rabbah in the Babylonian Talmud explains Rabban Shimon ben Gamliel’s ruling in a slightly different way: When a man betroths a woman, he does so subject to the will of the Rabbis, and [in this case] the Rabbis annul his betrothal (!"#$%&#'(!")*+!%,)#-'./%!$&'0!")*+&!/1-&/!$&'0&!(2 ,#)#0). The judicial act here is not validating the invalid get. The get is not valid since it was cancelled by the husband. However the couple are divorced since the betrothal 220 is annulled. Why should the Talmud make such a shift in explaining Rabban Shimon ben Gamliel’s ruling? As we have argued above, this case is the source of the concept of hafka‘at kiddushin, and its development is a result of the discussion between 221 Rav !isda and Rabbah. The current shift between validating the get and annulling the marriage is part of that dispute: according to the first approach we need to assume that the Sages have the authority to uproot the words of the Torah, as Rav !isda argues. Rabbah therefore explains Rabban Shimon ben Gamliel’s ruling as a 222 result of the unique character of Jewish marriage and thus rejects the view that the Sages can uproot the words of the Torah. Nevertheless, if the authority of the Sages is (only) in relation to the marriage, and not wider, we must explain Rabban Shimon ben Gamliel’s ruling as annulling the marriage, rather than validating the cancelled get, since these are the limits within which the Sages may act. Fascinatingly, although Rabbah’s explanation diminishes the authority of the sages (since, as he argues, the Sages do not have the wider authority to uproot the words 223 of the Torah ), regarding marriage his view (marriage annulment) gives much more authority to the Sages than according to Rav !isda’s view (validating an existing get). Thus, according to Rav !isda, divorce is performed by a writ of divorce, like any other Jewish divorce, while the Sages prevent its cancellation. According to Rabbah, on the other hand, in the messenger case the Sages constitutively annul the marriage without the (ritual) act of divorce. Rabbah however, prefers this approach as regards marriage, due to his objection to ascribe a wide and general authority to the Sages to uproot the words of the Torah. What is the meaning of marriage annulment in Rabbah’s teaching? I argue that ---the Yerushalmi in this way. Some scholars and classic commentators interpreted the Bavli in a similar way: see below. 220 221 222 223 See Edrei, ibid., 34-35. See supra, section 4.3. See supra, text at notes 176-177. Rabbah, however, agrees that the Sages do have the authority to uproot the words of the Torah in a passive manner, without any active act: see BT Yevamot, 90a-b (#)/$!,$-1!(/%!*$). Chapter Four: Marriage Annulment: From Mishnah to Talmud 55 it is more reasonable to understand it as prospective annulment rather than retroactive annulment. Interpreting annulment of marriage as retroactive is much more drastic both conceptually and practically (declaring cohabitation to be 224 promiscuity, the possible effect on the status of the children, etc.). Moreover, hafka‘ah in the talmudic context is usually prospective rather than retroactive. It means to cancel, to cause to cease, or the like, usually in the context of 225 (prospective) annulment of a legal status or voiding the validity of a legal act. 226 Thus, for example, the act of !alitsah nullifies the levirate bond (zikah); the 227 Sabbatical year cancels one’s debts and so on. Marriage annulment in our context is no different: the status of a betrothed couple is prospectively taken away from 228 the couple and the couple is not considered married from that moment on. Thus, in Rabbah’s teaching hafka‘ah means prospective annulment of the marriage. However, it is not the final meaning of that concept in the Talmud. Later, Rav Ashi implied Rabbah’s concept of annulment in a case of which “he acted 229 230 improperly”. In Rav Ashi’s case, the case of Naresh, the act of betrothal was performed improperly and the Sages sought to annul its validity. This was done by expropriating the betrothal money or by declaring the cohabitation to be promiscuity (bi’at zenut), as in the discussion Rav Ashi and Ravina, that follows Rav Ashi’s statement: 224 225 226 227 228 229 230 I.e. declaring them not to be mamzerim: see Tosafot, Gittin 33a, s.v. !"#$%&'(!, and elsewhere.) See Sokoloff, Dictionary, 158 (translating (*%&'( as suspension) and 925-926 (translating the root %+&' as to rupture, split, cease, cancel, confiscate). The root PQA (!"#),%+&') has a similar meaning both in Mandaic and Syriac; see Drower and Macuch, Mandaic Dictionary, 376; Brockelmann, Lexicon Syriacum, 590.) See Yevamot 52b. See also Rashi, ibid., 50a, s.v. "-$./, who uses the root %&' to describe prospective annulment of marriage by either get or !alitsah ()$%!&'(0)12)$%3!)"*&$40)(*%&'()"/$056 "$0$6)7$8!6&). See Shevuot 58b. For more examples see Sokoloff, ibid. Sokoloff, ibid., translates hafka‘ah in our context as follows: “the scholars (retroactively) nullified his betrothal”. The word in brackets is the translator’s addition based on the common interpretation of the concept. Nevertheless, the meaning of the concept in the specific talmudic context (as opposed to later stages; see below) is as I have argued above: prospective annulment of the status of the couple. See also Halivni, Mekorot, 530, according to whom hafka‘ah at this stage is retroactive, but there is still a distinction between this stage and the discussion of Ravina and Rav Ashi: here, since annulment is based on the prior consent of the husband (86&9)7#3.6)(*%6()86&96)0-), we do not need the Sages to declare his cohabitation to be promiscuity. See supra, text at nn.199-203. Case (a), supra, text at nn. 164-165. 56 Westreich: Talmud-Based Solutions to the Problem of the Agunah Said Ravina to Rav Ashi: [Your explanation is] satisfactory where the man betrothed 231 [her] with money; what [however, can be said where] he betrothed her by cohabitation? [Rav Ashi answers that] the Rabbis have declared his cohabitation to be an act of mere promiscuity (!"#$%!&'()%"!&'()&%*#)+%,"'"-). The discussion between Rav Ashi and Ravina is a discussion regarding how the 232 Sages could effect the legal validity of the act of betrothal, as argued above. The shift here is between annulling the status of marriage according to Rabbah, in a case of which the annulment is executed after the couple was already married (a delayed annulment), and annulling the validity of the act of betrothal, in a case of 233 which the betrothal itself is improper (an immediate annulment). At this stage however, annulment is still prospective, according to Rav Ashi as also according to Rabbah, although different in its meaning: hafka‘ah for Rabbah is annulment of marriage from the time of cancelling the messenger’s agency and breaching Rabban Gamli’el’s enactment, whereas for Rav Ashi hafka‘ah means annulment of the act of betrothal which occurs immediately when the husband betroths the 234 woman improperly. But the development of the concept of marriage annulment was not completed by Rav Ashi. Rav Ashi and Ravina’s discussion is cited in all of the five talmudic 235 cases of hafka‘at kiddushin, after arguing for the annulment of the marriage. It is unlikely that this discussion occurred five times. One of its occurrences (case (b)) 236 refers according to many textual witnesses to a statement of Mar bar Rav Ashi, who was the son of Rav Ashi – which makes the possibility of an original discussion between earlier Amoraim (his father, Rav Ashi, and Ravina) even less 237 likely. Moreover, as Tosafot indicate, there are some interpretative difficulties 231 232 233 234 235 236 237 Kesef (money) and bi’ah (cohabitation) are two of the forms of betrothal (Mishnah, Kiddushin 1:1). The Sages have the authority to confiscate a man’s property (+./,% *'0% !')% +./,), so they might regard the money given by the man as a mere gift to the woman. See Rashi, Yevamot, 110a, s.v. 1#'!. See supra, text at notes 208-213. See Lifshitz, Afke‘inhu, 318-319. See Riskin, Response, 44, and compare Wieder, Rebuttal, 37. In my opinion, annulment in Rav Ashi’s case is prospective, as argued here (see also Westreich, Hafka‘at Kiddushin, n.89). See supra, text at nn.164-175. Supra, n.169. There were two or three Amoraim named Ravina. Ravina in our case is Rav Ashi’s disciple-friend (5th generation and perhaps later: see below). As for the other Ravina, his dates are unclear and disputed amongst scholars (see Albeck, Mavo, 421; Cohen, Ravina, 256-261). According to Albeck, the Ravina who had relations with Mar bar Rav Ashi is a later Ravina, from the 7th generation (see Albeck, ibid., 448-450). According to him, the Ravina of our discussion certainly never met Mar bar Rav Ashi, and couldn’t discuss his statement. According to Cohen, ibid., the Ravina of our discussion (the main Ravina of the Talmud) died after Rav Ashi and had relations with 6th and 7th Chapter Four: Marriage Annulment: From Mishnah to Talmud 57 when we read the discussion in the context of the delayed annulment cases – cases 238 (c)-(e). We may conclude therefore, as some scholars have already indicated, that the discussion occurred originally in one case, probably the case of Naresh (case 239 [a]), or according to some, in the case of the coerced betrothal, “the hanging” 240 241 (case [b]) , where it follows a statement of Rav Ashi himself. Indeed, we clearly need the explanation of “the Rabbis have declared his cohabitation to be an act of mere promiscuity” (!"#$%!&'()%"!&'()&%*#)+%,"'"-) in these two cases: the annulment is required to invalidate the improper act of betrothal of the “husband”. The legal construction here is therefore hafka‘ah by expropriating the betrothal money or by declaring the cohabitation to be promiscuity (bi’at zenut). Later, a talmudic redactor added Ravina and Rav Ashi’s discussion to all other occurrences of the concept of annulment. According to Rav Ashi the Sages “declared his cohabitation to be an act of mere promiscuity”, and as a result of the talmudic redactional work it became the reasoning of annulment in all talmudic cases. Yet, in cases (c)-(e), the delayed annulment cases, the annulment takes place some period of time after the couple married. If indeed his cohabitation was declared as an act of mere promiscuity, it must mean that the marriage is retroactively annulled. Prospective annulment of marriage does not require declaring the cohabitation to be promiscuity, but rather leads to termination of an actual marriage and the past cohabitation does not affect (and is completely irrelevant to) the annulment. It is a dramatic conceptual change: hafka‘ah at this final stage became retroactive annulment of the act of betrothal. The change between annulling the status of marriage and annulling the marriage ---generation Amoraim, including Mar Bar Rav Ashi. However, even according to Cohen it is unlikely in my opinion that Ravina and Rav Ashi had a discussion regarding a statement of Rav Ashi’s son, Mar bar Rav Ashi. 238 239 240 241 See Tosafot, Ketubbot 3a, s.v. .#'!: Tosafot implicitly ask why we need the explanation that “the Rabbis have declared his cohabitation to be an act of mere promiscuity” only for betrothal by cohabitation – it is surely necessary also for betrothal by money, since after betrothal there would have been some cohabitation which needed to be declared to be promiscuity! (For explanation of this Tosafot, see Maharam Schiff, ibid.) BT Yevamot, 110a. BT Bava Batra, 48b. See Sho!etman, Hafka‘at Kiddushin, 354-355; idem, Ones, 118-119; Halivni, Mekorot, 530 n.2. I. Franzus argues that the case of the coerced betrothal (the “hanging”) should properly refer to “Rav Ashi” and not “Mar bar Rav Ashi” (see supportive textual witnesses supra, n.169), and the source for the statement and the following discussion is this case; see Franzus, Od LeKol Demekadesh, 9192. His view was later accepted by Atlas, Netivim, 242. In my opinion, Rav Ashi’s original teaching and the discussion that follows occurred in the case of Naresh (see supra, n.169). Anyway, all these scholars agree that the discussion between Rav Ashi and Ravina originally occurred in none of the second group of cases – cases (c) to (e). The discussion below is therefore consistent with both approaches. % 58 Westreich: Talmud-Based Solutions to the Problem of the Agunah act was first made by Rav Ashi, who applied Rabbah’s concept to his case, probably in the case of Naresh (a), and was adopted by his son, Mar bar Rav Ashi, 242 in the case of the coerced betrothal. They did not apply it to the other talmudic cases of marriage annulment. However, Rav Ashi’s move made the next step of the talmudic redactor possible: viewing hafka‘ah in all the five cases in a similar way and consequently understanding it as a retroactive annulment. Thus, the transfer of the discussion to cases (c)-(e), as described above, entails our explaining hafka‘ah as retroactive annulment. What motivated this quite significant conceptual change? While the first development in understanding hafka‘ah, from validating the get to annulling the marriage, is the result of a conceptual process (i.e. the debate between Rav !isda and Rabbah), and the next development, from annulling the status to invalidating the validity of the legal act, is the result of applying the concept of hafka‘ah in new circumstances (i.e. by Rav Ashi in a case of improper betrothal), the last move, from prospective to retroactive annulment, is merely a result of redactional work. Nevertheless, I assume that it was done with awareness. Transmitting the discussion to a group of cases reflects a quest for harmonization: since a similar concept is mentioned in these few cases, the later talmudic view sought harmony in its meaning and implications. Thus hafka‘ah became a legal concept which refers to the act of betrothal not only in cases of improper betrothal, but also in the cases of improper divorce. In those cases the meaning of hafka‘ah thus became 243 retroactive annulment of the marriage. To sum up, the concept of annulment of marriage (hafka‘at kiddushin) was developed in the talmudic sources through four main stages: (a) At the first stage, annulment (or, better: “quasi-annulment”) means that the Rabbis validate an (externally flawed) get. This refers to a case in which the husband gave his wife a valid get and later invalidated it, but the Rabbis revalidated the get. It need hardly be said that the question of prospective vs. retroactive annulment is irrelevant according to this view, since “annulment” here is in fact divorce performed by a get as in all normal cases, the involvement of the Rabbis simply being the validation of that get. (b) At the second stage Rabbah, due to wider questions of the authority of the Rabbis, interpreted the concept of annulment as prospective annulment of marriage. Here, the Rabbis assume the authority to terminate marriage without any act of the husband, and the termination is valid from that point onward. Their authority is derived from the unique character of marriage: 242 243 See supra, n.241. Lifshitz, Afke‘inhu, 317 n.1, 317-319. ! Chapter Four: Marriage Annulment: From Mishnah to Talmud 59 “Everyone who betroths [a woman], does so subject to the consent of the Rabbis, and [in this case] the Rabbis annul his betrothal”. (c) At the third stage hafka‘at kiddushin became an annulment of the act of betrothal, and was applied by Rav Ashi to cases of improper betrothal. This was justified since “he acted improperly; they, therefore, also treated him improperly, and the rabbis annulled his betrothal”. According to our suggested analysis, the authority for marriage annulment according to Rav Ashi, following Rav !isda, is derived from the wide and general authority of the Sages to uproot the words of the Torah. (d) Finally, hafka‘ah becomes retroactive annulment of the marriage. This conceptual change was made by applying hafka‘ah as an annulment of the act of betrothal after the betrothal has taken place, which means that the betrothal is retroactively annulled. This significant change results from talmudic redactional work, which applied Rav Ashi’s reasoning of hafka‘ah in all talmudic cases, including those after the betrothal has taken place. In conclusion, hafka‘at kiddushin can be a halakhic tool by which divorce initiated by the wife is performed. More precisely, it is a halakhic construction by which marriage is terminated and thus recalcitrance (sarvanut get) can be by244 passed. It, however, depend on the way we understand the concept of marriage annulment, and within the talmudic text we have found several possible understandings. How, if so, was hafka‘at kiddushin treated by post-talmudic commentators? Can it, in principle, be adopted for practice? Was it adopted for practice? We turn now to the discussion of these questions. 244 I define it as termination of marriage since according to stages (b)-(d) of the Talmudic sugya there is no formal divorce (a proper get given by the husband to the wife) but rather an annulment (retroactive or prospective) of marriage executed by the court. ! Chapter Five Marriage Annulment in Post-Talmudic Times 5.1 Introduction In the previous chapter we discussed the concept of annulment of marriage in the Talmud, arguing that it developed in the talmudic passage through several stages. The next question to be discussed is whether post-talmudic traditions were familiar with the concept of annulment, and if so, did they use it in practice. A.H. Freimann has discussed at length the severe debates around the practical 245 use of marriage annulment since the Talmud to modern generations, a debate 246 which continues up to our days. It would not be possible to discuss here the practical use of marriage annulment throughout the whole history of the halakhah. Rather, the focus is on two unique traditions which we have already encountered: those of the Palestinian Genizah and the Babylonian Geonim. These two traditions 247 were discussed above in relation to get compulsion. The current discussion completes our survey by providing a broader account of these traditions. Though focused on two old traditions, this discussion nevertheless is relevant to the contemporary debate about marriage annulment: it emphasizes the way that marriage annulment was understood and perhaps practiced in the past. Were these two traditions familiar at all with the concept of annulment (even if they did not use it in practice)? In fact, it is difficult to find explicit indications of familiarity with the concept of hafka‘ah in any of its forms (besides the divorce clause itself, if indeed the latter used hafka‘ah, which we shall presently discuss) in the Palestinian tradition at the time of the Cairo Genizah ketubbot. This may result from the character of the sources: legal documents rather than theoretical writings. Nevertheless, the predecessors of this tradition were familiar with some version of annulment. The Palestinian Talmud was familiar with hafka‘ah in its preliminary form (that found also in the earliest stratum of the Babylonian Talmud), i.e., annulment of marriage by validating an invalid get: “the Torah said that [the get] is void [when the husband cancels it], while they [= the Sages] said that it is not void [i.e., the 245 246 247 Freimann, Seder Kiddushin. See supra, text at n.158. See supra, Chapter Three. Chapter Five: Marriage Annulment in Post-Talmudic Times 61 248 husband’s cancellation is invalid]”. However, the Palestinian Talmud based this view on the wide concept that: “their [i.e., the Sages’] words uproot the words of the Torah” (!"#$% &"'(% )&"*#+% )!&"'(), which means that the Sages have the authority (in appropriate cases) to rule against Torah laws. Since the Sages have authority to “uproot the words of the Torah”, we may theoretically assume that the Palestinian tradition could even accept an expanded version of hafka‘ah, i.e., complete annulment of marriage, prospective or retroactive, and even without a get. According to either option – a limited version of annulment or an expanded one – it is plausible that later generations in this tradition (such as the Genizah tradition) accepted this concept, following the Palestinian Talmud. Turning to the geonic tradition, we do find explicit references to annulment of 249 marriage, as in the following geonic responsum: Our grandfather, teacher, and Rabbi, Judah Gaon, enacted for them that they should not betroth other than by the Babylonian procedure: with ketubbah, witnesses’ signature, and betrothal blessing. And as for one who does not follow this procedure, he enacted that [we] disregard his betrothal [lit. him], since we say: “everyone who betroths, does so subject to the will of the Rabbis, and the Rabbis annul his betrothal.” You should cancel such a custom [which does not follow Judah Gaon’s procedure] as well. This responsum deals with a case of improper betrothal (i.e., when the betrothal was not according to the geonic enactment), in which hafka‘ah can be applied 250 more easily. Nevertheless, the Gaon here uses the concept as found in the Babylonian Talmud but for a case other the five cases mentioned in the Talmud, and this concept in principle gives him a wider authority, including termination of marriage long after its creation, without any hesitation or limitations. To conclude this section, the sources do not provide us with a direct proof of the use of retroactive annulment in the traditions here discussed; rather they reveal different levels of familiarity with it. Nevertheless, they do potentially validate its wider use. The question now to be discussed is whether annulment in its wider form was applied in our two specific traditions: those of the Palestinian ketubbot and the geonic rebellious wife. 5.2 Mere Annulment or Coercion? Rabbenu Asher ben Je!iel (Rosh) describes the geonic rule of the rebellious wife as 248 249 250 PT, Gittin, 4:2, 45c; stage (a), supra, text at nn.182-185.% Rav Hai Gaon, Otsar HaGeonim, Ketubbot, 7b, 18-19. This point is strongly reflected in the modern disputes regarding retroactive annulment versus annulment at time of marriage: see supra, n.162. % 62 Westreich: Talmud-Based Solutions to the Problem of the Agunah follows (She’elot uTeshuvot [Shut] haRosh, 43:8): […] And they enacted that the husband should divorce his wife against his will when she says: I do not want my husband […] For they relied on this [dictum]: “Everyone who betroths, does so subject to the will of the Rabbis” ($%&'# ,-./%# )*+%)# $%&'(# !") and they agreed to annul the marriage when a woman rebels against her husband (# ('0"1(2 (!+.#!+#($)(#%2/'*$"#,0$2%0&(#+0&4(!#3*+%). According to Rosh, the geonic enactment of coerced divorce in the case of a rebellious wife is based on annulment of the marriage. One may argue that the annulment does not even require a get given by the husband, i.e., in this kind of case there is a constitutive verdict of the bet din that the marriage is annulled, and this decision effects the couple’s divorce. 251 Michael Broyde, amongst others, accepts this view as historically correct (i.e., as an accurate description of the geonic view). According to Broyde, following Rosh, the geonic ruling of the rebellious wife was based on annulment, and if the husband refused to divorce his wife and coercion was not possible, the marriage could be annulled even without compelling him to give a get. Broyde emphasizes, 252 however, that it is not possible to adopt this view for practice today: “such annulments remain a dead letter in modern Jewish law”, and “…the nearly insurmountable halachic objections to a return to halachic rules that have not been normative for 800 years.” Bernard Jackson seems to agree with Broyde’s historical conclusions, while – as opposed to Broyde – giving this precedent a dogmatic 253 weight, even for the halakhic practice today. Some have suggested that support for this interpretation may be derived from 254 the plural formulation of the law of the rebellious wife in some geonic writings. Accordingly, statements such as “they write her a get immediately” (#56#(!#0.*"2 255 /*!)!) are understood as a writ of divorce written and given by the bet din, which means that divorce is executed without the participation of the husband. Rosh’s quotation above gives the theoretical basis for this possible interpretation: divorce by constitutive annulment of marriage upon the wife’s demand. As we 256 have concluded from the geonic responsum cited above, the concept of annulment was known and used. The current interpretation of the law of the rebellious wife is therefore based on a possible expansion of the concept of 251 252 253 254 255 256 Broyde, Marriage, 19-20, 60-61, 160 n.3. Ibid., 20, 61. See Jackson, Papyri, 162; Jackson, Agunah, 96-97.# See Westreich, Divorce on Demand, 347-348. Cited in Shut Maharam meRothenburg, Prague ed., 261, in the name of Teshuvot HaGeonim. Supra, text at notes 249-250. Chapter Five: Marriage Annulment in Post-Talmudic Times 63 annulment. Similarly, some scholars have argued that the Palestinian tradition is based on a 257 variation of marriage annulment. One of the Cairo Genizah ketubbot states the 258 divorce clause as follows: And if this ‘Aziza, the bride, should hate this Mevasser, her husband, and not desire his partnership [...] and she will go out by the authorization of the court and with the consent of our masters, the sages. “By the authorization of the court and with the consent of our masters, the sages” (!"#$%&'()&"(#*$'(%$+(,"-()&) means, according to this view, a constitutive divorce by the court. Accordingly, this stipulation gives the authority to the bet din to decide when marriage should be terminated, similarly to the plural formulation of the geonic dicta above. Thus, when the wife “hates” her husband and unilaterally desire a separation, she may exit the marriage, based on the court’s final decision 259 (which very likely means a prospective termination of the marriage ). Interpreting the Babylonian and the Palestinian traditions as using constitutive annulment produces the following model: we have a positive law basis for constitutive annulment of marriage by the court with no get given by the husband, but we need to clarify the authority for applying it in practice to a recalcitrant husband. At this point the tradition develops into two branches (without any necessary historical connection between them): on the one hand, annulment based on agreement of the spouses (Land of Israel); on the other, annulment based on a legal decree (Geonim). Yet from a historical point of view, in my opinion, this description is doubtful, in regard to both the Palestinian divorce clause and the Geonim. Rosh’s explanation of the enactment of the Geonim as marriage annulment (hafka‘ah) is anachronistic. The Geonim based their view on the Talmud (or at least on a decree of the Saboraim, as cited at the final stage in the talmudic 260 passage), which legitimated coercion of a get in cases of a rebellious wife, without relating it to annulment. This is explicitly stated in some geonic responsa, 257 258 259 260 See Friedman, Jewish Marriage, 336 n.78 (this was Friedman’s initial view, but he abandoned it entirely; see ibid.); Jackson, Papyri, 161-162; Jackson, Agunah, 130-131.( Ketubbah no. 1, lines 23–24, in Friedman, Jewish Marriage, II, 9 (Heb.); 13 (translation). I.e., the second stage of the development of the concept of annulment (see above, text at note 243). The authority for annulment in the Palestinian divorce clause, however, is not the authority to “uproot the words of the Torah”, but rather a contractual agreement between the spouses. See above, text at nn.123-130. 