The Talmudic Argument

THETALMUDICARGUMENT
A study in Talmudic reasoning and
methodology
LOUIS JACOBS
Rabbi of the NewLondon Synagogue
Visiting Professor at Harvard Divinity School
The right of the
University of Cambridge
to print and sell
all manner of books
was granted by
Henry VIII in 1534.
The University has printed
and published continuously
since 1584.
CAMBRIDGEUNIVERSITY PRESS
Cambridge
London New York New Rochelle
Melbourne Sydney
CAMBRIDGEUNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521263702
Cambridge University Press 1984
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 1984
Re-issued in this digitally printed version 2008
A catalogue record for this publication is available from the British Library
Library of Congress Catalogue Card Number: 84-4351
ISBN978-0-521-26370-2 hardback
ISBN978-0-521-26948-3 paperback
FOR DANIEL
ON THE OCCASION OF HIS BARMITZVAH
Contents
Preface page ix
Note on transliteration xi
Note on Biblical translations xi
Note on the Babylonian Talmud xii
List of Abbreviations xv
1 The Talmudic argument i
2 The literary form of the Babylonian Talmud 18
3 Bererah: retrospective specification 24
4 Teush she-lo mi-da'at: unconscious abandonment of
property 34
5 Rubba: probability 50
6 Davar she-lo ba le-'olam: conveyance of a thing not yet
in existence 64
7 Kol she-eyno be-zeh ahar zeh afilu be-vat ahat eyno:
whatever cannot be established in a consecutive
sequence cannot be established even in a
simultaneous sequence 75
8 Tesh horesh telem ehad: a single act of ploughing can
result in a number of penalties 83
9 Simanin de-oraita 0 de-rabbanan: whether reliance on
distinguishing marks for the purpose of identification
is Biblical or Rabbinic 91
I o Devarim she-be-lev eynam devarim: mental reservations
in contracts are disregarded 101
I1 Hazakah: presumptive state 11 o
12 Gadol kevod ha-beriot: the law and regard for human
dignity 115
viii Contents
13 Hazmanah milta: whether the designation of an object
for a particular use is effective 122
14 Mitzvat 'aseh she-ha-zeman geramah: positive precepts
dependent on time from which women are exempt 133
15 Heyzek she-eyno nikar: indiscernible damage to
property 144
16 Kinyan hatzer: acquisition by means of a domain 160
17 Palginan be-dibbura: admission of part of a testimony
even though another part of the same testimony is
rejected 168
18 Tadir u-mekuddash: which takes precedence: the more
constant or the more sacred? 174
19 Palga nizka: the nature of the payment of
half-damages to which the owner of a goring ox is
liable ^2
20 Patur mi-diney adam ve-hayyav be-diney shamayim: cases
where there is liability in the eyes of God even though
the human courts cannot enforce payment 191
21 Mahal 'al kevodo kevodo mahul: renunciation of honour
by one to whom it is due 198
22 Conclusions 203
Glossary 214
Bibliography 216
Preface
This book seeks to combine the methodological and the literary
approach to the study of the reasoning and argumentation typical
of the Babylonian Talmud. In the great, traditional Yeshivot, most
of which today are influenced by the methods of the Lithuanian
Talmudic giants R. Hayyim Soloveitchik and his son R. Velvel;
R. Simeon Skopf; R. Moshe Mordecai Epstein of Slabodka; R.
Baruch Baer of Kamenitz; R. Joseph Judah Laib Bloch of Telz;
R. Meir Simhah of Dvinsk; R. Elhanan Wasserman and others too
numerous to mention - the Talmud is subjected to acute, logical
analysis of the utmost rigour, to yield important results for the full
understanding of the work. But, in these institutions, historical and
literary questions are virtually ignored. Modern critical investiga-
tion is considered not so much to be taboo as irrelevant. As the old
Yeshiva gibe has it, the Professor of Semi tics can tell you what kind
of clothes Abbaye wore but the lamdan ('traditional scholar') can
tell you what Abbaye said. Modern critical scholars, on the other
hand, in addition to their massive historical researches, have, in
recent years, subjected the Talmudic texts to keen literary analysis,
in which the pre-history of the Talmudic pericopae and their later
formulation have been studied in order to shed fresh light on how
the literary traditions have emerged and to try to solve the mystery
of how the Babylonian Talmud came to assume its present form,
who its authors were and how they went about their task. One
thinks of scholars like I. H. Weiss; Isaac Halevy; C. Albeck;
Abraham Weiss; David Halivni; Saul Lieberman; Hyman Klein;
J. N. Epstein; Benjamin De Friess; Shamma Friedman; Jacob
Neusner and his pupils. Quite naturally, these scholars, in their
concentration on context, tend to leave content to take care of itself
x Preface
to some extent. Whatever originality this book may possess lies in
its attempt to combine the two methods by analysing both the logic
of the arguments and their literary form.
The first two chapters are introductory: one on Talmudic
argumentation in general, its nature, terms and history; the other
on the central literary problem of how the Babylonian Talmud was
composed. The bulk of the work consists of the analytical treatment
of a number of typical sugyot (plural of sugya, the name for a
complete Talmudic unit). Each sugya is paraphrased, the thrust of
the argument noted and an attempt made to subject the material
to literary analysis. The passages in smaller type call attention to
further points requiring elucidation and provide some information
regarding the discussion of the same theme in other parts of the
Talmud as well as in the post-Talmudic works. For this purpose
the two works: Kesef Nivhar and the Entziklopedia Talmudit have
proved invaluable. The final chapter seeks to tie all the material
together and to suggest the conclusions which have been reached
in the investigation.
In reply to the question, allegedly put by the combative person
who witnessed two men fighting: ' Is this a private quarrel or can
anyone join in?' , I hope that the answer, so far as this book is
concerned, is that non-professional students of the Talmud and, for
that matter, anyone interested in Jewish thought, will gain
something from reading the book. At the same time, I am
sufficiently presumptuous to offer the work as a small contribution
towards the solution of the kind of problems faced by full-time
Talmudic scholars.
Note on transliteration
This is not a work on Hebrew and Aramaic philology. The
transliteration from these languages has, consequently, been kept
as simple as possible. In the main the scheme followed is that of
the Encyclopedia Judaica (vol. i, p. 90), e.g. het is represented by h
and hey by h (without the dot) but k represents both koj and kaf.
However, for the sake of greater accuracy, the sign ' is used for' ayin,
to distinguish it from alef (which is not marked). The letter e is used
to represent both the segol and the tzeri {eh and ay) but where
otherwise the e might be read as part of a single syllable the tzeri
is marked e, e.g. nose reads no-say, the mark denoting that it is not
to be read as nose. The sign ' is placed over a vowel letter to denote
that the letter has to be sounded on its own and is not part of a
diphthong or a single syllable with previous letters, e.g. le-hoil is
to be read as le-ho-il, kanesu is to be read as ka-ne-su not kane-su.
NOTE ONBIBLICAL TRANSLATIONS
Quotations are usually from the Authorised Version of the Bible.
In certain cases, however, a free translation is given, where the
particular exegesis of the Talmud requires it.
Where alternative verse numbers are given, they are quoted from
The Torah: A New Translation, Philadelphia, 1962.
Note on transliteration
This is not a work on Hebrew and Aramaic philology. The
transliteration from these languages has, consequently, been kept
as simple as possible. In the main the scheme followed is that of
the Encyclopedia Judaica (vol. i, p. 90), e.g. het is represented by h
and hey by h (without the dot) but k represents both koj and kaf.
However, for the sake of greater accuracy, the sign ' is used for' ayin,
to distinguish it from alef (which is not marked). The letter e is used
to represent both the segol and the tzeri {eh and ay) but where
otherwise the e might be read as part of a single syllable the tzeri
is marked e, e.g. nose reads no-say, the mark denoting that it is not
to be read as nose. The sign ' is placed over a vowel letter to denote
that the letter has to be sounded on its own and is not part of a
diphthong or a single syllable with previous letters, e.g. le-hoil is
to be read as le-ho-il, kanesu is to be read as ka-ne-su not kane-su.
NOTE ONBIBLICAL TRANSLATIONS
Quotations are usually from the Authorised Version of the Bible.
In certain cases, however, a free translation is given, where the
particular exegesis of the Talmud requires it.
Where alternative verse numbers are given, they are quoted from
The Torah: A New Translation, Philadelphia, 1962.
Note on the Babylonian Talmud
For readers totally unfamilar with the Talmud this brief outline
gives the essential facts. Three easily remembered (but only
approximate) dates are: 200 G.E. when the Mishnah was compiled;
400 G.E. when the Palestinian Talmud (also known as Terushalmi,the
Jerusalem Talmud) was compiled; and 500 G.E. when the Babylo-
nian Talmud was compiled. The Mishnah ('Teaching') is a digest
of what the Rabbis call ' the Oral Torah' , i.e. the detailed
expositions, as these developed in the Palestinian schools, of' the
Written Torah' , the Pentateuch and the rest of the Hebrew Bible.
The teachers whose views are recorded in the Mishnah and in other
works from the same period are known as Tannaim ('teachers').
The Mishnah is divided into six Orders and these are subdivided
into tractates and chapters. Each chapter contains a number of
smaller units, each of which is also called a Mishnah. Thus
Mishnah is the term for the smallest unit as well as for the work
as a whole. The Mishnah is in Hebrew, this being the scholarly
language in Palestine in the first two centuries.
In the schools of Palestine and Babylon the Mishnah became
accepted as a sacred text, taking its place by the side of the Bible.
In these schools the Mishnah was discussed and interpreted so that
the name given to the post-Mishnaic teachers is Amoraim
('interpreters'). Around the year 400 C.E. the debates and
discussions of the Palestinian Amoraim were collected and edited
to form the Palestinian Talmud. (The word ' Talmud' also means
'teaching' or 'study'.) Around the year 500 the material from the
Babylonian schools was collected and edited to form the Babylonian
Talmud. Both the Palestinian and the Babylonian Talmuds are in
Aramaic (the language of discourse even among the scholars in this
Note on the Babylonian Talmud xiii
period) but the former is in the Western dialect of Aramaic, the
latter in the Eastern. The Babylonian Talmud is far more
comprehensive than the Palestinian and is often referred to as the
Talmud, although, strictly speaking, there are two Talmuds.
Although the Mishnah exists both in manuscript and in print on
its own, it is always presented, too, as part of the Talmud. Thus
the Babylonian Talmud in its present form consists of the Mishnah
together with the Talmud proper, the material from the Babylonian
schools.
There was constant scholarly exchange between the two centres
of Palestine and Babylon with the result that teachings of the
Babylonian Rabbis are found in the Palestinian Talmud and
teachings of the Palestinian Rabbis in the Babylonian Talmud.
Eventually, out of fear of the censor (for whom the Talmud was
suspect as allegedly containing attacks on Christianity), a substitute
name for ' Talmud' was used. This is the term ' Gemara' (also with
a meaning of'teaching' but here of a particular text). This term
is used in the Talmud frequently for a particular passage or text
but is now used for the Talmud as a whole. Thus, nowadays, and
in all printed versions, the Babylonian Talmud is called ' The
Gemara', consisting, as above, of the Mishnah and the Gemara.
The complete Talmud was first printed in Italy in the sixteenth
century, since when it has gone into numerous editions. The
standard edition of the Babylonian Talmud is that printed and
reprinted (and now frequently photocopied) in Vilna by the printing
house of Romm. There is a uniform pagination in all editions of
the Babylonian Talmud. Each folio page has two sides - a and b.
The method used to quote a passage from the Talmud is to refer
to the tractate, the page and the side; thus Bava Kama i ia means
tractate Bava Kama, page 11, side a; Toma 23b means tractate Toma,
page 23, side b. The Babylonian Talmud has been translated into
English, under the editorship of Dr I. Epstein, and published by
the Soncino Press.
The following is a list, in order, of the tractates of the Babylonian
Talmud. (There is no Babylonian Talmud on a number of tractates
of the Mishnah.) The list gives the name of the tractate, its
translation and a brief description of its subject matter.
Berakhot: 'Blessings'; the Shema' and the daily prayers
Shabbat: ' Sabbath' ; the Sabbath laws
'Eruvin: 'Mixtures'; the laws of Sabbath boundaries
xiv Note on the Babylonian Talmud
Pesahim: 'Passovers'
Toma: ' The Day' ; the laws of the Day of Atonement, Tom Kippur
Sukkah: 'Booth', the laws of Tabernacles
Betzah: ' Egg' (after the opening word); the laws of the Festivals
Rosh ha-Shanah: ' Beginning of the year' ; the laws of the New Year
festival
Tctanit: ' Fast' ; the laws of Fast Days
Megillah: 'Scroll (of Esther)'; the laws of the Purim Festival
Mo'ed Katan: 'Minor Festival'; the laws of the intermediate days
of the Festivals
Hagigah: 'Festival offering'; the special offering for the Pilgrim
Festivals
Tevamot: 'Sisters-in-law'; the laws of levirate marriage
Ketubot: 'Marriage settlements'
Nedarim: ' Vows'
Nazir\ 'Nazirite'
Sotah: ' Wife suspected of adultery'
Gittin: 'Divorces'
Kiddushin: ' Marriage'
Bava Kama: 'First gate', dealing with torts
Bava Metzi'd'. 'Middle gate', dealing with civil law
Bava Batra: 'Last gate', dealing with property law
Sanhedrin: procedures of the Sanhedrin and other Law Courts
Makkot: 'Floggings', dealing with criminal law
Shevitot: ' Oaths'
'Avodah arah: 'Strange worship'; laws concerning idolatry
Horayot: 'Court decisions', given in error
Zevahim: 'Sacrifices', of animals in the Temple
Menahot: ' Meal offerings'
Hullin: 'Non-sacred', animals, and their preparation as food
Bekhorot: ' First-borns'
'Arakhin: 'Assessments', of gifts to the Temple
Temurah: 'Exchange', of a sacred animal
Keritot: ' Extirpations'
Me'ilah: 'Trespass-offering'
Niddah: ' Menstruant'
Abbreviations
ARN
b.
BSOAS
C.E.
Dor
DS
EJ
ET
Gilyon ha-Shas
ha-Gra
HUCA
JE
JJS
JQR
JSS
Kesef Nivhar
Maggid Mishneh
Maharsha
Meiri
Mevuot
R.
R.
Radbad
Ramban
Ran
Avot de-Rabbi Natan
ben, 'son of
Bulletin of the School of Oriental and African Studies
Common Era
Dor Dor ve-Doreshav by I. H. Weiss
Dikdukey Soferim by R. Rabbinovicz
Encyclopedia Judaica
Entziklopedia Talmudit
Notes of R. Akiba Eger (i 761-1837)
Notes of Elijah Gaon of Vilna (1720-97)
Hebrew Union College Annual
Jewish Encyclopedia
Journal of Jewish Studies
Jewish Quarterly Review
Journal of Semitic Studies
by Baruch Benedict Goitein (c. 17 70-1842)
Commentary on Maimonides' Tad by Vidal
Yom Tov of Tolosa (fourteenth century)
Novellae of R. Samuel Edels (1555-1631)
Bet ha-Behirah by Menahem Meiri of Perpignan
(1249-1316)
Mevuot le-Sifrut ha-Amoraim by J. N. Epstein
Rabbi
(after Genesis, Exodus etc.) Midrash Rabbah
R. Abraham ben David of Posquires (c. 112598)
R. Moses ben Nahman, Nahmanides
(1194-1270)
R. Nissim Gerondi (d. c. 1375)
xvi List of abbreviations
Rashba R. Solomon Ibn Adret (c. 1235-c. 1310)
Rashi R. Solomon Yitzhaki (10401105)
Riban R. Judah ben Nathan, son-in-law of Rashi
Ritba R. Yom Tov Ishbili (c. 1250-1330)
Rosh R. Asher ben Yehiel (c. 12501327)
SM Shitah Mekubbetzet by Bezalel Ashkenazi
{c. 1520-91)
Soncino The Babylonian Talmud, translated into English,
Soncino Press
Tosafists Glosses to the Talmud by the French and
German schools, twelfth to fourteenth century
Tur The Turim of Jacob ben Asher (d. c. 1340)
Tad Tad ha-Hazakah, Code of Maimonides
(1135-1204)
I
The Talmudic argument
The Babylonian Talmud consists almost entirely of arguments
having as their aim the elucidation of the law, ruling, religious
teaching or ethical idea. Theories are advanced and then contra-
dicted. They are examined from many points of view and qualified
where necessary. One argument leads to another when logic
demands it. The claims of conflicting theories are investigated with
great thoroughness and much subtlety. Fine distinctions abound
between apparently similar concepts. The whole constitutes reas-
oning processes which have received the most careful study on the
part of generations of Jewish scholars and have contributed more
to the shaping of the Jewish mind than any other factor.
No serious student of the Babylonian Talmud can be unaware
that, for all the variety of topics discussed in the work, there is a
formal pattern to the argumentation. Whatever the subject matter,
the moves open to the debaters are comparatively few in number
and these are always expressed in the same stereotyped formulae.
There is much originality in Talmudic argumentation but this
consists in the application to new situations of conventional
responses, not in the invention of new responses. The game is always
played according to the rules.
These formal methods of argumentation occur with the utmost
frequency in the Babylonian Talmud yet, although there is to be
observed a complete consistency in their use, nowhere in the
Talmud itself is any attempt made at their enumeration and
classification. Part of this task was left to the famous post-Talmudic
methodologies, largely concerned with the classification of Tal-
mudic method. However, in the main, the Talmudic method-
ologies deal with the precise definition of the terms used rather than
2 The Talmudic argument
with the actual types of argument. Of these there has been very
little detailed, systematic treatment.
For Talmudic methodology see W. Bacher, Exegetische Terminologie der
jiidischen Traditionsliteratur, Hebrew translation by A. Z. Rabbinowitz
under title ' Erkhey Midrash (Tel-Aviv, 1924); H. Strack, Introduction to the
Talmud and Midrash (Philadelphia, 1945), pp. 135-9; M. Meilziner,
Introduction to the Talmud, 4th edn with new bibliography by A. Guttmann
(New York, 1968), part i n: ' Talmudic terminology and methodology', pp.
190-280; and I. H. Weiss' survey of the methodological literature in Bet
Talmud, vol. 1 (Vienna, 1881) and vol. 11 (Vienna, 1882). The most
important of the methodologies are: Mevo ha- Talmud attributed to Samuel
ha-Naggid (d. 1055), printed with commentaries in the Vilna edition of
the Talmud after tractate Berakhot; Sefer Keritut by Samson of Chinon (end
of thirteenth century), ed. J. Z. Roth with commentary (New York,
1961), and by S. B. D. Sofer with commentary (Jerusalem, 1965); Halikhot
' Olam by Joshua ha-Levi of Tlemcen (compiled in Toledo c. 1467), edition
Warsaw, 1883, with commentaries: Kelaley ha-Gemara by Joseph Karo
(1488-1575) and Tavin Shem'va by Solomon Algazi (seventeenth century);
Sheney Luhot ha-Berit by Isaiah Horowitz (d. c. 1630) (Amsterdam, 1649),
section Torah she-be-'al Peh; Tad Malakhi by Malachi Ha-Kohen of
Leghorn (early eighteenth century) (Jerusalem, 1976).
