Rabbinic Traditions between Palestine and Babylonia

Ancient Judaism and

Early Christianity
Arbeiten zur Geschichte des antiken Judentums
und des Urchristentums

Founding Editor

Martin Hengel † Tübingen

Executive Editors

Cilliers Breytenbach, Berlin

Martin Goodman, Oxford

Editorial Board

Lutz Doering, Münster – Pieter W. van der Horst, Utrecht

Tal Ilan, Berlin – Judith Lieu, Cambridge
Tessa Rajak, Reading/Oxford – Daniel R. Schwartz, Jerusalem
Seth Schwartz, New York

VOLUME 89

The titles published in this series are listed at brill.com/ajec

 Rabbinic Traditions
between Palestine and Babylonia

Edited by

Ronit Nikolsky and Tal Ilan

LEIDEN | BOSTON
Library of Congress Cataloging-in-Publication Data

Rabbinic traditions between Palestine and Babylonia / edited by Ronit Nikolsky and Tal Ilan.
  pages cm. — (Ancient Judaism and early Christianity, ISSN 1871-6636 ; v. 89)
 Includes index.
 Summary: “In this book various authors explore how rabbinic traditions that were formulated in the Land
of Israel migrated to Jewish study houses in Babylonia. The authors demonstrate how the new location and
the unique literary character of the Babylonian Talmud combine to create new and surprising texts out of
the old ones. Some authors concentrate on inner rabbinic social structures that influence the changes the
traditions underwent. Others show the influence of the host culture on the metamorphosis of the traditions.
The result is a complex study of cultural processes, as shaped by a unique historical moment”—Provided by
publisher.
 ISBN 978-90-04-26789-3 (hardback) — ISBN 978-90-04-27731-1 (e-book) 1. Rabbinical literature—History
and criticism. I. Nikolsky, Ronit. II. Ilan, Tal.
 BM496.6.R345 2014
 296.1’24—dc23
2014014738

This publication has been typeset in the multilingual ‘Brill’ typeface. With over 5,100 characters covering
Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities.
For more information, please see brill.com/brill-typeface.

issn 1871-6636
isbn 978-90-04-26789-3 (hardback)
isbn 978-90-04-27731-1 (e-book)

Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands.

Koninklijke Brill nv incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided
that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive,
Suite 910, Danvers, ma 01923, usa. Fees are subject to change.

This book is printed on acid-free paper.

Dedicated to our dearly departed friend and colleague Hanan Eshel

∵
Contents

1 ‫מהתם להכא‬, from There to Here (bSanh 5a), Rabbinic Traditions between
Palestine and Babylonia: An Introduction  1
Ronit Nikolsky and Tal Ilan

2 Now You See it, Now You Don’t: Can Source-Criticism Perform Magic on
Talmudic Passages about Sorcery?  32
Shamma Friedman

3 No Boundaries for the Construction of Boundaries: The Babylonian

Talmud’s Emphasis on Demarcation of Identity  84
Moshe Lavee

4 Midgets and Mules, Elephants and Exilarchs: On the Metamorphosis of

a Polemical Amoraic Story  117
Geoffrey Herman

5 Rescue from Transgression through Death; Rescue from Death through

Transgression  133
Christiane Tzuberi

6 A Tale of Two Sinais: On the Reception of the Torah according to

bShab 88a  147
Amram Tropper

7 Heaven and Hell: Babylonia and the Land of Israel in the Bavli  158
Tal Ilan

8 From Disagreement to Talmudic Discourse: Progymnasmata and the

Evolution of a Rabbinic Genre  173
David Brodsky

9 The Misfortunes and Adventures of Elihoreph and Ahiah in the Land

of Israel and in Babylonia: The Metamorphosis of a Narrative Tradition
and Ways of Acculturation  232
Reuven Kiperwasser
viii contents

10 Commercial Law in Rome and Ctesiphon: Roman Jurisconsults, Rabbis

and Sasanian Dastwars on Risk  250
Yaakov Elman

11 From Palestine to Babylonia and Back: The Place of the Bavli and the
Tanhuma on the Rabbinic Cultural Continuum  284
Ronit Nikolsky

12 Was Rabbi Aqiva a Martyr? Palestinian and Babylonian Influences in the

Development of a Legend  306
Paul Mandel

Index of Sources  355
Index of Authors  363
Index of Rabbinic Names  367
Index of Place Names  369
General Index  371
‫מהתם להכא‬, from There to Here (bSanh 5a),
Rabbinic Traditions between Palestine
and Babylonia: An Introduction
Ronit Nikolsky and Tal Ilan

This collection of studies is about differences between the rabbinic Palestinian

literature and its parallels in the Babylonian Talmud (Bavli). The fact that dif-
ferences exist may be considered banal, and therefore it is legitimate to inquire
why such a volume is at all necessary. To explain this it is helpful to note that
a vast part of the Bavli comprises texts that have Palestinian parallels. These
include citations of the Mishnah, of baraitot, and of sayings attributed to
Palestinian sages up to the fourth century (which include biblical interpreta-
tion, midrash, as well as legal discussion, halakhah). Recently, scholars have
even argued that Palestinian tractates form the foundation of Babylonian
ones.1 The reason for this enormous volume of Palestinian traditions which the
Babylonians incorporated in their composition is likely their self-perception
as being the heirs to this sacred canon. In a tradition formulated in extreme
terms, the Babylonian rabbis convey the notion that they are responsible not
only for the texts they composed in their own land after the Mishnah was pub-
lished, but also that all the wealth of knowledge that was produced in Palestine
between the biblical period and the publication of the Talmud would not have
been possible without intensive Babylonian support: “In the beginning, when
the Torah was forgotten from Israel, Ezra went up (to the Land of Israel) from
Babylon and restored it. (When) it was again forgotten, Hillel the Babylonian
went up and restored it. (When) it was again forgotten Rabbi Hiyya and his
sons went up and restored it” (bSuk 20a). The ideology voiced in this tradition
is that the entire rabbinic Palestinian corpus is in fact the Torah of Babylonia.
This of course makes it authoritative for the Babylonians and worthy of cita-
tion in their own composition (the Bavli).

* This volume began as a session in the International SBL meeting in Rome 2009. However, not
all presenters in the session contributed to this volume and not all contributors to this vol-
ume participated in the session. The editors take this opportunity to thank Amram Tropper
for reading the contributions of the non-English speakers and suggesting corrections. We are,
of course, responsible for the final product and all its shortcomings.
1 A. M. Gray, A Talmud in Exile: The Influence of Yerushalmi Avodah Zarah on the Formation of
Bavli Avodah Zarah (Providence RI 2005) 199.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�01

2 NIKOLSKY AND ILAN

In addition, over the course of time, the Babylonians who composed the
Bavli became convinced that not only is their knowledge the foundation of the
Torah of the Land of Israel, but also that their interpretation of the Mishnah
(the Bavli) is more authentic and divinely blessed than that of their Palestinian
brethren (the Yerushalmi).2 Historically speaking, this claim gained credence
in Babylonian eyes as the Bavli triumphed over its Palestinian counterpart in
becoming the most prominent canonical Jewish text.

1 Theorizing Diaspora

The Babylonian–Palestinian axis is a particular case of what was conceptual-

ized in the last decades among scholars of cultural studies as “diaspora studies”;
in fact, diaspora studies were constructed first and foremost with the Jewish
case in mind,3 though not necessarily the Palestinian–Babylonian example.
Rather, the Jewish diaspora in Medieval through Modern times was in view.
When more than 150 people are declared a unified group, a tribe, a nation,
we have an imagined community.4 The imagined community, regardless of
whether it was created because of political, economic or other reasons, drives
or motivations, shares a narrative that serves as a basis for its self-definition as a
group; this could be a story of a common ancestor, stories about shared events
in the past, or a divine command. Throughout human history, the imagined
community to which people belonged constantly grew in size. From a natural
group of some 150, it grew to a few tens of thousands in a common ancient
city (Babylon, Nippur), several hundreds of thousands in an ancient nation
(Solomon’s kingdom), or a few millions in an ancient empire (the Assyrian).
The importance of narrative in creating a sense of a unified group becomes
very clear in cases of diaspora. Migrations happened, of course, throughout
the history and prehistory of the human race, and groups of humans sepa-
rated from their land of birth, migrated and intermixed with other groups

2 I. M. Gafni, Land Center and Diaspora (Sheffield 1997) 96–7; J. L. Rubenstein, The Culture of
the Babylonian Talmud (Baltimore 2003) 159–60.
3 R. Brubaker, “The ‘Diaspora’ Diaspora,” Ethnic and Racial Studies 28 (2005) 1–19.
4 Scientists calculated that, up until this group size, people can know each other personally,
and are therefore a “natural” group; for larger groups one need a story that will explain why
a person belongs to the same group as people that he or she does not know and will prob-
ably never know; see D. S. Massey, “A Brief History of Human Society: The Origin and Role of
Emotion in Social Life,” American Sociological Review 67 (2002) 1–29; B. Anderson, Imagined
Communities: Reflections on the Origin and Spread of Nationalism (London 2006—new edi-
tion); the concept of the “imagined community” is the major thesis of the book.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 3

(or killed each other). For a migration to become a diaspora, a narrative is

needed. According to Brubaker’s balanced definition, being a diaspora means
having the following three criteria: (1) dispersion in space, (2) orientation to a
homeland, and (3) boundary maintenance of the group in the host country.5
Aside from physical grounds such as skin color, and resistance to integration
into the host-society’s ways, a story of self-separation serves as a significant
cultural tool for demarcation. This story usually involves cultural memory of
the home-country, but the demarcation also entails continuing to use the orig-
inal cultural canon that migrated with the people. All this holds true for the
Jewish diaspora in Babylonia. If the second and third criteria mentioned above
are not maintained, the group will assimilate into the local society within a few
decades, with old customs being forgotten, and physical diversity being toler-
ated after some time.
Exile, or rather the contrast between Land and Exile, is part and parcel of
the mega-narrative of the Deuteronomistic culture of the Hebrew Bible. It is
part of the God–People–Land triangle as the following quotation shows:

And now, Hear O Israel, unto the statutes and unto the ordinances, which
I teach you, to do them; that you may live, and go in and possess the land
which the Lord, the God of your fathers, gives you (Deut 4:1).

The Deuteronomist here makes the connection between adhering to God’s

statutes and possessing the land. Later, he reveals what will happen if the
Children of Israel should not adhere to his command­ments:

When you shall beget children, and children’s children, and you shall
have been long in the land, and shall deal corruptly, and make a graven
image, even the form of anything, and shall do that which is evil in the
sight of the Lord your God, to provoke Him . . . the Lord shall scatter
you among the peoples, and you shall be left few in number among the
nations, where the Lord shall lead you away (Deut 4:25–7).

Indeed, this is how the Bible explains the exile of the northern tribes in the
time of Shalmaneser:

And the king of Assyria carried Israel away to Assyria, and put them in
Halah, and in Habor, on the river of Gozan, and in the cities of the Medes,
because they hearkened not to the voice of the Lord their God, but

5 Brubaker, “Diaspora,” 5.
4 NIKOLSKY AND ILAN

transgressed His covenant, even all that Moses the servant of the Lord
commanded, and would not hear it, nor do it (2Kgs 18:11–2).

Later on, this is how the Bible explains the exile of the Judeans, in the hand of
Nebuzaradan, the army commander of Nebuchadnezzar:

And he (Zedekiah) did that which was evil in the sight of the Lord,
according to all that Jehoiakim had done . . . And the residue of the peo-
ple that were left in the city, and those that fell away, that fell to the king
of Babylon, and the residue of the multitude, did Nebuzaradan the cap-
tain of the guard carry away captive. (2Kgs 24:19 and 25:11).

Regardless of what Shalmaneser thought when he exiled the northern tribes

into Mesopotamia, or what Nebuchadnezzar thought when he brought the
Judeans there, or what the rulers of the Roman or the Parthian empires thought
when they allowed freedom of movement of Jews (and others) throughout the
empire, the Jews knew for certain that their exile was a God-sent punishment,
and that they were bound to return to their land when God saw fit. They there-
fore held to and developed the idea of keeping themselves as a separate group.
The awareness of, and the particular view on, the conditions of exile and
return in Jewish culture made the Jews construct strategies and ideologies for
dealing with their circumstances. These perspectives pronounced already in
the biblical text, first and foremost in the words of the Prophet Jeremiah, who
instructed the exiles of Judah how to behave in the host country:

Thus said the Lord of Hosts, the God of Israel, to all the captivity, whom
I have caused to be carried away captive from Jerusalem unto Babylon:
Build houses, and dwell in them, and plant gardens, and eat their fruit;
take wives, and beget sons and daughters; and take wives for your sons,
and give your daughters to husbands, that they may bear sons and daugh-
ters; and multiply there, and be not diminished. And seek the peace of
the city where I have caused you to be carried away captive, and pray to
God for it; for in its peace shall you have peace. For thus says the Lord
of hosts, the God of Israel: Let not your prophets that are in the midst
of you, and your diviners, beguile you, neither hearken to your dreams
which you dreamed, for they prophesy falsely in My name; I have not sent
them, said the Lord (Jer 29:4–9).

The message of Jeremiah here is twofold. On the one hand, he calls the exiles to
continue their physical and emotional existence to the best of their ability in
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 5

their host country, to participate in the country’s economy (build houses), and
to connect themselves emotionally to the new mother-land (pray for the well-
being of the city of their dwelling). On the other hand, Jeremiah warns them
against developing culturally on their own: no prophet from among the exiles
should be heeded. This restriction asserts that the people in the diaspora will
not be able to develop their own separate culture, and would therefore remain
dependent on the original culture that had developed in their homeland and
was brought with them when they came to their new home.
Perhaps the Jews followed Jeremiah’s advice in the early period of their stay
in Babylonia, but by the time we read about them in the Bavli, their view on
diaspora and on themselves as carriers of authentic Jewish culture is certainly
a turn away from the biblical instruction; their move challenges the idea of
diaspora as a secondary place. Their assertiveness about their own culture,
and the legitimacy of their creativity, found its expression in the vast literature
embedded in the Bavli. How they managed the balance between the original
holy cultural canon—one that should be taken authoritatively—and the new
ideas they developed throughout their new homeland, is a complicated and
sophisticated process. Here we will only dwell on one aspect among many
ways of juggling cultural heritage. It has to do with the exegetical approach of
Rabbi Aqiva.
In early Palestinian sources we know of two schools of rabbinic thought,
the one of Rabbi Yishmael and the one of Rabbi Aqiva.6 The major difference
between the way they approached Scripture was that while Rabbi Yishmael
interpreted difficulties in the biblical story in light of their literary context,
assuming that “the Torah spoke in a language of humans” (SifNum 112), Rabbi
Aqiva did not hesitate to deconstruct the texts of the Bible (to use the term
anachronistically), identifying in banal texts obscurities which he interpreted
as wholly meaningful. This method of Rabbi Aqiva opened up the Bible to
extreme interpretation, at the price of losing some of the original meaning of
the biblical narrative. However, this approach enabled the rabbis to keep the
Bible as an unchanged canonical text, while interpreting it to fit their changing
circumstances.7
The Yerushalmi kept the memory of the two schools of thought. In the Bavli,
by contrast, Rabbi Yishmael’s school is hardly mentioned, while Rabbi Aqiva,

6 See M. I. Kahana, “The Halakhic Midrashim,” in: S. Safrai, Z. Safrai, J. Schwartz, and P. J.
Tomson (eds.), The Literature of the Sages: Second Part: Midrash and Targum, Liturgy, Poetry,
Mysticism, Contracts, Inscriptions, Ancient Science and the Languages of Rabbinic Literature
(Assen 2006) 3–105.
7 See description of the two schools in Kahana, “Halakhic Midrashim,” 17–39.
6 NIKOLSKY AND ILAN

and his method of interpretation, is highly honored, frequently used, and even
developed further.8 The freedom this approach provided the rabbis for deal-
ing with the ancient sanctified, and therefore crystallized texts, found favor in
the eyes of the Babylonians, and allowed them to maintain a balance between
claims of authenticity and the introduction of necessary innovations.

2 A Brief History of Babylonian Jewry

As shown by many and summed up convincingly by I. Gafni, what we know

about Babylonian Jewry between the return from the Babylonian Exile (end of
the sixth century BCE) and up to the time when the Mishnah was edited and
published (ca. 200 CE)—a period of almost a millennium—is next to nothing.9
From around 200 CE, a wealth of information is provided by the Bavli, which,
although a later composition, describes a reality and quotes scores of traditions
from the third century onward. Thus, we know almost nothing about Jewish
society in Babylonia prior to the infiltration of rabbinic culture into it, but
we receive the impression that we know a lot about it once rabbinic Judaism
became established in Babylonia. The Bavli describes, inter alia, a diaspora
concentrated in many cities and villages in Mesopotamia, one which enjoyed a
large measure of autonomy. It had its own leadership—the exilarchate, which
functioned like a royal dynasty to all intents and purposes—and its own civil
and criminal courts of law, which in extreme cases also administered capital
punishment. And of course it also nurtured a wide network of study houses,
all busy studying and interpreting the Mishnah, keeping in touch with each
other, exchanging ideas and maintaining constant contact with the rabbinic
study houses in Palestine through a number of traveling scholars (nehutei)
who transmitted the most recent innovations from one center to another.
As time went by, a new self-perception of themselves developed among
Babylonian rabbis, which included the assertion that they received their abso-
lute halakhic authority together with the Mishnah, which was brought from
Palestine by Rav (Abba Arikha). The latter had studied with Rabbi Yehudah

8 Kahana, “Halakhic Midrashim,” 59–60.

9 I. M. Gafni, The Jews of Babylonia in the Talmudic Era: A Social and Cultural History (Jerusalem
1990) 20–35 [Hebrew]; G. Herman, “The Jews of Parthian Babylonia,” in: P. Wick and
M. Zehnder (eds.), The Parthian Empire and its Religions: Studies in the Dynamics of Religious
Diversities (Gutenburg 2012) 141–50.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 7

Ha-Nasi, the father of the Mishnah, and in 219 CE returned to Babylonia and
founded the rabbinic study houses where the Mishnah was interpreted.10

3 Scholarship on the Bavli and Yerushalmi Parallels

A significant amount of Palestinian material is included in the Bavli, and

there are substantial differences between the way these texts appear in the
Yerushalmi and other Palestinian sources and the way they are presented in
the Bavli. These differences have interested scholars for decades and this vol-
ume is another link in the chain of scholarship. It takes into account dramatic
shifts in recent scholarship of the Talmud and the influence these have had
on the study and understanding of talmudic parallels, making the Bavli the
focus of the discussion. The following is a general, and by no means compre-
hensive, overview of the scholarship on the issue. We concentrate here on two
approaches to the question of Palestinian and Babylonian parallels—the his-
torical and the literary—because we believe they demonstrate both the leaps
scholarship has made since the middle of the 20th century, and how the gap
between the two approaches, which at one time seemed unbridgeable, is
slowly closing.
In order to demonstrate how attitudes to the way rabbinic texts should be
read have changed over the years, we will look at one story that has repeatedly
drawn scholarly attention, both because of its literary qualities, and because of
its supposed historical significance. It relates the events that led to the depo-
sition of Rabban Gamaliel of Yavneh (from what? patriarchate? head of the
assembly? head of the law court? head of the yeshivah?) and the appointment
of another sage, Eleazar ben Azariah, in his place. The story is related both in
the Yerushalmi (yBer 4:1, 7d; cf. yTaan 4:1, 67d) and in the Bavli (bBer 27b–28a),
making it ideal for the demonstration of the relationship between parallel
sources in the two centers.
Here is the outline of the story as it is told in the Bavli (some interpretations
having been inserted in the summarizing process):

1. In the form of a tannaitic baraita in Hebrew: Rabban Gamaliel and Rabbi

Yehoshua differ on whether the evening prayer is obligatory or not. When
this question is addressed publicly in the study house, Rabbi Yehoshua
does not openly oppose Rabban Gamaliel’s position. Nonetheless Rabban
Gamaliel openly puts Rabbi Yehoshua to shame by forcing him to stand

10 On this foundational event, see Gafni, Jews of Babylonia, 31, 256.

8 NIKOLSKY AND ILAN

while he himself sits and teaches. The people resent this and ask Hutspit
the translator to stop [translating].
2. In Aramaic: This attitude of Rabban Gamaliel catalyzes a rebellion and
his subsequent deposition. The position is then offered to Rabbi Eleazar
ben Azariah, since he is wise, rich, and a tenth generation descendant of
(the biblical) Ezra.
3. In a mix of Hebrew Aramaic, tannaitic and amoraic: As soon as Rabbi
Eleazar ben Azariah becomes the head of the Study House, the number
of students rises significantly. The change occurs because of Rabban
Gamaliel’s strict rules of entry into the study house, according to which
a student whose external and internal intentions are not equal is barred,
are lifted.
4. In Hebrew: Rabban Gamaliel attends the study house as well, and on that
day another halakhic debate arises between Rabban Gamaliel and Rabbi
Yehoshua, on whether or not a converted Ammonite is allowed to join
the congregation of Israel. Rabbi Yehoshua’s opinion—to embrace the
Ammonites—is accepted (cf. mYad 4:4).
5. In Aramaic: Rabban Gamaliel now decides to apologize to Rabbi
Yehoshua, and goes to his house for this purpose. After some discus-
sion, and when the honor of Rabban Gamaliel’s house is invoked, Rabbi
Yehoshua is appeased.
6. In Aramaic: First a messenger, and then Rabbi Yehoshua himself, come
to the rabbis in the study house to inform them (in a form of a riddle)
that Rabban Gamaliel and Rabbi Yehoshua have made peace, and that
Rabban Gamaliel will now regain his position.
7. In Hebrew: The rabbis, represented by Rabbi Aqiva, fear Rabban Gamaliel
and are reluctant to re-establish his authority, but consent when they
realize that Rabbi Yehoshua has indeed been appeased,
8. In Aramaic: Rabbi Eleazar ben Azariah willingly steps down from his
position; still, he is not altogether demoted, but is appointed to preach
once every four Shabbatot.
9. The Bavli adds here that the student (who started the whole sequence of
events rolling) was Rabbi Shimeon ben Yohai.

The story in the Yerushalmi is different. It is much shorter, lacking many of

the narrative details, but still presents a very lively and focused story. Some
of the differences are relatively minor; for instance the figure mentioned in
the Yerushalmi is Zenon the Cantor, whereas the Bavli refers to Hutzpit the
Translator. There are, however, some more significant points of difference:
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 9

•
In the Bavli Rabban Gamaliel himself provokes Rabbi Yehoshua by forcing
him to either enter into a debate or retract what he had said earlier; in the
Yerushalmi, while still putting Rabbi Yehoshua to shame, Rabban Gamaliel
is not the one to provoke Rabbi Yehoshua—he “only” instructs the student
to raise the question in the assembly house (‫בית הועד‬, not the study house,
‫ישיבה‬, as in the Bavli).
•
After the appointment of Rabbi Eleazar ben Azariah, in the Yerushalmi (but
not the Bavli) Rabbi Aqiva becomes jealous, pointing out the fact that Rabbi
Eleazar was chosen instead of him not because of his better scholarship, but
because of his lineage.
•
In the Bavli, Rabban Gamaliel and Rabbi Yehushua quarrel over the conver-
sion of an Ammonite, and later Rabban Gamaliel visits Rabbi Yeshoshua
in order to appease him. By contrast, in the Yerushalmi, the story of the
Ammonite comvert is missing altogether, and Rabban Gamaliel visits all his
students in order to appease them.
•
In the Yerushalmi it is Rabbi Eleazar ben Azariah who asks whether Rabban
Gamaliel and Rabbi Yehoshua have made up, and not Rabbi Aqiva, as in the
Bavli.11
•
In the Bavli an arrangement is described, according to which Rabban
Gamaliel and Rabbi Eleazar ben Azariah preach on alternating Shabbatot.
In the Yerushalmi, there is no such arrangement, and after being demoted,
Rabbi Eleazar ben Azariah is appointed head of the court.
•
This story will accompany us as we move through the various stages of
Bavli–Yerushalmi scholarship.

In the 19th and first half of the 20th century, scholars used the differences
between the Babylonian Talmud and its Palestinian Vorlage in order to explain
the latter. In other words, if the Palestinian text displayed difficulties, scholars
believed that the explanations supplied by the Bavli were the ones intended
by the original authors of the texts.12 In his article, “Status and Leadership in

11 The different role assigned to Rabbi Aqiva in the Bavli in comparison to the Yerushalmi
is interesting, but there is no room here to study it; apparently it has to do with the
Babylonian favoritism of Rabbi Aqiva, an issue that still has to be further studied.
12 On this story see e.g. H. Graetz, History of the Jews vol. 2 (Philadelphia 1956 [English trans-
lation—the original German was composed in the 1850s]) 338–40; J. H. Weiss, dor dor
ve-dorshav vol. 2 (Vilna 1904) 80–9 [Hebrew]; I. Halevy, dorot ha-rishonim vol. 2 (Vienna
1923) 313–39 [Hebrew]; G. Alon, The Jews in their Land in the Talmudic Age vol. 1 (Tel Aviv
1954) 319–22.
10 NIKOLSKY AND ILAN

the World of the Palestinian Sages,” E. E. Urbach discussed the story of the
deposition of Rabban Gamaliel in some detail. How he did this, and how he
used the various sources at his disposal demonstrates the premise that was
widely accepted by scholars up to the 1970s. First he states that

the stories concerning the clashes between Rabban Gamaliel and

R. Yehoshua ben Hananiah and R. Eliezer ben Hyrcanus testify to the
intensification of the struggle between divergent principles: The free-
dom of halakhic decision against the claim on the part of the Patriarch
to the right of supervision and organization, that is, the concept of the
Patriarch’s jurisdiction as national leader—as an authority independent
of the Sanhedrin—in opposition to the view of the Sages who did not
consider that the Patriarch enjoyed a special status, but regarded him as
the head of the Beth Din from which he derived his authority.13

If we analyze what Urbach wrote here we see first of all that he reads our story
together with other stories, both those told about Rabbi Yehoshua and Rabban
Gamaliel and those told about Rabbi Eliezer and Rabban Gamaliel, as deriving
from the same sort of source and as describing the same historical situation (in
this instance Urbach does not even cite the sources that support his assertion,
perhaps assuming that they are so well known that no reference is necessary).
In the next lines he cites Rabban Gamaliel’s attitude toward students based
on our story in the Bavli (“a student whose external and internal intentions
are not equal should not study”—no. 3 in the summary above) but instead
gives examples of confrontations between Rabban Gamaliel and Rabbi Aqiva,
citing three separate Tosefta traditions and one from the Yerushalmi. Then,
Urbach describes the actual deposition, not bothering to state whether he is
basing his description on the Bavli or on the Yerushalmi. Then he describes the
sage chosen to replace Rabban Gamaliel, based on the Bavli’s description that
he was both rich and wise and one of exalted lineage (unlike the Yerushalmi
description, which emphasizes lineage only—see no. 2 in the summary above)
without stating this clearly and without explaining why the Bavli’s description
should be preferred. Urbach continues to paraphrase the story of the deposi-
tion and reinstatement of Rabban Gamaliel, suggesting motives for this rever-
sal of policy (“A decisive factor in the matter, doubtless, was the consideration
that any weakening of the Patriarch’s position held the danger of weakening
the authority of the Beth Din and of the entire Centre [obviously the center at

13 E. E. Urbach, Collected Writings in Jewish Studies (Jerusalem 1999) 456, based on a Hebrew
article first published in 1966.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 11

Yavneh is intended—Eds.] which still lacked stability . . .” p. 457) and bolstering

his arguments on the basis of a mixture of Mishnah, Tosefta and Yerushalmi
passages. Only when he arrives at the end of the story does Urbach concede
that the Bavli and Yerushalmi do not quite agree one with the other. He writes:
“According to the Babylonian Talmud, R. Eleazar b. Azaryah continued to
lecture once every third Sabbath, and according to the Palestinian Talmud
he remained the Av Beth Din” (see in text above, under no. 8 and the sum-
mary of the Yerushalmi). However, he refrains from explaining why there is
a difference between the two sources, and to what institution or institutions
they are referring. He ends his short description of the event by stating that
Rabbi Eliezer refused to accept the settlement the rabbis reached with Rabban
Gamaliel and was, therefore, eventually excommunicated by the rabbis. In a
footnote, Urbach refers to a tradition in the Yerushalmi ( yMQ 3:1, 81c) and to
two in the Bavli (bBM 59b; bSanh 68a) which describe a conflict between Rabbi
Eliezer and Rabban Gamaliel, without explaining the difference between them
or without explaining why he thinks that, chronologically, the events they
describe post-dated the deposition of Rabban Gamaliel that he read in other
sources embedded in the same compositions—the Bavli and the Yerushalmi.
Little has to be added to this description to make it abundantly clear that
scholars do not read rabbinic texts like this anymore.14 Only six years after this
article by Urbach was published, a completely different treatment of these par-
allel traditions was published by Robert Goldenberg in the Journal of Jewish
Studies. In order to understand the revolution that was underway during these
years, and which eventually produced Goldenberg’s article, we must make a
small detour.
The father of this revolution was Jacob Neusner. Between his book A Life
of Rabban Yohanan ben Zakkai (Leiden 1962), which was a conventional rab-
binic biography, telling the history of a sage based on stories told of him in a
jumble of rabbinic source, and his other book on the same sage, Development
of a Legend (Leiden 1970), in which he denied any possibility to write such a
biography, Neusner broke faith with all he had known about how one reads
rabbinic texts. It was Neusner who firmly informed us that rabbinic traditions
had authors, audiences and a Sitz im Leben, and that they are hardly a faithful
historical record of events they purport to report. It was he who taught us that
the Bavli does not fill in where the Yerushalmi is unclear, for there is no rea-
son that its authors knew better than we do what the Yerushalmi had omitted

14 See also L. Ginzberg, A Commentary on the Palestinian Talmud vol. 3 (New York 1971) 174–
220 [Hebrew], who interprets the story in the Yerushalmi completely in light of the Bavli.
12 NIKOLSKY AND ILAN

or corrupted.15 Baraitot—i.e. earlier authoritative Palestinian traditions

embedded in the Bavli—according to Neusner are, unless proven otherwise,
most probably pseudepigraphic Babylonian compositions. Parallel tradi-
tions should be studied side by side synoptically, not in order to see how they
complement one another, but rather in order to mark the differences between
them. Synopsis, a tool taken over from Christian Gospel studies, became a key
word in Neusner’s studies.
Goldenberg’s article is all about synopsis. He first presents the two stories
about the deposition of Rabban Gamaliel one after the other, highlighting their
literary character. About the Yerushalmi story he concludes that, aside from
interpolations and additions, it is probably composed of two original strands—
one that reported events associated with Rabban Gamaliel and Rabbi Yehoshua
(nos. 1 and 5 in the summary of the Bavli above), and one that reported events
associated with Rabban Gamaliel and Rabbi Eleazar ben Azariah (nos. 2, 3 and
8 in the Bavli summary above). This conclusion is reached because nowhere in
the entire text is there any interaction between Rabbi Eleazar ben Azariah and
Rabbi Yehoshua, and Goldenberg claims that each story could, under certain
circumstances, stand alone. He also distinguished between the languages used
in the composition, singling Aramaic out as the redactor’s language, and texts
composed in it as late. Thus he concludes regarding the chronological context
of the Yerushalmi version that

the central interest of the account is political. It revolves around the pres-
idency over the Rabbinic gathering and who might legitimately exercise
it . . . From the time of Rabbi Judah the Patriarch, the hold of the Hillelite
dynasty on the Patriarchate was secure. Political issues would thereafter
have naturally shifted to other issues. The main story seems therefore to
antedate the ascendancy of Judah the Patriarch ca. 185 C.E.16

Next he presents the Bavli version, characterizing it by stating that “here

the Rabbis are described as acting exclusively in Joshua’s interest [no. 7 in
the above text—Eds.] . . . This change reduces the political aspect of the nar-
rative, and turns it into a story about a personal dispute.”17 The bulk of the

15 Actually this is an extreme formulation. For a recent attempt to explain a difficult word
in the Yerushalmi in light of its Bavli parallel, see A. Amit, “On the Contribution of the
Comparison of the BT and the PT to Lexicography: ‫ אגיסטון‬and ‫האסטו‬,” Lešonenu 72
(2010) 135–53 [Hebrew].
16 R. Goldenberg, “The Deposition of Rabban Gamaliel II: An Examination of the Sources,”
Journal of Jewish Studies 23 (1972) 171.
17 Goldenberg, “Deposition,” 175.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 13

article, however, is the synopsis. He divides the story into several parts and
discusses the Yerushalmi and Bavli versions. His discussion of the first part is
very promising, since he begins by stating categorically: “These tables leave no
doubt as to show how the . . . sources are related. In only one case . . . is it even
remotely possible that the PT is an expansion of the BT.”18 This claim suggests
that the author wishes to show us in which direction the story was altered and
argues that the Bavli reworked the Yerushalmi. This, in the final analysis is the
direction research of parallels has gone since Neusner’s days. However, the rest
of the article somehow melts into insignificant, self-evident differences, and
when a conclusion is sought, more in the realm of literary and source criticism
than in any historical context, we learn that in fact the scholar wishes to show
us what he considers the “historical kernel” of the story. He concludes:

Taking the whole evidence into consideration, we may conclude that

some serious disturbance interrupted the period of Raban Gamaliel’s
leadership. This conclusion is supported by the fact that the three cen­tral
characters—the Hillelite Gamaliel, the priest Eleazar and Yohanan ben
Zakkai’s disciple Joshua—represent three of the major political factions
of the early post-Destruction period. It is highly plausible that an intense
power-struggle should have revolved around these men.19

One may well ask if analyzing the Bavli version of the story has added anything
to this conclusion. Certainly, nothing is stated here about the significance of
Babylonia as the location where the alterations were carried out.
The reason why Neusner, who was a declared historian of Jewish Babylonia,
having produced a five-volume study of it,20 and his student Goldenberg were
not able to utilize their analysis of the texts to arrive at meaningful conclu-
sions on the reasons for the differences between the two versions, is the fact
that Neusner was a student of the historical school whose methodological
approach was described above. Neusner was able to see through the fallacies
of the methodology he had been taught, and abandon it, but was not yet able
to reconstruct something historically meaningful instead.21

18 Goldenberg, “Deposition,” 179.

19 Goldenberg, “Deposition,” 190.
20 J. Neusner, A History of the Jews of Babylonia (Leiden 1965–70; 5 volumes).
21 Another of Neusner’s students who wrote on this episode is David Goodblatt. See
D. Goodblatt, The Monarchic Principle (Tübingen 1994) 253, though he only referred to it
as part of a general discussion of the leadership in Palestine, and uses this story to dismiss
the Babylonian evidence.
14 NIKOLSKY AND ILAN

Aside from Neusner’s synoptic approach, which undermined the way his-
torians had read these texts, the 1960s and 70s saw an attack on the historical
approach from another angle. The chief representative of this approach was
the literary scholar, Yonah Fraenkel. This scholar argued that, regardless of who
the rabbis were, about whom rabbinic stories are told, or the supposed histori-
cal background against which they are told, the rabbis who composed rabbinic
literature never wrote history. They wrote literature. They created small literary
units which are stylistically and linguistically formulated, each self-contained
in a world of its own. And this world is that of the rabbinic academy. The rabbis
told these stories about themselves for themselves.22 Reading them one with
the other, or against a common background, is to do damage to the literary
integrity of the text. Fraenkel would rather ignore the existence of parallels
and read each story as a separate composition. Or as put by Hillel Newman:
“While Fraenkel nowhere explicitly objects to the trend of current scholarship
to seek out and underscore the social and cultural distinctions between the
Jews of Babylonia and Palestine as reflected in the stories of the two talmu-
dim . . . neither does he express much interest in such inquiry. This is prob-
ably a result . . . of his methodological insistence on ‘closure’ of the rabbinic
story.”23 Time and again Fraenkel demonstrated his methodology, but he never
employed the story of the deposition of Rabban Gamaliel for this purpose,
even though there is hardly a more fitting composition against which to test
the rabbis’ attitude to their study house and the relations therein.
Not so long ago (2004), however, a student of Fraenkel, J. Klikstein, pub-
lished an article on this episode from a purely literary point of view.24 His
approach demonstrates nicely the kind of message Fraenkel instilled in his
students, for he begins by stating: “From a historical point of view it is possible
to derive much material on the spiritual—and perhaps even political—lead-
ership in the period after the destruction of the Temple . . . The existence of
the story in the Yerushalmi has encouraged scholars to attempt tracing cus-
toms of Babylonia and the Land of Israel concerning the laws of the study

22 We use Hillel I. Newman’s review article on Fraenkel to describe his work, see H. I.
Newman, “Closing the Circle: Yonah Fraenkel, the Talmudic Story and Rabbinic History,”
in: M. Kraus (ed.), How Should Rabbinic Literature be Read in the Modern World?
(Piscataway NJ 2006) 118.
23 Newman, “Closing the Circle,” 112. For an example of how Fraenkel discusses parallel sto-
ries, in order to argue that the fact that they are parallels is literarily unimportant, see J.
Fraenkel, “Hermeneutic Problems in the Study of the Aggadic Narrative,” Tarbiz 47 (1978)
139–72 [Hebrew], esp. pp. 146–57.
24 J. Klikstein, “The Deposition of Rabban Gamaliel from his Presidency,” Shma‘atin 156
(2004) 110–34 [Hebrew].
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 15

house, as reflected in the analysis and comparison between the two stories.”25
Immediately following these words, however, the author assures us that this
is not his scholarly project, for he writes: “In this article we do not deal with
the historical aspects. We shall analyze this aggadah in order to learn what
lessons it was bent on teaching. This study will be based on an analysis of the
structure of the story and its composition.” Klikstein’s loyalty to Fraenkel’s les-
son begins by his refusal even to look at the Yerushalmi version, while analyz-
ing its Bavli counterpart. The scholar shows compositional elements found in
the story, based on linguistic repetitions (like the use of ‫“—עמוד‬stand”—both
for Rabban Gamaliel’s punishment of Rabbi Yehoshua and for the command-
ment to Hutspit the Translator to cease translating—both under no. 1 in the
text above), and explains how each part of the story functions together with
the other (for example, how the first part of the story describes the study house
of Rabban Gamaliel [nos. 1 and 2 in the Bavli summary above] and the second
part, that of Rabbi Eleazar ben Azariah [nos. 3 and 4 in the Bavli summary
above] and how they can be compared one to the other, almost element for
element.26 This sort of literary analysis is employed in order to present the
psychology of the characters involved and to convey a clear didactic message.
The story is told as a whole, in the words of Klikstein, in order to demonstrate
that “paradoxical as it may appear, it was precisely on the day that all those
students ‘whose external and internal intentions are not equal’ were allowed
in that the best study results were achieved . . . It follows that the results pro-
duced by the study house are not a direct function of the quality and level of
students. The number of students has its own power, which is translated into
academic results. The good students were not required to compromise as a
result of integration in the study house. On the contrary, the added value for
the study house as a result of the opening of its doors to all students had a good
influence on the choice ones.”27
Yet, as Newman showed regarding Fraenkel himself,28 Klikstein too assumes
some a priori historical information acknowledged by the reader when
approaching this story. Thus he writes: “We are dealing with Rabban Gamaliel
II, namely Rabban Gamaliel of Yavneh. The people of Israel have just under-
gone a traumatic crisis—the destruction of the Temple . . . [T]he Sanhedrin
which instructed all of Israel goes into a series of ten exiles and loses its status

25 Klikstein, “Deposition,” 110.

26 Klikstein, “Deposition,” 118, 128.
27 Klikstein, “Deposition,” 129.
28 Newman, “Closing the Circle,” 121.
16 NIKOLSKY AND ILAN

and authority . . .”29 In other words, despite his disavowal of any interest in the
historical value of the story, Klikstein takes a certain historical background for
granted. This is important because it shows how, even when doing their best
to avoid any historical reference, the truth is that scholars of literature cannot
really do without it. Yet, unlike the discerning historian, the literary critic takes
for granted historical information (like the existence of a Sanhedrin) that is
neither mentioned in the story, nor accepted unquestioningly by historians.
Thus, closest to the goal of the present volume comes the article published
more than a decade ago by the historian Haim Shapira. Dealing with the very
story we have been discussing, Shapira, using both the synoptic and literary
approaches, deconstructs the two versions of the story, inquires where, when,
and why they were composed, and explains the differences between the
Yerushalmi and the Bavli versions on the basis of their different historical and
cultural backgrounds.
Shapira begins by stating categorically that the story is not tannaitic. It con-
tains references to tannaitic stories and tannaitic sources (the ones described
above as in Hebrew and tannaitic), but it does not, in and of itself, contain a
tannaitic kernel, and therefore reflects no specific historical event that took
place in the tannaitic period.30 Shapira claims that the focus of the Yerushalmi
version of the story is the political question whether a Nasi of the House of
Hillel may be deposed, and the negative answer provided suggests a time when
the patriarchate was firmly established, while being strongly criticized by the
rabbis. He suggests the third century, under the sage Rabbi Yohanan and the
Patriarch Rabbi Yudan Nasia, as just such a time, and claims this would be
the period when such a story would have been told.31
When he turns to the Bavli version, Shapira claims that the story the
Babylonian editors had before their eyes was similar to the one we find in the
Yerushalmi and that the differences between them derive from one of three
sorts of alterations the editors made: (1) alterations based on literary consider-
ations; (2) additional tannaitic sources; (3) alterations made based on the influ-
ence of a Babylonian context.32 Since the last of the three is the one that most
interests us, let us observe what Shapira understood as Babylonian influence.
First, he identifies the leadership role that is disputed in the text as described
differently from its Palestinian counterpart. While in the Yerushalmi the lead-
ership is political, the Babylonians present it as the position of head of the

29 Klikstein, “Deposition,” 122.

30 Shapira, “Deposition,” 13–4.
31 Shapira, “Deposition,” 19–23.
32 Shapira, “Deposition,” 30.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 17

yeshivah (“Would the master accept to become head of the yeshivah?”—no. 2

in the Bavli summary above).33 Secondly, Shapira claims that the fear Rabbi
Aqiva expresses of Rabban Gamaliel’s servants (no. 7 in the text above) reflects
the relationship between the exilarch and the rabbis in Babylonia, rather than
anything Palestinian.34 Thirdly, Shapira interprets the different outcome of the
story as deriving from the Babylonian setting of the yeshivah (no. 8 in the sum-
mary above and the Yerushalmi summary). The preaching that Eleazar Ben
Azariah is allowed once every three weeks is the famous Babylonian pirqa in
their yeshivot.35 Finally, Shapira points to the image of the doorkeeper, and
of the limited accessibility to the study house (no. 3, in the Bavli summary
above: “A student whose external and internal intentions are not equal should
not enter the study house”) as typically Babylonian, the latter statement even
assigned to Rava elsewhere in the Bavli (bYom 72b). He concludes that the
Babylonian story displays Babylonian rabbis and their concerns vis à vis the
head of the yeshivah in the 4th century.
Jeffrey Rubenstein, in his book The Culture of the Babylonian Talmud, blends
his discussion of this story into a chapter on the elitism of the Babylonian rab-
bis. He does not conduct a long analysis of this story and its comparison with
the Yerushalmi version, for in this book, only the end-product, as it appears in
the Bavli, is of interest to him, and he uses what he concluded from this story,
together with the conclusions derived from many other stories, in order to
draw a picture of Babylonian rabbinic society, based on clearly Babylonian pre-
sentations of themselves, and on the Babylonian changes that were introduced
into Palestinian stories. His conclusion in this chapter is: “The sages perhaps
perceived their academic world of Torah study as increasingly professional-
ized, elitist, and isolated from the general population. As a result, non-rabbis
outside of the academy were viewed as ‘Others’ and even included with other
categories of ‘Others’—slaves, gentiles, and animals.”36 Yet in his discussion of
our story he basically accepts Shapira’s observations, for he writes: “The story
thus contains an internal debate concerning the merits of limiting access to the
academy. While the storytellers ultimately reject Rabban Gamaliel’s restrictive
policy, the internal debate and the motivation for the story suggest that some
sages—perhaps the leadership of the academy itself—thought that the acad-
emy should be closed to outsiders.”37 That Rubenstein agrees with Shapira’s

33 Shapira, “Deposition,” 31.

34 Shapira, “Deposition,” 31–2.
35 Shapira, “Deposition,” 32–3.
36 Rubenstein, Culture of the Babylonian Talmud, 141.
37 Rubenstein, Culture of the Babylonian Talmud, 140.
18 NIKOLSKY AND ILAN

conclusions is significant, because this story actually contradicts the thesis he

is promoting in this chapter.
Elsewhere, Rubenstein acknowledges his debt to Fraenkel’s enormous con-
tribution to the literary study of talmudic stories by stating: “Fraenkel must be
acclaimed for the scope of his work, the depth of his analysis, and his success
in defining and describing the literary characteristics of the rabbinic story.”38
Yet several pages later he goes beyond Fraenkel’s approach by stating: “Stories
embedded in the BT should be considered in relation to talmudic halakhah in
general, as well as with the particular halakhot to which they are juxtaposed—
which brings us back to the immediate literary context. These considerations
contest the view of talmudic stories as closed and self-contained texts.”39 In his
first book, Talmudic Stories, Rubenstein analyzed talmudic stories according
to the principles outlined by Fraenkel, while also adding the context element
just outlined. In his second book, The Culture of the Babylonian Talmud he ana-
lyzes our story. Here he goes a step further and uses the central messages he
identifies in the stories he analyzes in order to describe and assess the rabbinic
values and culture in which these stories were composed. Thus, from a pure
literary critic, Rubenstein has become a cultural historian.
From our point of view, what is interesting to note is that all three major
analyses of the story we are following since the 1990s have come to the same
conclusions about its message. Klikstein, the literary critique, concluded that
the message of the story is didactic. He thinks the story promoted egalitarian
education, criticizing elitism. Shapira, the historian, agreed, claiming that this
is a reflection of a historical debate in the Babylonian academies of the fourth
century. Even Rubenstein, whose main thesis is that the Babylonian rabbis
who told these stories (the stammaim of the 5th–6th centuries in his opinion)
were elitists, who thought the academy should not allow within its walls “A
student whose external and internal intentions are not equal,” agrees that our
story’s main message is a criticism of this approach. All that is left is to agree
with Newman’s assessment that “our two heroes, the historian and the literary
critic, had difficulty finding a common language. Yet increasingly the method-
ological boundaries separating the two disciplines are becoming less distinct.
Historians are conceding the significance of the literary components of rab-
binic literature; literary criticism is rediscovering the importance of context,
including historical context.”40

38 Jeffrey L. Rubenstein, Talmudic Stories: Narrative Art, Composition and Structure

(Baltimore 1999) 10.
39 Rubenstein, Talmudic Stories, 15.
40 Newman, “Closing the Circle,” 133.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 19

4 Post-Amoraic Contact between Palestine and Babylonia

The Babylonian acceptance of Palestinian rabbinic authority lasted through-

out the amoraic period. Yet, as we have already intimated, the two centers were
different to begin with, regarding customs, lifestyles, even language. During the
amoraic period, when constant contact between the centers was maintained,
an effort was invested in keeping halakhah in both centers common—not nec-
essarily unified, but accommodating both its Palestinian and its Babylonian
versions. Throughout the fifth century, however, communication between
the two centers became increasingly sporadic, and both centers continued to
develop independently.
The Jews in the Land of Israel suffered greatly under the pagan Romans
and even more under the decrees of the Christian Byzantine authorities. After
the death of Rabban Gamaliel VI, the office of Nasi was abolished (425 CE).
Nonetheless, Jewish culture continued to flourish in Palestine, as is apparent
from archeological evidence (synagogues and amulets) and literary activity
(aggadah, piyyut, apocalyptic and hekhalot literature, as well as halakhah).41
While the second half of the first millennium witnessed a huge upsurge of
aggadah and piyyut, less is known about the development of the halakhah of
Eretz Israel after the Yerushalmi was edited. It is obvious that halakhic activity
did take place there, especially in Tiberias, but Eretz Israel suffered a decline,
relative to its dominance in the first half of the first millennium.42
In Babylonia, with the exception of magic bowls, the only literary and intel-
lectual source known to us today from the end of Late Antiquity is the Bavli,
which in spite of being of an encyclopedic nature, is oriented toward halakhah.
Based on David Weiss-Halivni’s intensive study of the Bavli, it is now a
scholarly consensus that its main text was created by the stammaim (mean-
ing “anonymous”), a term coined by Halivni for the authors of the anonymous
give-and-take added to discussions of the amoraim. According to Halivni, the
early stammaim (mid-fifth century) could still reconstruct the original hal-
akhic give-and-take, based on remembered oral traditions. Toward the end of
the period (8th century) stammaitic editorial work was founded more on the
initiative of these scholars. Halivni goes on to argue that by the end of the
talmudic period, the savoraim, the very last rabbis of the stammaitic period,

41 See now H. I. Newman, The Ma‘asim of the People of the Land of Israel: Halakhah and
History in Byzantine Palestine (Jerusalem 2011) 1–119 [Hebrew].
42 Newman, Ma‘asim, 10.
20 NIKOLSKY AND ILAN

added short editorial remarks to the text of the Bavli, which was by then almost
in its final form. The very last layer of the Bavli was added in the Gaonic period.43
The situation with regard to the aggadic material in the Bavli is slightly dif-
ferent. The aggadah is less authoritative, less accurate in its transmission, not
always attributed to a rabbi even in the early periods, and not limited to one
inference from one verse;44 much aggadah was also written earlier than the
talmudic text of the Bavli.45
The period during which contact was scarce between the centers in Babylo-
nian and the Land of Israel lasted at the most 200 years (5th to 7th centuries).
Although the halakhah was basically the same in the two Jewish cultures, the
differences that developed in this period are highly meaningful because of the
all-encompassing and exclusive nature of halakhah. Palestinian halakhah was
prominent, aside from the Land of Israel, also in other parts of the Byzantine
Empire, and later in Ashkenaz.46 In the 10th–12th centuries, the Bavli became
prominent not just in the Muslim countries but also in the Jewish communi-
ties of Europe. The image of the halakhically weak Palestinian sages came into
vogue as part of the promotion of the Bavli.47
But what about the Land of Israel itself? Scholars usually maintain, quite
vehemently, that Jewish literature from Eretz Israel does not exhibit any
influence of the Babylonian culture or of the Bavli.48 This attitude is sur-
prising, since some contact between the two centers did exist throughout
the period under discussion, and the wave of Babylonian immigrants to the
Land of Israel increased dramatically once it was conquered by the Muslims

43 D. Halivni, Introduction to “Sources and Traditions”: Studies in the Formation of the Talmud
(Jerusalem 2009) 3–8 [Hebrew].
44 On this concept see the contribution to the present volume by D. Brodsky, pp. 188–250.
45 Halivni, Introduction, 141–3.
46 See e.g. M. Gil, “Between Two Worlds: The Connections between Babylonia and the
Communities in Europe,” in: Daniel Carpi et al. (eds.), Shlomo Simonsohn Jubilee Volume:
Studies of the History of the Jews in the Middle Ages and Renaissance Period (Tel-Aviv 1993)
45–52 [Hebrew]; I. M. Ta-Shma, Studies in Medieval Rabbinic Literature, Volume 3: Italy
and Byzantium (Jerusalem 2005) 232; Newman, Ma‘asim, 19, 117; but see A. Grossman, The
Early Sages of Ashkenaz: Their Lives, Leadership and Works (900–1096) (Jerusalem 2001)
426–7 [Hebrew], who restricts this influence to Jews of Italian origin, in contrast to the
Jews in France who came from Spain.
47 R. Bonfil, “Between Eretz Israel and Babylonia: Contribution to the Study of the Culture of
the Jews in South Italy and in Christian Europe in the Early Middle-Ages,” Shalem: Studies
in the History of the Jews in Eretz Israel 5 (1987) 14; Ta-Shma, Studies, 231–7.
48 See Newman, Ma‘asim, 46 in the footnotes, referring to the scholarly discussions about
evidence of Babylonian influence on liturgy and Torah reading.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 21

(mid-7th century) and both centers were part of the same empire. We know
(from Genizah documents) that communities of rabbinic Babylonians were
present in various cities of Palestine.49
It seems that research has not yet begun discussing the question of Babylo-
nian influence on Palestinian Jewish culture of the Byzantine period. Most of
the research is still busy identifying and defining the unique Palestinian cul-
ture which had been hidden away until the discovery of the Genizah.50 Thus,
when Babylonian influence is recognized, it is usually discarded as a late inter-
polation. The present volume suggests, however, that it is now the right time
to begin taking this topic more seriously. Historically, we should probably talk
about Palestinian resistance to Babylonian culture more than about a lack of
acquaintance with it, and Babylonian influences can certainly be clearly iden-
tified. Some of the contributions in this book begin to do exactly this.

5 This Book

5.1 Internal Development

Jeffrey Rubenstein is one of the recent scholars whose approach to the Bavli
and to its reworking of Palestinian sources is the same as that adopted by sev-
eral of our contributors to this volume.51 His approach—which suggests that

49 J. Mann, The Jews in Egypt and Palestine under the Fatimid Caliphs: A Contribution to their
Political and Communal History Based Chiefly on Genizah Material hitherto Unpublished
(London etc. 1920) 148–50. Mann states (p. 171): “That Palestine had a considerable num-
ber of Babylonian Jews already in the eighth century can be gathered from an interesting
Halakhic fragment in Geonica II (see J.Q.R., N.S., VII, 474). We have also seen above (p. 148)
that Ramlah possessed a ‫כנסת אלשאמיין‬, presupposing a ‘synagogue of the “Babylonians”
with s at the end.’ Also Tiberias probably had two such congregations and likewise
Damascus (above, pp. 150, 167).” See also: M. Margalioth, The Halakha of the Land of Israel
from the Geniza, with Introductions, Notes, Commentaries, and Reference, Compiled by I. M.
Ta-Shma (Jerusalem 1973) 8; A. Grossman, The Early Sages 433; idem, “The Yeshiva of Eretz
Israel, its Spiritual Work and the Affinity to It,” in: Y. Prawer (ed.), The History of Jerusalem,
Early Islamic Period (638–1099) (Jerusalem 1987) 191.
50 Newman, Ma‘asim, 9.
51 See also a representative sample: S. Safrai, “The Tales of the Sages in Palestinian Tradition
and the Babylonian Talmud,” in: J. Heinemann and D. Noy (eds.), Studies in Aggadah
and Folk-Literature (Scripta Hierosolymitana 22; Jerusalem 1971) 209–32; D. Goodblatt,
“The Beruriah Traditions,” Journal of Jewish Studies 26 (1975) 68–85; S. J. D. Cohen, “The
Conversion of Antoninus,” in: P. Schäfer (ed.), The Talmud Yerushalmi and Greco-Roman
Culture vol. 1 (Tübingen 1997) 140–71; C. E. Hayes, Between Babylonian and Palestinian
Talmuds: Accounting for Halakhic Difference in Selected Sugyot from Tractate Avodah
22 NIKOLSKY AND ILAN

if a phenomenon is present often enough in the reworked Palestinian texts of

the Bavli, but absent from the same texts in their original setting, this reflects a
Babylonian phenomenon—serves as a starting point for the discussion of the
articles presented in this volume.
Shamma Friedman discusses a number of tannaitic traditions about magic
in which important tannaim are involved in studying the legal status of sorcer-
ers and the way to combat sorcery. He demonstrates from these stories, which
view magic negatively, that the Yerushalmi, and even more so the Bavli, trans-
form the rabbis involved first into experts in magic itself and then into power-
ful magicians. From this Friedman concludes that in Babylonia magic was to a
far greater extent incorporated into, and tolerated by, rabbinic culture than in
Palestine. This assertion is implicit in the words of the Bavli, which purports to
describe the Palestinian tanna, Rabbi Eliezer as a great magician.
Moshe Lavee shows how Palestinian traditions about conversion into
Judaism are systematically and tendentiously reworked in the Bavli so that it
is made more difficult (from a halakhic perspective), and at the same time is
looked upon negatively (in narratives on conversion). The Babylonians pres-
ent these alterations as authentic Palestinian halakhah. Lavee suggests that the
Babylonians here are implicitly referring to their own culture, in which they
tried to discourage conversion to Judaism.
There are only a handful of Palestinian sources that describe and dis-
cuss Babylonia and Babylonians. It is of special interest, therefore, to see
how these traditions are adopted and reworked by the Baby­lonians because
these are obviously cases where the latter know much more about the sub-
ject matter than their Palestinian counterparts. We may assume that when
they make alterations they are explicitly (in this case) presenting their view
of themselves. Geoffrey Herman discusses just such a tradition. In a story told
in the Yerushalmi about tannaim who travel to Arsacid Babylonia, the local
Babylonians are described as wicked and violent hosts. The Bavli chooses to
reproduce the Palestinian story, but in its version the Babylonians are substi-
tuted for Nehardeans, members of the exilarch’s court, transforming the story
into one about internal Babylonian strife, and the wicked Babylonians become
specific Babylonians who are wicked. The Babylonian point of view here is not

Zarah (Oxford 1997); R. Kalmin, The Sage in Jewish Society in Late Antiquity (New York
1999) 27–79 and see his other books as well; S. Friedman, “The Further Adventures of Rav
Kahana: Between Babylonia and Palestine,” in: P. Schäfer (ed.), The Talmud Yerushalmi
and Greco-Roman Culture vol. 3 (Tübingen 2002) 247–71; T. Ilan, “Stolen Water is Sweet:
Women and their Stories between Bavli and Yerushalmi” in: ibid., 185–223; Gray, Talmud
in Exile, esp. 199–234.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 23

necessarily historically more accurate, since the Palestinians tell of their con-
temporary Arsacid dynasty, and the Baby­lonians, who rework this story, live
under Sasanian domination. The Babylonians’ version of the story thus reflects
their own understanding of their own culture.
Martyrdom is the topic of the next contribution to this volume. Christiane
Tzuberi shows how the Bavli faithfully quotes the Palestinian halakhic sources
on martyrdom, which claim that a person should always die rather than trans-
gress commandments against bloodshed, sexual immorality, and idolatry, and
in times of religious persecutions, should die rather than transgress any com-
mandment. Tzuberi shows how the Bavli then elegantly overturns this hal-
akhah so as to claim that one may transgress any commandment rather than
die. They do this in a short midrash on the actions of Queen Esther, who saves
herself and her people by committing sexual transgression.
The book of Esther was indeed of major significance for the Babylonian
rabbis, as can be seen from the fact that it is the only biblical book on which
an entire verse-for-verse midrash was composed in the Bavli (bMeg 11a–17a),
which usually quotes most of its midrashim from collections of Palestinian
origin. The prominence of the Esther story in the eastern provinces is also evi-
dent from its appearance among the frescoes of the Dura Europos synagogue.52
This should come as no surprise, because the events recounted in the book of
Esther take place in a Persian setting under a Persian monarchy, a situation
similar to the one current in the Babylonian rabbis’ time and well known to
them.
“Heroizing” local figures in a diasporic setup is a known phenomenon in
Jewish history down to the present. It is found, for example, in the way Joseph
was “lionized” in Jewish Hellenistic Egypt, as is evident from books like Joseph
and Aseneth, the Testament of Joseph, Philo, and a composition quoted by
Josephus.53 More recent examples may include the way Moses Mendelssohn
is admired and idolized among German Jews in Berlin or Spinoza among
Dutch Jews to this day.54 Perhaps more relevant in this context is the way
the Palestinian Talmud was revived and revered among scholars of Talmud

52 In contrast to Qumran, where Esther is not found at all. On the Dura Europos frescoes
see e.g. K. Weitzmann and H. L. Kessler, The Frescoes of the Dura Synagogue and Christian
Art (Washington 1990) 114–7.
53 On the last two see e.g. M. Niehoff, The Figure of Joseph in Post-Biblical Jewish Literature
(Leiden 1992).
54 The editors of this volume are the source for this observation. Ronit Nikolsky teaches
Jewish Studies in Groningen, Netherlands, and Tal Ilan teaches Jewish Studies in Berlin,
Germany.
24 NIKOLSKY AND ILAN

in the modern State of Israel. Thus, scholars like Saul Lieberman and Yaacov
Sussmann devoted their major academic output to this composition and its
superior linguistic and literary qualities.55 Joshua Efron even declared it a his-
torical source far superior to the Bavli.56
The use of the book of Esther is a definite marker of a Babylonian point
of view, and therefore it is not surprising to find it used to counter the
Palestinian halakhah on martyrdom as demonstrated in Tzuberi’s article. The
book of Esther is also the focus of Amram Tropper’s article. Tropper reads two
Palestinian traditions and their transformation in the Bavli. Although the book
of Esther is mentioned already in the tannaitic source (tSot 7:7), its use is so
drastically transformed in the Bavli, that with it, the Babylonians deny any
validity to the Torah before it was willingly accepted by the Jews in the time of
Mordechai and Esther, presumably in the Babylonian–Persian diaspora. Thus,
Tropper shows how one of the most defining moments in Jewish history—
receiving the Torah on Mount Sinai—is downgraded in favor of another defin-
ing moment that takes place in the eastern diaspora.
The book of Esther is not the only biblical text that the Babylonians pro-
moted because of its Babylonian setting; it is also true regarding other biblical
accounts that take place in Mesopotamia. The Baby­lonians endow their place
of residence with meaning derived from the Jewish cultural canon, first and
foremost the Bible. For example, they see themselves as residents in the place
of origin of Abraham, the father of the Jewish people. Thus in bPes 87b, as one
among three answers to the question “Why did God exile the Jews to Babylonia,
of all places?,” we read: “Said Rabbi Yohanan: Because He sent them [back] to
their mother’s house. This is like a person who becomes angry with his wife. To
where does he banish her? To her mother’s house.”57 Here we see the rabbis’
understanding of Babylonia as the birthplace and cradle of the Jewish people.
In addition, the Babylonians identify their homeland as the landscape in
which human beings first trod the earth—the Garden of Eden. In Gen 2:14, one

55 E.g. Y. Sussmann, “pirqei yerushalmi,” in: M. Bar-Asher and D. Rosenthal (eds), Mehqarei
Talmud: Talmudic Studies Dedicated to the Memory of Professor Eliezer Shimshon Rosenthal
vol. 2 (Jerusalem 1993) 220–77, and esp. “Before and After the Leiden Manuscript of
the Talmud Yerushalmi” Bar Ilan 26–7 (1995) 203–20 [both in Hebrew]; S. Lieberman,
ha-yerushalmi kiphshuto: A Commentary (New York 1935), and other studies by these two
prominent scholars.
56 See J. Efron, Studies on the Hasmonean Period (Leiden 1987) 143–218.
57 And on this text see also J. L. Rubenstein, “Addressing the Attributes of the Land of
Israel: An Analysis of Bavli Ketubbot 110b–112a,” in: I. M. Gafni (ed.), Center and Diaspora:
The Land of Israel and the Diaspora in the Second Temple, Mishnah and Talmud Periods
(Jerusalem 2004) 159–88 [Hebrew].
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 25

reads that two of the rivers watering the Garden of Eden are the Euphrates and
the Tigris. This is the exact geographical location in which the main centers
of the Babylonian diaspora were situated. In her essay, Tal Ilan discusses the
Babylonians’ local patriotism which identifies their residence as the Garden
of Eden. She shows how, when manipulating a text from the Yerushalmi
that showers scorn on wicked Babylonia, the Babylonian rabbis claim, based
on explicit biblical verses, that they live at the entrance of Paradise, while
claiming at the same time, that the Jews of Jerusalem live at the entrance of
Hell (Gehenna).

5.2 External Influence

By the end of the 19th and throughout the 20th century, rabbinic literature
has been read side by side with classical Graeco-Roman texts. This move
placed rabbinic literature in its wider cultural context, since it was composed
in Hellenistic and later Roman-dominated Palestine. Historians of rabbinics
tended to study Greek and Latin alongside their immersion in rabbinic lit-
erature. The foremost representative of this approach was undoubtedly Saul
Lieberman.58 Lieberman demonstrated the rabbinic acquaintance with the
Graeco-Roman culture, and the penetration of this culture into rabbinic litera-
ture. Many followed in his footsteps, and the presentation of parallels between
Graeco-Roman and rabbinic texts is still widely practiced.59 This methodology
proved very productive in studying Palestinian sources. However, it is problem-
atic when applied in the same vein to the Babylonian Talmud.
So what happens when a Palestinian text, heavily influenced by Graeco-
Roman culture, is adopted by the Babylonians? Obviously many such texts
made this journey from Palestine to Babylonia. David Brodsky’s essay addresses
this question on a grand scale. He searches for the origins of the rabbinic give-
and-take (shakla ve-tarya) in the Babylonian sugya. This is perhaps the most
defining literary character­istic of the Bavli, found, practically, on every one of
its pages. Scholars have claimed that this genre is the most important literary
invention of the Bavli.60 Brodsky convincingly demonstrates not only that this

58 See S. Lieberman, Greek in Jewish Palestine (New York 1942), and Hellenism in Jewish
Palestine (New York 1950).
59 E.g. D. Sperber, Essays on Greek and Latin in the Mishna, Talmud and Midrashic Literature
(Jerusalem 1982); E. Halevy-Epstein, The World of the Aggadah in Light of Greek Sources
(Tel Aviv 1972) [Hebrew]. These are two prominent examples, and there are many others.
60 See A. Goldberg, “The Babylonian Talmud,” in: S. Safrai and P. J. Tomson (eds.), The
Literature of the Sage, First Part: Oral Tora, Halakha, Mishna, Tosefta, Talmud, External
Tractates (Assen 1987) 337–9 for a discussion and literature.
26 NIKOLSKY AND ILAN

genre is to be found in Palestinian halakhic midrashim and in the Yerushalmi,

but that it was deeply influenced by contemporary Hellenistic rhetorical prac-
tices. In this he is clearly following Lieberman’s scholarly heritage. But Brodsky
goes further, to demonstrate how the Bavli develops and appropriates this
rhetorical tool in a unique manner. What started as an external influence on
the Palestinian rabbinic scholastic culture, ended as an internal development,
characterizing the Babylonian one.
Recently, Lieberman’s model of studying Palestinian rabbinic literature
in light of Greek and Latin sources has been copied by scholars onto the
Babylonian milieu. Just as the Palestinian rabbinic culture is embedded in a
Graeco-Roman context, and just as it proved methodologically sound to resort
to studying the parallels between the two, so too we must assume that the
Babylonian rabbinic culture is embedded in the Sasanian milieu, and must
have been deeply influenced by it. It is, therefore, unsound to assume that
the Graeco-Roman backdrop is a priori the correct one against which to study
the Bavli,61 and indeed, immersing oneself in the study of Middle-Iranian lan-
guages and literature can be most rewarding.62 A new school of scholarship is
steadily developing in this direction, under the visionary guidance of Yaakov
Elman, who has been forcefully promoting the study of Middle-Persian and its
literatures among the students of rabbinics in the last decade.63

61 We should stay open to the possibility of Graeco-Roman influence originating in the

Christian academies of Syria. This influence increased in the academy of Nisibis as of
the middle of the 6th century. See Rubenstein, Culture of the Babylonian Talmud, 35–8;
G. J. Reinik, “ ‘Edessa Grew Dim and Nisibis Shone Forth’: The School of Nisibis at the
Transition of the Sixth-Seventh Century,” in: H. J. W. Drijvers and A. MacDonald (eds.),
Centres of Learning: Learning and Location in Pre-Modern Europe and the Near East
(Leiden 1995) 77–89; A. Becker, Fear of God and the Beginning of Wisdom: The School of
Nisibis and Christian Scholastic Culture in Late Antique Mesopotamia (Philadelphia 2006).
62 It should, however, be noted that the major part of this literature was put down in writing
a century or two later than the stammaitic layer of the Bavli, see M. Macuch, “Pahlavi
Literature,” in: R. E. Emmerick and M. Macuch (eds.), The Literature of Pre-Islamic Iran
(History of Persian Literature 17; London 2009) 116–96, especially 116–21.
63 Elman wrote many articles on the topic; see e.g. Y. Elman, “Acculturation to Elite Persian
Norms and Modes of Thought in the Babylonian Jewish Community of Late Antiquity,”
in: E. B. Halivni and Z. A. Steinfeld (eds), Neti’ot Le-David: Jubilee Volume for David Weiss
Halivni (Jerusalem 2004) 31–56; “The Other in the Mirror: Iranians and Jews View One
Another: Questions of Identity, Conversion and Exogamy in the Fifth-Century Iranian
Empire,” Bulletin of the Asia Institute 19 (2005) 15–25; “Who are the Kings of East and
West in Ber 7a?: Roman Religion, Syrian Gods and Zoroastrianism in the Babylonian
Talmud,” in: S. J. D. Cohen and S. Schwartz (eds.), Studies in Josephus and the Varieties
of Ancient Judaism (Leiden 2007) 43–80; “Middle Persian Culture and Babylonian
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 27

While it is still too early to assess the impact of this scholarly approach, in
this volume too there are contributions that demonstrate how its application
to traditions that originated in Palestine enriches our insights into Babylonian
rabbinic culture. In his article, Reuven Kiper­wasser discusses a Palestinian
midrash, which, when transferred to Babylonia, while keeping its Palestinian
theological message, acquired some enigmatic additions. Kiperwasser shows
that these additions are best understood when read against the background
of Iranian myth. Kiperwasser’s study presents a Palestinian tradition which
is highly embellished by the Babylonian rabbis, reflecting their own culture.
In this case, however, unlike in the other studies in this volume, we can see
unambiguously the influence of the external Iranian culture on the formation
of the story.
And yet, the study of parallels between Zoroastrian texts and the Bavli is not
in itself as straightforward as one would expect. In other words, not every par-
allel between the two points to Zoroastrian influence. Yaakov Elman’s article in
the present volume is an example of this complexity. He shows that in the field
of fraud legislation the moral concerns shown by the Bavli are similar to those
of Zoroastrian law; both protecting the victim of fraud. On the face of it one
could assume that the Bavli is here influenced by Zoroastrian law; however,
these same concerns for the weaker party in a case of fraud are already voiced
in the Mishnah, and it is doubtful whether this can be assigned to Zoroastrian
influence (not just because the rabbis of the Mishnah acted beyond the pale of
Zoroastrian influence, but also because the Mishnah was composed during the
Parthian period, when the empire exercised little influence on local cultures).
Was the Mishnah here, as elsewhere, influenced by Roman law? Evidently not.
Elman shows that Roman law had a completely different attitude toward fraud,
favoring the stronger party.

Sages: Accommodation and Resistance in the Shaping of Rabbinic Legal Tradition,” in:
C. E. Fonrobert and M. S. Jaffee (eds.), The Cambridge Companion to the Talmud and
Rabbinic Literature (Cambridge 2007) 165–97; “Toward an Intellectual History of Sasanian
Law: An Intergenerational Dispute in ‘Herbedestan’ 9 and its Rabbinic and Roman
Parallels,” in: C. Bakhos and M. R. Shayeqan (eds.), The Talmud in its Iranian Context
(Tübingen 2010) 21–57; see also S. Secunda, “Talmudic Text and Iranian Context: On the
Development of Two Talmudic Narratives,” AJS Review 33 (2009) 45–69; “Reading the
Bavli in Iran,” JQR 100 (2010) 310–42; R. Kiperwasser, “Rabba bar Bar Channa’s Voyages,”
Jerusalem Studies in Hebrew Literature 22 (2007–8) 215–42 [Hebrew].
28 NIKOLSKY AND ILAN

5.3 Back from Babylonia to Palestine

We began this overview by arguing that the Babylonian alterations of Pales-
tinian traditions were based on the Babylonians’ strong conviction concern-
ing the authenticity of their traditions and superiority of their learning. This
approach is of course a polemical stance taken against the intuitive under-
standing that the Palestinian rabbinic culture is the authentic one.
The Tanhuma, a late Palestinian midrashic corpus, forcefully demon­strates
Palestinian acknowledgment of Babylonian superiority claims. In a text whose
origin is not quite clear and still much debated64 the Tanhuma states the
following:

‫ואף הקב״ה כרת ברית עם ישראל שלא תשכח תורה שבע״פ מפיהם ומפי זרעם‬
‫ ״ואני זאת בריתי אותם אמר ה׳ רוחי אשר עליך ודברי‬:‫ שנא׳‬,‫עד סוף כל הדורות‬
‫ ולפיכך קבע הקב״ה שתי ישיבות‬. . . )‫אשר שמתי בפיך לא ימושו וגו׳״ (ישעיה נט כא‬
‫לישראל שיהיו הוגין בתורה יומם ולילה ומתקבצין שתי פעמים בשנה באדר ובאלול‬
‫ ואותן ב׳ ישיבות לא ראו שבי‬. . . ‫מכל המקומות ונושאין ונותנין במלחמתה של תורה‬
‫ והוציאן הקב״ה י״ב שנה קודם‬,‫ ולא שלט בהן לא יון ולא אדום‬,‫ולא שמד ולא שלל‬
‫חרבן ירושלים בתורתן ובתלמודן שכך כתיב ״והגלה את כל ירושלים ואת כל השרים‬
‫ואת כל גבורי החיל עשרת אלפים גולה וכל החרש והמסגר ולא נשאר זולת דלת עם‬
‫ וכי מה גבורה יש בבני אדם ההולכים בגולה? אלא אלו גבורי‬,)‫הארץ״ (מלכים ב כד‬
,‫ חרש‬. . . )‫ ״על כן יאמר בספר מלחמות ה׳״ (במדבר כא יד‬:‫ שכך נאמר בה‬,‫תורה‬
‫ כיון שאחד מהן סוגר דברי‬,‫ מסגר‬.‫שבשעה שאחד מהן מדבר נעשו הכל כחרשין‬
‫ וכתיב‬. . . ‫ אין בעולם שיכול לפתוח לטהר ולהתיר‬,‫טומאה וטהרה או איסור והתר‬
‫ וכי משום‬.)‫״וישקד ה׳ על הרעה ויביאה עלינו כי צדיק ה׳ אלהינו״ (דניאל ט יד‬
‫דצדיק ״וישקד ה׳ על הרעה ויבא את הרעה״? אלא צדקה עשה הקב״ה עם ישראל‬
,‫שהקדים והגלה את גלות יכניה לגלות צדקיה כדי שלא תשתכח מהן תורה שבע״פ‬
‫ ולא גזרו‬,‫ ולא שלט בהן לא אדום ולא יון‬,‫וישבו בתורתן בבבל מן אותה שעה עד היום‬
.)‫ ואף לימות המשיח אין חבלי של משיח רואין (תנחומא נח ג‬.‫עליהם שמד‬

And the Holy One blessed be He also made a covenant with Israel that
the Oral Torah would never be forgotten from their mouths and from the
mouths of their descendants till the end of all generations, as it is writ-
ten: “And this shall be my covenant with them, said the Lord: My spirit
which is upon you, and the words which I have put in your mouth shall
not be absent [from your mouth, nor from the mouth of your children,
nor from the mouth of your children’s children—said the Lord—from
now on for all time]” (Isa 59:21) . . . thus the Holy One blessed be He insti-
tuted two yeshivot for Israel in which one would contemplate the Torah

64 So we were informed by I. Gafni in a personal conversation.

‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 29

night and day and congregate twice a year in Adar and in Elul65 of all
places and negotiate in the war of Torah . . . and these two yeshivot have
not been persecuted or looted and were ruled neither by Greece nor by
Edom (i.e. Rome), and the Holy One blessed be He removed them from
Jerusalem twelve years before the destruction with all their Torah and
Talmud, as it is written: “He exiled all of Jerusalem: all the commanders
and all the valiant heroes—ten thousand exiles—as well as the craftsmen
and smiths; only the poorest people in the land were left” (2Kgs 24:14).
What is valiant about people going into exile? These are the heroes of
the Torah, as it is written of it: “Therefore it is said in the Book of the
War of the Lord” (Num 21:14) . . . “the craftsman” (harash) because when
one of them speaks all become as the deaf (heresh); “smiths” (masger)
because if one of them locks (soger) impurities and purities or permis-
sions and prohibitions, no one in the world can open to purify or permit
(SifDeut 321) . . . And it is written: “Hence the Lord was intent on bringing
calamity upon us, for the Lord our God is righteous” (Dan 9:14). Because
he is “righteous” he “was intent on bringing calamity upon us”?! Rather,
the Holy One blessed be He was righteous with Israel by first exiling
Yechonia and only then Zedekiah, so that the Oral Torah not be forgotten
from them, and so that they reside with their Torah in Babylonia from that
time on to this day and they were not subjected to Edom (Rome) and to
Greece, and they were not persecuted, and even in the messianic future,
they will not suffer the pangs of the Messiah (cf. bKet 111a) (Tan Noah 3).

This tradition, although not found anywhere in the Bavli, is obviously Baby­
lonian. Scholars agree that in the beginning, when speaking of scholarly gath-
erings twice a year, it describes the institute of yarhei kalah, known from the
Gaonic period. Scholars have even used this Palestinian source to argue that
this institution dates back to the amoraic period.66 But in itself, this passage in
the Tanhuma is a masterful literary composition, one which employs a pleth-
ora of earlier rabbinic motifs.
The history of this passage’s components can be reconstructed as follows:
In Sifre on Deuteronomy we find a midrash that interprets the harash and
the masger of 2Kgs 24:14 as two types of Torah scholars. The Bavli takes up
this midrash several times (bGit 88a; bSanh 38a; bHag 14a), creating a narra-
tive about the two exiles of the Judeans (one under Yehoyachin and one under

65 These are the famous kallah months known from Gaonic tradition, see next note.
66 Gafni, Jews of Babylonia, 213–26; R. Brody, The Geonim of Babylonia and the Shaping of
Medieval Jewish Culture (New Haven 1998) 43–8.
30 NIKOLSKY AND ILAN

Zedekiah), asserting that the first exile was of Torah scholars (harash and mas-
ger), and the second was exiled while these scholars were still alive and active,
so that a continuous line of transmission remained unbroken. This is acknowl-
edged as divine grace.
The Tanhuma Buber, the more “Palestinian” of the two Tanhuma
compilations,67 includes this midrash in the pericope of Tazria. It adds but
a few words here and there, asserting that the midrash refers to the schol-
ars in Babylonia, and adding that those who descended to Babylonia trusted
Jeremiah’s words (probably referring to Jeremiah’s letter to the exiles that was
discussed above). The assertion that Babylonia will be spared the suffering
connected to the coming of the Messiah, which it also incorporates in its text,
is an idea already found in the Bavli (bKet 111a). If the line of development of
this passage that we are suggesting here is correct, we observe that the penetra-
tion of Babylonian culture into Palestinian literature can be followed within
the Tanhuma corpus itself. Evidently the Tanhuma here is not only presenting
a Babylonian agenda; it is also using the text of the Bavli itself in its aggrandize-
ment of Babylonia and its culture.
The final two contributions to this book deal with Babylonian traditions trav-
eling back to Palestine. Ronit Nikolsky discusses two midrashic traditions that
travel from early Palestinian midrashic compilations to the Bavli, and that
are later found together with Babylonian embellishments in the Tanhuma.
Evidently the Bavli has left many markers in the text of this midrash, as can be
seen both from Nikolsky’s essay, and the quote from the Tanhuma discussed
above.
Paul Mandel’s project is of a similar nature. He goes back to discussing the
famous story about Rabbi Aqiva’s martyrdom as it is related in the two tal-
mudim. Surprisingly, he shows, based on manuscripts, that certainly in the
Yerushalmi, and probably also in an early version of the Bavli, this was not a
martyrdom story at all. The Yerushalmi story tells about Rabbi Aqiva’s verbal
contest with the governor of Judaea Tineius Rufus; in the Bavli, although Rabbi
Aqiva is executed, this too is not a martyrdom story but turns into a moment of
teaching, as is typical in Babylonian stories. Having seen the Baby­lonian atti-
tude toward martyrdom as presented in the article of Christiane Tzuberi, it
should come as no surprise that the Babylonians did not voluntarily change a
simple story about a debate between a foreign potentate and a Jew into martyr-
dom. Only when the story, together with major sections of the Bavli, returned
to Palestine in the late Byzantine period was it influenced by Christian mar-
tyrdom stories, and updated to acquire the martyrdom plot that is so familiar

67 M. Bregman, The Tanhuma-Yelammedenu Literature: Studies in the Evolution of the Versions

(New Jersey 2003) 6, 12, 17 and 18.
‫מהתם להכא‬, FROM THERE TO HERE (bSanh 5a): AN INTRODUCTION 31

to us. Mandel illustrates, therefore, the Bavli’s widening circle of influence in

Palestine but also its own ingestion of Palestinian culture tropes on the way to
acquiring its canonic version.68
The phenomenon identified in these last two essays supports what we
claimed above, that some time after the Bavli was already in an advanced stage
of formation, its traditions traveled back to Palestine and were eventually
accepted as authoritative. This is perhaps best illustrated through the prolif-
eration of the Esther midrashim in Palestine in very late antiquity and the early
Middle Ages. As we argued above, Esther became a favorite role-model for dias-
pora Jews. Two contribu­tions to this book highlight her formative role in shap-
ing Babylonian Jewish identity. In a number of recent articles, Arnon Atzmon
conjectures69 that the abundance of what he terms neo-classical midrashim70
on Esther stems from a process that he designates “Babylonization” of the later
Palestinian material.71
We hope that this volume will constitute an additional contribution to
ongoing discourse on the relations between the culture of the Land of Israel
and the Babylonian culture; or to use a different metaphor, perhaps will build
another bridge over the divide between the two. We have no doubt, though,
that in this book, we have not even uncovered the tip of the iceberg of the
fascinating cultural, literary, and historical relationship between these two rab-
binic centers of learning.

68 Mandel identified Babylonization also with regard to the Palestinian midrash Lamentations
Rabba, which he claimed had two recensions, one of them heavily Babylonized, see
P. Mandel, “Between Byzantium and Islam: The Transmission of a Jewish Book in the
Byzantine and Early Islamic Periods,” in: Y. Elman and I. Gershoni (eds.), Transmitting
Tradition: Orality, Textuality and Cultural Diffusion (New Haven 2000) 94–5.
69 A. Atzmon, “Haman is the Satan: The Development of a Midrash Narrative of the Esther
Scroll,” in: N. Ilan et al. (eds.), Carmi Sheli: Mehkarim Ba’agada U-ve-Pharshanuta (Boston
2012) 162 [Hebrew].
70 This term was coined by Arnon Atzmon in his article “Old Wine in New Flasks: The Story
of Late Neo-Classical Midrash,” European Journal of Jewish Studies 3 (2009) 183–203, and
it refers to the early medieval midrashim of the 11th century. The last half of the 11th
century was also the time that saw the end of the process of the Babylonian “takeover”
of the European Jewish culture of Byzantion, Italy, and Ashkenaz, that was essentially
of Palestinian nature. See e.g. the work of A. Grossman, The Early Sages of Ashkenaz
(Jerusalem 2001) 424–35 and the addenda and corrigenda [Hebrew]; or I. M. Ta-Shma,
Studies in Medieval Rabbinic Literature vol. 3: Italy and Byzantium (Jerusalem 2005) 9–19,
177–87, and 231–7 [Hebrew].
71 A. Atzmon, “Hagadata de-Megilat Esther: Toward the Anthologist’s Methodology,” in:
B. J. Schwartz, A. Melammed, and A. Shemesh (eds.), Iggud: Mivhar Ma’amarim Be-Mada‘ei
Ha-Yahadut vol. 1 (Jerusalem 2008) 35 [Hebrew].
Now You See it, Now You Don’t:
Can Source-Criticism Perform Magic on Talmudic
Passages about Sorcery?
Shamma Friedman

for Tomer

In this essay I will deal with three major talmudic passages of rabbinic teaching
relating to witchcraft: the story of Rabbi Hanina’s encounter with a sorceress in
the context of Rabbi Yohanan’s and Rabbi Hanina’s conflicting positions on the
nature of magic; the pericopae that treat Rabbi Eliezer as a magician; and the
story about Rabbi Hanina and Rabbi Oshaya, who studied Sefer Yetzirah and
created a calf, which they then prepared as a sumptuous meal. My approach to
these sources is outlined briefly below.
The central talmudic passages dealing with sorcery (for convenience’s sake
we will also call it “magic”), including how it was viewed in the theological
and legal thinking of the sages, have been cited in all the major works on the
subject, from Brecher,1 Joël,2 Blau,3 Trachtenberg,4 and onward. I feel, however,
that a new contribution to this subject can emerge through the application to
these passages of more recently developed methodologies for the analysis and
interpretation of talmudic texts, essentially, but not limited to, redaction criti-
cism. These tools can lend a historical dimension, an awareness of conflicting
approaches, and especially help uncover evolutionary processes, unobserved
by a superficial or fundamentalist reader, or by the non-critical scholar.5
Methodologies of talmudic research have developed profusely in recent
decades, although they have yet to be described in a com­prehensive fashion,6
and often remain the secret lore of the initiated. These methodologies do

1 G. Brecher, Das Transzendentale, Magie, und magische Heilertarten im Talmud (Vienna 1850).
2 D. Joël, Der Aberglaube und die Stellung des Judenthums zu demselben (Breslau 1881–3).
3 L. Blau, Das altjüdische Zauberwesen (Budapest 1898) 19–23.
4 J. Trachtenberg, Jewish Magic and Superstition: A Study in Folk Religion (foreword by Moshe
Idel, Philadelphia 2004; originally published 1939) 15–6, 19–22.
5 See S. G. Wald, “On the Historical Study of the Talmud,” in: D. Golinkin et al. (eds.), Torah
Lishma: Essays in Jewish Studies in Honor of Professor Shamma Friedman (Jerusalem and
Ramat Gan 2007) xiv.
6 But see now a partial addressing of this task in A. Yorav, Transmission and Methodology
(Jerusalem 2002) [Hebrew].

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�02

TALMUDIC PASSAGES ABOUT MAGIC 33

not always reach all the scholars in the field, and they are even less accessi-
ble to scholars in adjacent fields, or those who must use rabbinic literature
for broader purposes. In the study below scholarship of earlier periods will be
recorded in order to set former methodologies in contrast with new ones.
A major contribution of these critical methodologies is exposing the perva-
sive evolution of texts and of the concepts imbedded within them. This serves
as a vast corrective for the earlier critical approach of viewing similar accounts
on a parallel alignment rather than upon a developmental continuum. We sug-
gest that with such an approach (the methodology which exposes evolution)
we can find “the divine in the [contrasting] details,” for they reveal the tenden-
zen of the editorial reworking, and by contrast those of the original form.
This approach is particularly apt for the issue of Bavel and Eretz Israel,
and their overlapping talmudic literature—a perfect situation for evaluating
the “different,” which stands out in contrast to the “similar.” Appreciation of
this factor goes far beyond chronology in focusing upon the distinctive men-
talities of talmudic society produced by the two communities, the one (Eretz
Israel) earthly, indigenous, and integrated within other social classes; the other
(Bavel), often exhibiting opposite characteristics, among them rabbinization
and scholasticism.
The rise of these critical methodologies, and their flourishing since the
second half of the twentieth century, mark a watershed in the correct under-
standing of tannaitic material embedded in the Bavli. Previous to that, and
to some degree among those from various disciplines who accept the earlier
conclusions, we are witness to a type of scholarly fundamentalism which takes
accounts about tannaim in the Bavli as a historical record, or at least as a reflec-
tion of tannaitic beliefs and mentalities, thus falling far short of current stan-
dards. This is quite true for the appreciation of reigning concepts and practices
in Eretz Israel and Bavel—such as the existence and study of Sefer Yetzirah,
and practices based upon it, and other magical practices attributed to tannaim
in the Bavli—as we attempt to demonstrate in the following pages.
Sorcery is forbidden in Deut 18:10 and the sorcerer (‫ )מכשף‬is punishable
by death according to Exod 22:17.7 Prevalent and pervasive in the ancient

7 Through “stoning,” according to the Mishnah (mSanh 7:4), following Rabbi Aqiva (MdRY,
neziqin 17, p. 309). The same opinion is held there by Rabbi Yosi Ha-Gelili and Rabbi Yehudah
ben Betera, versus Rabbi Yishmael, who stipulates “beheading.” See in detail S. Friedman,
“Real and Illusory,” (Hebrew, in preparation) and Appendix B.
34 Friedman

Near East,8 sorcery (or witchcraft9) is also a capital offence, by ordeal, accord-
ing to the Middle Assyrian Laws and the Laws of Hammurabi:

If either a man or a woman should be discovered practicing witchcraft

(kišpī), and should they prove the charges against them and find them
guilty, they shall kill the practitioner of witchcraft. A man who heard
from an eyewitness to the witchcraft, that he witnessed the practice of
the witchcraft, who said to him: “I myself saw it,” that hearsay-witness
shall go and inform the king. If the eye­witness should deny what he (i.e.
the hearsay-witness) reports to the king, he (i.e. the hearsay-witness)
shall declare before the divine Bull the-son-of-the-Sun-God: “He surely
told me”—and thus he is clear. As for the eyewitness who spoke (of wit-
nessing the deed to his comrade) and then denied (it to the king), the
king shall interrogate him as he sees fit in order to determine his inten-
tions; an exorcist shall have the man make a declaration when they make
a purification, and then he himself (i.e. the exorcist) shall say as follows:
“No one shall release any of you from the oath you swore by the king and
by his son; you are bound by oath to the stipulations of the agreement to
which you swore by the king and by his son.”10

If a man charges another man with practicing witchcraft (kišpī) but can-
not bring proof against him, he who is charged with witchcraft shall go to
the divine River Ordeal, he shall indeed submit to the divine River Ordeal;
if the divine River Ordeal should overwhelm him, his accuser shall take
full legal possession of his estate; if the divine River Ordeal should clear
that man and should he survive, he who made the charge of witchcraft
against him shall be killed; he who submitted to the divine River Ordeal
shall take full legal possession of his accuser’s estate.11

8 I. T. Abusch, Babylonian Witchcraft Literature (Atlanta 1987).

9 The Akkadian verb and noun are cognates of the biblical ‫כש״ף‬: kašāpu = “to bewitch, to
cast an evil spell” (CAD, K, p. 284); kišpū = “witchcraft, sorcery” (p. 454).
10 A, §47: translation from M. T. Roth, Law Collections from Mesopotamia and Asia Minor
(Atlanta 1997) 172; cf. G. Cardascia, Les Lois Assyriennes (Paris 1965) 230–6, and especially
G. R. Driver and J. C. Miles, The Assyrian Laws (Oxford 1935) 118–21.
11 §2; translation from Roth, Collections, 81.
TALMUDIC PASSAGES ABOUT MAGIC 35

Witchcraft can be considered a crime in this culture12 and others,13 not for
theological reasons, but presumably for inflicting criminal damage.14 In the
context of Deut 18:9–14, sorcery in subsumed, together with other acts, under

12 See also, E. Neufeld, Hittite Laws (Michigan 1951) 44b, 111, 163,
13 “[T]here is also another kind which persuades the more daring class that they can do injury
by sorceries, and incantations, and magic knots, as they are termed, and makes others
believe that they, above all persons, are injured by the powers of the magician . . . [H]e
who seems to be the sort of man who injures others by magic knots, or enchantments,
or incantations, or any of the like practices, if he be a prophet or diviner, let him die;
and if, not being a prophet, he be convicted of witchcraft, as in the previous case, let the
court fix what he ought to pay or suffer” (Plato, Laws, XI, translated by Benjamin Jowett,
http://classics.mit.edu/Plato/laws.11.xi.html; see P. S. Alexander, “The Talmudic Concept
of Conjuring (’ahizat ‘einayim) and the Problem of the Definition of Magic (kishuf ),” in:
R. Elior and P. Schäfer (eds.), Creation and Re-Creation in Jewish Though: Festschrift in
Honor of Joseph Dan on the Occasion of His Seventieth Birthday (Tübingen 2005) 7.
14 “Almost as unpredictable as the activities of demons or ghosts were the machinations
of hu­man sorcerers and many rituals dealing with this problem are known. It has to
be stressed, however, that black magic as a category never existed in Mesopotamia;
sorcerers used exactly the same techniques and spells for their illegiti­mate purposes
that the victims might use to defend themselves legitimately. The only dif­ference is that
evil sorcery was done by se­cretly invoking the gods or manipulating other supernatural
powers, while the defense relied on the openness of its acts. It is not easy to under­stand
how the gods themselves could be fooled by this simple distinction, but they obviously
were believed to act on behalf of the illegitimate rites as long as the victim failed to point
out to them . . . In a lengthy nocturnal ceremony an effigy of the witch has to undergo a
trial to determine the criminal nature of her acts, after which her likeness is destroyed by
fire . . . [A]nonymity, seemingly an important part of the Babylonian belief in witchcraft,
is also reflected in the fact that we have no evidence for witches being actually criminally
prosecuted, although several ‘law codes’ mention such procedures. The machinations of
a person recognized as the witch were probably no longer magically dan­gerous and were
therefore not mentioned or counteracted in extant rituals. In any case, an accusation of
sorcery after the fact was extremely hard to prove, could easily lead to the death of the
accuser himself, and was thus probably avoided whenever possible: the standard proce­
dure in such a case was not a trial by human judges, but rather an ordeal by immersion
into the ‘Divine River’ who could pronounce the accuser guilty by drowning him, or
innocent by letting him survive” (W. Farber, “Witchcraft, Magic, and Divination in
Ancient Mesopotamia,” in: J. M. Sasson [ed.], Civilizations of the Ancient Near East, vol.
3 (Peabody 1995) 1898; “The attitude of the ruling classes towards magic was at best
ambivalent, but usually negative. Magic was almost by definition ‘forbidden,’ and in legal
texts it is included in lists of forbidden things. ‘Magic’ tended to function sociologically
as a category of disapproval and control, deployed to marginalize and even criminalize
certain activities that were not acceptable to the religious and political elites” (Alexander
“Conjuring” 9).
36 Friedman

the category of abominations practiced by the Canaanites,15 but not neces-

sarily as a violation of monotheism, as indeed we find in one of the opinions
recorded in the Bavli (see below).16 Modern Bible scholarship has also tended
to view the prohibition of witchcraft as an aspect of the dismissal and elimina-
tion of idolatry.17 However, consideration of the fact that the same punishment,
namely, death, was also meted out to witchcraft/sorcery in the ancient Near
East, casts a new light upon this question. One must conclude that the prohi-
bition is not inherent to monotheistic theology,18 but is taken over from the
ancient Near East much like the “goring ox,” yet is presented in the Pentateuch
in Israelite categories: “abomination,” “defilement,” that is, illegitimate avenues
to the divine. Closer to the mark would be the view of the anti-social, indeed,
deadly force of sorcery as the root cause for its outlawing it and for legislating
the death penalty for it.19 In terms of its literary formulation, “You shall not let

15 In Lev 19:31 the prohibition, which includes some of the practices under discussion, is
explained as defiling. Two practices are punishable by the death penalty (Lev 20:27).
16 A view in that direction is taken by Tigay: “Although the reason divination and magic are
unacceptable ways of learning God’s will is nowhere explicitly stated, it is inferably because
they rely, or seem to rely, on powers other than God, both human and supernatural. Magic
is frequently predicated on the belief that there are powers independent of the gods, and
even superior to them that may be employed without their consent or even against their
will. Even where magic is assumed to rely on divine assistance, the spells uttered by pagan
magicians leave room for the impression that it is their own power, not the gods’ that is
operating” (J. H. Tigay, The JPS Torah Commentary, Deuteronomy [Philadelphia 1996] 174;
for talmudic references see: pp. 375–6, n. 34).
17 “In biblical religion, sorcery in any form was, by definition, deemed ineffectual since all
events were under the control of the one God. It was also deemed heretical since any
attempt to alter the future purported to flout and overrule the will of God” (J. Milgrom,
The JPS Torah Commentary, Numbers [Philadelphia 1989], 471 = J. Milgrom, Leviticus 17–22
[The Anchor Bible; New York 2000] 1186–7). According to the Book of Watchers, the black
arts were passed on to mortals by the fallen angels, see F. García Martínez and E. J. C.
Tigchelaar (eds.), The Dead Sea Scrolls Study Edition vol. 1 (Leiden 1997) 403.
18 “[T]he antisocial character of black magic . . . led many societies to ban it. That may be the
reason behind the prohibition of sorcery in Exodus 22:17” (Tigay, Deuteronomy, 174).
19 In other words, black magic. For a critique of overdoing the conceptualization of
magic solely in terms of monotheistic standards, although from a different aspect, see
R. Schmitt, “The Problem of Magic and Monotheism in the Book of Leviticus,” Journal
of Hebrew Scriptures 8 (2008) 3–6. For an investigation of the route traveled by legal
material from Mesopotamia, see S. Greengus, “Biblical and Mesopotamian Law,” in: R. E.
Averbeck, M. W. Chavalas, and D. B. Weisberg (eds.), Life and Culture in the Ancient Near
East (Bethesda 2003) 67, n. 14.
TALMUDIC PASSAGES ABOUT MAGIC 37

a sorceress (‫ )מכשפה‬live” (Exod 22:17) is not substantially different from “they

shall kill the practitioner of witchcraft (kišpī).”20
Regarding the Assyrian Laws, we note that the damning formulation of the
witness is: “I myself saw it.” Below we will deal with the tannaitic distinction
between illusion and actual results. This distinction cannot be read back into
the Assyrian Laws. Certainly “I myself saw it” refers to the act of witchcraft and
not its tangible results.

1 Rabbi Yohanan and Rabbi Hanina: Two Conflicting Positions

on Magic

In discussion of the rabbinic material, I will begin non-chronologically.

Analyzing amoraic material first will not only provide an opportunity to dem-
onstrate the use of one of the primary methodological tools at our disposal,
but also reveal a conceptual and ideological divide between the two periods,
that is, tannaitic and amoraic.
According to my reading, the amoraim, Rabbi Yohanan and Rabbi Hanina,
present two fundamental and directly conflicting positions on the nature of
magic and its relationship to the divine realm. One of them documents the
emergence of full legitimacy for certain magical acts,21 which, in the amoraic
period, eventually becomes a most respected action, and its practitioners
heroic. I will explain why this deep contrast of opinions became invisible to
many scholars, and how it can disclose the redactional tendency in reworking
tannaitic material.
Our passage occurs in two locations with slight variation in BT:

20 Cf. D. P. Wright, Inventing God’s Law: How the Covenant Code of the Bible Used and Revised
the Laws of Hammurabi (Oxford 2009) 200.
21 “Abbayye said: The laws of sorcerers are like those of the Shabbat: Certain actions are
punished by stoning, some are exempt from punishment, yet forbidden, whilst others are
entirely permitted. Thus, if one actually performs magic, he is stoned; if he merely creates
an illusion, he is exempt, yet it is forbidden; whilst what is entirely permitted? Such as was
performed by Rabbi Hanina and Rabbi Oshaia, who spent every Sabbath eve studying the
Laws of Creation, by means of which they created a third-grown calf [better: three-year-
old calf, see below] and ate it” (bSanh 67b). For more on this text see below.
38 Friedman

bSanh 67b22 ‫ב׳ סנהדרין סז ע״ב‬

Rabbi Yohanan said: Why are they [sor- ?‫ למה נקרא שמן כשפים‬:‫אמר רבי יוחנן‬
cerers] called keshafim? Because they 23‫שמכחישין פמליא של מעלה‬
lessen the power of the Divine agencies.
“There is none else besides Him” (Deut :‫ אמר רבי חנינא‬.)‫״אין עוד מלבדו״ (דברים ד לה‬
4:35). Rabbi Hanina said: Even by sorcery. .‫ כשפים‬24‫אפילו‬
A woman once attempted to take earth ‫ההיא איתתא דהות קא מהדרא למשקל‬
from under Rabbi Hanina’s feet: He said ‫ אי‬:‫ אמר לה‬.‫עפרא מתותי כרעיה דרבי חנינא‬
to her: If you succeed in your attempts, .‫ אין עוד מלבדו כתיב‬25.‫מסתייעת זילי עבידי‬
go and practice it [i.e. sorcery]: it is writ-
ten: “There is none else beside Him”
(Deut 4:35).
But that is not so, for did not Rabbi ‫ למה נקרא שמן‬:‫איני? והאמר רבי יוחנן‬
Yohanan say: Why are they called kes- .‫ שמכחישין פמליא של מעלה‬26?‫כשפים‬
hafim? Because they lessen the power of
the Divine agencies?
Rabbi Hanina was in a different category, .‫ דנפיש זכותיה‬,‫שאני רבי חנינא‬
owing to his abundant merit.

22 Soncino translations, with slight adaptations. A parallel text is found in bHul 7b. Significant
differences are noted below.
23 Rashi: ‫ שעל מי שנגזר לחיות ממיתין‬,‫( כשפים—נוטריקון כחש פמליא של מעלה‬keshafim—
this is an acronym, they lessen the power of the Divine agencies, killing those who were
destined to live). Yad Ramah: ‫ למה נקרא שמן כשפים? שמכחישין פמליא של‬:‫אמר ר׳ יוחנן‬
‫ שכן כשפים כש לשון‬,‫ מחוברת היא משתי מלות‬,‫ וכשפים אעפ״י שמלה אחת היא‬.‫מעלה‬
‫ כענין שם (ב׳ שבועות‬,‫ כאדם שמכחיש דברי חבירו‬,‫ ולשון מכחישין‬.‫ פם לשון פמליא‬,‫מכחש‬
‫ ויש לפרש לשון דבר כחוש וחלש כלומר‬.‫מז ע״ב) ״שני כתי עדים המכחישות זו את זו״‬
‫שניתן להם רשות לבטל גזרת פמליא של מעלה שגזר הגזירה טובה לצדיק שאינו גמור שאם‬
‫( נעשו לו כשפים אין אותה גזירה עומדת לו מפני עונותיו‬Said Rabbi Yohanan: Why are they
[sorcerers] called keshafim? Because they lessen the power [makhishin] of the Divine
agencies. And keshafim, even though it is one word, is composed of two words, for [in]
keshafim kash derives from “lie” [kahash] and makhishin is like a person who contradicts
the words of his fellow as there [bShevu 47b] “two groups of witnesses that contradict
one another.” Or it should be interpreted as something thin and weak, namely that they
were given permission to repeal the decrees of the Divine agencies, who decreed positive
decrees for the righteous who is not complete[ly righteous], that if he is bewitched, that
decree does not protect him because of his sins”).
24 Printed editions add: ‫לדבר‬.
25 The version in bHul 7b has ‫ לא מסתייעא מילתיך‬,‫( שקולי‬Take [as much as you will], you
will not succeed in your attempts).
26 Printed editions: ‫מכשפים‬.
TALMUDIC PASSAGES ABOUT MAGIC 39

Rabbi Yohanan says that the practitioners of sorcery deny or weaken27 the very
power of God. They are working against God from the outside of His realm.28
In contrast, Rabbi Hanina quotes the verse, “There is none besides Him” (Deut
4:35), alleging that it is impossible to operate outside of God’s realm, even
through sorcery.
When a witch tried to retrieve some earth from under Rabbi Hanina’s feet
in order to use it magically to harm him,29 he said: “If you succeed in your
attempts, go and practice sorcery. It only works if God is behind it” (so in
bSanh), or “Try as you will, you will not succeed, because God is not behind it!”
(in bHul). This is perfectly consistent with Rabbi Hanina’s position that sorcery
only works or does not work according to God’s will; there is only one realm
(‫)אין עוד מלבדו‬, in contrast to Rabbi Yohanan, who holds that sorcery operates
against God from the outside.
This is the primary material of this sugya, in which, I suggest, we find two
diametrically opposing views. However, the talmudic redactor uses Rabbi
Yohanan’s statement to challenge Rabbi Hanina’s action, assuming that Rabbi
Hanina accepts Rabbi Yohanan’s basic premise. He then removes the difficulty,
by assuming that although Rabbi Hanina agreed with Rabbi Yohanan’s world
view, namely, that magic succeeds even when working against God’s will,
Rabbi Hanina considered himself an exception, owing to his abundant merit,
which protects him.30 In other words, Rabbi Hanina offers a position à la Rabbi
Yohanan: sorcery operates outside of the realm of the holy, challenging the

27 See Y. Harari, Early Jewish Magic: Research, Method, Sources (Jerusalem 2010) 283, n. 48.
In order to decide between these two possibilities it is necessary to uncover the more
significant meaning behind the rhetoric which is dictated by the notariqon. This would
seem to relate to power of action rather than theology. ‫ כחש‬in the hifīl is a transitive
verb meaning to “reduce, weaken, impair” (M. Jastrow, A Dictionary of the Targumim, the
Talmud Babli and Yerushalmi and the Midrashic Literature [New York 1926] 629); “to make
lean, to cause deterioration” (M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic
[Ramat Gan and Baltimore 2002] 568), which seems to be standing behind the Hebrew
verb in the Bavli in our passage.
28 Alexander writes: “Behind this statement lies a dualistic view of the world as a theatre of
conflict between good and evil forces” (Alexander, “Conjuring,” 23).
29 The practice was well known in ancient Mesopotamia. An Akkadian text reads (in
translation): “for purposes of magic, a witch will take dust from a man’s footprints” (CAD,
vol. 4 E, 185b). See R. C. Thompson, Semitic Magic, its Origins and Development (London
1908) 146 (cited by S. Lieberman, Greek in Jewish Palestine [New York 1942] 113, n. 136), for
whom the Akkadian material was not yet available.
30 “Der Talmud kann den Muth Chaninas nur so begreifen, dass er sich auf die Kraft seiner
vielen Verdienste stützte” (Blau, Zauberwesen, 22).
40 Friedman

power of God, but merit is a counter force, neutralizing the power of magic.
The redactor assumes that Rabbi Hanina’s lack of concern that the witch’s spell
would harm him was not due to a blanket denial that magic could operate from
the outside. Rather, he trusted in the counter power stored in the merits he had
amassed through performing mitzvot. In other words, according to the redac-
tor, were the witch to have acted against a normal person, Rabbi Hanina would
not have said that the witch would not succeed. He said so about himself only
due to his own merit.
The redactor has reconciled the conflict by assimilating Rabbi Hanina’s
position to that of Rabbi Yohanan. However, the unadulterated contrast of the
two diametrically opposed theological positions can be retrieved by concen-
trating on the primary material, in isolation from the editorial stratum.31
The redactor abandons Rabbi Hanina’s position and also his rhetoric by rec-
onciling and assimilating it both to Rabbi Yohanan’s position and to his rheto-
ric. If sorcery works against God’s will, as Rabbi Yohanan said, how could Rabbi
Hanina have originally said, “Try as you will, you will not succeed!” The redac-
tor’s answer is given within Rabbi Yohanan’s concepts: even though sorcery
is an autonomous power that overrules Heaven (as Rabbi Yohanan claimed),
meritorious piety overrules sorcery!
I claim, however, that Rabbi Hanina himself would never have accepted the
redactor’s interpretation of his doctrine or of his action. It is not merit that
overcomes magic; rather, all magic works or does not work through God’s will.
The practical difference is that, according to Rabbi Hanina, God could block
harmful magic even against a sinner. According to the redactor, He could not.
Rabbi Hanina’s significant ideological statement is thus blunted and diluted by
the redactor’s scholastic reconciliation. In connection with this passage, Saul
Lieberman wrote in the 1940s:

The Rabbis displayed the same intelligent attitude towards magic and
charms as towards astrology. Magic is effective in the case of the ordi-
nary man only, but not in that of the really righteous, whose merit is
great (‫ ;)דנפישא זכותיה‬it is powerless in face of the virtuous man. So TB32
relates: ‫ההיא איתתא דהות קא מהדרא למישקל עפרא מתותיה כרעי דר׳ חנינא‬

31 The following passage may indicate that the two positions have been reversed in
transmission, a not uncommon phenomenon in talmudic literature: “It was stated. Rabbi
Hanina said: The planetary influence gives wisdom, the planetary influence gives wealth,
and Israel stands under planetary influence. Rabbi Yohanan maintained: Israel is immune
from planetary influence” (bShab 156a).
32 Note 134 here: “Hullin 7b (and parallel).”
TALMUDIC PASSAGES ABOUT MAGIC 41

‫ אין עוד מלבדו כתיב‬,‫אמר לה שקילי לא מסתייעא מילתיך‬, “A woman tried to

take the sand from under the footsteps of R. Hanina.33 He said to her:
Take it; thy sorcery shall not succeed, for it is written: ‘There is none else
besides Him’ ” (Deut. IV.35).34

Lieberman cites Rabbi Hanina, but interprets him according to the redactor, by
saying: “Magic is effective in the case of the ordinary man only”—thus drawing
Rabbi Hanina into Rabbi Yohanan’s position of automatic effectiveness. The
power of “merit” is the redactor’s bending of Rabbi Hanina’s response to the
witch in order to accommodate it to Rabbi Yohanan’s view. However, Rabbi
Hanina’s citation of the verse “There is none else beside Him” as quoted by
Lieberman could only serve Rabbi Hanina’s original position, but not as proof-
text for the redactor’s explanation, which posits that there is a power beside
Him, but a counter-force (Rabbi Hanina’s merit) neutralizes it.
Lieberman opens with the redactor’s position, even quoting the Aramaic
text (‫)דנפישא זכותיה‬, but keeps it separate from the formal quotation of the
episode of Rabbi Hanina and the witch, as if signaling his awareness that
this is a position based on the redactor’s reinterpre­tation, a position which,
however, Lieberman completely accepts. Writing in the 1940s, this was more
of an option than from the 70s onwards, when awareness brought with it a
requirement to flesh out the original pre-redactional import of amoraic
materials,35 a methodologi­cal switch which Lieberman himself subsequently
acknowledged.36
In most contemporary scholarship on magic the positions of Rabbi Hanina
and Rabbi Yohanan are also presented through the eyes of the redactor, and as
representing, in their combined form forged by the redactor, a general rabbinic
stance on magic.37 Thus a golden opportunity for bringing these contrasting

33 Note 135 here: “A Palestinian sage of the III c.”

34 Lieberman, Greek in Jewish Palestine (New York 1965) 113.
35 See S. Friedman, “A Critical Study of Yevamot X with a Methodological Introduction,” in:
H. Z. Dimitrovsky (ed.), Texts and Studies: Analecta Judaica vol. 1 (New York 1977) 283–314
[Hebrew].
36 Private written communications, Feb. 19, 1975 (‫אור ליום ה׳ לסדר זכור אשר עשה לך עמלק‬
‫ ;)תשל״ה‬May 31, 1977 (‫)יום ג׳ פרש׳ בהעלותך תשל״ז‬.
37 See Urbach, Sages, 97–8, where both opinions are cited together to indicate a general
“opposition to sorcerers,” without any indication of the specific conflicting ideologies. The
general position is presented closer to what we have associated with Rabbi Hanina only:
“The Rabbinic doctrine [. . .] excludes the possibility of the existence of a magical power
capable of influencing the laws of nature and the decrees of God.” Thus, he assimilates
Rabbi Yohanan to Rabbi Hanina, in the opposite direction taken by the redactor. Veltri,
42 Friedman

ideological opinions into sharp focus is forfeited, eclipsed by redactional

harmonization.38

2 Rabbi Eliezer Portrayed as Magician

Scholarly opinion would have it that just as Saul practiced necromancy

according to the first book of Samuel (1Sam 28:8–25), so, according to tan-
naitic literature, Rabbi Eliezer practiced the magical planting and harvesting
of cucumbers, and recounted his accomplishments in a detailed first-person
narrative. The relevant passage follows:

bSanh 68a ‫ב׳ סנהדרין סח ע״א‬

Moreover, I have studied three hundred ,‫ולא עוד אלא שאני שונה שלש מאות הלכות‬
(or, as others state, three thousand) ‫ בנטיעת‬,‫ואמרי לה שלשת אלפים הלכות‬
laws39 about the planting of cucumbers ,‫קשואין ולא היה אדם שואלני בהן דבר מעולם‬
[by magic] and no man, excepting Aqiva .‫חוץ מעקיבא בן יוסף‬
ben Joseph, ever questioned me thereon.
For it once happened that he and I were :‫ אמר לי‬.‫פעם אחת אני והוא מהלכין היינו בדרך‬
walking together on a road, when he said .‫ למדני בנטיעת קשואין‬,‫רבי‬
to me: My master, teach me about the
planting of cucumbers.
I pronounced a spell, and the whole field .‫ נתמלאה כל השדה קשואין‬,‫אמרתי דבר אחד‬
was filled with cucumbers.

on the other hand, unites Rabbi Hanina and Rabbi Yohanan in the position assigned by
the redactor: “Die Zauberei hatte über R. Hanina wegen seiner Verdienste (‫ )זכותיה‬keine
Macht. Dieses talmudische Diktum bekräftigt einmal mehr, dass R. Yohanan—entgegen
der Interpretation Urbachs—den Zauberem die Macht einräumt, sich der oberen
Familia zu widersetzen; R. Hanina ist davon ausgenommen, was der Talmud seinen
Verdiensten zuschreibt” (G. Veltri, Magie und Halakha: Ansätze zu einem empirischen
Wissenschaftsbegriff im spätantiken und frühmittelalterischen Judentum [Tübingen 1997]
38–9). J. H. Chajes (“Rabbis and their [In]Famous Magic: Classical Foundations, Medieval
and Early Modern Reverberations,” in: R. S. Boustan, O. Kosansky, and M. Rustow [eds.],
Jewish Studies at the Crossroads of Anthropology and History: Authority, Diaspora, Tradition
[Philadelphia 2011], 59) mentions only Rabbi Yohanan, apparently assimilating Rabbi
Hanina to that view, and finding a different approach only with Abbayye.
38 See Harari, Magic, 283, who builds both statements into his description of a rabbinic
stance.
39 See the Yerushalmi cited below.
TALMUDIC PASSAGES ABOUT MAGIC 43

Then he said: Master, you have taught me .‫ למדני עקירתן‬.‫ למדתני נטיעתן‬,‫ רבי‬:‫אמר לי‬
how to plant them. Now teach me how .‫ נתקבצו כולן למקום אחד‬,‫אמרתי דבר אחד‬
to pluck them up. I pronounced another
spell and all the cucumbers gathered in
one place.

This cucumber episode is part of a longer passage introduced by ‫תניא‬, marking

it as a baraita. As is well known, the terms ‫ ברייתא‬and ‫ תניא‬apply to tannaitic
pericopae embedded in the talmudic discourse. Were we to accept a categori-
zation of this passage on the basis of the introductory term (‫)תניא‬, it would be
valid to ascribe this report to an authentic tannaitic tradition. We would then
conclude, as is indeed the position of contemporary scholarship on rabbinic
magic, that the tannaim viewed Rabbi Eliezer as actively performing cucum-
ber magic. However, let us review the entire baraita40 in which this report is
but a small part:

bSanh 68a ‫ב׳ סנהדרין סח ע״א‬

Surely it has been taught: :‫והתניא‬
When Rabbi Eliezer fell sick, Rabbi Aqiva ‫כשחלה רבי אליעזר נכנסו רבי עקיבא וחביריו‬
and his companions went to visit him. He ‫ הוא יושב בקינוף שלו והן יושבין‬.‫לבקרו‬
was seated in his canopied four-poster, .‫בטרקלין שלו‬
whilst they sat in his salon.
That day was Shabbat eve, and his son ‫ ונכנס הורקנוס בנו‬,‫ואותו היום ערב שבת היה‬
Hyrcanus went in to him to remove his .‫ גער בו ויצא בנזיפה‬.‫לחלוץ תפליו‬
phylacteries. But his father rebuked him,
and he retreated crestfallen.
It seems to me, said he to them, that my ‫ כמדומה אני שדעתו של אבא‬:‫אמר להן לחביריו‬
father’s mind is deranged. But [Rabbi ‫ היאך‬.‫ דעתו ודעת אמו נטרפה‬:‫ אמר להן‬.‫נטרפה‬
Eliezer] said to them: His mind and his ?‫מניחין איסור סקילה ועוסקין באיסור שבות‬
mother’s mind are deranged. How can one
neglect a prohibition which is punished
by stoning, and turn attention to some-
thing which is merely forbidden as shevut?

40 An expanded and reworked parallel is found in AdRN A 25 (see A. Goshen-Gottstein,

“A Lonely Sage on his Deathbed: The Story of the Death of Rabbi Eliezer—an Ideological
Analysis,” in: M. Bar-Asher, A. Edrei, J. Levinson, and B. Lifshitz (eds.), Studies in Talmudic
and Midrashic Literature in Memory of Tirzah Lifshitz [Jerusalem 2005] 79–112 [Hebrew]),
based mostly on the Bavli, but also drawing on the Yerushalmi (arranged synoptically in
Appendix A below), and other sources. The 300 laws taught are not laws of cucumber
planting, but laws of witchcraft, as in the Tosefta and ySanhedrin. See also AdRN A 19.
44 Friedman

The sages, seeing that his mind was clear, ‫ נכנסו‬,‫כיון שראו חכמים שדעתו מיושבת עליו‬
entered his chamber and sat down at a .‫וישבו לפניו מרחוק ארבע אמות‬
distance of four cubits.
Why have ye come? Said he to them. To ‫ ללמוד תורה‬:‫ למה באתם? אמרו לו‬:‫אמר להם‬
study the Torah, they replied; And why ?‫ ועד עכשיו למה לא באתם‬:‫ אמר להם‬.‫באנו‬
did ye not come before now? He asked. ‫ תמיה‬:‫ אמר להן‬.‫ לא היה לנו פנאי‬:‫אמרו לו‬
They answered: We had no time. He then :‫ אמר לו רבי עקיבא‬.‫אני אם ימותו מיתת עצמן‬
said: I will be surprised if these die a nat- ‫ נטל שתי‬.‫ שלך קשה משלהן‬:‫שלי מהו? אמר לו‬
ural death. Rabbi Aqiva asked him: And ‫ אוי לכם שתי‬:‫ אמר‬,‫זרועותיו והניחן על לבו‬
what will my death be? And he answered: ‫ הרבה‬.‫ שהן כשתי ספרי תורה שנגללין‬,‫זרועותיי‬
Yours will be more cruel than theirs. He ‫ הרבה תורה‬.‫ והרבה תורה לימדתי‬,‫תורה למדתי‬
then put his two arms over his heart, and ‫ ולא חסרתי מרבותי אפילו ככלב המלקק‬,‫למדתי‬
bewailed them, saying: Woe to you, two ‫ ולא חסרוני תלמידי‬,‫ הרבה תורה לימדתי‬.‫מן הים‬
arms of mine that have been like two .‫אלא כמכחול בשפופרת‬
Scrolls of the Law that are wrapped up.
Much Torah have I studied, and much
have I taught. Much Torah have I learnt,
yet have I but skimmed from the knowl-
edge of my teachers as much as a dog
lapping from the sea. Much Torah have I
taught, yet my disciples have only drawn
from me as much as a painting stick from
its tube.
Moreover, I have studied three hundred ‫ אלא שאני שונה שלש מאות הלכות‬,‫ולא עוד‬
laws on the subject of a deep bright spot, ‫ ולא היה אדם ששואלני בהן דבר‬,‫בבהרת עזה‬
yet no man has ever asked me about them. .‫מעולם‬
Moreover, I have studied three hundred ,‫ולא עוד אלא שאני שונה שלש מאות הלכות‬
(or, as others state, three thousand) ‫ בנטיעת קשואין‬,‫ואמרי לה שלשת אלפים הלכות‬
laws41 about the planting of cucumbers ‫ חוץ‬,‫ולא היה אדם שואלני בהן דבר מעולם‬
[by magic] and no man, excepting Aqiva ‫ פעם אחת אני והוא מהלכין‬.‫מעקיבא בן יוסף‬
ben Joseph, ever questioned me thereon. .‫ למדני בנטיעת קשואין‬,‫ רבי‬:‫ אמר לי‬,‫היינו בדרך‬
For it once happened that he and I were .‫אמרתי דבר אחד נתמלאה כל השדה קשואין‬
walking together on a road, when he said .‫ למדני עקירתן‬,‫ למדתני נטיעתן‬,‫ רבי‬:‫אמר לי‬
to me: My master, teach me about the .‫ נתקבצו כולן למקום אחד‬,‫אמרתי דבר אחד‬
planting of cucumbers. I pronounced a
spell, and the whole field was filled with
cucumbers. Then he said: Master, you
have taught me how to plant them. Now

41 See Yerushalmi cited below.

TALMUDIC PASSAGES ABOUT MAGIC 45

teach me how to pluck them up. I pro-

nounced another spell and all the cucum-
bers gathered in one place.
His visitors then asked him: What is the ‫ הכדור והאמוס והקמיע וצרור‬:‫אמרו לו‬
law of a ball, a shoemaker’s last, an amu- ‫ הן‬:‫המרגליות ומשקולת קטנה מהו? אמר להן‬
let, a leather bag containing pearls, and ‫ מנעל שעל גבי האמוס‬.‫טמאין וטהרתן במה שהן‬
a small weight? He replied: They can .‫ הוא טהור‬:‫מהו? אמר להן‬
become impure, and if impure, they are
restored to their purity just as they are.
[Then they asked him:] What of a shoe
that is on the last? He replied: It is pure.
And in pronouncing the word “pure” his .‫ויצאה נשמתו בטהרה‬
soul departed in purity.
Then Rabbi Yehoshua arose and ex- ,‫ הותר הנדר‬:‫עמד רבי יהושע על רגליו ואמר‬
claimed: The vow is annulled, the vow is .‫הותר הנדר‬
annulled!
On the conclusion of Shabbat Rabbi ‫ מן קיסרי‬,‫למוצאי שבת פגע בו רבי עקיבא‬
Aqiva met his bier being carried from .‫ היה מכה בבשרו עד שדמו שותת לארץ‬.‫ללוד‬
Caesarea to Lod. [In his grief] he beat ‫ ״אבי אבי רכב ישראל‬:‫פתח עליו בשורה ואמר‬
his flesh until the blood flowed down ‫ הרבה מעות יש לי ואין‬,)‫ופרשיו״ (מלכים ב ב יב‬
upon the earth. [Then Rabbi Aqiva] com- .‫לי שולחני להרצותן‬
menced [his funeral address, the mourn-
ers being] lined up about the coffin, and
said: “My father, my father, the chariot of
Israel and the horsemen thereof” (2Kgs
2:12); I have many coins, but no money
changer to accept them.

The extreme length of this baraita alone would suffice to support the conclu-
sion that it is not an authentic tannaitic source. Further investi­gation dem-
onstrates that it is an amoraic conglomerate composition, in part based upon
reworked tannaitic material.42

42 It bears a relationship to the examples cited by Rubenstein as “pseudo-baraitot,”

“pseudoteachings” (J. L. Rubenstein, Stories of the Babylonian Talmud [Baltimore 2010]
261–2), see his general description there; S. Friedman, Talmud Arukh, BT Bava Metzi’a VI:
Critical Edition with Comprehensive Commentary, Text Volume and Introduction (Jerusalem
1996) 8–9; idem, “Towards a Characterization of Babylonian Baraitot: ‘ben Tema’ and
‘ben Doretai’,” in: Y. Elman, E. B Halivni, and Z. A. Steinfeld (eds.), Neti‘ot Ledavid: Jubilee
Volume for David Weiss Halivni (Jerusalem 2004) 200–3 [Hebrew]; idem, “Wonder Not at
46 Friedman

The Yerushalmi has a similar extended baraita43 portraying Rabbi Eliezer’s

death scene, in a similar, but shorter version. Notably, he cucumber episode is
completely absent! This episode, then, is to be taken as a further Babylonian
expansion of the Palestinian amoraic composition.44 A comparison of the
Bavli and Yerushalmi passages follows:

‫ב׳ סנהדרין סח ע״א‬ ‫ ה ע״ב‬,‫י׳ שבת ב ב‬

‫כשחלה רבי אליעזר נכנסו רבי עקיבא וחביריו‬ .‫מעשה ביר׳ ליעזר שהיה גוסס‬
‫ הוא יושב בקינוף שלו והן יושבין‬.‫לבקרו‬
.‫בטרקלין שלו‬
‫ ונכנס הורקנוס בנו‬,‫ואותו היום ערב שבת היה‬ ‫ערב שבת עם חשיכה ונכנס הורקנוס בנו לחלוץ‬
.‫ גער בו ויצא בנזיפה‬.‫לחלוץ תפליו‬ .‫את תפיליו‬
‫ כמדומה אני שדעתו של אבא‬:‫אמר להן לחביריו‬ ,‫ הינחת מצות הנר שהיא שבות‬,‫ בני‬:‫אמ׳ לו‬
.‫נטרפה‬ ‫ שאינן‬,‫ ובאתה לחלוץ תפילין‬,‫וחייבין עליה כרת‬
.‫אלא רשות ואינן אלא מצוה‬
‫ היאך מניחין‬,‫ דעתו ודעת אמו נטרפה‬:‫אמר להן‬ ‫ אוי לי שנטרפה דעתו‬:‫יצא לו והיה צועק ואמ׳‬
?‫איסור סקילה ועוסקין באיסור שבות‬ ‫ דעתי‬.‫ דעתך היא שנטרפה‬:‫ אמ׳ לו‬.‫שלאבא‬
.‫היא לא נטרפה‬
‫ נכנסו‬,‫כיון שראו חכמים שדעתו מיושבת עליו‬ ,‫כיון שראו תלמידיו שהשיבו דבר שלחכמה‬
.‫וישבו לפניו מרחוק ארבע אמות‬ .‫נכנסו אצלו‬
‫ ללמוד תורה‬:‫ למה באתם? אמרו לו‬:‫אמר להם‬
?‫ ועד עכשיו למה לא באתם‬:‫ אמר להם‬.‫באנו‬
‫ תמיה‬:‫ אמר להן‬.‫ לא היה לנו פנאי‬:‫אמרו לו‬
:‫ אמר לו רבי עקיבא‬.‫אני אם ימותו מיתת עצמן‬
‫ נטל שתי‬.‫ שלך קשה משלהן‬:‫שלי מהו? אמר לו‬
‫ אוי לכם שתי זרו־‬:‫ אמר‬,‫זרועותיו והניחן על לבו‬
‫ הרבה‬.‫עותיי שהן כשתי ספרי תורה שנגללין‬

a Gloss in which the Name of an Amora is Mentioned: The Amoraic Statements and the
Anonymous Material in the Sugyot of the Bavli Revisited,” in: S. Friedman (ed.), Talmudic
Studies: Investigating the Sugya, Variant Readings and Aggada (New York and Jerusalem
2010) 117–30 [Hebrew]; idem, “The Baraitot in the Bavli and their Relation to their Parallels
in the Tosefta,” in: D. Boyarin et al. (eds.), Atara l’Haim: Studies in the Talmud and Medieval
Rabbinic Literature in Honor of Professor Haim Zalman Dimitrovsky (Jerusalem 2000) 163–
201 [Hebrew].
43 See J. N. Epstein, Studies in Talmudic Literature and Semitic Languages, vol. 2 (Jerusalem
1988), 68–9 [Hebrew] for comparison with a parallel to one of its parts.
44 See Friedman, Talmud Arukh, Text Volume, 14, n. 50; Goshen-Gottstein, “Ideological
Analysis,” 84–90.
‫‪TALMUDIC PASSAGES ABOUT MAGIC‬‬ ‫‪47‬‬

‫תורה למדתי‪ ,‬והרבה תורה לימדתי‪ .‬הרבה תורה‬

‫למדתי‪ ,‬ולא חסרתי מרבותי אפילו ככלב המלקק‬
‫מן הים‪ ,‬הרבה תורה לימדתי‪ ,‬ולא חסרוני תלמידי‬
‫אלא כמכחול בשפופרת‪.‬‬
‫ולא עוד אלא שאני שונה שלש מאות הלכות‬
‫בבהרת עזה‪ ,‬ולא היה אדם ששואלני בהן דבר‬
‫מעולם‪.‬‬
‫ולא עוד אלא שאני שונה שלש מאות הלכות‪,‬‬
‫ואמרי לה שלשת אלפים הלכות‪ ,‬בנטיעת‬
‫קשואין ולא היה אדם שואלני בהן דבר מעולם‪,‬‬
‫חוץ מעקיבא בן יוסף‪ .‬פעם אחת אני והוא מהל־‬
‫כין היינו בדרך‪ .‬אמר לי‪ :‬רבי‪ ,‬למדני בנטיעת‬
‫קשואין‪ .‬אמרתי דבר אחד נתמלאה כל השדה‬
‫קשואין‪ .‬אמר לי‪ :‬רבי‪ ,‬למדתני נטיעתן‪ ,‬למדני‬
‫עקירתן‪ .‬אמרתי דבר אחד‪ ,‬נתקבצו כולן למקום‬
‫אחד‪.‬‬
‫אמרו לו‪ :‬הכדור והאמוס והקמיע וצרור המרג־‬ ‫והיו שואלין אותו והיה או׳ להן על הטמא טמא‬
‫ליות ומשקולת קטנה מהו? אמר להן‪ :‬הן טמאין‬ ‫ועל הטהור טהור‪.‬‬
‫וטהרתן במה שהן‪ .‬מנעל שעל גבי האמוס מהו?‬
‫אמר להן‪ :‬הוא טהור‪.‬‬
‫ויצאה נשמתו בטהרה‪.‬‬ ‫ובאחרונה אמ׳ טהור‪ ,‬ונסתלקה נשמתו‪ .‬אמרין‪:‬‬
‫ניכר רבי שהוא טהור ‪. . .‬‬
‫עמד רבי יהושע על רגליו ואמר‪ :‬הותר הנדר‪,‬‬ ‫נכנס ר׳ יהושע וחלץ את תפיליו והיה מגפפו‬
‫הותר הנדר!‬ ‫ומנשקו ובוכה ואו׳‪ :‬רבי רבי‪ ,‬הותר הנדר‪.‬‬
‫למוצאי שבת פגע בו רבי עקיבא מן קיסרי ללוד‪,‬‬
‫היה מכה בבשרו עד שדמו שותת לארץ‪ .‬פתח‬
‫עליו בשורה ואמר‪:‬‬
‫״אבי אבי רכב ישראל ופרשיו״ (מל״ב ב יב)‪,‬‬ ‫רבי‪ ,‬״רכב ישראל ופרשיו״ (מל״ב ב יב)‪.‬‬
‫הרבה מעות יש לי ואין לי שולחני להרצותן‪.‬‬

‫‪bSanh 68a‬‬ ‫‪yShab 2:2, 5b‬‬

‫‪When Rabbi Eliezer fell sick, Rabbi Aqiva‬‬ ‫‪It happened that Rabbi Liezer was dying.‬‬
‫‪and his companions went to visit him. He‬‬
‫‪was seated in his canopied four-poster,‬‬
‫‪whilst they sat in his salon.‬‬
‫‪That day was Shabbat eve, and his son‬‬ ‫‪[It was] the eve of Shabbat, and Hyrcanus,‬‬
‫‪Hyrcanus went in to him to remove his‬‬ ‫‪his son, entered to remove his (Rabbi‬‬
‫‪phylacteries. But his father rebuked him,‬‬ ‫‪Eliezer’s) phylacteries (which are forbid-‬‬
‫‪and he retreated crestfallen.‬‬ ‫‪den to be worn on a Shabbat).‬‬
48 Friedman

It seems to me, said he to them, that my He (Rabbi Eliezer) said to him: My son,
father’s mind is deranged. But [Rabbi you abandoned the command of lighting
Eliezer] said to them: His mind and his the candle, which is shevut [a major com-
mother’s mind are deranged. How can mandment of the Shabbat] and is pun-
one neglect a prohibition which is pun- ished by death at the hands of heaven,
ished by stoning, and turn attention to and you came to remove my phylacteries,
something which is merely forbidden as which is merely reshut, and is merely a
shevut? mitsva?!
He left shouting and said: O, my father’s
mind is lost. He said to him: Your mind is
lost. My mind is not lost.
The sages, seeing that his mind was clear, When his pupils saw that he answered
entered his chamber and sat down at a him with words of wisdom, they entered
distance of four cubits. [his chamber]
Why have ye come? said he to them. To
study the Torah, they replied; And why
did ye not come before now? He asked.
They answered: We had no time. He then
said: I will be surprised if these die a nat-
ural death. Rabbi Aqiva asked him: And
what will my death be? And he answered:
Yours will be more cruel than theirs. He
then put his two arms over his heart, and
bewailed them, saying: Woe to you, two
arms of mine that have been like two
Scrolls of the Law that are wrapped up.
Much Torah have I studied, and much
have I taught. Much Torah have I learnt,
yet have I but skimmed from the knowl-
edge of my teachers as much as a dog
lapping from the sea. Much Torah have I
taught, yet my disciples have only drawn
from me as much as a painting stick from
its tube.
His visitors then asked him: What is the And they busied themselves asking him,
law of a ball, a shoemaker’s last, an amu- and he was telling them, that impure
let, a leather bag containing pearls, and thing impure, and pure is pure.
a small weight? He replied: They can
become impure, and if impure, they are
restored to their purity just as they are.
TALMUDIC PASSAGES ABOUT MAGIC 49

[Then they asked him:] What of a shoe

that is on the last? He replied: It is pure.
And in pronouncing the word “pure” his And in the end he said “pure,” and his soul
soul departed in purity. departed. They said: It is apparent that
Rabbi is pure.
Then Rabbi Yehoshua arose and exclaimed: Rabbi Yehoshua came in, and removed
The vow is annulled, the vow is annulled! his phylacteries, and embraced him and
kissed him and cried and said: My master,
my master, the vow is annulled.
On the conclusion of Shabbat Rabbi
Aqiva met his bier being carried from
Caesarea to Lod. [In his grief] he beat
his flesh until the blood flowed down
upon the earth. [Then R. Aqiva] com-
menced [his funeral address, the mourn-
ers being] lined up about the coffin, and
said:
“My father, my father, the chariot of Israel My master, “The chariot of Israel and the
and the horsemen thereof” (2Kgs 2:12); horsemen thereof” (2Kgs 2:12).
I have many coins, but no money changer
to accept them.