64 Westreich: Talmud-Based Solutions to the Problem of the Agunah such as that of Rav Sherira: 261 262 The original law was that [the bet din] does not oblige the husband to divorce his wife 263 if she asks to divorce […] [Later the Rabbis] enacted that when she demands divorce [the bet din] makes her wait twelve months [in case] perhaps they reconcile, but if they do not reconcile after twelve months [the bet din] compels the husband and he writes her a get. After the Saboraim […] [the Geonim] enacted […] and [the bet din] coerces the husband and he writes her a get immediately [upon her demand] and she gets the hundred or two hundred [zuz, of her ketubbah]. This is the way that we have ruled for three hundred years and more. You should also act in this way. According to Rav Sherira, the procedure of divorce in the case of a rebellious wife is by a coerced get, and the talmudic passage of the rebellious wife is the source for it. This passage is a sufficient basis for this ruling, and no additional normative 264 basis is required. As for the use of the plural formulation, this should be understood in the light of Rav Sherira’s explicit statement as referring to the act of coercion which is performed by the court. !"#$%&'%()#*&# or "$%&'%#*+, is a short formulation for “we (i.e., the bet din) coerce the husband and he writes (or: gives) her a get”, as in this 265 geonic responsum ("$%&'%*+-,-%'.*&%+!%(#/-,). It is remarkable that in some responsa plural and singular formulations are used together in reference to the giving of the get, without intending any distinction between them. For example: “the Geonim enacted […] we try to make peace between them, and if she does not accept [we, the court] give [plural] her a get immediately (0+'!'% "$% &'% (#)+-)) […] and so wrote Rav Hai […] the earlier Geonim enacted that [we, the court] compel her husband immediately to give a get 261 262 263 264 265 Teshuvot HaGeonim, Sha‘are Tsedek, Vol. 4, 4:15 (cited above, text at notes 123-130). Both Friedman and Brody assume that this view was largely accepted by the Geonim: see Friedman, Jewish Marriage, 324-25; Brody, HaGeonim, 298-99. Verb in plural ((#*##12). Similarly, all the judicial acts below are formulated in the plural (%3(#*##12 (#/-,%3(#4#0,2), in contrast with the actual writing of the get: "$%&'%*+-,- (see infra). According to the final Saboraic section of the talmudic passage: !"#$!% !+5% #10#% 06#0+% &'% ()#&52(Ketubbot 64a). % Even if we interpret annulment as validating an invalid get, as some have suggested (see Sho!etman, Hafka‘at Kiddushin, 356–365; supra text at note 243), it is still not required for our case. The get is a valid get since we deal here with a legitimate coercion.% Perhaps the plural formulation was also influenced by the talmudic style of the passage of the rebellious wife, which uses a plural formulation: !"#$!% !+5% #10#% 06#0+% &'% ()#&52- (we make her wait twelve months for her divorce). Thus, just as the waiting period is executed by the bet din, so is the giving of the get, but the actual giving is done by the husband who is compelled to do so by the bet din.% Chapter Five: Marriage Annulment in Post-Talmudic Times 65 266 (./%(&)"%0)1% !"#$%&'%()*+,-)” The plural formulation of “giving her a get” thus means “compelling her husband to give a get”. To be sure, this and the other 267 responsa are based on different geonic sources. Nevertheless, it is implausible to assume that they reflect a dispute between the Geonim regarding the procedure of the law of the rebellious wife (or different traditions regarding the actual enactment of the Geonim). If indeed this significant dispute had taken place, it would have been reflected more sharply and in a more explicit way. The plural formulation, therefore, reflects different styles and formulations of the same ruling: compelling the husband to give a get. So why did Rosh mention annulment? The dogmatic halakhah had developed in a direction different from that of the Geonim. The geonic view was totally rejected by Rabbenu Tam, who argued that there is no basis in the Talmud for compelling 268 269 divorce in such a case. Rabbenu Tam’s view was largely accepted; therefore the geonic view needed justification. Rosh very limitedly accepted the geonic view 270 (only in certain bedi’avad [ex post facto] cases), and attempted to provide some 271 justification for it by interpreting it as entailing annulment. In this way, Rosh could both adhere to Rabbenu Tam’s view, that a coerced get in a case of a rebellious wife is not found in the Talmud, while at the same time legitimating the geonic measures (ex post facto). In any case, Rosh did not intend to introduce a different procedure for cases of a rebellious wife, but rather to base the problematic 272 coercion enacted by the Geonim on their authority to annul marriage. Historically, therefore, it is hard to accept Rosh as a support for the view which sees the geonic rule of the rebellious wife as based on annulment. The procedure of divorce in the law of the rebellious wife is merely the performance of a compelled 266 267 268 269 270 271 272 Shut Maharam meRothenburg, Lemberg ed., 443. See also Shut Maharam meRothenburg, Prague ed., 261: the first part of the responsum (cited in some manuscripts in the name of Rabbenu Gershom Me’or haGolah) uses a singular formulation (2&"'"%'.)/%!"%$)!)+), the middle part (in the name of Teshuvot HaGeonim) uses a plural formulation (2&"'"%./%!"%)$&,+), and the last part (in the name of Halakhot Gedolot) uses a singular formulation again (2&"'"% ./% !"% $&+,+). The same phenomenon is documented in Shut Maharam meRothenburg, Prague ed., 443. See previous note.% See Sefer haYashar leRabbenu Tam, Teshuvot, 24. It is arguable whether this total rejection of the geonic view was indeed held by Rabbenu Tam (see Abel, Morgenstern, 18 n.57; idem, Za‘aqat Dalot, 10-11). However, later Rishonim attributed that view to Rabbenu Tam, and largely accepted it (see E. Westreich, Rise and Decline, 212-218). This fact led to the reinterpretation of the geonic view, as described infra. See E. Westreich, ibid.% See infra.% Me’iri’s Teachers’ Teachers’ were motivated by similar considerations in making a link between the geonic moredet and the Palestinian divorce clause; see supra, text at notes 147-151. See also Sho!etman, Hafka‘ah Kiddushin, 377.% 66 Westreich: Talmud-Based Solutions to the Problem of the Agunah get, and, according to the geonic responsum cited above, this get was a regular get given by the husband (although under the pressure of the bet din). This law is based on the talmudic passage of the rebellious wife. Yet, in order to reconcile it with different views regarding those sources, Rosh anachronistically suggested the reasoning of annulment. However, the view of Rosh is important from a dogmatic point of view, as discussed below. According to the analysis of Rosh here suggested, the procedure of the rebellious wife is not merely an annulment of marriage but rather a divorce by a coerced get, while the authority for it is derived from the authority to annul 273 marriage. Another responsum of Rosh supports this view. In it, Rosh justifies coercion of a get due to special circumstances: the husband was suspected as an immodest man who cheated his betrothed wife (see citation below). Rosh argues that this case is similar to that of Naresh in which, according to Rav Ashi, the 274 Rabbis applied annulment since “he acted improperly”. Rosh then discusses the possibility of annulment: But if it looks to you, my masters, who are close to this matter, that the betrothing man is not a person worthy and decent to marry this girl of good descent, and that he has persuaded her by fraud and cheating, and that it is reasonable to compare [this case] to the case of Naresh (BT Yevamot, 110a) where we learned that since it (the betrothal) was done improperly [the Rabbis] annulled the betrothal – [then in the case of] this [person] as well, who acted improperly, although we would not annul the betrothal, nevertheless we should follow in this case the view of some of our Rabbis who ruled in the law of the moredet that [the bet din] should compel him to divorce her. Annulment, according to Rosh, should not be applied here. However, the partial similarity between the talmudic case of annulment and the current case legitimates coercion in the latter. Due to its special circumstances, Rosh argues, we can follow the view that supports coercion in cases of a rebellious wife, i.e., the geonic view, which was normally rejected by Rosh. His reasoning is probably that in such a case it is right to apply the geonic rule of the rebellious wife since there is no “moral 275 fear” which usually prevents it. ! If the geonic law of the rebellious wife were merely a procedure of annulment, Rosh’s discussion in this responsum would be superfluous or even internally contradictory: we cannot apply annulment, but we can apply the rule of the rebellious wife – which is the same. We must assume therefore that they are 273 274 275 Shut ha-Rosh, 35:2. BT Yevamot, 110a, and see supra, text at nn.164-165. See Shut haRosh, 43:8. On the role of the “moral fear” in Rosh’s view, see Knol, Agunah and Ideology, ch.3, mainly at 89-91. Chapter Five: Marriage Annulment in Post-Talmudic Times 67 different halakhic procedures: the one is coercion of a get, i.e., a divorce executed by the husband (against his will), while the other is annulment executed by the bet din. However, we can see here that there is a relationship between the two, since they are ultimately based on the same reasoning. This is reflected also in Rosh’s view, which supports the geonic coercion by the concept of annulment, as discussed above. Thus, integrating Rosh’s two responsa (35:2, which exceptionally authorizes coercion, and 43:8, which explains the geonic rebellious wife on the basis of annulment) produces the following explanation: the law of the rebellious wife is partially based on annulment (specifically, in terms of the authority for it), but the procedure includes a coerced get. Since it includes a get, hafka‘ah can be more easily applied than can termination by mere annulment of marriage. The case in responsum 35:2 is similar to the talmudic annulment, but for particular reasons 276 does not admit of annulment. However, the second possibility, coercion based on annulment, may be applied in such a case. Considering annulment in the Palestinian divorce clause, although “by the authorization of the court” (!"#$%&#'%()*%+,) might be interpreted as a constitutive decision of the court without a get, it is more likely that the divorce clause does not replace a get but rather enforces it. The phrase “by the authorization of the court” is thus required, as Katzoff put it, “… to make it crystal clear that no right or powers 277 of divorce are provided the wife other than those in rabbinic law”. I accept that I have not found decisive support for either of the possible procedures (annulment or coercion). However, the history of the halakhah further supports the option of coercion. Get in the rabbinic tradition is a central matter and 278 difficult to ignore. Accepting the annulment theory requires us to assume that a condition (in the Land of Israel) or a decree (in Babylonia) adopted such a radical practice, which dispenses with the need for a get, with no explicit discussion and with no reservations. I suspect that if such a decision had been taken, it would not have been left in silence, with no explicit mention either in the decree or in the ketubbah, without being accompanied by a deep halakhic discussion and without at 276 277 278 Rosh does not detail the reasons for not applying it. It might be that this case is not as improper as the talmudic case, or it may reflect a hesitation to apply annulment in practice: see infra.% See Katzoff, Papyrus, 246. For further support see Westreich, Divorce on Demand, 353-354. Reflected, for example, in BT Ketubbot 74a: “and the rabbis did not have the power to release her without a get”, i.e., even in cases where there is some theoretical basis for annulling the marriage, the Rabbis do not have the authority to release the wife without a get. Interestingly, amongst some Karaite sages around the 15th century there was a practice of authorizing annulment of marriage by a bet din without requiring a get: see Falk, Gerushin, 25-26. 68 Westreich: Talmud-Based Solutions to the Problem of the Agunah 279 least some objections. What is the significance of our conclusion? As we have shown, Rosh links the geonic moredet to the concept of marriage annulment. While we had great doubt whether we could consider his view as historically accurate, rather than as an anachronistic justification for an earlier halakhah, his responsum has important dogmatic implications. Rosh here legitimates marriage annulment in practice at least in bedi’avad cases, and has no doubt that it may be used. This is particularly meaningful in a halakhic environment in which the practical use of hafka‘at kiddushin has become subject to major dispute, from the Geonim until our own 280 days. As we have argued, it is hard to assume that Rosh understood the geonic rule of moredet as a judicial act of annulling the marriage without participation of the husband, namely as hafka‘ah without a get. Rosh here tries to justify coercion of a get, rather than to revive a practice different from that in his own day. Therefore, the implication of Rosh’s writing is that it legitimates hafka‘at kiddushin at least when it is accompanied by a coerced get. We cannot however prove that Rosh demanded a get as a necessary condition for hafka‘ah. Rosh does not discuss the typical cases of hafka‘ah, but is concerned only to provide support for the rule of moredet; in regard to classic hafka‘ah he may well have accepted it even without a 281 get. Another implication of defining Rosh’s view as anachronistic relates to the opponents of the geonic tradition. In a recently published paper, Rabbi U. Lavi argued, based on Rosh’s reasoning, that the Rishonim who disagreed with the 282 Geonim regarding moredet (mainly, Rabbenu Tam) rejected hafka‘ah as well. According to the analysis above, this is a false conclusion. The element of hafka‘ah is a later one, added by Rosh, while the dispute between the Geonim and Rabbenu Tam relates to the authority for coercing a divorce, without taking hafka‘ah into consideration. Rosh’s second responsum, which we have discussed in relation to the historical 283 aspects of his view, supports our current conclusions regarding its dogmatic implications. In this responsum, Rosh does not reject the possibility of annulment. Moreover, it seems that Rosh would agree that in principle annulment can be 279 280 281 282 283 The later objections relate to the legitimation of coercion. Only Rosh raised the issue of annulment, and even he, as analyzed above, treated it as a support for coercion. See infra, Chapter Six; Westreich Gatekeepers.! See further infra, text at nn.302-316.! Lavi, Ha’im, 308. Shut HaRosh, 35:2; see supra, text at notes 273-276. Chapter Five: Marriage Annulment in Post-Talmudic Times applied even when no get is given. As he writes: 69 284 [I]t is reasonable to compare [this case] to the case of Naresh (BT Yevamot, 110a) where we learned that since it (the betrothal) was done improperly [the Rabbis] annulled the betrothal – [then in the case of] this [person] as well, who acted improperly, although we would not annul the betrothal, nevertheless we should follow in this case the view of some of our Rabbis who ruled in the law of the moredet that [the bet din] should compel him to divorce her. The case is similar to the case of Naresh, in which the Sages annulled the betrothal. In principle, we could annul the betrothal here as well, although no get was given. However, for an unmentioned reason (perhaps because the case here discussed is not as radical as kidnapping the betrothed girl from her former husband in the case 285 of Naresh, or maybe because of a more general hesitation to apply annulment in practice), Rosh was not willing to apply annulment here, but rather preferred coercion. Accordingly, a distinction should be made between the possibility (and validity) of retroactive hafka‘ah in principle and its practical implementation. 286 While Rosh avoids the latter, he does not reject the former. We return to the practical use of hafka‘ah in the next chapter. 284 285 286 See full quotation above. See supra, text at nn.164-165. Cf. Sho!etman, Hafka‘at Kiddushin, 369 n.54. Chapter Six Hafka‘at Kiddushin in Practice 6.1 The Various Talmudic Approaches as Bases for the Later Disputes: Does Hafka‘at Kiddushin Require a Get? Within the talmudic text we observe a tension between different approaches. This tension is reflected in the contradictory interpretations of the concept of annulment 287 amongst both later poskim and modern scholars. Some poskim follow one approach, others take an opposite view, explaining the contradictory parts of the 288 sugya by means of several different hermeneutical approaches. While total 289 rejection of other views is common in this kind of debate, the following analysis shows that this would be incorrect. 290 Let us repeat the main approaches found in the Talmud: (a) At the first stage, Rabban Shimon ben Gamliel’s decision on the husband who cancelled his writ of divorce before the wife received it was that the Sages essentially validate the (externally flawed) writ of divorce (a quasiannulment). (b) At a second stage, the Talmud (following Rabbah) explains this ruling as annulling the betrothal. At this stage, an annulment is understood as a prospective annulment of the status of the couple as a betrothed (and married) couple (a “delayed annulment”). As claimed by Rabbah, the Sages’ authority is derived from the unique character of marriage: “everyone who betroths [a woman], does so subject to the consent of the Sages, and [in this 291 case] the Sages annul his betrothal”. (c) Rav Ashi at a later stage applied annulment where the betrothal was improperly entered into (an “immediate annulment”), either by expropriating the betrothal money or by declaring the cohabitation to be mere promiscuity 287 288 289 290 291 The dispute continues since the redaction of the Talmud, not only amongst classic commentators but even in modern days, amongst rabbis and dayanim, as well as Jewish Law Researchers. See Riskin v. Goldberg and Lifshitz v. Lavi, supra, text at n.162; Berkovits v. Sho!etman, infra, n.303. See Westreich, Kama Milim, 245-261.! Especially when the possibility of practical use of hafka‘ah is under discussion; see Rabbi Goldberg and Rabbi Lavi, supra, n.287. These approaches are discussed at length above: see supra, text at nn.216-244. See supra, text at nn.186-189. Chapter Six: Hafka‘at Kiddushin in Practice 71 (bi’at zenut). As regards the authority of the Sages, Rav Ashi accepts an expanded 292 view of the authority of the Sages to uproot the words of the Torah. (d) Finally, as a result of later talmudic redaction, annulment was applied to invalidate the act of betrothal in both “immediate” and “delayed” cases. Invalidation of the act of betrothal quite a while after it took place (the “delayed” case) makes this a retroactive rather than prospective annulment of the marriage. All these approaches are found in the Rishonim and A!aronim. While recent debates frequently discuss how hafka‘ah was interpreted by Rishonim and A!aronim, the conceptual distinction between the various views has not always been clearly defined. The following discussion contributes to a more accurate understanding of the approaches amongst classic writers and their basis in the Talmud. 293 The last view (d) reflects the final talmudic stage, and is therefore the 294 dominant view amongst Rishonim and A!aronim. Indeed, some elements vary within writers of this group, as will be shown below. Nevertheless, the basic attitude (i.e. viewing hafka‘ah as a retroactive annulment of the betrothal, being interpretatively influenced by Ravina and Rav Ashi’s discussion) is common to them. Nevertheless, although (d) is the dominant view, we do find some Rishonim who suggest different interpretations for the concept of hafka‘ah, focusing on other talmudic stages. Stage (a) is found in Ri Halavan’s Tosafot, whose explanation of hafka‘at kiddushin is as follows: !"#$!% &'% ($#")*% #+,'-!% ($./$0% &.0"#, i.e.: the Sages in 295 their decree made [the get] valid according to Torah law. He was followed by 296 some scholars, who were influenced in their analysis by that talmudic stage. 292 293 294 295 296 See, supra, text at notes 199-215. Stage (c) is the basis for the conceptual development in stage (d). Commentators who interpret annulment as retroactive, following stage (d), would adhere to stage (c) as well, as regards the immediate annulment. See for example Rashi, Gittin 33a, s.v. 1.,$ and !#,#); Tosafot, ibid., s.v. #!.,-/23#; Ramban, Ketubbot, 3a, s.v. !#,#) and elsewhere. The Rishonim however were partly influenced by stage (a): see the discussion below on Rashi’s commentary and the discussion regarding the demand for a get in the process of hafka‘ah.% Tosafot Ri Halavan, London, 1954, Ketubbot 3a, s.v. )+/'+%4*. See Atlas, Netivim, 211-214; Sho!etman, Hafka‘at Kiddushin, 355. This view is found also in Teshuvut Be’anshe ‘Aven, 13 (cited by Mar’e Kohen, Yevamot, 90b; Atlas, ibid.). But if we follow this interpretation, it becomes difficult to integrate the other parts of the sugya with the suggested understanding for hafka‘ah. An interesting reflection of this difficulty is found in Teshuvut Be’anshe ‘Aven, 13, who suggests that we amend the Talmudic text and read: )"5'% &.0"+% 3$-+3% )"5'+% 4* (“everyone who divorces [his wife] does it subject to the will of the Sages” [who can prevent him 72 Westreich: Talmud-Based Solutions to the Problem of the Agunah This view should be distinguished from the view of Rashbam, followed by some 297 Rishonim. Rashbam argues that in the three talmudic cases the get in fact is valid and the marriage is not retroactively annulled, since the husband fears that his marriage may be annulled, and therefore cancels the annulment of the agency (Gittin 33a), forgoes his condition (Ketubbot 3a) or, in the case of a dying person (Gittin 73a), agrees that the get should not be annulled even if he recovers. Although according to Rashbam the get is valid, in principle Rashbam admits that annulment of marriage is retroactive: if the Sages did have the need to use hafka‘ah (which they do not) it would be applied retroactively. Therefore from a conceptual point of view, Rashbam’s view follows stage (d). Some Rishonim cited by Ritva in the Shitah Mekubetset followed stage (b) in 298 their understanding of the concept of hafka‘ah. This interpretation is also 299 discussed by !atam Sofer, and I assume that it was also the understanding of 300 Rashi’s teachers. Amongst Jewish Law scholars it was recently suggested by 301 Arye Edrei. It is not clear according to this interpretation why the Sages should ---from canceling the get]). This suggestion has no basis in any textual witnesses or any of the Rishonim (as correctly mentioned by Atlas, ibid.), and of course – as analyzed here – the text should not be amended since it views the hafka‘ah in quite a different way. The simple meaning of ! "# $%&'% is the basis for Teshuvut Be’anshe ‘Aven’s critic, R. Yits!ak Z. Margireten (Tokef HaTalmud, Ofen: Konigl Ungarischen Universitats Buchdruckerei Print, 3:1; 3:4), who argues that hafka‘ah must be understood as retroactive annulment (see also Shitah Mekubetset, Ketubbot 3a, end of s.v. ve’od katav). 297 298 299 300 301 See Ramban in the name of Rashbam (“Rabbi Shmuel Ramrogi”; i.e. from Ramerupt), Ketubbot, 3a, s.v. ()*)$; ibid., Gittin, 33a, s.v. "# (Ramban however seems not to accept this interpretation); Rashba, Ketubbot, 3a, s.v. "#; ibid., Responsa, 1162 (regarding his view see n.303). See also Pene Yehoshu‘a, Ketubbot, 3a, s.v. )(+*,&-. and s.v. "#. Shitah Mekubetset, Ketubbot 3a, s.v. .3/0*1(!/2#), in the name of *412'%!2*.: “when we say that the Rabbis annul his betrothal, it does not [apply retroactively] from the time of betrothal but [it applies] now, at time of the act” (!.2$(!.".!5*$)%*&!2,$'!)."!(*+*'!5*$)%*&"!5+/1!)(+*,&-.!5+*1'.!*#% ($,'!2,$'). The get mentioned later in the Ritva (08/)!6"*.)!5.#'!7.!*#!5*"0/!5*$)%*&(!5*.) has a similar meaning according to the view that hafka‘ah is a retroactive annulment (discussed earlier by the Ritva): it is an element required for applying hafka‘ah, but this doesn’t mean that the Sages validate the get (as according to Ri Halavan). !iddushe !atam Sofer, Gittin, 33a, s.v. 9+*2. Rashi’s teacher’s view is cited – and strongly rejected – by Rashi in the various sugyot of hafka‘ah. Rashi indicates that according to his teacher’s (mistaken) understanding the betrothal is prospectively annulled, as opposed to his interpretation: see Rashi, Ketubbot, 3a, s.v. ()*)$: !691#!", ./(")!5.#'!.")!51&*,'!5*1&,+!5*$)%*&($!*2$1*-$!)'#!$1-"!(2.!6*14, i.e. you must take the view that the betrothal is retroactively annulled and not only from now on [as his teachers argue]). Rashi’s teacher’s view requires more investigation and is beyond the scope of the current discussion; see Westreich, Hafka‘at Kiddushin, text at notes 118-123. Edrei, Koa! Bet Din, 34-35. Chapter Six: Hafka‘at Kiddushin in Practice 73 302 declare the cohabitation to be promiscuity. ! Indeed, one major implication of the range of conceptual constructions encountered in the Talmud is the question of the necessity of a valid get for the process of hafka‘ah. According to the first stage ([a] above), in order to perform hafka‘ah there should obviously be a valid get, which became invalid only due to external reasons, such as being cancelled by the husband. This I assume is the view 303 of Ri Halavan and his followers. But at the other stages, in principle there is no need for a get in order to annul the marriage. Based on the analysis of the different talmudic meanings of the concept, we can now understand better the motivation behind some integrated – and more complex – approaches, such as that of Rashi. Thus Rashi on the one hand explains hafka‘ah as a retroactive annulment, while on the other still regards a get as a necessary element in this process. While Rashi’s view is not completely clear, for our purposes it is sufficient to emphasise the two elements which Rashi integrates together: the get on the one hand and the retroactive annulment on the other. One 304 quotation from Rashi sharply reflects this integration: ! "#$%! #$""&! '()%*! "+$+, "-! ./! $0$! &1! 1*23&, i.e. the retroactive (1*23&) declaration of the cohabitation as promiscuity is effected by the get. His view accordingly is a result of mediation between different views which are found in the talmudic text itself: one which bases hafka‘ah on (validation of) the get; the other which bases it on retroactive annulment of the act of marriage. I assume that the ambiguity in Rashi’s interpretation is the interpretative price he is willing to pay for integrating contradictory parts of the sugya. The result is that the exact object of the get according to Rashi is still disputed. Rashi mentions the existence of a get several times, and this led some (e.g. the Israeli rabbinical court Dayan, Rabbi Uriel Lavi) to understand the get as a substantive element in 305 the process of annulment. However, other passages of Rashi indicate that the get 302 303 304 305 See !atam Sofer, Gittin, 33a, s.v. 4)$5. Harmonizing all the parts of the sugya is quite difficult according to this approach and would apparently (like Ri Halavan’s, above) need to use an historical approach, according to which Ravina and Rav Ashi’s discussion is explained as a transferred part of the sugya, and therefore isn’t consistent with its meaning (as argued by all the above writers: Teshuvut Be’anshe ‘Aven, Atlas and Sho!etman, supra, n.296). Nevertheless, a harmonious solution is possible but quite complicated. See the view of Rashi’s teachers and their explanation of the declaration of the cohabitation as promiscuous, cited by Rashi, Ketubbot, 3a, s.v. "+$+, (and more), discussed in Westreich, Hafka‘at Kiddushin, ibid. As regards Rashba, Berkovits and Sho!etman dispute whether he held the view that hafka‘ah is effected by the get (Sho!etman, Hafka‘at Kiddushin, 359-360), or rather is used in cases when a get exists (Berkovits, Tnai, 123-133). Rashi, Gittin, 33a, s.v. "+$+,. See Rashi, Ketubbot, 3a: "-!./!