In every complete Talmudic unit - the sugya the differing
views are presented in the form of a debate. The protagonists may
be actual teachers expressing their opinions, Rav and Samuel,
R. Johanan and Resh Lakish, Rabbah and R. Joseph, Abbaye and
Rava and so forth. Very frequently, however, the thrust and parry
of the debate is presented anonymously. It has long been
conventional among students of the Talmud to give a kind of
fictitious personality to the arguments by attributing questions to
an alleged 'questioner' (the makshan) and answers to an alleged
'replier' (tartzan). In a particular sugya different types of argument
may be produced as the course of the debate demands, e.g. an
appeal to authority, the detection of flaws in an analogy, the
readmission of a rejected plea and so forth. The unit as a whole
consists of all the arguments, together with any extraneous material
which may arise as the discussion proceeds.
On the use of the terms makshan and tartzan see e.g. Rashi, Shabbat 104a top;
Kiddushin 2a, s.v. ve-khesef minalan; Tosqfists to Berakhot 44a s.v. inhu and
Toma 43b s.v. amar R. Tehudah. Cf. ben Yehudah's Thesaurus, vol. vn, p.
3295 and vol. xvi, p. 7923.
The sustained argument, particularly in the form of question and
answer, as presented in the Babylonian Talmud, has strong
The Talmudic argument 3
antecedents in the earlier sources. There are numerous instances
in the Pentateuch and in the historical books of the Bible: Eve's
debate with the serpent (Genesis 3: 1-5); God's accusation of Adam
(Genesis 3: 9-13); God's interrogation of Cain (Genesis 4: 9-15);
Abraham's plea to God to spare Sodom (Genesis 18: 2333);
Abimelech's upbraiding of Abraham (Genesis 20: 9-17);
Abraham's reproof of Abimelech (Genesis 21: 22-30); Abraham's
purchase of the field and bargaining with Ephron (Genesis 23:
316); Jacob's dialogue with his wives (Genesis 31:416) and with
Laban (Genesis 31: 26-53); the arguments presented by Schechem
and Hamor (Genesis 34: 4-23); the arguments and counter-
arguments in the Joseph saga (Genesis 42: 7-38; 43: 2-14; 44:
6-34); Joseph's bargain with the Egyptians (Genesis 47: 15-25);
Pharaoh's complaint against the midwives (Exodus 1: 15-19);
Moses' confrontation with God (Exodus 3: 4 to 4: 17); Moses'
dialogue with Pharaoh (Exodus 10: 1-11); Jethro's plea for reforms
(Exodus 18: 1323); Moses' entreaty on behalf of his people
(Exodus 32: 7-14 and 33: 12-23); Moses' questioning of Aaron and
his sons (Leviticus 10: 16-20); Moses' complaint (Numbers 11:
11-23);
t n e
episode of Moses, Aaron and Miriam (Numbers 12:
1-14); the debate between Caleb and the spies (Numbers 13: 27
to 14: 10); the rebellion of Korah (Numbers 16: 119); Moses and
the King of Edom (Numbers 20: 1421); Balaam and the ass
(Numbers 22: 28-35); the account of the sons of Reuben and Gad
(Numbers 32: 1-32); the two lawsuits concerning the daughters of
Zelophehad (Numbers 27: 1-7 and 36: 1-10). The major portion
of the book of Deuteronomy consists of a sustained argument in
which Moses reminds the people of their history and their oblig-
ations in the future. Among arguments of this type in the historical
books are: Rahab and the spies (Joshua 2: 1-21); the debate
between the other tribes and the sons of Reuben and Gad (Joshua
22: 13-14); Jotham's argument (Judges 9: 7-20); Jephthah and
the sons of Ammon (Judges 11: 12-28); Samuel against the
appointment of a king (I Samuel 8: 10-21); his rebuke of the people
in this matter (I Samuel 12: 1-24); his castigation of Saul (I Samuel
13: 10-14); Eliab and David (I Samuel 17: 28-37); Jonathan and
Saul (I Samuel 20: 21-32); David at Nob (I Samuel 21: 2-7);
David and Saul (I Samuel 24: 9-22); David and Abigail (I Samuel
25: 23-35); David and Uriah (II Samuel 11: 10-12); David and
the death of his child (II Samuel 12: 17-23); the woman of Tekoa
4 The Talmudic argument
(II Samuel 14: 4-24); Barzillai and David (II Samuel 19: 32-40);
Bath-sheba and David (I Kings 1: 11-27); Solomon and the two
harlots (I Kings 3: 16-27) ' > Obadiah and Elijah (I Kings 18: 7-15);
Elijah and the prophets of Baal (I Kings 18: 21-7); the four lepers
(II Kings 7: 3-4); Rab-shakeh and Eliakim (II Kings 18: 19-35).
As for the prophetic books, the whole of the prophetic message
is in the form of a sustained argument. The passages especially to
be noted are: Isaiah 40: 12-26; 44: 9-20; 49: 14-26; 51: 12-13;
58: 2-14; 66: 1-2; Jeremiah 2: 4-37; 12: 113; Ezekiel 18: 1-29;
Amos 3: 3-8; 6: 1-2; 9: 7; Jonah 1: 6-15; 4: 2-11; Micah 6: 3-8;
Zechariah 4: 1-7; Malachi 1: 2-14; 2: 10-17; 3: 13-16.
The same type of sustained argument is found in the book of
Psalms. Psalm 10 in its entirety is a plea for the salvation of the
righteous from the hands of the wicked. In both Psalm 15 and Psalm
24 the way of righteousness is prescribed and expressed as a reply
to a question. Psalm 23 draws the conclusion that man should trust
in God from the premiss that the Lord is his Shepherd. Psalm 50
is an argument in favour of the view that it is righteousness that
God wants, not sacrifices. Psalm 96 is a mighty plea that God be
praised and Psalm 100 that He is to be thanked. These themes are
repeated in Psalms 104 and 105. Psalm 112 is an argument for
righteous living and Psalm 115 against idolatry. Psalm 119 exhausts
the letters of the alphabet eight times in calling attention to the need
for man to be loyal to God's law. In Psalm 136 there are a number
of'proofs' that God's mercy 'endureth for ever'. Psalm 139 argues
that it is impossible to escape from God. Psalm 146 argues that it
is better for man to put his trust in God than in princes. And the
book of Psalms in general is mainly an appeal by argument to God,
expressed in poetry, that He should pay heed to the cry of the poor
and oppressed.
In the other books of the Hagiographa the same phenomenon
is to be observed. In Proverbs we find arguments for the cultivation
of wisdom (3: 13-23); against harlotry (5: 1-20; 7: 5-23); against
sloth (6: 6-15); against wickedness (10: 2-32); and for the worth
of a good wife (31: 1031). The book of Job has the arguments of
Satan (1: 6 to 2: 6); of Job's friends and his replies (4: 1 to 37: 24);
of God to Job (38 to 41) and Job's reply (42: 1-6). The book of
Ruth contains Naomi's argument to her daughters-in-law (1:
8-170) and that of Boaz with his kinsman (4: 1-12). In addition
to the argument for pessimism, the general theme of Ecclesiastes,
The Talmudic argument 5
there are arguments in favour of melancholia (7: 26); of wisdom
(7: 10-12); of mirth (8: 15); and of effort (11 :i-8). Esther contains
the arguments advanced by Memucan for deposing Vashti (1:
16-22); Haman's arguments for destroying the Jews (3: 8-11);
Esther's dialogue with Mordecai (4: 10-14) and Esther's plea for
her people (7: 3 to 8: 6).
The dialectical tone of the above passages was no doubt familiar
to the Babylonian Amoraim and, for that matter, to many of their
fellow-Jews, from infancy. In addition, so far as we can tell, the
hermeneutical principles laid down by the Tannaim were widely
discussed and accepted by all the Amoraim. These principles are
themselves largely ways of argumentation and references to them
abound in the Babylonian Talmud.
It should be noted that it is acknowledged in the Rabbinic literature itself
that the argument from the minor to the major is found in the Bible; see
my article: 'The "qal va-homer" argument in the Old Testament', BSOAS,
35: 2 (1972), 221-7. A. Schwarz, in Der Hermeneutische Syllogismus in der
Talmudischen Litteratur (Karlsruhe, 1901); Die Hermeneutische Antinomie in der
Talmudischen Litter atur (Vienna, 1913); Die Hermeneutische Quantitdtsr elation
in der Talmudischen Litteratur (Vienna, 1913) and in other works, has
examined the thirteen principles of R. Ishmael with the utmost attention
to detail. These principles are found in Sifra, Introduction. The seven
principles attributed to Hillel are found in Sifra, Introduction, end; ARJV
37; Tosefta Sanhedrin, 7, end. The twelfth-century Karaite author Judah
Hadassi argued for Greek influence on the hermeneutic principles and this
matter has been discussed by David Daube, 'Rabbinic methods of
interpretation and Hellenistic rhetoric', HUCA, 22 (1949), 239-64 and by
Saul Lieberman, Hellenism in Jewish Palestine (New York, 1950), pp. 47-82.
Cf. J. Z. Lauterbach 'Talmud hermeneutics', JE, vol. xn, pp. 30-3; Ghaim
Hirschensohn, Berurey ha-Middot (New York, 1929-31); M. Ostrowsky,
ha-Middot she-ha-Torah JVidreshet ba-Hem, (Jerusalem, 1924); and my article
'Hermeneutics', EJ, vol.vra, pp. 366-72. The Palestine Talmud uses some
of the methods of argument that are found in the Babylonian Talmud but
these are in a much less finished form in the Palestinian Talmud; see I.
H. Tavi ob, ' Talmudah shel Bavel ve- Talmudah shel Eretz TisraeV in his Collected
Writings (in Hebrew) (Berlin, 1923), pp. 73-88 and the Introduction,
Homat Terushalayim by S. Feigensohn (Sha/an ha-Sofer), based on Z.
Frankel's work, in the Vilna edition of the Palestinian Talmud.
The Babylonian Amoraim thus had a long tradition behind them
of skill in debate and argument. The study of the Torah was their
consuming purpose in life to which they applied themselves with
ruthless devotion and dedication. Over the years, the exercise of
their minds in these dialectics seems to have produced an automatic
6 The Talmudic argument
response to the problems they were concerned to solve. Naturally,
it is necessary to distinguish between the use of argument by the
Amoraim themselves and the use by the final editors. This problem
will be considered in the following chapter. But even if, as seems
extremely probable and as we shall see there, the final form of these
arguments owes much to the redactors or compilers of the Talmud,
the methods must have had a history and some were almost
certainly used even by the earliest of the Babylonian Amoraim.
On the more general question of attitudes towards skill in
debating matters of Torah, it is clear that such skill was highly
praised. In the Mishnah (Avot 2: 8) we find a report that Rabban
Johanan b. Zakkai praised his disciple R. Eliezer b. Hyrcanus,
saying that he was ' a plastered cistern which loses not a drop' , i.e.
he had an extraordinarily retentive memory and was able to recall
everything he had been taught. He praised another disciple, R.
Eleazar b. Arakh, saying that he was ' an ever-flowing spring', i.e.
he had the ability to advance fresh, original arguments and theses.
Two versions are then recorded. According to one of these, the
master declared that if all the Sages of Israel were in one scale of
the balance and Eliezer b. Hyrcanus in the other he would out-
weigh them all. The other version, in the name of Abba Saul, is:
if all the Sages of Israel were in one scale of the balance, together
with Eliezer b. Hyrcanus, and Eleazer b. Arakh was in the other
scale, he would outweigh them all. A Talmudic report ('Erwin 13b)
about the second-century teacher R. Meir says that no one in his
generation could be compared to him in brilliance but that, none
the less, the law does not follow his opinions because his colleagues
were incapable of penetrating to the depths of his mind and the
law always follows the majority opinion. Of R. Meir it is also said
that he was able to produce arguments to render the clean unclean
and the unclean clean. This statement was puzzling to the Tosajists
to the passage. ' What is so meritorious', they ask, 'in arguing
against the laws of the Torah?' Is it possible that we have here an
echo of an institution in which disciples were taught to test their
skills in argumentation by arguing for positions known to be false
because contradicted by the Torah? In the same Talmudic passage
it is stated, R. Judah the Prince declared that the reason his intellect
was sharper than his colleagues' was because he had had the privi-
lege of sitting in the lecture-hall behind R. Meir. Had he sat in
front of the master when he taught, his brilliance would have been
The Talmudic argument y
even greater; no doubt a reference to the use of gestures and facial
expressions by the teacher in order to convey the teachings more
effectively. In support the verse is quoted: 'But thine eyes shall
see thy teachers' (Isaiah 30: 20). In the legend told {Shabbat 33b)
regarding R. Meir's contemporary, R. Simeon b. Yohai, it is said
that this teacher spent twelve of privation in a cave, which had the
effect of heightening his intellectual powers. Before that time, when
R. Simeon suggested a problem his son-in-law was able to provide
12 different solutions, but after the experience in the cave the roles
were reversed: to every problem set by his son-in-law R. Simeon
was able to offer 24 different solutions. It is said [Bava Metzi' a 84a)
of the two third-century Palestinian Amoraim, R. Johanan and R.
Simeon b. Lakish, that 'Resh Lakish' was able to produce 24
objections to every statement made by R. Johanan. When Resh
Lakish died, the Sages sent R. Eleazar b. Pedat as a substitute for
Resh Lakish, but R. Johanan found him very unsatisfactory. R.
Eleazar was able to do no more than produce 24 proofs in support
of R. Johanan's statements and this was of no help to the master,
who preferred to be challenged, as he was by Resh Lakish.
To be noted is the number 24, in this narrative and the one about R.
Simeon b. Yohai and his son-in-law. In the latter story the number is
perhaps a play on the number 12, the period of years R. Simeon spent in
the cave. The number 24 is a formal number, almost certainly corres-
ponding to the 24 books of the Bible with which the scholar is expected to
be familiar; see Exodus R. 41: 5 and Tanhuma, ed. Buber, to Exodus
31:18 and Buber's notes. Cf. the number 48 (twice 24) in the account of
Symmachus' reasoning powers in 'Eruvin 13b and the 48 days in which
the Torah is acquired, Avot 6: 5 (Kinyan Torah). Gf. Ta'anit 8a: ' R. Adda
b. Ahavah used to arrange his lessons in proper order 24 times,
corresponding to the number of books of the Torah, Prophets and
Hagiographa, before he appeared in the presence of Rava.'
The two types of scholar referred to, as above, as ' the plastered
cistern' and ' an everflowing spring' were called, in the Amoraic
period (Berakhot 64a), ' Sinai' (one who knows the whole Torah as
it was given at Sinai) and 'oker harim ('uprooter of mountains').
In an age when teachings were transmitted orally, the scholar with
vast stores of information was highly regarded but his claim to
pre-eminence was hotly contested by admirers of the less knowl-
edgeable but more original and brilliant scholar. Rabbah was such
a scholar while his colleague, R.Joseph, belonged to the other type.
The scholars of Palestine, when asked for their advice, sent a
8 The Talmudic argument
message that R. Joseph was to be preferred as the Sinai type
[Horayot 14a). This same Rabbah, it is said, was fond of encouraging
his disciples to cultivate sharpness of mind by appearing, on
occasion, to act contrary to the law in order to see whether the
disciples would be sufficiently alert to spot his mistakes {Berakhot
33b; Hullin 43b; Niddah 4b). This method of'alerting the mind of
the disciples' {le-hadded et ha-talmidim) is also said to have been
practised by Samuel ('Eruvin 13a); by R. Akiba (see Niddah 45a);
and by the latter's teacher, R.Joshua (Nazir 59b). In another
passage (evahim 13a) the method is attributed to Rabbah's
teacher, R. Huna. This idea of sharpening the wits of the disciples
must not be confused with the demand that words of Torah should
be ' sharp' in the scholar's mouth (Kiddushin 30a) even though the
same term (mehuddadim) is used, since there the meaning is a sharp
clear utterance ('when someone asks you a question do not
stammer when you tell him the answer'; cf. Sifre to Deuteronomy
6: 7, where the reading is mesudarim, 'well-ordered', see edn.
Friedmann, p. 74a and Friedmann's note).
The Talmudic debate and argument reached its apogee in the
work of Abbaye and Rava, Rabbah's disciples. Hundreds of de-
bates between these two are recorded in the Babylonian Talmud.
Later generations considered the work of these two to be so typical
of Rabbinic learning that when they wished to list the many themes
with which Rabban Johanan b. Zakki was conversant they referred,
anachronistically, to his being familiar with ' the arguments of
Abbaye and Rava' (havayot de-Abbaye ve-Rava), though these are
described as a 'small thing' in comparison with the 'great thing',
the mystical study of the Heavenly Chariot seen by Ezekiel (Sukkah
28a; Bava Batra 134a).
A study of Abbaye and Rava, listing every reference to them in the
Talmud, is: Abbaye ve-Rava by J. L. Maimon (Jerusalem, 1965). In the
twelfth century, Maimonides identified ' the arguments of Abbaye and
Rava' with the typical approach of the whole range of Talmudic study,
Tad, Tesodey ha-Torah 4: 13. David Kimhi remarked in a letter (Kovetz
Teshuvot ha-Rambam, ed. A. Lichtenberg (Leipzig, 1859), partm, pp. 4c-d)
that, for all his love of philosophy, he was thoroughly familiar with havayot
de -Abbaye ve-Rava. Cf. Frank E. Talmage, David Kimhi The Man and His
Commentary (Harvard University Press, 1975), pp. 378. In later ages this
identification became a commonplace so that the term havayot de-Abbaye
ve-Rava was used as a synonym for the Halakhic discussions of the Talmud.
Some time before the Amoraic period, the debate in Tor ah
matters was described in military terms - milhamtah shel Torah
The Talmudic argument g
(Sanhedrin 111 b). On the verse:' And he carried away all Jerusalem,
and all the princes, and all the mighty men of valour' (II Kings
24: 14) the Sifre (to Deuteronomy 32: 25) comments: ' What
mighty deeds could have been accomplished by men taken into
captivity and what kind of warfare could men bound in chains have
engaged in? But "all the mighty men of valour" means, in the
warfare of the Torah.' This enabled the Rabbis to interpret Biblical
verses glorifying military prowess as referring to the battles of the
mind. For instance, on the verse: ' Happy is the man that hath his
quiver full of them: they shall not be ashamed, but they shall speak
with the enemies in the gate' (Psalm 127:5), a Rabbi commented:
even father and son, master and disciple, become enemies of one
another when they are on opposing sides in the Torah debates
(Kiddushin 30b). The debates were said to take place even in
Heaven, the scholars in the Yeshivah on High having the right to
disagree even with God Himself (Bava Metzi' a 86a). Both motifs,
of military metaphor and debate in Heaven, are present in the
comment {Bava Kama 92a) on: 'Hear, Lord, the voice of Judah, and
bring him unto his people; let his hands be sufficient for him; and
be thou an help to him from his enemies' (Deuteronomy 33: 7).