As noted above, the cucumber episode is completely absent from the

Yerushalmi version.45 However, there is indeed an authentic tannaitic parallel

45 Levinson writes regarding this baraita in the Bavli: “the unique version of the narrative
of R. Eliezer’s death that appears here emphasizes how magic is part of the Torah. In
distinction from the version recorded in Palestinian texts (yShab 5b, 382), where magic
does not even appear, here R. Eliezer is praised for his magical knowledge, as one who
knows three hundred laws of magic. What is important is not so much the hyperbolic
quantity but the actual transformation of magic into law, into a type of legal tradition
that must be studied and transmitted. It would seem that the Babylonian sages are less
concerned with other magicians who can perform the same magical feats than with
protecting and distinguishing their knowledge of magic from non-sage magicians”
(J. Levinson, “Enchanting Rabbis: Contest Narratives between Rabbis and Magicians in
Late Antiquity,” Jewish Quarterly Review 100 [2010] 73). It cannot be said, however, that
only the Bavli portrays Rabbi Eliezer as “one who knows three hundred laws of magic.”
Tannaitic literature and the Yerushalmi describe Rabbi Eliezer as transmitting three
hundred laws regarding ‫( מכשפה לא תחיה‬see below), laws which are in essence about
specific magical practices.
50 Friedman

regarding Rabbi Eliezer and cucumber magic, but the differences are telling.
tSanh 11:5 reads:

Cod. Erfurt Ed. Prin. Cod. Vienna

‫ שלש מאות‬:‫אמ׳ ר׳ עקיבא‬ ‫ שלש מאות‬:‫א״ר עקיבא‬ ‫ שלש מאות‬:‫אמ׳ ר׳ עקיבא‬
‫הלכות היה ר׳ אליעזר דורש‬ ‫הלכות היה ר׳ אליעזר שונה‬ ‫הלכות היה ר׳ אליעזר שונה‬
‫(בנטי) ב״מכשפה לא תחיה״‬ ‫ ולא למדתי ממנו‬,‫במכשפה‬ ‫ב״מכשפה לא תחיה״ (שמות‬
‫(שמות כב יז) ולא למדתי‬ ‫ שנים מלקטין‬:‫אלא שני דברי׳‬ ‫ ולא למדתי ממנו אלא‬,)‫כב יז‬
‫ שנים‬:‫הימנו אלא שני דברים‬ ‫ ואחד‬,‫ א׳ לוקט פטור‬,‫קשואין‬ ‫ שנים מלקטין‬:‫שני דברים‬
‫ אחד לוקט‬,‫מלקטין קישואין‬ ‫ העושה מעשה‬.‫לוקט חייב‬ ,‫ אחד לוקט פטור‬,‫קישואין‬
.‫ ואחד לוקט פטור‬,‫חייב‬ .‫ האוחז את העני׳ פטור‬,‫חייב‬ ‫ העושה מצוה‬.‫אחד לוקח חייב‬
‫ והאוחז‬,‫העושה מעשה חייב‬ .‫ האוחז את העצים פטור‬.‫חייב‬
.‫את העינים פטור‬

Said Rabbi Aqiva: Three hundred halakhot Rabbi Eliezer used to expound
on “Thou shalt not suffer a sorceress to live” (Exod 22:17), but I have only
learnt two things from him: Two may be gathering cucumbers, of whom
the one is innocent and the other guilty; he who actually does the deed is
guilty, and he who only appears to do so is innocent.46

The three hundred laws that Rabbi Eliezer could teach according to the Tosefta
in this regard were not “three hundred laws about the planting of cucum-
bers,” as they became in the Bavli’s baraita, but originally three hundred laws47
regarding Exod 22:17, “Though shalt not suffer a sorceress48 to live”49 (‫ְמ ַכ ֵּׁש ָפה‬

46 Translation: H. Danby, Tractate Sanhedrin, Mishnah and Tosefta, Translated from the
Hebrew with Brief Annotations (London 1919).
47 The exaggerated round number 300 is quite common in rabbinic literature (see
I. Zeligman, The Treasury of Numbers [New York 1942] 340–3 [Hebrew]), has been
observed by medieval commentators, and often discussed in secondary literature; see
recently Ch. Gafni, “Hyperbolic Language in the Mishnah?,” Jewish Studies: An Internet
Journal 8 (2009) 153–66 [Hebrew] and the literature cited there.
48 Y. Bazak voiced the far-reaching claim that Rabbi Eliezer wished to avoid mentioning
sorcery (‫ )כשפים‬by name, so he called it “cucumbers” (‫)קשאים‬, in that the gematria
of both is 451 (when using the ‫ כולל‬system for the first = counting the word itself as 1
(Y. Bazak, “Cucumbers and Witchcraft,” Bar Ilan 6 (1968) 165–6 [Hebrew]).
49 Cf. Goshen-Gottstein, “Ideological Analysis,” 88. Goshen (n. 29) considers the possibility
that the Bavli’s baraita changed the subject of the 300 laws from witchcraft in general
to cucumbers in particular under the influence of the Mishnah, which speaks about
harvesting cucumbers. This is, of course, not necessary, for the very Tosefta passage which
served as the source of the baraita in the Bavli already contains the same language, i.e.
TALMUDIC PASSAGES ABOUT MAGIC 51

‫)לֹא ְת ַחּיֶ ה‬.50 If the sorcerer is to be sentenced by the court to execution, the
judges must be versed in the legal status of various acts of sorcery. This is
exactly the reason for the requirement that appointment to the Sanhedrin
was made conditional upon knowledge of sorcery: “Rabbi Yohanan said: None
are to be appointed members of the Sanhedrin but men of stature . . . with a
knowledge of sorcery” (bSanh 17a).51
Of the many (not necessarily three hundred) laws about sorcery that Rabbi
Eliezer knew, only one dealt with cucumber magic, or as the Tosefta calls it,
“two things,” namely, if one uses magic actually to harvest real cucumbers, that
person is liable to the death penalty by stoning as a sorcerer, as mandated by
Exod 22:17. However, if the fantastic flying cucumbers were nothing but an
optical illusion, s/he is exempt.52
Furthermore, this more original form of the tradition does not have Rabbi
Eliezer himself perform this magical act at all, but rather he uses the category

cucumbers, and the baraita may be harmonizing the two parts of the Tosefta passage
(see the second possibility raised by Goshen). Of course, the mention of cucumbers in
the Mishnah may be encouraging this process. It is common that the Babylonian baraitot
introduce language from the Mishnah into the original texts they received from the Tosefta,
etc. (see Friedman, “Baraitot”; idem, “Baraita,” The Oxford Dictionary of the Jewish Religion
[Oxford 1997] 98). Goshen hesitates as to whether the Mishnah was indeed the factor
influencing the baraita here: “The fact that the gemara cites the story as an independent
source, and uses it to raise a difficulty against the Mishnah, lends more probability to the
theory that changing ‘Thou shalt not suffer a sorceress to live’ into reaping cucumbers
as the subject of the instruction is an independent process.” I would submit, though, that
the sage who produced the reworked baraita was probably much earlier and not identical
to the author of the anonymous stratum (stam hatalmud) who raised the contradiction.
I refer to the former as ‫( תנא בבלאה‬see Friedman, Talmudic Studies, ‫)יז–ט‬.
50 The scribe of Cod. Erfurt began writing ]‫בנטי[עת קישואין‬, the language of the Bavli, struck
out the four letters, and wrote ‫במכשפה לא תחיה‬. Sefer Ha-Yuhasin (ed. Filipowski 33b)
cites the Tosefta as: ‫ שלש מאות הלכות‬:‫ובפרק עשירי דסנהדרין בתוספתא אמר רבי עקיבא‬
‫היה דורש רבי אליעזר במכשפה לא תחיה ולא למדתי אלא שנים במלקטי קישואין אחד‬
‫לוקט חייב אחד לוקט פטור‬. The emphasized words are certainly the compiler’s addition.
It is interesting to note that Kallah Rabbati (Higger, p. 305), in the course of presenting
a parallel which is a shortened and reworked form of the baraita in the Bavli, reads: ‫ולא‬
‫ ושלש מאות הלכות פסוקות במכשפה‬,‫עוד אלא שאני שונה שלש מאות הלכות בבהרת עזה‬
!‫ חוץ מעקיבא בן יוסף‬,‫ ולא היה אדם ששאלני בהם‬,‫( לא תחיה‬regarding the number, see
Higger’s note there to l.77). Cf. below n. 128 and Appendix B.
51 Cf. n. 81. Rashi gave a less direct explanation for this requirement.
52 The “two things” that Rabbi Aqiva learned in the Tosefta are two laws of a theoretic
nature (real versus illusion). In the Bavli’s baraita they become two hands-on requests:
show me cucumber planting and cucumber harvesting (Goshen-Gottstein, “Ideological
Analysis,” 89).
52 Friedman

of cucumber-harvesting as a paradigmatic example in order to present the

legal distinction between actually performing versus creating an optical illu-
sion. In choosing cucumber magic as a fitting paradigm, we would expect that
Rabbi Eliezer was making use of a well-known magical feat. Magical harvest-
ing of crops fits the bill as an accomplishment that an ancient sorcerer could
certainly boast about, as Gideon Bohak has already pointed out.53
Simon Magus, also known as Simon the Sorcerer, was a Samaritan proto-
Gnostic in the first century CE.54 Mentioned in Acts 8:9–24 and accused of
being a demon in human form, he was said to possess the ability to levitate
and fly at will, with stories of his fantastic accomplishments persisting into the
Middle Ages. Simon plied his wares shamelessly:

I can make myself invisible to those who would seize me, and again, if I
wish to be seen, I can appear before them. If I should wish to flee, I would
bore through mountains and pass through rocks as if they were clay. If I
should hurl myself from a high mountain, I should be brought to earth
unharmed, as if borne up. If I be bound, I will lose myself, and those who
fettered me I will lay in bonds; if confined in prison, I will make the doors
open of themselves. I will animate statues, so that those who behold
them will suppose them to be living men. I will make new trees spring
up at once and cause thickets to grow up suddenly. I will throw myself
into the fire and I shall not be burned. I change my countenance so as
not to be recognized; nay I can show to men that I have two faces. I will
turn myself into a ewe or a she-goat. I will cause beards to grow on the
faces of little boys. I will fly up into the air, I will produce gold in great
quantities, I will make kings and cast them down . . . and once when my
mother Rachel ordered me to go to the field to reap, and I saw a sickle
lying, I ordered it to go and reap; and it reaped ten times more than the
others. Lately, I produced many new sprouts from the earth, and made
them bear leaves and produce fruit in a moment.55

53 G. Bohak, Ancient Jewish Magic: A History (New York 2008) 360.

54 Regarding a possible reflection of Simon Magus in Rabbinic literature, see H. J. Schoeps,
“Simon Magus in der Haggadah?,” Hebrew Union College Annual 21 (1948) 257–74. My
suggestion is not contingent on Schoeps’ hypothesis.
55 Pseudo-Clementine Recognitiones, II, 9. Cf. Bohak, Ancient Jewish Magic, 360, in connection
with Rabbi Eliezer. For further examples of the magician’s boast genre see C. Bonner,
“A Papyrus Describing Magical Powers,” Transactions and Proceedings of the American
Philological Association 52 (1921) 111–8.
TALMUDIC PASSAGES ABOUT MAGIC 53

Thus the Bavli’s baraita expanded and extended the ancient category of har-
vesting magic mentioned in the Tosefta to a more picturesque and exaggerated
level: it offers three hundred laws, not just one; it supplies narrative context
(“he and I were walking together on a road” etc.);56 and, furthermore, it records
that Rabbi Eliezer did more than teach the law—he recounted his actual
exploits of magically planting and harvesting cucumbers as a demonstration
for Rabbi Aqiva’s benefit.
We have compared the Bavli baraita with the Tosefta parallel in detail, yield-
ing a clear evolutionary relationship; indeed, we have “a close genetic link”57
with man-made mutations. A competing model understands similar parallel
passages which conflict on certain details, as “separate traditions.”58 I have

56 Cf. mAZ 2:5 etc.

57 See M. Kister, “Divorce, Reproof, and Other Sayings in the Synoptic Gospels: Jesus
Traditions in the Context of ‘Qumranic’ and Other Texts,” in: R. A. Clements and D. R.
Schwartz (eds.), Text, Thought, and Practice in Qumran and Early Christianity: Proceedings
of the Ninth International Symposium of the Orion Center for the Study of the Dead Sea
Scrolls and Associated Literature (Leiden 2009) 202.
58 Alexander writes: “The juxtaposition of the two texts also raises a serious issue which
troubled the Bavli’s redactor. If, as the Bavli story implies, Eliezer ‘did the deed,’ is he not
then culpable in terms of the Mishnaic ruling? ‘Did we not learn,’ the redactor asks, ‘that
he who does the deed is culpable? [However,] if it is only to teach, it is different. For it
has been said, ‘You shall not learn to do after the abominations of these nations’ (Deut
18:9): you may not learn in order to practice, but you may learn in order to understand.”
This explanation simply does not work, for while Aqiva might at a stretch be seen as
studying or learning, Eliezer cannot: he actually “does the deed.” This analysis strongly
suggests that we have here two traditions which, despite superficial similarities, origi-
nally had nothing to do with each other. The one does not explain the other. The only
substantial resemblances between them happen to be the involvement of Aqiva, and
the presence of cucumbers” (p. 12). Alexander goes on to establish an understanding of
the Mishnah separate from and independent of its close parallels: “The Mishnah has to
be explained in another way. I would suggest that it envisages basically an act of magic
whereby an object—a coin, an egg, here a cucumber—is suddenly produced to the sight
of an onlooker. ‘Gathering’ here has a technical sense and graphically describes the action
of ‘plucking’ something out of thin air” (Alexander, “Conjuring,” 13). Alexander continues
with an attempt to buttress this explanation: “This interpretation of ‘gathering’ as a tech-
nical term borrowed from magic gains some support from an obscure passage in Bavli
Sanhedrin in the middle of folio 67b, which has been commonly misunderstood . . . Rav
Papa exclaims: ‘By God! Though something the size of a camel he [a magician] cannot cre-
ate, the former [an object the size of a barley-corn] he can [at least] collect. However, the
latter [an object the size of a camel] he cannot collect . . . The Mishnah, then, I would sug-
gest, envisages a situation in which a magi­cian has produced a cucumber apparently from
nothing: if he has actually created the cucumber he is guilty” etc. (pp. 13–4). However, in
54 Friedman

argued for the advantages of the evolutionary model, in light of much con-
scious editorial activity that we are now able to perceive in rabbinic works, an
Occam’s Razor approach.59
The laudatory attitude to Rabbi Eliezer for performing acts through sor-
cery is not to be ascribed to a tannaitic conceptualization, whose legitimate
representatives blame sorcery for distancing Israel from God’s holy bounties:
“Fornication and sorceries have made an end of them altogether” (mSot 9:13).
As a reworking of genuine tannaitic material it reflects the tendenz of the
redactor of baraitot in the Bavli, and expresses the positive attitude towards
magic performed by sages as reflected in amoraic narrative, and in the unadul-
terated form of Rabbi Hanina’s stance: all magic is “under God.”
I will now address three issues relating to this analysis: (1) the relation-
ship between the relevant passages in the Mishnah and Tosefta; (2) whether
Rabbi Aqiva received this teaching from Rabbi Eliezer or from Rabbi Yehoshua;
(3) the expansion of the permission to study magic from theory to practice.

2.1 The Relationship between the Mishnah and Tosefta

The Mishnah runs as follows:

mSanh 7:11 60‫מ' סנהדרין ז יא‬

A sorcerer, if he actually performs magic .‫ לא האוחז את העיניים‬,‫ העושה מעשה‬,‫המכשף‬
[is liable to the death penalty], but not if
he merely creates an illusions.

the original the verbs are different. The Mishnah’s ‫ לקט‬is­indeed an agriculture term for
“harvest”; Aramaic ‫ והאי—לא מיכניף ליה( כנף‬,‫“ = )האי—מיכניף ליה‬assemble,” etc. Sokol-
off, Babylonian, 589, translates this passage: “(the demon) can (magically assemble this
[i.e. large objects] but not that [i.e. small objects].”
59 See a summary of this approach in S. Friedman, “Hanukkah in the Scholion to Megilat
Ta‘anit,” Zion 71 (2006) 6–8 [Hebrew]; idem, “Uncovering Literary Dependencies in the
Talmudic Corpus,” in: S. J. D. Cohen (ed.), The Synoptic Problem in Rabbinic Literature
(Providence RI 2000) 73–6; idem, “Historical Aggadah in the Babylonian Talmud,” in Saul
Lieberman Memorial Volume (Jerusalem and New York 1993) 1–46 [Hebrew]. Kimberly
Stratton writes, “The Jerusalem Talmud, however, records only R. Eliezer’s teaching that
the one who does an actual act of magic is guilty of crime while the one who performs
an illusion is not.” At the same time she opens, saying, “In the Jerusalem Talmud (Sanh.
41a) R. Eliezer’s use of ‘magic’ (makhashefa) to plant cucumbers is explicitly stated”
(K. Stratton, “Imagining Power: Magic and the Miracle and the Social Context of Rabbinic
Self-Representation,” Journal of the American Academy of Religion 73 [2005] 369, n. 22).
60 According to Cod. Kaufmann. Cf. SifDeut 171.
TALMUDIC PASSAGES ABOUT MAGIC 55

Rabbi Aqiva said in Rabbi Yehoshua’s ‫ שנים לוקטים‬:‫ר׳ עקיבה או׳ משם ר׳ יהושע‬
name: Of two who gather61 cucumbers .‫ אחד לוקט פטור ואחד מלקט חייב‬.‫קישואים‬
[by magic] one may be punished and the .‫ והאוחז את העיניים פטור‬.‫העושה מעשה חייב‬
other exempt. He who really gathers them
is punished; whilst he who produces an
illusion is exempt.

It would appear that the passage preserved in the Tosefta served as a source for
the compiler of our Mishnah.62 He presented the distinction between an actual
action and an illusion (‫)האוחז את העיניים‬63 anonymously, and in the context of
the general law of sorcery. However, he retained the statement attributed to
Rabbi Aqiva of exactly what he had learned from his teacher, probably due to
its literary attractiveness. Thus, the Mishnah ends up repeating this insight,
presenting it first anonymously, and then as transmitted by Rabbi Aqiva64
(albeit in Rabbi Yehoshua’s name, see below). The tendency in scholarship is

61 The “gathering” referred to (‫ )לקיטה‬is harvesting (see below; cf. ‫[ ליקט אתרוג‬tTer 2:6]).
The accomplishment would be extended beyond the text more than is warranted, by
paraphrasing the Mishnah, as done by Alexander: “On the one occasion ‘the deed is
done,’ that is to say, the ‘gathering’ involves actually creating cucumbers which were not
there before, whereas on the other occasion, the ‘gathering’ involves tricking the observer
into thinking that the cucumbers were being created, when, in fact, they had been there,
hidden, all the time” (Alexander, “Conjuring,” 11; italics added). See in greater detail,
ibid, p. 12.
62 On this phenomena see S. Friedman, Tosefta Atiqta: Pesah Rishon (Synoptic Parallels of
Mishna and Tosefta Analyzed with a Methodological Introduction) (Ramat Gan 2002) 15–95
[Hebrew].
63 This phrase is also used to define ‫( מעונן‬Deut 18:10), “soothsayer” (JPS) in the anonymous
opinion quoted along with the opinions of Rabbi Aqiva and Rabbi Yishmael (Sifra
qedoshim, pereq 6:2 (90c); SifDeut 171 (Finkelstein, pp. 218–9); tShab 7:14 (Lieberman, p. 27);
see bSanh 65b (note the change of location among the parallels and textual witnesses).
Ostensibly, Rabbi Eliezer would not hold that position (connected by popular etymology
‫ עיניים > מעונן‬as R. Aqiva’s: > ‫)עונה‬.
64 Cf. Friedman, Tosefta Atiqta, 99–100, regarding mPes 1:1. Malbim, apparently disturbed
by the repetition in the Mishnah, assigned opposing halakhic stances to each statement.
On this see further Friedman, “Real and Illusory,” in n. 7 above. Alexander, stating the
problem, wrote: “The extreme compression of the language, as so often in the Mishnah,
creates problems. The unit clearly falls into two parts, the first of which makes a statement
of principle, while the second provides a concrete illustration [. . .] The illustration adds
little to the general statement” (Alexander “Conjuring,” 10–1).
56 Friedman

always to present the Mishnah as the “key text.”65 In this case and, I suggest,
in many others, the Mishnah is a secondary text. Here it first paraphrases the
Tosefta and then repeats it almost verbatim.

2.2 Did Rabbi Aqiva Receive this Teaching from Rabbi Eliezer or from
Rabbi Yehoshua?
Regarding who taught Rabbi Aqiva magic the Mishnah is at odds with the
Tosefta (and the lengthy pseudo-baraita in the talmudic sugya). According to
the Mishnah,66 Rabbi Aqiva learned the cucumber reality/illusion distinction
from his teacher Rabbi Yehoshua, whereas according to the Tosefta and baraita
his mentor in this was Rabbi Eliezer. The Bavli itself raises this question:

bSanh 68a ‫ב׳ סנהדרין סח ע״א‬

But did Rabbi Aqiva learn this from Rabbi :‫והא רבי עקיבא מרבי יהושע גמיר לה? והתניא‬
Yehoshua? Surely it has been taught: ‫ מרבי אליעזר‬,‫ אלמא‬. . . ‫כשחלה רבי אליעזר‬
When Rabbi Eliezer fell sick etc . . . Thus !‫גמרה‬
from this story we see that he learned this
from Rabbi Eliezer.
He learned it from Rabbi Eliezer, but did ‫ הדר גמרה מרבי‬,‫גמרה מרבי אליעזר ולא סברה‬
not grasp it; then he learned it from Rabbi .‫יהושע ואסברה ניהליה‬
Yehoshua, who made it clear to him.

The Bavli resolves this contradiction by positing two separate events in suc-
cession, a standard talmudic resolution for maintaining both of two conflict-
ing passages describing the same episode.67 Rabbi Aqiva first learned this
distinction from Rabbi Eliezer and, subsequently, learned it again from Rabbi
Yehoshua. It should be clear, however, that we are dealing here with an edi-
torial revision, rather than multiple events. The original tradition had Rabbi
Aqiva learning from Rabbi Eliezer. The compiler of the Mishnah, or one of his
sources, preferred to attribute the teaching to Rabbi Yehoshua rather than to
Rabbi Eliezer, who had fallen out of rabbinic grace when he was placed under
a ban. As Yitzhak Gilat put it, “sometimes the attribution of a halakhah was

65 Alexander, “Conjuring,” 9, in relation to our passage, without mentioning the Tosefta; “The
Rabbinic discussion of the definition of magic begins, then, with Mishnah Sanhedrin 7:11.”
66 The language of the Mishnah also appears in SifDeut 171 (p. 219), which also reads
“R. Yehoshua” (as does a Genizah fragment, see M. I. Kahana, The Genizah Fragments of
the Halakhic Midrashim [Jerusalem 2005] 279 [Hebrew]).
67 See Friedman, “Historical Aggadah,” 121, n. 6; 135, n. 68; 130; 143, n. 136; 147. Often:
‫תרי עובדי הוו‬.
TALMUDIC PASSAGES ABOUT MAGIC 57

deliberately changed from Rabbi Eliezer to Rabbi Yehoshua or some other

tanna, in order that Rabbi Eliezer’s opinion might be accepted as the law
without being cited in his name. Thus, the fact that the halakhah had been
decided in accordance with his teaching was con­cealed.”68 Here we observe
Gilat’s description of a general tendency. Accordingly, our mishnah records
an opinion, and even the fact that Rabbi Aqiva heard this from his teacher,
in the name of Rabbi Yehoshua, despite the fact that earlier sources, namely,
the above Tosefta, and its reflection in the parallel baraitot of the Yerushalmi
and Bavli, had Rabbi Eliezer as the source of this halakhah.69 This conclusion
regarding the mishnah under study was already reached by J. N. Epstein.70 The
“real or illusion” distinction is often cited in scholarly literature in the name
of “the Mishnah” or “the Sages.” We can now, with relative certainty, identify
Rabbi Eliezer as the source of this teaching.
There is a further name confusion. The Yerushalmi reports the episode and
content in language similar to the Tosefta passage (namely, the three hundred
are witchcraft laws, not three hundred cucumber laws), but as a law learned
from Rabbi Eliezer, not by Rabbi Aqiva, but by Rabbi Yehoshua:

‫ שלש מאות פרשיות היה ר׳ ליעזר דורש בפרשת מכשפה‬:‫אמ׳ ר׳ יהושע בן חנניה‬
‫ ואחד‬,‫ אחד לוקט פטור‬,‫ שנים לוקטין קישואין‬:‫ומכולם לא שמעתי אלא שני דברים‬
‫ תשע‬:‫ אמ׳ ר׳ דרוסא‬.‫ פטור‬,‫ והאוחז את העינים‬,‫ חייב‬,‫ העושה מעשה‬.‫לוקט חייב‬
‫ ושלש מאות לחיוב שהוא‬,‫ ושלש מאות לפטור‬,‫מאות פרשיות היו; שלש מאות לחיוב‬
.)‫ כה ע״ד‬,‫פטור (י׳ סנהדרין ז יג‬

Said Rabbi Yehoshua: Three hundred laws did Rabbi Eliezer expound
concerning the verse, “Thou shalt not suffer a witch to live” (Exod 22:17),
and of all of them I have heard only two things: Two may gather cucum-
bers. One gatherer may be exempt, and one gatherer may be liable. He
who does a deed is liable, but he who merely creates an illusion is exempt
( ySanh 7:13, 25d).71

68 Y. D. Gilat, R. Eliezer ben Hyrcanus: A Scholar Outcast (Ramat Gan 1984) 486; see J. N. Epstein,
Introduction to the Mishnaic Text (Jerusalem 2000), 6 [Hebrew]; idem., Prolegomena as
Litteras Tannaiticas, 66–7 [Hebrew]. The primary sources are cited in these studies.
69 SifDeut 171, which has Rabbi Yehoshua, is reasonably a quotation from the Mishnah
(see Friedman, “Real and Illusory,” as above, n.7). Conversely, if the Mishnah is dependent
on the Sifre, it or its antecedent is the source for substituting Rabbi Yehoshua’s name for
Rabbi Eliezer.
70 Epstein, Introduction, 69.
71 Translation by J. Neusner.
58 Friedman

This reworking of the Tosefta passage by the Yerushalmi probably reflects a

parallel effort to resolve the same contradiction.72 The Yerushalmi may mean
by this: Rabbi Yehoshua heard it from Rabbi Eliezer, and Rabbi Aqiva heard it
from Rabbi Yehoshua! This account portrays the relationship between these
famous colleagues, Rabbi Eliezer and Rabbi Yehoshua, as a master/disciple
relationship. We can show that indeed Rabbi Yehoshua is portrayed elsewhere
in the Yerushalmi as a disciple of Rabbi Eliezer.73
Our conclusion is unmistakable: the primary sources of this tradition ascribe
the teaching about witchcraft to Rabbi Eliezer, and not Rabbi Yehoshua. Rabbi
Eliezer is portrayed, not as possessing three hundred laws regarding cucumber
magic, but three hundred legal insights regarding the subject of sorcery, and
the proper juridical interpretation of the death penalty prescribed for sorcery
in Exodus. Only one of these three hundred applies to cucumber magic. The
insight transmitted to Rabbi Aqiva is especially enlightening and, we may even
say, liberal. Rabbi Eliezer is keenly aware of the fact that much of what is called
sorcery is, in truth, merely an optical illusion or slight-of-hand.74 Seeing is not
believing. Therefore, a so-called sorcerer or magician who is simply perform-
ing optical illusions is exempt from the death penalty.
How different is this approach from that of Shimon ben Shetah, who, as
Rabbi Eliezer himself recalled, executed eighty women in Ashkelon. According
to an extraordinary account in the Yerushalmi, they were executed as practitio-
ners of witchcraft!75

72 If this is the case, we can modify Levinson’s statement that “only the Bavli addresses
this contradiction concerning the proper lines of rabbinic transmission” (“Enchanting,”
57, n. 9).
73 See Appendix B.
74 “As early as the 5th century BC[E], Greek magos had spawned mageia and magike to
describe the activity of a magus, that is, it was his or her art and practice. But almost
from the outset the noun for the action and the noun for the actor parted company.
Thereafter, mageia was used not for what actual magi did, but for something related to
the word ‘magic’ in the modern sense, i.e. using supernatural means to achieve an effect
in the natural world, or the appearance of achieving these effects through trickery or
slight of hand. The early Greek texts typically have the pejorative meaning, which in
turn influenced the meaning of magos to denote a conjurer and a charlatan” (http://
en.wikipedia.org/wiki/Magi). On Origen, Pliny (“if there is even a shimmer of truth in it,
that shimmer owes more to chemistry than to magic”), etc., see Levinson, “Enchanting,”
58 and n. 11.
75 mSanh 6:4; SifDeut 221; ySanh 6:6, 23c.
TALMUDIC PASSAGES ABOUT MAGIC 59

2.3 Expanding the Permission to Study Magic from Theory to Practice

On the primary level of our sources Rabbi Eliezer was never portrayed as per-
forming cucumber magic himself, but merely as teaching Rabbi Aqiva that
the illusionist is exempt! In referring to the cucumber feat recounted in the
Bavli’s expanded baraita, we must now be aware of how far removed this is
from the primary sources, and ipso facto removed from any historical reality.76
It can document the history of ideas in amoraic Babylonia, but definitely not
the historical action of a first-century tanna,77 or, more importantly, not even
any Palestinian tannaitic tradition or conceptualization.78 When the amoraim
freed magic from the realm of the devil (Rabbi Hanina: “There is none besides
Him”), the path was cleared for reworking the Rabbi Eliezer account to a more
hands-on story. By that time “sorcery” had become “magic.”
However, the Bavli’s anonymous scholastic voice (stam) is uncomfortable
with the [reworked] baraita’s claim that Rabbi Eliezer performed magical acts:

76 See Friedman, “Historical Aggadah,” 122.

77 As earlier scholarship might have had it. I believe it has become commonplace today to
see talmudic aggadah as literature and not necessarily history.
78 Boyarin reads the Babylonian account as reflecting the conceptuali­zation of Rabbi Eliezer
as a sectarian-type magician, building on the Jesus figure: “On one reading at least, for this
text, Rabbi Eli‘ezer’s magical activity of planting and harvesting cucumbers with a word
continues to mark his liminality, indeed, his closeness with ‘sectarianism’ . . . As we have
seen, early rabbinic texts repeatedly refer to Jesus as a magician” (D. Boyarin, Dying for
God: Martyrdom and the Making of Christianity and Judaism [Stanford, 1999] 38). Even
though “most, if not all, of what we ‘know’ about him is the stuff of legends told centuries
after his life” (p. 148, n. 87), and the assumption of anti-Christian polemic in Babylonian
texts “is seemingly a questionable practice, since the Palestinian texts were redacted
in a society in which Christianity was dominant, while the Babylonian texts weren’t,
and this ought to make a big difference” (p. 147. n. 86), still Boyarin counters by citing
current research, regarding orientation to Christianity in Babylonia, and flirts with the
idea that “the stories of Rabbi Eli‘ezer’s arrest and of Ben Dama’s near fall into heretical
behavior, appear as doublets, the formal similarity between the two death stories in the
Babylonian Talmud—the use of the phrase (attested in only one other place) ‘his soul
left him in purity’—also suggests that the two were once a pair in an earlier corpus,
apparently a variation of the two forms in which the stories appear together in the early
Palestinian texts” (p. 40). The inclusion of the death-bed account with the arrest story
bears comment. Since we have the Yerushalmi version of the death-bed story (quoted in
the text above), and the cucumber magic is absent there, and is thus clearly inserted by
the Bavli through reworking the tannaitic account, that reworking cannot support the
Christianity theory, in that the very acceptance and high regard of magical practice in the
amoraic period, especially regarding Rabbi Eliezer and Rabbi Yehoshua, is a simpler and
sufficient statement and explanation of the background of this conceptualization.
60 Friedman

bSanh 68a ‫ב׳ סנהדרין סח ע״א‬

But how might [Rabbi Eliezer] do so? ?‫היכי עביד הכי‬
Did we not learn that if he actually per- !‫ העושה מעשה חייב‬:‫והאנן תנן‬
forms the deed he is liable?
If it is only to teach, it is different. For ‫ דאמר מר ״לא תלמד לעשות״‬.‫להתלמד שאני‬
it has been said: “Thou shalt not learn ‫ אבל אתה‬,‫ לעשות אי אתה למד‬.)‫(דברים יח כ‬
to do [after the abominations of these .‫למד להבין ולהורות‬
nations]” (Deut 18:20). You may not learn
in order to practice, but you may learn in
order to understand and teach.

The price the Bavli pays for rescuing Rabbi Eliezer requires trampling the
simple meaning of this wonderful derashah. We find its original form in
SifDeut 170:79

SifDeut 170 ‫ספרי דברים קע‬

“For thou art entering the Land which the ‫״כי אתה בא אל הארץ אשר ה׳ אלהיך נותן לך‬
Lord thy God is giving you. Thou shalt ‫לא תלמד לעשות כתועבות הגוים ההם״ (דברים‬
not learn to do after the abominations of .)‫יח ט‬
these nations” (Deut 18:9):
One may have thought that you are not ‫יכול אי אתה רשיי ללמוד להורות ולהבין? תלמוד‬
permitted to study, teach, or understand. ‫ לעשות אי אתה למד אבל אתה‬.‫ ״לעשות״‬:‫לומר‬
Therefore the verse spells out: “to do.” .‫למד להורות ולהבין‬
You may not learn in order to practice,
but you may learn in order to teach and
understand.

This rabbinic Magna Carta of academic freedom holds that one may research
and teach about a forbidden act, but not perform it, even in a closed instruc-
tional setting.80 Following our conclusions above, we now know that Rabbi

79 In Hoffmann’s Midrash Tannaim (p. 109) he uses large type to represent the reconstituted
Mekhilta to Deuteronomy (note 4: “Sifre in a different formulation”) as follows: ‫לא תלמד‬
‫ למד אתה על מנת לישב בדין‬:‫ ר׳ יהושע אומר‬.‫לעשות‬. I doubt that this is a tannaitic
formulation. Perhaps it is a reworking of Rav Hai Gaon’s comment cited by the Tosafot at
bMen 65a: ‫( בעלי כשפים—פירש רב האי גאון כדי לידע את הדין‬cf. Rav Hai’s responsum,
Otzar Ha-Geonim, Sanhedrin [Jerusalem 1966] 409, par. 940).
80 Contra Bohak, Ancient Jewish Magic, who, with reference to our sugya, concludes:
“Returning to the Babylonian Talmud, we might be tempted to suggest that the final
statement, that while one may not practice magic, one may learn it and teach it (including
hands-on experimentation [italics added—S.F.]), is just an ad hoc invention intended to
save R. Eliezer from the grievous charges, or just a Babylonian add-on to earlier Palestinian
TALMUDIC PASSAGES ABOUT MAGIC 61

Eliezer did not uproot cucumbers, even to show Rabbi Aqiva how it is done.
The permission to do so is an invention of the editorial voice reinterpreting the
Sifre in order to resolve a contradiction, just as the very magical act documented
derives from an expansion of the Tosefta by a Babylonian tanna-teacher.

3 Sefer Yetzirah and the Creation of a Calf

The calf-creating episode is a remarkable tale, and much can be learned from it:

bSanh 65b (= 67b) )‫ב׳ סנהדרין סה ע״ב (= סז ע״ב‬

Rav Hanina and Rav Oshaia ‫רב חנינא ורב אושעיא‬
spent every Sabbath eve in studying the ,‫הוו יתבי כל מעלי שבתא ועסקי בספר יצירה‬
Book of Creation,
by means of which they created a third- .‫ ואכלי ליה‬,‫ומיברו להו עיגלא תילתא‬
grown calf and ate it.

Joshua Levinson sees in this studying, creating, and then eating sequence a
Babylonian transformation of a Yerushalmi passage, where gourds are changed
into stags:

ySanh 25d: Rabbi Yehoshua ben Hananya said: I can take gourds and
melons and transform them into stags and deer, and these then produce
more stags and deer.

bSanh 67b: Rav Hanania and Rav Hoshaia would study the Book of
Creation every Sabbath eve and create a third-grown calf and eat it.

traditions. This, however, clearly is not the case, for the assumption that magic is a body
of knowledge which requires detailed study, and the claim that one indeed may (or even,
must) study it, are central components of the rabbinic view of this subject from very early
times, and are attested both in Palestine and in Babylonia” (p. 360). Here (note 19) Sifre
Deuteronomy cited above is referenced as presenting “the same exegesis of Deut 18.9” as the
Bavli’s defense of Rabbi Eliezer. Thus Bohak makes no distinction between the tannaitic
and Babylonian forms of this tradition. I have claimed that “hands-on experimentation” is
indeed “just an ad hoc invention intended to save R. Eliezer from the grievous charges . . . a
Babylonian add-on to earlier Palestinian traditions.” I do not understand Levinson’s
comment: “Thus the entire discussion in the Bavli concludes with an editorial (stammaitic)
attempt to circumvent the biblical and tannaitic prohibition on magic by declaring that
while it is prohibited to perform magical acts, ‘you may study it in order to understand it’ ”
(Levinson, “Enchanting”). Certainly the ruling in this form is tannaitic.
62 Friedman

In the Yerushalmi, R. Yehoshua boasts that he can transform melons into

animals. The Bavli relates how the sages created a calf every Sabbath
eve by studying the Book (or Laws) of Creation. If it is sufficient for the
Yerushalmi to praise the sages’ magical powers, the Bavli transform[s]
this culinary feat into a type of study.81

It is certainly reasonable to see this Bavli account as a reworking of an earlier

[Palestinian] aggadah,82 applying to it rabbinization and intellectualizetion
tendencies.83 However, I would suggest that the point of departure and base
text for this rewriting is rather the following account, dealing directly with calf
magic:

ySanh 7:19, 25c–d ‫ כה ע״ד‬,‫י׳ סנהדרין ז יג‬

Rabbi Yannai said: I was walking in the ‫ מהלך הוינא בהדא איסרטא‬:‫אמר רבי ינאי‬
square of Sepphoris and I saw a heretic ‫דציפורי וחזית חד מיניי נסיב צריר וזרק לי׳‬
take a stone and throw it into the air, . . . ‫ והוה נחת ומתעביד עגל‬,‫לרומא‬
and when it came down it turned into a
calf . . .
Rabbi Hinnena bar Rabbi Hananiah said: ‫ מטייל הוינא באילין‬:‫אמ׳ ר׳ חיננא ביר׳ חנניה‬
I was walking in the Gufata of Sepphoris ‫גופתא דציפורין וחמית חד מיניי נסב חדא‬
and I saw a heretic take a skull and throw ‫ והיא נחתא ומתעבדא‬,‫גולגלא וזרקה לרומא‬
it into the air, and when it came down it .‫עגל‬
turned into a calf.84
I came and reported this to my father. He ,‫ אין אכלת מינה‬:‫ אמ׳ לי‬.‫אתית ואמרת לאבא‬
told me: If you ate from it, it is a real deed. .‫ אחיזת עינים הוא‬,‫ ואילא‬,‫מעשה הוא‬
If not, it was merely an illusion.

Rabbi Hananiah the father of Rabbi Hinnena may not appear elsewhere in
talmudic literature,85 but this story suffices to demonstrate the sharp wit he
used as a method of instruction: “Look my son, don’t believe it unless you
can eat it. I don’t think you’ll get even one bite, so you don’t have to worry if
it is kosher or not.” I think this is the source for the Bavli account. Both deal

81 Levinson, “Enchanting,” 72. Regarding the intellectual character of religious Sasanian texts
in general and in the Bavli, see: S. Secunda, “Reading the Bavli in Iran,” Jewish Quarterly
Review 100 (2010) 313.
82 See in general, Friedman, “Historical Aggadah.”
83 In this account “magic itself is appropriated as a type of Torah” (Levinson, “Enchanting,” 72).
84 Alexander explains to us how it is done: “The conjurer distracts the attention of the
audience, here by throwing an object into the air, which allows him or an accomplice to
introduce unobserved a new object into the scene” (“Conjuring,” 15).
85 See A. Hyman Toldoth Tannaim ve-Amoraim (London 1912) 474 [Hebrew].
TALMUDIC PASSAGES ABOUT MAGIC 63

with a calf 86 created magically, and eaten, but in the Bavli the humor is gone.
Whatever a heretic can do, the rabbis can do better. They do it as rabbis, in a
context of studying the holy books. They produce the calf in order to fulfill the
religious duty of eating a scrumptious Shabbat banquet.87
The similarity of the two accounts goes far beyond producing a calf and the
idea of eating it. Even the names are similar. The Yerushalmi’s account takes
place in Sepphoris, between Rabbi Hinnena and his father Rabbi Hananiah.

3.1 Rabbi Hanina and Rabbi Oshaia

The two researchers of mystical lore in the Bavli are Rav Hanina and Rav Oshaia.
Although each of these names applies to more than one amora, the pair are
mentioned together several times in the Bavli88 with style overlapping our
passage.89 They are described as acting together,90 largely in a Babylonian set-
ting, but sometimes in Palestine. The following tradition is an example thereof:

Said Rava: [. . .] Rav Hanina and Rav Oshaia, who were cobblers in Eretz
Yisrael and dwelt in a street of harlots and made shoes for harlots;91 they
[the harlots] looked at them, but they [these scholars] would not lift their
eyes to look at them, and their [the harlots’] oath was: By the life of the
holy Rabbis of Eretz Yisrael (bPes 113b).92

86 For the Golden Calf as a living magical creation in later sources see Lieberman, Greek,
113–4.
87 See Torat Hayyim to bSanh 67b; Margaliot HaYam to bSanh 65b.
88 See B. Kosowski, Thesaurus Nominum quae in Talmude Babilonoco Reperiunter (Jerusalem
1976–83) vol. 1, 125; vol. 2, 509 [Hebrew]. Regarding this pair, see Albeck, Introduction,
221–2, 241–3. Albeck places them in the third generation of Palestinian amoraim.
89 At bMak 19b they are sitting together, see Appendix D.
90 M. Margalioth takes them as brothers (Encyclopedia of Talmudic Sages and Geonim
[Tel Aviv 1995] 62–3 [Hebrew]). This conclusion may simply be a deduction based on
chronologists cited by Hyman, Toldoth, 500–1. Although he could find no basis for this
opinion, Hyman attempted to create such a source through an emendation, which,
however, has no basis.
91 See R. Rabbinovicz, Variae Lectiones in Mischnam et in Talmud Babylonicum ad loc, 351, n. 8.
92 In Margalioth’s Encyclopedia, 63 we find: “The two of them went up to visit ruins of
Jerusalem. They didn’t enter it [the city—S.F.] but sat next to one of its gates” (‫שניהם עלו‬
‫ ולא נכנסו לתוכה אלא ישבו על יד אחד שעריה‬,‫)לבקר בירושלים החרבה‬. This conclusion,
based on bMak 19b, is not warranted, and flows from an inferior textual tradition and the
commentary of RIVAN (R. Yehuda bar Natan as printed in place of Rashi ad loc.). The
correct understanding is presented in the Soncino translation: “R. Hanina and R. Hoshaia
sat and raised the [following] question: What would be the case [regarding redeeming
Second Tithe—S.F.] [where a pilgrim had just reached] the very entrance to Jerusalem”
(see Rabbinovicz, Variae Lectiones, 35, n. ‫ ;ת‬M. Friedmann, Babylonischer Talmud: Tractat
64 Friedman

At bSanh 14a they are described as disciples of Rabbi Yohanan, to whom he

failed to grant ordination. This passage fits the above-mentioned genre, and
it has consequently been concluded that these are the “Rav Hanina and Rav
Oshaia” of the calf episode.93
However, at least regarding some of the actions or dicta attributed to this
pair, the possibility may be raised that they refer to an earlier Rabbi Oshaia
and Rabbi Hanina. These sages, both of outstanding stature and fame, do not
appear together as a pair, and are not exact (but probably overlapping) con-
temporaries. Rabbi Hanina ([bar Hama], called “the Great”) belonged to the
earlier group of first-generation amoraim in Palestine,94 and Rabbi Oshaia
(also called “the Great” and “the father of the Mishnah,” ‫)אבי המשנה‬95 belonged
to the second group of that generation.96 In bPes 87b we find:

And this is what a certain sectarian said to Rabbi Hanina: . . . Said he to

him: If you agree, a disciple will debate it with you. [Thereupon] Rabbi
Oshaia debated it with him.

In place of “Rabbi Hanina,” an alternate tradition has “Rabbi Yehudah Nasia.”97

This latter tradition was adopted by Lieberman. Thus he took Rabbi Oshaia
as the sage of that name belonging to the earlier pair (Oshaya the Great).98
On the other hand, according to the reading “Rabbi Hanina,” we can certainly
view this as an occurrence of the later pair (of the calf episode)!99
Rabbi Hanina (bar Hama) and Rabbi Oshaya (the Great), a renowned sage-
pair, in the literary tradition, may, under certain conditions, replace a tradition

Makkoth, kritische Edition [Vienna 1888] 49, n. 10). The primary textual witnesses are
recorded below, Appendix C. Regarding bPes 87b, see Appendix D.
93 Hyman, Toldoth, 501; Kosowski, Thesaurus, loc. cit.
94 C. Albeck, Introduction to the Talmud Babli and Yerushalmi (Tel Aviv 1969) 155–6 [Hebrew].
95 We find at least three times in the Yerushalmi: ‫כך פירשה רבי הושעיה אבי המשנה‬.
96 Albeck, Introduction, 163–4; Epstein, Introduction, 40.
97 See detailed discussion in Appendix E.
98 “R. Juda the Patriarch (III c.) commissioned R. Oshaia to debate with this gentile. The
latter, who apparently was in a position to harm the Jews, was a Roman official who lived
in the same place as R. Oshaia, the head of the school in Caesarea” (Lieberman, Greek, 141).
The Rabbi Oshaia at Caesarea was the earlier of that name, namely, of the first period.
99 And so it was always understood by those who used the printed version only. Lieberman
seems to interpret this tradition as involving the earlier pair: “The reading of the editions
and ms. Munich is erroneous, for R. Hanina would not call R. Oshaia ‘a disciple,’ whereas
the Patriarch, R. Juda the Second, by virtue of his office may have permitted himself to
term R. Oshaia ‘disciple’ in order to show the gentile that even his pupil will be able to give
him the right answer” (Lieberman, Greek, 141, n. 95).
TALMUDIC PASSAGES ABOUT MAGIC 65

involving the first pair, a flexibility we must be aware of in interpreting our

calf episode.

3.2 The Calf

Another marker of the calf-creation story as being a literary formation is the
use of an exceptionally literary description of the animal they created. ‫עגלא‬
‫תילתא‬, or ‫לתא‬ ָ ‫ ִעיגְ ָלא ֻת‬as in the Yemenite MS,100 is exactly the Peshitta’s transla-
tion (‫ )עגלא תולתא‬of ‫ ֶעגְ ָלה ְמ ֻׁש ֶּל ֶׁשת‬in Gen 15:9 )and in 1Sam 1:24101 and of ‫ֶעגְ ַלת‬
‫ ְׁש ִל ִׁשּיָ ה‬in Isa 15:5, and the same in Jer 48:33), a form of translation not found in
the Jewish Targumim to those verses. The phrase is used commonly as a deli-
cacy in the Bavli’s accounts of rabbis being served meals, and here it is a literary
borrowing from those contexts.102

100 Lieberman Institute Website, http://www.lieberman-institute.com/.

101 MT has ‫ֹלׁשה‬ ָ ‫ ְּב ָפ ִרים ְׁש‬, but Qumran fragments and versions have ‫בפר בן ]בקר משלש‬
)‫(=מ ֻׁש ָּלׁש‬
ְ , see Appendix E.
102 The forms ‫תילתא‬/‫ תולתא‬are not recorded in talmudic literature outside of this phrase,
which therefore must be taken as a fossilized literary usage in Aramaic, based on the
Akkadian šulušû = “three-year-old,” used regarding animals, see CAD, Š 3, 2634. There we
find: “said of cattle: six minas of silver [. . .] purchase price of three-year-old oxen (text:
calves).” The final remark indicates that even though a three-year-old is not really a
calf, Akkadian literary usage has it thus as a fixed phrase (‫)ביטוי כבול‬, which was taken
over in Syriac and used to translate ‫( ֶעגְ ָלה ְמ ֻׁש ֶּל ֶׁשת‬Gen 15:9, cf. Ibn Ezra) and taken
over in Babylonian Aramaic, probably from one of the eastern Aramaic dialects as a
gastronomic phrase, much as French is used today in English to signify “a beef delicacy.”
This meaning, “three-year old,” is corroborated by the explanation given by Lieberman to
the correct reading of mBM 15:9 (S. Lieberman, Hellenism in Jewish Palestine [New York
1962] 198). Thus we have a clear determination of the root meaning, a term with which
the commentators struggled, and to which they gave various explanations: grown a third
of its growth (cf. Rashi, bBM 68a), third born of its mother, or in general, of good quality.
Contemporary writers also made use of traditional ideas. Schäfer writes: “Rav Hanina
and Rav Oshaya spent every Sabbath eve in studying . . . because of our iniquities Rava
could not create a fully fledged man (his creature lacked speech, the most important
characteristic of a human being), let alone a world, and Rav Hanina/Rav Oshaya could
not even create a fully fledged calf. The latter example again is an enhancement of the
first, this time an ironical one: despite their fervent study of the instructions concerning
creation, they did not succeed in creating a man, let alone a world, they only succeeded
in producing a calf one third of the natural size of a calf. What could they do with it? They
just ate it-which certainly is the climax of the irony because how did they do this? Would
they have cooked it on the eve of Sabbath?” (P. Schäfer, “The Magic of the Golem: The
Early Development of the Golem Legend,” Journal of Jewish Studies 46 [1995] 253). May
I point out that “Sabbath eve” (‫ )מעלי שבתא‬refers to Friday before the start of Shabbat,
see S. Friedman, “Evening Metaphors” (‫ ״אור ליום ארבעה עשר״‬:)‫)״לשון המועדים (א‬
in: M. Bar-Asher et al. (eds.), Studies in Talmudic and Midrashic Literature in Memory of
66 Friedman

Thus we have all the hallmarks of a reworked literary source, where ele-
ments from the original and elsewhere are used as building blocks in a new
creation,103 in order to recast the sectarian’s purported success in bringing life
into being (in the Yerushalmi account) as acts performed by rabbis in a wholly
holy context, and as a meritorious religious act.104
There are many talmudic accounts of rabbis’ magical feats outdoing sectar-
ian rivals (heretics, minim).105 Our suggested understanding of the calf episode
uncovers a new weapon in this contest: intertextuality. Without even having to
appear in the same scene, these rabbis outdo their rivals intertextually, taking
aim at someone as though through a time tunnel across the Yerushalmi/Bavli
divide.

3.3 Sefer Yetzirah

Perhaps the most interesting cultural datum in this calf-creation account
involves the “Book of Creation”—‫ספר יצירה‬. The power of creation was
achieved, according to the commentators,106 through combinations of letters

Tirzah Lifshitz (Jerusalem 2005) 475–519 [Hebrew]. Idel, has correctly translated: “a calf
of three years old” (M. Idel, Golem: Jewish Magical and Mystical Traditions on the Artificial
Anthropoid [Albany 1990] 31).
103 See the primary examples of this phenomenon in Friedman, “Historical Aggada,”
etc., and recently idem, “Three Stories,” Oqimta 1 (2013) 133–9. This approach has been
applied successfully in subsequent studies by Rubenstein, Stories; Tropper, and others.
See A. Tropper, Like Clay in the Hands of the Potter: Sage Stories in Rabbinic Literature
(Jerusalem 2011) 11–26 [Hebrew], regarding three methodological approaches.
104 Alexander compares these passages regarding eating as a test of real results, without
discussing literary dependencies: “But how does one distinguish between ‘real’ magic and
illusion? The Talmud offers two rather curious tests. The first is that if the object pro­duced
is edible then the magic is real. This test emerges most clearly from the continuation of
the story, quoted above, about the skull that turned into a calf. When Rabbi Hinnena
recounts the incident to his father, Rabbi Hananiah, the latter comments: ‘If you could
have eaten it [the calf], then it would have been real magic, but if not, then it was only a
case of holding the eyes’ ” (ySanh VII, 19 [25c.9 from bottom]). The same idea lies behind
the Bavli’s insistence that Hanina and Oshaia ate the calf which they had made: the eating
proved that it was a real calf, produced by real magic” (Alexander, “Conjuring,” 24).
105 And they have been dealt with in detail in scholarly research. See G. Bohak, “Magical
Means for Handling Minim in Rabbinic Literature,” in: P. J. Tomson and D. Lambers-
Petry (eds.), The Image of the Judeo-Christians in Ancient Jewish and Christian Literature
(Tübingen 2003) 267–79; Levinson, “Enchanting.”
106 Rashi 65b, 67b; Yad Ramah 65b. On Sefer Yezirah (and combining letters), see Rashi, bBer
55a s.v. ‫ ;אותיות‬bShab 104a s.v. ‫אמר שר‬, bEruv 63a s.v. ‫עגלא‬, bHag 13a s.v. ‫סתרי‬, bMen 29b
s.v. ‫אחת‬.
TALMUDIC PASSAGES ABOUT MAGIC 67

from the divine name. The creator Himself, according to amoraic aggadah,
created the world through the use of the letters yod and he.107 Gershom
Scholem was open to reading our talmudic passage as a source for historical
information, at least in terms of basing upon it the existence of the medieval
Sefer Yetzirah already in talmudic times: “We should not dismiss out of hand
the possibility that the hilkhot yezirah mentioned in Sanhedrin 65b and 67b
could be one early version of this text.”108 Joseph Dan109 and others leave the

107 GenR 12:4 (Theodor Albeck, pp. 107–9).

108 Encyclopedia Judaica 10 (Jerusalem 1974) 507; cf. 16, 785; G. Scholem, J. Garb, and M. Idel,
“Kabbalah,” Encyclopedia Judaica 11 (Detroit 2007) 595. Thus he follows Louis Ginzberg,
who wrote: “The title of two esoteric books. Of these the older is also called “Hilkot
Yezirah” (Rules of Creation), and is a thaumaturgical work that was popular in the
Talmudic period. On the eve of every Sabbath, Judah ha-Nasi’s pupils, Rab Ḥanina and
Rab Hoshaiah, who devoted themselves especially to cosmogony, used to create a three-
year-old calf by means of the ‘Sefer Yezirah,’ and ate it on the Sabbath (Sanh. 65b, 67b)”
Jewish Encyclopedia 12 (1905) 602. Scholem’s position and style is based on his “Jezira,”
Encyclopedia Judaica (1934), 107, where he writes: “Darüber hinaus ist, trotz aller hiergegen
vorgebrachten Bedenken, zu erwägen, ob nicht der talmudische Bericht in Sanh. 65b, 67b
sich auf den J.-Text oder dessen Prototyp bezieht; es heisst dort, dass R. Chanina und R.
Oschaja (im 4. Jh.) an jedem Freitag das ‘Buch der Schöpfung,’ oder nach anderen LAA: die
‘Halachot der Schöpfung,’ studierten und dadurch ein Kalb erschufen, das sie verzehrten.
Ein solcher thaumaturgischer Gebrauch des J.-Buches wäre durchaus denkbar.” Scholem
makes it clear here that he was being pulled in two directions, and that raising this
possibility was against his better sense of kabbalistic history. In his Major Trends in Jewish
Mysticism (already in 1941, p. 75) he uses the same general language as in the encyclopedia
articles (“Written probably between the third and sixth century”), but does not mention
his feeling about the relationship to the talmudic passage. I do not know whether this is
just for brevity’s sake or whether he had changed his mind on this matter, and the English
Encyclopedia Judaica article simply drew from the German, with less revision. In any case,
in 1960 he wrote: “That the Book Yetsirah should be mentioned in this passage does not
strike me as quite so impossible as numerous authors have assumed. We do not know the
exact date of this enigmatic text . . . We can only be sure that it was written by a Jewish
Neo-Pythagorean some time between the third and the sixth century” (G. Scholem, On
the Kabbalah and its Symbolism [New York 1965] 167). “As I shall explain elsewhere, I now
[1960—S.F.] incline toward the earlier dating” (ibid., 167 n. 3). “If Jewish esoterics as early
as the third century—in case the Book Yetsirah really comes from this period—believed
Abraham to be capable of such miraculous creation on the strength of his insight into
the hilkhoth yetsirah, we shall be justified in drawing a parallel between these views and
certain others held at roughly the same time” (p. 172). “it is not at all impossible that it is
referred to by this name in the Talmud” (G. Scholem, Origins of the Kabbalah [ed. R. J. Z.
Werblowsky; translated from the German by A. Arkush; Philadelphia 1987] 25).
109 For his most recent discussion see next note. Previously he gave the subject more
nuanced treatment. See J. Dan, Jewish Mysticism in Late Antiquity (Northdale NJ 1998) 155,
where he opens with an assessment of the authorship as “probably in the third century,”
68 Friedman

time of composition of Sefer Yetzirah, as “unknown, but the general thrust of

their discussions is to throw into question the possibility of a talmudic prov-
enance entertained by Scholem for this work which went unrecorded before
the ninth century.”110 Liebes (2000) was absolutely certain that the talmudic
account refers to the same Sefer Yetzirah known since Saadia.111 A comprehen-
sive and convincing evaluation was penned by Ezra Fleischer, arguing from the
cultural milieu in which Sefer Yetzirah credibly fits, that there is no compelling

even though this is followed by a long series of surprises over, and attempts to explain,
its being ignored for six centuries. He also gives our talmudic passage general credence:
“This problem is closely related to the question of whether the reference to ‫הלכות יצירה‬
in the Talmud is indeed connected with our Sefer Yezira, but it is a rather com­plicated
relationship, because we should distinguish between the possibility that Rabbi Hanina
and Rabbi Hoshaia (and, by implication, Rava as well), indeed used the Sefer Yezira in a
version close to the one we have before us . . . If these scholars in the fourth century not
only knew Sefer Yezira but studied it and made practical use of it, [they] reflect a positive
atti­tude toward this work and making this attitude known and even famous by the ‫עגלא‬
‫ תולתא‬they presumably created . . .” (p. 157).
110 J. Dan, History of Jewish Mysticism and Esotericism vol. 2 (Jerusalem 2009) 554–61; idem,
Jewish Mysticism in Late Antiquity, 155–70. I would agree with Peter Schäfer that “in the
case of the creation of the calf, when the Hilkhot of Sefer Yeşirah are mentioned (whatever
this may be) there is no evidence that the Sefer Yezirah in the technical sense is alluded to,
let alone the technique of the permutation of letters) . . . I obviously do not subscribe to its
early, pretalmudic dating” (“Magic of the Golem,” 254–5). Cf. I. Gruenwald, “Some Critical
Notes on the First Part of SĒFER YEZĪRĀ,” REJ 132 (1975) 475–512.
111 Y. Liebes, Ars Poetica in Sefer Yetsira (Jerusalem 2000) 67–9; 231–2 [Hebrew]: “I see no
reason to doubt that Sefer Yetzira mentioned here [in bSanh] is our SY” (p. 67); “the
mention of the book by its name, together with a reliable description on its contents . . . is
in my mind an absolute philological proof, which supplies the entire measure of certainty
required for philological demonstrations in general, to establish that the book SY existed
in the hands of the Amoraim” (p. 231). Aslanoff, in his review of Liebes’ book, accepts
the premise that the talmudic passage refers to Sefer Yetzira, questioning only the dating
of the passage: “The mention of Sefer Yetsira in BT Sanhedrin 65b does not constitute
sufficient evidence for an early dating of the mystical treatise. It only provides a terminus
ad quem, which is the 4th century CE” (C. Aslanoff, “Review of Liebes, Yetsira,” Tarbiz
71 (2002) XII [Hebrew]). Liebes, in a previous study referring to the mention of double
resh in Sefer Yetzira, concluded: “These instances all suggest that Sefer Yezira was written
under Greek influence, that is to say around the third century C.E., and in a provenance
such as Palestine” (Y. Liebes, “The Seven Double Letters BGD KFRT: On the REISH and the
Background of Sefer Yeżira,” Tarbiz 61 (1992) X [Hebrew]). This conclusion was challenged
by Morag, see next note.
TALMUDIC PASSAGES ABOUT MAGIC 69

reason to date the work long before the time Saadia came across it, in the first
third of the 10th century.112
Although the story of Rav Oshaia and Rav Hanina creating a calf can hardly
be considered historical, in any case the story may have inspired the name of
the later work Sefer Yetzirah. Scholem remarks that “[i]n early manuscripts
it is called Hilkhot Yezirah (‘Halakhot on Creation]’), and later Sefer Yezirah.”113
I would like to add that this fact has a tantalizing correspondence to the
range of variant readings preserved for bSanh 65b and 67b as presented in the
following table:114

65b
‫ועסקי בהילכות יצירה‬ ‫יתבי כל מעלי דשבתא‬ ‫הוה‬ Klosterneuburg 127–128
‫עסקי בהלכות יצירה‬ ‫יתבי‬ ‫כי הוו‬ ‫הרצוג‬
‫עסקי בהלכות יצירא‬ ‫יתבי בכל מעלי שבת׳‬ ‫פ‬
‫ועסקי בספר בריאה‬ ‫יתבי כל מעלי שבתא‬ ‫הוו‬ ‫ר‬
‫והוו עסקי בהלכו׳ יציר׳‬ ‫יתיבי כל מעלי שבת׳‬ ‫מ‬
‫ועסקי בספר יצירה‬ ‫יתבי כל מעלי שבתא‬ ‫הוו‬ ‫דפוס ברקו‬

67b
‫הוו עסקי בהילכו׳ יצירה‬ ‫כל מעלי שבתא‬ Klosterneuburg 127–128
>. . . < ‫הוו‬ ‫כל מעלי שבתא‬ Ebr 602
‫הוו עסקי בספר יצירא‬ ‫כל מעלי שבתא‬ ‫פ‬
‫הוו עסקי בהילכות יצירה‬ ‫כל מעלו שבתא‬ ‫ר‬

112 E. Fleischer, “On the Antiquity of Sefer Yeżira: The Cyrilian Testimony Revisited,” Tarbiz 71
(2002) 423–5, 432 [Hebrew]; “In this author’s opinion, the possibility that Sefer Yezira was
composed in Rav Sa‘adya’s times or shortly before should be seriously considered” (p. 8).
Fleischer added: “This conclusion converges with the opinions of Y. Tzvi Langermann and
Steven Wasserstrom” (= S. M. Wasserstrom, “Further Thoughts on the Origins of ‘Sefer
Yeṣirah’,” Aleph 2 [2002] 201–21; Y. T. Langermann, “On the Beginning of Hebrew Scientific
Literature and on Studying History Through ‘Maqbilot’ [Parallels],” Aleph 2 [2002] 169–89;
see also S.M. Wasserstrom, “Sefer Yeṣira and Early Islam: A Reappraisal,” Jewish Thought
and Philosophy 3 [1993] 1–30). Similarly Morag, in response to Liebes, placed the work as
probably in a Babylonian provenance, under Arabic influence, fifth or sixth century at the
earliest (S. Morag, “Response: ‘On the Seven Double Letters BGD KFRT, and the Names
SARAH-SARAI, AVRAM-AVRAHAM’,” Tarbiz 63 (1993) 137–42 [Hebrew], esp. p. 140). Cf.
N. Aloni, “The Time of Composition of Sefer Yezirah,” in: I. Weinstock (ed), Temirin: Texts
and Studies in Kaballa and Hasidism (Jerusalem 1984) 41–50 [Hebrew].
113 Cf. Gruenwald, “Critical Notes,” 475–6, regarding the name of this work, including “The
Letters of Abraham the Patriarch” (‫)ספר אותיות דאברהם אבינו‬.
114 Some of which have been noted in Schäfer, “Magic of the Golem,” 253, n. 20.
70 Friedman

‫הוו עסקי בהלכו׳ יציר׳‬ ‫כל מעלי שב׳‬ ‫מ‬

‫הוו עסקי בהלכו׳ יצירה‬ ‫כל מעלי שבתא‬ ‫ד׳ ברקו‬

The version ‫ הילכות יצירה‬definitely seems original.115 The form ‫ ספר יצירה‬is in a
minority, and is most probably a gloss, ostensibly with an Ashkenazic connec-
tion. In fact, each of the witnesses recording ‫ ספר‬do so only in one (and not
always the same) of the two occurrences, demonstrating inconsistent glossing116
(and ‫ ספר בריאה‬would indicate a further change during the glossing proce-
dure). It is possible that the name of our Sefer Yetzirah was patterned after
the fanciful calf-creation account in the Bavli117 and its allusion to ‫הלכות יצירה‬.
Later, when the medieval work came to be called ‫ספר יצירה‬, this form of the
name was glossed into the Talmud MSS in Ashkenaz.

4 Conclusion

The calf vignette is a fanciful literary creation, inspired by the sectarian prow-
ess in calf magic as described in the Yerushalmi, and the witticism voiced there
that it is real only if you can eat it. The narrative has rabbis outdo the sectar-
ian, and create an edible calf, indeed a delicacy, through holy halakhic means.
This is a polished literary creation, polemic rhetoric in the form of narrative.
Scholarship can devote itself to analysis of its composition, and need not

115 This language is also quoted by early authorities: Saadia, Emunot veDeot, First Article;
Geonic Responsa, Harkavi, par. 29; Sekhel Tov to Exodus 7; Yalqut Shimoni, Exodus, par. 182;
Yad Ramah, Sanh 25b; Sefer Eshkol, Albeck, 7a; etc. “Sefer” is in the minority, but used in
the contem­porary editions of Rashi, Ibn Ezra, Radaq, Nahmanides, and many more.
116 It is thus obvious that the different readings originate in the trans­mission of the text (i.e. a
lower-critical phenomenon), and not as an original difference between the two passages
(which would be a higher-critical issue), contra Stratton, who claims that one of the
passages relates to the other as “another version of this report, [in which] Rav Hanina and
Rav Oshaia are said to be studying from the ‘book of creation’ (sefer yetzira) when they
create the calf (b. San. 65b).” K. Stratton, “Imagining Power: Magic, Miracle, and the Social
Context of Rabbinic Self-Representation,” Journal of the American Academy of Religion
73 (2005) 361–93 (p. 366 n. 13).
117 I see that Gruenwald has written: “The name Sēfer Yeżīrā given to the whole of the book
may be due to its identification with the ‫ ספר יצירה‬and ‫ הלכות יצירה‬mentioned in
B. Sanh. 65b and 67b respectively. Yet, this identification has nothing substantial to justify
itself, and it is, therefore, noteworthy in this respect that Sa‘adya Gaon, who is the first
known commentator on the book, refers to it as ‘The Book of the Beginnings’ ” (“Critical
Notes,” 476).
‫‪TALMUDIC PASSAGES ABOUT MAGIC‬‬ ‫‪71‬‬

‫‪overly concern itself with documenting the appearance of Sefer Yetzirah or for-‬‬
‫‪mulating a theology for the sages mentioned here or for their contemporaries.‬‬
‫‪What we do document in this survey is the heroic status accorded by the‬‬
‫‪amoraim to magical rabbis, evinced in detail by the fabulous tales in the‬‬
‫‪Yerushalmi, and in the Bavli expressed in the theological stance annunciated‬‬
‫‪by Rabbi Hanina, through the attribution to Rabbi Eliezer of flashy cucumber‬‬
‫‪magic by reworking a tannaitic tradition, and elegantly presented in the calf-‬‬
‫‪creation episode ascribed to the sage-pair, Rav Hanina and Rav Oshaia.‬‬

‫ ‬ ‫‪Appendix A‬‬

‫ ‬ ‫)‪Synopsis with AdRN A 25 (above, n. 40‬‬

‫ב׳ סנהדרין סח ע״א‬ ‫אבות דרבי נתן א כה עמ׳‬ ‫י׳ שבת ב ה ע״ב‬
‫‪81–80‬‬
‫כשחלה רבי אליעזר נכנסו רבי‬ ‫כשחלה רבי אליעזר‬ ‫מעשה בי רבי ליעזר שהיה‬
‫עקיבא וחביריו לבקרו‪ .‬הוא‬ ‫גוסס‬
‫יושב בקינוף שלו והן יושבין‬
‫בטרקלין שלו‪.‬‬
‫ואותו היום ערב שבת היה‪,‬‬ ‫אמרו‪ ,‬אותו היום ערב שבת‬ ‫ערב שבת עם חשיכה‪.‬‬
‫היה ונכנס רבי עקיבא וחביריו‬
‫לבקרו והיה ישן בתוך חדרו‪,‬‬
‫והוא יושב בקינוף‪ ,‬והם יושבין‬
‫בטרקלין שלו‪.‬‬
‫ונכנס הורקנוס בנו לחלוץ‬ ‫נכנס הורקנוס בנו לחלוץ‬ ‫ונכנס הורקנוס בנו לחלוץ את‬
‫תפליו‪ .‬גער בו ויצא בנזיפה‪.‬‬ ‫תפילין שלו‪ ,‬ולא הניחו‪ ,‬והיה‬ ‫תפיליו‪ .‬אמר לו‪ :‬בני‪ ,‬הינחת‬
‫בוכה‪.‬‬ ‫מצות הנר‪ ,‬שהיא שבות וחייבין‬
‫עליה כרת‪ ,‬ובאתה לחלוץ‬
‫תפילין‪ ,‬שאינן אלא רשות‪ ,‬ואינן‬
‫אלא מצוה‪.‬‬
‫אמר להן לחביריו‪ :‬כמדומה אני‬ ‫ויצא הורקנוס ואמר לחכמים‪:‬‬ ‫יצא לו והיה צועק ואמר‪ :‬אוי לי‬
‫שדעתו של אבא נטרפה‪ .‬אמר‬ ‫רבותי‪ ,‬דומה לי שנטרפה‬ ‫שנטרפה דעתו של אבא‪ .‬אמר‬
‫להן‪ :‬דעתו ודעת אמו נטרפה‪.‬‬ ‫דעתו של אבא‪ .‬אמר לו‪:‬‬ ‫לו‪ :‬דעתך היא שנטרפה‪ .‬דעתי‪,‬‬
‫היאך מניחין איסור סקילה‬ ‫בני‪ ,‬לא דעתי נטרפה‪ ,‬אבל‬ ‫היא לא נטרפה‪.‬‬
‫ועוסקין באיסור שבות?‬ ‫דעתך נטרפה‪ ,‬שהנחת‬
‫הדלקת הנר שנתחייבת עליה‬
‫מיתה לשמים‪ ,‬והיית מתעסק‬
‫בתפילין‪ ,‬שאין אתה מתחייב‬
‫עליהן אלא משום שבות‪.‬‬
‫‪72‬‬ ‫‪Friedman‬‬

‫כיון שראו חכמים שדעתו‬ ‫כיון שראו חכמים שדעתו‬ ‫כיון שראו תלמידיו שהשיבו‬
‫מיושבת עליו‪ ,‬נכנסו וישבו‬ ‫מיושבת עליו‪ ,‬נכנסו וישבו‬ ‫דבר של חכמה‪ ,‬נכנסו אצלו‪.‬‬
‫לפניו מרחוק ארבע אמות‪.‬‬ ‫לפניו ברחוק ד׳ אמות‪.‬‬
‫אמר להם‪ :‬למה באתם? אמרו‬
‫לו‪ :‬ללמוד תורה באנו‪ .‬אמר‬
‫להם‪ :‬ועד עכשיו למה לא‬
‫באתם? אמרו לו‪ :‬לא היה לנו‬
‫פנאי‪ .‬אמר להן‪ :‬תמיה אני אם‬
‫ימותו מיתת עצמן‪ .‬אמר לו רבי‬
‫עקיבא‪ :‬שלי מהו? אמר לו‪:‬‬
‫שלך קשה משלהן‪ .‬נטל שתי‬
‫זרועותיו והניחן על לבו‪ ,‬אמר‪:‬‬
‫אוי לכם שתי זרועותיי שהן‬
‫כשתי ספרי תורה שנגללין‪.‬‬
‫הרבה תורה למדתי‪ ,‬והרבה‬
‫תורה לימדתי‪ .‬הרבה תורה‬
‫למדתי ולא חסרתי מרבותי‬
‫אפילו ככלב המלקק מן הים‪.‬‬
‫הרבה תורה לימדתי—ולא‬
‫חסרוני תלמידי אלא כמכחול‬
‫בשפופרת‪.‬‬
‫ולא עוד אלא שאני שונה שלש‬
‫מאות הלכות בבהרת עזה‪ ,‬ולא‬
‫היה אדם ששואלני בהן דבר‬
‫מעולם‪.‬‬
‫ולא עוד אלא שאני שונה‬
‫שלש מאות הלכות‪ ,‬ואמרי לה‬
‫שלשת אלפים הלכות‪ ,‬בנטיעת‬
‫קשואין‪ ,‬ולא היה אדם שואלני‬
‫בהן דבר מעולם‪ ,‬חוץ מעקיבא‬
‫בן יוסף‪ .‬פעם אחת אני והוא‬
‫מהלכין היינו בדרך‪ .‬אמר לי‪:‬‬
‫רבי‪ ,‬למדני בנטיעת קשואין‪.‬‬
‫אמרתי דבר אחד‪ ,‬נתמלאה כל‬
‫השדה קשואין‪ .‬אמר לי‪ :‬רבי‪,‬‬
‫למדתני נטיעתן‪ ,‬למדני עקירתן‪.‬‬
‫אמרתי דבר אחד‪ ,‬נתקבצו כולן‬
‫למקום אחד‪.‬‬
‫‪TALMUDIC PASSAGES ABOUT MAGIC‬‬ ‫‪73‬‬

‫אמרו לו‪ :‬הכדור והאמוס‬ ‫אמרו לו‪ :‬רבי כסת עגולה‬ ‫והיו שואלין אותו והיה אומר‬
‫והקמיע וצרור המרגליות‬ ‫והכידור והאימום והקמיע‬ ‫להן על הטמא טמא‪ ,‬ועל‬
‫ומשקולת קטנה מהו? אמר‬ ‫ותפילין שנקרעו‪ ,‬מהו מקבלין‬ ‫הטהור טהור‪.‬‬
‫להן‪ :‬הן טמאין וטהרתן במה‬ ‫טומאה? אמר להם‪ :‬מקבלין‬
‫שהן‪ .‬מנעל שעל גבי האמוס‬ ‫טומאה‪ ,‬והטבילו אותן כמו‬
‫מהו?—אמר להן‪ :‬הוא טהור‪.‬‬ ‫שהן‪ ,‬והזהרו בהן שהן הלכות‬
‫גדולות שנאמרו למשה בסיני‪.‬‬
‫והיו שואלין לו בטהרות‬
‫בטומאות במקוואות‪ .‬אמרו‬
‫לו‪ :‬רבי מה הוא זה? אמר‬
‫להם טמא‪ .‬מה הוא זה? אמר‬
‫להם טהור‪ .‬והיה משיב על‬
‫טמא טמא ועל טהור טהור‪.‬‬
‫אחר כך אמר רבי אליעזר‬
‫לחכמים‪ :‬תמה אני על‬
‫תלמידי הדור‪ ,‬שמא יענשו‬
‫מיתה לשמים‪ .‬אמרו לו‪ :‬רבי‬
‫מפני מה? אמר להם‪ :‬מפני‬
‫שלא באו ושמשו אותי‪ .‬ואחר‬
‫כך אמר לעקיבא בן יוסף‪:‬‬
‫עקיבא מפני מה לא באת‬
‫לפני ושמשת אותי? אמר לו‪:‬‬
‫רבי‪ ,‬לא נפניתי‪ .‬אמר לו‪ :‬תמה‬
‫אני עליך אם תמות מיתת‬
‫עצמך‪ .‬ויש אומרים‪ :‬לא אמר‬
‫לו כלום‪ .‬אלא כיון שאמר‬
‫רבי אליעזר לתלמידיו כך‪,‬‬
‫מיד נמס דמו בקרבו‪ .‬אמר‬
‫לו רבי עקיבא‪ :‬רבי מיתתי‬
‫במה‪ .‬אמר לו‪ :‬עקיבא‪ ,‬שלך‬
‫קשה מכולן‪ .‬נכנס רבי עקיבא‬
‫וישב לפניו ואמר לו‪ :‬רבי‪,‬‬
‫מעתה שנה לי‪ .‬פתח ושנה לו‪:‬‬
‫ש׳ הלכות בבהרת‪ .‬באותה‬
‫שעה הגביה רבי אליעזר‬
‫שתי זרועותיו והניחן על חזה‬
‫שלו ואמר‪ :‬אוי לי על שתי‬
‫זרועותי [שהן כשני] ספרי‬
‫תורות שנפטרין מן העולם‪.‬‬
‫שאם יהיו כל הימים דיו וכל‬
‫‪74‬‬ ‫‪Friedman‬‬

‫האגמים קולמוסים וכל בני‬

‫אדם לבלרין‪ ,‬אינן יכולין‬
‫לכתוב כל מה שקריתי ושניתי‬
‫ומה ששמשתי לחכמים‬
‫בישיבה‪ .‬ולא חסרתי מרבותי‬
‫אלא כאדם שטובל אצבעו‬
‫בים‪ .‬ולא חסרתי מתלמידי‬
‫אלא כדי שיכחול המכחול מן‬
‫השפופרת‪ .‬ועוד‪ ,‬שאני שונה‬
‫ג׳ מאות הלכות ב״מכשפה‬
‫לא תחיה״ (שמות כבי ז)‪.‬‬
‫ויש אומרים‪ :‬שלשת אלפים‬
‫הלכות‪ .‬ולא שאלני אדם בם‬
‫דבר מעולם‪ ,‬חוץ מעקיבא‬
‫בן יוסף‪ ,‬שפעם אחת אמר‬
‫לי‪ :‬רבי‪ ,‬למדני איך נוטעים‬
‫קישואין ואיך עוקרין אותן‪.‬‬
‫אמרתי דבר אחד נתמלאת‬
‫כל השדה קישואין‪ .‬אמר לי‪:‬‬
‫רבי‪ ,‬למדתני נטיעתן‪ ,‬למדני‬
‫עקירתן‪ .‬אמרתי דבר אחד‬
‫נתכנסו כל הקישואין כולם‬
‫למקום אחד‪ .‬אמר לו רבי‬
‫אליעזר בן עזריה‪ :‬רבי‪ ,‬מנעל‬
‫שעל גבי האימום [מהו]?‬
‫אמר לו‪ :‬טהור‪ .‬והיה משיב על‬
‫טמא טמא ועל טהור‪.‬‬
‫טהור‪ ,‬עד שיצתה נשמתו‬
‫בטהרה‪ .‬מיד קרע רבי‬
‫אליעזר בן עזריה [את] בגדיו‬
‫ובכה ויצא ואמר לחכמים‪:‬‬
‫רבותי באו וראו ברבי אליעזר‬
‫שטהור הוא לעולם הבא‪.‬‬
‫ויצאה נשמתו בטהרה‪.‬‬ ‫לפי שיצתה נשמתו בטהרה‪.‬‬ ‫ובאחרונה אמר טהור ונסתלקה‬
‫נשמתו אמרין ניכר רבי שהוא‬
‫טהור ‪. . .‬‬
‫עמד רבי יהושע על רגליו ואמר‪:‬‬ ‫נכנס רבי יהושע וחלץ את‬
‫הותר הנדר‪ ,‬הותר הנדר!‬ ‫תפיליו והיה מגפפו ומנשקו‬
‫ובוכה ואומר רבי רבי הותר‬
‫הנדר‬
TALMUDIC PASSAGES ABOUT MAGIC 75

‫למוצאי שבת פגע בו רבי‬ ‫לאחר שבת בא רבי עקיבא‬

‫ היה‬,‫עקיבא מן קיסרי ללוד‬ ‫ומצאו באריסרטיא שבא‬
‫מכה בבשרו עד שדמו שותת‬ ‫ מיד קרע את‬.‫מקסרי ללוד‬
.‫לארץ‬ ‫בגדיו ותלש בשערו והיה דמו‬
‫שותת ונופל לארץ והיה צועק‬
‫ובוכה ואומר אללי רבי עליך‬
‫אללי עליך מרי שהנחת כל‬
.‫הדור יתום‬
:‫פתח עליו בשורה ואמר‬ ‫פתח עליו בשורה ואמר‬
,‫אבי אבי רכב ישראל ופרשיו‬ ‫אבי אבי רכב ישראל ופרשיו‬ .‫רבי רכב ישראל ופרשיו‬
‫הרבה מעות יש לי ואין לי שול־‬ ‫מעות יש לי ואין לי שולחני‬
.‫חני להרצותן‬ :‫לרצותן‬

Clearly, the Bavli is the base text of ARNA, augmented and significantly deteriorated.

Appendix B

Rabbi Eliezer and Rabbi Yehoshua as Master/Disciple (above, n. 73)

Rabbi Eliezer and Rabbi Yehoshua are regularly taken as contemporary colleagues.
They appear first among the five disciples of Rabban Yohanan ben Zakkai enumerated
in mAv 2:8. Rabbi Yosi refers to the pair as “the early fathers.”118
In seven places the post-classical Midrash Mishle portrays Rabbi Eliezer as a disciple
of Rabbi Yehoshua, as e.g.:

“He reserves sound wisdom for the upright, He is a shield for those that live
blamelessly” (Prov 2:7). Rabbi Eliezer asked Rabbi Yehoshua: What is [the mean-
ing of] this verse? Rabbi Yehoshua replied: My son, from the time a person is
formed in his moth­er’s womb, the Torah which he is to learn is reserved for him,
and that is why Scripture says, “He reserves sound wisdom for the upright, He is
a shield for those that live blamelessly.” Just as the shield protects a person, so
Torah shields all who study it, and that is why Scripture says: “He is a shield for
those that live blamelessly” (Prov 2:7) (Midrash Mishle 2:7).119

118 tYT 1:10.

119 Visotzky, p. 29 and n. 28; idem, Eng., p. 30.
76 Friedman

The impropriety of this casting was observed by Zunz,120 Buber121 and others, who
point out that talmudic usage has them as colleagues.
Amoraic sources refer to Rabbi Yehoshua’s actions after the death of Rabbi Eliezer,
and his reverence towards him (see tNid 1:5).122 Another source reports:

Once Rabbi Yehoshua entered [Rabbi Eliezer’s Study Hall after the latter’s death]
and began kissing the stone [that served as Rabbi Eliezer’s chair] and said: This
stone is like Mount Sinai, and the one who sat on it was like the Ark of the
Covenant (SongR 1).

This reverential attitude may not be inconsistent with a collegial relationship, in that
Rabbi Eliezer is in any case senior, and mentioned first in the list of Rabban Yohanan
ben Zakkai’s disciples (mAv 2:4, as above). However we will try to demonstrate that
there are sources, predominantly the Yerushalmi, which hint to a master/disciple
relation­ship. In several places Rabbi Eliezer responds to Rabbi Yehoshua, saying “What
is this, Yehoshua!” (‫מה זה יהושע‬, e.g. mPes 6:2 and parallel in SifZut 9; SifDeut 38), a
phrase which Rabbi Eliezer uses when addressing his disciples, Rabbi Aqiva (mNaz 7:4;
tOhil 3:7)123 and Rabbi Ilai (tSuk 2:1, and also perhaps in tBer 1:4).124 Compare it being
disparagingly used toward an underling in tYom 1:4 and tBQ 7:13.
In the Yerushalmi account of the death-bed scene, although Hyrcanus, son of Rabbi
Eliezer, approached him to remove his phylacteries, it was Rabbi Yehoshua who finally
carried this out (perhaps a symbolic act of a disciple)125 and called Rabbi Eliezer “mas-
ter” three times: “My master, my master, the vow is annulled; my master, the chariots
of Israel and his horsemen” (‫ רכב ישראל ופרשיו‬,‫ רבי‬,‫ הותר הנדר‬,‫)רבי רבי‬, while echoing
Elisha’s departing words to his master Elijah: “Oh father, father, Israel’s chariots and
horsemen” (2Kgs 2:12; ‫) ָא ִבי ָא ִבי ֶר ֶכב יִ ְׂש ָר ֵאל ָּופ ָר ָׁשיו‬.126

120 L. Zunz, Die gottesdienstlichen Vorträge der Juden: historisch entwickelt: Ein Beitrag zur
alterthumskunde und biblischen Kritik, zur Literatur- und Religionsgeschichte (Berlin 1919)
281 and n. d.
121 Midrash Mishle, Vilna 5653, p. 11.
122 P. 641; Friedman, Igud Gittin, sugya 2.
123 Compare the structure of the above-mentioned SifZut, with Rabbi Yehoshua-Rabbi Aqiva
here in reverse order.
124 SifDeut 34, Rabbi Yishmael to Rabbi Eleazar ben Azariah.
125 Serving the master (‫)שימוש חכמים‬.
126 Yonah Fraenkel noted that Rabbi Yehoshua turns to Rabbi Eliezer “as the disciple Elisha
to his master Elijah” (Y. Fraenkel. “Time and its Shaping in Aggadic Narratives,” in: J. J.
Petuchowski and E. Fleischer (eds.), Studies in Aggadah, Targum and Jewish Liturgy in
Memory of Joseph Heinemann (Jerusalem 1981) 153 [Hebrew]). At the same time he refers
to Rabbi Eliezer as Rabbi Yehoshua’s “erstwhile colleague” (‫געגועיו של ר׳ יהושע אל חברו‬
TALMUDIC PASSAGES ABOUT MAGIC 77

The synoptic comparison of the Yerushalmi and the Bavli versions of the account
discussed above reads thus:

‫ב׳ סנהדרין סח ע״א‬ ‫ ה ע״ב‬,‫י׳ שבת ב ב‬

‫עמד רבי יהושע על רגליו‬ ‫נכנס ר׳ יהושע וחלץ את תפיליו והיה מגפפו‬
!‫ הותר הנדר‬,‫ הותר הנדר‬:‫ואמר‬ ‫ומנשקו ובוכה ואו׳ רבי רבי הותר הנדר‬
,‫למוצאי שבת פגע בו רבי עקיבא מן קיסרי ללוד‬
‫ פתח‬.‫היה מכה בבשרו עד שדמו שותת לארץ‬
:‫עליו בשורה ואמר‬
,)‫״אבי אבי רכב ישראל ופרשיו״ (מל״ב ב יב‬ .)‫רבי ״רכב ישראל ופרשיו״ (מל״ב ב יב‬
.‫הרבה מעות יש לי ואין לי שולחני להרצותן‬

The last two paragraphs quoted here from the Bavli are translated (Soncino) as follows:

On the conclusion of the Sabbath Rabbi Aqiva met his bier being carried from
Caesarea to Lod. [In his grief] he beat his flesh until the blood flowed down
upon the earth. Then Rabbi Aqiva commenced his funeral address, the mourn-
ers being lined up about the coffin, and said: “My father, my father, the chariot of
Israel and the horsemen thereof” (2Kgs 2:12); I have many coins, but no money
changer to accept them.

This translation “met his bier” expands the original ‫פגע בו‬, literally: “met him.”
Furthermore, we see that Rabbi Aqiva is considered the subject of all the verbs in this
passage. This determination was already made in Tractate Semahot 9:2:

‫ והיה מכה על לבו והדם‬,‫מעשה כשמת רבי אליעזר וחלץ רבי עקיבא לפניו את שתי ידיו‬
‫ הרבה מעות יש לי ואין‬,)‫ ״רכב ישראל ופרשיו״ (מל״ב ב יב‬,‫ רבי רבי‬:‫ וכך היה אומר‬,‫שותת‬
.‫שולחני להרצותן‬

Now it happened that when Rabbi Eliezer died, Rabbi Aqiva bared both arms
and beat his breast, drawing blood. And thus he spoke: My master, my master,
“the chariots of Israel and the horsemen thereof!” (2Kgs 2:12). A multitude of
coins have I, but no money-changer to sort them!127

‫)הותיק ר׳ אליעזר‬, thus explaining his charged words. Also, Fraenkel presents a different
explanation for his removal of the phylacteries.
127 D. Zlotnick, The Tractate “Mourning” (Semahot): Regulations related to Death, Burial and
Mourning (New Haven 1966) 67; Higger, p. 169.
78 Friedman

Alon Goshen-Gottstein, in his analysis of the Bavli’s baraita, also presents the last
scene as revolving around Rabbi Aqiva only, and interprets this as the Bavli baraita’s
desire to portray Rabbi Aqiva as the disciple and spiritual heir of Rabbi Eliezer.128
Now in the Yerushalmi, certainly a more original form of the baraita than that
in the Bavli, it is clear that the same person who says “The vow is annulled, the vow
is annulled!” went on to deliver the eulogy “the chariot of Israel and the horsemen
thereof,” namely, Rabbi Yehoshua! Indeed Tractate Semahot and ARNA129 made Rabbi
Aqiva the subject of the verbs, based perhaps on an Aqiva-centered bias. However,
in the baraita of the Bavli itself it is possible to view Rabbi Yehoshua as the subject
of the verbs. It all boils down to who is the referent of ‫“( בו‬him”) in the phrase ‫פגע בו‬
(“he met him”).130 Simple style and syntax should indicate that Rabbi Aqiva met Rabbi
Yehoshua (and not Rabbi Eliezer’s bier). Thus, Rabbi Yehoshua would be the subject of
the following verbs, and, as in the Yerushalmi, he is the one who eulogizes. In order to
make Rabbi Aqiva the referent in ‫ בו‬one would have to read: ‫פגע בו בר׳ עקיבא‬.131
Thus the central role of Rabbi Yehoshua in this episode in the Yerushalmi’s more
primary parallel is still observable through a close reading of the Bavli. The later recen-
sions (Semahot and ARNA) and commentators enhanced the position of Rabbi Aqiva
in this narrative, thus eclipsing Rabbi Yehoshua.
In two magical accounts in the Yerushalmi Rabbi Eliezer turns to Rabbi Yehoshua to
perform magical salvation, saying: “Now Yehoshua ben Haninah, see what you can do”:

‫ חמתון חד‬.‫דלמא רבי ליעזר ורבי יהושע ורבי עקיבה עלון למיסחי בהדין דימוסין דטיבריא‬
‫ חמי מה‬,‫ מה יהושע בן חנינה‬:‫ א״ר ליעזר לר׳ יהושע‬.‫ אמר מה דמר ותפשיתון כיפה‬.‫מיניי‬
‫ אמר רבי ליעזר‬. . . ‫ מי נפיק אהן מינייא אמר רבי יהושע מה דמר ותפש יתיה תרעה‬.‫דאת עבד‬
.‫ מה יהושע בן חנניה חמי מה דאת עביד‬:‫לרבי יהושע‬

128 ‫ הצגה זו מלמדת‬.‫ הלא הוא ר׳ עקיבא‬,‫ יש גיבור אחד בלבד‬,‫ החותמת את הסיפור‬5 ‫באפיזודה‬
‫ מפרספקטיבה זו הסיפור כולו חותר לשיא שהוא‬.‫על מרכזיותו של ר׳ עקיבא בסיפור הבבלי‬
‫( המונולוג החותם של ר׳ עקיבא‬Goshen-Gottstein, “Ideological Analysis,” 83; and see also
p. 86, where the author struggles somewhat with this idea).
129 Even though its text is essentially based on the Bavli (see Appendix A), the emphasized
words in the following quote make in clear that Rabbi Aqiva is the subject of the verbs:
‫לאחר שבת בא רבי עקיבא ומצאו באריסרטיא שבא מקסרי ללוד מיד קרע את בגדיו ותלש‬
‫בשערו והיה דמו שותת ונופל לארץ והיה צועק ובוכה‬.
130 ‫למוצאי שבת פגע בו רבי עקיבא‬.
131 As Daniel Boyarin seems to have done, when he translates: “On the going out of the
Sabbath, he met Rabbi Akiva” (Boyarin, Dying, 38).
TALMUDIC PASSAGES ABOUT MAGIC 79

Once Rabbi Liezer and Rabbi Yehoshua and Rabbi Aqiva went to the baths in
Tiberias. They saw a heretic. He said what he said,132 and they were caught in
by the dome (of the bathhouse). Said Rabbi Eliezer to Rabbi Yehoshua: Now
Yehoshua ben Haninah, see what you can do. When that min tried to leave, Rabbi
Yehoshua said what he said, and the doorway of the bath seized and held the
heretic firm . . .
Said Rabbi Eliezer to Rabbi Yehoshua: Now, Yehoshua ben Hananiah, let us see
what you can do. (ySanh 7:19, 25d)

On this Levinson writes:

Throughout the talmudic discussion here, R. Yehoshua ben Hananya is presented

as the premier magician, while in the Bavli this position is occupied by R. Eliezer.
I do not know why R. Yehoshua takes the lead when both masters are present,
with Eliezer saying, “Show him what you can do.” In any case, it is interesting that
a similar situation and locution occurs in the Apocryphal Acts of Peter and Paul
(VII): “Paul said to Peter: Do at once what you doest” [ANF VIII: 484].133

However, we have seen that the Yerushalmi portrays Rabbi Yehoshua (and not Rabbi
Aqiva) as the disciple of Rabbi Eliezer, regarding the many laws about sorcery (“Said
Rabbi Yehoshua: Three hundred laws did Rabbi Eliezer expound concerning the verse:
‘Thou shalt not suffer a witch to live’ [Exod 22:17], and of all of them I have heard only
two things”). In the Yerushalmi’s aggadic anecdotes where the two appear in a set-
ting of magical acts (performed for noble purposes), it is the “master” who charges the
“disciple” with the task of performing magic. Rather than Levinson’s conclusion that
“R. Yehoshua takes the lead when both masters are present,” I would claim that Rabbi
Eliezer “takes the lead” as master vis à vis Rabbi Yehoshua, who is not an equal (“both
masters”), but rather disciple (the same could apply to Paul vis à vis Peter).134 Certainly
the salutation “Now Yehoshua ben Haninah” does not seem to be one fitting for a col-
league, but rather for a disciple.135 It is the master magician who instructs the disciple
(serving as apprentice and shamash) actually to perform the deed.
The Yerushalmi concludes the aggadic anecdotes with a full-blown statement
ascribed to Rabbi Yehoshua, as if connecting it to them (e.g: “and those stories
correspond with what he himself said”): “Rabbi Yehoshua ben Hananiah said: I can

132 D. Sperber, “On a Meaning of the Word ‫מלה‬,” in D. Sperber, Magic and Folklore in Rabbinic
Literature (Ramat Gan 1994) 63. Cf. “He did what he did” in Aramaic (‫= )עבד מאי דעבד‬
“he performed a magical practice” (Sokoloff, Babylonian, 813; Bohak, “Minim,” 271, n. 15).
133 Levinson, “Enchanting,” 60, n. 17.
134 See http://executableoutlines.com/gal/ga2_6.htm.
135 The common address forms are first name or ben x.
80 Friedman

take gourds and melons and transform them into stags and deer, and these then pro-
duce more stags and deer” (ySanh 7:13, 25d).136 This assertive claim is similar in style to,
and its formulation may be patterned upon:

‫ היך? עבידא זבן בתרין מניי זרע‬.‫ יכיל אנא כתב כל קרייא בתרין מניין‬:‫ר׳ חייה רבה אמר‬
.‫דכיתן וזרע ליה וחצד ליה ועבד חבלין ותפש טביי וכתב כל קרייה על משכיהון‬

Rabbi Hiyya the Great said: I can write down all the verses (of Scripture) for the
price of two maneh. How do I do this? I buy flax seed for two maneh and sow
it and harvest it and make ropes and trap deer and write all the verses on their
skins (yMeg 4:1, 74d).

By describing Rabbi Yehoshua’s magical accomplishments the same way Rabbi Hiyya’s
Torah accomplishments are described, Rabbi Yehoshua is being advanced beyond the
disciple status and is now cast as a major magic figure, proud of his magical feats and
even bragging about them, reminiscent of Simon Magus. This admiration, esteem,
and according heroic status, is not earlier than the amoraic period, part and parcel
of the growing approval of magic during that time. It cannot be assigned to tannaitic
conceptualization, and certainly not to the beginning of the tannaitic period, as some
earlier scholars tended to do; we note, for insance, “Soon after 70, R. Yehoshua ben
Hananiah boasted of his ability to transform cucumbers and melons into living deer.”137 It
is the amoraic Yerushalmi which casts Rabbi Yehoshua as learning the laws of sorcery from
Rabbi Eliezer, and transmits the expanded aggadic anecdotes in which Rabbi Eliezer is the
master and Rabbi Yehoshua the disciple, who finally comes into his own right as master.
It is possible that the subsequent raising of the figure of Rabbi Yehoshua ben
Hananiah to that of paradigmatic magician was facilitated by the aim of having him
serve as a Jesus (Yeshua) competitor/substitute, as Yehoshua ben Perahiah138 was cast
as the teacher of Jesus139 due to the common name “Yehoshua” and the “Alexandria”
connection.140 Yehoshua ben Perahiah then served as a competitor-substitute for Jesus
in magical literature. Our Rabbi Yehoshua ben Hananiah’s persona may have been
treated the same way.

136 ‫ יכיל אנא נסיב קריין ואבטיחין ועביד לון איילין טבין והידנון עבידין‬:‫אמ׳ ר׳ יהושע בן חנניה‬
‫איילין וטבין‬.
137 S. W. Baron, A Social and Religious History of the Jews, vol. 2 (New York 1952) 21.
138 About whom see B. L. Sherwin, Workers of Wonders: A Model for Effective Religious
Leadership (Lanham 2004) 75–7.
139 bSot 47a and cf. MSS = bSanh 107b.
140 For Jesus’ magic, see bShab 104b = bSanh 67a and Yehoshua ben Perahya regarding wheat
from Alexandria being impure, tMakh 3:4 (J. Klausner, Jesus of Nazareth (trans. H. Danby;
New York 1925] 25–6; J. Rubenstein, Stories of the Babylonian Talmud [Baltimore 2010] 138
and references).
TALMUDIC PASSAGES ABOUT MAGIC 81

Appendix C

bMakkot 19b, Variant Readings (above, n. 92)

‫מיבעיא להו אפיתחא דירושלם‬ ‫יתיב רב חנניה ורב הושעיה וקא‬ ‫הרצוג‬
[‫מיבעיא להו אפיתחא דיר‬ ] [‫יתיב ר‬ ‫מודינא‬
‫מיבעי להו אפיתחא דירושלם‬ ‫יתיב רב חנניא ורב הושעיא וקא‬ ‫ד״ו‬
‫בפיתח׳ דירושל׳ מיבעי לי׳‬ ‫יתיב רב חנינ׳ ורב אושעי׳ וקאי‬ ‫מינכן‬

Modena: Archivio di Stato Fr. Ebr. 472

Appendix D

Rabbi Hanina or Rabbi Juda Nasia? (see above, n. 92)

The following passage exemplifies the fluidity of these traditions:

‫ ״צדקות פרזונו בישראל״ (שופטים ה יא)? צדקה עשה הקדוש‬:‫ מאי דכתיב‬:‫אמר רבי אושעיא‬
‫ אנן מעלינן‬:‫ והיינו דאמר ליה ההוא מינא לרבי חנינא‬.‫ברוך הוא בישראל שפזרן לבין האומות‬
‫ איתינכו גבן‬,‫ ואלו אנן‬,)‫ ״כי ששת חדשים ישב שם וגו׳״ (מלכים א יא טז‬:‫ כתיב בכו‬.‫מינייכו‬
‫ נטפל ליה רבי‬.‫ יטפל לך תלמיד אחד‬,‫ רצונך‬:‫ אמר לו‬.‫ ולא קא עבדינן לכו מידי‬,‫כמה שני‬
‫ מאי דאיכא‬.‫ ליתנהו גבייכו‬,‫ תכלינן כולהו‬:‫ משום דלא ידעיתו היכי תעבדו‬:‫ אמר ליה‬.‫אושעיא‬
‫ גפא דרומאי! בהא נחתינן ובהא סלקינן (ב׳‬:‫ אמר ליה‬.‫ קרי לכו מלכותא קטיעתא‬,‫גבייכו‬
.)‫פסחים פז ע״ב‬

Rabbi Oshaia said: What is meant by the verse, “Even the righteous acts of His
Ruler in Israel” (Judg 5:11)? The Holy One, blessed be He, showed righteousness
[mercy] unto Israel by scattering them among the nations. And this is what a
certain sectarian said to Rabbi Hanina: We are better than you. Of you it is writ-
ten: “For Joab and all Israel remained there six months, until he had cut off every
male in Edom” (1Kgs 11:16); whereas you have been with us many years, yet we
have not done anything to you! Said he to him: If you agree, a disciple will debate
it with you. [Thereupon] Rabbi Oshaia debated it with him, [and] he said to him:
[The reason is] because you do not know how to act. If you would destroy all,
they are not among you. [Should you destroy] those who are among you, then
you will be called a murderous kingdom! Said he to him: By the [Love—S.F.] of
Rome!141 With this [care] we lie down and with this [care] we get up (bPes 87b).

141 = Isis, see Lieberman, Greek, 140.

82 Friedman

The name “Rabbi Hanina,” in bold above, appears in some MSS as Rabbi Judah the
Patriarch II, a reading adopted by Saul Lieberman:142

The oath ‫ גפא דרומא‬is recorded once more in TB in the name of a gentile. R. Juda
the Patriarch (III c.) commissioned R. Oshaia to debate with this gentile.

In note 95 (to “Rabbi Juda the Patriarch”), Lieberman comments:

This is the only correct reading, see ‫ דקדוקי סופרים‬ad loc., p. 268 n. 200. This read-
ing is also corroborated by Seder Eliyyahu Rabba (XI ed. Friedmann, p. 54) which
drew from a different source. The reading of the editions and ms. Munich is erro-
neous, for R. Hanina would not call R. Oshaia “a disciple,” whereas the Patriarch,
R. Juda the Second, by virtue of his office may have permitted himself to term
R. Oshaia “disciple” in order to show the gentile that even his pupil will be able
to give him the right answer.

Today we can document the reading “Rabbi Hanina” from more textual witnesses than
the two (Munich MS and printed edition) cited by Lieberman. The witnesses divide
into two discrete families,143 and consequently we should consider the nature and
mechanism of the emergence of these two traditions. The attestations among full tex-
tual witnesses are as follows:

‫יהודה נשיאה אנן עדיפינן מיניכו‬ ‫כי הא דאמ׳ ליה ההוא מינא לר׳‬ 6‫מ‬
‫יהודה נשיאה אנן מעלינן מניכו‬ ‫והינו דאמ׳ ליה ההוא מינא לר׳‬ 1623
‫יהודה נשיאה אנן מעלינן מניכו‬ ‫והינו דאמ׳ ליה ההוא מינא לר׳‬ ‫קול‬
‫חנינה אנן מעלינן מיניכו‬ ‫והיינו דאמ׳ ליה ההוא מינה לר׳‬ 125‫ו‬
‫חנינא אנן מעלי מיניכו‬ ‫והינו דא״ל ההוא מינאה לר׳‬ ‫אוק‬
‫חנינא אנן מעלינן מינייכו‬ ‫והיינו דא״ל ההוא מינא לר׳‬ 95‫מ‬
‫חיננא אנן מעלינן מיניכו‬ ‫כי הא דאמ׳ ליה ההוא מינא לר׳‬ 109‫ו‬
‫חנינא אנן מעלינן מניכי‬ ‫והינו דא״ל ההוא מינא לר׳‬ 134‫ו‬
‫חנינא אנן מעלינן מיניכו‬ ‫והיינו דאמ׳ ליה ההוא מינא לרבי‬ ‫ד״ו‬

142 Lieberman, Greek, 141.

143 For research on the dynamics of the textual traditions of bPesahim see S. G. Wald, Pesahim II
(‫ ניו יורק וירושלים‬,‫ מהדורה מדעית וביאור מקיף‬,‫תלמוד בבלי מסכת פסחים פרק אלו עוברין‬
‫ )תש״ס‬269–83; A. Amit, “The Place of Yemenite Manuscripts in the Transmission-History
of bPesahim,” Hebrew Union College Annual 73 (2002) ‫לא‬-‫[ עז‬Hebrew]; idem, ,‫א׳ עמית‬
‫ מהדורה ביקורתית של פרק רביעי ממסכת פסחים על פי‬:‫״נוסח פרק מקום שנהגו בבבלי‬
‫ תשנ״ה‬,‫אילן‬-‫ אוניברסיטת בר‬,‫ עבודת מוסמך‬,‫ כתבי היד וקטעי הגניזה בצרוף מבוא״‬pp.
3–114; Friedman, Tosefta Atiqta, 87–8.
TALMUDIC PASSAGES ABOUT MAGIC 83

The first group, reading “Rabbi Judah the Patriarch” (II) is composed of a Spanish MS and
Yemenite MSS. The second, testifying to “R. Hanina,” is represented by five Ashkenazic
or Ashkenazic-related MSS and the first printed edition.144 One of the manuscripts, Vat
125, exhibits early and remarkable linguistic forms and quality text.145 Now that it has
been established that we are dealing with two time-honored and eminent traditions,
we must go beyond the categories of “correct” and “erroneous,” and entertain “original”
and “reworked” editorially. Whichever of the two groupings belongs to the second cat-
egory (reworked), it should be clear that its reading emanates from an early, scholastic
emendation.146 If the original reading was “Rabbi Yehudah Nasia,” the scholastic who
emended to “Rabbi Hanina” did so because Rabbi Hanina occurs at the beginning of
the passage, and Rabbi Hanina and Rabbi Oshaia were a well-known sage-pair; if the
original text was “Rabbi Hanina,” the switch to “Rabbi Yehuda Nasia” was inspired by
the account in bAZ 6b147 of a “sectarian” presenting Rabbi Yehuda Nasia with a chal-
lenging situation. In either case, support is given to a working hypothesis that our
two rabbis were a renowned sage-pair, readily available for use in literary creativity,
whether appearing here in the original text or the emended one.

Appendix E

1Sam 1:24 at Qumran (above, n. 101)

‎ (4Q51 2a_d:6) 1Sam 1:24 ‫שמואל א א כד‬

)4Q51 2a_d:7( ‫כאשר‬
֯ ‫ותעל אׄותו שילה‬ ‫וַ ַּת ֲע ֵלהּו ִע ָּמּה ַּכ ֲא ֶׁשר‬
‫[יעלה אלקנה לזבח ליהוה‬ ‫ּגְ ָמ ַלּתּו‬
}}°°°°{{ )4Q51 2a_d:8( ‫בפר בן ]בקר משלש ולחם‬ ‫ְּבפָ ִרים ְׁש ̇ל ׁ ָשה‬
‫‏[ואיפה אחת קמח ונבל יין‬ ‫יפה ַא ַחת ֶק ַמח וְ נֵ ֶבל יַ יִ ן‬ ָ ‫וְ ֵא‬

144 This corresponds exactly to the observed division of the witnesses to Pesahim into two
major families.
145 See Friedman, Talmudic Studies, 319.
146 On the phenomenon of variant readings in the Bavli see Friedman, Talmudic Studies, 192–
246; idem, “On the Origin of Textual Variants in the Babylonian Talmud,” Sidra: Journal for
the Study of Rabbinic Literature 7 (1991) 67–102 [Hebrew].
147 ‫ הוה יתיב ריש לקיש‬.‫ההוא מינאה דשדר ליה דינרא קסריאנא לר׳ יהודה נשיאה ביום אידי׳‬
‫ הויא ליה איבה! א״ל ריש‬,‫ אזיל ומודה! לא אשקליה‬,‫ היכי אעביד? אשקליה‬:‫ אמר‬,‫קמיה‬
‫ כל שכן דהויא ליה איבה! כלאחר יד הוא דקאמינא‬:‫ אמר‬.‫ טול וזרוק אותו לבור בפניו‬:‫לקיש‬.
No Boundaries for the Construction of Boundaries:
The Babylonian Talmud’s Emphasis on
Demarcation of Identity
Moshe Lavee

The growing awareness of the Talmud’s gradual construction has afforded

scholars new tools for examining historical developments during the talmu-
dic period. Equipped with this new awareness, we are now able to investigate
otherwise undetected social and cultural developments in Talmudic times.
Unique tendencies of the Babylonian Talmud are manifested in three differ-
ent ways:

1. in the structure and arrangement of materials, which hints at the distinct

agenda of the redactors;
2. in later layers of additions and comments by the savoraim or stammaim;
3. on “Babylonian baraitot” and meimrot, namely, Babylonian versions of
tannaitic and early Palestinian amoraic statements, whose innovative
character can be discerned from a comparison with parallels in earlier
rabbinic texts from the Land of Israel.1

At first sight, the methods used for identifying the last two phenomena seem
to contradict one another. The effort to identify late additions, as in phenom-
enon 2, is based on the assumption that they were added to earlier materials
that were preserved verbatim, and still reflect the authentic sayings of their
authors. Synoptic comparisons, however, reveal at times that the cited texts,
as we have them in the Bavli, are comprehensive reworkings of the Palestinian
version to the extent that they may be identified as Babylonian baraitot.

1 For recent surveys of pertinent methodologies see the introduction in J. L. Rubenstein,

Creation and Composition (Tübingen 2005) 1–22; R. L. Kalmin, “The Formation and Character
of the Babylonian Talmud,” in: S. T. Katz (ed.), The Cambridge History of Judaism, vol. 4
(Cambridge 2006) 840–76; Y. Brody, “The Literature of the Gaonim and the Talmudic Text,”
Talmudic Studies 1 (1990) 237–303; E. Fonrobert and M. S. Jaffee, The Cambridge Companion
to the Talmud (Cambridge 2007) 1–37; S. Friedman, Talmudic Studies: Investigating the Sugya,
Variant Readings and Aggada (New York 2010).

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�03

THE BAVLI ON DEMARCATING IDENTITY 85

It seems that the conceptual and social circumstances that led to introduc-
ing Babylonian changes had worked in both channels. At times they led to
minor but significant rephrasing of earlier materials, (Babylonian baraitot and
meimrot) and at times they are recognized because they appear in late anony-
mous Aramaic additions (stam).
Together, the three phenomena produce the organizing or lecturing voice of
the Bavli. It is important to clarify that I do not claim here an authorial coher-
ence of this voice; neither do I accept its perception as the product of inten-
tional redactors. Rather, this voice is the conglomerated product of centuries
of evolution of the text in a Babylonian environment. The “Babylonization” of
rabbinic culture began during the last generations of the Babylonian amoraim,
who contributed to, and shaped, the nature of Babylonian discussions, and
hardly left any influence on the rabbinic works of the Land of Israel (from Rava
and Abbayye onwards). This activity continued with those who were involved
in organizing sugyot, in the late additions of comments and questions; and it
ended (but was never sealed) with minor changes to the text during the long
period of its oral transmission, well into the Gaonic era. I doubt whether the
talmudic text permits a more delicate and detailed account of its formation
than this sketchy presentation.
In the following pages I use these three categories—arrangement of materi-
als, late comments on earlier texts, and reworking of earlier texts—to portray
briefly the unique voice of the Bavli in issues related to the demarcation of
identity and the marking of social boundaries. Social boundaries are expressed
in various manners: they are declared in the imagined gates of the city, the act
of conversion; this is the procedure of entrance into the group, and the way
this procedure is designed, performed, and conceived. They are stated in the
perception of newcomers. They are manifested in engagement in, or avoidance
of, seeking new adherents. They may also be noticed in the liminal social areas
between “us” and “them,” representing the extent to which the group permits
the existence of those who are neither here nor there. The following examples
will demonstrate a relatively consistent tendency towards further fortifying
the boundaries of Jewish identity in the Bavli. This tendency is manifested
in all three modes described in the first paragraph of the present study: the
structuring of larger units of texts (no. 1); commenting upon and contextu-
alizing earlier traditions (no. 2), and rephrasing and reworking them (no. 3).
The overall cultural work of this process offers something which is beyond the
introduction of a unique Babylonian agenda. These textual processes divert
the chronological prism, and portray later developments as the fruits of earlier
generations.
86 Lavee

1 The Procedure of Conversion and the Formal Acceptance

of Converts

First we shall take a look at the procedure of conversion.2 In the Babylonian

Talmud we find a long and well-developed literary unit, devoted to the proce-
dure of conversion and to the validity of conversions that were not performed
properly.

1.1 The Structure of the Literary Unit

The literary unit is arranged as a kind of minor tractate within Tractate Yevamot,
represented by a set of talmudic discussions around baraitot on the topic.3
This form places each of these baraitot in the same literary role as a mishnah,
and hence grants them a sense of mishnaic authority. The very structuring of
this unit hints at the importance and centrality of the procedure of conver-
sion. Nevertheless, while organized around tannaitic materials, and includ-
ing a set of early Palestinian amoraic statements (attributed to Rabbi Hiyya
bar Abba in the name of Rabbi Yohanan) it is evident that the unit is a late
Babylonian product. Firstly, most of the discussions and comments are anony-
mous. Furthermore, the unique Babylonian nature of the structure of the unit
can be discerned from the different role the same tannaitic material plays in
the parallel Yerushalmi sugyot. In the latter we find them merely quoted within
the relevant discussion. It is only in our “minor tractate” in the Bavli that they
serve as the starting point for the discussion.4 The last two baraitot have an

2 For a more detailed account see M. Lavee, “The ‘Tractate’ of Conversion—BT Yeb. 46–48 and
the Evolution of the Conversion Procedure,” European Journal of Jewish Studies 4 (2010) 169–
213; idem, “A Convert is Like a Newborn Child: The Concept and its Implications in Rabbinic
Literature” (Ph.D. diss., Beer Sheva 2003), ch. 4.1 [Hebrew].
3 These are the baraitot: (1) ‫ גר שמל ולא טבל‬. . .—“A convert who was circumcised but did not
immerse . . .” (bYev 46a); (2) ‫ מלתי ולא טבלתי‬:‫ הרי שבא ואמר‬. . .—“If one came and stated:
I was circumcised but did not immerse . . .” (bYev 46b); (3) ‫ גר אני‬:‫ מי שבא ואמר‬. . .—“One
who came and stated: I am a convert . . .” (bYev 46b); (4) ‫ ושפטתם צדק‬. . .—“And judge
righteously . . .” (bYev 47a); (5) ‫ גר שבא להתגייר‬. . .—“A proselyte who came to convert . . .”
(bYev 47a); (6) ‫ מפני מה גרים מעונין‬. . .—“Why do converts suffer . . .” (bYev 48b).
 The second baraita is introduced as part of the discussion of the first, but it is plausible
that at earlier stages of the redaction of the sugya it served as an opening point to the follow-
ing discussion, like the other baraitot in the text.
4 The first baraita is presented in the Yerushalmi in the discussion of the status of the offspring
of mixed relations (a slave or a non-Jew with a Jewess) see yQid 3:9, 64b–c. The Babylonian
parallel to this discussion is the text preceding our “minor tractate.” Here this baraita is no
longer part of the former discussion, but rather serves as an opening to our “minor” tractate.
THE BAVLI ON DEMARCATING IDENTITY 87

important structural function, presenting first the full procedure and then
an aggadic baraita with the statement “a proselyte is like a newborn child,” to
be discussed below. As such, they adhere to common structural conventions:
placing the description of procedures towards the end of a unit and sealing
with aggadah. The text concludes with a verse of consolation, “The Lord rec-
ompenses thy work, and be thy reward complete from the Lord, the God of
Israel, under whose wings thou art come to take refuge” (Ruth 2:12)—another
mark of a well-designed, autonomous and structured unit.

1.2 The Content of the Unit

Not only the form but also the content of this Babylonian “minor tractate” pres-
ents developments regarding the procedure of conversion, demonstrated in a
further institutionalization of the procedure, and a hardening of the legal con-
ditions for conversion, and for approval of the convert’s status. These changes
testify to a transition from a model of “witnessing” the procedure to a court-
controlled conversion. The institution of a “court of conversion,” repeatedly
referred to in this unit, appears in other sources in the Bavli, but is unknown to
the earlier Palestinian rabbinic sources.5

1.3 The Move toward a Court-Governed Procedure

The Babylonian move from witnessed conversion to a court-governed
procedure is evident in late annotations in our “mini-tractate” which
comment on the tannaitic or early amoraic materials. The procedure of con-
version, as presented in the Bavli, states that “two disciples of the sages must
stand beside him” (bYev 47b), while a late talmudic comment insists that Rabbi
Yohanan amended the version to three (ibid.).6 This minor change is intro-
duced in order to befit the amoraic statement attributed to Rabbi Yohanan
earlier in our unit: “A proselyte needs three [persons to facilitate his conver-
sion]. The word ‘Law’ (Num 15:16) was written regarding to him” (bYev 46b).

The fourth baraita is found in yQid 4:7, 66a in the discussion of the validity of one’s testimony
regarding the status of his sons. It is only in the Bavli that it is presented as the beginning of
a discussion devoted to the testimony of a doubtful convert regarding the status of his sons.
The emphasis has shifted from the discussion of the validity of testimony regarding one’s
sons to the question of the validity of conversion.
5 See also J. Kulp, “The Participation of a Court in the Jewish Conversion Process,” Jewish
Quarterly Review 94 (2004) 437–70.
6 The parallel in Gerim 1:1 refers neither to sages as the facilitators of the procedure, nor does it
mention their number.
88 Lavee

Despite the claim to Palestinian authorship, this text is unknown beyond the
realm of the Bavli.
The same literary mechanism is at work in the first occurrence of this state-
ment of Rabbi Yohanan in our unit. It is presented as one of the halakhic les-
sons of a case in which rabbis waited until morning for the sake of a convert’s
immersion. This act is explained as rooted in the juristic nature of the proce-
dure, as court acts are not performed during the night. Here also we have a
Palestinian source for synoptic com­parison. A similar case in the Yerushalmi
refers to a female convert, but it is not in any way explained as related to the
participation of a court in the conversion procedure:

yYev 8:1, 8d bYev 46b

:‫ואמ׳ ר יצחק בר נחמן‬ 8:‫אמר רבא‬
‫ והוה תמן ר׳ יודן‬.‫ר׳ יהושע בן לוי היה בלודקיא‬ 9‫ ור׳ יוסי‬,‫עובדא הוה בי רבי חייא בר רבי‬
.‫ בעא מיקרוץ‬.‫נשייא‬ ‫ ר׳‬:‫ ורב ספרא מתני‬,‫ רבי אושעיא בר רבי‬:‫מתני‬
‫אמ׳ ליה המתן ואנו מטבילין את הגיורת הזו‬ ‫אושעיא בר׳ חייא דאתא לקמיה גר שמל ולא‬
.‫למחר‬ 10.‫ א״ל שהיי כאן ונטבילנן למחר‬.‫טבל‬
‫ למה? מפני‬:‫ר׳ זירא שאל לר׳ יצחק בר נחמן‬
‫כבוד הזקן או משום שאין מטבילין את הגיורת‬
‫בלילה? אמ׳ ליה משום שאין מטבילין את‬
.‫הגיורת בלילה‬
‫אתא עובדא קומי דר׳ יוסי מהו להטביל את‬ ‫ וש״מ אינו עד‬,‫ ש״מ גר צריך שלשה‬.‫ש״מ תלת‬
7.‫הגרים בלילה? ולא הודו‬ .‫ וש״מ אין מטבילין גר בלילה‬,‫עד שימול ויטבול‬

Rabbi Isaac bar Nahman said: Rava said:

Rabbi Joshua ben Levi was in Laodicea, There was a case in which Rabbi Hiyya
and Rabbi Yudan Nasia was there, and he bar Rabbi was involved, and Rabbi Yosi
wished to leave. said: It happened to Rabbi Hoshaya bar

7 Assis suggests amending the text: ‫“( ולא הורי‬and he did not instruct them”). See M. Assis,
A Concordance of Amoraic Terms Expressions and Phrases in the Yerushalmi, vol. 2 (New
York–Jerusalem 2010) 810, n. 232; S. J. D. Cohen, The Beginnings of Jewishness: Boundaries,
Varieties, Uncertainties (Berkeley 1999) 198–233.
8 As in all MSS, except for Oxford 367 and the Vilna edition, which read ‫רבה‬. Unless stated
otherwise the quotations of the Bavli are from the Vilna edition. Only selected and sub-
stantial variants are presented. The texts from the Yerushalmi are presented on the basis
of MS Leiden with completion of abbreviations and initials. Gerim—on the basis of
Higger’s edition.
9 As in MSS Munich 95 and Vatican 111. MSS Moscow Guenzburg 594, Oxford 367 and
Cambridge Add. 3207 read ‫רב יוסף‬.
10 As in MS Munich 95. See MSS for variants.
THE BAVLI ON DEMARCATING IDENTITY 89

He said to him: Please wait here and we Rabbi, and Rav Safra said: It happened to
shall immerse this female convert tomor- Rabbi Hiyya bar Rabbi Hoshaya: A con-
row. Rabbi Zeira asked Rabbi Isaac bar vert who immersed and was not circum-
Nahman: Why? Was it because of the cised came to him. He said to him: wait
honor of the sage, or was it because one here and we shall immerse you tomorrow.
should not immerse a female convert at You should learn three things from this
night? The case came before Rabbi Yosi: case:
What is the law regarding immersion (1) A convert needs three [judges]; (2) he
of converts at night? And they made no is not a valid convert until he has both
decree. immersed and circumcised; (3) we should
not immerse a convert at night.

One may suggest that the reason for waiting till morning in the Yerushalmi
may be related to the fact that the case deals with a woman: there is potential
promiscuity involved in the immersion of a female convert at night-time. The
implied presence of the sages in the immersion can be explained as rooted
in the demand for a witnessed conversion and is not to be understood as evi-
dence supporting the existence of a Palestinian court for conversion. It is only
in the Bavli that the event is presented as deriving from the explicit demand for
the presence of a court. One should also note that the Yerushalmi suggests that
the question of whether immersion of converts at night was at all necessary
was still open during the days of the fourth-generation Palestinian amora Yosi
(ben Zevida), while the Bavli suggests a different chronology, presenting the
same law as deduced from the deeds of first-generation Palestinian amoraim.
The introduction of the institution of a court in a late Babylonian com-
ment referring to an earlier Palestinian source, which did not mention it at all,
appears again in the words of the Babylonian amora, Rav Sheshet, on the third
baraita in our talmudic unit:

‫ שמענו שנתגייר בבית‬:‫ דאמרי‬:‫ קרא למה לי? אמר רב ששת‬.‫״בא הוא ועדיו עמו״‬
.)‫ קמ״ל (ב׳ יבמות מז ע״א‬,‫ ליהמניה‬12‫ לא‬11‫ מהו דתימא‬.‫דינו של פלוני‬

If he came and had witnesses with him. Why is a proof text needed here?
Rav Sheshet said: [The baraita refers to a case where the witnesses]
stated: We heard that he converted at a certain particular court. Since you

11 As in all MSS except for Moscow Guenzburg 107 and Vilna edition that reads ‫סלקא דעתך‬
‫אמינא‬.
12 MS Cambridge Add. 3207 omits this crucial word.
90 Lavee

might have said, that we are not to believe them, we were taught [that we
do believe them] (bYev 47a).

According to the baraita, if a person comes with his witnesses and claims that
he is a convert, his testimony is accepted. The Bavli questions the necessity for
such a statement, claiming that it is obvious that when a convert comes with
his witnesses his testimony is accepted. Rav Sheshet explains that the tannaitic
text refers to a case when the witnesses only testify to having heard that he
was converted in a certain court; hence, even in the case of indirect testimony
the conversion is valid. Implied in Rav Sheshet’s explanation is the assump-
tion that a valid testimony is one in which the witnesses testify that they were
present at this person’s court of conversion. This is a beautiful example of the
Bavli’s tricky “rhetoric of the obvious.” The text proclaims itself to be lenient,
approving also indirect witnesses, but in fact, the idea of court conversion is
introduced by Rav Sheshet into a tannaitic text, which knew nothing of it! The
tannaitic text referred to a convert who came with his witnesses, probably
because this was the model of conversion with which it was familiar: the act
of immersion should be observed by witnesses. The Babylonian amora trans-
ferred it to a new realm: the whole procedure is run by a court.