$0$!&1, which can be understood as: “[the annulment is effected] by this get”. Rabbi Uriel Lavi seems to understand Rashi in this way: see Lavi, Ha’im, 306: ! &6% 74 Westreich: Talmud-Based Solutions to the Problem of the Agunah is (merely) described in the cases in which the Sages enacted hafka‘ah, but is not a 306 necessary element in the legal process of hafka‘ah. Accordingly, some (e.g. Berkovits) argue that Rashi does not refer to a get as a necessary condition for applying hafka‘ah, but merely says that hafka‘ah is applied in such a case, but can 307 be applied in some other cases as well. Nevertheless, following the repeated mentioning of get in Rashi’s commentary, it is hard to say that a get doesn’t have any role according to Rashi. On the other hand, Rashi could not be taken as far as is done by Rabbi Lavi, as maintaining that a get is a substantive element in the process of annulment. It seems to me that according to Rashi hafka‘ah does not validate the get. Rather, Rashi sees the get as a supportive element for the process of hafka‘ah, which is indeed required (see the possible reasoning below), but not substantive, and thus could be replaced by other elements. Therefore hafka‘at kiddushin could be initiated with the “support” of one witness, without any get: releasing a wife on the basis of one witness to her 308 husband’s death is, according to Rashi, based on hafka‘at kiddushin. Historically the mentioning of a get in the various talmudic passages could result from the integration of the different talmudic stages – both the first stage, according to which the get is a substantive element in the process of hafka‘ah, and the later stages, which negate it. Purely historically, it was possible to argue that for the second, third and fourth approaches the get was not necessary at all. Some 309 Rishonim held this view, as explicitly stated by Me’iri. However, from a classical dogmatic point of view, all parts of the sugya are meaningful, even if different in their origin. Thus, the main challenge becomes integrating the different meanings of the talmudic text into one harmonious approach. Since a get is mentioned in all of the talmudic cases of hafka‘ah, many writers deduced that a get is always 310 needed for the process of hafka‘ah. Thus many commentators, even though following stage (d) in their interpretation of the concept of hafka‘ah, claimed that a get is a necessary element in this process. Nevertheless, hafka‘ah does not mean that the get is validated. So, ---#6%)78#'$9#07&#:0(;4#+%+#3%/+ "*%1)!#4,1$# 5#-.+#%/%#01#'%!12#3%!(/%*+#'1*4+!#%,!"#"$%&#'()(*)+ 3%!(/%*+#'1*4+#$(+#("!7+#61-(#<("%!7+0#(2*'. 306 307 308 309 310 See Rashi, ibid., s.v. 3%!(/%*0# 32&"# +(1*4$(: 6+%"8$# +9# -.# $&%!7 (i.e. the Rabbis annulled the marriage when it is followed by such a get); see Sho!etman, Hafka‘at Kiddushin, 360. See Berkovits, Tnai, 133-141; Riskin, Hafka‘at Kiddushin (English version), 12-14; 33-34, notes 2326; ibid., 46-47 (a response to Wieder, Rebuttal, 39-40). Shabbat, 145b, s.v. '(/10, and see also below. Me’iri, Ketubbot 3a, s.v. (")$!# 07. Rabbis Riskin and Wieder are in dispute as to whether Me’iri accepted this view: see Riskin, Response, 47-48 (in my opinion - correctly); or withdrew from it: see Wieder, Rejoinder, 64-67. Rashi according to some; Ri Migash; Ramban and his disciples: Ra’ah and Rashba, and others. Chapter Six: Hafka‘at Kiddushin in Practice 75 by contrast to the first approach, a get according to them is not an essential element 311 of hafka‘ah itself. It may be necessary, but this is due to external reasons, for 312 example prevention of a “slippery slope” in the use of hafka‘ah or creation of a similarity between hafka‘ah and the normal halakhic way of terminating 313 marriage. Accordingly, we would not necessarily demand a proper get which was merely externally flawed (i.e. due to cancellation, an unexpected event in case of conditional divorce, etc.) in order to apply hafka‘ah. On the contrary, !"#$%&$'(, i.e. 314 any get can fulfil those objectives, and is sufficient for applying hafka‘ah. It could even be replaced by other halakhic devices. Therefore, according to both 315 Rashba and Rashi, hafka‘ah is applied when one witness testifies to the husband’s death: the additional required element for hafka‘ah, which normally 316 means a get, is replaced here by one witness. Had hafka‘ah been conceived as a means of validating an externally invalid get, this would not have been possible. In modern discussions, it is often suggested that hafka‘at kiddushin be assisted by other means of terminating the marriage, as described in what follows. Thus, the above discussion becomes significant for practice: do these means fulfil the demand for an additional supportive element in the process of hafka‘ah? 311 312 313 314 315 316 As opposed to Sho!etman, Hafka‘at Kiddushin, 397. Sho!etman’s conclusion is neither historically nor dogmatically decisive. Historically, it may reflect a specific stratum of the Talmud, but is not unanimous, as I have shown, so that the opposite view cannot be ignored. Dogmatically his view reflects Ri Halavan’s approach, but many other Rishonim understand hafka‘ah as a retroactive act, while both are rooted in the ambiguity of the talmudic text, as shown here. As a matter of fact, from a dogmatic point of view the last talmudic stage, which is rejected by Sho!etman, is many times more authoritative (Sho!etman doesn’t accept it in our case, following his analysis of Rambam’s view regarding coerced marriage, according to which late talmudic strata cannot stand against “Talmud arukh”: see Sho!etman, Ones, 117-121). The fear of the “slippery slope” is described in Shut Mishpete Uzzi’el, Part 2, Even Ha‘Ezer, 87. This might be the reasoning of the view that explains the demand for a get by the mere fact that the marriage was properly effected: see for example Ra’ah (Shita Mekubetset, Ketubbot 3a, s.v. $)*&$+&! ".,-"): “but in a case in which the marriage was effected in accordance with the Sages, she cannot go out without a get” ('($,%)$",/0!"%$-12,$/,$3/4&5$+!0-)$,1#64#$,&/"$%),) for the term $"7$'(% -1&$,!"1 used by Ramban in this context: see Lifshitz, Afke‘inhu, 320. See also Wieder, Rejoinder, 73-74 n.1. I assume that this view is also the rationale of the argument that after hafka‘ah the couple is still bound by rabbinical marriage, and this is the formal reason for a get. See Rabbi Ovadya Yosef, Kol Ha-mekadesh, 100-101. See Ri Migash, cited in Me’iri, Ketubbot 3a, s.v. !-4,1$%&, as “Ge’one Sefarad”. Similar is Rashba’s term: ,'/($8-9, See Rashba, Ketubbot, 3a, s.v. 1#64#$%&. Rashba, ibid.; Rashi, Shabbat, 145b, s.v. *!#:%. Rashi’s view here contributes to the uncertainty about the exact meaning of his approach regarding hafka‘ah: see above, text at notes 303-308.$ See also Berkovits, Tnai, 127-139 (discussing Rashi and Rashba’s views). 76 Westreich: Talmud-Based Solutions to the Problem of the Agunah 6.2 Marriage Annulment in Practice: Can hafka‘ah be applied today? As already mentioned, this question has been repeatedly debated amongst writers in the last decades. The above discussion contributes to the issue by revealing the basis for the contradictory approaches as residing within the talmudic texts themselves. Indeed, the origin of the concept was quite limited in its application (stage [a], above), but was expanded in a process of several steps which culminated at a late talmudic stage. As regards the authority of the sages, late amoraic generations re-enforce their authority to annul marriages which are valid according to the Torah. In practice, however, hafka‘ah was hardly used. Halakhic sources deal extensively with hafka‘ah in cases of improper betrothal, such as fraud or betrothal in breach of requirements of communal enactments (takkanot hakahal). While the 317 main halakhic writers rejected the practical use of hafka‘ah, some did accept it, at 318 least where other considerations were involved. The question now arises as to whether we can take hafka‘ah a step further, and apply it also after a proper marriage took place (the “delayed annulment” situation). This application is indeed much more radical and difficult to use in 319 practice. It is also doubtful whether it was ever used in the past in practice at all. This solution therefore has met severe objections, frequently total rejection, from opposing scholars, who have punctuated their analysis with strong emotional 320 reactions. Nevertheless, some classic writers have mentioned the use of retroactive 321 annulment as a supportive argument for problematic rulings. This latter approach accepts in principle a wide use of hafka‘ah, and appears to be a potential way to use hafka‘ah in the quest for a remedy to the problem of agunot. Yet many writers demand that hafka‘ah in practice not be used without any other elements, but rather requires some support such as an “externally flawed” get (but not necessarily that support). Hafka‘ah therefore could be accompanied by different (but still otherwise halakhically problematic) forms of termination of marriage. Thus, it could serve as 317 318 319 320 321 Even Rema, Even Ha‘Ezer, 28:21, who accepts hafka‘ah in principle negates its practical use: “even so, we should be strict in practice” (!"#$%&'(#)%*'$+!)%"'%',!%-)'./). See the famous case of the Egyptian enactment of 1901: Freimann, Seder Kiddushin, 338-344. See supra, Chapter Five, and Westreich, Divorce on Demand. See Goldberg, Lavi and others, supra, text at n.162. The debates around marriage annulment contain three levels which may be defined as hermeneutical, political, and sociological. I have analyzed these aspects elsewhere: see Westreich, Gatekeepers. The most famous examples are Rosh regarding the geonic moredet (Shut HaRosh, 43:8; see discussion supra, text at nn.279-286); Ran regarding teme’ah ’ani (Nedarim 90b, s.v. /,'/-) and Rema regarding the Austrian pogroms (Darkhe Moshe, EH 7:13).% Chapter Six: Hafka‘at Kiddushin in Practice 77 a complement to a compelled get, making the latter a (permitted) form of 322 coercion. It is mostly relevant to the Israeli context: whereas rabbinical courts have legal authority to use sanctions against a recalcitrant spouse, they are too often reluctant to use it in practice due to the fear of an illegitimate coerced get (get me‘useh). An enactment which combines difficult cases of coerced get with marriage annulment would make the coercion easier to apply. If the get is invalid (me‘useh), annulment would be applied; if annulment cannot be applied on its own and needs support, the coerced get provides us with such a support – it would surely be considered as at least a get kol dehu, which according to many Rishonim is sufficient in order to make hafka‘ah legitimate. We may also suggest that annulment may be accompanied by other forms of termination of marriage, such as conditional marriage or kiddushei ta‘ut. This suggestion adheres to the rationale behind Rabbi Prof. Michael Broyde’s recent proposal of a Tripartite Prenuptial Agreement, in which annulment is supported by an advance authorization to give a 323 get, conditional marriage and kiddushei ta‘ut. At any rate, the issue is still debated. And as in many other issues, hafka‘ah still awaits the proper halakhic authority for its application in practice. 322 323 As Rosh argues. In some cases Rosh supports coercion on this basis even in practice, although he usually rejected the geonic view which enacted coercion: see Shut HaRosh, 35:2. On the issue of hafka‘at kiddushin as a support for a coerced get see Rabbi Ovadya Yosef, Kol Hamekadesh, 96103. This model was suggested by Rabbi Wieder, Rejoinder, 76-77 n.28, as an alternative to Rabbi Riskin’s more radical proposal. See Broyde, Tripartite Agreement, 3-11. For a suggested version of the tripartite agreement see Broyde, ibid., 12-15. Chapter Seven Mistaken Marriage, Conditional Marriage: The Talmudic Basis of an Innovative Approach 7.1 Introduction This chapter discusses a different form of marriage annulment: a declarative marriage annulment on the basis of mistaken transaction (meka! ta‘ut, or mistaken 324 betrothal; kiddushei ta‘ut) and some other related halakhic constructions, mainly terminative conditions. Retroactive cancellation of marriage due to a mistake in its creation (kiddushei ta‘ut) has been extensively discussed since the talmudic period up to our days, and prompts deep halakhic and meta-halakhic disputes. The use of terminative conditions as a possible solution to the agunah problem is no less discussed and no less accompanied by halakhic and meta-halakhic debate, sometimes quite emotional. Both remedies seek to render the marriage void retroactively, but there is an important difference between them. Whereas retroactive cancellation due to a mistake is based on a fact which existed at the time of the betrothal, a terminative condition (in our context) is based on an event which will occur later. In the Babylonian Talmud we find a case which is introduced as a theoretical possibility, ultimately to be rejected (hava amina), although later poskim applied it in practice in some circumstances as a possible paradigm for retroactive cancellation of marriage. In that case a new circumstance which did not exist at the time of the marriage is the reason for voiding the marriage, and this ruling is justified by a legal presumption: had the wife known that this circumstance would develop, she would never have married her husband (!"#$% !"&'% ()% *+!&% (,-&(, 325 i.e., “on this assumption she did not betroth herself”). Such a case is defined in the rabbinic literature as a case of umdena (literally: assessment). Umdena is a concept commonly used by halakhic sources to fill in gaps in a wide range of halakhic areas: civil law (dinei mamonot), criminal law (including 326 capital jurisdiction), family law, etc. It can be applied as an assumption regarding factual events or as an assessment of the intentions of the parties. By applying 324 325 326 For the current purpose (as indicated above regarding marriage annulment: see supra, n.157), I refer to mistaken marriage and mistaken betrothal as synonymous unless otherwise explicitly specified. BT B.K., 110b-111a. See Talmudic Encyclopaedia, 1, s.v. ($&./(, 295-302. Chapter Seven: Mistaken Marriage, Conditional Marriage 79 umdena, various legal constructions become possible, depending upon the nature of the particular case. Thus, umdena can substitute for factual evidence, serve as 327 the basis for validating constitutive acts, and so on. In our case the function of the umdena is to reveal retrospectively the intention of the wife at the time of the marriage: “On this assumption she did not get 328 married.” But exactly what legal construction is used when applying the umdena? Do we assume the presence of an implied condition which retroactively annuls the 329 marriage? Or might the umdena reflect the existence of a mistaken transaction: the betrothal was initially based on a mistake and therefore never took place? In fact, we find three conceptual understandings of umdena in halakhic literature: that it is (a) an implicit condition; (b) a mistaken transaction; (c) a 330 combination of both these notions. This chapter explores each approach at length, and discusses the main halakhic writings which adopted each view. One responsum will be examined separately in the final section of this chapter: a responsum of Rabbi Moshe Feinstein, who makes a unique and in my view brilliant use of the integrated conception of umdena.! Our discussion derives from the Talmudic source, but focuses on post-talmudic literature. Its main object is the conceptual aspects of the umdena of ada‘ata dehakhi lo kidshah nafshah, and it is therefore a dogmatic analysis rather than an 327 328 329 330 See Ben-Meir, Re’ayah Nesibatit, 95-108, regarding umdena in betrothal and divorce. Ben-Meir, following H.S. Hefetz, focuses on umdena in the law of evidence, which is beyond the scope of this chapter, and therefore has a limited contribution to the current discussion (see the next note). It should be emphasized that the current discussion does not seek to analyze umdena as a general concept, as is done in some studies (see previous notes). My object here is the study of a specific umdena: ada‘ata dehakhi lo kidshah nafshah, and similar cases in other realms of the halakhah. However, this discussion has broader implications for the study of marriage and divorce in Jewish Law. Berachyahu Lifshitz argues that the Palestinian tradition as reflected in the Tosefta and the Yerushalmi, which was adopted by Rambam and the Geonim, distinguishes between two kinds of conditions: conditions of the type “if” (im) and conditions of the type “in order that” (al menat). The legal function of the two is different: “if” means that the legal validation of the action is not completed until the condition is fulfilled (for example, the couple is not fully considered married), whereas “in order that” means that the action is legally valid from the outset (the couple is married), but if the condition is not fulfilled, the action is retroactively annulled: see Lifshitz, Asmakhta, 140148, and see ibid., chap. 2, for more implications of this distinction. For the linguistic meaning of al menat see ibid., 162-169. Accordingly, our case is similar to a case of the type “in order that”: the marriage was fully valid, but there is an implied condition (condition subsequent) that a later event can retroactively annul it. One possible practical implication of defining a case as a condition rather than a mistake (ta‘ut) is that it may lead to certain formal halakhic requirements to which conditions are subjected (see Resp. Me’il Tsedakah, 2, 4b, s.v. "#"$, at the end). However, this is not a necessary conclusion, as shown by commentators who define umdena as a condition (see below), but ignore these formal requirements. See for example Resp. Beit HaLevi (Vilna, 1863), 3:3, 22-24, regarding umdena demukhakh; Talmudic Encyclopaedia, s.v. %#&'$%), 296-297. Westreich: Talmud-Based Solutions to the Problem of the Agunah 80 historical one: we seek to explore the significance of the three conceptions within halakhic literature and not to discuss the time and place of each conception’s evolution. Hence this presentation does not necessarily reflect the chronological order and historical importance of the halakhic sources. ! A further preliminary remark: classical mistaken marriage (kiddushei ta‘ut) and the use of conditional marriage (tnai) have been extensively discussed in previous 331 scholarship, and I do not seek to repeat well-known facts and conclusions. The focus here is therefore on the issue of umdena, which has not yet received proper conceptual clarification. Nevertheless, umdena in our context is closely related to the issues of kiddushei ta‘ut and tnai. Therefore, analysis of its conceptual significance may shed new light also on kiddushei ta‘ut and tnai themselves, both with regard to their conceptual substance and their use in practical halakhah. 7.2 Mistaken Marriage; Conditional Marriage Marriage in Jewish law reflects a contractual relationship: it requires the consent of both spouses and without such consent there is no marriage. Consent here means informed consent: when consent is based on misleading facts, there is no real 332 consent and thus no marriage. This is the core of the notion of mistaken marriage (kiddushei ta‘ut). More 333 specifically, according to many halakhic authorities, when betrothal is based on an error which existed at the time of the betrothal, and one spouse – for the present 334 purposes, the wife – was unaware of this, the marriage is void ab initio. From a conceptual viewpoint, the basis for kiddushei ta‘ut is the same as that for nullifying any other commercial transaction contracted in error: a transaction based on a mistake is void from the outset, and is considered as if it had never taken place. 331 332 333 334 For kiddushei ta‘ut see for example Broyde, Marriage, 89-102; Hacohen, Oppressed. For terminative conditions: see Berkovits, Tnai; Abel, Confronting ‘Iggun, Chapter One. Broyde, Marriage, 89-102. See also ibid., 33-35 (implications of a private law model of marriage). See Bass, Meka! Ta‘ut. Some poskim reject this notion due to considerations of halakhic policy, or because of a broad adoption of Resh Lakish’s presumption of tav lemeitav: see Bass, ibid., 195-201, and the discussion regarding Resh Lakish infra, text at nn.347-349. In a recently published article, Rabbi Eitam Henkin challenged Rabbi Bass’ view that many poskim support kiddushei ta‘ut. Bass however responded, in my opinion – correctly. See Henkin, Ha’omnam, 282-290; Bass, Batlut Nissu’in, 291 and further. In principle these claims are relevant to both spouses. However, according to some poskim it is harder to apply such claims to the husband since he has the option of unilateral divorce without paying a ketubbah, an option which does not exist for the wife. Therefore the claim of meka! ta‘ut was more easily applied in the woman’s favour. See Shut Noda BiYehudah, Mahadura Tinyana, Even Ha‘Ezer, 80 (last paragraph of the responsum). See also Broyde, Kiddushei Ta‘ut, 214 and 215 n.24, who ascribes this view to Rabbi !ayyim Ozer Grodzynsky and Rabbi Moshe Feinstein. Chapter Seven: Mistaken Marriage, Conditional Marriage 81 The concept of conditional marriage is different. Conditional marriage applies where a condition was made regarding a future event (conditions subsequent), whose occurrence (or non-occurrence, if the condition is formulated in a negative manner) retroactively voids the marriage, which originally was valid. Another kind of condition may also be mentioned here, a condition regarding facts at the time of the marriage (conditions precedent), which is a branch of the concept of mistaken 335 transaction. For example, if one betroths a woman on condition that she hasn’t made any past vows which still obligate her, and it was discovered that she had made such vows, her betrothal is not binding. Although in both types of condition the relevant sources use the expression ‘al menat (“on condition that”), the conceptual significance of this expression differs: in the first type of case the transaction is made subject to the occurrence of a future event, whereas in the 336 second this expression refers to a present situation. Accordingly, the latter type of condition is an integral component of the transaction, and violating it entails a mistake in the creation of the transaction, which in our context constitutes 337 kiddushei ta‘ut. This conceptual distinction also has practical implications. For example, the concern about voiding the condition between kiddushin and nissu’in, according to some writers, is relevant only in conditions which refer to the present status of the 338 wife and not in conditions which refer to a future event. This has significant consequences. According to this view it is possible to make a condition which refers to a future event. Thus, the notion that there is no conditional marriage (ein tnai benissu’in; see BT Yevamot, 107a) is merely a descriptive statement: people usually make conditions which apply only to betrothal, and ordinarily such conditions are either waived before marriage, or they prevent the couple from 339 getting married in the first place. In our case, that of conditional marriage, the marriage was created properly. Both spouses were aware of every important fact and fully agreed to the marriage. The marriage is therefore halakhically valid. However, the spouses made a condition which may lead to retroactive cancellation of the marriage. The continuing validity of the marriage therefore depends on that condition, even after 335 336 337 338 339 BT Ketubbot, 72b-75a. Cf. Lifshitz, Asmakhta, 166-167. Therefore the term ta‘ut is used in the Talmud and by talmudic commentators in this context; see BT Ketubbot, 73b and Alfasi, Ketubbot, 34a. This is also the meaning of kiddushei ta‘ut in the dictum of R. Judah in the name of Samuel in the name of R. Ishmael, ibid. See also Rashi, ibid., 51b, s.v. "#$%&'(#)!. See Berkovits, Tnai, 23; Abel, Confronting ‘Iggun, 13-16. See Tosafot on Ketubbot 73a, s.v. *+. Westreich: Talmud-Based Solutions to the Problem of the Agunah 82 the marriage took place. This kind of condition may be defined as a terminative condition. Indeed, not all conditions regarding a future event necessarily lead, when violated, to cancellation of the marriage. There are conditions which include an agreement on acts which must be done or not be done, but they do not affect the 340 validity of the marriage. We deal with such conditions later, when discussing some unique approaches to umdena. However, for the conceptual analysis here in general, and for comparison of the concepts of kiddushei ta‘ut, tnai and umdena in particular, we focus on terminative conditions. The last century witnessed great debates, generated by a variety of halakhic problems and approaches to halakhic policy, as to whether conditional marriage 341 could be entered as a matter of actual practice (halakhah lema‘aseh). However, the conceptual basis of conditional marriage was generally accepted as viable, at least in principle, and we shall focus on this dimension, without considering for the present problems of its concrete application. The latter question is mainly one of authority: are the halakhic decisors of our days allowed to adopt such a solution, despite its possible problems? 7.3 “And the Sages Did Not Have the Power to Release Her without a Get” (BT Ketubbot 74a) Before discussing the retroactive annulment of marriage based on ada‘ata dehakhi lo kidshah nafshah (on this assumption she was not married), I wish to indicate the difficulty, both substantive and epistemological level, in retroactive annulment of marriage based on a stipulation or on error. The Mishnah (Ketubbot 7:7) states: If a man betrothed a woman on the condition that she was under no vow and she was found to be under a vow, her betrothal was not valid. If he married her [!"