Moses prayed that Judah be admitted to the Heavenly Yeshivah
but Judah was unable to understand the debates in order to
participate in them. Moses' prayer for Judah to participate was
granted, but a further plea by Moses was required before Judah
was able to argue so convincingly that his decisions in matters of
law could be followed. In the same vein are the statements regard-
ing King David when he rendered legal decisions (Berakhot 4a).
R. Judah, in the name of Rav, interprets (Sanhedrin 93b) the verse
praising David's qualities (I Samuel 16: 18) as referring to his skill
in debate:' that is cunning in playing' knowing the right questions
to ask; ' a mighty valiant man' knowing the correct answers; ' a
man of war' - knowing how to give and take in the battle of the
Torah; ' prudent in matters' - knowing how to deduce one thing
from another; ' and a comely person' -who demonstrates the
proofs for his opinions; ' and the Lord is with him' the ruling is
always in accordance with his views.
Skill in Torah debate was also compared to the skill exhibited
by a competent craftsman. The 'craftsmen and smiths' carried
away into captivity (II Kings 24: 14) were identified with scholars
gifted with great reasoning powers (Sifre to Deuteronomy 32: 25).
On the basis of this the keen debater was compared to a carpenter.
i o The Talmudic argument
Of a text presenting severe problems of interpretation it was said
that neither a carpenter nor his apprentice could remove the
difficulties (' Avodah arah 50b, cf. Palestinian Talmud Tevamot 8:
2, 9b). In similar vein scholars were compared to builders (Berakhot
64a), possibly because scholars ' built up' their arguments, as in the
very frequent binyan av, 'father construction', for an argument by
inference where the premiss is the 'father' to the conclusion reached
by a process of' building' . The expression: ' Do you weave them
all in the same web' (Berakhot 24a; Shabbat 148a; Pesahim 42a; Hullin
58b) suggests that scholars were compared to weavers, as does the
use ofmassekhet, ' web' , for a tractate. The purveyor of the difficult
Halakhic teachings was compared to a dealer in precious stones for
the connoisseur whereas the more popular but less profound
Aggadic teacher was compared to the retailer of cheap tinsel goods
which all can afford to buy (Sotah 40a).
The keen scholar was called a harif 'sharp one'. Thus there is
a discussion as to which is the superior scholar, the harif, capable
of raising objections, or the more cautious debater who is less quick
in refutation but can arrive more readily at a correct solution
(Horayot 14a). When a particularly pungent argument was seen to
solve a problem far more effectively than more learned but
pedestrian attempts, it was said that one grain of sharp (harifta)
pepper is worth more than a basket-full of pumpkins (Megillah 7a).
The acuteness of a scholar's reasoning process was spoken of as his
'sharp knife' (Hullin 77a, cf. Tevamot 122a). The scholars of
Pumbedita were especially renowned for their sharpness. Among
these were Efa and Avimi, described as ' the sharp ones of
Pumbedita' (Sanhedrin 17b; Kiddushin 39a; Menahot 17a). The
brilliance of the Pumbeditans was, however, somewhat suspect in
that it bordered on the eccentric, so that they acquired the
reputation of ' causing an elephant to pass through the eye of a
needle', i.e. of producing far-fetched, improbable arguments (Bava
Metzi'a 38b, cf. Berakhot 55b).
There are found in the Babylonian Talmud a number of formal
terms for the moves in an argument and for the argument itself,
some of them of earlier usage. The earlier term for argumentation
and debate is nose ve-noten, ' give and take' (Sifre to Deuteronomy
32: 25). This term is also used (e.g. in Shabbat 31a) for business
dealings,' buying and selling', as in the idiomatic English expression
'selling an idea'. The Aramaic equivalent is shakla ve-taria (Bava
The Talmudic argument 11
Kama 92a; Sotah 7b). The reasoning by means of which an
argument is supported is sevara ('theory', 'reasoning', 'common-
sense') with a root meaning of' t o think'. Very frequently in the
Talmud one finds the expression:' mar savar... u-mar savar...',' this
master holds. . . and this master holds. . . ' . The term Saboraim, for
the post-Talmudic teachers is derived from this term; perhaps ' ex-
pounders'. An objection is a kushia (Bava Kama 117a; Bava Metzi'a
84a and very freq.), from a root meaning of 'hardness', hence,
a difficulty. A stronger term, used when the difficulty is insur-
mountable, is tiyuvta, 'refutation' (Bava Kama 15b and freq.). The
term meteve, ' an objection was raised' (from the same root) and
similar expressions are used when the refutation is from a Tannaitic
source. When the objection is based on Amoraic reasoning, the term
used is matkif, from a root meaning ' to seize', i.e. A seized hold of
B's theory and sought to refute it. The reply to an objection raised
is tirutz, ' an answer', from a root meaning of' t o make straight'
(Gittin 4b; Bava Metz? a 14b and freq.), generally used when a diffi-
culty is 'straightened out', e.g. by emending a text and the like.
A more direct reply is piruka, ' a reply', from a root meaning of' to
break', i.e. to shatter the objection (Bava Kama 117a; Bava Metzi'a
84a; ' Avodah arah 50b and freq.). The term for a proof advanced
in support of a theory is found in the Mishnah (' Eduyot 2:2) and
used by the Amoraim (Pesahim 15a). The term is radyah, literally
' a seeing'. The Aramaic equivalent is sa'ya,' a support', with a root
meaning of to assist' (Sanhedrin 71b; Hullin 4a and very freq.). The
unit of argument and counter argument is shema'ta, from a root
meaning ' to hear' (Kiddushin 50b; Sanhedrin 38b). The shema'ta
generally consists of the text, gemara, and its exposition, the severa.
The bringing of an argument to a successful conclusion so that it
results in a correct application of the law is ' asukey shema
i
ta aliva
de-hilkheta, i.e. bringing the argument to a final ruling, to the
halakhah, the actual ruling in practice (e.g. in Bava Kama 92a). An
abstract problem of definition in which the two ways of looking at
the matter are so equally balanced that, without proof from some
authority, it is impossible to decide which is correct, is known as
a ba'ya, from a root meaning of to request', i.e. to request a solution
to the problem (Berakhot 2b; Pesahim 9b and very frequently
throughout the Talmud). Solutions to this kind of problem are
described by terms taken from the rootpashat, ' to make clear', ' to
smooth out' a difficulty (Berakhot 26b; Kiddushin 9b and very freq.).
12 The Talmudic argument
Where no solution is forthcoming the term used is teyku, ' it remains
standing', i.e. the two possibilities are so equally balanced and, in
the absence of proof from authority, there is no solution to the
problem; it is by nature insoluble (Berakhot 8a; 25b; and very freq.).
There are over 300 instances of this phenomenon in the Babylonian
Talmud but none in the Palestinian. A doubt about the facts used
in an argument or about the correctness of a theory that has been
advanced is safek, ' a doubt ' (Berakhot 3b; Bava Metzi'a 83b; Keritot
21b and freq.).
On the teyku phenomenon see my study: (London-New York, 1981), in
which all the instances are noted. The debate between scholars is called
a mahaloket, 'division', 'controversy', e.g. the debates between the House
of Shammai and the House of Hillel are described in the Mishnah (Avot
5: 7) as: 'controversy for the sake of Heaven', mahaloket le-shem shamayyim.
The whole subject of the mahaloket is treated by B. De Friess, Mehkarim
be-Sifrut ha-Talmud (Jerusalem, 1968), pp. 172-8. The locus classicus for the
problem is Tosefta Sanhedrin 7: 1 (ed. Zuckermandel, p. 425). This reads:
' R. Jose said: "At first there was no controversy (mahaloket) in Israel except
in the Court of the Seventy in the Chamber of Hewn Stones . . . When the
disciples of Hillel and Shammai who had not served (their masters)
sufficiently increased, controversy increased and there were two Torot in
Israel.' " In the Babylonian Talmud (Sanhedrin 88b) this is quoted as:
' R. Jose said: " At first they did not increase controversy in Israel. . . When
the disciples of Shammai and Hillel, who had not served sufficiently,
increased, controversy increased and Torah became as if it were two
Torot. "' Thus, according to the reading in the Tosefta, controversy
between the Sages is said to have been completely unknown before the rise
of the disciples of Shammai and Hillel, whereas according to the reading
in the Babylonian Talmud it is said that controversy was known before
this period, but not to any large extent. De Friess discusses how far R. Jose's
view is historical. In any event, the term mahaloket became in our literature
the normal one for the controversy or debate with its Aramaic equivalent,
pelugta. When, for instance, the Talmud states, as it does very frequently,
that A and B disagree on this or that point, it is usually expressed as: be-mai
ka-mippalgi, 'on what point do they disagree?'
Two terms are, at times, ambiguous. These are: ve-ha-tanya, ' and
we have learnt in a Baraita* and peshita, ' i t is obvious' . In the
majority of instances these are questions: ' But have we not l ear nt ?' ;
' Is it not obvious?' Occasionally, however, they are statements:
' We have l ear nt ' ; ' 11 is obvious t hat ' . Rashi usually helps the student
by pointing out when these terms are used as simple statements.
There are no punctuation marks in the Tal mud so that the reader
has to supply these by inflections of the voice. To obtain the best
results in detecting the various moves in a Tal mudi c sugya the
The Talmudic argument 13
Talmud has to be ' sung' , as it is in Yeshivot today with the
traditional 'Gemara niggun\ There can be no doubt that melody
was used in Talmud study from the earliest times and it seems
certain that the compilers themselves relied on melody as a means
of punctuation. There is even some evidence for a system of
cantillation with musical notes as in the Biblical books.
The various kinds of arguments found in the Babylonian Talmud
can be classified according to a number of formal types or patterns.
All the main moves will be found to belong to one or other of these
categories. The argument from authority consists of a proof or support
of the correctness of a theory by an appeal to an incontrovertible
source, i.e. Scripture or a Tannaitic source; a Mishnah or a Baraita.
Where the attempted proof is from Scripture the only way to refute
it is to interpret the relevant verse or verses differently. Where the
attempted proof is from a Tannaitic source two moves are open to
the contestant who wishes to engage in refutation. He can either
demonstrate that an alternative interpretation is possible, or,
perhaps, necessary, or he can adduce another Tannaitic source
which disagrees with the one quoted by his opponent and on which
he can now rely. This is based on the generally accepted view, at
least by the later Amoraim, that an Amora cannot disagree with
a Tanna unless he can find support for so doing in the opinions of
another Tanna.
The terms used to introduce a proof from authority are: minalan, ' How
do we know this?' (Berakhot 7a; Pesahim 7b; Kiddushin 3b and freq.); mena
hanney miley, 'How do we know these things?' (Hullin 24b and very freq.);
mina ha milta, 'How do we know that?' (Hullin 10b; 1 ia); mina amina lah?,
'How do I know this?' (Sanhedrin 61b); Mai ta
i
ama, ' What is the reason?
(Bava Batra 173b and very freq.); ta shema
i
, ' Come and hear' (Bava Kama
22a and very freq.); dikhetiv, 'For it is written in Scripture' (Sanhedrin 61b
and very freq.); ditenan, 'For we have learnt in a Mishnah', Yevamot 57a
and very freq.); de-tono rabbanan,' For our Rabbis have taught' (Hullin 24b
and very freq.) introducing a Baraita; de-tanya, 'For we have learnt in a
Baraita' (Bava Kama 30a and very freq.); she-neemar, 'For it is said in
Scripture' (Berakhot 7b and very freq.); ve-ha-tenan, 'But we have learnt
in a Mishnah' (Bava Batra 30a and very freq.); qf anan nami teninan, 'We
have also been taught this' (Berakhot 27a and freq.); tenituha, ' We have
learnt it in a Mishnah' (Bava Kama 22a and freq.); amar kera, 'Scripture
said' (Kiddushin 3b and freq.); amar rahamana ' The All-Merciful said', used
for a Scriptural proof (Bava Metzi'a 3b and very freq.); de-amar / ?. . . , ' For
Rabbi . . . said' (Pesahim 7b and very freq.). This last term is sometimes used
even for a proof from an Amora, provided the Amora is an established
authority.
14 The Talmudic argument
The argument by comparison is the deduction of a rule, not stated
explicitly, from an accepted teaching to which it bears a strong
resemblance. The refutation of this consists in demonstrating that
although the two cases do appear to be analogous they are, in fact,
different. This can be termed an argument by differentiation. The
either /or argument seeks to demonstrate that whichever one of two
possible interpretations of a given premiss is adopted, it will lead
to the desired conclusion. The on the contrary argument seeks to
demonstrate that, far from the premiss yielding the suggested
conclusion, it yields the exactly opposite conclusion. The acceptance
of an argument in part seeks to demonstrate that a particular
conclusion will follow from one construction of the phenomenon
under consideration but not from a different construction. The
argument based on an opponent's position seeks to demonstrate
that even if the suggested premiss is true, which, in fact, it is not,
the suggested conclusion does not follow from it. The argument
exposing the flaws in an opponent's argument is a suggestion to the
opponent that if he will only examine carefully the steps in his
argument he will see for himself that his case is faulty. An argument
is, at times, put forward only to be rejected.
The following are the terms used: (a) Argument by comparison: hainu, ' that
is', i.e. 'this is the same as that' (Pesahim 9b10a and freq.) ;yalfinan, 'we
learn' (this from that) (Tevamot 57 and freq.); shema'minah,' hear from this',
i.e. compare that to this (Pesahim 5b and very freq.). (b) Argument by
differentiation: shani hatam, 'there it is different' (Kiddushin 50a and very
freq.); ve-R., ' and what will R. . . say in reply?' (Kiddushin 51a and freq.);
mi damey, ' are the two cases alike?' (Pesahim 14b and freq.); hakhi hashta,
lit. 'how now', i.e. 'what is the comparison?' (Hagigah 13b and freq.). (c)
Either/or argument: mi-mah nafshakh, 'whichever way you see it' (Hullin 29a
and freq.). (d) On the contrary argument: ipkha mistabbera,' it is more plausible
to see it in the opposite way' (Pesahim 28a and freq.); aderabbah, ' on the
contrary' (Pesahim 28a and freq.). Sometimes, as in Pesahim here, the two
terms are combined: aderabbah ipkhah mistabbera; sometimes they are used
separately. In J. S. Roth's edition of Samson of Chinon's Sefer Keritut, p.
434, note 2, there is a fairly comprehensive list of the instances of aderabbah
in the Babylonian Talmud, (e) Acceptance of an argument in part: bishel-
ama ... ela...,' This is correct according to . . . but . . . ' (Pesahim 7a and very
freq.); hanihah, 'this is appropriate' (Bava Kama 12a and very freq.). (f)
Argument based on an opponent'sposition: u-le-ta ' amekh,' and according to your
reasoning' (Berakhot 43a and very freq.); li-devarav de-R...', 'according to
the opinion of R. . . ' (Kiddushin 51a and freq.). (g) Argument exposing a flaw
in an opponent's argument: ve-tisbera, ' and even according to your theory'
(Bava Kama 32a and freq.); mi sabbarit, ' do you hold' (Berakhot 27a and
f r eq. ) . (h) Argument put forward only to be rejected: mahu de-tema^ ' i t mi g h t h a v e
The Talmudic argument . 15
been said' ('Arakhin 21b and freq.); saleka da'atekh amina, ' I might have
argued' (Kiddushin 34b; Sotah 44a and very freq.); ka-saleka da'atekh, 'you
might have supposed' (Pesahim 26a and very freq.).
In addition to arguments based on pure reason there are to be
found arguments based on the facts of the case or the interpretation
of the facts. An example of this is the argument based on historical or
geographical conditions, in which an attempt is made to demonstrate
that these conditions affect the law and limit its application.
Another example is the argument based on the analysis of states of mind,
in which the law is said to depend on how human beings normally
react psychologically in a given situation.
The terms are: (a) Argument based on historical or geographical conditions: bimey
R. nishnet..., ' this was (only) taught in the days of R. . . ' (e.g. Bava Kama
94b); ha Ian ve-ha le-hu, 'this is according to us (the Babylonians) and that
according to them (the Palestinians)' (Berakhot 5b; Kiddushin 29b and
freq.). (b) Argument based on analysis of states of mind: hazakah, ' i t is an
established fact that' (Bava Metzi'a 3a; Bava Batra 6b and freq.). For a full
treatment of this argument see ET, vol. xm, s.v. hazakahj, pp. 693-713.
At times there occurs the readmission of an argument that has been pre-
viously rejected. An argument that has been rejected in favour of what
seemed to be a more convincing argument is now reinstated as
offering, after all, the best solution of the difficulty, the reason for
the original rejection then being shown to be unsound. The argument
against a statement of the obvious is presented whenever a statement
is made that appears to be quite superfluous since no one would
have thought otherwise. The defence is to demonstrate that what
seemed so obvious is not so at all. Reasons are given why it might
have been thought otherwise and the statement requires, therefore,
to be stated.
The terms are: (a) Readmission of an argument that has been previously rejected:
le-olam..., 'in reality', 'actually' (Berakhot 3a-b and very freq.); ela
mehavrata ke..., ' but it is better to say' (Pesahim 55b and freq.). (b)
Argument against a statement of the obvious: mai ka-mashma' Ian, ' what does he
tell us?' (Shabbat 108a and freq.); peshita, 'is it not obvious?' (Bava Batra
137a and freq.).
The argument to resolve a contradiction between sources occurs where
two Scriptural verses or two Tannaitic sources appear to contradict
one another. Where the contradiction appears to be between two
Scriptural verses the only way open is to show that, rightly
understood, there is no contradiction, that it is only apparent not
real. Where two Tannaitic sources appear to be in contradiction
16 The Talmudic argument
this solution is open but here, on occasion, the argument may
proceed to demonstrate that there is, indeed, a contradiction and
we must conclude that there is a debate on the matter between
Tannaim. The attempt is then made to identify the Tannaim
involved by referring to other Tannaitic sources. The argument by
textual emendation seeks to demonstrate that the text of a Mishnah
or Baraita cannot possible be accepted as it stands, that it is
obviously corrupt. The correct text is then established by means
of emendation. The argument from the principle of literary economy
proceeds on the assumption that the earlier, classical texts have
been so carefully worded that any apparently superfluous statement
is not coincidental or due to mere literary style but is contrived and
the text is then examined in order to discover what the apparently
superfluous statement intends to teach.
The terms are: (a) Argument to resolve a contradiction between sources: mar amar
hada u-mar amar hada ve-lofeligey, ' this master says one thing and the other
master another and they are not in disagreement' (Hullin 105a and freq.);
terey tannai, 'there are two Tannaim' (i.e. who disagree on the matter
(Berakhot 3a and freq.). (b) Argument by textual emendation: hesurey mehasara
ve-hakhi ka-teni, 'something is missing and this is how it should read'
(Pesahim 10b; Bava Kama 16a and very freq.). The standard methodologies
differ as to whether the intention in such instances is really to emend the
text or is simply a way of explaining the text, i.e. the text does not really
have to be read differently but this is what it means. For the first view see
Rashi to Berakhot 1 ib, s.v. afli-gemara; Rashi to Megillah 28b, s.v. hey tzana
and the other sources quoted by M. Higger, Otzar ha-Baraitot, vol. x (New
York, 1948), pp. 130-1. For the opposite view see Isaiah Horowitz: Shelah,
Torah she-be-
1
al Peh s.v. be-khamah mekomot. Other terms for the same device
are: meshabeshta hi,' the text is erroneous' (Gittin 73a); al tinney,' do not learn
thus' (Sotah 49b); teni, 'learn it thus' (Bava Kama 4a-b and freq.). (c)
Argument from the principle of literary economy: tzerikhey, 'both are necessary',
generally after lama li le-mitney, 'why do I have to state?' (Gittin 8a and
very freq.).