So far, I presented two cases with a similar dynamic—later sources introduce

the model of a court-governed conversion, whereas earlier sources, quoted
in the Bavli, do not actually refer to it. We also noticed that parallel materi-
als in the Yerushalmi were not familiar with this institution. However, in the
fourth baraita of our unit the term “court” is included, implying a tannaitic
origin of the institution:

:‫ ״ושפטתם צדק בין איש ובין אחיו ובין גרו״ (דברים א טז) מכאן א״ר יהוד׳‬:‫תנו רבנן‬
.‫ בינו לבין עצמו אינו גר‬.‫גר שנתגייר בב״ד הרי זה גר‬
‫ יש לך‬:‫ א״ל‬.‫ נתגיירתי ביני לבין עצמי‬:‫ באחד שבא לפני ר׳ יהוד׳ ואמ׳ לו‬13‫ומעשה‬
‫ נאמן אתה לפסול את עצמך ואין אתה נאמן‬.‫ הן‬:‫ יש לך בנים? א״ל‬.‫ לאו‬:‫עדים? א״ל‬
.)‫לפסול את בניך (ב׳ יבמות מז ע״א‬

Our Rabbis taught: “And judge righteously between a man and his
brother, and the proselyte that is with him” (Deut 1:16); from this text
Rabbi Yehudah deduced that a man who converted in the presence of a
Court is a convert; but he who does so privately is not a convert.

13 The Vilna edition omits the ‫ ו‬against all MSS. The ‫ ו‬is important in establishing the alleged
unity of the two parts of the source.
THE BAVLI ON DEMARCATING IDENTITY 91

It once happened that a man came before Rabbi Yehudah and told him:
I have converted privately. Rabbi Yehudah said: Have you witnesses? He
said: No. Have you children? He said: Yes. (He said to him): You are trusted
as far as your own disqualification is concerned but you cannot be relied
upon to disqualify your children. (bYev 47a)

This tannatic text in the Bavli has two parts. The first is a halakhic midrash,
suggesting that the biblical verse Deut 1:16 was the source of Rabbi Yehudah’s
demand for a court of conversion; the second part is a case in which Rabbi
Yehudah did not accept a convert who claimed that he had converted privately
and has no witnesses (bYev 47a). I claim that a significant gap between the
two parts of the baraita is evident and can arouse suspicions. In both parts
“private conversion” is presented as the wrong option and is contrasted with
another model, which is assumed to be the appropriate one. However, in the
second part witnessed conversion is presented as the appropriate option. Only
in the first part, in the halakhic midrash, is it suggested that Rabbi Yehudah
demanded a court for conversion. This seems like a classical case of deducing
the view of a rabbi from a tradition about his ruling in a particular case and
attributing to him something he never said. And indeed the first part of the
baraita is completely Babylonian, with no parallels in Palestinian rabbinic lit-
erature. In a parallel discussion in the Yerushalmi we find another tanaitic text,
discussing a slightly different case:14

‫ ומטבילין אותו בשבת‬,‫ נאמן‬,‫ מלתי ולא טבלתי‬:‫ ואמ׳‬,‫ והיו לו בני׳‬,‫גר שמל ולא טבל‬
.)]‫(י׳ קידושין ד ז [סו ע״א‬

A convert that was circumcised but was not immersed and had children,
and he said: I was circumcised but did not immerse—he is trusted, and
he may be immersed on Shabbat ( yQid. 4:7, 66a).

In the Yerushalmi the issue is the status of the sons of a convert who testifies
that he was only circumcised and not immersed—the question of private con-
version is not brought up at all. It seems that this is an example of the rephrasing

14 The claim for seeing these texts as parallels, namely as revised versions of one original
text, is based on the talmudic discussions in their entirety. Since the amoraic materials
reveal that the sugyot as a whole probably originated in the same nucleus discussion, it
is reasonable to believe that the tannaitic texts cited in the talmudim are also parallels,
despite the immense difference. For a detailed presentation of this argument see the
appendix in Lavee, “The ‘Tractate’ of Conversion.”
92 Lavee

of tannaitic materials in a manner that makes them fit the new agenda. We can
present three steps in the development of this tannaitic source:

The case of a convert who underwent partial fulfillment of conversion

Palestinian Talmud The case, as presented in The halakhic midrash

the Bavli (second part of presented in the Bavli as the
the baraita) source for the ruling ( first
part of the baraita)
A convert who testifies A convert who testi- A convert who was con-
that he was circumcised fies that he converted verted privately, without
but not immersed. privately, and has no a court, is not a valid
witnesses. convert.

We are confronted here with a confusing literary situation in which the very
same talmudic unit presents products of different literary processes. In three
cases the court of conversion is being read into earlier texts in which it was not
included at all, and in one case the court of conversion was probably inserted
into the early text and is presented as an early authoritative view. The existence
of both textual processes is also evident in the stringent measures our literary
unit requires concerning the acceptance of converts. I will first present two
cases in which the Bavli presents reworked tannaitic material, and then one
case in which it preserves the tannaitic material as is, but explicitly suggests an
emendation for it. The final outcome of both literary processes is the same: an
imagined tannaitic stringency.
The baraitot in this unit tend to present stringent views not documented in
the Palestinian parallels, or at least to attribute the stringent view to the major-
ity of the sages.

i. The first baraita in our ‘tractate’:

yQid 3:12, 64a (cf. Gerim 1:2) bYev 46a

‫ הכל הולך אחר‬,‫ טבל ולא מל‬,‫גר שמל ולא טבל‬ ‫ הרי זה‬:‫ אומר‬16‫ רבי אליעזר‬,‫ שמל ולא טבל‬15‫גר‬
.‫ דברי רבי אליעזר‬,‫המילה‬ .‫ שכן מצינו באבותינו שמלו ולא טבלו‬,‫גר‬
,‫טבל ולא מל‬

15 MS Moscow Guenzburg 1017 reads ‫גוי‬.

16 MSS Munich 95 and Vatican 111 read ‫אליעזר‬.
THE BAVLI ON DEMARCATING IDENTITY 93

.‫ אף הטבילה מעכבת‬:‫רבי יהושע אומר‬ ,‫ שכן מצינו באמהות‬,‫ הרי זה גר‬:‫ר׳ יהושע אומר‬
.‫שטבלו ולא מלו‬
‫ אין‬,‫ מל ולא טבל‬,‫ טבל ולא מל‬:‫וחכמים אומרים‬
.‫ עד שימול ויטבול‬,‫גר‬

A convert who was circumcised but did A convert who was circumcised but did
not immerse, immersed but was not cir- not immerse, Rabbi Eliezer says: He is
cumcised, all (cases) follow his circum- a convert, as we found that our fathers
cision, [these are] the words of Rabbi were circumcised but did not immerse.
Eliezer. [A convert who] immersed but was not
circumcised:
Rabbi Yehoshua says: Immersion is also Rabbi Yehoshua says: He is a convert, as
indispensable. we found that our mothers immersed but
were not circumcised.
And the sages said: A convert who
Immersed but was not circumcised, [or]
was circumcised but did not immerse—
he is not a convert unless he was both cir-
cumcised and immersed.

Note that it is only in the Babylonian version of the baraita that the majority
of sages demand both conversion and immersion. Surprisingly, their words are
the same as those attributed to Rabbi Yohanan at the end of the discussion of
this baraita.

ii. The third baraita in our “tractate”:

Sifra qedoshim, pereq 8:1 bYev 46b–47a

]. . .[ ?‫ יכול קבלו‬,‫ גר אני‬:‫בא ואמר לך‬ ]. . .[ ?‫ יכול נקבלנו‬18,‫ גר אני‬:‫מי שבא ואמר‬
‫ ובחוץ לארץ‬,‫בארץ אינו צריך להביא ראייה‬ ‫ בח״ל אין צריך להביא‬,‫בארץ צריך להביא ראיה‬
17.‫צריך להביא ראייה‬ .‫ דברי ר׳ יהודה‬,‫ראיה‬
,‫ בין בארץ בין בחוצה לארץ‬:‫וחכמים אומרים‬
.‫צריך להביא ראיה‬

17 As in MSS Vatican 66 and Vatican 31. MS NY reads: ‫בארץ צריך להביא ראיה ובחוץ לארץ‬
‫אינו צריך להביא ראיה‬.
18 MS Vatican reads ‫איני‬, probably an error.
94 Lavee

If a man came and said to you (singu- If a man came and stated, “I am a con-
lar): I am a convert, should you (singular) vert,” should we accept him? [. . .]
accept him? [. . .]
In the Land proof must be provided; In the Land proof must be provided;
abroad, no such proof need be provided. abroad, no such proof need be provided.
These are the words of Rabbi Yehudah.
But the Sages say: Both within the Land of
Israel and abroad, proof must be provided.

Here again, it is only in the Bavli that the majority of the sages demand that
a man claiming to have been converted should bring evidence for his con-
version both in the Land of Israel and abroad and these are the same words
as those attributed to Rabbi Yohanan. In both cases the last sentence in the
baraita attributing this stringent view to the majority of the sages is missing
in the Palestinian parallels of the tannaitic text (I suspect that in these cases
the tannaitic sources were reworked in accordance with views attributed to
the leading Palestinian amora). We should recall that, in the discussion of the
fifth baraita which we saw above, the Bavli had suggested that Rabbi Yohanan
amended the number of sages participating in the conversion procedure to
fit the statement attributed to him. So, once again we witness two different
mechanisms in the same talmudic unit: in the first and the third baraita we
find a reworked Babylonian version that fits the view of Rabbi Yohanan; in the
fifth baraita we find a version that probably preserved an old model and an
explicit suggestion that Rabbi Yohanan amended it.
We should ask how it happened that the first and the third baraitot were
transmitted to a new stringent versions that purports to fit the view of Rabbi
Yohanan, while regarding the fifth, the Bavli reports (a real or imagined) process
in which the text was changed according to his opinion. Is it possible that, his-
torically, the other baraitot were also reworked by Rabbi Yohanan, but the Bavli
only reported his act in the case of the procedure of conversion? Or maybe the
baraitot were amended even later, in order to fit the views attributed to Rabbi
Yohanan? My reconstruction is that the stringency attributed to Rabbi Yohanan,
probably rooted in authentic Palestinian traditions, was shaped into a series of
meimrot attributed to him.19 This stringency was later perceived as representing

19 None of them have a verbatim parallel in Palestinian sources! The demand for both
immersion and circumcision is in accordance with his position in a case presented in
both talmudim (bYev 46a—before the beginning of our “minor tractate”; the threefold
story in the Bavli has three separate parallels in the Yerushalmi: yQid 3:9, 64b; yAZ 2:8,
42a, 4:8, 44b, and hence might be a statement that was phrased on the basis of Rabbi
Yohanan’s acts in this narrative, and not on the basis of his actual saying.
THE BAVLI ON DEMARCATING IDENTITY 95

the opinion of the majority of the tannaim, and hence was also introduced
into the tannaitic texts or was suggested explicitly as an emendation of these.
We see here incoherence, in terms of the modes of development of the text,
that reflects coherence and consistency in terms of the conceptual and social
constrains that were involved in shaping the text. These constrains promoted
the new strict and institutionalized agenda, in some cases by reworking tan-
naitic materials, in others by reporting amendments of tannaitic materials and
in others by adding the new values only in late comments referring to the ear-
lier texts.

The fifth baraita, to which I have already referred above, adhered to the model
of witnessed conversion by preserving the number of two sages involved in the
procedure. However, as discussed in detail by Shaye Cohen, this version is sig-
nificantly different from another one, preserved in the minor Tractate Gerim.
Cohen summarized the differences between the versions as follows: “In Gerim
the ceremony is becoming less of a vehicle to ensure the compliance of legal
norms and more of an initiation ritual.”20 He too sensed that the Babylonian
version leans towards the institutionalization of conversion.

1.4 The Chronological Prism

The process of reworking tannaitic material has another implication. Not only
does the text promote a certain agenda, it also creates a certain chronologi-
cal prism, which presents this agenda as having been already suggested and
legislated in earlier days, in the time of the tannaim. If we were to write the
history of the conversion procedure only on the basis of the Bavli, accepting
the naïve assumption that the texts represent the original statement of their
transmitters, we would have argued that already in the 1st century, in the days
of Rabbi Eliezer and Rabbi Yehoshua the majority of sages demanded both
circumcision and immersion, as suggested in the first baraita, and two genera-
tions later, in the days of Rabbi Yehudah, the institution of a conversion-court
already existed, as can be observed in the fourth baraita. This picture cannot
be supported on the basis of Palestinian rabbinic sources.

2 The Perception of Conversion: “A Convert is like a Newborn Child”

I mentioned earlier that the last baraita of the Babylonian “minor tractate”
of conversion contains the statement: “A convert is like a newborn child.”

20 See S. J. D. Cohen, Beginnings of Jewishness, 198–233. We differ in our assessment of the

history of these texts, but agree about the content.
96 Lavee

The inclusion of the Aramaic suffix: “‫גר שנתגייר כקטן שנולד דמי‬,” suggests that
this is not an original tannaitic phrase, but rather a Babylonian version of it.
Indeed, the development of the motif of the convert as a new-born, its evolu-
tion into this specific phrase and the legal and conceptual values attached to it
reveal again the unique voice of the Bavli.
The phrase “A convert is like a newborn child” appears in bYev 48b as part
of the claim that the convert is not punished for sins performed prior to his
conversion, since he is now considered as a newborn person, who did not exist
previously. The deeds of the other person, who was there before his conver-
sion, are not his. In addition, this phrase is repeated in the Bavli in halakhic
passages pertaining to the legal status of the convert’s kin. If the convert is a
new being that had not existed previously, then the relatives he had prior to
his conversion are no longer to be considered to be his relatives (bYev 22b; 62b;
97b; bBekh 47a).

2.1 The Aggadic Context: Atonement of Sins or Abolishment of the Sinner?

I will first discuss the issue of punishment for wrongdoings before a person’s
conversion. In this realm the image of the convert as a newborn is also found in
Palestinian rabbinic sources, including the parallel baraita to the one conclud-
ing Tractate Gerim (2:5):21

Gerim 2:5 bYev 48b

‫ רבי יהודה‬.‫ דברי רבי יוסי‬,‫נפרעין מן הגרים‬
.‫ אלא הרי הוא כבן יומו‬,‫ אין נפרעין ממנו‬:‫אומר‬
‫ מפני מה הגרים‬:‫אמר רבי חנינא בן גמליאל‬ ‫ מפני‬:‫רבי חנניא בנו של רבן גמליאל אומר‬
‫ קודם לשנה וקודם‬,‫ מפני שהן מחשבין‬22?‫מעונין‬ ?‫מה גרים בזמן הזה מעונין ויסורין באין עליהן‬
.‫ שאגבה את חובי ושאעשה את צרכי‬,‫לשתים‬ ‫ רבי יוסי‬.‫מפני שלא קיימו שבע מצות בני נח‬
‫ הרי‬,‫ אלו ממון הן נחסרין‬:‫אמר לו רבי יוסי‬ ‫ אלא מפני‬.‫ כקטן שנולד דמי‬,‫ גר שנתגייר‬:‫אומר‬
‫ אלא שהם קוברים בנים ובני בנים‬,‫הוא כדבריך‬ ‫מה מעונין? לפי שאין בקיאין בדקדוקי מצות‬
‫ מפני מה‬,‫ וייסורין באין עליהן‬23‫וחלאים רבים‬ :‫ אומר משום ר׳ אלעזר‬24‫ אבא חנן‬.‫כישראל‬
.‫מעונין? מפני שבע מצות שנצטוו עליהן בני נח‬ ‫ אחרים‬.‫לפי שאין עושין מאהבה אלא מיראה‬
‫ מפני ששהו עצמם להכנס תחת כנפי‬:‫אומרים‬
.‫השכינה‬

21 For a more detailed comparison of the parallels see Lavee, “The Convert,” 97–8.
22 MS Munich 95 reads here ‫ מאמינין‬and the second occurrence of this word ‫מאמינים‬.
23 MS NY JTS Ena 2237 rab. 34 reads ‫רעים‬.
24 MSS Moscow Guenzburg 1017 and Munich 95 read ‫חנין; אבא חנן—משום ר׳ אלעזר‬: MS
Moscow Guenzburg 594 reads ‫וי״א‬.
THE BAVLI ON DEMARCATING IDENTITY 97

Converts are punished in the opinion

of Rabbi Yosi; Rabbi Yehudah says: He is
not punished, but he is like a one-day-old
(infant).
Rabbi Hanina ben Gamaliel said: Why It was taught: Rabbi Hananiah ben
are converts afflicted? Because they cal- Rabban Shimeon ben Gamaliel said: Why
culate a year or two years before their are converts at the present time oppressed
conversion, saying: When I have collected and visited with afflictions? Because they
my debts and when I have dealt with my had not observed the seven Noahide com-
needs [I will convert]. Rabbi Yosi said mandments. Rabbi Yosi said: One who has
to him: If they were to lose money, you become a proselyte is like a child newly
would be right; but they bury their chil- born. Why then are proselytes oppressed?
dren and grandchildren and sore diseases Because they are not so well-acquainted
and chastisements befall them. Why with the details of the commandments
are they afflicted? Because of the Seven as the Israelites. Abba Hanan said in the
Noahide laws they were commanded name of Rabbi Eleazar: Because they do
upon. not do it out of love but out of fear. Others
said: Because they delayed their entry
under the wings of the Shekhinah.

The two texts vary significantly, but their nature as parallel adaptations of an
original Urtext cannot be doubted. Both texts deal with the question of afflic-
tion of converts and present the motif of the convert as a newborn. Both texts
share the same answers: the Noahide laws, and the delay of conversion. The
materials are, however, organized in a different manner: as two separate dis-
putes in Gerim, and as one dispute in the Bavli. For the sake of our discussion
we should note that the Bavli reworks the text using unique terminology, dif-
ferent from the one found in the Palestinian source. There we find: ‫הרי הוא‬
‫“( כבן יומו‬behold, he is like a one day old infant”). The phrase ‫ כבן יומו‬is differ-
ent semantically from the Babylonian ‫כקטן שנולד‬. The Babylonian phrase sug-
gests an analogy between the convert and the newborn, as well as between the
processes of conversion and the process of birth. The Palestinian phrase only
refers to the convert’s new age. The Babylonian phrase is a literary reflection of
the further emphasis on the legal process formerly presented. The comparison
to the act of birth is substantial. The word ‫ נולד‬here seems to carry the legal
connotation it has in discussions of things that were created suddenly and did
not exist previously, as in the context of Shabbat and Festivals (“an egg that was
born in . . .,” mBes 1:1).25 This phrasing implies that the convert is not punished

25 The word ‫ נולד‬is used as a category, as is common in later layers in the Bavli, see bBes 2a.
98 Lavee

for his former deeds because they are not his deeds. He is a new person. The
argument in favor of this connotation of the Babylonian phrasing is also based
on the legal use of the phrase, which will be discussed later on. This is a nice
example for the halakhic resonance of haggadic expressions in the Bavli.
In contrast, the Gerim term ‫( בן יומו‬and in related sources ‫ )בן שנתו‬is used
in Palestinian sources to mark the status of the convert after his sins were
forgiven and atoned. There is a perception of a continuity of the personality
here: he sinned, he converted, his sins were atoned, and now, like a one-day-old
baby, he does not carry the burden of any sin. This argument is based on many
sources that describe atonement as forgiveness of the sins of various groups
of people who are involved in rites of passage, including a convert, a groom,
a newly nominated scholar, an anointed king, and the Israelites in the annual
and the daily rites of atoning sacrifices.26 In addition to the slight but signifi-
cant difference in meaning between the Babylonian and Palestinian phrasings
of the image, we should note the semantic of separation created in the Bavli.
Here we find a unique phrase used only for converts, in contrast to the variety
of phrases used in the Palestinian literature applying the image interchange-
ably for Jews and converts alike.

Other Palestinian sources did preserve an image, which is slightly closer to

the motif of birth in the Bavli, referring to the convert as ‫( בריה חדשה‬a new
creature or a new creation).27 In one case we find that the Bavli preserved a
halakhic midrash that speaks about the atonement of a convert’s sins, and
hence adheres to what I presented here as continuity of the personality of the
convert.28 Given the collective and fragmentary nature of rabbinic works we
should not expect absolute coherence of a given sub-corpora. What we can
identify are the dominant trends in each centre and the delicate ways in which
they were then expressed; in this case the wording of one motif.

2.2 The Halakhic Context: The Severing of Former Family Ties

The unique nature of the Babylonian phrase is evident also from its use in the
context of a convert’s severing his relationship with his former kin. Neither
the specific phrase nor the image of the newborn is used in this context in
Palestinian rabbinic literature. The legal idea itself, however, is not missing.
Palestinian rabbinic sources suggest that some family relations of the convert

26 yBik 3:3, 65c–d: atonement for the sins of a convert, a groom, a king, and a scholar;
yRH 4:8, 59c and PRK et qorbani, 4: atonement of the sins of Israel.
27 See GenR 39:11; PRK: 4:8.
28 bZev 116b; MdRY Yitro 1; MdRSbY 18:1 (based on Midrash Ha-Gadol).
THE BAVLI ON DEMARCATING IDENTITY 99

have no legal validity. However, they differ from the Babylonian sources in their
phrasing, their extent and in the different conceptual framework in which
they are presented. As already stated, they are not marked or explained by the
phrase “a convert is like a newborn child.”
Tannaitic literature formulates laws in which the convert is assumed not
to be related to his prior relatives: he is not considered the offspring of his
father for the sake of inheritance,29 levirate marriage,30 transgressions against
parents31 and the status of the first-born.32 Amoraic Palestinian literature in
this context also testifies to tannaitic materials concerning incest.33 However,
in the rare cases where the later Palestinian sources depart from casuistic lan-
guage and turn towards a conceptualization of these laws, they do not use the
image of the convert as a newborn and of conversion as rebirth. Rather, they
suggest a different conceptual framework that correlates the lack of kinship
of the convert with the lack of valid halakhic kinship among gentiles: “There
is no paternity for gentiles.”34 A legally valid kinship between a gentile and his
son is never established. Hence, upon conversion, this situation continues, and
the convert is still not considered the offspring of his biological father, just as
he was not considered such before conversion. There is no severing of familial
relations due to the act of conversion, but rather continuity in the lack of legal
valid kinship.
Let me present the most explicit example:

bYev 62a yYev 2:6, 3d

,‫ היו לו בנים בגיותו ונתגייר‬:‫איתמר‬ :‫גוי שבא על גויה וילדה‬
,‫ קיים פריה ורביה‬:‫ר׳ יוחנן אמר‬ .‫ גוים יש להם יחסים‬:‫ר׳ יוחנן אמ׳‬
.‫ לא קיים פריה ורביה‬:‫וריש לקיש אמר‬ ‫ גוים אין להן יחסין‬:‫ר׳ שמעון בן לקיש אמ׳‬
‫ דהא הוו‬,‫ קיים פריה ורביה‬:‫רבי יוחנן אמר‬
‫ גר‬,‫ לא קיים פריה ורביה‬:‫ליה; וריש לקיש אמר‬
.‫שנתגייר—כקטן שנולד דמי‬

29 As in the field of inheritance, i.e mShevi 10:9; mBQ 9:11; tBQ 10:17 and many others. See also
Lavee, “The Convert,” 29–44.
30 mYev 11:2, tYev 2:5–6; 12:2; SifDeut 289.
31 Sifra qedoshim, pereq 9:9;. cf. yYev 11:2, 11d; MdRSbY 21:17.
32 mBekh 8:1; cf. MdRSbY 13:1; SifZut 18:15.
33 yYev 11:2, 12a; GenR 18:5.
34 yYev 2:6, 3d; GenR, ibid.
100 Lavee

It has been stated: If someone had chil- A gentile who had sexual intercourse
dren while he was a gentile, and then he with a gentile women, who gave birth:
was converted:
Rabbi Yohanan says: He has already ful- Rabbi Yohanan says: Gentiles have
filled the obligation to be fruitful and kinship.
multiply, since he had [children].
Resh Laqish said: He has not fulfilled the Rabbi Shimon ben Laqish says: Gentiles
obligation to be fruitful and multiply—a have no kinship.
convert is like a newborn child.

At first sight these seem to be two different disputes, one dealing with the legal
validity of the relations of non-Jews (Yerushalmi), and the other with the legal
validity of kinship of converts. However, the one leads to the other—if indeed
there is no valid kinship for gentiles, then the convert has no valid paternity
of offspring born prior to his conversion. The common origin of these two dis-
putes is evident when considering the wider sugya. The same halakhic issues
are raised—validity of firstborn laws and fulfillment of procreation; the same
verses are presented in both talmudim. Since the verses focus on gentiles, it is
evident that the original issue was the status of relationships of non-Jews, as
in the Yerushalmi, and the Bavli has to make some efforts to explain the verses.
Within these dialectical maneuvers, the Bavli explains that Resh Laqish con-
siders only converts as having no legal kinship to their previous relatives, but
non-Jews do establish legal kinship among themselves.
It is possible that the concept we find in Palestinian sources, and maybe also
the original motivation for this legislation is close in nature to Greco-Roman
legal perception of citizenship and kinship. Only legal sons may inherit. If one
had intercourse with any of his female-slave and she gave birth, the offspring
would be a slave and would not be considered an heir.35
In contrast, the Bavli not only justifies these laws with the slogan “the con-
vert is like a newborn child,” but also suggests a wider scope for its validity,
applying it also to the realms of testimony, procreation, and the inheritance of
the firstborn (bYev 62a; bBekh 47a; bYev 22a). The Bavli stresses that, according
to this new perception, the lack of kinship validity also applies to the relation-
ship of the convert to his biological mother (bYev 22a), even in the realm of
incest (ibid., and bYev 97b). In this area the phrase “The convert is like a new-
born child” is always quoted in the late anonymous strata of the Bavli, and in
many cases the views that support this concept or discuss it are ascribed to late

35 See Lavee, “The Convert,” 275–9.

THE BAVLI ON DEMARCATING IDENTITY 101

amoraim from the generations of Rava, his students and onward.36 In addition,
the Bavli grants the convert’s lack of kinship the status of ‫מדאורייתא‬, a biblical
law (bYev 22a).37
By granting this principle the value of biblical law, the Bavli applies the same
dynamic identified above, in the saying of Rav Sheshet, regarding the institu-
tion of a conversion court: The rationale of leniency is used for the purpose of
establishing a novel conceptual framework. The idea that the lack of kinship is
from the Torah is discussed in the Bavli in a context that limits the farfetched
implications of lack of kinship: according to the Torah, one would go as far as
permitting incest between a convert and his mother, as he is a new person, and
she is not considered his mother. However, the sages deny this possibility, and
rule that this is prohibited according to their view (‫)מדרבנן‬.38 This talmudic
move enables the harmonization of the earlier view—according to which the
convert’s lack of previous kinship applies only with regard to paternal relations
(probably because of the idea of no paternity for a gentile)—with the later per-
ception that maternal relations are also included. As such, this talmudic move
reflects the understanding that the later concept is a governing principle, and
laws that only focus on paternal relations can now be understood as based on
a rabbinic limitation of the (imagined) biblical law, and not as originating from
a different (tannaitic) concept. As in the case of Rav Sheshet and the court for
conversion, here too a claim for later leniency enables the application of the
new concept on earlier sources and traditions.
Another phenomenon attested with regard to the image of the convert as a
newborn is the dual expression of the Babylonian tendencies, both in rephras-
ing earlier materials, and in the additions in the anonymous layers. The Bavli
quotes the view of Resh Laqish that a convert is obliged to procreate even if
he had sired offspring prior to his conversion because “the convert is like a
newborn child” and his previous offspring are not considered his.39 Were we
to adopt a simplified analysis, based on textual “archeological excavations,”
aimed at removing the later layers, we may have claimed that Resh Laqish
ruled that the convert had to fulfill the obligation of procreation, and that the
later stam explained that he said so because of his concept regarding converts
(in italics in the chart above). We could have claimed that his original words
about converts can easily be explained on the basis of the concept of “no pater-
nity for gentiles” and that it is only the later anonymous layer of the Bavli that

36 As in bYev 22a; bQid 17b.

37 A hint found in Shevi 10:9, 39d in the context of inheritance.
38 yYev 2:6, 3d.
39 bYev 62a.
102 Lavee

explained them on the basis of the new Babylonian standard. However, in the
parallel passage in the Yerushalmi we discover that the concept “no paternity
for gentiles” is explicitly attributed to Resh Laqish. Yet, as explained above,
the whole discussion in the Yerushalmi was only about non-Jews.40 It is hence
possible to suggest that not only did the governing voice of the Bavli attribute
the new concept to Resh Laqish, but also that in order to do so it reworded
his statement somewhere on its long geographical and chronological journey
from Palestine to Babylonia.41

To conclude, the concept of the convert as a newborn child is significantly

developed in the Bavli. It is applied to new fields and laws, is granted a higher
status and validity, and is rephrased in a manner that further emphasizes the
nature of conversion as a birth and of the convert as a person who did not exist
before the conversion. As such it seems to echo the evolution of the conversion
procedure that was presented in the first part of the present study. Conversion
to Judaism is perceived as a significant irreversible step, an utter severing of the
convert’s contact with his non-Jewish past.
We also noticed that the Bavli developed unique language regarding con-
verts, suggesting a unique semantic sphere that separates them from other
Jews. By doing so it demonstrates how limited the proclamation “once he has
immersed he is an Israelite in every respect” is, it being documented only in
the Babylonian version of the baraita on the conversion procedure (bYev 47b).

40 yYev 2:6, 3d. The difference was explained as related to other issues see J. Cohen, “Be Fertile
and Increase, Fill the Earth and Master It” The Ancient and Medieval Career of a Biblical
Text (Ithaca 1989) 144–52.
41 The textual evolution is slightly more complicated than presented in the body of the
present study. We should notice that the views of the amoraim in the Bavli are presented
in a casuistic manner, and only then does the stam suggest their conceptual explanation.
The Yerushalmi attributed to them conceptual claims, and not casuistic laws. Thus, in
terms of the form of their sayings there is an argument for the precedence of the Bavli.
The Babylonian form seems to be authentic, while the Palestinian one hints at the
possibility of a later rephrasing of the original saying. This form-based conclusion stands
in contradiction with the context based conclusion: it is very clear that the original issue
was that of non-Jews, as in the Yerushalmi, and that the Bavli had to maneuver in order
to make the sources fit his new agenda. I thus suggest the following textual development:
The original statement was a casuistic statement dealing with non-Jews. The Yerushalmi
only preserved the conceptual summary of it. The Bavli preserved a different version of
it, maybe with the minor change of the word ‫ גוי‬for ‫גר‬. Adhering to its dialectical style,
the Bavli added the conceptual layer as a stam comment, next to the reworked amoraic
material.
THE BAVLI ON DEMARCATING IDENTITY 103

Indeed, in the next part of this study I will show how the Bavli develops a nega-
tive perception of converts, hinting at support for the avoidance of marriage
between Israelites and converts.

3 The Negative Approach towards Converts

The rabbinic ambivalence towards converts is a broad and complicated topic.42

However, a novel comparative examination enables us to notice the unique
cultural work of the Bavli also in this regard. Here I will depart from my discus-
sion of the “minor tractate” on conversion in bYevamot and will focus on the
phrase “converts are as hard to Israel as a scab (‫[ )ספחת‬is to/in the skin].”43 The
use of this phrase demonstrates dynamics similar to the ones found regard-
ing the phrase “the convert is like a newborn child.” For instance, we find: a
phrase that is not documented in Palestinian rabbinic literature (even though
it is attributed to a Palestinian rabbi—Rabbi Helbo); that the image itself is
hinted at in Palestinian sources, but in a significantly different context; that
the phrase is used in the Bavli in two different contexts, demonstrating an
expansion of its application; a broadening of the semantic sphere related to
the phrase, correlating it with other negative statements on converts.
It seems that the original use of the phrase is found in bQid 70b, in a dis-
cussion that has already been recognized by others as reflecting unique
Babylonian anxieties concerning purity of genealogy.44 The use of the phrase
there hints to a tendency to avoid marriage with converts, and maybe even
with their descendents, as God only graces families of pure genealogy with

42 See for example the abundance of sources and references in G. Porton, The Stranger
within Your Gates: Converts and Conversion in Rabbinic Literature (Chicago 1995).
43 bNid 13b; bQid 70b and variants in MS Munich and in the 1489 Spanish print edition. bYev
47b and variants in MS Moscow Ginzburg 594 and in MS Cambridge 3207; bYev 109b and
variants in MS Moscow Ginzburg 594 and the Pesaro and Vilna editions. See also the
seemingly Gaonic halakhic section quoted in M. Ben Shlomo, Sekhel Tov, to Gen 17, Buber
edition, p. 19. This is part of a long treatise on circumcision and conversion. Buber already
pointed to other details in this section reflecting Gaonic traditions (see p. 20, n. 70); Alfasi,
Shab 39b. In Zohar, Va-Yehi 215b we find another variation, ‫בעור החי‬, probably a corrupt
conflation of ‫ בעור‬and ‫בבשר החי‬.
44 See R. Kalmin, “Genealogy and Polemics in Rabbinic Literature of Late Antiquity,” Hebrew
Union College Annual 67 (1996) 77–94; A. Oppenheimer and M. Lecker, “The Genealogical
Boundaries of Jewish Babylonia,” in: A. Oppenheimer and N. Oppenheimer (eds.),
Between Rome and Babylon: Studies in Jewish Leadership and Society‫( ‏‬Tübingen 2005)
339–55. J. L. Rubenstein, The Culture of the Babylonian Talmud (Baltimore 2003) 83–7.
104 Lavee

his presence. It is possible that this passage indicates an actual broadening of

the halakhic prohibition of intermarriage between priests and converts, apply-
ing it also to all Israelites.45 It is also probable that avoidance of marriage with
converts was actually practiced in Gaonic days on the basis of this sugya.46 Yet,
in that context the phrase “converts are as hard to Israel as a scab [to/in the
skin]” still carries only a limited meaning. The image of a scab, as a relatively
minor kind of impurity caused by leprosy, might mark the fact that bQid 70b
is dealing with a custom that is not as blemishing as the strictly prohibited
marriages of priests and first generation converts or Israelites and mamzerim.47
However, the other appearances of the phrase mark an expansion of its
application. It is used in order to justify the view that “evil after evil will befall
those who accept converts” (bYev 109b); it is used as an explanation for the
requirement to interrogate the convert as part of his conversion procedure,
which is now perceived as a deterrent to the potential convert from under-
taking the act (bYev 47a); and it is used in order to fortify the claim that con-
verts and those who play with children (namely pedophiles) delay the advent
of the Messiah (bNid 13b). The last claim is probably based on a correlation
between the messianic times and a certain number of pure Israelite souls: both
pedophiles and converts do not add souls to the pure multitude of Israelites
required for the dwelling of the Shekhinah in the midst of Israel.48 While this
use of this phrase is the most severe, the two other usages are more significant
in identifying the literary process typical to the Bavli. In bYev 109b and bNid
13b we find the expansion of the application of the image of the scab from the
limited (and probably original) issue of marriage to the question of the very
acceptance of converts into the midst of Israel.

45 The only other use of the term ‫( משפחות מיוחסות‬genealogically traced families) in the
Bavli refers to non-priestly families who took upon themselves the practice of priests
regarding the payment of a higher marriage price in the ketubbah (bKet 12b).
46 See Sheiltot Exodus, 43, Mirsky edition, pp. 28–9, and G. Blidstein, “4Q Florilegium and
Rabbinic Sources on Bastard and Proselyte,” Revue Qumran 8 (1974) 434, n. 19.
47 See Sifra tazria‘, nega‘im, pereq 1:4, quoted in bSot 5a–b. Note also the absence of ‫ספחת‬
from the mishnaic taxonomy in mNeg 7:1–2.
48 Also here it seems that the Babylonian idea is based on Palestinian building blocks, which
reflect a different agenda in their unique presentation and use in the Bavli. Cf. SifNum 84;
MdRY, bahodesh, 3. These Palestinian tannaitic midrashim were familiar with the image
of the Shekhinah dwelling in the midst of the multitude of Israel. However, it is only in
the Bavli (here and in other places) that the concept acquires practical implications, jus-
tifying a continuous requirement to multiply, beyond the minimum number of offspring
defined by the Mishnah.
THE BAVLI ON DEMARCATING IDENTITY 105

When searching for the Palestinian roots of the phrase, we discover once
again the novelty of its Babylonian setting and application. For example, in
the Palestinian Midrash, Pesiqta de-Rav Kahana, we find the use of the image
of a scab in the context of preventing inter­marriage. The daughters of Zion,
who were arrogant enough to believe that intercourse with conquering officers
would save them from harm, were afflicted by scabs. The scab deterred the
officers from having sex with the women, and they violently threw them out
of their chariots and ran them over. We should note the huge semantic change
the term undergoes: the illness that deterred the gentile officers and prevented
them from having intercourse with Israelite women becomes in the Bavli an
image of the convert. This reflects an attempt to dissuade Jews from marrying
converts, by applying to them the same disgusting characteristics used in the
Palestinian source to explain how unlawful intercourse between Jews and non-
Jews was averted.49
There is another case in which the term “scab” is used in Palestinian litera-
ture to single out non-Jews. A homily in LevR 15:9 identifies the different kinds
of leprosy mentioned in the Bible with various groups of non-Jews:

LevR 15:9 bQid 70a–71b

[1] The “other” defined as Gentiles: Proselytes:
‫ספחת‬ A rising (‫ )שאת‬is Babel Rabbi Helbo said: Converts
[. . .] A scab (‫ )ספחת‬is are as hard for Israel as scab
Media [. . .] A bright spot is to the skin . . .
(‫ )בהרת‬is Greece [. . .]
The plague of leprosy
(‫ )נגע צרעת‬is Edom
[. . .] In this world the
priest examines for lep-
rosy; but in the World
to Come, says the Holy
One, blessed be He, I will
render you clean.
[2] Eschatological purity Symbolic: Genealogic:
of Israel Thus it is written: “And I When the Holy One, blessed
will sprinkle clean water be He, purifies the tribes, He
upon you, and ye shall will first purify the Tribe of
be clean” (Ezek 36:25) Levy . . .

49 See PRK 11:6.

106 Lavee

In contrast to the former texts presented, these are not parallel sources, in
the strict sense. There is no ground to claim that they originated in the same
nucleus Urtext. In each source there are two haggadic treatments of the motif
of leprosy: the first is the use of leprosy for marking the “other” and the sec-
ond is the use of leprosy as marking the negative situation in the past, and
hence the hope for purification in the messianic future. However, the specific
application of this common sequence in the two texts (leprosy as marking; the
future messianic purification) suggests a significantly different context. In the
Bavli the scab serves as a metaphor for converts, while in the Palestinian work
it is applied as a metaphor for non-Jews; in the former the messianic future
promises the genealogical purification of the Jews from blemished social com-
ponents of impure origin, while in the latter it symbolizes purification, imply-
ing the cleansing of the heart.
However, we do have a parallel to the Babylonian image of purification,
which shows that the presentation of God first purifying the Tribe of Levi is an
expansion of a very limited exclamation in the Palestinian sources:

yYev 8:2, 9d bQid 70b–71a

‫ אף לעתיד לבוא אין הקב׳ה‬:‫ר׳ יסא בשם ר׳ יוחנן‬ ‫ כשהקדוש ברוך הוא‬:‫אמר ר׳ חמא ברבי חנינא‬
‫ מאי טע׳? ״וישב‬.‫נזקק אלא לשבטו שללוי‬ ‫ שבטו של לוי מטהר תחילה‬,‫מטהר שבטים‬
‫[מצרף ו]מטהר כסף וטהר את בני לוי ויזקק‬ ‫ ״וישב מצרף ומטהר כסף וטיהר את בני‬:‫שנא‬
.)‫אותם״ (מלאכי ג ג‬ ‫לוי וזיקק אותם כזהב וככסף והיו לי״י מגישי‬
.)‫מנחה בצדקה״ (מלאכי ג ג‬

Rabbi Yasa in the name of Rabbi Yohanan: Rabbi Hama bar Hanina said: When the
Also in the future, the Holy One, blessed Holy One, blessed be He, will purify the
be He, will only deal with the tribe of tribes, he will first purify the tribe of Levi
Levi. What is the reason? “The refiner will first, as it is written: “The refiner will sit,
sit, the purifier of silver, and he will purify the purifier of silver, and he will purify
the sons of Levi and refine them as gold the sons of Levi and refine them as gold
and silver” (Mal 3:3). and silver” (Mal 3:3).

First, the parallel passage in yYev 8:2, 9d is only concerned with cleansing the
Israelites of mamzerim, while the context in the Bavli implies extension of the
act of cleansing to include converts (note the complementary nature of this
idea to the other Babylonian claim about conversion delaying the advent of
the Messiah). Secondly, according to the Yerushalmi, God will only cleanse the
Tribe of Levi, while according to the Bavli He will begin with this tribe, that is,
He will continue with others. The Bavli thereby extends levitical genealogical
purity to the Israelites as a whole. Last, in the Bavli, cleansing seems to be part
THE BAVLI ON DEMARCATING IDENTITY 107

of an expected and desired eschatological future, a realization and restoration

of an ideal situation missing from the mundane present. The Yerushalmi, in
sharp contrast, presents the act of cleansing as problematic: Rabbi Hoshaya
responds to the claim with a rhetorical question: “because we are Levites,
should we lose?” This question marks the cleansing as an act of harassment,
and confirms, once again, that it will not include other tribes.

To conclude, the rabbinic statement “Proselytes are as hard to Israel as a scab

is (to/in the skin)” played a significant role in the development of a negative
attitude towards converts elaborated in the Bavli. The phrase is attributed to a
Palestinian rabbi, Rabbi Helbo, but is only documented in the Bavli. It is cloned
again and again, and used for justifying other negative statements concerning
converts. Its meaning is expanded from a determent of marriage with converts
to an objection of their very acceptance. The original Palestinian phrase prob-
ably referred to gentiles and not to converts, and conveys a much more relaxed
approach towards converts.

4 Converting the Mission

The negative Babylonian attitude toward converts and conversion is expressed

also in the realm of missionary traditions found in rabbinic literature.50 I will
not discuss here the question of the historicity of these traditions and whether
they reflect the memory of a period in which missionary activity was per-
formed by Jews; I will concentrate on literary presentations. On the literary
level we find a relatively consistent phenomenon. Palestinian rabbinic sources
portray positively the spread of Jewish customs and beliefs among the gentiles.
This is mostly portrayed through the presentation of biblical figures, especially
Abraham, as positive exemplars. They drew gentiles near, brought them to
recognize God, the value of the people of Israel and their norms, and in rare
cases even stimulated their conversion.51 Babylonian parallels of those sources
tend to “convert” these traditions, internalize them, and present them as an
inner-Jewish issue. Instead of portraying the Jew who spreads “the gospel”
among the gentiles, they present the sage who distributes Jewish values and

50 For additional readings see: M. Lavee, “Converting the Missionary Image of Abraham:
Rabbinic Traditions Migrating from the Land of Israel to Babylon,” in: G. van Kooten (ed.),
Abraham, the Nations, and the Hagarites: Jewish, Christian, and Islamic Perspectives on
Kinship with Abraham (Leiden 2011) 203–22.
51 I.e. SifDeut 32; 336; GenR 39:2, 14; 82:12; 84:4.
108 Lavee

promotes the status of the scholars among the people of Israel. In the follow-
ing example, it is not Abraham whose fragrance is spread among the gentiles,
but rather the sage whose perfume is carried to the Jews:

GenR 39:252 bAZ 35b53

‫ ״לריח שמניך טובים וגו׳״‬54:‫ר׳ ברכיה פתח‬ :‫ מאי דכתיב‬:‫דרש רב נחמן בריה דרב חסדא‬
)‫(שה״ש א ג‬ )‫״לריח שמניך טובים״ (שה״ש א ג‬
?‫ למה היה אברהם דומה‬:‫אמר ר׳ ברכיה‬ ,‫ מגולה‬.‫למה ת״ח דומה? לצלוחית של פלייטין‬
‫ מוקפת צמיד פתיל‬55.‫לצלוחית שלפפלוסימון‬ .‫ אין ריחה נודף‬,‫ מכוסה‬.‫ריחה נודף‬
‫ כיון‬.‫ ולא היה ריחה נודף‬,‫ומונחת בזוית‬
56.‫שנטלטלה נדף ריחה‬
‫ טלטל עצמך ממקום‬:‫כך אמר הקב״ה לאברהם‬
.‫למקום ושמך מתגדל בעולם‬

Rabbi Berekhiah commenced: Rav Nahman bar Rav Hisda expounded:

“Your ointments have a goodly fragrance” What is the meaning of the verse: “Your
(Song 1:3). ointments have a goodly fragrance”
(Song 1:3)?
Said Rabbi Berekhia: To what was To what may a scholar be compared?
Abraham compared? To a flask of polia- To a flask of poliatum: When opened, its
tum, closed with a tight fitting lid, and odor is diffused, but if covered up its odor
lying in a corner, so that its odor does not does not diffuse;
diffuse; as soon as it was lifted (shaken),
its odor was diffused.
Similarly, the Holy One, blessed be He,
said to Abraham: Travel from place to
place, and thy name will become great in
the world.

Similarly the “making of souls” mentioned in Gen 12:5 is not understood by the
Bavli as the conversion of the household of Abraham, but rather the giving of
spiritual birth to students by their teacher, as the following synopsis between
Palestinian sources and the Bavli demonstrates:

52 Later Palestinian parallels: TanB lekh lekha 3; SongR 1:3:3.

53 The text is presented on the basis of MS NY JTS 15.
54 MS Vatican 60 reads ‫א׳ ר׳ ברכיה‬. Other variants are noted by Theodoer-Albeck.
55 MS Vatican 60 reads ‫שלפופילסמון מוקפת צמיד ופתיל‬.
56 MS Vatican 60 reads ‫שהיתה מיטלטלת היה ריחה נודף‬. Both versions seem to imply passiv-
ity and not a proactive missionary effort.
THE BAVLI ON DEMARCATING IDENTITY 109

SifDeut 32 GenR 39:14 AdRN B, 26 bSanh 99b

‫״את הנפש אשר עשו‬ ‫״ואת הנפש אשר‬ ‫כל המכניס בריה‬ ‫ כל‬:‫אמר ריש לקיש‬
‫בחרן״ (בראשית‬ ‫עשו״ (בראשית‬ ‫אחת תחת כנפי‬ ‫המלמד את בן חבירו‬
‫ והלא אם‬.)‫יב ה‬ ‫יב ה)? אלא אילו‬ ‫ המקום‬,‫שמים‬ ‫תורה מעלה עליו‬
‫מתכנסים כל באי‬ ‫ ויאמר‬.‫הגרים‬ ‫מעלה עליו כאילו‬ ,‫הכתוב כאילו עשאו‬
‫העולם לבראות יתוש‬ ‫ למה‬,‫שגיירו‬ ‫בראה וכאילו יצרה‬ ‫ ״ואת הנפש‬:‫שנאמר‬
‫אחד ולהכניס בו‬ ‫שעשו? כל המכניס‬ ‫ ״ואת הנפש‬:‫שנאמר‬ ‫אשר עשו בחרן״‬
‫ אלא‬.‫נשמה אין יכולין‬ ‫בריה אחת תחת‬ ‫אשר עשו בחרן״‬ .)‫(בראשית יב ה‬
‫שהיה אברהם אבינו‬ ‫כנפי שמים מעלה‬ .)‫(בראשית י״ב ה‬
‫ תחת‬57‫מגייר ומכנסן‬ ‫עליו הכתוב כאילו‬
.‫כנפי השכינה‬ ‫ כאילו יצרה‬,‫בראה‬
‫ ״ואת הנפש‬:‫שנאמר‬
‫אשר עשו בחרן״‬
.)‫(שם‬

“And the persons “And the persons Another Resh Laqish said:
that they had that they had interpretation. Everyone who
made in Haran . . .” made” (Gen 12:5)? Everyone who teaches Torah to
(Gen 12:5), . . . Rather, these are brings one his friends’ son,
Hence we learn the proselytes . . . person under Scripture counts
that Abraham Everyone who the wings of him as one who
converted people, brings a gentile the shekhinah, made him, as it
and gathered them near, Scripture counts is written: “And
under the wings of is as one who him as though the persons that
the shekhinah. created him. he created him they had made
and formed him, in Haran . . .”
as Scriptures (Gen 12:5)
says: “And the
persons that
they had made
in Haran . . .”
(Gen 12:5).

In another case the internalization of the missionary image is expressed by

a demonstration of how the act of martyrdom of Hananiah, Mishanel and
Azariah, and their miraculous rescue, influenced the devotion of Jews rather

57 As in M. Kahana, The Genizah Fragments of the Halakhic Midrashim (Jerusalem 2005) 253
[Hebrew], against ‫מכניסם‬, the Genizah version has stronger missionary connotations,
perhaps implying the gathering of people rather than their entrance into the group.
110 Lavee

than non-Jews. In the Palestinian source this act leads many non-Jews to con-
vert (SongR 1:3:3). In the Babylonian subversion of this story, the non-Jews
approach the Jews, saying: “How come you have such a great God and you
worship idols?” (bSanh 93a). The impact of the event results in a criticism of
the Jewish people.
Another example is the exegetical narrative concerning the birth of Isaac.58
All versions of the story share the same kernel: gentile women brought their
sons to Sarah for breastfeeding, and learned that she indeed gave birth in her
old age. It is the different contexts in which the plot is portrayed that grant it its
renewed meanings. In the Palestinian versions Sarah’s act is presented as part of
an effort to spread the knowledge of God through God’s miracles among non-
Jews (GenR 53:9) or even to show that non-Jews also gained benefit from God’s
miracles (PRK 22:1). This act can be interpreted as symbolizing the removal
of the babies from their former gentile family to the family of Sarah, turning
some of them into God-fearers (GenR 39:14) or even converts (PRK 43:5). In the
Bavli the plot is transferred to a polemical setting. The breast-feeding serves as
a proof of the mistake the non-Jews made in their suspicion of Isaac’s mater-
nity and their accusations that he could not have been born of Sarah. This is a
narrative of a polemical nature, which does not suggest any positive influence
of Jews on gentiles, but rather their victory. Such a polemical setting conveys
a sense of a continuous debate, preserving and emphasizing the boundaries
of identity.59 One must admit that the polemical setting found in the Bavli is
based on traditions that also appear in other places in Palestinian source as in
GenR 53:5–6 and in the versions of our exegetical narrative in the two Pesiqtas.
However, the Bavli preserved only this tendency and nothing of the missionary
undertones of the earlier attestation of the narrative in Genesis Rabba.
In all the cases presented here (and also in some others), missionary
motifs that are found in Palestinian literature disappear or are subverted in
the Babylonian parallels. This is not to say that we do not find rare cases in
which the Bavli preserves a hint of the missionary approach, as its midrash on
Gen 21:33—“And Abraham planted a tamarisk-tree in Beer-Sheba”—suggests.60

58 For a detailed account see M. Lavee, “ ‘Sarah Would Have Suckled Sons’: Diverting
Tendencies Toward non-Jews in the Development of One Midrashic Narrative,” in:
U. Ehrlich (ed.), Al Pi Ha-Be‘er: Jubilee Volume for Prof. G. J. Blidstein (Beer-Sheva 2008)
269–91 [Hebrew].
59 G. G. Stroumsa and O. Limor, Contra Iudaeos: Ancient and Medieval Polemics between
Christians and Jews (Tübingen 1996).
60 See R. Hayward, “Abraham as Proselytizer at Beer-Sheba in the Targums of the Pentateuch,”
Journal of Jewish Studies 49 (1998) 24–37.
THE BAVLI ON DEMARCATING IDENTITY 111

It is also true that in a complementary contrast, we can trace the use of some
of those motifs in internalized Jewish contexts in the literature of the Land of
Israel.61 Just as the Bavli preserved the model dominant in the Land of Israel
in one case, so were some Palestinian sources familiar with the model that
was dominant in Babylonia and cited them. As in former examples presented
in the present study, we are speaking here about a fragmentary and gradually
evolving literature, which never underwent coherent and consistent editing,
and hence preserves also other voices than those prevailing in each center.
Yet, the overall perspective, suggested by these geo-culturally separated sub-
corpora of rabbinic literature, is significantly different, showing a strong pres-
ence of missionary trends in Palestinian literature, in contrast with the almost
complete silence on the issue in the Bavli.

5 The Representation of Liminal Groups: The Case of the Samaritans

The missionary traditions in the Palestinian literature are related to another

phenomenon which is unknown to the Bavli: the status of God Fearers.62 The
missionary model presented in Palestinian rabbinic literature is not neces-
sarily aimed at full conversion, and hence permits the presence of a liminal
group of non-Jews that partially accept Jewish beliefs and customs.63 This
group, which is known from non-rabbinic literary sources as well as inscrip-
tions, can be identified in the Palestinian rabbinic sources.64 However, in the
Bavli the term ‫( ירא שמים‬God Fearer) loses its liminal meaning. The term no
longer points to this group, but rather describes a pious Jew (once again an

61 I.e. tHor 2:7. For further examples cf. bBer 10b with yBer 4:4, 8b; ySan 10:2, 28c. Consider
also the following partial list of examples of traditions only known in Palestinian litera-
ture: Joseph coerced the Egyptians to circumcise (GenR 90:6; 91:5. See L. H. Feldman, Jew
and Gentile in the Ancient World: Attitudes and Interactions from Alexander to Justinian
(Princeton 1993) 298, 409; Jethro went back to his place to convert his people (MdRY, Yitro
2 and parallels). Many non-Jews were astonished by the successful agricultural produc-
tion of the tribe of Issachar, and when they realized it came as a result of devotion to the
Torah, they converted (GenR 78:12).
62 See also L. H. Feldman, “Conversion to Judaism in Classical Antiquity,” Hebrew Union
College Annual 74 (2003) 115–56.
63 I must admit that my presentation here tends to use Babylonian categorization.
64 See S. Lieberman, Greek in Jewish Palestine (New York, 1942) 64–90; J. Reynolds and
R. Tannenbaum, Jews and God-Fearers at Aphrodisias (Cambridge 1987); Feldman, Jew and
Gentile; J. Levinson, The Twice Told Tale: A Poetics of the Exegetical Narrative in Rabbinic
Midrash (Jerusalem 2005) 137–48.
112 Lavee

internalization of the concept).65 This is one example of a wider tendency in

the Bavli to reject liminal models of identity, and to mark people as either fully
here or completely there. The rejection of the liminal or ambiguous status is
also exemplified in the case of the Samaritans, as I will show presently.66 It is
highly possible that the development of laws, narratives and images in Babylon
was done on a theoretical level, as the Jews of Babylonia were not exposed to
the group. However, the cultural and social function of those texts is the rejec-
tion of liminal identities.
Jews (or should we say Judeans?)67 and Samaritans perceived themselves as
distinct groups already before the rabbinic period, as is evident from Josephus’
reports on quarrels and debates between the two groups.68 However, it is well
known that early rabbinic literature preserved a very ambiguous approach
towards the Samaritans, in which they were considered Jews with regard to
some issues, and as non-Jews regarding others. This presentation of ambigu-
ity toward the Samaritans already takes into account the binary division of
people into Jews and non-Jews, and blurs the possibility—to which some tan-
naitic sources hint at an earlier stage of categorization—that the Samaritans
were not perceived as essentially different from other Second Temple sects and
inner-Jewish divisions, such as the Sadducees and the Pharisees. However, rab-
binic sources from Palestine testify to a gradual shift in the halakhic status of
the Samaritans, as evident from comparing materials in the Tosefta and the
Yerushalmi with texts from the Mishnah, an exercise that, due to lack of space,
I will not be able to examine here. The Bavli documents a further intensifica-
tion of the halakhic demarcation and exclusion of the Samaritans, amount-
ing to their full rejection. It is here that we find them classified as “completely
non-Jews” (‫)גוים גמורים‬.69 This tendency of the Bavli is also demon­strated in its
detailed discussions of the Samaritans. A nice example is the interpretation
suggested for the mishnaic permission to join with Samaritans in the blessing
after meal. This inclusion is presented in contrast to the exclusion of gentiles,
marking the Samaritans as different from the latter. The Bavli wonders how is it

65 See also Feldman, Jew and Gentile, 353

66 For a more detailed account and bibliography see M. Lavee, “The Samaritan May be
Included: Another Look at the Samaritan in Talmudic Literature,” in: M. Mor and F. V.
Reiterer (eds.), Samaritans: Past and Present (Berlin 2010) 147–73.
67 See S. J. D. Cohen, “Ioudaios: ‘Judaean’ and ‘Jew’ in Susanna, First Maccabees, and Second
Maccabees,” in: P. Schäfer (ed.), Geschichte—Tradition—Reflexion: Festschrift für Martin
Hengel zum 70. Geburtstag I: Judentum (Tübingen 1996) 211–20.
68 See R. Pummer, The Samaritans in Flavius Josephus (Tübingen 2009).
69 bHul 6a.
THE BAVLI ON DEMARCATING IDENTITY 113

possible that Samaritans are included in a ritual in which Jews, who are defined
as am ha-aretz (Jews who avoid that or another aspect of rabbinic observance),
are not included. The force of the question is much stronger than any answer
suggested later by Babylonian amoraim. The question represents the rhetoric
of the obvious. It sets down the conceptual grid according to which one has to
assess Samaritans in all cases, and indeed, post-talmudic legislation followed
the logic of the question and not of the answers, and banned Samaritans from
the ritual while including the am ha-aretz.70
Another Samaritan-related issue is the function of their imagined false con-
version in the past, which serves as the means for defining their identity. Here
we can trace the following rhetorical development: tannaitic sources use the
expressions ‫כישראל‬/‫( כגוי‬as a Jew/as a gentile) to characterize halakhah regard-
ing Samaritans in specific cases.71 The Yerushalmi already used this phrase to
explain exclusion or inclusion in legislation regarding them:72

.‫ כותי כגוי‬:‫ תיפתר כמאן דאמ׳‬:‫ אמר רבי בא‬.‫ אין מקבלין מידן‬,‫הנכרי והכותי ששקלו‬
.‫ כותי כישראל לכל דבר‬:‫ רבן שמעון בן גמליאל אומר‬.‫ דברי רבי‬,‫דאיתפלגון כותי כגוי‬

If a non-Jew or a Samaritan paid the sheqel, we do not receive it from

them. Rabbi Ba said: This should be explained according to the view that
a Samaritan is like a non-Jew, as they disputed: The Samaritan is like a
non-Jew; these are the words of Rabbi. Rabban Shimeon ben Gamaliel
says: The Samaritan is like a Jew in every aspect ( ySheq 1:4, 46b).

The rhetorical difference is significant, since from now on the expression can
be seen not as characterizing the halakhah in one or another specific case, but
as defining the status of Samaritans as well as explaining the legislation as fol-
lowing their defined identity. This mode of explanation is further developed in
the Bavli, where the contrast “as a Jew/as a gentile” is replaced with /‫גרי אריות‬
‫“( גרי אמת‬true converts/lion converts”), implying that the legislation about the
Samaritans is based on the assessment of the validity of their imagined past
conversion.73 This Babylonian setting demonstrates the centrality of proce-
dural, approved and legal conversion as the marker of the boundary of identity,
but it also serves as another stage in the growing exclusion of the Samaritans.

70 bBer 47b; Isaac Alfasi, Halakhot, Ber 35b.

71 E.g. tPeah 4:1.
72 Other examples are: yDem 4:3, 23c; yKet 3:1, 27a.
73 bBQ 38a; bNid 56b; bSanh 85b. An exception is bNid 44a.
114 Lavee

The novelty of the Bavli here is not in raising the issue of their past conver-
sion, but rather in the consistent use of it in various halakhic questions. The
question of the legal validity of the conversion of the Samaritans is also found
in the Yerushalmi, but only in the context of marriage. A comparison of this
discussion to its parallel in the Bavli reveals some phenomena already noted
above. Once again, we find the Babylonian tendency to create a new chrono-
logical prism, projecting the later and gradually evolved perception onto ear-
lier generations. The discussion of the imagined problematic conversion of the
Samaritans is found in the Yerushalmi in amoraic explanations of tannaitic
views. In the Bavli the explanations are ascribed to tannaim, as already noticed
by Elizur.74
Finally, I wish to present the Babylonian development of the narrative of the
exclusion of Samaritans. The following chart sum­marizes the parallel discus-
sions between the Yerushalmi and the Bavli:

yAZ 5:4, 44d bHul 5b–6a

The context: Status of Samaritan wine The context: Slaughter by Samaritans
Rabban Gamilel and his court issued a
decree excluding the Samaritans.
Rabbi Shimon ben Eleazar was warned They did so because of an event in which
regarding Samaritans’ wine. Rabbi Shimon ben Eleazar was warned
regarding their wine, and consequently
Rabbi Meir decreed to exclude them.

Rabbi Abbahu, in the name of Rabbi

Hiyya, Rabbi Ami and Rabbi Assi75
prohibited Samaritans’ wine.
They did so because of He did so because
(1) a case in which a non-Jew who was
suspected regarding his wine was seen in
a Samaritan locale;
(2) or (because of) an event in which
the Samaritans probably used wine of
non-Jews;

74 Cf. bQid 75a–b with yGit 1:5, 43b–c. Y. Elizur, “The Samaritans in Tannaitic Words,” in: Israel
and the Bible: Studies in Geography (Ramat Gan 1999) 393–414.
75 As in MS Vatican 122.
THE BAVLI ON DEMARCATING IDENTITY 115

(3) or (because of) an event in which the

Samaritans were forced by the Romans to
pour libation wine;
(4) or because (the Samaritans) have They found an image of a dove on Mount
a kind of a dove and they pour libation Gerizim, which they worshiped,
wine for it.

The setting in the Bavli is declarative; a court of Rabban Gamaliel ruled to

exclude the Samaritans since it was discovered that they worshiped an image
of a dove. The Palestinian text has a few amoraic explanations for the ban on
their wine; only one of them is the claim that they pour libation to a dove. Only
in the Bavli does this text become the foundational story for their exclusion,
and the text ends with the above-mentioned expression declaring them “fully
non-Jews.” Attributing this declaration/expression to Rabban Gamaliel76 once
again creates a chronological prism that presents the outcome of a long evolu-
tion as the product of a revolutionary decision by earlier rabbis.77

6 Conclusion: The Chronological Prism: From Revolution

to Evolution

The study of the details of each of the issues discussed here teaches that a
clear-cut presentation of the division between Palestinian and Babylonian
sources is not possible. Almost every “Babylonian” tendency presented here
can be traced to its earlier sources or heralds in Palestinian literature. At the
same time, one may find sources in the Bavli that still preserve or adhere to
the models prevailing in Palestinian literature. These sources were not sub-
ject to the constraints that yielded new versions and explanations manifesting
the later agenda documented mainly in the Bavli. I cannot suggest a defini-
tive chronology for the textual develop­ments presented here. This inability
to define chronologies is inherent to the nature of rabbinic literature as a
collective, fragmentary and gradually evolving literature. The processes pre-
sented here should not be perceived as the product of well-planned and orga-
nized literary activity of a certain redactor (or a generation of redactors) that
intentionally reworked former materials to fit their new agenda. Rather, we
should see it as the final product of a gradual process in which new percep-
tions diffused through creative and fluid transmission. The talmudic text was

76 Even if the intended “Gamaliel” is not Rabban Gamaliel of first century Yavneh.
77 Cf. bHul 5b–6a with yAZ 5:4, 44d.
116 Lavee

open to rephrasing, commenting and rearranging of materials over centuries

of transmission. This also enabled the induction and inclusion of influences
from the cultural milieu in which it evolved—that of Sasanian Babylonia of
the 4th to the 6th centuries, which is another important aspect, although I
have not had occasion to address it here.
I have also demonstrated the attribution of later developments to earlier
generations and to sages from the Land of Israel. I suggest here the thesis that
the Bavli has shaped the chronological prism of the rabbinic period, constantly
making the ripe fruits of a prolonged and gradual evolution seem to be the
work of earlier “founding fathers.” The pleasure of identifying Babylonian ten-
dencies projected upon earlier generations resulted in a scholarly discourse
that speaks about the “invention.”78 To a certain extent this is a modern reac-
tion to the revolutionary narrative that the Bavli hints at, consistently applying
later tendencies to the founding fathers of the rabbinic corpora of the early
generations of the tannaim, a narrative that was embraced by modern scholar-
ship until quite recently.79 My reading here does not aim to transfer the locus
of the invention from the tannaim to the later redactors of the Bavli. Rather,
it wishes to move from the discourse of revolution to that of evolution. The
sources examined here suggest that we should speak about a final product of
a gradual development that tends to attribute itself to the early stages of the
process. Instead of replacing one invention for another (for instance, present-
ing the late stammaim, as opposed to the tannaim, as the inventors of rabbinic
culture), I suggest an evolutionary model, understanding the Bavli as repre-
senting the final product of a prolonged process.

78 See M. Lavee, “Welfare and Education vs. Leadership and Redemption: The Stories about
Rabbi and Rabbi Hiyya as an Example of the Image of the Tannaitic Past in the Babylonian
Talmud,” Jewish Studies Internet Journal 8 (2009) 93, n. 140 [Hebrew].
79 An oft-cited example is the attitude towards am-haaretz. See L. J. Rubenstein, The Culture
of the Babylonian Talmud (Baltimore 2005) 124–42.
Midgets and Mules, Elephants and Exilarchs:
On the Metamorphosis of a Polemical
Amoraic Story
Geoffrey Herman

1 Introduction

The study of Babylonian Jewry has undergone considerable fluctuation in

recent decades. One of the more significant changes pertains to the way the
legendary sources are to be read. There has been a critical shift from a his-
torical towards a literary reading of the many talmudic legends that, in earlier
years, constituted the mainstay of historical narratives on this period. With
the emergence of the study of talmudic stories as a discipline in its own right,
with its incumbent methodlogical implications,1 historians have been revis-
ing their assumptions and reassessing the value of such sources for their own
concerns.2 In view of the need to submit such sources to close literary analysis,
one might wonder about the kind of use a historian could derive from them.

* Earlier versions of this study were read at What’s New, Changes and Developments in the
Study of Jews and Judaism, The Annual Graduate Conference in Jewish Studies, Hebrew
University, Jerusalem, April 30–May 1, 2003; and at the Midwest Jewish Studies Association,
The Seventeenth Annual Conference, University of Madison, Wisconsin, September 18–19,
2005.
1 See, for instance, J. L. Rubenstein, Talmudic Stories: Narrative Art, Composition, and Culture
(Baltimore 1999). Many have been inclined to avoid historical conclusions, and be satisfied
with the treatment of the sources from a literary perspective, alone. See Rubenstein, ibid.,
4–5, 209. O. Meir (Rabbi Judah the Patriarch: Palestinian and Babylonian Portrait of a Leader
[Tel-Aviv 1999] 19 [Hebrew]) announces that she does not relate to the historical aspect of
her subject, despite the fact that historical consequences are inevitably the result of her anal-
yses. On the erosion of our capacity to provide a useful record of the history of the Jews in
Babylonia cf. S. Schwartz, “Historiography on the Jews in the ‘Talmudic Period’ (70–640 CE),”
in: M. Goodman (ed.), The Oxford Handbook of Jewish Studies (Oxford 2002) 109; D. Goodblatt,
“Towards the Rehabilitation of Talmudic History,” in: Baruch M. Bokser (ed.), History of
Judaism, The Next Ten Years (Chicago 1981) 31–44, and cf. A. Baumgarten, “Rabbinic Literature
as a Source for the History of Jewish Sectarianism in the Second Temple Period,” Dead Sea
Discoveries 2 (1995) 34, n. 68, on the apposition between stories and history.
2 See, most recently, for example, A. Tropper, Like Clay in the Hands of the Potter: Sage Stories
in Rabbinic Literature (Jerusalem 2011) [Hebrew]. The prospective historian of Babylonian

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�04

118 HERMAN

In this short study I subject one such story to source-critical analysis. It

is, however, a story that has played a vital role in earlier historical studies of
Babylonian Jewry. I shall probe the kind of adjustments the historical narrative
of Babylonian Jewry must make in light of the results that a genre-sensitive
analysis will produce. The story I shall examine exists in two versions,3 one
in the Yerushalmi and the other in the Bavli. The existence of two versions
stemming from geo-culturally distinct corpora will provide an opportunity
to observe the subtle manner in which the respective cultures have left their
mark. I shall first subject the sources to a literary analysis and then turn to
explore the potential historical implications of such an analysis.

Jewry has been offered plenty of helpful advice on how to practice her or his craft, particu-
larly when handling the kind of sources that are particularly susceptible to literary study.
Friedman urged such an historian to exhaust a source-critical analysis of his sources before
embarking upon historical investigation. Particularly, in the case where parallels exist for a
story such as in the two talmudim—which is indeed the case we shall be looking at here—he
maintains that it is imperative to determine the relationship between them before any his-
torical conclusions might be sought. Friedman did not offer his judgment on the historical
reliability of the sources—even according to the primary version (the “literary kernel”), as he
says: “and each historian will decide according to his methodology.” His conclusion was that
“while it is indeed hard to prove with respect to any aggadic datum that it reflects a historical
fact, even if such a conclusion is very reasonable . . . [B]ut one can note the opposite, that is,
a considerable likelihood that there is no historical basis for information that is only to be
found in the expanded parts of the Bavli, and especially when it is possible to trace the liter-
ary factor for that expansion.” Therefore, Friedman advised the historian: “Before you seek
out the historical kernel . . . you should seek out the literary kernel and base your historical
study on it” (S. Friedman, “La-Agada ha-historit ba-Talmud ha-bavli,” in: S. Friedman [ed.],
Saul Lieberman Memorial Volume [New York 1993] 122 [Hebrew]—emphasis in the original).
Y. Fraenkel (e.g. “Response,” Tarbiz 49 [1980] 429 [Hebrew]), on the other hand, had argued
for the autonomy of the rabbinic stories as a genre and for their imperviousness to valu-
able historical study. On an evaluation of the challenge presented by Fraenkel to the histori-
cal use of rabbinic sources see now H. I. Newman, “Closing the Circle: Yonah Fraenkel, the
Talmudic Story, and Rabbinic History,” in: M. Kraus (ed.), How Should Rabbinic Literature be
Read in the Modern World (Piscataway NJ 2006) 105–135. [And see also the introduction to
this volume—Eds.]
3 The only scholar who does not relate to these two versions as parallels of the same event is
J. Neusner, “Arda and Arta and Pyly Bryš,” Jewish Quarterly Review 53 (1962/3) 298–305.
the metamorphosis of a polemical amoraic story 119

2 Sources

yQid 3:4, 64a4

6‫ איתאמר‬.‫ לחבריא תמן‬5‫ ר׳ דוסתי ביר׳ ינאי ור׳ יוסי בן כיפר נחתון לגביה‬.‫דלמא‬
‫ אמרין‬10.‫ מיפקא מינהון‬9‫ אתון בעון‬8.‫ הוון בעיין דלא יהיויין כלום‬.7‫עליהון לישנא בישא‬
:‫ אמרין לון‬.13‫ מנן בעי תקמינון טבאות‬:‫ אמרין לון‬12.‫ ומנן‬:]‫ אמרן לון‬.11‫ [כבר זבנון‬:‫לון‬
‫ נסבון‬16.‫ אהן הוא כולא‬:‫ אמ׳ לון‬.‫ לגבי ר׳ דוסתי ביר׳ ינאי‬15‫ אזלון‬.‫ חינם אנחנו‬14‫שומרי‬
:‫ אמ׳ ליה‬.‫ אתא לגבי אבוי‬18,‫ כד סלקון להכא‬.‫ ואפקון מיניה‬17‫לר׳ יוסי בן כיפר וכפתון‬
‫ עימי‬20‫ אילו אשוי‬:‫ מה עבד לך? אמ׳ ליה‬:‫ אמ׳ ליה‬.]‫ מה עבד(ת כן) [לי ברך‬19]‫[חמי‬
22‫ ראיתי אותן בית דין שוה‬:‫ מה עבדת כן? אמ׳ ליה‬:‫ אמ׳ ליה‬.‫ מינן כלום‬21‫לא הוון מפקה‬

4 According to MS Leiden. There is a parallel in yGit 1:5, 43d. I shall present the variants from
Gittin below in the notes on the basis of MS Leiden.
5 ‫מיגבי ]לגביה‬.
6 ‫ואיתאמרת ]איתאמר‬.
7 ‫לשן ביש ]לישנא בישא‬.
8 ‫ ]הוון בעיין דלא יהיויין כלום‬missing.
9 ‫בעיי ]בעון‬.
10 ‫מיניה ]מינהון‬.
11 ‫זכינן ]זבנון‬.
12 ‫ ]אמרן לון ומנן‬missing.
13 ‫]אנן בעי תקימינון טבאת] ]מנן בעי תקמינון טבאות‬.
14 ‫שומר ]שומרי‬.
15 ‫אתון ]אזלון‬.
16 ‫ההנו כולה ]אהן הוא כולא‬.
17 ‫ופ(?ר?כ)[טר]וי ]וכפתון‬.
18 ‫[כד סלקון להכא] ]כד סלקון להכא‬.
19 [‫ לית את חמי ]חמי‬.
20 ‫אשוויי ]אשוי‬.
21 ‫מפקין ]מפקה‬.
22 On ‫שוה‬, see S. Lieberman, Greek in Jewish Palestine (New York 1965) 176–7. He holds that
we are dealing with a law court of one opinion. Cf. J. Neusner, A History of the Jews in
Babylonia II (Leiden 1965) 303. On the other hand, E. S. Rosenthal, “For the Talmudic
Dictionary—Talmudica Iranica,” Irano-Judaica 1 (1982) 86, understands it in the sense
of “worthy,” ἄξιος, dingus, as in the phrase ‫( בית דין יפה‬bSanh 32b). He discusses there
the change between ‫ יפה‬and ‫שוה‬. Accordingly ‫ בית דין שוה‬is not a derogatory designa-
tion (cf. bSanh 32b ‫)צדק צדק תרדוף‬. It is, however, then necessary to explain why such a
Palestinian source would praise a Babylonian court! It may be that the Bavli editor did not
understand this Palestinian expression and simply ignored it, but it seems more probable
that the Babylonian redactor was not interested in attaching to the exilarchate the quality
of a law court.
120 HERMAN

24:‫ ואמרתי‬.‫ ויוסה אחי כפות ורצועה עולה ויורדת‬23‫וכובעיהן אמה ומדברין מחציין‬
.‫שמא דוסתי אחר יש לאבא‬

A story: Rabbi Dosethai bar Rabbi Yannai and Rabbi Yose ben Kipar went
down there [i.e. to Babylonia] to collect [money] for the sages. Ill [inten-
tions] was said of them. They [the Babylonians] sought [now] not to give
to them anything.25 They came and sought to take back from them [what
they had already given]. They said to them: We have already acquired
possession [of the money].26 They said to them: We want you to under-
take liability [for any loss]. They replied: We are in the status of unpaid
guardians [who have no legal liability for loss]. They went over to Rabbi
Dosethai bar Rabbi Yannai. He said to them: Here is the lot. They took
Rabbi Yose ben Kipar and bound him and [forcibly] took [the money]
from him.
When they went up here [i.e. to Palestine], [Rabbi Yose ben Kipar] came
before [Dosethai’s] father. He said to him: See what your son did to me!
He asked him: What did he do to you? He answered: Had he agreed with
me they would not have taken anything from us. [Dosethai’s father] said
to him: Why did you act in this way? [Dosethai] replied: I saw them; a
proper court27 and their hats a cubit high, and they were speaking from
their middle, and Yose, my brother, bound, and the whip rising and com-
ing down, and I asked [myself]: Does my father have another Dosethai?

bGit 14a–b28

‫ אמר להו לר׳ דוסתאי בר׳‬.‫ר׳ אחי בר׳ יאשיה הוה ליה איסקפא דכספא בנהרדעא‬
‫ אמרי‬.‫ אזול יהביה ניהליה‬29.‫ינאי ולר׳ יוסי בר כיפר בהדי דאתיתו אייתוה ניהליח‬
‫ ר׳ דוסתאי ברבי ינאי‬.‫ אהדריה ניהלן‬:‫ אמרי להו‬.‫ לא‬:‫ אמרי להו‬.‫ נקני מינייכו‬:‫להו‬

23 ‫מחציים ]מחציין‬.
24 ‫ואמרית ]ואמרתי‬.
25 This phrase, missing in the Gittin parallel, makes little sense and seems to be corrupt.
26 I have not translated the words ‫אמרן לון ומנן‬. They are missing in the Gittin parallel and
seem to be a corruption.
27 On this expression see above, n. 22.
28 The source is presented here according to the editio princeps with limited changes on the
basis of the manuscripts as noted. Significant manuscript variants are cited in the notes.
29 ‫]אייתוה ניהליה‬. Thus according to a number of manuscripts. In the printed edition: ‫אתיוה‬
‫ניהלי‬.
the metamorphosis of a polemical amoraic story 121

:‫ א״ל‬30.‫ הוו קא מצערי ליה‬.‫ כפתוה‬.‫ לא‬:‫ ר׳ יוסי בר כיפר אמר להו‬.‫ אין‬:‫אמר להו‬
‫ חזי מר לא‬:‫ א״ל‬,‫ כי אתו לגביה‬31.‫ טב רמו ליה‬:‫חזי מר היכי קא עביד? אמר להו‬
‫ הכי? א״ל‬33‫ אמאי עבדת‬:‫ א״ל‬.‫ טב רמו ליה‬32:‫ אלא אמר להו‬,‫ דלא סייען‬,‫מיסתייה‬
‫ ארדא‬,‫ ושמותיהן מבוהלין‬,‫ ומדברין מחצייהן‬,‫ וכובען אמה‬,‫ הן אמה‬,‫אותן בני אדם‬
‫ אילו הרגו את‬.‫ הורגין‬,‫ הרוגו‬:‫ אומרין‬.‫ כופתין‬,‫ כפותו‬:‫ אומרין‬.‫וארטא ופילי בריש‬
?‫ בני אדם הללו קרובים למלכות הן‬:‫ כמותי? א״ל‬34‫ מי נתן לינאי אבא בן‬,‫דוסתאי‬
.‫ אי הכי שפיר עבדת‬.‫ אין‬:‫ יש להן סוסים ופרדים שרצים אחריהן? א״ל‬35.‫ אין‬:‫א״ל‬

Rabbi Ahai bar Rabbi Josiah had a silver goblet in Nehardea. He said to
Rabbi Dosethai bar Rabbi Yannai and to Rabbi Yose bar Kipar: When
you come [there] bring it [back to me]. They went and retrieved it. [The
Nehardeans] said to them: Make [legal] acquisition from us. They replied
to them: No. [The Nehardeans] said to them: [Then] return it to us! Rabbi
Dosethai bar Rabbi Yannai said to them: Yes. Rabbi Yose bar Kipar said to
them: No. They bound him [Rabbi Yose bar Kipar]. They were torment-
ing him, saying: Does Sir observe how things are done [here]? [Rabbi
Dosethai bar Rabbi Yannai] said to them: Beat him well! Beat him well!
When they came to him [Rabbi Ahai bar Rabbi Josiah, Rabbi Yose bar
Kipar] said: Does Sir note that not only did [Rabbi Dosethai] not help me,
but he said “beat him well.” He said: Why did you act in this way? [Rabbi
Dosethai bar Rabbi Yannai] said: Those people are a cubit (high) and their
hats a cubit (high), and they speak from the middle, and their names are
outlandish, Arda, Arta, and PYLYBRYŠ. [If] they say: bind [him]! They
[surely] bind [him]; [would] they say: kill! They would [surely] kill! Had
they killed Dosethai, who would give Yannai my father a[nother] son like
me?
[Rabbi Ahai] said to him: Are those people close to the sovereign
powers? He said: Yes. Do they possess horses and mules that run after
them? He replied: Yes. [Rabbi Ahai then said:] If so, you acted well.

30 ‫]כתפוה הוו קא מצערי ליה‬. Corrected on the basis of the manuscripts, but in the printed
edition: ‫הוו קא מצערו ליה‬.
31 ‫]טב רבו ליה‬. This statement is repeated in a number of manuscripts.
32 ‫]אמר להו‬. Thus according to all the manuscripts. The printed edition adds: ‫נמי‬.
33 ‫]עבדת‬. This according to most of the manuscripts, in the printed edition: ‫תיעבד‬.
34 ‫]בן‬. Thus in the manuscripts. In the printed edition: ‫בר‬.
35 ‫]אין‬. Thus in the manuscripts. In the printed edition: ‫הן‬.
122 HERMAN

In this story, according to the version presented in the Yerushalmi, two sages
are sent to Babylonia to collect debts for the Palestinian sages.36 They get into
trouble and are required to surrender the money that they have received. After
a brief exchange of a halakhic nature37 one of the sages declares his refusal to
cooperate with the locals and is conse­quently beaten. The second rabbi coop-
erates and remains unharmed. Upon their return to Palestine they speak about
this Babylonian experience. The second sage defends his action whilst describ-
ing the intimidating court judges and his fear of personal injury.
In the Bavli, however, the background to the story is somewhat different.
The same two sages are sent to retrieve a silver goblet belonging to a third
sage, Rabbi Ahai bar Rabbi Josiah, a local38 rabbi, whilst they are passing
through the city of Nehardea. Once collected, the locals demand the return of
the object. After their initial joint refusal (the halakhic discussion is different
from the Yerushalmi version and briefer) one of the two consents to hand over
the object while the other, who refuses, is beaten by the locals, cheered on by
his colleague. Upon their return to Rabbi Ahai bar Rabbi Josiah, the one sage
complains about his colleague’s behavior. The second sage defends himself
by describing the intimidating nature of these people. Rabbi Ahai bar Rabbi
Josiah’s inquiries lead him to the conclusion they are “close to the sovereign
powers.”
Scholars have been drawn to this story for its vivid portrayal of the
Babylonians.39 A Babylonian law court with its apparent capacity to carry

36 Gedaliah Alon wished to find here evidence of a delegation intended to collect contribu-
tions for the support of the rabbis in Palestine. However the term used, ‫למגבי‬, relates to
the collection of debts and not donations. Likewise, the Talmudic context and the conver-
sation that ensues between the rabbis and the Babylonians seems better suited to the col-
lection of debts than donations. On the collection for the sages in Palestine see G. Alon,
The Jews in their Land in the Talmudic Age vol. 1 (Jerusalem 1980) 248–52; M. Beer, “Torah
and Derekh Eretz,” Bar-Ilan 2 (1964) 148–51.
37 This constitutes the legal context for the appearance of the story in this locus within the
Yerushalmi.
38 See bShab 152b; bQid 72a. He seems to have originated in Palestine, see A. Hyman Toldoth
Tannaim ve-Amoraim vol. 1 (Jerusalem 1964) 136 [Hebrew]. Jacob Neusner (A History of
the Jews in Babylonia, vol. 1 [Leiden 1965] 136) thought that he dwelled in Palestine at the
time when the story is set.
39 This story connects both to the question of the collection of monies for the rabbis and
the issue of the jurisdiction over capital cases in Babylonia in the Parthian era, and more
generally the judicial system in the Arsacid era and its possible relationship to the exi-
larchate. Because of these many scholars have referred to this source. Earlier studies
the metamorphosis of a polemical amoraic story 123

out both corporal and capital punishment is particularly striking. The dress
and personal names of the Jews imply a remarkably acculturated, Persianized
Jewry. Mention of horses and mules suggests that some Jews were part of the
elite, perhaps filling a role in the military makeup of the region. Indeed ele-
phants, typically associated with the army, might also be hinted at, as we shall
see below. More generally, the allusion to Babylonian Jews in close contact
with the kingdom suggests a striking degree of confidence and autonomy.
The Babylonian villains of the story have usually been identified with the
exilarchate. Set in the second century CE, this story has served as a potent
source for the early history of the exilarchate and for Babylonian Jewry as a
whole.40 Taken at face value it might suggest that this exilarchal authority
enjoyed at this point in time a status unparalleled in the course of the talmu-
dic era, perhaps with a military role and a mandate for capital punishment.41
Some have therefore imagined the exilarchate as attaining its zenith precisely
in the Arsacid era.

include (a partial list): F. Lazarus, Die Häupter der Vertriebenen (Frankfurt a. Main 1890)
66, n. 5; N. Brüll, “Zur Geschichte der babylonischer Exilarchen,” Central-Anzeiger für
Jüdische Litteratur (Frankfurt a. M. 1891) 91; Alon, The Jews in their Land, 1, 249; M. Beer The
Babylonian Exilarchate (Tel-Aviv 19762) 45, 58–60; Neusner, History of the Jews in Babylonia
1, 94–7, 100–3; Rosenthal, “Talmudica Iranica,” 86–7; I. Gafni, The Jews in Babylonia in the
Talmudic Era: A Social and Cultural History (Jerusalem 1990) 97 [Hebrew]; D. Goodblatt,
The Monarchic Principle (Tübingen 1994) 140, n. 43; A. Oppenheimer, “Enforcement in
Palestine and in Babylonia in the Late Tannaitic Era,” in: D. Gera and M. Ben-Zeev (eds.),
The Paths of Peace: Studies in Honor of Israel Friedman Ben-Shalom (Beer-Sheva 2005)
366–70.
40 On this source as evidence for the success of the Jews to integrate into the Parthian nobil-
ity, and likewise, on the weakness of rabbinic authority in Babylonia in the second cen-
tury, and generally for the Parthian era see Neusner, History of the Jews in Babylonia, 1,
94–7, 100.
41 It has been widely assumed that in the Parthian era the Jews were empowered to try capi-
tal cases, whether by official permission or as a result of a general anarchical situation that
is attributed to this period. See, for instance, J. Mann, “Seqira historit ‘al dinei nefashot,”
Ha-Sofe le-Hokhmat Yisrael 10 (1925) 201–3 [Hebrew]; and recently, Oppenheimer,
“Enforcement,” 366–70. This source and the story of the execution of an informer by Rav
Kahana are among the limited pieces of evidence for this view. On the problems with
using the Rav Kahana story as evidence for the Parthian era see G. Herman, “The Story of
Rav Kahana (BT Baba Qamma 117a–b) in Light of Armeno-Persian Sources,” Irano-Judaica
6 (2008) 53–86, with references to earlier scholarship.
124 HERMAN

3 Yerushalmi and Bavli: A Comparison

While there is considerable divergence between the two versions of the story,
it has been common in the past to treat them as synoptic and complimentary.42
One account is supplemented with details from the other as if together they
can contribute to the recovery of an hypothesized Urtext.43 Language has also
played a role in this assessment. Of the two languages used in both versions
of this story, the Hebrew has been deemed as preserving an earlier stratum of
the story than the Aramaic.44 Analysis can show, however, that the function

42 Needless to say, this was the practice of traditional commentators. An early example is
the 18th-century German commentator, David Hirschel Fränkel, who, in his commentary
on the Yerushalmi, Qorban ha-‘Eda, explains that the two sages came “to collect money
for the rabbis, that they had there as a deposit, and the depositors delivered to them the
money.” Similarly, the 19th-century Moses Margoliot wrote in his commentary on the
Yerushalmi, Pnei Moshe, that “they went down to collect a debt for their fellow rabbi.” See
also Rosenthal, “Talmudica Iranica,” and further below.
43 Rosenthal (“Talmudica Iranica,” 87), for instance, surmises that we have an ancient source
upon which the talmudim have based themselves. As such, ‫“ בית דין שוה‬is completely
missing in the Bavli,” and on the other hand, “the Babylonian tradition teaches us another
detail . . . that was lost to the Yerushalmi . . .,” and he draws his data from both versions.
The differences between the two versions have been interpreted in diverse ways by schol-
ars. Neusner pointed out that the Babylonian version reveals a greater local familiarity
than the Yerushalmi version. The assumption being that this version preserved details
that the Yerushalmi version had generalized, or forgotten. It has also been argued that
since the story takes place in both Palestine and Babylonia the usual presuppositions
about attributing greater credibility to the version that is geographically the closest, is not
tenable in this case.
44 Cf. Rosenthal’s judgment (“Talmudica Iranica,” 87) that Rabbi Dosethai’s Hebrew com-
ments “are the self-same formulated kernel of an ancient and reliable tradition around
which the story of the event has been fashioned, whilst transmitted in a popular and free
style.” It is necessary to probe the function of the choice of language, whether Hebrew
or Aramaic, in the talmudic story. It is evident, at any rate, that the notion that Aramaic
reflects a popularity or vulgarity is unsatisfactory. It seems that the rules for the distinc-
tion between Hebrew and Aramaic that have been applied to the distinction between
the amoraic meimra and the anonymous stratum (see S. Friedman, “Pereq ha-Isha
Rabba ba-Bavli,” in: H. Z. Dimitrovsky [ed.], Texts and Studies vol. 1 [New York 1978] 301–2
[Hebrew]) are not equally applicable to such stories. For a discussion of the function of
language, albeit in the legal strata of the Bavli, see J. Neusner, “Language as Taxonomy:
The Rules for Using Hebrew and Aramaic in the Babylonian Talmud,” in: Z. J. Kapera
(ed.), Intertestamental Essays in Honour of J. T. Milik (Krakow 1992) 327–42. Cf., also
A. Baumgarten, “Rabbi Judah I and his Opponents,” Journal for the Study of Judaism 12
the metamorphosis of a polemical amoraic story 125

of the two languages within the story is not consistent with this assumption.
In both versions Aramaic is dominant and has the role of the language of the
narrator. It is clear that we cannot be dealing with an early (tannaitic) kernel
of the story in Hebrew, upon which one finds Aramaic glosses and accretions,
but rather the parts of the story that appear in Aramaic are integral to the story
as a whole, and apparently belong there from the start. The function of the two
languages within the story must be explained differently and does not support
the above-mentioned theory of an Urtext.45
More critically, it will become clear that we are dealing with two quite sepa-
rate stories, and that the relationship between them must be viewed as linear,
with the Bavli as derivative.46 Each version, in fact, can be seen as a separate
piece pursuing distinct objectives.
The Yerushalmi is an anti-Babylonian story. Its Palestinian perspective
has full articulation. It relates a journey from Palestine to a foreign land of

)1981( 140; A. Bendavid, Biblical Hebrew and Mishnaic Hebrew (Tel Aviv 1967) 134–5
[Hebrew]; E. Margolies, “Ivrit ve-aramit betalmud uva-midrash,” Leshonenu 27 (1963–4)
20–33 [Hebrew].
45 One discerns that in the Yerushalmi, in the conversation between the sages and the
Babylonians, the sages tend to speak Hebrew, whereas the Babylonians address them in
Aramaic. This may be deliberate, perhaps since they claim superior halakhic authority,
although even here there is no absolute consistency, and at the end of the story the rabbis
speak both Hebrew and Aramaic with Rabbi Yannai. In the Bavli only the final conver-
sation between Rabbi Ahai bar Rabbi Josiah and Rabbi Dosethai is (almost entirely) in
Hebrew.
46 Even if it was not necessarily the selfsame Yerushalmi text that lay before the redactors of
the Bavli. It is generally acknowledged that the Yerushalmi in its present completed state
was not available to the redactors of the Bavli. See J. N. Epstein, Introduction to Amoraic
Literature (Jerusalem 1962) 290. Fraenkel (“Response,” 429) had already advised: “If the
story has parallels, one must listen attentively to what one narrator leaves out and does
not bring in his story. An interpretation of combining [italics in original] parallels assumes
that there was one (historical) fact that must be reconstructed through the combination.”
Shamma Friedman has argued, on the other hand, that generally speaking, and in prin-
ciple, “one must always give preference to the Palestinian parallel” (Friedman, “La-aggada
ha-historit,” 121). This is in view of the much earlier date of redaction of the Palestinian
material as a whole and the numerous attested examples of such a relationship. Cf. too,
Rubenstein, Talmudic Stories, 311, nn. 97–8. And this is in opposition to a vintage scholarly
approach. See, for instance, M. D. Judelowitz, Sefer Nehardea (Vilna 1906) 84 [Hebrew]:
.‫ עלינו להשען ביתר שאת על הבבלי מאשר על הירושלמי״‬,‫ כמובן‬,‫״והמעשי׳ אשר קרו בבבל‬.
S. D. Galante, “De-vei Resh Galuta,” Ha-Osem (St. Petersburg 1898) 36: ‫״ואף שנוכל לאמר‬
‫״‬. . . ‫ עדות הבבלי מאומתת יותר מן הירושלמי‬,‫כדברי המהר״ץ חיות דמאורע הנעשית בבבל‬
126 HERMAN

two distinguished Palestinian rabbis. When false accusations undermine their

mission, they apply acceptable Jewish legal principles but are countered by
intimidation. A feeling of foreignness pervades all that relates to Babylonia.
That country is seen as a dangerous place, the law there is arbitrary, and the
external appearance and manner of the people is distinctly odd.
The Bavli indeed shares some of these details. It, too, features the same cou-
ple of sages from Palestine, but nothing is said of a journey from Palestine—
they are already in Babylonia from the beginning. Their Palestinian origin
functions differently, providing naive characters unfamiliar with the local
scene. They, too, are roughly treated—here by the Nehardeans—though they
do not return to Palestine but to Rabbi Ahai who sent them. He reveals his own
familiarity with the villains. In this version, then, the focus is not Babylonia but
Nehardea.
The differences between the two versions cannot be in the way of minor
details that are added or dropped in the course of transmission. Furthermore,
there is an uneven relationship between the two versions. The Bavli is lengthier
and contains significant additions to the Yerushalmi version that advance its
objective. They reflect its conscious elaboration, development, and dramatiza-
tion of the Yerushalmi version. By deleting and adding details it sets the story
on a new path. We shall outline all this presently.

3.1 The Bavli’s Story—Inner-Babylonian Polemic

The Bavli is telling us a different story. It does not rewrite the story but
reworks the existing account with overt and subtle changes. It dramatizes
the plight of the bound sage by including the concrete threat of capital pun-
ishment which also recurs in the words of Dosethai at the end of the story:
“had they killed Dosethai.” The Yerushalmi had been far less explicit in say-
ing “Does Father have another Dosethai?” The dramatization comes to play in
the response of the second sage. In the Yerushalmi Rabbi Yose is harmed but
no harm is done to Rabbi Dosethai. The Bavli, however, has Rabbi Dosethai
actively encourage the villains.
The halakhic context of the story has been diluted in the Bavli. Throughout
the Yerushalmi the halakhic question remains central and halakhic terminol-
ogy is employed. The Babylonians are a “proper court,” and the complaint by
Rabbi Yose at the end, too, is based on the assumption that the bet-din would
not have been empowered to decide against them had the second sage not dif-
fered from his colleague. The Bavli, on the other hand, seems little interested
in legalities. It portrays the meeting between the emissaries and the locals as
one of intimidation and little more. There are no halakhic explanations and
after Rabbi Dosethai is portrayed as encouraging the Nehardeans while they
the metamorphosis of a polemical amoraic story 127

torment Rabbi Yose, this action becomes the focus of Rabbi Yose’s complaint
against him to Rabbi Ahai bar Rabbi Josiah.47
The changes introduced by the Bavli function both as a response to the
Yerushalmi, and as an opportunity to pursue its own agenda. It counters the bla-
tant broadly anti-Babylonian tone of the story in its Yerushalmi caste whereby
a Palestinian embassy sets off to the Diaspora to collect debts for its sages in
Palestine. Uncomfortable with the implicit hierarchical assumptions of this
scenario, whereby the Babylonian rabbis are subjected to their Palestinian
counterparts, the background has been reformulated. The purpose of the visit
becomes a private matter for a local resident. Here, the Bavli takes Rabbi Ahai
bar Rabbi Josiah, a well-known figure connected elsewhere to Babylonia.48 It
utilizes the style and structure of the adjacent legal case cited immediately
beforehand in the Talmud to redraft its opening. This refers to money that Rav
Sheshet has in Mehoza. He asks Rav Joseph bar Hama to collect it when he
next passes through. The story in the Bavli is localized. It takes place in the city
of Nehardea.49 The dialogue appended to the end steadily draws the audience
closer to the true target of the story, who are not just Nehardeans, but those
bearing such foreign names as Arda and Arta and with equestrian habits. The
Bavli is branding specific people within the Babylonian Jewish community.
The Bavli has converted the story into a polemic against this Nehardean
group. It marks them culturally with distinctive names and other details link-
ing them to the powers that be, distancing them from the rabbis and their legal
system. Furthermore, by means of certain intertextual allusions it conveys
judgment on this group and its identity.

3.2 Intertextuality in the Bavli’s Story

To the Yerushalmi’s description of the menacing Babylonians the Bavli has
added that “they are a cubit [high]”!50 However, there is nothing necessarily

47 It is not so rare in cases where a Bavli parallel exists for a Palestinian source for the Bavli
to zoom in on a detail that is relatively marginal in the Palestinian text. Cf. the story on
Shmuel and Qarna: yBB 5:5, 15a–b; bBB 89a.
48 See above, n. 38.
49 Neusner observed that the Babylonian version contains inner and deeper acquaintance
with what goes on in Babylonia. On the competition between the cities see I. Gafni,
“Expressions and Types of ‘Local Patriotism’ among the Jews of Sasanian Babylonia,”
Irano-Judaica 2 (1990) 63–71; Neusner, Jews of Babylonia, 119–25.
50 The Yerushalmi version in Gittin, found in the Krotoschin edition has been corrupted here
on the basis of the Bavli.
128 HERMAN

intimidating about people one cubit high.51 The expression would appear to
have been borrowed from bMQ 18a:

Pharaoh in the days of Moses was one cubit [high], and his beard was one
cubit [long], and his PRMŠTQ52 a cubit and a (finger) span [in length],
to fulfill what was said: “that he may set over it even the lowest of men”
(Dan 4:14).

The source continues “Pharaoh in the days of Moses was a Magus.”53 This
description was said about non-Jews, after the verse from Dan 4:14. The rab-
bis saw in this verse the ruler Nebuchadnezzar, and employed the Magi for
an up-to-date example. Our source, however, deals with Jews! In the light of
the adoption of this physical trait, with its exegetical burden, we may pon-
der for a moment over whether the Bavli has in mind describing these people
as non-Jews, further supported perhaps by their Persian names, thus totally
removing any stain from Babylonian Jewry. However, it seems to me that they
did not go in that direction. The phrase “they are a cubit [high]” was added
associatively due to its common ring with the previous “their hats were a cubit
[high].” Nevertheless, the Bavli editor wished to emphasize their foreignness to
ordinary Jews by their Persian names and their life-style—hence allusion to a
scriptural verse with gentile associations for those people is not totally out of
character.
This appears to add ridicule to the portrayal of the villains. Whatever the
precise cultural sense of the expressions already found in the Yerushalmi, “their

51 Neusner paraphrases this sentence: “the hats of those people are as long as themselves.”
See also Rosenthal, “Talmudica Iranica,” 87. Cf., too, A. Goldberg, relating to the descrip-
tion of Pharaoh in the days of Moses in bMQ 18a (“The Babylonian Talmud,” in: S. Safrai
[ed.], The Literature of the Sages, First Part [Philadelphia 1987] 336): “what we have here is
a description of the fighting appearance of the Parthians or Persians of the time.”
52 On this word see B. Geiger, Additamenta ad librum Aruch Completum (Wien 1937), s.v.
‫פרמשתק‬, 340–1, who sums up the earlier efforts to interpret this word by saying that “all
the explanations that have hitherto been proposed for this word are either unlikely or
impossible.” He concludes that the word seems to be Persian but its origins are unknown.
James R. Russell, however, in a lecture presented before the Association Internationales des
Etudes Arméniennes in Budapest, October, 2011, following up on an etymology proposed
by Martin Schwartz, identified the word as behind an Iranian loan in Armenian, with the
sense of membrum virile. RNL Evr I 187 has actually included the phrase ‫ופרמשתרקן אמה‬
between the margins in our story. Note its alternative spelling.
53 There is also a rabbinic tradition that Nebuchadnezzar was a midget. See GenR 16:4, and
there many references.
the metamorphosis of a polemical amoraic story 129

hats are a cubit in height, and they speak from their middle” (and whether
they were understood by the Bavli or not), the Bavli has converted the frightful
court into a circus populated by dwarfs.
Apart from intertextually evoking the image of Pharaoh as he is described
elsewhere in the Bavli, the foreign personal names, too, contribute to this
image of hostile gentileness appended to the villains here. The strange names,
described as ‫ושמותיהן מבוהלין‬, would seem to be an echo of a similar expres-
sion, ‫שמות מובהקין‬, names that when encountered on a legal document can
be assumed to belong to gentiles. This topic is, in fact, raised only a few pages
prior to our story in Tractate Gittin, in a discourse set on the mishnah that
speaks of gentile courts. It lists seven names typically associated with gentiles.54
Here we have three additional names: Arda, Arta, and Pilibaresh.55 Ardā is a
common enough Persian name.56 Artā would merely be the archaic phonetic
form of the same. The audience surely knew this.57 Did it also recognize the
subtle irony in the choice of names? Ardā in Middle Persian means “righteous,
truthful,” while the court in which he serves is far from being that.58 The third
name ‫ פיליבריש‬is more problematic and lacks a satisfying etymology.59 If some
association with elephants (‫ )פיל‬is intended, such a connection would blend
well with the theme of proximity to the “the sovereign powers,” since elephants
were royal animals and their management a royal prerogative. This would also
then provide further support for dating the Bavli’s version in the Sasanian era
since the use of elephants by the Sasanians is historically proven. The Arsacids,

54 This statement is, however, neutralized a few lines away with the comment that in the
Diaspora Jews tend to share the same names as their gentile neighbors (bGit 11b).
55 On these names see already S. Y. L. Rapoport in Kerem Hemed 7 (1843) 198–9; idem, (1852)
193–4, 278; Lazarus, Die Häupter der Vertriebenen, 66–7, 132–5; Geiger Additamenta, s.v.
‫ארדא‬, 64; Neusner, “Arda and Arta”; idem, “How Much Iranian in Jewish Babylonia,” in
Talmudic Judaism in Sasanian Babylonia (Leiden, 1976) 144, esp. n. 4.
56 It is also attested among the graffiti on the walls of the Dura synagogue written in Pahlavi.
See B. Geiger, “The Middle-Iranian Texts,” in: A. R. Bellinger et al. (eds.), The Excavations
at Dura-Europos. Final Report VIII, part 1 (New Haven 1956) 311.
57 This dialectical playfulness also occurs in a separate story in bQid 70a.
58 It originated in the Avestan areta, aša = judgment. For the Middle-Persian see Mackenzie,
A Concise Pahlavi Dictionary (London 1971) 11.
59 Neusner’s proposal was early on rejected on philological grounds by Ilya Gershevitch,
as Neusner notes, correcting in his book what he had written in this article, Neusner,
“Arda and Arta,” 95, n. 2. Neusner was unfortunately unaware of the entries by Geiger in
Additamenta. He sought a Parthian background to the names but they need to be inter-
preted in light of the onomastic reality of the Sasanian and not the Parthian era.
130 HERMAN

on the other hand, did not employ elephants in their army.60 Despite allusions
to sources that evoke gentiles, the villains in our story are clearly identified as a
Jewish group, perhaps similar to those affiliated with the House of Bar Elyashiv,
mentioned but a few lines from our story, who when owed something, aggres-
sively “bind and take” their due.
The Jewishness of the villains, however, is further implied in another
intertextual allusion. They are identified by the expression, “close to the
sovereign powers.” The tannaitic sources use this expression in reference to
the patriarchal dynasty and we may assume that it is evoked here precisely
to draw a parallel with the Palestinian patriarch.61 It refers to the Babylonian
equivalent—the exilarch.
The creation of a new Babylonian story out of old materials has come at a
price. While the setting for the story makes less sense with two Palestinians
wandering around Babylonia, the role played by Rabbi Ahai bar Rabbi Josiah
is strange. According to the Bavli, we are supposed to understand that the two
sages return to Rabbi Ahai and the final conversation is conducted between
the three of them. With such a threesome, the humorous finale found in
the Yerushalmi, “Does Father have another Dosethai?,” is all but lost. Rabbi
Dosethai is talking to Rabbi Ahai about his father, but is not actually speaking
with his father. However, why should Rabbi Ahai care for the pain felt by Rabbi
Yannai, Rabbi Dosethai’s father?!62

60 Elephants were an important component in the Sasanian army. See Scriptores Historiae
Augustae (LCL; ed. W. Heinemann; New York 1922), Alexander Severus, 56, for elephants
in the army of Ardašīr I. On the period of Šābuhr II see Hamza al-Isfahani, Kitab tārīkh sinī
mulūk al-ard wa al anbiyā’, Berlin, 1922, 37; Sozomen, The Ecclesiastical History (Nicene
and Post-Nicene Fathers of the Christian Church, 2nd Series; Michigan 1890), 2:14. For
the latter part of the Sasanian era see Chronica Minora 19 (T. Nöldeke, “Die von Guidi
herausgegebene syrische Chronik übersetzt und commentiert,” Sitzungsberichte der
Philosophisch-historischen Classe der Kaiserlichen Akademie der Wissenschaften, b. 128, IX
[Vienna 1893] 12). See also Maçoudi, Les Prairies d’Or II (ed. De Maynard and De Courteille;
Paris 1861–77) 230; and A. Christensen, L’Iran sous les Sassanides (Copenhagen 1936) 208.
61 The expression ‫ קרובין למלכות‬appears in tSot 15:8, according to MS Wien (but in MS
Erfurt: ‫ )זקוקין‬with reference to the Gamalielian dynasty; and also there, in tAZ 3:5
(Zuckermandel edition, 463) in MS Erfurt, although there, in MS Wien, and similarly
in the editio princeps we have ‫ זקוקין‬in place of ‫קרובין‬. In yAZ 2:2, 41a, the tradition is
ascribed to ‫בית רבי‬, who are ‫קרובין למלכות‬.
62 The way the Bavli has “spoiled” a humorous ending recalls a similar case—the Yerushalmi
and Bavli traditions on the deposition of Rabban Gamaliel. The Yerushalmi concludes
the story with a declaration of submission. The rebels express their acceptance of the
rule of the patriarch by means of the salutatio ritual greeting in the following words: ‫אני‬
‫( ואתם נשכים לפתחו‬yBer 4:1, 7d). This is how the story ends. The Bavli, on the other hand,
the metamorphosis of a polemical amoraic story 131

4 Historical Conclusions

The literary analysis confirms that the Yerushalmi is the primary version of this
source and the Bavli secondary. Historical conclusions follow suit and each ver-
sion must therefore be assessed separately for its historical input. Although it
features rabbis who belong to the tannaitic era, there is little reason to date the
Yerushalmi version earlier than the period when the Yerushalmi was authored
in the 3rd–4th centuries. This Yerushalmi version does not contain any allusion
to the exilarchate. Although its assessment of Babylonia is somewhat hostile it
still recognizes a Jewish legal court system there.
The Bavli version, on the other hand, has an agenda of its own, and pur-
sues it with vigor, diverting the Yerushalmi account to defend the integrity of
its Jewish population as a whole, and at the same time settling a local score.
Derivative from the Yerushalmi, it is certainly no source for the Arsacid era but
it enriches our knowledge of Babylonian Jewry in the later period.
The Bavli’s story targets the exilarchate. It has, in fact, created a caricature
of the exilarchate. With its choice of names and reference to horses the exi-
larchs’ Persianized features have been inflated.63 The (exilarchic) authority’s
harsh treatment of the visiting rabbis is familiar from similar conduct endured
by local rabbis. We read elsewhere in the Bavli, for instance, how Rav Amram
Hasida was tormented, Rav Gada and Eleazar Zeira were imprisoned, and Rav
Zevid was poisoned with a chalice of vinegar.64 Our story has been denuded of
its legal context and demeans the legal authority claimed by the exilarchate.
This is achieved by removing both the term “a proper court” and the assertion
appearing towards the end that “had he agreed with me they would not have

after stating ‫למחר אני ואתה נשכים לפתחו‬, continues with a short discussion. It is appar-
ent that the Bavli did not regard the statement as signifying submission, since it offers
some options, including the division of the patriarchal functions in various ways. The
Bavli apparently did not recognize this as a technical term for such submission, perhaps
it being particular to the Roman cultural sphere, and interpreted it merely as the rabbis
agreeing to inform the deposed patriarch of their decision as soon as possible. [On this
episode see also the Introduction to this volume—Eds.]
63 On the degree of integration of Babylonian Jewry into Iranian culture, and especially the
Persian language, see Gafni, The Jews of Babylonia, 156–61; Neusner, “How much Iranian”;
Y. Elman, “Middle Persian Culture and Babylonian Sages: Accommodation and Resistance
in the Shaping of Rabbinic Legal Tradition,” in: C. E. Fonrobert and M. S. Jaffee (eds.), The
Cambridge Companion to the Talmud and Rabbinic Literature (Cambridge 2007) 165–97;
G. Herman, “The Persians Are More Expert than You in Banquet Protocol: Table Etiquette
and Persian Culture in the Babylonian Talmud,” Zion 77 (2012) 149–88 [Hebrew].
64 E.g. bGit 67b; bEruv 11b; bAZ 38b.
132 HERMAN

taken anything from us.” This anti-exilarchal polemic reveals to us how the exi-
larchate was perceived by its Babylonian rabbinic discontents. The geographi-
cal factor here is noteworthy. Nehardea is identified with the exilarchate.65 The
authors of this Babylonian version are not only opponents of the exilarchate
but presumably stem from a different region of Babylonia.66
The conclusions arrived at in this case lead to a historical narrative quite dif-
ferent from much of the earlier historical writing on Babylonia. The historian,
following a literary analysis of these parallels, is left with a more precise under-
standing of the sources. The story is now located in its appropriate setting and
we are better informed of current international and local perceptions. In this
case, where parallel versions stem from Babylonia and Palestine the separa-
tion of the stories has been particularly valuable in identifying Babylonian
concerns. We have observed how this story, upon its journey from Palestine
to Babylonia, has been radically transformed. From a simple tale of “high-
hatting” Babylonians it has come to speak of midgets and mules, of elephants
and of exilarchs.

65 Other sources that connect the exilarch with Nehardea: bBQ 59a–b and bQid 70a.
66 More generally on the exilarchate see, now, G. Herman, A Prince without a Kingdom: The
Exilarch in the Sasanian Era (Tübingen 2012).
Rescue from Transgression through Death;
Rescue from Death through Transgression

Christiane Tzuberi

A short text, part of the gemara on mSanh 8:7 (cf. bSanh 74a–b), juxtaposes two
different approaches to a transgression committed in the face of death: accord-
ing to the one, a transgression is legitimate if it rescues one’s life; according to
the other, death “saves” an individual from committing a transgression. These
two approaches are attributed by the Babylonian stam to particular places of
origin: those advocating rescue from a transgression through death are rab-
bis active in the Land of Israel; those advocating rescue from death through a
transgression are Babylonians.

1 The Sages of Lod and a Question on Esther

Let me begin with the text:

‫ כל‬:‫ נימנו וגמרו בעליית בית נתזה בלוד‬:‫אמר רבי יוחנן משום רבי שמעון בן יהוצדק‬
‫ חוץ מעבודת‬,‫ יעבור ואל יהרג‬,‫ עבור ואל תהרג‬:‫ אם אומרין לאדם‬,‫עבירות שבתורה‬
.)‫] (ב׳ סנהדרין עד ע״א‬. . .[ .‫ ושפיכות דמים‬,‫ וגילוי עריות‬,‫כוכבים‬
‫ אבל בשעת‬.‫ לא שנו אלא שלא בשעת גזרת המלכות‬:‫כי אתא רב דימי אמר רבי יוחנן‬
‫ אפילו‬:‫ כי אתא רבין אמר רבי יוחנן‬.‫ יהרג ואל יעבור‬,‫ אפילו מצוה קלה‬,‫גזרת המלכות‬
,‫ אפילו מצוה קלה‬,‫ אבל בפרהסיא‬,‫ לא אמרו אלא בצינעא‬,‫שלא בשעת גזרת מלכות‬
?‫ והא אסתר פרהסיא הואי‬.‫יהרג ואל יעבור‬
‫ דאי לא תימא‬,‫ הנאת עצמן שאני‬:‫ רבא אמר‬.‫ אסתר קרקע עולם היתה‬:‫אמר אביי‬
‫ הנאת‬,‫ הכא נמי‬.‫ הני קוואקי ודימונקי היכי יהבינן להו? אלא הנאת עצמן שאני‬,‫הכי‬
.(‫ב‬-‫עצמן שאני )ב׳ סנהדרין עד ע״א‬

Rabbi Yohanan said in the name of Rabbi Shimeon ben Yehozadak: By a

majority vote it was resolved in the upper chambers of the house of Nitza
in Lod that in every law of the Torah, if a man is commanded: “Transgress
and suffer not death” he may transgress and not suffer death, excepting
idolatry, forbidden sexual relations and bloodshed (bSanh 74a).1 [. . .]

1 For a parallel in the Yerushalmi cf. Shevi 4:2, 35a–b and ySanh 3:6, 21b. Cf. A. Gray, “A
Contribution to the Study of Martyrdom and Identity in the Palestinian Talmud,” Journal of
Jewish Studies 54 (2003) 243–56.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�05

134 tzuberi

When Rav Dimi came [from the Land of Israel] he said in Rabbi Yohanan’s
name: [The decision of the sages of Lod] was taught only if there is no
royal decree (of persecution), but if there is a royal decree, one must be
killed rather than transgress even a minor precept. When Rabin came
[from the Land of Israel], he said in Rabbi Yohanan’s name: Even without
a royal decree it was only permitted in private; but in public one must be
killed rather than transgress even a minor precept [. . .].
But was not Esther’s case in public?
Abbayye said: Esther was natural soil.2 Rava said: Their personal pleasure
is different, for otherwise, how dare we yield to them our braziers and
coal shovels?3 But their personal pleasure is different; so too [in Esther’s
case] (bSanh 74a–b).

According to this tradition, by a majority vote in the “upper chambers of the

house of Nitza in Lod”4 it was decided that someone, who is forced to choose
between committing an act of idolatry, a sexual trans­gression or bloodshed
and being killed, should rather let him/herself be killed than commit one of
the three enumerated transgressions.5 With relation to this tradition Rav Dimi,
a nehutei (i.e. a scholar who traveled regularly between Babylonia and the
Land of Israel in order to report traditions from the Land of Israel to the schol-
ars of Babylonia), transmits in the name of Rabbi Yohanan that the majority
vote of the Lodian sages appertains solely to a situation in which there is no
“royal decree” (‫ )גזרת המלכות‬forbidding the practice of Judaism. If, however,

2 I explain this term in the analysis below.

3 Rashi explains that ‫ קוואקי ודימונקי‬are utensils made of metal, into which one puts coals
and which one places in front of the kings’ tables so as to warm them (cf. Rashi, s.v. ‫הני‬
‫)קוואקי ודימוניקי‬. That the Jews were asked to give their utensils to the unspecified “them” is
explained by Rashi as follows: there was a festival day of the Persians, at which the “priests
of idolatry” took by force the lights from every household, in order to place that light in their
“house of idolatry” (cf. Rashi, s.v. ‫)היכי יהבינן להו‬.
4 On the dating of this meeting in Lod, cf. M. D. Herr, “Persecutions and Martyrdom in Hadrian’s
Days,” Scripta Hierosolymitana 23 (1972) 109, n. 79.
5 The decision of the Lodian sages installs limits to an act, which constitutes, according to mod-
ern juridical thought, an act of self-defense. For example, a physician may not kill a human-
being even if this saves the life of five other patients. This, however, does not mean that when
the physician’s own life is threatened through some pursuer, he may not kill that man. His
deeds would be categorized as acts of self-defense, and the physician could, if he cannot
stop the pursuer through lesser means, also legitimately kill the offender. (Cf. W. Sinnott-
Armstrong, “Consequentialism,” in: Stanford Encyclopedia of Philosophy [Stanford 2006]; on
the so-called “Transplant-problem” see http://plato.stanford.edu/entries/ consequentialism.)
rescue from transgression through death 135

such a decree was issued, even the most insignificant religious custom must be
defended at all costs: “One must be killed rather than transgress even a minor
precept.” The idea “to transgress and not to be killed” (‫)יעבור ואל יהרג‬, that is,
the saving of one’s life by means of a transgression, applies solely to a situation
in which Judaism itself is not “officially persecuted” by the state.6
Rabin, another routine traveler between Babylonia and the Land of Israel,
transmits a different restriction to the majority vote of Lod, again, in the name
of Rabbi Yohanan: whether or not an official decree of persecution, a “royal
decree,” was issued is irrelevant; what matters is solely whether the person,
who is forced to choose between a transgression and death, commits the trans-
gression in public or in private.7 If one is forced to make that choice in public,
any trans­gression, even a minor transgression, is forbidden; but if one is forced
to choose between a transgression and death in private, one may commit a
minor transgression in order to save one’s life.8
The subsequent anonymous question about Esther—“But was not Esther’s
case in public?”—may be understood as a stammaitic editorial reaction to the
restriction which Rabin transmits in the name of Rabbi Yohanan on someone
who is forced to choose between death and a transgression in public: if, as
Rabin argues, a Jew, who is forced to choose between a minor, public transgres-
sion and death, should rather let himself be killed than commit that minor
public transgression, then, according to the same logic, Esther should also have
let herself be killed, rather than commit a minor transgression in public! But
since the actions which Esther undertook are accepted by all the rabbis per se
as righteous9—if she behaved in a particular fashion, her behavior cannot be

6 This tradition coheres with the toseftan “Nothing stands against mortal danger, except for
idolatry, forbidden sexual relations and bloodshed” (tShab 15:16), which is equally embedded
in a context of persecution. Following this general declaration, the Tosefta lets an anony-
mous voice ask, “Under what circumstances?” and answers, that the permissibility to trans-
gress every commandment except for the three mentioned here applies solely to normal
times, “but in a time of persecution, for even the slightest of any of the lesser commandments
a man must give his life” (tShab 15:17).
7 That a Jew should even die for a minor transgression he committed in public is transmitted
also in the Yerushalmi, in Shevi 4:2, 35a–b and in ySanh 3:6, 21b.
8 On a possible reconstruction of the developments which led to these restrictions cf. Herr,
“Persecutions,” 116–21.
9 J. R. Baskin, “Erotic Subversion: Undermining Female Agency in bMegillah 10b–17a,” in:
T. Ilan et al. (eds.), A Feminist Commentary on the Babylonian Talmud: Introduction and Studies
(Tübingen 2007) 222, writes: “The rabbis were certainly hesitant to condone the union of a
Jewish woman and a gentile man. However, Esther’s salvific role in her eponymous biblical
story is so strongly drawn that she was, essentially, above direct rabbinic criticism.”
136 tzuberi

construed as being guided by a fallacy—the question about Esther is an argu-

ment (or more precisely, a rhetorical question) directed against Rabin’s view:
the assumption of this question is that Esther did commit a minor transgres-
sion in public, which indicates that a Jew who is forced to choose between
death and a minor, public transgression, in fact may commit that minor trans-
gression in public rather than let himself be killed.
In order to defend the tradition Rabin transmits in the name of Rabbi
Yohanan, both Abbayye and Rava attempt to construe the minor public trans-
gression Esther is supposed to have committed as not being identical with the
minor public transgression with which Rabin is concerned.
According to Abbayye, Esther was qarqa olam (‫)קרקע עולם‬, literally, the
“soil of the world.” This term designates soil which is not considered tainted
by an idolater even if his idol was attached to it (cf. bSanh 47b and bAZ 54b).10
Therefore, the “soil of the world” is also not forbidden to be used by Jews after
the idol has been detached from it. In addition to bSanh 47b and bAZ 54b, the
expression appears in the Bavli three more times (in bBQ 28b, 30a and 50b),11
and in each instance it is used to indicate that for an injury caused by “the soil
of the world” nobody can be held liable.12
Thus, according to Abbayye, Esther did not need to save herself from a
transgression through her own death, because being “the soil of the world,”
her “attachment” to the idol—Ahasuerus—does not make her tainted by him;
she is like the “soil of the world” not forbidden to be “used” by Jews after the

10 “It was the practice of people to take earth [= Esther] from Rav’s grave and apply it on the
first day of an attack of fever [even though the use of an object belonging to the dead is
forbidden, cf. bAZ 29b]. When Shmuel was told of it, he said: They do well; it is the soil of
the world and the soil of the world does not become forbidden, for it is written: ‘And he
cast the dust thereof (of the Ashera) upon the graves of the common people’ (2Kgs 23:6).
Thus he compares the graves of the common people to idols. Just as [the use of] idols is
not forbidden when they are ‘attached’ [the technical term for soil, mountains, etc.] [. . .],
so here too, what is ‘attached’ [what belongs to the dead] is not forbidden” (bSanh 47b).
“He, who worships a piece of ground, does not render it prohibited” (bAZ 54b).
11 In bNid 57b the meaning of the expression is simply “the ground.”
12 Cf. for example, bBQ 50b, “If a man digs a pit on public ground and an ox or an ass falls
into it, he becomes liable [. . .] (mBQ 5:5). Rav stated: The liability imposed by the Torah
in the case of a pit is for the unhealthy air created by the excavation, but not for the blow
given by it. It could hence be inferred that he held that so far as the blow was concerned
it was the soil of the world that caused the damage (for which the defendant has not to be
held liable).” The expression occurs one more time in bNid 57b; here, however, the term
signifies soil: “If a woman examined the soil of the world and after sitting on it, found on
it some blood, she remains clean . . .”
rescue from transgression through death 137

idol has been detached from it.13 Esther never becomes prohibited to her “real
husband,” Judaism. Moreover, being the soil of the world, Esther cannot be
held liable for any damage: the soil cannot choose, but is “acted upon.” Esther,
accordingly, was an entirely passive object, not in a situation of relative, but of
absolute coercion.14 She could not choose to marry Ahasuerus.15
Rava too regards Esther’s minor public transgression as consisting of her
marriage to a gentile, but follows a different strategy of justifi­cation. He
explains that Esther did not need to save herself by death, because Ahasuerus
did not marry her in order to violate her faith, but for the sake of his personal
pleasure.16 According to Rava, only if a tyrannical ruler commands a Jew to
transgress in order to violate his faith does the Jew have a duty to save himself
from transgression rather than to transgress and thereby save his life.17
Both Rava and Abbayye understand the question about Esther as an argu-
ment against Rabin, according to whom a Jew, who is forced to choose between
a minor, public transgression and death should let himself be killed rather than
commit that minor, public transgression. According to the two rabbis, then,
Esther is comparable to a “tyrant’s victim,” who is forced to decide between a
transgression and death—yet, since the actions which Esther undertook are
by definition righteous, both rabbis attempt to construe her transgression to

13 It is unlikely to assume that Esther’s case is understood by Abbayye to be an act of adul-

tery. First of all, intermarriages are not considered adulterous relations. Only if one
assumes that Esther was married to Mordechai and he did not divorce her before she
married Ahasuerus (cf. bMeg 13b) would the relation to the latter be adulterous. And
even if Abbayye would understand the question on Esther as “But was not Esther’s case
[= adultery] in public?” the reference to public adultery does not make sense since adul-
tery, even if it is committed in private, is absolutely prohibited. Furthermore, Esther’s
previous marriage to Mordechai is understood by the amoraim to have been kept a secret.
Her adultery thus did, in no way, take place in public.
14 “ ‘Absolute coercion’ means someone throws me down in front of an idol; ‘relative coer-
cion’ means I choose to bow down to the idol because I fear otherwise being murdered. In
the former, the individual’s body is the object of another’s action; in the latter, the person’s
will is the object of coercion, for in relative coercion the individual must freely choose to
actively abjure his religion to avoid death” (H. Soloveitchik, “Halakha, Hermeneutics and
Martyrdom in Medieval Ashkenaz [Part I and II],” Jewish Quarterly Review 94 [2004] 80).
15 Cf. also Rashi, s.v. ‫קרקע עולם היא‬: ‫ הוא עושה בה מעשה‬,‫( אינה עושה מעשה‬She does not
act, he acts upon her).
16 This argument is in a way related to the one of Abbayye: Esther is here not the mountain,
or the soil upon which the idol stands, but the idol itself, that is, Ahasuerus’s “idol.”
17 The idea that the tyrant’s intention bears upon the permissibility to transgress is to
be found also in Shevi 4:2, 35a–b and in ySanh 3:6, 21b. Cf. on this subject also Gray, “A
Contribution to the Study of Martyrdom.”
138 tzuberi

be not identical with the transgression Rabin is concerned with. They aim at
dissolving the equivalence between her and a Jew, who is forced to choose
between death and a minor public transgression. In other words, they attempt
to explain why Esther rightfully did not need to let herself be killed (or kill
herself); why she is an exception which proves the general rule. The majority
vote of the Lodian sages, and in particular, the addition, which Rabin transmits
in the name of Rabbi Yohanan is that a Jew, who is forced to choose between
bloodshed, a forbidden sexual act, idolatry and death, should rather let himself
be killed, than commit the transgression, and in public, he may not even com-
mit a minor transgression.

2 Questioning the Reading of Abbayye and Rava

However, in what follows, the gemara questions Abbayye’s and Rava’s reading
of the anonymous question:

‫ מעשה באדם אחד שנתן עיניו באשה אחת והעלה לבו‬:‫אמר רב יהודה אמר רב‬
,‫ ימות‬:‫ אמרו חכמים‬.‫ אין לו תקנה עד שתבעל‬:‫ ואמרו‬,‫ ובאו ושאלו לרופאים‬.‫טינא‬
‫ תספר עמו‬.‫ ואל תעמוד לפניו ערומה‬,‫ לו תעמוד לפניו ערומה? ימות‬.‫ואל תבעל‬
.‫ ולא תספר עמו מאחורי הגדר‬,‫מאחורי הגדר? ימות‬

Rav Yehudah said in Rav’s name: A man once had set his eyes on a certain
woman, and his heart was consumed by his burning desire. When the
physicians were consulted, they said: His only cure is that she be pen-
etrated by him. Thereupon the sages said: Let him die rather than that
she should be penetrated by him. [Then the physicians suggested:] What
if she stands nude before him? [The sages answered:] Let him sooner
die rather than that she stand naked before him. [Then, said the physi-
cians:] What if she converses with him from behind a fence? Let him die
[the sages replied,] rather than that she should converse with him from
behind a fence (bSanh 75a).

According to the reasoning of Rava and Abbayye, Esther was permitted to marry
Ahasuerus because Ahasuerus did not intend to violate her religion (Rava), or
because she was “the soil of the world” (Abbayye). It follows logically, then,
that any other woman, whom a man desires without attempting to violate her
faith, should be given to that man. Yet, the sages’ severity as displayed in this
story implies the opposite: Even if the desiring man pines away because of his
unfulfilled passion, the woman should not be given to him. He may not even
rescue from transgression through death 139

converse with her from behind a fence, even though he, too, simply desires her
and has absolutely no intention of violating her faith, and even though she,
too, may be considered the “soil of the world” just as much as Esther.
This story thus questions the way Rava and Abbayye read the question
about Esther’s minor, public transgression (“But was not Esther’s case in pub-
lic?”), for it implies, firstly, that Esther was not, as assumed by Abbayye, found
in a situation of absolute coercion, like “the soil of the world.” In the case of the
man who desires a woman, there obviously exists a choice either to give her to
him or to refuse to do so. The physicians interrogate the rabbis on the matter. If
the woman was in a situation of absolute coercion, her body would be subject
to some ulterior force, a force majeure, and no interrogation would have been
possible at all. Neither the case of Esther, nor the case of the unnamed woman,
is thus one in which a man acts upon a woman’s body.
The story implies also that, contrary to Rava’s explanation of “Esther’s case,”
the woman is not given to him regardless of the man’s motivation—regardless
of whether he attempts to satisfy his desire, to violate her religion, or even to
save his life. The fact that Esther was “given” to Ahasuerus thus cannot, as Rava
claims, be related to the fact that Ahasuerus simply desired her. Also, the man
in the story simply desires the woman, but the woman is, nevertheless, not
given to him. Thus, the legitimacy of “Esther’s case” cannot depend on the fact
that Ahasuerus simply desired her.
What, then, is the justification of “Esther’s case”? Why may she be given to
Ahasuerus, and why may the unnamed woman of the story not be given to
the man? Rav Jacob bar Idi and Rav Shmuel bar Nahmani dispute the sages’
reasoning:

‫ וחד‬,‫ אשת איש היתה‬:‫ חד אמר‬.‫פליגי בה רב יעקב בר אידי ורב שמואל בר נחמני‬
‫ שפיר; אלא למאן דאמר‬,‫ בשלמא למאן דאמר אשת איש היתה‬.‫אמר פנויה היתה‬
‫ רב אחא בריה דרב‬.‫ משום פגם משפחה‬:‫ מאי כולי האי? רב פפא אמר‬,‫פנויה היתה‬
‫ ולינסבה? מינסב לא מייתבה‬.‫ כדי שלא יהו בנות ישראל פרוצות בעריות‬:‫איקא אמר‬
‫ מיום שחרב בית המקדש ניטלה טעם ביאה‬:‫ דאמר רבי יצחק‬,‫ כדרבי יצחק‬,‫דעתיה‬
)‫ ״מים גנובים ימתקו ולחם סתרים ינעם״ (משלי ט יז‬:‫ שנאמר‬,‫וניתנה לעוברי עבירה‬
.)‫(ב׳ סנהדרין עה ע״א‬

Rav Jacob bar Idi and Rav Shmuel bar Nahmani dispute therein. One said
that she was a married woman; the other that she was unmarried. Now,
this is intelligible on the view, that she was a married woman, but on the
latter, that she was unmarried, why such severity? Rav Papa said: Because
of the disgrace to her family. Rav Aha bar Rav Iqa said: So that the daugh-
ters of Israel may not be immorally dissolute. Then why not marry her?
140 tzuberi

Marriage would not assuage his passion, as Rabbi Isaac said: Since the
destruction of the Temple, sexual pleasure has been taken [from those
who practice it lawfully] and given to sinners, as it is written: “Stolen
waters are sweet, and bread eaten in secret is pleasant” (Prov 9:17) (bSanh
75a).