#$] making no conditions [%&"], and she was found to be under a vow, she may be divorced ['(&, lit., 340 341 Those conditions are mainly monetary, dealing with ketubbah, alimony, etc. Some prenuptial agreements are based on such conditions: for example, if the husband refuses to give a get to his wife, he will pay a large sum of alimony. We may also include in this category the monogamy condition, which sometimes includes an agreement for a coerced get when the husband marries another wife: see E. Westreich, Temurot, 26-28. See Abel, Confronting ‘Iggun, 6-45. Two halakhic scholars recently published a series of articles which reject any use of conditional marriage: see Gertner and Karlinsky, Ein Tnai BeNissu’in. These articles emphasize that the main arguments are based on halakhic policy, while substantive issues are hardly mentioned, and most of these can be satisfactorily resolved. See, for example, the writers’ objection to Rabbi Berkovits’s suggestion, a suggestion which was supported in principle by Rabbi Y.Y. Weinberg. Gertner and Karlinsky’s objections to this approach are total, no matter what halakhic basis may be found for it (see the final conclusion to the third and last article, s.v. )*+)%,-,./ %.,/0). Chapter Seven: Mistaken Marriage, Conditional Marriage 83 go out] without [receiving] her ketubbah. [If he betrothed her] on the condition that there were no defects in her, and defects were found in her, her betrothal is not valid. If he married her making no conditions, and defects were found in her, she may be divorced without [receiving] her ketubbah. This Mishnah teaches that when the husband made an explicit stipulation, and it became known after the betrothal that his condition was not fulfilled and the 342 marriage is based on an error, the betrothal is retroactively voided. However, as regards the status of the error after the wedding, the picture is by no means clear. According to the Mishnah, if “he married her making no conditions”, the woman requires a get, and she is in an inferior position regarding monetary matters (“without receiving her ketubbah”). Consequently, if the husband had explicitly stipulated the condition prior to the wedding, the condition is valid, and the marriage is retroactively voided. The Mishnah, however, does not state this expressly, and the views in the Babylonian Talmud on this point are not unanimous. This Mishnah is the subject of an extensive discussion in the Babylonian Talmud. The sugya centres around the question of whether a stipulation at the time of the betrothal is valid after the wedding as well. Thus, using the example given in the Mishnah, if the husband stipulated at the time of the betrothal that the betrothal is effected on condition that the woman is under no vows, and later married her, but then she was found to be under vows, is the marriage annulled, with no get necessary; or does she require a get from her husband? The sugya, which extends over about two pages in the Talmud, presents a number of views as to the validity of the marriage, and a number of interpretations for the various positions. Rav and Shmuel, for instance, dispute this point: while Rav demands a get, Shmuel argues that a get is not required. Towards the end of the sugya (BT Ketubbot, 74a) the following position is cited, and supported by an actual case: Rav Kahana said in the name of Ulla: If a man betrothed a woman on a certain condition and then had intercourse with her, she requires a get from him. Such a case once occurred, and the Sages did not have the power to release her without a get. According to this opinion, the act of intercourse (which effects marriage) cancels the stipulation made at the time of the betrothal. The decision in this case is quite forceful: “and the Sages did not have the power to release her without a writ of divorce”. That is, in this discussion the Sages tried with all their might to release the woman from marriage without the need for a get. Their efforts, however, were 342 The sugya is concerned with stipulation (!"#$), but its legal construct is that of mistaken betrothal. See supra, text at notes 335-337. 84 Westreich: Talmud-Based Solutions to the Problem of the Agunah not fruitful, and they required the issuance of a get. In opposition to the stance of Rav Kahana in the name of Ulla, the Talmud cites the position attributed to Rabbi Ishmael (74a-b): [The dictum by R. Kahana is meant] to exclude [the view] of the following Tanna. For Rav Judah said in the name of Shmuel in the name of Rabbi Ishmael: “And she be not seized” [Num. 5:13] – [only then is she] forbidden; if, however, she was seized, she is permitted. There is, however, another [kind of woman] who is permitted, even though she was not seized. And who is she? A woman whose betrothal was mistaken, and who may – even if her son sits riding on her shoulder – make a declaration of refusal [against her husband] and go away [i.e., leave the marriage]. Talmudic literature contains several expositions of the law of “and she be not seized” (such as the deduction that the wife of a kohen who engaged in sexual relations with another, even if raped, is forbidden to her husband: “but there is another class of woman who is forbidden, even though she was seized. And who is that? The wife of a kohen” [BT Ketubbot, 51b]). In our case, R. Ishmael expounded the verse as follows: the sotah (wife accused of infidelity) who was not seized is forbidden to her husband. If, however, she was seized (that is, raped), she is permitted to her husband. All this, R. Ishmael establishes, refers to the sotah; but there is “another class of woman,” who is permitted, even if she willingly engaged in sexual relations. This permitted woman is one who was betrothed in error (such as in our case: the husband made a condition regarding facts existed at the time of the betrothal, and it turned out to be a mistaken betrothal). In this case, the marriage is retroactively not valid, even if much time has passed, so that she already gave birth and “her son sits riding on her shoulder.” In such a case, the first marriage is retroactively voided, and she is therefore permitted to either of these men, even if she engaged in relations with another. All that she must do is to declare that she does not desire to remain married, and she can go on her way unhindered. From a stylistic perspective, and also in terms of the literary structure of the sugya, it seems that the case mentioned in the statement by Rav Kahana is part of his dictum, and is not an interpretive-redactional addition. The case plays a significant halakhic role, providing additional support for Rav Kahana’s halakhic ruling. Rav Kahana uses this case to show that, despite those who permitted releasing the wife without a get, and despite the Sages’ inclination to do so, in practice, the Sages did not have legal grounds to do so. The interpretive and redactional elements in this sugya – here referring not necessarily to later redaction, but rather to the formulations by the Amoraim themselves – are highly significant, and the role they play in Rav Kahana’s statement cannot be dismissed. A strong argument can be raised, inter alia, that the halakhah presented in the name of R. Ishmael, too, is based on an actual case, Chapter Seven: Mistaken Marriage, Conditional Marriage 85 possibly the one that was mentioned by Rav Kahana elsewhere (see BT Niddah 343 52a). We cannot, however, disregard the force of the teaching by Rav Kahana, that the practical attempt to void the marriage failed, for “the Sages did not have the power to release her without a get”. Eliezer Berkovits extensively examined this issue in his attempt to establish the talmudic and post-talmudic foundations for conditional marriage. He based the halakhic sources that justify this solution on the various opinions set forth in the sugya and the differing interpretations given to it. This was recently supported by 344 Yehudah Abel. Notwithstanding these attempts, it is possible that the impression made by R. Kahana’s case influenced the post-talmudic halakhah, in its objection to mistaken betrothal and stipulation in marriage, because of the difficulty in the attempt “to release her without a get”. The halakhah, therefore, found a different way to establish the basis for the institutions of erroneous betrothal and stipulation in marriage: “on this assumption she did not get married”. This path is complex and sophisticated, contains elements of both stipulation and ta‘ut, and possibly of both together. As argued in the next section, it also facilitates the use of the error argument in relation even to a future event. 7.4 Ada‘ata dehakhi lo kidshah nafshah (Umdena) – Mistake or Condition? We find the umdena that ada‘ata dehakhi lo kidshah nafshah (“on this assumption she did not betroth herself”) in a well-known talmudic source (BT Bava Kamma, 110b–111a). The talmudic discussion deals with a few different transactions: consecrating an animal as an offering, marriage (or betrothal, according to 345 Tosafot), etc. In each of the cases discussed here a later event occurs, and we assume that, had it been known at the time of the transaction that such an event would take place, the transaction would not have been entered. The term used here is ada‘ata dehakhi lo…, i.e. “on this assumption [that such an event would take place] one would not have entered the transaction.” As to marriage, the concept is 346 discussed with reference to the case of a levir who has a severe skin disease: 343 344 345 346 I intend to expand on this in the future. For now, see Westreich, Halakhic Story. See Berkovits, Tnai, Chapter 1; Abel, Confronting ‘Iggun, Chapter 1. These cases are not classic contracts; see infra, text at nn. 375-377. The term used here, mukkeh she!in (!"#$% &'()), literally means “afflicted with boils”. This expression probably refers to a kind of leprosy: see Preuss, Biblical and Talmudic Medicine, 346347. 86 Westreich: Talmud-Based Solutions to the Problem of the Agunah But still, according to your argument why should a deceased brother’s wife on becoming bound to one affected with leprosy [mukkeh she!in] not be released [even] without the act of !alitsah, for she did not betroth herself on this assumption? Levirate marriage (yibbum) and !alitsah refer to the relationship which arises between a childless widow and the brother of her deceased husband. Under Jewish law they are obliged to marry one another, and indeed even prior to any ceremony they are connected by the bond of a “levirate marriage”. Their quasi-marital condition, however, can be dissolved by a ceremony called !alitsah, in which (among other things) the widow removes one of her brother-in-law’s shoes. The brother-in-law may, however, refuse to perform !alitsah, leaving the widow bound and unable to remarry. In this talmudic passage the Talmud argues that if the levir is affected with leprosy, no yibbum or !alitsah would be required at all, since the wife did not betroth herself on the assumption that she would be required to marry a brother-in-law with such a physical condition. The assumption ada‘ata dehakhi lo is a legal presumption – one that claims to discover the implicit thoughts which were part of the transaction. In rabbinic 347 literature this presumption is defined as umdena. Before exploring the meaning of umdena, one comment should be made. The sugya rejects the wife’s claim by citing Resh Lakish: In that case we can all bear witness that [the wife] was prepared to accept any situation, as we learn from Resh Lakish. For Resh Lakish said: It is better to dwell as two than to dwell in widowhood (!"#$%&'()#"#&!*&+,&'()#"&',). Resh Lakish presumes that a woman always prefers to be married, and thus can never claim ada‘ata dehakhi lo kidshah nafshah. However, numerous 348 commentators argue that this presumption is not always applicable, and there are cases where women do prefer to remain unmarried. In such cases it is legitimate 349 therefore to claim ada‘ata dehakhi lo. 347 348 349 The term is common in the writings of more recent commentators (A!aronim) who discuss the conceptual basis for this talmudic discussion (Beit Halevi, R. Shimon Shkop and others: see below), but it is also found in earlier sources dealing with this issue. See e.g. Rosh, Ketubbot, 9, in connection with Ket. 47b, who defines the approach which accepts the claim that “he wrote for her [i.e. he obligated himself in the ketubbah] only [on the assumption that] he would marry her” (&%"./0!1"& (0#& "2& %"%& ."& '(1, Hebrew equivalent of the Aramaic %"& )1.*& %(2*%) as “he follows ... umdena” (%0*#!%&$('&333")4%). See e.g. Broyde, Marriage, 98-100 and 175-176 n.62; Hacohen, Oppressed, 45-92. One of the well known opponents of the (widespread?) use of ada‘ata dehakhi lo in actual halakhic practice is Rabbi J.B. Soloveitchik, who suggests an ontological understanding of Resh Lakish’s presumption, according to which it is not subject to sociological or psychological changes, and is indeed immutable. See Bleich, Kiddushei Ta‘ut, 106-107 and 124-125 n.28. Broyde, Marriage, 174 n.55, argues on the basis of a complex of halakhic sources that total rejection of ada‘ata dehakhi lo cannot be accepted and that even Rabbi Soloveitchik would agree with this. R. Soloveitchik’s view Chapter Seven: Mistaken Marriage, Conditional Marriage 87 It is difficult to propose a conceptual definition of the umdena discussed here. On the one hand, this umdena addresses the moment of the transaction (for our purposes, marriage) by arguing that if the event or the changed fact – which, to be sure, occurs in the future – had been known at that moment, the spouse would never have agreed to the marriage. From this perspective, our umdena seems to be 350 a kind of kiddushei ta‘ut. On the other hand, the fact that this is a future event brings the umdena close to the concept of condition, i.e., the marriage is valid, but a future occurrence changes its status retroactively, based on an implicit stipulation of the couple at the time of marriage. Thus, many writers seem to construe the 351 umdena of our case as an implicit condition. A clear conceptual discussion is offered by Rabbi Shimon Shkop in his Sha‘arei 352 Yosher. What makes his analysis of great importance for our discussion is the fact that Rabbi Shkop applies his conceptual arguments to our sugya, i.e. to the claim of ada‘ata dehakhi lo with regard to the case of a leprous levir. Rabbi Shkop criticizes the view that identifies umdena as a mistaken transaction, ascribing it to Maharit Al-Gazi. Maharit deals with mistaken donation of a firstborn animal (bekhor behemah tehorah) to a priest (kohen), where the owner was not obliged to give the firstborn to the priest. He then compares this case to umdena regarding a future event: The same applies when a person rents a boat and [the boat] sinks in the middle of the journey [i.e. during the rental] ... [Here] also he did not give [the payment] on this presumption [that it would sink]. In the last case we do not accept the claim that the renter didn’t pay to use the boat for only half a journey. Maharit compares this case to a mistaken donation of a ---is therefore “limited to opposing the wholesale abandonment of the principle [=of Resh Lakish] rather than merely asserting that it did not apply in any given case or set of cases” (Broyde, ibid.). In any case, according to Broyde, many poskim do view this presumption as subject to socio-cultural changes (see also the analysis of Rabbi Y.E. Spektor and Rabbi M. Feinstein’s views regarding Resh Lakish’s presumption by Halperin-Kaddari, Tav Lemeitav, 21-24). It is noteworthy that Broyde proposes a similar argument with regard to interpreting Rabbi Yosef Eliyahu Henkin’s objection to kiddushei ta‘ut in a limited way and not as an objection on principle, an argument which was rejected by Bass: see Bass, Meka! Ta‘ut, 197 n.12. 350 351 352 See for example Shut Sho’el uMeshiv, Mahadura Kamma, 1:197, who explains that it “retroactively becomes kiddushei ta‘ut.” Additional sources which adopt this view are cited below. This seems to be a widespread view regarding umdena in cases similar to ours in various realms of the halakhah. See for example Shut HaRosh, 34:1. See further Talmudic Encyclopaedia, s.v. !"#$%!, 295-302, according to which umdena is an implicit alternative to an explicit condition (different kinds of umdena are discussed in the Talmudic Encyclopaedia; the relevant one for our discussion is that which entails an assessment of the intention of the actor). See especially ibid., 296-297, part 3. Rabbi Shimon Shkop, Sha‘arei Yosher (New York: HaVa’ad Lehotsa’at Sifre HaGaon Rabbi Shimon z”l, 1959), 5:18, 68-70. 88 Westreich: Talmud-Based Solutions to the Problem of the Agunah firstborn animal, and therefore argues that one cannot claim for a mistaken giving 353 of a firstborn to a kohen. However, the mistake regarding the firstborn takes place at the time of its being given, and therefore it is a pure case of mistaken transaction, so how can it be compared to renting a boat that later sank? R. Shkop thus deduced that Maharit identifies umdena as a mistaken transaction, and does not distinguish between a mistake with regard to an existing fact and a mistake with regard to a future one. But R. Shkop rejects this view. According to him, the case of the firstborn is a mistaken transaction, where the transaction was based on an error and is cancelled ab initio. By contrast, the case of the sinking boat cannot be regarded as a mistaken transaction, since there was no error at the moment of the transaction (in R. Shkop’s words: “For it is impossible to know the future”). Umdena can therefore only be regarded as an implicit condition. R. Shkop applies the distinction between umdena and mistaken transaction to marriage: Regarding what is stated in the Talmud, where it is proposed that a deceased brother’s wife (yevamah) who found herself bound to a leprous levir should be free without !alitsah, because “on this assumption she did not get married”, it is not possible to explain that this means that it is retrospectively clarified that there was an error in the marriage, for how is it possible to say that there was an original error? What it means is that the [marriage] was in suspense, as if it had been a conditional marriage. R. Shkop’s argument seems to be convincing: one cannot claim that there was a mistake at the time of the marriage due to a fact which was not then in existence. Nevertheless, there are a number of authorities who take the opposite view and 354 define umdena as a mistaken transaction. The most explicit is She’elot uTeshuvot 355 356 Me’il Tsedakah. Discussing the case of a levir, he writes: It seems to me that the questioner [in the Talmud] also did not mean to ask whether [the wife] should go free without !alitsah because this is considered as if there had been a condition [in the marriage]. What he meant was that it was like a mistaken transaction, a 353 354 355 356 Maharit ascribes his view to Ramban and Rosh, contrary to Rema, Yoreh De‘ah, 315:1, and Shakh, Tekafo Kohen, 62 (cited in Sha‘arei Yosher, ibid.). In addition to the writers cited below, see Shut Sho’el uMeshiv (supra, n.350), who probably holds this view. However the conceptual approach of Sho’el uMeshiv is not fully clear, since at the same time he discusses the relation between umdena and regular conditions, and seems to understand umdena as a form of the latter (see ibid.). Rabbi Yonah Landsofer, Shut Me’il Tsedakah (Prague, 1757), 2, 3b. Me’il Tsedakah (p.4a) also ascribes this view to Maharam of Rothenburg, according to the version of Maharam’s teaching cited in Shut Maharam, ed. Prague, 564, but this conclusion is questionable. BT B.K., 110b, quoted above. Chapter Seven: Mistaken Marriage, Conditional Marriage 89 marriage contracted in error, as when he says to her “[I marry you] on the understanding that I am a kohen” and he was discovered to be a Levite. Thus, disregarding the fact that the significant event occurred after the transaction, this approach defines the transaction as a mistaken one. In fact, this view is found in sources much earlier than Me’il Tsedakah. 357 Ra’avyah defines umdena as a mistaken transaction, and applies this notion to the case of a leprous levir (“there too she was betrothed in error”). According to Ra’avyah, the umdena here is not based on a condition, since the parties were not aware of the future occurrence that might require such a condition. Hence it is only 358 “we” – the poskim or court – who assess the parties’ intention. We accordingly assume that the parties would not have agreed to the transaction had they contemplated this occurrence, and this is therefore considered a mistaken transaction. Astonishingly, Ra’avyah supports his view with the same argument and even uses the same terminology as was used centuries later by R. Shimon Shkop in support of exactly the opposite view, namely that umdena is a condition and not an error! Ra’avyah writes: But umdena is a mistaken transaction (literally: giving in error) ... and so all cases of umdena entail error, [because] no one knows the future at the time of the transaction. Thus, whereas R. Shkop argued that the fact that one didn’t know the future at the time of marriage (“for it is impossible to know the future”) makes it impossible to define a case as one of ta‘ut, Ra’avyah argues that this very fact makes it impossible to define the case as a condition! However, many other writers, both talmudic commentators and halakhic decisors in their responsa, define our umdena as an implicit condition and accordingly reject the equation of umdena and ta‘ut. This view can already be found in the medieval commentators, although it is not always expressed as clearly there as by R. Shimon Shkop. For example, it is found in the writings of the 359 Tosafists, as we may conclude from the following remarks of Tosafot HaRosh: For if [the wife] had wanted to make a condition at the time of betrothal that if [the husband] should die before he married her, then the betrothal would be annulled, so that 357 358 359 Rabbi Eliezer b. R. Joel Halevi, Sefer Ra’avyah, Teshuvot UVe’urei Sugyot (ed. David Deblitski, Bene Brak, 2000), §1032, s.v. !"!#$%&'(, 321-322. “[In cases of] umdena it is not in the person’s mind to make a condition, so at the time of the transaction he did not consider making any condition ... but we assess his [subconscious] intention ... therefore what reason could he have had to make a condition [as he was unaware of future developments]?!” Cited in Shitah Mekubetset, B.K. 110b. This is an expanded version of Tosafot on B.K., 110b, s.v. '!)*'; see below. Rosh himself holds the same view: see Shut HaRosh, 34:1, regarding a case of breaching a marriage agreement due to conversion of the bride’s sister. 90 Westreich: Talmud-Based Solutions to the Problem of the Agunah she would not find herself bound to his leprous brother, the husband would not have objected. Therefore it is considered as if she had made such a condition. What is the content of this implicit condition? The most widespread interpretation sees it as follows: if a future event occurs, the marriage will be regarded as 360 invalidated from its beginning, even though it was created properly. According to this interpretation umdena is comparable to a terminative condition. However, we also find a different view, which understands umdena as a normal (nonterminative) condition, without any retroactive invalidation, comparable to standard conditions pertaining to certain aspects of the marriage contract. This 361 view is found in Terumat HaDeshen, who defines the umdena in our sugya as “a manifestation of [the husband’s] intention that the levirate bond should not take effect”. Accordingly, the umdena here does not invalidate the marriage, but 362 releases the wife from the levirate obligation. Thus, the marriage is valid even if 363 the umdena is applied, and only the duty of levirate (zikah) is terminated. But there is an even more far-reaching view of the function of the implicit 364 condition. In She’elot uTeshuvot Binyamin Ze’ev we find the view that the umdena (the implicit condition) refers to the levirate marriage itself rather than to the levirate bond (zikah). Accordingly, the implicit condition was that if such circumstances occur the wife would not be obliged to enter into a levirate marriage with her apostate brother-in-law. However, although she is not obliged to have yibbum, she has in principle the duty to undergo !alitsah. But she is exempted due to another halakhic principle: “whoever is subject to the obligation of levirate marriage is also subject to !alitsah and whosoever is not subject to the obligation 365 of the levirate marriage is not subject to !alitsah”. According to this view, 360 361 362 363 364 365 See Sha‘arei Yosher, supra, n.352, ibid. Many more recent commentators (A!aronim) interpret matters this way, following Mahari Bruna against Terumat HaDeshen: see below. Shut Terumat HaDeshen, 223. According to Terumat HaDeshen the Bavli accepts this condition, whereas the Yerushalmi considers it a stipulation in violation of Torah law, rendering the condition void. See Terumat HaDeshen, ibid., and infra, n.370. Terumat HaDeshen’s argument was rejected by Rabbi Yosef Karo: see below.! See Shut Maharam Schick, 70, 35; Berkovits, Tnai, 29-32. However, some more recent commentators argue that even Terumat HaDeshen deals with a condition which invalidated the marriage. See for example Rabbi Avraham Brody, cited in Shut Me’il Tsedakah, 1, 1a-3a; Maharam Schick, ibid. In my opinion, however, the first approach is more persuasive. Shut Binyamin Ze’ev, 71. )*#&+&! )&(,! ("#$! -(.##&! )&(,! ("#$%! &'(! )*#&+&! )&(,! -(.##&! )&(,)! &'. This principle explains why a woman who is forbidden to the levir on the grounds of incest (more accurately, in the specific talmudic context: the second wife of the dead man, who herself is not forbidden; /.)!/0*) is exempt not only from the yibbum, but also from !alitsah: see Bavli, Yevamot, 3a and elsewhere. Chapter Seven: Mistaken Marriage, Conditional Marriage 91 umdena is a regular condition, and – quite unlike kiddushei ta‘ut – does not invalidate the marriage at all, but only avoids the obligation of yibbum and consequently the obligation of !alitsah as well. Analyzing Terumat HaDeshen is important also with regard to a related issue: 366 whether an explicit terminative condition is valid. In this responsum, Terumat HaDeshen informs us of a concrete custom in his days of making an explicit condition at the time of marriage in order to avoid levirate marriage in the case of an apostate levir (yavam mumar). Later, this condition was rejected by Rabbi Yosef Karo, but Rema accepts a very similar condition in the name of Mahari 367 Bruna: Shul!an Arukh: If [a woman] found herself bound to an apostate levir, there is someone who permits [her release without !alitsah] if [the brother-in-law] had been an apostate when his brother married her, but one should not rely on this. Gloss of Rema: ... a person who wishes to marry and has an apostate brother is permitted 368 to marry with a double condition stating that if she finds herself bound to the apostate brother, then she will not have been married [in the first place]. Nevertheless, there is a clear difference between the views of Terumat HaDeshen 369 and Mahari Bruna. According to the former, the condition (whether explicit or implicit, by umdena) cancels the duty of levirate, and thus is problematic in that it entails a stipulation made in violation of Torah law (matneh al mah shekatuv 370 baTorah), which is void. The latter, on the other hand, suggests a condition 371 which explicitly invalidates the marriage, and therefore is not matneh al mah 366 367 368 369 370 371 For full and detailed discussion regarding Terumat HaDeshen’s condition vs. Mahari Bruna’s condition (below), see Freimann, Seder Kiddushin, 386-394; Berkovits, Tnai, 29-32. Shul!an Arukh, EH 157:4, and Rema, ad loc. As indicated, the source of Rema’s ruling is Mahari Bruna, who is cited fully in Rema’s Darkhei Moshe on Tur, EH 157:5. Rabbi Yosef Karo, on the other hand, explicitly rejects Terumat HaDeshen’s condition in his Bet Yosef on Tur, EH 157, end of s.v. !"#$%&'()&*%*(&+, but does not refer to Mahari Bruna's condition. A “double condition” (i.e., if A then B, and if not A, then not B) is required due to the formal rules of conditions: see Maimonides, Ishut, 6:1-6. However, this is not the core of the distinction between this condition and that of Terumat HaDeshen; see Berkovits, Tnai, 31, citing Shut Tsela’ot HaBayit, and see below. See Bach, EH 157, s.v. ,)-)"%*)%*&+; Resp. Maharam Schick, 70, 35; Berkovits, Tnai, 29-32. Some writers reject this view for various reason: see Rabbi Avraham Brody, cited in Resp. Me’il Tsedakah, supra n.363. This principle is widely accepted in rabbinic literature; see for example T Kiddushin 3:7-8 and BT Kiddushin 19b. The details of this principle, however, are disputed, so that Terumat HaDeshen might argue that according to the Bavli his condition is valid; see supra n.362.