Frequently in the Talmudic debate a statement is presented in
more than one form. The different versions of an argument are due to
the difficulties in transmitting accurately reports of what the earlier
teachers actually said. Similarly, there is the argument presented by
different teachers where the statement itself is not in doubt but the
doubt is about who made the statement. In both these instances
there is generally an attempt to demonstrate the consequences of
different arguments, i.e. the practical differences which result from
looking at the matter in one way rather than another. The
The Talmudic argument 17
Talmudic debate also frequently calls attention to the limited
application of an argument. The suggestion here is that the argument
is sound so far as it goes but when examined turns out to be limited
in scope.
The terms ar e: (a) Different versions of an argument: ika de-amrey, ' others say'
(Hullin 3b and very freq.). (b) Argument presented by different teachers'.
' R. . . says . . . and R. . . says' (Berakhot 3b-4a and very freq.); ve-ibbayit ema,
' and if you want I can say' (Berakhot 3b and very freq.). (c) The consequences
of different arguments: mai beynayhu... ika beynayhu...,' what is the difference
between them? The difference between them i s . . . ' (Kiddushin 50a; Bava
Kama 23a; Bava Batra 174b and freq.). (d) Limited application of an argument:
haney miley, 'when are these words applied?' (Berakhot 15a and very freq.);
lo nitzrekha, 'it is not necessary (to state it except in the following instance)'
(Bava Metzi'a 30b and very freq.); lo amaran ela, 'we do not state it
except...' (Berakhot 8b and freq.).
A large portion of the Talmudic debate is taken up with the
posing of purely academic problems. These are either set by
individual Amoraim or anonymously and the aim of the exercise
is to discuss theoretically the principles upon which the law is based.
As we have noted in connection with the term ba'ya (see p. 11),
in this type of problem the two halves are so equally balanced that
no reason exists for favouring one over the other. There are more
than a thousand of these problems scattered through the Talmud.
It is highly probable that such contrived problems were set
consciously as an intellectual exercise, especial skill being required
to see that the two halves were, in fact, equally balanced.
M. Guttman,' Sheelot Akademiot ba- Talmud''in Dvir, 1 (Berlin, 1923), 38-87;
2 (Berlin, 1924), 101-64, has assembled all the material in the Talmud
on the purely academic question - the ba'ya - which he compares to
mathematical puzzles or philosophical conundrums such as Zeno's problem
of Achilles and the tortoise.
It has to be said that there is no actual classification of the different
types of argument in the Talmud itself and the names for them are
our invention. Nevertheless a close study of the Talmud reveals that
the patterns we have noted are there. Certain 'ploys' are always
used as the occasion demands. These are strictly limited but the
richness and variety of the Talmudic debate are nevertheless
preserved, because these depend not on the number of the moves
available but on the ingenuity of the protagonists in making the
right move at the right time.
2
The literary form of the Babylonian Talmud
How was the Babylonian Talmud compiled and by whom? These
are among the most intractable problems in Jewish literature.
This many-volumed work, bearing all the marks of a finished
literary product, replete with the names of Amoraim and with their
opinions and debates, from the beginning of the third to the end
of the fifth century, remains completely silent on the questions every
student feels bound to ask: Who recorded all these opinions? Is it
a literary work at all or was it originally a verbal compilation,
committed to writing at a later date? Is it correct to speak of
'editors' of the Talmud, or did the work simply grow by stages?
What are we to make of the mediaeval tradition that R. Ashi and
Ravina were the compilers of the Talmud, in view of the immense
portion of the work that must have been added, at least, after these
teachers? Assuming that there were editors, what principles guided
them in their selection of the material and how did they shape it?
Is there evidence of different hands shaping the material in
different tractates or even in the same tractate? Why is there no
Talmud to many of the tractates of the Mishnah? Mediaeval and
modern scholars down to the present day have tried to supply
answers to these and similar questions, but the basic problem
remains as stubborn as ever.
The basic mediaeval text around which much of the question has centred
is the famous letter of Sherira Gaon which dates from the tenth century,
i.e. some five centuries after the ' close of the Talmud'. Moreover this letter
is in two recensions, a Spanish and a French, which contradict one another
on some of the most important issues, e.g. when the Talmud was
committed to writing. The best edition of the Iggeret De-Rav Sherira Gaon
is that of B. M. Lewin (Haifa, io,2i),J. N. Epstein's Prolegomena Ad Litteras
Amoraiticas (Mevuot), ed. E. Z. Melamed Jerusalem, 1972), has a full
18
The literary form of the Babylonian Talmud 19
discussion on the implications of the letter (Appendix, pp. 610-15), and
this work as a whole deals with our problem. A very full treatment of the
letter is in M. A. Tenenblatt's The Formation of the Babylonian Talmud (in
Hebrew, Tel-Aviv, 1972), chapter 15, pp. 276-94. Tenenblatt's work is
an admirable summary of all the views on our question, relying heavily
on the very fine study by Julius Kaplan, The Redaction of the Babylonian
Talmud (New York, 1933). Tenenblatt refers to the work of S. M. Rubin-
stein, le-Heker Siddur ha-Talmud (Kovno, 1932), a work which I have not
been able to consult but which is extensively quoted by Tenenblatt
(chapter 5, pp. 81-111). Kaplan's very thorough work opens with an
analysis and critique of the view of the earlier scholars: Graetz, Frankel,
Rapoport, Briill, I. H. Weiss, Halevy and W. Jawitz. Other works on the
subject are those of Hyman Klein (see works cited in the bibliography);
C. Albeck, Introduction to the Talmud (in Hebrew, Tel-Aviv, 1969); David
Halivni, Sources and Traditions (in Hebrew, Tel-Aviv, 1968); B. M. Lewin,
Rabbanan Savor ai ve-Talmudan in Azkarah (in memory of Rabbi A. I. Kook),
vol. iv (Jerusalem, 1937), pp. 145-208; Abraham Weiss, Hithavutha-Talmud
Bishlemuto (New York, 1943) and le-Heker ha-Talmud (New York, 1954);
Louis Jacobs: Studies in Talmudic Logic and Methodology (London, 1961);
Jacob Neusner, ed., The Formation of the Babylonian Talmud: Studies in the
Achievements of Late Nineteenth and Twentieth Century Historical and Literary-
Critical Research (Leiden, 1970); Shamma Friedman, Perek ha-Ishah Rabbah
be-Bavli (Jerusalem, 1978).
The three periods relevant to the problem are: (1) the Amoraic;
(2) that of the Saboraim (the meaning of which term is itself very
problematical); (3) the Geonic. While there are very few Geonic
additions to the Talmud the degree of Saboraic participation is
hotly debated. Some scholars hold that the Talmud in its present
form is the work of the Amoraim, the Saboraim only adding
occasional notes and the like. Others look upon the Saboraim as
the real editors of the Talmud. Others again see a lengthy process
of editorial work, beginning with the Amoraim and completed by
the Saboraim. Abraham Weiss' is a lone voice. Weiss denies that
the Talmud was ever 'edited' at all. In his view there are strata
of Talmudic material, the earliest of which belong to the early
Amoraic period, one being added to the other until, eventually, the
Talmud emerged in the form in which we have it now.
Hardly any of the scholars who have discussed the question, with
great subtlety and profound understanding, and whose work is
indispensable, have noted adequately the feature to which this book
repeatedly calls attention - the literary structure of the Talmudic
sugya. It is hoped that, after a careful study of the material as
presented here, the reader will become convinced not only that the
20 The Talmudic argument
Talmud was put together by a series of anonymous editors, in the
post-Amoraic period, but that these men were, in fact, far more
than mere editors or compilers. They were literary artists of a high
order who, undoubtedly, used earlier material and actual
statements of the Amoraim, but who worked this material into
shape. There has been hardly any detailed literary analysis of the
Talmudic sugya as a whole. When such analysis is undertaken, it
becomes abundantly clear that the sugya is presented in such a way
that argument leads on to further argument, in neat and logical
sequence. We have referred in this connection repeatedly to the
'final editors' but, in reality, the compilers were creative artists,
reshaping all the earlier material to produce a literary work.
Our thesis does not seek to deny that there are earlier strata to
be detected in the sugya. It is obvious that the final editors did use
much earlier material and there is a considerable body of evidence,
marshalled by J. N. Epstein and Abraham Weiss in particular, for
the existence of such strata. Our thesis calls attention, however, to
the re-working of all this material, in the process of which some of
the material has been left in its original form. Nor can we deny
that there are later additions, i.e. made after the bulk of the work
had been completed. Sherira Gaon and other mediaeval writers
report a tradition that, for example, the first sugya in tractate
Kiddushin is Saboraic and the different style of this passage supports
the tradition. Abraham Weiss has succeeded in detecting other
opening passages of tractates that are late and here again, though
Weiss does not stress this, the style is revealingly different. Our
concern is with what the mediaeval scholars referred to as setama
de-gemara, ' the anonymous Gemara', i.e. the framework of the
sugyot. This anonymous material, comprising the bulk of the
Talmud, is, we maintain, not merely a framework provided for the
earlier material but embraces that material as well, which has been
re-shaped. Shakespeare, for example, used earlier chronicles and
the like, but was far less interested in conveying these verbatim and
with regard to historical truth than in using them as bricks with
which to construct his completely literary product. We maintain
that the Babylonian Talmud is a work of this order.
If we are correct the following three positions will have to be
maintained. First, the Talmud is a literary work, contrary to the
opinion of some mediaeval and some modern scholars that it was
not at first written down. It is impossible for a literary work of this
The literary form of the Babylonian Talmud 21
nature, in which there are such things as literary device and the
working up of the material to a carefully calculated climax, to be
carried by successive generations only in the mind and expressed
by word of mouth. True, even today, there are those who do know
the whole Talmud by heart, but that is because they know the
completed work. The shaping of the material in this way can only
have been done, originally, in writing.
Secondly, this framework is, as we have already suggested, far
more than a mere framework. It is the Talmud, in a very real
sense. It is difficult, perhaps impossible, at this late hour, to know
how much of the material put into the mouth of the Amoraim is
authentic, how much their own reported words; but it is clear from
our analysis that some, at the very least, of' their' words have been
put into their mouths so as to promote the on-going argument
of the sugya. Some of it, at least, is 'fictitious', contrary to
Abraham Weiss, who disagrees with the Tosafists who pointed out
long ago the existence of fictitious sugyot.
Thirdly, since the framework is uniform in style and literary form
and embraces the teachings of even the latest of the Amoraim, it
follows that the final editors were post-Amoraic; 'Saboraic', in
fact, though this, after all, is only a convenient term for the
remarkable body of men who, for whatever reason, preferred to
remain anonymous.
It would seem that something of this kind happened towards the
end of the fifth and the beginning of the sixth century, possibly even
later in the sixth century. A number of teachers felt themselves
obliged to collect the Amoraic material and to shape it so that a
coherent work on the Mishnah would result. Although, no doubt,
their concern was also with practical law - there are many instances
of case law in the Babylonian Talmud - their main aim was purely
academic. They evidently wished to provide argument and debate
as an intellectual exercise, precisely because, as stated in the
previous chapter, the study of the Torah was by this time the
established and acknowledged supreme way of worship. From this
point of view the nature of the raw material they used was
irrelevant. The only condition necessary for its inclusion was that
it should have to do with one or other aspect of Jewish teaching.
In all probability there were a number of centres in which these
men flourished. An indication of this is the different style and
language of the tractates Nedarim, JVazir, Temurah and MeHlah, as
22 The Talmudic argument
has frequently been noted. But even here, and for all the differences
in style, basically the method was the same. A comparison of the
Babylonian Talmud with the Palestinian demonstrates that the
flavour of the former is all its own: the use of keen questioning and
reply, of the thrust and parry of debate, of the liveliness of the
arguments, especially of the stock framework into which lengthy
discussions have been fitted. Had the writing of Introductions been
in vogue at that time we would, no doubt, have had an Introduction
in which the aim of the work was stated. But, as with the Biblical
books, the Talmud gives all the appearance of having dropped
down from Heaven complete, as it were. The very problem of how
and by whom the Talmud was compiled is itself the most eloquent
testimony to the teachers' skill. Instead of introducing us to their
own work they introduce us into the world of the Amoraim, in some
respects an imaginary world, but one in which the spirit of the
Amoraim actually lived on.
On the pseudepigraphic nature of a good deal of the Talmudic material,
ie. on the conscious attempt to use earlier teachers as figures in a
truth-conveyed-through-fiction narrative, cf. my ' How much of the
Babylonian Talmud is pseudepigraphic?', JJS, 28, 1 (Spring, 1977),
4659. From the evidence adduced there, it emerges that in a large number
of instances the editors clearly reveal their intention of writing this kind
of material and in an even larger number of instances the pseudepigraphic
nature can easily be detected. The majority of scholars who have dealt so
comprehensively with our problem have concentrated in the main on those
Talmudic passages which purport to give some information on the scope
of the work of the teachers, Amoraic for the most part, who feature as the
heroes of the work. But this is to miss the point that they are the ' heroes'
not the authors of the work. It seems to me that the closest analogy to the
genre of much of the Babylonian Talmud is that of an historical novel,
where the novelist may well draw on actual events of the past and even
use the recorded words of real historical characters and yet, for all that,
what happened in the past was not quite like that. The work of fiction tries
to succeed in bringing the past to life but in a way really larger than life.
Unless this is appreciated and if one persists, as some do, in seeing the
Babylonian Talmud as the ' Hansard' of the Amoraic debates, the whole
point is missed. One may even hazard a guess that some of the confusions
regarding Amoraic chronology are due to this peculiar nature of the
Talmudic material in which the argument counts for practically all, the
historical details for very little; though one should not be so sceptical as
to deny the usual, conventional reconstruction, in broad outline at least,
of the history of the Amoraic period. Cf. David Goodblatt's Rabbinic
Instruction in Sassanian Babylon (Leiden, 1975), for a very helpful reminder
of how precarious it is to reconstruct the actual conditions obtaining in
The literary form of the Babylonian Talmud 23
the Babylonian ' schools' on the basis of alleged ' traditions' first mentioned
hundreds of years after the events. The correct method for all inquiries of
this kind is to see what the sources themselves say, always bearing in mind
that in the Babylonian Talmud we have a very unusual type of literary
product.
3
Bererah: Retrospective specification
The following (from Gittin 25a26a) is one of the main Talmudic
passages in which the bererah theme is developed.
The term bererah is Amoraic. The root meaning of brr is ' to
choose'; here the meaning is: a choice, among possibilities, which
determines an act, the choice being made not when the act is
performed but subsequently. The legal problem is whether the later
' choice' validates the original act as if the choice had been made
at the time of the act. For this reason it is best to paraphrase bererah
as 'retrospective specification', i.e. the subsequent 'choice' or
specification operates retrospectively. In the Talmudic debate,
when an authority is said to accept the bererah principle, this is
formulated as: ' he holds bererah' (it leh bererah) and when an
authority is said to reject the bererah principle, this is formulated
as: ' he does not hold bererah' (let leh bererah).
Our sugya opens with a problem set by the third-century
Babylonian Amora, R. Hoshea, and presented to his teacher,
R. Judah. The rule is that a get ('bill of divorce') to be valid
requires to be written specifically for that husband and wife. If, for
instance, a man had two wives with the same name and the get has
been written for the purpose of divorcing wife A, that get is invalid
as a bill of divorce for wife B, even though the name is the same.
R. Hoshea's problem is as follows. The husband instructed the
scribe to write out the get 'for the wife who will first come out
through the door', i.e. the get is written for that particular wife, the
one who subsequently does emerge first, but at the time of writing
it is not known whether it is wife A or wife B who will in fact, emerge
first. Is such a get invalid, since at the time of writing it cannot be
said to have been written specifically for that wife, or do we rather
24
Retrospective specification 25
say that since it later became clear t hat wife A or B did, in fact,
emerge first this operates retrospectively, so that the g^ is considered
to have been written for her and is valid? In other words, is the
bererah principle accepted?
R. J udah replies that the solution to the problem is to be found
in the Mishnah (Gittin 3 : 1 , 24b, to which this passage is appended).
The Mishnah rules that if the husband instructs the scribe to write
the get 'for the wife I will decide to divorce' , the get is invalid since
at the time of writing the husband has not yet decided which of
the two he will divorce. This is, suggests R. J udah, exactly
analogous to R. Hoshea' s case. Just as in the case dealt with in the
Mishnah ('for the wife I will decide to divorce' ) the get is invalid
because, evidently, the bererah principle is rejected, so, too, in
R. Hoshea' s case ('for the wife who will first come out through the
door' ) the get is invalid because the bererah principle is rejected.
To this R. Hoshea objects that he has discovered another
Mishnah {Pesahim 8: 3) from which it appears t hat the bererah
principle is accepted. This Mishnah deals with the Paschal l amb.
Before the Paschal l amb is slaughtered those who are to eat of it
must be ' count ed' or ' number ed' for that l amb; i.e. a man cannot
fulfil his obligation to eat of the meat of the Paschal l amb unless
he has been allotted a share in that l amb at the time of its
slaughtering. Now the Mishnah rules: ' If a man says to his sons:
"Behold I will slaughter the Paschal l amb for whichever of you
arrives first in Jerusal em", the first one to enter with his head and
most of his body acquires thereby his portion and with it he
acquires on their behalf the portions of his brothers. ' From this
Mishnah it appears that, although at the time when the l amb was
slaughtered it was not known which of the sons would, in fact, be
the first to enter Jerusalem, the son who did reach Jerusalem first
acquires his portion in the l amb. Hence it is clear that the Mi shnah
does accept the bererah principle.
To this R. J udah retorts: ' Hoshea, my son! What has Pesahim
to do with Gittin?\ i.e. the two cases are not analogous. A com-
ment of the third-century Palestinian Amora is then quoted. R.
Johanan, commenting on the Pesahim Mishnah, explains it as ' i n
order to encourage the sons to be energetic in the performance of the
precepts' , i.e. the Mishnah does not deal, as R. Hoshea thought,
with the case of a father who had no t ' count ed' his sons before the
lamb had been slaughtered but rather with the case of a father who
26 The Talmudic argument
had, in fact, 'counted' all his sons for the lamb before it had been
slaughtered. The father only pretended to make his 'counting'
depend on the son who first entered Jerusalem and this pretence
was for the sake of encouraging his sons to be energetic in the
performance of the precepts. Thus the reason why the son who first
enters Jerusalem acquires his share is not at all because of the bererah
principle, as R. Hoshea imagined, but because the son had been
'counted' from the beginning. It is not at all a question of
retrospective specification. The specification, albeit unknown to the
sons, had been made by the father from the beginning, from the
time of the slaughtering when it was required to become operative.