In this discussion one of the rabbis holds that the sages were so severe because
the woman of the story is a married woman. If she is married, the man’s inter-
course with her would constitute an act of adultery, a transgression which
may be committed neither in public nor in private, neither with nor without a
decree of persecution. In this case, then, the sages’ severity would cohere to the
toseftan dictum, that there is “no healing through bloodshed, forbidden sexual
relations (including inter­course with a married woman) and idolatry,” or alter-
natively to mSanh 8:7: he who attempts to rape a betrothed girl is “saved by his
life,” that is, killed before committing the transgression; the man who desires a
married woman should rather die than have intercourse with her.
However, one of the sages holds the opinion that the woman of the story is
not married. How can the sages’ severity be explained in this case? Why do the
sages not simply suggest—as Scripture itself does—that he marry her? After
all, intercourse with her would, after marriage, not involve any transgression.
The anonymous answer offered to this problem is the following: if the man
marries her, he does not desire her anymore. His survival requires a transgres-
sion. “Bread not eaten in secret” or “water not stolen” would not only have a
different, less appetizing taste, but would also not save the man’s life. There
simply is no way to save his life through legitimate intercourse.18
This explanation for the sages’ severity links the story of the unnamed woman
to “Esther’s case”: Both the rescue of the Jews requires a transgression—inter-
course between Esther and Ahasuerus—and the rescue of the love-sick man
requires a transgression—illegitimate intercourse with the unnamed woman.
The difference between the two cases pertains to the identity of the endan-
gered: the unnamed woman who is confronted with a lusting man, does not
need to save the life of an innocent victim. If she does not give in to the man’s
desire, if she is not “given” to him, the one who possibly loses his life is the
“offender” himself, the lusting man. His rescue from death requires a transgres-
sion, just as the Jews’ rescue from death requires a transgression; but in contrast
to the threatened Jews of Esther’s story, this lusting man is the very source of
his own dangerous situation. He is therefore not to be compared to the threat-
ened Jews, but rather to the pursuer of mSanh 8:7: in this mishnah the pursuer

18 For an identical interpretation of this verse, cf. bNed 91b, bSot 7a and bSanh 26b.
rescue from transgression through death 141

is actively saved from sin through the bystander’s lethal intervention. So too in
this case the lusting man—who himself is some kind of a pursuer—is passively
saved from sin through death, that is, through not being given the woman he
desires. In other words: the pursuer—be he someone who attempts to murder,
or someone who attempts to have illegitimate intercourse—is “saved by his
life” from sin. Whether the pursuer is a tyrannical ruler, or is situated within a
man himself, he may not save his own life through committing one of the three
crimes, but, rather, should be saved from sin through his death. The prohibition
of murder, forbidden sexual relations and idolatry is absolute and limits the
power of the tyrannical ruler, as much as it limits the power of the impulses,
which are in man himself, and which turn him into a “pursuer.”
In contrast to the case of the desiring man, in the case of Esther those whose
lives are threatened are not themselves “pursuers,” but innocent victims.
Esther’s transgression is thus legitimate not because she is akin to the entirely
passive “soil of the world,” or because Ahasuerus does not aim at the violation
of her faith, but for the very opposite reason—her submission to Ahasuerus
is legitimate because of her motivation: the rescue of innocent lives from the
decree of persecution. Accordingly, in the Esther Midrash in the Bavli, the
point of time when Esther transgresses is fixed precisely at the moment when
she goes to Ahasuerus voluntarily, that is, at the very moment she approaches
him in order to save the Jews:

‫ שלא‬:‫ אמר רבי אבא‬.)‫״לך כנוס את כל היהודים וגו׳ עד אשר לא כדת״ (אסתר ד טז‬
‫ ״וכאשר אבדתי אבדתי״‬.‫ ועכשיו ברצון‬,‫ שבכל יום ויום עד עכשיו באונס‬,‫כדת היה‬
.‫ כשם שאבדתי מבית אבא כך אובד ממך‬.)‫(שם‬

“Go, gather together all the Jews etc. which is not according to the
custom”19 (Est 4:16). Rabbi Abba said: It will not be [she said] according
to the custom of every other day. Till now [I was with Ahasuerus] under
compulsion, but now I will do so of my own will. “And if I am lost, I am
lost” (ibid.). As I am lost to my father’s house, so I shall be lost to you
(bMeg 15a).

In this talmudic text Esther’s transgression takes place at the moment she goes
voluntarily and actively to Ahasuerus, which is exactly at the moment she saves

19 The JPS translation reads: “Go, gather together all the Jews who are present in Shushan,
and fast for me, and neither eat nor drink three days, night or day; I also and my girls will
fast likewise; and so will I go to the king, though it is against the law; and if I perish, I
perish.”
142 tzuberi

her people. Only through this voluntary deed is she forever forbidden to her
legitimate husband, Mordechai and Judaism: “If I am lost, I am lost.” Before, her
relation to Ahasuerus could be halakhically legitimized through the assump-
tion that it was forced, but now, precisely at the time of persecution, she com-
mits a voluntary transgression; she is no longer a passive, untainted “soil of the
earth.”20
If one follows the tradition, according to which Esther was married to
Mordechai prior to her marriage to Ahasuerus and was not divorced before
entering a second marriage (cf. bMeg 13b), then her voluntary approach to
Ahasuerus amounts to an act of adultery.21 In any case, however, her volun-
tary approach to Ahasuerus inherently involves a public transgression of the
law as the anonymous question “But was not Esther’s case in public?” implies:
Even if one assumes that the Jews themselves did not know about Esther being
a Jewess, and that they were, accordingly, unaware of her intercourse with a
gentile—when she decides to go to Ahasuerus voluntarily, she ordains a public
fast in spite of the feast-days of Pesah: “Go, gather together all the Jews who are
present in Shushan, and fast for me, and neither eat nor drink three days, night
or day; I also and my girls will fast likewise; and so will I go to the king, though
it is against the law; and if I am lost, I am lost” (Est 4:16).22
To the stam of the gemara, therefore, Esther is no passive victim forced to
transgress by a tyrannical ruler, but rather a rescuer, forced to transgress by the
obligation to save life. Thus, at the sugya’s end, there stands an implicit, yet

20 Cf. Rashi, s.v. ‫אבדתי ממך‬: ‫ואסורה אני לך דאשת ישראל שנאנסה מותרת לבעלה וברצון‬
‫“( אסורה לבעלה‬I am lost to you.” Forbidden am I as an Israelite woman, for one who was
raped is permitted to her husband, but one who willingly [had intercourse with another
man] is prohibited to her husband).
21 B. D. Walfish, “Kosher Adultery? The Mordecai–Esther–Ahasuerus Triangle,” Prooftexts 22
(2002) 306 observes: “This interpretation [that she was married to Mordechai—C.T.], far
from smoothing over Esther’s intermarriage, in fact raised the level of transgression from
one of intercourse with a gentile to possible adultery.” The tradition, that Esther was mar-
ried to Mordechai prior to her marriage to Ahasuerus is found only in the Babylonian
Esther Midrash (bMeg 10b–17a) and its derivatives; cf. E. Segal, The Babylonian Esther
Midrash: A Critical Commentary (Atlanta, 1994) 2:51. However, it is corroborated indepen-
dently in the Septuagint: “The presence of this tradition in the Septuagint, which dates
from the third century B.C.E., suggests that it is indeed very ancient” (Walfish, “Kosher
Adultery,” 307).
22 “Then Esther bade them return answer unto Mordecai (Est 4:15). She said to him: Go,
gather all the Jews that are present in Shushan, and fast for me, and neither eat nor drink
three days” (Est 4:16): These were the thirteenth, fourteenth and fifteenth of Nisan. He
sent back word to her: But these include the first day of Pesah? She replied: Elder of Israel,
why is there a Pesah? Mordecai thereupon acceded to her request [. . .]” (EstR 8:7).
rescue from transgression through death 143

obvious, rejection of Abbayye’s and Rava’s reading of Esther’s transgression:

that Esther’s transgression was legitimate is not due to her being “the soil of the
world,” or due to Ahasuerus’ intentions. Rather, it was Esther’s intention, her
aim, which turned her transgression into a legitimate transgression.

3 Implications of this Second Reading

This alternative reading, according to which Esther is a rescuer who volun-

tarily transgresses in order to save the Jews’ lives, is not formulated explicitly.
The reason for this avoidance of an explicit formulation may be found in the
implications this reading has regarding the majority vote of the Lodian sages.
Once Esther’s deed is legitimized on account of its effect—the rescue
of innocent lives—one could argue that also a Jew who is forced to choose
between a transgression and death should be permitted to commit the trans-
gression in order to save his own life—provided the latter does not kill or rape
another innocent human being.
Of course, if a Jew refuses to commit rape or murder in order to save
his own life, one may argue that this refusal is motivated by an attempt to not
kill or dishonor an innocent human being. As Aryeh Cohen notes, the stam of
the gemara does not categorize the obligation to die rather than to murder or
rape one’s fellow as an instance of qiddush ha-shem, but as a result of logical
reasoning.23 For just before our sugya we read “And how do we know this of
murder [e.g. the prohibition to commit murder in order to save one’s own life]
itself? It is logical reasoning (‫)סברא הוא‬.24 Even as one who came before Rava
and said to him: The governor of my town has ordered me: Go and kill so and
so; if not, I will slay you. He answered him: Let him rather slay you than that
you should commit murder; who knows that your blood is redder? Perhaps his
blood is redder?” (bSanh 74a).25

23 “The obligation to die rather than murder one’s fellow is not necessarily an instance of
Qiddush ha-Shem. It is a result of sevara—what greater right to live do you have than any-
one else? . . . The idea of Qiddush ha-Shem is only introduced with the prooftext from Lev
22:32: ‘Neither shall you profane my name, but I will be hallowed [ve-niqdashti] amongst
the children of Israel. . . .’ Qiddush ha-Shem, for this sugya at least, is inherently tied to
idolatry” (A. Cohen, “Response to Elizabeth S. Alexander’s ‘Dialogues on the Theme of
Martyrdom’,” Textual Reasoning: The Postmodern Jewish Philosophy Network 5/1 [1996]).
24 Cf. also D. Weiss Halivni, “The Meaning of S’vara,” S’vara 1 (1990) 3–6.
25 “Esther’s case” thus does not contradict the rule, that one may not kill or rape another
innocent human being in order to save oneself.
144 tzuberi

However, in contrast to murder and rape, if a Jew refuses to commit

idolatry, his refusal is obviously not motivated by an attempt to not harm
another innocent human being. An act of idolatry does, after all, not endanger
any human being, and a Jew who is forced by a tyrant to choose between idola-
try and death is therefore “killed and does not transgress” not in order to avoid
harm to other human beings. Why, then, should he sacrifice his life for the sake
of avoiding idolatry, whereas Esther may commit a transgression, possibly as
severe as adultery, in order to save the Jews? If Esther was forced by the obliga-
tion to save life to commit a transgression, then why should the same argument
not be used of a Jew, who is forced by a tyrant to choose between idolatry and
death? Is not the latter, too, forced to commit the transgression by the obliga-
tion to save his own life?26
The fact that Esther commits a transgression in order to save her fellow Jews
thus contradicts the majority-vote of the Lodian sages, according to whom a
Jew, when being forced by a tyrant to choose between idolatry and death, has to
prefer death. Even more so does the legitimacy of Esther’s deed contradict the
conditions which Rav Dimi and Rabin transmit concerning the Lodian major-
ity vote: as mentioned above, according to Rav Dimi, a Jew may not commit
even a minor transgression in order to save one’s life if there is a royal decree
forbidding the practice of Judaism; according to Rabin, a Jew may not commit
even a minor transgression if the latter is about to occur in public.
The legitimacy of “Esther’s case” thus questions the majority vote of the
Lodian sages: if Esther commits a transgression in order to save the Jews, so
the stam might have reasoned, then also a Jew who is forced to choose between
transgression and death may commit the transgression for the sake of his own
life, provided that this transgression does not cause harm to another innocent
human being. Whereas the Lodian sages reason that in times of persecution
one may not commit idolatry even if one is forced to choose between idola-
try and death, and Rav Dimi adds that in times of persecution one may not
save oneself even from a minor transgression, Esther transgresses precisely in
a time of persecution. The editors of the Babylonian sugya seem to argue that
just as the obligation to save life legitimizes the transgressions of Esther, so
too it might also legitimize the transgression of a Jew, who is forced to choose
between idolatry and his death.
The Lodian majority vote is thus not overtly dismissed. Nevertheless, there
runs a decisively veiled, yet strong, “unofficial” layer beneath this “official”

26 One could even imagine a qal va-homer such as “if the voluntary idolater does not need to
be saved by his life, as mSanh 8:7 claims, then how much more so should a person, who is
forced to choose between death and idolatry, not need to be saved by his life?”
rescue from transgression through death 145

reading, which uses Esther as its emblematic icon and propagates not res-
cue from transgression through death, but instead, rescue from death through
transgression.

4 The Sages of Lod and the Babylonian Stam

All the traditions which are cited by the stam of bSanh 74a concerning the
subject of transgressions a Jew may not commit even at the cost of his own life,
are also to be found virtually without alterations in the Yerushalmi.27 These
traditions are, moreover, explicitly and repeatedly ascribed by the Babylonians
themselves to different rabbis from the Land of Israel. The tradition according
to which someone who is forced to choose between the three cardinal crimes
and death should rather let himself be killed, appears first in Tosefta (tShab
15:16–7), and is ascribed in the above sugya to a majority vote of the sages, who
met at a private gathering in Lod. All the following restrictions and qualifica-
tions of this majority-vote are explicitly identified by the stam as originating in
the Land of Israel. Aside from the tradition ascribed to the gathering in Lod,
also the view that one may commit idolatry in a situation of relative coercion
is ascribed to a Palestinian rabbi, Rabbi Yishmael (cf. bSanh 73b). Rav Dimi and
Rabin, who add a restriction on the Lodian majority vote, assign it to Rabbi
Yohanan—another Palestinian rabbi—“when they come [from the Land of
Israel].” The Babylonian rabbis seem not to add anything “new” to these tradi-
tions. They only “bring” them from the Land of Israel, unambiguously refrain-
ing from claiming them as their “own,” original creation. As Soloveitchik
observes: “The rules of martyrdom given in the Bavli come from Palestinian
sources [. . .] Even subsidiary discussions in the Bavli are either by a Palestinian
amora or echo a Palestinian position.”28

27 Cf. Shevi 4:2, 35a–b and its parallel in ySanh 3:6, 21b, and yAZ 2:2, 40d–41a and its parallel
in yShab 14:4, 14d–15a. For an analysis of these texts, cf. Gray, “A Contribution to the Study
of Martyrdom.”
28 Soloveitchik, “Halakha, Hermeneutics and Martyrdom,” Part II, 282. Cf. also R. Kalmin,
“Rabbinic Traditions about Roman Persecutions of the Jews: A Reconsideration,” Journal
of Jewish Studies 54 (2003) 22: “In most respects Babylonian and Palestinian traditions
describe the Romans interfering with the same set of Jewish practices. This is no argu-
ment in favor of the historicity of these traditions, but it does show that much was
absorbed into the Bavli from Palestine without substantive change. Babylonian rabbis,
for the most part, did not invent stories and traditions, or at least motifs about Roman
persecutions, and attribute them to Palestinian rabbis, nor did the Babylonians, for the
most part, extensively doctor Palestinian traditions.”
146 tzuberi

Explicitly identifying the origin of the traditions on the absolutely pro-

hibited transgressions, the Babylonian stam weaves these traditions into the
sugya, quotes them faithfully—but in addition, it lets the sugya end with
a Babylonian heroine of exile, Esther, and the narrative on the lusting man.
This passage does not appear anywhere else in rabbinic literature, but solely
in Bavli Sanhedrin,29 and as explained above, it is this particular passage which
questions the premise that a Jew forced to choose between a transgression and
death has to choose death.
To ascribe explicitly the traditions on the transgressions a Jew may not
commit under any circumstance to the sages from the Land of Israel and to
let the discussion about them end with “Esther’s case” is, I maintain, a con-
scious editorial decision. The Babylonian stam, through concluding this
sugya with a Babylonian icon, juxtaposes the traditions from the Land of
Israel with an implicit, yet powerful, counter-tradition. While unwilling to dis-
card the inherited traditions from the Land of Israel, it seems as if the edi-
tor was not wholly content with them either. Something, he seems to think, is
“wrong” with letting oneself be killed for the sake of avoiding idolatry or a
minor transgression.30 Yet, something may have seemed to him “wrong,” too,
with an outward rejection of the Palestinian traditions on those transgressions
a Jew may not commit even for the sake of saving a life. After all, these tradi-
tions are far from being questioned or rejected in an outspoken, immediate
manner.

29 The story on the love-sick man appears also in yAZ 2:2, 40d–41a and in yShab 14:4, 14d–
15a. The idea that only a transgression is “sweet” is commonly associated with the verse
“Stolen waters are sweet, and bread eaten in secret is pleasant” (Prov 9:17) (cf. bNed 91b;
bSot 7a; bSanh 26b). However, only in the Bavli is the story associated with the question of
“Esther’s case” and thus only here does it convey the message outlined above.
30 According to Gray’s analysis of the halakhah on martyrdom, as represented in the
Yerushalmi (“A Contribution to the Study of Martyrdom,” 243), the editor(s) of this latter
work associated martyrdom not only with a public denial of committing a transgression,
but also viewed the great hardship of observing the commandments as “quasi-martyrdom.”
A Tale of Two Sinais: On the Reception of the
Torah according to bShab 88a

Amram Tropper

A short but well-known talmudic discussion found on bShab 88a culminates

in a shocking and radical derashah. In this concluding derashah, Rava makes
the bewildering claim that the Torah only became legally binding in the time
of Mordechai and Esther. In other words, this Talmudic exchange concludes
that the true Sinaitic moment, that is, the authentic reception of the Torah’s
binding authority, occurred in the time of Mordechai and Esther, long after the
reception of the Torah on Mount Sinai as narrated in the Pentateuch.1
This short and surprising discussion is comprised of three segments. It
opens innocently with an evocative derashah, follows with a legal query, and
concludes with Rava’s astonishing statement. Each segment of the discus-
sion stems from elsewhere in rabbinic literature and through a comparison
of each segment to its literary backdrop I hope to reveal how the Bavli has
interpreted its sources anew. Indeed, the revolutionary nature of the Bavli’s
discussion is greatly intensified when read in light of its sources, for the Bavli’s
use and reinterpretation of these sources illustrate how it conveys an entirely
new and unprecedented idea by recycling and recasting already extant literary
materials.

‫ מלמד‬:‫ אמר רב אבדימי בר חמא בר חסא‬:)‫״ויתיצבו בתחתית ההר״ (שמות יט יז‬

‫ אם אתם מקבלים‬:‫ ואמר להם‬,‫שכפה הקדוש ברוך הוא עליהם את ההר כגיגית‬
.‫ ואם לאו—שם תהא קבורתכם‬,‫התורה—מוטב‬
!‫ מכאן מודעא רבה לאורייתא‬:‫אמר רב אחא בר יעקב‬
‫ דכתיב ״קימו וקבלו היהודים״‬2,‫ הדור קבלוה בימי אחשורוש‬,‫ אף על פי כן‬:‫אמר רבא‬
.)‫ (ב׳ שבת פח ע״א‬.‫ קיימו מה שקיבלו כבר‬,)‫(אסתר ט כז‬

“And they stood at the bottom of the mount” (Exod 19:17): Rav Avdimi
bar Hama bar Hasa said: This teaches that the Holy One, blessed be He,
overturned the mountain upon them like an [inverted] cask, and said to

1 See e.g. Exod 24; Deut 5. See also Tosefot ad loc., s.v. moda‘a rabah le’oraita.
2 Various textual witnesses refer to Mordechai and Esther rather than Ahaseurus. See, for
example, MS Munich 95 and MS Vatican 108. See also R. Rabbinovicz, Variae Lectiones in
Mischnam et in Talmud Babylonicus: Tract. Sabbath (Munich 1875) 185.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�06

148 tropper

them: If you accept the Torah, it is well; and if not, there shall be your
burial. Rav Aha bar Jacob said: This furnishes a strong protest against the
Torah! Said Rava: Yet even so, they re-accepted it in the days of Ahasuerus,
for it is written “the Jews confirmed and irrevocably accepted” (Est 9:27):
[i.e.] they confirmed what they had accepted long before3 (bShab 88a).

Before comparing the three segments of the text to their respective sources,
let us first consider the thrust and meaning of each segment. The text opens
with Rav Avdimi bar Hasa bar Hama’s derashah on the clause from Exodus
which describes where the Israelites stood when Moses ascended Mount Sinai
to receive the Torah: ‫ויתיצבו בתחתית ההר‬, “And they stood at the bottom of the
mount.” As the Maharsha explains, Rav Avdimi’s derashah apparently inter-
prets the preposition ‫( ב‬bet) in the word ‫ בתחתית‬as “in” rather than as “at,”
rendering the biblical clause “And they stood at the bottom of the mount” as
“And they stood in the bottom of the mount.”4 In fleshing out his derashah, Rav
Avdimi offers the frightening picture of the Israelites receiving the Torah within
Mount Sinai as God overturned “the mountain upon them like an (inverted)
cask.” According to Rav Avdimi, God threatened His people with certain death
should they refuse the Torah; so, with no viable alternative, they received the
Torah under severe duress.5
In the wake of Rav Avdimi’s derashah, Rav Aha bar Jacob questions the con-
stitutional validity of the compelled reception of the Torah at Mount Sinai
since, in rabbinic law, contractual commitments, such as the Jewish commit-
ment to observe the laws of the Torah, are only valid when made willingly.
According to Rav Aha bar Jacob, if the Jews had committed themselves to the
Torah under threat of death, as Rav Avdimi claims, then they would have a
“strong protest,” that is, an excellent excuse, for not fulfilling the Torah’s com-
mandments. In other words, if God should ever summon the Jewish people to
the heavenly court for neglecting the Torah’s commandments, the Jews would
have an ideal defense, since contracts or agreements made unwillingly are null
and void.6

3 bShab 88a. Cf. bAZ 2b. The translation is that of H. Freedman, The Hebrew—English Edition of
the Babylonian Talmud: Shabbath (London 1972), ad loc., slightly modified.
4 See Maharsha ad loc., s.v. she-kafah ‘aleyhen ha-har ke-gigit.
5 It seems likely that Rav Avdimi’s derashah was designed to convey the notion that the chil-
dren of Israel, who had witnessed God’s mighty hand in the form of the miraculous plagues
in Egypt and the splitting of the Red Sea, and whom God Himself had addressed directly at
Mount Sinai, had no choice but to accept the Torah. In other words, the image of the over-
turned and threatening mountain seems to be a picturesque way of saying that with all they
had witnessed the generation of the wilderness could not have refused the Torah.
6 See Rashi ad loc., s.v. moda‘a rabah.
the reception of the torah according to bShab 88a 149

Responding to Rav Aha bar Jacob, Rava concludes our short talmudic dis-
cussion with a derashah on a clause from Est 9:27: ‫קימו וקבלו היהודים‬, “the Jews
confirmed and irrevocably accepted.” In the context of Esther, this clause refers
to the process wherein the Jews in the time of Mordechai and Esther assumed
the obligation to celebrate the holiday of Purim. By means of his derashah,
Rava detaches the clause from its immediate context, the celebration of Purim,
and links it to an alternative context, the reception of the Torah. Rava does not
explicitly divulge the midrashic technique enlisted by his derashah, but pre-
sumably it exploits the seemingly problematic sequence of verbs in Est 9:27.7
The word ‫ קימו‬was translated above as “they confirmed,” but it also carries the
meanings “they ratified” and “they fulfilled,”8 and in Est 9:27 it precedes the
word ‫קבלו‬, “they took upon themselves.” However, how could the Jews possi-
bly confirm, ratify or fulfill the celebration of Purim before they took it upon
themselves? Surely only after the Jews first assumed the obligation of observ-
ing Purim could they confirm, ratify or fulfill the obligation. Since the verb
order does not seem to fit the local biblical context, Rava extricates the phrase
from the immediate context and interprets it in light of a different context,
concluding that the Jews willingly “confirmed” in the time of Mordechai and
Esther what they had unwillingly “accepted long before” at Mount Sinai. This
understanding of Rava’s derashah was apparently shared by the editor of the
ambient sugya because derashot employing the very same hermeneutic tech-
nique appear in the continuation of the sugya:

‫ באו ששים ריבוא של מלאכי‬,‫ בשעה שהקדימו ישראל נעשה לנשמע‬:‫דרש רבי סימאי‬
‫ אחד כנגד נעשה ואחד כנגד‬,‫ לכל אחד ואחד מישראל קשרו לו שני כתרים‬,‫השרת‬
. . . ‫ ופירקום‬,‫ ירדו מאה ועשרים ריבוא מלאכי חבלה‬,‫ וכיון שחטאו ישראל‬.‫נשמע‬
:‫ בשעה שהקדימו ישראל נעשה לנשמע יצתה בת קול ואמרה להן‬:‫אמר רבי אלעזר‬
‫ ״ברכו ה׳ מלאכיו גברי כח‬:‫מי גילה לבני רז זה שמלאכי השרת משתמשין בו? דכתיב‬
.‫ והדר לשמֹע‬,‫ ברישא עושי‬,)‫עשי דברו לשמע בקול דברֹו״ (תהלים קג כ‬
)‫ מאי דכתיב ״כתפוח בעצי היער וגו׳״ (שיר השירים ב ג‬:‫אמר רבי חמא ברבי חנינא‬
—‫ אף ישראל‬,‫ מה תפוח זה פריו קודם לעליו‬:‫למה נמשלו ישראל לתפוח? לומר לך‬
.‫הקדימו נעשה לנשמע‬

7 See Maharsha ad loc., s.v. qiymu mah sheqiblu kevar. Cf. F. W. Bush, Ruth, Esther (WBC 9;
Dallas 1996) 468.
8 See F. Brown, S. R. Driver and C. A. Briggs, A Hebrew and English Lexicon of the Old Testament
(Oxford 1907) s.v. qum (878–9); M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the
Talmudic and Geonic Periods (Ramat Gan 2002) s.v. qum (996).
150 tropper

,‫ דחזייה לרבא דקא מעיין בשמעתא ויתבה אצבעתא דידיה תותי כרעא‬9‫ההוא מינא‬
‫ עמא פזיזא דקדמיתו פומייכו‬:‫ אמר ליה‬,‫ וקא מבען אצבעתיה דמא‬,‫וקא מייץ בהו‬
—‫ אי מציתו‬,‫ אכתי בפזזותייכו קיימיתו! ברישא איבעיא לכו למשמע‬,‫לאודנייכו‬
‫ כתיב בן ״תומת‬,‫ דסגינן בשלימותא‬,‫ אנן‬:‫ אמר ליה‬.‫ ואי לא—לא קבליתו‬,‫קבליתו‬
‫״וס ֶלף בוגדים‬
ֶ ‫ כתיב בהו‬,‫ אינשי דסגן בעלילותא‬,‫ הנך‬.)‫ישרים ַתנחם״ (משלי יא ג‬
.)‫יש ֵדם״ (שם‬
ָ

Rabbi Simai lectured: When the Israelites gave precedence to “we will do”
over “we will hearken,” six hundred thousand ministering angels came
and set two crowns upon each one of Israel, one as a reward for “we will
do,” and the other as a reward for “we will hearken.” But as soon as Israel
sinned, one million two hundred thousand destroying angels descended
and removed them . . . Rabbi Eleazar said: When the Israelites gave pre-
cedence to “we will do” over “we will hearken,” a Heavenly Voice went
forth and exclaimed to them, Who revealed to My children this secret,
which is employed by the ministering angels, as it is written, “Bless the
Lord, ye angels of His: Ye mighty in strength, that fulfill His word, That
hearken unto the voice of His word” (Ps 103:20): first they fulfill and then
they hearken?
Rabbi Hama bar Rabbi Hanina said: What is meant by “As the apple tree
among the trees of the wood, [so is my beloved among the sons]” (Song
2:3): why were the Israelites compared to an apple tree? To teach you just
as the fruit of the apple tree precedes its leaves, so did the Israelites give
precedence to “we will do” over “we will hearken.”
There was a certain heretic who saw Rava engrossed in his studies while
the fingers of his hand were under his feet, and he ground them down, so
that his fingers spurted blood. You rash people, he exclaimed, who gave
precedence to your mouth over your ears: you still persist in your rash-
ness. First you should have listened; if within your powers, accept; if not,
you should not have accepted. Said he to him: We who walked in integrity,
of us it is written, “The integrity of the upright shall guide them” (Prov
11:3). But of others, who walked in perversity, it is written, “but the per-
verseness of the treacherous shall destroy them” (ibid.) (bShab 88a).

In light of these similar derashot, we may conclude that the editor of our sugya
also understood that Rava’s midrashic exegesis exploits the puzzling sequence
of verbs in Est 9:27, reinterpreting the passage to mean that even though the
Jews had accepted the Torah on Mount Sinai under duress, they freely and will-
ingly re-accepted it in the time of Mordechai and Esther. In this way, Rava sets

9 The term “Sadducee,” which appears here in the Vilna edition, is a product of the Christian
censor’s influence. See Rabbinovicz, Variae Lectiones, 186.
the reception of the torah according to bShab 88a 151

forth the radical claim that the Torah became legally binding only in the time
of Ahasuerus, Esther and Mordechai, long after the reception of the Torah on
Mount Sinai. Though Rava agrees with Rav Aha bar Jacob that the reception
of the Torah on Mount Sinai had no constitutional legitimacy, since the Jews
of the time were under severe duress, Rava nonetheless concludes that con-
temporary Jews have no excuse for neglecting the Torah’s precepts since their
ancestors in the time of Mordechai and Esther freely and willingly accepted
it “upon themselves and upon their descendants” (Est 9:27). By concluding
this short discussion with Rava’s position, the Talmud maintains that the true
Sinaitic moment, the actual reception of the Torah’s legal authority, took place
when the Jews followed their exiled leadership and willingly accepted the
Torah in the time of Mordechai and Esther.
Let us turn now to the literary backdrop for each of the three segments in our
short talmudic exchange in order to see how each segment reworked already
extant literary materials. The central idea of the first segment, the notion that
Mount Sinai was uprooted and raised over the Jews, appears already in tan-
naitic literature, in the Mekhilta de-Rabbi Ishmael:

‫ מלמד שהיו ישראל מתיראין מפני הזיקין מפני‬.‫ נצפפו‬:)‫״ויתיצבו״ (שמות יט יז‬
‫ ״בתחתית ההר״ (שם) מלמד שנתלש‬.‫הזועות מפני הרעמים מפני הברקים הבאים‬
‫ וקרבו ועמדו תחת ההר שנאמר ״ותקרבון ותעמדון תחת ההר״ (דברים‬,‫ההר ממקומו‬
‫ עליהם מפורש בקבלה ״יונתי בחגוי הסלע בסתר המדרגה הראיני את מראיך‬.)‫ד יא‬
‫השמיעיני את קולך כי קולך ערב ומראיך נאוה״ (שיר השירים ב יד) (מדר״י יתרו‬
.)‫דבחודש ג‬

“And they stood” (Exod 19:17): They were huddled together. This teaches
that they were afraid of the winds, of the earthquakes, of the thunders
and the lightnings that came. “At the bottom of [or under] the mount”
(ibid.): Scripture indicates that the mount was pulled up from its place
and the people came near and stood under it, as it is said: “And ye
came near and stood under the mountain” (Deut 4:11). Of them it is
declared in the traditional sacred writings: “O my dove, in the cranny of
the rocks, hidden by the cliff, let me see your face, let me hear your voice;
for your voice is sweet and your face is comely” (Song 2:14) (MdRY Yitro,
ba-hodesh 3).10

10 The translation is that of J. Z. Lauterbach, Mekhilta de-Rabbi Ishmael (Philadelphia 1933–

5), ad loc, only slightly modified. The Mekhilta de Rabbi Shimeon bar Yohai published by
Epstein and Melamed includes a derashah practically identical to that of Rav Avdimi, but
Melamed himself already questioned whether this quotation from the Midrash Hagadol
was an authentic part of the Mekhilta de Rabbi Shimeon bar Yohai (see Mekhilta de-Rabbi
Simeon Bar Yohai [ed. J. N. Epstein and E. Z. Melamed; Jerusalem 1955] 143). Since the
152 tropper

Just like the amora Rav Avdimi in our Babylonian sugya, the Mekhilta also
introduces the image of an uprooted Mount Sinai hanging over the Israelites in
a derashah on the biblical phrase “And they stood at the bottom of the mount”
(Exod 19:17). However, three differences between the parallel texts reveal how
Rav Avdimi has significantly revised the tannaitic derashah. First, Rav Avdimi
and the Mekhilta apparently enlist different hermeneutic techniques. Whereas
Rav Avdimi interprets the preposition bet of the word ‫ בתחתית‬as “in” rather
than as “at,” rendering the biblical phrase “And they stood at the bottom of the
mount” as “And they stood in the bottom of the mount,” the Mekhilta interprets
the word ‫ תחתית‬in a highly literal vein, rendering the phrase “And they stood
at the bottom of (= below) the mount” as “And they stood under the mount.”
This understanding of the Mekhilta’s hermeneutic technique is confirmed by
the corroborating passage it cites, ‫וַ ִּת ְק ְרבּון וַ ַּת ַע ְמדּון ַּת ַחת ָה ָה ְר‬, “And ye came near
and stood under the mountain,” since the word ‫תחת‬, “under,” appears there
without the preposition bet.
Second, Rav Avdimi introduces the image of the life-threatening overturned
cask which does not appear in the Mekhilta. Indeed, save for Rav Avdimi’s
derashah, this image appears only one other time in all of early rabbinic litera-
ture, in an amoraic meimra cited on bSanh 77a:

. . . ‫ כפה עליו גיגית ופרע עליו מעזיבה‬:‫איתמר‬

It has been stated: If one overturned a cask upon a man [who then died
of suffocation], or broke open a ceiling above him . . .11

The use of the overturned cask image in the Bavli alone suggests that the lan-
guage and image of the inverted cask is part of the Babylonian recasting of the
Mekhilta’s idea.
Third, the uprooted mountain plays opposite roles in the two derashot. In
Rav Avdimi’s derashah, God threatens the Israelites by inverting the mountain
upon them and confronting them with the ultimatum: “If you accept the Torah,
it is well; and if not, there shall be your burial.” For Rav Avdimi, the uprooted

quotation from the Midrash Hagadol brought by Epstein and Melamed continues with
more material also unattested elsewhere in tannaitic literature and since, as I shall argue
below, the image of the life-threatening inverted cask is Babylonian, we may conclude
that that the Midrash Hagadol parallel was dependent on Rav Avdimi’s derashah from
the Bavli.
11 The translation is that of H. Freedman, The Hebrew–English Edition of the Babylonian
Talmud: Sanhedrin (London 1969) ad loc., slightly revised.
the reception of the torah according to bShab 88a 153

mountain is a terrifying spectacle designed to browbeat the Israelites into

submission. In the Mekhilta, by contrast, the uprooted mountain provides
a secure hiding place for the Israelites who are terrified by the winds, earth-
quakes, thunder and lightning. The Israelites, according to the Mekhilta,
will­ingly come near and stand under the uprooted mountain because they
welcome the security supplied by the large overhanging mass just as the dove
of Song 2:14 seeks security in the cranny of the rock. Hence, the protective over-
hanging mountain of the Mekhilta is transformed in the Bavli into a life-threat-
ening inverted cask. In short, Rav Avdimi’s derashah replaces the Mekhilta’s
hermeneutic technique with an alternative one, introduces the image of the
inverted cask absent from the Mekhilta and transforms the Mekhilta’s comfort-
ing and protective overhanging mountain into a terrifying and life-threatening
massive casket.
The second segment of our talmudic discussion is Rav Aha bar Jacob’s ques-
tion concerning the legality of the Israelites’ compelled acceptance of the
Torah. In light of earlier Palestinian sources, two notable features of his ques-
tion merit attention. First and foremost, the Palestinian source for the image of
the uprooted and overhanging Mount Sinai, that is, the Mekhilta parallel cited
above, does not pose the question raised by this Babylonian amora. Indeed,
since the Mekhilta lacks God’s alarming ultimatum and envisions the over-
hanging mountain as a source of comfort and security, there is no place for
Rav Aha bar Jacob’s constitutional query in the Mekhilta. Moreover, there is no
similar query anywhere in early Palestinian rabbinic literature. Second, the key
term employed by Rav Aha bar Jacob, ‫מודעא‬, “protest,” is a technical term in the
Bavli that never appears in early Palestinian rabbinic literature. According to
Michael Sokoloff’s definition, the word moda‘a denotes “a document of protest
made in advance before witnesses in order to invalidate a transaction or a legal
action to be made under duress.”12 Thus, Rav Aha bar Jacob framed his consti-
tutional question in terms familiar to him from the ambient legal environment
in Jewish Babylonia.
The third and final segment of our talmudic discussion is Rava’s surpris-
ing derashah and it was apparently modeled on a parallel tannaitic derashah
which also extricates the clause ‫הּודים‬ִ ְ‫ ִקּיְ מּו וְ ִק ְּבלּו ַהּי‬, “the Jews confirmed and
took upon themselves,” from its local biblical context in order to link it to the
ancient past. The earliest extant form of this baraita appears in tSotah, and
bShevuot cites the baraita with some editorial modifications and clarifications.

12 Sokoloff, A Dictionary of Jewish Babylonian Aramaic, s.v. mod‘a (pp. 645–6).

154 tropper

‫ב' שבועות לט ע״א‬ )‫ו (על פי כ״י וינה‬-‫ת' סוטה ז ד‬

‫ הוי יודע שלא על‬:‫וכשמשביעין אותו אומרים לו‬ ‫ הוי יודע שלא על תנאי שבלבך אנו‬:‫[ד] אומ׳ לו‬
‫ אלא על דעת המקום ועל‬,‫דעתך אנו משביעין אותך‬ ‫ וכן‬.‫ אלא על תנאי שבלבבנו‬,‫משביעים אותך‬
‫ כשהשביע את‬,‫ שכן מצינו במשה רבינו‬.‫דעת ב״ד‬ ‫ כשהשביע משה את ישראל בערבות‬,‫מצינו‬
‫ דעו שלא על דעתכם אני משביע‬:‫ישראל אמר להן‬ ‫ לא על תנאי שבלבבכם אני‬:‫מואב אמ׳ להם‬
:‫ ועל דעתי שנאמר‬,‫אתכם אלא על דעת המקום‬ ‫ ״ולא‬:‫ שנ׳‬,‫משביע אתכם אלא על תנאי שבלבי‬
.)‫״ולא אתכם לבדכם וגו׳״ (דברים כט יג‬ .)‫אתכם לבדכם וגו׳״ (דברים כט יג‬
‫״כי את אשר ישנו פה״ (דברים כט יד); אין לי אלא‬ )‫״כי את אשר ישנו פה וגו׳״ (דברים כט יד‬
‫ דורות הבאים וגרים‬,‫אותן העומדין על הר סיני‬ ‫ מניין לדורות הבאים‬,‫[ה] אין לו אלא אתם‬
‫ ״ואת אשר איננו״‬:‫העתידין להתגייר מנין? ת״ל‬ ‫ ״ולא‬:‫אחריכם ולגרים שנתוספו עליכם? שנ׳‬
,‫(שם); ואין לי אלא מצוה שקיבלו עליהם מהר סיני‬ ‫אתכם לבדכם״ (שם) אלא ״ואת אשר איננו פה‬
?‫ מנין‬,‫ כגון מקרא מגילה‬,‫מצות העתידות להתחדש‬ ‫ [ו] אין לי אלא מצות שנצטוו‬.)‫עמנו היום״ (שם‬
‫ קיימו מה שקבלו‬,)‫ ״קימו וקבלו״ (אסתר ט כז‬:‫ת״ל‬ ?‫ מניין לרבות מקרא מגלה‬,‫ישראל על הר סיני‬
.‫כבר‬ ‫ ״קיימו וקבלו וגו׳״ ״ולא יעבר״ (אסתר‬:‫ת״ל‬
.)‫ט כז‬

bShevu 39a tSot 7:4–7 (according to MS Vienna)

And when they adjure him [who takes an [4] They say to him [who takes an oath]:
oath], they say to him: Know that we do not Know that we do not adjure you accord-
adjure you according to your own mind, but ing to the condition in your heart but
according to the mind of the Omnipresent, according to the condition in our heart.
and the mind of the Beth Din; for thus we And thus we find that when Moses
find in the case of Moses our teacher: When adjured Israel in the Plains of Moab he
he adjured Israel, he said to them: Know said to them: Not on the condition in
that not according to your own minds do your heart do I adjure you, but on the
I adjure you, but according to the mind of condition in my heart, as it is said: “Not
the Omnipresent, and my mind: as it is said: with you only” etc. (Deut 29:13).
“Not with you only” etc. (Deut 29:13).
“But with him that stands here with us” (ibid. “But with him that stands here with us”
14): hence I know only those who were stand- (ibid. 14): [5] hence he only (adjured)
ing by Mount Sinai [were adjured]; the com- you [who were standing in the Plains
ing generations and proselytes who were of Moab]; how do we know [that Moses
later to be proselytized, how do we know also adjured] the coming generations
[that they were adjured also then]? Because and the proselytes that will join you?
it is said: “and also with him that is not here Because it is said: “Not with you only”
with us this day” (ibid.). And from this we but “and also with him that is not here
know only [that they were adjured for] the with us this day” (ibid.). [6] From this
commandments which they received at we know only [that they were adjured
Mount Sinai; how do we know [that they for] the commandments which they
the reception of the torah according to bShab 88a 155

were adjured for] the commandments received at Mount Sinai; how do we

which were to be promulgated later, such know [that they were also adjured for]
as reading the Megillah? Because it is said: reading the Megillah? Because it is said:
“[The Jews] confirmed and irrevocably “[The Jews] confirmed and irrevocably
accepted” (Est 9:27): accepted” (Est 9:27).
They confirmed what they had accepted
long before.

The baraita opens with the procedural rule that when administering an oath,
the court informs the oath-taker that his oath is not conditioned or limited by
any of his own unstated intentions but is conditioned by the unstated intentions
of the judges. The rest of the baraita then illustrates this point with derashot
on Deut 29:13–4 and, more importantly for our purposes, with a derashah on
Est 9:27. In response to the question how the Israelites in the time of Moses
could have obligated their descen­dants to fulfill the precept of reading Esther
when they contracted the covenant on the Plains of Moab according to tSot or
when they received the Torah at Mount Sinai according to bShevu, the bara-
ita concludes that the Jews “confirmed” in the time of Mordechai and Esther
“what they had accepted long before” in the time of Moses. In other words, the
Israelites’ commitment to observe the Torah in the time of Moses, made on the
Plains of Moab or at Mount Sinai, was predicated on the unstated intentions
in Moses’s heart13 and since Moses knew that reading Esther on Purim would
become a precept in the future, the Israelites’ commitment included the read-
ing of Esther as well.14
Like Rava’s derashah, the baraita’s derashah also exploits the puzzling verb
order in the phrase “the Jews confirmed and irrevocably accepted,” question-
ing how the Jews could “confirm” (“ratify” or “fulfill”) the celebration of Purim
before they “accepted” it “upon themselves.” Since confirmation (ratification or
fulfillment) surely follows acceptance, the verb order does not seem to fit the
local context.15 Consequently, the baraita introduced an alternative context

13 MS Erfurt of tSot 7:4 has God adjure the Jews.

14 See also yMeg 1:4, 70d; bMeg 19b.
15 The Tosefta does not divulge the midrashic technique underlying its conclusion and Saul
Lieberman raised the possibility that the Tosefta’s derashah does not address the puzzling
verb order in Est 9:27 but rather an apparent redundancy in the verse. Since ‫ קימו‬can
also be translated as “they took an oath” (see Sokoloff, Dictionary of Jewish Babylonian
Aramaic, s.v. qum [p. 996]), Est 9:27 easily could have read “the Jews took an oath” or
“the Jews took upon themselves” and in both cases the meaning of the verse would have
156 tropper

and taught that the Jews actually assumed the obligation to read Esther in the
time of Moses and only confirmed this obligation in the time of Mordechai
and Esther.
Despite the noted similarities between the two derashot, a comparison
of Rava’s derashah to the baraita’s derashah reveals that Rava has actually
turned the earlier derashah on its head. Whereas the baraita argues that the
legal obligation to celebrate Purim was already created in the time of Moses,
Rava argues that the legal obligation to observe the entire Torah was only cre-
ated in the time of Mordechai and Esther! Thus, though both the baraita and
Rava employ the same language and enlist the same midrashic technique to
interpret the very same biblical passage, their conclusions are diametrically
opposed to one another. While the baraita argues that already in the time of
Moses the Jews assumed the legal obligation to read Esther and, by extension,
to fulfill all post-Pentateuchal precepts, Rava argues that the Jews only truly
assumed the legal obligation to observe the Pentateuch’s precepts in the time
of Mordechai and Esther. In my opinion, Rava’s radical stance becomes even
more shocking when we realize that he formed his derashah on the basis of an
earlier derashah whose language and logic he borrowed but whose conclusion
he reversed entirely.
In sum, the three segments of our short talmudic discussion extensively
modified and reworked their sources. The first segment transformed the pro-
tective overhanging mountain of the Mekhilta into a life-threatening casket.
The second segment drew out the constitutional ramifications of the first seg-
ment with the help of a technical legal term familiar to Babylonian amoraim
and the third segment flipped an extant derashah on its head in the formula-
tion of an innovative, surprising and counter-intuitive claim. Thus, a discus-
sion with an unprecedented and radical conclusion was formed in a process
wherein earlier materials underwent extensive and intensive revision.

remained (largely) unchanged. The seemingly extraneous verb, Lieberman suggested,

prompted the Tosefta to teach that “the Jews took an oath” on the Plains of Moab to cel-
ebrate Purim “and took upon themselves” in the time of Mordecai and Esther to fulfill
that oath. Lieberman, however, rejected his own suggestion and the Bavli’s gloss (“they
confirmed what they had accepted long before”), which reverses the chronological order
of the verbs, indicates that the Bavli is troubled by the word order and not by an apparent
redundancy (otherwise it would not have had to reverse the chronological sequence of
the verbs. See S. Lieberman, Tosefta ki-fshutah, 8 [New York, 1973] 678.)
the reception of the torah according to bShab 88a 157

By way of conclusion, let us note that perhaps it is no coincidence that Rava’s

striking statement, which dates the authentic Sinaitic moment to a time when
the Jewish leadership lived in Persia, is attributed to a Babylonian amora and
appears in the Bavli only. Indeed, perhaps this short talmudic discussion not
only attests to the literary creativity and extensive recasting of earlier materials
typical of Babylonian rabbinic sages, but also embodies the local patriotism of
Babylonian rabbis who sought to transfer the authentic reception of the Torah
to the Persian sphere in which they flourished.
Heaven and Hell: Babylonia and the Land of Israel
in the Bavli

Tal Ilan

The Land of Babylon takes up an important historical and theological position

within the Hebrew Bible. Historically, it is the land of Nebu­chadnezzar, the
wicked king, who destroys Jerusalem and exiles the Judeans (see e.g. 2Kgs 25:1–
21). Theologically, this king is the tool in the hand of God with which to punish
his people for disobeying him (Jer 27). Theologically, the Babylonian captivity
ended with the fall of Babylon in the hands of the Persians and their king,
Cyrus, whom Deutero-Isaiah labels Messiah (Isa 45:1–7), and with his decree,
allowing and urging the Judeans to return to Jerusalem and rebuild the Temple
(Ezra 1:1–6). This king too is a tool in the hand of God—a tool signifying his
forgiveness. Historically, of course, some Jews took the opportunity to return
to the Land of Israel (e.g. Ezra 2), but the vast majority probably did not, essen-
tially creating the existential reality of Judaism from then on—namely, that of
a dispersed nation. The creation of two commentaries on the Mishnah, one in
Babylonia and one in the Land of Israel, is a logical conclusion of this historical
development. Quite a different question is how Jewish sages in these two lands
handled the theological question of Babylonia in the ancient Jewish Scriptures
and in their present day. It is the working hypothesis of the present paper that
people still living in Babylonia as a result of the Babylonian exile could not
treat that land in the same disinterested, academic manner in which their
compatriots in the Land of Israel did.1 I will examine this issue with the help of
several discussions that appear in Palestinian and in Babylonian sources refer-
ring to the same tannaitic texts, with the hope of demonstrating how these
amoraic centers tackled differently the thorny questions of Babylonian wick-
edness and perfidiousness.
Let us begin with a mishnaic text. In mBer 9:1 we read: “Whoever observes
a place in which miracles were performed for our forefathers says: Blessed be
He who performed miracles for our forefathers in this place.” Both talmudim
use this passage as a pretext for the discussion of a large number of miracles

1 And see on this primarily: I. Gafni, “Expression and Types of ‘Local Patriotism’ among the
Jews of Sasanian Babylonia,” Irano-Judaica 2 (1990) 63–71; idem, “How Babylonia Became
‘Zion’: Shifting Identities in Late Antiquity,” in: L. I. Levine and D. R. Schwartz (eds.), Jewish
Identities in Antiquity: Studies in Memory of Menahem Stern (Tübingen 2009) 333–48.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�07

babylonia and the land of israel in the bavli 159

that, when observed, should be blessed, and a large number of places where
blessing should be pronounced. The Yerushalmi discussion includes Babylonia
in this instance:

Whoever sees Babylonia should bless five blessings:

1. Seeing the Euphrates one should say: Blessed is He who created the
beginning.
2. Seeing Markulis one says: Blessed be He who is patient.
3. Seeing the House of Nebuchadnezzar, one says: Blessed is He who
destroyed the house of that wicked [man].
4. Whoever sees the fiery furnace [into which Daniel’s companions
were thrown] and the lions’ den [into which Daniel himself was
hurled], says: Blessed is He who performed miracles for our forefa-
thers in this place.
5. Seeing a place from which earth is removed, one says: Blessed is He
who says and does, blessed be He who decrees and fulfills.
6. Seeing Babylon, one says: “And I shall sweep it with the broom of
destruction” (Isa 14:23) ( yBer 9:1, 12d).

Let us first note what the Yerushalmi is doing here. It begins by stating that
Babylon merits five blessings. The third and fourth blessings refer the reader
directly to biblical events where Babylonians had inflicted suffering on Jews
and were duly punished for it. Their punishment should be seen as the mira-
cles which should, according to Mishnah Berakhot, be blessed: These are the
ruined house of the wicked Nebuchadnezzar, and the places where Daniel and
his companions suffered persecution in Babylonia (cf. Dan 3:19–27; 6:17–25).2
According to this tradition, these places are still visible. The second and fifth
blessings are less conspicuously biblical, or Babylonian. The Mishnah describes
Markulis as some form of idolatry represented by heaps of stones (mAZ 4:1)
in Palestine, and this is usually associated with some form of worship of the
Greek god Hermes, whose Latin name was Mercury, which could have been
(intentionally or unintentionally) distorted.3 Yet here, the Yerushalmi ties the

2 On the realia behind these terms see A. Oppenheimer, Babylonia Judaica in the Talmudic
Period (Wiesbaden 1983) 52–60. On Babylon in ruins see specifically pp. 53–54. On visits of
Jews to the location where the lions’ den of Daniel was located, according to Muslim sources,
see p. 59, n. 41.
3 On Markulis see I. Pintel-Ginsberg, “ ‘Throwing a Stone at Markulis’: Symbolizing the ‘Other’
in a Jewish Cultural Context,” in: J. Dan (ed.), Gershom Scholem (1897–1982): In Memoriam,
vol. 2 (Jerusalem 2007) 455–68 [Hebrew].
160 ilan

Markulis directly to Babylonia. Also, the Yerushalmi understands the removal

of earth (apparently in Babylonia) as another indication of God’s will being
fulfilled on earth. The first and last blessings in this list merit a different judg-
ment. First, it should be noted that although the tradition begins by pronounc-
ing that five blessings should be recited over Babylon, six are enumerated.
One could, however, argue that the last one, a general blessing pronounced
on (apparently the ruins of) Babylon as such, relates directly to the introduc-
tory phrase and forms an inclusio. Secondly, one may argue that it is not really
formulated as a blessing, but is rather a citation of a biblical verse, taken from
Isaiah’s destruction prophecy pronounced against Babylon. Thus, it could be
seen as a general verse, applicable to all specific instances mentioned in the
list above.
However, if this is true, we are confronted with a major problem regarding
the first blessing in the list. If we see the entire list as a celebration of God’s
destruction of Babylon, what are we to do with the first clause, which instructs
a person, when observing the Euphrates River, to utter a blessing where God’s
creative (rather than destructive) powers are praised? The city of Babylon is
located on the Euphrates River, mentioned in the creation narrative as one of
the four water sources for the Garden of Eden (Gen 2:14). Praising God for his
works of creation certainly does not fall within the scope of praising him for
his destructive power as exemplified in his treatment of Babylon.
Obviously the Babylonian rabbis were confronted with the same problem.4
Let us see how they resolved these difficulties:

Rav Hamnuna expounded: He who sees wicked Babylon should bless five
blessings:
1. Seeing Babylon, one says: Blessed is He who destroyed wicked Baby-
lon.
2. Seeing the House of Nebuchadnezzar, one says: Blessed is He who
destroyed the house of wicked Nebuchadnezzar.

4 This tradition, according to both the Yerushalmi and the Bavli, is not a baraita. It is, how-
ever, obvious that the Babylonian rabbis have taken it over from the Yerushalmi, or from
a Yerushalmi-like composition. And see also T. Ilan, “‘Stolen Water is Sweet’: Women and
their Stories between Bavli and Yerushalmi,” in: P. Schäfer (ed.), The Talmud Yerushalmi and
Greco-Roman Culture, vol. 3 (Tübingen, 2002) 185–223, esp. p. 186, n. 3. On this concept see
more recently A. M. Gray, A Talmud in Exile: The Influences of Yerushalmi Avodah Zarah on the
Formation of Bavli Avodah Zarah (Providence RI 2005) 1–33.
babylonia and the land of israel in the bavli 161

3. Whoever sees the lions’ den [into which Daniel was hurled] and the
fiery furnace [into which his companions were thrown], says:
Blessed is He who performed miracles for our forefathers in this
place.
4. Seeing Markulis one says: Blessed is He who gave patience to those
who transgress his will.
5. Seeing a place from which earth is removed, one says: Blessed is He
who says and does, decrees and fulfills (bBer 57b).

The tradition is followed by two short anecdotes that are attached to the last
blessing and demonstrate its application:

When Rava saw donkeys transferring earth, he would slap their backs
with his hand and say: Run, you righteous beings, to perform the com-
mandments of your master.

The commandment intended here is obviously the removal of earth. This story
is followed by another one:

Mar bar Ravina, when he would come to Babylonia, he would pick up

earth in his clothing and throw it out in order to fulfill the words of the
verse: “And I shall sweep it with the broom of destruction” (Isa 14:23).

At the very end Rav Ashi expresses an opinion on this list of curses one should
voice when confronted by Babylon:

I do not know this tradition assigned to Rav Hamnuna (he says). I uttered
all these as blessings of my own accord.

I shall now present the two traditions synoptically, side-by-side:

‫ב׳ ברכות נז ע״ב‬ ‫ יב ע״ד‬,‫י׳ ברכות ט א‬

:‫דרש רב המנונא‬
:‫הרואה בבל הרשעה צריך לברך חמש ברכות‬ :‫הרואה בבל צריך לברך חמש ברכות‬
.‫ ברוך שהחריב בבל הרשעה‬:‫ראה בבל אומר‬ .‫ ברוך עושה בראשית‬:‫ אומר‬,‫ראה פרת‬
‫ ברוך שהחריב‬:‫ראה ביתו של נבוכדנצר אומר‬ .‫ ברוך ארך אפים‬:‫ אומר‬,‫ראה מרקוליס‬
.‫ביתו של נבוכדנצר הרשע‬
‫ ברוך‬:‫ראה גוב של אריות או כבשן האש אומר‬ ‫ ברוך שהחריב‬:‫ אומר‬,‫ראה ביתו של נבוכדנצר‬
‫שעשה נסים לאבותינו במקום הזה‬ .‫ביתו של אותו רשע‬
162 ilan

‫ ברוך שנתן ארך אפים לעו־‬:‫ראה מרקוליס אומר‬ ‫ ברוך‬:‫ אומר‬,‫ראה מקו׳ כבשן האש וגוב אריות‬
.‫ברי רצונו‬ .‫שעשה ניסים לאבתינו במקום הזה‬
‫ ברוך אומר‬:‫ראה מקום שנוטלין ממנו עפר אומר‬ ‫ ברוך‬:‫ אומר‬,‫ראה מקום שנוטלין ממנו עפר‬
.‫ועושה גוזר ומקיים‬ .‫ ברוך גוזר ומקיים‬,‫אומר ועושה‬
‫ טריף‬,‫רבא כי הוה חזי חמרי דשקלי עפרא‬ ‫ ״וטאטאתיה במטאטא השמד״‬:‫ אומר‬,‫ראה בבל‬
‫ רהוטו צדיקי למעבד‬:‫להו ידא על גבייהו ואמר‬ .)‫(ישעיהו יד כג‬
.‫רעותא דמרייכו‬
‫ הוה שקיל‬,‫מר בריה דרבינא כי הוה מטי לבבל‬
:‫ לקיים מה שנאמר‬,‫עפרא בסודריה ושדי לברא‬
.)‫״וטאטאתיה במטאטא השמד״ (ישעיהו יד כג‬
,‫ אנא הא דרב המנונא לא שמיע לי‬:‫אמר רב אשי‬
.‫אלא מדעתאי בריכתינהו לכולהו‬

bBer 57b yBer 9:1, 12d

Rav Hamnuna expounded:
He who sees wicked Babylon should bless Whoever sees Babylon should bless five
five blessings: blessings:
Seeing Babylon, one says: Blessed be He Seeing the Euphrates one should say:
who destroyed wicked Babylon. Blessed is He who created the beginning
(‫)בראשית‬.
Seeing the House of Nebuchadnezzar, one
says: Blessed be He who destroyed the
house of wicked Nebuchadnezzar.
Whoever sees the lions’ den and fiery fur- Seeing Markulis one says: Blessed be He
nace, says: Blessed be He who performed who is patient.
miracles for our forefathers in this place.
Seeing Markulis one says: Blessed be He Seeing the House of Nebuchadnezzar,
who gave patience to those who trans- one says: Blessed be He who destroyed
gress his will. the house of that wicked [man].
Seeing a place from which earth is Whoever sees the fiery furnace and the
removed, one says: Blessed be He who lions’ den, says: Blessed be He who per-
says and does, decrees and fulfills. formed miracles for our forefathers in
this place.
Whenever Rava saw donkeys transferring Seeing a place from which earth is
earth, he would slap their backs with his removed, one says: Blessed be He who
hand and say: Run, you righteous beings, says and does, blessed be He who decrees
to perform the commandments of your and fulfills.
master.
babylonia and the land of israel in the bavli 163

Mar bar Ravina, when he would come Seeing Babylon, one says: “And I shall
to Babylon, he would pick up earth in sweep it with the broom of destruction”
his clothing and throw it out in order to (Isa 14:23).
fulfill the words of the verse: “And I shall
sweep it with the broom of destruction”
(Isa 14:23).
Rav Ashi said: I do not know this tradition
assigned to Rav Hamnuna. I uttered all
these as blessings of my own accord.

When looking at these two traditions one next to the other we immediately
note three glaring differences between them. First, in the Babylonian tradi-
tion the introductory phrase refers specifically to wicked Babylon and not to
Babylon in general. This may suggest that there is also another, less wicked
Babylon. Second, the list here really does only include five clauses—that is,
one of the clauses from the Yerushalmi list is missing. This, however, is not
the last one, which was, as I suggested, an inclusio and should not have been
counted among the five. The blessing on the destruction of Babylon, as such,
is found in the Bavli text—significantly, first in the list—but following the for-
mal blessing pattern, with no biblical text attached. The blessing that is absent
in the list is the one where God is praised for creating the Euphrates River.
Obviously the Babylonians did not think it right to include a blessing for God’s
works of creation among blessings of his works of destruction.
The Babylonians also do not forget the biblical verse on the destruction of
Babylon. Yet, instead of attaching it to one of the blessings, it is attached to
the second of the two anecdotes describing the removal of earth. What the
Yerushalmi had intended by blessing the place from which earth is removed is
not spelt out. We can, perhaps, imagine that the description entails a mound
of ruins, a tel, where a city had once stood, and to which people come, collect
and carry away remains, such as ceramic bricks, in order to recycle them in
their new building activities, perhaps in Bursif, which the Babylonians identify
as in direct vicinity to the ruined Babylon (e.g. bShab 36a), although this is not
specifically stated. If my interpretation is correct, this blessing is intended to
describe the complete ruin of the city. This blessing, followed in the Yerushalmi
by the verse from Isaiah applied to the whole of Babylon, may confirm my sus-
picion. The Babylonians, however, treat this entire pericope differently. They
describe the removal of earth observed by the Babylonian Rava as something
banal, an agricultural activity performed on donkeys’ backs and as a symbolic
act performed by Mar bar Ravina. This is where they incorporate the verse
from Isaiah, applying it to this activity only, rather than to the destruction of
all Babylon, as we saw in the Yerushalmi.
164 ilan

Finally, it should also be noted that at the outset, this entire tradition in
the Bavli is assigned to Rav Hamnuna—a Babylonian sage. This is unexpected,
since we know it to be Palestinian, and since it is so clearly in Hebrew. The final
sentence associated in the Bavli with this tradition may explain this unusual
assignation. Here we find the prominent Babylonian Rav Ashi exclaiming that
he does not know of any tradition which expects one to curse Babylonia. He
claims he has his own blessings to utter on seeing these and similar places.
Ultimately we find here a Babylonian rebellion against this hostile Palestinian
tradition.
Of course, the major question we should be asking here is what have the
Babylonians done with the one positive blessing found in the Yerushalmi, the
one referring to the Euphrates River as exemplifying God’s works of creation.
This is not difficult to discover. The next Mishnah in chapter 9 of Berakhot,
the one coming after that which instructs a person to bless miracles, encour-
ages Jews to bless the wonders of nature. With reference to these, the Mishnah
singles out “the mountains and the hills, the seas and the deserts” as places
about which “one says: Blessed be He who created in the beginning (‫עושה‬
‫( ”)מעשה בראשית‬mBer 9:2). Both talmudim comment on this Mishnah too.
Both do not present an orderly list of various phenomena which should be
viewed as the works of “the beginning,” namely, works of creation, but rather
here and there the expression is repeated in the discussions. I present here
these points very briefly, and note only the marked differences between the
Yerushalmi and the Bavli.

‫ע״ב‬-‫ב' ברכות נט ע״א‬ ‫ד‬-‫ יג ע״ג‬,‫י' ברכות ט ב‬

‫ התם מברך‬:‫ רבא אמר‬. . . ‫על ההרים ועל הגבעות‬ ‫על ההרים ועל הגבעות ועל הימים ועל הנהרות‬
‫ ועושה מעשה‬,‫ ברוך שכחו מלא עולם‬:‫תרתי‬ .‫ ברוך עושה בראשית‬:‫ אומר‬,‫ועל המדברות‬
. . . ‫בראשית‬
:‫ אומר‬,‫ הרואה את הים הגדול‬:‫רבי יהודה אומר‬
,‫ בזמן שרואהו לפרקים‬.‫ברוך שברא את הים‬
.‫ ברוך עושה בראשית‬:‫אומר‬
.‫על הרוחות מהו אומר? ברוך שכחו מלא עולם‬
,‫ אבל שבאים בנחת‬,‫מתניתא בשבאין בזעף‬
.‫ ברוך עושה בראשית‬:‫אומר‬
,‫ הרואה רקיע בטהרתה‬:‫אמר רבי יהושע בן לוי‬ ‫ ואת הלבנה‬,‫הרואה את החמה בתקופתה‬
.‫ ברוך עושה בראשית‬:‫אומר‬ ‫ ברוך‬:‫ אומר‬,‫ ואת הרקיע בטיהרו‬,‫בתקופתה‬
‫ הרואה חמה בתקופתה לבנה בגבורתה‬:‫תנו רבנן‬ .‫עושה בראשית‬
‫ ברוך‬:‫ אומר‬,‫וכוכבים במסילותם ומזלות כסדרן‬
. . . ‫עושה בראשית‬
babylonia and the land of israel in the bavli 165

‫ אמר רמי‬. . . ‫ הרואה הים וכו׳‬:‫רבי יהודה אומר‬

‫ אגשרא‬,‫ הרואה פרת‬:‫בר אבא אמר רב יצחק‬
‫ ואמר רמי‬. . . ‫ ברוך עושה בראשית‬:‫ אומר‬,‫דבבל‬
,‫ אגשרא דשביסתנא‬,‫ הרואה דגלת‬:‫בר אבא‬
.‫ ברוך עושה בראשית‬:‫אומר‬

bBer 59a–b yBer 9:2, 13c–d

On the mountains and the hills  . . . Rava On the mountains and the hills, the
says: There one blesses two (blessings): seas and the rivers and the deserts one
Blessed be He whose might fills the whole says: Blessed be He who created in the
world and who created in the beginning. beginning.
Rabbi Yehudah says: He who sees the
great sea says: Blessed be He who created
the sea. If one sees it occasionally one
says: Blessed be He who created in the
beginning.
When one sees the winds, what does
one say: Blessed be He whose might fills
the whole world. The Mishnah speaks of
when they come in rage, but when they
come in calm one says: Blessed be He
who created in the beginning.
Said Rabbi Joshua ben Levi: He who sees He who sees the sun in its period, and the
the firmament in its purity says: Blessed moon in its period, and the firmament in
be He who created in the beginning. its purity says: Blessed be He who created
in the beginning.
Our Rabbis taught: He who sees the sun in
its period, the moon in its virility and the
stars in their route and the Zodiac signs in
their order one says: Blessed be He who
created in the beginning . . .
Rabbi Yehudah says: He who sees the
sea etc. . . . said Rami bar Abba, said Rav
Yitshaq: He who sees the Euphrates on the
Bridge of Babylon says: Blessed be He who
created in the beginning . . . said Rami bar
Abba Rav Yitshaq: He who sees the Tigris
on the Bridge of Bistana says: Blessed be He
who created in the beginning . . .
166 ilan

In addition to the Mishnah’s list of natural phenomena, which should be

blessed as works of creation, for which the Yerushalmi presents an improved
text, we are also informed by it that one should bless the works of creation for
calm winds and for the heavenly bodies together with the firmament (‫)רקיע‬.
In passing it should be noted that this blessing is an authentic baraita, also
found in tBer 6:6, which, however, does not include the firmament, but only
the sun, the moon and the stars.
The Bavli too decrees blessing the works of creation for the mountains
and the hills, for the firmament and the heavenly bodies. Note, however, that,
in accordance with the Tosefta version, the Bavli separates between the heav-
enly bodies and the firmament. Also in the Bavli we find no blessing for calm
winds. Instead, the Babylonians apparently take note of the rivers, mentioned
in the mishnaic text next to the sea, for it is here that they insert the text
on the Euphrates, which they removed from the list of curses on Babylonia.
And next to the Euphrates, they also do not neglect blessing the Tigris River
(‫דגלת‬/‫)חדקל‬, also mentioned in Gen 2:14 as one of the rivers that water the
Garden of Eden. When doing this, the Bavli is also very exact geographically.
One blesses the Euphrates on a bridge in the city of Babylon, and the Tigris on
another specific bridge known to them. The Bavli here has clearly differenti-
ated between the biblical stories that curse Babylon, and the ones that bless it.
For the Babylonians, as opposed to the sages of the Land of Israel, Babylonia
is positively assessed. It is important for them to place its attributes among
the natural features that dominated the earth before the advent of mankind,
among the perfect subjects of God’s creation.
I will not elaborate here on the extent to which this topic interested the
Babylonians.5 Let me just mention here that the Euphrates River, as a site not
of the cursed Babylon, but of the blessed Garden of Eden, is a favorite topic
of the complex, dialectic way in which the Babylonians expound biblical
verses.6 Thus, for example, the geographical reality, which was exemplified in
the exact, familiar location where one is expected to bless the Euphrates River,

5 And see further Gafni.

6 On an actual geographic location for the Garden of Eden see K. A. Kitchen, “Wither Eden?
Recent Attempts to Visit the Garden,” Buried History 34 (1998) 101–3. See also D. Neiman,
“Gihon and Pishon: Mythological Antecedents to the Two Enigmatic Rivers of Eden,”
Proceedings of the Sixth World Congress of Jewish Studies 1 (Jerusalem 1977) 321–8. The issue
of the rabbinic perception of Eden, and its geographical location has not been a favorite
topic of recent scholarship. In the collection of articles edited by G. L. Luttikhuizen, Paradise
Interpreted: Representations of Biblical Paradise in Judaism and Christianity (Leiden 1999) rab-
binic attitudes to the issue are only marginally investigated, see W. J. van Bekkum, “Paradise
as Paradigm: Good and Evil in Rabbinica and Kabbalah,” 116–27; and in K. Schmid and
babylonia and the land of israel in the bavli 167

is repeated in the Babylonian interest in the exact location of the Garden of

Eden. For the Babylonians this is not an abstract question. Thus, in one tradi-
tion, which is found only in the Bavli, we are informed that the location of the
entrance to the Garden of Eden is disputed.

—‫ ואם בערביא‬,‫ אם בארץ ישראל הוא—בית שאן פתחו‬:‫ אמר ריש לקיש‬,‫גן עדן‬
‫ אביי משתבח בפירי‬.‫ בבבל‬,‫ ואם בין הנהרות הוא—דומסקנין פתחו‬,‫בית גרם פתחו‬
.)‫ רבא משתבח בפירי דהרפניא (ב׳ עירובין יט ע״א‬,‫דמעבר ימינא‬

The Garden of Eden, said Resh Laqish: If it is in the Land of Israel, Beit
Shean is its entrance, and if in Arabia—Beit Garem is its enterance and
if it is between the rivers—Dumascnin is its entrance, in Babylonia.
Abbayye recommends [in this context] the fruits of the Passage of Yamina
and Rava recommends [in the same context] the fruits of Harpania
(bEruv 19a).

A rabbi from the Land of Israel—Resh Laqish—is here reputed to have sug-
gested three possible locations for the entrance to the Garden of Eden—in
the Land of Israel, in Beit Shean,7 in Arabia (in an unknown place called Beit
Garem)8 and between the rivers. This last location obviously refers to the area
of Mesopotamia—greater Babylonia for the rabbis. Interestingly, this entrance
is identified by the same Resh Laqish as in a place called Dumascnin, usually
translated as Damascus, a city in Syria that lies well west of the Euphrates.
However, we should note that, on the one hand, the authors of the Bavli were
well-acquainted with Damascus and its geography, and on the other hand that
the spelling suggested here for this place between the rivers through which
one enters the Garden of Eden is singular (namely, it appears nowhere else
in rabbinic literature) but is repeated in exactly this form in all manuscripts
of the Bavli for bEruv 19a. This indicates, in my opinion, that this is not a cor-
rupted reading of Damascus, but rather that the rabbis have here preserved the

C. Riedweg (eds.), Beyond Eden: The Biblical Story of Paradise (Genesis 2–3) and its Reception
History (Tübingen 2008) not at all.
7 This is a singular tradition. On an attempt to connect it with rabbinic traditions about the fer-
tility of Beth Shean see M. Avi-Yonah, “Scythopolis,” The Beth Shean Valley (Jerusalem 1962)
58 [Hebrew].
8 On attempts to identify this place in such locations as Yemen or Trans­jordan see G. Reeg, Die
Ortnamen Israels nach der rabbinischen Literatur (Wiesbaden 1989) 97.
168 ilan

correct name of an otherwise unknown place in Mesopotamia.9 We have no

way of knowing if this is an authentic Palestinian tradition, because it has no
parallel in Palestinian sources, but it is interesting to note that it is immediately
followed by the opinions of the two leading Babylonian rabbis—Abbayye and
Rava, about the location of the entrance to the Garden of Eden in Babylonia.
The location is to be identified, according to these rabbis, based on the high
quality of the fruit one can purchase in these locations, either in the Passage of
Yamina or in Harpania. Perhaps these rabbis too were not quite decided about
where Dumascnin was, but they were certain it was to be located in Babylonia.
Perhaps it is no surprise that this tradition, about the precise geographi-
cal location of the Garden of Eden, follows closely on the heels of another
one, also only found in the Bavli, where the location of the entrance to hell
(Gehinnom—‫ )גהינום‬is located.

‫ אחד במדבר ואחד בים ואחד‬,‫ שלשה פתחים יש לגיהנם‬:‫ואמר רבי ירמיה בן אלעזר‬
‫ בירושלים דכתיב ״נאם ה׳ אשר אור לו בציון ותנור לו בירושלים״ (ישעיה‬. . . ‫בירושלים‬
‫ וזו‬. . . ‫ שתי תמרות יש בגי בן הנום ועולה עשן מביניהן‬:‫ דבי רבי יוחנן בן זכאי‬.)‫לא ט‬
.)‫היא פתחה של גיהנם (ב׳ עירובין יט ע״א‬

Said Rabbi Yirmiah ben Eleazar: There are three entrances to Gehinnom,
one in the desert and one in the sea and one in Jerusalem . . . In Jerusalem,
as it is written: “So spoke the Lord, who has a fire in Zion and an oven in
Jerusalem” (Isa 31:9) . . . (In the name of) the school of Rabban Yohanan
ben Zakkai: Two palm trees are located in the Hinnom Valley and smoke
rises between them . . . this is the entrance to Gehinnom (bEruv 19a).

In this tradition the Bavli transmits, again in the name of a Palestinian sage
(Yirmiah ben Eleazar), that there are three possible locations for the entrance
to Gehinnom—the desert, the sea and Jerusalem. Another, even more promi-
nent Palestinian sage, Rabban Yohanan ben Zakkai specifically identifies this
location in Jerusalem in the Hinnom Valley.10 While the desert and the sea are

9 For the opinion that Damascus actually refers to Babylonia and the Babylonian exile,
albeit with reference to Damascus mentioned in the Dead Sea scrolls see I. Rabinowitz, “A
Reconsideration of ‘Damascus’ and ‘390 Years in the ‘Damascus (‘Zadokite’) Fragments’,”
Journal of Biblical Literature 73 (1954) 11–35.
10 On the location of a geographically realistic Gehinnom see L. R. Bailey, “Gehenna: The
Topography of Hell,” Biblical Archaeologist 49 (1986) 187–91. On the development of this
tradition from a physical to a metaphysical location see C. Milikowsky, “Gehenna and
‘Sinners of Israel’ in Light of Seder ‘Olam,” Tarbiz 55 (1986) 311–43 [Hebrew].
babylonia and the land of israel in the bavli 169

neutral places, Jerusalem is the most central theological location in the Land
of Israel. According to the Bavli, two Palestinian sages identify the entrance to
Hell in their region. No corresponding tradition locates it in Babylonia.
In passing, one may note that another tradition, also only found in the Bavli,
may be locating the entrance to Hell elsewhere in the Land of Israel. This tra-
dition is a response to a mishnah which describes the people of Tiberias as
warming water on Shabbat in a warm aqueduct:

‫ אמרו להן‬.‫מעשה שעשו אנשי טבריא והביאו סלון של צונן לתוך אמה של חמין‬
.)‫ אם בשבת כחמין שהוחמו בשבת אסורין ברחיצה ובשתיה (מ׳ שבת ג ד‬:‫חכמים‬

There was the case in which the people of Tiberias brought a stream of
cold [water] into an aqueduct of warm [water]. The sages said to them: If
it is on Shabbat, it is like water that was warmed on Shabbat and is forbid-
den for washing and drinking (mShab 3:4).

The Bavli comments that the origin of this aqueduct is in the warm springs
of Tiberias (Hama) and goes on to observe that these are warm because they
flow by the entrance to hell (‫ דחלפי אפיתחא דגיהנם‬,‫)ההוא תולדות אור הוא‬. Thus,
according to the Bavli, if the tradition that locates hell in Jerusalem is not
enough to convince one that that institution is located in the Land of Israel,
another one, which identifies its entrance in Tiberias, may clinch the argu-
ment.11 On points, one can consider Babylonia a winner, in hosting the more
hospitable of the two eternal resting places.12

11 An earlier Palestinian tradition identifies the entrance to both Gan Eden and Gehinnom
respectively in the vicinity of Jericho and Zoar, both mentioned in the same verse (Deut
34:3), see SifDeut 357. On this text see R. Nikolsky, “Gog in Two Rabbinic Narratives,” in:
W. J. van Bekkum, J. W. Drijvers, and A. C. Klugkist (eds.), Syriac Polemics: Studies in Honour
of Gerrit Jan Reinink (Leuven 2007) 21–40. On this specific topic see pp. 31–3.
12 I have found in the literature another trend, one attempting to identify Jerusalem with
the Garden of Eden. Based on the name of the third river of Eden, Gihon, it has been
suggested that this is the Gihon mentioned in the Bible (e.g. 1Kgs 1:33) as the water source
of Jerusalem; see L. E. Stager, “Jerusalem and the Garden of Eden,” Eretz Israel 26 (1999)
183*–94*. One scholar suggested that this concept is also developed in rabbinic literature;
see S. R. Simhoff, “Gardens: From Eden to Jerusalem,” Journal for the Study of Judaism 26
(1995) 145–55. However, the evidence she brings from the rabbis alternates between slim
to non-existent. If she is right, however, and “a common theme throughout these [mysti-
cal texts—T.I.] is the status of Jerusalem as the eschatological location of the Garden”
(a claim which she makes no attempt to prove textually), then we may view it as a late
polemical reaction to the Babylonian claim about the location of Eden.
170 ilan

What we were able to see here is how the Babylonians, without ignoring
or denying Babylon’s negative role in the biblical history of Israel, use biblical
verses and traditions to exalt rather than denigrate Babylonia. To demonstrate
that this is not an isolated incident, let us briefly look at another tradition,
which begins by praising the Land of Israel, but ends praising Babylonia at the
expense of the former. The tradition is found in Tractate Ta‘anit. I bring it here
with its parallel in a Palestinian Midrash:

‫ב' תענית י ע״א‬ ‫ספרי דברים לז‬

‫ ארץ ישראל נבראת תחילה וכל העולם‬:‫תנו רבנן‬ ,‫ נבראת לפני כל‬,‫ שחביבה מכל‬,‫ארץ ישראל‬
‫ ״עד לא עשה ארץ‬:‫ שנאמר‬,‫כולו נברא לבסוף‬ ‫ ״עד לא עשה ארץ וחוצות״ (משלי ח‬:‫שנאמר‬
‫ ארץ ישראל משקה‬.)‫וחוצות״ (משלי ח כו‬ . . . )‫כו‬
‫ וכל העולם‬,‫אותה הקדוש ברוך הוא בעצמו‬
‫ ״הנתן מטר על פני‬:‫ שנאמר‬,‫כולו על ידי שליח‬
‫ ארץ‬.)‫ארץ ושלח מים על פני חוצות״ (איוב ה י‬
‫ וכל העולם כולו מתמ־‬,‫ישראל שותה מי גשמים‬
‫ כל העולם כולו‬:‫ אמר רבי יהושע בן לוי‬. . . ‫צית‬
‫ ״ונהר יוצא‬:‫ שנאמר‬,‫מתמצית גן עדן הוא שותה‬
:‫ אמר רבי אושעיא‬. . . )‫מעדן וגו״ (בראשית ב י‬
‫ ״שוכנת על מים רבים רבת אוצרת״‬:‫מאי דכתיב‬
‫(ירמיהו נא יג)? מי גרם לבבל שיהו אוצרותיה‬
‫ מפני ששוכנת על מים‬,‫מלאות בר? הוי אומר‬
.‫ עתירה בבל דחצדא בלא מיטרא‬:‫ אמר רב‬.‫רבים‬
‫ כשבא לפתות את‬,‫וכן אתה מוצא בסנחריב‬
‫ישראל מה אמר להם? ״עד בואי ולקחתי אתכם‬
‫ אל ארץ יפה‬.)‫אל ארץ כארצכם״ (מל״ב יח לב‬
‫ אין כתוב כאן אלא ״אל ארץ כאר־‬,‫מארצכם‬
‫ אם מי שבא‬:‫ והלא דברים קל וחומר‬.‫צכם״‬
‫לומר שבחה של ארצו לא אמר גנייה של ארץ‬
.‫ קל וחומר לשבחה של ארץ ישראל‬,‫ישראל‬

bTaan 10a SifDeut 37

It was taught: The Land of Israel was cre- The Land of Israel, which is more beloved
ated first and the rest of the world last, as than all, was created before all, as it is
it is written: “Before he created Land and written: “Before he created Land and
beyond” (Prov 8:26). The Land of Israel is beyond” (Prov 8:26).
watered by the Holy One blessed be He
himself and all the rest of the world by
a messenger, as it is written: “Who gives
rain to the Land, and sends water over
babylonia and the land of israel in the bavli 171

the beyond” (Job 5:10). The Land of

Israel drinks rain water, and the rest
of the world from the leftover . . . Said
Rabbi Joshua ben Levi: All the world
drinks from the leftovers of the Garden
of Eden, as it is written: “And a river flows
from Eden etc.” (Gen 2:10) . . . Said Rabbi
Oshaya: What is the meaning of “She
dwells on many waters and her ware-
houses are full of grain” (Jer 51:13)? What
makes Babylonia’s warehouses full of
grain? It is because she dwells on many
waters. Said Rav: Babylon is rich because
it is watered without rain.
And also with regard to Sennacherib,
when he came to seduce the Israelites,
what did he say to them? “Till I come and
take you to a land like your land” (2Kgs
18:32). It is not written “a land more beau-
tiful than your land,” but rather “a land
like your land.” This is an a fortiori case: If
this person, who came to praise his own
country, did not [do this by] degrading
the Land of Israel, how much more so
[would one who came] to praise the Land
of Israel do.

The Babylonian tradition here compares the hydrological qualities of the two
countries: the Land of Israel is watered by rain, Babylonia by rivers and irriga-
tion canals. The tradition begins favorably toward the Land of Israel, singling it
out as having been created first and as being personally watered by God while
other countries are being watered by an agent, or by leftovers of the waters of
the Garden of Eden. The reference to the Euphrates and Tigris rivers is not lost.
Consequently, the Babylonian version of this midrash ends with a description
of the richness and plenty of Babylonia, achieved by citing a verse from Jer 51.
This chapter in Jeremiah, it should be noted, is a destruction prophecy, in
which the prophet predicts the utter destruction of the rich Babylonia. Verse 13
describes this wealth: “She dwells on many waters and her warehouses are full
of grain.” In the next verses Jeremiah describes how this wealth is destined to
be destroyed; taken on its own, however, verse 13 is an especially positive verse.
172 ilan

Our text inquires: Why are Babylon’s warehouses full of grain? And answers the
question with the words: “because she dwells on many waters.” The question
and answer are presented in the present tense. There is no suggestion that this
wealth is a thing of the past, or that it was ever destroyed. The tradition ends
with the judgment of the Babylonian Rav: Babylon is rich because it is watered
without rain. In other words, what began as a compliment for the Land of
Israel—being personally watered by God—ends as a condemnation. In the
eyes of the Babylonian rabbis, Babylonia is more blessed because it dwells on
water and its wealth, unlike that of the Land of Israel, is not dependent on
God’s capricious good will.
It is particularly instructive to compare this tradition with the Palestinian
midrash that lies at its basis. In Sifre Deuteronomy we find the tradition that
describes the Land of Israel as having been created first and as being favored
above and beyond all other countries. This long song of praise, only part of
which is here presented, also ends with a comparison of the Land of Israel
with Mesopotamia, albeit its Assyrian region. The Palestinian midrash cites
the words of Sennacherib to the inhabitants of Jerusalem—“I shall come and
take you to a land like your land”—and reacts to it with the words, “This is an
a fortiori case: If this person, who came to praise his own country, did not [do
this by] degrading the Land of Israel, how much more so [would one who
came] to praise the Land of Israel do.” This midrash does not end by compar-
ing Babylonia to the Land of Israel favorably. Yet based on the same tradition,
the Babylonians did.

To conclude: the Babylonian rabbis, who composed the Bavli, were deeply
indebted to, and wholly dependent on, the Torah of the Land of Israel and
on the Hebrew Bible in their traditions.13 Yet they also saw themselves as
local patriots of their adoptive country. By not rejecting the traditions they
received, but rather integrating and subtly manipulating them, they were able
to tone down the negative attitude found in these sources toward their land,
Babylonia. Especially useful to them in this rhetoric is the biblical assertion
that the two easily recognizable rivers of Babylonia are described in the cre-
ation story as watering the Garden of Eden. The Babylonians make us aware
that they are now living in what is the closest location to the Garden of Eden
as humanly possible.

13 See e.g. A. Goldberg, “The Babylonian Talmud,” in: S. Safrai and P. J. Tomson (eds.),
The Literature of the Sages 1 (Philadelphia 1987) 323–35.
From Disagreement to Talmudic Discourse:
Progymnasmata and the Evolution of a
Rabbinic Genre
David Brodsky

1 Introduction

How did rabbinic literature develop from the simpler structure of the Mishnah
in tannaitic Palestine (ca. 70 to 220 CE), with its statements of laws sometimes
followed by a dissenting opinion, to the complex structure of the Babylonian
Talmud (the Bavli) several centuries later in the Persian Empire, with its
lengthy give-and-take? What facilitated this change in the style of composi-
tion? While many answers to these two questions exist, two key answers are
able to account for much of this change. The first is the introduction of a model
of composition found in Greek primers of compositional writing. The second
is a subtle but significant change in the rules of biblical exegesis that facilitated
the increasing complexity of many sugyot.
The quintessential genre of the Talmud is the sugya, which in its basic form
is a statement with a support (usually a scriptural or tannaitic proof-text), fol-
lowed by a challenge (qushya ‫)קושיה‬, a resolution (teiruts ‫ )תירוץ‬of the chal-
lenge, another challenge, another resolution, and so forth. bEruv 52b is a classic
example:1

,‫ לא יכנס‬,‫ רגלו אחת בתוך התחום ורגלו אחת חוץ לתחום‬:‫[ אמ׳ ר׳ חנינא‬Statement]
.‫ ״אם תשיב משבת רגל[י]ך״ (ישעיהו נח יג) ״רגלך״ כת׳‬:‫[ משו׳ דכת׳‬Support]
.‫ יכנס‬:‫[ והתניא‬Challenge]
.‫ למקום שרובו שם נזקר‬:‫ אחרים אומ׳‬:‫ דתניא‬.‫[ הא מני? אחרים היא‬Resolution]

[Statement] Rabbi Hanina said: If [a person has] one foot within the
Sabbath limit [past which it is forbidden to walk on the Sabbath], and
one foot outside the Sabbath limit, he should not enter.

* I would like to thank Miryam Segal, Sipai Klein, Ron Naiweld, Michael Peachin, Tal Ilan, and
Ronit Nikolsky for their insightful comments on earlier drafts of the present study.
1 MS Vatican 109.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�08

174 BRODSKY

[Support] Because it is written [mishum di-khetiv]:2 “If you return your

feet [raglekha] from the Sabbath” (Isa 58:13). It is written “your foot
[regelkha].”
[Challenge] But is it not taught [ve-ha-tanya]: He may enter [if one foot
is within the Sabbath limit and one is outside of it]?!
[Resolution] Who taught that [?‫ ?]הא מני‬Others [that is, not the major-
ity], for it is taught: Others say: He is considered to be in the place in
which his majority resides. [So, if only one foot is past the Sabbath
limit, but the rest of his body is within it, he may still return home.
Since this is quoted as the minority opinion, it proves that the majority
must have held that such a person may not return home.]3

This is a fairly simple sugya. It begins with an opening statement with a bibli-
cal proof-text to support it, followed by a challenge to this statement, based
on a tannaitic ruling, and a final resolution, arguing that the tannaitic ruling
represents a minority opinion. More complex sugyot have a lengthy series of
challenges and resolutions. Some use the Mishnah as the opening statement,
others open with a baraita (a statement from the tannaitic period) or meimra
(a statement from the amoraic period), with the series of challenges posed
thereon. Some begin with two opposing laws and pose a series of challenges
and resolutions from one to the other and vice-versa. A variety of sugyot exist,
but the fundamental, basic structure underlying most of these sub-genres of
the sugya is the one present in bEruv 52b.
This structure is particularly known from the Bavli, and, for that reason, as
Daniel Boyarin recently stated, “the common sentiment [has been] that the
Bavli is a very strange book indeed, a unicum even on the rabbinic scene, a for-
tiori in world literature.”4 Yet, as Boyarin has noted, “The composition is rarely
discussed, and it seems that most scholars believe (without ever having spelled
it out) that the Babylonian Talmud is indeed sui generis.”5

2 The shift in language from Hebrew to Aramaic (‫ )משום דכתיב‬suggests that Rabbi Hanina’s
statement likely ended and the anonymous voice of the Bavli (the stam) has picked up. While
not all of the manuscripts have the word ‫משום‬, they all contain the word ‫דכתיב‬, confirming
the change in language from Hebrew to Aramaic.
3 Except when otherwise indicated, all translations are my own.
4 D. Boyarin, Socrates and the Fat Rabbis (Chicago 2009) 23.
5 Boyarin, Socartes, 21. On the sugya and its development, see, e.g. A. Weiss, The Talmud in its
Development (New York 1954) [Hebrew]; idem, Studies in the Literature of the Amoraim (New
York 1962) [Hebrew]; idem, Mehqarim ba-talmud (Jerusalem, 1975) [Hebrew]; J. N. Epstein,
Prolegomena ad Letteris Amoraiticas Talmud Babylonicum et Hierosolymitanum (Jerusalem
1962) [Hebrew]; C. Albeck, Introduction to the Talmud Babli and Yerushalmi (Tel Aviv 1969)
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 175

Interestingly, this genre of statement, support, challenge, and resolution is

not common in Persian literature of the Sasanian Empire, in spite of Yaakov
Elman’s extensive and perspicacious demonstration of the close relationship
between the Bavli and Middle Persian literature.6 While Persian literature has

[Hebrew]; D. Weiss Halivni, Sources and Traditions, 7 vols. (Tel Aviv and Jerusalem, 1968–2007)
[Hebrew]; S. Friedman, “A Critical Study of Yevamot X with a Methodological Introduction,”
in: H. Z. Dimitrovsky (ed.), Texts and Studies: Analecta Judaica vol. 1 (New York 1977) 275–
441 [Hebrew]; idem, Talmud Arukh: BT Bava Meẓi‘a VI, Critical Edition with Comprehensive
Commentary (2 vols.; New York 1993, 1997) [Hebrew]; J. Hauptman, Development of the
Talmudic Sugya: Relationship Between Tannaitic and Amoraic Sources (Lanham 1988);
R. Kalmin, The Redaction of the Babylonian Talmud: Amoraic or Saboraic? (Cincinnati 1989);
idem, Sages, Stories, Authors, and Editors in Rabbinic Babylonia (Atlanta 1994); idem, Jewish
Babylonia between Persia and Roman Palestine (New York 2006); D. Kraemer, The Mind of
the Talmud (New York 1990); idem, Reading the Rabbis: The Talmud as Literature (New York
1996); A. Cohen, Rereading Talmud: Gender, Law, and the Poetics of Sugyot (Atlanta 1998);
L. Moscovitz, “Between Casuistics and Conceptualization: On the Term ‘ameru davar ehad’ in
the Palestinian Talmud,” Jewish Quarterly Review 91 (2000) 101–42; idem, Talmudic Reasoning:
From Casuistics to Conceptualization (Tübingen 2002); idem, “‘Designation is Significant’:
An Analysis of the Conceptual Sugya in bSanh 47b–48b,” AJS Review 27 (2003) 227–52; and
Boyarin, Socrates. Most of this scholarship traces the genre back to its tannaitic and amoraic
rabbinic roots. Tracing the genre back to its potential non-Jewish origins, or even finding
close non-Jewish parallels, has proven quite difficult.
6 Y. Elman, “Marriage and Marital Property in Rabbinic and Sasanian Law,” in: C. Hezser
(ed.), Rabbinic Law in its Roman and Near Eastern Context (Tübingen 2003), 227–76; idem,
“Acculturation to Elite Persian Norms and Modes of Thought in the Babylonian Jewish
Community of Late Antiquity,” Neti’ot Ledavid (2004) 31–56; idem, “‘Up to the Ears’ in Horses’
Necks (B.M. 108a): On Sasanian Agricultural Policy and Private ‘Eminent Domain’,” Jewish
Studies, an Internet Journal 3 (2004) 95–149; idem, “Rav Yosef in a Time of Anger,” Bar Ilan
Annual 30–31 (2006) 9–20 [Hebrew]; idem, “Babylonian Yeshivot and Courts during the
Amoraic and post-Amoraic Era,” in: E. Etkes (ed.), Yeshivot and bate midrash (Jerusalem
2007) 31–55 [Hebrew]; idem, “ ‘He in His Cloak and She in Her Cloak’: Conflicting Images
of Sexuality in Sasanian Mesopotamia,” in: R. Ulmer (ed.), Discussing Cultural Influences:
Text, Context, and Non-Text in Rabbinic Judaism (Lanham 2007) 129–63; idem, “Who are
the Kings of East and West in Ber 7a?: Roman Religion, Syrian Gods and Zoroastrianism
in the Babylonian Talmud,” in: S. J. D. Cohen and J. J. Schwartz (eds.), Studies in Josephus
and the Varieties of Ancient Judaism: Louis H. Feldman Jubilee Volume (Leiden 2007) 43–80;
idem, “Middle Persian Culture and Babylonian Sages: Accommodation and Resistance in
the Shaping of Rabbinic Legal Tradition,” in: M. Jaffee and C. Fonrobert (eds.), Cambridge
Companion to Rabbinic Literature (Cambridge 2007) 164–97; idem, “The Socioeconomics of
Babylonian Heresy,” Jewish Law Association Studies XVII (2007) 80–126; idem, “Returnable
Gifts in Rabbinic and Sasanian Law,” Irano-Judaica 6 (2008) 139–84; idem, “The Other in
the Mirror: Iranians and Jews View One Another: Questions of Identity, Conversion and
Exogamy in the Fifth-Century Iranian Empire, Parts 1 and 2,” Bulletin of the Asian Institute 19
176 BRODSKY

much in common with the Bavli, including certain stylistic elements,7 techni-
cal terminology,8 hermeneutics,9 orality, and recording of differences of legal
opinion among scholars (i.e. what in the Bavli is called mahloqet), and while
it has proven an invaluable tool for unlocking the cultural context of many
passages and legal concepts in the Bavli, it lacks this specific genre (statement,
support, challenge, resolution). This is not to say that no form of give-and-
take ever appears in Middle Persian literature (indeed, the Pahlavi Videvdad
contains some extensive and quite interesting give-and-take that deserves fur-
ther study), but rather that such give-and-take is rare within the corpus as a
whole, and that it does not match the precise form we find in the Bavli.10 This,

(2009) 15–25; and 20 (2010) 25–46; idem, “Toward an Intellectual History of Sasanian Law:
An Intergenerational Dispute in ‘Herbedestan’ 9 and Its Rabbinic and Roman Parallels,”
21–57, in: C. Bakhos and M. R. Shayegan (eds.), The Talmud in Its Iranian Context (Tübingen
2010); G. Herman, “Ahasuerus the Former Stable-Master of Belshazzar, and the Wicked
Alexander of Macedon: Two Parallels between the Babylonian Talmud and Persian
Sources,” AJS Review 29 (2005) 283–98; and idem, “The Story of Rav Kahana (BT Baba
Qamma 117a–b) in Light of Armeno-Persian Sources,” in Irano-Judaica 6 (2008) 53–86;
R. Kipperwasser and D. J. Shapira, “Irano-Talmudica I—The Three-legged Ass and ‘Ridya’
in B. Ta’anith: Some Observations about Mythic Hydrology in the Babylonian Talmud and
in Ancient Iran,” AJS Review 32 (2008) 101–16; S. Secunda, Dashtana- ‘Ki Derekh Nashim
Li’: A Study of the Babylonian Rabbinic Laws of Menstruation in Relation to Corresponding
Zoroastrian Texts (Ph.D. diss., Yeshiva University, 2008); idem, “Reading the Bavli in Iran,”
Jewish Quarterly Review 100 (2010) 310–42.
7 Among the stylistic elements, we can include the fact that it is a commentary primarily
in one language (Pahlavi/Aramaic) on a text in another language (Avestan/Hebrew). This
is an element that we shall pick up in more detail below regarding Hellenistic Roman
legal scholia and their relationship to the Yerushalmi. Yet, even here, differences abound.
The Pahlavi literature is by and large a translation of the Avestan with, relative to the two
Talmudim, brief commentaries scattered throughout, while the two Talmudim, especially
the Bavli, can hardly be called a translation of the Mishnah, and, with the bulk of the
effort being commentary (in an expansive sense of that term). In this sense, the Pahlavi
literature is generally closer in genre to Targum than it is to Talmud.
8 E.g. ast kē ēdōn gowēd = ‫אית דאמר‬/‫איכא דאמרי‬.
9 Shai Secunda, for example, has shown that Pahlavi literature can be found interpreting
Avestan literature using a hermeneutic of omnisignificance much like rabbinic literature
does to biblical literature. That is, the Zoroastrians assume that verbosity and redundancy
in the Avestan text requires exegesis, and that each synonym should be interpreted as
adding something (Secunda, Dashtana—‘Ki Derekh Nashim Li’ 132–43).
10 Yishai Kiel pointed me recently specifically to PV 3.16, and Yaakov Elman has recently
highlighted a few passages from the PV and ZfJ that could potentially be read as give-and-
take not unrelated to that found in the Bavli. We shall take this point up further in the
conclusion.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 177

of course, makes sense when we note that the genre did not first appear in rab-
binic literature in Babylonia, but in Roman Palestine;11 and when we turn to
“world literature” of the time and place, we find the genre there as well.12
Thus, while the style of the sugya further developed in Sasanian Babylonia,
the present study will show that it originated in Greco-Roman Palestine and
that progymnasmata (Greco-Roman primers for written composition) from the
period confirm that the basic structure of the sugya was already well known in
the Greco-Roman world and was part of the educated elite’s basic education.13

11 Of course, the Bavli is composed of numerous styles and genres. Nevertheless, that of
statement, support, challenge, and resolution is at the same time ubiquitous and paradig-
matic. On the general genre of the sugya, see the sources cited in n. 5, above.
12 For previous studies on the relationship between rabbinic and Greco-Roman liter-
ary forms, see S. Lieberman, Greek in Jewish Palestine (New York 1942); idem, Hellenism
in Jewish Palestine (New York 1950); D. Daube, “Alexandrian Methods of Interpretation
and the Rabbis,” in: Festschrift Hans Lewald (Basel 1953) 27–44; E. Halevy-Epstein, “The
Writers of the ’Aggada and Greek Grammarians,” Tarbiz 29 (1959) 47–55 [Hebrew];
H. Fischel, “Studies in Cynicism and the Ancient Near East: The Transformation of a
Chria,” in: J. Neusner (ed.), Religions in Antiquity (Leiden 1968) 372–411; idem, “Story
and History: Observations on Greco-Roman Rhetoric and Pharisaism,” in: H. Fischel
(ed.), Essays in Greco-Roman and Related Talmudic Literature (New York 1977) 443–72;
C. Hezser, “Die Verwendung der hellenistischen Gattung Chrie im frühen Christentum
und Judentum,” Journal for the Study of Judaism 27 (1996) 371–439; B. Visotzky, “Midrash,
Christian Exegesis, and Hellenistic Hermeneutics,” in: C. Bakhos (ed.), Current Trends in
the Study of Midrash (Leiden 2006) 120–6; R. Hidary, “Classical Rhetorical Arrangement
and Reasoning in the Talmud: The Case of Yerushalmi Berakhot 1:1,” Association of Jewish
Studies Review 34 (2010) 33–64. On the relationship, or lack thereof, between Aphthonius’
and Hermogenes’ use of “syncrisis with the equal” in their progymnasmata and the
rabbinic use of gezerah shava, see Lieberman, Hellenism, 59 (see also, p. 94); and L. H.
Feldman, Jew and Gentile in the Ancient World: Attitudes and Interactions from Alexander
to Justinian (Princeton 1993), 35.
13 On the ubiquity of the Greco-Roman educational system by the rabbinic period, see
D. Clark, Rhetoric in Greco-Roman Education (New York 1957), 60, quoting Juvenal, Satire
xv.110; and H.-I. Marrou, Histoire de l’éducation dans l’antiquité (Paris 1948) 360. On Greco-
Roman education and the progymnasmata in particular, see Clark, Rhetoric, 177–212;
S. F. Bonner, Education in Ancient Rome: From the Elder Cato to the Younger Pliny (Berkeley
1977), esp. 250–76; G. Kennedy, Greek Rhetoric under Christian Emperors (Princeton 1983)
52–73; H. Cichocka, “Progymnasma as a Literary Form,” Studi italiani di filologia clas-
sica 10 (1992) 991–9; R. Cribiore, Writing, Teachers, and Students in Graeco-Roman Egypt
(Atlanta 1996); eadem, Gymnastics of the Mind: Greek Education in Hellenistic and Roman
Egypt (Princeton 2001); R. Webb, “The Progymnasmata as Practice,” in: Y. L. Too (ed.),
Education in Greek and Roman Antiquity (Leiden 2001) 289–316; M. Heath, “Theon and the
History of the Progymnasmata,” Greek, Roman, and Byzantine Studies 43 (2002–3) 129–60;
178 BRODSKY

The fact that debates existed has never been claimed to be unique to rab-
binic literature, and indeed can be found in dispute poems from the Middle
East as far back as ancient Sumeria. But that an author should introduce a
law and then walk the audience through a series of challenges and resolutions
has often been considered unique to rabbinic literature and particularly to
the Bavli. Yet, this style can already be found in Midrash Halakhah and the
Yerushalmi, and is even found in the Greek progymnasmata.
What is new in the Bavli is the level of complexity achieved at the hands of
the anonymous redactors. One of several factors contributing to this develop-
ment was a change in the rules of redundancy that governed biblical exegesis.
In Midrash Halakhah (whose material dates from the tannaitic and perhaps
early amoraic periods) rabbinic hermeneutics required that every aspect of a
verse teach something. No verse could have an extra or unnecessary word or
statement, and no two extra words or verses could come to teach the same law.

G. Kennedy, Progymnasmata: Greek Textbooks of Prose Composition and Rhetoric (Atlanta

2003); and C. Gibson, “Learning Greek History in the Ancient Classroom: The Evidence of
the Treatises on Progymnasmata,” Classical Philology 99 (2004) 103–29. See also R. Hock
and E. O’Neil, The Chreia and Ancient Rhetoric: Classroom Exercises (Atlanta 2002).
To my knowledge, except for Jenny Labendz’s recent dissertation (“Socratic Torah:
Non-Jews in Rabbinic Intellectual Culture” [PhD diss., The Jewish Theological Seminary,
2010]) and Daniel Boyarin’s recent book (Socrates and the Fat Rabbis), scholars have not
traced the genre of the sugya beyond rabbinic literature. Labendz argues for a relationship
between a specific sub-genre found in rabbinic literature (namely, dialogues between rab-
bis and non-Jews) and that of Socratic dialogue. She is not attempting to trace the style of
the sugya (i.e. statement, support, challenge, resolution) to the Socratic method. Boyarin
finds correlations between Plato’s and the Bavli’s general projects. Be that as it may, the
progymnasmata are stylistically much more closely related to the rabbinic genre than are
Plato’s dialogues, as I shall demonstrate below. David Daube compares rabbinic dialogic
discourse with that of the gospels (D. Daube, The New Testament and Rabbinic Judaism
[London 1956] 141–57), and even connects it to Hellenistic rhetoric (pp. 154–5), particu-
larly Aristotle’s Rhetoric (3.18.2–5) and Quintilian’s Institutio Oratoria (5.11). While Daube
is on the right track, his examples from Hellenistic rhetoric are all of dialogues between
individuals, and not of the narrator bringing both sides of an argument all under the same
voice. This is a key difference that sets the talmudic sugya and the progymnasmata apart.
Moreover, as a teacher, Quintilian taught the progymnasmata, and, therefore, his rhetoric
ought to be seen in that context.
For comparisons of the Bavli with other rabbinic sources, see, e.g. J. Neusner, The
Bavli’s Unique Voice: A Systematic Comparison of the Talmud of Babylonia and the Talmud
of the Land of Israel, 7 vols. (Atlanta 1993); idem, The Reader’s Guide to the Talmud (Leiden
2001) 295–364; J. Lightstone, The Rhetoric of the Babylonian Talmud: Its Social Meaning
and Context (Canada 1994), esp. 49–75 and 173–245; and A. Cohen, “A Unique Form of
Argumentation in Talmudic Rhetoric,” Rhetoric 15 (1996) 42–51.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 179

This rule remained in place throughout the rabbinic period. By the time of the
stam of the Bavli (ca. 500–600 CE), however, another rule had been added to
it: each redundancy or anomaly could teach one, and only one, law. This subtle
change multiplied the complexity of Babylonian sugyot. The present study will
survey these various genres, tracing how the genre of the talmudic sugya devel-
oped from these simpler Palestinian forms into the complex sugyot we find in
the Bavli.

2 The Origins of the Sugya in Greco-Roman Palestine

The basic structure of the sugya—an opening statement followed by a series of

challenges and resolutions—can already be found in tannaitic Palestinian rab-
binic documents, though the genre is less common in those documents than
it is in the Bavli.
The Mishnah (edited in Galilee around 220 CE) is generally composed of
lists of legal pronouncements, often with the dissenting opinion included.14
Thus, for example, with relation to the talmudic sugya discussed above, mEruv
4:11 states:

.‫[ מי שיצא חוץ לתחום אפילו אמה אחת לא יכנס‬Statement 1]

.‫ שלש לא יכנס‬,‫ שתים יכנס‬:‫[ רבי אליעזר אומר‬Statement 2]

[Statement 1] One who went outside of the limit [beyond which it is for-
bidden to walk on the Sabbath] even by one cubit may not [re-]enter.
[Statement 2] Rabbi Eliezer says, “Two [cubits], he may [re-]enter. Three
[cubits], he should not [re-]enter.”

While dissent is included in this literature, the give-and-take, so familiar from

talmudic literature, is largely absent from the Mishnah.
Of course, while give-and-take in the Mishnah is infrequent, it does exist.
One of the longest examples can be found in mPes 6:1–2, with its lengthy dis-
pute between Rabbi Eliezer, Rabbi Yehoshua, and Rabbi Aqiva:

14 On the Mishnah and its genre, see, e.g. the classic studies of J. N. Epstein, Introduction to
the Mishnaic Text (Jerusalem 1948; reprint, 2000) [Hebrew]; C. Albeck, Introduction to the
Mishna (Jerusalem 1959) [Hebrew]; and J. Neusner, Judaism: The Evidence of the Mishnah
(Atlanta 1988). For the Mishnah’s own explanation of the reason for and meaning of such
dissent, see mEduy 1:4–6. For a comparison of the Mishnah’s style with that of the Bavli,
see Lightstone, Rhetoric, 174–88.
‫‪180‬‬ ‫‪BRODSKY‬‬

‫משנה א‬
‫]‪ [Statement 1‬אלו דברים בפסח דוחין את השבת‪ :‬שחיטתו‪ ,‬וזריקת דמו‪ ,‬ומיחוי‬
‫קרביו‪ ,‬והקטר חלביו‪ .‬אבל צלייתו והדחת קרביו אינן דוחין את השבת‪.‬‬
‫]‪ [Statement 2‬הרכבתו והבאתו מחוץ לתחום וחתיכת יבלתו אין דוחין את השבת‪.‬‬
‫]‪ [Statement 3—Opposing Statement 2‬רבי אליעזר אומר‪ :‬דוחין‪.‬‬

‫משנה ב‬
‫]‪ [Challenge to Statement 2‬אמר רבי אליעזר‪ :‬והלא דין הוא‪ .‬מה אם שחיטה‪,‬‬
‫שהיא משום מלאכה‪ ,‬דוחה את השבת‪ ,‬אלו‪ ,‬שהן משום שבות‪ ,‬לא ידחו את השבת?‬
‫]‪ [Resolution‬אמר לו רבי יהושע‪ :‬יום טוב יוכיח‪ ,‬שהתירו בו משום מלאכה ואסור בו‬
‫משום שבות‪.‬‬
‫]‪ [Challenge‬אמר לו רבי אליעזר‪ :‬מה זה יהושע‪ ,‬מה ראיה רשות למצוה?‬
‫]‪ [Resolution‬השיב רבי עקיבא ואמר‪ :‬הזאה תוכיח‪ ,‬שהיא מצוה‪ ,‬והיא משום‬
‫שבות‪ ,‬ואינה דוחה את השבת‪ .‬אף אתה אל תתמה על אלו‪ ,‬שאף על פי שהן‬
‫מצוה והן משום שבות‪ ,‬לא ידחו את השבת‪.‬‬
‫]‪ [Challenge‬אמר לו רבי אליעזר‪ :‬ועליה אני דן‪ .‬ומה אם שחיטה‪ ,‬שהיא משום‬
‫מלאכה‪ ,‬דוחה את השבת‪ ,‬הזאה‪ ,‬שהיא משום שבות‪ ,‬אינו דין שדוחה את השבת?‬
‫]‪ [Resolution‬אמר לו רבי עקיבא‪ :‬או חלוף‪ .‬מה אם הזאה‪ ,‬שהיא משום שבות‪ ,‬אינה‬
‫דוחה את השבת‪ ,‬שחיטה‪ ,‬שהיא משום מלאכה‪ ,‬אינו דין שלא תדחה את השבת?‬
‫]‪ [Challenge‬אמר לו רבי אליעזר‪ :‬עקיבא‪ ,‬עקרת מה שכתוב בתורה ״בין הערבים‬
‫במועדו״ (במדבר ט ג)‪ ,‬בין בחול בין בשבת‪.‬‬
‫]‪ [Resolution‬אמר לו‪ :‬רבי‪ ,‬הבא לי מועד לאלו כמועד לשחיטה‪.‬‬
‫]‪ [Conclusion‬כלל אמר רבי עקיבא‪ :‬כל מלאכה שאפשר לעשותה מערב שבת אינה‬
‫דוחה את השבת‪ .‬שחיטה‪ ,‬שאי אפשר לעשותה מערב שבת‪ ,‬דוחה את השבת‪.‬‬

‫‪Mishnah 1‬‬
‫‪[Statement 1] These things on Passover postpone the Sabbath [i.e. they‬‬
‫‪are to be done even though it is the Sabbath and generic form of these‬‬
‫‪activities is normally forbidden on it]: the slaughtering [of the Paschal‬‬
‫‪lamb], the sprinkling of its blood, and the cleaning out [lit., wiping] of‬‬
‫‪its innards, and the burning of its fats; but its roasting and the rinsing‬‬
‫‪of its innards do not postpone the Sabbath.‬‬
‫‪[Statement 2] Carrying and bringing it from outside of the limit [beyond‬‬
‫‪which it is forbidden to travel on the Sabbath], and cutting off its warts‬‬
‫‪do not postpone the Sabbath.‬‬
‫‪[Statement 3—Opposing Statement 2] Rabbi Eliezer says: They do‬‬
‫‪[postpone the Sabbath; i.e. are permitted to be done on the Sabbath].‬‬

‫‪Mishnah 2‬‬
‫‪[Challenge to Statement 2] Rabbi Eliezer said: But is it not deriv-‬‬
‫]מה עם[ ‪]? If‬והלא דין הוא ;‪able by logical inference [from another law‬‬
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 181

slaughtering which is [forbidden on the Sabbath] on account of work

[mela’khah; i.e. it is one of the thirty-nine categories of work] post-
pones the Sabbath [i.e. is permitted on the Sabbath], these that are
[merely] on account of rest [shevut; and do not fall under one of the
thirty-nine categories of work], should they not postpone the Sabbath
[i.e. should they not be permitted]?”
[Resolution] Rabbi Yehoshua said to him: The festal day [ yom tov]
will prove it [yokhiah], for they permitted [work] on it on account of
mela’khah [i.e. some work that fell under the thirty-nine categories is
permitted on yom tov] and [yet] it is forbidden on it to do [work that
falls under the category of] rest [shevut].
[Challenge] Rabbi Eliezer said to him: What is this, Yehoshua [,‫מה זה‬
‫ ?]יהושע‬What evidence does that which is voluntary [i.e. shevut] have
to do with that which is commanded [i.e. mela’khah]?
[Resolution] Rabbi Aqiva responded and said [‫]השיב ר׳ עקיבא ואמר‬:
Sprinkling [of the blood of the sacrifice] will prove it [‫]תוכיח‬, for it
is a commandment and it is [forbidden on the Sabbath] on account
of rest [shevut] and it does not postpone the Sabbath. So you, too, do
not be dismayed [‫ ]אף אתה אל תתמה‬regarding these that even though
they are commanded and they are [forbidden on the Sabbath] on
account of rest [shevut], they do not postpone the Sabbath [i.e. they
are forbidden].
[Challenge] Rabbi Eliezer said to him [‫]אמר לו ר׳ אליעזר‬: And regard-
ing it I apply the following logical inference [‫]ועליה אני דן‬: If [‫]ומה אם‬
slaughtering which is [forbidden on the Sabbath] on account of work
[mela’khah] postpones the Sabbath, sprinkling which is [merely for-
bidden on the Sabbath] on account of rest, is it not logical [‫ ]אינו דין‬that
it should postpone the Sabbath [i.e. be permitted on the Sabbath]?
[Resolution] Rabbi Aqiva said to him [‫]אמר לו ר׳ עקיבא‬: Or the opposite
[’o haluf ]: If [‫ ]מה אם‬sprinkling which is [forbidden on the Sabbath]
because of rest [shevut] does not postpone the Sabbath, slaughtering
which is forbidden on account of work [mela’khah] is it not logical
[‫ ]אינו דין‬that it should not postpone the Sabbath [i.e. be forbidden]?
[Challenge] Rabbi Eliezer said to him [‫]אמר לו ר׳ אליעזר‬: Aqiva, you have
uprooted what is written in the Torah: “at eventide . . . at its appointed
time” (Num 9:3). Be that on a weekday or on the Sabbath!
[Resolution] [Rabbi Aqiva] said to him [‫]אמר לו‬: My teacher [‫]רבי‬, give
me an appointed time for these like [i.e. as good as] the appointed
time for slaughtering! [That is, what is more of an appointed time than
slaughtering the holiday sacrifice?!]
182 BRODSKY

[Conclusion] Rabbi Aqiva stated a general ruling [‫]כלל אמר ר׳ עקיבא‬:

All work [kol mela’khah] that can be performed before the Sabbath
does not postpone the Sabbath [and thereby remains forbidden on
the Sabbath]. Slaughtering, which cannot be done before the Sabbath,
postpones the Sabbath [i.e. is permitted on the Sabbath].

While this mishnah already exhibits extensive debate—including repetitive

use of key terminology of give-and-take—even mishnayot like this are quali-
tatively different from a standard Babylonian sugya, with one of the key differ-
ences being that the Babylonian sugya generally offers the give-and-take all in
the same voice (usually that of the stam).15
The earliest rabbinic sources to contain such discourse are the various
works of Midrash Halakhah, especially the Sifra, as in the following example
(Sifra, dibbura de-hovah, parashah 2:1–3):16

.‫[ ״נפש בני ישראל״ הרי צבור כיחיד‬Statement with biblical support]
‫ אף צבור לא יביא‬,‫מה יחיד אינו מביא אלא על דבר שזדונו כרת ושגגתו חטאת‬
.‫אלא על דבר שזדונו כרת ושגגתו חטאת‬
,‫ שהוא מביא על שגגת המעשה‬,‫ מה אם היחיד‬:‫[ הלא דין הוא‬Challenge]
‫ שאין מביא על שגגת‬,‫ צבור‬,‫אינו מביא אלא על דבר שזדונו כרת ושגגתו חטאת‬
?‫ אינו דין שלא יביא אלא על דבר שזדונו כרת ושגגתו חטאת‬,‫המעשה‬
‫ תאמר‬,‫ שאין מביא חטאת על לא הודע‬,‫ אם אמרת ביחיד‬.‫[ לא‬Resolution]
?‫ שהוא מביא חטאת על לא הודע‬,‫בצבור‬
‫ ואינו מביא על‬,‫ שהוא מביא חטאת על לא הודע‬,‫[ משיח יוכיח‬Challenge]
.‫דבר שזדונו כרת ושגגתו חטאת‬
‫ שאין מביא חטאת על לא הודע במזבח‬,‫ אם אמרת במשיח‬.‫[ לא‬Resolution]
‫ שהוא מביא חטאת על לא הודע במזבח החיצון? הואיל‬,‫ תאמר בצבור‬,‫החיצון‬
‫ יביא על דבר שזדונו כרת ועל‬,‫והוא מביא חטאת על לא הודע על מזבח החיצון‬
.‫דבר שאין זדונו כרת‬

15 For other mishnayot that exhibit debate between sages, see e.g. mGit 1:6; mPea 6:6, 7:7;
mTer 9:2; mPes 6:5; mNed 10:6; mShevu 3:5 and 6; mZev 7:4; mMen 12:5; mHul 4:4; and mKer
3:9–10. I have found very few examples in the Mishnah that can be categorized as give-
and-take in one voice. Most of them, like mKer 3:10 or mAZ 2:5, are reports by a single
person of a give-and-take between rabbis. To the extent that they are reported by a single
individual, the give-and-take can be considered delivered by one voice. Yet, since they are
reports of disputes between earlier rabbis, the give-and-take is still between people. mZev
12:3 is one of the few mishnayot that contains give-and-take that is entirely anonymous,
but it is extremely brief. It does, however, start us in this important direction.
16 On the Bavli’s stylistic similarities to Midrash Halakhah, see Lightstone, Rhetoric, 215–35.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 183

‫ מה יחיד אינו מביא‬.‫ הרי צבור כיחיד‬.)‫[ ת״ל ״נפש בני ישראל״ (שם‬Conclusion]
‫ אף צבור לא יביא אלא על דבר שזדונו‬,‫אלא על דבר שזדונו כרת ושגגתו חטאת‬
.‫כרת ושגגתו חטאת‬

[Statement with biblical support] “An individual . . . the children of

Israel” (Lev 4:2–3).17 Behold [‫ ]הרי‬the public is like the individual [i.e.
the verse connects the two, equating them for legal purposes]: just as
[‫ ]מה‬the individual only brings [a sacrifice] for [a sin] for which the
purposeful transgression receives karet (and the accidental transgres-
sion requires a guilt offering),18 so [‫ ]אף‬the public only brings [a sac-
rifice] for [a sin] for which the purposeful transgression receives karet
(and the accidental transgression requires a guilt offering).
[Challenge] But is it not [derivable from] logic [‫ ;הלא דין הוא‬and there-
fore the verse was not required to establish the connection between
the two]? If [‫ ]מה אם‬the individual, who brings [a sacrifice] for an
accidental transgression of a deed [for which they were misled by the
court; ‫]שגגת מעשה‬, only brings [a sacrifice] for a matter for which the
purposeful transgression receives karet (and the accidental transgres-
sion requires a guilt offering), the public which does not bring [a sac-
rifice] for an accidental transgression of a deed [for which they were
misled by the court; ‫]שגגת מעשה‬, is it not logical that [‫ ]אינו דין ש‬it
should only bring a sacrifice for a matter for which the purposeful
transgression receives karet (and the accidental transgression requires
a guilt offering)?
[Resolution] No. If you said regarding [‫ ]לא אם אמרת ב‬an individual
who does not bring [an offering] for what is not known to him, would
you say [the same] regarding [‫ ]תאמר ב‬the public that does bring [an
offering] for that which is not known to it?
[Challenge] The anointed [priest] will prove the matter [‫]יוכיח‬, for he
brings a guilt offering for that which is not known to him, but he does
not bring [a sacrifice] for a matter for which the purposeful transgres-
sion receives karet (and the accidental transgression requires a guilt
offering).

17 These two phrases appear in the opposite order in the verse. I am preserving the order
quoted in the Sifra.
18 In his commentary on the Sifra, Louis Finkelstein argues that the words in parentheses
here and in what follows were added by mistake. The logic of the passage suggests that
he is correct, since, as he states, this is what the passage is attempting to determine. See
L. Finkelstein, Sifra on Leviticus (New York 1990), 4:108.
184 BRODSKY

[Resolution] No. If you say regarding [‫ ]לא אם אמרת ב‬the anointed

[priest] who does not bring a guilt offering on the outer altar for that
which is not known to him, would you say [the same] about [‫]תאמר ב‬
the public who does bring a guilt offering on the outer altar for that
which is not known to it? Since [the public] is bringing a guilt offering
on the outer altar for that which is not known to it, let it bring [a sacri-
fice both] for a matter the purposeful transgression of which receives
karet and for a matter the purposeful transgression of which does not
receive karet.
[Conclusion] Therefore, the Torah teaches [‫]תלמוד לומר‬, “An individ-
ual . . . the children of Israel” (Lev 4:2) [i.e. therefore, the Torah needed
to connect these two in this verse in order to teach the following
principle]: Behold [‫ ]הרי‬the public is like the individual: just as the
individual [‫ ]מה יחיד‬only brings [a sacrifice] for [a sin] for which the
purposeful transgression receives karet (and the accidental transgres-
sion requires a guilt offering), so too [‫ ]אף‬the public only brings [a sac-
rifice] for [a sin] for which the purposeful transgression receives karet
(and the accidental transgression requires a guilt offering).

Here we find an extensive give-and-take in a singular, anonymous voice, which

begins with an opening statement, develops through a series of challenges and
resolutions, and culminates in the conclusion, which is the restatement of the
initial position. Unlike mPes 6:1–2, the give-and-take is not part of an “actual”
debate, with each side arguing its case. Rather, the give-and-take all serves the
purpose of the author: to introduce the ruling and walk the audience through
the potential objections to it with their ready-made resolutions, culminating
in a full defense of the ruling. In Bakhtinian terms, this new genre is monologi-
cal not dialogical.19 Though most well-known from the Bavli, this genre first
appeared in rabbinic literature in Palestine in the late tannaitic period. It con-
tinued to develop in the amoraic period, first in Palestine and later reaching its
apex in Babylonia.
Likewise, in yBer 3:1, 5d we find the following kind of give-and-take:

.‫ אין שומעין לו‬,‫ אם רצה להחמי׳ על עצמו‬:‫[ תני‬Statement]

.‫ או משום שאין לו מי שישא משואו‬,‫[ למה? מפני כבודו של מת‬Support]

19 M. Bakhtin, The Dialogic Imagination: Four Essays by Mikhail Bakhtin (ed. M. Holquist;
transl. M. Holquist and C. Emerson; Austin 1981). On the relation of Bakhtin to rabbinic
literature, see Boyarin, Socrates, esp. 140–6. See also n. 35 below.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 185

‫ ואין תימר מפני‬,‫[ מה נפוק מביניהן? היה לו מי שישא משואו‬Clarification]

‫ הרי יש לו מי‬,‫ מפני שאין לו מי שישא משואו‬:‫ ואם תאמר‬.‫כבודו של מת אסור‬
.‫שישא משואו‬
.‫ פטור מנטילת לולב‬:‫[ והתני‬Challenge to B]
.‫[ תיפתר בחול‬Resolution]
.‫ פטור מתקיעת שופר‬:‫[ והתני‬Challenge to B]
.‫ בחול לא ביום טוב‬:‫[ אית לך מימר‬Resolution]
‫ כיי‬,‫ מכיון שהוא זקוק לו להביא לו ארון ותכריכין‬:‫[ אמ׳ ר׳ חנינא‬Conclusion]
‫ מחשיכין על התחום לפקח על עיסקי הכלה ועל עיסקי המת להביא‬:‫דתנינן תמן‬
.‫ כמי שהוא נושא משאו‬,‫ חלילים ומקוננות‬,‫לו ארון ותכריכין‬

[Statement] It is taught [tani], “If [a person whose close relative died]

wanted to be stringent on himself [and recite the shema‘ and/or put on
phylacteries even though he is exempt], we do not listen to him.
[Support] Why [‫( ?]למה‬A) To honor the dead or (B) because he has no
one to take on his burden [of responsibility]?
[Clarification] What is the practical difference between them [‫מה נפוק‬
‫[ ?]מביניהן‬A case in which] he has someone to take on his burden [of
responsibility]. If you say [the law] is (A) in order to honor the dead, it
would be forbidden, but if you say [the law] is (B) because he has no
one to take on his burden, behold [‫ ]הרי‬he has someone to take on his
burden.
[Challenge to B] But it was taught [‫]והתני‬: He is exempt from taking
the lulav! [which would make it a holy day, and therefore he would be
exempt from his burden anyway.]
[Resolution] Explain it as referring to [‫ ]תיפתר ב‬a regular day [of the
festival not a holy day].
[Challenge to B] But it was taught [‫]והתני‬: He is exempt from blowing
the shofar!
[Resolution] You can say [‫]אית לך מימר‬: on a regular day, not on a holy
day [i.e. Rosh ha-Shannah].
[Conclusion] R. Hanina said, “Since he is required to bring him a cof-
fin and shrouds, as we teach there ‘They may await nightfall at the
Sabbath limit to care for the needs of a bride or a corpse, to bring him
a coffin or shrouds, flutes or lamenters’ [mShab 23:4], that fits [with the
explanation in B above] that he is taking on his burden.”

The only key structural difference between this sugya and what I have termed
the classical structure is that here the support is given some clarification. The
186 BRODSKY

other difference is that only one part of the support is challenged. These vari-
ants are neither unique to, nor particularly typical of, the Yerushalmi. Rather,
they are part of the standard variation found in the Bavli as well. Every sugya
varies from the classical template, but that template still underlies most give-
and-take in both talmudim.
We have already seen an example of a simple sugya from the Bavli in the
Introduction, and in Part 2 below we shall soon see a classic complex type of
Bavli sugya. For the moment, suffice it to say that even beginner students of
Talmud quickly become familiar with the dizzying complexity into which this
genre developed at the hands of the Bavli’s redactors. We shall therefore skip
the Bavli for now and turn directly to the Greco-Roman sources.
When we look at progymnasmata, Greco-Roman primers of the same gen-
eral time and place of the Sifra and Yerushalmi used to train school children
in composition, we find a perfect match. The two styles in the progymnasmata
that are an exact fit with the rabbinic sugya are the last two, “On Thesis” (the-
sis) and “On Introduction of a Law” (nomou eisphora).20 The progymnasmata
were widespread in the late Roman Empire.21 Theon, Pseudo-Hermogenes,
Libanius, his student Aphthonius, and Nicolaus authored treatises and sample
exercises on the progymnasmata preserved to this day.22 Catherine Chin has
even found evidence that the progymnasmata were known to, and likely used
in, Syriac schools of education.23
Significantly, these two exercises are found in Libanius’s primer (among
others), and Libanius was in correspondence with the nasi (the Jewish patri-
arch) of his day.24 Thus, a direct link can be established between a known

20 Some consider it a gymnasma rather than a progymnasma. That is, it is on the border
between the end of the training to the gymnasmata and the beginning of the gymnas-
mata themselves (Kennedy, Greek Rhetoric, 65).
21 See the sources in n. 13 above.
22 Their primers often differ little between them, even using the same examples with at times
quite the same wording. For a collection of these primers, see Kennedy, Progymnasmata.
23 C. Chin, “Rhetorical Practice in the Chreia Elaboration of Mara bar Serapion,” Hugoye:
Journal of Syriac Studies 9:2 (2006). Online: <http://www.bethmardutho.org/index.php/
hugoye/hugoye-author-index/173.html>.
24 For Libanius’s progymnasmata, see C. Gibson, Libanius’s Progymnasmata: Model
Exercises in Greek Prose Composition and Rhetoric (Atlanta 2008). For his letters to the
patriarch see M. Stern, Greek and Latin Authors on Jews and Judaism (Jerusalem 1980)
2:580–99. On this correspondence see the sources cited in n. 26 below. On Libanius and
his school see Kennedy, Greek Rhetoric, 150–63; and R. Cribiore, The School of Libanius in
Late Antique Antioch (Princeton 2007). Probably one of the most famous treatises on the
progymnasmata was that of Aphthonius, who was known to have been one of Libanius’s
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 187

teacher of this genre and a member of the rabbinic community at the general
time the genre appeared in rabbinic literature.
Libanius gives three examples of theses. To conserve space, I shall excerpt
from one of them:

2. Εἰ τειχιστέον.
[Statement] (1) Ὁ τὴν φυλακὴν τοῦ βίου ζητῶν αἱρείσθω τὰ τείχη.
[Support] πᾶν γὰρ ὃ ταῖς πόλεσι σώζεται, φρουρεῖται τοῖς τείχεσι. τειχίζει μὲν
γὰρ οὐρανὸς τοὺς θεούς, τὰ δὲ θνητῶν διατηροῦσι περίβολοι, καὶ ὃ τοῖς θεοῖς
οὐρανός, τὰ τείχη θνητοῖς. (2) καὶ ἔγωγε τοὺς ποιητὰς ἄγαμαι τειχιστὰς τοὺς
θεοὺς πεποιηκότας τῆς Τροίας. οὕτως καὶ ὅσοι τειχῶν οὐ δεδέηνται, τεῖχος
κατεσκευάσαντο. . . . (4) καὶ μὴν τὰ τείχη σωφροσύνην διατηρεῖ. . . . (5) εἰ
δὲ . . . τὰ δίκαια σώζεται τείχεσι καὶ μετὰ τούτων τὸ σωφρονεῖν, πῶς οὐκ εἰς
ὅσοv οἷόν τε τιμητέον τὰ τείχη;
[Challenge] (6) Ναί, φησίν, ἀλλὰ δειλίας τὰ τείχη γίνεται πρόφασις.
[Resolution] δειλίαν ὀνομάζεις τὴν πρόνοιαν καὶ τὰ κάλλιστα τῶν ἔργων
αἰσχίστοις περιβάλλεις ὀνόμασι. . . . ἀλλὰ καθίστασαι τῆς ἀσφαλείας
κατήγορος ὄκνον ὀνομάζων τὴν φυλακήν. (7) τοὐναντίον, ἄνθρωπε, θαρρεῖν
τὸ τεῖχος, οὐ δεδιέναι παρέχεται. καὶ γὰρ ὅπλων ἀποσοβῆσαι παρασκευὴν
καὶ ξίφος ἀπαμύναι ἐχθρῶν . . . τοῖς μαχομένοις μόνη τῶν τειχῶν παρέσχεν
ἡ τέχνη. . . .
[Challenge] (9) Ναί, φησίν, ἀλλὰ δαπανηρὰ τοῖς κτωμένοις τὰ τείχη. τείχη
τὸν ὄντα πόρον ἀνάλωσε.
[Resolution] καὶ πόθεν ὁ τῶν ἀνθρώπων τετήρηται πλοῦτος; τὴν περιουσίαν
φρουροῦσι περίβολοι. . . .
[Conclusion] (13) καὶ τοίνυν, εἰ νόμοι μὲν κρατοῦσιν ὑπὸ τειχῶν, διὰ δὲ
περιβόλων πολιτεύεται δίκαια καὶ συνοίσουσιν οἷς ἂν εἶεν ταῦτα γινόμενα,
καὶ τειχίσαι πᾶσι τιμητέον τὰ τείχη τιμᾶν εἰδότα καὶ πολιτείαν καὶ βίον.