% See Rema, above: “If she finds herself bound to the apostate brother then she will not have been married [in the first place]”. Westreich: Talmud-Based Solutions to the Problem of the Agunah 92 shekatuv baTorah, so that the condition is valid. Accepting this distinction (i.e. the conceptual distinction between the rejected view of Terumat HaDeshen and the accepted view of Mahari Bruna) may legitimate the use of explicit terminative 372 conditions in order to prevent cases of agunot, as suggested by some 373 contemporary writers. To conclude this section, both ancient and modern authorities are divided in defining umdena. Some writers compare it to kiddushei ta‘ut, arguing that had the bride known about this fact at the time of the marriage, even though it did not yet exist, she would not have married her husband, and therefore the whole transaction is regarded as mistaken. However, other writers classify umdena as an implicit condition, according to which the marriage is valid at the moment of its creation, 374 but can be invalidated retroactively by this condition. 7.5 The Integrated Approach Alongside the two basic approaches to umdena there is a third, which combines the other two. Finding evidence of this approach requires a close reading of different passages of Tosafot, to which we now turn. This integrated approach has been accepted in practice in our days by Rabbi Moshe Feinstein, as will be shown in the final section. At first sight, Tosafot on the sugya of a leprous brother-in-law (on Bava Kamma, 110b, s.v. !"#$!) reflects an understanding of umdena as a condition. Here Tosafot interprets the initial part of the talmudic discussion about ada‘ata dehakhi lo as referring only to the case of a betrothed wife (%&'(!). Tosafot then raises the following question: And if you ask: when a person buys an item from another person and it [later] breaks, could he cancel the sale because [he can claim] that he did not buy it on this understanding [that it would eventually break]?! The answer is that there the matter is not dependent only on the buyer but also on the seller, and the seller sold it to him on that understanding [that if it is broken in the future, this is the buyer’s risk]. But here, the betrothal depends on [the woman], as he is not 372 373 374 A few halakhic authorities, including followers of the Sephardic tradition (e.g. Rabbi Yosef Shmuel Modiano from Salonica and !ikrei Lev: see the citations and discussion in Freimann, Seder Kiddushin, 387-388) argue that even the opponents of Terumat HaDeshen’s condition (including R. Yosef Karo) would have changed their minds had they been aware of the difference between Terumat HaDeshen and Mahari Bruna. See mainly Berkovits, Tnai, 49-51. According to Terumat HaDeshen, the condition does not retroactively annul the marriage, but rather releases the wife from the levirate obligation.) Chapter Seven: Mistaken Marriage, Conditional Marriage 93 concerned how [i.e., under what conditions] she wants to become betrothed. Similarly regarding a person who consecrated [an offering], everything depends on him, and so is it with one who restores money that was stolen from a proselyte [who died] – it all depends on him. According to Tosafot, the umdena of ada‘ata dehakhi cannot stand on its own, but, being a term of the contract, requires the agreement of both parties to the transaction. Therefore, in a normal commercial transaction the buyer cannot cancel the sale based on the argument of ada‘ata dehakhi when something happens after the transaction. Nevertheless, in the cases discussed in the sugya (giving priests 375 money stolen from a proselyte who subsequently died, consecrating something as an offering, and a leprous levir) such agreement is either not required (as in the 376 first two cases, which are not mutual transactions, but unilateral acts ) or is assumed implicitly to exist (as in the case of a sister-in-law potentially subject to 377 levirate ). From a conceptual viewpoint, as already noted, in order to determine that a transaction is mistaken, one needs only to consider the viewpoint of one of the parties to it: if that party was not aware of an important fact pertaining to the transaction, the transaction is deemed mistaken and is invalidated ab initio. On the other hand, a condition reflects an agreement between both parties to an initially valid transaction. Every condition must therefore be made with the agreement of both parties. According to Tosafot, we can claim ada‘ata dehakhi lo only if we assume that both sides would agree to the condition. This requirement clearly reflects the view of umdena as an implicit condition: the assumption that the marriage will be invalidated on such an occurrence must be based on a preliminary mutual agreement, i.e., it should be part of the contract between the two parties to the transaction. If one spouse does not agree, the marriage remains valid but is unconditional – unlike, for example, the case where the wife was not aware of a serious disease of her husband, in which case she could claim kiddushei ta‘ut without her husband’s agreement. The basis of Tosafot’s discussion is therefore the understanding of umdena as a condition, and in principle we need at least the implicit agreement of both parties 375 376 377 The Mishnah (B.K. 9:11-12) discusses the first case. The Babylonian Talmud compares this case with the other two cases.! See Tosafot on Ket. 47b, s.v. "#$: “[the transaction] depends only on the giver”. According to Tosafot, ibid., the Talmud refers here to a betrothed woman and not to a married one, and therefore the husband would not mind invalidating the betrothal in the event of his death, if his brother is afflicted with a skin disease. By contrast, had the husband been married, he might oppose this option, due to concern that nullification of the marriage would render his cohabitation promiscuous. 94 Westreich: Talmud-Based Solutions to the Problem of the Agunah 378 to the transaction. We find this view almost explicitly in Tosafot HaRosh, which 379 at this point may be regarded as a better-explained version of our Tosafot. However, the interpretation of umdena as a condition is not the only analysis 380 found in Tosafot. Elsewhere, we find an additional element, which reflects a more complicated conception of umdena. In Ketubbot 47b, the Talmud discusses the case of a betrothed woman who got divorced or became a widow. R. Elazar Ben Azaryah maintains that the betrothed wife receives her basic ketubbah (100 or 200 zuz), but is not entitled to any additional ketubbah payments that her husband promised. R. Elazar Ben Azaryah’s reasoning is that the husband obligated himself only on the assumption that he would marry her, and therefore if the marriage does not take place, whether because of death or divorce, his wife does not receive any of these additional payments. Tosafot ad loc. (s.v. !"#) raises a question similar to that raised in connection with the sugya of a leprous levir: And if you ask: If so, [when] anyone buys a cow from someone else and it becomes terefah [nonkosher] or dies, we can testify that he did not buy it on that understanding [and he should therefore always be able to obtain a refund] …! But contrary to Tosafot in Bava Kamma, Tosafot here suggests a different answer: The answer is that there we can be sure that [the buyer] would have been willing to take that chance [literally: he would have been willing to enter into that doubt, i.e., the possibility that the cow might die]). Indeed, even if the [seller] said to him: “If it becomes terefah you must accept the loss”, he would have bought it. Here, however, he wrote [the addition to the ketubbah] only on the understanding of marrying her, and he had no intention whatsoever of entering into a doubt[ful situation]. The umdena here depends on the following argument. We ask whether the party to the contract would have accepted the agreement even if s/he had been aware of the possibility that a certain future event might occur. If we assume that had s/he known this possibility at the time of the transaction s/he would have accepted the 378 379 380 .'0.%(/"'!+(*/#4(5+"%( $%&'(")(*+),(")*%(%'%(!"(---./0.%"(%1/2(%.'%(3!&. For the full quotation see supra text at n.359. Tosafot HaRosh is a collection of some corpora of earlier Tosafot: Those of R“i, Rash of Sens and others. On the redaction of Tosafot by the Rosh (the “Tosafot HaRosh”), compared to “our” Tosafot (which are cited in the printed editions of the Babylonian Talmud), see Urbach, Ba‘aley HaTosafot, Vol. 2, 585-599. The following is “our” Tosafot of Ketubbot while the former is Tosafot of Bava Kamma. According to Urbach, Ba‘aley HaTosafot, 625-629; 639-645 (especially at 642), both were redacted by Rabbi Eliezer of Touques (5/7,(28)'"!( 62), but for Bava Kamma Rabbi Eliezer based his text on Tosafot Rabbi Yehuda Sirleon, whereas for Ketubbot (and usually) he based it on Tosafot Rash of Sens. Chapter Seven: Mistaken Marriage, Conditional Marriage 95 agreement anyway, taking the “risk” of the future event (as s/he would in the case of any normal transaction), s/he cannot cancel the transaction later on the basis of ada‘ata dehakhi lo. But if we assume that had the party known that a future event might occur s/he would not have accepted the agreement due to that future possibility, s/he can claim ada‘ata dehakhi lo. This explanation is not conceptually clear. Is it an implicit condition or a sort of meka! ta‘ut? And what is the relation between this concept and the mutual agreement required by Tosafot in Bava Kamma on the sugya of the leprous levir? Are these explanations compatible, or are they different and contradictory 381 approaches? Further on in Ketubbot, Tosafot (ibid.) discuss the need for mutual agreement, in terms similar to those presented in Bava Kamma. But in Ketubbot the discussion regarding the need for mutual agreement is related to the previous discussion, i.e., to the question of whether the party to the contract would have accepted the agreement even if he or she knew that a particular event might occur in the future: And regarding what [the Talmud] asks at the end of the first [chapter] HaGozel: “A yevamah who found herself bound to a leper should be released without !alitsah, because she did not [agree to] marry on that understanding” – although she would probably have taken that risk at the time of betrothal, Rabbenu Yits!ak explains... that since it depends 382 only upon the one who gives (i.e. the wife), we should follow her intention, and since it depends upon her, [we may say] that she certainly does not want to take any chances. This cannot be compared to one who buys an object to which an accident occurs, in which case we do not say that he did not buy it on that understanding, so he can annul the transaction. For [in that case, the matter] does not depend on the mind of the buyer only, as there is also the mind of the seller, who [we presume] would not sell to him in accordance with [the buyer’s] intention unless he expressly [agrees]. The widow, according to Tosafot here and in Bava Kamma, is in a different position from the normal buyer, who cannot cancel the transaction based on a future event. The widow can claim ada‘ata dehakhi lo due to the implicit consent of her deceased husband, which does not exist in regular commercial transactions. However, Tosafot on Ketubbot links this notion with the earlier discussion: mutual consent, although implicit, becomes a significant element in the case of the 381 382 Below I argue that the two explanations are not mutually exclusive, but rather reflect a single, combined approach which was followed by Tosafot on Ketubbot. Historically, the Tosafot may be based on different sources, i.e. different Tosafists, or on different compositions used by the redactor of our Tosafot (probably Rabbi Eliezer from Touques: see Urbach, Ba‘aley HaTosafot, 625-629; 639-645). However, the redactor harmonized those sources, and his view accordingly reflects the adoption of an integrated approach which is worthy of clarification. The method used here thus follows the general conceptual-dogmatic approach adopted in this chapter. ! In the English translation I use the feminine forms instead of masculine forms used by Tosafot. 96 Westreich: Talmud-Based Solutions to the Problem of the Agunah yevamah only since we assume that if she were asked at the time of betrothal, she would probably have agreed to the marriage even with the possibility of finding herself obliged to marry a leprous levir. Because of the assumption that she might have accepted that possibility, Tosafot argues, we need to base her claim for voiding the betrothal on mutual agreement. And in the case of the yevamah, contrary to normal commercial transactions, this mutual agreement implicitly exists. This analysis leads to the following conclusion: if we know that if the woman had known about the possibility of this future event, she would not have got married, her claim to void the marriage will be accepted even without her husband’s (implicit) agreement. This has an extremely significant conceptual implication: the question of whether she did or did not know about the possibility of this future event brings us back to the concept of kiddushei ta‘ut: it focuses on only one party to the transaction and on that party’s assumption at the moment of 383 the creation of the contract. But isn’t this a mistake with regard to a future event? Isn’t it the same as the argument that we saw in Maharit Al-Gazi and Me’il 384 Tsedakah? I would claim that the answer to these two questions is negative. Maharit Al-Gazi and Me’il Tsedakah deal with a future event, lack of knowledge about which renders the transaction mistaken. Tosafot on the other hand does not deal with that event on its own, but rather with the party’s perception of its possibility: would she accept the possibility that this event might happen? This difference is of the utmost importance. According to Tosafot, the discussion relates purely to the present: in the present we can have only a doubt that such an event may occur, or we may be aware that there is a statistical possibility that this kind of thing might happen. Both here and in Bava Kamma, the formulation of 385 Tosafot’s view found in Tosafot HaRosh makes this argument clearer: had he or she known about this possibility, s/he would not have refrained from the transaction. A case of ta‘ut exists then only when one party was unaware of this possibility, and when, had he or she known of it, s/he would have refrained from the transaction. The view of Maharit Al-Gazi and Me’il Tsedakah is problematic, for how can anyone claim a ta‘ut on the basis of an event that has not yet happened? But Tosafot’s view solves this problem: the mistake applies to the way each side views the present. In the present what exists is a doubt or a statistical possibility that an event might happen. If one of the parties was aware of this future possibility, the 383 384 385 For a different view see Me’il Tsedakah, 4, 6b-7a, who suggests, inter alia, that ta‘ut also requires the agreement of both parties. See supra, text at nn. 352-359.! Tosafot HaRosh on Ket. 47b, s.v. shelo. Chapter Seven: Mistaken Marriage, Conditional Marriage 97 transaction is not a ta‘ut, but if they were not, and we assume that knowing about this possibility would have prevented the transaction, the transaction is considered 386 mistaken. However, the present perception of the parties (at the time of making the contract) is not the only element to be considered. In the future, when the event actually does occur, the transaction may be invalidated. But then it cannot be done on the basis of the concept of ta‘ut, but only on the basis of an implicit condition. And this latter concept requires the consent of both parties to the transaction. To summarize and clarify the above discussion: according to Tosafot, the umdena of ada‘ata dehakhi lo kidshah nafshah combines the notions of implicit stipulation and erroneous transaction. When the wife claims ada‘ata dehakhi and wishes to void the marriage, we must ask two questions: (1) Is this a mistaken transaction? (2) If it is not a mistaken transaction, is there an implicit condition? In (1) we deal with the possibility at the time of making the contract (the betrothal): was the woman aware then of the chance (or risk) that such an occurrence might happen? If not, the transaction is void because it is mistaken. If she was aware of this option but nevertheless accepted the marriage, the transaction is valid. Yet in this latter case we must also ask question (2): Was there an implicit condition? The answer to this question depends on the kind of transaction. In a regular commercial transaction there is no implicit condition, since the seller would never agree to cancel the transaction in a case where, for example, his cow becomes nonkosher (terefah). But in a case of a betrothed widow when the levir is a leper, according to the possibility initially suggested in BT B.K., 110b, there was such an implicit agreement. In that case, therefore, the marriage can be voided in principle, since, in Tosafot HaRosh’s words, “It is considered as if she had made such a condition”. In practice, however, it is problematic to apply umdena in marriage, according to the Talmud’s conclusion: In that case we can all bear witness that [the wife] was prepared to accept any situation, as we learn from Resh Lakish. For Resh Lakish said: It is better to dwell as two than to dwell in widowhood (!"#$%&'()#"#&!*&+,&'()#"&',). However, the commentaries, both earlier and later, have noted numerous cases in which the notions of ta‘ut, implicit conditions and umdena were invoked, while Resh Lakish’s assumption was there considered irrelevant, as briefly discussed 387 above. 386 387 A similar construction was proposed by Daniel Friedman for the concept of ta‘ut in modern Israeli Law. According to Friedman, an argument of a mistaken transaction may be applied even for a future event, if the party (at the time of the transaction) was not aware of the risk he or she assumes. See Friedman, Contractual Risk, 467-471, 475-476. Supra, text at nn. 348-349. Westreich: Talmud-Based Solutions to the Problem of the Agunah 98 The next section examines one interesting responsum, which is remarkable for its practical use of the integrated approach of Tosafot. 7.6 Iggrot Moshe: an Application of the Integrated Approach One famous halakhic decisor who accepted the claim of kiddushei ta‘ut in actual 388 halakhic practice (halakhah lema‘aseh) is Rabbi Moshe Feinstein. This fact is 389 well known and has been discussed by a number of scholars. However, it is usually claimed that Rabbi Feinstein accepted only a limited version of kiddushei ta‘ut, which demands inter alia that the basis for the wife’s claim is a fact which 390 had been in existence at the time of the marriage, similar to our conceptual 391 analysis of kiddushei ta‘ut above. Nevertheless, the responsum discussed here reveals a more complicated approach, which is closer in many aspects to Tosafot’s 392 integrated approach. 393 The responsum deals with a communist levir, who refused to perform !alitsah for his sister-in-law. As background to a proper understanding of this responsum, we briefly summarize relevant halakhic discussion regarding an apostate levir. According to some Geonim, where the levir is an apostate there is no obligation 394 of yibbum. Traditionally, until Maharam of Rothenburg, this was explained as a result of the fact that the apostate, having converted out of the faith, is not 395 considered Jewish and is therefore not bound to his brother’s wife. Some Rishonim discussed the significance of the date of the brother’s 388 389 390 391 392 393 394 395 Iggrot Moshe, Even Ha‘Ezer, 1:79; 80 and elsewhere (see references in the following note). See Broyde, Marriage, 89-102; Bass, Meka! Ta‘ut, 208-213; Halperin-Kaddari, Tav Lemeitav, 1924; Hacohen, Oppressed, 57-60. Broyde, Marriage, 90; Broyde, Error, 51-52 n.34 (English section).! See text at nn.332-334. In correspondence with the Agunah Research Unit, Rabbi Prof. M. Broyde argues that Rabbi Feinstein uses umdena regarding a future event only in order to cancel a levirate bond, as in the responsum discussed below, but not in order to release a married wife without a get. In a private discussion, Rabbi Prof. David Bleich raised a similar argument, i.e. that this is a unique responsum of Rabbi Feinstein, in which he retroactively annulled marriage on the basis of a later event (becoming bound to an apostate levir). Although this might be true in practice, from a theoretical point of view there is no difference between marriage and levirate: in both cases the marriage is retroactively annulled. Indeed, the practical hesitation in applying umdena to a married woman is understandable, in the light of concerns about bastardy (mamzerut) and adultery (!umrat eshet ish) which would apply in such a case. ! Iggrot Moshe, Even Ha‘Ezer 4:121. Shut Maharam MeRothenburg, ed. Prague, 564. Ibid. Chapter Seven: Mistaken Marriage, Conditional Marriage 99 conversion according to this geonic view, whether before his brother’s marriage or after. This is related to the halakhic dispute as to when the legal relationship (zikah) between the brother-in-law and the wife is created: is it at the moment of the brother’s marriage or at his death (&!'!()* &!"#$!% vs. +'()* ,+!)). If the legal relationship is created at the moment of the brother’s marriage, the brother-in-law must already be converted at that moment for the widow to be exempt from yibbum. Otherwise his conversion would not affect the levirate bond – if he converted after the marriage but before the brother’s death, the obligation of yibbum would still exist. On the other hand, if the legal relationship (zikah) between the widow and her brother-in-law is created only at the moment of the brother’s death, then provided that he converted before this moment (even after 396 marriage) the widow would be exempt from yibbum. Rashi strongly objected to the geonic view and considered the apostate as Jewish, so that there is a levirate obligation in this case, even if the brother-in-law converted before his brother’s marriage. But 200 years later, Maharam of Rothenburg suggested an innovative explanation of the geonic view: the reason there is no levirate obligation is not the halakhic status of the apostate, but rather because the widow did not marry her husband on the assumption that she might find herself subject to a levirate bond with an apostate. To be sure, the Talmud rejects this claim in the case of the leprous levir, based on Resh Lakish’s presumption that a woman prefers to be married. However, an apostate husband, according to Maharam, is worse than a man afflicted with leprosy. Hence Resh Lakish’s presumption is not applicable when the levir is an apostate, and the 397 widow is exempt from levirate. This is a clear case of umdena – the yibbum occurs after the marriage, and only then does the wife claim that had she known that this would happen, she would not 398 have married her husband. Therefore, this is not a standard meka! ta‘ut, since the widow’s claim is related to the later obligation of yibbum (which did not exist at 399 time of marriage) and not to the current status of her brother-in-law. Defining this case as an umdena regarding a future event is not affected by the time of the apostasy. Even if the brother had already converted at the time of the marriage, the yibbum occurred only later, and the wife’s claim refers to this later event. Indeed, Maharam’s students dealt with the question of the time of the apostasy according to his reasoning: some accept the umdena even if the apostasy 396 397 398 399 Hagahot Mordekhai, Yevamot, 107, and see also below, explaining this dispute according to Maharam’s reasoning. See references supra, n.394. See Beit HaLevi, 3:3; Iggrot Moshe, Even Ha‘Ezer 1:79, §1. See Shut Maharam, supra n.394; Teshuvot Maimoniyyot, Nashim, 29. 100 Westreich: Talmud-Based Solutions to the Problem of the Agunah 400 took place after the apostate’s brother’s betrothal, while others dispute it, arguing that in order to apply the umdena the brother-in-law must be an apostate at the time 401 of his brother’s betrothal. However, this dispute, as correctly explained by Rabbi Joel Sirkes in his commentary to the Tur (Bach), focuses on the strength of the 402 umdena. For if the brother became an apostate only after the marriage, the wife’s claim runs as follows: had she known that this brother would convert and that she might become subject to levirate, she would never have married her husband. This is quite a weak argument, since such occurrences are very rare, unlike the case where the brother was already an apostate at the time of the marriage and the wife claimed that had she known that she might have to marry him in the framework of levirate, she would never have married her husband. However, even in the former 403 case some sources argue that Maharam’s umdena is valid, as mentioned above. Neither Maharam nor later halakhic authorities accepted the claim of ada‘ata dehakhi lo kidshah nafshah in practice, and this approach was rejected almost 404 totally in normative halakhah. Furthermore, according to Hagahot Mordekhai, based on Tosafot, Maharam’s ruling applies only to a betrothed woman and not to 405 a married one. However, in some cases this ruling was applied in actual practice, even to a married woman. In his responsum, Rabbi Feinstein distinguishes the case he discussed from that of the Maharam: This case is different from usual acts of marriage in general because she was married to him after it was already known to him and to her and to the witnesses and to everyone that he had to go to the army, where there was a great likelihood that he would die. Since he had this brother it was obvious that she would be bound to this levir if her husband died in battle, since he was a member of the [Communist] party and denied all matters of Torah and would not release her through !alitsah ... Besides, from the manner in which he responded to her it is clear that he is a man of bad character and an apostate out of spite, and she surely knew him [as such]... This case, he argues, is extraordinary since the danger of the husband’s death was very real and well known. Accordingly, Rabbi Feinstein argues that Resh Lakish’s 400 401 402 403 404 405 See Hagahot Mordekhai, Yevamot, 107, and Shut Maharam, 1022. Teshuvot Maimoniyyot, Nashim, 29. See Bach, Even Ha‘Ezer, 157, s.v. !"#"$%&"%&'(. Cf. Hacohen, Oppressed, 39-41; Broyde, An Unsuccessful Defense, 8-9; Westreich, Book Review. See Shul!an Arukh, Even Ha‘Ezer 157 (quoted above, text at note 367). Even Maharam himself did not rely on this approach in actual practice: see Terumat HaDeshen, 223; Mordekhai, Yevamot, 29. However, this was the basis for the explicit condition in the case of a converted levir, which was partly accepted; see supra, text at nn. 366-373. See Hagahot Mordekhai, Yevamot, 107, based on Tosafot on B.K. 110b, s.v. !')*!