The Talmud proceeds to demonstrate that this interpretation
must be correct. For the Mishnah rules that the other sons, too,
acquire their portion. But how can they acquire the lamb, if the
reason why the son who enters Jerusalem first acquires it is because
of the bererah principle? This principle can only operate in favour
of the son who enters first. No bererah principle can operate on behalf
of the other sons and we have learnt (Mishnah Pesahim 8:3, the
same Mishnah) that once the lamb has been slaughtered there can
be no further 'counting'. It must follow that the Mishnah deals
with the case of a father who had previously ' counted' all his sons
so that the bererah principle is entirely irrelevant to the issue. A
Baraita is then quoted in further support of the contention that the
issue is not one of bererah but of prior ' counting' and a pretence
of the father in order to encourage his children to be energetic. The
Baraita states: ' It once happened that daughters entered Jerusalem
before the sons so that the daughters were found to be energetic
and the sons lazy.' This Baraita supports the argument that the
whole purpose of the father's declaration is in order to encourage
his children to be energetic and is not for the purpose of having
them 'counted' i.e. since the 'counting' had already been done on
their behalf.
Thus far we have had the debate between R. Hoshea and
R. Judah and the discussion around this. The comments of the
Babylonian Amoraim Abbaye and Rava are now introduced.
These lived a century after R. Hoshea and R. Judah but were
evidently familiar with the earlier debate. From an analysis of this
debate it emerges that neither R. Hoshea nor R. Judah draws a
distinction which, according to Abbaye, can legitimately be drawn.
Three cases have so far been considered:
Retrospective specification 27
(1) R. Hoshea' s: The wife who will first come out through the door
(2) The Gittin Mishnah: The wife I will decide to divorce
(3) The Pesahim Mishnah: The son who will first enter Jerusalem
Throughout the whole of the previous discussion it is assumed that
the three cases are analogous, but Abbaye perceives a clear
distinction between (1) and (3) on the one hand and (2) on the
other. In cases (1) and (3)the required specification depends on
another the wife who will first come out through the door in (1)
and the son will first enter Jerusalem in (3). But case (2) depends
on the husband's own decision. It is the husband himself who will
eventually specify which wife he wishes to divorce. Why should we
not say, argues Abbaye, that the bererah principle does operate in
cases (1) and (3) - where the man concerned makes the specification
depend on another (toleh be-da" at aherim) - but does not operate in
case (2) - where the man concerned makes the specification depend
on his own ultimate decision {toleh be-da' at' atzmo) ? If this distinction
is accepted all difficulties are removed. In case (2) the bererah
principle does not operate - hence the ruling of the Gittin Mishnah
that the get is invalid; whereas in cases (1) and (3) it does
operate - hence the ruling in the Pesahim Mishnah that the son who
first enters Jerusalem does acquire his portion in the Paschal lamb
and hence the solution of R. Hoshea's problem fails to be
contradicted by the Gittin Mishnah.
We are helped by Rashi, the great French commentator, to
appreciate the distinction drawn by Abbaye. It must be postulated
that the less uncertainty there is in the original specification, i.e.
at the time of the act, the more readily can the final specification
operate retrospectively. So much has been achieved, as it were, at
the beginning that the bererah principle requires no severe applic-
ation for the final specification to be achieved. Conversely, the
greater the degree of uncertainty there is at the beginning the less
readily can the bererah principle be put to work so as to achieve
retrospective specification. So much has already been specified in
the first instance that even a weak specification such as that
provided by bererah will suffice, whereas so little has been achieved
in terms of specification in the second instance that the bererah
principle is called upon to do practically all the work, and this it
cannot do. Now where, as in cases (1) and (3), the final specification
is made to depend on another this very fact means that the man
concerned has, at the time of his act, virtually made up his mind,
28 The Talmudic argument
leaving only the final clarification to another. The man who states
his intention of divorcing whichever wife will first come out through
the door has decided definitely to divorce that wife; the lack of
clarity in the specification being due solely to ignorance of which
wife will eventually emerge first. Similarly, the father has already
decided quite definitely that the portion of the Paschal lamb is to
be acquired by the son who first enters Jerusalem. In case (2), on
the other hand, the husband is himself uncertain whether he will
divorce wife A or wife B. In Rashi's words: ' he halts between two
opinions'. Consequently, at the time of the act, the writing of the
get, the degree of specification is exceedingly weak. In the famous
words of Sam Goldwyn, it is only ' a definite maybe'. Here all
depends on the final specification operating retrospectively and it
may well be that even though in instances (1) and (3) the bererah
principle operates here it does not.
Rava, however, defending R. Hoshea and R. Judah, refuses to
draw Abbaye's distinction. According to Rava the issue is simple:
does the bererah principle operate or does it not? No distinction is
to be made regarding bererah, so that it is accepted as operative in
some instances and not in others. As our sugya puts it: ' Rava said:
What is the difficulty? Possibly, whoever accepts the bererah
principle does so whether it depends on his own decision or on the
decision of others, whereas whoever does not accept the bererah
principle rejects that principle whether it depends on his own
decision or on the decision of others.'
In the next stage of the sugya R. Mersharsheya addresses Rava,
seeking to prove to him that Abbaye's distinction is well-founded
and that he, Rava, is ill-advised to refuse to make the distinction.
This R. Mesharsheya seeks to do by quoting two opinions of the
second-century Tanna, R. Judah (not, of course, to be confused
with the Babylonian teacher of R. Hoshea mentioned at the
beginning of the sugya). An examination of R. Judah' s views as
recorded (a) in a Baraita and (b) in a Mishnah (Gittin 7: 4) seems
to show that in the case recorded in the Baraita he does not accept
the bererah principle, whereas in the case recorded in the Mishnah
he accepts it. Thus R. Judah seems to contradict himself. But the
case in the Baraita is one in which the final specification depends
on the person himself, whereas in the case of the Mishnah it depends
on others. This shows the correctness of Abbaye's distinction and
refutes Rava.
Retrospective specification 29
First the Baraita is quoted. The case recorded here is of one who
buys wine from Samaritans who are suspected of failing to tithe
their wine. The buyer is obliged to separate the tithes before being
allowed to drink the wine. Supposing he wishes to drink from the
j ar of wine he has bought and has no vessels to hand into which
he can pour the separated tithes. What remedy is there for him,
to allow him to drink the wine? Is he allowed to declare that the
wine he will later separate as the tithes, is now separated as tithe
wine, to be removed later from the j ar, even though the actual
separation will not have taken place until later? In other words,
does the bererah principle operate so that the subsequent separation
of the tithe wine operates retrospectively and the wine he drinks
from the j ar before the actual separation is wine from which the
tithes have been separated? The Baraita states that R. Meir permits
it but R. J udah, R. Jose and R. Simeon forbid it. Thus it is clear
that R. J udah does not accept here the bererah principle and here
the subsequent specification depends not on others but on himself;
it is the buyer himself who will be responsible for the later
specification when he eventually separates the tithes by pouring
them out into his vessels. It is the buyer himself who decides
ultimately which part of the wine he will separate as his tithes.
In the Mishnah in question there is a ruling of R. J udah
regarding a law recorded in the previous Mishnah (Gittin 7: 3).
The case considered in this previous Mishnah is that of a sick man
who delivers a get to his wife (in order to release her from the levi-
rate bond - Deuteronomy 25: 5 - 1 0 - i n the event of his deat h),
stipulating that the get will only be valid if he dies from the illness
from which he is at present suffering, but that in the event of his
death the get will be valid retrospectively, i.e. from the time of its
delivery (as it must be if the wife is to avoid the levirate bond). Now
R. J udah in the following Mishnah discusses what the status of the
wife will be during the period of the illness and rules t hat until the
man actually dies she has the full status of the married woman. But
R. J udah, by discussing the status of the wife during the period of
the husband' s illness, clearly implies his agreement with the
previous ruling of the Mishnah that when the husband does
eventually die from that illness the get is valid retrospectively. Now
for the get to be retrospectively valid the principle of bererah must
be invoked and this shows that R. J udah does accept the bererah
principle. How, then, can we square this with his opinion as
30 The Talmudic argument
recorded in the Baraita? We must perforce accept Abbaye's dis-
tinction. In the Mishnah the later specification depends on ' others'
(i.e., as Rashi comments, on God in whose hand is life and death).
Thus we see that there is an authority, namely, R. Judah, who
accepts the bererah principle when the subsequent specification
depends on others, but who rejects that principle when the
subsequent specification depends on the person himself.
R. Mesharsheya (so in current editions, variant reading, as in
marginal note, Ravina) proceeds further to refute Rava and to
support Abbaye's distinction. In the Samaritan Baraita it is stated
that R. Simeon, too, as well as R. Judah, forbids the drinking of the
wine and so evidently rejects the bererah principle. But another
Baraita records the following case. A man cohabits with a woman
(the act of cohabitation can effect a valid marriage if performed for
that purpose) declaring that the act is for the purpose of effecting a
valid marriage, but only if his father will subsequently approve of
the match. R. Simeon here rules that if subsequently the father is
pleased with the match the marriage is retrospectively valid. Thus,
although R. Simeon rejects the bererah principle in the case of the
wine, here he accepts it. The solution must surely be on the basis
of Abbaye's distinction. In the case of the wine the subsequent
specification depends on the man himself and here R. Simeon
refuses to accept the bererah principle. But in the case of the
marriage the subsequent specification depends on ' others' , i.e. on
the father and here R. Simeon does accept the bererah principle.
Rava has now been forced into a corner, but manages to
extricate himself. In reply to R. Mesharsheya, Rava stoutly persists
in his refusal to draw the distinction suggested by Abbaye. No such
distinction, argues Rava, is made by any authority. Those who
reject the bererah principle do so even when the subsequent
specification depends on others. Those who accept the bererah
principle do so even when the subsequent specification depends on
the person involved himself. In that case why do R. Judah and
R. Simeon forbid the drinking of the wine and so evidently refuse
to accept the bererah principle? To this Rava replies that the reason
why R. Judah and R. Simeon prohibit the wine is not because the
tithing is invalid. It is perfectly valid because the bererah principle
is accepted. The reason why they prohibit the wine is out of fear
that the jar may split before the buyer returns home to separate
the tithes. If that were to happen no tithes would have been given,
Retrospective specification 31
the wine having been lost, and this would mean that he had
imbibed untithed wine. Rava suports this interpretation by quoting
the conclusion of the Baraita in which R. J udah and R. Simeon, in
fact, ask R. Meir why he is not apprehensive that the j ar will split
and R. Meir retorts that he has no such apprehensions. Thus,
argues Rava, all three authorities, R. J udah, R. Simeon and
R. Meir, accept the bererah principle, the sole mat t er on which they
disagree being whether the buyer is allowed to take the risk of the
j ar splitting before he can pour out the tithe wine.
The commentators are puzzled by the implications of our passage, that
the validity of the get in R. Judah' s case and of the marriage in
R. Simeon's depend on the bererah principle. Surely all authorities accept
the validity of a conditional declaration. The rule of condition (tennai) is
treated everywhere in the Talmudic literature with unqualified acceptance.
Nowhere else is there the slightest suggestion that tennai depends on bererah,
so that those authorities who reject bererah reject tennai. And yet here the
proof that R. Judah and R. Simeon reject bererah is from cases that deal
with tennai. In order to cope with this difficulty, Rashi draws a distinction
between a condition the fulfilment of which depends solely on the person
who makes it (e.g. 'Be betrothed to me on the condition that I give you
200 zuz') and a condition the fulfilment of which is beyond his control
('This is your get if I die from this illness' and 'Be betrothed to me if my
father will be pleased with the match' ). In the former case the act is
complete from the beginning, since the intention to satisfy the requirements
of the condition and the fulfilment of the condition depend entirely on
the person who makes the condition. This does not depend at all on the
principle of bererah. The condition is here treated as something external
to the act so that the validity of the act depends on no retrospective
specification. But in those cases where the fulfilment of the condition
depends on others the act itself is indeterminate until the condition has
been fulfilled and this does depend on bererah, although, as a specification
which depends on others, it can possibly operate more readily than when
bererah depends solely on the person's own decision. Thus, as Rashi sees it,
there are three cases of doubtful validity in descending order:
(1) A condition the man can fulfil himself ('on condition that I give
you 200 zuz')- This does not depend on bererah and is valid, if the
condition is fulfilled, even according to those authorities who reject
the bererah principle.
(2) A condition the man cannot fulfil himself ('on condition that I die
of this illness' and ' on condition that my father is pleased with the
match' ). This does depend on bererah, but here the principle can
be the more readily employed precisely because all the indefiniteness
is due to the decision of others, not to the indecisiveness of the
person himself.
(3) ' Let the get be written for whichever wife I shall decide to divorce.'
32 The Talmudic argument
This is not a condition. The act itself is indecisive and so depends
on bererah. It is, moreover, a form of bererah less easily applied since
all the uncertainty is in the mind of the man himself. Other
commentators (see e.g. Meiri) offer different interpretations, but
Rashi's seems entirely satisfactory.
In the form we now have it this sugya is a complete unit, evidently
built up by the final editors out of earlier material extending from
the second century in Palestine (Mishnah and Baraita) down to
the second half of the fourth century (R. Mesharsheya) in Babylon,
In schematic form the sugya can be thus presented:
1i) R. Hoshea's problem: get written for wife who first emerges
(2) R. Judah' s reply: Mishnah: 'For wife I shall decide to divorce'
(3) * From which we see that bererah is not accepted' (this is in Aramaic)
(4) R. Hoshea's objection from Mishnah Pesahim
(5) R. Judah' s reply: What has Pesahim to do with Gittin
(6) Elaboration: R. Johanan: to encourage them to be energetic
(7) Proof that this is correct from latter part of the same Mishnah (this
is in Aramaic)
(8) Further support from the Baraita: daughters energetic (introduced
in Aramaic)
(9) Abbaye's objection: distinction between toleh be-da'at 'atzmo and
toleh be-da' at aherim
(10) Rava: possibly no distinction should be made
(11) R. Mesharsheya's objection to Rava from R. Judah in Mishnah
and Baraita
(12) R. Mesharsheya's second objection to Rava from R. Simeon
(13) Rava's reply: not because of bererah but fear that jar may split
It can be seen from the above that the framework of the sugya
has been provided by the final editors, who shaped the earlier
material in order to provide a carefully constructed progression of
thought, working it all up into a complete unit in which argument
follows on argument in the right order. No doubt the debate
between R. Hoshea and R. Judah is authentic but (3), (7) and (8),
and probably (6), do seem to have been added by the final editors
as part of their framework. There is no conclusive evidence that
the term bererah itself was used by R. Hoshea or R. Judah. Even
allowing, however, for editorial reworking of the arguments of
Abbaye and Rava, it would seem probable that in the original
debate between these two Amoraim the term bererah was actually
used. In any event the term bererah belongs to the Amoraic period
and is one of the many examples of Amoraic abstractions by means
of which the views of the Tannaim were summarised.
Retrospective specification 33
For bererah as an Amoraic formulation and as part of the general Amoraic
thrust towards abstraction see EJ, vol. II, p. 867. ET, vol. IV, p. 216
defines bererah as: ' Of something unspecified (she-eyno mevurar) now but
which becomes specified at a later date; whether or not we say that the
specification operates retrospectively.' Cf. pp. 216-46 for a comprehensive
account of all the legal discussions on the subject. Kese/JVivhar, no. 33, 1-37,
pp. 46a-54.b, lists all the Talmudic references to bererah as well as the
comments of the major post-Talmudic authorities. The main Talmudic
passages, in addition to the above, in which the bererah theme is considered
are: Gittin 25a (the sugya which precedes ours); 47a~48b;'Eruvin 36b~38a;
71b; 82a; Toma 55b~56b; Sukkah 24a; Betzah ioa-b; 37b-38a; 3ga-b; Sotah
18a; Bava Kama 51b; 6ga-b; Bava Batra 2 7b; Hullin 14a-b; 135b; Bekhorot
56b57a. A classic analysis of the bererah theme is given by Ran to Nedarim
45b.
4
Teush she-lo mi-da'at: unconscious
abandonment of property
This sugya, one of the best-known in the Talmud, is found in Bava
Metzfa 21 a22b. The basis of the sugya is a debate between the two
fourth-century Babylonian Amorafm, Abbaye and Rava.
Teush is the technical term for abandonment of property when
it has been stolen or lost without hope of recovery. The basic
meaning of yeush is despair. Although in some circumstances yeush
of stolen property gives title to the thief, this matter is dealt with
at length elsewhere in the Talmud. In our sugya the reference is to
yeush of lost property. The legal principle here is that if a man who
has lost some of his property despairs, i.e. gives up all hope of ever
recovering that property, the property is held to have become
ownerless and the finder may keep it. According to the rule as stated
in the Mishnah (Bava Metzi'a 2: 1 and further) the normal procedure
when one finds lost property is to have the find proclaimed in a
public place such as a synagogue. This serves as a kind of lost-
property office. There finders and losers repair, the loser stating any
distinguishing marks - siman, 'sign', 'means of identification'-
and if this tallies with the siman of the lost article the article is
restored to its owner. Thus, since there is a procedure for its
recovery, it is assumed that there has been no yeush where the lost
article has a siman. Why should the loser despair of recovering his
article since he knows of the procedure and stands a good chance of
identifying his property and having it returned to him? Yet if he,
none the less, did declare in the presence of witnesses that he had
abandoned all hope of ever recovering his property, then yeush
does come into operation and the finder, subsequent to the act of
yeush by the owner, may keep what he finds. Thus the general
rule is that finding is keeping if the article has no siman (because
34
Unconscious abandonment of property 35
it can then be assumed thatyeush has taken place), but finding is not
keeping if the article does have a siman (because here yeush cannot
be assumed) unless, as above, there is evidence of an explicit
declaration ofyeush.
The debate between Abbaye and Rava concerns yeush she-lo
mi-da'at, literally, ' unconscious abandonment of propert y' . Trans-
lated thus baldly into English the term makes no sense. How can
abandonment of property be otherwise than conscious? The
meaning of this technical term is, as Rashi explains it, ' the status
of a lost article, of which it can be assumed t hat when the owner
will become aware that he has lost it he will abandon it but of whose
loss the owner is unaware at the time when the article is found' .
The point here is thatyeiish only operates to give title to the finder
if it takes place before the finder takes possession of the article. If
there had been no yeush beforehand, the finder, when he takes the
article, is held to have become a bailee acting on behalf of the owner
since, in the absence of any prior yeush, the finder has no legal title
to the article. An article held in trust for its owner by a bailee is
legally held to be still in the owner' s possession so t hat yeush cannot
come into effect. Yeush only takes effect when an article has been
lost, not when it is still in the owner' s possession. If, for instance,
a man has an article in his house which he imagines to have been
lost, any yeush he may express will be ineffective and it would
obviously be wrong for another, hearing of the yeush declaration,
to enter the owner' s house and appropriate the article for himself.