Thesis 2: Whether One Should Build a Wall

[Statement] (1) Let him who seeks to protect life choose walls;
[Support] for everything that is kept safe for cities is guarded by walls;
for while heaven serves as a wall for the gods, perimeter walls watch
over the affairs of mortals, and what heaven is to the gods, walls are to
mortals. (2) And I admire the poets for having depicted the gods as the

students (Kennedy, Greek Rhetoric, 59–60), and whose treatise also contains the lesson on
how to introduce a law with an example nearly identical to that found in Libanius’s edi-
tion, complete with give-and-take. Aphthonius’s progymnasmata have been translated in
Kennedy, Progymnasmata, 89–127.
188 BRODSKY

wall-builders of Troy. Thus even those who have not needed walls have
built a wall for themselves . . . (4) Moreover, walls watch over chas-
tity . . . (5) [I]f justice is kept safe by walls, and self-control along with
it, how must one not honor walls as much as possible?
[Challenge] (6) Yes, one might say, but [nai, phēsin, alla] walls become
an excuse for cowardice.
[Resolution] You are calling forethought “cowardice” and surrounding
the finest of works with the most shameful of names . . . [I]nstead you
stand as an accuser of security, calling protection “cowering.” (7) To
the contrary, sir, a wall allows us to take courage, not to be afraid; for
indeed, only the art of wall-building allows fighters to frighten away
the armament of weapons and ward off the enemy’s sword . . .
[Challenge] (9) Yes, one might say, but [nai, phēsin, alla] walls are
expensive for those who own them. Walls use up the existing revenue.
[Resolution] And what has watched over people’s riches? Perimeter
walls guard the wealth . . .
[Conclusion] (13) Well then, if the laws have power because of walls,
and if it is through perimeter walls that the government is conducted
justly, then their construction will benefit whoever has them, and
everyone must honor the building of walls, as they know how to honor
both government and life.25

This structure is uncannily similar to the talmudic sugya. It opens with a state-
ment (“Let him who seeks to protect life choose walls”), followed by a sup-
port (“for everything that is kept safe for cities is guarded by walls”), followed
by a challenge (“‘Yes,’ one might say, ‘but walls become an excuse for coward-
ice’”), a resolution (“You are calling forethought ‘cowardice’ . . .”), another chal-
lenge (“‘Yes,’ one might say, ‘but walls are expensive . . .’ ”), a final resolution
(“Perimeter walls guard the wealth . . .”), and a conclusion that reasserts the
initial statement (“everyone must honor the building of walls”). This style was
adopted by the Greco-Roman schoolteachers as the way to introduce a law, pre-
cisely what talmudic literature is trying to do in its sugyot. Any student trained
in one of these schools would be trained to introduce a law using this style. One
of the letters written by Libanius may indicate that the patriarch’s son was with
Libanius in his school, and that he had come to Libanius already educated in
the rhetorical arts. To whom the letter was addressed, however, is not entirely

25 Translated by C. Gibson (with original Greek) in Libanius’s Progymnasmata, 520–3.

FROM DISAGREEMENT TO TALMUDIC DISCOURSE 189

clear.26 Whether or not that letter refers to the patriarch’s son, clearly members
of the patriarch’s household would have to have studied in such schools and
would have been trained with the progymnasmata. The responsibilities of the
patriarch and his household to correspond in writing with important mem-
bers of the empire would have necessitated such education, and was one of
the purposes of such an education in the late antique Roman world.27 In fact,
the Tosefta states outright: ‫התירו להם לבית רבן גמליאל ללמד בניהן יונית מפני שהן‬
‫קרובים למלכות‬, “They permitted the house of Rabban Gamaliel to teach their
children Greek because they were close to the government” (tSot 15:8).28 That
is, part of the job of the patriarch’s household was to correspond in Greek with
governmental officials, necessitating such rhetorical education.
On the introduction of a law, Libanius gives the following example, which I
shall reduce to conserve space:

1. Συνηγορία νόμου τοῦ κελεύοντος τὰς τῶν ἀδελφῶν γαμετὰς μὴ γαμεῖν.

[Statement and Support] (1) Ἔφθασε μὲν ἡ φύσις τὸν νόμον καὶ τοὺς
ἅμα γεγονότας εἰς ὑμέναιον οὐκ ἐᾷ συνελθεῖν, βεβαιοῖ δὲ ὅμως τὴν φύσιν
ὁ νόμος καὶ δι’ ἑκατέρων ἄμφω προῆλθε, φύσις μὲν νόμῳ καταβάλλουσα
γένεσιν, νόμος δὲ βεβαιῶν ἃ τῇ φύσει δοκεῖ. δεῖ δ’ ἐπὶ τῷ νόμῳ βραχέα
φιλοσοφῆσαι. . . . τὰ δὲ συγγενῆ συγγένειαν βεβαιοῖ διιστάμενα. ἃ καλῶς
εἰδὼς ὁ τὸν παρόντα νόμον εἰσενεγκὼν ἀδελφοὺς οὐκ ἐᾷ ταῖς τῶν ἀδελφῶν
συναφθῆναι συνοίκοις.
[Challenge] (5) Ναί, φησίν, ἀλλ’ ἐπικουρίας δεῖ χηρευούσαις ταῖς γυναιξίν.
[Resolution] ἀδικεῖς, ἄνθρωπε, τὴν συμμαχίαν ἡδοναῖς ὁριζόμενος, ὥσπερ
οὐκ ἐνὸν συμμαχεῖν, εἰ μὴ γάμος παρῆν. . . . (6) τί δὲ χηρείαν ἐρεῖς γυναικός,
ἔνθα τοῦ συνόντος ἀδελφὸς καταλείπεται; μενέτω πατὴρ τοῖς παρ’ ἐκείνου
γεγονόσι παισὶ καὶ μὴ συγχωρείτω τῆς χηρείας αἰσθάνεσθαι <ταύτην ἐκ> τῶν
παρ’ αὑτοῦ συμμαχῶν. . . . καὶ μὴ ποιείτω γυναῖκα μὲν δοκεῖν ἀποβλέπειν εἰς
ἡδονάς, αὑτὸν δὲ γάμῳ συγγενείας ἀλλότριον. οἷς γὰρ συνάπτεται, διεῖλε τὸ
γένος.

26 See Stern, Greek and Latin Authors, 2:595–6; W. Meeks and R. Wilken, Jews and Christians
in Antioch in the First Four Centuries of the Common Era (Missoula 1978) 62, 65, n. 9, and
66, nn. 10–11; Visotzky, “Midrash, Christian Exegesis,” 120–1; Cribiore, Libanius, 76, 321;
C. Hezser, Jewish Literacy in Roman Palestine (Tübingen 2001) 106; and Hidary, “Classical
Rhetorical Arrangement,” 36–7, n. 16.
27 Kennedy, Greek Rhetoric, 70–1.
28 See Lieberman, Hellenism, 102.
190 BRODSKY

[Challenge] (7) Ναί, φησίν, ἀλλὰ γυναιξὶν ἡδοναὶ πρόχειροι. κρείττους δὲ

τῶν οὐ προσηκόντων οἱ συγγενεῖς. ἀμείνους εἰς παράπαν οἱ συγγενεῖς καὶ τὸ
παρ’ αὐτῶν κομιζόμενον ἐκ τοῦ προσήκοντος γίνεται δίκαιον.
[Resolution] (8) οὔπω τεθέαμαι μᾶλλον τὸ παρανομεῖν ἐπαινούμενον. τί
δὲ τῶν γυναικῶν κατηγορεῖς ἡδονὰς καὶ ταύτας βεβαιοῖς συναπτόμενος;
εἰ γυναῖκες νοσήσασαι πάσχουσιν, οἱ προσήκοντες εἴργουσι καὶ κωλυταὶ
γίνονται τῆς ἐκείνων ἀκρασίας, οὐ σύμμαχοι. (9) καὶ ἔμοιγε δοκεῖς
οὐδὲν ἐπαισθάνεσθαι τῶν ἐν γάμῳ δεινῶν. . . . (10) μέγα τοῖς ἀνθρώποις
τὸ γένος. . . . (11) καὶ συμμαχεῖν μὲν ἀλλήλοις, ἤν που δέῃ, προῄρηνται,
προανακόπτουσι δὲ τὰς τῶν ἐπιόντων ἐπιβουλάς, καὶ συγγένεια γίνεται ἡ
πλείστη τῆς ἐπικουρίας συντέλεια καλῶς ὁριζομένοις <καὶ> βεβαίως τὰ
νόμον τῆς φύσεως οὐ παραβαίνοντα νόμιμα.

Introduction of a Law 1: In Support of a Law Bidding Men Not to Marry

Their Deceased Brothers’ Wives
[Statement and Support] (1) Nature has anticipated law and does not
allow those born together to join in marriage, but law nevertheless
reinforces nature, and both advance through each other, nature estab-
lishing an origin for law, but law reinforcing what seems best to nature.
And it is necessary to philosophize briefly upon the law . . . [F]amilies
reinforce the family relationship by staying apart. Knowing this well,
the one who introduced the present law is preventing men from being
joined with their brothers’ wives.
[Challenge] (5) Yes, one might say, but [nai, phēsin, alla] wives need
help if they become widows.
[Resolution] You, sir, are wrong to define assistance in terms of plea-
sure, as if it were impossible to lend assistance unless there is a mar-
riage . . . (6) Why are you going to mention the wife’s widowhood
when the husband’s brother is left behind? Let him remain a father
to the children born of that man, and by assisting <with>29 his own
resources, let him not allow <her> to perceive her widowhood . . . . and
let him not make the woman seem to be looking toward pleasure, and
let him not alienate himself from the family by marrying her; for inas-
much as he joins with her he destroys the extended family.
[Challenge] (7) Yes, one might say, but [nai, phēsin, alla] pleasures are
readily available to women. And family members are better than those

29 These markings indicate that Gibson’s Greek text and translation differ in these words
from Richard Foerster’s Greek edition of 1903–27.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 191

who are not related. Family members are altogether better, and a child
raised by them becomes just from their being related.
[Resolution] (8) I have never seen acting illegally being praised more.
Why do you condemn the pleasures of women and reinforce them by
joining with one? If women suffer from this disease, their relatives
shut them in and become preventers of their lack of self-control, not
assistants. (9) And to me, at least, you seem to be aware of none of
the terrible things in marriage. . . . (10) The extended family is impor-
tant to people . . . (11) And they choose to assist one another, if ever it is
needed, and they beat back the plots of attackers, and the family rela-
tionship becomes the greatest joint contributor of help for those who
nobly <and> firmly define customs for themselves that do not violate
the law of nature.30

This exercise models for the student the way to introduce a new law: to open
with a statement of the law and its main basis (here primarily philosophical
rather than scriptural, as the talmudim are wont to do), and to follow that law
with a series of challenges and resolutions that leave the audience with a clear
sense that the law has stood up to scrutiny. This scrutiny is part of the defense
of the law that is essential to its introduction. In this progymnasma we have
the basic structure of composition that talmudists generally call a sugya, and
that most scholars consider unique to talmudic discourse. This suggests the
progymnasmata as one of the key points of dissemination of this genre in the
culture at large, including in rabbinic culture, whether or not Libanius and
the nasi themselves were the conduits. Since the progymnasmata were graded
by level, and since the progymnasma on how to introduce a law was the last
level in the first course book, it would seem that whoever the people were
who introduced the genre to rabbinic literature they had attained at least the
advanced elementary stage in the Greco-Roman educational system.31 Here,

30 Translated by C. Gibson (with original Greek) in Libanius’s Progymnasmata, 528–31. For

the reasons why I label this last section “resolution” rather than splitting off the last part
into a separate section labeled “conclusion,” see n. 52 below.
31 On the graded levels of the progymnasmata and the tripartite Greco-Roman educa-
tional system, see Clark, Rhetoric, 60 and 181; Kennedy, Greek Rhetoric, 52–103; Bonner,
Education, 250–308. Cribiore (Writing; cf. also her Gymnastics) has recently argued that
the lines between these three echelons were often blurred. For our purposes, this need
not concern us. However distinct or blurred, the exercises on thesis and on the introduc-
tion of a law were not at the beginning of a child’s study, nor were they at the highest
levels of training of a rhetor. Even after mastery of the Progymnasmata, the advanced
student went on to develop the skills of give-and-take further.
192 BRODSKY

I do not want to enter into the complicated question of whether the rabbis
were open to allowing their children to study Greek language and literature.
The Yerushalmi (yPea 1:1, 15c and ySot 9:15, 24c) answers initially in the nega-
tive, though it equivocates on this point. Indeed, the very need for such state-
ments is often read as evidence for the existence of the activity which is being
proscribed, and, as already mentioned, the Tosefta states that the patriarch’s
family was permitted to study Greek for their professional needs.32 Whether
or not rabbinic children were studying in such schools, and whether or not the
progymnasmata were the direct source of the rabbinic genre, the genre can no
longer be seen as exclusively rabbinic. Rather, this quintessentially rabbinic
style must be seen as part and parcel of its non-Jewish context.33
Of course, the progymnasmata are likely not themselves the originators
of the genre (i.e. statement–support–challenge–resolution) but its widest
disseminators. Here we enter the difficult question of how to understand the
relationship between these sources. Recent scholarship has moved away from
thinking in terms of influence of one text or culture upon another and towards
thinking of these texts and cultures much more intertextually, as having mutual
influences upon one another in a complex web of interactions. Among the
many scholars who have critiqued the notion of “Romanization” of the prov-
inces, Jane Webster has argued that a better model is that of “Creolization,”
that is, thinking of a mixing and confluence of cultures.34 As Martin Millett
put it:

32 tSot 15:8. See also mSot 9:14 (though there is question surrounding the authenticity of this
latter section of mSot in general; see Epstein, Introduction to the Mishnaic Text 2, 976–7)
and bSot 49a–b. Of course, rabbis like Rabbi Abbahu are known in rabbinic literature to
be quite well versed in Greek, and the Greek language has a special status in rabbinic law
(e.g. mMeg 1:8). In bSot 49b, Rabbi Judah the Patriarch is quoted as privileging Greek and
Hebrew over Aramaic as the languages to be spoken in the Land of Israel. Saul Lieberman
has already dealt with this issue at length; see Lieberman, Greek, 21; and idem, Hellenism,
100–14; see also Hezser, Literacy, 90–94 and 106.
33 See his Socrates, esp. 11–2, 140–78. V. Tejera (“The Hellenistic Obliteration of Plato’s
Dialogism,” in: G. Press [ed.], Plato’s Dialogues: New Studies and Interpretations [Lanham
1993] 129–43) argues that the dialogic nature of Plato’s dialogues was largely ignored
throughout the Hellenistic period.
34 J. Webster, “Creolizing the Roman Provinces,” American Journal of Archaeology 105 (2001)
209–25. See also R. Roth, “Introduction: Roman Culture between Homogeneity and
Integration,” in: R. Roth and J. Keller (eds.), Roman by Integration: Dimensions of Group
Identity in Material Culture and Text (JRA Supplementary Series 66; Portsmouth RI 2007):
7–10; M. Millet, The Romanization of Britain: An Essay in Archaeological Interpretation
(Cambridge 1990); D. Mattingly (ed.), Dialogues in Roman Imperialism (JRA Supplement
23; Portsmouth RI 1997); idem, “Vulgar and Weak ‘Romanization’: Or Time for a Paradigm
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 193

“Roman” culture was by definition a cosmopolitan fusion of influences

from diverse origins rather than purely the native culture of Rome itself.
We must thus see Romanization as a process of dialectical change, rather
than the influence of one “pure” culture upon others. Roman culture
interacted with native cultures to produce the synthesis that we call
Romanized.35

These scholars argue that it is more fruitful to think in terms of shared contexts
than to look for precise antecedents or direct influences.36 In many ways, this
study is no exception, and yet, if any genre were intended to have direct influ-
ence upon a larger culture, school primers are it! Primers were used to inculcate
children in styles of composition. The children then took these styles home to
their local communities. These school primers, then, were likely the point of
dissemination of this genre to the culture at large, rabbinic Judaism included.
Here, therefore, we ought to speak in terms of influence. Nevertheless, school
primers are rarely the origins of a genre, but rather reflect and teach the literary
norms of the culture. Here we come full circle to the contemporary scholastic
position. While we must speak of influence, the influence we are speaking of is
both of the school primers upon the culture and of the culture upon the prim-
ers. We are left understanding both the culture and the primers as equally set
in each other’s context. Following Webster’s notion of Creolization, Palestinian

Shift?,” Journal of Roman Archaeology 15 (2002): 541–6; G. Woolf, Becoming Roman: The
Origins of Provincial Civilization in Gaul (Cambridge 1998); R. Hingley, Globalizing Roman
Culture: Unity, Diversity and Empire (London 2005). As Millet has shown, this accultura-
tion was especially pronounced among the local elites, though as Woolf, Mattingly, and
Roth have shown, we should not think that acculturation was limited to the elite classes.
Important for this study, R. Roth has shown that acculturation was not unilateral, but that
a form of hybridization can be identified which may have even strengthened local identi-
ties (R. Roth, “Black-Gloss Wares from the Acropolis of Capena [LaCivitucola, Provincia
di Roma],” Papers of the British School at Rome 74 [2009] 119–62; idem, “Introduction”;
and idem, Styling Romanisation: Pottery and Society in Central Italy [Cambridge Classical
Studies; Cambridge, 2007]).
35 Millett, Romanization, 1. Webster has argued against Millet’s notion of cultural emulation,
pushing against a concept of unidirectional ideology to embrace a notion of Creolization
of all groups involved. While Webster may be correct that elsewhere Millet’s understand-
ing of this cultural exchange can be too unidirectional, in the above quote I think Millet
seems to share her general perspective.
36 See especially M. Satlow, “Beyond Influence: Toward a New Historiographic Paradigm,”
in: A. Norich and Y. Z. Eliav (eds.), Jewish Literatures and Cultures: Contexts and Intertext
(Providence 2008), 37–54.
194 BRODSKY

rabbinic Judaism should similarly be viewed in this context, both influenced

by and potentially influencing its cultural setting.37
While this is a fruitful way to think of rabbinic Judaism and these primers,
we should still trace the Greco-Roman sources of this genre. In Socrates and
the Fat Rabbis, Daniel Boyarin has attempted to trace the genre of the give-and-
take of the talmudic sugya to Plato’s dialogues. While Boyarin correctly notes
the platonic dialogue between “that” and “have” that have a certain monologi-
cal dialogism not entirely foreign to Talmudic discourse, they lack this specific
format of statement–support–challenge–resolution–challenge–resolution
that is so charac­teristic of the Bavli. In fact, the role of Socrates’s interlocutor is
often simply to offer agreement and to be a sounding board for Socrates’s ideas.
For example, in the Euthyphro, after stating his initial proposition, Euthyphro’s
only statements from 7a–8a are:

“No doubt, then [πάνυ μὲν οὖν]”;

“Just so, then [Οὗτω μὲν οὖν]”;
“I agree, Socrates; it was said [Δοκῶ, ὦ Σώκρατες˙ εἴρηται γάρ]”;
“It was said [εἴρηται γάρ]”;
“Definitely [πάνυ γε]”;
“That it is [Ἔστι ταῦτα]”;
“How not? [Πῶς γὰρ οὔ]”;
“But that is where the difference lies, Socrates, on these matters [Ἀλλ’
ἔστιν αὕτη ἡ διαφορά ὦ Σώκρατες, καὶ περὶ τούτων]”;
“Very necessarily [Πολλὴ ἀνάγκη]”;
“You say correctly [Ὀρθῶς λέγεις]”;
“Definitely [πάνυ γε]”;
“Thus [Οὗτως]”;
“It seems so [Ἔοικεν]”;
“It may be [Κινδυνεύει].”

This kind of dialogue rarely if ever appears in the Bavli, certainly not as part
of the usual give-and-take of the sugya. In fact, such “discourse” is overtly
denigrated in the famous story about Rabbi Yohanan and Resh Laqish in the
Bavli, in which Rabbi Yohanan finally laments the loss of his counterpart, Resh
Laqish, noting that his substitute, Rabbi Eliezer, is just a “yes man.”38 Of course,

37 For a similar understanding of Hellenistic Judaism (including Palestinian Jewry), see

E. Gruen, “Hellenistic Judaism,” in: D. Biale (ed.), Cultures of the Jews: A New History (New
York 2002) 77–132.
38 bBM 84a. I would like to thank Lynn Kaye for mentioning this point to me.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 195

in many places Socrates’ interlocutor has much more of a role. Yet, even in such
places, the dialogue generally remains what Boyarin, following Bakhtin, calls
“monological dialogue.”39 As Jeremy Barris notes, “The philosopher’s interlocu-
tor often seems just to be going along with the flow, with a very lengthy and
unrelieved series of affirmatives like ‘yes,’ ‘of course,’ ‘certainly.’”40
While Boyarin challenges this monological reading of Plato, he acknowl-
edges that the dialogues between Socrates and his interlocutors are monologi-
cal. Where he finds his dialogism is “between Plato the ‘author’ and the speech
of his hero, Socrates, throughout the corpus.”41 Whether Plato’s dialogues are
purely monological or contain a hidden and key dialogical message, as Boyarin
posits, the style of the corpus is still a far cry from the genre of give-and-take
(statement–support–challenge–resolution) typical of talmudic literature. As
the primers used in the schools of rhetoric, the progymnasmata point to these
schools and to the schools of rhetoric rather than those of philosophy as the
more likely source of the genre.
Rivka Ulmer has done a better job of showing the existence of the style in
the philosophical genre of diatribe, but, even here, the widespread dissemina-
tion of the genre seems best attributed to the progymnasmata, which were a
product of the schools of rhetoric. Nevertheless, when we turn to those schools,
we find a long history of the genre that can be traced back to Hermagoras and
stasis theory in the late second century BCE.42 Unfortunately, except for a few
fragments, Hermagoras’ own works are now lost, but his theory has been pre-
served by later rhetoricians—in fact, by rhetoricians much closer in time (and
therefore more relevant) to the rabbinic texts in which this genre is found.
As Malcolm Heath summarizes Cicero’s presentation of stasis theory in Inv.
1.18–19:

39 Boyarin, Socrates, 38; M. Bakhtin, Problems of Dostoevsky’s Poetics (ed. and transl.
C. Emerson; Minneapolis 1984) 100 n. 1.
40 J. Barris, The Crane’s Walk: Plato, Pluralism, and the Inconstancy of Truth (New York 2009),
147; also quoted in Boyarin Socrates, 57. In her recent dissertation “Socratic Torah,” Jenny
Labendz correctly focuses on the relationship of the Socratic dialogue with passages
in rabbinic literature that record (or purport to record) dialogues between rabbis and
non-Jews.
41 Boyarin, Socrates, 125.
42 R. Ulmer, “The Advancement of Arguments in Exegetical Midrash Compared to that of
the Greek ΔΙΑΤΡΙΒΗ,” Journal for the Study of Judaism 28 (1997) 48–91. I would like to
thank Sipai Klein for pointing me to Hermagoras and to Stasis theory as key parts of the
history of the development of this genre.
196 BRODSKY

P[rosecution]1 You killed your mother.

D[efense]1 I killed her justly.
J[ury]1, zêtêma Did he kill her justly?
D2 aition For she killed my father.
P2 But your mother should not have been killed by you,
her son; her crime could have been punished without
your committing a crime.
J2 krinomenon Was it right for Orestes to kill his mother because she
had killed his father?
D3 sunekhon My mother’s disposition . . . was such that her children
above all were bound to exact the penalty.43

“P[rosecution]1” can be effectively understood as what I have been labeling the

opening “Statement.” The “Defense” (D1, D2, and D3) corresponds with what I
have been labeling “Challenge.” “P2” corresponds to what I have been labeling
“Resolution.” Seen this way, Cicero’s presentation of stasis theory is quite close
to the format found in the progymnasma on how to introduce a law and in
rabbinic literature, with the main difference being the Jury’s two questions in
Cicero’s model which are absent from the progymnasma and from the generic
rabbinic model. Of course, such questions can be found in rabbinic sugyot,
though it remains to be seen whether this precise style (P1, D1, J1 [zêtêma],
D2 [aition], P2, J2 [krinomenon], D3 [sunekhon]) can be identified in a rabbinic
text. Of course, as Heath notes, this genre was still developing, and had several
sub-genres, with the format presented above being what Heath calls “Model 1.”
Heath offers two examples of what he calls “Model 2.” The first, from Cicero’s
Partitiones Oratoriae 101–6, is P1, D1, J1 [zêtêma], D2 [aition], P2 [sunekhon],
J2 [krinomenon].44 The second, from Rhetorica ad Herennium 1.26 is P1, D1, D2
[aition], P2 [sunekhon], J2 [krinomenon].45 This latter format, with only one
jury question at the end, is much closer to the progymnasmata and what I
deem to be the generic form of the rabbinic genre. It is basically statement–
challenge–support–resolution–concluding question, with the support being
simply a sub-section of the challenge. Heath’s Model 3, especially in the form
he identifies in pseudo-Augustine, is quite close to a common rabbinic style:

43 M. Heath, “The Substructure of Stasis-Theory from Hermagoras to Hermogenes,” The

Classical Quarterly 44 (1994) 117.
44 Heath, “Stasis-Theory,” 118.
45 Heath, “Stasis-Theory,” 118.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 197

P1 kataphasis The general is guilty of murder

D1 apophasis The killing was lawful.
J1 zêtêma Was the killing lawful?
P2 aition He killed a soldier.
D2 sunekhon I killed him because he swore he would desert.
J2 krinomenon Was it lawful for the general to kill the soldier because
he swore he would desert?46

If J1 is simply changed from a legitimate jury’s question to a rhetorical question

asked as part of P2 as a challenge to the Challenge (making it a Resolution in
the sugya), and if J2 is deleted, then we have the following format quite com-
mon to the Bavli:

P1 [Statement] The general is guilty of murder

D1 [Challenge] The killing was lawful.
P2 [Resolution] Was the killing lawful?! But, he killed a soldier!
D2 [Challenge] He killed him because he swore he would desert!

While the Bavli would rarely leave the give-and-take there, this could easily be
the opening of a talmudic sugya. Thus, while the genre was most broadly dis-
seminated through the exercise on how to introduce a law in the progymnas-
mata, we can trace the development of the genre over several centuries in the
schools of rhetoric that composed these primers.47
This does not mean that we should see the origins of the more general
genre as having formed exclusively within the Greek and Roman spheres. In its
most basic form, the give and take of the rabbinic sugya is a form of sustained
dispute between two positions, and this genre can be found in Mesopotamia

46 Pseudo-Augustine, RLM 143.25–145.33; Heath, “Stasis-Theory,” 119.

47 Demosthenes (“On the Crowns,” 24) evinces a brief example of his own rhetorical give-
and-take, though this does not seem like a mainstay of his style generally. As mentioned
in n. 13 above, David Daube has pointed out the similarity between Quintilian and the
rabbis, and while in the Institutio Oratoria 5.11.5 Quintilian advocates that the orator offer
both sides of a give-and-take, the give-and-take he suggests is the use of rhetorical ques-
tions and their answers, not a presentation of a series of challenges and resolutions on
a given legal proposition. Of course, as an educator, Quintilian seems to have used the
progymnasmata, including the last on how to introduce a law (Institutio Oratoria 2.4, esp.
2.4.33; see also Kennedy, Greek Rhetoric, 55). Unfortunately, Quintilian does not discuss
the details of this progymnasma enough for us to know if he taught it with give-and-take.
To the extent that he did, however, we should understand him as part of the context of the
progymnasmata.
198 BRODSKY

stretching back to ancient Akkadian and Sumerian dispute poetry, in which

oxen and horses, birds and fish, ploughs and hoes, ewes and wheat pair off in
sustained verbal debate for supremacy.48 The dispute poem was also popular
in Christian Syriac literature, notably by such important authors as Ephrem
and Jacob of Sarug, and one such poem even survives in Pahlavi.49 Notably, in
Syriac, this kind of poetry is known as a soghīthā, the Syriac counterpart to the
rabbinic Aramaic term sugya, though the rabbinic term is primarily of post-
talmudic usage.50 Wout Jac. van Bekkum has noticed the genre in Jewish/
Israelite literature dating as far back as the book of Judges, including in
midrashic works such as Mekhilta de-Rabbi Yishmael, Genesis Rabba, and Esther
Rabba, and Sebastian Brock identifies the genre in the Targumim.51 Neither of
them claims that the genre is particularly common in rabbinic literature, not-
ing only a few examples. In any case, the notion of dispute or debate should
hardly be seen as unique to Greek and Roman societies or to their school prim-
ers. Nevertheless, the precise form of the rabbinic genre analyzed in the pres-
ent study (statement–support–challenge–resolution—with the challenges

48 The full bibliography is too large to cite here. See, e.g. J. Bottéro, “La ‘tenson’ et la réflexion
sur les choses en Mésopotamie,” in: G. J. Reinink and H. L. J. Vanstiphout (eds.), Dispute
Poems and Dialogues in the Ancient and Medieval Near East: Forms and Types of Literary
Debates in Semitic and Related Literatures (Leuven 1991) 7–22; H. L. J. Vanstiphout, “Lore,
Learning and Levity n the Sumerian Disputations: A Matter of Form, or Substance?,” in
ibid., 23–46; M. E. Vogelzang, “Some Questions about the Akkadian Disputes,” in ibid.,
47–57.
49 S. P. Brock, “Syriac Dispute Poems: The Various Types,” in: G. J. Reinink and H. L. J.
Vanstiphout (eds.), Dispute Poems and Dialogues in the Ancient and Medieval Near East:
Forms and Types of Literary Debates in Semitic and Related Literatures (Leuven 1991)
109–19; idem, “A Dispute of the Months and Some Related Syriac Texts,” Journal of Semitic
Studies 30 (1985) 181–211; idem, “The Dispute Poem: From Sumer to Syriac,” Journal of the
Canadian Society for Syriac Studies 1 (2001): 1–10; idem, “The Dispute between the Cherub
and the Thief,” Hugoye: Journal of Syriac Studies 5/2 (2002); and R. Murray, “Aramaic and
Syriac Dispute-Poems and Their Connections,” in: M. J. Geller, J. C. Greenfield, and M. P.
Weitzman (eds.), Studia Aramaica (JSS Supplement 4; Oxford 1995), 157–87.
50 The word sugya does not appear in the Mishnah, Tosefta, Midrash Halakhah or the
Yerushalmi. It appears only three times in the Bavli (bShab 66b; bSanh 33a, 51b), though
only the latter two have the possibility of meaning something like the give-and-take of a
debate. See also bSanh 6a. N.B. that this term seems to be connected to Rav Papa.
51 W. J. van Bekkum, “Observations on the Hebrew Debate in Medieval Europe,” in: G. J.
Reinink and H. L. J. Vanstiphout (eds.), Dispute Poems and Dialogues in the Ancient and
Medieval Near East: Forms and Types of Literary Debates in Semitic and Related Literatures
(Leuven 1991) 77–9; Brock, “Dispute of the Months,” 184, n. 17, 209–11; Murray, “Aramaic,”
165–72.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 199

and resolutions in a single, anonymous voice, especially as a way of introduc-

ing law) matches the progymnasmata (and some diatribe as Ulmer has shown),
and is not found in that precise form in any other literature, at least not to my
knowledge. It is therefore to the rhetorical schools of the Roman Empire that
we should look for a better understanding of the genre, its origins and its goals.
Indeed, like the rabbinic sugya, while the progymnasma on how to introduce
a new law is dialogical in form, it is essentially monological, that is, all “voices”
ultimately serve the author’s goal of introducing and defending the new law.
Of course, as similar as the rabbinic sugya is to this Greco-Roman genre, dif-
ferences do exist. One difference is that rabbinic sugyot rarely have a stated
conclusion at the end (though the example from the Sifra did, and Libanius’s
sole example on “the introduction of a law” ends on a resolution, rather than
having a separate concluding section as the thesis did).52 Instead, by ending on
a resolution, talmudic sugyot usually infer the conclusion (though, of course,
not all sugyot end on a resolution). Another difference is that the very style
of writing in rabbinic sources is often exceedingly terse and dense, frequently
requiring the reader to flesh out the point her/himself, whereas Libanius spells
out the details quite explicitly.
Another difference is that the support offered in the Greco-Roman genre
is more likely to be based on logic than it is to be based on an authoritative
source such as Aristotle or Homer, while rabbinic sources are more likely to use
a biblical or rabbinic proof-text as a support than logic. To the extent that this
genre can be seen as borrowed, then, the adaptation of the proof-text shows
evidence of rabbinization. In line with Webster’s replacement of the concept
of “Romanization” with that of “Creolization,” our findings show that while
the origins of this genre may be Greco-Roman, in the hands of the rabbis it
becomes thoroughly rabbinic.53
Another difference is that rabbinic sources use the genre as part of a run-
ning commentary on an earlier source (the various works of Midrash Halakhah
use it as part of their running commentary on the Torah, and the Yerushalmi

52 I label the last section of the example above “conclusion” and do not include it in the pre-
vious resolution because it turns back to the introductory theme of honoring the building
of walls, rather than the main question of whether one should build a wall. Conversely, I
do not consider the end of Libanius’s example on how to introduce a law to be a conclu-
sion, since it seems primarily to be continuing the main defense of why one should not
marry one’s brother’s widow. Of course, whether to consider a section the conclusion or
the continuation of the final resolution is a fine distinction and somewhat semantic.
53 Cf. Henry Fischel’s observations about the rabbinization of the chreia, when it was bor-
rowed from Greco-Roman culture (Fischel, “Cynicism,” 407–11).
200 BRODSKY

and Bavli use it as part of their running commentary on the Mishnah), whereas
the progymnasmata offer it as a stand-alone passage. Of course, as with the
other dozen or so exercises in the progymnasmata, the intention is for the form
to be incorporated into whatever literature the student ultimately engages in
when s/he comes of age, with some rhetorical forms being more useful for cer-
tain genres than others. Thus, Nicolaus had already recommended refutation
(ἀνασκευή) and confirmation (κατασκευή) as particularly useful for judicial
argumentation (γένος δικανικόν).54 Yet, nowhere do the progymnasmata hint
that it should be incorporated into a running commentary, nor do we find any
Greco-Roman examples of such legal commentary. Nevertheless, it may not
have been as out of the ordinary in late antique Palestine as we might think.
First, Catherine Hezser has already found stylistic similarities between
Roman legal texts and the Yerushalmi. For example, she demonstrates that
both share the same basic structure for case stories: they begin with a state-
ment of the case (casus), which is then brought as a legal question (quaestio)
before a rabbi/jurist, who then renders his legal opinion (responsum).55 While
this is a good example of stylistic commonalities between the two corpora,
and while the format case–question–response may sound similar to the for-
mat statement–challenge–resolution, the origins of which the present study
is attempt­ing to trace, the two styles are actually significantly different. Case–
question–response is a brief structure that lays out a legal situation that is
brought before a jurist for a ruling, which he then gives. The format statement–
challenge–resolution begins always already with the legal ruling, which is then
challenged, only to be resolved. Legal questions are not challenges to a law, and
challenges need not be posed as a question. Resolutions to a difficulty need not
be recorded as a legal ruling (responsum), and responsa rarely come to resolve
a challenge to a law. While both structures can be found in rabbinic literature,
it is that of statement–challenge–resolution that came to dominate the genre.
Hezser also offers examples of Roman legal commentary and demon-
strates its similarities to rabbinic legal commentary.56 Neverthe­less, we have

54 See Nicolaus’ introduction to his progymnasmata, translated in Kennedy, Progymnas­

mata, 129–72, esp. 146 and 154; and see Clark, Rhetoric, 181 and 192; Kennedy, Greek
Rhetoric, 68; and Cichocka, “Progymnasma,” 994.
55 C. Hezser, “The Codification of Legal Knowledge in Late Antiquity: The Talmud
Yerushalmi and Roman Law Codes,” in: P. Schäfer (ed.), The Talmud Yerushalmi and
Graeco-Roman Culture vol. 1 (Tübingen 1998) 588–91; eadem, “Roman Law and Rabbinic
Legal Composition,” in: C. Fonrobert and M. Jaffee (eds.), The Cambridge Companion to
the Talmud and Rabbinic Literature (Cambridge 2007) 153–5.
56 Hezser, “Codification,” 596–602; eadem, “Roman law,” 156–9.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 201

yet to find Roman legal commentaries that used the format statement–
challenge–resolution, though it is completely plausible that such commentar-
ies existed. Indeed, when we look beyond the legal literature, we find that the
various styles taught in the progymnasmata were adopted by Greco-Roman
commentaries.57 In the time and place that the rabbis were composing
commentaries on their classical religious and legal texts (the Torah and the
Mishnah), their non-Jewish contemporaries were writing scholia (marginal
commentaries) on their religious, literary, and legal texts.58 Lebanon and
Palestine in particular saw the creation of scholia on Roman law codes, and
Alexandria was a center of scholia on Homer.
Significantly, the legal scholia shared a key innovation with talmudic lit-
erature: they were in a different language from the legal text on which they
commented. The Roman legal documents were primarily in Latin, but as the
students of the eastern empire (particularly those students of the important
law school of the East in Beirut), became progressively less proficient in Latin,
brief commentaries accrued in the margins of these texts to help these stu-
dents understand the Latin base-text.59 Thus, in the Scholia Sinaitica, which
are roughly contemporary with the Yerushalmi, we find the following marginal
comments:

§18 Nos generaliter: ὁρᾷς, πῶς καὶ Ulpianos κανον[ίζει] ἡμῖν, ὅσα
δαπανήματα πεποίηκεν ὁ ἀνὴρ [π]ρόσκ(αι)ρα τῶν καρπῶν ἕνεκεν, ταῦτα
τοί[ς κ]αρπο[ῖς] compensatε ύεται, οὐ μὴν ποιεῖ τ(ὴν) retention(a).
§23 Mora: σημείωσαι ὅτι moras γενομένης ἐπὶ τῆς a[dven]ticias προικὸς
ὁ πατὴρ ἔχει τὴν rei uxoriae. μά[θ]ε, [ὅτι] τελευτησάσης τῆς κόρης ὁ
πατὴρ ἔχει τὴν.60

Note that the lemmata (and thus the language of the base text on which these
marginal notes are commenting) are all in Latin, while the base language of
the commentary is Greek. While this shift in language may seem slight, it was

57 Cichocka, “Progymnasma,” 991.

58 The literature on the Homeric scholia is vast. For an overview, see G. Nagy, “Homeric
Scholia,” in: I. Morris and B. Powell (eds.), A New Companion to Homer (Leiden 1997) 101–
22. Nagy does a good job of citing and summarizing prior scholarship. This scholarship
focuses more on the importance of the scholia for reconstructing and understanding the
Homeric text than it does on the scholia as a genre. Regarding the legal scholia, see K.
McNamee, “Another Chapter in the History of Scholia,” The Classical Quarterly 48 (1998)
269–88.
59 McNamee, “Another Chapter,” 273.
60 Copied from McNamee, “Another Chapter,” 274–5.
202 BRODSKY

groundbreaking, as Kathleen McNamee has shown. Moreover, it came pre-

cisely as rabbinic commentaries on the Mishnah shifted from the Hebrew of
the Mishnah to the vernacular Aramaic of the talmudim. In fact, the anony-
mous commentaries of both the talmudim and the scholia are composed in a
similar mixture of the two languages, with the base language being Aramaic or
Greek, and the technical legal terminology being in Hebrew or Latin, respec-
tively. In the scholia, this shift can be seen graphically as the scribe moves from
one script to the other. Thus, in §23, we see the commentator shift into Latin
for such terms as moras, adventicias, and rei uxoriae, and in §18, we find the
author even using a Greek grammatical suffix (εὔεται) on a Latin legal term
(compensat). While the technical terminology may be preserved in Latin, the
language of discourse—and therefore the language in which even the Latin
is conjugated—is decidedly Greek. While the talmudim rarely go so far as to
conjugate a Hebrew word with an Aramaic grammatical suffix,61 they shift
between the technical legal terminology in Hebrew and the Aramaic language
of the commentary in similar ways. Once again, the rabbinic genre is consis-
tent with legal literature of its time and place.62
The main difference between the Yerushalmi and the legal scholia contem-
porary with it is that the scholia have brief legal notes and explanations with-
out the kinds of longer sugyot that we find in the Yerushalmi. Nevertheless,
if we combine the genre of the scholion with that of the progymnasmata on
how to introduce a law, and if we apply that to the Mishnah as the base text,
we effectively have invented Talmud. All of the building blocks of talmudic
literature were in place in the Greco-Roman legal community of late antique
Palestine and we must view the Yerushalmi as part of that fabric.63

61 Perhaps some words that have been labeled as Hebraisms by philologists ought to be seen
instead as a Hebrew term given an Aramaic conjugation, though this is much more dif-
ficult to determine in Hebrew/Aramaic than in Latin/Greek, since the former share the
same alphabet and script, while the latter do not.
62 For previous discussion of why the talmudim use Hebrew in some places and Aramaic
in others, see A. Bendavid, Biblical Hebrew and Mishnaic Hebrew (Tel Aviv 1967), 134–5
[Hebrew]; Friedman, “Critical Study of Yevamot X,” 301–2; and J. Neusner, Language as
Taxonomy: The Rules for Using Hebrew and Aramaic in the Babylonian Talmud (Atlanta
1990). In the present study, I am not addressing the question of why some meimrot (attrib-
uted statements from the amoraic period) are in Hebrew while others are in Aramaic.
Rather, I am contextualizing the stam’s use of Aramaic and Hebrew in light of Greek scho-
lia on Latin legal texts, showing that once again the rabbinic genre is not unparalleled.
63 That the rabbis had some knowledge of Roman law (or at least some relationship to it)
has been demonstrated by several scholars. See S. Lieberman, “Roman Legal Institutions
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 203

Moreover, the progymnasmata have more in common with their later tal-
mudic counterparts than just the overall structure of the argument. Both use
technical language to indicate the shift from one part of the give-and-take to
the next. Each challenge is introduced with a phrase that identifies and marks
it off from what came before: ναί, φησίν, ἀλλὰ, “Yes, one might say, but . . .” The
talmudim also create unique phrases to introduce challenges that set them
apart from the surrounding passage, words like ‫איני‬, ‫מיתיבי‬, ‫ורמינהו‬, and ‫והתניא‬.
While talmudic technical terminology lacks a close correlative to the Greek ναί,
φησίν, ἀλλὰ in Aramaic—the language of most talmudic technical terminology
that governs the give-and-take of the sugya—one Hebrew phrase (‫ואם לחשך‬
‫ )אדם לומר‬comes rather close to the Greek.64 It is used in give-and-take to intro-
duce a challenge, and it is always attributed to a tanna or an amora, that is, to
the earlier, often Palestinian, layer of the sugya.65 It is found in both Palestinian

in Early Rabbinics and in the Acta Martyrum,” Jewish Quarterly Review 35 (1944) 1–57;
B. Cohen, “Civil Bondage in Jewish and Roman Law,” in: S. Lieberman et al. (eds.), Louis
Ginzberg Jubilee Volume (New York 1945) 113–32; idem, “Letter and Spirit in Jewish and
Roman Law,” in: M. Davis (ed.), Mordechai M. Kaplan Jubilee Volume (New York 1953)
109–35; idem, “Conditions in Jewish and Roman law,” in: S. Lieberman et al. (eds.),
Harry Austryn Wolfson Jubilee Volume (Jerusalem 1965) vol. 1, 203–32; D. Daube, “Texts
and Interpretation in Roman and Jewish Law,” Jewish Journal of Sociology 3 (1961) 3–28;
R. Katzoff, “Sperber’s Dictionary of Greek and Latin Legal Terms in Rabbinic Literature—A
Review Essay,” Journal for the Study of Judaism 20 (1989) 195–206; idem, “Roman Edicts and
Taanit 29a,” Classical Philology 88 (1993) 141–4; idem, “Children of Intermarriage: Roman
and Jewish Conceptions,” in: C. Hezser (ed.), Rabbinic Law in its Roman and Near Eastern
Contexts (Tübingen 2003) 277–86; and C. Hezser, “Introduction,” in ibid., 1–15. Amram
Tropper even suggests that at least one rabbinic passage may reflect a dialogue between
a rabbi and a Roman legal scholar comparing and contrasting Roman and rabbinic law
(“Roman Contexts in Jewish Texts: On Diatagma and Prostagma in Rabbinic Literature,”
Jewish Quarterly Review 95 [2005] esp. 222–7).
64 While the Aramaic phrase ‫“( וכי תימא‬and if you say . . .”) is close, the less well-known
Hebrew phrase ‫ ואם לחשך אדם לומר‬lomar (“and if someone misleads you saying . . .”)
is much closer to the Greek since it retains the third-person element of the Greek word
φησίν.
65 By saying this, I am not claiming that it actually is the Palestinian layer of the sugya, only
that the text identifies it as that layer. Whether it actually is from that layer is difficult to
decipher with certainty. On this general issue, see Rubenstein, Culture, esp. pp. 7–13; and
Kalmin, Jewish Babylonia. Neusner has a different take on the Bavli’s sources (Neusner,
Language; idem, How the Talmud Shaped Rabbinic Discourse [Atlanta 1991]).
204 BRODSKY

and Babylonian texts, though primarily in the Bavli.66 bBer 7b offers our earli-
est attribution:67

‫ מותר להתגרות ברשע׳ בעו׳ הזה‬:‫[ וא״ר יוח׳ משום ר׳ שמע׳ בן יוח׳‬Statement]
.)‫ ״עזבי תורה יהללו רשע ושומרי תורה יתגרו בם״ (משלי כח ד‬:‫[ שנ׳‬Support]
.‫תניא נמי הכי‬
,‫ מותר להתגרות ברשעים בעולם הזה‬:‫[ ר׳ דוסתי בר׳ מתון או׳‬Statement]
.)‫ ״עוזבי תורה יהללו רשע ושומרי תורה יתגרו בם״ (שם‬:‫[ שנ׳‬Support]
‫] אל תתחר במריעים״‬:‫ ״לדוד‬. . . :‫ [והכת׳‬:‫[ ואם לחשך אדם לומ׳‬Challenge]
.)‫(תהלים לז א‬
‫ אלא ״אל תתחר במרעים״ (שם) להיות‬.‫[ מי שלבו נוקפו או׳ כן‬Resolution]
‫ ״אל יקנא לך‬:‫ ואו׳‬.‫כמרעים ״ואל תקנא בעושה עול״ (שם) להיות כעושה עולה‬
.)‫בחטאים כי אם ביראת יי׳ כל היום״ (משלי לג יז‬

[Statement] Rabbi Yohanan68 said in the name of Rabbi Shimeon ben

Yohai:69 It is permitted for a person to attack the wicked in this world,
[Support] as it is said: “Those who abandon the Torah will praise the
wicked, and those who keep the Torah will attack them” (Prov 28:4).
It is also taught thus in a baraita:70
[Statement] Rabbi Dostai bar Matun said: It is permitted to attack the
wicked in this world,
[Support] as it is said: “Those who abandon the Torah will praise the
wicked, and those who keep the Torah will attack them” (Prov 28:4).

66 It is found once in the Minor Tractate Derekh Eretz, half a dozen times in the Bavli, and
once in Midrash Psalms.
67 I am here following MS Florence. This passage has a parallel in Derekh Eretz (pereq ha-
minin 29) and in bMeg 6b. It is unclear which of the three is the earliest version.
68 A first to second generation Palestinian amora.
69 A third generation Palestinian tanna.
70 This is where the most variation exists among the manuscripts. The printed editions
(Soncino d. 1484, and Venice) agree with the Florence MS, which I used here. After quoting
Prov 28:4, MSS Paris and Oxford, on the other hand, skip directly to the give-and-take (“If
someone misleads you saying . . .”). This still attributes the give-and-take to Palestinians,
and ultimately to a tanna (Rabbi Shimeon ben Yohai), but it no longer officially quotes it
as a baraita. Of course, MSS Paris and Oxford’s version is likely an error based on homoio-
teleuton in which the scribe simply skipped from one quotation of Prov 28:4 to the next.
MS Munich preserves the attribution to Rabbi Dostai b. Matun, but it does not preserve
the introduction of his statement as a baraita (‫)תניא נמי הכי‬.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 205

[Challenge] And if someone misleads you saying [‫]ואם לחשך אדם לומר‬:
[But is it not written: “to David,]71 do not become angry with evildoers”
(Ps 37:1)?
[Resolution]72 One whose heart wounds him [Rashi: he who fears his
own sins] says thus. Rather, “Do not become angry with the evildoers”
(Ps 37:1) to be like the evildoers, and “Do not become envious of those
who do wrong” (ibid.) to become like those who do wrong. And it says,
“Do not become envious of sinners, but rather of the God-fearing all
day long” (Prov 23:17).

In this example, the statement–support–challenge–resolution are all given by

the same sage, purportedly dating this give-and-take to the tannaitic period.73
Whether it is tannaitic or not, the fact that this formula is always attributed to
sages of the tannaitic and amoraic periods (rather than the stam) strengthens
the likelihood of its relatively early provenance. Significantly, as in the pro-
gymnasmata, the challenge is introduced as a hypothetical statement that a
third party might make: “Yes, one might say, but . . .” in the progymnasmata,
and “And if someone misleads you saying, but . . .” in the rabbinic formula. Even
in the way they introduce these challenges, rabbinic sources and their counter-
parts in the progymnasmata seem closely related.
The structure of the Bavli, then, is not unique, neither in its larger rab-
binic context, nor among contemporary non-Jewish sources. While the Bavli
remains a singular and important text in Jewish history and of central impor-
tance for the development of Jewish culture, its style should be appreciated
in its cultural context. This is not to say that the Yerushalmi or, especially, the
Bavli are necessarily directly influenced by these Greco-Roman sources (par-
ticularly not by the legal scholia which were less ubiquitous than the progym-
nasmata, though neither is it to preclude such a possibility), but that they must
be viewed and understood in that context. Boyarin recently articulated this
general argument quite well:

71 The phrase in brackets is absent from the text of MS Florence—the MS used here—but
added in the margin in a different hand. For this reason, I include it in brackets to indicate
the marginal addition. Furthermore, the phrase is found in MSS Munich and Paris and in
the printed editions, though it is absent from MS Oxford.
72 All of the other MSS and the printed edition add here “say to him [’emor lo],” although
the parallel in Derekh Eretz lacks the phrase. When the formula appears in other places,
it often—though not always—contains this phrase to introduce the resolution. It seems,
therefore, generally—though not always—to be a part of the larger formula.
73 Be that Rabbi Yohanan in the name of Rabbi Shimeon ben Yohai or Rabbi Dostai in the
baraita.
206 BRODSKY

My point is not to argue that in general rabbinic texts are “influenced”

by Greco-Roman texts, but to use this particular incidence as evidence
for a claim of cultural interdependence between the Sasanian East and
the Byzantine West in late antiquity sufficient to understand and make
plausible my attempts to read the Bavli within the context of literary and
cultural moves taking place in that broader context.74

Even the Bavli in the Sasanian East must be understood as part of a larger
Greco-Roman literary culture that had reached as far east as Sasanian
Babylonia. Of course, Yaakov Elman has convincingly demonstrated that the
Bavli must be understood in its Persian context, but that does not negate the
fact (nor does Elman claim that it does) that the Bavli must also be under-
stood in the Greco-Roman context in which rabbinic literature developed as
a whole.75 Indeed, the genre’s relative absence in Pahlavi literature demon-
strates that the process of rabbinization of the Babylonian Jewish community
was also a process of Hellenization. That is, Palestinian rabbinic literature was
a conduit to the Babylonian Jewish community not only for rabbinic Jewish
ideology, but also for certain Hellenistic and Roman genres and notions, with
the basic style of the sugya being chief among them. These findings are con-
sistent with Martin Millett’s argument that native elites played an important
role in “Romanization.”76 While scholars have challenged the use of this loaded
term and have since pushed this theory to include non-elite native cohorts,77
Millett’s point that this process should be seen as taking place through native
conduits is still valid, and the elite nature of rabbinic literature78 makes his
thesis particularly apt.

74 Boyarin, Socrates, 181.

75 Richard Kalmin ( Jewish Babylonia, 173) recently made a similar point when he wrote:
“Babylonian Jewish society between the third and seventh centuries ce conformed to
Persian models. At no time, however, was it sealed off from influences from the west,
most recognizably in the form of traditions deriving from Palestinian rabbis, but also in
the form of non-rabbinic and non-Jewish traditions deriving from Palestine, and perhaps
from elsewhere in the Roman Empire.”
76 Millet, Romanization; and M. Millet, N. Roymans, and J. Slofstra, “Integra­tion, Culture
and Ideology in the Early Roman West,” in: M. Millett, N. Roymans, and J. Slofstra (eds.),
Integration in the Early Roman West: The Role of Culture and Ideology (Luxembourg 1995)
1–5.
77 Woolf, Becoming Roman; Roth, “Introduction”; idem, “Black-Gloss”; idem, Styling
Romanisation; Mattingly, Dialogues; idem, “Vulgar.”
78 This is not to say that the rabbis were all of the elite class in a Marxist or capitalist sense,
but merely that the rabbis who were trained to transmit and engage in rabbinic literature
were an elite group.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 207

3 The Rules of Redundancy: How a Midrashic Hermeneutic Changed

over Time and Its Impact on the Development of the Sugya

While the basic structure of the talmudic sugya had its origins in Palestine, the
genre did not reach its apex until the redaction of the Bavli in post-amoraic
Babylonia. There, the sugya would reach dizzy­ing heights of complexity and
convolution.79 Leib Moscovitz has been instrumental in showing that changes
in hermeneutics and in abstract language are key to unlocking the differences
between tannaitic, amoraic, and post-amoraic literature.80 Christine Hayes
has shown that exegetical differences are essential to unlocking the halakhic
differ­ences between the Palestinian and Babylonian Talmudim.81 However,
one of the elements that helped facilitate and contribute to this growing com-
plexity, a subtle but significant change in the hermeneutics of biblical exegesis,
has garnered little attention to date.
Already in the tannaitic period, superfluous language in the Torah required
interpretation (particularly by Rabbi Aqiva and the school attributed to him).
The notion was that for the Torah to be a perfect text, it could not contain any
unnecessary words. The standard explana­tion for wordiness in the Torah was
to interpret each unnecessary word as coming to teach something beyond the
simple meaning of the text. This hermeneutic rule remained in place through-
out the rabbinic period.82

79 Much has been written on the Bavli’s relative complexity when compared to other rab-
binic literature. See the sources cited in n. 5 above, especially the recent work by Leib
Moscovitz.
80 Moscovitz, Talmudic Reasoning.
81 C. E. Hayes, Between the Babylonian and Palestinian Talmuds: Accounting for Halakhic
Differences in Selected Sugyot from Tractate Avodah Zarah (New York 1997).
82 See J. Kugel, The Idea of Biblical Poetry: Parallelism and Its History (New Haven 1981)
103–4; D. Stern, “Midrash and Indeterminacy,” Critical Inquiry 15 (1988) 132–61; idem,
Midrash and Theory: Ancient Jewish Exegesis and Contemporary Literary Studies (Evanston
1996) 29; idem, “The First Jewish Books and the Early History of Jewish Reading,” Jewish
Quarterly Review 98 (2008) 174–5; R. Steiner, “Meaninglessness, Meaningfulness, and
Super-Meaningfulness in Scripture: An Analysis of the Controversy Surrounding Dan
2:12 in the Middle Ages,” Jewish Quarterly Review 82 (1992) 431–9; Y. Elman, “Midrash
Halakhah in its Classic Formulation,” in: L. Teugels and R. Ulmer (eds.), Recent
Developments in Midrash Research: Proceedings of the 2002 and 2003 SBL Consultation on
Midrash (Piscataway NJ 2005) 7–12; idem, “ ‘It is No Empty Thing’: Nahmanides and the
Search for Omnisignificance,” Torah U-Madda Journal 4 (1993) 1–83; idem, “The Rebirth
of Omnisignificant Biblical Exegesis in the Nineteenth and Twentieth Centuries,” Jewish
Studies Internet Journal 2 (2003); idem, “Classical Rabbinic Interpretation,” in: A. Berlin,
M. Brettler, and M. Fishbane (eds.), The Jewish Study Bible (New York 2004), 1848–58;
208 BRODSKY

Thus, when the Bible states three times, “Do not cook a kid in its mother’s
milk” (Exod 23:19; 34:26 and Deut 14:21), tannaitic rabbinic sources already felt
a need to explain each occurrence of the statement as coming to teach a sepa-
rate rule. In the Mishnah (mHul 8:4), Rabbi Aqiva states:

‫ פרט‬,‫ ״לא תבשל גדי בחלב אמו״ שלש פעמים‬:‫ שנאמר‬,‫חיה ועוף אינם מן התורה‬
.‫לחיה ולעוף ולבהמה טמאה‬

Deer and fowl are not [prohibited to cook/eat with milk] from the Torah,
for it says “Do not cook a kid in its mother’s milk” three times, [once] to
exclude deer [from the prohibition], [once to exclude] fowl, and [once to
exclude] the impure animal [i.e. a non-kosher species].

Even in the Mekhilta (MdRY mishpatim, massekhta de-kaspa, parashah 20) of

the (so-called) school of Rabbi Yishmael, which has often been considered
more lax on these issues,83 rabbi after rabbi is recorded as offering an explana-
tion for “why it is said in three places,” with Rabbi Yoshaya offering the answer
(contra Rabbi Aqiva above) that one instance comes to include cattle, one deer,
and one fowl in the prohibition. Other rabbis offer different answers, but what
nearly every answer has in common is that it comes to explain each instance
as coming to teach something separate. Some rabbis recorded in this pas-
sage subscribe to the rule that the first instance of a law is not redundant and
therefore need not be interpreted—only the second and third instances need
interpretation—while others interpret all three statements. While the details
of this ideology vary slightly from rabbi to rabbi, what is consistent throughout
is that redundancies in the Bible (however that is defined) must be explained.84
This basic aesthetic that a well-written text (which the Torah ought to be!)
ought not to have redundancies is itself consistent with the Greco-Roman con-
text of the Palestinian tannaitic rabbis, and it is even taught in the progymnas-
mata. Thus, in the section on Narration of his progymnasmata, Aelius Theon
explains that the “Virtues [aretai] of a narration are three: clarity, conciseness,

M. Chernick, A Great Voice that Did Not Cease: The Growth of the Rabbinic Canon and its
Interpretation (Cincinnati 2009) 29–30, and 268.
83 See, J. N. Epstein, Mevo’ot le-Sifrut ha-Tanna’im: Mishnah, Tosefta u-Midrashei-Halakhah
(ed. E. Z. Melamed; Jerusalem 1957) 521–36; Albeck, Introduction to the Talmud, 39–43,
79–143; M. Elon, Jewish Law: History, Sources, Principles vol. 1 (transl. by B. Auerbach and
M. Sykes; Philadelphia, 1994) 374–80; and A. Yadin, Scripture as Logos: Rabbi Ishmael and
the Origins of Midrash (Philadelphia 2004).
84 See also MdRSbY 23:19; SifDeut 104; PRK 10; yAZ 5:12, 45b; bQid 57b; and bHul 115b.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 209

credibility,”85 and in his lesson on Fables, Theon explains that they can be criti-
cized, among other reasons, for redundancy.86
In his lesson on How to Introduce a Law, Theon explains that lack of clarity
can come, among other reasons, from polyonymy, “whenever only one thing is
signified but many names are used for it; for example, ‘weapon, blade, sword,
dirk.’ There is a lack of clarity when someone thinks there are as many signi-
fied [objects] as names.”87 That is, for Theon, listing so many synonyms leads
the reader to assume that each word is coming to add some extra element to
the overall rule. Poor legal writing includes multiple synonyms when only one
signified is intended, whereas good legal writing would only include multiple
words when each word is intended to have a separate signified. The rabbis, of
course, would come to read the Bible as intended to be interpreted in the latter
manner.
In his final lesson, explaining how to pose a challenge (counter-statement),
Theon instructs:

“Contradiction (antirrhêsis) is discourse that attacks the credibility

of another discourse.” Try to show that the other discourse is obscure,
impossible, incredible, deceitful, or inadequate in thought or expression;
or, conversely, redundant or lacking vigor, or confused, or that the dis-
course is contradictory, or departs from what is legal . . .88

Redundancy and contradiction are two of the main challenges that rabbinic
literature tends to pose, with the charge of redundancy mainly being reserved
for the biblical text, necessitating the obvious response that the text is not
redundant, but that each statement was needed for some further teaching.
What eventually was added to this Palestinian/Greco-Roman aesthetic
in amoraic and post-amoraic Babylonia was that each scriptural anomaly
could correspond to one and only one legal teaching. That is, in tannaitic and
amoraic Palestine, a single scriptural anomaly could result in as many laws
as the exegete required. By the time of the stam of the Bavli, however, this
had changed so that each new law could be justified by only one scriptural
anomaly. A perfect one-to-one correspondence was needed for laws to account
for scriptural anomalies and for scriptural anomalies to be the basis for laws
that otherwise were not grounded in the Torah. Rabbi Abraham ben David

85 Aelius Theon, Progymnasmata 79, translated in Progymnasmata (ed. Kennedy, p. 29).

86 Aelius Theon, Progymnasmata 77 (ed. Kennedy, p. 27).
87 Aelius Theon, Progymnasmata 130 (ed. Kennedy, p. 63).
88 Aelius Theon, Progymnasmata 111P (ed. Kennedy, p. 72).
210 BRODSKY

of Posquières (twelfth century, Provence) already noticed this difference,

although he attempted to harmonize it with the talmudic methodology.89 This
was a major change in rabbinic hermeneutics and would drive one classic type
of Bavli sugya.
For example, the Sifra (qedoshim, pereq 4:8) has no problem reading three
different laws out of the same statement in Lev 19:16, “do not stand upon the
blood of your fellow”:

:‫ אין אתה רשאי לשתוק עליה? תלמוד לומר‬,‫( ומנין שאם אתה יודע לו עדות‬A)
.)‫״לא תעמוד על דם רעך״ (ויקרא יט טז‬
,‫ או חיה רעה באה עליו‬,‫ או ליסטים באים עליו‬,‫( ומנין אם ראית טובע בנהר‬B)
.)‫ ״לא תעמוד על דם רעך״ (שם‬:‫חייב אתה להצילו בנפשו? תלמוד לומר‬
‫ חייב אתה‬,‫ ואחר נערה המאורס׳‬,‫ ואחר הזכור‬,‫( ומנין לרודף אחר חבירו להורגו‬C)
.)‫ ״ולא תעמד על דם רעך״ (שם‬:‫להצילו בנפשו? תלמוד לומר‬

(A) And from where do we derive [‫ ]ומנין‬that if you have testimony to

testify on someone’s behalf, that you are not free to keep silent? The
Torah teaches, “Do not stand on the blood of your fellow” (Lev 19:16).
(B) And from where do we derive [‫ ]ומנין‬that if you saw [your fellow]
drowning in the river, or bandits coming upon him, or a wild animal
coming upon him, that you are obligated to save him? The Torah
teaches, “Do not stand on the blood of your fellow” (Lev 19:16).
(C) And from where do we derive [‫ ]ומנין‬that if someone was chasing
after his fellow to kill him or after a male or after an engaged girl [to
rape them] that you are obligated to save him by means of his soul

89 See e.g. his commentary to Sifra, metzora‘, parasha 1:12, and the other citations in Albeck,
Introduction to the Talmud, 84, n. 12. Albeck correctly notes this point when quoting
R. Abraham b. David in n. 12, but then fails to apply the point in the body of his book
(Albeck, Introduction to the Talmud, 84–6). In n. 12 he states that “the Talmud” (presum-
ably the Bavli) differs from the works of Midrash Halakhah in that the Talmud does not
permit one law to be derived from multiple verses, while Midrash Halakhah does. In the
body of his book, Albeck implies that the Bavli permits a single verse to derive multiple
laws. He takes this from a quote attributed to Abbayye (fourth generation, Babylonian
amora) that “one verse goes out for many explanations [‫]טעמים‬, but a single explanation
[‫ ]טעם‬cannot be derived from multiple verses” (bSanh 34a). Whether or not Abbayye
intended this statement to be applied to halakhic contexts (or merely aggadic ones), the
stam of the Bavli clearly did not apply this hermeneutic principle to halakhic contexts,
as we shall see below. That is, the stam did not permit multiple laws to be derived from a
single verse. Whether Babylonian amoraim permitted it is a more complicated issue also
to be addressed below.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 211

[i.e. by killing him if necessary]?90 The Torah teaches, “Do not stand
on the blood of your fellow” (Lev 19:16).

The Sifra does not derive these three laws from three separate scriptural anom-
alies, redundancies, or parts of the verse. Rather, it derives all three from the
same part of the verse. That this rule changed drastically by the time of the
stam of the Bavli can be seen from bSanh 73a, in which two of the above three
statements appear:91

‫ מנין לרודף אחר חבירו להרגו שניתן להצילו‬:‫[ ת׳ר‬Statement and Support]
.)‫ ״ולא תעמד על דם רעך״ (ויקרא יט טז‬:‫בנפשו? ת׳ל‬
‫ מנין לרואה את‬:‫[ והאי להכי הוא דאת?א? האי מיבעי ליה לכד()[ת]ניא‬Challenge]
‫ שהוא חייב להצילו‬,‫ או לסטים באין עליו‬,‫ או חיה גוררתו‬,‫חבירו שהוא טובע בנהר‬
.)‫ ״לא תעמד על דם רעך״ (שם‬:‫בנפשו? ת׳ל‬
?‫ ואלא ניתן להצילו בנפשו מנא לן‬.‫ הכי נמי‬.‫[ אין‬Rejection of Statement]

[Statement and Support] Our rabbis taught [‫]תנו רבנן‬: From where do
we derive [‫ ]מנין‬that if someone was chasing after his fellow to kill him
that he is given to save him by means of his soul [i.e. by killing him
if necessary]? The Torah teaches: “Do not stand on the blood of your
fellow” (Lev. 19:16).
[Challenge] But is it for this that [the verse] came? This [verse] is
needed [for the following law] as it is taught [‫]לכדתניא‬: From where
do we derive [‫ ]מנין‬that one who sees his fellow drowning in a river, or
a wild animal dragging him, or bandits coming upon him, that he is
obligated to save him by sacrificing himself [if need be]?92 The Torah
teaches: “Do not stand on the blood of your fellow” (Lev 19:16).
[Rejection of Statement] Yes, indeed it is so [‫ הכי נמי‬.‫]אין‬. Rather, that
[in the case of the pursuer] he is given to save him by means of his
soul, from where do we derive this [‫ ?]מנלן‬. . .

90 That is, one saves the perpetrator from the sin by killing him before he commits it.
91 The text and my translation are from MS Yad ha-Rav Herzog.
92 I am here staying faithful to MS Yad ha-Rav Herzog, which adds the word ‫בנפשו‬, “by
means of his soul,” here. This is my best understanding of what this would mean in con-
text. Nevertheless, the word ‫ בנפשו‬may well have been added by error from the phrase as
it appears above, since it is not to be found in the Sifra. MS Karlseruehe supports MS Yad
ha-Rav Herzog’s reading.
212 BRODSKY

Whereas the Sifra was content to list these statements next to each other, see-
ing no apparent conflict in them, the stam of the Bavli quotes one (B in the
Sifra) to disprove the other (C in the Sifra). They accept the difficulty and reject
the first baraita (the Sifra’s C) based on the fact that the verse was already used
to derive a different law (the Sifra’s B).93 The sugya goes on to derive the law
from a different verse.
A typical genre of the sugya in both the Yerushalmi and the Bavli is one in
which the rabbis attempt to find the verses that underlie the laws delineated
in the Mishnah. If there is a dispute in the Mishnah or if a conflict of law or
hermeneutic is perceived to exist, the sugya will often attempt to explain what
each side does with the other’s verse, and then how the other side derives that
law, if not from that verse, and so forth. A prime example of this genre as found
in the Bavli can be found in bNid 40a:94

.‫מתני‬
.‫ ואין חייבין עליו קרבן‬,‫( יוצא דופן אין יושבין עליו ימי טומאה וימי טהרה‬X)
.‫ הרי זה כילוד‬:‫( ר״ש אומר‬Y)

.‫גמ׳‬
‫ עד שתלד‬.‫ ״אש׳ תזרי׳ וילד׳ זכר״‬:‫> מאי טעמ׳ דרבנן? א׳ קר׳‬..<‫(  מני בר פ‬1)
.‫ממקו׳ שמזרע׳‬
?‫א( ור׳ שמע׳‬2)
.‫ אמו טמא׳ ליד׳‬,‫ אל׳ כעין שהזריע׳‬,‫ב(  ההי׳ אפי׳ לא ילד׳‬2)
.‫ לרבו׳ יוצ׳ דופן‬,)‫ ״תלד״ (ויקרא יב ה‬:‫ א׳ קר׳‬:‫(  ור׳ שמע׳ מאי טע׳? א׳ ריש לקי׳‬3)
‫ דסד״א זכר ונקב׳ כתי׳? ״זכר״‬.‫(  ורבנן? האי מיבעי להו לרבו׳ טומטו׳ ואנדרו׳‬4)
.‫ קמ״ל‬.‫ ולא טומטו׳ ואנדרוגי׳‬,‫(שם ב ז) ודאי ״נקב׳״ (שם ה ז) ודאי‬
.‫ ״לבן״ (שם ו) לבן מ״מ‬:‫ תני בר ליואי‬.‫(  ור׳ שמע׳? נפק׳ ליה מדתני בר ליואי׳‬5)
.‫״לבת״ (שם) לבת מ״מ‬

93 For the nature of various kinds of talmudic passages that comment on the Sifra, see
Neusner, Talmud, 21–51. Neusner sees passages that comment on the Sifra independently
of the Mishnah as part of an earlier Talmud on the Sifra that has been preserved in the
Bavli, while he sees those passages that comment on the Sifra as part of its discussion of
the Mishnah as authored by the editors of the Bavli.
94 My translation is from the Munich MS. This type of sugya can be found throughout the
Bavli. Too many examples exist to cite them all here. See, e.g. bBer 13a, 50a; bShab 131b;
bEruv 15b; bPes 28b; bSuk 24b, 43a; bMQ 8a; bYev 8b–9a, 101a–b; bSot 32b–33a, 44b–45a,
45a (second sugya); bGit 21b; bQid 14b–15a; and bTem 7a. Similarly, see bPes 77a–b and
bQid 9b–10a. A related though different version of it can be found in the Yerushalmi, and
is exemplified by yNed 1:1, 36c quoted below—for a list of examples of this type of sugya
in the Yerushalmi and its contemporary midrashic literature, see n. 109 below.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 213

.‫(  ורבנן? ההו׳ מיבעי להו לחייב על כל בן ובן ועל כל בת ובת‬6)

:‫ דתני תנ׳ קמי דרב ששת‬,‫(  ור׳ שמע׳? נפק׳ ליה מדתני תנ׳ קמי׳ דרב שש׳‬7)
‫ יכול‬.‫ מלמ׳ שמביאה קרבן אחד על יולד׳ הרבה‬,)‫״תור׳ היולד׳ לזכר״ (שם ז‬
.)‫ ״זאת״ (שם‬:‫תבי׳ על ליד׳ שלפני מלאת ועל ליד׳ שלאחר מלאת כאחד? ת״ל‬
,‫ סד״א ה״מ בתרי עיבורי‬.‫ ״זאת״ (שם) איצטריך לבן או לבת‬:‫(  ורבנן? אע״ג דכתי׳‬8)
.‫ קמ״ל‬.‫ אימ׳ בחד קרבן תסגי‬,‫ כגון יהוד׳ וחזקיה בני ר׳ חייא‬,‫אבל בחד עיבורא‬

Mishnah:
(X) [Statement] One born by caesarean section, [the mothers] do not
sit for him the days of impurity or the days of purity [as prescribed
in the Torah], and one is not obligated [to bring] a sacrifice for him
[i.e. the sacrifice that is brought when a child is born].
(Y) Rabbi Shimeon says: Behold this one is like one born [by natural
means].

Gemara:
(1) [Support for X] Rabbi Mani bar Pa[tish said]:95 What is the
[scriptural] basis for the rabbis [‫ ?]מאי טעמייהו דרבנן‬The Bible states
[‫]א(מר) קרא‬, “A woman, when she gives forth seed and bears a
male” (Lev 12:2). [She is not obligated to bring the sacrifice] until
she gives birth from the place that she emits seed.
(2a) [Challenge to Y] And Rabbi Shimeon?
(2b) [Resolution] That [verse is needed for:] even if she only gave birth
like her seed [i.e. even if she miscarried and the fetus was still in an
early stage of development, and looked more like semen than like a
baby], its mother would still be impure.
(2) [Support for Y] And Rabbi Shimeon, what is his [scriptural] basis
[)‫ ?]ור׳ שמע׳ מאי טע(מיה‬Resh Laqish96 said: The Bible said [‫אמר‬
‫]קרא‬: “she shall give birth” (Lev 12:5). To include [‫ ]לרבות‬one born
by caesarean section.

95 MS Munich is hard to read here. I have filled in the missing material from the other MSS
and editions.
96 MS Vatican 111 reads, “Rabbi Shimeon said” The rest of the MSS and printed editions, how-
ever, all read: “Resh Laqish” here. I am following the Munich MS here as I do throughout
this sugya. Resh Laqish is sometimes referred to as Rabbi Shimeon ben Laqish, which
likely accounts for the anomalous variant in MS Vatican 111. In the continuation of the
sugya, Rabbi Yohanan picks up the discussion, supporting the likelihood that Resh Laqish
is credited in the text with this statement, since the two of them are often found together
in the same sugya.
214 BRODSKY

(4a) [Challenge to X] And the rabbis [what do they do with this

word]?
(4b) [Resolution] They need it [‫ ]האי מבעי ליה‬to include the tumtum
[a person with ambiguous genitalia] and the androgynus. For, you
might think [that since] it is written “male and female” (Lev 12:2,
5, and 7), [this implies] one who is definitely male and definitely
female, and not the tumtum and the androgynous. Therefore it
teaches us [“she shall give birth” to teach you to include the tumtum
and the androgynous in the law].
(5a) [Challenge to Y] And Rabbi Shimeon? [How does he derive this
law?]
(5b) [Resolution] He derives [the law] from [‫ ]נפק(א) ליה מדתני‬that
which Bar Livai taught. For Bar Livai taught: “as a son” (Lev 12:6), as
a son in any case [i.e. even if he is an androgynous or tumtum], “as
a daughter” (Lev. 12:6) as a daughter in any case [even if she is an
androgynus or tumtum].
(6a) [Challenge to X] And the rabbis? [What do they do with the
words “as a son” and “as a daughter”?]
(6b) [Resolution] They need it [‫ ]מבעי להו‬to obligate [a sacrifice for]
every son and every daughter.
(7a) [Challenge to Y] And Rabbi Shimeon? [From what scriptural
source does he obligate the case of every son and daughter?]
(7b) [Resolution] He derives this from [‫ ]נפקא ליה מדתני‬that which
the reciter [‫ ]תנא‬taught before Rav Sheshet. For the reciter taught
before Rav Sheshet: “[This is] the teaching of the one who gives
birth” (Lev 12:7). This teaches that she brings one sacrifice [even]
for multiple births [in the same pregnancy].97 Could it be that she
brings one sacrifice for a birth that is before the days are fulfilled
and for a birth that is afterwards? The Torah teaches: “This” (Lev
12:7) [i.e. she must bring a separate sacrifice for each].
(8a) [Challenge to X] And the rabbis? [What do they do with this
word?]

97 The printed editions add here: “Could it be that she brings one sacrifice for a birth and
for a gonorrheal flow? Rather, a woman who gave birth who ate blood, and one who ate
forbidden fat, will one sacrifice satisfy them? Rather . . .” (‫יכול תביא על לידה ועל זיבה‬
. . . ‫ בחד קרבן תסגי לה? אלא‬,‫ ויולדת דאכלה חלב‬,‫)כאחת? אלא יולדת דאכלה דם‬. It may
be an addition in the printed editions or it may have fallen out of the MSS by means of
homoiteleuton. MS Vatican 113 has the passage added in the margins.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 215

(8b) [Resolution] Even though it says: “This” (Lev 12:7), “as a son” (Lev
12:6) and “as a daughter” (Lev 12:6) are still needed [‫—איצטריך‬to
derive the law that multiple births from the same pregnancy require
separate sacrifices]. [If you only had the word “This”], you might
have thought [‫]סלקא דעתך אמינא‬: This is regarding a case in which
there were two pregnancies [e.g. the twins were conceived at sepa-
rate times] but in one pregnancy, like Yehudah and Hizqiyah the
sons of Rabbi Hiyya [who, though born several months apart, were
conceived in the same pregnancy—see bYev 65b and bNid 27a], I
might say that one sacrifice would suffice. Therefore it teaches us
[‫“—קא משמע לן‬as a son” and “as a daughter” to teach us that one
sacrifice would not suffice even in such a situation].

Put in a generic format, this sugya looks as follows:

Mishnah:
Rabbi X (Rabbanan in this sugya): Law is A
Rabbi Y (Rabbi Shiemon in this sugya): Law is B (= not A)

Gemara:
(1) From where does Rabbi X derive Law A? From Verse 1
(2) Then what does Rabbi Y do with Verse 1? He uses it to derive Law C.
(3) From where does Rabbi Y derive Law B? From Verse 2.
(4) What does Rabbi X do with Verse 2? He uses it for Law D.
(5) Then from where does Rabbi Y derive Law D? From Verse 3.
(6) What does Rabbi X do with Verse 3? He needs it for Law E.
(7) Then from where does Rabbi Y derive Law E? From Verse 4.
(8) What does Rabbi X do with Verse 4? (He holds that) Verse 4 is not
enough to derive Law E on its own, requiring Verse 3 for clarification.

This type of sugya keeps going until it is able to come to a resolution like the
one found in number 8. Many people misunderstand this type of sugya, assum-
ing that it sets out to prove one position and disprove the other. The ending of
such sugyot, then, seems anti-climactic: the sugya ends without determining
who won. Some conclude from this that the Bavli does not like to wrap things
up in a clean way; the Bavli avoids neat conclusions. They conclude from this
that the Bavli is pluralistic.98

98 See Kraemer, Mind, esp. pp. 103–4 and 126. See also Yaakov Elman’s review article
(“Argument for the Sake of Heaven: ‘The Mind of the Talmud’,” The Jewish Quarterly
Review 84 [1993–4] 261–82), in which he critiques Kraemer on this point.
216 BRODSKY

Yet, seen from a different light, the sugya wraps up quite cleanly and has a
very neat conclusion. The concern of the stam of the Bavli is not to determine
who won, whose halakhah is authoritative. After all, mishnaic hermeneutics
already determine that the majority opinion (Law A) trumps the minority
opinion (Law B).99 The law was never in doubt. Rather, the stam’s concern is to
ensure that both sides’ systems are sound. This is not legal pluralism, but a need
to believe that no rabbi would be unintelligent enough to have an unsound
legal system. In other words, I am arguing that the motivation behind such sug-
yot in the Bavli is neither to challenge the assumption that “truth” can be iden-
tified, as David Kraemer would have it,100 nor is it for “legal determination,”101
but rather, as Elman has argued, to defend the intelligence of the stam’s prede-
cessors and the internal consistency of those rabbis’ respective legal systems.102
Here, the progymnasmata can once again help us understand the genre. The
goal of the exercise on How to Introduce a New Law in the progymnasmata is
not to decide law, but to introduce it. That is, it is not to engage in an actual
debate, coming to a final juridical decision at its completion, but to introduce
a law with its support and enough challenges resolved that the audience can
leave feeling that the law indeed held up to scrutiny. In this modified genre of
the talmudim, we encounter something of a mash-up of the introduction of
two competing laws. Nevertheless, the goal remains to leave both introduced
properly, not to determine which one wins. As a mash-up, each acts as the chal-
lenge to the other, and each resolution becomes at the same time a challenge
to the other. Yet the very fact that the sugya only ends when both systems are
left intact shows us that they have not strayed far from the intention of the les-
son in the progymnasmata.
To spell this out, if Rabbi X agrees that Verse 2 comes to teach Law B, then he
cannot hold that the law is A (= not B). Therefore, (in number 4) he must use
Verse 2 to derive some other law (Law D). Once he does so, Rabbi Y must either
not hold by Law D, or he must derive Law D from a different verse (Verse 3; see
number 5 above). On and on this goes until the author can bring the sugya to

99 See Eduy 1:4–6.

100 Kraemer, Mind, esp. pp. 6–7.
101 See Elman, “Argument,” 273.
102 Elman, “Argument,” 275–7. Consistency was important in the Roman legal system as well,
as Quintilian explains; see, e.g. Clark, Rhetoric, 207, quoting Quintilian, Institutio Oratoria,
2.4.37: “for the fault of any law must be either in words or in matter . . . As to matter,
whether the law is consistent with itself.” See also, 2.4.35. The progymnasmata pick up
on the issue of consistency as well. See exercises 2, 12, and 17 of Aelius Theon in Kennedy,
Progymnasmata, 13, 65, and 72, respectively.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 217

a resolution like that found here in number 8, in which the sugya comes to a
place in which both Rabbi X’s system and Rabbi Y’s system work. Rabbi X can
hold by Laws A, D, and E, deriving Law A from Verse 1, Law D from Verse 2, and
Law E from Verses 3 and 4.103 Rabbi Y can hold by Laws B, C, D, and E, deriving
Law B from Verse 2, Law C from Verse 1, Law D from Verse 3, and Law E from
Verse 4. While their systems result in slightly different laws, and derive the laws
they share from different verses, by number 8, both are able to explain what
they do with each of the verses and how they derive each of the laws. From
the perspective of the lesson in the progymnasmata, we have effectively intro-
duced both rabbis’ competing legal systems. Determining which law is legally
binding and which is not binding is not the purpose of the genre.
At the same time, this Babylonian genre differs from its Greco-Roman and
Palestinian forebears. Consistent with the stricter rules of redundancy in the
Bavli’s stam” seen above in bSanh 73a, in this sugya, as in others of its kind
in the Bavli, at no time is a scriptural anomaly used to derive more than one
law.104 Interestingly, this requirement can only be identified in this sugya to
be coming from its stammaitic layer. Numbers 1 and 3 are the only parts of the
sugya attributed to amoraim, and both come to answer only the most basic
of questions: From where does the sage in the mishnah derive his law? While
tannaim and amoraim can (and are) quoted in other parts of the sugya, they
seem to be brought into the sugya by the stam. That is, without the stam to
bring them there, they do not on their own seem to be engaged in a discussion
of this particular mishnah. The two amoraic sages quoted in numbers 1 and 3,
however, are directly engaged in inquiry on this mishnah, and they take the
sugya to its first level: the scriptural derivation of both laws. Yet, we do not
find amoraim engaged in this sugya in the next level of discourse: determin-
ing what the other sage does with those two verses.105 Beginning in Babylonia
in the third amoraic generation, we find amoraim arguing that a verse is not
free to teach one law because it was already used to teach another law. Thus,
the third-generation Babylonian amora, Rav Safra, makes such an argument
in bZev 7b, and Rav Yaaqov the brother of Rav Aha bar Yaaqov (also a third

103 That Law E is derived from verses 3 and 4 is not breaking the rule that one verse cannot
be derived from multiple verses, since the stam goes to great pains to prove that one verse
alone would not have been enough to prove this Law. The multiple verses are not redun-
dant but mutually necessary.
104 For the stam of the Bavli, more than one law may be derived from a single verse, but
always from distinct parts or anomalies within that verse.
105 At least, I have not found any sugyot in which amoraim engage in such an activity. For a
list of relevant sugyot, see n. 94 above.
218 BRODSKY

generation Babylonian amora) makes the same argument in bSanh 76a. Both
use the phrase . . . ‫ האי להכי הוא דאתא? האי מיבעי ליה ל‬to raise this objection.
Otherwise, this phrase is used exclusively by the stam of the Bavli.106 While this
new hermeneutical rule did not develop in the amoraic period into the kinds
of complex sugyot into which the stam of the Bavli would later develop it, a key
ingredient had been put in place.107

106 The stam uses this phrase in bBer 35a; bPes 24a, 44a; bMQ 5a; bQid 4a, 14a, 17b, 57b; bBQ
64b; bSanh 73a; bMak 14b; bShevu 14a; bAZ 51a; bZev 4a, 4b, 33b, 36a, 36b–37a, 49a; bMen
51b, 55a, and 72b. See also bEruv 58a; bYom 27a; bMeg 8b; bQid 9b; bBM 11a; bSanh 42b, 46b,
68b, 71b; bAZ 5b; bZev 11b, 56a–b, 106a; bMen 61a; bHul 82b; bBekh 16a; bTem 7a, and the
examples in n. 94 above. All of these examples are stam except for bSanh 76a and bZev 7b
mentioned above. This strongly suggests that while the concept may have existed in the
amoraic period, it did not take hold until post-amoraic times.
While these two amoraic cases stand alone (as far as I have found), MS evidence none-
theless supports both cases. In bZev 7b, all of the MSS (Munich, Columbia, and Vatican)
agree that Rav Safra raised this question using the language ‫ להכי הוא דאתא? האי‬. . . ‫האי‬
. . . ‫מיבעי לי(ה) לכ‬. Similarly, all of the MS evidence for bSanh 76a (Florence, Jerusalem,
Karslruhe, Munich, and Barko) agrees that an amora posed the challenge that the verse
was already used and therefore not free to teach the point. All of the MSS except MS
Jerusalem record that this amora was Rav Yaaqov the brother of Rav Aha bar Yaaqov,
while MS Jerusalem records the sage as Rav Yaaqov the father of Rav Aha bar Yaaqov. This
would change the result by one generation, but still identify the sage within the same
family. Likewise, all of the MSS except one record the phrase he used as ‫ להכי הוא‬. . . ‫האי‬
‫דאתא? האי מיבעי ליה לכדתניא‬. The only exception is MS Karlsruhe, which lacks the
question ?‫ להכי הוא דאתא‬. . . ‫האי‬, though this does not affect the overall meaning. With
or without this question, the sage’s statement still challenges how the verse can be used
for one law since it has already been used to teach another. The variant in MS Karlsruhe
is likely the result of homoioteleuton from one ‫ האי‬to the next, and therefore the error
likely lies with that MS.
107 The talmudic phrase “Is it for this [law] that this [verse] comes? This [verse] is needed
for . . . [. . . ‫ ”]האי להכי הוא דאתא? האי מיבעי ליה ל‬ostensibly sounds similar to the tan-
naitic phrase (found especially in those midrashim attributed to the school of Rabbi
Yishmael) “Do you say X? Or is it rather Y? [. . . ‫ או אינו אלא‬. . . ‫]אתה אומר‬,” which also
challenges whether the verse ought to be read to teach Law A when it seems to teach Law
B. Nevertheless, they are different on exactly the point that this section of the present
study is addressing. The talmudic phrase (. . . ‫)האי להכי הוא דאתא? האי מיבעי ליה ל‬,
which is unique to the Bavli, claims that a verse has already been used, and therefore is
not free to teach this other principle. The tannaitic phrase, however, simply questions
whether the suggested interpretation of the verse is correct, since it seems to say some-
thing else. It does not claim that the verse cannot be read the former way (or both ways),
but merely that since it can more effectively be read another way, the exegete needs to
justify his/her reading. This is followed by a proof that supports the exegete’s original
interpretation. This subtle difference is extremely significant for the purposes of the pres-
ent study.
‫‪FROM DISAGREEMENT TO TALMUDIC DISCOURSE‬‬ ‫‪219‬‬

‫‪An interesting intermediary can be found in the stam of the Yerushalmi,‬‬

‫‪which shows a concern with what each does with the other’s verses and how‬‬
‫‪each derives the other’s laws, but allows multiple laws to be derived from a‬‬
‫‪single verse or scriptural anomaly. An excellent example of this can be found‬‬
‫‪in the opening sugya of yNed 1:1, 36c:‬‬

‫משנה‪ :‬כל כינויי נדרים כנדרים‪ ,‬חרמים כחרמי׳‪ ,‬שבועות כשבועות‪ ,‬נזירות כנזירות ‪. . . .‬‬
‫גמרא‪:‬‬
‫)‪(1‬‬
‫)‪ (Α‬כתיב‪ :‬״איש כי ידר״ (במדבר ל ג)—מה תלמוד לומר ״נדר״? אלא‪ ,‬מיכן‬
‫שכינויי נדרי׳ כנדרים‪.‬‬
‫)‪  (Β‬״או השבע״ (שם)—מה תלמוד לומר ״שבועה״? אלא‪ ,‬מיכן שכינויי שבועות‬
‫כשבועות‪.‬‬
‫)‪ (C‬״אך כל חרם״ (ויקרא כז כח)—מה ת״ל ״יחרים״? אלא‪ ,‬מיכן שכינויי‬
‫חרמים כחרמים‪.‬‬
‫)‪ (D‬״נדר נזיר״ (במדבר ו ב)—מה תלמוד לומ׳ ״להזיר״ (שם)? אלא‪ ,‬מיכן‬
‫שכינויי נזירות כנזירות‪.‬‬
‫)‪ (2‬עד כדון כר׳ עקיבה‪ ,‬דאמ׳‪ :‬לשונות ריבויין הן‪ .‬כר׳ ישמעאל‪ ,‬דאמ׳‪ :‬לשונו׳ כפולין‬
‫הן והתורה דיברה כדרכה‪ :‬״הלוך הלכת״ (בראשית לא ל)‪ ,‬״נכסף נכספתה״ (שם)‪,‬‬
‫״גונב גונבתי״ (שם מ טו)‪ ,‬מנלן?‬
‫)‪(3‬‬
‫)‪ (A+B‬״איש כי ידר נדר ליי׳ או השבע שבועה לאסור אסר על נפשו לא יחל‬
‫דברו״ (במדבר ל ג)—מה תלמוד לומר ״ככל היוצא מפיו יעשה״ (שם)? אלא‪,‬‬
‫מיכן שכינויי נדרים כנדרים וכינוי שבועה כשבועה‪.‬‬
‫)‪ (C‬ומניין שכינוי חרמים כחרמים? ״נדר נדר״ (ויקרא כז ב)—מה ״נדר״ שנאמ׳‬
‫להלן (במדבר ל ג) כינוי נדרים כנדרים וכינוי שבועה כשבועה‪ ,‬אף ״נדר״‬
‫שנאמר כאן כינוי חרמין כחרמין‪.‬‬
‫)‪ (D‬ומניין שכינויי נזירות כנזירות? ״נדר נדר״ (ויקרא כז ב)—מה נדר שנאמר‬
‫להלן (במדבר ל ג)‪ ,‬כינוי שבועה כשבועה‪ ,‬אף נדר שנאמר כאן כינוי נזירות‬
‫כנזירות‪.‬‬
‫)‪ (4a‬מה מקיים רבי עקיבה ״ככל היוצא מפיו יעשה״ (שם)?‬
‫)‪ (4b‬מיכן לנדר שבטל מקצתו‪ ,‬בטל כולו‪.‬‬
‫)‪  (5a‬ולית ליה לרבי ישמעאל כן?‬
‫)‪ (5b‬כולה מן תמן‪ .‬אית ליה—מיכן שכינוי נדרים כנדרים‪ ,‬וכינוי שבועות כשבועות‪.‬‬
‫אית ליה—מיכן לנדר שבטל מקצתו‪ ,‬בטל כולו‪.‬‬
‫)‪ (6a‬מה מקיים רבי ישמעאל ״נדר נזיר להזיר״ (שם ו ב)?‬
‫)‪ (6b‬מיכן שאדם קובע עליו נזירות בתוך נזירותו‪.‬‬
‫)‪ (7a‬ולית לרבי עקיבה כן?‬
‫)‪ (7b‬אית ליה‪ .‬כולה מתמן‪ .‬אית ליה—מיכן שאדם קובע עליו נזירות בתוך נזירותו‪.‬‬
220 BRODSKY

Mishnah: All euphemisms for vows are like vows, for herem are like herem,
for oaths are like oaths, for Nazirite vows are like Nazirite vows. . . .

Gemara:
(1) [Support X]
(A) It is written, “If a man vows . . .” (Num 30:3). Why does the
Torah say “. . . a vow” (Num 30:3)? Rather, from here [we
learn] that euphemisms for vows are like vows.
(B) “. . . or swears . . .” (Num 30:3). Why does the Torah say “. . . an
oath” (Num 30:3)? Rather, from here [we learn] that euphe-
misms for oaths are like oaths.
(C) “But every ostracized thing [herem]” (Lev 27:28)—Why
does the Torah say “[that] he shall ostracize”? Rather, from
here [we derive] that all euphemisms for herem are like
herem.
(D) “A vow of nazir” (Num 6:2)—Why does the Torah say, “to
nazir”? Rather, from here [we derive] that all euphemisms
for Nazirite vows are like Nazirite vows.
(2) [Question] So far, [we have explained] according to Rabbi Aqiva.
For he said: [Doubled] language is inclusive [of extra laws]. But for
Rabbi Yishmael, who said: They are [merely] double language, and
the Torah was speaking according to its fashion,—[e.g.] “get going
[halokh halakhta]” (Gen 31:30), “you have truly shamed [nikhsof
nikhsaftah]” (Gen 31:30), “I have been stolen” [gunav gunavti] (Gen
40:15), from where [does he derive these laws]?
(3) [Support Y]
(A+B) “When a man vows a vow to the Lord, or swears an oath
to forbid something upon himself, he shall not violate his
word” (Num 30:3). Why does the Torah say, “according to all
that goes out of his mouth shall he do” (Num 30:3)? Rather,
from here [we derive] that euphemisms for vows are like
vows and euphemisms for oaths are like oaths.
(C) And from where [do we derive] that euphemisms for herem
are like herem? “Vow” (Lev 27:2) “Vow” (Num 30:3)—Just
as [the word] “vow” said there [in Num 30:3 implies that]
euphemisms for vow are like vows, so [the word] “vow” said
here [in Lev 27:2 implies that] euphemisms for herem are
like herem.
(D) And from where do we derive that euphemisms for Nazirite
vows are like Nazirite vows? “Vow” (Num 6:2) “Vow”
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 221

(Num 30:3)—Just as [the word] “vow” said there [in Num 30:3
implies that] euphemisms for oaths are like oaths, so [the
word] “vow” said here [in Num 6:2 implies that] euphemisms
for Nazirite vows are like Nazirite vows.
(4a) [Challenge to X] What does Rabbi Aqiva do with [‫מה מקיים רבי‬
‫“ ]עקיבה‬according to all that goes out of his mouth shall he do”
(Num 30:3)?
(4b) [Resolution] From here he derives that [‫ ]מיכן ל‬a vow that part of
it has been nullified, all of it has been nullified.
(5a) [Challenge to Y] Does Rabbi Yishmael not hold thus?
(5b) [Resolution] He derives both from there [‫כולה מן תמן‬, i.e. from the
same verse]. He holds [‫ ]אית ליה‬from here we derive [‫ ]מיכן ש‬that
euphemisms for vows are like vows and euphemisms for oaths are
like oaths, [and] he holds from here [we derive] that a vow that part
of it has been nullified, all of it has been nullified.
(6a) [Challenge to Y] What does Rabbi Yishmael do with [‫מה מקיים רבי‬
‫“ ]ישמעאל‬a vow of nazir to nazir” (Num 6:2)?
(6b) [Resolution] From here he derives [‫ ]מיכן ש‬that one may make
himself a Nazirite during his Nazirite vow.
(7a) [Challenge to X] Does Rabbi Aqiva not hold thus?
(7b) [Resolution] He does [‫]אית ליה‬. He derives both from there [‫כולה‬
‫]מן תמן‬. He holds [‫ ]אית ליה‬From here we derive that [‫ ]מיכן ש‬one
may make himself a Nazirite during his Nazirite vow. [And he holds:
From here we derive that all euphemisms for Nazirite vows are like
Nazirite vows.]

Put in generic format, the sugya looks as follows:

Mishnah:
Law A, Law B, Law C, and Law D.

Gemara:
(1) (Rabbi X):
(A) Law A from Part 1 of Verse 1.
(B) Law B from Part 2 of Verse 1.
(C) Law C from Verse 2.
(D) Law D from Verse 3.
(2) This works for Rabbi X, but how does Rabbi Y derive these laws
(since he has a different hermeneutic)?
222 BRODSKY

(3) Rabbi Y:
(A+B) Laws A and B from Part 3 of Verse 1.
(C) Law C from the fact that Verses 2 and 1 share the same word
(neder).
(D) Law D from the fact that Verses 4 and 1 share the same word
(neder).
(4) What does Rabbi X do with Part 3 of Verse 1? He uses it to derive
Law E.
(5) Doesn’t Rabbi Y hold by Law E? He derives both from there (i.e.
Laws A+B and Law E from Part 3 of Verse 1).
(6) What does Rabbi Y do with Verse 3 (specifically, the part of Verse 3
used by Rabbi X in number 1D)? He uses it to derive Law F.
(7) Doesn’t Rabbi X hold by Law F? He derives it, too, from Verse 3
(along with Law D).

In number 1, the stam lays out a midrash/halakhah-style baraita (or pseudo-

baraita) that derives each of the four laws delineated in the mishnah from four
scriptural sources (two of which are from different parts of the same verse).
Number 2 states that the derivation works for Rabbi Aqiva, but that Rabbi
Yishmael’s hermeneutics do not allow for that derivation. Number 3 thereby
quotes a baraita or pseudo-baraita that finds scriptural bases for all four laws
that are consistent with Rabbi Yishmael’s hermeneutic system. So far, whether
the stam of the Yerushalmi has quoted actual baraitot or made up pseudo-
baraitot,108 this layer of the sugya has done little more than what the amoraic
layer of bNid 40a had done (numbers 1 and 3 in that sugya): find the verses that
two different tannaim use to derive a law or laws.
Of course, the two sugyot differ insofar as the mishnayot differ. The mishnah
in Tractate Niddah had opposing positions regarding the law. In Nedarim, the
Mishnah records a single law, and it is only in the gemara of the Yerushalmi
that two different tannaitic positions are expressed—and even that dispute is
not about the law but about the hermeneutical principles to be used to derive
that law. While bNid 40a and yNed 1:1, 36c differ in this technical way, they are
similar in the key way that they each contend with two conflicting tannaitic
positions on which verses are able to give rise to which laws. This affords us the
opportunity to compare and contrast those systems.
In numbers 4 through 7, the stam of the Yerushalmi attempts to determine
what each sage would do with the other’s verse or law. While it is uncannily
similar to the genre found in the Bavli (exemplified here in bNid 40a), yNed

108 I found no known parallels, although this is not decisive.

FROM DISAGREEMENT TO TALMUDIC DISCOURSE 223

1:1, 36c has one important difference: multiple laws can be derived from the
same verse or part of the verse. In numbers 5 and 7, the stam of the Yerushalmi
finds an easy way of resolving the potential conflict of the sugya by stating that
Rabbi Yishmael (in number 5) and Rabbi Aqiva (in number 7) simply derive
both laws from the same place. Such a solution would not have been accept-
able to the stam of bNid 40a.109
In fact, yNed 1:1, 36c has something of a parallel in bNed 3a–b, which is gen-
erally similar to its Yerushalmi counterpart, with one of the chief exceptions
being that it does not permit two laws to be derived from a single scriptural
anomaly. Significantly, bNed 3a–b retains the resolution that two laws were
derived from the same verse (as in yNed 1:1, 36c, numbers 5 and 7), but with
the subtle yet all-important alteration that the two laws derive from separate
anomalies, albeit in the same verse. In this way, while the stam of the Bavli
retains the resolution, it has made sure to modify it to fit the new rules. The
passage is too long to deal with in full here. The heart of the parallel can be
found in the following excerpt from bNed 3a–b:110

,‫(  ומאי שנא גבי נזירות‬1a)

)‫(  דכת׳ ״נזיר להזיר״ (במדבר ו ב‬1b)
?‫ הקישא למה לי‬.)‫( נדרים נמי הכת׳ ״לנדור נדר״ (שם לג‬2a)
‫ ״נזיר להזיר״ (שם ו ב)? מי? נדר קאמרת ולא צריך‬:‫ כי דכת׳‬,‫ נדר לנדור‬:‫( אי כת׳‬2b)
.‫ דברה תורה בלשון בני אדם‬,)‫ ״לנדור נדר״ (שם ל ג‬:‫ דכת׳‬,‫ השתא‬.‫הקישא‬

109 Like yNed 1:1, 36c; yNed 10:1, 41d also states outright that the sage derives both laws from
the same scriptural anomaly (that is, it says explicitly ‫כולה מתמן‬, “[he derives them] both
from there”). In addition, the Yerushalmi derives multiple laws from single scriptural
anomalies in other passages as well—even if it does not state as categorically that that
is what it is doing, as it did in yNed 1:1, 36c and 10:1, 41b. See, e.g. yPes 9:2, 36d and yYom
4:6, 42a; 5:4, 42c–d and 8:1, 44d. Midrash Halakhah does so extensively, including in the
Sifra passage quoted above. The Yerushalmi and amoraic Palestinian works of Midrash
Aggadah have numerous passages in which they ask what each sage does with the other
sage’s verse. When such sugyot appear in the Bavli, it often provokes the stam of the Bavli
to ask from which verse each sage learns the other law, since that verse is no longer free
for that sage to teach from it the other law, as we saw in bNid 40a. In amoraic Palestinian
sources, this additional development never occurs. For examples of the stam of the
Yerushalmi asking what each sage does with the other sage’s verse without going on to ask
how each sage then derives the other sage’s subsequent law or exegetical point, see, e.g.
yBer 1:3, 3b; yPea 6:1, 14b; yMS 1:1, 52c; yHal 2:1, 58b; yShab 1:5, 3d; 9:1, 11d; yPes 7:6, 34c; yYom
8:7, 45c; ySot 3:2, 18d and yQid 1:8, 61c–d = Shevi 6:1, 36b; for this general phenomenon in
amoraic works of Midrash Aggadah, see, e.g. GenR 71:9, LevR 9:6 = SongR 4:16.
110 Translated from MS Vatican 487.
224 BRODSKY

‫ לא דברה‬:‫ אלא למאן דאמ׳‬,‫ דברה תורה כלשון בני אדם‬:‫( הניחא למאן דאמ׳‬3a)
?‫תורה כלשון בני אדם ״לנדור נדר״ (שם) מאי עביד ליה‬
)‫( דדריש ליה לעשות ידות כנדרים ומקיש נזירות לנדרים ״ונזיר להזיר״ (שם ו ב‬3b)
.‫דריש ליה (מ)ללמד שהנזירות חלה על הנזירו׳‬
‫ ״ונזיר להזיר״ דרשיה לעשות ידות‬,‫ דברה תורה כלשון בני אדם‬:‫( ולמאן דאמ׳‬4a)
‫ מלמד שהנזירות חלה על הנזירות מנא ליה? הא ניחא אי‬,‫נזירות כנזירות‬
:‫ אלא אי סבירא ליה כמאן דאמ׳‬,‫סבירא ליה דאין נזירות חלה על הנזירות‬
?‫ מנא ליה‬,‫נזירות חלה על הנזירות‬
.‫ מאי ״להזיר״ (שם)? שמעת מינה תרתי‬.‫ לנזיר‬:‫( לימא קרא‬4b)
)‫ אית תנא דמפיק להון ידות מ״לנדור נדר״ (שם ל ג‬:‫ במערבא אמרין‬:‫איב׳ אימ׳‬
.)‫ואית תנא דמפיק להון מ״מכל היוצא מפיו יעשה״ (שם‬

(1a) [Question] How do Nazirite vows differ?

(1b) [Answer] Because it is written: “a Nazirite to vow [nazir le-hazir]”
(Num 6:2).
(2a) [Challenge] About vows also it is written “to vow a vow [lindor
neder]” (ibid.)! Why does the Torah juxtapose them?
(2b) [Resolution] If it had written “a vow to vow [neder lindor]” as it
had written “a Nazirite to nazirite [nazir le-hazir],” then as you say111
it would not have needed to juxtapose them. But now that it has
said “to vow a vow [lindor neder],” the Torah has spoken using the
style of language that people use.
(3a) [Challenge] That works for the one who holds [‫הניחא למאן‬
)‫ ]דאמ(ר‬that the Torah spoke using the style of language that peo-
ple use, but for the one who holds that the Torah did not speak in
the style of language that people use, what does he do with [‫מאי‬
‫“ ]עביד ליה‬to vow a vow [lindor neder]”?
(3b) [Resolution] He uses it [‫ ]דדריש ליה‬to make abbreviations of vows
and then Nazirite vows are connected to vows through scriptural
juxtaposition. And he uses “a Nazirite to nazirite [nazir le-hazir]”
to teach that Nazirite vows can fall on Nazirite vows [i.e. one who
is already a Nazirite can make a Nazirite vow and it will take hold
when his current vow is up].
(4a) [Challenge] And for the one who holds that the Torah spoke
using the style of language that people use, and who [therefore]
uses “a Nazirite to nazirite [nazir le-hazir]” to make abbreviations
of Nazirite vows have the same status as Nazirite vows, from where

111 MS Vatican 487 is difficult to read here. Consequently, I am using MS Vatican 110 to help
make sense of it. Slight variants exist in each of the MSS here, but the gist is the same.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 225

does he derive that Nazirite vows can fall on Nazirite vows? That
works if he holds that Nazirite vows do not fall on Nazirite vows,
but if he holds like the one who says that Nazirite vows fall on
Nazirite vows, from where does he derive this?
(4b) [Resolution] Let the verse say: “to vow [lizor; in the G-stem].”112
Why [does it say] “to cause to vow [le-hazir; in the C-stem]”? We
learn both points from it [‫]שמעת מינה תרתי‬.
(4b’) [Alternate Resolution] If you want, I’ll say, in the West [i.e.
Palestine] they say: There is a transmitter [of the baraita] who
derives abbreviations for them from “to vow a vow [lindor neder]”
and there is a transmitter who derives it for them from “from every-
thing that comes out of his mouth he shall do” (Num 30:3).113

4b and 4b’ are significant. In 4b’ we learn that the stam of the Bavli does indeed
know that the sugya originated in Palestine, regardless of whether he knows
it in exactly the form which we saw above in yNed 1:1, 36c. This tells us that it
is probably no coincidence that both yNed 1:1, 36c (in numbers 5 and 7 there)
and bNed 3a–b (in 4b) resolve the challenge by deriving both points from the
same verse. Nevertheless, while the stam of bNed 3a–b is willing to preserve
this resolution, it makes sure to modify it to accord with the new rules: the
extra word, plus the additional complexity of the binyan give us two anoma-
lies, thus accounting for the two rules. Both points may derive from the same
verse, but multiple anomalies must be found in that verse to accommodate
them. This new rule, among others, pushed the Bavli into more complex sugyot
than its predecessors.
Thus, bNid 40a, which we saw towards the beginning of this section, was
more complex than yNed 1:1, 36c, though they were Babylonian and Palestinian
versions of the same type of sugya. This progression of complexity of the devel-
opment of this kind of sugya can be found both by comparing bNid 40a and
yNed 1:1, 36c, and internally within bNid 40a itself. The amoraic layer as rep-
resented in numbers 1 and 3 in bNid 40a shows the most rudimentary devel-
opment of this kind of sugya, one that simply attempts to find a verse basis
for each position, and no more. The stam of the Yerushalmi takes this one step

112 I am here following MSS Vatican 110 and Ginzberg 1134. MS Vatican 487 has le-nazir, which,
ostensibly, should be the preposition plus the noun, rather than an infinitive verb. This
makes less sense in context, however. The basic point is the same in either case: the verse
could have been written differently, and therefore more than one scriptural anomaly is
open for interpretation here.
113 I have followed MS Vatican 487 except when noted.
226 BRODSKY

further by requiring that each position acknowledge what it does with the
other side’s verses and laws. Nevertheless, since the stam of the Yerushalmi
allows more than one law to be derived from the same verse or part of the
verse, it is easily able to accomplish this. Finally, the stam of the Bavli adds a
level of difficulty to the task by requiring that only one law be derived from
each scriptural anomaly. This added difficulty creates the lengthier and more
complex sugyot of this genre so familiar in the Bavli. Nevertheless, even the
Bavli’s more complex sugyot have in common with the parallel Yerushalmi
genre that they end when each side’s system works. In this way, regarding this
genre of sugya, at least, neither the Yerushalmi nor the Bavli is more “pluralis-
tic” than the other, they simply have slightly different rules governing the same
basic task.

4 Conclusion

The genre of the sugya, then, should be understood in the context of Greco-
Roman school primers of late antiquity, primers that were widely used in the
time and place in which the genre first began to appear in rabbinic literature.
In fact, if we combine the progymnasmata with legal scholia of the same gen-
eral place and time, we can basically account for Talmud as we know it. This
is not to argue that Talmud is simply a wholesale borrowing of two genres of
Greco-Roman literature, but rather that we should no longer view Talmud as a
unique genre that has no relationship to the genres of its surrounding cultures.
Both in terms of its give-and-take and as a commentary on a legal base text in a
different language, Talmud can and should be seen as part of its Greco-Roman
cultural context.
Nevertheless, while the genre may owe its origins to Palestine and the Greco-
Roman world, it was in Babylonia that it reached its apex, at least among rab-
binic sources. One key reason for this was a subtle but significant change in the
hermeneutics of scriptural analysis, in the rules of redundancy, which resulted
in at least one kind of sugya becoming much more complex than it had been
among its earlier Palestinian counterparts.
What led the Babylonian rabbinic community to this greater literary com-
plexity? Obviously, the change in the hermeneutic rules was one proximate
cause, but it only explains the means of this increase in complexity, not the
reason for it. Of course, any attempt at explaining the reasons behind historical
phenomena is inherently speculative. It is often difficult enough to trace the
historical events themselves; explaining their “causes” is a subjective endeavor.
Nevertheless, the data do point in certain directions.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 227

First and foremost, following Jane Webster, I am arguing that creolization,

or hybridization, creates a cultural confluence in which the sum total is greater
than the individual parts. Babylonian Judaism was in a unique position, situ-
ated at the nexus of several cultures: it was born of ancient Israelite culture;
it was raised for a millennium in Zoroastrian Persia; it underwent a signifi-
cant reculturation as rabbinic culture was imported from Palestine, bringing
Hellenistic culture and literary modes of composition along with it (as we saw
in the first part of the present study). In addition, this new culture butted up
against a Syriac Christian culture which was making significant inroads into
the Persian Empire.
This confluence of cultures not only produced a more worldly and cos-
mopolitan atmosphere, but also compelled some of these rabbis to respond
to sophisticated attacks on their practices and ideologies. The rise of scho-
lasticism in Babylonian culture—rabbinic Jewish, Syriac Christian, and
Zoroastrian Persian—has already been well noted by such eminent scholars
as Daniel Boyarin, Adam Becker, and Yaakov Elman.114 This form of scholasti-
cism was creating literary texts similar to the Bavli in a number of ways, includ-
ing the production of the Zand: Pahlavi translation and commentary on older
Avestan texts, complete with sometimes complex commentary, dispute, and
the give-and-take most notable in the Pahlavi Videvdad (e.g. PV 3.16).115 Thus,
the Zoroastrian and Christian cultural context of the Bavli offered a ripe atmo-
sphere for the production of a scholastically sophisticated text.
What I would like to add to this growing scholarly consensus is that rabbinic
scholasticism not only became the most important mitzvah on which all other
mitzvot were made to stand in the Babylonian rabbinic Weltanschauung, but
it also formed the central response to the Pauline attack against justification
through acts. While the precise nature of Paul’s antinomism is hotly debated to
this day, what is clear is that Paul mounted an attack against those who would
seek justification through the fulfillment of the Law in addition to, or in lieu of,
faith through Christ, and that this argument formed one of the backbones of
the various Christian attacks against (rabbinic) Judaism. Thus, whatever Paul’s
original intention, his letters stand at the heart of Christian–Jewish polemics
in the talmudic period. Paul’s main argument against the Law is laid out in
Romans and Galatians, where he posits that it brings death because it is impos-
sible to fulfill it perfectly. Only faith in Christ can redeem one from this fate.

114 Boyarin, Socrates; A. Becker, Fear of God and the Beginning of Wisdom (Philadelphia 2006).
For a list of Elman’s publications, see the works cited in n. 6 above.
115 I would like to thank Yishai Kiel for pointing me to this passage.
228 BRODSKY

Galatians 3:10 articulates this position quite succinctly, stating: “For all who
rely on works of the law are under a curse; for it is written, ‘Cursed be everyone
who does not abide by [ἐμμένει] all things written in the book of the law to do
them [πᾶσιν τοῖς γεγραμμένοις ἐν τῷ βιβλίῳ τοῦ νόμου τοῦ ποιῆσαι αὐτά]’ (Deut
27:26).” Following the Septuagint,116 Paul is using Deut 27:26 as the crux of his
polemic against those who would claim that one can be right with God by fol-
lowing the Law. The Septuagint, however, differs from the Masoretic version in
two key ways. First, it renders the Hebrew ‫“( יקים‬to establish, set up, or make
stand”) as ἐμμένει: “to abide, be faithful to, cleave to.” Second, it has the added
word πᾶσιν, “all,” before τοῖς λόγοις τοῦ νόμου (“the matters of the law”).117 The
Masoretic version merely states ‫ּתֹורה ַהּזֹאת ַל ֲעׂשֹות‬
ָ ‫ָארּור ֲא ֶׁשר לֹא יָ ִקים ֶאת ִּד ְב ֵרי ַה‬
ָ , “one who shall not ‫( יָ ִקים‬set up, fulfill, stand by) the matters of [‫ ] ִּד ְב ֵרי‬this
‫אֹותם‬
teaching [‫ּתֹורה ַהּזֹאת‬
ָ ‫ ] ַה‬shall be cursed.” Paul accentuates the Septuagint’s ver-
sion by rendering it πᾶσιν τοῖς γεγραμμένοις ἐν τῷ βιβλίῳ τοῦ νόμου, “all that is
written in the book of the Law,” focusing on “all.” Thus, Deut 27:26 is the biblical
proof that it is only through grace (which for Paul comes only through faith in
Christ) that one can be saved. As Paul explains in his letter to the Romans, to
live under the Law is inherently to live in a system of sin; faith bypasses this by
allowing us to live under grace.
It is hard to imagine that this key Pauline argument never came up among
the many encounters between Christians and Jews in the early centuries of
the Common Era in the eastern Roman and Persian Empires. A statement
attributed to a key 3rd-century Babylonian amora and preserved in a 5th-
century Palestinian text articulates an artful response to Paul’s polemic and
also happens to help explain how and why scholasticism came to be so cen-
tral to Babylonian rabbinic ideology. In LevR 25:1, the second-generation
Babylonian amora, Rav Huna, states:

‫ מה יעשה ויחיה? אם היה למוד‬.‫ מיתה בידי שמים‬118‫ נתחייב‬,‫נכשל אדם בעבירה‬
‫ ואם אינו‬.‫ ואם היו למוד לשנות פרק א׳ ישנה ב׳ פרקים‬.‫לקרות דף א׳ יקרא ב׳ דפין‬

116 Paul’s version is not identical to the Septuagint, but it is faithful to that version, whether
because he is paraphrasing it or because he is quoting/using another Greek version that
is itself heavily influenced by the Septuagint.
117 The Samaritan Pentateuch also has the added word ‫כל‬, “all,” here.
118 Of the nine MS witnesses and the printed edition, seven witnesses have ‫נתחייב‬, London
has ‫מתחייב‬, the printed edition has ‫חייב‬, and Munich has ‫שהוא מתחייב‬. Though alone in
its reading, the Munich MS is significant. The other MSS can (and I believe ought) to be
translated as I have below: if a person stumbles over a sin, he is culpable of death by the
hands of heaven. The Munich MS, on the other hand, ought to be understood as saying
“If a person stumbles over a sin for which he is culpable of death by heaven . . .” This has
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 229

‫ וגבאי‬,‫ מה יעשה ויחיה? ילך ויעשה פרנס על הציבור‬,‫למוד לא לקרות ולא לשנות‬
‫ אלא ״ארור‬.‫ לא היתה תקומה‬,‫ ארור אשר לא ילמד‬:‫ שאילו נאמ׳‬.‫ והוא חי‬,‫צדקה‬
‫ לא היתה‬,‫ עץ חיים היא לעמלים בה‬.)‫אשר לא יקים את דברי״ וגו׳ (דברים כז כו‬
119.)‫ אלא ״עץ חיים היא למחזיקים בה״ (משלי ג יח‬.‫תקומה‬

If a person stumbles over a sin, he is culpable of death by the hands of

heaven. What should he do that he may live? If he is learned [enough]
to read a folio [of Bible], he should read two folios; and if he is learned
[enough] to recite one chapter [of Mishnah], he should recite two chap-
ters. But if he is not learned neither to read nor to recite, what should
he do that he may live? Let him go and become a monetary sustainer of
the community or a treasurer of charity and he shall live. For, had it said,
“Cursed is the one who shall not learn,” there would be no remedy, but
[it says], “Cursed is the one who shall not set up these matters . . .” (Deut
27:26). [And if it had said]120 “It is a tree of life for those who toil in it,”
there would be no remedy, but [it says], “It is a tree of life for those who
take hold of it” (Prov 3:18).

From its opening line, Rav Huna’s statement is strikingly uncharacteristic of a

rabbinic text: since when does a person deserve death at the hands of heaven
for any and every transgression? While this is not particularly reminiscent of
rabbinic Jewish conceptualization of the law, it is a perfect representation
of the Pauline conception, based on the very verse that Rav Huna goes on
to quote. It is significant that this statement is attributed to Rav Huna. Peter
Schäfer has argued that a number of statements attributed to Rav Huna’s stu-
dent Rav Hisda can be read as responding to core Christian tenets,121 and I have
found this same phenomenon regarding Rav Huna himself.122
Here, in Leviticus Rabbah, Rav Huna seems to be using the rabbinic trope
of arguing on the terms of one’s opponent: even if I accept your position (that
every transgression theoretically yields death), there is a remedy (and that

implications for my analysis of how I believe Rav Huna might be responding to Paul, as I
shall note in the footnotes below.
119 I am here following MS Sassoon. Most MSS are fairly similar to this MS, though a few sig-
nificant variants and their implications are noted in the footnotes.
120 In a number of MSS, these words are added in brackets. Of course, some MSS lack this
second prooftext altogether. Nevertheless, the fact that this verse is found in the (loose)
parallel in the Yerushalmi (ySot 7:4, 21d) suggests that early on it was already associated
with it.
121 P. Schäfer, Jesus in the Talmud (Princeton 2007) 16–7, 26–8, 30–1, 42, and 114.
122 To be published in a forthcoming article.
230 BRODSKY

remedy comes from a correct reading of the very verse on which you base your
position). The Hebrew word Rav Huna uses for remedy (‫—תקומה‬tequmah)
adumbrates the first of the two changes in the Septuagint and in Paul (which
thereby alerts us to pay closer attention to the text) and intimates the second
of the two changes (that of the “all” itself), which is the crux of the Pauline
position. Tequmah (“remedy”) points us back to the word yaqim in Deut 27:26.
If it had not said yaqim (as Paul and the Septuagint would have it), Rav Huna
is saying, then perhaps his anonymous interlocutor would have a point: there
would be no remedy (tequmah) from the Torah. However, it does say yaqim:
the transgressor need merely establish the Torah even if vicariously by sus-
taining others (presumably, in the study of Torah), and that is his remedy for
transgression.123
But Rav Huna is doing something even more subtle and poignant. Where
does Rav Huna’s hypothetical “had it said: ‘Cursed is the one who shall not
learn’” come from? Why “Learn”? And, indeed, what if the verse had said
“Cursed is the one who shall not learn”? Is it so impossible to learn? Need learn-
ing be so perfect that it is impossible to fulfill this command? According to
Marcus Jastrow, at its core, lamad means “to join” or “to cleave,” and this mean-
ing was retained in rabbinic usage.124 That is, in Hebrew, learning is inherently
a process of joining two concepts together. Whether by coincidence or not,
lamad is thereby a proper correlative of the Septuagint’s emmenei, also mean-
ing to join or cleave, though understood by Paul to mean “cleave to” as in “to
stick with” or “be faithful to.”

123 Following the Munich MS, Rav Huna is not acknowledging that all laws theoretically
result in death as Paul would have it. Rather, he is merely taking up the issue of those
laws which get the penalty of karet (often understood as death by heaven). Even in this
reading, however, Rav Huna’s question is strange. Why even ask the question, what can
one do to avoid eternal damnation? The answer in rabbinic law is unanimous: the sinner
should repent and then the divine punishment will not be meted out. Even repenting
one moment before death is enough to avoid such punishment. While I do not think the
Munich MS’s version is the correct reading here, it is important to notice that even in this
reading Rav Huna is creating a problem that is more easily resolved than he would have us
believe. I think Rav Huna is purposely entertaining a problem not in order to entertain the
issue for its own sake, but in order to answer Paul on his own terms: granted your position
(though we ourselves do not hold it), here are the flaws with your exegesis, and here is the
“correct” solution.
124 To the extent that limmudim refer to shingles and boards used as junctions in construc-
tion, see M. Jastrow, Dictionary of the Targumim, Talmud Bavli, Talmud Yerushalmi and
Midrashic Literature (Philadelphia 1971) 708, s.v. limmudim and 712, s.v. lamad.
FROM DISAGREEMENT TO TALMUDIC DISCOURSE 231

While Rav Huna might have had better Hebrew words to choose from to
correlate precisely and uniquely to the Greek emmenei and the concept of
“cleaving to,” he could not have chosen a better correlative with which to
respond to Paul. Read alongside the Septuagint and Paul, Rav Huna is redirect-
ing Paul’s exegesis of Deut 27:26, and using Paul to argue for the preeminence
of rabbinic scholasticism. He is saying: if it had said emmenei, if it had said
lamad, they would truly be in a difficult position, not because the violation
of even a single command­ment yields death as Paul’s reading would have it,
but because limmud, Torah study, the rabbis’ preeminent cultural value, is so
impossible to achieve. Fortunately (Rav Huna continues), God only required
that the Israelites establish (yaqim) the Torah, which, in the causative (as it is
here found), can imply even through others. Nevertheless, Rav Huna seems to
be saying that Paul’s midrash has a point: the ultimate value is limmud, and he
suggests that ideally, one can obtain a remedy (tequmah) for one’s transgres-
sions. For, following this reading, ultimately, Prov 3:18 should have mandated
that we “toil” in the Torah. It was lenient for those who cannot, providing a
tree of life even for them, but the true value, for those who can, is to go beyond
merely taking hold of the Torah. And, it goes without saying for Rav Huna that
Prov 3:18 did not intend that the tree of life is Christ, as many of his Christian
counterparts were articulating around his time.125
If the attribution to Rav Huna is correct, then we seem to have an early
Babylonian amoraic position, articulating rabbinic scholasticism (and the
more sophisticated the better) both as a central tenet of rabbinic Judaism and
as the cornerstone of his response to Christian antinomism.
I do not wish here to be reductionist or a positivist. I am not arguing that
Babylonian rabbinic scholasticism is merely a response to Christianity, or that
it is solely a product of the rabbis’ encounter with it, any more than I wish to
reduce the sugya to a mere reproduction of the progymnasmata. Torah study
was an integral part of rabbinic culture and had value independent of its use as
a response to the Pauline critique of rabbinic Judaism. Nevertheless, Rav Huna’s
use, early in the amoraic period, of this central rabbinic tenet as a response to
Paul would seem to be one piece in the puzzle of how and why it came to
be so prominent in Babylonian rabbinic culture (beyond the place it held in
Palestine, as scholars have noted before me) and why that culture would come
to augment it with ever-greater complexity of thought and hermeneutics.

125 Origen, Homilies on Lev. 16.4.3; idem, Homilies on Joshua 8.6; Ephrem, Hymns on the
Nativity 1.
The Misfortunes and Adventures of Elihoreph and
Ahiah in the Land of Israel and in Babylonia:
The Metamorphosis of a Narrative Tradition and
Ways of Acculturation1

Reuven Kiperwasser

One characteristic of rabbinic literature is the transformation that its liter-

ary traditions undergo when migrating from the various Palestinian rabbinic
milieus to the academies of Babylonia. This phenomenon is very complicated
and different explanations for it have been offered by scholars.2 In this study
I will discuss two narrative traditions that stem from a relatively ancient
Palestinian story; I will explain how they metamorphosed in the various redac-
tions both in the literature of the Land of Israel and in that of the Babylonian
Talmud. The case under study is the story, attested in various versions, of the
misfortunes of Elihoreph and Ahiah, the sons of Shisha (1Kgs 4:3). Analyzing
the rich plot in the Babylonian version of the tradition, I propose to reveal
remnants of an Iranian myth and common topoi of storytelling from the
ancient Orient.
This story, as an example of the migration of narrative traditions from
Palestine to Babylonia, was discussed at the dawn of talmudic research, in the
Wissenschaft des Judentums, by Z. Frankel in his introduction to the Palestinian
Talmud (Yerushalmi).3 He preferred the more developed Babylonian Talmud

1 A version of this paper, was presented on July, 2009, at the SBL conference in Rome. Words
of thanks go to all the participants for their notes and questions and especially to the read-
ers of the various drafts—Ronit Nikosky, Tal Ilan, Amram Tropper, Dan Y. Shapira, Geoffrey
Hermann, and Maria Kaspina.
2 See e.g. J. N. Epstein, Mevo’ot le-Sifrut ha-Amoraim (Jerusalem 1962) 292, 391 [Hebrew];
D. Rosenthal, “The Transformation of Eretz Israel Traditions in Babylonia,” Cathedra 92
(1999) 7–16 [Hebrew]; Y. Sussmann, “Ve-shuv le-Yerushalmi Nezikkin,” in: Y. Sussmann and
D. Rosenthal (eds.), Mehqerei Talmud: Talmudic Studies (Jerusalem 1990) 55–134 [Hebrew];
I. Gafni, “Between Babylonia and the Land of Israel: Ancient History and the Clash of
Ideologies in Modern Jewish Historiography,” Zion 62 (1997) 213–42 [Hebrew]; V. Noam,
“A Story that Was Captivated: The Evolvement of a Tale between Eretz Yisrael and Babylon,”
Jerusalem Studies in Hebrew Literature 19 (2003) 9–23 [Hebrew]; J. L. Rubenstein, The Culture
of the Babylonian Talmud (Baltimore 2003). [See also the Introduction to the present
volume—Eds.].
3 See Z. Frankel, Mevo ha-Yerushalmi (Breslau 1870) 44a [Hebrew].