*. Chapter Seven: Mistaken Marriage, Conditional Marriage 101 principle that “a woman prefers to be married” is irrelevant in this case, and thus the wife may claim ada‘ata dehakhi lo kidshah nafshah. Rabbi Feinstein explains: It is clear to everyone that no woman would agree to get married for the sake of so short a period – days or even months – even though [as a rule] “it is better to live as two people” (tav lemeitav tan du). Hence, it is legitimate to claim ada‘ata dehakhi lo kidshah nafshah here, and this claim cannot be rejected by Resh Lakish’s presumption. At this point Rabbi Feinstein discusses two issues. The first pertains to the wife and to her assumptions regarding her marriage: And therefore, since in [our] case [she knew] her marriage was [merely] for days, because she knew that there was a great likelihood that he would be taken to the army and [that] he would die in battle, we are forced to say that she did not know that she would be bound to this levir for levirate or !alitsah. Since, however, it is far-fetched to say that she did not know that there is such a thing as a bond of yibbum and !alitsah, which is a matter well-known even to women and ignoramuses, [we must say that she assumed that there would not be a levirate bond] because she thought that [the levir] was not considered a Jew, since he had become a member of the [Communist] party, for he and the [Jewish] community regard him as separated from the general communal body. Therefore she thought that such a brother does not generate a levirate obligation. Had she known that he does generate a levirate obligation, she would not have got married even to the kosher brother for a [mere] few days – even twice twenty days – and certainly [for] more. The second issue pertains to both parties: And it is logical [to say] that also on the husband’s side this umdena applies, because he also certainly knew that for the sake of a few days [of marriage] he would not find any woman in the world who would marry [him], unless it were on condition that if he should die childless she would not be married to him. There is a clear assumption that [the validity of the marriage was based] specifically on a condition [agreed to] by both sides. This last argument views umdena as an implicit condition, namely, that if the husband dies without children, the marriage is retroactively invalidated. So why do we need the previous argument, which claims that the wife did not know that she would be obliged to undergo levirate with her brother-in-law? Isn’t it sufficient to say that although she knew about these obligations, she made an implicit condition that the marriage would be invalidated if she became subject to a levirate bond? The condition is indeed an acceptable condition since the husband agrees, as indicated by the second argument! Rabbi Feinstein does not base his halakhic decision on a happenstance group of unrelated considerations which could support his decision regarding this difficult case. I suggest that these two arguments work in tandem. This double 102 Westreich: Talmud-Based Solutions to the Problem of the Agunah argumentation confronts both aspects of Tosafot’s umdena, that of a mistaken transaction and that of a terminative condition. In the first argument, Rabbi Feinstein deals with ta‘ut, addressing the position of the wife only, whose claim that there was a mistake stems from the fact that she was unaware of the possibility of yibbum. But here R. Feinstein takes a very innovative approach: whereas Tosafot deal with unawareness of the possibility of a factual situation (in a normal case of yibbum, for example, unawareness of the possibility that the woman might be left a childless widow and be subject to levirate, a claim that would not be accepted), Rabbi Feinstein extends this to a 406 mistake concerning the law. According to Rabbi Feinstein, in this unique case, the wife was unaware of the possibility that she might be subject to levirate, since she thought that such a brother would not be halakhically regarded as a levir. This is quite a sophisticated argument: the woman’s mistake is not simple unawareness of a specific halakhah, since, as Rabbi Feinstein claims, it is difficult to assume that she didn’t know the basic law of yibbum. Rather, her mistake lies in adopting the geonic view that the converted brother is considered a non-Jew, thereby freeing the wife from the levirate obligation. But had she known at the time of the marriage that the apostate brother was subject to the obligation of levirate, she would never have agreed to the marriage. Therefore the marriage is based on a mistake, and is void ab initio. It should be emphasized, according to our previous 407 discussion, that this is not a mistake with regard to a future event, but rather a mistake with regard to the present. Rabbi Feinstein finds the ta‘ut to be a mistake regarding the woman’s current knowledge of the law – a very innovative analysis. In the second argument, Rabbi Feinstein deals with an implicit condition. When the husband died the widow became subject to levirate marriage. This event had not occurred at the time of marriage (only its statistical possibility), so the marriage cannot be regarded as a mistaken transaction. But was there an implicit condition which invalidates the marriage when such an event occurs? Maharam suggested such an argument regarding an apostate brother, but his view was rejected by 408 normative halakhah. However, Rabbi Feinstein distinguishes his case from those of Maharam and Tosafot, and argues that since the marriage was only for a short 406 407 408 The use of mistakes concerning the law was introduced to support the practice of R. Rackman’s bet din in cancelling marriage on grounds of kiddushei ta‘ut, as explained by Susan Aranoff: “Kiddushei ta‘ut III emphasizes that a woman would not knowingly consent to a domestic partnership based on gufah qanui...” (Aranoff, Response, 2). Rabbi Feinstein’s argument is quite different, since he applies it in a specific case, where the mistake pertained to a specific detail of the halakhah, whereas a general mistake which broadly applies cannot be accepted as grounds for annulling a marriage. See also Rabbi Bleich’s critique of Aranoff’s argument in Bleich, Kiddushei Ta‘ut, 108-116. Supra, text at nn.383-386.! See supra, n.404. Chapter Seven: Mistaken Marriage, Conditional Marriage 103 period, we assume that both the husband and the wife agreed to cancel the marriage in these circumstances. The wife thus can claim that ada‘ata dehakhi lo kidshah nafshah, and the marriage is retroactively invalidated. Rabbi Feinstein hence uses double argumentation in order to support his claim to annul the marriage, based on the two aspects of umdena. But his two arguments appear at first sight to contradict each other: the first is based on a mistake about the obligation of levirate, while the second is based on the fact that an implicit terminative condition (if the woman needed levirate, the marriage would be cancelled) may be upheld in practice. But how can we assume that the couple made a condition to invalidate the marriage in a case of yibbum (the second argument), if they thought that the obligation of yibbum was halakhically irrelevant (the first argument)? The last paragraph of Rabbi Feinstein’s responsum answers this question: It is obvious that even though she married him without explicit specification (bistam), and no condition was made, because they did not know that [in this case] there would be a levirate bond, and not because they knew the rules of conditions [i.e., that the umdena would render the marriage conditional], even so the marriage is annulled just as if they did know of the laws of conditions. [This is because] we do not need them to be knowledgeable of the law, but [only that] they had no desire to get married on condition that she would need !alitsah from this man if her husband died [childless]. If [the situation] is such that we can clearly assume this, it is considered as if they made a condition and the [marriage] is annulled. This is because in the cases of assumptions (umdena) mentioned in the gemara nowhere is a distinction made between ordinary people [who do not know the details of the law] and scholars. [emphasis added] Rabbi Feinstein’s conclusion takes the definition of an implicit condition a step forward: not only can this be a condition which was not made explicitly by the two spouses (but one which, we may assume, they would have adopted had they been asked), but this is a condition the need for which was rejected by the couple (since the wife thought that she would not be bound to this brother-in-law for levirate marriage or !alitsah). This is thus an extremely complex situation: there is an implicit condition that the marriage would be cancelled in a case of a bond to an apostate levir, but making this condition explicit would not be possible, since it would conflict with the first argument. For, as stated in the first argument, the wife was unaware of the levirate bond, and thus could not know that there is anything to make a condition about. This is accordingly a condition implied by the law: it is sufficient, according to Rabbi Feinstein, that the couple did not want the result (being bound to the apostate levir), while the legal construction of the condition and its imputation to the couple (unawareness of the obligation on the one hand; awareness with an implicit condition to cancel this obligation on the other) is the work of the posek. 104 Westreich: Talmud-Based Solutions to the Problem of the Agunah 7.7 Conclusion Three concepts have been discussed above: (a) “terminative conditions”, i.e. cases where an event which occurs during married life renders a marriage retroactively void, based on an explicit stipulation of the spouses at the time of marriage; (b) “mistaken transaction”, i.e. cases in which a fact obtained at the time of marriage, but one spouse was unaware of it, and if he or she had been aware of it, they would not have married. In this case, the awareness of that fact at a later time reveals the actual status of the marriage: the marriage was based on a mistake, and therefore was never valid; (c) umdena (an assessment – in our context, of one’s intentions) that “on this assumption she did not get married”. This possibility occupies a middle ground between the previous two: it is based on a fact which we assume could lead one of the spouses to cancel the marriage. But this fact did not exist at the time of the marriage, so no mistake actually occurred at that time. Consequently, the commentators do not agree about how to define this case: as an implicit condition, as a kind of mistaken transaction, or as a combination of both: (a) a mistake regarding the “facts” that obtained at the time of the marriage, i.e., a mistake regarding the possibility of a particular future occurrence; and (b) an implicit condition in regard to the later actual occurrence of that fact. Beyond the conceptual discussion, these three concepts share one common function: nullifying the marriage. In some cases – and get refusal is a typical one – halakhic authorities seek a solution which will void the marriage without a get or !alitsah. The formal halakhic approach which is used varies from case to case. Sometimes two contradictory arguments can be used even in the same case (as by Rabbi Feinstein above), but the goal is identical. But discovering the three concepts and finding their legitimation in the talmudic literature and the poskim is not the end of the quest for a solution. Although there is a halakhic basis for them, the move from theory to practice is not always an easy one. Kiddushei ta‘ut and umdena were accepted in practice by some poskim, but were rejected in practice by others, and this seems to be a common view in the 409 rabbinical courts, at least in Israel. With regard to conditional marriage, the rejection is almost total. It is not used today at all (maybe only in private, rare cases), although it is sometimes discussed theoretically. Nevertheless, as I have shown, in its implicit appearance – as the basis for umdena according to some views – we do find a use of conditional marriage, both in theory and in practice. The three concepts here discussed reflect (in different measures) a declarative annulment of the marriage. The court here reveals (sometimes indeed fictionally) the intention of the parties and on this basis defines the marriage as annulled. The 409 Cf. Bass and Henkin, supra, n.333. Chapter Seven: Mistaken Marriage, Conditional Marriage 105 same outcome, i.e. annulling the marriage, can be achieved in a different way: by a constitutive annulment of the court, based on the talmudic maxim that “everyone who betroths does so subject to the consent of the Rabbis, and the Rabbis annul his betrothal” (!"#"$%&"'()"*+%&#,-%(!#".*/0(%')*$%&#,-)%01.)0%')*$)%+2). An explicit, constitutive, hafka‘at kiddushin is the ultimate means of voiding marriage without a get in exceptional cases. Yet, as discussed in previous chapters, the concept of hafka‘at kiddushin is much more radical from a halakhic point of view, and its application, both in theory and practice, is largely disputed since the Rishonim to 410 our days. The previous discussion therefore opens another route for annulling marriage, not in every case, but when the marriage developed into an unsustainable situation, in which the wife might claim that on this assumption she did not get married (ada‘ata dehakhi lo kidshah nafshah). 410 See supra, text at nn.317-323. Chapter Eight From Theory to Practice: Will and Ability 8.1 The Pluralistic Nature of the (Normal) Halakhic Discourse Halakhic discourse is characterized by an internal pluralism: in spite of the numerous controversies, this legal system recognizes the significance and value of opposing views. The basis for this approach can already be found in the talmudic literature, which affords theoretical, normative, and practical legitimization to a diverse spectrum of positions. On the theoretical (or, may we say, theological) level, the pluralistic view is anchored in such statements as “these and those are the words of the Living 411 God”, which imply that there may be a number of equally legitimate interpretive 412 positions. This view is linked to, and even based on, a pluralistic hermeneutical attitude, which posits multiple possible interpretations of the texts of the Torah; the 413 classical formulation is that there are “seventy faces to the Torah”. The pluralistic approach is also grounded on an understanding of the hermeneutical process as a dynamic and creative one, which ascribes an essential role to the Sages in the process of interpreting the Torah, which was given “as wheat, to bring forth from it 414 fine sifted flour”. In legal practice, the Sages afford validity to minority and rejected views in certain circumstances, at least after-the-fact (bedi‘avad) or in emergency situations 415 (bishe‘at hade!ak). The coexistence of different halakhic approaches was legitimized by the Sages in their practice of recognizing the particular halakhic 411 412 413 414 415 BT Eruvin, 13b. This position is not the only view. We see in the talmudic literature the tension between the view expressed in “these and those are the words of the Living God” and those emphasizing the importance of halakhic traditions, with its consequent difficulty in recognizing opposing opinions. Nonetheless, the dominant trend is to acknowledge the authority of the expositions of the Sages. See BT Bava Metsia, 59b (“it [the Torah] is not in Heaven”; !"#$ %"&'($ !)), a statement which may legitimate a range of views (cf. Statman, Autonomy, 639-662). See also Sagi, The Open Canon, 1-7. This conclusion is supported by the two additional aspects, normative and practical, described infra, text at nn.415-417. BT Shabbat, 88b. On the term “seventy faces to the Torah”, see Mack, Shiv‘im Panim, 449. Seder Eliyahu Zutta, 2 (ed. Ish-Shalom) 172. See, e.g., BT Berakhot, 8b-9a (“R. Shimon is worthy [i.e., sufficiently authoritative] to rely upon in a case of emergency”). Chapter Eight: From Theory to Practice: Will and Ability 107 authority of the local community. Sometimes local customs that went against the mainstream view were legitimated and even encouraged by the Sages, as local 416 communities followed their leader, as shown by this passage: The Sages taught: In R. Eliezer’s locality they would cut timber to make charcoal for making iron on the Sabbath. In the locality of R. Yoseh the Galilean they would eat the flesh of fowl with milk [both of which are prohibited by normative Jewish law] [...] R. Isaac said: There was one town in the Land of Israel where they followed R. Eliezer, and they died at the proper time. Moreover, the wicked authorities once issued a decree against Israel concerning circumcision, but did not decree this against that town. From a practical point of view, the Sages were capable of bridging the immense legal gulfs between the opposing views. This commitment to pluralism is sharply reflected in the mishnaic statement that despite deep controversies in matters of marriage and divorce, “[the men of] the House of Shammai did not abstain from marrying women of the House of Hillel, nor did [the men of] the House of Hillel 417 abstain from marrying women from the House of Shammai”. These talmudic foundations find practical expression in post-talmudic Jewish 418 law. In the consciousness of the rabbis, the statement that “these and those are the words of the Living God” has become a guiding principle which reflects the essence of the halakhah and the nature of halakhic deliberations. Many rabbis have even invoked extreme pluralistic positions, holding that all positions expressed in 419 halakhic discourse are equal, reflecting halakhic truth (or halakhic truths). Conceptually rather than practically, however, others take a monistic position that does not recognize a multiplicity of halakhic truths, but strives for one halakhic 416 417 418 419 BT Shabbat, 130a.! M Yevamot 1:4. It is noteworthy that the talmudic interpretation of the mishnaic statement limited its pluralistic meaning. According to the Talmud (BT Yevamot, 13b-16a), the Houses did not abstain from marrying women from the other’s House in general, but they did abstain from marrying each other’s women if they were the subject of a dispute between the Houses, and they notified each other about such women. The Houses, according to the Talmud, respected the other’s view, which they regarded as legitimate – as may be expected following a pluralistic approach – although they did not adopt it for themselves. In fact, there might be a difference on this issue between the Babylonian Talmud and the Palestinian Talmud, where the Babylonian Talmud reflects a more pluralistic attitude, although “The Bavli, like the Yerushalmi, is not willing to entertain the possibility that they would put tolerance above risking mamzerut”: see Hidary, Legal Pluralism, 213, and cf. Levine, Review. See Ben Menahem et al., Hama!loket, Vol. 2, 855-964. Sagi, The Open Canon, 67-107. Alongside recognition of the authority of diverse positions, it is necessary for practical reasons to decide that a particular position is the law, a decision left to the rabbis (see ibid. at 71-87). The significance of internal pluralism from the point of view of halakhic decision-making is that it justifies recognition of the possibility of change in the law from accepted practice (ibid., at 93-97; M Eduyot 1:4-6). 108 Westreich: Talmud-Based Solutions to the Problem of the Agunah 420 verity. Yet even for those adhering to the monistic view, halakhic truth is unknown and can only be revealed by means of halakhic discourse, using a framework of equality of interpretation. Thus, even according to the monistic view, halakhic discourse is necessarily pluralistic, though its objective – admittedly an 421 ideal – is monistic. The pluralistic nature of Jewish law in practice is well-reflected in the extremely influential Jewish law codex compiled in the past half-millennium, the sixteenthcentury Shul!an Arukh by R. Yosef Karo, together with the glosses by R. Moses 422 Isserles (Rema). The code does not produce one single ruling on all specific issues; rather, it is often possible to derive differing legal rulings from this code, resting on differing views in Karo’s core rulings and Isserles’ dissenting addenda. The codex and its interpretations suggest that all are valid and legitimate, reflecting 423 the nature of the dispute as “the words of the Living God”. 8.2 Legitimizing Solutions to the Problem of ‘Iggun Creative solutions for new problems are an integral part of the world of Jewish law. Such solutions are based upon an interpretation of halakhic sources viewed in the light of the contemporary situation. As an example of how the problem of the agunah was dealt with in a different historical period, fifteenth-century Ashkenazic communities began to add a stipulation to the betrothal ceremony that would prevent the bride from becoming an agunah, should she become subject to the levirate bond but the brother-in-law is unable or unwilling to release her (by performing !alitsah). In this case, the rabbis of Ashkenaz interpreted the relevant talmudic sources as permitting a stipulation that – under certain circumstances – 424 retroactively annuls the marriage. Halakhic controversies, at times consequential, intense and acrimonious, are also quite routine, and this is the nature of halakhah since its inception. It is also natural that novel ideas deriving from halakhic sources will encounter some type of opposition. Indeed, this Ashkenazic stipulation permitting retroactive annulment 420 421 422 423 424 Sagi, ibid., 11-65. This is “weak pluralism” as defined by Sagi, ibid., at 185-186. See!Elon, Jewish Law, ch.36. The pluralistic view of the halakhah, here justified from a theological viewpoint, requires a hermeneutical justification as well. See Westreich, Gatekeepers, section II, text at notes 58-65. Shut Terumat HaDeshen, 223; Rema, Even Ha‘Ezer 157; idem., Darkhei Moshe, Even Ha‘Ezer, 157:5. On this issue see supra, text at notes 366-373. Chapter Eight: From Theory to Practice: Will and Ability 109 425 was greeted with criticism (for example, by R. Yosef Karo). Nonetheless, for many years, numerous Jewish communities adopted it in practice, although its use 426 has nearly faded into history. Similarly, throughout the history of Jewish law, the rabbis have disputed the nature of marriage annulment and their authority to employ it in practice. Most of these deliberations up to the present day focus upon cases of marriage improperly performed or not in accordance with community regulations (triggering what we termed “immediate annulment”). In spite of the profound halakhic debates on the propriety of these annulments, Jewish communities have still passed marriage annulment regulations relating to such cases, especially in the modern age with the loss of Jewish communal and judicial autonomy and limitations placed upon the 427 ability of the rabbis to act by alternative methods. Today’s agunah problem is no less severe and no less may be regarded as an emergency situation, or at least as a one of she‘at hade!ak. These concepts have halakhic implications, mainly by legitimizing lenient views which in our case may support the use of solutions to the problem of the agunot, despite the usual tendency towards stringency in matters of marriage and divorce, as discussed at 428 length in the Agunah Research Unit’s analysis of the agunah problem. The above study of the pluralistic nature of Jewish Law strengthens this conclusion. However, it is not unusual to encounter views that object to particular solutions (and sometimes to any solution) to the agunah problem. Interestingly, the opponents may sometimes accept these very solutions in practice, but not for the agunot themselves. For example, some critics of the marriage annulment proposal for agunot accept it, at least on a post factum level, to prevent a person’s 429 stigmatization as a mamzer. This solution was first suggested (in theory, rather than for practical use) by Rabbi Shalom Mordekhai ben Moses Shwadron 430 (Maharsham) at the end of the 19th century (known as Get Maharsham). Yet the 425 426 427 428 429 430 See R. Yosef Karo, Bet Yosef, Even Ha‘Ezer 157, s.v. !"#$%&'()&*%*(&+; supra, note 367.% Freimann, Seder Kiddushin, 386-388. Abraham H. Freimann assembled most of the material in his book, Seder Kiddushin. The modern period is discussed at 310-345; as Freimann states (at 345): “what internal pressure did not accomplish, external pressure did [...] [W]ithin the period of about a hundred years (1804-1901) seven regulations annulling illegally performed marriages (in accordance with state law) were passed and practiced in various countries, including Italy and France, Algeria and Egypt”. See mainly Jackson, Agunah, 44-83. Rabbi David Malka, a severe critic of the marriage annulment proposal on behalf of agunot, admits that in extreme cases, to prevent mamzerut, rabbinical courts use marriage annulment. See Malka, Ein Hafka‘at Kiddushin, end of second section; Westreich, Gatekeepers, Section III. Preventing mamzerut using marriage annulment was mentioned quite a few times by Rishonim and A!aronim (see, e.g., Tosafot, Gittin, 33a, s.v. ($,-./01() in the context of the talmudic messenger 110 Westreich: Talmud-Based Solutions to the Problem of the Agunah 431 exegetical basis for the use of this solution for mamzerut is highly untenable. In fact, the mamzerut solution is an extremely novel view and from some aspects it is easier to apply marriage annulment to solve the problem of an agunah than to 432 prevent mamzerut. Truly, Jewish law considers releasing a married woman from her marital bond to be a very serious matter, which, if not performed according to the halakhah, may lead to promiscuity and mamzerut. Therefore, the halakhic authorities are discouraged from following any innovative routes for terminating marriage other than by a writ of divorce, while employing these same avenues to prevent mamzerut is legitimate and encouraged. Yet, the legal construction is quite similar, and, as mentioned, from a purely exegetical point of view, is even easier to apply it to agunot. Nevertheless, the option of marriage annulment for agunot has been forcefully rejected by the opponents, although some acceptance, at least for emergency cases or on a post factum level (i.e. stating that if an annulment has been granted it would be recognized “after the fact”, even though they might not have granted one had the case been presented to them earlier), could have been expected. What, then, might explain this hostile reaction? Elsewhere I have discussed the debate over marriage annulment, identifying three different levels which characterize it: hermeneutical, political and 433 sociological. Here, however, I limit myself to discussion of one central aspect, which may clarify the wider objections of any of the above solutions to the problem of agunot. This aspect is the ideological aspect, i.e. whether Jewish Law recognizes that women have a right to divorce (and, if so, in what circumstances). For if the answer to that question is negative, we do not have a problem in the first place. ---case (supra, text at nn.173-175, fifth Talmudic case). Following theses sources Maharsham proposed a theoretical solution by deliberately simulating such circumstances: see Shut Maharsham, Part 1, 9, and discussion in Jackson, Agunah, 227-230 (§§5.23-24). 431 432 433 See, e.g., Auerbach, Afke‘inhu; Sho!etman, Hafka‘at Kiddushin. The issue is quite complex, and requires an intensive discussion of the Talmudic sources. In brief, marriage annulment to prevent mamzerut is based on the view of Rabban Shimon ben Gamliel (fifth Talmudic case, supra, text at nn.173-175), but according to important halakhic decisors (presumably including Maimonides), the law was determined according to his disputant, Rabbi Judah haNasi. The supporters of marriage annulment to prevent mamzerut use innovative interpretations in order to ascribe the concept of annulment to R. Judah haNasi as well. On the other hand, marriage annulment on behalf of agunot is proposed to be based on an explicit enactment, one not necessarily linked to the rejected view of Rabban Shimon ben Gamliel. Westreich, Annulment, section IV. Chapter Nine Who is an Agunah? 