The debate between Abbaye and Rava concerns, as we have noted,
an article which the owner would abandon when he becomes
aware that he has lost it but of whose loss the owner is unaware
at the time when the finder takes it. Abbaye argues t hat there has
been no actual, positive yeush before the finder took it and hence the
finder becomes a bailee and any subsequent yeush is inoperative.
Rava argues that, since if the owner had been aware of his loss he
would certainly have abandoned all hope of recovery, it is treated
as if there had heenyeush when he lost it. There is, for Rava, a kind
ofyeush state, even though there is no conscious awareness on the
part of the loser. The loser's lack of awareness of his loss is
irrelevant, according to Rava, since when he does become aware of
his loss he will certainly engage in yeush. Thus our sugya opens:
' Yeush she-lo mi-da'at: Abbaye said: It is notyeush but Rava said: It
is yeush.'
36 The Talmudic argument
After this bare statement of the opinions of Abbaye and Rava,
there is an anonymous Talmudic elaboration which seeks both to
explain the views held by the two Amoraim and to delineate the
boundaries of their debate. This addition reads: ' With regard to
an article that has a siman no one in the world [the usual technical
term for both protaganists, i.e. here Abbaye and Rava] disagrees
that it is not yeush. And even though we did eventually hear that
he engaged in yeush it is still not yeush. For when it came into his
hand [i.e. into the finder's possession], it came into his hand at a
time when it was prohibited to him, since when he [the loser]
becomes aware of his loss he does not engage in yeush, for he says:
" I have a siman so I will state the siman [at the ' lost-property office']
and take it." Where the article has been found after it has been
swept away by the sea tides or by the overflow of a river, the Torah
permits it even if it has a siman, as we shall have occasion to state
later on. When do they [Abbaye and Rava] disagree? It is with
regard to an article that has no siman. [Here] Abbaye said: It is
notyetlsh. For he [the loser] does not know that he has lost it. Rava
said: It is yeush. For as soon as he becomes aware that he has lost
it he engages in yeush, saying: "I have no siman" Hence it is from
now [from the time of his loss] that he engages in yeush.'
The meaning of this addition is that there are in all three distinct
cases. In one of these Rava agrees with Abbaye that there is no
yeush; in another Abbaye agrees with Rava that there is yeush; and
in the third there is a debate between Abbaye and Rava. The case
where both agree that there is noyeush, i.e. thatyeiish is inoperative,
is where the lost article has a siman. Such an article is not normally
abandoned (since it can be reclaimed). Hence even Rava will agree
ihatyeush she-lo mi-da'at will be inoperative and the finder may not
keep it even if the owner eventually did engage in yeush. The reason
is obvious. Here there is, in fact, no 'unconscious yeush' at all. When
yeush does eventually take place the article is already in the finder's
possession. He has acquired it illegally and is a bailee for whomyeiish
is inoperative. There is not the slightest reason for suggesting that
the loser's subsequent yeush should operate retrospectively.
This has nothing to do with the bererah principle, considered in the previous
chapter; see Radbad quoted by SM who states that the discerning will note
the difference between our case and bererah. The difference appears to be
this. In the case of bererah the act takes place now. It is only the exact
specification that is required to work retrospectively. For instance, in the case
Unconscious abandonment of property 37
of: 'Write a get for the wife who will first come out through the door', the
act, the writing of the get with that intention, takes place at once. All that
bererah has to do is to clarify the significance of the act. But here there was
no act of yeush at all at the time the owner lost his article. A subsequent
act of yeush is in no way connected with the status of the article at the time
it was lost and there is no reason whatever for holding that the subsequent
act ofyeush should operate retrospectively.
The case where both Abbaye and Rava agree that there is yeush
is where the sea or a river has swept away an article and that article
is subsequently deposited elsewhere by the sea or river. Here there
is an express permission of the Tor ah (to be described later on in
the sugya, hence ' as we shall have occasion to state later on' ) for
the finder to keep the article. The principle here would appear to
be that once the article has been swept away beyond human ken
it becomes automatically ownerless; it is as if it had vanished from
the owner' s possession. The finder may here keep the article even
if it has a siman, since the whole purpose of a siman is for the
identification and here there is no point in the identification, the
Tor ah having declared the article to be ownerless. Consequently,
if the owner is unaware that his article has been swept away, it is
still permitted and even Abbaye would agree. Abbaye only
demands that the owner be aware of his loss where the right of the
finder to keep the lost article depends on yeush. Here Abbaye
demands an actual yeush, not an ' unconscious' one. But the right
of the finder to keep an article swept away by the sea is not because
of the yeush principle but because the Tor ah has permitted it.
Hence the owner' s awareness of his loss is entirely irrelevant and
Abbaye will agree that the finder may keep the article. Thus there
are three cases where the owner is ignorant of his loss at the time:
(1) Where there is a siman - even Rava agrees - because there is no
yeush state
(2) Where the sea has swept it away - even Abbaye agrees - because
yeush is irrelevant
(3) Where there is no siman: Abbaye forbids, because there is no actual
yeush; Rava permits, because there is a.yeush state
The Tal mud now seeks to prove the case one way or another,
in favour of Rava or of Abbaye. First there is a series of attempted
proofs from the Mishnah to which this sugya is appended (Bava
Metzi'a 21: 1). Here a number of items are listed which, if lost, may
be kept by the finder, evidently because they have no siman, so that
38 The Talmudic argument
yeush can be assumed. The point behind this series of proofs is that
according to Abbaye's reasoning the finder is only allowed to keep
a lost article that has no siman if he knows for certain that the loser
has become aware of his loss before he, the finder, takes possession
of the article. If it is doubtful whether or not the loser has become
aware of his loss it is illegal for the finder to keep the article, since
there may not have been the positive j wi ^ Abbaye demands if the
finder is to keep the article. Now first in the list recorded in the
Mishnah are scattered sheaves (literally, 'scattered fruit
5
). These
have no siman; there are no means of identification. But presumably
these sheaves have fallen off the owner's cart and he will not become
aware of his loss until a considerable time has elapsed. Why, then,
according to Abbaye, is the finder allowed to keep them? The
Mishnah appears to support Rava and refute Abbaye. The reply
given on Abbaye's behalf is that the fourth-century Babylonian
Amora, Ukba bar Hama, has in any event previously (Bava Metzi'a
21 a) explained the Mishnah as referring not to sheaves that have
been lost but to sheaves left scattered at the threshing floor. The
Mishnah means, on Ukba bar Hama's interpretation, that if
sheaves are found stacked up at the threshing floor, it is proper
to assume that their owner will return to fetch them; but where the
sheaves are found scattered at the threshing floor it is safe to assume
that they have been intentionally abandoned there by the owner
after he had completed his threshing. The owner evidently attached
no significance to these scattered sheaves and will not bother to
gather them up. They are thus in the category of' articles that have
been lost intentionally' [avedah mi-da
1
at), i.e. it has nothing
whatsoever to do with theyeush question. The reason why the finder
may keep the sheaves is because the owner has left them there
intentionally for whoever wishes to take them. (Rashi used the
expression, 'they are hefkerV, hefker being 'ownerless property',
property over which the owner has intentionally relinquished his
ownership.)
The second proof is from another item in the Mishnaic list -
'scattered money'. Money has no siman (even if, for instance, coins
have a mark denoting that they belonged to a particular person,
this cannot count as a means of identification since money is not
kept by the owner but is used for purchasing goods). But according
to Abbaye, who requires positive yeush, the owner may not have
Unconscious abandonment of property 39
been aware that he has lost the money. The reply is given t hat the
Palestinian Amora, R. Isaac, has stated that it is in the nat ure
of men constantly to be examining their money-bags to see if these
are intact and it can therefore be assumed that the loser has become
aware of his loss before the money has come into the possession of
the finder.
The third attempted proof from the Mishnah is from two
further items in the list: ' cakes of figs and baker' s loaves' . These,
to be sure, have no siman but the owner is unlikely to have become
aware of his loss before the finder takes them, which supports Rava
and refutes Abbaye. To this the reply is given that it is safe to assume
that the loser does know of his loss almost at once because they are
' heavy' and he quickly senses the lessening of the load he carries.
The fourth attempted proof is from another item in the Mi shnah:
' strips of purple wool' . Surely here the owner does not become
aware of his loss until much later (i.e. since these are mere ' st ri ps'
and are not ' heavy' ) ? The reply is that these are valuable so t hat
a similar principle to that stated by R. Isaac applies. Just as men
constantly check their money-bags to see whether or not anything
is missing so do they check these valuable strips of wool. The owner
will have become aware of his loss almost immediately.
There is a difficulty with regard to this series of attempted proofs from the
Mishnah (see SM and the marginal note of the Vilna Gaon). The Mishnah
gives the following list of items the finder may keep: (1) scattered fruit; (2)
scattered money; (3) small ears of corn in the public domain; (4) cakes
of figs; (5) baker's loaves; (6) strings offish; (7) pieces of meat; (8) bundles
of wool shearings; (9) bundles of flax; (10) strings of purple wool. The
Talmud in our passage refers only to the (1), (2) and (4) and (5) treated
together, and then (10). Why are the others omitted? The Vilna Gaon
explains the omission of (3) on the grounds that, in any event, the Talmud
later on (Bava Metzi'a 22b) understands item (3) as referring to ears of corn
that have been left there intentionally. As for the omission of the other items
in the list, the Gaon refers to a variant reading adopted by the Rosh. If
our text is correct, however, the explanation of the omission would seem
to be that when the Talmud quotes (4) and (5) together, not as separate
items, as they are in the Mishnah, the meaning is: ' cakes of figs, baker's
loaves etc.
9
There is no equivalent in Talmudic language for ' et c' and it
is plausible to suggest that it is here implied. The progress of thought would
thus be: You have dealt adequately with ' scattered fruit' by saying that
it is 'fruit' left at the threshing floor, but what will you do with 'scattered
money'? The reply is given, because of R. Isaac's principle. But what of
cakes of figs, baker's loaves etc. (i.e. items (6) to (9)) where this does not
40 The Talmudic argument
apply? Reply: all these are heavy and their loss is known. But what of (io)
'strips of wool', which are not 'heavy'? Reply: they are valuable and so
R. Isaac's principle applies.
A further proof is now attempted from a Baraita. This reads: ' If
one finds coins ['monies'] in synagogues, houses of study or any
public place, they belong to him [the finder] because the original
owners give up all hope of recovery.' This seems to refute Abbaye,
since the owner may not have been aware of his loss. The reply is,
as above, that R. Isaac has said that men are in the habit of
constantly examining their money-bags and hence the loser will
have become aware of his loss.
The obvious difficulty here is that we have already been given R. Isaac's
principle in reply to the first objection regarding scattered money in the
Mishnah. Why repeat it in connection with the Baraita'? Moreover, from
the expressions used (' As R. Isaac said' in the Mishnaic instance;' R. Isaac
said' in the instance of the Baraita) it would appear that R. Isaac made
his original statement not on the Mishnah but on the Baraita, from which
it was later applied to the Mishnah. Why? (See Tosqfists s.v. ta shema\) It
should be noted that R. Isaac's principle is quoted, too, in Bava Kama 118b,
but there also as a quote. From all this it would appear (see Ritba quoted
by SM) that, indeed, once one knows of R. Isaac's principle one knows
that the same reply can be given when the attempted refutation is from
the Baraita, but that originally the questions were put at different times
and owe their present arrangement to the final editors {Ritba calls them:
R. Ashi and Ravina, the traditional editors of the Talmud) who presented
them in this order so as to work down, as it were, from the Mishnah to
a Baraita.
An attempt is now made to support Rava from another Mishnah
{Peak 8: i). Here the ruling is given that once the 'stragglers'
(namoshot) have been in a field, any gleanings left behind are
permitted to anyone who cares to appropriate them even if he is
not a poor man. The gleanings, the sheaves which fall during the
reaping, belong to the poor by the law of the Torah (Leviticus 23:
22) but once the 'stragglers' have been through the field the
gleanings belong to whoever wishes to take them. The third-century
Palestinian Amoraim, R. Johanan and Resh Lakish, are quoted as
defining namoshot either as ' old men who walk [slowly] with the aid
of their sticks' (and so are able to see all the gleanings that are
available, Rashi) or as 'gatherers after gatherers'. In any event, the
Mishnah states that there is a stage at which all the poor of that
town have had their choice of the gleanings. Whatever remains is
the result oiyeush on the part of these poor to whom the gleanings
Unconscious abandonment of property 41
belong. But the gleanings really belong, as it were, to all the poor
everywhere. Now it is true that if the poor of other towns were to
know that the namoshot had passed through that field they would
engage in yeush, but they cannot know this and hence, according
to Abbaye, the gleanings ought to be prohibited as yeush she-lo
mi-da!'at. The reply is given, the poor of other towns engage myeush
beforehand, since they know full well that the gleanings in that field
will be gathered by the poor of the town in which it is found.
Another Mishnah (Ma'aserot 3: 4) is now quot ed: ' If dried figs are
found by the wayside, even if they are found beside a field in which
figs have been laid out to dry, and so, too, if a fig tree overhangs
a road and figs are found underneat h it, they are permitted and
there is no prohibition because of the laws against theft and [as
ownerless property that has been appropriated] they are exempt
from tithing. In the case of olives and carobs, however, they are
forbidden.' The Tal mud observes that the first law recorded in this
Mishnah, that the dried figs and the figs underneat h the tree are
permitted, affords no refutation of Abbaye. These dried figs are
valuable and, as in R. Isaac' s principle, the owner constantly
examines his field and soon becomes aware of his loss. So, too, with
regard to the figs underneat h the tree, the owner is aware that figs
do constantly fall from the tree and hence engages inyeush from the
beginning. But the second law in the Mishnah appears to refute
Rava. It is at present assumed that the reason why the olives and
carobs are forbidden is because these do not drop so easily from
the tree and so there has been no actual, positive yeush. But when
the owner does become aware of his loss it is assumed t hat he does
engage inyeush. We thus have here an instance oiyeush she-lo mi-da'at.
The reply is given by the late-third-century Palestinian Amora, R.
Abbahu. (The chronology is here somewhat odd. R. Abbahu could
hardly have replied to a difficulty presented on the debate between
Abbaye and Rava. The meaning is either that the issue debated
by Abbaye and Rava was known in the time and place of
R. Abbahu or, more probably, that R. Abbahu made his comment
independently on the Mishnah and this fortuitously was of help to
Rava in his debate with Abbaye.) According to R. Abbahu the
reason why the olives and carobs are forbidden is because everyone
can see that they belong to the owner of the tree underneat h which
they are found and hence there is no yeush at all, even when the
owner discovers his loss. The owner will not despair because he
42 The Talmudic argument
knows that people will appreciate that the olives and carobs are
his and will not take them for themselves. In that case, it is objected,
why not say the same with regard to figs ? To this the reply is given
by Rava's disciple, R. Pappa, that when figs fall from the tree onto
the ground they become repulsive and uneatable and hence there
is yeush.
A further attempt at proof is from another Baraita: ' If a ganav
[' thief] took from one man and gave to another and so, too, if a
gazlan ['robber'] took from one and gave to another and so, too,
if the River Jordan took from one and gave to another, that which
has been taken has been taken and that which has been given has
been given', i.e. and the one to whom it has been given may keep
it, presumably because the owner has engaged inyeush. Now a. ganav
is a sneak-thief who steals by stealth while a gazlan is a robber who
snatches property from its owner with the owner's knowledge but
under protest. Hence, the Talmud remarks, there is no refutation
of Abbaye from the cases of gazlan and the Jordan because the
owner is aware of his loss. But in the case of the ganav the owner
in unaware at the time of his loss and this is a refutation of Abbaye.
The reply is given that R. Pappa has 'interpreted' (tirgemah) this
clause as referring not to a sneak-thief, the usual meaning of ganav,
but to an armed robber. In that case, the Talmud objects, it is the
same as the gazlan, to which the reply is given that the Baraita is
recording two kinds of gazlan, the one who snatches the object from
its owner and the one who threatens the owner with violence if he
does not surrender the object. In both cases the owner is aware of
his loss. The point of this discussion is, it is assumed that when an
object has been stolen the owner will engage inyeush so that even if
the object has a siman it is treated here like an object which has no
siman with regard to the law of lost property (see Tosafists s.v. mah
she-natan).
Another Baraita is now quoted: ' If a river swept away a man's
beams, wood or stones and deposited these in a neighbour's field,
they are his [the neighbour's] because the owners have engaged in
yeush", i.e. it can be assumed that there has been yeush right from
the beginning, when the river swept the goods away. But according
to Rava, since it is evidently a case where the owner does engage
in yeush when he becomes aware of his loss, why does he have to
be aware of it at the time? The Talmud replies that the Baraita deals
with the case of an owner who can save his goods, hence they are
Unconscious abandonment of property 43
only permitted when it is known that he has engaged inyeush from
the beginning. But in that case, the Tal mud objects, why does the
Baraita go on to state: ' If the owners were running after them [the
beams etc.] they are forbidden' ? Why are they only forbidden
then? Since we are dealing with objects that can be rescued, they
ought to be forbidden in any event. The reply is that the objects
can be rescued but only if great effort is taken. Consequently, where
the owners are running after them this demonstrates t hat there is
noyeush, otherwiseyeush can be assumed.
The discussion now takes a new turn. A Baraita is quoted which
does not deal With yeush at all and is yet said to be relevant to the
debate between Abbaye and Rava. This Baraita deals with the
separation of terumah, the tithe given to the priest (Numbers 18: 12).
The Baraita reads: ' In what manner did the Sages say that when
one separates terumah without the knowledge of the owner the
terumah becomes such? If a man went into the field of his neighbour
and gathered sheaves and separated terumah without the owner' s
permission, then, if the owner suspects theft [i.e. if the owner is
displeased, imagining t hat the one who has separated terumah is
robbing him in being too generous to the priest] it is not terumah,
otherwise it is terumah. How can he know whether or not the owner
is suspicious of theft? If the owner came along and said: "Why did
you not go to the better ones?", then, if there are better ones there
it is terumah, otherwise it is not terumah [the owner is being sarcastic].
If the owner himself gathered further sheaves and added these to
the ones gathered it is in any event terumah.' From this Baraita it
follows that when the owner of the field later declares: ' Why did
you not go to the better ones?' , it is terumah because this demon-
strates t hat he is the sort of person who is only too pleased t hat his
terumah has been separated on his behalf. Thus, as it were, there
is a state of willingness with regard to terumah and hence the
separation is valid even though at the time of separation there was
no positive awareness by the owner of what was being done on his
behalf. This refutes Abbaye who holds that for yeush to be effective
positive awareness is required at the time.
On the face of it, what has terumah to do With yeush? The real
point of this stage of the discussion is as follows. It is here being
suggested that the debate between Abbaye and Rava is not
specifically about yeush but about ' knowledge' or ' awareness'
where the law requires these. Their debate does not revolve around
44 The Talmudic argument
the legal definition ofyeush but rather around the legal definition
of'knowledge' or 'awareness', wherever these are required by a
particular law. In both cases -yeush and terumah - awareness is
required on the part of the person involved. By definition bothy eush
and separation of the terumah depend on the person's awareness of
what is being done. Now Rava does not demand a positive aware-
ness for yeush to be effective but, as it were, only an awareness state.