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�09

THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 233

(Bavli) version, and characterized it as of a higher literary quality than the

“poor” version of the Yeru­shalmi. In this study I will emphasize the secondary
character of the Bavli version, and will suggest the possible factors that influ-
enced its late redaction.
Let us look at the Palestinian versions of the story:

5‫קהלת רבה ג ב‬ 4‫ לב ע"ג‬,‫י' כלאים ט ג‬

)‫״עת ללדת״ (קהלת ג ב‬ ‫ ״ואתה פשחור וכל יושבי ביתך‬:‫ כתיב‬1
‫תלכו בשבי ובבל תבא ושם תמות ושם‬
.)‫תקבר״ (ירמיהו כ ו‬
‫משעה שאדם נולד נגזר לו באיזה חיים יחיה באי‬ ‫ אמר ר׳ חלבו ור׳ חמא‬,‫ר׳ אבא בר זמינא‬
.‫זו מיתה ימות‬ ,‫ מת שם ונקבר שם‬:‫ חד אמר‬.‫בר חנינא‬
‫ מת שם ונקבר כאן יש בידו‬.‫יש בידו שתי׳‬
‫ וחרנה אמר קבורה שבכאן מכפרת‬.‫אחת‬
.‫על מיתה שלהן‬
‫ רגלוי דבר אניש אנון ערבין‬:‫ר׳ יונה בר׳ חנינא‬ ‫ ריגליי‬:‫ רבי יונה בשם ר׳ חמא בר חנינה‬2
.‫ לאתר דמתבעי תמן מובלין יתיה‬.‫ביה‬ ‫דבר נשא ערבתיה למיקמתיה כל הן דהוא‬
.‫מתבע‬
,‫ חבלים טמונים בארץ וקשורים בארץ‬:‫[ויש אומ׳‬
])‫ ״טמון בארץ חבלו״ (אויב יח י‬:‫הה״ד‬
‫ ״ואתה פשחור וכל יושבי ביתיך תלכו‬:‫כת׳‬
‫ושם תקבר״ (ירמיה כ ו) ״ולא ימות‬ . . . ‫בשבי‬
.‫בתוך ביתו״ לא כת׳ תמן‬
)‫ יי מי יפתה את אחאב וכו׳׳׳ (מל״א כב ב‬:‫״ויאמר‬ ‫ ״ויאמר מי יפתה את אחאב‬:‫כתיב‬
.‫עניינה ולא <כתיב> ״מת בתוך ביתו״‬ ‫ויעל ויפל ברמות גלעד״ (מל״א‬
.‫ ״וימות בתוך ביתו״ ולא תמן‬.)‫כב כ‬
:‫תמן תנינן‬ 3
.‫אליחורף ואחיה הוו איסקבטוריי דשלמה‬ ‫איסקבטיריי‬ ‫תרין‬ ‫ואחיה‬ ‫אליחורף‬
.‫דשלמה‬

4 According the Academy of the Hebrew Language edition, p. 176. Compare parallel in yKet
12:3, 35b.
5 On this midrash, see, e.g. L. Grünhut, Kritische Untersuchung des Midrasch Kohelet
Rabbah (Berlin 1892); J. Wachten, Midrasch-Analyse: Strukturen in Midrasch Qohelet Rabba
(Hildesheim 1978); M. Hirshman, “Midrash Qohelet Rabbah: Chapters 1–4” (Ph.D. diss.; New
York 1983) [Hebrew]; R. Kiperwasser, “Midrashim on Kohelet: Studies in their Redaction and
Formation” (Ph.D. diss.; Ramat Gan 2005) 43–72 [Hebrew]. The text is according to MS JTS
and see Hirshman’s edtion, pp. 231–3 (see above).
‫‪234‬‬ ‫‪kiperwasser‬‬

‫<והוה שלוחהון דבריתא נהיג שאיל שלמא‬

‫דשלמא‪ >6‬כל יומא‪.‬‬
‫חד זמן על למשאול בשלמיה וחמא יתהון יתיהין‬ ‫חמא מלאך מותא מסתכל בון וחריק‬
‫גביה שריה מחריק בשיניה כל קבליהון‪.‬‬ ‫בשינוי‪.‬‬
‫אמ׳ שלמה‪ :‬דילמא דאינון מסרין בידיה‪.‬‬
‫מה עבד? אמ׳ מילה‪ ,‬ינסב יתהון וגנז יתהון‬ ‫אמר מילה ויהבון בחללא‪.‬‬
‫בחללה‪.‬‬
‫אזל ונסב ית נשמיתהון מן תמן‪.‬‬ ‫אזל ונסתון מן תמן‪.‬‬
‫חזר למשאל בשלמיה דשלמה ושרי חייך כל‬ ‫אתא קאים ליה גחיך לקבליה‪.‬‬
‫קבליה‪.‬‬
‫אמ׳ ליה שלמה‪ :‬אמור לי מה עיסק? בזמנא‬ ‫אמ׳ ליה‪ :‬ההיא שעתא הויתא איחרוק‬
‫קדמייא דעלת לגבי‪ ,‬הות מחריק בשיניך‪ ,‬וכדון‬ ‫בשנייך וכדון את גחיך לן?‬
‫את קיים גחיך?‬
‫אמ׳ ליה‪ :‬חייך‪ ,‬שליחותך דמרך עבדת‪.‬‬
‫אמ׳ ליה‪ :‬מאריה דעלמא דאזיל ואיסב נשמתהון‬ ‫אמר ליה‪< :‬אמר לי> רחמנא דינסב‬
‫דאליחורף ואחיה מן גו חללה‪.‬‬ ‫לאליחורף ואחיה מן חללא‪.‬‬
‫ואמרית בלבי‪ :‬מאן יהיב לי אילין מן הן אשתלחית‬ ‫ואמרית מאן יהיב לי אילין להן‬
‫דנסיב יתהון?‬ ‫דאישתלחית מיסבינין‪,‬‬
‫ואתייהיב בלבך את דתעבד מן דנעביד שליחותי‪,‬‬ ‫ויהב בליבך למיעבד כן בגין דנעביד‬
‫איזיל ואיטפל לון מתמן‪,‬‬ ‫שליחותי‪.‬‬
‫ואזל ואיטפללון מתמן‪.‬‬ ‫אזל ואיטפל בון (ז)מן תמן‪.‬‬
‫תרתין בנוי דר׳ ראובן בר איסטרובולוס הוון‬ ‫‪ 4‬תרי׳ ברויי דר׳ ראובן בר איסטרובילו׳‬
‫למידוי דר׳‪ .‬פגע בהון שליחתהון דברייתא‪.‬‬ ‫תלמידוי דרבי‪.‬‬
‫חמא יתהון עימיה ושרי מחריק שיניה לקבליהון‪.‬‬ ‫חמא מלאכא דמותא מסתכל בון וחריק‬
‫אמ׳ דילמא דאינון מסירין בידיה?! מה עבד? נסב‬ ‫בשינוי‪.‬‬
‫יתהון ואגלי יתהון לדרומא‪ ,‬לומ׳ הגלות מכפרת‬ ‫ואמ׳ ניגלינון לדרומא שמא הגלות‬
‫עון‪ .‬אזל ונסב יתהון מתמן‪.‬‬ ‫מכפרת‪.‬‬
‫אזל ונסתון מן תמן‪.‬‬

‫‪QohR 3:2‬‬ ‫‪yKil 9:4, 32c‬‬

‫”‪1 “A time to be born, and a time to die‬‬ ‫‪It is written: “As for you Pashhur, and all‬‬
‫)‪(Qoh 3:2‬‬ ‫‪who live in your house, you shall go into‬‬
‫;‪captivity. You shall come to Babylon‬‬
‫‪there you shall die and there you shall‬‬
‫‪be buried” (Jer 20:6).‬‬

‫‪6 The emendations are according the Vatican MS (Biblioteka Apostolica ebr. 291).‬‬
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 235

From the hour that a person is born he Rabbi Abba bar Zemina said: Rabbi
is sentenced to which life he will live Helbo and Rabbi Hama bar Hanina, one
and in which death he will die. said: He died there and was buried there,
these are two. Died there and buried
here this is one. The other said: Burial
here atones for their death.
2 Rabbi Yonah in the name of Rabbi Rabbi Yonah in the name of Rabbi Hama
Hanina said: A person’s feet are his bar Hanina: A person’s feet are his surety.
surety. They bring him to the place They place him in the location where he
where he is sought. is sought.
(And some say: Nooses are hidden in
the earth and bound to the earth, as it
is written: “A noose is hidden for him in
the ground, and a trap for him on the
way” [Job 18:10]).
It is written: “And you, Pashhur, and
all that dwell in your house shall go
into captivity; and you shall come to
Babylon, and there you shall die, and
there shall you be buried, you, and all
your friends, to whom you have proph-
esied falsely” (Jer 20:6) and it is not
written that he will die in his home . . .
It is written: “And the LORD said: Who It is written: “Who shall entice Ahab, that
shall entice Ahab, that he may go up he may go up and fall at Ramot-Gilead”
and fall at Ramot-Gilead” (1Kgs 22:20). (1Kgs 22:20). “And die in his house” is not
And it is not written “die in his home.” written there!
3 It is taught:
Elihoreph and Ahiah were Solomon’s Elihoreph and Ahiah were two of
secretaries. Solomon’s secretaries (?).
And it was customary for the
“Dispatcher of the People” to inquire
after Solomon’s health every day.
<Once he was entering to inquire after He (Solomon) saw the Angel of Death
his health>7 and saw them entering and looking at them and grinding his teeth.
began to grind his teeth. Solomon said:
Probably they are in his possession!
What did he do? He said a word and He said a word and put them into a cave.
put them into a cave.

7 Emended according to the Vatican MS.

236 kiperwasser

(The Angel of Death) went and took (The Angel of Death) went and took
their souls from there. them from there.
He returned Solomon to inquire after (Then) he came laughing in his face.
his health and began laughing in his
face.
Solomon said to him: Tell us what were Solomon said to him: Previously you
your deeds? The last time you came you were grinding your teeth and now you
were grinding your teeth and now you are laughing at us?
are laughing?
He said to him: I have carried out the
commission of the Lord!
4 And he said: The Master of the Universe He said to him: It was related to me
said to me that I should go and take the by the Master of the Universe to take
souls of Elihoreph and Ahiah from the Elihoreph and Ahiah from the cave.
cave.
And I said in my heart: Who will give And I said: Who would put them there,
them to me in any place other than where I was sent to take them?
here? Thus I received the commission
to take them!
And it was put into your mind to do so, He put it into your mind to do so, in
in order that I could fulfill my mission, order that I could fulfill my mission.
that I go and deal with them there.
So I went and dealt with them there. He/I went and dealt with them there.
4 The two sons of Rabbi Reuven ben The two sons of Rabbi Reuven ben
Strobilos were students of Rabbi. Rabbi Strobilos were students of Rabbi.
saw the Dispatcher of the people look- (Rabbi) saw the Angel of Death looking
ing at them and grinding his teeth and at them and grinding his teeth. He said:
he said: It seems that they belong to let us exile them to the South, maybe
him! What did he do? He took them exile atones. The Angel of Death went
and exiled them to the South, maybe and took them from there.
exile atones. The Angel of Death went
and took them from there.

First, let us consider the structure of the text. Three elements are noticeable
here: i. the aphorism of Rabbi Yonah (1); ii. the short midrashim about the
false-prophets Pashhur and the one who enticed King Ahab, and thus about
sinners whose fate was to die away from their homes (2); and iii. two short
stories (3, 4) about people who tried to escape death but ran directly into their
appointment with him. The first of these stories, about King Solomon’s scribes,
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 237

is the one in which we are interested. Behind the two Palestinian works, the
Yerushalmi and Qoheleth Rabba (QohR), lies a common tradition, probably an
exegesis on a prophecy by Jeremiah on the false prophet Pashur and his admir-
ers (Jer 20), that strongly emphasizes the idea of predestination.
Let us now analyze this narrative. The story has a midrashic nucleus, based
on the verse “. . . Elihoreph and Ahiah sons of Shisha—scribes . . .” (1Kgs 4:3).
The Yerushalmi version is shorter than the QohR one, more elliptic and rela-
tively poor in plot details. It probably also contains errors of transmission, but
in the absence of any other textual witness we have no way to correct them.
The QohR version is relatively late, but it still stems from the complex of live
midrashic traditions of Palestine. Thus, we read both versions together in an
attempt to reconstruct a common prototype of the Palestinian story. The mes-
sage of the story is that one cannot prevent the predestined moment of meet-
ing with Death.
Let us list the heroes: King Solomon, according to an ancient tradition
from the Second Temple period, was endowed with unnatural powers;8 fur-
thermore, various stories circulated in the ancient world about his dominion
over demons and his conversations with angels.9 In the story under discus-
sion, King Solomon is reported to have held a normal and peaceful conver-
sation with the Angel of Death, despite the fact that mortals do not usually
have the opportunity to meet this angel more than once in their lifetime.10
Two other protagonists of the story, Elihoreph and Ahiah, appear only once in
Scripture, where they are both described as noble people: soferim11 and sons
of the prominent sofer of King David. According to Scripture, a sofer was a

8 See P. A. Torijano, Solomon the Esoteric King: From King to Magus, Development of
a Tradition (Leiden 2002) and see also L. Ginzberg, The Legends of the Jews, vol. 4
(Philadelphia 1947) 125–69.
9 See D. C. Duling, “The Eleazar Miracle and Solomon’s Magical Wisdom in Flavius
Josephus’s Antiquitates Judaicae 8.42–49,” Harvard Theological Review 78 (1985) 1–25;
idem, “The Testament of Solomon: Retrospect and Prospect,” Journal for the Study of the
Pseudepigrapha 2 (1988) 87–112; H. M. Jackson, “Notes on the Testaments of Solomon,”
Journal for the Study of Judiaism 19 (1988) 19–60; B. A. Pearson, “Jewish Haggadic
Traditions in the Testimony of Truth from Nag Hammadi (CG IX,3),” in: B. A. Pearson
(ed.), Gnosticism, Judaism, and Egyptian Christianity (Minneapolis, 1990), 39–51; S. Shalev-
Eyni, “Solomon, his Demons and Jongleurs: The Meeting of Islamic, Judaic and Christian
Culture,” Al-Masaq: Islam & the Medieval Mediterranean 18 (2006) 145–60.
10 See below.
11 See H. Nieher, “Sōpēr,” in Theological Dictionary of the Old Testament, vol. 10 (Grand Rapids
Mich. 1999) 323–6; M. Z. Kaddari, A Dictionary of Biblical Hebrew (Ramat Gan 2007) 751
[Hebrew].
238 kiperwasser

high diplomatic position, second after the sar, but in the rabbinic story, their
office is not described in this way. The narrator of the midrash uses a relatively
rare term for the two brothers’ profession, ‫אסקפטריא‬/‫איסקבטיריי‬, which stems
from the Greek σκέπτορ (“sceptor”). This term means specifically a scribe of
the court or a scribe attached to a magistrate,12 which is not a very high social
position. The decrease in social status remains unexplained, but the lack of
biographical detail concerning the two biblical heroes will be resolved in the
aggadic story, for new biographies will be created for them.13 The last hero, the
Angel of Death, is a typical talmudic protagonist. In Scripture there is no single
angel to lead dying people to the netherworld. There are different angels who
execute God’s death verdict. The names of the divine executioners are ‫ַה ַּמ ְׁש ִחית‬
(Exod 12:23), ‫( ַּמ ְל ָאְך ַה ַּמ ְׁש ִחית‬2Sam 24:16; 1Chr 21:15), ‫( ַמ ְל ַאְך ה׳‬2Kgs 19:35), or
in plural, ‫( ַמ ְל ֲא ֵכי ָמוֶ ת‬Prov 16:14). Only in rabbinic thought does death receive
a personification in the image of a particular Angel of Death.14 This Angel is
Death itself; his role is to take the living souls away from dying people. There
is no way for a person to release himself from the Angel of Death, but the per-
son who is permanently busy with studying the Torah and performing good
deeds is well-protected against this angel, until the moment he desists from
his pious practice.15 A person who desists from these good deeds offers him/
herself as a victim to the Angel of Death. In our story, there is a difference

12 See D. Sperber, Greek and Latin Words in Rabbinic Literature (Ramat Gan 1986) 119–21.
13 About this phenomenon in rabbinic literature see Y. Fraenkel, Midrash ve-Agagdah, (Tel-
Aviv 1996) 296 [Hebrew].
14 A. P. Bender, “Beliefs, Rites, and Customs of the Jews, Connected with Death, Burial,
and Mourning,” Jewish Quarterly Review 6 (1894) 317–47, esp. 322–3; P. Schäfer, Rivalität
zwischen Engeln und Menschen: Untersuchungen zur rabbinischen Engelvorstellung
(Berlin 1975) 27, 59, 66–7. See also B. Rebiger, “Angels in Rabbinic Literature,” in: F. V.
Reiterer, T. Nicklas, K. Schöpflin (eds.), Angels: the Concept of Celestial Beings: Origins,
Develop­ment and Reception (Berlin 2007) 629–44. It seems to me that the nature of this
theological innovation is not well understood.
15 See MdRY, masekhta de-ba-hodesh 9, p. 237. Unambiguous language of the principle
according to which good deeds prevent the Angel of Death from making his case we find
in Tractate Derekh Eretz 7:16, 30, but these are rather late texts of Palestinian provenance.
For the motif that no-one can escape the Angel of Death, except the scholar who is
permanently busy studying Torah see yMQ 3:5, 82d and further in bShab 30b, bMQ 28a.
The topic of combat between chosen mortals and the Angel of Death is represented in
rabbinic narratives and discussed by the scholars, see Y. Fraenkel, The Aggadic Narrative:
Harmony of Form and Content (Tel Aviv, 2001) 273–94 [Hebrew]. See also R. Kiperwasser,
“The Visit of the Rural Sage: Text, Context and Intertext in a Rabbinic Narrative,” Jerusalem
Studies in Jewish Folklore 26 (2009) 3–24 [Hebrew].
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 239

between the king and his secretaries in their relationship to the Angel of
Death. Since, according to rabbinic perception of Solomon’s image, the king
spends all his time in good deeds and in Torah study, and the Angel cannot take
his soul, they are sort of friends—the Angel, comes for a daily visit, inquiring
after the king’s health, but, perhaps, also playing the role of memento mori at
the royal court.
In the beginning of the story in QohR, which is absent from the Yerushalmi,
we hear that the Dispatcher to the creatures (i.e. the Angel of Death) visited
the royal court every morning to salute the admired king. There is a funny
paranomasiac element, built on the nearness of the name of the king (Shlomo)
to the word “peace” or health (shalom), but the pun is “lost in translation.”
Discovering that the Angel of Death makes an angry gesture toward two
of his servants, Solomon assumes that their death is near. Perhaps the king
concludes that they are sinners and that is the reason for the angelic wrath.
The king, an experienced wonder-maker, decides to rescue his secretaries from
death through his magic art. By uttering the magic word (‫)מילא‬,16 the wise king
sends his servants to a cave. Why a cave? Perhaps simply because a cave is a
hiding place; but the motif of escape to a cave, or the model of living in the
cave in opposition to urban life, serves as refuge for the persecuted in many
biblical stories.17 It is also common in rabbinic narratives: a sojourn in a cave
retains this role as a place of refuge, but acquires further functions—as a place
of atoning for sins,18 and an abode for ascetics. The cave also plays an impor-
tant role in constructing the image of the rabbinic hero.19
Regarding our story, we need to bear in mind that the main usage of caves
in Palestine was for burial purposes. Therefore, if the narrator attempts to find
shelter for his heroes in a cave, it must have a symbolic meaning. Placing these
living people in a burial place is a symbolic replacement of death, its purpose
being the atonement of sin and prevention of death. As in the next story in the
Yerushalmi and QohR, about the sons of Rabbi Reuven, who were sent to exile,
because exile atones, in this story too, preserving the two brothers in the cave

16 See D. Sperber, Magic and Folklore in Rabbinic Literature (Ramat Gan 1994) 60–6; and see
Y. Harari, “The Sages and the Occult,” in: S. Safrai (ed.), The Literature of the Sages vol. 2
(Assen 2006) 521–64.
17 There are prophets which were hidden by Obadiah in a cave (1Kgs 18:4), and five kings
who hid in one (Josh 10:16) etc.
18 See e.g. GenR 41:30 (p. 537).
19 On the case of Rabbi Shimeon bar Yohai and his son, and the prophet Elijah see
R. Shoshany, “Rabbi Shimon ben Yohai in the Cave and Elijah in the Wilderness: A
Comparison between Talmudic and Biblical Narratives,” Jewish Studies: An Internet
Journal 6 (2007). Online: http://www.biu.ac.il/JS/JSIJ/6-2007/Shoshany.pdf.
240 kiperwasser

implies an act of atonement, and its intention is to prevent the brothers’ death.
But this magic trick, despite its successful performance, does not achieve its
goal. The Divine Dispatcher fulfills his mission and takes the lives of Elihoreph
and Ahiah even in their hiding place. The Angel of Death explains to the king
that the cave was their designated place of death, from where he was expected
to take their souls, and that the reason for his previous bewilderment was see-
ing the king’s servants in court and not in their designated place. The fatalis-
tic ending of the story resembles an ancient motif that appears, for example,
already in Josephus’ writings, relating the death of prince Antigonus, which
was foretold by Judas of the sect of the Essenes:

But here one may take occasion to wonder at one Judas, who was of the
sect of the Essenes, and who never missed the truth in his predictions;
for this man, when he saw Antigonus passing by the temple, cried out to
his companions and friends, who abode with him as his scholars, in order
to learn the art of foretelling things to come, that it was good for him to
die now, since he had spoken falsely about Antigonus, who is still alive,
and he sees him passing by, although he had foretold he should die at the
place called Strato’s Tower that very day, while yet the place is six hun-
dred furlongs off, where he had foretold he should be slain; and still this
day is a great part of it already past, so that he was in danger of proving a
false prophet. As he was saying this, and that in a melancholy mood, the
news came that Antigonus was slain in a place under ground, which itself
was called also Strato’s Tower, or of the same name with that Caesarea
which is seated at the sea (Josephus, Antiquities 13.311–3).

Here we meet a similar idea—the place of death (and needless to add, the
time) is already predestined and its name is known to God, but people, even
the seers and sages, usually make mistakes in interpreting God’s words.
The story related by Josephus belongs to a well-known type of folkloristic
motifs, “Vain attempts to escape fulfillment of prophecy” and “Prophecy of
death fulfilled,”20 with this motif being also incorporated by the narrator into
our midrash for the purpose of filling in the biographical lacuna found in the
scriptural representation the Elihoreph and Ahiah. The rabbinic narrator was
familiar with ancient popular traditions regarding King Solomon the magician
and wonder-maker, and his dealings with angels and demons. But the story

20 According to the Aarne-Thompson’s Index it is M370 (Cf. †M341.2.10, †M343, †M344) and
M370.1. †M370.1.
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 241

he creates here is not a simple mirabilia or panegyric of the wise king, but a
skeptical and ironic narrative. The narrator meditates on the known, but per-
manently surprising, fact that death will take the living from every place and
at any stage of life, and that no trick or deed can prevent this sad outcome. In
a wider sense the story expresses the existential fear of the unavoidable end.
There is no way to win in a combat with death. Even the wisest Jewish arch-
magician was unable to do so. Note that the narrator of the later Palestinian
story (QohR) maintained the typological characteristics of its folktale proto-
type and at the same time added to it another type of folk tale: “Futile moving
to avoid death.”21
And now to the Babylonian parallel.

bSuk 53a22
Rabbi Yohanan stated: A person’s feet .‫ רגלוהי דבר איניש אינון ערבין ביה‬:‫אמ׳ ר׳ יוחנן‬
are his surety. They transport him to the .‫ לתמן מובילין יתיה‬,‫לאתרא דמתבעי‬
place where he is sought.
There were once two Cushites who :‫הנהו תרי כושאי דהוו קיימי קמיה דשלמה‬
attended on Solomon, as it is written: ‫״אליחורף ואחיה בני שישא סופרים״‬
“. . . Elihoreph and Ahiah sons of Shisha .‫(מל״א ד ג) דשלמה‬
scribes . . .” (1Kgs 4:3) of Solomon.
One day [Solomon] observed that the .‫יומא חד חזייה למלאך המות דהוה עציב‬
Angel of Death was sad. ?‫אמ׳ ליה אמאי עציבת‬
He said to him: Why are you sad? .‫ דקיימי קמא‬,‫ דקא מתבעי מנאי הני‬:‫אמ׳ להו‬
Because, he answered him, I am
demanded [to take] these two who are
in your presence.

21 Ibid., M382. †M382: Man told by Death he will die where he stands sells everything and
moves to another town. He goes for a ride on a mare which runs away with him and
throws him on the spot he so dreads, killing him. India: Thompson-Balys.
22 The comparative analysis of the textual versions of the passage is based on the following
mss: London—BL Harl. 5508 (400), Munich 140, Munich 95, Oxford Opp. Add. fol. 23,
Vatican 134, Oxford—Bodl. heb. e. 51 (2677), New York—JTS Rab. 218 (EMC 270) and as the
preferable version Oxford Opp. Add. fol. 23 was chosen.
242 kiperwasser

[Solomon] put them in the hands of a .‫מסרינהו לשידא‬

demon.
He said to him: Conceal them in the dis- .‫ אמטונהו למחוזא דלוז‬:‫אמ׳ להו‬
trict of Luz.
When they reached the gate of the city .‫ שכיבו‬,‫כי מטו אבבא‬
they died.
[The Angel of Death] took them. .‫שקלינהו‬
On the following day [Solomon] .‫ חזייה דהוה קא בדח‬,‫לאחר‬
observed that the Angel of Death was
laughing.
Why, he said to him, are you laughing? ?‫ אמאי קא בדחת‬:‫אמ׳ ליה‬
He replied: You sent them to the place .‫ התם שדרתינהו‬,‫ היכא דאיבעו מנאי‬:‫אמ׳ ליה‬
where I expected them!
Solomon uttered the saying: A person’s ‫ רגלוהי דבר איניש אינון ערבין‬:‫פתח שלמה ואמ׳‬
feet are his surety. They transport him to .‫ לתמן מובילין יתיה‬,‫ לאתרא דמתבעי‬.‫ביה‬
the place where he is sought.

As is evident from this text, the story is framed within the aforementioned aph-
orism, which even preserves its Galilean Aramaic linguistic features. But, as a
result of the editing of the Babylonian redactor,23 the aphorism is more closely
associated with the story, and receives additional meaning: if one decides to go
anywhere, even by one’s free will, one will not arrive at the place one intends,
but rather at the place that was predestinated.
The narrator of the Bavli story makes the scribes kushim, that is, dark-
skinned people. The reason for this metamorphosis is unclear. Goldenberg
proposed that the change stems from a process of misunderstanding in which
the relatively rare Greek word (is)sceptorei (“scribes”) which we encountered
in the Palestinian sources was misunderstood as deriving from scotos, plural
scotoi, which means dark-skinned people.24
The gloom of the Angel of Death, who is instructed to kill Solomon’s scribes,
is unexplained in the beginning of the story and falsely interpreted by Solomon
as implying that the Angel is ready to kill the scribes then and there. Therefore
the king decides to rescue his servants. The Angel’s strange behavior was inter-

23 This feature can probably be interpreted as evidence for the existence of a version of
the “Palestinian Talmud in exile” in the hands of the editor, on which see especially
A. M. Gray, A Talmud in Exile: The Influence of Yerushalmi Avodah Zarah on the Formation
of Bavli Avodah Zarah (Providence RI 2005).
24 See D. M. Goldenberg, The Curse of Ham: Race and Slavery in Early Judaism, Christianity,
and Islam (Princeton 2003) 201–2.
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 243

preted by the king as follows: as a good friend of the king, the Angel wants to
warn him. Taking the hint, the king does something. He mobilizes his demonic
servants and sends his human servants to the City of Luz.
The motif of King Solomon, the ruler of the demons, is an ancient one.25 But
the motif of using demons to do one’s bidding was especially widespread in
Sasanian Babylonia.26 The Iranian narrators told stories about Iranian heroes,
such as Taxmorup, who used Ahariman or demons as servants.27 Against this
background, the flowering of stories about Solomon and demons28 in the Bavli
seems explicable. The idea that with demonic help it is possible to resolve prob-
lems of Divine judgment seems to me part and parcel of the common culture
against the background of which the Bavli was composed. The narrator of our
story seems completely indifferent to the theological problem of attempting to
neutralize the divine messenger with demonic help. But the most interesting
motif that was added by the Babylonian narrator is that of the City of Luz.
A city named Luz is known from scripture, first as the ancient name of
Bethel (Judg 1:23–26):

While the House of Joseph were scouting at Bethel, the name of the town
was formerly Luz, their patrol saw a man leaving the town. They said to
him: Just show us how to get into the town, and we will treat you kindly.
He showed them how to get into the town: They put the town to the
sword; but they let the man and all his relatives go free. The man went
to the Hittite country. He founded a city and named it Luz, and that has
been its name to this day.29

The location of the new Luz is not known, but what was the reward promised
to the collaborator of old Luz? Was it only to let him and his relatives leave
the city and stay alive? An aggadic tradition developed around this short plot,

25 See above nn. 8–9.

26 See A. Kohut, Über die jüdische Angelologie und Daemonologie in ihrer Abhängigkeit vom
Parsismus (Leipzig 1866) 55–62.
27 “After that it came to Taxmorup the well-armed, and through that glory the demon and
evil mankind, the wizard and witch, were smitten by him; idolatry was also cast out by
him, and he propagated in his time the reverence and service of the creator; the evil spirit,
converted into the shape of a horse, was also carrying him for thirty winters” (Denkard VII,
1:19, in E. W. West [trans.], Sacred Books of the East, vol. 5 [Oxford 1897] 8–9). See also E. W.
West, Dina-i Mainog-I Khard, vol. 24, pp. 58–9.
28 E.g. Building the Temple with the help of Ashmedai (bGit 68a–b).
29 JPS translation, pp. 514–5.
244 kiperwasser

found only in the Bavli,30 in which the admirable actions of the collaborator
from Luz are reflected in the excellent qualities of his new city. Into the new
city Death cannot enter, and everyone lives there forever. The narrator in bSot
46b quotes the above-mentioned verses and inquires regarding the word “we
will treat you kindly”:

bSot 46b31
What was the kindness they did to him? ?‫ומה חסד עשו עמו‬
They slew the whole of the city at the ‫ ואת האיש ואת‬,‫כל העיר הרגו לפי חרב‬
edge of the sword, but let that man and ‫ ״וילך האיש ארץ החתים‬:‫משפחתו שלחו שנ׳‬
his family go, as it is written: “The man ‫ויבן עיר ויקרא את שמה לוז [היא שמה עד‬
went to the Hittite country. He founded )‫היום הזה]״ (שפטים א כו‬
a city and named it Luz, and that has
been its name to this day” (Judg 1:26)
It has been taught: That is Luz in which :‫תאנא‬
they dye purple (‫)תכלת‬. .‫היא לוז שצובעין בה תכלת‬
That is the Luz against which ‫ בא נבוכדנצר‬,‫לוז שבא סנחריב ולא בלבלה‬
Sennacherib marched without disturb- ‫ ואף מלאך המות אין לו רשות‬,‫ולא החריבה‬
ing it, against which Nebuchadnezzar .‫ליכנס בתוכה‬
marched without destroying it, and even
the Angel of Death has no permission to
enter it, but when old men and women ‫ מוציאין‬,‫ בזמן שדעתן קצה עליהן‬,‫וזקנים וזקנות‬
there become tired of life they are taken .‫אותם חוץ לחומה ומתים‬
outside the wall and they die.

Thus in the Babylonian aggadah, the biblical toponym Luz became the city of
immortality. This was the reward given to the collaborator of the old Luz. It is
evident that the purpose of this tradition is to complement the short biblical
narrative and resolve the moral problems that could be raised after reading it.
This city, created as a midrashic compensation for the biblical hero, became
mythological. And in this far-away, mythological destination King Solomon
wanted to hide his beloved servants. It is a good refuge for the poor victims
of the Dispatcher, since the Angel of Death cannot enter its gate. But even the

30 The publisher of the printed edition of Genesis Rabba was influenced by the following
text from the Bavli and edited the Palestinian tradition in GenR 69:8 according to the
Bavli, but all the manuscripts show no evidence of Babylonian influence about the city of
immortals. See Theodor-Albecks edition p. 798.
31 The text is according to Oxford—Bodl. heb. d. 20 MS with the minor corrections.
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 245

king does not understand the way predestination works: death happened at
the gate of the city.
Now I would like to draw attention to the resemblance of Luz, the city of
immortality, to another city, which was the object of fantasies in Sasanian
Babylonia. The Sasanian messianic hero Pešyōtan, “who is without death,”32
lives his eternal life in Kangdiz, the “fortress of Kang/Gang.”33 It is said to be a
region other than Xwaniras (the central continent), situated in the direction
of the east, above the ocean Fraxhard.34 In Pahlavi literature, it is said that
Pešyōtan son of Wištasp will come from the direction of the fortress of Kang
with one hundred and fifty righteous men.35 He will destroy the place of idola-
try. The fortress was built, according to the Denkard, “for the retention and pro-
tection of the great power and glory and mystery of the religion, through which
is manifest the ordering of time and restoration of the sovereignty of Iran and
the restitution of power and triumph to the religion of Ohrmazd.”36 The image
of the city of Luz in the Bavli was probably influenced by the image of Kangdiz.
The Babylonian narrator’s representations of the City of Luz played a role in
Jewish eschatology as well; it is no coincidence that the tekhelet (translated in
the text above as “purple”) necessary for rituals in the eschatological future was
dyed in this city.37
The mythologemma of the city of immortals was formulated in the Bavli
under the influence of Iranian myth. In the magic city where Pešyōtan was for-
gotten by Death, King Solomon wanted to hide his servants, but came to grief.
In the Bavli the name of the Iranian mythic city was Hebraized and biblically

32 K. Anklesaria, “Dādestān ī Dīnīk,” in: M. Boyce and I. Gershevitch (eds.), Henning Memorial
Volume (Lund 1970) 12.
33 See P. Lurje, “Kangdez,” in: Encyclopedia Iranica (www.iranica.com/articles/kangdez);
M. Boyce, “On the Antiquity of Zoroastrian Apocalyptic,” Bulletin of the School of Oriental
and African Studies 47 (1984) 61–2.
34 For descriptions of the remote location of the city, see The Bundahishn, in E. W. West
(trans.), Sacred Books of the East, vol. 5 (Oxford 1897) 12:1, p. 34; 20:31, p. 82; 29:10, p. 119.
Regarding the list of immortal heroes living in different remote locations see ibid. 29:4,
pp. 117–8.
35 See Dadestan-i Denig (“Religious Decisions”) 90, in E. W. West (trans.), Sacred Books of the
East, vol. 8 (Oxford 1882) 256–7.
36 See Denkard in E. W. West (trans.), Sacred Books of the East, vol. 5 (Oxford 1897) Book 7,
1:38, p. 14, and 5:12, pp. 76–7.
37 Here is the appropriate place to mention the fact that, according to another aggadic story
from the Bavli (bBB 74a), tekhelet is stored in the depths of the sea in a special basket
protected by mythological monsters and awaiting its eschatological usage.
246 kiperwasser

supported. But the Jewish narrator kept the tremendous power of the Angel of
Death intact, because according to Jewish myth, even death derives from the
Creator, and not from the Evil one.
At this juncture another city of immortals described in the Bavli may be
mentioned, which is not identified with Luz, but which I propose also stems
from the Iranian myth of a miraculous city.

bSanh 97a
Rava said: I used to think at first that ‫ ליכא קושטא‬:‫ מריש הוה אמינא‬:‫אמר רבא‬
there is no truth (kushta) in the world. ,‫ אמר לי ההוא מרבנן ורב טבות שמיה‬.‫בעלמא‬
But one of the students of the sages, by ‫ דאי הוו יהבי‬,‫ רב טביומי שמיה‬:‫ואמרי לה‬
the name of Rav Tabuth, others say, by :‫ליה כל חללי דעלמא לא הוה משני בדיבוריה‬
the name of Rav Tavyomi, who, even if he ‫ וקושטא‬,‫זימנא חדא איקלעי לההוא אתרא‬
were given all the content of the world, ‫ ולא הוה‬,‫ ולא הוו משני בדיבורייהו‬.‫שמיה‬
would not lie, told me that he once came ‫ נסיבי איתתא‬.‫מיית איניש מהתם בלא זימניה‬
to a place called Kushta, in which no one ‫ יומא חד‬.‫ והוו לי תרתין בנין מינה‬,‫מינהון‬
ever told lies, and where no man ever died ‫ אתאי‬.‫הוה יתבא דביתהו וקא חייפא רישה‬
before his time. Now, he married one of ‫ לאו אורח‬:‫ סבר‬.‫שיבבתה טרפא אדשא‬
their women, by whom he had two sons. ‫ שכיבו ליה‬.‫ ליתא הכא‬:‫ אמר לה‬.‫ארעא‬
One day his wife was sitting and wash- ‫ אמרו‬.‫ אתו אינשי דאתרא לקמיה‬.‫תרתין בנין‬
ing her hair, when a neighbour came and .‫ הכי הוה מעשה‬:‫ מאי האי? אמר להו‬:‫ליה‬
knocked at the door. Thinking to himself ‫ ולא‬,‫ פוק מאתרין‬,‫ במטותא מינך‬:‫אמרו ליה‬
that it would not be proper etiquette [to .‫תגרי בהו מותנא בהנך אינשי‬
tell her that his wife was washing], he
said: She is not here. His two sons died.
Then people of that town came to him and
questioned him: What is the cause of this?
So he related to them what had happened.
They said: We pray you, quit this town, and
do not incite Death against us.

Another type of city of the immortals is represented in this story. The city
should not be identified with Luz, but certain similarities between the two
exist and its characteristics are obviously influenced by the same Sasanian tra-
dition. The citizens of Kushta, the name of which means “The Truth,” were
awarded immortality because of their uncompromising commitment to tell-
ing the truth and nothing but the truth.
In the beginning of the story the narrator, Rava, expresses the well-known
rabbinic idea: absolute truth does not exist in the world of humans, only the
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 247

seal of the Holy One is truth.38 Rava’s viewpoint is challenged by his opponent,
whose identity is unclear—Tabuth or Tavyomi.39
This character is known by his modesty and his refusal to utter a lie even in
exchange for all the wealth in the world. Yet when this proud person arrives at
a certain place named Truth (Kushta) where all the inhabitants tell nothing
but the truth, when residing there, he does not withstand the test of uncom-
promising truth-telling. Seemingly, the narrator is skeptical about human
beings’ ability to avoid uttering lies. The typical structure of the plot, in which
the hero fails the test and tells a lie, as a result of which he is excommunicated
and exiled from the dwelling place of the righteous truth-tellers, resembles a
Christian fable about the traveler Zosimos and his sin in the utopian city of the
permanently happy Rechabites:40

Therefore they were asking me all things and I was answering them, and
I became faint in spirit and in body, and besought the man of God that
served me, and said: I beseech thee, brother, if any come to see me, tell
them he is not here, so that I may rest a little. And the man of God cried
out, saying: Woe is me that the story of Adam is summed up in me, for
Satan deceived him through Eve, and this man by his flattery desires to
make me a liar while he is here. Take me away from hence, for I shall flee
from the place. For behold, he wishes to sow in me seeds of the world of
vanity. And all the multitude and the elders rose up against me, and said:
Depart from us, man; we know not whence thou art come to us.

Because of the polite lie which the hero of The Journey of Zosimos wished to
tell, he was immediately expelled from the wonderful place to which he had
come, and was excommunicated from the society of the chosen people. As in
the talmudic story, a polite lie of the hero was received by the inhabitants of
the remote city as the worst of sins. In the Christian story the sin is compared

38 I am paraphrasing the well-known saying from the Bavli (bShab 55a; bYom 69b; bSanh
64a), but a similar idea is found in Palestinian literature too, see yBer 1:5, 3c; ySanh 1:1, 18а;
LevR 6:1 (p. 145 = 26:1, p. 487); PRK 4:2 (p. 55).
39 The last person is a Babylonian sage of fifth generation see C. Albeck, Introduction to the
Talmud Bavli and Yerushalmi (Tel Aviv 19892) 409 [Hebrew].
40 M. R. James, “On the Story of Zosimos” and “Narratio Zosimi,” in Apocrypha Anecdota: A
Collection of Thirteen Apocryphal Books and Fragments, Now First Edited from Manuscripts,
Texts and Studies 2.3 (Cambridge 1893) 86–108, 118; C. H. Knights, “The Story of Zosimos or
The History of the Rechabites?,” Journal for the Study of Judaism 24 (1993) 235–45. For an
analysis of this story see R. Nikolsky, “The Provenance of ‘The Journey of Zosimos’ ” (Ph.D.
diss.; Jerusalem 2003) 153–63 [Hebrew].
248 kiperwasser

to the sin of Adam and therefore the man is expelled from the “Garden of Eden”
in which this community resides. The narrator wishes thereby to create the
foundation for a negative attitude to human deception in Christian theology.
Deception, even if it is necessary from a social point of view, is inappropriate
in an ideal society of true believers. Therefore the motif of the liar in the city of
truth seems primary in the Christian story, and secondary in the talmudic one.
The story about Zosimos was duplicated in various versions: Greek, Syriac and
Armenian and was widespread in the ancient Orient.41 It seems plausible that
the motif of the wondering stranger in the land of truth, and his typical failure
to comply with the rules of such a world was taken over by the talmudic story-
teller from this popular oriental environment and converted by him to comply
with his objectives. The process of the story’s conversion was powerful enough
to paint the foreign literary motif with the characteristic colors of talmudic
culture. Therefore we have in the talmudic story a different model of the city of
the immortals. In bSanh, the city is not identified with the City of Luz of bSuk.
Moreover, the inhabitants of Kushta are not really immortals, only long living,
and their longevity is a divine gift for their selfless love of truth.42 It should be
noted that the love of truth and the rejection of lies, as well as the perception
of a lie as something demonic is characteristic of the Zoroastrian doctrine.43
Lie, named mihōxt, is one of the demons in the Zoroastrian army of evil, cre-
ated by evil Ahariman for his harmful purposes.44 According to Zoroastrian
ethics, lying is a sin, even when performed with good intentions, and

41 See A. Zanolli, “La Leggenda di Zosimo secondo le Redazione Armena,” Giornale della
Societa Asiatica Italiana NS 1 (1925–8) 146–62; F. Nau, “La légende inédite des fils de
Jonadab, fils de Réchab, et les iles fortunées, Jacques d’Edesse,” Revue Semitique 7 (1899)
54–75; Ronit Nikolsky, “The ‘History of the Rechabites’ and the Jeremiah Literature,”
Journal for the Study of the Pseudepigrapha 13 (2002) 185–207.
42 Such is also the case in The Journey of Zosimos, see Nikolsky, “The Provenance,” 90–102.
43 See Čīdag andarz ī pōryōtkēšān, which is also known by the title Pand-nāmag ī Zardušt
J. M. Jamasp-Asana, Pahlavi Texts [Bombay 1913], 41–50), see Encyclopedia Iranica (http://
www.iranica.com/articles/cidag-andarz-i-poryotkesan-selected-precepts-of-the-ancient-
sages-a-post-sasanian-compendium-of-apothegms-intended). English translations are
available in: M. F. Kanga (ed.), Čītak Handarž i Pōryōtkēšān: A Pahlavi Text (Bombay 1960),
and H. S. Nyberg, Hilfsbuch des Pehlevi I (Uppsala, 1928) 17–30.
44 See P. O. Skjærvø, “Truth and Deciption in Ancient Iran,” in: C. Ceretu and F. Vajifdar
(eds.), Atas-e Dorun: The Fire Within: Jamshid Soroush Soroushian Commemorative Volume,
vol. 2 (Bloomington 2003) 383–434, and also J. S. Mokhtarian, “Rabbinic Depictions of the
Achaemenid King Cyrus the Great: The ‘Babylonian Esther Midrash’ (bMeg 10b–17a) in
its Iranian Context,” in: C. Bakhos and M. R. Shayegan (eds.), The Talmud in its Iranian
Context (Tübingen 2010) 129–30.
THE MISFORTUNES AND ADVENTURES OF ELIHOREPH AND AHIAH 249

metonymic proximity between a lie and death is also observed in Zoroastrian

scriptures.45
But let us return to the City of Luz. The mythological city of the Mazdean
religion was “converted” to Judaism in the aggadah of the Babylonian sages. The
Babylonian narrators accepted the nucleus of the story from the Palestinian
transmitter. The Palestinian story was about fate, about not being able to avoid
the Last Judgment. The Babylonian narrators enriched its content with con-
cepts of the culture to which they belonged. The Palestinian tradition was
shortened; the aphorism became the frame of the story and in the end was
attributed to King Solomon himself. All these additions and changes add a
philosophical meaning to the story—a person is guided to his death by his
deeds, and every step of his life paves the way to it. Even if he decides to run
away from fate, the escape route will bring him directly to his tomb. Thus, the
Jewish-Babylonian storyteller preserved the power of the Angel of the Death,
who remained undefeated.
By studying the metamorphosis of this story we saw a typical development.
The story starts as a short Palestinian plot. After it migrated to Babylonia, it
acquired a few features of the local culture, but in general maintained the pur-
pose of the Palestinian story, taking its ideas through a new stage. When told
against the background of Iranian culture it acquires a further level, that of the
Iranian mythologemma about immortality, but keeps the traditional values of
the original Palestinian story.

45 Mēnōg ī xrad 10, p. 50, E. W. West, The Book of the Mainyo-I-Khard (Stuttgart, 1871); idem,
Pahlavi Texts III: čē ahreman mihōxt ī wad menēd u-š kār xešm ud kēn ud anāštīh, see
n. 35.
Commercial Law in Rome and Ctesiphon:
Roman Jurisconsults, Rabbis and Sasanian
Dastwars on Risk
Yaakov Elman

It has often been observed that, but for the dedication, Justinian’s Code con-
tains little that can be understood as “Christian,” and Alan Watson has tried
to explain the severing of Roman private law from Roman religious law on
historical grounds.1 In the following, I wish to approach this question from a
comparative point of view by examining Roman, rabbinic and Sasanian regu-
lation on three issues of commercial law: overreaching, buyer-protection, and
land fraud—three areas in which rabbinic and Sasanian law are surprisingly
similar, while Roman law, despite its acute awareness and avowal of justice,
is governed by the rule of caveat emptor, and affords little protection to the
weaker party.
Needless to say, this was not hypocrisy on the part of the Roman jurists,
no more than the dependence of the Roman economy on slave labor was
hypocritical in light of Julian’s reference to the “principle favoring freedom”
(D.40.4.17.2),2 or Justinian’s declaration that slavery is “contrary to the law of
nature” (J.1.3.2).3 Slavery was an accepted institution of long standing and
universal validity; it would take a long time for the human race to arrive at
a consensus outlawing it—and even then it still exists!4 Roman commercial
law was the outgrowth of a complex confluence of factors, not least of which
was the existence of an unlegislated area of life, which would only slowly yield
to the idea that it was for the betterment of the individual and society that it
be regulated. But it was also the consequence of a view of human interaction
and government responsibility very different from that of rabbinic Judaism
and Zoroastrianism. Despite the ringing declaration that opens Justinian’s

1 See A. Watson, The Spirit of Roman Law (Athens GA 1995) 42–56, and especially his comments
on page 45.
2 All citations from Justinian’s Digest are made from the translation edited by A. Watson: The
Digest of Justinian (Philadelphia 1998) (no pagination).
3 See P. Birks and G. McLeod (trans.), Justinian’s Institutes (Ithaca NY 1987) 39.
4 See O. F. Robinson’s comments in “Persons,” in: E. Metzger (ed.), A Companion to Justinian’s
Institutes (Ithaca NY 1998) 18, and see G. MacCormack’s discussion of that point in “Sources,”
in the same volume, pp. 4–5.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�10

ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 251

Institutes, “justice is the unswerving and perpetual determination to acknowl-

edge all men’s rights,” and “learning in the law . . . entails knowledge of God
and man, and mastery of the difference between justice and injustice,” such
assertions remained, as so often, ideals rather than principles for social pol-
icy. And even after the victory of Christianity, with its Jewish base, Roman law
remained essentially agnostic on the issue of legal intervention regarding the
issues that will be discussed below, though it should be noted that in some
areas of the Empire, particularly Egypt, things were different.5
It is not my intention to trace the history of the particular laws that will be
examined, even if that were possible. The mishnaic law of ona’ah (overreach-
ing), which can be dated to the early second century, had a prehistory that
we cannot trace, but that does not impair prospects for a fruitful comparison
of the rabbinic view with others of late antiquity.6 Likewise, since our major
source for Sasanian law dates to the early sixth century (although the rules
themselves may be earlier), we cannot trace their prehistory. As a result, I will
not argue for, or trace, cultural or legal influences in one direction or the other.
The following is intended as a comparative study that seeks to highlight
similarities or differences in economic policy that may stem from deeper
cultural attitudes. However, because one pole of my concern is the policy of
the Sasanian jurists, it seems wise to confine this study, whenever possible,
to rabbinic sources of the Babylonian Talmud (hereafter: the Bavli). Though
the Bavli, Justinian’s Code and the Sasanian Lawbook were all compiled dur-
ing the period 530 to 620 C.E., each of them is a compilation that includes
earlier material and betrays a good deal of continuity. This is also true of our
first case, which is centered on mishnaic legislation of the early second cen-
tury, but which remained in force, at least in theory, until the close of the Bavli
three centuries later. Since I am more concerned with values than with actual
practice (though I am not indifferent to it either), and in this case the Bavli
certainly retains the values of the earlier compilation, such a comparison will
be both valid and fruitful.

5 Thus, in Roman Egypt, overreaching could be remedied; see R. Taubenschlag, The Law of
Graeco-Roman Egypt in the Light of the Papyri, 332 B.C.–640 A.D. (Warsaw 1954) 330–1.
6 B.-Z. Rosenfeld and J. Menirav, “On the Meaning of Fraud and the Calculation of ‘Sixth’ (shtut)
in Tannaitic Law,” Diné Israel: Studies in Halacha and Jewish Law 22 (2003) 151–5 [Hebrew].
For a general view of rabbinic efforts to control the cost of living, see B.-Z. Rosenfeld and
J. Menirav, Markets and Marketing in Roman Palestine (Leiden 2005) 137–70, 171–210, for the
administrative mechanisms put in place for enforcing the legal determinations. For a more
socioeconomically oriented view of the material, see H. Lapin, Early Rabbinic Civil Law and
the Social History of Roman Galilee: A Study of Mishnah Tractate Baba’ Mesi’a’ (Atlanta GA
1995) 175–86.
252 elman

My goal is hardly as ambitious as J. A. Cook or D. Johnston, in their Law and

Life of Rome: 90 B.C./–A.D. 212 and Roman Law in Context, respectively, or as
focused on marriage sociology and legal history. I am hopeful that this rela-
tively crude—but useful—comparative study of juristic values as expressed in
law will allow us a hold on essential views on human nature and society that
lay at the roots of each of these legal systems.
Finally, the question of what power the rabbinic courts of Roman Palestine
may or may not have had in the third century, or the Babylonian rabbinic courts
of the fourth century, a matter that has been debated with much passion for
over a century, and particularly in the last generation, is also not my concern.
Again, I wish to get at the cultural/religious roots of questions of social and
economic policy, and not necessarily at how they were implemented. It would
seem that in all the cases considered below, the remedy is (or is not) legal
action undertaken by the plaintiff, not by the agronomos in charge of weights
and measures, or some similar official.
I will first take up these issues separately, and compare them afterwards.

1 Overreaching

The rabbinic rules for ona’ah (overreaching) are set out in two mishnayot:
mBM 4:3–4. Here is the Hebrew text along with the translation to be found in
the Soncino edition of the Bavli, with some minor changes.

.‫ שתות למקח‬,‫ האונאה ארבעה כסף מעשרים וארבעה כסף לסלע‬:‫משנה‬

.‫עד מתי מותר להחזיר? עד כדי שיראה לתגר או לקרובו‬
,‫ שליש למקח‬,‫ האונאה שמונה כסף מעשרים וארבע כסף לסלע‬:‫הורה רבי טרפון בלוד‬
‫ יניח לנו רבי טרפון‬:‫ אמרו לו‬.‫ כל היום מותר לחזור‬:‫ אמר להם‬.‫ושמחו תגרי לוד‬
.‫ וחזרו לדברי חכמים‬,‫במקומינו‬
‫ כשם שאונאה להדיוט כך‬.‫ יש להן אונאה‬,‫ אחד הלוקח ואחד המוכר‬:‫משנה‬
.‫ אין אונאה לתגר‬:‫ ורבי יהודה אומר‬.‫אונאה לתגר‬
‫ או תן לי מה‬,‫ תן לי מעותי‬:‫ אומר לו‬,‫ רצה‬.‫ ידו על העליונה‬,‫מי שהוטל עליו‬
.‫שאניתני‬

Mishnah: Fraud is constituted by [an overcharge of] four silver [ma’ahs]

in twenty-four, which is a sela, [hence,] a sixth of the purchase.
Until what time is one permitted to revoke [the sale]? Until he can show
[the article] to a merchant or a relative.
Rabbi Tarfon ruled in Lod that fraud is constituted by eight silver [ma’ahs]
in twenty-four, which is a sela, [hence] a third of the purchase, at which
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 253

the merchants of Lod7 rejoiced. But, said he to them, one may retract
the whole day. Then let Rabbi Tarfon leave us in the status quo, they
requested, and so they reverted to the ruling of the Sages.

Mishnah: Both the buyer and the seller can claim for overreaching; just
as the law of overreaching holds good in the case of a layman, so it holds
good in the case of a merchant. Rabbi Yehudah said: There is no [law of]
overreaching for a merchant.
He who was deceived has the upper hand; if he wishes, he can either say:
Give me my money, or: Return what you overcharged me.

Although I have retained Soncino’s rendering of ona’ah as “fraud,” which

reflects the biblical sense of the phrase lo tonu (‫ )לא תאנו‬of Lev 25:4, the rabbis
extended the rule to encompass unintentional overreaching; hence the three-
day grace period for either party to consult an expert on the value of the object
in question. We should also note that the Mishnah takes private sales into
account; hence the mention of a “layman” in the second passage. However, a
merchant is not necessarily considered more knowledgeable than a layman,
and so he too is permitted the grace period, as well as the possibility of recov-
ering the disputed amount. It is important to note that this point is disputed
by Rabbi Yehudah, and that both his position and its rationale are disputed by
later Babylonian authorities.

?‫ משום שהוא תגר אין לו אונאה‬.‫ אין אונאה לתגר‬:‫רבי יהודה אומר‬
,‫ מאי טעמא? מידע ידע זבינתיה כמה שויא‬.‫ בתגר ספסר שנו‬:‫ אמר רב‬,‫אמר רב נחמן‬
‫ והשתא‬,‫ והאי דזבנא הכי—משום דאתרמיא ליה זבינתא אחריתי‬.‫ואחולי אחיל גביה‬
!‫מיהא קא הדר ביה‬
‫ שאפילו פחות מכדי‬,‫ מאי אין לתגר אונאה? אינו בתורת אונאה‬:‫רב אשי אמר‬
.‫אונאה—חוזר‬
.‫ מפני שהוא בקי‬,‫ תגר אין לו אונאה‬:‫ רבי יהודה אומר‬:‫תניא כוותיה דרב נחמן‬

Rabbi Yehudah said: There is no [rule of] overreaching for a merchant.

[The redactors ask:] Because he is a merchant, has he? no claim for
overreaching?
Said Rav Nahman in Rav’s name: This was taught of a safsar [broker, mid-
dleman]. Why? Because he well knows the value of what he sells, but
foregoes [part of it] to [the buyer]; the reason that he sells thus [cheaply]

7 For the nature of their profession, see Rosenfeld and Menirav, Markets and Marketing, 15–70.
254 elman

being that he has chanced upon another purchase, and now wishes to
retract.
Rav Ashi said: What is meant by: There is no overreaching for a merchant?
He is not subject to the law of overreaching, i.e. he can withdraw even for
less than the [recoverable] standard of overreaching.
It has been taught in accordance with Rav Nahman: Rabbi Yehudah said:
There is no overreaching for a merchant, because he is an expert.8

Soncino chose Jastrow’s “speculator” to render the Hebrew safsar, but Sokoloff’s
“broker, middleman,” based on New Persian safsir is more to the point.9 The
differing reasons given for this exception by the third-generation Rav Nahman
and the sixth-generation Rav Ashi may reflect the narrowing margins of profit
available in the late fourth and early fifth centuries, perhaps in the wake of the
Roman invasion of 363, over conditions at the beginning of the century; how-
ever, the essential point is that the Babylonian amora Rav Nahman interprets
the words of the tanna Rabbi Yehudah as based on his considering a merchant
to be an expert who knows the price of whatever he sells. Rav Ashi’s interpre-
tation of the passage negates this reasoning by pointing to his narrow profit-
margin. In any case, whatever the details of the application of the rabbinic law
of ona’ah, the essential point is that they attempted to regulate sales in terms
of the “correct price.”
The two essential points of the mishnaic law of overreaching are: (1) the
setting of a rule for calculating overreaching, and thus, to some extent, maxi-
mum profit-margin, and (2) the establishment of a period of three days (more
or less, depending on circumstances) for withdrawal. Both aspects of the law
are to be found in the so-called Sasanian Lawbook, more properly, Mādāyān
ī Hazār Dādestān, “The Book of a Thousand Decisions.” The book is a collec-
tion of laws set down around 620 CE by one Farroxmard ī Wahrāmān, and sur-
vives in only one manuscript, which, in the course of time, had been divided
between two heirs; the two parts are designated as MHD and MHDA. The entire
work has benefited from two editions in the last generation, one in Russian

8 A word should perhaps be said about the translations used in the present study. For the Bavli,
I have taken the Soncino Talmud as my basis, though I have made minor stylistic changes,
and in a few places, substantive ones. For Sasanian law, I have relied on Maria Macuch’s edi-
tion of the so-called Sasanian Lawbook (see n. 10), and have translated her German and made
a few stylistic alterations, especially regarding names. The citations from Justinian’s Digest
are from Watson’s edition, and are used without change.
9 See M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic Periods
(Ramat Gan 2002) 826b, s.v. safsara.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 255

(published in an English translation in 1969) by Anahit Perakhanian, with very

few notes. The second, in German, by Maria Macuch, contains a voluminous
commentary which deals with both philological and legal issues; it was pub-
lished in two parts (in 1981 and 1993).10 MHD is our major source for Sasanian
law, which, though dating from the first quarter of the seventh century, con-
tains provisions dating back to at least the fourth century, as we know from
parallels in the Bavli.11
The Sasanian laws regarding overreaching appear in two places, one in each
part of the manuscript. The more detailed description appears in MHD 37.2–10.

MHD 37.2–10
ka šōy ō zan gōwēd kū guharīn kun zan (3) guharīn rāst pādixšāy kard
ud ka šōy nē guft ēstēd kū pad čahār ē(w) bahr pad (4) šōy mīrmadtar
pādixšāy kardan kū ka frāy kū pad čahār bahr ē(w) bahr az wēš pad ān ī
az wēš (5) duz čē guharīn ēdōn pādixšāy kardan ka frāy kū čahār ēk-ē(w)
az wēs nēst ud ziyānag (6) abāz nē pādixšāy ēstād. ud ān kē guharīn abāg
hamē kunēd sē šabag abāz pādixšāy (7) ēstād paydāg kū ka mard ē(w) ō
kas ē(w) gōwēd kū guharīn kun ān kē awiš gōwēd (8) guharīn rāst pādixšāy
kardan ud andar-iz sē šabag abāz nē pādixšāy ēstād. ud ēn-iz kū (9) ān kē
ōy kas pad guharīn kardan pādixšāy kunēd andar sē šabag pādixšāy ēstād
(10) ōh abāyēd uskārdan.12

Here is my translation of Macuch’s German, slightly modified in order to con-

form to English syntax.

If a husband says to (his) wife: Undertake an exchange (of this thing),

(then) the wife is authorized to undertake an even exchange. And when
the husband has made no declaration (in that regard), (the wife) is
authorized to undertake (the exchange), so long as it brings the husband
a profit of four parts to one part. And when (the profit is) greater than one

10 Anahit Perikhanian, The Book of a Thousand Judgements (trans. N. Garsoian; Costa Mess
CA 1997) and M. Macuch, Das sasanidische Rechtsbuch “Mātakdān i Hazār Dātistān,”
Part II (Wiesbaden 1981), and idem, Rechtskasuistik und Gerichtspraxis zu Beginn des
siebenten Jahrhunderts im Iran: Die Rechtssammlung des Farrohmard i Wahrāmān
(Wiesbaden 1993).
11 See the cases discussed in Y. Elman, “Returnable Gifts in Rabbinic and Sasanian Law,”
Irano-Judaica 6 (2008) 150–95, and “ ‘Up to the Ears’ in Horses’ Necks: On Sasanian
 

Agricultural Policy and Private ‘Eminent Domain’,” Jewish Studies: An Internet Journal 3
(2004) 95–149.
12 Macuch, Das sasanidische Rechtsbuch, Part I, 273 (text), 274 (translation), and 275–6, n. 2.
256 elman

fourth, (i.e. it) exceeds that (az wēš), (then), on the basis of that (trans-
action), which exceeds it, she is (considered) a thief, since she is (only)
authorized to undertake an exchange when (the profit) is not greater
than a fourth, and does not exceed that. The wife is not authorized to
withdraw. But the one with whom the exchange was made is authorized
to withdraw (within the space) of three nights.
Thus, it is clear that when a man says to someone: Undertake this
exchange (of this thing), the one to whom the declaration was given is
(then) authorized to undertake an even exchange, and is not authorized
to withdraw (in the space) of three nights. And regarding (the case) of
that one who is authorized to undertake an exchange, s/he is entitled
to withdraw within the space of three nights—considerations must
be made.

Macuch’s analysis of the passage, as detailed in her n. 2, is as follows:

The case is as follows: A man can give his wife authorization to con-
clude an exchange. In this case the wife is given the right to undertake
an even exchange (guharīn rāst), by which she gains no profit from the
man’s property. She is however authorized, even without a declaration
by her husband, to conclude an exchange, when the business provides a
profit for the spouses and a gain of a fourth of the worth of the exchange,
yielding a profit (ka pad čahār bahr ē[w] bahr pad šōy mīrmadtar). The
following restriction is noteworthy: the profit produced by the exchange
must not exceed a fourth of the value of the property exchanged. When it
produces more, the wife is considered a thief in regard to this excess (pad
ān i az wēš duz). She is not authorized to withdraw from the exchange
transaction after its conclusion, since it is clear from the following expla-
nation, that someone who gives authorization for an exchange transac-
tion with a third party has basically only the right for an even exchange,
but no right to cancel the transaction. The partner in the exchange, in
contrast, can withdraw within three nights (sē šabag). If the one who
gave authorization for the exchange (in our case, the spouse) has the
right to withdraw within the specified time of three nights, each case
must be evaluated (that at least is my understanding of the meaning
of abāyēd uskārdan at the end of the sentence). See in this connec-
tion MHDA 12.3–5 regarding this exchange transaction and the “three
nights” time of reconsideration . . . DKM [an edition of the Dēnkard,
a ninth-century encyclopedia of Zoroastrianism—Y.E.] also speaks
of sē šabag in a section regarding an “exchange transaction” (brīnag i
guharīgstān).
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 257

abar guharīg ānēran kardan kū čiyōn ān i kā (13) pādihšāy, abar gōspand

guharīg andar sē šabag pas az sē (14) šabag kuštan . . .

Regarding the conclusion of an exchange (transaction) with non-Iranians,

how the (transaction) (should be) properly (concluded): Regarding the
exchange of a sheep, (regarding whose) slaughter within three nights,
after the (lapse) of three (14) nights . . .

Before dealing with MHD’s legislation proper, let me point out the univer-
salistic goals of Zoroastrian legislation. The passage that Macuch quotes
from the Dēnkard applies the rules of exchange to non-Iranians, that is, non-
Zoroastrians, a theme that is sounded in the Avestan Vidēvdād, the Zoroastrian
scriptural book that deals primarily with matter of purity and pollution, but
also includes a chapter on contracts. This provision of Zoroastrian legislation
will be of help to us in limning its contours in comparison to those of Roman
law. It should be noted that for the Iranians, as rulers of a vast multi-ethnic
empire, these provisions were hardly utopian!
The casting of the law of overreaching in terms of a wife’s transaction as
authorized by her husband may seem strange to us; why not formulate the
rules of sale directly? And why deal with a case of exchange or barter when
in the end the issue seems to have been a question of lawful profit margins?
Unfortunately, the answer to this question is impossible because the line
before this passage is broken, and it has been estimated that as many as two
folios are missing. However the chapter, of which our passage is the end, seems
to have been devoted to the principles of exchange, and more basic scenar-
ios may have been considered in the missing folios. Because the anonymous
author of this passage wished to deal with the various possibilities engendered
by a wife’s acting for her husband in an exchange transaction, he dealt with the
issues of allowable profit-margin and the three-day time of reconsideration,
when one or both parties may withdraw.
It may be noted, however, that the other part of the manuscript contains a
paragraph partially relevant to our passage.

MHDA 12.3–5
Dād-Farroxw13 ī Farroxw-Zurwān bōzišn ēwag ēn guft (4) kū ka zan guharīn
kunēd ud šōy andar sē šabag ī zan rasēd u-š abar (5) ēstēd ka sē šabag ī zan
šud bē bawēd.14

13 I have slightly altered Professor Macuch’s rendering of compound names and also
substituted an “x” for her underscored “h” for Pahalavī het.
14 Macuch, Das sasanidische Rechtsbuch, Part II, 125.
258 elman

Dād-Farroxw ī Farroxw-Zurvān gave the following solution (= decision)

of this (case): If a wife undertakes an exchange (of property) and the hus-
band desists (from protesting) during the three nights (of consideration)
that is due the wife, it remains in custody. When the three nights (of con-
sideration) of the wife are over (the transaction is valid).

The details are not clear, but the general rules are, and they resemble those
of the Mishnah to a marked degree. The Mishnah allows for a maximum of a
sixth variation from either the generally accepted price or the cost price, MHD
provides for a profit of a quarter, and both allow for a period of three days (or
nights) of reconsideration. In the passages now before us there is no indication
of how that margin is calculated. For the possibilities, let us look at the Bavli’s
analysis of the question. Once again I reproduce the Soncino translation with
minor changes, but the comments are mine.

.‫ שתות מעות נמי שנינו‬:‫ ושמואל אמר‬,‫ שתות מקח שנינו‬:‫ רב אמר‬:‫ אתמר‬:‫גמרא‬
,‫ שוי שיתא בשבעה—כולי עלמא לא פליגי דבתר מקח אזלינן‬,‫שוי שיתא בחמשא‬
:‫ לשמואל דאמר‬.‫ ושוי שבעה בשיתא‬,‫ שוי חמשא בשיתא‬,‫ כי פליגי‬.‫ואונאה הויא‬
‫ שוי חמשא‬,‫ בתר מקח אזלינן‬:‫ לרב דאמר‬.‫ אידי ואידי אונאה הוי‬,‫בתר מעות אזלינן‬
:‫ כי אמרינן‬:‫ ושמואל אמר‬.‫ שוי שבעה בשיתא מחילה הויא‬,‫ ביטול מקח הויא‬,‫בשיתא‬
‫ אבל היכא דאיכא שתות מצד‬.‫מחילה וביטול מקח היכא? דליכא שתות משני צדדים‬
.)‫ אונאה הויא (ב׳ בבא מציעא מט ע״ב‬,‫אחד‬

GEMARA. It has been stated: Rav said: We learnt, A sixth of the [true]
purchase price. Shmuel said: A sixth of the money [actually] paid we also
learnt.
Now, if that which is worth six [ma’ahs] was sold for five or seven, all
agree that we follow the purchase price [in assessing whether there
was overreaching or not. Since one out of six, plus or minus, equals a
sixth] and there was overreaching. Where do they differ? [They differ in
a case where] something worth five or seven [ma’ahs] was sold for six.
According to Shmuel, who maintained that we follow the money paid [as
well], both cases constitute overreaching. But according to Rav, viz., that
we follow only the purchase price, if something worth five is sold for six,
the sale is null, [for this constitutes 20% over the value, and the sale is
null, so that either party can withdraw]; but if what is worth seven is sold
for six, it is renunciation [since a seventh, about 14%, is less than a sixth,
16.6%]. But Shmuel maintained: When do we say that there is renun-
ciation or annulment of the sale? Only if there is not a sixth on either
side [that is, whether calculated from the “true” (or perhaps wholesale)
price, or the purchase price], but if there is a sixth on [only] one side, it is
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 259

overreaching, [and what is done in that case is a matter of debate among

medieval commentators] (bBM 49b).

The rabbis thus considered two means of assessing ona’ah, one calculated
according to the “true” price, and one according to the purchase price. There
is another way of expressing the margin, although the terminology is not used
here, but rather regarding the various additions of a biblical (and occasional
rabbinic) fifth (‫ )חומש‬for calculating the amount of money needed for pur-
poses of redemption or the biblical fine of a fifth added for misuse of sacred
property. The question is raised below (bBM 53b–54a), whether the calculation
is made “inside” (calculating the sixth according to the lower, “true” price) or
“outside” (calculating the sixth from the higher price actually paid, the pur-
chase price). Here is the passage:

?‫ חומשא מלגיו או חומשא מלבר‬:‫איבעיא להו‬

‫ וכל אדם בעשרים—הבעלים‬,‫ הבעלים אומרים בעשרים‬:‫ תא שמע‬:‫אמר רבינא‬
‫ הרי עלי בעשרים ואחד—הבעלים נותנין‬:‫ אמר אחד‬.‫ מפני שמוסיפין חומש‬,‫קודמין‬
—‫ בעשרים ושלש‬.‫ עשרים ושנים—הבעלים נותנין עשרים ושבע‬.‫עשרים ושש‬
.‫ בעשרים וארבע—הבעלים נותנין עשרים ותשע‬.‫הבעלים נותנין עשרים ושמונה‬
.‫ לפי שאין מוסיפין חומש על עילוי של זה‬,‫עשרים וחמש—הבעלים נותנין שלשים‬
,)‫ ״ויסף חמשתו עליו״ (ויקרא כז כז‬:‫ כתנאי‬.‫ שמע מינה‬,‫ חומשא מלבר‬:‫שמע מינה‬
‫ רבי יונתן אומר‬.‫ דברי רבי יאשיה‬,‫שיהא הוא וחומשו חמשה‬:x ‫״חמשתו״—חומשו‬
.‫של קרן‬

The scholars propounded: Is the fifth calculated on the inner sum [sc. the
principal] or on the outer [sc. the principal plus the addition]?
Said Ravina: Come and hear: If the owners value it at twenty [selas]
and every other person [values it at] twenty—the owners have prior-
ity, since they add a fifth. If a stranger declared: I accept it for twenty-
one, the owners must give twenty-six; for twenty-two, the owners must
give twenty-seven; for twenty-three, the owners must pay twenty-eight;
for twenty-four, the owners must pay twenty-nine; for twenty-five, the
owners must pay thirty; because a fifth is not added on this man’s higher
valuation. This proves that the fifth is calculated on the outer sum. This
proves it. As it is taught: “And he shall add his fifth to it” (Lev 27:27), so
that him and his fifth be five, these are the words of Rabbi Yoshayah.
Rabbi Yonatan says: “his fifth” (ibid.)—a fifth of the investment.

This raises the question: How, then, is the Sasanian fourth calculated? In addi-
tion, we may ask: Were Jewish merchants then constrained by rabbinic law
to sell at about 8% below the Sasanian market level? This in turn leads us to
260 elman

another issue: One sixth is a particularly uncomfortable fraction when dealing

with a currency based on the decimal system. As Daniel Sperber noted long
ago in discussing land prices in Roman Palestine, the standard rabbinic system
of currency was of course not really in use; there was no real-world sela made
up of four dinarim, or a golden dinar made up of four selas, or in the case of our
mishnah, a silver ma’ah, 24 of which would make up a sela.15 Did Babylonian
merchants use such an inconvenient fraction? We cannot answer this ques-
tion, but we can hazard a guess as to how the rabbis came to use it. Note that
almost all the rabbinic systems of weights, measures and coinage are based
on the sexagesmal system, and are fossils from ancient Babylonian times. The
same goes for the (Babylonian) rabbinic use of the numbers “60” and “300”
for expressing “a great number.” Whether the use of a sixth as a fraction in the
Mishnah may be explained as a survival of high antiquity may be debated, but
such survivals do exist, as Yohanan Muffs and Baruch Levine pointed out a
generation ago.16 Rosenfeld and Menirav have suggested a perhaps less exotic,
though less inclusive, theory.17
This then explains, at least in part, the discrepancy between the Sasanian
and rabbinic estimates of overreaching. The Sasanian fourth reflects an alle-
giance to the Greek system, as for example in the case of the drachma and the
tetradrachma; the rabbinic sixth to the Babylonian sexagesmal. The dang was
a sixth of a drachma, and a drachma was a fourth of a sesterce; in both sys-
tems, the fractions chosen for overreaching were designed for easy calculation
in terms of their monetary systems, and the Sasanian system was convenient
enough for rabbinic usage. In this case, the fact that the sexsegismal system
was not in general use anymore was irrelevant, since the numbers “four” and
“six” were! Thus, viewing the rabbinic sixth as 16.666 . . . % (an irrational num-
ber, mathematically speaking) is to impose our notions of a decimal system on
two monetary systems, neither of which was strictly decimal.
I now move on to Roman legislation on the issue. Commenting on two laws,
Alan Watson has observed: “in classical Roman law in the contracts for sale and
hire there was no notion of a just price. The price had to be seriously intended,
but there was no requirement that it should correspond in value in some way

15 See D. Sperber, Roman Palestine, 200–400: Money and Prices (Ramat Gan 1974) 27–30.
16 See Y. Muffs, Studies in the Aramaic Legal Papyri from Elepantine (Leiden 1969), and
B. A. Levine, “Mulugu/melug: The Origins of a Talmudic Legal Institution,” Journal of the
American Oriental Society 88 (1968) 271–85. Levine also published a prolegomenon to a
reprint of Muffs’ work by Brill in 2003.
17 See Rosenfeld and Menirav, “On the Meaning of Fraud,” 175.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 261

to the thing sold or hired, or the services provided. This is very explicitly set out
in the Digest:

D.19.2.22.3 PAULUS libro trigesimo quarto ad edictum: Quemadmodum

in emendo et uendendo naturaliter concessum est quod pluris sit minoris
emere, quod minoris sit pluris uendere et ita inuicem se circumscribere, ita
in locationibus quoque et conductionibus iuris est:18

D.19.2.22.3 [Paul, On the Edict, book 14]: Just as in the contract of sale it
is naturally permitted to buy for less what is worth more, and to sell for
more what is worth less, and thus to take advantage of one another, so
that is the law in the case of hire.

D.4.4.16.4 ULPIANUS libro undecimo ad edictum: Idem Pomponius ait

in pretio emptionis et uenditionis naturaliter licere contrahentibus se
circumuenire.

D.4.4.16.4 [Ulpian, On the Edict, book 11]: Likewise, Pomponius says with
regard to the price in sale, that the contracting parties are naturally per-
mitted to take advantage of one another” [italics mine—Y.E.].

Note the last words of D.19.2.22.3. They imply that it is not only natural for
people to take advantage of one another, but it is naturally permitted. However,
let us here note that neither the rabbis nor the Sasanian jurists held this view of
law. Indeed, ancient Babylonian cuneiform law too concerns itself with price-
fixing and the proper rental to be charged for the rental of oxen in a team. The
latter may be found in the Laws of Lipit Ištar, fifth ruler of the Second Dynasty
of Isin (r. 1934–1924 BCE),19 and the former in the Laws of Eshnuna (ca. 1770),
whose first four paragraphs deal with the proper prices for various commodi-
ties and the rental for oxen.20 Indeed, we may go further, for cuneiform law, like
rabbinic and to some extent Sasanian law—but unlike Roman law—attempts
to regulate marriage even beyond the matter of marital property. In these
respects, as in others, the Roman jurists adopted a laissez faire policy.

18 The Latin text here, and in succeeding passages, is taken from the Mommsen-Krueger
edition as reproduced in Alan Watson, The Digest of Justinian, Latin Text Edited by Theodor
Mommsen with the aid of Paul Krueger, English Translation (Philadelphia 1985).
19 See M. T. Roth, Law Collections from Mesopotamia and Asia Minor (Atlanta GA 1995) 26.
20 Roth, Law Collections, 59–60.
262 elman

This is not as heartless as it may sound, since ancient empires were very
far from the totalitarian regimes that modern technology has enabled. Even in
much less far-reaching matters, such as registration of land sales (see below),
the very use of rules of land-ownership that depended on occupation rather
than title search indicate as much. But the Romans seem to have been much
more hard-headed than other ancient legislators, and did not even attempt to
express a negative opinion on overreaching.
However, that does not mean that such behavior was approved; here is
Cicero in de Officiius:

Sordidi etiam putandi, qui mercantur a mercantoribus, quod statim ven-

dant; nihil enim proficiant, nisi admodum mentiantur; nec vero est quiquam
turpius vanitate. . . . Mercantura autem, si tenuis est, sordida putanda est;
sin magna et copiosa, multa undique apportans multisque sine vanitate
impertiens, non est admodum vituperanda. . . .

Vulgar we must consider those also who buy from wholesale merchants
to retail immediately; for they would get no profits without a great deal of
downright lying; and verily, there is no action that is meaner than misrep-
resentation . . . Trade, if it is on a small scale, is to be considered vulgar;
but if it is wholesale and on a large scale, importing large quantities and
distributing it to many without misrepresentation, it is not to be greatly
disparaged . . .21

In other words, what can we expect from the lower classes? In Book III Cicero
tells of Quintus Scaevola, who paid 100,000 sesterces more for a farm than was
asked because he felt that is was worth more; Cicero laments that “here then is
that mischievous idea—the world accounting some men upright, others wise”
(haec igitur est illa pernicies, quod alios bonos, alios sapientes existimant).22 But
then Cicero, as a representative of old Roman republican values, was on the los-
ing side, and the Empire was an intensely practical affair of an intensely prac-
tical people. As Cicero would have put it, expediency won out over morality.
But practicality cuts both ways, and no one likes to be cheated in one’s own
affairs, and few if any people can be forever vigilant; a society which operates
under a sort of collective paranoia would consume itself in a few generations.
The theme of misrepresentation and fraud that runs like a red thread through

21 Cicero, de Officiis I, xlii (LCL, with an English translation by Walter Miller; Cambridge MA
1913) text, 150, 152, translation, 151, 153.
22 Ibid., text, 330, 332, translation, 331, 333, the quote is from 332–3.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 263

Cicero’s de Officiis indicates that this problem was recognized already at the
cusp of empire. And yet almost six centuries later, even under the pressure of
“law in the crisis of empire,” as Tony Honoré has put it, this problem of buyer
protection had not been addressed.23
This is true even though it seems that this attitude did change somewhat,
at least in this matter and to a minor extent, at least for a time. Watson notes
that “a change of some kind occurred at some time after the classical period
[that is, after 250 CE—Y.E.] to introduce in some cases rescission of a con-
tract for gross inequality of exchange—laesio enormis . . . Both texts appear in
Justinian’s Code, both are rescripts attributed to the emperor Diocletian, and
both were ostensibly issued on account of a sale.” Watson then presents the
two texts (C.4.44.2 and C.4.44.8), both relating to the sale of a farm, where an
enormous loss is constituted by a price that was half of the true worth. At this
point it should be noted that for the rabbis, ona’ah does not apply to land sales.
At any rate, this development in Roman law was only temporary, as Watson
continues:

The two rescripts, of A.D. 285 and 293 respectively, are virtually the only
evidence for laesio enormis in the Roman legal sources and they seem to
be contradicted by rescripts of only a few years later. Since these two texts
are so close together in date, and since they say the same thing about
the law, and since there is no other legal evidence, we cannot plot the
historical development of laesio enormis. At most we may say, in view of
what seem to be the contrary texts in the Digest, that the doctrine does
not appear to have been known to the classical jurists. To complicate the
issue, from the time of Christianus Thomasius (1655–1728) the claim has
frequently been made that the rescripts are interpolated and that the
doctrine is due to Justinian. In this instance the issue of interpolation
is of relevance to us. Interpolations are changes made subsequently by
another person to juristic texts or imperial rescripts. They may affect the
substance or only the form. With these rescripts we can produce argu-
ments both in favor of and against substantive interpolation, but none is
conclusive.24

Without sufficient data, Watson is unable to determine the motives for this
temporary change of policy, whether “moral and philosophical” or “social

23 I refer of course to T. Honoré, Law in the Crisis of Empire 379–455 AD: The Theodosian
Dynasty and its Quaestors with a Palingenesia of Laws of its Dynasty (Oxford 1998).
24 Watson, The Spirit of Roman Law, 175–7.
264 elman

and economic.” For our purposes, however, the important facts are that the
attitude of the classical jurists was strictly laissez faire, no matter the results,
and that even the advent of Christianity seems to have made little difference,
while in the East, both the rabbis and the Sasanian jurists—and the ancient
Babylonians—adopted a more activist moral economic policy.

2 Buyer Protection

Our next venture into the ethics of commercial law involves buyer protection,
that is, the buyer’s assurance that the seller will safely deliver his sale to him, so
that, unlike the situation in Roman law, as we shall see, possession and assump-
tion of risk came together. Undoubtedly the most ancient and most widespread
form of transference of ownership was the symbolic handing over of the object
sold—in Latin traditio, “handing over,” or Hebrew meshikhah, “pulling.” In her
edition of the Sasanian Lawbook, Anahit Perikhanian takes the Middle Persian
dāt, or dād, as equivalent to traditio (in MHD 17.1; 30.10, 12; 31.6, 8; 90.4; 95.1,
MHDA 9.8; 27.13; 28.1).25 However, these instances are interpreted by Maria
Macuch, in her much more comprehensive edition, as referring to “transfers”
of various sorts, including through testament or donatio mortis causa (see
MHD 31.6–8 and see Macuch’s commentary).26 Thus, in all these cases, dād can
be parsed more generally—and safely—as simply referring to transfer in gen-
eral and not necessarily transfer by traditio. That of course does not mean that
traditio did not exist in Sasanian law; chances are that it did, simply because
of its convenience and universality. Though of course there is a rough equiva-
lence between the Roman traditio, the rabbinic meshikhah and our presumed
Sasanian dād, we cannot assume that they agreed in detail; for example, we
know that traditio could be used to acquire land, which meshikhah could not,
or traditio longa manu, “pointing out the thing to the transferee, and authoriz-
ing him to take it, in such conditions that it was in his immediate power to
do so”27 was not a form of meshikhah.

25 A. Perikhanian, Book of Thousand Judgments, 351 s.v. dāt. However, Macuch (see next
note) does not recognize dād as a noun, but rather as part of a verbal phrase denoting a
transfer however it takes place, and Perikhanian herself is not consistent on this matter;
see for example her rendering of MHDA 9.7–9, Perakhanian, ibid., 260–1.
26 The discussion of transfers occurs in Macuch, Das sasanidische Rechtsbuch, Part II, 126,
n. 3.
27 See W. W. Buckland, A Text-Book of Roman Law from Augustus to Justinian (Cambridge
1932) 227.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 265

The Bavli explains the basic motive for the requirement of meshikhah as
a means of protecting the buyer by ensuring that the object bought passes
safely into his possession; if ownership were transferred without meshikhah,
the seller would not necessarily exert himself to guarantee the object’s safety
in cases of danger, classically conceived as “wheat stored in an upper storey”
and the danger being fire. In order to accomplish this goal, the rabbis required
meshikhah for transfer of possession; “mere” payment of the sale-price was not
sufficient. Here is the talmudic passage from bBM 47b:

:‫ אמר רבי יוחנן‬.‫כיצד? משך הימנו פירות ולא נתן לו מעות—אינו יכול לחזור בו וכו׳‬
‫ נשרפו‬:‫ שמא יאמר לו‬,‫ משיכה קונה? גזירה‬:‫ ומפני מה אמרו‬.‫ מעות קונות‬,‫דבר תורה‬
‫ שמא תפול‬,‫ גזירה‬,‫ מאן דשדא דליקה בעי שלומי! אלא‬,‫ סוף סוף‬.‫חטיך בעלייה‬
‫ לא מסר‬,‫ ואי לא‬.‫ טרח ומציל‬,‫ מסר נפשיה‬,‫ אי מוקמת להו ברשותיה‬.‫דליקה באונס‬
.‫ טרח ומציל‬,‫נפשיה‬
. . . ‫ משיכה מפורשת מן התורה‬:‫ריש לקיש אמר‬

And here is bBM 47b–48a:

.‫ כל שהכסף בידו—ידו על העליונה‬:‫ רבי שמעון אומר‬:‫תנן‬

,‫ בשלמא מעות קונות‬:‫ אי אמרת‬.‫ לוקח לא מצי הדר ביה‬.‫מוכר הוא דמצי הדר ביה‬
‫ מעות אינן‬:‫ אלא אי אמרת‬.‫ לוקח לא מצי הדר ביה‬.‫משום הכי מוכר מצי הדר ביה‬
,‫ אליבא דרבי שמעון לא קאמינא‬:‫ לוקח נמי ליהדר ביה! אמר לך ריש לקיש‬,‫קונות‬
.‫כי קאמינא אליבא דרבנן‬

[Quoting the Mishnah:] For example, if [A] drew into [B’s] possession
without paying him the money, he cannot retract. Rabbi Yohanan said: By
biblical law, [the delivery of] money effects possession.

Why then was it said meshikhah effects possession?

Lest [the seller] say to [the buyer]: Your wheat was burnt in the loft.
But after all, whoever causes the fire must pay compensation!
Rather [for fear] lest a fire accidentally break out [and no one would be
responsible].
Now, if the ownership is [still] vested in [the seller], he will wholeheart-
edly take pains to save it; if not, he will not do so.
Resh Laqish said: Meshikah is explicitly provided by biblical law. [The
remainder of this analysis deals with the omnisignificant niceties of the
two positions, and is irrelevant to our concerns] . . .
266 elman

We learnt: Rabbi Shimeon said: He who has the money in his hand has
the advantage. [This means that] only the seller can retract, but not the
buyer.
Now, should you say that [according to biblical law the delivery of]
money effects possession, it is well; therefore the seller can retract, but
not the buyer. But if you say that [the delivery of] money does not effect
a title [even according to biblical law], then the purchaser too should be
able to retract!
Resh Laqish can answer you: I [certainly] did not state [my view] on the
basis of Rabbi Shimeon’s opinion, but according to the Rabbis’ [opinion].
Now, as for Resh Laqish, it is well, for precisely therein do Rabbi Shimeon
and the Rabbis differ. But according to Rabbi Yohanan, wherein do Rabbi
Shimeon and the Rabbis differ?
In respect to Rav Hisda’s dictum, viz.: Just as [the Rabbis] enacted the law
of meshikah in respect of the seller, so did they institute it in respect to
the buyer.
Thus, Rabbi Shimeon rejects this dictum of Rav Hisda, while the Rabbis
agree therewith.

As I pointed out in another context,28 in two places in the Bavli there is an

animated discussion as to whether this enactment applies to transactions
involving non-Jews, to whom meshikhah is held not to apply, at least by some
authorities. Since I have discussed the sources in a previous study, I will not
repeat the analysis here. Instead, I want to focus on another question: Why
would the Babylonian rabbis strip Jews dealing with non-Jews of the protec-
tion of meshikhah? If meshikhah does not apply to such transactions, we are
left with the principle that the transaction is concluded when money changes
hands, even thought the merchandise is not delivered, and the buyer has no
guarantee! Could it be that they hoped thereby to discourage such dealings?
That is hardly likely for a minority with close economic ties to the majority!
The answer I proposed is that this ruling would have had no practical effect,
since Sasanian law itself provided protection for the buyer and its commercial
practice made the whole discussion moot. The Sasanian Lawbook provides
us with the rule of drust dārišnīh, which provided precisely the guarantee that

28 Y. Elman, “Shopping in Ctesiphon: A Lesson in Rabbinic and Sasanian Commercial Law,”

presented at a conference on Talmudic Archaeology, University College, London, June
15–17, 2009, and to be published with the proceedings.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 267

the rabbis attributed to meshikhah. Here too, for convenience, I have translated
Macuch’s German translation into English.

MHD 6.13–16
ka mard 2 āganēn xwāstag-ē(w) ō mard-ē(w) frōšēnd ud nibišt kunēnd kū
drust dārēm Wahrām guft kū drust dārišnīh az har(w) kē kāmēd xwāst
pādixšāy29

If two men together sell one thing to a(nother) man, and prepare one
deed: “We guarantee (the goods).” Wahrām said: Then is (the buyer) per-
mitted to demand the guarantee from whichever (of the two sellers) he
wishes.

As Macuch notes:

That is, the sellers of some goods make themselves responsible to pro-
vide a guarantee, drust dārēm, “we provide a guarantee.” The buyer can
demand fulfillment of the contract from each of the representatives, in
the event that the buyer is in dispute with a third party.30

Macuch goes into greater detail in her comments on a later text, MHD 64.9–15,
regarding a somewhat more complicated case.

MHD 64:9–15
Rād Ohrmazd guft kū ka Mihrēn gōwēd kū ēn xwāstag ōy kē Farroxw xwēš
būd rāy gōwēd xwēš ē bawēd ud pas az Farroxw xwāstag-ē(w) bē ō Dād
Farroxw frōšēd ud paymān kunēd kū hamē ka drust nē dārom anšahrīg kē
Mihrēn ān gōwišn abar guft tō xwēš ē bawēd ud ka-z gōwēd kū hamē ka ān
xwāstag drust nē dārom ēg ān anšahrīg tō xwēš būd rāy guft ka xwāstag
drust nē dārēd ān anšahrīg pad gōwišn bē šawēd. Wahrām <guft> kū wēš
uskārdan abāyēd 31

Rād Ohrmazd said: When Mihrēn said: “This thing shall belong to whom-
ever Farroxw designates,” and then Farroxw sells something to Dād
Farroxw and makes this agreement: “Whenever I do not keep (this thing)
undamaged, this slave, regarding which Mihrēn made this declaration,

29 Macuch, Das sasanidische Rechtsbuch, Part I, text, p. 88, translation, p. 93, see n. 10 on p. 99.
30 Macuch, Das sasanidische Rechtsbuch, Part I, 99, n. 10.
31 Ibid., text, p. 434, translation, p. 437, see n. 6 on p. 442.
268 elman

will belong to you.” And also when (Farroxw) said: “Whenever I do not
keep that thing undamaged, will that slave be designated to you as pos-
session”: If (Farroxw) does not keep that thing undamaged, then that
slave goes (to Dād Farroxw) on the basis of that declaration. Wahrām
said: Here other considerations should be made.

As Macuch comments:

The expression drust dāštan, “to hold safe, undamaged, to attend to its
undamaged state” is predominantly employed in connection with the
sale: the seller undertakes the responsibility, according to the terms of
the contract, upon the settlement of the sale price, that the goods remain
in his possession for specific, but not more closely defined reasons, to
hold them undamaged until the final transfer to the buyer. So the juris-
consult Rād Ōhrmazd in this sentence: The one giving the gift, Mihrēn,
bestows on Farroxw the right to take a slave (Mihrēn) from whomever
(Farroxw) chooses as a possession with the formula: ēn xwāstag ōy kē
Farroxw xwēš būd rāy gōwēd xwēš ē bawēd, “This thing shall belong to
whomever Farroxw designates as his possession.” Then Farroxw, whom
Mihrēn had designated in his declaration as having the right of disposal,
sold something to Dād Farroxw, which does not have the guarantee of
perfection for the buyer, but remains in the first instance in the posses-
sion of the seller (we do not learn the reason; it is conceivable that this is
because the object must be selected from a storehouse, or has yet to be
set out by the seller, or has yet to be finished). The seller, Farroxw, obli-
gates himself to the buyer, Dād Farroxw, to keep the object undamaged
until it is given over to him; otherwise, he would provide a substitute, the
slave, who Mihrēn has placed at his disposal to designate to whomever
he wished. Two possible types of guarantees are formulated: in the first
declaration, the seller Farroxw specifies the origin of the slave (anšahrīg
kē Mihrēn ān gōwišn abar guft tō xwēš), in the second the original owner,
Mihrēn, is not named (ān anšahrīg tō xwēš). Both declarations are valid
according to Rād Ōhrmazd.

What is the question at issue here? Macuch plausibly suggests that the object
sold is not immediately at hand, but perhaps in a storehouse. In other words,
it was not available for traditio! Now, obviously, if the object is handed over by
traditio (or, let us assume that the Middle Persian term for it was, as Perikhanian
suggested, dād), there is no need for the protection of drust dārišnīh,
since then the object is no longer in the seller’s possession. Note that the
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 269

transaction comes along with a penalty clause. The question that Rād Ohrmazd
must decide, as Macuch implies in her commentary, is how carefully defined
the penalty-clause must be in order to be effective. Does this origin of the slave
have to be defined as having belonged to Mīhren? Rād Ohrmazd’s answer is
in the negative. Thus, the question that Rād Ohrmazd had to decide relates
more to the penalty clause than to the guarantee of drust dārišnīh itself. In the
absence of the penalty clause, it would seem that drust dārišnih was univer-
sally applied to cases of sale. As the Bavli’s redactors note in passing in bGit 10b,
and as Rād Ohrmazd’s case suggests as well, the laws of sale and those of gifts
are not the same, as indeed we might have expected.32 Since the gift-giver has
shown his generosity by giving the gift, the law does not want to burden him
with such a guarantee.
We turn to the Roman law again. The influential if mysterious Roman jurist,
Gaius, refers to traditio in his Institutions:

II. 18: Magna autem differentia est inter mancipi res et nec mancipi. 19. Nam
res nec mancipi ipsa traditione pleno iure alterius fiunt, si modo corporales
sunt et ob id recipiunt traditionem. 20. Itaque si tibi uestem uel aurum uel
argentum tradidero siue ex uenditionis causa siue donationis siue quauis
alia ex causa, statim tua fit ea res, si modo ego eius dominus sim.

II. 18. Now, there is a great difference between things capable of manci-
pation [a sort of fictitious sale] and those which are not. 19. For the lat-
ter become the full property of someone else by the very act of delivery,
provided that they are corporeal and so capable of delivery. 20. And so, if
I deliver to you clothing or gold or silver, whether on the basis of a sale or
a gift or on any other basis, it immediately becomes yours, provided that
I am owner of it.33

Or, as Alan Watson puts it, “Things were further classified as corporeal or incor-
poreal, a distinction whose sole importance was that incorporeal things could
not be transferred by the legal method that required actual physical delivery
(traditio).”34 For our purposes here, however, the essential point is whether the

32 Why a gift is not automatically invested with the guarantee is indeed a question, since we
must assume, along with the bBB 53a, that a gift is given “with a generous eye” (‫)בעין יפה‬.
33 So W. M. Gordon and O. F. Robinson (trans.), The Institutes of Gaius, with an Introduction
(Ithaca NY 1988) 56, 58 (Latin text), 131, 133 (translation).
34 Watson, Spirit of Roman Law, 16.
270 elman

buyer was protected from loss or damage, in the absence of traditio, when s/he
did not have possession of the article in question. We will return to this point
below, after discussing the rabbinic and Sasanian rules governing this case.
As noted above, according to rabbinic law, meshikhah was instituted—
either biblically or rabbinically—in order to place responsibility for safe deliv-
ery of the purchase on the shoulders of the seller, who, were ownership to
have been transferred by any means short of actually handing over the mer-
chandise, would not exert himself to keep it safe, since it was no longer his.
This concern was common to rabbinic and Sasanian law, but not to Roman
law, where, as Buckland stated, “if after the contract, the thing was damaged
by accident, or wholly or partially ceased to exist, without his fault, he was
bound only to deliver what was left and the buyer must still pay the price; the
risk was on him”35 (see D.18.6.9; 21.2.11). Roman law was not totally oblivious to
such concerns, but was much more circumspect in providing the buyer with
such protection. Here, for example, is Ulpian’s discussion of this matter in the
Digest:

D.18.6.1 ULPIANUS libro uicesimo octauo ad Sabinum. si uinum uenditum

acuerit uel quid aliud uitii sustinuerit, emptoris erit damnum, quemadmo-
dum si uinum esset effusum uel uasis contusis uel qua alia ex causa. sed si
uenditor se periculo subiecit, in id tempus periculum sustinebit, quoad se
subiecit: quod si non designauit tempus, eatenus periculum sustinere debet,
quoad degustetur uinum, uidelicit quasi tunc plenissime ueneat, cum fuerit
degustatum. aut non igitur conuienit, quoad periculum uini sustineat, et
eatenus sustinebit, aut non conuenit et usque ad degustationem sustinebit.

D.18.6.1: Ulpian, Sabinas, book 28: If wine which has been sold goes sour
or goes off in some other way, the loss is the purchaser’s, as it would be if
the wine were spilled, whether through the casks being staved or for some
other reason. But if the vendor has undertaken the risk, he will sustain it
for so long as he has undertaken it; if he has specified no period, then he
bears the risk until the wine is tasted, wine, of course, being regarded as
absolutely sold only when it has been tasted. Consequently, it will have
been agreed how long the vendor bears the risk in the wine or, in the
absence of agreement, he will bear it until the wine is tasted.

35 Buckland, Text-Book of Roman Law, 484–5.

ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 271

Whatever protection the buyer has is due to the implied stipulation that contin-
ues until the wine is tasted. This is clear from D.18.6.8:

D.18.6.8 PAULUS libro trigesimo tertio ad edictum: Necessario sciendum

est, quando perfecta sit emptio: tunc enim sciemus, cuius periculum sit:
nam perfecta emptione periculum ad emptorem respiciet. et si ed quod
uenierit appareat quid quale quantum sit, sit et pretium, et pure uenit,
perfecta est emptio: quod si sub condicione res uenierit, si quidem defecerit
condicio, nulla est emptio, sicuti nec stipulatio: quod si exstiterit, Proculus
et Octauenus emptoris esse periculum aiunt: idem Pomponius libro nono
probat.

D.18.6.8 Paul, Edict, book 33: It is essential to know when a sale is perfect
because we then know who bears the risk in the thing; for once the sale is
perfect, risk is on the purchaser. And if the thing sold be identified, what
it is, its nature, and quantity, the price be fixed, and the sale be subject to
no condition, the sale is perfect. But if it be conditional and the condition
not yet satisfied, there is no sale, any more than there would be a stipula-
tion. If, though, the condition has been realized, Proculus and Octavenus
say that the risk is on the purchaser and, in his ninth book, Pomponius
approves that view.

As Alan Watson puts it:

For a long time, the Roman contract of sale contained no inherent war-
ranty of quality, especially with regard to psychological issues. A buyer
who wished protection, as many did, against hidden defects, would have
to take an express promise by the formal verbal contract of stipulatio. The
point here is that for some jurists, a stipulatio in the form “I promise that
the slave is not a thief and is not a corpse robber” was ineffectual. If the
slave did not have these defects, the guarantee would produce no effect
and was thus unnecessary. If the slave did have these defects, the prom-
ise was of an impossibility, and a stipulatio for an impossibility was void.
Again we have reasoning remote from the real world [sic]. But this time
we have the unusual feature that the author of the text, Ulpian, brings
in the real world: the point is, he says, the interest of buyers. But what
he says next is most revealing. First, he says that stipulation would have
more validity if the guarantor had said “reparation will be made if.” So
he is making some concession to the force of the formalistic argument.
Next he brings in the edict of aediles that a seller in the marketplace must
272 elman

declare a slave’s faults and promise he had no others. Ulpian’s point is that
the second provision in the edict involves a stipulation in the form that
some jurists regard as valueless, and so he declares that that juristic opin-
ion cannot be right. We are back to a very formalistic argument. Lastly,
Ulpian overdoes his response: no sane person, he says, would accept that
the stipulation proposed by the aediles would be useless. He clearly does
not find it easy to rebut the logic of the jurists he is disagreeing with.36

Buckland sums up the situation as follows:

Duties of the vendor: Sale was a bonae fidei contract in which both par-
ties benefitted. The vendor must abstain from dolus [fraud—Y.E.], and
must take care of the thing until delivery, being liable for culpa levis, i.e.
he must show the care of a bonus paterfamilias. He was also bound to
“custodia” [see D.18.6.3 and 19.1.31; 47.2.14.pr—WWB]. He was not liable
for casus [accident—Y.E.], apart from agreement or mora [delay—Y.E.].
Thus if, after the contract, the thing was damaged by accident, or wholly
or partially ceased to exist, without his fault, he was bound only to deliver
what was left and the buyer must still pay the price; the risk was on him.37

Buckland notes that the fact that the buyer assumes the risk even before deliv-
ery is difficult to understand or justify, and suggests that it may be a “mere
traditional survival.” David Johnston makes a valiant attempt at explaining this
survival.

It has been regarded as strange that the buyer became liable at such an
early point in the transaction, since he would not become owner until
the goods were actually conveyed to him, for example by delivery. The
result is that, at a time at which he did not own the goods, he was at risk
if they were lost or destroyed in certain circumstances. The explanation
for this again seems to lie in the fact that sale was originally cash-sale,
and conclusion of contract, conveyance and transfer of risk all took place
at the same time. That would not account for the survival of the rule at
a time when the contract was concluded well before the conveyance of
the goods was made. But the risk which passed to the buyer was only
for events which were not preventable by the seller, and this restriction

36 Watson, The Spirit of Roman Law, 87. See also the discussion in Buckland, Text-Book of
Roman Law, 486.
37 Buckland, Text-Book of Roman Law, 286–7.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 273

was interpreted strictly; the seller was still liable if the goods were stolen,
unless the theft involved overwhelming force. So the sorts of risks which
the buyer assumed on conclusion of the contract were for destruction
by earthquake, flooding or fire (so long as the seller was not responsible
for it). The risk rule, odd though it may seem, was therefore kept within
strict limits.38

As we know, however, fire is far more frequent than earthquakes, and so the
rule was not really very effective. Indeed, fire is the reason the rabbis gave for
their assignment of risk to the seller even after payment had been made though
delivery had not. It is therefore hard to credit the absence of such a clause in
Roman law as a mere “survival” of an ancient custom when so much else had
changed. However, whatever its cause, this is an expression of the Roman lais-
sez faire policy as regards many things, not just commercial law, and may be
due in part to the nature of Roman law as private law rather than public, gener-
ated in large part by jurists rather than legislators or administrators—but then
this was true of the rabbis as well. As we see, in marked contrast to rabbinic
and Sasanian/Zoroastrian law, Roman law provided relatively few protections
to the buyer. This is particularly telling, since rabbinic law had no government
to enforce it
We observe, however, that Roman law, while leaving the buyer more or
less on his/her own in regard to wine, afforded some protection by means
of an implied stipulation. Note that the Roman remedy remained in the orbit
of private law, unlike the rabbinic and Sasanian jurists, who adopted a more
interventionist policy. As we shall see in the next section, there is at least one
opinion in the Bavli that assumes that something akin to the Roman policy was
contemplated.

3 Land Fraud

Our third case is one in which Roman law and rabbinic law approach a prob-
lem in similar ways, but the Sasanian jurists provide for it a peculiar solution of
their own, based on Zoroastrian mores. Ancient economies were in large part
subsistence economies, and agriculture was of prime importance. The impor-
tance of land can be gauged from the fact that the Rabbis excluded it from the
prohibition of ona’ah, something that Justinian did not, at least in the case of
laessio enormis, as we have seen, though in the end it made little difference.

38 D. Johnston, Roman Law in Context (Cambridge 1999) 82.

274 elman

The importance of land points up an interesting common denominator among

all three legal systems that stand at the center of our attention: all of them
had to deal with fraudulent land sales. Public order required that the original
owner be compensated in some way; however, what might seem straightfor-
ward to the modern mind, conditioned to legal action by modern states, was
not at all straightforward in late antiquity.
As we have done in the other cases, we will begin with the Bavli, proceed to
the Sasanian Law Book, and end with Roman law. As we shall see, each system
deals with the problem in its own way, shedding light on the scale of values and
ethics that undergirded each of them.
The locus classicus of the Bavli’s discussion is at bBM 14b:

,‫ יש לו מעות ויש לו שבח‬:‫ רב אמר‬,‫ המוכר שדה לחבירו ונמצאת שאינה שלו‬:‫איתמר‬
.‫ שבח—אין לו‬,‫ מעות—יש לו‬:‫ושמואל אמר‬
‫ פירש לו את השבח מהו? טעמא דשמואל—משום דלא פירש‬:‫בעו מיניה מרב הונא‬
‫ טעמיה דשמואל כיון דלית ליה קרקע—מחזי‬:‫ או דלמא‬.‫שבחא? והכא הא פירש לה‬
.‫ ורפיא בידיה‬,‫ אין ולאו‬:‫כריבית? אמר ליה‬
‫ אף על פי שפירש‬,‫ שבח—אין לו‬,‫ מעות—יש לו‬:‫ אמר שמואל‬,‫ אמר רב נחמן‬:‫איתמר‬
.‫ עומד ונוטל‬,‫ מאי טעמא? כיון דקרקע אין לו שכר מעותיו‬.‫לו את השבח‬

It was stated: If one sells a field to his neighbor and it turns out not to
be his own, Rav says: [The buyer] is entitled to [the return of the money
which he paid for the field] and to [compensation from the seller for the]
improvement [which he made in the field]. But Shmuel says: He is entitled
to the money [he paid] but not to [compensation for the] improvement.
Rav Huna was asked: If [the seller] expressly stated [that he would com-
pensate the buyer for the] improvement [if the field were taken away],
what is the law then? Is Shmuel’s reason [for withholding compensa-
tion] that [the seller] did not expressly state [that he would compensate
the buyer for the] improvement? [Then it would not apply to this case,
for] here [the seller] did state expressly [that he would compensate the
buyer]. Or is Shmuel’s reason that, in view of the fact that [the seller]
really had no land [to sell, the money received by the buyer as compensa-
tion for the improvement] would appear like usury?
Rav Huna answered: Yes and No, for he was hesitant.
It was taught: Rav Nahman said in the name of Shmuel: [The buyer] is
entitled to [have returned to him] the money [paid for the field], but
not to [compensation for] improvements [made by the buyer], even
if [the seller] stated expressly that [he would compensate the buyer
for the] improvement, the reason being that, in view of the fact that
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 275

[the seller] really had no land to sell, [the buyer] would be taking profit
for his money[, which is equivalent to usury].

The Bavli’s discussion continues for several folios, but the essential principle
of compensation for fraud is clear. As Jacob Neusner noted, the rabbis were
most concerned with maintaining a just status quo;39 the defrauded buyer is
compensated, the real owner retains his ownership of the land. Sasanian law
attempts to provide a similar solution but in a very Zoroastrian way.

MHDA 8.13–17
Anī gyāg-ēw (14) nibišt kū Farroxw daskard-ēw ī-š nē xwēš rāy kunēd kū-m
pas az 10 sāl (15) ō Mihrēn dād ud ōy kē ān daskard-ēw xwēš pas az ān ōy
daskrad rāy (16) kunēd kū-m pas az 10 sāl ō Farroxw dād ān xwastāg pad
dād ī Farroxw (17) ō Mihrēn rasēd.40

It is written in another place: (If) Farroxw promises regarding a field

which he does not own: “It should be given by me to Mihrēn after ten
years,” and the one, to whom the field belongs, afterwards stipulates
regarding the field: “It should be given by me to Farroxw after ten years,”
then this thing shall go to Mihrēn through the gifting of Farroxw.41

MHDA 8.17–9.1
Ud ka gōwēd kū-t dahom u-š nē xwēš bē xrīnišn (1) ud bē dahišn.42

And if someone said: “I give you (this thing),” and it is not his own, he
must buy it (9.1) and hand it over.43

And in a decision parallel to the talmudic one that we began with:

MHDA 9.5–7
Wahrām az Wahrāmšāt ud Rād Ohrmazd bē ōwōn guft kū (6) ka daskard-ēw
ī nē xwēš rāy kunēd kū ō tō dahom u-š pad wahāg xrīt-ēw ud (7) bē dād nē
tuwān arz ī ān daskard bē dahišn.44

39 J. Neusner, The Economics of the Mishnah (Chicago 1990) 141–2.

40 For the text, see Macuch, Das sasanidische Rechtsbuch, Part II, 29–30.
41 See Macuch, Das sasanidische Rechtsbuch, 121.
42 For the text, see Macuch, Das sasanidische Rechtsbuch, 30.
43 Macuch, Das sasanidische Rechtsbuch, Part II, 122.
44 See Macuch, Das sasanidische Rechtsbuch, 30 for the text.
276 elman

Wahrām said, following Wahrāmšāt and Rād Ohrmazd: If one agrees in

regard to a field which does not belong to him: “I give it to you,” and it is
not possible for him to buy it for a (certain) price and transfer it as a gift,
he should transfer its monetary worth as a gift.45

In consonance with the Zoroastrian triad of “good thought, good speech and
good action,” the would-be defrauder is hoist with his own petard: he is held to
his promise and must make good on it, even at the cost of buying the land he sold
fraudulently or his giving the buyer the equivalent!46
In contrast, Roman law is more concerned with the buyer’s good faith; the
original owner does not have the protection of the law:

D.41.3.38 GAIUS librto secundo rerum cottidianarum siue aureorum: Quam

rem ipse quidem non potest usucapere, quia intelligent alienum se possidere
et ob id mala fide possidet. sed si alii bona fide accipienti tradiderit, poterit
is usucapere, quia neque ui possessum neque furtiuum possidet: abolita
est enim quorundam ueterum sententia existimantium etiam fundi lociue
furtum fieri.

D.41.3.38: GAIUS, Common Matter or Golden Things, book 2: There is land

which a man cannot usucapt47 for himself, because he knows that he
possesses someone else’s thing and is thus in possession in bad faith. But
if he delivers it to someone else who accepts it in good faith, the latter
can usucapt it, because he possesses what has come into his possession
neither by force nor by stealth; for we have discarded the view of some
of the earlier jurists who took the line that there can also be theft of land
or sites.

Or, as Buckland states:

There was nothing to prevent the sale of a third person’s property. It

might be difficult to carry it out, but that was the vendor’s fault. It is plain

45 See Macuch, Das sasanidische Rechtsbuch, 122 for the German translation.
46 It is interesting to note that the Persians were known for keeping their promises, but at
least one rabbi thought that it was a result of arrogance; see bAZ 71a.
47 Usucapion is the right to, or ownership of, property on the basis of possession of it for a
prescribed period of time, equivalent to the rabbinic hazaqah. But it seems to have fallen
out of use after Constantine, though in the end it retained some of the characteristics of
the latter; see H. F. Jolowicz and B. Nichols, Historical Introduction to the Study of Roman
Law, 3rd edition (Cambridge 1972) 506.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 277

that such a sale might be in good faith, with full knowledge of the facts;
the vendor might intend to acquire from the owner, or induce him to
convey to the buyer [see D.18.1.28—WWB].48

Buckland’s summation is a restatement of D.18.1.28:

D.18.1.28 ULPIANUS libro quadragensimo primo ad Sabinum: Rem alienam

distrahere quem posse nulla dubitatio est: nam emptio est et uenditio: sed
res emptori auferri potest.

D.18.1.28 Ulpian, Sabinus, book 41: There is no doubt that one can sell a
third person’s property; there is a valid sale and purchase, even though
the thing may be taken away from the purchaser.

The essential criterion here is good faith, as Gaius already noted above. As
David Johnston explains, “What this (i.e. good faith) meant was that the par-
ties’ dealings with one another were assessed in any eventual litigation on
the basis of what good faith demanded; and so, without any need for adding
further express promises or undertakings, the law on sale kept pace with the
customs of trade and commerce: it was open to a judge to find that the failure
of a party to act in accordance with ordinary commercial standards was not
consonant with good faith and therefore amounted to a breach of contract.
The standard of good faith therefore gave the contract extraordinary vitality
and flexibility.”49
But, as we have seen, this reliance on good faith had its weak points, as when
one bought in good faith from a fraudster, because ultimately land cannot be
considered as stolen without imperiling the whole system of usucaption. Thus,
as Buckland sums up the situation,

It might happen that, from some cause, the holder lost actual possession.
Some other person might, during his absence, enter on the property, and
refuse to give up possession. How was the bonitary owner [i.e. the owner
in good faith—YE] to recover it? . . . The Praetor provided an action,
called the actio Publiciana, based on the fact that the bonitary owner
would in the ordinary way become dominus [i.e. owner—Y.E.] in the
course of time by what was called usucapio . . . The actio Publiciana was
also available, perhaps primarily available, to an ordinary b.f. possessor,
one who has received the property in good faith, but who, in fact, was

48 Buckland, Text-Book of Roman Law, 484.

49 Johnston, Roman Law in Context, 81.
278 elman

not so-capable, e.g. a non-owner. His protection must be less complete:

it was not the object of the Praetor to destroy the law of usucapio, or
bar an owner from recovering his property. Such a possessor was not
protected against the true owner, but only against third parties.50

So far, so good. But Buckland then admits:

The b.f. possessor had the actio Publiciana against all but the dominus; in
a sense even against him. The facts alleged in his claim were true in that
case also, but the owner was allowed to plead in reply the exceptio iusti
dominii, an allegation that he was the true owner. [The reason for this is
that—Y.E.] in strictness a b.f. possessor cannot know that he is one; he
thinks he is really entitled: as used here, however, the expression means
one whose possession began in good faith, one who is in via usucapiendi
[on the way to ownership by usucapio—Y.E.]. It was immaterial had he
discovered his mistake.51

In other words, the one who bought the land from the fraudster could defend
his claim against the original owner—because he had acted in good faith.
One might well assume that this was truly a conqueror’s law! But it need
not be seen as such, since ownership—Latin domininium or Middle Persian
xwēšīh—must be distinguished from mere possessio or dārišn; though “pos-
session is nine-tenths of the law,”—it is not the whole law. The governments
of late antiquity—even the high and mighty empires—could not maintain
central depositories of land-ownership; that was why they all depended on
usucapio or hazaqah, the occupation of the land in question without protest
from a competing owner over a period of years. The rabbis required three years
(mBB 3:1; bBB 28a–b52), the Romans two, the Persians 40 (bBB 55a53). Still, each
legal system worked out its own response to land fraud. The rabbis required

50 Buckland, Text-Book of Roman Law, 192–3. For more on the actio Puliciana, see Jolowicz
and Nichols, Historical Introduction, 263–7. They make it very clear that it was based on
a legal fiction. See also: L. Moskovitz, “Legal Fictions in Roman Law and Rabbinic Law:
Some Comparative Observations,” in: C. Hezser (ed.), Rabbinic Law in its Roman and Near
Eastern Context (Tübingen 2003) 105–32.
51 Buckland, Text-Book of Roman Law, 193.
52 See the discussion on bBB 29a; three years is the limit at which a farmer is expected to
guard his deeds.
53 So the report of the exilarch, Uqban bar Nehemiah, in the name of Shmuel, as reported by
Rava. To my knowledge, we have no Persian sources for this.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 279

compensation, the Persians required that the seller fulfill his promise and pro-
vide land or its equivalent to the buyer—and the Romans relied on the good
faith and honor of the secondary seller, even in the face of fraud on the part of
the primary vendor.
D.6.2.17 makes this abundantly clear:

D.6.2.17 NERATIUS libro tertio membranarum. Publiciana actio non ideo

comparata est, ut res domino auferatur: eiusque rei argumentum est primo
aequitas, deinde exceptio ‘si ea res possessoris non sit’: sed ut is, qui bona
fide emit possessionemque eius ex ea causa nactus est, potius rem habeat.

D.6.2.17 NERATIUS, Parchments, book 3: The Publician action was not

introduced to take the thing away from the owner [that is, the one who
bought it from the fraudster—Y.E.]. The justification for the action was
first equity and then the defense “if the thing does not belong to the pos-
sessor.” It aimed to insure that the party who bought something in good
faith, and on that ground, acquired possession, should keep it.

The talmudic discussion of fraudulent land sale also includes cases of stolen
land, or land stolen and then “bought” by the robber, etc., and the discussion
is not all that different from the one between Roman jurists in, say, D.41.2.3,
where the roles of intent and physical possession in determining possession
are debated. Land fraud shades off into land robbery, with all sorts of combina-
tions and permutations of forced sale and the like. A discussion of such mat-
ters would take us too far afield. But the fact remains that the rabbis insured
equity by forcing the fraudster to recompense the buyer for his outlay, while the
land presumably returned to its original owner. The Sasanian jurists insured
the same result by forcing the fraudster either to buy land to the same value,
or, if that was not possible, to compensate the buyer—and the land remained
that of the original owner. Only the Romans, in the end, preferred process
over truth.
This was not a slip of the pen. As Alan Watson has noted,

. . . I have stressed the jurists’ central role in shaping the law, their distance
from the courts, their artificial mode of reasoning (for private law), and
their conceptualization, even at the expense of sensible development.54

54 Watson, Spirit of Roman Law, 98.

280 elman

This element in the formation of Roman law cannot be ignored, and, indeed,
it may shed light on a contentious issue in talmudic studies, the idea that the
rabbis were ivory-tower theoreticians divorced from the day-to-day concerns
of their communities.55 Perhaps a more balanced statement of the real situ-
ation is that of Jeffrey L. Rubenstein, that “throughout Jewish history rabbis
exhibited both elitist and populist tendencies.” He states:

On the one hand, they saw themselves as the religious leaders of the
entire Jewish people and perceived their role as, at least in part, to edu-
cate, inspire, and guide their fellow Jews. On the other, they naturally felt
a sense of distance from their less learned and less pious brethren, and
intermittently resentment and frustration. In antiquity, before rabbinic
piety became the normative form of Judaism it became in later times,
these poles were exaggerated.56

This may explain some of the anomalies we have noted, and mitigate somewhat
the paradox that the jurists of the most powerful empire of the time consid-
ered justice in commercial relations to be the responsibility of the individual,
while the Sasanian Empire, hardly its equal in military might and administra-
tive ability, and the Jewish rabbis, the equal of neither in worldly terms, sought
redress for the individual in terms reminiscent of Hammurabi in the prologue
and epilogue of his Code.
However, having reviewed these contrasting policies, or lack of them, it
is difficult not to see the Roman reluctance to intervene in these matters an
aristocrat’s disdain for “business,” as we noted in the case of Cicero. Watson
has called attention to the aristocratic dominance of legal thought, and of its
distance from reality, certainly the reality of commerce. This was fueled by
the ideal of autarky, where an aristocratic paterfamilias lived on his farm, self-
sufficient in his own domain. We should add to this the Jewish and Zoroastrian
concept of a God who stands for and mandates justice. The interesting ques-
tion is why the dominance of Christianity did not affect Roman social and
economic policy; it may be that the emphasis on the next world and the idea
of this world as unredeemed may have had something to do with this. Still,
Judaism and Zoroastrianism shared these theological ideas with Christianity.
Could the energy expended on theological controversies and heresy-hunting
have diverted the Romans from attempting to apply Christian ideals to
the real world?

55 See most trenchantly R. Kalmin, The Sage in Jewish Society of Late Antiquity (London 1999).
56 J. L. Rubenstein, The Culture of the Babylonian Talmud (Baltimore 2003) 123.
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 281

I would like to add another factor that may have impelled such a denial of
an unpleasant reality on the part of legislators: their view of the Roman people
as honorable. Note that in cases of land fraud the innocent buyer is protected
because he acted in good faith. Unfortunately, by acknowledging his good
faith, the lawmaker does not provide easy restitution for him, and the fraudster
escapes. We are reminded of the story Cicero tells of Gaius Canius, “a Roman
knight, a man of considerable wit and literary culture,” who was cheated in
buying a vacation estate in Syracuse by a banker of that city, one Pythias, who
suggested that the estate’s location assured a sumptuous supply of fresh fish—
something that was untrue. Cicero comments:

Stomachari Canius; sed quid faceret? nondum enim C. Aquilius, college et

familiaris meus, protulerat de dolo malo forumulas; in quibus ipsis, cum
ex eo quaereretur, quid esset dolus malus, respondebat” cum esset aliud
simulatum, aliud actum. Hoc quidem sane luculente u tab homine perito
definiendi. Ergo et Pythius et omnes aliud agentes, aliud simulantes perfidi,
improbi, malitiosi. Nullum igitur eorum factum potest utile esse, cum sit tot
vitiis inquinatum.

Canius was furious [when he discovered the fraud]; but what could he
do? For not yet had my colleague and friend, Gaius Aquilius, introduced
the established forms to apply to criminal fraud. When asked what he
meant by “criminal fraud,” as specified in these forms, he would reply:
“Pretending one thing and practicing another”—a very felicitous defini-
tion, as one might expect from an expert in making them. Pythius, there-
fore, and all others who do one thing while they pretend another are
faithless, dishonest, and unprincipled scoundrels. No act of theirs can be
expedient, when what they do is tainted with so many vices.57

But despite six centuries more of experience with fraud and misrepresenta-
tion, and with the lack of buyer protection, no legislative remedy was applied,
despite the rise of Christianity, although the state of Christian business ethics
in the sixth century is itself a subject for study. Though Romanists have been
surprised that the assumption of risk on the part of the buyer preceded taking
possession of the property in question, it is clear that this blind spot was not
restricted to that issue alone.

57 Cicero, text, 330, 332, translation, 331, 333.

282 elman

Could this then be the difference between Republic and Empire? After all,
Rome underwent many changes between Cicero’s death in 43 BCE and the
second edition of Justinian’s Code in 534 CE. Yet a Roman’s view of empire
was not that of a conqueror (or not only that of a conqueror), but one who
brought civilization and peace to the world; in 212 citizenship was extended in
a universal manner. The failure of Roman initiative in the area under discus-
sion here remains a puzzle, but not, I think, an unsolvable one. Since other
problems were addressed by Roman jurists, a comparative study of which
problems were dealt with and which were not will perhaps help us understand
why Roman commercial law seems to diverge from a common pattern of other
ancient legal systems. But such an investigation lies beyond the scope of the
present study.
In my discussion of land fraud I noted the very Zoroastrian solution offered
by Sasanian law: in accordance with the basic Zoroastrian ethical principle
of “good thought, good speech and good deed,” the fraudster must make
good on his promise and acquire the land he had proposed to sell, or its
equivalence—no matter the expense—and offer it to the defrauded buyer.
Since Zoroastrianism and Judaism share a belief in a benevolent creator who
will in the end renew creation and rid it of evil, that is, both are “salvation
religions”58 as opposed to Roman paganism, it might be thought that the answer
to our conundrum lies in the theological sphere. Indeed, one could even argue
that Christianity had not yet developed to the point of taking such charge of
the world it came to dominate as to impose its ethical system on Roman cul-
ture. However, we must still contend with the fact that the ancient Babylonian
pagan monarchs felt the responsibility to ensure that prices remained “fair,” as
we found in the Laws of Eshnuna, and that, moreover, quite like the biblical
sabbatical year, kings of the Old Babylonian era issued edicts remitting debts
on the occasion of their second regnal years (the mīšarim-acts). The prologue
and epilogue of the Laws of Hammurabi enunciate that king’s responsibility
to insure that the powerful do not trample the weak. As the prologue has it,
“the gods Anu and Enlil, for the enhancement of the well-being of the people,
named me . . . to make justice prevail in the land, and to abolish the wicked

58 J. Scurlock and F. Al-Rawi (“A Weakness for Hellenism,” in: Ann K. Guinan et al. [eds.], If
a Man Builds a Joyful House: Assyriological Studies in Honor of Erle Verdun Leichty [Leiden
2006] 357–81) stress “the interchange of cultural ideas across the Hellenized world and
the concomitant emergence of three of the four great salvation religions” (p. 381).
ROMAN JURISCONSULTS, RABBIS & SASANIAN DASTWARS ON RISK 283

and the evil, to prevent the strong from oppressing the weak . . .”59 It appears
that one need not be an adherent of a salvation religion to express such senti-
ments. Nor is the Roman drive for conquest the answer, since the Sasanians
had that drive as well. In the end, then, we discover that a complex of ethical
and cultural attitudes and calculations influenced the decision to intervene or
to refrain from intervening in rectifying the all-too-human propensity to take
advantage of one another. Ironically, the sentiment expressed by Rabbi Hanina
Segan ha-Kohanim, to “pray for the welfare of the [Roman] government, for
were it not for it, men would swallow each other alive” (mAv 3:2), did not work,
at least in these areas. The perplexity expressed by students of Roman law as to
why this was so, has yet to find its solution, but I hope to have made progress in
that direction by adding several comparative aspects to its description.

59 Roth, Law Collections, 76–7. See J. J. Finkelstein, “Ammisaduqa’s Edict and the Babylonian
Law Codes,” Journal of Cuneiform Studies 15 (1961) 91–104. This is despite Finkelstein’s
conclusion that the law collections are more royal apologia with more “moral than legal
force” (p. 103), for the “original mīšarum act” was considered “as real as the building of a
temple, a canal, or a military success, and so deserving of commemoration in the date-
formulae” (ibid.). See more recently K. R. Veenhof, “The Relation between Royal Decrees
and ‘Law Codes’ of the Old Babylonian Period,” Jaaresbericht “Ex Orienta Lux” 35–6 (1997–
2000) 49–83.
From Palestine to Babylonia and Back:
The Place of the Bavli and the Tanhuma on the
Rabbinic Cultural Continuum
Ronit Nikolsky

One can almost say that as the number of scholars, so the numbers of con-
structed images of rabbinic culture; on the basis of a relatively small number
of literary sources (talmudim and midrashim), with a limited amount of his-
torical information (names of rabbis, place names, elements belonging to the
beit-midrash, choice of proof-texts, halakhot etc.), each scholar constructs her
or his own picture of rabbinic society, its social classes, its economy, the study
institutions and so forth.
I am no exception. Shifting my focus away from the external reality, I address
rabbinic culture first and foremost as a semiosphere, a constructed reality, or
world of meaning. Thus, without claiming to know the socio-historical circum-
stances in great detail, I can still study the culture by studying the exogram
(that is, the literature which is extant).1
The model that I use to envisage rabbinic culture—polysystem2—explains
that although the elements of the cultural cannon—cultural knowledge and
customs—are common to all parts of society, each group adapts it to fit its
own narrative, resulting in the canon being active in the semiosphere of this
group. Thus, although any particular group within the culture creates its own
exogram to fit its point of view, this particular exogram is not oblivious to
other groups, and traces of the exogram of one group can be found in the lit-
erature of another. Except for the social or geographical differences between
groups within one culture, the polysystem model also talks about how cultures
develop and change through time. This image of a culture as a dynamic multi-
faceted developing phenomenon can be applied to rabbinic society.

1 The term “exogram,” coined by Merlin Donald (Origins of the Modern Mind: Three Stages in
the Evolution of Culture and Cognition [Cambridge MA and London 1993] 308–32), refers to
the materialization of the semiosphere with the intent of recording it for others to decipher.
2 Developed by Itamar Even-Zohar in his “Polysystem Theory (Revised),” in: I. Even-Zohar
Papers in Culture Research (2005). As far as I know the book is only available in an electronic
version under http://www.tau.ac.il/~itamarez/works/books/ez-pss1990-toc.pdf; for the arti-
cle see: http://www. tau.ac.il/~itamarez/works/papers/papers/ps-revised.pdf.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�11

FROM PALESTINE TO BABYLONIA AND BACK 285

The advantage of this model is that it allows us to pay the same level of
attention to two types of rabbinic material. On the one hand, we have mate-
rial that was purposely and meticulously documented and disseminated by
the rabbis traveling from the Land of Israel to Babylonia in the first centuries
of our era (the nehutei). On the other hand, we have the ideas and parts of
the narratives that were diffused en passant, originally not intended to be
recorded, but which somehow found their way into the extent rabbinic texts,
as vague statements and hints. Such hints are found in the examples discussed
in the present study.
The differences between the Jewish culture of the Land of Israel and that
of Babylonian can be seen as a part of an on-going cultural dynamic, develop-
ing in two trajectories: the Palestinian rabbinic culture claims to be continuity
from the Bible to the tannaitic corpus, from there to the amoraic Palestinian
one, and then to the Tanhuma, which belongs to a later stratum of rabbinic
literature in the Land of Israel. A geographically parallel route is the one lead-
ing from the Land of Israel to Babylonia, where the Bavli reuses tannaitic and
amoraic material to assert its own world view.
In the present study I intend to show how the later stratum of the midrash
from the Land of Israel, the Tanhuma, incorporates material from the Palestin-
ian amoraim as well as from Babylonia. This use of Babylonian materials dem-
onstrates that the influence of the Bavli can already be detected in the latest
rabbinic Palestinian culture.
I will address two narratives. In each case I will start with an analysis of a
Palestinian midrash. Then I will analyze the reworking of the narrative in the
Bavli, and in the end I will show that traces of a particular Babylonian rework-
ing of each narrative is found in the late Palestinian midrash Tanhuma Buber.
Since much of the aggadic material in the Bavli is a reworking of Palestinian
material, in order to show influence of the Bavli on the Tanhuma, the latter has
to exhibit acquaintance with unique Babylonian reworkings.3

3 The lateness of the Tanhuma corpus and its reworking of earlier material has been dealt in
the scholarly literature, see e.g. C. Milikowsky, “Jacob’s Punishment: Studies in the Editing
of Midrash Tanhuma,” Bar Ilan 18/19 (1980) 144–9 [Hebrew]; idem, “Seder Olam and the
Tosefta,” Tarbiz 49 (1980) 246–63 [Hebrew]; M. Bregman, “Early Sources and Traditions in
the Tanhuma-Yelamedenu Literature,” Tarbiz 60 (1990) 269–74 [Hebrew]; J. L. Rubenstein,
“From Mythic Motifs to Sustained Myth: The Revision of Rabbinic Traditions in Medieval
Midrashim,” Harvard Theological Review 89 (1996) 131–59; D. Steinmetz, “Beyond the
Verse: Midrash Aggadah as Interpretation of Biblical Narrative,” Association of Jewish Studies
Review 30 (2006) 325–45.
286 NIKOLSKY

The present study adopts a narratological point of view; this means look-
ing not only at the verbatim similarity or dissimilarity between the corpora,
but also into (1) similarities on the level of the “story,” (2) structural similari-
ties (where one text follows the sequence of the narrative of another), and
(3) similarities in exegetical motifs.4 Verbatim similarities point to acquain-
tance of one fixed text with another; structural similarities, or common exe-
getical motifs, point to an acquaintance with an oral rendering of a narrative
which is not yet fixed. Rabbinic culture as a learning society developed a mech-
anisms for the transfer of both fixed and non-fixed.
The two narratives that I will study are (1) an interpretation of a verse
from the story of the destruction of Sodom and Gomorrah, and (2) The
prophecy of Eldad and Medad.