9.1 Introduction According to a statement released by the Rabbinical Courts’ Administration 434 (26.6.2007), the number chained wives was 180 and of chained husbands 190. On the other hand, according to research conducted by Israeli women’s organizations and presented to a special meeting of the Israeli Parliament Commission of the Advancement of the Status of Women, there are around 435 100,000 chained wives (23.3.2005). Surely, the gap between these two countings is not merely statistical. Rather, it reflects a deep conflict over the definitions of get refusal and aginut: a wide definition according to women’s organizations on the one hand; a narrow definition according to the Israeli rabbinical courts on the other. This reflects an ideological conflict on the right to divorce, which directly influences the way each side views the agunah problem. Bridging the gap would not only accord recognition to many chained spouses (mainly chained wives), but would also legitimate the use of various solutions, as discussed in the previous chapter. In the current chapter, I discuss the deep, often emotional, debate surrounding the right to divorce, which, in other words, is the disagreement between those who favour a (limited version of) no-fault divorce in cases of irretrievable breakdown of marriage, and those who regard it as invalid under Jewish law. Some claim that this is a debate between a traditional Jewish approach on the one hand and a modern liberal secular approach on the other. But this would be wrong: no fault divorce has a stable basis in classical Jewish law sources, as is recognised by some contemporary dayanim. Nevertheless, in current rabbinical court decisions it has become the subject of keen debate: there is also a significant school of rabbinical judges which rejects any such right to divorce, and thus opposes compulsion of divorce (or other solutions) in such cases. In what follows I analyze the various legal and hermeneutic methods used in order to establish the view that no-fault divorce has no basis in or validity under Jewish law. There are several possible explanations of the roots of this debate. An important 434 435 See http://www.rbc.gov.il/statistics/2007/1.doc. See http://www.knesset.gov.il/protocols/data/html/maamad/2005-03-23.html. 112 Westreich: Talmud-Based Solutions to the Problem of the Agunah one is political. Israeli family law is based on a unique system, which gives jurisdiction either to rabbinical or civil courts, and sometimes to both. As a result, the system is characterized by a contest (or race) for jurisdiction: an implicit or explicit competition between the two institutions to expand their authority in matters of family law (even though jurisdiction in divorce itself is exclusive to the 436 rabbinical courts). However, I argue that any such political explanation is not 437 sufficient, and therefore suggest a supplementary ideological reason. I argue that some decisions of the school that opposes irretrievable breakdown of marriage as a ground for divorce are based on a deep ideological rejection of this approach. This rejection is absolute. It rests on the assertion that divorce for “irretrievable breakdown” belongs to “the laws of the nations”; that is, that it arises from nonJewish sources and lacks roots in Jewish law. This argument is incorrect. But it reveals the ideological nature of the controversy regarding the right to divorce. 9.2 No-Fault Divorce? A prominent halakhic authority, Rabbenu Yero!am (Provence-Spain, fourteenth 438 century), wrote: My teacher Rabbi Abraham ben Ishmael wrote that […] a woman who says: “I do not desire him, [I demand that] he will give me a writ of divorce and ketubbah”, and he says: “I also do not want you, but I do not want to give you a writ of divorce”, it seems that we do not rule her to be a rebellious wife [who is subject to social and financial sanctions], so that she would lose her basic ketubbah payment and dowry. Rather, we have her wait twelve months for her divorce, [in case] perhaps they will be reconciled; and after a year we coerce him to divorce her and she loses the addition [that is, extra obligations of the husband in the ketubbah] and all that he gave her from his own [property], since he did not give it to her on the understanding that she takes it and divorces [lit., goes forth]. According to Rabbenu Yero!am, if both spouses desire to divorce, but the husband refuses to give a writ of divorce (possibly because he wishes to impose some 439 financial conditions), after 12 months (during which time they might reconcile) he will be coerced to divorce his wife. This view is of extreme importance: even 436 437 438 439 See Rosen-Zvi, Forum Shopping, 347-396; Westreich, The Right to Divorce, 184-187. The conflict, however, is not necessary: see Daichovsky, The Rabbinical Courts; Shifman, Safa Datit. Pin!as Shifman has indicated the importance of considering both of these two aspects – the political and the ideological – as each plays a significant role in matters of family law. See Shifman, HaHalakhah, 27-46, and see also Shifman, Safa Datit, 423-425. Yero!am ben Meshulam, Mesharim, Netiv 23:8 (Venice, 1553), vol. 3, 60b–61. Unfortunately, this is common today: see, e.g., Rabbinical Court (Ariel), file no. 057140493-21-1 (12.3.2006) (HaDin vehaDayan 18 [2008], 6–7). Chapter Nine: Who is an Agunah? 113 were there no specific cause for divorce, the husband would under certain circumstances be coerced to divorce his wife. This ruling is hereafter referred to as the “Yero!am Ruling”. Rabbenu Yero!am does not speak explicitly about the state of the marriage. We may, however, reasonably understand his ruling as based on marital breakdown. We might therefore conclude that, according to Rabbenu Yero!am, in a case of an irretrievable breakdown of marriage, which is reflected in the failure to reconcile over the course of 12 months, a divorce shall be executed. A more explicit source that takes a similar view is a ruling by R. !ayyim Palaggi (Izmir, nineteenth century; sometimes rendered “Palache”), who rules as 440 follows: Whenever it seems to the rabbinical court that they are separated for a long time and cannot reconcile, contrary [despite efforts to make peace between them (discussed earlier in the responsum)]: great efforts must be taken to separate them one from the other and to issue a writ of divorce, so that they would not commit many sins, both the husband and the wife. And I give the measure of time for this: if a dispute were to arise between husband and wife, they failed to make peace between themselves, and there is no hope for them, they are to wait a period of 18 months. And if […] it appears to the court that there is no hope of making peace between them, [the court] is to separate the couple, and coerce them to grant a writ of divorce until they would say “I am willing” [i.e., to give the writ of 441 divorce]. Accordingly, after 18 months of a court-imposed reconciliation period following the breakdown, the court is obliged to coerce the couple to divorce. Needless to say, in such a case the court has the authority to impose sanctions in order to compel the recalcitrant spouse to consent to the divorce (as Palaggi explicitly writes: “and coerce them to grant a writ of divorce”). This ruling has been widely discussed, and is hereafter referred to as the 442 “Palaggi Ruling”. It has been endorsed by several halakhic decisors, and indeed 443 rabbinical courts in Israel often adopt similar views. 440 441 442 443 !ayyim Palaggi, !ayyim veShalom 2, Even ha‘Ezer 112 (Izmir, 5632 [1872]), folio 148b. Coercion is effected by imposing sanctions upon the recalcitrant spouse, until she or he says “I am willing [to be given/give a writ of divorce].” See Maimonides, Mishneh Torah, Gerushin 2:20. The influential and leading twentieth-century American posek, R. Moshe Feinstein, seems to accept this view: see Iggrot Moshe, Vol. 8, Yoreh De‘ah 4:15. For a discussion and further references to modern Jewish law decisors: see Broyde, Marriage, 23, 25-28. Rabbinical courts take this approach not only when the wife refuses to divorce but also when it is the husband who is recalcitrant, despite the possible fear of get me‘useh in an unlawful compulsion. See, e.g., High Rabbinical Court, file no. 028127447-21-1 (1.4.2008) (HaDin vehaDayan 24 [2010], 6); 114 Westreich: Talmud-Based Solutions to the Problem of the Agunah The Yero!am Ruling, as well as the Palaggi Ruling, are far removed from the severely restrictive doctrinal pattern as regards the wife’s right to divorce under the main sources of talmudic law. However, talmudic law itself – in the case of the rebellious wife (moredet) – approves unilateral divorce, as we have seen in 444 previous chapters. Throughout the history of Jewish law several traditions have expanded the legitimation of unilateral no-fault divorce, e.g. the Genizah tradition 445 and the geonic tradition, and both Yero!am and Palaggi may be regarded as branches of those views. The Yero!am and Palaggi rulings are however important (compared to the earlier traditions), since in some sense they have been accepted as authoritative for modern halakhic discourse, either as requiring full acceptance or at least as challenging us to engage in sophisticated hermeneutical tactics, as discussed below. The rationale behind these views is that the formal status of marriage has an instrumental object: to be the necessary condition for a stable marital relationship. When such a relationship does not exist, there is no justification for forcing the spouses to be formally bound one to the other. R. Shlomo Daichovsky, a former 446 High Rabbinical Court judge, describes this rationale thus: 447 The wife is not entitled to continue residing in the marital home […] in a case of death of marriage, no matter whose fault it is. We do not deal with “resurrection of the dead”, and there is no reason to perform “artificial respiration” on dead marriages. ---Rabbinical Court (Tel Aviv), file no. 8564-21-1 (22.10.2003); Rabbinical Court (Tel Aviv), file no. 307426676-21-1 (21.12.2005) (HaDin vehaDayan 18 [2008], 3). 444 445 446 447 See supra, Chapter One. Regarding the relations between the sources see ibid., text at nn.93-96. Indeed, some additional support for the Yero!am and Palaggi rulings may be found in the law of the rebellious wife. Although the view that approved coerced divorce in the case of a rebellious wife was rejected by later Jewish law decisors following Rabbenu Tam’s criticism (see ibid.), some rabbinical courts rule that a husband is obliged to divorce his rebellious wife, especially when there is an apparent justification for the wife’s claim that her husband is repulsive to her (see, e.g., High Rabbinical Court, file no. 011926961-21-1 [23.10.2007] [HaDin vehaDayan 18 [2008], 3). This ruling implies that while the husband may not be directly coerced, indirect measures may be applied, such as social sanctions, the denial of a driver’s license, or a ban on leaving the country. Israeli Rabbinical courts derive their authority to apply these and other measures from the Rabbinical Courts Law (Enforcement of Divorce Decrees) 5755-1995. Supra, Chapter Three. High Rabbinical Court, file no. 21-02887447 Niago v. Niago (24.10.02). The verdict was not published, but it was partly cited in Daichovsky, Batei Hadin, 283 n.463. This statement was cited a few times by civil judges, who expressed their agreement; see, e.g., Judge Yehudah Granit in Family Court (Tel Aviv), file no. 94740/00 K. Sh. v. K. S. (3.3.2003). According to Jewish law, the wife is entitled during the marriage to a pleasant home, which usually means that she cannot be forced to leave the marital home, and that, in practice, the couple’s home cannot be sold without her consent. Rabbi Shlomo Daichovsky argues that in a case of “death of marriage” the wife would not be entitled to continued residence in the marital home, which could be sold without her consent (and the proceeds distributed between the spouses), even though the formal Chapter Nine: Who is an Agunah? 115 Western legal systems have broadly adopted a liberal conception of marriage and 448 have provided for “no-fault” divorce. This approach is applied in several systems in an extreme way, according to which any demand for divorce is accepted, without taking into consideration broader elements or other contextual considerations. There is also, however, significant support for a more moderate conception, which involves social and communal considerations, in addition to the 449 liberal basic right of freedom. This view is consistent with a “death of marriage” divorce model, similar to that of the halakhic decisors and rabbinical judges 450 discussed above. Yet, a large group of rabbinical judges rejects the “death of marriage” divorce model. Thus, for example, in one case of a recalcitrant wife, Rabbis A. Sherman 451 and H. Izirer wrote: Rabbi H. Palaggi’s ruling does not create an obligation of the wife [to accept] a writ of divorce, and [does not result in her] losing her alimony. It is a ruling regarding the obligation of the rabbinical court to act for divorce, and possibly also an obligation of the couple to heed the court, but there is no obligation on the wife vis-à-vis the husband [to accept a writ of divorce]. The judges use here the concept of obligation with various meanings. First, in “an obligation […] to accept a writ of divorce”, it refers to an enforceable legal requirement. According to the judges’ interpretation, the Palaggi Ruling imposes no such obligation. Second, in an “obligation of the rabbinical court”, it means an instruction, or possibly a word of guidance, for the court, without any legal 452 implications for the spouses. Third, in “an obligation of the couple”, it refers to ---divorce had not yet taken place. His view is disputed by other panels of judges in the High Rabbinical Court. See Rabbinical Court (Tel Aviv), file no. 8801-21-1 (24.6.2009) (available at www.rbc.gov.il/judgements/docs/400.doc). Among those challenging R. Daichovsky’s view are Rabbis Sherman and Izirer, who are the leading opponents of the “death of marriage” model (see text at n.451, infra). See High Rabbinical Court, file no. 007998479-24-1 (29.10.2007) (HaDin vehaDayan 21 [2009], 8–9). 448 449 450 451 452 See Wardle, International Marriage, 511-512. See Lifshitz, The Liberal Transformation, 28-44. Israeli civil law has recently adopted this approach regarding property distribution, and it is reflected clearly in the 2009 amendment of the Spouses (Property Relations) Law, 5733-1973, 27 LSI 313. Nonetheless, executing divorce (rather than the monetary aspects that are ancillary to divorce) is under the exclusive jurisdiction of the rabbinical courts. High Rabbinical Court, file no. 059133397-21-1 (25.12.2007) (HaDin vehaDayan 18 [2008], 11). In another verdict (High Rabbinical Court, file no. 034524637-21-1 [4.11.2007] [HaDin vehaDayan 18 (2008), 8–9]), issued a few weeks before the verdict cited in previous note, the High Rabbinical Court (Rabbis Sherman [chairman], Hashay, and Algrabli) characterized the Palaggi Ruling as “an advice” (!"#) to the court lacking authority to coerce divorce. This characterization is very similar to that suggested by the phrase “an obligation of the rabbinical court”. 116 Westreich: Talmud-Based Solutions to the Problem of the Agunah an obligation which cannot be legally enforced and which has no legal 453 implications; perhaps it alludes to a duty of an educational or religious nature to obey the rabbinical court’s instructions. In “an obligation on the wife vis-à-vis the husband”, the writers finally return to the first meaning, the legal duty to divorce, and emphasize once again that none is imposed. The judges’ interpretation empties the Palaggi Ruling of any effective legal implication. If it imposes no legal duty, but only guidance for the court or for the couple, the ruling lacks real enforcement power, and possesses only declarative importance. Similarly, the Yero!am Ruling, that if both spouses do not desire each other but the husband still refuses to formally divorce his wife he should be compelled to divorce, receives a new limiting interpretation, according to which the “right” under discussion, the wife’s right to divorce, is significantly limited. In several rulings rabbinical judges argue that Rabbenu Yero!am distinguishes between cases in which both spouses desire divorce owing to an objective reason and those in 454 which one spouse consents only as a result of the other’s attitudes. The Yero!am Ruling is interpreted as referring only to the former case, not the latter. The reasoning is that in the latter case the consenting spouse does not really want the divorce, though he or she had at one time agreed to it owing to the impossible situation in which he or she had been placed. (For example, when the wife had been insisting upon divorcing for a long time, and the husband finally agreed: we may surmise that his agreement resulted from his wife’s not leaving him any real choice.) In the last group of cases, Rabbenu Yero!am would not allow the coercing of the husband to divorce his wife. Or so it is argued. This distinction is not clear. A marriage breakdown is usually the result of continuing bad relationships between the couple. Applying the reasoning described above, a recalcitrant spouse could avoid the compulsion of a divorce in almost any instance by claiming that his or her agreement was the result of the other spouse’s behaviour. The group of cases in which the coercion could be applied would be negligible. 453 454 The religious nature of this duty is made explicit in the other High Rabbinical Court’s decision (supra, n.452), which identifies the Palaggi Ruling as “based on the principle of ‘a compulsion to fulfill the commandments of the Torah’” (!""#$%& '(& )*+",& )*-$; an expression which refers to a religious duty). See, e.g., Rabbinical Court (Tel Aviv), file no. 023559859-21-1 (11.6.2007) (HaDin vehaDayan 18 [2008], 4–5); High Rabbinical Court, file no. 323397786-22-1 (8.8.2007) (HaDin vehaDayan 18 [2008], 6). The judges in the High Rabbinical Court were, again, Rabbis Sherman and Izirer, here together with Rabbi N. Shaynin. Chapter Nine: Who is an Agunah? 9.3 117 The Right to Impose Conditions on Divorce The strict approach regarding no fault divorce goes hand-in-hand with a strict approach in a related matter – the husband’s right to impose conditions upon divorce, a right which in fact gives him the legal ability to evade divorce. According to this view, even when the husband is required to agree to the divorce, 455 he is entitled to stipulate conditions, for example, relating to money and property 456 and the custody or education of the children. Such doctrines limited the effective scope of the wife’s (apparent) entitlement to divorce. 457 Some rabbinical rulings (mainly from recent years) accept in principle the right of the husband to impose such conditions. The basis of these rulings is the following statement by one of the leading sixteenth-century halakhic authorities, 458 R. Shmuel ben Moses de Medina of Salonika (Maharashdam): [T]here is no doubt, that even in those cases in which the Sages ruled in the Mishnah […] “and these [men] are coerced [to divorce]: one who is affected with boils, etc.”, they say that he is coerced to divorce only when he does not want to divorce at all. Nevertheless, if he is desirous of divorcing, but wishes to impose some condition in the writ of divorce, in this case they surely did not say that [the court] coerces him to divorce unconditionally. According to Maharashdam, even though coerced divorce is legitimate in certain cases, the husband may impose conditions. Divorce should not be coerced unless the conditions were fulfilled. Margins have an important purpose: being out of the mainstream, the margin 459 assists us in defining the centre. This approach is valid in sociological research, as well as in the humanities. It is also helpful in legal theory: legal concepts and principles may be better defined and better understood by examining marginal cases. For our purposes, the margin sheds light on the fundamental question of the right to divorce in Jewish law. 455 456 457 458 459 E.g., he might stipulate that the property be divided according to the ruling of the rabbinical court, rather than in a civil family court. For example, he might stipulate that the children be given an ultra-Orthodox education, rather than a modern one. In this instance (and in the previous one) a rabbinical court, as a religious tribunal, might see the stipulation as legitimate owing to the husband’s beliefs. See infra, notes 460-462, and accompanying text. Shmuel ben Moses de Medina, Shut Maharashdam, Even ha‘Ezer 41 (1959, photocopy ed. of Lvov [Lemberg] 1862 ed., folio 22b-23). See Durkheim, Division of Labour, 291: “[T]he study of deviant forms will allow us to determine better the conditions for the existence of the normal state.” See also Durkheim, Rules, 47-75 at ch. 3, “Rules for Distinguishing between the Normal and the Pathological”. 118 Westreich: Talmud-Based Solutions to the Problem of the Agunah Maharashdam’s case is a marginal one, owing to the fact that entitlement to coerced divorce is not disputed. However, we have now seen that even in such a case, the husband can evade his obligation to divorce his wife. The right to divorce, accordingly, is as a practical matter quite limited, and would be difficult to exercise. We may take this a step further. If the husband can avoid divorce even when he is obliged to do so owing to specific faults, it is unlikely that a general right to divorce exists in no-fault cases. The wife therefore would not be entitled to divorce in such cases. 460 Maharashdam’s view is a minority view amongst decisors. The paucity of 461 precedent can be surmised from Maharashdam’s own words. After presenting his innovative decision, Maharashdam tells us about his quest for finding some support for his unique ruling (“I wonder whether I could find support for what I wrote”). He admits that his view is unusual amongst halakhic decisors (“up to now I have not found any place supporting this”), and expresses great satisfaction when finally he finds some support for it (“I saw and rejoiced”). The support, it should be noted, is a deduction from an earlier authority, which is not explicit and not decisive, and, anyway – it is still a single view among other halakhic authorities. Moreover, even according to Maharashdam, there is a limit to the sort of condition that a husband may impose. Maharashdam discusses two extreme conditions: one is the sort of condition that is almost impossible to fulfill and the other the sort that can easily be fulfilled. The first is not accepted, while the second is. But what is the status of the majority of conditions, which are possible but not easy to fulfill (for example, monetary conditions)? Some writers have argued that even according to Maharashdam only a minor 462 condition, which can easily be fulfilled, should be accepted. This view is supported by the context of Maharashdam’s response. That case involved not a marriage formed in the usual way, but rather a brother-in-law who refused to allow the !alitsah ceremony except on condition that the widow would not thereafter marry her uncle. (The uncle was already married to one of the brother-in-law’s relatives, and he feared that if the !alitsah took place, the uncle would divorce his first wife and marry the widowed niece.) Fulfilling the brother-in-law’s condition may be defined as easy (and might even be justified as intended to protect his relative). 460 461 462 ! See Daichovsky, Ba‘al Hamatne, 156-159; Bass, Tena’im, 149-162; Jackson, Agunah, 16-18. Shut Maharashdam, supra, n.458. Daichovsky, Ba‘al Hamatne. This opinion (or even a complete rejection of Maharashdam’s view) seems to have been reflected in the common practice among rabbinical courts until the last decade or two. See Bass, Tena’im. Rabbi Daichovsky, who served as a rabbinical judge for more than 30 years, including 20 years on the High Rabbinical Court, has expressed himself as having the same impression: see Daichovsky, Batei Hadin, 19-27. Chapter Nine: Who is an Agunah? 119 Nevertheless, some rabbinical courts adopt Maharashdam’s view, not only for minor conditions, but also for more demanding ones. Those courts apply Maharashdam’s ruling to conditions regarding matters such as custody and finances, which cannot easily be fulfilled. What stands behind that position? An important factor in this issue in the Israeli context is the deep conflict over jurisdiction, as well as over substantive matters, between civil and rabbinical legal 463 authorities. Accordingly, some rabbinical courts use Maharashdam’s view in 464 order to expand their authority, as in the following decision: ! We have written and proved in several rulings, that even when the husband is obliged to divorce [lit., to give a writ of divorce], if the husband is willing to divorce, but demands to receive the property and rights to which he is entitled by Torah law and the wife refuses, he is not to be obliged to divorce, but rather the obligation [to divorce] should be postponed. The political explanation seems to be attractive (as may be the nature of this kind 465 of explanation), but it is not completely satisfactory. It misses a strong ideological aspect, which links together the death of marriage conflict and the debate around Maharashdam’s view. We shall now analyze this aspect. 9.4 The Motivating Force: Restricting the Right to Divorce In previous sections we directed attention to the desire behind some rabbinical judges’ interpretations to limit the scope of the Yero!am and Palaggi Rulings. There is much in common between the devices used to limit these rulings and the device of limiting the right to divorce by authorizing the imposition of conditions on the basis of Maharashdam’s view. In each of these instances an apparent right to divorce has been circumvented by one or another legal and hermeneutic method: establishing a new right of the husband (the right to impose conditions) that enables him to evade divorce; interpreting the recalcitrant spouse’s duty to divorce as legally nonbinding (in an interpretation of the Palaggi Ruling); or limiting the scope of the cases in which coercion can be applied (in an interpretation of the Yero!am Ruling). Those methods appear to be tightly connected, like two sides of the same coin. 463 464 465 See supra, n.436. Rabbinical Court (Tel Aviv), file no. 054331665-21-1 (3.7.2008) (HaDin vehaDayan 22 [2009], 7). For an expanded discussion see Westreich, The Right to Divorce, 192-195. The political explanation may, however, well clarify other strict rulings of (some) rabbinical courts. A recent and sharp example is the relatively new willingness among rabbinical courts to retroactively cancel a divorce on the ground of mistake when the wife, supported by the civil courts, breaches a divorce agreement that was made in the rabbinical court. See Radzyner, Get Mut‘eh, 215-229. 120 Westreich: Talmud-Based Solutions to the Problem of the Agunah 466 In one case, the husband stipulated certain monetary arrangements as a condition to agreeing to divorce his wife. The court accepted the stipulation, justifying its decision on the ground that the Yero!am Ruling did not apply because the husband did not really desire the divorce, but agreed only owing to his wife’s incessant demands to end the marriage. In this case, limiting the Yero!am Ruling had the same object and implication as the expansion of Maharashdam’s ruling: namely, it justified conditions stipulated by the husband. The result of those hermeneutical and legal moves is quite significant. There are several cases that afford a stable basis for a right to divorce, and some of these cases even recognize a limited version of the concept of no-fault divorce, on the basis of the irretrievable breakdown of marriage. In practice, however, this right is significantly limited, and is close to being completely eliminated. The husband’s right to impose conditions on divorce and actually to evade the legal duty to divorce is supported by only a minority, and perhaps only a single view, among the post-talmudic commentators and decisors, but is adopted by several rabbinical courts as binding. Both Rabbi !ayyim Palaggi and Rabbenu Yero!am legitimize divorce in “death of marriage” cases, but their rulings have been effectively undermined. The rabbinical judges who adopt the restrictive approaches are undoubtedly aware of the fact that they are based on a minority view (Maharashdam), as well as on an innovative interpretation of the sources (Palaggi and Rabbenu Yero!am). This was cogently expressed by R. Zion Bo’aron, 467 who opposed the restrictive interpretation of the Palaggi Ruling: I have reservations about the interpretation of what the distinguished personage of his generation, Rabbi H. Palaggi, wrote, since the interpretation written here is completely the opposite of what was explicitly stated in R. H. [Palaggi]’s Responsa (and his teaching was cited many times in rabbinical verdicts). What, then, motivates the restrictive approach? We mentioned above the political motivation of the expansion of Maharashdam’s ruling (“forum shopping”, with reference to the contest or race for jurisdiction between rabbinical and civil 468 courts). But the political motivation provides only a partial explanation. It clarifies the use of Maharashdam’s opinion when this ruling has assisted the ambitions of a rabbinical court in a jurisdictional competition with a civil family 466 467 468 Rabbinical Court (Tel Aviv), file no. 023559859-21-1, supra, n.454. High Rabbinical Court, file no. 059133397-21-1 (25.12.2007) (HaDin vehaDayan 18 [2008], 11, emphasis added). R. Bo’aron was in the minority in this case, as well as in other aspects of the “death of marriage” conflict (see supra, n.447). See supra, n.436, and the accompanying text. Chapter Nine: Who is an Agunah? 