It follows that since Abbaye does require positive awareness for
yeush to become effective he will require positive awareness for
terumah to be effective and, conversely, since the Baraita evidently
does not require it in the case of terumah it will not require it in the
case ofyeush, which refutes Abbaye.
The reply is given by Rava himself- ' Rava interprets [tirgemah]
it on behalf of Abbaye'; as if to say, Rava admits that Abbaye can
easily find his way out of the difficulty through a reinterpretation
of this Baraita. Rava suggests that the Baraita does not deal, as was
first thought, with a man who goes into the field of his neighbour
to separate terumah without the owner's permission but with a man
who has been delegated by the owner to separate his terumah.
'Without the knowledge of the owner' in the Baraita does not mean
that the owner is totally unaware but simply that the owner, when
he delegated the other man to separate his terumah, did not inform
him of the quality of the sheaves he should separate. The Talmud
adds that this interpretation must be given in any event since
according to a Rabbinic understanding of Numbers 18: 28 an agent
cannot act without the knowledge of his principal, i.e. without
being authorised by the principal to act as such. Thus, the Talmud
continues, the Baraita refers to an owner who did instruct his agent
to separate terumah on his behalf. The owner, however, failed to give
his agent instructions regarding the quality of the sheaves he
intended him to separate. Normally a man gives average quality
for his terumah, neither bad nor good, but here the agent took it upon
himself to give the better quality. Here the statement: ' Why did
you not go to the best?' suffices to demonstrate that this was the
owner's intention right from the beginning, that he left it entirely
to the agent to determine which quality he will separate and that
whatever is done is done with the full approval of the owner.
This is followed by a digression from the main theme in the form
of a narrative. The fifth-century Babylonian Amoraim, Amemar,
Mar Zutra and R. Ashi, paid a visit to the orchard of one Mari
Unconscious abandonment of property 45
bar Isak, whose tenant-farmer brought dates and pomegranates for
the Rabbis to eat. The other two Rabbis saw no har m in eating
but Mar Zutra, renowned for his scrupulous piety, refused to eat
since Mari was unaware of the gift. When Mari did come along
he remarked to his t enant : ' Why did you not give the Rabbis those
better fruits?' When Mar Zut ra still refused to eat, his colleagues
reminded him of the ruling in the Baraita. Mar Zut ra, however,
quoted a ruling of Rava that the principle, stated in the Baraita,
of ' Why did you not go to the better ones?' applies only to terumah,
the giving of which is a religious obligation, whereas Mari may
really have been displeased and only said what he did in order to
avoid being embarrassed.
The Tal mud now reverts to the main theme. Another Baraita is
quoted, dealing with the following. Leviticus 11: 38 reads: ' But if
any water be put upon the seed, and any part of their carcase fall
thereon, it shall be unclean unto you. ' The Rabbi ni c understanding
of this verse is that food only becomes contaminated, through
contact with a source of ritual contamination such as a dead reptile,
if the food has first been in contact with some liquid. Fruit while
still on the tree cannot become contaminated. If the fruit has been
plucked from the tree it only becomes susceptible to contamination
if it has first been ' prepared' (hekhshar) by coming into contact with
liquid after it has been plucked from the tree. This is sometimes
referred to as the law of ' be put ' - hi yuttan - after the verse in
Leviticus. There is the further law that in order for the liquid to
' prepare' the food the owner of the food must be willing for the
liquid to be there. We are now in a position to understand the
relevance of the Baraita quoted to our discussion. The Baraita reads:
' if the dew is still on them [some fruit] and he was pleased [that
the fruit should be moistened] then it is a case of be put. But if the
dew had dried, then even though he was pleased [that the fruit had
been moistened] it is not a case of be put.' In other words, the owner
of the fruit must be pleased with the moisture while it is still there.
But if Rava is correct, the fact that the owner was later pleased
demonstrates that there was a state of' unconscious awareness' at
the time, when the dew was on the fruit. This supports Abbaye.
As the Tal mud spells it out: ' What is the reason? Is it not because
we do not say: since it has become evident t hat he is now pleased
he is pleased from the beginning?'
This section, too, has to be understood on the basis of our
46 The Talmudic argument
previous remarks regarding the ' awareness' issue between Abbaye
and Rava. It has to be said once again that the Talmud understands
our debate as not specifically about yeush but about any law in
which 'awareness' is required for something to take effect. Does
'unconscious awareness', i.e. a state of'willingness' but no actual,
positive 'willing' suffice. According to Rava it does but according
to Abbaye it does not. The Baraita thus supports Abbaye and refutes
Rava.
To this the Talmud replies by referring to the words ' be put'
in our text. Traditionally this is written without the vav, although
it is read as ki yuttan. Now according to this traditional way of
writing the words can be read not as ki yuttan, ' be put' , but as ki
yitten, 'he puts' . Hence here even Rava will agree that positive
awareness - as when he actually puts it there - is required. The
Talmud asks, in that case, even if he is pleased, it ought not to be
effective since we require him actually to put it there and in this
case the dew came onto the fruit without human agency. In reply to
this a piece of exegesis by R. Pappa is quoted. R. Pappa contrasts
the traditional reading - ' be put' - with the traditional writing - ' he
puts' In order to reconcile the two, the rule is stated that it is
effective even if the liquid came onto the food by itself (as in ' be
put') but (like 'he puts') it is only effective if he knows about it.
Thus even though as a general principle Rava holds that ' uncon-
scious awareness' suffices, it does not suffice with regard to the law
of hekhsher because of the special implications of the ' be put' and
'he puts' verse. The Baraita of hekhsher thus deals with an exceptional
case.
A final proof is quoted to support Abbaye and refute Rava. This
is from an exposition by R. Johanan in the name of the Tanna
R. Ishmael (variant reading R. Simeon) ben Yehozadok. This
reads: 'Whence do we know that a lost article swept away by a
river is permitted? Because it is written: "And so shalt thou do with
his ass; and so shalt thou do with his garment; and so shalt thou
do with every lost thing of thy brother's, which is lost from him,
and thou hast found" (Deuteronomy 22: 3). When it is "lost from
him" but "found " so far as other men are concerned but excluding
this [where the river swept it away] where it is lost from him but
is not found so far as others are concerned.' The Torah law only
demands that a lost article be returned when it is only 'lost' so far
as the loser is concerned. If it is 'lost' to all, i.e. when the river has
Unconscious abandonment of property 47
swept it away, the Torah allows the eventual finder to keep it, it
is treated by the Torah as ownerless property. (This is the reference
at the beginning of our sugya where it says: ' as we shall have
occasion to state later on' . )
Now the Tal mud takes all this to mean t hat the Tor ah is record-
ing two cases: (1) issura (' that which is forbidden' ), when it is only
lost from him; (2) hetera (' t hat which is permi t t ed' ), when it is
swept away by the river. Moreover, it is as if Scripture declares:
hetera is always permitted, whether or not the article has a siman
and issura is always forbidden whether or not it has a siman. This
is as if the Torah had stated explicitly that a lost article is forbidden,
even if there is no siman, if the loser is unaware of his loss (as he
is unaware in the case of hetera when the river sweeps it away). This
favours Abbaye.
The conclusion of the whole discussion is: ' The refutation [tiyuvta]
of Rava is, indeed, a refutation. And the ruling is in accordance
with Abbaye in the cases Y' ALKGM. ' The latter is a mnemonic,
referring to six disputes between Abbaye and Rava in the Tal mud.
Although, generally, where Abbaye and Rava debate an issue, the
ruling follows the opinion of Rava, in these six cases it follows the
opinion of Abbaye.
According to Rashi Y'ALKGMdenotes:
(1) Y = yeush she-lo mi-da'at, our case
(2) 'A = 'ed zomem, 'false witness', Sanhedrin 27a
(3) L= lehi ha-'omed me-elav, ' a post put up accidentally', 'Eruvin 15a
(4) K = kiddushin she-lo nimseru le-viah, ' betrothal which cannot lead to
cohabitation', Kiddushin 51a
(5) G= gilluy da'at-be-gittin, 'revealing one's attitude indirectly in
divorce', Gittin 34a
(6) M= mumar okhel nevelot lo-hakh'is, ' an apostate who eats forbidden
food out of spite', Sanhedrin 27a
The Talmud, in fact, only records Y'ALKGMin (1), (2) and (4), but
for (5) see Gittin 34b top and Rashi ad loc. Cf. Tosqfists Kiddushin 52a s.v.
be-fal kgm for other suggestions.
The sugya concludes with a note following on the ruling in favour
of Abbaye. The fifth-century Babylonian Amora, R. Aha son of
Rava, asked R. Ashi: ' Since Rava has been refuted, how can we
eat dates the wind has blown from the t ree?' , i.e. the owner of the
tree does not necessarily know of it at the time. R. Ashi replies t hat
since the owner knows that any dates t hat will be blown off the
48 The Talmudic argument
tree will be eaten by vermin he engages in yeush right from the
beginning, i.e. he knows that some dates will be blown off the tree
to be eaten by vermin so he engages in yeush for these dates even
while they are on the tree. But, R. Aha continues, what if the trees
belong to orphans who are minors and, since a minor has no powers
of consent in law, hisyeush is ineffective. R. Ashi replies that we are
not bound to assume, of all the trees on that large plot of ground,
that the date-palms belong to orphans. R. Aha then asks, but what
if we do know that the palms belong to orphans or what if there
is a fence to keep out the vermin? In these cases, R. Ashi replies,
the dates are indeed forbidden.
Note that Aha son of Rava states in his question: ' Since Rava has been
refuted', not: 'Since my father has been refuted', which shows that this
Amora is not the son of our Rava but of another Rava, see Hyman: Toledot,
vol. 1, pp. 130-2.
In this sugya there is material from second-century Palestine (the
Mishnah and Baraita) down to fifth-century Babylon (R. Ashi and
his colleagues), all fitted into the framework by the later editors.
Although in some instances of tiyuvta, ' it is a refutation', this is
generally held to be the work of the post-Talmudic Saboraim, here
it would seem to have been inserted prior to R. Aha son of Rava,
who refers to it, though it is possible that these are not R. Aha's
actual words and were added later. The mnemonic at the beginning
of the sugya, found in our editions in brackets and listing the whole
series of proofs, is undoubtedly Saboraic and supplied after our
whole sugya had received its final form. There is some evidence of
two successive editing processes, e.g. the words 'as we shall have
occasion to state later on' seem to suggest that these were supplied
when the main sugya had already been edited. It is, indeed,
probable that this whole explanatory comment at the beginning
of the sugya, describing in detail the limits of the debate, was
supplied later and the same applies to the digression regarding Mari
bar Isak, which is a unit on its own and seems to have been inserted
by the later editors. The series of proofs are presented in a logical
arrangement: first the Mishnah, then the Baraitot dealing Withy eush
directly, then the indirect Baraitot or terumah and ki yuttan, and,
finally, the conclusive proof which is left to the end; all of which
demonstrates clearly that the material has been given literary shape
by the editors. It is to be noted that there are no references elsewhere
Unconscious abandonment of property 49
in the Tal mud to yeush she-lo mi-da'at, except for the Y in Y' AL
KGMand this is almost certainly a later addition, made when the
whole Babylonian Tal mud had been completed. It is puzzling t hat
Y' ALKGMonly occurs in three instances. It is possible t hat it did
originally occur in all six places.
5
Rubba: probability
The principle that the Torah allows reliance on probability in cases
of doubt is found in numerous Talmudic passages. The sugya we
here examine Hullin 11 a12a is concerned with the derivation
of the principle from Scripture. What is the Scriptural warrant for
relying on probability rather than demanding certainty? The term
rubba is in Aramaic (Hebrew: rov) with the meaning of'majority',
hence the majority or most likely opinion. It can be translated
literally into English as 'majority' but in most instances the word
'probability' comes nearest to the meaning. The sugya does not
question the application of the principle. That is everywhere
assumed. Its concern is solely with the question of Scriptural
authority for the principle.
The sugya first sets the stage for the discussion. The opening words
are: 'How do we know that which the Rabbis say: "Follow [lit.
'go after'] probability" [rubba]?' The Talmud immediately
interjects: how can you ask, 'how do we know?'; surely there is
no problem since Scripture states explicitly: ' Turn after the
majority' (Exodus 23: 2)? That is to say, the Talmud is puzzled
by the request for Scriptural warrant for the principle, as if this were
unclear. There is an explicit verse which states that rubba is
followed. (Actually, the plain meaning of the verse, in its context,
is ' Do not turn after the majority to do evil' but in its traditional,
Rabbinic, interpretation is: ' Do not follow a multitude to do evil
(but otherwise) turn after the majority'.) The verse is applied by
the Rabbis to a Court of Law, e.g. if two judges in a civil case
declare a man to be liable and one judge declares him not to be
liable or, in the case of a Lower Sanhedrin in a capital charge, if
Probability 51
12 declare him to be innocent and 11 guilty, the majority opinion
is followed.
In the reply, three illustrations of rubba are given. One of these
is that of the Sanhedrin, where the majority opinion is decisive.
Another is that of nine butcher's shops selling meat that has been
slaughtered properly by the method of shehitah and is hence kasher
and one shop selling non-kasher. By the rubba principle any meat
found in that town is kasher since a majority of the shops is kasher.
The meat is assumed to have come from one of the shops belonging
to the majority rather than from the single, minority shop. (The
same law would apply if only two shops sold kasher meat and a third
non-kasher, since there is still a simple majority.) The third
illustration is from a case of levirate marriage (Deuteronomy 25:
5-10) debated by R. Meir and the Sages (Tevamot 61b). The
purpose of levirate marriage is for the levir and the widow to have
a child. Normally it is forbidden for a man to marry his deceased
brother's wife but the Torah enjoins it where the brother has died
without issue. It follows that if either the levir or the widow is sterile
the prohibition of marrying a brother's wife applies. Now the
condition of sterility, which in this context refers to physical defects
in the generative organs of man or woman, cannot be determined
while male or female are still minors. Consequently, the question
arises, may levirate marriage take place where the levir is a boy
under age or the widow a girl under age? It can be argued, and
R. Meir does so argue, that this is forbidden since it may turn out
later than the levir or the widow is really sterile. But the Sages
disagree with R. Meir and permit the marriage on the grounds
that the majority of persons are not sterile. It is improbable that
either is sterile and the rubba principle is accepted. There are thus
three cases of the application of rubba: (1) nine shops 5(2) Sanhedrin;
(3) minor boy or girl.
We now turn to the reply given in the Talmud to the question,
why not derive rubba from the verse: ' Turn after the majority'?
Indeed, the Talmud replies, one type of rubba - that represented
by the nine shops and the Sanhedrin requires no further Scriptural
proof since it is derived from our verse. This is a case of"'rubba before
us' . Our question, however, is: what Scriptural warrant is there for
relying on a different kind of rubba, that which is ' not before us' ,
such as the case of the boy or girl?
52 The Talmudic argument
Superficially, the distinction between the two types of rubba is
clear. The Scriptural verse deals with the Sanhedrin, 12 members
of which declare the man on trial to be innocent, 11 guilty. Because
of the rubba principle he is declared innocent. Here the majority
is 'before us', i.e. we can observe the majority which decided for
acquittal. Similarly, in the case of the nine shops, we can observe
the nine kasher shops on the basis of which the meat found is
permitted to be eaten. But in the case of the boy or girl the majority
is 'not before us'. We do not observe the majority on the basis of
which we decide that the boy or girl is not sterile. So far so good.
But what is the logical ground for the distinction? Why should it
make any difference whether or not the majority is 'before us'?
Much has been written on this question but the distinction would
appear to be the following. In the case of the Sanhedrin the decision
comes from the majority. It is basically a procedural matter, the
Torah ruling that the majority opinion be followed. The man
declared innocent by the majority is not said to be innocent, as if the
majority were always right, but he is to be acquitted, since otherwise
no case could be carried to a successful conclusion without a
unanimous verdict, which the Torah does not demand. Thus if a
majority decide that the man on trial is innocent he is acquitted,
even if, in fact, as the minority holds, he is really guilty. He is
innocent in law. Conversely, if the majority decides that he is guilty
he is sentenced, even though, in fact, he may be innocent, as the
minority holds. He is guilty in law. It is all a matter of legal
procedure, not a determinative matter. From this case of the
Sanhedrin that of the nine shops is derived. This, too, is a question
of legal procedure in religious law. Since there is a doubt whether
or not the meat is kasher, the procedure to be adopted is to resolve
the doubt by relying on the probability principle, that the meat
comes from the nine shops, though, in fact, it may well have come
from the non-hasher shop. Here, too, the matter is procedural, not
determinative. The meat is kasher in law no matter what its true
nature, since the Torah allows us to rely on probability and that
which the Torah allows is kasher. But in the case of the boy or girl
it is not a procedural matter but one of determination arrived at
by means of empirical investigation into the majority of instances.
When we say that the marriage of the boy or girl is permitted
because the majority of persons are not sterile, this does not mean
that at any given moment there is an observable majority of
Probability 53
non-sterile persons from which the boy or girl derive. That is
nonsense. It is rather that we say, since our observation demon-
strates that to be sterile is not the normal human state, it is safe
to assume that the boy or girl will have the normal human nature
and will not be sterile. We rely, in other words, on empirical
investigation and the high degree of probability this affords. That
this kind of rubba is to be relied on does not at all follow from the
fact that the Torah permits us to rely on the very different kind
of rubba that is 'before us' , that is, purely procedural. From the fact
that the Torah allows us to rely on a purely mathematical
probability in the case of the Sanhedrin and the nine shops, as a
matter of procedure, it cannot follow that where the laws of
probability have to be relied on to determine the status of an
individual they are so relied on. Consequently, the question in our
passage is, what is the Scriptural warrant for relying on probability
in such cases as that of the boy or girl which is an instance of' rubba
that is not before us' ?
Another illustration of' rubba that is not before us' is given in the sugya as
it proceeds. If an animal has a defect in one of its vital organs, this is known
as terefah. Eighteen such defects are recorded (Mishnah Hullin 3:1) that
render the animal terefah. If the internal organs of an animal that had had
shehitah were left unexamined the animal would still be kasher on the
grounds that' the majority of animals are kasher' (rov behemot kesherot). This
is another example of'rubba that is not before us', which is the concern
of our sugya. Rabbi Hayyim Soloveitchik, R. Hayyim Brisker, is reported
to have put the distinction between the two kinds of rubba neatly in this
way. Even if all the animals in the world had died and only one animal
were left, that animal would be kasher because of the principle ' the majority
of animals are kasher'. An illustration based on the ideas of Rabbi Simeon
Skopf of Grodno is the following. A coin has turned up heads three times.