1 “God rained upon Sodom”

The first rabbinic interpretation is of a verse from the biblical story of the
destruction of Sodom and Gomorrah. The problem the midrash faces in this
narrative is triggered by a difficult verse ‫וַ ה׳ ִה ְמ ִטיר ַעל ְסד ֹם וְ ַעל ֲעמ ָֹרה ּגָ ְפ ִרית וָ ֵאׁש‬
‫“( ֵמ ֵאת ה׳ ִמן ַה ָּׁש ָמיִ ם‬Then the Lord rained upon Sodom and upon Gomorrah
brimstone and fire from the Lord out of heaven,” Gen 19:24). The difficulty is
that the word “Lord” appears in this verse twice, once as a subject, “the Lord
rained,” and once as an indirect object “from the Lord.” From the solutions
offered by the rabbis to this enigma, we can conjecture that in various periods
and cultures, the rabbis were troubled by a diversity of problems posed by this
verse, and not only the grammatical one mentioned above.
The verse is discussed in Sifre Zuta from the tannaitic period, Genesis Rabba
from the “classical” amoraic period, the Bavli from the amoraic (or even stam-
maitic) Babylonian culture, and the Tanhuma, of the late rabbinic Palestinian
period.

1.1 The Tannaitic Discourse—Two Entities in Heaven?

The appearance of the word “Lord” twice in this verse can be interpreted as
referring to two entities in heaven, in which the first “Lord” is different from the

4 The concept “exegetical motif” was developed by James Kugel (In Potiphar’s House: The
Interpretive Life of Biblical Text [Cambridge MA 1994] 3–9, and The Ladder of Jacob: Ancient
Interpretations of the Biblical Story of the Sons of Jacob [Princeton 2009] 4–7), and was
described by him as “the underlying idea about how to explain a Biblical verse, that becomes
the basis, or part of the basis, of a narrative expansion” (Kugel, In Potiphar’s House, 8).
FROM PALESTINE TO BABYLONIA AND BACK 287

second one. This seems to be the theological problem with which the rabbis of
the tannaitic period struggled.
Peace is the focus of the tannaitic discussion in Sifre Zuta, since it appears
in connection with the priestly blessing “God will turn his face to you and give
you peace” (‫)ישא ה׳ פניו אליך וישם לך שלום‬. The discussion leads to the rabbis’
assertion that peace is great in God’s eyes. Next, Rabbi Shimeon describes how
God’s actions are different from human actions: in contrast to humans, God,
when performing a peaceful act, arrives accompanied by others, but when
He fights—He does so alone. Some verses are quoted to prove both manners
of God’s action. Our verse is quoted to prove that God fights alone; here is
the quote:

SifZut 6:26 (Horowitz-Rabin edition, )249 ‫ עמ׳‬,‫רבין‬-‫ספרי זוטא ו כו (הורויץ‬

p. 249)
When He took revenge of the five cities ‫ לא‬,‫ כשפרע מחמשת כרכי סדום‬,‫וכן את מוצא‬
of Sodom, He did it alone, as it is stated: ‫ ״וה׳ המטיר על סדום‬:‫פרע מהן אלא יחידי שנא‬
“Then God rained upon Sodom and upon ‫ועל עמורה גפרית ואש מאת ה׳ מן השמים״‬
Gomorrah brimstone and fire from God .)‫(בראשית יט כד‬
out of heaven” (Gen 19:24).

Truth be told, our verse does not really prove that God fights alone. Even worse:
the double usage of the word “God” in this verse might point to the contrary.
The very use of this verse as proof that God works alone points both to the
problem the rabbis faced in this verse, as well as to the solution they offered:
it is God alone who rained brimstone and fire upon Sodom and Gomorrah;
the midrash uses here a “rhetoric of the obvious”5 to assure the reader that no
multiplicity in heaven is found in this verse.

1.2 The Palestinian Amoraic Midrash—Socio-Moral Values

A shift or discontinuity with the tannaitic culture, or at least its agenda, is
apparent in the amoraic composition Genesis Rabba, which asks different
questions about the verse, gives different answers, and is even oblivious to the
issue that was so important to the tannaitic sage. Here is the text:

5 The term was coined by Moshe Lavee; see his contribution to the present volume.
288 NIKOLSKY

GenR 51:2 (Vilna print)6 )‫בראשית רבה נא ב (דפוס וילנא‬

1. “And the Lord rained upon Sodom [and .)‫״וה׳ המטיר על סדום׳ וגו׳״ (בראשית יט כד‬
upon Gomorrah brimstone and fire from
the Lord out of heaven]” (Gen 19:24).
This is similar to two cities that rebelled :‫ אמר המלך‬.‫משל לשני מדינות שמרדו במלך‬
against the king. The king ordered [that] .‫ וא׳ תשרף מטמיון‬,‫תשרף אחת משלה‬
One [city] should be burned by using its
own [money], and the other should be
burned from the king’s treasury.
Thus it is said: “And its [the city’s] ‫כך להלן ״ונהפכו נחליה לזפת ועפרה לגפרית״‬
streams shall be turned into pitch, and ‫ ברם הכא ״וה׳ המטיר על סדום‬,)‫(ישעיה לד ט‬
the its dust into brimstone . . .” (Isa 34:9). .)‫ועל עמורה׳ וגו׳״ (בראשית שם‬
but here: “And the Lord rained upon
Sodom and upon Gomorrah . . .” (Gen
19:24)
2. Rabbi Abun said: This is similar to a .‫ לשפחה שהיתה רודה פת בתנור‬:‫אמר ר׳ אבון‬
slave-woman who was taking out bread 7‫ בא בנה‬.‫בא בן גברתה ורדת פת ונתנה לו‬
from the oven. The son of her mistress .‫ורדת גחלים ונתנה לו‬
came [to her], and she took out bread and
gave it to him. When her own son came to
her, she gave him coals.
Thus is it stated later: “And God said to ‫ ״ויאמר ה׳ אל משה הנני ממטיר לכם‬:‫כך להלן‬
Moses: I am raining bread from heaven for ‫ ברם הכא‬.)‫לחם מן השמים״ (שמות טז ד‬
you” (Exod 16:4) but here “And the Lord ‫״וה׳ המטיר על סדום ועל עמורה גפרית ואש״‬
rained upon Sodom and upon Gomorrah .)‫(בראשית יט כד‬
brimstone and fire” (Gen 19:24).
3. Rabbi Helbo ben Rabbi Hilfi bar ‫ר׳ חלבו בן ר׳ חילפי בר סמקאיי בשם ר״י בר ר׳‬
Samkai in the name of Rabbi Yehudah ‫ ״מאת‬.‫ ״וה׳ המטיר על סדום״ זה גבריאל‬:‫סימון‬
ben Rabbi Simon: “And the Lord rained ‫ כל מקום‬:‫ אר״א‬.‫ זה הקב״ה‬,‫ה׳ מן השמים״‬
upon Sodom”—this is [the angel] .‫ הוא ובית דינו‬,‫שנאמר ׳וה׳׳‬
Gabriel; “from the Lord out of heaven”—

6 This is the version of MSS ‫ד ת ש‬. In Theodor-Albeck’s chosen version for the text (p. 532) the
slave woman gives the better loaf to her son and the burned one to her mistress’ son. I find
this version inferior because the sharpness of the parable, pointing to the tragic situation of
the slave-woman, is lost; further, the Sodomites in this inferior version are represented by the
mistress’ son, while the Israelites are represented by the slave woman; such a representation
does not make sense.
7 The version in the mss is ‫בן בנה‬, which is clearly a graphic error.
FROM PALESTINE TO BABYLONIA AND BACK 289

this is the Holy One Blessed Be He. Rabbi

Aha said: Whenever it is stated “and the
Lord” [it means] Him and his [heavenly]
court.
4. Rabbi Yitshaq said: We find in the Torah, ‫ בנביאים וכתובים מצינו‬,‫ בתורה‬:‫א״ר יצחק‬
in the Prophets and in the Writings that a .‫שההדיוט מזכיר שמו ב׳ פעמים בפסוק אחד‬
simple person mentions his name twice.
In the Torah: “And Lemech said to his .)‫ ״ויאמר למך לנשיו״ (בראשית ד כג‬:‫בתורה‬
wives: [‘Adah and Tsillah, hear my voice; ‫״נשיי״ אין כתיב כאן אלא ״נשי למך האזנה וגו׳״‬
wives of Lemech, hearken unto my .)‫(שם‬
speech’]” (Gen 4:23). It does not say:
“my wives” but rather “wives of Lemech.”
In the Prophets: “And the king said to ‫ ״ויאמר המלך להם קחו עמכם את‬:‫בנביאים‬
them: Take with you the servants of your ‫עבדי אדוניכם והרכבתם את שלמה בני על‬
lord, and cause Solomon my son to ride .)‫הפרדה אשר לי וגו׳״ (מל״א א כג‬
upon my own mule etc” (1Kgs 1:33).
It does not say here [“my servants,” but ‫ ״את שלמה‬:‫ אלא‬,‫״את בן המלך״ אין כתיב כאן‬
“the servants of your lord”].8 .‫בני״‬
In the Writings, as it is written: “Since a ‫ דכתיב ״כי כתב אשר נכתב בשם‬,‫בכתובים‬
command that was written in the king’s .)‫המלך ונחתם בטבעת המלך״ (אסתר ח ח‬
name and was signed with the king’s
ring” (Est 8:8).
And you wonder why the Holy One ‫ואת תמיה שהקב״ה מזכיר שמו שני פעמים‬
Blessed Be His Name mentions His name !?‫בפסוק אחד‬
twice?!

Genesis Rabba answers three questions regarding this verse, two concerning
the word “rained,” and one concerning the repetition of the word “Lord”; the
first question (which is not explicitly phrased, but is implied from the answer)
is how is it possible that the infliction of the horrible substances of brimstone
etc. can, according to one verse, originate in heaven (“and the Lord rained
brimstone”) and according to another verse originate from below (“the streams
thereof shall be turned into pitch, and the dust thereof into brimstone,” Isa
34:9); Genesis Rabba answers with a parable about a city that needs to be
destroyed, and one time God destroys it with his own might, and at another

8 Here I emend the text. The example is quoted to refer to the words “the slaves of your master”
instead of “my slaves,” and not to the end of the verse as is specified in the Hebrew cited in
the parallel column.
290 NIKOLSKY

time, with the help of the his treasury,9 thus the same result is achieved by
using different forces.
The second question about the word “rained” (again, only implied) is how
is it that this word appears in one verse to indicate something negative (“rain-
ing brimstone and fire”) while in another verse it points to something posi-
tive (“Then said the Lord unto Moses: Behold, I will cause bread to rain from
heaven for you,” Exod 16:4). Genesis Rabba answers this with a parable about a
maidservant standing near the oven, taking bread out for the son of the king,
but ashes for her own son. Although explaining how two objects with com-
pletely different value can come out of the same “oven,” this parable does not
seem to be constructed to answer the question that emerges from the verse;
the parable accentuates a situation where a mother gives her own son some-
thing inferior, while giving someone else’s son something better. This is not the
problem that the verse from Genesis posed for the rabbis, although the ques-
tion how can the same word, “to rain,” relates to both positive and negative
phenomena is answered.
The third difficulty which Genesis Rabba addresses is the appearance of the
word “Lord” twice in the verse, once as a subject “and the Lord rained,” and
once as an indirect object “from the Lord”; it offers two basic solutions for this
difficulty. One interprets it as referring to an entourage that accompanied the
Lord when he rained brimstone and fire on Sodom. This is an exact opposite
solution to the one offered by the tannaitic composition, but it seems that for
the amoraim, God with an entourage poses no difficulty. Genesis Rabba quotes
the opinions of two rabbis, claiming that in this action the angel Gabriel
accompanied God (so Rabbi Helbo bar Simlai) or that God was joined in this
by his heavenly court, ‫( בית דינו‬Rabbi Aha).
Another solution is that of Rabbi Yitshaq, who says: “In the Torah, in the
Prophets and in the Writings we find that even a layman (a hediot, the Greek
idiotes) refers to himself twice in one verse” (‫ בתורה בנביאים וכתובים‬:‫א״ר יצחק‬
‫)מצינו שההדיוט מזכיר שמו ב׳ פעמים בפסוק אחד‬, so obviously God can do so too.
The text goes on to quote verses from the three sections of the Bible. One is
from the Torah (about Lemech and his wives), another from the Prophets (of
King David talking to his servants) and a third from the Writings (from the
book of Esther) to exemplify this phenomenon. The obvious conclusion is that
the fact that God mentions his name twice in the verse should not come as a
surprise; it is proper for Him to do so.

9 The word ‫טימיון‬, derived from the Greek τεμαῖον, referring to the Latin fiscus (imperial
treasury).
FROM PALESTINE TO BABYLONIA AND BACK 291

However, things are not as clear cut as I just presented them. In fact, there
is a gap between what Rabbi Yitshaq says and the substance of the verses here.
In none of them does the speaker mention his name twice (‫ההדיוט מזכיר שמו‬
‫)ב׳ פעמים בפסוק אחד‬. In fact, the connecting factor in all these verses is that
the speaker speaks of himself in third person, instead of using a first person
singular pronoun to refer to himself. Thus Lemech says “the wives of Lemech,”
instead of saying “my wives”; David says “the son of the king,” instead of “my
son”; and Ahasuerus says “with the ring of the king,” instead of “with my ring.”
This phenomenon is certainly parallel with what is found in our verse, “God
rained . . . brimstone and fire from God” instead of saying “God rained . . . brim-
stone and fire from him/himself.” This is indeed a grammatical problem, but it
is not an example of the problem the rabbis raised, which is “a hediot mention-
ing himself twice in one verse.”
The manner in which the quoted texts are introduced points to what the
question really was for the rabbis; it was not that the same name appears twice,
nor that the same person appears as the subject and as the direct object in
the same verse. It seems that, for the rabbis, referring to oneself in the third
person is considered boasting. And yet, given the occurrences of laymen refer-
ring to themselves in the biblical text in the third person, we should not ques-
tion the legitimacy of God doing so. We recognize that this was the problem
from the phrase which introduces the verses: ‫מצינו שההדיוט מזכיר שמו ב׳ פעמים‬
‫בפסוק אחד‬.
Let me summarize the sequence of narratives in Genesis Rabba as this will
be important later on when compared with Midrash Tanhuma.

1. In answering the question, how can brimstone and fire come from heaven
in one case and from rivers in another, a parable about two cities is pre-
sented as a solution.
2. In answering the question, how the word “to rain” can be positive in one
case, and negative in another, a parable about a maidservant is related.
3. Three answers are offered to the question of why does the word “Lord”
appear twice in the verse. The answers offered are:
a. He had the angel Gabriel with Him
b. He had His hosts with Him
c. Other people refer to themselves in the third person in biblical texts,
and thus God can do so too. These are the other verses which do the
same in the Bible:
• Lemech in the Torah (Gen 4:23)
• King David in the Prophets (1Kgs 1:33)
• King Ahasuerus in the Writings (Est 8:8)
292 NIKOLSKY

Obviously, the third answer (c above) does not address the problem of dualism
as the first two do.

1.3 The Babylonian Version

The Bavli incorporates part of the narrative from Genesis into a discussion
about how to answer the minim (heretics);10 the passage is constructed out
of four separate sections which, when put together, create a flowing discus-
sion and produce a unified message. In what follows I will summarize the Bavli
sugya in bSanh 38b.
(1) It begins with a statement of Rabbi Yohanan: “every verse to which the
minim pointed as asserting multiplicity in heaven, there is a parallel which dis-
proves this conviction” (‫ כל מקום שפקרו המינים תשובתן בצידן‬:‫)אמר רבי יוחנן‬. A list
of verses that can be suspected as acclaiming multiple heavenly entities and
verses to the contrary, are quoted. This list ends abruptly11 when a new section
is inserted, one which has a parallel in bHag 14a.
(2) In this new section we find a baraita in the name of Rabbi Aqiva, inter-
preting Dan 7:9 in which “thrones” (in the plural) of God are mentioned. Rabbi
Aqiva explains that these are chairs for God and for King David. This expla-
nation is violently denied by Rabbi Yosi (‫ עד מתי אתה עושה שכינה חול‬,‫עקיבא‬,
“Aqiva, how long will you be turning holy into secular?”); Aqiva accepts Rabbi
Yosi’s interpretation, that the two thrones refer to judgment and grace, but
this interpretation too is refuted, this time by Rabbi Eleazar Ben Azariah, who
suggests that Rabbi Aqiva cease to engage in aggadah, and stay with halakhah
(‫ מה לך אצל הגדה? כלך אצל נגעים ואהלות‬,‫ עקיבא‬:‫אמר לו רבי אלעזר בן עזריא‬, “Aqiva,
what are you doing with aggadah, go to [study] purity laws”).
(3) In the next section we learn from Rav Nahman that “those who can refute
the heretics as well as Rav Idit could—should do it; those who cannot—should

10 I will not discuss here the issue of the minim; see the following bibliography about them:
W. Horbury, “The Benediction of the Minim and Early Jewish-Christian Controversy,”
Journal of Theological Studies 33 (1982) 19–61; R. Kimelman, “Birkat ha-minim and
the Lack of Evidence for an Anti-Christian Prayer in Late Antiquity,” in: E. P. Sanders,
A. I. Baumgarten, and A. Mendelson (eds.), Jewish and Christian Self-Definition, vol. 2
(Philadelphia 1981) 226–44; S. J. D. Cohen, “The Significance of Yavne: Pharisees, Rabbis,
and the End of Jewish Sectarianism,” Hebrew Union College Annual 55 (1984) 27–53;
M. Goodman, “The Function of ‘Minim’ in Early Rabbinic Judaism,” in: H. Cancik,
H. Lichtenberger, and P. Schäfer (eds.), Geschichte—Tradition—Reflexion: Festschrift für
Martin Hengel zum 70 Geburtstag vol. 1 (Tübingen 1996) 501–10; S. S. Miller, “The ‘Minim’
of Sepphoris Re-considered,” Harvard Theological Review 86 (1993) 377–402.
11 With the verse from 2Sam, 7:23, which is not being refuted.
FROM PALESTINE TO BABYLONIA AND BACK 293

not answer them” (‫ ואי לא—לא‬,‫האי מאן דידע לאהדורי למינים כרב אידית—ליהדר‬
‫)ליהדר‬.
(4) Only at this stage is our verse about the Lord raining fire and brimstone
introduced:

A min said to Rabbi Yishmael bar Rabbi :‫אמר ליה ההוא מינא לרבי ישמעאל ברבי יוסי‬
Yosi: It is written: “Then God rained upon ‫כתיב ׳וה׳ המטיר על סדם ועל עמרה גפרית‬
Sodom and upon Gomorrah brimstone ‫ואש מאת ה׳׳ (בראשית יט כד) מאתו מיבעי‬
and fire from God” it should have said !‫ליה‬
“from Him” (Gen 19:24).
A washer said to him: Leave him to me, ‫ אנא מהדרנא‬,‫ שבקיה‬:‫אמר ליה ההוא כובס‬
I will answer him; as it is written: “And ‫ ״ויאמר למך לנשיו עדה וצלה‬:‫ דכתיב‬:‫ליה‬
Lemech said to his wives, Adah and Tsillah ‫ ״נשיי״ מיבעי ליה! אלא‬."‫שמען קולי נשי למך‬
listen to me, wives of Lemech [hearken to ‫ הכא נמי—משתעי קרא‬.‫משתעי קרא הכי‬
my voice]” (Gen 4:23). It should have said .‫הכי‬
“my wives” [instead of “wives of Lemech”]!
But this is how a verse speaks. Here also—
this is how the verse speaks.
(Rabbi Yishmael bar Rabbi Yosi) said to ?‫ מנא לך הא‬:‫אמר ליה‬
him: From where did you learn this?
(I heard it) from the teaching of Rabbi .‫מפירקיה דרבי מאיר שמיע לי‬
Meir.
As Rabbi Yohanan said: When Rabbi Meir ‫ כי הוה דריש רבי מאיר‬:‫דאמר רבי יוחנן‬
was teaching in the rabbinic assembly ‫ תילתא‬,‫ הוה דריש תילתא שמעתא‬,‫בפירקיה‬
(pirqa), he would expound one third from .‫ תילתא מתלי‬,‫אגדתא‬
what he heard from his teachers, one third
aggadah, and one third parables.

The Babylonian passage is not focused on the importance of peace (as was the
tannaitic Sifre Zuta), nor is it restricted by the sequence of the biblical narra-
tive (as Genesis Rabba was); rather, it utilizes part of the Palestinian amoraic
discussion for its own purpose, which is to refute difficulties in biblical verses
pointed out by heretics.
In this context, Rabbi Aqiva is presented as unable to conduct such a discus-
sion with heretics, and Rabbi Meir is praised for his method of teaching in the
rabbinic assembly (pirqa),12 because a launderer who studied with Rabbi Meir

12 About pirqa, a Babylonian initiated custom of rabbis teaching the public, see: I. M. Gafni,
“On Public Lectures in Talmudic Babylon: The Pirqa,” in: S. Elizur et al. (eds.), The Assembly
of Ezra (Jerusalem 1994) 121–9 [Hebrew].
294 NIKOLSKY

was able to answer an annoying question put forward by a min better than
Rabbi Aqiva.13
The question which the washer-man could answer concerns our verse,
where the word “Lord” appears in two different functions; in his answer, the
washer-man quotes the verse about Lemech from Gen 4:23, which Genesis
Rabba also quotes, and reasons that “this is how a verse speaks.”
All that is left in the Bavli from the Palestinian amoraic discussion is the
argument that the possible duality in the verse about brimstone and fire can
be refuted by the verse in which Lemech addresses his wives as “the wives of
Lemech.” The Bavli exhibits knowledge of the Palestinian exegetical tradition,
but no knowledge of a composition similar to Genesis Rabba.

1.4 The Tanhuma Reworkings (TanB Vayera 19)

In light of this, it is interesting to observe the Tanhuma’s rendering of this
midrash. The Tanhuma exhibits a structural similarity with Genesis Rabba,
namely, it includes most (but not all) the elements from the Genesis Rabba pas-
sage, but it also reworks them intensively. I will enumerate the places where
Tanhuma alters the Genesis Rabba narrative, first comparatively, in the form of
a table, and later by explaining the elements presented in the table.

Element in the narrative Genesis Rabba Tanhuma

1 Answering the question of how A parable about (Lacking in the

brimstone and fire can come two cities Tanhuma)
from heaven in one case, and
from the rivers in another
2 Answering the question of how A parable about a A different,
the word “to rain” can be maidservant “improved” version
positive in one case, and of the parable
negative in another (about a king)

13 The fact that Rabbi Aqiva, a most celebrated rabbinic master, was not able to utilize
aggadah properly in a polemic against a heretic, but Rabbi Meir, whose status in the
Bavli is usually much more problematic is interesting in its own right, but will not be
discussed here. For Rabbi Meir’s status as a loyal disciple of Elisha ben Abbuya (bHag 15a),
being involved in a problematic story with his wife Beruriah’s sister (bAZ 18a–b), taking
part in an attempt to overthrow Rabban Shimeon ben Gamaliel (bHor 13b–14a), see J.
L. Rubenstein, Talmudic Stories: Narrative Art, Composition, and Culture (Baltimore 1999)
176–211.
FROM PALESTINE TO BABYLONIA AND BACK 295

Table (Cont.)

Element in the narrative Genesis Rabba Tanhuma

3 Three answers to the question of a) He had the a) + b) God and

why the word “God” appears angel Gabriel a court of law
twice in the verse with Him decided to punish
b) He had His the Sodomites
hosts with Him c) A more linguisti-
c) Other people cally oriented
mention their answer to this
name twice in question
a verse, so God
can too.
4 Examples from the Bible to I. Lemech in the The same exam-
answer c) above (people Torah (Gen 4:23) ples, but not
mentioning their name twice) II. King David in according to the
the Prophets order of the Bible:
(1Kgs 1:33) I. King David
III. King II. Lemech
Ahasuerus in the III. King Ahasuerus
Writings (Est 8:8)

Unlike Genesis Rabba, the Tanhuma offers only one (not two) explanations for
the word “rained.” It answers the question of how two contrasting phenomena
can be expressed by the same word. The parable about the maidservant taking
bread out of the oven for the son of the king, and coal for her own son, is trans-
formed in the Tanhuma into a parable about a baker standing at his oven and
handing out bread to his lover and coal to his enemy. 14
A version of the parable is found in TanB beshalah 20, also in the context of
the verse about God raining brimstone. I will analyze it first, and then the ver-
sion which appears in our passage.

14 Except in our passage (TanB vayera 19), the parable is found in other tanhumic texts:
TanB beshalah 20; a different version of this parable is found in TanB vayera 23, in Tan
vayera 10 and in the Yelammedenu (of Mann, from Yalkut Talmud Tora) 54b (beginning of
number 98).
296 NIKOLSKY

TanB beshalah 20 ‫ בשלח כ‬,‫תנחומא בובר‬

To what is this similar? To a baker who ‫למה הדבר דומה? לנחתום שהיה עומד על‬
was standing by his oven; when his enemy .‫ נכנס שונאו וחתה גחלים ונתן על ראשו‬,‫הכבשן‬
entered, he put coal on top of it [i.e. a loaf ‫ ונתן לו הגחלים‬,‫נכנס אוהבו רדה פת חמה‬
which is already cold], when his friend .‫ שניהם יצאו מתוך הכבשן‬.‫והפת‬
entered, he took out a warm loaf and gave
him the coal and the bread. Both came
from the [same] oven.

Tanhuma distinguishes between a baker, who gives a fresh warm loaf to his
loved one, while to his enemy he gives a loaf re-heated by placing coal on top of
it—thus stating that the quality of the bread received depends on the baker’s
relationship with the client. The parable here is better suited for our passage
than the one from Genesis Rabba, because the one here focuses on how the
thing given depends on the recipient and not on the issue which Genesis Rabba
illustrates, i.e. the horrible fact that a woman has to treat the king’s son better
than she does her own son. Tanhuma is, thus, improving on the Genesis Rabba
text, while keeping the context identical and the content similar to what was
before it. The inclusion of the parable in its context, and the changing of it to
illustrate the argument better, both point to the Tanhuma’s lively and attentive
milieu and exposes the more literary nature of Genesis Rabba.
Still, this version of the parable is a bit awkward, since if the stove is the
source of everything given (i.e. God), who is represented by the baker? And
if the baker is God (which in itself is strange, given that it is considered a low
class occupation), what is the stove?
A version of this parable which is even better suited to our passage about
the punishment to the Sodomites is the one found in TanB vayera 19:

TanB vayera 19 ‫ וירא יט‬,‫תנחומא בובר‬

Rabbi Berachia said: To what is this simi- ‫ למה הדבר דומה? למלך שהיה‬:‫אמר ר׳ ברכיה‬
lar? To a king who was standing by an oven ‫ נכנס אוהבו‬.‫עומד אצל הכבשן של נחתום‬
of a baker; when his loved one entered, he —‫ נכנס שונאו‬.‫אצלו—רדה פת חמה נתן לו‬
took out a warm loaf and gave him. When .‫חתה גחלים ונתן לו‬
his enemy entered, he took out coal and
gave him.

This version seems to have been reworked especially for our passage; it under-
mines the semiosis of the low-class baker, by placing the king near the oven,
who hands out bread to his loved ones and coal to his enemies. The king as the
FROM PALESTINE TO BABYLONIA AND BACK 297

protagonist of the parable is a better-suited image for representing God than a

baker (or a female slave, as in Genesis Rabba). And it is he, the God-king, who
rained bread on the Israelites and brimstone and fire on the Sodomites.
But as much as the parable suits better the nimshal (the referent of the par-
able), the narrative of the parable itself becomes less logical: it is less logical to
have a king stand by the oven than a baker. And further—all realism which is
found in the “baker” version, which differentiates between the desired warm
loaves and the cold one which was artificially warmed by the coal—is lost by
the crude non-realistic behavior of the king who hands out coal.
Thus, the version of the parable, with the baker as the protagonist, seems to
be the original one, and the one in our passage, where the king stands by the
oven, is a reworking of it.
Returning to the comparison of the Genesis Rabba parable, we find the
Tanhuma parable very similar in principle, but very different in details, point-
ing to a cultural, but not necessarily a textual, borrowing.
In the next section of the sequence (number 3 in the table), instead of men-
tioning Gabriel or the heavenly host joining God, as Genesis Rabba did, the
Tanhuma depicts God sitting in the Sanhedrin of seventy-one (‫מהו וה׳? שישב‬
‫ )עליהם בדין סנהדרין של שבעים ואחד וחייבן‬discussing the Sodomites’ actions
and sentencing them to be burnt by fire. To fulfill the Sanhedrin’s verdict,
God rained brimstone and fire upon them. This image presents God head-
ing a court of justice, rather than as a mythic persona surrounded by other
heavenly beings.
The next issue (number 4 in the table) is the double appearance of the
word “Lord” in the verse. Unlike Genesis Rabba, which focuses on the social
aspect, and shows that even simple people mention their name twice—so
we should not wonder about God doing this (‫ואת תמיה שהקב״ה מזכיר שמו שני‬
!?‫פעמים בפסוק אחד‬, “and you wonder that The Holy One Blessed Be He men-
tions his name twice in one verse?!”)—the Tanhuma relates to the linguistic
problem, how can the subject of a sentence also be the indirect object, stating
that “there is such a similar phenomenon in other biblical verses” (‫יש בפסו־‬
‫)קין כיוצא בדבר‬. An example proves the case, but the Tanhuma, in contrast to
Genesis Rabba, does not state that three examples will now be given—it simply
quotes King David referring to himself in the third person, this being the best
verse to parallel God’s action, if only one example is to be used. It seems to
me that originally this example sufficed. Later, someone who was acquainted
with the Genesis Rabba narrative added to the Tanhuma the other two exam-
ples, resulting in three verses being quoted, though not according to the
biblical sequence.
298 NIKOLSKY

The above analysis shows that the Tanhuma was well-acquainted with the
Palestinian amoraic midrash about the verse in question. It also shows that
the Tanhuma does not quote the midrash verbatim, but rather reworks it, in
this case also improving on the amoraic material, at least from a narratological
point of view.
However, there also seems to be an influence of the Babylonian narrative on
this text as well. This can be seen in two points: (1) the Tanhuma addresses the
linguistic problem in the verse, as did the Bavli, not the social one as in Genesis
Rabba; (2) the words inserted before solving the linguistic problem of the verse
in question are of special significance:

If a person comes and asks you: What is :‫אם בא אדם לשואלך מה כתיב בפסוק הזה‬
written in this verse: “and the Lord rained ,)‫״וה׳ המטיר וגו׳ מאת ה׳״ (בראשית יט כד‬
etc. . . . from the Lord” (Gen 19:24) tell .‫ יש בפסוקין כיוצא בדבר‬:‫אמור לו‬
him, you can find in other verses such a
phenomenon

Presenting the solution to the linguistic problem in the verse as part of a dia-
logue with another person is not taken from Genesis Rabba, but is part and
parcel of the discussion in the Bavli, which presents a polemic with heretics
(sections 3 and 4 of the Bavli discussion, as was summarized above). Thus,
the Tanhuma and the Bavli tell a similar story, which is not the one told in
the Palestinian sources. At the same time, in its basis, the Tanhuma narrative
remains very close to the Genesis Rabba composition, as is evident from the
structural similarity between the two corpora.

2 Nominating the Seventy Elders, and the Prophesying of

Eldad and Medad

The biblical story about nominating seventy elders to assist Moses in adminis-
tering the Israelite coalition (Num 11:16–17, 25) is combined, in the Bible, with
the story about Eldad and Medad, who remained in the camp and prophesized
(Num 11:26), and with the story about the quail which God supplied the glut-
tonous Israelites (Num 11:31).
In the midrash, these three stories are handled in two tannaitic sources,
Sifre on Numbers and Sifre Zuta, in the Bavli and in the Tanhuma literature
(Yelamedenu, the two Tanhumas, and Numbers Rabba). This sequence of
stories is therefore another case in which we can observe the development
FROM PALESTINE TO BABYLONIA AND BACK 299

and change of a motif, one particle of the cultural canon, through time and
geographical locations. From the late midrash I will only discuss the Tanhuma
Buber version since the other versions do not add information which is signifi-
cant for our purpose.15

2.1 The Tannaitic Midrash

SifNum 95 ‫ספרי במדבר צה‬
(Horovitz Rabin edition, pp. 95–96)
1. (“And Moses came out, and talked to ‫וירד ה׳‬ . . . ‫ ״ויצא משה וידבר אל העם‬.1
the people . . . And God came down in . . . ‫בענן‬
the cloud, [and spoke unto him, and
took of the spirit that was upon him, and
put it upon the seventy elders . . .] But ‫וישארו שני אנשים במחנה״ (במדבר יא‬
there remained two men in the camp”) .)‫כו‬-‫כד‬
(Num 11:24–26).
Some say that they remained in a ballot ‫ לפי שאמר לו‬,‫ בקלפי נשתיירו‬:‫יש אומרים‬
(kalpe), because when God told Moses :‫ אמר משה‬,‫הקב״ה למשה לבור לו שבעים זקנים‬
to nominate seventy elders, Moses said: ‫מה אני עושה? הרי הם נופלים ששה וששה לכל‬
What shall I do? [Should I nominate] six ‫שבט ושבט וחמשה וחמשה לשני שבטים? איזה‬
from each tribe, and five from two tribes? ?‫שבט מקבל עליו לבור ממנו חמשה‬
Which tribe will be ready to have only
five people nominated from it?
So Moses found a solution: He took sev- ‫ נטל שבעים פיתקים וכתב‬:‫עשה משה תקנה‬
enty notes and wrote on them “elder” and ‫ ונטל שני פיתקים חלקים ובללן‬,‫עליהם זקן‬
took two notes that are empty, and mixed ‫ באו וטלו‬:‫ אמר להם‬.‫והטילם לתוך קלפי‬
them in the ballot, and told them [the .‫פיתקיכם‬
elders] come and take your notes.
Whoever took a note on which it was ‫כל מי שנטל פיתק וכתוב עליו ׳זקן׳ היה אומר לו‬
written “elder,” Moses said to him: God ‫ וכל מי שהיה נוטל‬.‫ כבר קידשך המקום‬:‫משה‬
has chosen you; and whoever took an ‫פיתק שלא היה כתוב בתוכו ׳זקן׳ היה משה אומר‬
empty note, Moses told him: It is from ?‫ מן השמים הוא מה אני יכול לעשות לך‬:‫לו‬
heaven, I cannot help you.
In the same manner you can understand ‫ ״ואת פדויי השלשה‬:‫כיוצא בו אתה אומר‬
the case of the “redemption of the three .)‫והשבעים והמאתים״ (במדבר ג מו‬
and the seventy and the two hundred”
(Num 3:46).

15 The Yelammedenu is too scant, the Warsaw edition does not differ much from the Buber
edition, and Numbers Rabba seems contaminated by the Bavli.
300 NIKOLSKY

2. Moses said: What shall I do now? ‫ מה אני אעשה עכשיו? כל אחד‬:‫ אמר משה‬.2
Everyone will tell me—the Levite has .‫ואחד אומר כבר פדאני לוי‬
already redeemed me.
Moses came up with a solution. He took ‫ ׳לוי׳‬:‫ נטל פיתקים וכתב עליה‬.‫עשה משה תקנה‬
notes and wrote on them “Levite,” and ‫ ׳כסף חמשת שקלים׳‬:‫ונטל פיתקים וכתב עליהם‬
other notes on which he wrote “five silver .‫ובללם והטילם בקלפי‬
sheqalim,” and he mixed them and them
in a ballot.
He told them: Come take your notes. ‫ כל מי שנטל‬.‫ באו וטלו פיתקיכם‬:‫אמ׳ להם‬
Whoever took a note on which it was ‫ כבר‬:‫ ׳בן לוי׳ אומר לו‬:‫ וכתוב היה עליו‬,‫פיתקו‬
written: “son of a Levite,” Moses told him: ‫ ׳כסף‬:‫ ומי שנטל פיתקו וכתוב עליו‬.‫אתה פדוי‬
You are redeemed already; and whoever ‫ צא ותן‬:‫חמשת שקלים׳ היה משה אומר לו‬
took a note on which it was written: “five .‫פדיונך‬
silver sheqalim” Moses would tell him:
Pay your redemption money.
3. Rabbi Shimeon says: “Remained in the ‫ ״במחנה נשתיירו״ (במדבר‬:‫ ר׳ שמעון אומר‬.3
camp” (Num 11:26), since they saw Moses ,‫ לפי שראו את משה שמברר לו הזקנים‬.)‫יא כו‬
choosing the elders, they said: We are not ‫ הלכו והטמינו את‬.‫ אין אנו כדיי לגדולה זו‬:‫אמרו‬
worthy of such greatness; they went and .‫עצמם‬
his themselves.
God told them: Since you belittled your- ‫ אני‬,‫ אתם מעטתם את עצמכם‬:‫אמר להם המקום‬
selves, I will make you greater than all of ‫ בשבעים זקנים אומר‬.‫אגדל אתכם יותר מכולם‬
[the elders], [as] it says of the elders that .‫ נתנבאו לפי שעה‬,‫״ויתנבאו״ (שם כה) ולא יספו‬
they “prophesied” (Num 11:25) [one time] ‫באלדד ומידד הוא אומר ״ויתנבאו במחנה״ (שם‬
and no more, and [that they] prophesied .‫כו) שהיו מתנבאים עד יום מותן‬
about issues of this time; and about Eldad
and Medad it says that they “prophesied
in the camp” (Num 11:26) [meaning that]
they prophesied until the day they died.
And what were they saying [in their
prophecy]? That Moses will die, and ‫ומה היו אומרים? משה מת ויהושע מכניס את‬
Joshua will bring the Israelites into the .‫ישראל לארץ‬
Land.

The Sifre Numbers passage gives two explanations for the words “there
remained” in the verse “there remained two men in the camp, the name of
the one was Eldad, and the name of the other Medad; and the spirit rested
upon them . . .” (Num 11:26). The first (anonymous) explanation is that when
God commanded Moses to elect seventy elders, Moses was faced with an
FROM PALESTINE TO BABYLONIA AND BACK 301

arithmetic problem: if he were to nominate six elders of each tribe, he would

end up with seventy-two elders, which is two too many; if he were to pick five
of each tribe, he would have sixty elders, which is not enough; if he were to pick
six from some tribes and five from others, it would arouse jealousy between the
tribes. Moses then devised a method to make this nomination just: he put into
a ballot (kalpe) seventy notes with the word “elder” and two empty notes. All
elders were asked to take a note each. Those who picked a note with the word
“elder” became elders. The two who picked the empty notes did not become
elders. When the verse states that Elded and Medad “remained,” it means, so
Sifre Numbers, that they remained out of the list of 70 elders.16
This method of election, says Sifre Numbers, works also in the case of the
redemption of the first born, described in Num 3:44–51. There Moses had to
select those Israelites not to be redeemed by members of the tribe of Levi
who were dedicated to God, and therefore had to pay five shekels for their
redemption.
The second way to explain the words “there remained” is quoted in the
name of Rabbi Shimeon; in his version, there is an implied conviction that
seventy-two elders were already a known institution among the Israelites; hav-
ing to select only seventy of them posed a problem for Moses, but, says Rabbi
Shimeon, when Eldad and Medad heard that Moses was looking for only sev-
enty elders, they humbled themselves, claiming to be unworthy of the position,
leaving Moses with only the seventy elders which he needed. They, therefore,
were rewarded by being endowed with prophecy more than the other elders:
the other elders prophesied one day, while Eldad and Medad prophesied for
the rest of their lives. What did they prophesy about? About Joshua succeed-
ing Moses.
Sifre Zuta follows Sifre Numbers structurally and verbally, but it is not iden-
tical; some of the differences between the two are important because they
enable us to see which of the two versions was later incorporated into the
Bavli. For example, while Sifre Numbers states that Moses told the elders who
picked an empty note, “It is from heaven, so how can I help you?” (‫מן השמים‬
?‫ מה אני יכול לעשות לך‬,‫)הוא‬, Sifre Zuta says, “God doesn’t want you, so what can
I do for you?” (?‫ אני מה אעשה לך‬,‫)המקום לא חפץ בך‬. Sifre Zuta also omits the

16 The text says that they remained “in the box,” i.e. the box with the notes; this makes the
story slightly incomprehensible—does it mean that they did not pick out their notes?
Perhaps this refers to the explanation that will be apparent later, that they withdrew
themselves from the nomination process.
302 NIKOLSKY

comparison between the methods of nominating seventy elders and of redeem-

ing the first-born. It also adds another opinion about the content of Eldad and
Medad’s prophecies: they prophesied about the quail (a narrative directly fol-
lowing in the biblical chapter). All of the features of Sifre Zuta are also found
in the Bavli.

2.2 The Reworking in the Bavli

The Bavli (bSanh 11a) follows Sifre Zuta almost verbatim; only a few words or
expressions are slightly different (for example, the sentence “I will [uninten-
tionally] make the tribes to be jealous of each other” is ‫נמצא אני מטיל קנאה‬
‫ בין השבטים‬in Sifre Zuta and ‫ הריני מטיל קנאה בין השבטים‬in the Bavli); but the
Bavli adds new narratives between some passages in the sequence. One of
the new narratives is the paragraph found in Sifre Numbers, namely, the com-
parison between the nomination of the elders and the election the first-born
redemption.
A second and, for our purposes, an important addition in the Bavli is found
at the very end of the discussion (which is the locus most susceptible to addi-
tions). While Sifre Numbers ends the discussion claiming that Eldad and
Medad prophesied on one topic, and Sifre Zuta knows of two such prophesies,
the Bavli quotes both and adds a third prophecy. In the name of Babylonian
amora Rav Nahman, the Bavli asserts that Eldad and Medad prophesied about
Gog and Magog, the kings from the east who will storm with their armies the
Land of Israel at the End of Days.17 The Bavli derives support for this opinion
from a verse from Ezekiel, in the chapter that speaks of the coming of Gog
and Magog:

17 Gog, the leader of the land Magog (Ezek 38:2ff.), is the king to attack Israel at the end
of days (Ezek 38:16), and is bound to be defeated (Ezek 39:6) and him and his army
will end up buried in a valley in Jezreel (Ezek 39:11). In Genesis, Magog is mentioned as
one of the sons of Yefet (Gen 10:2). In the New Testament the names of the king and
the country became names of two leaders, Gog and Magog (Rev 20:7–10). In rabbinic
literature we find reference both to Gog alone (e.g. in Shevi 4:8, 35c; LevR 11:2), or to Gog
and Magog (mEduy 2:8). Gog is identified in Genesis Rabba as the nation of the Romans,
see R. Nikolsky, “Gog in Two Rabbinic Narratives,” in: W. J. van Bekkum, J. W. Drijvers,
and A. C. Klugkist (eds.), Syriac Polemics: Studies in Honour of Gerrit Jan Reinink (Leuven
2007) 21–40.
FROM PALESTINE TO BABYLONIA AND BACK 303

bSanh 11a ‫ב׳ סנהדרין יא ע״א‬

Rav Nahman said: They prophesied on ,‫ על עסקי גוג ומגוג היו מתנבאין‬:‫רב נחמן אמר‬
the issue of Gog and Magog, as it is writ- ‫ ״כה אמר ה׳ אלהים האתה הוא אשר‬:‫שנאמר‬
ten: “Thus said the Lord God: Are you he ‫דברתי בימים קדמונים ביד עבדי נביאי ישראל‬
of whom I spoke in old times through ‫הנבאים בימים ההם שנים להביא אתך עליהם‬
My servants the prophets of Israel, that .)‫וגו׳׳ (יחזקאל לח יז‬
prophesied in those days for many years,
that I would bring thee against them?”
(Ezek 38:17)
Read not “years” (shanim) but “two” .‫אל תיקרי ׳שנים׳ אלא ׳שנים׳‬
(shnayim).
And who are the two prophets who ‫ואיזו הן שנים נביאים שנתנבאו בפרק אחד‬
prophesied in the same prophecy in the ‫נבואה אחת? הוי אומר אלדד ומידד (ב׳ סנהדרין‬
same period? They are Eldad and Medad. .)‫יא ע״א‬

The Babylonian amora quotes a midrashic motif on the verse from Ezekiel
connecting Eldad and Medad’s prophecy to the eschatological event of Gog
and Magog. The rationale behind the association is that the word “years” (‫שנים‬
in the verse) can be read as “two” (‫)שניים‬, and the only two prophets who
prophesized the same prophecy at the same time were Eldad and Medad. This
exegetical motif is not found in any source before the Bavli, but is found in the
later Palestinian corpus, the Tanhuma.

2.3 The Tanhuma

The Tanhuma exhibits a close structural similarity to the early Palestinian tan-
naitic midrashim, with some additions and reworkings, and some lexical dif-
ferences that are only to be expected, located at a later point in the reception
continuum.
Side-by-side with the structural similarity, however, the Tanhuma tells a
story quite different from all previous versions, completely altering the sys-
tem of nomination: in the Tanhuma, the tannaitic system (putting seventy
notes with the word “elder” and two empty notes in a ballot) is attributed to
the tanna Rabbi Yehudah. A critique of the system is quoted in the name of
Rabbi Nehemiah, the well-known opponent of Rabbi Yehudah. The critique
claims that the method is unjust, because one could conceive of a situation
where an elder puts his hand in the box, but there would be no note saying
“elder” anymore. In such a case he would have no chance of being nominated,
which would render the procedure unjust. Rabbi Nehemiah claims that Moses
devised a different method: he put seventy-two notes with the word “elder,”
and seventy-two empty notes in a box; once seventy notes with the word
304 NIKOLSKY

“elder” were picked, the nomination process ended. Moses would then have
his seventy elders, and all seventy-two of them had a chance to be nominated
(assuming all elders had a chance to take a note out of the box at least once).
This critique and the new method of nomination show both the Tanhuma’s
acquaintance with the earlier traditions together with its independence in
amending and reworking the narratives, and making the nomination process
more just. This situation is similar to the one found in the previous narrative,
where the Tanhuma was creative with regard to the parable of the baker and
some other issues, while still following the structure of Genesis Rabba.
The last major change in the Tanhuma story is that it relates that Eldad and
Medad exceeded the other elders in prophecy in respect to five points (a num-
ber we have not encountered before). It enumerated the following aspects:
(1) the elders prophesied about the next day, but Eldad and Medad prophesied
about what would happen forty years in the future; (2) the elders did not enter
the Land of Israel, but Eldad and Medad did; (3) the elders’ names were not
specified, but Eldad and Medad’s were; (4) the elders’ prophetic power ceased,
but Eldad and Medad’s did not; and (5) Eldad and Medad received the spirit
directly from God, while the elders received it through Moses.
When discussing the first of these five aspects, the Tanhuma wonders about
the contents of Eldad and Medad’s prophecies. One (anonymous) opinion is
that they prophesied about Moses’ death, which is what is found in all previous
sources, but another (‫ )יש אומרים‬is that they prophesied about Gog and Magog,
which in the Tanhuma appears without the midrashic context we saw in the
Bavli. It would have been hard to explain how this statement came about had it
not been for the passage in the Bavli which we saw above, quoted in the name
of Rav Nahman, explaining the verse from Ezekiel as referring to Eldad and
Medad prophesying about Gog and Magog. This statement is, again, proof that
the Tanhuma was acquainted with this Babylonian tradition.

3 Conclusions

In this study I followed the continuum—the development that results from

both preserving the tradition and allowing it to become relevant for current
life—of two exegetical motifs in rabbinic literature. In the case of the verse
about Sodom and Gomorrah, where the word “Lord” appeared twice, we found
the Bavli utilizing part of the Palestinian narrative in a totally different con-
text. Rather than introducing the Palestinian narrative to solve an exegetical
problem, the Bavli introduced it in a discussion of how aggadah can be used
to answer arguments of heretics. In the Tanhuma, the core narrative is based
FROM PALESTINE TO BABYLONIA AND BACK 305

on earlier Palestinian sources, but the dialogic setting, in which the problem of
the verse is solved is taken over from the Babylonian usage of the Palestinian
narrative.
Similarly, in the second case, concerning the nomination of the Elders and
the prophecy of Eldad and Medad, we find distinct Babylonian influences on
the Tanhuma. The Bavli adds at the end of the discussion, which otherwise is
almost a verbatim rendering of the Palestinian midrash, additional narratives
(assigned by the Bavli to Rav Nahman); this tradition, in turn, is incorporated
into the Tanhuma, which is otherwise, for the most part, a reworking of earlier
Palestinian material.
These are only two minor examples of Babylonian traces in a corpus of a
Palestinian origin. Regardless of whether the text of the Tanhuma was com-
posed in the Land of Israel or in other areas of the Byzantine Empire, it was part
of a culture which as a rule resisted Babylonian influences, as was described
in the introduction. We see, then, that in spite of this resistance, Babylonian
traditions penetrated this barrier and entered Palestinian culture even before
the Gaonic period, as is usually assumed. It is, indeed, an aggadic text, with
no stated halakhic consequences, but it does help to clarify the cultural atmo-
sphere in Palestine of the Byzantine era.18

18 The Jewish Palestinian culture seems to have lost the battle in Medieval Europe to the
Babylonian hegemony. But traces of it can still be detected in a central medieval corpus
the Zohar, see R. Nikolsky, “Traces of Tanhuma in the Zohar,” in: A. Rapoport-Albert and
W. Smelik (eds.), The Literary and Linguistic Context of the Zohar (in preparation) and see
the works of Ronit Meroz about the Byzantine context of early Jewish mysticism, as for
example R. Meroz, “Der Afbau des Buches Sohar,” PaRDeS: Zeitschrift der Vereinigung für
Jüdischen Studien 2 (2005) 16–36.
Was Rabbi Aqiva a Martyr? Palestinian and
Babylonian Influences in the Development
of a Legend
Paul Mandel

1 Introduction: Martyrdom among Jews and Christians

in Late Antiquity

The persecution of the Jews of Palestine during the Hadrianic period, both
before and after the Bar Kokhba Revolt of the years 132–135 CE, is well docu-
mented. The attempt of the Roman government to coerce Jews to abandon
their laws and customs resulted in many Jews being tortured and killed for
obeying these laws; among the forbidden practices were circumcision, the
reading and teaching of the written and oral laws, and the recital of the Shema.1
During this same period, Christians throughout the Roman Empire were
often persecuted and arrested by the Roman authorities and sometimes exe-
cuted after confessing their faith. Attested already in the first century, such
executions continued in various places within the empire until the conver-
sion of Rome to Christianity in the early fourth century. Admittedly, persecu-
tion of Christians on a wide scale did not occur until the middle of the third
century under the Emperor Decius, with a second wave of widespread per-
secutions in the days of Diocletian, at the beginning of the fourth century.
However, it is in the middle of the second century, notably in the account of

1 See S. Lieberman, “The Martyrs of Caesarea,” Annuaire de l’Institut de Philologie et d’Histoire

Orientales et Slaves 7 (1934–4) 395–446, and esp. 423–32; M. D. Herr, “Persecutions and
Martyrdom in Hadrian’s Days,” in: D. Asheri and I. Shatzman (eds.), Studies in History (Scripta
Hierosolymitana 23; 1972) 85–125, see esp. 103ff.; S. Safrai, “Martyrdom in the Teachings of
the Tannaim,” in: T.C. de Kruijff and H. van der Sandt (eds.), Sjaloom: Ter nagedachtenis
van Mgr. Dr. A.C. Ramselaar (Arnhem 1983) 145–64; S. Lieberman, “redifat dat Yisrael,” in:
S. Lieberman (ed.), Salo Baron Jubilee Volume (Jerusalem 1976) 234–45 [Hebrew]. For an alter-
nate view, see P. Schäfer, Der Bar Kokhba-Aufstand: Studien zum zweiten jüdischen Krieg gegen
Rom (Tübingen 1981) 194–235; idem, “The Bar Kokhba Revolt and Circumcision: Historical
Evidence and Modern Apologetics,” in: A. Oppenheimer (ed.), Jüdische Geschichte in helle-
nistisch-römischer Zeit (Munich 1999) 119–32. See also the more recent study by R. Kalmin,
“Rabbinic Traditions about Roman Persecutions of the Jews: A Reconsideration,” Journal
of Jewish Studies 54 (2003) 21–50 (also printed in idem, Jewish Babylonia Between Persia and
Roman Palestine [Oxford 2006] 19–36).

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004277311_�12

WAS RABBI AQIVA A MARTYR ? 307

the martyrdom of Polycarp, that the word martyr, Greek for “witness,” began
to be used to describe one, specifically a Christian, who accepts death, usu-
ally a violent death, at the hands of the authorities, rather than comply with
demands counter to his or her religious beliefs. Thus begins the history of a
literary genre, in which the trial and execution of a Christian is described in
detail for the express purpose of celebrating his or her actions, which are seen
to provide a “testimony” or “witness” to the faith. Common to many marty-
rological accounts are the description of the arrest and trial of the martyr,
sometimes preceded by a description of the coercive decree, with an emphasis
on the interrogation of the defendant and the confrontation with the Roman
judge, a detailed list of tortures, and finally a description of the execution itself.
A central element in these texts is the active desire of Christian martyrs for
violent death as ultimate proof of their testimony of faith and as a goal, the ful-
fillment of which will bring them into the presence of God. Since the purpose
of the narrative is to show the martyr’s actions as an exemplum for the faith,
the martyr’s words under interrogation and torture are significant for lauding
his (or her) heroism and the steadfastness of the exemplified faith, and the
accounts often conclude with words testifying to the subsequent blessings
accorded the martyr or martyrs in heaven.2
The professors of the faith in Jesus as the Christ were convicted upon a
simple identification with the essence of “Christianity”; thus, the very words
“Christianus sum” (“I am a Christian”) became the cause for their conviction. In
contrast, Judaism was not an outlawed religion per se under Roman law; and,
especially after the Hadrianic persecutions, only those aspects of Jewish cul-
ture which were considered to represent a continued rebellion against Roman
rule were proscribed, and, at least formally, became reason for arrest, torture
and punishment.3

2 See the concise summary by J. W. van Henten, “Jewish and Christian Martyrs,” in: M. Poorthuis
and J. Schwartz (eds.), Saints and Role Models in Judaism and Christianity (Leiden and Boston
2004) 163–81; see also T. Baumeister, Die Anfänge der Theologie des Martyriums (Münster
1980) esp. 229–306; G. Bowersock, Martyrdom and Rome (Cambridge MA 1995); E. A. Castelli,
Martyrdom and Memory: Early Christian Culture Making (New York 2004).
3 Cf. S. Lieberman, “Palestine in the Third and Fourth Centuries,” Jewish Quarterly Review 36
(1945–6) 329–44, and esp. 341–2 (concerning the persecutions of Christians and Jews during
the third and fourth centuries CE: “There is absolutely no direct reference in the entire rab-
binic literature of the third and fourth centuries from which we might legitimately conclude
that the Roman government deliberately persecuted the Jewish religion during this time,”
342); cf. idem, “The Martyrs of Caesarea,” 416, and 428: “And herein lies the main difference
between the persecution of the Jews at that time [third and fourth centuries CE] and that of
308 mandel

2 Accounts of Rabbi Aqiva’s Death: Babylonian and Palestinian

Traditions in the Context of mBer 9:5 and Deut 6:5

With this in mind, the present study will discuss the accounts surrounding one
who is often regarded as the most famous Jewish martyr of the rabbinic period,
Rabbi Aqiva. The tales analyzed herein appear in the Bavli to Tractate Berakhot,
and in the Yerushalmi to Tractates Berakhot and Sotah; it is important to view
the passages in their redactional-literary contexts.4 In both talmudic passages
in Berakhot the narrative is appended to mBer 9:5, which prescribes praise
(“blessing”) of God in adverse as well as beneficial circumstances:

,‫חייב לברך על הרעה כשם שהוא מברך על הטובה‬

)‫ ״ואהבת את יי״ אלהיך בכל לבבך ובכל נפשך ובכל מאודיך״ (דברים ו ה‬,‫שנאמר‬
;‫ ביצר טוב וביצר רע‬,‫״בכל לבבך״—בשני יצריך‬
;‫״בכל נפשך״—אפילו הוא נוטל את נפשך‬
.‫״בכל מאודיך״—בכל ממוניך‬
‫ בכל הווי מודה לו בכל‬,‫ ״בכל מאודיך״—בכל מידה ומידה שהוא מודד לך‬:‫דבר אחר‬
.‫מאד מאד‬

One is bound to bless [God] for evil even as he blesses [God] for the good,
as it is written: “And you shall love the Lord your God with all your heart
and with all your soul and with all your might” (Deut 6:5).

the Christians. To be a Jew was not a crime.” See also A. M. Rabello, “On the Relation Between
Diocletian and the Jews,” Journal of Jewish Studies 35 (1984) 147–67.
4 The tale has been treated numerous times in scholarly literature. The following are relevant
references for the present discussion: E. E. Urbach, The Sages: Their Concepts and Beliefs
(transl. by I. Abrahams; Jerusalem 1975) 443ff.; Lieberman, “The Martyrs of Caesarea,” 420–2;
idem, “redifat dat Yisrael,” 234–45, and esp. 222–8; E. P. Sanders, “R. Akiba’s View of Suffering,”
Jewish Quarterly Review 63 (1972–3) 332–51; Safrai, “Martyrdom and the Teachings of the
Tannaim,” 156–7; A. Z. Goldberg, “Das Martyrium des Rabbi Aqiva: Zur Komposition einer
Märtyrererzählung (bBer 61b),” Frankfurter judaistische Beiträge 12 (1984) 1–82; M. Fishbane,
The Kiss of God: Spiritual and Mystical Death in Judaism (Seattle and London 1994): “The
Sanctification of God in Love,” 66–71; D. Boyarin, “ ‘Language Inscribed by History on the
Bodies of Living Beings’: Midrash and Martyrdom,” Representations 25 (1989) 139–51, later
appearing as chapter 8 in idem, Intertextuality and the Reading of Midrash (Bloomington
and Indianapolis 1990) 117–29; A. Tropper, Like Clay in the Hands of the Potter: Sage Stories in
Rabbinic Literature (Jerusalem 2011) Chapter 5 (“From Halakhah to Aggadah: The Formation
of Rabbi Aqiva’s Martyrdom Narrative”) 111–54, and cf. Excursus B (“On the Story of Rabbi
Aqiva and Tyranus Rufus and the Story of Rabbi Judah and the Roman Matron”) 208–11
[Hebrew]. On Boyarin’s transcription and interpretation of the texts of the tales in the
Yerushalmi and the Bavli, see Appendix III.
WAS RABBI AQIVA A MARTYR ? 309

“with all your heart”—with both your impulses, the good impulse and the
evil impulse;
“and with all your soul”—even if He takes away your soul;
“and with all your might”—with all your wealth.
Another explanation:
“with all your might (me’odecha)”—for every measure (mida) he metes
(moded) out to you, for all give Him thanks (modeh) exceedingly
(me’od me’od).

Since more than one proof seems to be offered for the injunction to praise God
under both good and bad circumstances, it is unclear which proof (if only one)
is intended: whether it is the injunction to love (and consequently praise) God
“with [one’s] evil inclination” as well as one’s “good inclination” (glossing the
words bekhol levavekha); or the injunction to praise God “even as He takes your
soul” (glossing bekhol nafshekha); or, as in the final comment in the mishnah
(i.e. davar aher), the demand to praise and thank God “for every measure He
metes out to you” (glossing bekhol me’odekha, and presuming that there are
different “measures” meted out to individuals under various circumstances—
some considered a “measure of goodness” [midat hatov], while others a “mea-
sure of retribution [or punishment]” [midat pur‘anut]5). In tannaitic parallels
the final midrashic comment is not found together with the previous ones;6 it
is only this latter one that is expressed in the name of Rabbi Aqiva, whereas the
teachings including love of God “in death” are attested in the name of Rabbi
Aqiva’s disciple, Rabbi Meir, especially in contexts referring to suffering and

5 On these “midot,” see the parallel passages cited in the following note.
6 The final midrashic comment cited in the mishnah has parallels in the tannaitic midrash to
Deuteronomy (SifDeut 32; cf. also MdRY bahodesh 10) in the name of Rabbi Aqiva and in the
context of good and bad events (‫)בין במידת הטוב ובין במדת פורענות‬, and similarly appears in
tBer 6:1 in the name of Rabbi Aqiva’s student, Rabbi Meir (but see Lieberman, Tosefta kifshuta
vol. 1, 102, to line 8, on the order of the Tosefta passages and their relationship to the Mishnah
text); the other expansions on the verse, including the injunction to love God “until / upon
death,” also appear in the tannaitic midrashim (as well as in tBer 6:7), but in neither parallel
are they connected to the name of Rabbi Aqiva (in the Tosefta they are cited in the name
of Rabbi Meir). Rabbi Aqiva’s teaching to love God under all circumstances is in keeping
with a similar statement repeated in his name, that of the “preciousness” of suffering (‫חביבין‬
‫ ;)יסורים‬see MdRY bahodesh, loc. cit., and SifDeut, loc. cit., and cf. GenR 33:1. On this idea and
its relationship to Rabbi Aqiva’s thought, see E. E. Urbach, “Ascesis and Suffering in Talmudic
and Midrashic Sources,” in: S. Baron et al. (eds.), Yitzhak F. Baer Jubilee Volume (Jerusalem
1960) 60–1 [Hebrew]; idem, The Sages, 443–6; and cf. Sanders, “R. Akiba’s View of Suffering,”
332–51.
310 mandel

deaths encountered by Jews during the Hadrianic persecutions and the Bar
Kokhba Rebellion.7 It would thus seem that the authenticity of the tannaitic
citation of Rabbi Aqiva as the author of the midrashic comment on bekhol
nafshekha (“even in death”) cited in the Bavli’s discussion of this mishnah, both
as an independent saying introducing the pericope and as part of the tale con-
cerning his execution, is suspect: it is probable that this attestation has been
influenced by the interpretation of Deut 6:5 in the mishnah.8
In the following discussion I will evaluate and compare the two parallel
narratives surrounding Rabbi Aqiva’s presumed martyrdom, as represented in
the accounts of the Bavli and the Yerushalmi, in the context of the genre of
martyrology of the period. In particular, I will investigate how Rabbi Aqiva’s

7 The midrashic expansion of ‫בכל נפשך‬, enjoining love of God “even as he takes your soul,” is
followed by the verse from Ps 44:23, ‫כי עליך הורגנו כל היום נחשבנו כצאן טבחה‬, in tBer 6:7 and
SifDeut 32; the latter verse (and its adjacent verses, esp. 21–22) appears elsewhere in reference
to the Hadrianic persecutions and the Bar Kokhba revolt: cf. MdRY shirta 3; SifZutDeut on
Deut 6:5 (M. I. Kahana, Sifre Zuta on Deuteronomy: Citations from a New Tannaitic Midrash
[Jerusalem 2002] 147; see below on this passage). While nowhere except in the Bavli (see
below) is Rabbi Aqiva cited as the author of the midrashic comment on Deut 6:5 regarding
the love of God at the moment of death, the verse is cited in the name of his students in the
context of the Hadrianic persecutions; the comment connecting the commandment to love
God “with all one’s soul” (‫ )בכל נפשך‬to the moment of God’s “taking one’s soul” may thus
have originated as a result of these persecutions in the wake of the Bar Kokhba revolt. See
further on “giving one’s soul” and on the use of the verses from Ps 44 below.
8 The attestation occurs in bBer 61b appended to this mishnah, in a passage introduced as a
tannaitic source (‫ ;)תניא‬the beginning of that passage, Rabbi Eliezer’s discussion of Deut
6:5, is indeed found in the tannaitic midrash SifDeut 32, followed by Rabbi Aqiva’s teach-
ing, mentioned above, concerning the phrase bekhol meodekha of Deut 6:5. In the Bavli,
Rabbi Aqiva’s teaching is equivalent to that found in the Mishnah on the phrase bekhol
nafshekha. This passage is then followed by another, also marked as a tannaitic source (‫תנו‬
‫)רבנן‬, which includes the narrative (in Hebrew) of Rabbi Aqiva’s arrest along with Pappos
ben Yehudah and the tale concerning Rabbi Aqiva’s subsequent execution, the subject
of the present paper. A comparison of the entire pericope and the language paralleled in
other sources has been treated by Tropper, Like Clay in the Hands of the Potter, 111–54, 208–11.
Tropper notes the many literary influences on the separate scenes, among which are also
later amoraic sources. Therefore, he concludes that these sections seem to be examples of a
“Babylonian baraita,” created by Babylonian authors of the amoraic (or post-amoraic) era in
the style of a tannaitic source; see also Lieberman, “The Martyrs of Caesarea,” 421, n. 34 (“The
source [= the tale of Rabbi Aqiva’s death in the Bavli] seems to be a later ‘Baraitha’ ”); and
see on later, Babylonian baraitot: S. Friedman, “Towards a Characterization of Babylonian
Baraitot: ‘ben Tema’ and ‘ben Dortai’,” in: Y. Elman, E. B. Halivni, and Z. A. Steinfeld (eds.),
Neti‘ot LeDavid: Jubilee Volume for David Weiss Halivni (Jerusalem 2004) 195–274, and esp.
201–3 [Hebrew]. Cf. Appendix II below.
WAS RABBI AQIVA A MARTYR ? 311

understanding of the verses of the Shema, prescribing the love of God “with
all your soul” relates to his impending death in each of these narratives. It will
be important to determine, as precisely as possible, the original texts of each
work; a comparison of the texts of the editio princeps of both talmudim with
the manuscript evidence will lead to an analysis of the transformations of
this tale; this, in turn, will reveal contrasting perspectives that can be related
to geographic and chronological contexts—from Palestine to Babylonia and
from late antiquity through the Byzantine and medieval periods.

3 The Palestinian Tradition: Rabbi Aqiva’s Trial as Political Drama

The following is the text of the Aqiva narrative as it appears in the Yerushalmi
to the above mishnah, yBer 9:5, 14b, and in ySot 5:7, 20c.9 The text is given

9 The narrative is embedded in an extended pericope in the Yerushalmi (appearing almost

identically in both Berakhot and Sotah) discussing the dual and sometimes contradictory
approaches of love and fear of God. This pericope appears in ySot 5:7, 20c, appended to the
discussion in the mishnah there regarding Job’s relationship to God, understood to have
been motivated by either love or fear. The pericope, which may originally have had an inde-
pendent existence (cf. a similar discussion of the love and fear of God at the beginning of
SifDeut 32, which may have provided a foundation for the development of the pericope),
exhibits an inner structure governed by these two aspects: it is introduced by a comment
contrasting two verses, “You shall love God . . .” (Deut 6:5), and “You shall fear the Lord your
God and worship Him” (Deut 6:13) [this introduction is missing in the passage in Tractate
Berakhot, but was probably deleted as a result of scribal error], and it ends with two sto-
ries regarding Rabbi Aqiva’s understanding of each of these verses: the narrative concern-
ing his trial before Tineius Rufus presents a paradox where love of God, exemplified by the
reciting of the Shema (and in particular verse Deut 6:5) at the time of trial and torture, is
experienced at the moment most conducive to fear, whereas in the second anecdote Rabbi
Aqiva responds to the query of his student, Nehemiah the Amsonite, concerning the verse
in Deut 6:13, where the particle et preceding the name of God, regularly expounded by Rabbi
Aqiva as implying some “addition” (which in the present context would be presumed blas-
phemous), is expounded to include the fear (reverence) of God’s Torah in addition to rever-
ence for God Himself. Thus, the entire pericope in the Yerushalmi, including the narrative
concerning Rabbi Aqiva, would seem to be naturally and primarily connected to that section
of the mishnah in Berakhot containing the principle expressing the obligation to praise God
“in all measures” (the final gloss to bekhol meodekha) including both the love and fear of
God, rather than to the section containing the midrashic comment to the words of Deut 6:5,
bekhol nafshekha. The appearance of the midrashic comment in the mishnah of Berakhot to
‫ בכל נפשך‬may have provided additional motivation for the inclusion of this pericope in the
Yerushalmi to this mishnah, since, of course, the recital of this verse is the central issue of the
Yerushalmi tale of Rabbi Aqiva; in the Bavli this latter connection seems primary.
312 mandel

according to the version appearing in Tractate Berakhot of the Leiden man-

uscript (which served as the basis for the editio princeps, Venice 1522) with
minor corrections based on parallel texts; this version includes two lines (sur-
rounded here by parentheses) that are clearly later interpolations: these lines
appear only in the Leiden manuscript to Berakhot and the first printed edition
based on it (Venice 1522), and not in the parallel passage in ySot; and they are in
Hebrew in what is otherwise a narrative purely in Palestinian Aramaic. While
clearly spurious, the inclusion of these passages here is relevant to the discus-
sion below of the development, reception and transmission of this tale.10

yBer 9:5 (Ven. ed.) 14b / ySot 5:7 (Ven. ed.) 20c
.‫ר׳ עקיבה הוה קיים מיתדין קומי טורנוסרופוס הרשע‬
.‫אתת ענתה דקרית שמע; שרי קרי וגחיך‬
!‫ סבא סבא! או חרש את או מבעט בייסורין את‬:‫אמר ליה‬
,‫ תיפח רוחיה דההוא גברא! לא חרש אנא ולא מבעט בייסורין אנא‬:‫אמר ליה‬
:‫אלא כל יומיי הייתי קורא את הפסוק הזה‬
)‫ אימתי יבואו שלשתן לידי‬:‫(והייתי מצטער ואו׳‬
.)‫״ואהבת את י״י אלהיך בכל לבבך ובכל נפשך ובכל מאודך״ (דברים ו ה‬
;‫רחמתיה בכל לבי; רחמתיה בכל ממוני‬
.‫ו״בכל נפשי״ לא הוות בדיקה לי‬
,‫ ״בכל נפשי״‬11]‫וכדון דמטת [לי‬

10 The usual paucity of evidence for the text of the Yerushalmi is here alleviated by the
existence of the parallel in Tractate Sotah. In addition, both Berakhot and Sotah tractates
are attested in an additional manuscript (MS Vat. ebr. 133), and the narrative is cited twice
in medieval collections: it is cited by the Tosafot to bSot 31a (incipit ‫ ;)גדול מצווה ועושה‬and
appears also in the Qundras Aharon (remez 246), a compilation of aggadic passages culled
from the Yerushalmi appended to the first printed edition of the Yalqut Shimoni (Salonica
1526–7; see L. Ginzburg, Seridei Yerushalmi [Jerusalem 1969] 342). All the versions are
very close to each other except for the two interpolated sentences, and consequently it
is possible to construct one version of the narrative. All the texts have been published by
P. Schäfer and H.-J. Becker, Synopse zum Talmud Yerushalmi. Band I/1–2: Ordnung Zera’im:
Berakhot und Pe’a (Tübingen 1991), 250–1, and Band III: Ordnung Nashim (Tübingen 1998),
113–4. (It should be noted that two additional manuscripts included in Band I which
provide attestation to the passage in Tractate Berakhot, MS Paris Heb. 1389 and MS London
BL Or. 2822 [cat. Margoliouth, no. 403], containing the commentary of R. Solomon Serilio
to the Yerushalmi [mid-16th century], were composed after the appearance of the printed
editions to the Yerushalmi and are wholly dependent upon that text; thus, these do not
provide independent evidence of the Yerushalmi text.)
11 The word ‫“( לי‬for me”) appears in all versions of Tractate Sotah, but is missing in the
parallel passage in Berakhot. There is little difference between the two variants; see,
however, n. 18 below.
WAS RABBI AQIVA A MARTYR ? 313

—‫ואתת ענתה דקרית שמע ולא אפלגית עליה‬

.‫בגין כן אנא קרי וגחיך‬
)‫(לא הספיק לומר עד שפרחה נשמתו‬

Rabbi Aqiva was being tried before the wicked Tineius Rufus.
The time came for the reciting of the Shema; he began to read and laugh.
He [Tineius Rufus] said to him: “Old man! Old man!
Either you are a sorcerer or you scorn sufferings!”
He [Rabbi Aqiva] said to him: “May you perish!
I am neither a sorcerer nor do I scorn sufferings.
Rather, all my life I have recited this verse:
(and I was troubled, saying, When will these three come to my reach?)
“And you shall love the Lord your God with all your heart and with all
your soul
and with all your might” (Deut 6:5)—
I have loved Him with all my heart; I have loved him with all my money;
but “with all my soul” I have not been tested (lit.: has not been tested for/
to me).
And now that “all my soul” has arrived [for me],
and the time for the reciting of the Shema has come and I did not desist
from it—
for this reason do I read and laugh.”
(He did not have a chance to say [the Shema] until his soul expired,)

This version depicts a confrontation occurring at a trial of Rabbi Aqiva before

the Roman governor of Judaea, Tineius Rufus.12 Tineius Rufus’ mention of

12 On (Q.) Tineius Rufus as the consular governor of Judaea from as early as 130 CE, see
E. Schürer, The History of the Jewish People in the Age of Jesus Christ (175 B.C.–A.D. 135)
vol. 1 (ed. G. Vermes, F. Millar, M. Black; Edinburgh 1973) 518, 547–9, and n. 153; W. Eck,
“The Bar Kokhba Revolt: The Roman Point of View,” Journal of Roman Studies 89 (1999) 79,
n. 18; H. M. Cotton and W. Eck, “Governors and their Personnel on Latin Inscriptions from
Caesarea Maritima,” The Israel Academy of Sciences and Humanities, Proceedings 7 (2001)
235–8, and esp. 237, n. 78. The form of the name of this governor as found in rabbinic
sources—Turnusrufus, or, as sometimes occurs, in two words: Turnus Rufus, but also:
Tunus Trufus—may reflect the local misconstruction of the Latin name Tineius as the
(Greek) word tyrannos = “ruler” (sometimes, but not necessarily, indicating a despotic
ruler), as an introductory title of the governor; see H. G. Liddell, R. Scott, and H. S. Jones,
A Greek–English Lexicon (Oxford 1996) 1836, s.v. τυραννος; M. Sokoloff, A Dictionary of
Jewish Palestinian Aramaic (Ramat Gan 20022) 224, s.v. ‫טירנוס‬, 224, and in “Addenda et
Corrigenda,” 833. On other rabbinic passages describing disputations between Tineius
314 mandel

“sufferings” (‫ )יסורים‬implies that some form of torture was used during the trial.13
Indeed, since these trials were usually public, with the express intent of mak-
ing a dramatic spectacle of the process and of the defendant’s punishment,14 it
was important that torture have a visible effect on the accused. Tineius Rufus
is enraged by Rabbi Aqiva’s murmurings, which he interprets as either some
type of sorcery (itself a crime, and especially prosecuted in the first centuries
of Imperial Rome15) intended to ward off the effect of the tortures, or perhaps
as a reflection of a Stoic denial of the emotion of fear.16 Rabbi Aqiva’s retort,
in describing the propitious and happy “coincidence” of the prescribed time

Rufus and Rabbi Aqiva, see below. See bTaan 29a, where Tineius Rufus is mentioned as
the one who destroyed the Temple (or “ploughed the city”).
13 While the verb ‫ מיתדין‬literally means “being tried” (see Sokoloff, Dictionary of Jewish
Palestinian Aramaic, s.v. ‫דון‬, 141, see the previous note, and “Addenda,” 830, where he
supplants his earlier translation of this passage, “being judged,” with “[being] tortured”;
but see below), it is commonly found in contemporary rabbinic Palestinian Aramaic
sources to denote the torture of defendants in Roman trials (quaestio per tormentum—
investigation through torture; see Brill’s New Pauly: Encyclopedia of the Ancient World,
vol. 1 [Leiden 2008] 319–20, s.v. “quaestio per tormentum,” and vol. 14 [2009] 795–6,
s.v. “Torture”), a prevalent practice that provided an important source of evidence in
Roman criminal proceedings, even in some instances, for Roman citizens, and all the
more so during the investigation of slaves and non-Romans. See S. Lieberman, “Roman
Legal Institutions in Early Rabbinics and in the Acta Martyrum,” Jewish Quarterly Review
35 (1944–5) 15, n. 99.
14 See D. Potter, “Martyrdom as Spectacle,” in: R. Scodel (ed.), Theatre and Society in the
Classical World (Ann Arbor 1993) 53–88, who deals specifically with the trials and
executions of Christian martyrs in the early Roman Imperial period.
15 See Lieberman, “The Martyrs of Caesarea,” 420, n. 25: “The Christian martyrs were also
accused of defying torture by means of sorcery.”
16 On the Stoic claim for the elimination of the passions (Latin: affectus), of which fear is
one of the four main passions (the others being desire, pleasure and displeasure), and
the ideal for the Stoic sage as one completely free from its effects, see Brill’s New Pauly:
Encyclopedia of the Ancient World, vol. 13, s.v. “Stoicism,” 856–7. Clearly a defendant who
has mastered his passions to the extent that he can deny fear and cancel the feelings
associated with torture (‫)מבעט בייסורין‬, or one who, through sorcery, can cancel the
effect of torture, would raise the ire of a judge, who depends upon the administration
of torture both for extracting admission of guilt and/or for the public effect of the trial.
See Potter, “Martyrdom as Spectacle,” 63: “Such hearings [large public trials of martyrs]
were moments of the highest drama, moments which provided the defendant with a
vital and exciting forum for self-display. He might even defeat his persecutors, or, in the
words of the Carthaginian confessors of 250, to become the judge of his judge. If a person
persisted in denial, even under torture, he or she was a victor in the struggle, and indeed,
the treatment of trials in the literature of the high empire reflects this.”
WAS RABBI AQIVA A MARTYR ? 315

of reciting the verse demanding love of God at the very occasion of ‫בכל נפשך‬,
presents what seems to be a bold reproach to the authority of the judge.17
While the narrative as it appears in yBerakhot (Leiden manuscript and
printed edition) includes the final statement (in Hebrew) telling of Rabbi
Aqiva’s death, there is no doubt that this is not part of the original Yerushalmi’s
version, which, in fact, is not a martyrological story at all: Rabbi Aqiva is being
tried and tortured, but nowhere is it hinted that he knows that he is going to
die, or that he will be or has been executed. To be sure, Rabbi Aqiva mentions
a “trial” or “test” concerning bekhol nafeshi: “with ‘all my soul’ have I not been
tested; and now that ‘all my soul’ has arrived [for me] . . .”18 However, nowhere
in tannaitic or early Palestinian amoraic texts does the term nefesh, and in

17 The tale not only depicts Rabbi Aqiva exhibiting utter contempt for the governor (“May
you perish!”), but moreover, the very recital of the Shema had most probably been
outlawed precisely because of its supposed anti-imperial nature (crimen laesae maiestas);
see Lieberman, “Martyrs of Caesarea,” 425: “The first verse of the Shema . . . sounded like a
direct challenge to the ambitions of Hadrian . . . It is quite reasonable to assume that the
prohibition to recite the Shema was the precursor of the other [Hadrianic] restrictions.”
It should also be noted that the normative time for the recital of the Shema was in the
early morning hours just before sunrise, or, at the latest, during the early hours of the
morning (mBer 1:2); the time for the beginning of criminal proceedings under Roman
law was normally not before daybreak, but this may have been extended under special
circumstances (cf. T. Mommsen, Römisches Strafrecht [Leipzig 1899] 364–5, nn. 5–6; apud
Lieberman, “redifat dat Yisrael,” 233, n. 63).
18 The verb b-d-q in Palestinian Aramaic refers to an investigation, inspection or test; see
Sokoloff, Dictionary of Palestinian Jewish Aramaic, s.v., pp. 85–6. In using the passive
participial form of the verb here (“[the love of God] ‘with all my soul’ has not been tried
for/to me”), Rabbi Aqiva may mean that he was “not sure” of this type of love of God
until now, or that he has not “been tried” regarding this type of love (although note that
in any case, the subject of the passive participle here, bediqah, is the scriptural phrase
and not Aqiva himself). There is a difference in nuance between the two senses: the
latter may be closer to an expression of martyrdom than the former. However, since the
phrase bekhol nafshi in tannaitic and Palestinian amoraic texts does not necessarily—
and, indeed, usually does not—imply death (see next notes), the second sense, of “being
tried” concerning love of God at risk of death, is almost identical to the first sense.
Concerning the phrase “now that ‘all my soul’ has arrived [for me]” (‫כדון דמטת [לי] בכל‬
‫)נפשי‬, it is significant that the verb ‫ מטי‬is used in Palestinian Aramaic texts to connote the
“arriving” at a particular passage in one’s biblical reading (see M. Sokoloff, A Dictionary
of Palestinian Jewish Aramaic, 302); thus, this locution here may be a double allusion to
Rabbi Aqiva’s actual recital of the text, “arriving” at the phrase ‫בכל נפשי‬, and to the fact
of the content of the recited text “befalling” him. The other added Hebrew phrase (“and I
was troubled, etc.”) originated, no doubt, as a copy of the similar phrase appearing also in
variant traditions of the Bavli tale; see below.
316 mandel

particular, the idea of “giving one’s nefesh,” imply a certainty of impending

death; rather, it is used consistently to express the extreme effort expended by
an individual in the fulfillment of a specific act at the risk of personal injury
but not necessarily death. This is the only meaning of this phrase as it appears
in the Hebrew text of Ben Sira.19 In the tannaitic midrashim Moses is consid-
ered as one who has “given his soul for the Torah” (‫ )נתן נפשו על התורה‬as well
as for Israel and for the institution of judges, meaning that he had encoun-
tered dangers and risked his life while acting on behalf of these.20 Similarly,
Judah is extolled in an early midrashic comment for having “given his soul,” i.e.
expended great effort and risked his life, to protect his brothers during famine.21

19 See Ben Sira 7:20 (supported by the Hebrew Genizah manuscripts of the book [MS A and
MS C]—see P. C. Beentjes, The Book of Ben Sira in Hebrew: A Text Edition of All Extant
Hebrew Manuscripts and a Synopsis of All Parallel Hebrew Ben Sira Texts [Leiden 1997]),
and see also 39:1, 51:51; cf. 30:21.
20 See MdRY shirta 1; SifDeut 306: ‫ שמא אין אתם יודעים כמה צער‬:‫אמר להם משה לישראל‬
‫נכנסתי לבין המלאכים‬ . . . ‫נצטערתי על התורה וכמה עמל עמלתי וכמה יגיעה יגעתי בה‬
‫ונכנסתי לבין החיות ונכנסתי לבין השרפים שאחד מהם יכול לשרוף כל העולם כולו על‬
‫ דמי נתתי עליה‬,‫נפשי נתתי עליה‬ . . . ‫יושביו‬. (cf. DeutR ed. Lieberman, eqev 17, p. 91); and
cf. yPea 1:1, 15b; yShab 1:7, 3d and parallels: ‫שכל דבר שבית דין נותנין נפשן עליו סוף הוא‬
‫—מתקיים‬where the intent is to work hard and intensively. Cf. also GenR 82:8.
21
GenR 93:9. See also tBer 6:7, where Ben Azzai interprets Deut 6:5 in a similar fashion:
‫ בכל נפשך—תן נפשך על המצוות‬:‫בן עזיי אומר‬, and see S. Lieberman, Tosefta kifshuta vol. 1,
111, to line 40: ‫״תן נפשך״ כאן אינה אלא מליצה‬, and see his references there. This midrashic
comment of Ben Azzai on Deut 6:5 is found also in SifDeut 32 in slightly different form:
‫ ״בכל נפשך״—אהבהו עד מצוי נפש‬:‫ ;שמען בן עזיי אומר‬the unique phrase ‫מצוי נפש‬, “the
wringing [/ squeezing / draining] of the soul” (see E. Ben Yehuda, A Complete Dictionary
of Ancient and Modern Hebrew, s.v. ‫מצה‬, 3235–7), refers indeed to a call for personal
sacrifice in the sense of physical and/or spiritual hardship and risk, but falls short of
implying a necessary sacrifice of one’s life. The attestation of this comment to Ben Azzai,
a contemporary of Rabbi Aqiva who may not have outlived the Bar Kokhba war (he is
listed in LamR 2:2, along with Rabbi Aqiva, as one of ten sages killed, apparently, during
the Bar Kokhba revolt), would thus accurately reflect the prevalent attitude of his day, as
reflected in the attested sayings of Aqiva himself (see above), of the need for undergoing
personal hardship in order to continue observing God’s commandments, but should not
be understood to have already included a martyrological view “celebrating” the sacrifice
of life, whether actively sought after or imposed by an outside force. It is significant that
it is precisely in later contexts mentioning the sacrificing of life during the Hadrianic
persecutions (‫ )דורו של שמד‬that the idea of “hardship” and personal risk becomes
coupled with a testament to those who were murdered in the course of observing God’s
commandments; thus, as mentioned above, the comment to Deut 6:5, ‫״בכל נפשך״—אפילו‬
‫הוא נוטל את נפשך‬, appears in tannaitic texts other than mBer 9:6 only and always in
tandem with the verse Ps 44:23 (“for your sake we are killed all day long; we are accounted
WAS RABBI AQIVA A MARTYR ? 317

It thus becomes clear that this version does not imply that Rabbi Aqiva was
executed; rather, the narrative depicts one of a series of scenes included in
Palestinian rabbinic texts describing a confrontation between Rabbi Aqiva and
Tineius Rufus.22 What is significant here for our narrator is the fact that Rabbi

as sheep for the slaughter”), which became a standard text describing those murdered
during the Hadrianic persecutions: see tBer 6:7 and SifDeut 32 (cf. also MdRY beshalah
3 on Exod 15:2). It would seem that this coupling reflects a later development of the
meaning of the phrase ‫לתת את הנפש‬, so that it became commonly used in the context of
giving one’s life: see the later amoraic midrash, PRK beshalah 14: ‫ רבון‬:‫אמר דוד לפני הקב״ה‬
‫העולמים! תשוח נפשי על אילו שהן עתידין ליתן את נפשן על קידוש השם; ואיזה זה? זה דורו‬
‫( של שמד‬and cf. there, loc. cit.: ‫אם יאמר לך אדם תן נפשך על קידוש השם‬, clearly referring
to the sacrifice of one’s life [see the context there]). See also the apparently late reworking
of Ben Azzai’s comment in SifZutDeut, ed. Kahana, p. 147: ‫ אהביהו‬:‫בכל נפשך—שנו חכמים‬
‫ ״אם שכחנו שם אלהים״ וג׳ ״הלא אלהים‬:‫ שנ׳‬,‫כנפשך ותן נפשך עליו; וכן עשו בני ישראל‬
)‫ ״כי עליך הורגנו כל היום״ (תהלים מד כא–כג‬:‫יחקור זאת״ וג׳ וכתי׳‬. Interestingly, the
two parallel phrases of Ben Azzai’s midrashic comment to Deut 6:5 appearing in (earlier)
tannaitic versions—in tBer: ‫תן נפשך על המצות‬, and in SifDeut: ‫—אהבהו עד מצוי נפש‬are
here conflated in such a way so that the phrase ‫לתת את הנפש‬, which in its original citation
retains the earlier meaning of personal hardship and risk (“for the commandments”),
is associated with (but not necessarily yet identical to) the sacrifice of life experienced
during the Hadrianic persecutions (“for Him” = God; as exemplified, again, by the use of
the verses there of Ps 44:21–23, and by explicit reference to the actions of the Jews [“so
did the Israelites . . .”]). In sum, the use of the phrase “to give one’s soul” in the context
of an active sacrifice of one’s life seems to be a late development: in all early texts the
basic ideal to be upheld is the love of God and the observance of His commandments
under great hardship and personal risk; even when contemplating the terrible fact of
the murder of Jews despite/because of their observance as happened in great numbers
during the Hadrianic persecutions, the love of God was not seen during this period as a
catalyst for active anticipation of and participation in martyrdom. And while later texts
suggest the use of the phrase ‫ בכל נפשי‬in the context of the ultimate sacrifice of life, there
is no evidence from the Yerushalmi narrative concerning Rabbi Aqiva that “‫ בכל נפשי‬has
arrived for me” implies a certainty on the part of Rabbi Aqiva of impending death; rather,
especially in the context of the tale which emphasizes his torture, it can just as well, and
perhaps better, be understood to refer to the undergoing of great hardship as a result of
his Jewish observance. Cf. also Safrai, “Martyrdom and the Teachings of the Tannaim,”
156–7.
22 Several rabbinic passages describe confrontations between Rabbi Aqiva and Tineius
Rufus: a (the?) trial of Rabbi Aqiva before Tineius Rufus is mentioned also in LamR 3:44,
while in GenR 11:5, with parallels in PesR 23 [ed. M. Ish-Shalom, 119b–120a], Tan ki tisa
33, and bSanh 65b, the two men engage in a battle of riddles (of which Rabbi Aqiva’s
retort, “What is a man among men?,” may be seen to be allusively provocative and
even potentially demeaning), resulting in a discussion of the value of the Sabbath; in
TanB tazria 7:7 the two debate the value of circumcision; and in bBB 10a Tineius Rufus
318 mandel

Aqiva was able to turn a simple act of reciting the Shema into political drama:
his amusement and joy at being able to perform this act at the very moment
that he is being tried becomes a weapon against the ruthless governor, in plac-
ing the love of God and His sovereignty above the honor due to Tineius Rufus
himself—precisely that which the governor is trying to subvert by creating his
act of drama through torture and inquisition.23

inquires of Rabbi Aqiva concerning the virtue of God’s allowing the existence of the poor.
Further tales concerning the wife of Tineius Rufus are found in bAZ 20a and bNed 50b
(see Rashi, ad loc.). See the discussion in Tropper, Like Clay in the Hands of the Potter,
142–5. While clearly these traditions have been variously shaped by the rabbinic redactors
(as have the talmudic stories discussed here), there is no reason to be overly skeptical
about the historical value of traditions concerning confrontations, perhaps some of a
theological/philosophical nature, between the Roman governor of the province and
a recognized public leader of the Jews, including a trial of Rabbi Aqiva. On the other
hand, there is no reason to connect these other dialogues with the trial; cf. Lieberman,
“The Martyrs of Caesarea,” 421, n. 29 (“There is, however, no hint in the sources that the
discussions took place during R. ‘Akiba’s imprisonment”). Later midrashim describe a
period of imprisonment of Rabbi Aqiva (in Caesarea? cf. Midrash Mishle 9:2, in some of
the manuscript versions; and see Midrash Shir Hashirim [ed. Grünhut], 5b [influenced
by the Midrash of the Ten Martyrs, see below]). An additional, elliptic reference to Jews
murdered by Tineius Rufus (in the context of the Bar Kokhba revolt?), who thereby
exhibited their love of God, appears in the midrash, SongZ to Songs 1:4 (Midrasch Suta
[ed. S. Buber; Berlin 1894], 13; also published as Aggadat Shir Hashirim [ed. S. Schechter,
Cambridge 1896], 15): ‫ הרבה‬:‫ על מי שהרגו טורנוסרופוס הרשע‬:‫יהושע בן יהונתן היה אומר‬
‫“( אהבו אותך יותר מן הצדיקים הראשונים—״מישרים אהבוך״‬Joshua ben Jonathan would
say about those who were killed by the wicked Tineius Rufus: Many have loved you more
than the former righteous ones, ‘uprightly do they love You’ [Song of Songs 1:4]”—the
midrashic comment based on the [creative] reading: me-yesharim ahevukha = “[more]
than the righteous [lit., ‘straight’] ones do they love You”). This citation, if early (the
text is dated by some as early as the third-fourth century CE, although the dating of the
midrash is contested; cf. G. Stemberger, Introduction to the Talmud and Midrash [trans.
and ed. M. Bockmuehl, Second edition, Edinburgh 1996], 319–20), may be connected with
the Yerushalmi narrative, implying that the trial described there indeed ended in Rabbi
Aqiva’s death; or, on the other hand, the name of Tineius Rufus may be used here only in a
general way as an appellation for the governor in charge during (part of) the Bar Kokhba
revolt.
23 See Potter, “Martyrdom as Spectacle.” On the theme of Rabbi Aqiva’s laughter, see below,
n. 57.
WAS RABBI AQIVA A MARTYR ? 319

4 The Babylonian Tradition: Rabbi Aqiva’s Execution as a

Teaching Experience

4.1 The Bavli Version and the Manuscript Evidence

The following is the text of the tale as it appears in the first printed editions of
the Bavli to Tractate Berakhot.24

bBer 61b
,‫בשעה שהוציאו ר׳ עקיבא להריגה זמן קריאת שמע היה‬
;‫והיו סורקין את בשרו במסרקות של ברזל‬
.‫והיה מקבל עליו עול מלכות שמים‬
?‫ עד כאן‬,‫ רבינו‬:‫אמרו לו תלמידיו‬
:‫ כל ימי הייתי מצטער על פסוק זה‬:‫אמר להם‬
.‫״בכל נפשך״ (דברים ו ה)—אפילו נוטל את נשמתך‬
?‫ מתי יבא לידי ואקיימנו‬:‫אמרתי‬
?‫ועכשיו שבא לידי—לא אקיימנו‬
.‫היה מאריך ב״אחד״ עד שיצתה נשמתו ב״אחד״‬
.‫ אשריך ר׳ עקיבא שיצתה נשמתך ב״אחד״‬:‫יצתה בת קול ואמרה‬
?‫ זו תורה וזו שכרה‬:‫אמרו מלאכי השרת לפני הקדוש ברוך הוא‬
.)‫״ממתים ידך י״י ממתים מחלד״ (תהלים יז יד‬
.)‫ ״חלקם בחיים״ (שם‬:‫אמר להם‬
.‫ אשרך ר׳ עקיבא שאתה מזומן לחיי העולם הבא‬:‫יצתה בת קול ואמרה‬

At the hour that they were bringing out Rabbi Aqiva for execution,
it was the time of the reciting of the Shema,
and they were flaying his flesh with iron combs;
and he was accepting upon himself the yoke of the heavenly kingdom.
His students said to him: Our master, so far?
He said to them: All my life I have been troubled over this verse:
“[You shall love the Lord your God with all your heart and] with all your
soul” (Deut 6:5)

24 The first (extant) printed edition of this tractate is the Soncino (Italian) edition of 1484
(the tractate was also printed in Spain in the fifteenth century, but our passage has not
survived from these early editions); the text of this tale in the Venice edition of 1520,
which served as the basis of all subsequent editions of the Bavli, is almost completely
identical with that of the Soncino edition, and, as is known, was copied from it. The text
offered here differs from that of the printed editions only in the filling out of some of the
abbreviations. The narrative appears in abbreviated form in a later edition of Tan ki tavo;
see TanB, ki tavo 4, 47.
320 mandel

—even if He takes your soul.

I said: When will [this] come to me, that I might fulfill it?
And now that it has come to me—shall I not fulfill it?
He drew out [the word] “One” until his soul expired with “One.”
A heavenly voice came forth and said:
Happy are you, Rabbi Aqiva, that your soul has expired with “One”!
The ministering angels said to the Holy One, blessed is He:
This is Torah and this is its reward?!
“They are killed from Your hand, O Lord! They are killed from this world!”
(Ps 17:14)
He said to them: “Their lot is with life” (ibid.).
A heavenly voice came forth and said:
Happy are you, Rabbi Aqiva; you are invited to the world to come!

For the narrative in the Bavli we fortuitously have a significant number of tex-
tual witnesses besides the version in the printed editions; these divide into
two families: one is represented by MS Oxford Bodleian, Opp. Add. fol. 23 (cat.
Neubauer, no. 366) and MS Paris 671.4—together these will be designated as
Version A; the other version is mainly represented by the text of MS Munich
95, and is supported by several other text witnesses—these, collectively, will
be designated Version B.25 I note that the Oxford and Paris manuscripts are

25 The texts comprising Version B are as follows: a fragment from the private collection of
Vittorio Vivanti (Rome) (microfilm roll number 31673 of the Institute for the Microfilm
of Hebrew Manuscripts of the National Library in Jerusalem; see on this private
collection, B. Richler, Guide to Hebrew Manuscript Collections [Jerusalem 1994] 261);
Genizah fragment Oxford, Bodleian Library, Heb. b 1/1 (cat. Neubauer, no. 2673.1) (both
of these fragments apparently derive from full manuscripts of Bavli Berakhot); Genizah
fragment Cambridge University Library T-S N.S. 291.76 (from a collection of aggadot); and
the citations of the text in the following medieval collections: Sefer Pitron Torah (ed. E.
E. Urbach [Jerusalem 1978] 292–3; the work is a collection of rabbinic texts, compiled
in Persia in approximately the 10th century); Menorat Hama’or of R. Isaac Aboab (ed.
Y. Horev, Jerusalem 1961), 527–8; Yalqut Shim‘oni Torah, remez 837 (to Deut 6:5) (ed.
Heiman, Mossad Harav Kuk, Jerusalem 1991) 119; Midrash Hagadol (to Deut 6:5) (ed. Fish,
Mossad Harav Kuk, Jerusalem 1972) 130; Haggadot Ha-Talmud (Berakhot ad loc., in the
printed edition of Constantinople 1511, and in an earlier manuscript from 1506 [Morocco],
MS Parma, Palatine Library 3010 [catalogue De Rossi, no. 156]); and Ein Yaaqov of R. Jacob
ibn Habib (first edition, Salonica 1516, Berakhot ad loc.). The text is also cited by the 10th-
to 11th-century North African scholar, Rabbenu Hannanel, in his commentary to Tractate
Berakhot (Otzar Hageonim to Berakhot, ed., B. M. Levin, vol. I, addendum, p. 69; ed.
D. Metzger [Jerusalem 1990] 139–40), and appears in Raymond Martini’s Pugio Fidei
(written in Spain around 1280; ed. prin., Paris 1651, second edition: Leipzig 1667; I have
collated the text with the early 13th- to 14th-century manuscript of this work in the MS
WAS RABBI AQIVA A MARTYR ? 321

of Spanish 13th-century and Oriental 15th-century provenance respectively,26

and may reflect the text as transmitted in a Spanish-Oriental recension of the
Bavli. Version B, on the other hand, seems to have been known and recorded
in the (later) Ashkenazi textual tradition of the Bavli, although it is also known
outside of this cultural milieu. Indeed, it should not be assumed that the split
between the two recensions occurred in the wake of the transmission of the
talmudic text in medieval Europe. The attestation of one section of Version B
in the commentary of the 11th-century French-Ashkenazi commentator, Rashi,
as well as the citation of a type of Version B by Rabbenu Hannanel, would point
to a relatively early textual split, probably already in the East.27 Thus both ver-
sions, whatever their later attestations, most probably originated in the East.
The text in the first editions of the Talmud represents a third, derivative recen-
sion. The three recensions appear below in parallel columns; the narrative
is divided into two separate units (“acts”—see below), designated by Roman
numerals, and these are divided each into four sections.

First printed edition Version B Version A

I
:‫אמרו‬ :‫אמרו‬ 1
‫בשעה שהוציאו ר׳ עקיבא‬ ‫כשהוציאו ר׳ עקיבא להריגה‬ ‫כשהוציאו את ר׳ עקיבה‬
‫להריגה‬ ,‫זמן קרית שמע היה‬ ‫להריגה‬
,‫זמן קרית שמע היה‬ ‫והיו סורקין את בשרו‬ ,‫זמן קרית שמע היה‬
‫והיו סורקין את בשרו‬ .‫במסרק של ברזל‬
.‫במסרקות של ברזל‬

Paris, St. Geneviève 1405; on the readings there see H. Merhaviah, “The Hebrew Versions
of ‘Pugio Fidei’ in the Sainte Geneviève Manuscript,” Qiryat Sefer 51 [1976] 283–8). As
described below, some of these texts represent partial developments of the transmission
of the text; therefore, what I call here “Version B” represents, in effect, a series of
developments of the text. However, the appearance of the martyrological elements
described below in all the main witnesses of this version (especially those deriving from
full manuscripts: the Vivanti fragment, the Genizah fragment in the Oxford Bodleian
Library, and MS Munich 95), and its absence from the two manuscripts representing
Version A justifies the description of these manuscript versions together. A full synoptic
presentation of the text witnesses to the Bavli narrative appears in Appendix I.
26 See M. Krupp, “Manuscripts of the Babylonian Talmud,” appendix to A. Goldberg, “The
Babylonian Talmud,” in: S. Safrai (ed.), The Literature of the Sages vol. 1 (Maastricht and
Philadelphia 1987) 355, 357.
27 Rashi, ad loc., cites the lemma ‫והיה מקבל עליו עול מלכות שמים‬, which represents one of
the latest stages in the development of Version B; this form of the passage is also found in
the text cited by Rabbenu Hananel (see Appendix I, line 3).
‫‪322‬‬ ‫‪mandel‬‬

‫והיה מקבל עליו‬ ‫והיה מקבל עליו‬ ‫והיה מכוון דעתו לקבל‬ ‫‪2‬‬
‫עול מלכות שמים‪.‬‬ ‫עול מלכות שמים‬ ‫עול מלכות שמים‬
‫באהבה‪.‬‬ ‫באהבה‪.‬‬
‫אמרו לו תלמידיו‪:‬‬ ‫אמרו לו תלמידיו‪:‬‬ ‫אמרו לו תלמידיו‪:‬‬ ‫‪3‬‬
‫רבינו עד כאן?‬ ‫רבינו עד כאן?‬ ‫רבינו עד כאן?‬
‫אמר להם‪:‬‬ ‫אמר להם‪:‬‬ ‫אמר להם‪:‬‬ ‫‪4‬‬
‫בניי‬
‫כל ימי הייתי מצטער על פסוק כל ימי הייתי מצטער על‬ ‫כל ימיי הייתי דורש את‬
‫פסוק‬ ‫זה‬ ‫המקרא הזה‬
‫זה‬ ‫״ואהבת את י״י אלהיך‬
‫״בכל נפשך״‪,‬‬ ‫״בכל נפשך״‬ ‫בכל לבבך ובכל נפשך״‬
‫אפילו נוטל את נשמתך‪.‬‬ ‫אפילו נוטל את נפשך‪.‬‬ ‫אפילו נוטלין את נפשך‬
‫אמרתי‪ :‬מתי יבא לידי‬ ‫אמרתי‪ :‬מתי יבא לידי‬
‫ואקיימנו?‬ ‫ואקימנו?‬
‫ועכשיו שבא לידי לא אקיימנו? ועכשיו שבא לידי לא‬ ‫ועכשיו שבא לידי לא‬
‫אקיימנו?‬ ‫אקיימנו?‬

‫‪II‬‬
‫אמרו‪:‬‬ ‫אמרו‪:‬‬ ‫‪1‬‬
‫היה מאריך ב"אחד"‬ ‫לא הספיק לגמור את הדבר לא הספיק לגמור את הדבר‬
‫עד שיצתה נשמתו ב״אחד״‪.‬‬ ‫עד שיצתה נשמתו ב״אחד״‪ .‬עד שיצתה נשמתו ב״אחד״‪.‬‬
‫יצתה בת קול ואמרה‪:‬‬
‫אשריך‬
‫ר׳ עקיבא שיצתה נשמתך‬
‫ב״אחד״‪.‬‬
‫אמרו מלאכי השרת לפני‬ ‫אמרו מלאכי השרת לפני‬ ‫אמרו מלאכי השרת לפני‬ ‫‪2‬‬
‫הקב״ה‪:‬‬ ‫הקב״ה‪:‬‬ ‫הקב״ה‪:‬‬
‫רבונו של עולם‪,‬‬ ‫רבונו של עולם‪,‬‬
‫זו תורה וזו שכרה?‬ ‫זו תורה וזו שכרה?‬ ‫זו תורה וזו שכרה?‬
‫״ממתים ידך ה׳ ממתים‬ ‫״ממתים ידך י״י וגו׳״‬ ‫״ממתים ידך י״י ממתים‬
‫מחלד״‬ ‫מחלד״‬
‫אמר להם‪ :‬״חלקם בחיים״‬ ‫אמר להם‪ :‬״חלקם בחיים״‬ ‫אמר להם ״חלקם בחיים״‬ ‫‪3‬‬

‫יצתה בת קול ואמרה‪:‬‬ ‫יצתה בת קול ואמרה‪:‬‬ ‫יצתה בת קול ואמרה‪:‬‬ ‫‪4‬‬
‫אשריך ר׳ עקיבא שאתה‬ ‫אשריך ר׳ עקיבא שאתה‬ ‫אשריך ר׳ עקיבה שאתה‬
‫מזומן לחיי העולם הבא‪.‬‬ ‫מזומן לחיי העולם הבא‪.‬‬ ‫מזומן לחיי העולם הבא‪.‬‬
WAS RABBI AQIVA A MARTYR ? 323

4.2 Comparison of the Manuscript Versions with that of the Printed

Edition of the Bavli
As a first step in the comparison between the three versions, we note that the
text of the printed editions generally follows Version B where it differs from
Version A, but it includes several additional elements. These are:

1. the absence of the introductory word, ‫אמרו‬, in two places: at the begin-
ning of this story, and before the description of Rabbi Aqiva’s death;
2. the added description of Rabbi Aqiva prolonging the word ehad until he
died;
3. the additional description of the bat qol praising him for this.

Although seemingly minor, the first is perhaps the most significant vari-
ant of the three. It may be assumed that the appearance of the introductory
term “they said” is original, and was deleted by later transmitters who saw no
import to a term which seems to interrupt the natural flow of the Hebrew tale.
However, the appearance of this term, attested in the first occurrence in almost
all text witnesses except the printed editions, and in the second occurrence in
the three main manuscript versions (MSS Oxford, Paris and Munich), actually
serves to divide the tale into two separate “acts,” numbered in the text above
by Roman numerals I and II: the first act describes Rabbi Aqiva’s dialogue with
his students, while the second act reports his death and its repercussions in the
celestial sphere. While the fact of Rabbi Aqiva’s impending death by execution
is mentioned at the very beginning of the first act, it serves only as a backdrop
to the theological discussion between Rabbi Aqiva and his students which is
the subject of this part. The second act begins with Rabbi Aqiva’s dramatic
death while reciting the very beginning of the Shema (which occurred, as we
shall presently see, after the discussion of the first act), and this death, taking
place during Rabbi Aqiva’s recital of the significant first verse of the Shema,
provides the basis for the ensuing dialogue between the heavenly retinue (mal-
akhei hasharet) and God, with the concluding triumphant statement of the bat
qol concerning Rabbi Aqiva’s entrance into the “world to come.”
The other two additions in the version of the printed edition provide dra-
matic flourishes, but do not significantly add to the plot. Both of them are
related to the statement concerning Rabbi Aqiva’s death at the recitation of
the word “one” of the Shema; the threefold repetition of the word ehad in
these versions seems to weigh inordinately on the dramatic effect, as does the
statement by the bat qol, which anticipates the similar appearance and state-
ment of the bat qol in the final line, without adding anything to its content or
324 mandel

form. We may therefore safely assume that these are late additions to a text
based on Version B.
As mentioned, the first act does not focus upon Rabbi Aqiva’s death, which
is the subject of act II. For a fuller analysis of the two sections and their rela-
tionship to each other I refer the reader to Yonah Fraenkel’s interpretation;28
indeed, in the context of the Bavli’s text both parts work together as a
sequel to the earlier tale concerning the arrest of Rabbi Aqiva and Pappus
ben Yehuda.29 However, while the wider analysis is significant for an under-
standing of the Bavli narrative as a whole, it will be sufficient for our present
discussion to concentrate on the first act of the tale in the Bavli, for reasons
that will become clear.

4.3 Analysis of the Texts in the Manuscript Versions

Concentrating now on the manuscript versions of the tale, we take note of
the particular variants that characterize Version A and differentiate it from the
texts representing Version B; the significant variants—five in number—are
found in three places in the narrative: (1) the mention in the manuscripts of
Version B of the torture administered to Rabbi Aqiva (flaying his skin with iron
combs) in the exposition;30 (2) the description of Rabbi Aqiva’s activity sur-
rounding the recitation of the Shema—Version A describes this as preparation
(‫ )מכוון דעתו‬for “receiving the yoke of the kingdom of Heaven” (the latter being
a standard phrase in rabbinic literature for the recital of the Shema31), while
some of the texts of Version B use the simpler phrase “accepted the yoke . . . ”;32
(3) three variants in Rabbi Aqiva’s statement to his students appear in the

28 Y. Fraenkel, Iyunim be-Olamo ha-Ruhani shel Sipur ha-Aggadah (Tel Aviv 1981) 49–52
[Hebrew]. See also ibid., Midrash and Agadah, vol. 2 (The Open University of Israel
19962) 440–2 [Hebrew], and Darkei ha-Aggadah ve-ha-Midrash (Ramat Gan 1991) 351–2
[Hebrew].
29 It is significant that this first tale ends with the phrase (stated by Pappos): ‫אשריך ר׳‬
‫עקיבא‬, which then is mirrored by the same phrase in the mouth of the bat qol at the
conclusion of the second act.
30 The torture appears in all manuscripts of Version B except its citation in Midrash Hagadol,
Rabbenu Hannanel and Pitron Torah; see Appendix I, line 2.
31 See Rashi, ad loc.
32 The phrase found in Version A appears in corrupt form in several textual witnesses of
Version B (the citations in Ein Yaakov, Haggadot Ha-Talmud, Menorat Hama’or, of Rabbi
Isaac Aboab, Yalqut Shimoni, and the Cambridge Genizah fragment). As mentioned above,
this may indicate an intermediate stage in the development of Version B. See Appendix I,
line 3.
WAS RABBI AQIVA A MARTYR ? 325

manuscripts of Version B—the absence of the appellation ‫בניי‬,33 Rabbi Aqiva’s

mention of having been “troubled” (‫ )מצטער‬by the verse of Deut 6:5,34 and his
statement that he has anticipated the fulfillment of this verse (‫אמרתי מתי יבוא‬
‫)לידי ואקיימנו‬.35
The first variant (Aqiva’s torture), along with the last two (his “trouble’ con-
cerning the verse and his anticipation of the fulfillment of the verse), clearly
differentiate Version B from Version A, marking it a martyrological tale.36 In
Version B, Rabbi Aqiva has actively anticipated the fulfillment of the require-
ment of the love of God at the moment of death. While Rabbi Aqiva may still
have hoped that his death be a peaceful one,37 it seems clear that the present
moment has presented him a long sought-after opportunity to fulfill a desire
only possible at death. Moreover, the query of the students, coming after the
mention of Rabbi Aqiva’s torture, must be construed here to refer to his recital
of the Shema under these trying circumstances. This is not the case for Version
A, where the students’ surprise is registered only in relation to Rabbi Aqiva’s
preparation for the reciting of the Shema, mentioned directly prior to their
query; I will discuss this below.
Turning to the other characteristic readings of Version A, we note that Rabbi
Aqiva addresses his students (‫ )בניי‬as he tells them that he has always taught
the implications of this verse—‫כל ימי הייתי דורש את המקרא הזה‬: namely, that
the biblical injunction to love God “with all one’s soul” implies that this love

33 This word appears only in MS Paris, and also in some of the versions of the parallel in
Midrash Aseret Harugei Malkhut. Its appearance in the latter, which otherwise has no
affinity with the Paris MS text, would seem to confirm the authenticity of this reading. See
Appendix I, line 5.
34 This appears in all text witnesses except MSS Oxford and Paris of Version A (Appendix I,
line 5.).
35 This phrase appears in all witnesses of Version B except for the citations in Ein Yaakov,
Haggadat Ha-Talmud, and Menorat Hama’or. These versions are related; comparison with
the manuscript of Haggadot Ha-Talmud leads to a possibility that this phrase was omitted
there (or in its Vorlage) due to scribal error, and restored in the later printed editions of
these anthologies; see the synopsis in Appendix I, line 7. It is also omitted in Sefer Pitron
Torah and the citation by Rabbenu Hananel; these may represent hybrid texts.
36 This has been noted also by Safrai, “Martyrdom in the Teachings of the Tannaim,” 155–6,
and Fraenkel, Iyyunim, note on p. 167.
37 In parallel rabbinic passages the verb ‫מצטער‬, when preceding the citation of a verse from
Scripture, does not imply “sorrow” but rather a “concern” or a lack of understanding; cf.
MdRY bahodesh 1; bMeg 24b; and cf. tYev 14:5 (and parallels in yYev 16:4, 15d and bYev 121a);
bBer 5b.
326 mandel

must even accompany a person at his/her death.38 The use of the verb doresh
is particularly apt when coupled with the verb ‫ מקיים‬used in the next phrase
(as appearing in Version A): ‫ועכשיו שבא לידי לא אקיימנו‬. These two verbs often
appear together in rabbinic literature, describing the necessity of acting in
consequence of a teaching.39 Thus, this narrative centers upon the perfect
agreement between Rabbi Aqiva’s actions at the time of his execution and his
teachings imparted throughout his life.
The remaining variant between Versions A and B concerns the description
of Aqiva’s recitation of the Shema: as mentioned above, in Version A Rabbi
Aqiva is described as preparing himself through concentration (‫)מכוון דעתו‬40
to recite the Shema, not actually reciting it (the recital of the Shema is the usual
meaning of “accepting upon himself the yoke of the kingdom of heaven”).
This has particular significance for the understanding of the ensuing drama,
for according to this reading it is not to Aqiva’s recitation per se, and certainly
not to his torture, that the students are responding and questioning, but to
his ability to direct his attention to the preparation of the commandment to
recite the Shema rather than to his impending execution. This solves an impor-
tant crux of the story as it appears in the other versions: How is it possible for
Rabbi Aqiva to have a discussion with his students at the very moment that
he is reciting the Shema (during which it is forbidden to divert one’s atten-
tion), especially if this recitation is taking place as he is being tortured?41 More
important, however, is the literary significance of this act, which lays the foun-

38 The use of the plural for “take (your life)” in Version A (‫ )אפילו נוטלין את נשמתך‬seems to
hint at death at the hands of others (without a direct reference to the hand of God), which
in the present context would imply Rabbi Aqiva’s impending execution. See Fraenkel,
Iyyunim, note on p. 167.
39 On the verb darash as “taught” and not “derived a lesson through (special) exegetical
techniques,” see P. Mandel, “ ‘Darash Rabbi Peloni’: ‘Iyun Hadash,” Dappim: Research in
 

Literature 16–17 (2007–8) 27–55, see esp. 37 on the phrase ‫( נאה דורש נאה מקיים‬tYev 8:7;
tHag 2:1), which should be understood in the context of teaching and not “interpretation”
(in the sense of “practice what one preaches”); idem, “The Origins of Midrash in the
Second Temple Period,” in: C. Bakhos (ed.), Current Trends in the Study of Midrash (Leiden
2006) 9–34; idem, “Legal Midrash Between Hillel and Rabbi Akiva: Did 70 CE Make a
Difference?,” in: D. Schwartz and Z. Weiss (eds.), Was 70 CE a Watershed in Jewish History?
On Jews and Judaism before and after the Destruction of the Second Temple (Leiden and
Boston 2012) 343–70. See also mSot 5:5 and tYom 1:8, where a similar phrase is found (see
Appendix II).
40 See the cognate Aramaic term in: Sokoloff, Dictionary of Jewish Babylonian Aramaic, 563.
41 Note that according to the following part of the narrative, he dies while reciting the first
verse of the Shema. It is very difficult to understand how his students engaged him in
conversation while he was actually reciting this verse and undergoing torture. This
WAS RABBI AQIVA A MARTYR ? 327

dation for the essence of the story, that being the dialogue between Aqiva and
his students. This can be properly appreciated if we understand the function
of the phrase ‫ כיוון הדעת‬in Talmudic context. On the one hand, kivvun da‘at as
“concentration,” “intention,” is the standard legal requirement for the reading
of the Shema: anyone performing the reading of Shema without paying at least
minimal attention to it has not fulfilled his religious duty.42 On the other hand,
it is intrinsic to such concentration that it is essentially an inner, subjective
act; thus, no one except Rabbi Aqiva himself can know whether it has been
done successfully. This act of concentration at the present moment, of course,
holds special significance for Rabbi Aqiva. Thus, while the students see their
teacher preparing to perform what must usually have been considered a per-
functory, regularly observed ritual act, and are amazed that their teacher can
direct his attention to this at such a time, Aqiva responds lovingly to them,
explaining that the very essence of the verses to be read contains precisely
the consciousness that one must have each day: that the love of God ‫בכל נפשך‬
must include all circumstances, including death, whether that death comes qui-
etly in one’s bed or whether one’s life is “taken” by a Roman executioner.43 It is this
inner meaning of the Shema, taught and experienced “with intent” throughout
his life (“all my days”), that allows Aqiva to approach this moment of certain
death directed towards the love of God and not towards his impending fate.
His response to the surprised question of his students is a lesson not in martyr-
dom, but in the significance of the daily love of God and the fulfillment of that
love throughout one’s life until its very end, and in the way Aqiva has taught
this lesson and has continued to impart its significance to his students at the
present moment.
Thus, Version A and Version B represent very different views of Rabbi Aqiva’s
death and his understanding of it as conveyed to his students. Version B clearly
introduces elements close to a martyrdom tale, not only producing a graphic
and horrible description of torture, but also emphasizing Aqiva’s emotional
attachment to the prescription of the love of God “even when He/they take/s
your soul away”: “all my days I was concerned . . . when will it arrive . . .” Version

question is also raised by the 18th-century talmudic commentator, R. Ezekiel Landau, in

his commentary to the Bavli, Ziyun LeNefesh Haya, ad loc.
42 See mBer 2:1; tBer 2:2; mRH 3:7; mMeg 2:2 where the phrase ‫ כיוון לבו‬refers to the
concentration necessary to perform a religious act properly. The corresponding phrase
‫ כוון דעתו‬does not appear in the tannaitic sources, but is prevalent in the Bavli: see bBer
30b, 34a; bRH 28b–29a; bMeg 25a; yBer 6:1, 10a; see also SongR to Song 8:13 (in the context
of the recital of the Shema). Cf. Sokoloff, Dictionary of Jewish Palestinian Aramaic.
43 See Fraenkel, Iyyunim, note on p. 167.
328 mandel

A concentrates on a sparser, less dramatic vision of Aqiva’s ability to recite the

Shema at the hour of his execution, focusing upon the double aspect of having
previously taught the meaning of Deut 6:5 (to these very students?) and upon
his own fulfillment of that teaching at his own death. Is Version B to be con-
sidered a later development of Version A, adding the martyrological elements
to the earlier, leaner tale, or were these elements part of the original tale and
deleted from Version A?44 A comparison between the Yerushalmi and Bavli
texts will help us determine the answer to this question, and will also highlight
an essential difference between the tales as told in the different geographic
centers.

5 Transformations of the Tale of Rabbi Aqiva

5.1 Comparison of the Narratives in the Bavli and Yerushalmi

While the Bavli and Yerushalmi versions depict two different plots, they clearly
are interrelated, both dealing with Rabbi Aqiva’s interaction with a third party
concerning the recital of the Shema at a moment of imminent danger. The
Yerushalmi’s version is anchored in what was possibly a known and histori-
cally accurate situation: a trial of Rabbi Aqiva before the Roman governor of
the Province of Judaea. The Bavli version assumes Aqiva’s execution (which is
used, in the broader narrative found in the Bavli here, as a sequel to the tale
concerning Aqiva’s arrest for the teaching of Torah), but exhibits no interest in
the external situation; rather, the narrator concentrates on a dialogue between
teacher and students seemingly suspended within the time-frame leading to
Aqiva’s execution. It is possible to see a parallel structure between these ver-
sions; however, it is apparent that only act I of the Bavli versions can be com-
pared to the Yerushalmi tale, since the latter stops short of any description of
harm done to Rabbi Aqiva, and act II of the Bavli tale begins with the graphic
description of Aqiva’s death. The following table demonstrates that it is spe-
cifically act I as it appears in Version A that presents the clearest parallel to
the Yerushalmi tale: while the additional information in Version B regarding
Rabbi Aqiva’s torture has no parallel in the Yerushalmi versions, Version A cor-
responds neatly with the Yerushalmi narrative; each narrative consists of four
concise sections (as numbered in the Bavli text presented above).

44 See the doubts raised by D. Boyarin, “ha-midrash ve-ha-ma‘aseh: ‘al heqer ha-historiya
shel sifrut hazal,” in: S. Friedman (ed.), Saul Lieberman Memorial Volume (New York and
Jerusalem 1993) 116–7, n. 34 [Hebrew].
WAS RABBI AQIVA A MARTYR ? 329

Yerushalmi Version Bavli Version A

1. Exposition Trial of Rabbi Aqiva b/T. Rufus Execution of Rabbi Aqiva

during the time for reading of the Shema
2. Activity of R. Aqiva R. Aqiva reads Shema and R. Aqiva “prepares himself”
laughs for reading Shema “with love”
3. Surprise registered Tineius Rufus objects/ Students are surprised/
by audience questions question
4. R. Aqiva’s answer a) address/retort to Roman a) address to students
and explanation governor
b) admission of recital of b) admission of teaching of
Deut 6:5 “all my days” Deut 6:5 “all my days”
c) significance of “love with c) significance of “love with
all your soul” all your soul”
d) “now it has come” d) “now it has come to/for me”
e) explanation of laughter: e) explanation of concentra-
“I have not desisted from tion of the recital of Shema:
the recital of the Shema!” “Should I not fulfill it?!”