121 469 court, but it is not sufficient for clarifying a broader right to impose conditions. Moreover, restrictive hermeneutical approaches regarding the Yero!am and Palaggi rulings have been applied in cases in which the main question is whether the spouse is entitled to divorce or not, where the opportunity for forum shopping or jurisdictional disputes was slight. Above all, it seems that the political explanation misses a strong ideological aspect behind the right-to-divorce debate. In what follows we shall therefore examine the ideological aspect of the debate. It should be noted, however, that this is not to say that the political aspect does not play any role in this dispute. Political motives, conscious or unconscious, might be part of the story. The argument here is that the ideological aspect is real and 470 substantive, rather than only rhetorical, and therefore needs to be explored. The “death of marriage” model sees marriage as a platform for a stable marital relationship. The right to exit marriage is applicable when such a relationship does not exist, as stated by Rabbi Shlomo Daichovsky: “We do not deal with ‘resurrection of the dead,’ and there is no reason to perform ‘artificial respiration’ 471 on dead marriages”. Those who oppose this point of view see marriage as a durable institution, perhaps due to its religious dimensions, which should be terminated only in very formal and limited cases. Otherwise, it is claimed, “there would be no restraints against this being permitted in any case”, i.e., marriage 472 would become too easily terminated upon one spouse’s demand. ! When the ideological dispute surfaced, the opponents went a step further. In a case of irretrievable breakdown of marriage, while rejecting the right to divorce, Rabbis Sherman and Izirer justified their view with the following trenchant and 473 powerful argument: 469 470 471 472 473 See Bass, Tena’im, 151-162. This suggestion might accord with some moderate critical theories of law, such as Gordon’s, according to whom the Critical Legal Studies school “doesn’t argue that law is just a mask for privilege and exploitation”, but takes seriously doctrinal legal discourse as “deliver[ing] real resources to get some leverage on social change” (Gordon, Critical Theories, 653ff.). This view (the “moderate external point of view”) has been adopted in Jewish legal scholarship as well: see, e.g., Jackson, Jewish and Islamic Law, 109-121. See supra, text at notes 446-448. See, e.g., High Rabbinical Court, file no. 5727/109 (18.12.1967), Piskei Din Rabaniyim 7, 111-113. In that case the High Rabbinical Court accepted a suit for divorce on the basis of the Palaggi Ruling. The court emphasized that the Palaggi Ruling is to be accepted only in exceptional cases, while usually unilateral divorce should be limited. High Rabbinical Court, file no. 059133397-21-1 (25.12.2007) (HaDin vehaDayan 18 [2008], 11). It should be noted that in this case the husband demanded divorce and was refused. The opponents’ attitude is therefore not necessarily “patriarchal” (see Shifman, HaHalakhah, 3), but rather reflects a fundamental attitude concerning divorce, and a basic view concerning marriage, as described here. 122 Westreich: Talmud-Based Solutions to the Problem of the Agunah The approach of “death of marriage” is not based on Torah law or [the laws of] the Sages, but rather on the law of the [non-Jewish] nations regarding civil marriage. … [In those legal systems] there is no need for a cause for divorce; rather, irretrievable breakdown of marriage and “death of marriage” suffice. The dispute reflected in this passage is acutely ideological. One side accepts divorce in cases of irretrievable marital breakdown. The other side argues that this approach is not halakhic, but rather is influenced by an external point of view, 474 namely, by “the laws of the nations”, and presents the strict view as the traditional and authentic Jewish law view of the socio-religious institution of marriage and divorce. The dispute can now be seen to have a new dimension. It is not a typical halakhic dispute, which focuses on interpretative questions concerning the classic sources. Rather, it is an ideological dispute about the correct image of divorce in Jewish law. While the supporters of the “death of marriage” model present sources that support their view, the opponents totally reject this understanding. According to the latter, this view has no basis in Jewish law, but rather comes from external, civil legal systems, and any apparent support within Jewish law for the “death of marriage” model is innovatively interpreted in a restrictive way. Had this been a normal halakhic dispute, the opponents of the “death of marriage” model would have considered all the sources, and based their decision on the relevant sources while rejecting the others. In our case, however, the opponents delegitimize the “death of marriage” view, defining it as an external approach, not rooted in Jewish law but rather as based on “the laws of the nations”. As we have shown, they do this using several hermeneutic and jurisprudential methods. The alternative to the “death of marriage” model can now, from the opponents’ point of view, be presented as the only Jewish law approach. In reality, however, the strict approach may be no less problematic, not only from a moral point of view (which is recognized by the halakhah as an important principle and part of its internal considerations, that is, the need for leniency in cases of ‘iggun), but also from a formal-halakhic point of view (i.e., the fear of 475 breaching the prohibition of eshet ish), and might no less threaten the stability of Jewish marriage, as discussed at length by the report of the Agunah Research 474 475 This phrase apparently makes reference to the adoption of no-fault divorce in many legal systems. See supra, n.448, and accompanying text. See, e.g., Cohen, Kefiyat HaGet, 195-196, who defines this situation as “the problem of our generation”. Chapter Nine: Who is an Agunah? 123 476 Unit. Despite the contrary arguments, then, the “death of marriage” model appears to remain an appropriate proper approach, which lives on in both classical and modern halakhic discourse. At the beginning of this chapter the question “who is an agunah?” was raised. The “death of marriage” conflict here revealed influences the way get refusal and aginut are defined: while the supporters of death of marriage as a ground for divorce would consider a death of marriage case (where no get is given) as a case of ‘iggun, the opponents would object it, using a narrow definition of aginut. As claimed above, bridging the gap between these approaches would not only accord recognition to many agunot, but would also legitimate the use of the various solutions discussed in this book in order to solve the problem of the agunah. 476 See Jackson, Agunah, chapter one. For further discussion (and rejection) of the argument that the “death of marriage” divorce regime contradicts Jewish law and may threaten the stability of Jewish marriage, see Jackson, ibid., 29-43. Epilogue Can we, from a halakhic point of view, use any method in order to solve the Agunah problem? Several halakhic options have been discussed in this book: coercion of a get in a no-fault-divorce case (on the basis of moredet ma’is alay), constitutive marriage annulment (on the basis of kol demekadesh – “everyone who betroths does so subject to the consent of the Rabbis”) and declarative annulment, using an implied condition or a retroactive assumption of mistaken marriage (on the basis of ada‘ata dehakhi lo kidshah nafshah – “on this assumption she did not betroth herself”). The analysis has shown a quite stable basis from Talmud to posttalmudic decisors for one or more of these solutions. Yet, their application in practice is disputed, sometimes totally rejected, accompanied by strong emotional reactions. The question, thus, is not merely a hermeneutical question, whether this or that kind of interpretation is correct or incorrect. Rather, it is an ideological question: do we acknowledge the right of the chained spouse to divorce, even if she (or he) cannot show a classical fault as the basis for the divorce suit. In other words, is this wide range of cases – cases of “death of marriage” which are not necessarily based on a specific fault, but create an unsustainable situation for the spouses – rightly defined as cases of ‘iggun. The question becomes sharper when a formal fault does exist, but the spouse (usually the husband) refuses to participate in the divorce process, and no means can be taken against him – isn’t this a case of ‘iggun? Were these cases to be defined as cases of ‘iggun, we might expect some leniency towards the use of halakhic solutions to the get refusal problem. The Talmud already mentions that the Sages were lenient in agunot cases (!"#$%&'!($)* 477 +#,-! .,! $/&0"), and such leniency was adopted in practice by post-talmudic scholars, accepting sometimes minority views or basing their decisions on 478 evidence which normally was not accepted. Several possible halakhic routes have been presented in this book, focusing on their talmudic origins and their development in post-talmudic literature. Some (the geonic moredet and Palestinian divorce clause) were discussed also historically, by examining their use in practice and the halakhic justifications given for so doing. All this reveals a wide range of possible remedies for current agunot. But the use of these solutions in practice still awaits their adoption by today’s halakhic authorities. 477 478 BT Yevamot, 88a, and elsewhere. See, e.g., Taz’s discussion on relying on minority views in she‘at hade!ak: Shul!an Arukh, EH 17:15, and Taz, ibid., sub-paragraph 15; and see Jackson, Agunah, 50-51 (§2.11). Bibliography References to talmudic and halakhic literature throughout the book are to the regular print editions (unless otherwise specified). References to the Palestinian Talmud are according to the Venice 1523 print edition page numbering. In quoting these sources I have used, when available, the Bar-Ilan Responsa Project CD and the hebrewbooks online database (www.hebrewbooks.org). Abel, Confronting 'Iggun Abel, Morgenstern Abel, Za‘aqat Dalot Albeck, Mavo Aranoff, Response Atlas, Kol Demekadesh Atlas, Netivim Auerbach, Afke‘inhu Yehudah Abel, Confronting ‘Iggun: A combination of three possible solutions to the problem of the chained wife in Jewish Law (Agunah Research Unit, vol.1), Liverpool: Deborah Charles Publications, 2011. Yehudah Abel, “Rabbi Morgenstern’s Agunah Solution”, Working Papers of the Agunah Research Unit, no. 5, 2008, available from http://www.manchesterjewishstudies.org/publi cations/. Yehudah Abel, “A Critique of Za‘aqat Dalot”, Working Papers of the Agunah Research Unit, no. 6, 2008, available from http://www.manchesterjewishstudies.org/publi cations/. !anoch Albeck, Mavo LeTalmudim, Jerusalem: Dvir, 1969. Susan Aranoff, “Continued Discussion of Agunah, Kiddushei Ta‘ut and Tears of the Oppressed – Response of Dr. Susan Aranoff”, The Edah Journal 5 (2005), 1-4. Samuel Atlas, “Kol Demekadesh Ada‘ata Derabanan Mekadesh”, Sinai 75 (1974), 119143; 102-116 (Addition). Samuel Atlas, Netivim Bamishpat Ha‘Ivri, New York: American Academy for Jewish Research, 1978. Rabbi Shlomo Zalman Auerbach, “Be‘inyan Afke‘inhu Rabanan Lekiddushin Mineh”, Torah Shebe‘al Peh 16 (1974), 36-54. 126 Westreich: Talmud-Based Solutions to the Problem of the Agunah Bass, Batlut Nissu’in Bass, Meka! Ta‘ut Bass, Tena’im Ben Meir, Re’ayah Nesibatit Ben Menahem et al., Hama!loket Ben Menahem, Hu ‘Asa Berkovits, Tnai Bleich, Kiddushei Ta‘ut Brockelmann, Lexicon Syriacum Brody, HaGeonim Brody, Stam HaTalmud Broyde, An Unsuccessful Defense Broyde, Error David Bass, “Batlut Nissu’in BeTa‘anat Meka! Ta‘ut”, Te!umin 31 (2011), 291-300. David Bass, “Hatarat Nissu’in BeTa‘anat Meka! Ta‘ut”, Te!umin 24 (5764), 194-218. David Bass, “Hatsavat Tena’im Al Yedei Ba‘al HaMe!uyav BeGet", Te!umin 25 (2005), 149162. Yehoshua Ben-Meir, “Re’ayah Nesibatit BaMishpat Ha‘Ivri”, Diné Israel 18 (19951996), 87-143. Hanina Ben-Menahem, Nathan Hecht & Shai Wozner (eds.), Hama!loket BaHalakha, Vol. 2, Boston: Boston University, 1994. Hanina Ben Menahem, “Hu ‘Asa Shelo Kahogen”, Sinai 81 (1977), 156-158. Eliezer Berkovits, Tnai BeNissu’in UVeGet, Jerusalem: Mossad Harav Kook, 1966. David J. Bleich, “Survey of Recent Halakhic Literature: Kiddushei Ta‘ut: Annulment as a Solution to the Agunah Problem”, Tradition 33 (1998), 90-128. Karl Brockelmann, Lexicon Syriacum, Hildesheim: Georg Olms, 1966. Y. Brody, “Kelum Hayu HaGeonim Me!okekim?”, Shenaton Hamishpat Ha‘Ivri 11-12 (1984-1986), 279-315. Y. Brody, "Stam HaTalmud Vedivre Ha’Amoraim”, Iggud – Selected Essays in Jewish Studies, Jerusalem: World Union of Jewish Studies, 2008, Vol. I, 216-227. Michael J. Broyde, “Review Essay – An Unsuccessful Defense of the Beit Din of Rabbi Emanuel Rackman: The Tears of The Oppressed by Aviad Hacohen”, The Edah Journal 4 (2004), 1-28. Michael J. Broyde, “Error in the Creation of Marriages in Modern Time under Jewish Law”, Diné Israel 22 (2003), 39-65 (English section). Bibliography Broyde, Kiddushei Ta‘ut Broyde, Marriage Broyde, Tripartite Agreement Cohen, Kefiyat HaGet Cohen, Ravina Daichovsky, Ba‘al Hamatne Daichovsky, Batei Hadin Dor, Torat Erets Israel Drower and Macuch, Mandaic Dictionary Durkheim, Division of Labour Durkheim, Rules Edrei, Koa! Bet Din Elon, Jewish Law: 127 Matityahu (Michael) Broyde, “Kiddushei Ta‘ut BiZemanenu”, Te!umin 22 (2002), 210-217. Michael J. Broyde, Marriage, Divorce and the Abandoned Wife in Jewish Law, Hoboken, NJ: Ktav, 2001. Michael J. Broyde, “A Proposed Tripartite Agreement to Solve Some of the Agunah Problems: A Solution without any Innovation”, Jewish Law Association Studies 20 (2010), 115. Rabbi Shear-Yashuv Cohen, “Kefiyat HaGet Bizman Haze”, Te!umin 11 (1990), 192-202. Avinoam Cohen, Ravina Ve!akhme Doro, Ramat Gan: Bar Ilan University Press, 2001. Rabbi Shlomo Daichovsky, “Ba‘al HaMatne Et Matan HaGet BeVitul !iyuvav HaKodmim”, Te!umin 26 (2006), 156-159. Rabbi Shlomo Daichovsky, “Batei Hadin Harabaniyim UVatey HaMishpat Ha’Ezra!iyim: Hirhurim Al Te!umei Ha!ikukh SheBeynehem Be‘Inyaney Mishpa!ah”, Mozne Mishpat 4 (2005), 261302. Zvi Moshe Dor, Torat Erets Israel BeBavel, Tel Aviv: Dvir, 1972. E.S. Drower and R. Macuch, A Mandaic Dictionary, Oxford: Clarendon Press, 1963. Emile Durkheim, The Division of Labour in Society (W. D. Halls, trans.), Macmillan, 1984. Emile Durkheim, The Rules of Sociological Method (G. E. G. Catlin, ed., S. A. Solovay and J. H. Mueller, trans.), Free Press, 1964. Arye Edrei, “Koa! Bet Din VeDine Nissu’in VeGerushin”, Shenaton Hamishpat Ha‘Ivri 21 (1998-2000), 1-35. Menahem Elon, Jewish Law: History, Sources, Principles (B. Auerbach & M.J. Sykes trans.), Philadelphia: Jewish Publication Society, 1994. 128 Westreich: Talmud-Based Solutions to the Problem of the Agunah Epstein, Mavo Lenusa! HaMishnah Falk, Gerushin Florsheim, Rav !isda Franzus, Od Le‘Kol Demekadesh Freimann, Seder Kiddushin Friedman, D., Contractual Risk Friedman, M.A. Jewish Marriage Friedman, M.A., Divorce Friedman, S.Y., Ha’Isha Rabbah Friedman, S.Y., Tosefta Atikta: Gertner & Karlinsky, Ein Tnai BeNissu’in Gilat, Perakim Ya‘akov N. Epstein, Mavo Lenusa! HaMishnah, Jerusalem: Magnes, 2001, 2nd ed. Ze’ev Falk, Tevi‘at Gerushin Mitsad Ha’Isha BeDine Israel, Jerusalem: Institute for Legislative Research and Comparative Law, 1973. Yo‘el Florsheim, “Letoldot !ayav Shel Rav !isda”, Sinai 71 (1972), 121-131. Israel Franzus, “Od Le‘Kol Demekadesh Ada‘ata Derabanan Mekadesh’ ” , Sinai 77 (1975), 91-93. Avraham H. Freimann, Seder Kiddushin VeNissu’in A!arei !atimat HaTalmud, Jerusalem: Mossad Harav Kook, 1964. Daniel Friedman, “HaSikun Ha!ozi VeTa‘ut Ve!at’aya BaKeda’iyut”, Iyune Mishpat 14 (1989), 459-478. Mordechai A. Friedman, Jewish Marriage in Palestine – A Cairo Geniza Study, Tel Aviv and New York: Tel Aviv University and the Jewish Theological Seminary of America, 1980. Mordechai A. Friedman, “Divorce upon the Wife’s Demand as Reflected in Manuscripts from the Cairo Geniza”, The Jewish Law Annual 4 (1981), 103-126. Shamma Y. Friedman, “Perek Ha’isha Rabbah BaBavli Betseruf Mavo Kelali Al Derekh !eker Hasugya", Me!karim UMekorot 1 (1978), 275-441. Shama Y. Friedman, Tosefta Atikta, Ramat Gan: Bar Ilan University, 2003. Rabbi Zvi Gertner and Rabbi Bezalel Karlinsky, “Ein Tnai Benissu’in: HaMa‘avak Al Kedushat HaYi!us BeYisra’el”, Yeshurun 8 (2001), 678-705; Yeshurun 9 (2001), 669-704; Yeshurun 10 (2002), 711-750. Yits!ak D. Gilat, Perakim Behishtalshelut HaHalakhah, Ramat Gan: Bar Ilan University, 1992. Bibliography Goldberg, Ein Hafka‘at Kiddushin 129 Rabbi Zalman N. Goldberg, “|Ein Hafka‘at Kiddushin Lelo Get”, Te!umin 23 (2003), 165168. Goldberg, Hafka‘at Kiddushin Enah Pitron Rabbi Zalman N. Goldberg, “Hafka‘at Kiddushin Enah Pitron La‘aginut”, Te!umin 23 (2003), 158-160. Goldberg, Tsimtsum Ma!lokot Avraham Goldberg, “Derakhim Shel Tsimtsum Ma!lokot Etsel Amora’e Bavel", Me!kerei Talmud 1 (1990), 135-153. Gordon, Critical Theories Robert W. Gordon, “Some Critical Theories of Law and Their Critics”, in The Politics of Law: A Progressive Critique (ed. David Kairys), New York: Pantheon Books, 1998, 3rd ed., 641-661. Grossman, !asidot Umordot Avraham Grossman, !asidot Umordot, Jerusalem: Merkaz Zalman Shazar, 2003, 2nd ed. Grossman, Pious and Rebellious Avraham Grossman, Pious and Rebellious – Jewish Women in Medieval Europe, Waltham: Brandeis University Press, 2004. Hacohen, Oppressed Aviad Hacohen, The Tears of The Oppressed, An Examination of the Agunah Problem: Background and Halachic Sources, Hoboken, NJ: Ktav, 2004. Halivni, Mekorot David Weiss Halivni, Mekorot Umasorot: Nashim, Toronto: Otsreinu, 1994. Halivni, Mevo’ot David Halivni, Mevo’ot Limkorot Umasorot, Jerusalem: Magnes, 2009. Halperin-Kaddari, Tav Lemeitav Ruth Halperin-Kaddari, “Tav Lemeitav Tan Du MiLemeitav Armalu: An Analysis of the Presumption”, The Edah Journal 4 (2002), 124. Henkin, Ha’omnam Eitam Henkin, “Ha’omnam Efshar Levatel Kiddushin Mishum Meka! Ta‘ut?", Te!umin 31 (2011), 282-290. Hidary, Legal Pluralism Richard Hidary, Dispute for the Sake of Heaven: Legal Pluralism in the Talmud, Providence, R.I.: Brown Judaic Studies, 2010. Jackson, Agunah Bernard S. Jackson, Agunah: The Manchester Analysis (Agunah Research Unit, vol.1), 130 Westreich: Talmud-Based Solutions to the Problem of the Agunah Jackson, Jewish and Islamic Law Jackson, Papyri Katzoff, Papyrus Knol, Agunah and Ideology Lavi, Ha’im Lavi, Hafka‘at Kiddushin Einah Ma’aneh Levine, Review Lieberman, Hilkhot HaYerushalmi Lieberman, Tosefta Kifshuta Lifshitz, Afke‘inhu Liverpool: Deborah Charles Publications, 2011. Bernard Jackson, “Comparing Jewish and Islamic Law” (review essay), Journal of Semitic Studies, 48 (2003), 109-121 (reprinted in Review of Rabbinic Judaism, 6 [2003], 333345). Bernard S. Jackson, “Some Reflections on Family Law in the Papyri”, Jewish Law Association Studies 14 (2002), 141-177. Ranon Katzoff, “Legal Commentary”, in N. Lewis, R. Katzoff, and J.C. Greenfield, “Papyrus Yadin 18: I. Text, Translation and Notes (NL), II. Legal Commentary (RK), III. The Aramaic Subscription (JCG)”, Israel Exploration Journal 37 (1987), 229-250. Shoshana Knol, Agunah and Ideology (Agunah Research Unit, vol.3), Liverpool: Deborah Charles Publications, 2011. Rabbi Uri’el Lavi, “Ha’im Nitan Lehafki‘a Kiddushin Shel Sarvan Get?", Te!umin 27 (5767), 304-310. Rabbi Uri’el Lavi, “Hafka‘at Kiddushin Einah Ma’aneh Lesarvanut Get”, Te!umin 29 (2009), 247-256. David Levine, “Review of Richard Hidary, ‘Dispute for the Sake of Heaven: Legal Pluralism in the Talmud’, H-Judaic, H-Net Reviews, January, 2012, https://www.hnet.org/ reviews/showrev.php?id=3D32972 Saul Lieberman, Hilkhot HaYerushalmi LehaRambam, New York: Bet Hamidrash LeRabanim Be’America, 1948. Saul Lieberman, Tosefta Kifshuta, vol. 6, Nashim, New York: Bet Hamidrash LeRabanim Be’America, 1967. Berachyahu Lifshitz, “Afke‘inhu Rabanan Lekiddushin Minayhu”, Miperot Hakerem, Yavne: Yeshivat Kerem BeYavne, 2004, 317324. Bibliography Lifshitz, Al Masoret 131 Berachyahu Lifshitz, “Al Masoret, Al Samkhut Ve‘al Derekh Hahanmakah”, Te!umin 28 (2008), 82-91. Lifshitz, Asmakhta Berachyahu Lifshitz, Asmakhta, Jerusalem: Magnes, 1988. Lifshitz, The Liberal Transformation Shahar Lifshitz, “The Liberal Transformation of Spousal Law: Past, Present and Future”, Theoretical Inquiries in Law 13 (2012), 15-73. Mack, Shiv‘im Panim Hananel Mack, “Shiv‘im Panim LaTorah: Lema!alacho Shel Bituy”, in Rabbi Mordechai Breuer Festschrift: Collected Papers in Jewish Studies (ed. Moshe Bar Asher), Vol. 2, Jerusalem: Academon, 1992, 449-462. Malka, Ein Hafka‘at Kiddushin Rabbi David Malka, “Ein Hafka‘at Kiddushin Limsoravot Get” (2010), www.psakdin.co.il/ fileprint.asp?FileName=/Mishpaha/Public/art_ ccaa.htm. Moscovitz, Talmudic Reasoning Leib Moscovitz, Talmudic Reasoning – From Casuistics to Conceptualization, Tübingen: Mohr Siebeck, 2002. Olszowy-Schlanger, Karaite Marriage Judith Olszowy-Schlanger, Karaite Marriage Documents from the Cairo Geniza, Leiden, New York and Cologne: Brill, 1997. Binyamin Porat, “Ha!oze HaKafuy Ve‘ikron Porat, Ha!oze Hakafuy HaTdedek Ha!ozi”, Diné Israel 22 (5763), 49110. Preuss, Biblical and Talmudic Medicine Julius Preuss, Biblical and Talmudic Medicine (trans. and edited by Fred Rosner), New Jersey: Jason Aronson, 1993. Radzyner, Get Mut‘eh Ami!ai Radzyner, "MiLvov LeTel-Aviv: Pesikot Get Mut‘eh Beshel Hafarat Heskem HaGerushin BeVatei HaDin HaRabani’im", Mishpatim 39 (2009), 155-231. Riskin, Divorce Rabbi Shlomo Riskin, Woman and Jewish Divorce, Hoboken, NJ: Ktav, 1989. Riskin, Hafka‘at Kiddushin (Engl. Version) Rabbi Shlomo Riskin, “Hafka‘at Kiddushin: Towards Solving the Aguna Problem in Our Time”, Tradition 36 (2002), 1-36. 132 Westreich: Talmud-Based Solutions to the Problem of the Agunah Riskin, Hafka‘at Kiddushin Riskin, Response Rozen-Zvi, Forum Shopping Sagi and Statman, Religion and Morality Sagi, The Open Canon Sagi, Yahadut Shapira, Gerushin Shifman, HaHalakhah Shifman, Safa Datit Sho!etman, Hafka‘at Kiddushin Sho!etman, Ones Sokoloff, Dictionary Statman, Autonomy Rabbi Shlomo Riskin, “Hafka‘at Kiddushin – Pitaron La‘aginut”, Te!umin 22 (5762), 191209. Rabbi Shlomo Riskin, “Response”, Tradition 36 (2002), 44-53. Ariel Rosen-Zvi, “Forum Shopping between Religious and Secular Courts (and Its Impact on the Legal System)”, Tel Aviv University Studies in Law 9 (1989), 347-396. Avi Sagi and Daniel Statman, Religion and Morality (trans. Batya Stein), Amsterdam: Rodopi, 1995. Avi Sagi, The Open Canon: On the Meaning of Halakhic Discourse (trans. Batya Stein), London and New York: Continuum, 2007. Avi Sagi, Yahadut: Bein Dat LeMusar, Tel Aviv: HaKibbuts HaMe’u!ad, 1998. Moshe Shapira, “Gerushin Begin Me’isa”, Diné Israel 2 (1971), 117-153. Pin!as Shifman, “HaHalakhah HaYehudit Bimtsi’ut Mishtana: Ma Me‘akev Et Me‘ukavot HaGet?”, Alei Mishpat 6 (2007), 27-46. Pin!as Shifman, “Safa Datit Vesafa Ezra!it Bedinei Mishpa!ah”, Mishpat Va‘asakim 10 (2009), 423-445. Eliav Sho!etman, “Hafka‘at Kiddushin: Derekh Efsharit Lefitron Ba‘ayat Me‘ukavot HaGet?”, Shenaton Hamishpat Ha‘Ivri 20 (1995-1997), 349-397. Eliav Shoxetman, “Kiddushin Mehamat ‘Ones”, Sinai 105 (1990), 109-124. Michael Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talmud and Geonic Periods, Ramat Gan: Bar Ilan University, 2002. Daniel Statman, “Otonomya Vesamkhut Betanuro Shel Achnay”, Me!kerei Mishpat 24 (2009), 639-662. Bibliography Talmudic Encyclopedia: Ta-Shma, Hasifrut Haparshanit Urbach, Ba‘alei HaTosafot Wardle, International Marriage Weiss, Le!eker HaTalmud Westreich A., Book Review Westreich A., Divorce on Demand Westreich A., Gatekeepers Westreich A., Hafka‘at Kiddushin Westreich A., Halakhic Story Westreich A., Kama Milim 133 Talmudic Encyclopedia (ed. Rabbi Shlomo Y. Zevin), vol. 9, Jerusalem: The Talmudic Encyclopedia Press and Mossad HaRav Kook, 1959. Israel M. Ta-Shma, Hasifrut Haparshanit LaTalmud, vol. 1, Jerusalem: Magnes, 2000, 2nd ed.. Ephraim E. Urbach, Ba‘alei HaTosafot, Jerusalem: Bialik Institute, 1995, 5th ed. Lynn D. Wardle, “International Marriage and Divorce Regulation and Recognition: A Survey”, Family Law Quarterly 29 (1995), 497-515. Avraham Weiss, Le!eker HaTalmud, New York: Feldheim, 1956. Avishalom Westreich, “Book Review: ‘The Tears of The Oppressed, An Examination of the Agunah Problem: Background and Halachic Sources’ by Aviad Hacohen", The Jewish Law Annual 17 (2007), 306-313. Avishalom Westreich, “Divorce on Demand: The History, Dogmatics, and Hermeneutics of the Wife’s Right to Divorce in Jewish Law”, Journal of Jewish Studies 62 (2011), 340-363 Avishalom Westreich, “The ‘Gatekeepers’ of Jewish Marriage Law: Marriage Annulment as a Test Case”, Journal of Law and Religion 27 (forthcoming, 2012). Avishalom Westreich, “Hafka‘at Kiddushin Bamekorot HaTalmudiyim”, Sidra 27 (forthcoming, 2012). Avishalom Westreich, “ ‘And the Sages did not have Power to Release her without a Get’ (BT Ketubbot, 74a): On the Literary Development of the Halakhic Story”, Paper delivered at the Fifteenth World Congress of Jewish Studies, 2009 (on file with the author). Avishalom Westreich, “Kama Milim Al Me!kar Hamishpat Ha‘Ivri”, Shenaton Hamishpat Ha‘Ivri 26 (2009-2011), 245-261. 134 Westreich: Talmud-Based Solutions to the Problem of the Agunah Westreich A., Moredet Westreich A., The Right to Divorce Westreich A., Torts Westreich A., Umdena Westreich E., Rise and Decline Westreich E., Temurot Wieder, Rebuttal Wieder, Rejoinder Yosef, Kol Hamekadesh Zussman, Veshuv LiRushalmi Nezikin Avishalom Westreich, “ ‘Al Kefiyat Get BeHilkhat HaMoredet BaMishnah UvaTalmud”, Me!kerei Mishpat 25 (2010), 563-591. Avishalom Westreich, “The Right to Divorce in Jewish Law: Between Politics and Ideology”, International Journal of the Jurisprudence of the Family 1 (2010), 177196. Avishalom Westreich, Hermeneutics and Developments in the Talmudic Theory of Torts as Reflected in Exceptional Cases of Exemption, PhD, Ramat Gan: Bar Ilan University, 2007. Avishalom Westreich, “Umdena as a Ground for Marriage Annulment: Between Mistaken Transaction (Kiddushei Ta’ut) and Terminative Condition”, Jewish Law Association Studies 20 (2010), 330-352 Elimelech Westreich, “The Rise and Decline of the Law of the Rebellious Wife in Medieval Jewish Law”, Jewish Law Association Studies 12 (2002), 207-218. Elimelech Westreich, Temurot Bema‘amad Ha’Isha Bamishpat Ha‘Ivri: Masa Bein Masorot, Jerusalem: Magnes Press, 2002. Rabbi Jeremy Wieder, “Hafka‘at Kiddushin: A Rebuttal”, Tradition 36 (2002), 37-43. Rabbi Jeremy Wieder, “Hafka‘at Kiddushin: Rejoinder”, Tradition 37 (2003), 61-78.! Rabbi Ovadya Yosef, “Kol Hamekadesh Ada‘ata Derabanan Mekadesh Ve’afke‘inhu Rabanan Lekiddushin Mine”, Torah Shebe‘al Peh 3 (5761), 96-103. Ya‘akov Zussman, “Veshuv LiRushalmi Nezikin”, Me!kerei Talmud 1 (1990), 55-103.