The symmetry between the two tales suggests that the Bavli Version A is depen-
dent upon the version of the Yerushalmi and is a conscious retelling of it.45
The transformation that the tale has undergone by the Bavli narrator includes
a switch in both the time and the dramatic situation of the plot: from a politi-
cal discussion between Aqiva and his judge in the Yerushalmi tale, the Bavli
narrator creates a dialogue between Aqiva and his students; and by switching
the time of the discussion from after the reading of the Shema to the moment
before this reading takes place, the narrator is able to center the tale on the
inner, subjective intention which Aqiva shows (and which he teaches his stu-
dents) must accompany every act at all times. Precisely these two topics—the
relationship of teacher to students, and the inner world of intent—are those
that appear repeatedly in narratives in the Bavli, and they reveal what is at the
heart of the Babylonian narrator’s interest.46

45 Especially with regard to historical narratives that take place in the Land of Israel, it is
usual to assume that among parallel narratives appearing in the Bavli and Yerushalmi, it is
the Bavli tale that is based on the Yerushalmi; cf. P. Mandel, “The Loss of Center: Changing
Attitudes Towards the Temple in Aggadic Literature,” Harvard Theological Review 99
(2006) 17–35, and see the following note.
46 Examples of stories in the Yerushalmi that are transformed into a teacher–student
narrative are: the death of Rabbi Judah the Prince in yBik 3:3, 65c and its parallel in bKet
330 mandel

In reworking the Palestinian tale, the Babylonian narrator (of Version A)

uses bits and pieces of language that appear in other places in rabbinic litera-
ture to create what is essentially a pastiche or mosaic of rabbinic phraseology,47
from which he creates a new drama. Similar reworkings of Palestinian material
by the narrators of Bavli tales have been noted by scholars, and are a hallmark
of narrative practices as found in the Bavli.48

5.2 Further Transmission of the Bavli Narrative: The Making

of a Martyrology
Version A would thus seem to be the first stage in the re-editing of the Aqiva
narrative by a Bavli narrator; in this version, Rabbi Aqiva’s death/execution is
viewed as perpetrated by the Roman authorities, apparently as a consequence
of his arrest for teaching Torah.49 This version, however, should not properly be

104a; the legend of Kamza and (ben) Kamz(ora) in LamR 4:2 and the parallel in bGit
58b; and the tale concerning the grandson of Honi Ha-Maagel in yTaan 3:10, 66d and
the parallel in bTaan 23a. See J. Rubenstein, Talmudic Stories: Narrative Art, Composition,
and Culture (Baltimore and London 1999) (see there especially Chapter 2 [“The Oven
of Achnai”] and Chapter 3 [“Elisha ben Abuya”]); idem, The Culture of the Babylonian
Talmud (Baltimore and London 2003) (and see there especially Chapter 1 [“The Rabbinic
Academy”]). See especially on the Bavli’s “greater emphasis on Torah study” in R. Kalmin,
“Rabbinic Traditions about Roman Persecutions,” and cf. his discussion concerning the
preceding Bavli narrative of Rabbi Aqiva and Pappos (pp. 30–1), and his conclusions
(p. 49). On the importance of the concept of intentionality in the Bavli, see the recent study
by S. Strauch-Schick, “Intention in the Babylonian Talmud: An Intellectual History” (Ph.D.
diss.; Yeshiva University 2011), esp. 138–172 (“Intention in the Performance of Rituals”), and
190–1 (“Recitation of the Shema,” and cf. there the citations of the work of Leib Moscovitz);
and cf. J. Levinson, “From Narrative Practice to Cultural Poetics: Literary Anthropology
and the Rabbinic Sense of Self,” in: M. R. Niehoff (ed.), Homer and the Bible in the Eyes of
Ancient Interpreters (Leiden and Boston 2012) 345–67, and see esp. 355 on the dominance
of intention in rabbinic texts. I thank Joshua Levinson for the last two references.
47 See Appendix II for a list of phrases in the Bavli version found especially in other Bavli
narratives as well as elsewhere in the rabbinic corpus.
48 See S. Friedman, “The Further Adventures of Rav Kahana: Between Babylonia and
Palestine,” in: P. Schäfer (ed.), The Talmud Yerushalmi and Graeco-Roman Culture, vol. 3
(Tübingen 2002) 247–71; idem, “A Good Story Deserves Retelling: the Unfolding of the
Akiva Legend,” in: J. L. Rubenstein (ed.), Creation and Composition: the Contribution of the
Bavli Redactors (Stammaim) to the Aggada (Tübingen 2005) 71–100.
49 Whatever the earlier Palestinian tradition may have assumed, the Bavli tradition clearly
views Rabbi Aqiva and other contemporary sages as having died deaths “al qiddush
hashem”: see the use of the phrase “Rabbi Aqiva and his companions” in bSanh 110b and
bRH 23a in the context of the verse from Ps 50:5 and what would seem to be an allusion to
the Hadrianic persecutions; and cf. tSanh 13:11. Cf. Safrai, “Martyrdom in the Teachings of
WAS RABBI AQIVA A MARTYR ? 331

considered as an example of the genre of martyrology: its subject is not Aqiva’s

martyrdom but rather his attitude to death and his standing as a teacher.
Version B, on the other hand, in which its narrator has made small additions
and subtle changes to the earlier narrative, continuing the use of phrases in the
service of narrative and plot in its further development, now includes mary-
trological elements.50 Thus, the narrator of Version B introduces into this nar-
rative a theme, which may have existed as an independent tradition and may
be found in a somewhat different form elsewhere in Bavli narratives,51 of the
torture of Rabbi Aqiva (mentioned only obliquely in the Yerushalmi narrative),
and deletes the emphasis on intention and preparation of Aqiva’s recital of the
Shema. In this way, the question of the students becomes redirected from a
query concerning his ability to concentrate properly on the intention for the
recital of the Shema to his emotional, physical and spiritual stamina in the face
of his final torture and impending death. This narrative is, indeed, now a mar-
tyrology, and can be seen against the backdrop of the Christian martyrologies
of the third and fourth centuries and later. In fact, flaying the flesh was typi-
cal of Roman tortures during the third and fourth centuries,52 and it is likely
that these elements were added to the narrative as the Bavli text, composed

the Tannaim,” 155–6. On the Bavli emphasis on the theme of Torah teaching, see Kalmin,
“Rabbinic Traditions about Roman Persecutions.”
50 See Appendix II, especially concerning the form of the phrase ‫כל ימי הייתי מצטער על‬
‫פסוק זה‬, as well as the added phrase ‫ מתי יבוא לידי ואקיימנו‬,‫ אמרתי‬in bYom 19b; and see
n. 69 below, there concerning the absence of the latter phrase in some (authoritative)
manuscripts in bYom. The introduction of this phrase here is possibly connected to the
introduction of the same phrase in versions of bYom; but further research is necessary.
It should be noted that the same conclusion was reached by Safrai (Martyrdom in the
Teachings of the Tannaim,” 156–7) in his comparison of the textual versions appearing in
manuscripts and in the citations.
51 See this motif in bMen 29b (“weighing [Rabbi Aqiva’s] flesh in the marketplace”) and see
Appendix II; cf. the following note.
52 See Lieberman, “The Martyrs of Caesarea,” 420, n. 23: “This [i.e. flaying the flesh with iron
combs] seems to have been the most common torture in Caesarea at the beginning of
the IV c.”; and cf. ibid., 421: “TB [= Babylonian Talmud] is the first source to inform us
that he died of being tortured by means of iron combs at the hands of the proconsul
of Caesarea”—but in the light of the manuscript data, this should be revised to refer
only to the later transmission of the Bavli tale. See also Lieberman, “redifat dat Yisrael,”
233 n. 67. The theme of flaying the flesh with iron combs appears also in the tale of
Nebuchadnezzar and the blood of Zechariah in the Babylonian versions in bGit 57b and
bSanh 96b (it does not appear in the Palestinian versions of this tale, but is interpolated
in later transmissions of the tale in LamR [see ed. Buber, petiha 23, p. 21 and notes there;
2:2; 4:13] and in QohR to 3:16 and 10:4).
332 mandel

originally outside of the Roman cultural influence, was recited and retransmit-
ted, whether in Palestine or in Babylonia, in the later Byzantine period.53
Indeed, it is no surprise that precisely the added elements of Version B are
found in the version of Rabbi Aqiva’s death in the late Byzantine Midrash of the
Ten Martyrs (Midrash Aseret Harugei Malkhut), which was put together from
various earlier accounts—from both talmudim—in the sixth or seventh cen-
tury, perhaps in Palestinian circles.54 The text as found in most recensions of
this work is as follows:

53 See Lieberman, “redifat dat Yisrael,” 421–2, who cites the trial and execution of the Christian
teacher of Eusebius, Pamphilus, one of the prominent Christian martyrs of the early
4th-century CE in Caesarea. The parallels between the events surrounding the execution
and death of this Christian martyr, as well as his teaching activities leading up to his arrest,
provide, as Lieberman points out, a “striking parallel” to the account of Rabbi Aqiva’s life
and death “in the prison of the same city.” The relationship between this tale, and others
like it, and the accounts in the Bavli requires further study. The evidence of martyrological
events in the fourth century need not lead to a dating of the (rewritten) Bavli narrative to
that time. Rather, it would seem more likely that the historical occurrences led to literary
influences over time; how much time might have elapsed is difficult to ascertain, given the
lack of other evidence. It is also unclear whether these later influences were introduced
in a re-narration/recitation of the Bavli text in a Palestinian venue, where, perhaps,
Christian traditions concerning the martyrs became known in Jewish circles, or in a
Babylonian setting: as is known, Christians were persecuted by the Zoroastrian Sasanian
Empire, especially during the reign of Shapur II in the fourth century, and martyrological
texts describing these persecutions were composed in Syriac in later centuries; these may
have influenced the later Jewish traditions. See a bibliographic summary of these texts
in G. Herman, “ ‘Bury My Coffin Deep!’: Zoroastrian Exhumation in Jewish and Christian
 

Sources,” in: J. Roth, M. Schmelzer, and Y. Francus (eds.), Tiferet leYisrael: Jubilee Volume
in Honor of Israel Francus (New York 2010) 33–4, n. 11; cf. the review by S. P. Brock of
G. Wiessner, Zur Märtyrerüberlieferung aus der Christenverfolgung Schapurs II (Abhand-
lungen der Akademie der Wissenschaften in Göttingen, phil.-hist. Kl. III.67; Göttingen
1967), in Journal of Theological Studies 19 (1968) 300–9, and idem, The History of the Holy
Mar Ma’in, with a Guide to the Persian Martyr Acts (Piscataway NJ 2008). I thank Isaiah
Gafni and Geoffrey Herman for their assistance in this area.
54 A full synopsis of the text of this work was published by G. Reeg, Die Geschichte von den
Zehn Märtyrern (Tübingen 1985) 70*–71*. The version presented here is an eclectic one,
based on recensions IV, V, VI, VII and IX of Reeg’s edition; see Reeg’s discussion there,
Chapter 4 (“Analyse und Vergleiche der Rezensionen”), pp. 33–57, and esp. 52–53 on the
special, edited character of Recension III. The sentences in square brackets appear in only
some of the recensions; the statement of the angels before God appears in two different
places in the different versions.
WAS RABBI AQIVA A MARTYR ? 333

‫וכשהוציאוהו להורגו זמן קרית שמע היה‬

;‫והיו מסרקין בשרו במסרקות של ברזל‬
.‫והיה מקבל עליו עול מלכות שמים‬
!‫ רבונו של עולם‬:‫אמרו מלאכי השרת לפני הק׳‬/[
.)‫״ממתיך ידך יי״ ממתים מחלד״ (תהלים טז ד‬
]/.)‫ ״חלקם בחיים״ (שם‬:‫אמר להם‬
?‫ רבינו עד כאן‬:‫אמרו לו תלמידיו‬
‫מקרא זה‬/‫ בניי! כל ימיי הייתי מצטער על פסוק‬:‫אמר להם‬
)‫להיך בכל לבבך ובכל נפשך״ (דברים ו ה‬-‫״ואהבת את ה׳ א‬
.‫אפילו הוא נוטל את נפשך‬
.‫ אימתי יבא לידי ואקיימנו‬:‫והייתי אומר‬
?‫ לא אקיימנו‬,‫ועכשיו שבא לידי‬
]‫[האריך באחד‬
.‫ויצתה נשמתו באחד‬
!‫ רבונו של עולם‬:‫אמרו מלאכי השרת לפני הקב״ה‬/[
!)‫״ממתים ידך י״י ממתים מחלד״ (תהלים טז ד‬
]/.)‫ ״חלקם בחיים״ (שם‬:‫אמר להם‬
‫ אשריך ר׳ עקיבא ש[גופך טהור ו]יצתה נשמתך בטהרה‬:‫יצתה בת קול ואמרה‬
,]‫באחד‬/[
.‫ומזומן אתה לחיי העולם הבא‬

The version in the Ten Martyrs reflects an obviously martyrologically inflected

tale. It is unclear which elements in this version of the tale, if any, were already
extant in a Bavli version and were later introduced into the Ten Martyrs from
a Bavli tradition, or whether some of these elements were inventions of the
author/redactor of this work, who may have had access to Christian martyro-
logical texts, and which were subsequently interpolated into a later transmis-
sion of the Bavli tale.55 It is likely that the more fantastical elements, of the
death of Rabbi Aqiva while prolonging the word “one,” and of the heavenly
voice praising him for this, were included in later versions of the tale,56 and

55 In a paper given at the plenary session of the 14th World Congress of Jewish Studies in
Jerusalem (August 2005), Hillel Newman demonstrated a close parallel between the
tradition in the work Ten Martyrs concerning Rabbi Aqiva’s burial under the tetrapylon
of Caesarea and a Christian martyrological narrative of the seventh century CE. Further
research (P. Mandel, “‘The Sacrifice of the Souls of the Righteous Upon the Heavenly
Alter’: A Medieval Rabbinic Transformation of and Early Apocalyptic and Christian
Tradition,” forthcoming) has revealed other late Christian traditions embedded in this
work.
56 On the extension of the word ehad in Deut 6:4, see bBer 13b and the parallel in yBer 2:1,
4a; on the praise of the bat qol at Rabbi Aqiva’s death, and specifically his death while
dramatically reciting the word “one”; cf. bBM 86a and bAZ 27b. These motifs do not
334 mandel

were then interpolated by medieval scribes into versions of the Bavli talmudic
text, which, finally, appeared in the first printed editions to Bavli Berakhot, and
were transmitted from there to all printed editions of the Talmud.

6 The Social and Political Contexts of the Two Talmudic Versions:

From Political Drama to Teaching Exemplum

In conclusion, it may be determined that the tale of Rabbi Aqiva is not, in its
origin, a story of a martyr and does not share common tropes with Christian
martyrological texts. In the version of the Yerushalmi, the narrative describes
a story of a political drama where the love of God, as exhibited in Aqiva’s joy-
ful performance of the commandment of the recital of the Shema, triumphs
over the sufferings of the trial. Rabbi Aqiva’s laughter is, in fact, a common
element in other tales about him; here it serves to help Rabbi Aqiva turn the
tables on his tormentor, creating his own political drama, on his terms.57 As
mentioned above, there is no indication in this narrative that Rabbi Aqiva
will be executed; the narrator is satisfied to portray Rabbi Aqiva’s victory over
his tormentor.
The narrator of the original Bavli tale, as in other instances of the Bavli’s
rewriting of Palestinian traditions, transfers the interest from the political
sphere to the educational one: Rabbi Aqiva is here the teacher par excellence,
practicing now, at the culmination of his life, what he has preached to his stu-
dents “all his days.” Here, indeed, is a tale of Rabbi Aqiva’s execution,58 and the
transfer by this narrator of the time frame from the moment of confrontation
during the trial as in the Yerushalmi tale to just seconds before the execution
elegantly allows this narrator to emphasize the element of intent and atten-
tiveness (interests typical of the Bavli), which is used in two directions: Aqiva’s
ability to concentrate on the commandment of reciting Shema instead of on
his imminent death, and, more deeply—and this is the essence of his teaching

appear in the manuscript versions of our narrative but are found in some manuscripts
of the Midrash Hagadol version; these may have been influenced by later traditions, and
are cited also in Pitron Torah, and by Rabbenu Hannanel and in Pugio Fidei. It may be
significant that these last three witnesses end the narrative at this point, attesting to
editorial intervention; see Appendix I, lines 9 and following.
57 On Aqiva’s laughter in other rabbinic tales, see SifDeut 43 and its parallels in LamR 5:18
and bMak 24a.
58 Aqiva’s execution was apparently understood by this author as a given; it may be indicated
in other rabbinic traditions: see LamR 2:2, and the passages cited in n. 49 above; cf. also
n. 22 above.
WAS RABBI AQIVA A MARTYR ? 335

to his students—his deeper understanding of the significance of that mitzvah

in relation to his performance of it during his life.
Both narratives present this central commandment and ritual as provid-
ing ultimate meaning for Rabbi Aqiva, not as a martyr’s credo but rather as
an event that provides significance for Aqiva’s life, becoming a fulcrum for
a transformation towards others—whether they be his tormentor or his
students. In all of this there is no clear evidence of influence from outside,
Christian sources. But the later widespread persecutions of the Roman Empire,
at first particularly against Christians, but later also perpetrated by the later
Christian Empire against Jews, influenced the form of this story in the later
Byzantine period, and martyrological elements were gradually introduced into
the Bavli tale in its further transmission; these elements, indeed, show affini-
ties to Christian martyrological texts. More research is needed to determine
the precise relationship between the martyrological texts of the Ten Martyrs
narrative and Christian influences in general, as well as the time and precise
provenance of the traditions in this work. It is plausible that the final stages of
the transformation of the Aqiva tale into a martyrology were influenced by this
work; certainly, some elements found in the narrative of the Ten Martyrs were
introduced into the narrative in the medieval period, and these then found
their way into the text of the printed editions of the Bavli as well as in other
medieval versions and citations of the Bavli. As we have seen above, the tale as
a martyrology also made an impact upon the form of the text as found in the
printed edition of the Yerushalmi to Tractate Berakhot (attested already in the
13th-century Leiden manuscript to that tractate), adding to the original text
the mention of Aqiva’s death as well as his anticipation of it. These additions
tend to obscure the original development of the tale in place as well as in time,
although they themselves attest to later attempts to read the Yerushalmi narra-
tive in the light of the known Bavli tradition.
336 mandel

Appendix I

The Execution of Rabbi Aqiva (bBer 61b)—Synopsis of

Textual Witnesses*

line 1
‫היה‬ ‫זמן קרית שמע‬ ‫להריגה‬ ‫ר׳ עקיב׳‬ ‫את‬ ‫כשהוציאו‬ ‫אמרו‬ ‫א‬
‫זמן קרית שמע‬ ‫להריגה‬ ‫ר׳ עקיבה‬ ‫את‬ 59‫כשהוציאוהו‬ ‫אמרו‬ ‫פ‬
‫היה‬ ‫זמן קרית שמע‬ ‫להריגה ואותה‬ ‫לר׳ עקיבה‬ ‫הוציאוהו‬ ‫ת‬
‫שעה‬
‫היה‬ ‫זמן קריאת שמע‬ ‫להריגה‬ ‫לרבי‬ ‫כשהוציאו‬ ‫אמרו‬ ‫ב‬
‫עקיבא‬
‫היה‬ ‫זמן ק״ש‬ ‫להריגה‬ ‫לר׳ עקיבא‬ ‫כשהוציאו‬ ‫אמרו‬ ‫ה‬
‫היה‬ ‫זמן קרית שמע‬ ‫להריגה‬ ‫לר׳ עקיב׳‬ ‫כשהוציאו‬ ‫אמרו‬ ‫הד‬

‫היה‬ ‫זמן ק״ש‬ ‫להריגה‬ ‫לר׳ עקיבא‬ ‫כשהוציאוהו‬ ‫אמרו‬ ‫ע‬

‫היה‬ ‫זמן קרית שמע‬ ‫להריגה‬ ‫ר׳ עקיבא‬ ‫את‬ ‫כשהוציאו‬ ‫ח‬
‫היתה‬ ‫זמן קרית שמע‬ ‫להריגה‬ ‫ר׳ עקיבה‬ ‫את‬ ‫שהוציאו‬ ‫אותה‬ ‫ג‬
‫שעה‬
‫היה‬ ‫זמן קרית שמע‬ ‫להריגה‬ ‫ר׳ עקיב׳‬ ‫כשהוציאו‬ ‫אמרו‬ ‫י‬
> < ‫ק‬
‫היה‬ ‫זמן קרית שמע‬ ‫להורגו‬ ‫ר׳ עקי׳‬ ‫כשהוציאו‬ ‫אמרו‬ ‫ר‬
‫היה‬ ‫זמן קרית שמע‬ ‫להורגו‬ ‫ר׳ עקיב׳‬ ‫כשהוציאו‬ ‫אמרו‬ ‫ו‬
‫היתה‬ ‫זמן קר< >ת‬ ‫להריגה‬ ‫לר< >יבא‬ ‫כשהוציאו‬ ‫אמרו‬ 2‫א‬
‫שמע‬
‫היה‬ ‫זמן ק״ש‬ ‫ר׳ עקיבא ליהרג‬ ‫כשהוציאו‬ ‫אמרו‬ ‫מ‬
‫היה‬ ‫זמן ק״ש‬ ‫ר׳ עקיבא להריגה‬ ‫שהוציאו‬ ‫בשעה‬ ‫דש‬

‫היה‬ ‫זמן ק״ש‬ ‫רבי עקיבא להריגה‬ ‫שהוציאו‬ ‫בשעה‬ ‫דו‬

* See the Legend of Sigla of Manuscripts and editorial notations below, page 364. The text
witnesses are ordered, as much as is feasible, in accordance with the similarity between
the versions. Attention should be drawn to the potentially qualitative distinction between
versions which presently are, or once were, part of full manuscripts of Bavli Tractate
Berakhot: ‫א‬, 2‫א‬, ‫ו‬, ‫מ‬, ‫( פ‬as well as the printed editions), on the one hand, and citations
of the text appearing in anthologies, books and commentaries: ‫ב‬, ‫ג‬, ‫( ה‬MS and printed
edition), ‫ח‬, ‫י‬, ‫ע‬, ‫ק‬, ‫ ר‬and ‫ת‬, on the other; the latter may have been more susceptible to
combination and contamination of different text versions.
59 A line drawn through the final two letters of the word ‫ כשהוציאוהו‬designates deletion
(not original scribal hand); compare other versions.
‫? ‪WAS RABBI AQIVA A MARTYR‬‬ ‫‪337‬‬

‫‪line 2‬‬
‫א‬
‫פ‬
‫ת‬
‫של ברזל‬ ‫במסרקות‬ ‫בשרו‬ ‫והיו סורקין את‬ ‫ב‬
‫של ברזל‬ ‫במסרקות‬ ‫בשרו‬ ‫והיו סורקין את‬ ‫ה‬
‫של ברזל‬ ‫במסרקות‬ ‫בשרו‬ ‫והיו סורקים את‬ ‫הד‬

‫של ברזל‬ ‫במסרקאות‬ ‫בשרו‬ ‫והיו סורקין את‬ ‫ע‬

‫ח‬
‫ג‬
‫של ברזל‬ ‫במסרקות‬ ‫בשרו‬ ‫והיו מסרקין‬ ‫י‬
‫של < >‬ ‫במסרק‬ ‫>שרו‬ ‫והיו <‬ ‫ק‬
‫של ברזל‬ ‫במסרק‬ ‫אותו‬ ‫והיו סורקין‬ ‫ר‬
‫של ברזל‬ ‫במסרק‬ ‫אתו‬ ‫והיו סורקין‬ ‫ו‬
‫של ברזל‬ ‫במסרק‬ ‫בשרו‬ ‫את‬ ‫והיו < >קין‬ ‫א‪2‬‬
‫של ברזל‬ ‫במסרק‬ ‫בשרו‬ ‫את‬ ‫והיו מסרקין‬ ‫מ‬
‫של ברזל‬ ‫במסרקות‬ ‫בשרו‬ ‫את‬ ‫והיו סורקין‬ ‫דש‬

‫של ברזל‬ ‫במסרקות‬ ‫בשרו‬ ‫את‬ ‫והיו סורקין‬ ‫דו‬

‫‪line 3‬‬
‫באהבה‬ ‫מלכות שמים‬ ‫עול‬ ‫שיקבל‬ ‫דעתו‬ ‫והיה מכוין‬ ‫א‬
‫באהבה‬ ‫מלכות שמים‬ ‫עול‬ ‫לקבל‬ ‫דעתו‬ ‫והיה מכוון‬ ‫פ‬
‫מלכות שמים‬ ‫לקבל עליו‬ ‫את לבו‬ ‫והיה מכוין‬ ‫ת‬
‫באהבה‬ ‫מלכות שמים‬ ‫עול‬ ‫לקבל‬ ‫והיה מתכוין‬ ‫ב‬
‫באהבה‬ ‫מלכות שמים‬ ‫[עול]‬ ‫לקבל עליו‬ ‫והיה מתכוין‬ ‫ה‬
‫באהבה‬ ‫מלכו׳ שמים‬ ‫עול‬ ‫לקבל עליו‬ ‫והיה מתכוין‬ ‫הד‬

‫באהבה‬ ‫מלכות שמים‬ ‫עול‬ ‫לקבל עליו‬ ‫והיה מתכוין‬ ‫ע‬

‫מלכות שמים‬ ‫מקבל עליו‬ ‫והיה‬ ‫ח‬
‫מלכות שמים‬ ‫מקבל עליו‬ ‫והיה‬ ‫ג‬
‫באהבה‬ ‫מלכות שמים‬ ‫לקבל עליו‬ ‫דעתו‬ ‫והיה מכוון‬ ‫י‬
‫>ה‬ ‫מלכות ש<‬ ‫כדי שיקבל עליו‬ ‫דעתו‬ ‫והיה מכוין‬ ‫ק‬
‫מאהבה‬ ‫מקבל‬ ‫והיה‬ ‫ר‬
‫מאהבה‬ ‫מלכות שמים‬ ‫מקבל עליו‬ ‫והיה‬ ‫ו‬
‫באהבה‬ ‫מל<>ות שמים‬ ‫מקבל עליו‬ ‫והיה‬ ‫א‪2‬‬
‫בכוונה‬ ‫מלכות שמים‬ ‫עול‬ ‫מקבל עליו‬ ‫והיה‬ ‫מ‬
‫מלכות שמים‬ ‫עול‬ ‫מקבל עליו‬ ‫והיה‬ ‫דש‬

‫מלכות שמים‬ ‫עול‬ ‫מקבל עליו‬ ‫והיה‬ ‫דו‬

‫‪338‬‬ ‫‪mandel‬‬

‫‪line 4‬‬
‫עד כאן‬ ‫רבנו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫א‬
‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫פ‬
‫עד כן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ת‬
‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ב‬
‫עד כאן‬ ‫רבי׳‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ה‬
‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫הד‬

‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמר׳ לו‬ ‫ע‬

‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ח‬
‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ג‬
‫עד כאן‬ ‫תלמידיו‬ ‫א״ל‬ ‫י‬
‫>אן‬ ‫ת<‬ ‫אמרו לו‬ ‫ק‬
‫עד כאן‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ר‬
‫עד כאן‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫ו‬
‫עד כאן‬ ‫ר׳‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫א‪2‬‬
‫עד כאן‬ ‫רבנו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫מ‬
‫עד כאן‬ ‫רבינו‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫דש‬

‫עד כאן‬ ‫רבי׳‬ ‫תלמידיו‬ ‫אמרו לו‬ ‫דו‬

‫‪line 5‬‬
‫הזה‬ ‫המקרא‬ ‫את‬ ‫דורש‬ ‫הייתי‬ ‫כל ימי‬ ‫אמ׳ להם‬ ‫א‬
‫הזה‬ ‫המקרא‬ ‫את‬ ‫דורש‬ ‫הייתי‬ ‫כל ימיי‬ ‫בניי‬ ‫אמ׳ להם‬ ‫פ‬
‫הזה‬ ‫הפסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמ׳ להן‬ ‫ת‬
‫זה‬ ‫מקרא‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמר ‪2‬להו‬ ‫ב‬
‫זה שנא׳‬ ‫מקרא‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמ׳ להם‬ ‫ה‬
‫זה שנא׳‬ ‫מקרא‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמר להם‬ ‫הד‬

‫הזה‬ ‫הפסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמר להם‬ ‫ע‬
‫דכתיב‬ ‫אהא‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמר להן‬ ‫ח‬
‫הזה‬ ‫המקרא‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמר להם‬ ‫ג‬
‫זה‬ ‫פסוק‬ ‫על‬ ‫מצטער >ער‬ ‫הייתי‬ ‫כל ימי‬ ‫א״ל‬ ‫י‬
‫זה‬ ‫פסוק‬ ‫על‬ ‫אמ׳ <‬ ‫ק‬
‫זה‬ ‫פסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמר להם‬ ‫ר‬
‫זה‬ ‫פסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫[ו]אמ׳ להם‬ ‫ו‬
‫זה‬ ‫פסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמ׳ להן‬ ‫א‪2‬‬
‫מקרא‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימיי‬ ‫אמ׳ להם‬ ‫מ‬
‫זה‬ ‫פסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמ׳ להם‬ ‫דש‬

‫זה‬ ‫פסוק‬ ‫על‬ ‫מצטער‬ ‫הייתי‬ ‫כל ימי‬ ‫אמ׳ להם‬ ‫דו‬
 ‫? ‪WAS RABBI AQIVA A MARTYR‬‬ ‫‪339‬‬

‫‪line 6‬‬
‫נוטלין את נפשך‬ ‫אפי׳‬ ‫ואהבת את י"י אלהיך בכל לבבך ובכל נפשך‬ ‫א‬
‫נוטלין את נפשך‬ ‫ואהבת את י״י אלהיך בכל לבבך ובכל נפשך ובכל מאדך אפי׳‬ ‫פ‬
‫הוא נוטל את נפשך‬ ‫ואפי׳‬ ‫ובכל נפשך‬ ‫ת‬
‫הוא נוטל את נפשך‬ ‫אפילו‬ ‫בכל נפשך‬ ‫ב‬
‫הוא נוטל את נפשך‬ ‫אפי׳‬ ‫בכל נפשך‬ ‫ה‬
‫הוא נוטל את נפשך‬ ‫ואפי׳‬ ‫בכל נפשך‬ ‫הד‬

‫ואפילו הוא נוטל את נפשך‬ ‫בכל נפשך‬ ‫ע‬

‫נוטל את נפשך‬ ‫ואפילו‬ ‫בכל נפשך‬ ‫ח‬
‫הוא נוטל את נפשך‬ ‫אפלו‬ ‫בכל נפשך‬ ‫ג‬
‫נוטל את נפשך‬ ‫אפי׳‬ ‫בכל נפשך‬ ‫י‬
‫נוטל את נפשך‬ ‫א< >לו‬ ‫בכל נפשך‬ ‫ק‬
‫נוטל את נפשך‬ ‫ואפילו‬ ‫בכל נפשך‬ ‫ר‬
‫נוטל את נפשך‬ ‫ואפיל׳‬ ‫בכל נפשך‬ ‫ו‬
‫נוטל את נפשך‬ ‫אפילו‬ ‫בכל נפשך‬ ‫א‪2‬‬
‫נוטל את נפשך‬ ‫אפי׳‬ ‫בכל נפשך‬ ‫מ‬
‫נוטל את נשמתך‬ ‫אפי׳‬ ‫בכל נפשך‬ ‫דש‬

‫נוטל את נשמתך‬ ‫אפי׳‬ ‫בכל נפשך‬ ‫דו‬

‫‪line 7‬‬
‫אקימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫א‬
‫אקיימנו‬ ‫לא‬ ‫בידי‬ ‫שבא‬ ‫ועכשיו‬ ‫פ‬
‫אקימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫עכשיו‬ ‫ת‬
‫אקיימנה‬ ‫שבאה לידי‬ ‫ועכשיו‬ ‫ב‬
‫(ו)אקיימנו‬ ‫[ולא]‬ ‫לידי‬ ‫בא‬ ‫ועכשיו‬ ‫ה‬
‫אקיימנו‬ ‫ולא‬ ‫לידי‬ ‫בא‬ ‫ועכשיו‬ ‫הד‬

‫ואקיימנו‬ ‫לידי‬ ‫בא‬ ‫ועכשיו‬ ‫ע‬

‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫ח‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫עכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫אמרתי אימתי יבא‬ ‫ג‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫ואמרתי מתי יבא‬ ‫י‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫<>‬ ‫ועכשיו‬ ‫ואיקימנו‬ ‫לידי‬ ‫אמרתי אימת< >‬ ‫ק‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫אמרתי מתי יבא‬ ‫ר‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫אמרתי מתי יבא‬ ‫ו‬
‫אקיימנה‬ ‫ולא‬ ‫ליד< >‬ ‫בא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫ואמרתי מתי יבוא‬ ‫א‪2‬‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫בא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫אמרתי מתי יבא‬ ‫מ‬
‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫אמרתי מתי יבא‬ ‫דש‬

‫אקיימנו‬ ‫לא‬ ‫לידי‬ ‫שבא‬ ‫ועכשיו‬ ‫ואקיימנו‬ ‫לידי‬ ‫אמרתי מתי יבא‬ ‫דו‬
‫‪340‬‬ ‫‪mandel‬‬

‫‪line 8‬‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫את הדבר‬ ‫לגמור‬ ‫הספיק‬ ‫לא‬ ‫אמרו‬ ‫א‬
‫באחד‬ ‫נשמתו‪60‬‬ ‫שיצאה‬ ‫עד‬ ‫לגמור‬ ‫הספיק‬ ‫אמרו לו לא‬ ‫פ‬
‫באחד‬ ‫נשמתו‬ ‫יצתה‬ ‫ת‬
‫באחד‬ ‫נשמתו‬ ‫יצתה‬ ‫ב‬
‫באחד‬ ‫נפשו‬ ‫יצאה‬ ‫ה‬
‫באחד‬ ‫נפשו‬ ‫יצאה‬ ‫הד‬

‫באחד‬ ‫נפשו‬ ‫יצתה‬ ‫ע‬

‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫את הדבר‬ ‫לגמור‬ ‫הספיק‬ ‫לא‬ ‫ח‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫היה מאריך באחד‬ ‫ג‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫את הדבר‬ ‫הספיק‬ ‫לא‬ ‫אמרו‬ ‫י‬
‫<>‬ ‫נשמתו‬ ‫שיצאתה‬ ‫עד‬ ‫את הדבר‬ ‫לגמור‬ ‫הספיק‬ ‫לא‬ ‫אמרו‬ ‫ק‬
‫באחד‬ ‫נשמתו‬ ‫יצתה‬ ‫ר‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫את הדבר‬ ‫לגמור‬ ‫הספיק‬ ‫לא‬ ‫ו‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫את הדבר‬ ‫לגמור‬ ‫הספיק‬ ‫ולא‬ ‫א‪2‬‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫לומר‬ ‫הפסיק‬ ‫לא‬ ‫אמרו‬ ‫מ‬
‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫היה מאריך באח׳‬ ‫דש‬

‫באחד‬ ‫נשמתו‬ ‫שיצתה‬ ‫עד‬ ‫היה מאריך באח׳‬ ‫דו‬

‫‪line 9‬‬
‫א‬
‫פ‬
‫שיצתה נשמתך באחד‬ ‫עקיבה‬ ‫רבי‬ ‫אשרך‬ ‫ואמרה‬ ‫בת קול‬ ‫יצתה‬ ‫ת‬
‫ב‬
‫ה‬
‫הד‬

‫ע‬
‫שיצתה נשמתך באחד‬ ‫עקיבא‬ ‫ר׳‬ ‫אשריך‬ ‫בת קול‬ ‫יצתה‬ ‫ח‬
‫שיצתה נשמתך באחד‬ ‫עקיבה‬ ‫ר׳‬ ‫אשריך‬ ‫ואמרה‬ ‫בת קול‬ ‫יצתה‬ ‫ג‬
‫י‬
‫ק‬
‫שיצתה נשמתך באחד‬ ‫עקיבא‬ ‫ר׳‬ ‫אשריך‬ ‫ואמרה לו‬ ‫בת קול‬ ‫יצתה‬ ‫ר‬
‫ו‬
‫א‪2‬‬
‫מ‬
‫שיצתה נשמתך באח׳‬ ‫עקיבא‬ ‫ר׳‬ ‫אשריך‬ ‫ואמרה‬ ‫בת קול‬ ‫יצתה‬ ‫דש‬

‫שיצתה נשמתך באח׳‬ ‫עקיבא‬ ‫ר׳‬ ‫אשריך‬ ‫ואמרה‬ ‫בת קול‬ ‫יצתה‬ ‫דו‬

‫ ‪60‬‬ ‫‪ appears as a catchword at the bottom of the page but is missing in the‬נשמתו ‪The word‬‬
‫‪continuation of the text on the next page; the word clearly belongs to the original text.‬‬
‫? ‪WAS RABBI AQIVA A MARTYR‬‬ ‫‪341‬‬

‫‪line 10‬‬
‫זו תורה וזו שכרה‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הקב״ה‬ ‫א‬
‫זו תורה וזו שכרה‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הקב״ה‬ ‫פ‬
‫**‪61‬‬ ‫ת‬
‫זו תורה וזו שכרה‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הקדוש ברוך הוא‬ ‫ב‬
‫זו תורה וזו שכרה‬ ‫ש״ע‬ ‫רבו׳‬ ‫אמרו מלאכי השרת לפני הב׳ה‬ ‫ה‬
‫זו תורה וזו שכרה‬ ‫אמרו מלאכי השרת לפני הק׳ב׳ה׳‬ ‫הד‬

‫זו תורה וזו שכרה‬ ‫ש׳ ע׳‬ ‫רב׳‬ ‫אמרו מלאכי השרת לפני ה׳ב׳ה‬ ‫ע‬
‫**‬ ‫ח‬
‫זו תורה וזו שכרה‬ ‫שלעולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני המקום ברוך הוא‬ ‫ג‬
‫זו תורה וזו שכרה‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הק׳‬ ‫י‬
‫> תורה וזו שכ< >‬ ‫<‬ ‫ק‬
‫**‬ ‫ר‬
‫זו תורה וזו שכרה‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הק׳ב׳ה׳‬ ‫ו‬
‫זו תורה וזו שכרה‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הק׳ב׳ה׳‬ ‫א‪2‬‬
‫זו תורה וזו שכר׳‬ ‫של עולם‬ ‫רבונו‬ ‫אמרו מלאכי השרת לפני הקב״ה‬ ‫מ‬
‫זו תורה וזר שכרה‬ ‫אמרו מלאכי השרת לפני הקב״ה‬ ‫דש‬

‫זו תורה וזו שכרה‬ ‫אמרו מלאכי השרת לפני הקב״ה‬ ‫דו‬

‫‪line 11‬‬
‫חלקם בחיים‬ ‫אמ׳ להם‬ ‫ממתים‬ ‫י״י‬ ‫ממתים ידך‬ ‫א‬
‫מחלד‬
‫חלקם בחיים‬ ‫אמ׳ להם‬ ‫ממתים‬ ‫י״י‬ ‫ממתים ידך‬ ‫פ‬
‫מחלד‬
‫**‬ ‫ת‬
‫וגו׳‬ ‫ממתים‬ ‫ה׳‬ ‫ממתים ידך‬ ‫ב‬
‫חלקם בחיים‬ ‫אמ׳ להם‬ ‫ממתי׳ ידך‪62‬‬ ‫ה‬
‫חלקם בחיים‬ ‫אמר להם‬ ‫ממתים ידך‬ ‫הד‬

‫חלקם בחיים‬ ‫אמ׳ ל׳‬ ‫ממתים‬ ‫אמרו ממתים ידך ה׳‬ ‫ע‬
‫מחלד‬
‫**‬ ‫ח‬
‫חלקם בחיים‬ ‫אמר להם‬ ‫ג‬

‫ ‪61‬‬ ‫)‪ (all of which are citations of the narrative by medieval authors‬ת ‪ and‬ר ‪,‬ח ‪Text witnesses‬‬
‫‪—and do not‬יצתה בת קול ואמרה אשריך ר׳ עקיבא שיצאה נשמתך באחד—‪end at line 9‬‬
‫‪include the text to lines 10 through 13.‬‬
‫ ‪62‬‬ ‫ממתים ‪In both the MS and printed editions of Haggadot Hatalmud, following the words‬‬
‫פי׳ מידך היה לו למות ולא מיד ‪ the following comment found in Rashi, ad loc., is added:‬ידך‬
‫‪.‬בשר ודם‬
‫‪342‬‬ ‫‪mandel‬‬

‫בחיים‬ ‫א״ל הק׳‪63‬‬ ‫וג׳‬ ‫י״י‬ ‫ממתים ידך‬ ‫י‬

‫חלק< >‬ ‫< > ל< >‬ ‫וג< >‬ ‫י״י‬ ‫‪ > <64‬ידך‬ ‫ק‬
‫**‬ ‫ר‬
‫חלקם בחיים‪65‬‬ ‫אמ׳ להם‬ ‫ו‬
‫הק׳ב׳ה׳‬
‫חלקם בחיים‬ ‫ממתים‬ ‫י״י‬ ‫ממתים ידך‬ ‫א‪2‬‬
‫מחלד‬
‫חלקם בחיים‬ ‫אמ׳ להם‬ ‫י״י‬ ‫ממתים ידך‬ ‫מ‬
‫חלקם בחיים‬ ‫אמ׳ להם‬ ‫וגו׳‬ ‫ממתים‬ ‫י״י‬ ‫ממתים ידיך‬ ‫דש‬

‫חלקם בחים‬ ‫אמר להם‬ ‫וגו׳‬ ‫ממתים‬ ‫ה׳‬ ‫ממתים ידך‬ ‫דו‬

‫‪line 12‬‬
‫עקיב׳‬ ‫ר׳‬ ‫אשריך‬ ‫בת קול ואמ׳‬ ‫א יצתה‬
‫עקיבה‬ ‫ר׳‬ ‫אשריך‬ ‫לו‬ ‫בת קול ואמרה‬ ‫פ יצתה‬
‫ת **‬
‫שיצאת נשמתך באחד‬ ‫רבי עקיבא‬ ‫אשריך‬ ‫לו‬ ‫ואמרה‬ ‫בת קול‬ ‫ב יצתה‬
‫באחד‬ ‫שיצתה נפשך‬ ‫ר׳ עקיבא‬ ‫אשריך‬ ‫לו‬ ‫ואמרה‬ ‫בת קול‬ ‫ה יצתה‬
‫שיצתה נשמתך באחד‬ ‫ר׳ עקיבא‬ ‫אשריך‬ ‫לו‬ ‫ואמר׳‬ ‫בת קול‬ ‫הד יצתה‬
‫שיצאתה נשמתך באחד‬ ‫ר׳ עקיבה‬ ‫אשריך‬ ‫לו‬ ‫ואמרה‬ ‫בת קול‬ ‫ע יצאתה‬
‫ח **‬
‫ג ***‪66‬‬
‫עקיב׳‬ ‫ר׳‬ ‫אשריך‬ ‫לו‬ ‫בת קול ואמרה‬ ‫י יצתה‬
‫>בא‬ ‫ר<‬ ‫אשריך‬ ‫לו‬ ‫> קול ו< >ה‬ ‫ק <‬
‫ר **‬
‫עקיב׳‬ ‫ר׳‬ ‫אשריך‬ ‫לו‬ ‫ואמר׳‬ ‫בת קול‬ ‫ו יצתה‬
‫באחד‬ ‫שיצתה נשמתך‬ ‫עק< >‬ ‫ר׳‬ ‫אשריך‬ ‫ואמרה‬ ‫בת קול‬ ‫א‪> < 2‬ה‬
‫עקיבא‬ ‫אשריך‬ ‫ואמרה‬ ‫בת קול‬ ‫מ יצתה‬
‫עקיבא‬ ‫ר׳‬ ‫אשרך‬ ‫ואמרה‬ ‫בת קול‬ ‫דש יצתה‬
‫עקיבא‬ ‫ר׳‬ ‫אשרך‬ ‫ואמרה‬ ‫בת קול‬ ‫דו יצתה‬

‫ ‪63‬‬ ‫‪.‬ו ‪, which is missing here; but cf. MS‬חלקם ‪ is perhaps a corruption of‬הק׳‬
‫ ‪64‬‬ ‫אמר להם חלקם בחיים ‪The first and second halves of the line are transposed in the MS:‬‬
‫‪.‬״<אמרו [?] ממתים> ידך י״י וג׳״ ‪appears before‬‬
‫ ‪65‬‬ ‫)‪ (from a passage below‬שאין ‪ the scribe mistakenly wrote the word‬בחיים ‪After the word‬‬
‫‪and marked it with points to designate deletion.‬‬
‫ ‪66‬‬ ‫‪,‬חלקם בחיים ‪ (citation in Midrash Hagadol) ends at line 11 with the words‬ג ‪Text witness‬‬
‫‪and is missing the text to lines 12 and 13.‬‬
WAS RABBI AQIVA A MARTYR ? 343

line 13
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫א שאתה‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫פ שאתה‬
** ‫ת‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫ב ואתה‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫ה ואתה‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫הד ואתה‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫ע ואתה‬
** ‫ח‬
*** ‫ג‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫שאתה‬ ‫י‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫ק שאתה‬
** ‫ר‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫שאתה‬ ‫ו‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫ ושאתה‬2‫א‬
‫העולם הבא‬ ‫לחיי‬ ‫מ שאתה‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫דש שאתה‬
‫העולם הבא‬ ‫לחיי‬ ‫מזומן‬ ‫דו שאתה‬

Legend
Sigla of Manuscripts and Text Witnesses
‫ א‬MS Oxford, Bodleian Library, Opp. Add. fol. 23 (catalogue Neubauer, no. 366)
2‫ א‬Oxford, Bodleian Library, Heb. b. 1/1 (catalogue Neubauer–Cowley, no. 2673.1)
(Genizah fragment)
‫ ב‬R. Isaac Aboab, Menorat Hamaor (ed. Horev–Katzenelebogen; Jerusalem 1961)
pp. 527–8
‫ ג‬ Midrash Hagadol, Deuteronomy (ed. Fish; Jerusalem 1975) p. 130
‫ דו‬ Babylonian Talmud, Tractate Berakhot, Venice ed. 1520
‫ דש‬ Babylonian Talmud, Tractate Berakhot, Soncino ed. 1484
‫ ה‬Haggadot Hatalmud, MS Parma, Palatine Library 3010 (catalogue De Rossi,
no. 156)
‫ הד‬ Haggadot Hatalmud, editio princeps, Constantinople 1511
‫ ו‬ fragment, private collection of Vittorio Vivanti, Rome (presently in the State
Archives of Viterbo)
‫ ח‬Rabbenu Hannanel, Commentary to Bavli Berakhot (ed. Metzger; Jerusalem
1990) pp. 139–40
‫ י‬ Yalqut Shimoni, MS Oxford, Bodleian Library (catalogue Neubauer 2637), Torah,
remez 837
‫ מ‬MS Munich, Bavarian State Library 95
344 mandel

‫ ע‬ R. Jacob ibn Habib, Ein Yaakov, editio princeps, Salonica 1516

‫ פ‬ MS Paris, National Library 671.4
‫ ק‬ Cambridge University Library (CUL), T-S N.S. 291.76 (Genizah fragment; collec-
tion of aggadot)
‫ ר‬ Raymond Martini, Pugio Fidei, MS Paris, St. Geneviève, no. 1405, p. 106
‫ ת‬ Sefer Pitron Torah (ed. Urbach; Jerusalem 1978), pp. 282–3

Editorial notations
< > lacuna or obliteration
[ ] scribal addition in margin or between lines: second hand
( ) scribal deletion: second hand

Appendix II

The Literary Pastiche of the Bavli Narrative of Rabbi Aqiva

The following list contains references to parallel rabbinic passages where many of the
phrases making up the Bavli narrative appear. The list is not exhaustive, and I do not
claim that these particular passages served as sources for the narrative in bBer 61b.
Nonetheless, the many parallels suggest that known phrases were used by the narrator
of this tale in his reworking of the Yerushalmi narrative.67

“They combed his flesh with iron combs” "‫"והיו סורקין את בשרו במסרקות של ברזל‬
.‫ ואי לאו—מסריקנא לבשרייכו במסרקי דפרזלי‬,‫ אי אמריתו לי—מוטב‬:‫אמר להו‬
(bGit 57b; cf. bMen 29b)

68"‫"והיה מכוון דעתו לקבל עול מלכות שמים באהבה‬

“He set his mind [= concentrated] to ‘accept the yoke of the kingdom of heaven’
[= recite the Shema] with love”
‫ ודילמא‬:‫ אמר ליה רבי ירמיה לרבי זירא‬.‫ אני ראיתי את רבי ינאי דצלי והדר צלי‬:‫אמר רבי יוחנן‬
!‫ חזי מאן גברא רבה דקמסהיד עליה‬:‫ ולבסוף כוון דעתיה? אמר ליה‬,‫מעיקרא לא כוון דעתיה‬
(bBer 30b)
‫״היושבת בגנים חברים מקשיבים״—כשישראל נכנסין לבתי כנסיות וקורין קריאת שמע בכיוון‬
,‫ כשאתם קורין‬,‫ ״היושבת בגנים״‬:‫ הקב״ה אומר להם‬,‫ בקול אחד בדעה [אחת] וטעם אחד‬,‫הדעת‬
‫ השמיעני״; אבל כשישראל קורין קריאת שמע בטירוף‬,‫ ״מקשיבים לקולך‬,‫״חברים״ אני ופמליא שלי‬
‫ ״ברח‬:‫ רוח הקדש צווחת ואומרת‬,‫ זה מקדים וזה מאחר ואינם מכוונין דעתם בקריאת שמע‬,‫הדעת‬
(SongR 8:13) ". . . ‫דודי ודמה לך לצבי‬

67 See also Tropper, Like Clay in the Hands of the Potter, 111–54, and Excursus B there, 208–11,
and the references in n. 48 above.
68 See also n. 42 above.
‫? ‪WAS RABBI AQIVA A MARTYR‬‬ ‫‪345‬‬

‫”?‪“His students said to him: Our teacher—so far‬‬ ‫"אמרו לו תלמידיו‪ :‬רבינו—עד‬
‫כאן?"‬
‫אמרו לו‪ :‬רבינו‪ ,‬ברכנו! אמר להם‪ :‬יהי רצון שתהא מורא שמים עליכם כמורא בשר ודם‪ .‬אמרו‬
‫לו תלמידיו‪ :‬עד כאן?—אמר להם‪ :‬ולואי! תדעו‪ ,‬כשאדם עובר עבירה אומר‪ :‬שלא יראני אדם‪.‬‬
‫)‪(bBer 28b‬‬

‫”‪“All my days I would teach‬‬ ‫״כל ימיי הייתי דורש״‬

‫בו ביום דרש ר׳ יהושע בן הורקנוס לא עבד איוב את הקדוש ברוך הוא אלא מאהבה‬
‫שנאמר‪ ,‬״הן יקטלני לו איחל״ (איוב יג טו)‪ .‬ועדיין הדבר שקול‪ :‬״לו אני מצפה״ או ״איני‬
‫מצפה״? תלמוד לומר‪ :‬״עד אגוע לא אסיר תומתי ממני״ (שם כז ה)—מלמד שמאהבה עשה‪.‬‬
‫אמר רבי יהושע‪ :‬מי יגלה עפר מעיניך‪ ,‬רבן יוחנן בן זכאי! שהיית דורש כל ימיך שלא עבד‬
‫איוב את המקום אלא מיראה‪ ,‬שנאמר‪ :‬״איש תם וישר ירא אלהים וסר מרע״ (שם א א)‪,‬‬
‫והלא יהושע תלמיד תלמידך למד שמאהבה עשה! )‪(mSot 5:5‬‬
‫מעשה בביתסי אחד שהקטיר עד שהוא בחוץ‪ . . . ‬כשיצא אמר לאביו‪ :‬כל ימיכם הייתם דורשין‬
‫ולא עושין‪ ,‬עד שעמדתי ועשיתי‪ .‬אמר לו‪ :‬אף על פי שאנו דורשין ואין אנו עושין שומעין אנו לדברי‬
‫חכמים‪(yYom 1:5, 39a–b; tYom 1:8]) .‬‬

‫”‪“All my days I was troubled by this verse‬‬ ‫״כל ימי הייתי מצטער על פסוק זה״‬
‫אמר לו‪ :‬כל ימי הייתי מצטער על המקרא הזה ״כי בענן אראה על הכפרת״ )ויק׳ יט‪ ,‬ב(‪.‬‬
‫[אמרתי‪ ,‬מתי יבוא לידי ואקיימנו?] עכשיו שבא לידי—לא אקיימנו? )‪(bYom 19b—see below‬‬

‫”‪“Even if He takes your soul‬‬ ‫"אפילו [הוא] נוטל את נפשך"‬

‫חייב אדם לברך על הרעה כשם שהוא מברך על הטובה‪ ,‬שנאמר‪ :‬״ואהבת את ה׳ אלהיך בכל לבבך‬
‫ובכל נפשך ובכל מאדך״ (דבר׳ ו‪ ,‬ה)‪ . . . ‬״ובכל נפשך״—אפילו הוא נוטל את נפשך‪(mBer 9:5) .‬‬

‫"[אמרתי‪ ,‬מתי יבוא לידי ואקיימנו?] ועכשיו שבא לידי לא אקיימנו?"‬

‫‪“[I said, When would this come into my hands so that I might fulfill it?] And now that‬‬
‫”?‪it has come to my hands, shall I not fulfill it‬‬
‫תנו רבנן‪ :‬מעשה בצדוקי אחד שהתקין מבחוץ והכניס‪ .‬ביציאתו היה שמח שמחה גדולה‪ .‬פגע בו‬
‫אביו‪ ,‬אמר לו‪ :‬בני‪ ,‬אף על פי שצדוקין אנו—מתיראין אנו מן הפרושים‪ .‬אמר לו‪ :‬כל ימי הייתי מצטער‬
‫על המקרא הזה ״כי בענן אראה על הכפרת״ (ויקרא טז ב)‪[ .‬אמרתי‪ ,‬מתי יבוא לידי ואקיימנו‪].‬‬
‫‪ (bYom 19b)69‬עכשיו שבא לידי—לא אקיימנו?‬

‫‪69 The bracketed phrase, which appears in Versions B and C of our narrative, appears in‬‬
‫‪only some of the manuscripts to bYom: MS Munich 95; MS Vatican, ebr. 134; MS Oxford,‬‬
‫‪Bodleian Library, Opp. Add. folio 23 (catalogue Neubauer, no. 366); and MS London,‬‬
‫‪British Library, Harley 5508 (catalogue Margoliouth, no. 400). It is missing in MS Munich 6‬‬
‫‪and MS JTS ENA 270.6. The last two manuscripts are known to preserve good and original‬‬
‫‪readings.‬‬
‫‪346‬‬ ‫‪mandel‬‬

‫"לא הספיק לגמור את הדבר עד שיצתה נשמתו"‬

‫”‪“He was not able to finish the utterance before his soul expired‬‬
‫מעשה בכהן אחד שהיה מתעסק וראה הרצפה שהיא משונה מחברותיה‪ .‬בא ואמר לחברו‪ ,‬ולא‬
‫הספיק לגמור את הדבר עד שיצתה נשמתו‪ ,‬וידעו ביחוד ששם הארון נגנז‪(mSheq 6:2) .‬‬
‫מעשה בבן דמא בן אחותו של ר׳ ישמעאל שהכישו נחש ובא יעקב איש כפר סכניא לרפאותו‪,‬‬
‫ולא הניחו ר׳ ישמעאל‪ ,‬ואמר לו‪ :‬ישמעאל אחי‪ ,‬הנח לו וארפא ממנו‪ ,‬ואני אביא מקרא מן‬
‫התורה שהוא מותר ולא הספיק לגמור את הדבר עד שיצתה נשמתו ומת‪ .‬קרא עליו ר׳ ישמעאל‪:‬‬
‫אשריך בן דמא‪ ,‬שגופך טהור ויצתה נשמתך בטהרה‪ ,‬ולא עברת על דברי חביריך‪(bAZ 57b) .‬‬

‫”’‪“He extended [the recitation of the word] ‘one‬‬ ‫"והיה מאריך באחד"‬
‫תניא‪ :‬סומכוס אומר‪ :‬כל המאריך באחד—מאריכין לו ימיו ושנותיו‪(bBer 13b) .‬‬

‫"אשריך‪ . . . ‬שיצתה נשמתך באחד"‬

‫”’‪“Happy are you . . . that your soul has expired with [the recitation of the word] ‘one‬‬
‫קרא עליו ר׳ ישמעאל‪ :‬אשריך בן דמא‪ ,‬שגופך טהור ויצתה נשמתך בטהרה‪(bAZ 57b) .‬‬

‫"יצתה בת קול ואמרה‪ . . . ‬אשריך‪ . . . ‬שיצתה נשמתך באחד"‬

‫‪“A bat qol went forth and said, . . . Happy are you . . . that your soul expired with [the‬‬
‫”’‪recitation of the word] ‘one‬‬
‫יצאת בת קול ואמרה‪ :‬אשריך רבה בר נחמני שגופך טהור‪ ,‬ויצאתה נשמתך בטהור‪(bBM 86a) .‬‬

‫”?‪“This is Torah and this is its reward‬‬ ‫"זו תורה וזו שכרה?"‬
‫אמר לפניו‪ :‬רבונו של עולם‪ ,‬הראיתני תורתו—הראני שכרו‪ .‬אמר לו‪ :‬חזור לאחורך‪ .‬חזר לאחוריו‪,‬‬
‫ראה ששוקלין בשרו במקולין‪ ,‬אמר לפניו‪ :‬רבש״ע‪ ,‬זו תורה וזו שכרה? )‪(bMen 29b‬‬

‫"יצתה בת קול ואמרה‪ . . . ‬אשריך‪ . . . ‬שאתה מזומן לחיי העולם הבא"‬

‫‪“A bat qol went forth and said, . . . Happy are you . . . that you are destined for [/invited‬‬
‫”‪into] the World to Come‬‬
‫יצתה בת קול ואמרה‪ :‬אשריך הגמון שאתה מזומן ל‪s‬חיי העולם הבא‪(70bTaan 29a) .‬‬
‫יצתה בת קול ואמרה‪ :‬ר׳ אלעזר בן דורדיא מזומן לחיי העולם הבא‪(bAZ 17a) .‬‬
‫ההוא יומא דאשכבתיה דרבי‪ ,‬נפקא בת קלא ואמרה‪ :‬כל דהוה באשכבתיה דרבי מזומן הוא לחיי‬
‫העוה״ב‪ .‬ההוא כובס כל יומא הוה אתי קמיה‪ ,‬ההוא יומא לא אתא‪ ,‬כיון דשמע הכי‪ ,‬סליק לאיגרא‬
‫ונפל לארעא ומית‪ ,‬יצתה בת קול ואמרה‪ :‬אף ההוא כובס מזומן הוא לחיי העולם הבא‪(bKet 103b) .‬‬

‫‪.‬אותו הגמון מזומן [‪/‬מתוקן] לחיי העולם הבא ‪70 So in MS Göttingen 3; in other manuscripts:‬‬
WAS RABBI AQIVA A MARTYR ? 347

Appendix III

Boyarin’s Interpretation of the Rabbi Aqiva Narratives

In an article published in 1989, Daniel Boyarin analyzed the Yerushalmi and Bavli nar-
ratives discussed in the present paper within the context of a study of midrash and its
relationship to text and history.71 While Boyarin’s analysis interprets both narratives
in the context of other midrashic texts,72 the two talmudic narratives are central to a
thesis elaborated upon in Boyarin’s later work, as will be seen below. These narratives
present, in Boyarin’s words, “two . . . stages in cultural history—the history of an idea,”
the idea of “erotic, mystic death.”73 Through Rabbi Aqiva’s midrashic reading of Deut
6:5 (“ ‘You shall love the Lord your God with . . . all your soul’—even though He takes
your soul”), among other Scriptural passages, the rabbinic figure is portrayed in these
narratives as “discovering that dying is the way to fulfill the commandment of loving
God.”74 In the Yerushalmi text, which Boyarin sees as chronologically preceding the
Bavli narrative, Rabbi Aqiva is “[caught] in the act, as it were, of discovering” this truth;
the Bavli narrative is then a culmination of the “join[ing] of Eros and Thanatos,” of
transforming martyrdom from what was considered in previous centuries as a “nega-
tive command­ment” (refusing to worship idolatry) into a “positive commandment,” in
actively “dying for [the love of] God”:

R. Akiva died for the love of God; indeed he died because he held that this was
the only way to fulfill the commandment “to love the Lord with all your soul.”75

71 “ ‘Language Inscribed by History on the Bodies of Living Beings’: Midrash and Martyrdom,”
 

Representations 25 (1989) 139–51. The paper later appeared as Chapter 8 in Boyarin’s book,
Intertextuality and the Reading of Midrash (Bloomington 1990) 117–129. The chapter in
the book, titled “Between Intertextuality and History: The Martyrdom of Rabbi Akiva,” is
almost a verbatim copy of the article; in the following, I will refer to the text according to
the page numbers of the chapter in the book. A Hebrew translation (with some changes
and additions) was published as “ha-midrash ve-ha-ma‘aseh: ‘al ha-heqer ha-histori shel
sifrut hazal,” in: S. Friedman (ed.), Saul Lieberman Memorial Volume (New York and
Jerusalem 1993) 105–17. At the conclusion of the Hebrew article Boyarin adds comments
regarding the variants in the manuscript versions of the Talmudic narratives; see n. 83
below.
72 In particular, MdRY beshalah 3, on Exod 15:2 (“this is my God and I will beautify Him”).
73 Boyarin, Intertextuality, 127. See below for a discussion concerning the theses presented
in Boyarin’s later published work in relationship to this analysis, specifically in his books
Dying for God: Martyrdom and the Making of Christianity and Judaism (Stanford 1999), and
Border Lines: The Partition of Judaeo-Christianity (Philadelphia 2004).
74 Boyarin, Intertextuality, 127.
75 Boyarin, Intertextuality, 125; cf. 126–7.
348 mandel

Boyarin’s reading of these narratives, and in particular the Bavli version, puts them
squarely within the genre of martyrology; indeed, they serve as foundation texts for
his assumption concerning the creation of a Jewish concept of martyrology showing
remarkable similarities to Christian martyrological texts.76 Clearly, this thesis runs
counter to that suggested in the present study, which was reached through an analysis
of these same texts. It is therefore important to address Boyarin’s thesis through an
exploration of the textual basis upon which it rests.
His text of the Bavli narrative is cited here in full in its English and later Hebrew
versions:77

In the hour that they took R. Aqiva out [to be executed], his disciples
said to him, “Our teacher, so far? [i.e. is this necessary]” He said to
them, “All of my life I was troubled by this verse, ‘And thou shalt love the Lord with
all thy soul’—even though He takes your soul, and I said, when will it come to
my hand that I may fulfill it? Now that it is come to my hand, shall I not fulfill it?”

‫שעה שהוציאו את ר׳ עקיבא אמרו לו תלמידיו רבינו עד כאן אמר להם כל ימי‬
‫הייתי מצטער על פסוק זה ב כ ל נ פ ש ך אפילו הוא נוטל את נשמתך אמרתי‬
‫מתי יבא לידי ואקיימנו ועכשיו שבא לידי לא אקיימנו היה מאריך באחד עד‬
‫שיצתה נשמתו באחד יצתה בת קול ואמרה אשריך ר״ע שיצאה נשמתך באחד אמרו‬
‫מלאכי השרת לפני הקב״ה זו תורה וזו שכרה מ מ ת י ם י ד ך ה' מ מ ת י ם וגו׳‬
‫יד) אמר להם ח ל ק ם ב ח י י ם יצתה בת קול ואמרה אשריך ר״ע שאתה‬:‫(תהלים יז‬
‫מזומן לחיי העוה״ב‬

What is particularly significant in this text is the fact that the query of the disciples
to Rabbi Aqiva appears directly after the exposition declaring his being taken out for
execution; there is no mention of the torture or of Rabbi Aqiva’s recital of the Shema at
this time. This means that the disciples’ alarmed question, “Our teacher, so far?,” must

76 Boyarin does not make historical claims for either narrative, preferring to view them as
cultural products to be understood in the context of other literary texts, providing data for
a cultural history of the Jews in the early centuries of the common era.
77 The English passage is found in the article, “Language,” on pp. 146–7, and is repeated, in
identical form, in Intertextuality, on p. 125. The Hebrew text, which includes the “second
act” of the narrative (see below), appears in the Hebrew article, “ha-midrash ve-ha-
ma‘aseh,” 114. (It should be noted that the passage is referenced in the English article
erroneously as “Berakot 66a” [instead of 61b]; in the chapter of the book Intertextuality
no reference is given. This was corrected in the version presented in the Hebrew
article, as well as in the recent Hebrew translation of Intertextuality [Midrash Tannaim:
Intertextualiut ukeri‘at Mekhilta (Jerusalem 2011) 196–7]).
WAS RABBI AQIVA A MARTYR ? 349

be taken to be a challenge to the very act of his impending death, as Boyarin indeed
explains in a bracketed addition: [“i.e. is this necessary?”], meaning “is this [acquies-
cence to your] execution necessary?” Rabbi Aqiva’s answer, based upon his midrashic
comment to Deut 6:5, thus becomes a forceful argument for the “joining of Eros and
Thanatos”; Rabbi Aqiva’s message to his students is: “Death is not only required of me
at this time [“it is necessary”], but all the more: I have actively sought out just this
martyrdom all my life as a fulfillment of the commandment to love God.” While in the
Bavli narrative, Rabbi Aqiva “already knew from before what it was he had to do, and
was just waiting for the opportunity,” in the earlier Yerushalmi narrative he “discover[s]
that dying is the way to fulfill the commandment of loving God.”78
We note that Boyarin’s text, in the second part of the cited passage at least (from
Aqiva’s response to his students), follows Version B or that of the printed edition of
the Babylonian Talmud (the two versions are identical here).79 But whence the strange
beginning of the narrative, which, in deleting both the torture and the mention of
the recital of the Shema, prepares the way for an interesting and novel view of Rabbi
Aqiva’s interpretation and understanding of Deut 6:5? Is this based on a manuscript
version not discussed above?80 And what is the meaning of the phrase appearing in
square brackets: “[to be executed]”? Is this not part of the narrative (‫ ?)להריגה‬Why is
it bracketed?
The answer concerning the origin of Boyarin’s text is as simple as it is shocking.
As easily demonstrated by a quick perusal of the page from the standard Vilna edi-
tion of Tractate Berakhot reproduced below,81 Boyarin’s text of the Bavli is culled
directly from this version, except that in copying the Hebrew text (from which
he made his English translation, and which he presented in the Hebrew version
of the study) he mistakenly skipped a full line, moving directly from the words
‫ר׳ עקיבא‬, at the end of the sixth full line, to the query of the disciples, ‫אמרו לו תלמידיו‬, at
the beginning of the eighth full line.

78 Boyarin, Intertextuality, 127.

79 The phrases in Boyarin’s cited text, “All of my life I was troubled by this verse . . . I said,
when will it come to my hand that I may fulfill it,” reflect both versions; see the synopsis
of textual versions in Appendix I, lines 5 and 7.
80 Elsewhere Boyarin shows awareness of the importance of manuscripts, and, indeed,
states that he has modified the Mekhilta text (see n. 72 above) “where my manuscripts
have a better reading” (Intertextuality, 156, n. 6).
81 The pagination and lines are precisely the same in all standard editions of the Bavli
(based on the 19th-century Vilna edition) to this page.
350 mandel

This hiatus accounts for the deletion of the elements of the torture, the mention of the
time for the recital of the Shema and the recital itself, all of which are found in the sev-
enth full line. The bracketed words in Boyarin’s English text, “[to be executed],” leave
no doubt as to what has happened in the process of the transcription: these words,
of course, are a translation of the word ‫להריגה‬, which in no text version, manuscript
or print, is missing; indeed, the statement ‫ בשעה שהוציאו את ר׳ עקיבא‬is incomplete
without this word. But the word must have been lost in Boyarin’s transcribed (Hebrew)
text, as it is the first word in that missing seventh full line. Since the phrase “In the hour
that they took R. Aqiva out” (the translation of the last five words in the sixth full line)
makes little sense, Boyarin was forced to make an editorial “emendation,” adding the
words “to be executed” in order to complete the sense of the sentence. Being true to
“his text,” he placed the completion of the elliptical phrase in square brackets to note
his editorial “addition.”
As can be seen by comparing Boyarin’s recorded text of the Bavli narrative in the
later Hebrew version of the article as found above, the line is still missing there.82
Boyarin provides no ellipsis to mark a possible “jump” in his text; it purports to be the
complete and proper Hebrew text of the Bavli narrative.83

82 As noted above, the Hebrew version of this study was published three and four years
after the publications of the English versions of the book and article, respectively. As
mentioned, the Hebrew version of the narrative is more complete than that of the English
version, as it includes the continuation of the narrative (see below); however, it is still
missing the seventh full line.
83 In an added note to the Hebrew version of this paper (ha-midrash ve-ha-ma’aseh, 116–7,
n. 34), Boyarin discusses the textual variants of the Yerushalmi and Bavli texts, citing
Safrai’s philological analysis in his study in Zion 44 (1981) (“Martyrdom in the Teachings
of the Tannaim,” 28–42; Safrai’s philological discussion is on pp. 37–8 in the version in
Zion, and on pp. 156–7 of the English version), which agrees basically with my conclusions
WAS RABBI AQIVA A MARTYR ? 351

This unfortunate and faulty understanding of the Bavli narrative, while serious
enough in itself, as it provides a forceful yet erroneous “proof” for Boyarin’s thesis of
Rabbi Aqiva’s quest for martyrdom, has had wider ramifications. For, as mentioned
above, the understanding of the cultural creation of an idea of martyrdom during the
foundational period from the second through the fifth centuries of the Common Era
(during which the original Yerushalmi and Bavli narratives must have been formed)
lies at the basis of Boyarin’s later studies of a shared cultural outlook between Jewish
and Christian cultures of this period. Indeed, the very same texts are presented in
Boyarin’s book Dying for God: Martyrdom and the Making of Christianity and Judaism
(Stanford 1999), published nine years after the publication of Intertextuality. This was
a seminal work, in which Boyarin disclosed what would become a dominant theory in
his own subsequent research, influencing also the studies of other contemporary his-
torians of Judaism and Christianity of the first centuries of the Christian era: namely,
that the older paradigm of the “parting of the ways” between the two religions and
cultures, previously assumed by historians to have occurred by the second century, did
not occur at that time, but was, rather, a significantly later phenomenon, as “border-
lines” were actively put in place by proponents of both cultures.84 At the culmination

in the present study. He agrees with Safrai’s consideration of the spurious nature of the
added passages in the Yerushalmi narrative (found only in the Berakhot text but missing
in the parallel text from ySot), but defends the possibility of an “authentic” aspect of the
martyrological additions to the Bavli narrative, where the authenticity is not of Rabbi
Aqiva’s historical situation but of the cultural expression of the narrative. This is of course
true: as I have noted in my discussion above, the narrative containing the additions indeed
expresses a martyrological consciousness, although not necessarily one consonant with
Christian martyrologies of the second to fourth centuries. The important issue, discussed
above, is precisely at what stage and under which influences these passages were added.
Later Byzantine influences, which, as I have suggested in the present study, may have
occurred in the later transmission of the Bavli narrative, would add nothing to our
understanding of the relationships between the Jewish and Christian cultures during the
early centuries of the first millennium.
84 The idea of the “blurring” of borders between the Jewish and early Christian cultures is
the basis for Boyarin’s later work, Border Lines. Although the idea of the commonality
of martyrdom is not prominent in this latter work, a clear line of thought concerning
the “borderlines,” and the lack thereof, between early Christianity and Judaism can be
drawn to this later work from the study of “shared” concepts of martyrology among early
Christians and Jews as presented in Dying for God. See Dying for God, “Introduction:
When Christians Were Jews: On Judeo-Christian Origins,” 1–21, and especially
pp. 6–7 (“The So-Called ‘Parting of the Ways’ ”) and pp. 16–9 (“Living on Borderlines”).
 

Note Boyarin’s emphasis there in his summary of the fourth chapter of the book which
contains the discussion of the Aqivan narratives (“The Plan of the Essay”): “Since the
entire passage that is read in the first three chapters hovers around the fraught question of
352 mandel

of this book, Boyarin reproduces texts and discussions from his earlier research,
including the Bavli and Yerushalmi narratives of Rabbi Aqiva’s presumed martyrdom.85
The Bavli narrative is cited (in English) here, as in the Hebrew version cited above, at
greater length, as it includes the second part of the passage concerning the cry of the
angels and the bat qol.86 Although Boyarin notes there that he has produced the text of
the Oxford manuscript,87 the first part of the narrative is again copied precisely from
the erroneous texts of the previous decade (reflecting the printed edition’s text, sans
seventh full line), including the bizarre bracketed words, “[to be executed].”88

martyrdom, in the fourth chapter, I . . . enter a more directly historiographical mode. The
major motif of this chapter is the entanglements of rabbinic Judaism and Christianity
with the discourse of martyrdom and its role in helping them invent themselves as
separate entities” (20, my emphasis). Cf. also n. 89 below. The influence of Boyarin’s study
in Dying for God on other cultural historians is notable, for example, in Judith M. Lieu’s
work; in particular, see her book Christian Identity in the Jewish and Graeco-Roman World
(Oxford 2004—published in the same year as Boyarin’s Border Lines). In a central chapter
of this book (“Boundaries,” 98–146), Lieu discusses, in terms strikingly similar to those of
Boyarin, the areas where the “crossing of boundaries” between Jewish and early Christian
cultures of the first centuries CE may be found, and similarly questions the adequacy of
current conceptions of the early “parting of the ways” between the two cultures. Lieu
explicitly expresses her debt to Boyarin in the concluding chapter, 307 n. 19: “My thinking
about this [i.e. the separate entities of Judaism and Christianity in the early centuries
of the current era and the question of the ‘parting of the ways’] has been stimulated by
Daniel Boyarin, both in conversation and in Dying for God.” It should be noted that Lieu’s
expertise is in the early Christian texts, and it is mainly from the perspective of these
texts that she draws her conclusions. See also the collection of papers, A. H. Becker and
A. Y. Reed (eds.), The Ways that Never Parted: Jews and Christians in Late Antiquity and
the Early Middle Ages (Tübingen 2003, Minneapolis 20072), where Boyarin’s discussions in
Dying for God and related articles are cited numerous times.
85 Chapter 4, “Whose Martyrdom Is This, Anyway?” 93–126. This chapter, as well as other
parts of the book, appeared previously as “Martyrdom and the Making of Christianity
and Judaism,” Journal of Early Christian Studies 6 (1998), 577–627 (see the asterisked note
at the beginning of that paper announcing “the forthcoming monograph, Dying for God,”
and see there 605, n. 90).
86 Boyarin, Dying for God, 106. In the earlier version of this chapter (“Martyrdom and the
Making of Christianity and Judaism,” 605), the text appears precisely as it appeared in the
versions of Intertextuality, without the additional second “act” (with the exception of a
corrected reference to “Berakhot 61b”).
87 The reference to the Oxford manuscript (“Opp. Add. fol. 23”) appears at the end of the
citation in Dying for God, 106. Some of the features of the manuscript version appear
in this translation, but the major part of the beginning of the cited text is not from the
manuscript version (see following note).
88 As noted above (see n. 86 above), the same faulty text appears in the previously published
article, which apparently was then copied directly into the prepared text of the book, to
WAS RABBI AQIVA A MARTYR ? 353

In sum, Boyarin’s study of early Jewish so-called martyrological texts, which

not only becomes the pivot for his work on martyrdom89 but also lays the founda-
tion for his subsequent research describing the gradual laying down of “border-
lines” between the Jewish and Christian religious cultures during the second to fifth
centuries CE, is not supported by the rabbinic texts. Borderlines should not be erected
on missing lines.90

which was then added the passage of the “second act” from the Oxford manuscript, thus
causing a hybrid text (see the previous note). In the recent Hebrew translation of the
book Intertextuality (Midrash Tannaim, 196–7), the faulty text is finally rectified. However
in this edition, where the English text of the book is followed slavishly in translation,
Boyarin again reverts to the standard text of the printed versions of the Bavli (and not
that of the Oxford manuscript, without including the second “act”) without further
explanation.
89 The centrality of the “shared concept of martyrdom” is emphasized by Boyarin in
the Hebrew summary of his thesis of Dying for God, published as “mashehu al toledot
ha-marterion be-yisrael,” in: D. Boyarin et al. (eds.), Atara l’Haim: Studies in the Talmud and
Medieval Rabbinic Literature in Honor of Professor Haim Zalman Dimitrovsky (Jerusalem
2000) 3–27 [Hebrew]. See also his recapitulation and summary of the thesis of the book
Dying for God, in idem, “Semantic Differences; or, ‘Judaism’/‘Christianity’,” in: Becker
and Reed (eds.), The Ways That Never Parted, 74. The Hebrew text of the Bavli narrative
presented in the above-mentioned Hebrew article (p. 16 there) faithfully represents the
version of the Oxford manuscript (see the reference there on p. 13); nonetheless, Boyarin
makes a point of saying that this story is “vital [‫ ]חיוני‬for my thesis in this paper, for it is
the clearest expression of the Rabbinic consciousness of martyrology (‫( )אצל חז״ל‬p. 16;
my translation, emphasis added). As I have shown above, the text of the narrative in the
Oxford manuscript contains no explicit martyrological elements.
90 While this is not the place for an extended critique of Boyarin’s work, it should be noted
(and this is pertinent to the present discussion) that his carelessness is not confined to
transcriptions and faulty references, but is evident also in his translation of Hebrew and
Aramaic. Thus, in all his English citations of the Yerushalmi narrative, Boyarin erroneously
translates the Palestinian Aramaic word ‫ חרש‬as “deaf” (Intertextuality, 126; Dying for God,
108; similarly in the versions of these chapters previously published as journal articles).
The correct translation, as noted in all translations and lexicons of the Yerushalmi
passage, is “sorcerer.” While this error was corrected in the translation of the passage in
the recent Hebrew edition of Intertextuality (Midrash Tannaim, 198), it is unfortunate
that this misunderstanding was repeated in English versions of Boyarin’s discussions, as it
misrepresents what may be construed as a historically significant aspect of Tinieus Rufus’
claim against Rabbi Aqiva; see above in my discussion of this text, and cf. Lieberman’s
comment cited there, n. 15 above.
Index of Sources

Hebrew Bible 6:2 220–1, 224 2 Samuel

9:3 181 7:23 292
Genesis 11:16–7 298 24:16 238
2:10 171 11:24 299
2:14 24, 160, 166 11:25 298–300 1 Kings
4:23 289, 291–5 11:26 298–300 1:33 169, 289, 291,
10:2 302 11:31 298 293
12:5 108–9 15:16 87 4:3 232, 237, 241
15:9 65 21:14 29 11:16 81
19:24 286–9, 293, 30:3 220–1, 225 18:4 239–40
298 22:20 235
21:33 110 Deuteronomy
31:30 220 1:16 90–1 2 Kings
40:15 220 4:1 3 2:12 45, 49, 76–7
4:11 151 18:11–12 3–4
Exodus 4:25–27 3 18:32 171
12:23 238 4:35 38–9, 41 19:35 238
15:2 317, 347 5 147 23:6 136
16:4 288, 290 6:4 333 24:14 29
19:17 147, 151–2 6:5 308, 310–1, 313, 24:19 4
22:17 23, 36, 37, 316–7, 319–20, 25:1–21 158
50–1,57, 79 325, 328–9, 333, 25:11 4
23:19 208 347, 349
24 147 6:13 311 Isaiah
34:26 208 14:21 208 14:23 159, 161, 163
18:9 53, 60–1 15:5 65
Leviticus 18:9–14 35 31:9 168
4:2–3 183–4 18:10 34, 55 34:9 288–9
12:2 213–4 18:20 60 45:1–7 158
12:5 213 27:26 228–31 58:13 174
12:6 214–5 29:13 154–5 59:21 28
12:7 214–5 34:3 169
19:16 210–1 Jeremiah
19:31 36 Joshua 20 237
20:27 36 10:16 239 20:6 234–5
22:32 143 27 158
25:4 253 Judges 29:4–9 4
27:2 220 1:23–6 243 48:33 65
27:27 259 1:26 244 51:13 171
27:28 220 5:11 81
Ezekiel
Numbers 1 Samuel 36:25 105
3:44–51 301 1:24 65, 83 38:2 302
3:46 299 28:8–25 42 38:16 302
356 index of sources

38:17 303 Ezra Eruvin

38:2 302 1:1–6 158 4:11 179
39:6 302
39:11 302 1 Chronicles Pesahim
21:15 238 1:1 55
Psalms 6:1–2 179, 184
17:14 320 6:5 182
37:1 205 Second Temple Literature
44:21–2 317 Betsah
44:23 310, 316 Ben Sira 1:1 97
50:5 330 7:20 316
103:20 150 Rosh ha-Shanah
Josephus 3:7 327
Proverbs Antiquities 13, 311–3 240
2:7 75 Megillah
3:18 229, 231 1:8 192
8:26 170 The New Testament 2:2 327
9:17 140, 146
11:3 151 Acts of the Apostles Nedarim
16:14 238 8:9–24 52 10:6 182
23:17 205
28:4 204 Galatians Nazir
3:10 228 7:4 76
Job
5:10 171 Revelations Sotah
18:10 235 20:7–10 302 5:5 345
9:13 54
Song of Songs 9:14 192
1:3 108 Mishnah
2:3 150 Gittin
2:14 151; 153 Berakhot 1:6 182
1:2 315
Ruth 2:1 327 Bava Qama
2:12 87 9:1 158 5:5 136
9:5 308, 345 9:11 99
Qohelet 9:6 3016
3:2 234 Bava Metsia
Peah 4:3–4 252
Esther 6:6 182 15:9 65
4:15 142 7:7 182
4:16 142 Bava Batra
8:8 289, 291, 295 Sheviit 3:1 278
9:27 148–51, 155 10:9 99
Sanhedrin
Daniel Terumot 6:4 58
3:19–27 159 9:2 182 7:4 33
4:14 128 7:11 54–5
6:17–25 159 Shabbat 8:7 133, 140,
7:9 292 3:4 169 144
9:14 29 23:4 185
index of sources 357

Shevuot Terumot Makhshirin

3:5–6 182 2:6 55 3:4 80

Eduyot Shabbat
1:4–6 179, 216 7:14 55 Midrashei Tannaim
2:8 302 15:16–7 135, 145
Mekhilta de-Rabbi Yishmael
Avodah Zarah Yoma beshalah shirta
2:5 53, 182 1:4 76 1 316
4:1 159 1:8 326, 345 3 310, 317,
347
Avot Sukkah Yitro, ba-hodesh 1 98, 325
2:4 76 2:1 76 2 111
2:8 75 3 104, 151
3:2 283 Yom Tov 9 238
1:10 75 10 309
Zevahim mishpatim,
7:4 182 Hagigah neziqin 17 33
12:3 182 2:1 326 mishpatim, massekhta
de-kaspa 20 208
Menahot Yevamot
12:5 182 2:5–6 99 Mekhilta de-Rabbi Shimeon
8:7 326 ben Yohai
Hullin 12:2 99 13:1 99
4:4 182 14:5 325 18:1 98
8:4 208 21:17 99
Sotah 23:19 208
Bekhorot 7:4–6 158
8:1 99 7:7 24, 155, 158 Sifra
15:8 130, 189, 192 Vayiqra dibbura de-hovah,
Keritot parashah 2:1–3 182
3:9–10 182 Bava Qama tazria, negaim,
7:13 76 pereq 1:4 104
Negaim 10:17 99 metzora‘ parashah
7:1–2 104 1:12 210
Sanhedrin qedoshim,
Yadayim 11:5 50 pereq 4:8 210
4:4 8 13:11 330 pereq 6:2 55
pereq 8:1 93
Avodah Zarah pereq 9:9 99
Tosefta 3:5 130
Sifre on Numbers
Berakhot Horayot 84 104
1:4 76 2:7 111 95 299
2:2 327 112 5
6:6 166 Oholot
6:7 309–10, 316–7 3:7 76 Sifre Zuta on Numbers
6:26 287
Peah Niddah 9 76
4:1 113 1:5 76 18:15 99
358 index of sources

Sifre on Deuteronomy 4:8, 35c 302 Yevamot

32 108–9, 309–11, 6:1, 36b 223 2:6, 3d 99–102, 113–4
316–7 10:9, 39d 101 8:1, 8d 88
34 76 8:2, 9d 106
37 170 Maaser Sheni 11:2, 11d–12a 99
38 76 1:1, 52c 223 16:4, 15d 325
43 334
104 208 Hallah Ketubbot
170 60 2:1, 58b 223 3:1, 27a 113
171 54–7 12:3, 35b 233
221 58 Bikkurim
289 99 3:3, 65c–d 98, 329 Nedarim
306 316 1:1, 36a 222–3, 225
321 29 Shabbat 1:1, 36c 212, 216, 219
336 108 1:5, 3d 223 10:1, 41b 223
357 169 1:7, 3d 316
2:2, 5b 47 Sotah
Sifre Zuta on Deuteronomy 9:1, 11d 223 3:2, 18d 223
Deut 6:5 310, 317 14:4, 14d–15a 145–6 5:7, 20c 311–2
7:4, 21d 229
Pesahim 9:15, 24c 192
Talmud Yerushalmi 7:6, 34c 223
9:2, 36d 223 Gittin
Berakhot 1:5, 43b–d 114, 119
1:3, 3b 223 Sheqalim
1:5, 3c 247 1:4, 46b 113 Qiddushin
2:1, 4a 333 1:8, 61c–d 223
3:1, 5d 184 Yoma 3:4, 64a 119
4:1, 7d 7, 130 1:5, 39a–b 345 3:9, 64b–c 87, 94
4:4, 8b 111 4:6, 42a 223 3:12, 65a 92
6:1, 10a 327 5:4, 42c–d 223 4:7, 66a 87, 91
9:1, 12d–13c 159, 132 8:1, 44d 223
9:2, 13c–d 165 8:7, 45c 223 Bava Batra
9:5, 14b 311–2 5:5, 15a–b 127
Rosh ha-Shanah
Peah 4:8, 59c 98 Sanhedrin
1:1, 15b 192, 316 1:1, 18a 247
6:1, 14b 223 Taanit 3:6, 21b 133, 135, 137,
3:10, 66d 330 145
Demai 4:1, 67d 7 6:6, 23c 58
4:3, 23c 113 7:13, 25d 57, 61, 80
Megillah 7:19, 25c–d 62, 66, 79
Kilayim 1:4, 70d 155 10:2, 28c 111
9:4, 32c 234–6 4:1, 74d 80
Avodah Zarah
Sheviit Moed Qatan 2:2, 40d–41a 130, 145
4:2, 35a–b 132, 135, 137, 3:1, 81c 11 2:8, 42a 94
145 3:5, 82d 238 4:8, 44b 94
index of sources 359

5:4, 44d 114–5 77a–b 234 Yevamot

5:12, 45b 208 87b 24, 64, 81 8b–9a 212
113b 63 22a 100–1
22b 96
Talmud Bavli Yoma 46a 86, 92–3
19b 331, 345 46b 86–8, 93–4
Berakhot 27a 218 47a 86, 89–91,
5b 325 69b 247 93–4, 104
7b 204 72b 17 47b 87, 102–3
10b 111 48b 86, 96–7
13a 212 Sukkah 62a 99–101
13b 338, 346 20a 1 62b 96
27b–28a 7 24b 212 65b 215
28b 345 43a 212 97b 96, 100
30b 327, 344 53a 241–2 101a–b 233
34a 327 109b 103–4
35a 218 Betsah 121a 325
47b 113 2a 97
50a 212 Ketubbot
55a 66 Rosh ha-Shanah 12b 104
57b 161–3 23a 330 103b 343
59a–b 165 28b–29a 327 104a 329–30
61b 310, 319–20, 111a 28–30
336, 344 Taanit
10a 170–1 Nedarim
Shabbat 23a 330 3a–b 223, 225
30b 238 29a 314, 346 50b 318
36a 163 91b 140, 146
55a 247 Megillah
66b 198 6b 204 Sotah
88a 147–57 8b 218 5a–b 104
104a 66 10b–17a 23, 135, 142, 7a 140, 146
104b 80 248 31a 312
131b 212 13b 137, 142 32b–33a 212
152b 122 15a 141 44b–45a 212
156a 40 19b 155 46a–b 192, 244
24b 325 47a 80
Eruvin 25a 327 49a–b 192
11b 131
15b 212 Moed Qatan Gittin
19a 167–8 5a 218 10b 269
52b 173–4 8a 212 11b 129
58a 218 18a 128 14a–b 120–1
63a 66 28a 238 21b 212
57b 331, 344
Pesahim Hagigah 58b 330
24a 218 13a 66 67b 131
28b 212 14a 29, 292 68a–b 243
44a 218 15a 294 88a 29
360 index of sources

Qiddushin 38b 292 51a 218

4a 218 42b 218 54b 136
9b–10a 212, 218 46b 218 57b 346
14a 218 47b–48b 175 71a 276
14b–15a 212 51b 198
17b 101, 218 64a 247 Horayot
57b 208, 239 65b 55, 61, 63, 69, 317 13b–14a 294
70a–71b 103–6, 129, 132 67a 80
72a 122 67b 37–8, 61, 63, Zevahim
75a–b 114, 218 69–70 4a 218
68a 11, 42–5, 47–9, 4b 218
Bava Qama 56, 60 7b 218
28b 136 68b 218 7b 217
30a 136 69b 331 11b 218
38a 113 71b 218 33b 218
50b 136 73a 211, 217–8 36a 218
59a–b 132 73b 145 36b–37a 218
64b 218 74a–b 133–4, 136, 143, 49a 218
145 106a 218
Bava Metsia 75a 138–40 116b 98
11a 218 76a 218
14b 274 77a 152 Menahot
47b–48a 265–6 85b 113 29b 66, 331, 344, 346
49b 259 93a 110 51b 218
53b–54a 258–9 97a 246 55a 218
59b 11 99b 109 61a 218
68a 65 107b 80 65a 60
84a 194 110b 330 72b 218
86a 333, 346
Makkot Hullin
Bava Batra 14b 208 5b–6a 114–5
10a 317 19b 63, 81 6a 112
28a–b 278 24a 334 7b 38, 40
29a 278 82b 218
53a 269 Shevuot 115b 208
55a 278 14a 218
74a 245 39a 154–5 Bekhorot
89a 127 47b 38 16a 218
47a 96, 100
Sanhedrin Avodah Zarah
5a 1 2b 148 Temurah
6a 198 5b 218 7a 212, 218
11a 302–3 6b 83
14a 64 17a 346 Niddah
17a 51 18a–b 294 13b 103–4
26b 140, 146 20a 318 27a 215
32b 119 27b 333 40a 212, 222–3, 225
33a 198 29b 136 44a 113
34a 210 35b 108 56b 113
38a 29 38b 131 57b 136
index of sources 361

Midrashei Amoraim Qohelet Rabba vayera 23 295

3:2 234 beshalah 20 295–6
Genesis Rabba 3:16 331 tazria 7:7 317
11:15 317 10:4 331 ki tavo 4, 47 319
12:4 67
16:4 128 Esther Rabba Printed Tanhuma
18:5 99 8:7 142 Noah 3 32
33:1 309 vayera 10 323
39:2 107–8 Pesiqta de-Rav Kahana ki tisa 33 347
39:9 316 4:2 247 ki tavo 348
39:11 98 4:8 98
39:14 107, 109–10 10 208 Midrash Mishle
41:30 239 11:6 105 2:7 75
51:2 288 22:1 110 9:2 318
53:5–6 110 43:5 110
53:9 110 beshalah 14 317 Midrash Shir Hashirim,
69:8 244 et qorbani, 4 98 ed. Grünhut
71:9 223 p. 5b 318
78:12 111
82:8 316 Minor Tractates Song of Songs Zuta
82:12 107 to Song 1:4 (ed.
84:4 107 Avot de-Rabbi Nathan S. Buber, Midrasch
90:6 111 A 19 43 Suta, 13) 318
91:5 111 A 25 43, 71
93:9 316 B, 26 109 Pesiqta Rabbati ed.
M. Ish-Shalom
Leviticus Rabba Semahot 23, 119b–120a 317
6:1 247 9:2 77
9:6 223 Sekhel Tov
11:2 302 Kallah Rabbati Genesis 17 103
15:9 105 Higger, p. 305 51 Exodus 7 70
25:1 228
Derekh Eretz Seder Eliyyahu Rabba
Deuteronomy Rabbah Pereq haminim 9 204 (XI ed. Friedmann)
eqev 17 316 7:16 238 p. 54 82
7:30 238
Song of Songs Rabba Midrash Hagadol
1 76 Gerim Deut 6:5 320
1:3:3 108, 110 1:1 87
4:16 223 1:2 92 Yalqut Shimoni,
8:13 327, 344 2:5 96–7 Exod 182 70
Deut 837 320, 343
Lammentation Rabba
2:2 316, 334 Later Midrashim Zohar
3:44 317 Va-Yehi 215b 103
4:2 330 Tanhuma Buber
5:18 334 lekh lekha 3 108
vayera 19 294–6
362 index of sources

Commentaries on the Bavli 72 216 Ephrem

77 209 Hymns on the
Rashi 79 209 Nativity 1 231
Berakhot 7b 205 130 209
Berakhot 55a 66 Origen
Berakhot 61b 321, 324, Aristotle Homilies on Leviticus
341 Rhetoric 3.18.2–5 178 16.4.3 231
Shabbat 88a 148 Homilies on Joshua
Megillah 15a 142 Cicero 8.6 231
Nedarim 50b 318 De Officiis I, xlii
Bava Metsia 68a 65 150 262 Pseudo-Augustine
Sanhedrin 152 262 RLM 143.25–145.33 196–7
17a 51 330–3 262,
50b 347 281 Pseudo-Clementine
65b 66 Inv. 1.18–19 195–6 Recognitiones, II, 9 52
67b 38, 66 Partitiones Oratoriae
74a–b 134 101–6 196 Sozomen
74b 137 Rhetorica ad Heren. Ecclesiastical History
Makkot 1.26 196 2:14 130
19b (RIVAN) 63
Avodah Zarah 20a 318 Digest
Gaius, Common Matter 2 Zoroastrian Literature
Torat Hayyim (D.41.3.38) 276
Sanhedrin 67b 63 Gaius, Institutions The Book of a Thousand
II.18 269 Judgements
Sheiltot Julian (D.40.4.17.2) 250 MHD
Exodus, 43 104 Justinian (D.1.3.2) 250 6.13–16 267
Neratius, Parchments, 12 290
Yad Ramah book 3 (D.6.2.17) 279 17.1 264
Sanhedrin 25b 70 Paulus, On the Edict, 30.10 264
Sanhedrin 65b 66 book 14 (D.19.2.22.3) 261 30.12 264
Sanhedrin 67b 38 book 33 (D.18.6.8–9) 271 31.6, 8 264
Ulpian, On the Edict, 37.2–10 255–6
Margaliot Ha-Yam book 11 (D.4.4.16.4) 261 64.9–15 267
Sanhedrin 65b 63 book 28 (D.18.6.1) 270 90.4 264
Ulpian, Sabinus, 95.1 264
Isaac Alfasi, Halakhot book 41 (D.18.1.28) 277
Berakhot 35b 113 MHD A
Shabbat 39b 103 Juvenal 8.13–9.1 275
Satire xv.110 177 9.5–7 275
9.7–9 264
Greek and Roman Plato 12.3–5 256–8
Literature Euthyphro 7a–8a 214 27.13 264
Laws, XI 40 28.1 264
Aelius Theon
Progymnasmata Pahlavi Videvdad
13 216 Church Fathers 3.16 227
65 216
Chronica Minora 19 130
Index оf Authors

Aarne, A. 240 Brody, R. 29

Abusch, I. T. 34 Brown, F. 149
Al-Rawi, F. 282 Brubaker, R. 2–3
Albeck, C. 63–4, 174, 179, 208, 210, 247 Brüll, N. 123
Alexander, P. S. 35, 39, 53, 55–6, 62, 66 Buckland, W. 264, 270, 272, 276–8
Alon, G. 9, 122–3 Bush, W. 149
Amit, A. 12, 82
Anderson, B. 2 Cardascia, G. 34
Anklesaria, K. 245 Castelli, E. A. 307
Aslanoff, C. 68 Chajes, J. H. 42
Assis, M. 88 Chernick, M. 208
Atzmon, A. 31 Chin, C. 186
Avi-Yonah, M. 167 Christensen, A. 130
Cichocka, H. 177, 200–1
Bailey, L. R. 168 Clark, D. 177, 191, 200, 216
Bakhtin, M. 184, 195 Cohen, A. 143, 175, 178
Baron, S. W. 80 Cohen, B. 203
Barris, J. 195 Cohen, J. 102
Baskin, J. R. 135 Cohen, S. J. D. 21, 88, 95, 112, 175, 292
Baumeister, T. 307 Cook, J. A. 252
Baumgarten, A. 117, 124 Cotton, H. M. 313
Bazak, Y. 50 Cribiore, R. 177, 186, 189, 191
Becker, A. 227
Becker, H.-J. 312 Dan, J. 67–8
Beentjes, P. C. 316 Danby, H. 50
Beer, M. 122 Daube, D. 177–8, 197, 203
Ben Yehuda, E. 310 Donald, M. 284
Bendavid, A. 125, 202 Driver, G. R. 34
Bender, A. P. 238 Duling, D. C. 237
Blau, L. 32, 39
Blidstein, G. 104 Eck, W. 313
Bohak, G. 52, 60–1, 66, 79 Efron, J. 24
Bonfil, R. 20 Elizur, Y. 114
Bonner, C. 52 Elman, Y. 26–7, 131, 175–6, 206–7, 215–6, 227,
Bonner, S. F. 117, 191 255, 266, 310
Bottéro, J. 198 Elon, M. 208
Bowersock, G. 307 Epstein, J. N. 46, 57, 64, 125, 151–2, 174, 179,
Boyarin, D. 59, 78, 174–5, 178, 184, 194–5, 192, 208, 232
205–6, 227, 308, 328, 347–53 Even-Zohar, I. 284
Boyce, M. 245
Brecher, G. 32 Farber, W. 35
Bregman, M. 30, 285 Feldman, L. H. 111–2, 177
Briggs, C. A. 149 Finkelstein, J. J. 283
Brock, S. 198, 332 Finkelstein, L. 183
Brodsky, D. 20, 25–6 Fischel, H. 177, 299
364 index оf authors

Fishbane, M. 308 Hayes, C. E. 21, 207

Fleischer, E. 68 Hayward, R. 110
Foerster, R. 190 Heath, M. 177, 195–7
Fonrobert, E. 84 Herman, G. 6, 22, 123, 131–2, 176, 232, 332
Fraenkel, H. 20 Herr, M. D. 134–5, 306
Fraenkel, Y. 14–5, 18, 76–7, 118, 125, 238, Hezser, C. 175, 177, 189, 192, 200, 203, 278
324–7 Hidary, R. 177, 189
Fränkel, D. H. 124 Higger, M. 51, 77, 88
Frankel, Z. 232 Hingley, R. 193
Freedman, H. 148, 152 Hirshman, M. 233
Friedman, S. 22, 33, 41, 45–6, 51, 54–7, 59, Hock, R. 178
62, 65–6, 76, 82–4, 118, 124–5, 175, 202, 310, Hoffmann, D. Z. 60
330 Honoré, T., 263
Friedmann, M. 63 Horbury, W. 292
Hyman, A. 62–4, 122
Gafni, Ch. 50
Gafni, I. 2, 6–7, 24, 28–9, 123, 127, 131, 158, Idel, M. 66–7
166, 232, 293, 332 Ilan, T. 22–3, 25, 160, 173, 232
Galante, S. D. 125
Garb, J. 67 Jaffee, M. S. 84
Geiger, B. 128–9 Jamasp-Asana, J. M. 248
Gershevitch, I. 129 James, M. R. 247
Gibson, C. 178, 186, 188, 190–1 Jastrow, M. 39, 230, 254
Gil, M. 20 Joël, D. 32
Gilat, Y. 56–7 Johnston, D. 252, 272–3, 277
Ginzberg, L. 11, 67, 237, 341 Jolowicz, H. F. 276, 278
Goldberg, A. 25, 128, 172 Judelowitz, M. D. 125
Goldberg, A. Z. 308
Goldenberg, D. M. 266 Kaddari, M. Z. 237
Goldenberg, R. 11–3 Kahana, M. I. 5–6, 56, 109, 310
Goldenberg, D. M. 242 Kalmin, R. 22, 84, 103, 145, 175, 203, 206, 280,
Goodblatt, D. 13, 21, 117, 123 306, 330–1
Goodman, M. 292 Kanga, M. F. 248
Gordon, W. M. 269 Kaspina, M. 232
Goshen-Gottstein, A. 43, 46, 50–51, 78 Katzoff, R. 203
Graetz, H. 9 Kaye, L. 194
Gray, A. M. 1, 22, 133, 137, 145–6, 160, 242 Kennedy, G. 177–8, 186–7, 189, 191, 197, 200,
Greengus, S. 36 209, 216
Grossman, A. 20–1, 31 Kessler, H. L. 23
Gruen, E. 198 Kiel, Y. 176, 227
Gruenwald, I. 68–70 Kimelman, R. 292
Grünhut, L. 233 Kiperwasser, R. 27, 333, 238
Kister, M. 53
Halevy, I. 9 Kitchen, K. A. 166
Halevy-Epstein, E. 25, 177 Klausner, J. 80
Halivni, D.: see Weiss Halivni Klein, S. 173, 195
Harari, Y. 39, 42, 239 Klikstein, J. 14–6, 18
Hauptman, J. 175 Knights, C. H. 247
index оf authors 365

Kohut, A. 243 Mommsen, T. 315

Kosowski, B. 63–4 Morag, S. 68–9
Kraemer, D. 175, 215 Moscovitz, L. 175, 207, 330
Krupp, M. 321 Muffs, Y. 260
Kugel, J. 207, 286 Murray, R. 198
Kulp, J. 87
Nagy, G. 201
Labendz, J. 178, 195 Naiweld, R. 173
Langermann, Y. T. 69 Nau, F. 248
Lapin, H. 251 Neiman, D. 166
Lauterbach, J. Z. 151 Neufeld, E. 35
Lavee, M. 22, 86, 91, 99–100, 107, 110, 112, 116, Neusner, J. 11–4, 57, 118–9, 122–4, 127–9, 131,
287 177–9, 202–3, 212, 275
Lazarus, F. 123, 143 Newman, H. I. 14–5, 18–21, 118, 333
Lecker, M. 103 Nichols, B. 276, 278
Levine, B. 260 Nieher, H. 237
Levinson, J. 49, 58, 61–2, 66, 79, 111, 330 Niehoff, M. 23
Lieberman, S. 24–6, 39–41, 55, 63–5, 81–2, Nikolsky, R. 23, 30, 169, 173, 248, 302, 305
111, 119, 155–6, 177, 189, 192, 202, , 306–10, Noam, V. 232
314–6, 318, 328, 331–2, 353 Nöldeke, T. 130
Liebes, Y. 68–9 Nyberg, H. S. 248
Lieu, J. M. 352
Lightstone, J. 178–9, 182 O’Neil, E. 178
Limor, O. 110 Oppenheimer, A. 103, 123, 159
Lurje, P. 245
Peachin, M. 173
MacCormack, G. 250 Pearson, B. A. 237
Macuch, M. 26, 254–7, 264, 267–9, 275–6 Perakhanian, A. 255, 264
Mandel, P. 30–1, 326, 329, 333 Porton, G. 103
Margaliot, M. 21, 63 Potter, D. 314, 318
Margolies, E. 125 Pummer, R. 112
Marrou, H.-I. 177
Martínez, F. G. 34 Rabbinovicz, R. 63, 147–8
Massey, D. S. 2 Rabinowitz, I. 168
Mattingly, D. 192 Rapoport, S. Y. L. 129
McNamee, K. 201 Rebiger, B. 238
Meeks, W. 189 Reeg, G. 167, 332
Meir, O. 117 Reinik, G. J. 26
Melamed, E. Z. 151 Reynolds, J. 111
Menirav, J. 251, 253, 260 Richler, B. 320
Merhaviah, H. 321 Riedweg, C. 167
Meroz, R. 305 Robinson, O. F. 250, 269
Miles, J. C. 34 Rosenfeld, B.-Z. 251, 253, 260
Milgrom, J. 36 Rosenthal, D. 232
Milikowsky, C. 168, 285 Rosenthal, E. S. 119, 123–4, 128
Miller, S. S. 292 Roth, M. T. 34, 261, 282
Millett, M. 192–3, 206 Roth, R. 192
Mokhtarian, J. S. 248 Roymans, N. 206
366 index оf authors

Rubenstein, J. L. 2, 17–8, 21, 24, 26, 45, 66, 80, Tejera, V. 192
84, 103, 116–7, 125, 203, 232, 280, 285, 294, Thompson, R. C. 39
330 Tigay, J. H. 36
Torijano, P. A. 237
Safrai, S. 21, 306, 308, 317, 325, 330–1, 350 Trachtenberg, J. 32
Sanders, E. P. 299, 308 Tropper, A. 1, 24, 66, 117, 203, 232, 308, 310,
Satlow, M. 193 318, 344
Schäfer, P. 65, 68–9, 229, 238, 306 Tzuberi, C. 24, 30
Schmid, K. 166–7
Schmitt, R. 36 Ulmer, R. 195
Schoeps, H. J. 52 Urbach, E. E. 10–1, 41, 308–9
Scholem, G. 67–9
Schürer, E. 313 van Bekkum, W. J. 166, 198
Scurlock, J. 282 van Henten, J. W. 307
Secunda, S. 27, 62, 176 Vanstiphout, H. L. J. 198
Segal, E. 142 Veenhof, K. R. 283
Segal, M. 173 Veltri, G. 41–2
Shalev-Eyni, S. 237 Visotzky, B. 177, 189
Shapira, D. Y. 176, 232 Vogelzang, M. E. 198
Shapira, H. 16–18
Sherwin, B. L. 80 Wachten, J. 233
Shoshany, R. 239 Wald, S. G. 32, 82
Simhoff, R. 169 Walfish, B. D. 142
Sinnott-Armstrong, W. 134 Wasserstrom, S. M. 69
Skjærvø, P. O. 248 Watson, A. 250, 254, 260–1, 263, 265, 271–2,
Slofstra, J. 206 279–80
Sokoloff, M. 39, 54, 79, 149, 153, 155, 254, Webb, R. 177
313–5, 326–7 Webster, J. 192–3, 199, 227
Soloveitchik, H. 137, 145 Weiss Halivni, D. 19–20, 143, 175
Sperber, D. 25, 79, 238–9, 260 Weiss, A. 174
Stager, L. E. 169 Weiss, J. H. 9
Steiner, R. 207 Weitzmann, K. 23
Steinmetz, D. 285 Wilken, R. 189
Stern D. 207 Woolf, G. 193, 206
Stern, M. 186, 189 Wright, D. P. 37
Stratton, K. 54, 70
Strauch-Schick, S. 330 Yadin, A. 208
Stroumsa, G. G. 110
Sussmann, Y. 24, 232 Zanolli, A. 248
Zeligman, I. 50
Ta-Shma, I. M. 20, 31 Zlotnick, D. 77
Tannenbaum, R. 111 Zunz, L. 76
Taubenschlag, R. 251
Index оf Rabbinic Names

Abba Hanan 97 Rabbi, see Rabbi Yehudah Ha-Nasi

Abba bar Zemina 235 Rabbi Abba 113, 141
Abbayye 37, 42, 85, 134, 136–9, 143, 167–8, Rabbi Abbahu 114, 192
210 Rabbi Abraham ben David of Posquières
Aqiva ben Joseph see Rabbi Aqiva 209–10
Rabbi Abun 288
bar Elyashiv 130 Rabbi Aha 289–90
bar Livai 214 Rabbi Ahai 140, 144–5
ben Azzai 316 Rabbi Ahai bar Rabbi Josiah 121–2, 125–7,
ben Dama 59 130
ben Kamz(ora) 330 Rabbi Ami 114
Beruriah 294 Rabbi Aqiva 5–6, 8–10, 17, 30, 33, 42–5,
47–51, 53–9, 61, 76–9, 179, 181–2, 207–8,
Dosethai bar Rabbi Yannai 120–1, 124–6, 130 220–3, 292–4, 306–54
Rabbi Assi 114
Eleazar Zeira 131 Rabbi Ba see Rabbi Abba
Elisha ben Abbuya 294 Rabbi Berekhiah 108
Rabbi Dosethai bar Rabbi Yannai see Dosethai
Hillel 1 Rabbi Dostai bar Matun 204
Hizqiyah son of Rabbi Hiyya 215 Rabbi Eleazar 150
Honi Ha-Maagel 330 Rabbi Eleazar ben Azariah 7–9, 12, 15, 17, 76,
Hutspit the translator 8, 15 292
Hyrcanus 43, 47 Rabbi Eliezer 10–1, 22, 32, 42–3, 46–61, 71,
75–80, 93, 95, 179–81, 194, 310
Jacob of Sarug 198 Rabbi Hama bar Rabbi Hanina 106, 150, 235
Joshua ben Jonathan 318 Rabbi Hananiah ben Rabban Shimeon ben
Judah the Patriarch see Rabbi Yehudah Gamaliel 97
ha-Nasi, the Prince Rabbi Hanina 32–42, 54, 59, 61, 63–71, 81–3,
173–4, 185, 235
Kamza 330 Rabbi Hanina (bar Hama) 64
Rabbi Hanina ben Gamaliel 97
Mar bar Ravina 161, 163 Rabbi Hanina Segan ha-Kohanim 283
Rabbi Helbo 103, 105, 107, 235
Nehemiah the Amsonite 311 Rabbi Helbo ben Rabbi Hilfi bar
Nitza 133 Samkai 288
Rabbi Helbo bar Simlai 290
Pappos ben Yehudah 310, 324 Rabbi Hinnena bar Rabbi Hananiah 62–3,
66
Qarna 127 Rabbi Hiyya 1, 80, 114, 116, 215
Rabbi Hiyya bar Abba 86
Rabban Gamaliel 7–15, 17, 115, 130, 189 Rabbi Hiyya bar Rabbi 88
Rabban Gamaliel VI 19 Rabbi Hiyya bar Rabbi Hoshaya 89
Rabban Shimeon ben Gamaliel 113, 294 Rabbi Hoshaya 32, 64–5, 88–9 (bar Rabbi),
Rabban Yohanan ben Zakkai 11, 13, 75–6, 107, 171
168 Rabbi Ilai 76
368 index оf rabbinic names

Rabbi Isaac see Rabbi Yitshaq Rav 6, 136, 138, 171–2, 253, 258, 274
Rabbi Isaac bar Nahman 88–9 Rav Aha bar Jacob, see Rav Aha bar Yaaqov
Rabbi Joshua ben Levi 88, 165, 171 Rav Aha bar Rav Iqa 139
Rabbi Juda, see Rabbi Yehudah Rav Aha bar Yaaqov 148–9, 151, 153, 217–8
Rabbi Liezer see Rabbi Eliezer Rav Amram Hasida 131
Rabbi Mani bar Patish 213 Rav Ashi 161, 163–4, 254
Rabbi Meir 114, 293–4, 30 Rav Avdimi bar Hasa bar Hama 147–8, 151–3
Rabbi Nehemiah 303 Rav Dimi 134, 144–5
Rabbi Oshaia see Rabbi Hoshaya Rav Gada 131
Rabbi Reuven ben Strobilos 236, 239 Rav Hai Gaon 60
Rabbi Shimeon 8, 204–5, 213–4, 231, 266, Rav Hamnuna 160–4
287, 300–1 Rav Hisda 229, 266
Rabbi Shimeon ben Eleazar 114 Rav Huna 228–31, 274
Rabbi Shimeon ben Laqish 112, 213 Rav Idit 292
Rabbi Shimeon ben Yehozadak 133 Rav Kahana 123
Rabbi Shimeon ben Yohai see Rabbi Shimeon Rav Jacob bar Idi 139
Rabbi Simai 150 Rav Joseph bar Hama 127
Rabbi Tarfon 252–3 Rav Nahman 253–4, 274, 292, 302–5
Rabbi Yannai 62, 125 Rav Nahman bar Rav Hisda 108
Rabbi Yasa 106 Rav Papa 53, 139, 198
Rabbi Yehoshua (ben Hananiah) 7–10, 12, Rav Sa‘adya Gaon 69
15, 45, 49, 54–9; 61–2, 75–80, 93, 95, 179, 181 Rav Safra 89, 217–8
Rabbi Yehudah 90–1, 94–5; 97, 138 165, Rav Sheshet 89–90, 101, 127, 214
253–4, 303 Rav Shmuel bar Nahmani 139
Rabbi Yehudah ben Betera 33 Rav Tabuth 246
Rabbi Yehudah ben Rabbi Simon 288 Rav Tavyomi 246
Rabbi Yehudah Ha-Nasi 6–7, 12, 67, 192, 329 Rav Yaaqov 217–8
Rabbi Yehudah (Yudan) the Patriarch II Rav Yehudah 138
(Nasia) 16, 64, 82–3, 88 Rav Yitshaq 165
Rabbi Yishmael 5–6, 33, 55, 76, 145, 208, 218, Rav Zevid 131
220–3 Rava 17, 63, 65, 68, 85, 88, 101, , 134, 136–9,
Rabbi Yishmael bar Rabbi Yosi 293 143, 147–51, 153, 155–7, 161–3, 165, 167–8,
Rabbi Yitshaq 140, 289–91 246–7, 278
Rabbi Yohanan 16, 24, 32, 37–42, 51, 64, Ravina 259
86–8, 93–4, 100, 106, 133–5, 138, 145, 194, Resh Laqish 100–2, 109, 167, 194, 213, 265–6
204–5, 213, 265–6, 292–3
Rabbi Yonah 235–6 Shimon ben Shetah 58
Rabbi Yonatan 259 Shmuel 127, 136, 258, 274, 278
Rabbi Yose bar Kipar 120–1, 126–7
Rabbi Yoshaya 208, 259 Uqban bar Nehemiah 278
Rabbi Yosi 75, 88–9, 97, 299
Rabbi Yosi ben Zevida 89 Yehoshua ben Perahya 80
Rabbi Yosi Ha-Gelili 33 Yehudah son of Rabbi Hiyya 215
Rabbi Zeira 89 Yirmiah ben Eleazar 168
Rabin 134–8, 144–5
Rami bar Abba 165 Zenon the Cantor 8
Index оf Place Names

Alexandria 80, 177, 201 Haran 109

Arabia 167 Harpania 167
Ashkenaz 20, 31, 70 Hinnom Valley see Gehenna
Assyria 4
Empire 2 Iran 245
region 172 Italy 31

Babylon 1–4, 158–63, 165–6, 170–2 Jericho 169

Bridge of 166 Jerusalem 4, 25, 29, 63, 158, 168–9, 172
Beer Sheba 110 Judaea 30, 313, 328
Beirut 201
Beit Garem 167 Kangdiz (fortress of Kang) 245
Beit Shean 167 Kushta 246–8
Bethel 243
Bridge of Bistana 165 Laodicea 88
Bursif 163 Lebanon 201
Byzantion 31 Lod 45, 49, 77, 133–5, 145, 252–3
Lodian sages 134, 138, 143–5
Caesarea 45, 49, 64, 77, 240, 318, 331–2 Luz 242–6, 248–9
Caesarea, tetrapylon of 333
Carthage 314 Medes, cities of 4
Ctesiphon 250 Mehoza 127
Mesopotamia 4, 24, 35–7, 39, 167–8, 172, 197
Damascus 21, 167–8 Moab, Plains of 154–5
Dumascnin 167 Mount Sinai see Sinai, Mount
Dura Europos 23
Nehardea 121–2, 126–7, 132
Eden, Garden of 24–5, 160, 166–9, 171–2, 248 Nehardeans 22, 126–7
Egypt 148, 251 Nippur 2
Hellenistic 23 Nisibis 26
Euphrates River 25, 159–60, 162–7, 171
Paradise see Garden of Eden
Fraxhard (ocean) 245 Passage of Yamina 167–8
Persia 157
Galilee 2, 23 Zoroastrin 227
Gehenna/Gehinnom 5, 168–9 Persians 158, 276, 279
Gerizim, Mount 115 Empire 173, 227–8
Gihon 169 context 23, 206
Gomorrah 286–8, 293, 304
Gozan 4 Qumran 23, 65, 83
Greece 29, 105
Ramot-Gilead 235
Habor 3 Red Sea 148
Halah 3 Roman Palestine 177,179 252, 260
Hama see Tiberias Rome 29, 81, 193, 250, 281, 306, 314
370 index оf place names

Sepphoris 62–3 West, the 225

Sinai, Mount 24, 76 147–55
Sodom 286–8; 290, 293, 304 Xwaniras 245
-ites 295–7
South, the 236 Yavneh 7, 11, 15, 115
Strato’s Tower 240 Yemen 167
Syria 26, 167
Zoar 169
Tiberias (Hama) 19, 21, 79, 169
Tigris River 25, 165–6, 171
Transjordan 167
Troy 188
General Index

Abode of the Blessed; see City of Immortals Ark of the Covenant 76

Abraham 67, 69, 107–10 Armenian 248
place of origin 24 Iranian loan word in 128
academy see study house Arsacids 129
Adam 247–8 dynasty 23
aediles 271–2 era 22, 122–3, 131
aggadah 15, 19–20, 62, 87, 249, 292–4 ascetics 239
agronomos 252 astrology 40
Ahab, King 235–6 Avestan literature 129, 176, 227, 257
Ahariman 243, 248
Ahasuerus 136–43, 148, 151, 291 Babylonization 31, 85
Ahiah son of Shisha 232–49 Bakhtin 184, 195
Akkadian 34, 39, 65 Bar Kokhba Revolt 304, 310, 313, 316, 318
dispute poetry 198 baraita, Babylonian 7, 43, 49–53, 59, 78,
am ha-aretz 113, 116 87–95, 150, 153, 155, 160, 166, 174, 204,
amulet(s) 19, 45–8 292
Ancient Near East 36–39, 177 pseudo-baraita (midrash-halakhah-
androgynous 214 style) 56, 222, 310
angel(s) 150, 237–40, 320, 332, 352 Yerushalmi 46, 160
cry of 352 bat qol see heavenly voice
destroying 150 beit-midrash see study house
divine agencies 39 beth din 10, 154
fallen 36 Av 11
Gabriel 288, 290–1, 295 Book (or Laws) of Creation see Sefer Yetzirah
ministering 150, 320 Book of a Thousand Decisions,
of Death 235–42, 244, 246 The (Mādāyān ī Hazār Dādestān) 254–5
angelic wrath 239 boundaries, social 14, 33, 85
anointed, king 98 Bundahishn (Zoroastrian encyclopedia)
priest 183 245
anonymous voice of the Talmud see stam Byzantine period 19–21, 30, 206, 305, 311,
Antigonus (last Hasmonean king) 240 332
antinomism 227
Christian 231 Canaanites 36
apocalyptic literature 19 Christ see Jesus
Arabic 69 Christianity 59, 264, 306, 251–3
Aramaic 8, 12, 18, 41, 54, 65, 79, 85, 124–5, academies 26
176, 192, 202, 326, 353 in Justinian’s code 251
commentary 125 Jewish polemics against 231
from Hebrew to 174 Roman 281
Galilean 242 taking charge of the world 251, 280, 282
grammatical suffix 96, 202–3 Cicero 262–3, 280–1
Palestinian 312, 314–5, 353 stasis theory 195–6
Ardašīr, King 130 circumcision 93–5, 103, 306, 317
Aristotle 199 City of Immortals 244–8
372 general index

code, legal Eschatology/-gical 105, 107, 169, 245, 303

of Hammurabi 34, 280 Eshnuna see law, of Eshnuna
of Justinian 250–1, 254, 263, 273, 381 Essenes 240
see also law Esther 133–47, 149–51, 155–6
Constantine 277 book of 23–4, 31, 149, 155, 290
convert/sion 22, 85–114 reading of 155
-ed Ammonite 8–9 Eusebius, Pamphilus 332
lion 113 Eve 247
cosmopolitan, “Roman” culture as 193 exilarch(s) 6, 17, 22, 117, 119, 122–3, 130–2,
counter-tradition 146 278
Creolization 192–3, 199, 227 anti-polemic 132
cultural heritage 5 Persianized features 131
ideas, exchange 6, 280 exile 3–6, 15, 24, 29–30, 146, 151, 158, 168,
Cushites (kushim) 241–6 236, 239, 247
exogram 284
dād (traditio in Middle Persian) 264 Ezra 1, 8
Daniel 159, 161
Muslim sources about his den 159 fable, Christian 247
David, King 237, 290–2, 295, 297
Decius, Emperor 306 Gabriel see angel(s)
demarcation, social 3 Gaius (Roman jurist) 269, 277
demon(s) 35, 52, 54, 237, 240, 242–3, 248, Institutions of 269
272 Gamalielian dynasty 130
servants of the Angel of Death 243 Gaonic 29, 103
Denkard 245, 257 period 20, 29, 85, 104, 305
dialogical 184 Gehinnom 168–169
message 195 genealogy 103–6
diaspora 2–3, 5–6, 24–25, 31 Genizah 21, 56, 109, 316, 320–1, 324, 343–4
autonomy of 6 gentile(s) 17, 64, 82, 99–102, 105, 107–10,
dispute poetry 198 112–3, 128–30, 135, 137, 142
divination 36 give-and-take see sugya
diviner 12, 35 Gnostic, proto 52
divine agencies see angel(s) God Fearers 110–11
domininium (Latin) 278 Gog and Magog 302–4
see also law, Roman Rome identified as 302
Greco-Roman world 177, 199, 202, 206, 208,
economy 5, 250–83 209, 217, 226
Edom 29 educational system 177, 191
Eldad and Medad 286, 298, 300–5 legal commentary 200–2
elders, institute of 298–304 legal perception 100
Elihoreph son of Shisha 232–49 literature 177, 186, 194, 199, 205–6, 226
Elijah, Prophet 76, 239 school primers 177, 186, 194, 205, 226
Elisha, Prophet 76 schoolteachers 188
elite/-tism 17–8, 35, 123, 177, 193, 206, 280 Palestine 177, 179
End of Days 302 Greek 25–26, 58, 189, 192, 201–3, 231, 238,
Ephrem 198 242, 248, 290, 307, 313, 317
Eros and Thanatos 349 god 159
in rabbinic literature 347 grammatical suffix 202
general index 373

influence 68, 197, 260 Iranian 257

primers of compositional writing 173, culture 27, 131, 249
198 myth 27, 232, 245–6, 249
progymnasmata 178 stories 243
see also Zoroastrianism
Hadrianic period/persecutions 306–7, 310, Isaac 110
315–17, 330 Isis 81
Halakhah/-khot/-khic 1, 6, 8, 10, 18–24, 50, Issachar, tribe of 111
55–7, 69–70, 88, 96, 98–100, 103–4, 112–4,
122, 125–6, 142, 14 6, 207, 210, 216, 284, 292, Jehoiakim, King 4
305 Jeremiah (prophet) 4–5, 30, 171, 237
Hammurabi see law of Jesus 59, 80, 307
Hannanel, Rabbenu 320 faith in 227, 228, 231
Hananiah, Mishanel and Azariah 109 professed as the Christ 307
hazaqah; see law, hazaqah Joab 81
heavenly court/host(s)/retinue see angels Job 311
heavenly voice 150, 320, 333 Joseph (biblical) 23, 111
Hebrew 7–8, 16, 39, 124–5, 164, 174, 176, 192, Josephus 23, 112, 237, 240
202–3, 228, 230–1, 254, 260, 310, 312, 315–6, Joshua 300–1
323, 347–53 Judas the Essene 240
Hekhalot Literature 19 Judeans 4, 29, 112, 158
hell see Gehinnom Julian 250
Hellenistic world see Greco-Roman world jurists, Sasanian 251, 261, 264, 273, 279
heretic arguments, answered by Justinian’s Code see law, of Justinian
aggadah 304
minim 66, 292, 326 kallah months 29
hermeneutic of 228 kalpe (ballot) 299, 301
hermeneutic 149, 152–3, 176, 178, 207, 210, karet 183–4, 230
212, 216, 218, 221–2, 226, 231, 243 ketubbah 104
rule, a phrase to raise an objection 218 kushya see sugya
Hermes (Greek god) 159
Hilkhot (Sefer) Yetzirah see Sefer Yetzirah late anonymous strata of the Bavli see stam
Hillel, House of 12–3, 16 Late Antiquity 19, 31, 206, 226, 251, 274, 278,
Hillelite dynasty see Hillel, House of 306, 311
Homer 199 Latin 25–6, 159, 202
scholia of 201 Roman legal documents 201
hybridization 193, 227 law, Assyrian 37
capital punishment 6, 123, 126
identity 31, 84–5, 110, 112–3, 127,140, 247 contract of sale 261
idolatry 23, 36, 133–5, 138, 140–1, 143–6, 159, cuneiform 261
243, 245 fourth, Sasanian 259, 260, 266, 270,
Rabbi Aqiva refusing 347 274–6
imagined community 2 handing over (traditio) 264
immortals/ity 244–6, 248–9 Middle Assyrian 34
incantations 35 new, dialogical or monological 199
inclination (evil, good) 309 of Eshnuna 261, 282
intermarriage(s) 104–105, 137, 142 of fraud 27, 250–3, 262, 272–83
intertextually 66, 129, 192 of Hammurabi 34, 280, 282
374 general index

of hazaqah 276, 278 Aqiva’s death not 334

of Justinian 250–1, 263, 281 genre of 310, 331, 348
of Lipit Ištar 261 tale 327
of meshikah 290–292, 296 medieval scribes 334
of ona’ah (overreaching) 251–4, 259, 263, meimra 124, 152, 174
273 Mercury (Roman god) 159
land ownership 262, 264–5, 275–6, 278 Messiah 29, 30, 104, 106, 158
Pauline conception of 227 Middle-Persian 26, 129, 176, 264, 268
Roman 27, 201, 250–2, 257, 260–1, see also Zoroastrianism
263–4, 269–70, 273–4, 276, 280, 283, military 123
307, 315 minim see heretics
Roman and rabbinic, compared 27, 202 missionaries, rabbinic 107–11
Sasanian (dād) 250–1, 254–5, 261, 264, monological 184
266, 268, 270, 274–5, 282 dialogism 194–5
Law Book 251, 254, 264, 266, 274 reading of Plato 195
slave 267–8, 269, 271–2 monotheism 36
socioeconomical view on 251 monsters, mythological 245
usucapio 276–8 Mordechai 24, 137, 142, 147, 149–51, 1155–6
meshikah; see law Moses 4, 128, 148, 154–6, 288, 290, 298–300,
Leiden manuscript 312, 315, 335 303–4
Lemech 289–91, 293–5 Muslim conquest 20
Levi, tribe of 106, 301
Levites 107, 300 nasi see patriarch
Libanius 186–9, 191, 199 nazir 220–1
liminal groups 111–112 -ite 220–1
social areas 85 vow(s) 224–5
lineage 9, 10 Nebuchadnezzar 4, 128, 158–60, 162, 244,
literary analysis 15, 117–8, 131–2 331
criticism 18 Nebuzaradan 4
lulav 185 nehutei 6, 134, 285
New Persian 254
magic 22, 32, 35–7, 39–44, 49–50, 51–56, 58, Noahide laws 97
58–62, 66, 70–1, 79–80, 239–40, 245
bowls 19 Obadiah 239
magicians 22, 49 Ohrmazd 245
pagan 36 ona’ah see law of
mamzerim 104, 106 orality, of the Talmud 176
Markulis 159–62 Orestes 196
throwing stones at 159 Origen 58
martyr/dom 23–4, 30, 109, 145–6, 306–7, 315, overreaching, law of see law of ona’ah
317, 321, 327, 335, 347, 349, 351–3
Christian 30, 307, 314, 331–5, 348, 351 Pahalavī 129, 176, 198, 226, 257
Jewish 308, 310 literature 206, 245
in Imperial Rome period 314 see also Zoroastrianism
of Polycarp 307 Parthian 27
of the Ten Martyrs (Midrash Aseret Empire 4
Harugei Malkhut) 325, 332, 333, 335 era 122–3, 129
quasi 146 nobility 123
martyrology 310, 330–1, 335, 348, 351, 353 parting of the ways, theory of 351
general index 375

Paschal lamb 180 purple (tekhelet) 244–5

Pashhur, false-prophets 234–36
Passover see Pesah qarqa olam 136
Patriarch 10, 12, 16, 186, 188–9, 192 qiddush ha-shem see martyrdom
dynasty of 130 Quintilian 178, 197, 216
functions of 145
in Palestinian 130 rabbinic academy see study house
patriotism, of Babylonian rabbis 157 rabbinization 33, 62, 199, 206
Paul 79 Rashi 321
pedophiles 104 Rechabites 247
Pentateuch 36, 147, 156, 228 rei uxoriae 201–2
persecution(s) 23, 134–5, 140–2, 144, 332 rhetor 191
by Roman 145, 335 -ic; questions 197
Hadrianic 306–7, 310, 316–7, 330 schools 195, 199
of Christian 306–7, 335 Roman(s) 25, 64, 115, 262, 273, 278–9
of Jews 306–7, 335 commercial law 250
Persia/n 23–4, 131, 134, 175, 254, 276, 278–9, cultural sphere 131
320 drive for conquest 283
context of the Bavli 206 Empire 4, 186, 199, 206, 281, 306, 335
Empire 173, 227–8 Empire, members of 189
language 157 governor 313, 318, 328–9
see also Pahlavi and Middle Persian invasion (363) 279
Persian names 128–9 investigation of non-Romans 314
-ized Jewry 123 Jurisconsults 250
Pesah 142 legal commentary 200
Pesiqta de-Rav Kahana 105 legal documents 201
Pešyōtan son of Wištasp (Iranian messianic law see law, Roman
hero) 245 legal scholar 203
Peter 79 paganism 19, 282
Pharisees 112 people honorable 281
Pharaoh 128–9 republic 281
Philosophy/pher/phical 190–1, 195, 249, 293, republican values 262
318 social and economic policy 280
phylacteries 43, 47–9, 76–7, 185 tortures 331
physician 134, 138–9 trials 314
pirqa 17, 293 -ization 192–3, 199, 206
piyyut 19 Rosh ha-Shannah 185
Plato 178, 194–5
Pliny 58 Sabbath 11, 37, 61–2, 65, 67, 77–8, 173–4,
praetor 277–8 179–82, 185, 317
presidency (nesiut) see patriarch banquet 63
Progymnasmata 173–231 limit 173
Prophecy/phet/s 4–5, 35, 160, 171, 239 Šābuhr II 130
false 235–7, 240 Sadducee(s) 112, 150
of Eldad and Medad 286, 298–305 safsar 253–4
Elijah 237 safsir, see safsar
Jeremiah 4, 267 Samaritan(s) 52, 111–5
proselytes 97, 105, 107, 109, 154 Pentateuch 228
Purim 149, 155–6 Sanhedrin 10, 15–6, 43, 50, 53, 67, 146, 297
376 general index

Sarah 110 Solomon 235–7, 239, 240–5, 249, 289

Sasanian 23, 26, 129 kingdom of 2
Babylonia 116, 177, 206, 243, 245 building the Temple with the help of
Dastwars 250 Ashmedai 243
Empire 175, 206, 280 sorcery 22, 32–6, 38–41, 50–1, 54–5, 58–9, 79,
era 129–30 80, 314
estimates of overreaching 260 evil 40
law see law, Sasanian sorcerer(s) 22, 35, 37–8, 41, 51–2, 54, 58,
market level 259 313, 353
tradition 246 sorceress 32, 37, 50–1
drive for conquest 283 source criticism 13, 32
religious texts 62 stam 18–9, 23, 26, 51, 59, 61, 84–5, 101–2, 116,
Satan 247 133, 135, 142–6, 174, 179, 182, 202, 205,
savoraim 19, 84 209–12, 216–9, 222–3, 225–6, 286
scholasticism 33, 227–8 of the Yerushalmi 222–3, 225–6
rabbinic 227, 231 Stoicism; elimination of passions 314
scholia 201–2, 205, 226 Storytelling; in the ancient Orient 232
in Aramaic on Hebrew texts 202 Christian 247–8
in Greek for Latin base-text 202 study house 6–9, 14–5, 17
legal, and the Yerushalmi 202 sugya 25, 39, 56, 60, 86, 100, 104, 142–6,
lemmata 201 149–50, 152, 173–231, 292
on Homer 201 give-and-take 1, 176, 179, 182, 184, 187, 191,
Sinaitica 201 194–5, 197–8, 203–5, 226–7
school primers see Greek qushya 173
Scriptores Historiae Augustae 130 shakla ve-tarya 25
Second Temple sects 112 teiruts 173
Sectarian(s) 64, 66, 70, 81, 83 Sumerian dispute poetry 198
-ism 59 synagogue(s) 19, 21, 23, 129
-ian-type magician 59 Synop/sis (Bavli and Yerushalmi) 12–3, 43,
Sefer Yetzirah 32–3, 61, 66–71 77, 84, 88, 108, 124, 161
Sennacherib 171–2, 244 -tic approach 14, 16
Septuagint 142, 228, 230–1 Syriac 65, 227, 248, 332
Shabbat see Sabbat literature (Christian) 198
shakla ve-tarya see sugya schools of education 186
Shalmaneser 3–4
Shekhinah 97, 104, 109 Tanhuma 28–30, 284–6, 291, 294–9, 303–305
Shema 185, 306–35 Buber 30, 285, 299
commandment of reciting 334 tannaitic parallels 49, 309
prohibition to recite 315 source 16, 24, 45, 92, 94, 112–3, 130, 298,
Shisha 232, 237, 241 310, 327
shofar 185 targumim, genre of dispute in 198
Simon Magus 59, 90 Taxmorup 243
slave 267–9, 271 teiruts see sugya
-s, investigation of 314 Theology, theological 36, 39, 71, 243, 280,
-ry in Justinian’s declaration 250 282, 287, 318, 323
social policy 251 Christian 248
Socrates 194–5 innovation 238
Socratic dialogue 178, 195 position 40, 158
general index 377

Tineius Rufus 30, 311, 313–4, 317–8, 329 yarkhei kallah see kallah months
as the (Greek) word tyrannos 308 Yefet (son of Noah) 302
Torah 1–2, 5, 17, 20, 24, 28–9, 44, 48–9, 62, Yehoyachin, King 29
75, 80, 101. 109, 111, 133, 136, 147–57, 172, 181, yeshivah see study house
184, 199, 201, 204, 207–11, 213–4, 220, 224, yom tov 181
230–1, 238–9, 289–91, 295, 311, 316, 320, 328,
330–1, 346 Zand 227
Oral 28–9 see also Zoroastrianism
scholar of 30 Zechariah, tale of the blood of 331
tortures, Roman see Roman tortures Zedekiah, King 4, 29–30
Tosefot 147 Zoroastrian/ism 27, 176, 227, 250, 256, 273,
tree of life 229–31 275, 280, 282, 332
tumtum 214 army 248
concept of God 280
Ulpian 201, 261, 270–2, 277 doctrine 248
ethics 248
Vidēvdād (the Pahlavi composition) 176, legislation 257
227, 257 law 273, 301
Vilna edition of Tractate Berakhot 349 scriptures 249, 257
triad good thought, good speech and good
witch 35, 39, 40–1, 57, 79, 243 action 276, 282
-craft 32, 34–7, 43, 50, 57–8 see also: Avestan literature, Bundahishn,
wizard 243 Denkard, Vidēvdād, Ahariman, Zand
Zosimos, Journey of 247–8
Xwēšīh (Middle Persian, ownership) 278
see also law, Sasanian