It would be a rash gambler who would bet on it turning up heads the fourth
time. The mathematical probability is that it will turn up tails on the fourth
toss, assuming that the coin is not loaded. But the reason that the odds
are in favour of the coin turning up tails the fourth time is not because
anything will happen to change the nature of the coin. But when a gambler
bets on the favourite in a horse race he does so because the odds on the
horse winning are determined by the horse itself, because of its previous
performance which demonstrates that it is a 'winning' type of horse. The
probability here is determinative, telling us something about the physical
nature of the horse. All this can be compared with modern philosophical
discussions on the theory of probability and with Hume's famous under-
standing of causality as empirical not metaphysical. The standard works
on the theoretical discussion of doubt and rubba in the Talmudic and
post-Talmudic literature are Aryeh Laib Heller's Shev ShemaHata and
54 The Talmudic argument
Simeon Skopf s Sha'arey Tosher. Goitein's KesefNivhar, no. 144 is also useful
for a survey of most of the Talmudic and important post-Talmudic sources.
Having defined the scope of the inquiry - the Scriptural warrant
for relying on rubba when it is ' not before us' - the Talmud
proceeds to give a list of Amoraim, each of whom tries his hand
at adducing conclusive proof from a Scriptural verse.
The first proof is attempted by R. Eleazar. We have noted that
an animal with a serious defect is terefah and forbidden to be eaten.
It is also forbidden to offer such an animal as a sacrifice. (Some
defects which render an animal terefah can be detected while the
animal is still alive.) R. Eleazar notes the verse dealing with a
burnt-offering: And he shall cut it into its pieces (Leviticus 1:6),
one of the pieces mentioned being the head. Since the verse stresses
'cut it into its pieces' the traditional explanation is: ' but not its
pieces into pieces', i.e. it is forbidden to cut up the head into smaller
pieces. The head must be burnt on the altar while it is whole. Now
if the membrane which covers an animal's brain is perforated, that
animal is terefah and cannot be offered as a sacrifice. How do we
know that the burnt-offering is not terefah, that its membrane has
not been perforated? It cannot be that the head is cut open before
it is burnt on the altar to see whether or not the membrane is intact,
since Scripture forbids the cutting of the head into pieces. And yet
the Torah does record this law of the head of the burnt-offering.
It can only be that the principle of rubba is relied on and the
majority of animals are not terefah. R. Eleazar rests his case.
No, the Talmud retorts, this proof is invalid. It may be that rubba
is not relied on and that, indeed, the head has first to be cut into
in order to examine the membrane. As for the objection, it can be
suggested that the Torah only prohibits the cutting of the head into
two or more 'pieces', but there is no objection, if the head is kept
whole, to cutting down into the head until one reaches the
membrane.
Another proof is attempted by Mar son of Ravina. Of the
Paschal lamb the Torah says: ' neither shall ye break a bone
thereof (Exodus 12: 46). But the membrane of the animal may
have been perforated and the skull must not be broken into in order
to conduct the examination to see whether the membrane is intact.
This shows that the Torah does rely on the rubba principle, that
the majority of animals are not terefah. The Talmud invalidates this
Probability 55
proof by suggesting that the examination can be carried out by
burning through the skull with a hot coal until the membrane is
reached. To do this would not offend against the law forbidding
the breaking of its bones. A Baraita is quoted in which it is stated
explicitly that there is no objection to burning through the bones
of the Paschal lamb, only to breaking them.
The proof attempted by R. Nahman bar Isaac is then quoted.
The verse regarding the offering of a sheep states: ' The fat thereof
and the fat tail entire' (Leviticus 3: 9). Thus the fat tail has to be
burnt entire on the altar and this implies that it must not be cut open
before it is burnt. Now one of the terefot is the severance of the spinal
cord and the spinal cord goes into the tail. The tail cannot be cut
open to see whether or not the spinal cord has been severed since
the Torah ordains that the tail must be entire. If it be objected, the
Talmud continues, that the tail is cut off lower down than that part
of the spinal cord the severance of which renders the animal terefah,
this cannot be since Scripture states (in the same verse) that the
tail must be removed ' hard by the rump bone' which is ' where the
kidneys give counsel' (a pun on 'etzeh, ' rump bone' and 'etzah,
'counsel'), which means above the part of the spinal cord the
severance of which makes the animal terefah. To this the reply is
given, there is no proof. It may be that the tail has first to be cut
open to see whether or not the spinal cord is intact. As for the
objection, ' entire' only excludes the complete division of the tail
into parts but there is no objection to cutting into the tail until
the spinal cord is reached, since the tail is still then ' entire' , in one
piece.
There follows the attempted proof of R. Sheshet son of R. Idi
from the law of the heifer whose neck is broken (Deuteronomy 21:
19). Scripture states (verse 6) 'the one whose neck was broken'
(with the definite article). According to the traditional explanation,
this means that when the heifer is buried it must still be in the state
it was when it was beheaded. The only way to determine that the
animal has no defects in its internal oregans is to examine these and
here this is not possible since the heifer has to be as intact at the
time of its burial as it was at the time of its beheading. This affords
conclusive proof that the Torah does rely on rubba and the majority
of animals are not terefah. It may be objected, observes the Talmud
further, that perhaps there is no need for the heifer to be hasher (and
hence the proof would fall away) but this is not so. The heifer must
56 The Talmudic argument
not be terefah since Scripture speaks of the heifer as producing
'forgiveness' (atonement', verse 8) and the same law that it has
to be kasher applies to it as to the sacrifices which also produce
'forgiveness'. This comparison is stated as a teaching taught in the
school of R. Jannai, i.e. there is, in any event, a rule that the heifer
must not be terefah. R. Sheshet's proof is not refuted.
Rabbah b. Shila seeks to prove it from the law of the red heifer
(Numbers 19) of which Scripture states: 'And he shall slaughter
i t . . . and he shall burn it' (verses 3 and 5). The meaning is said
to be that when the heifer is burnt it must be in the same state of
' wholeness' as it was when it was slaughtered and this implies that
it must not be cut up in between, which means that no examination
is possible to determine whether or not it is terefah. If it be objected
that there is no proof that the red heifer must not be terefah, the
proof is that Scripture calls the heifer ' a sin-offering' (verse 9) and,
like the sin-offering, which is a sacrifice, it must be kasher. Rabbah
b. Shila's proof is not refuted. The red heifer is valid even though
no examination is possible, because the Torah relies on rubba and
the majority of animals are kasher.
R. Aha bar Jacob seeks to prove it from the law of the scapegoat
on the Day of Atonement. Scripture states: 'And he shall take the
two goats' (Leviticus 16: 7), which means that both must be alike.
(Some texts do not have this latter phrase.) But one of them might
be terefah? This proves that the rubba principle is accepted and the
majority of animals are not terefah. And if it be objected that there
is no proof that the scapegoat has to be kasher (unlike the goat
offered as a sacrifice), two lots have to be cast to determine which
is the scapegoat and which the sacrifice, from which it follows that
both goats must be kasher at the time of the casting of the lots since
either may be chosen by lot for the sacrifice. If it be further
objected that the scapegoat can be examined after it has been
pushed off the mountain crag (Rabbinic understanding of Azazel
in verse 8) this cannot be, because the Mishnah (Toma6: 6) informs
us that the goat was broken into pieces before it reached half-way
down the mountain. R. Aha bar Jacob's proof stands.
So far all the attempted proofs have been from the sacrificial
system or matters connected with it. R. Mari's proof is from a
different law, that of smiting a father or mother, the penalty for
which is death by the Court (Exodus 21: 15). But the man smitten
by the ' son' may not really be his father? It must be that paternity
Probability 57
is established on the basis of rubba. The man is married to the
smiter's mother and the majority of acts of cohabitation would be
by the husband not by her lovers if she had had any. To this the
Talmud objects that the law may only apply where husband and
wife were imprisoned in a dungeon with no access to any other man
from before the child's conception until the woman was seen to be
pregnant. Paternity would here be established for certain and there
would be no need to rely on rubba. To this the Talmud replies by
quoting: ' There is no guarding against unchastity', i.e. even so it
is only by probability that paternity is established since, despite its
inaccessibility, another man may have gained access to the dungeon.
R. Mari's proof stands. (However, some texts do not have the quote
and on this reading R. Mari is refuted.)
R. Kahana' s proof is from the law that a murderer is executed
by the Court. Now just as an animal can be terefah so can a man
if he suffers from the same defects. A human terefah is seen in law
as a man already dead and though it is, of course, a serious crime
to murder him, the murderer is not executed by the Court because
he has, in the words of the Talmud, 'killed a dead man' . In that
case, how can a murderer ever be executed since his victim may
have been a terefah? It can only be because the Torah relies on the
rubba principle and the majority of men are not terefah. But,
continues the argument, you might say that the murderer is only
executed after we have cut up the body of the victim and
conducted a post-mortem examination to determine whether or not
he was a terefah. The reply is that it is forbidden to mutilate a human
corpse. But, the argument continues further, perhaps the
prohibiting of mutilating a corpse is set aside for the purpose of
saving the life of the murderer (i.e. so as to give the murderer the
possibility, albeit a remote one, of escaping the death penalty, as
he will do if the victim is discovered to be a terefah). There would
then be no proof from the law that the rubba principle is accepted.
No murderer would be executed unless the corpse were first
examined. To this the reply is given that even if the corpse were
to be examined for all the different forms of terefah it would still
be necessary to rely on rubba for, if we did not, the victim may have
been terefah^ because the organ penetrated by the sword which
killed him was already perforated and because of the sword-cut
this would not show up in the examination. R. Kanna' s proof
stands.
58 The Talmudic argument
As a point of interest it might be noted that this passage is discussed at
length in R. Ezekiel Landau's famous Responsum on autopsies, Responsa
JVoda' Bi-Thudah, Tinyana, no. 210, and has become the locus classicus for the
whole question. The implication of the Talmudic argument at this stage
seems to be that a corpse may be mutilated if the purpose is to save life,
even, in this case, the life of a murderer and again even, as here, where
there is only a very remote chance of the mutilation resulting in the saving
of a life.
Ravina's proof is from the law of false witnesses. Of a false witness
in a capital charge Scripture says: ' shall ye do unto him, as he had
thought to have done unto his brother' (Deuteronomy 19: 19). But
by an a fortiori argument a false witness against a terefah is not
executed since even if a man actually murders a terefah he is not
executed. Moreover, the false witness in a charge against a terefah
cannot be executed because the law is that the terefah, as a man
'already dead', would not be executed by the Court even if found
guilty, so that the false witness cannot have imposed on him the
penalty he intended to have imposed on his victim. In the case of
a terefah there cannot have been any such intention. It follows from
all this that a false witness against a terefah is never executed. In
that case, how can the law of false witnesses ever be applied since
the victim may have been a terefah? It can only be because the rubba
principle is accepted and the majority of men are not terefah. To
this it might be objected, observes the Talmud, that the corpse of
the victim is examined to see whether or not he was a terefah, but
this cannot be. The teacher Beribbi is quoted as saying; ' If they
[the false witnesses] did not kill him [the victim, i.e. they were
discovered to be false before the man they testified against had been
executed] they are killed but if they did kill him [i.e. if they were
not found to be false until the Court, acting on their testimony, had
executed the man on charge] they are not killed.' That is to say,
a false witness is only executed when he only proposed to commit his
offence, when his offence was discovered while his potential victim
was still alive. If the false witness had not merely proposed to do it
but had actually committed the crime, i.e. his crime had not been
discovered until the Court had executed the victim, then the false
witness is not executed. Thus the law that a false witness is executed
only applies if the victim is still alive. A living person cannot be
examined to determine whether or not he is a terefah so how can
the law of false witnesses ever be applied? It can only be because
we rely on the rubba principle and Ravina's proof stands.
Probability 59
This law of false witnesses, that they are executed if their evil designs have
not been realised but are not executed if they have, is obviously puzzling.
It should have been the other way round if a distinction is to be drawn
at all. If they are executed for only proposing to do it why should they not
be executed if they actually did it? See Mishnah and Gemara Makkot 5b
where the matter is said to have been debated by the Sadducees and
Pharisees and where Beribbi's statement is quoted. The post-Talmudic
teachers seek to provide a rationale for this strange law. Among the
solutions suggested are: God has allowed the victim to be executed so it
is his fate and he is probably guilty even though the witnesses are false;
or if the witnesses have carried out their evil designs they do not deserve
the atonement that would be theirs if they were executed; or if the man
has been executed the likelihood is that the witnesses are not, in fact, false
despite the evidence to the contrary. See e.g. Meiri to Makkot ad loc. and
David Hoffman, Responsa Melammed le-Hoil, part in, no. 101.
The final proof is R. Ashi's. R. Ashi argues that the rubba
principle can be derived from the act of shehitah itself. Shehitah is
performed by cutting through, with a sharp knife, the animal's
food-pipe and wind-pipe. Now the rule is that the slightest
perforation of the food-pipe renders the animal terefah. Conse-
quently, unless we are to rely on the rubba principle, that the
majority of animals are not terefah^ shehitah could never be effective
to permit the meat of the animal to be eaten since the animal
might have had a perforated food-pipe, through which the knife
has cut, so that any subsequent examination would be impossible.
R. Ashi's proof stands.
Halevy, Dorot ha-Rishonim, vol. 11, chapter 68 end, pp. 561-2, gives a novel
interpretation of R. Ashi's 'from shehitah itself. What is the significance of
'itself? Why does R. Ashi not simply say: ' It is derived from shehitah'?
Rashi understands it as : 'shehitah, the very subject we have been discussing
up to now', i.e. in this tractate Hullin, the preceding discussions of which
are all connected with the laws of shehitah. According to Halevy this
demonstrates conclusively that when R. Ashi presented his proof he must
have already had before him some, at least, of the material now in tractate
Hullin, arranged more or less in the order we now have it. He must have had
this rubba discussion, which, after all, is a general discussion not a shehitah
one, in its present position, i.e. coming after material already arranged as
part of tractate Hullin. Hence R. Ashi states: 'We derive it from shehitah
itself, i.e. from this very topic being considered in this part of tractate
Hullin. Two things require to be said. First, it is possible that the meaning
of'itself is not as Rashi has it but that the principle of rubba can be derived
from the impossibility of examining the animal to see whether or not it
is terefah - not, as in some of the earlier proofs, from circumstances which
obtain after the animal has had shehitah but from shehitah itself! Secondly,
60 The Talmudic argument
even if we do understand the passage as Rashi does, this affords no proof
that R. Ashi already had before him a rudimentary form of our present
tractate Hullin, as Halevy suggests, since the words ' shehitah itself may not
have been R. Ashi's own but those placed into R. Ashi's mouth by the
editors when they presented the series of proofs (cf. Neusner, Formation of
the Babylonian Talmud, p. 34), which, as we shall see, are, in any event,
contrived.
At this stage the matter has been proved by R. Ashi and some
of the previous Amoraim. But the Talmud now states that R. Ashi
repeated his argument to R. Kahana; in another version, it is said,
R. Kahana repeated it (the argument of R. Ashi? or, perhaps, his
own argument, as above) to R. Shimi, who objected that the
arguments presented only proved that the rubba principle was relied
on where there was no alternative, where otherwise the law would
be impossible of application. It by no means follows that rubba may
be relied on even when one can have certainty. For instance, in the
case of shehitah it is perfectly true that we must rely on rubba that
the food-pipe has not been perforated, but why should this prevent
our examining the rest of the animal to determine whether or not
it is terefah due to some defect in another of its organs? Or, in the
case of the boy or girl, why should we rely on rubba if we can wait
until they are no longer minors when it can be determined for
certain whether or not they are sterile? And, the Talmud continues,
unless this distinction is made, how could R. Meir, who, as above,
in the case of the boy or girl, does not rely on rubba, eat meat? And
even if it be argued that, in fact, he did not, what of the Paschal
lamb and the meat of other sacrifices it is a religious obligation to
eat? We must perforce make the distinction between where there
is and is not an alternative and by the same token how can rubba
be derived from the Scriptural proofs adduced, since in these there
is no alternative but to rely on rubba?
Here the sugya reaches its conclusion, leaving us somewhat in the
air. Despite all the attempted proofs there is none for relying on
rubba where there is an alternative. Even the successful proofs only
prove reliance on rubba where there is no alternative. Rashi suggests
that according to the conclusion of the sugya we do rely on rubba,
as the Talmud does everywhere, except according to R. Meir, but
it is not derived, as was thought throughout until the conclusion,
from any Scriptural verse. Rashi states that, according to the
conclusion, it is a 'law given to Moses on Sinai', i.e. it is an
Probability 61
established law with full divine authority and can be relied on even
where certainty can be had; but there is, in fact, no Scriptural
warrant for this. Thus the whole series of attempted proofs has no
practical application and is a purely academic exercise. It is even
possible that it was a kind of challenge in the days of the Amoraim
to seek to prove the rubba principle from Scripture, each Amora
trying his hand, the results being collected by the editors to be
presented in the form we now have them as part of a rubba sugya.
Our sugya provides us with an excellent illustration of how the
Talmud was put together. First, let us note the order in which the
proofs are presented:
(0
( 2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
i o)
Name
R. Eleazar
Mar son of Ravina
R. Nahman b.
Isaac
R. Sheshet son of
R. Idi
Rabbah b. Shila
R. Aha bar Jacob
R. Mari
R. Kahana
Ravina
R. Ashi
Date
d. 279
fourth century
d. c. 356
end of fourth
century
third to fourth
century
fourth century
beginning of fourth
century
uncertain which
fifth century
fifth century
Proof
head of burnt-
offering
head of Paschal
lamb
fat tail
beheaded heifer
red heifer
scapegoat
smiting father
murderer
false witnesses
shehitah
The Amoraim mentioned could not have presented their proofs
together in the same time and place. Some two hundred years
divide the earliest from the latest. Nor, as can be seen from the table,
are the proofs presented in chronological order. The only way to
understand the series is that each Amora offered his proof inde-
pendently of the others and these were later put together in this
order by the editors. Why did they choose this particular order?
A careful examination of the proofs and the discussion on them
uncovers the pattern. Thus: (1) is refuted because the head of the
burnt-offering can be split open, as the Talmud observes, and this
does not offend against the prohibition of cutting the pieces into
further pieces. Very well, it is implied, you have refuted proof (1)
but we now adduce proof (2), to which your refutation will not
62 The Talmudic argument
apply. Here the term is ' breaking' and would presumably include
splitting. The refutation is that the burning of the bone does not
constitute ' breaking'. Very well, we adduce proof (3). Here the tail
must be whole and this would presumably exclude any type of
severance. No, it would not include partial cutting. Very well, we
adduce (4). Here the heifer must be as intact afterwards as it was
at the time it was beheaded and this presumably excludes any
interference with it. This stands, though not without us having to
face the difficulty of proof that the heifer has to be kasher. (5) is now
adduced, which also stands but which has to face the same difficulty
of proof. (6) is now adduced. Again this stands but only after
difficulties have been raised. All these are proofs from kindred
matters, sacrifices and animals that are treated as sacrifices. We
now adduce (7), dealing with a law other than sacrifice, but to do
with execution. This stands, but has to face objections. We now
adduce another proof having to do with execution, (8), as if to say,
even if in (7) circumstances can be found where it is certain that
the man is his father, what of (8) where there can be no certainty,
only probability. (According to the reading which omits the
conclusion of (7), as above, (7) is, in fact, refuted, hence the need
for further proof.) This stands, but has to face the objection that
possibly an examination can be made after death. We now adduce
(9), where the victim must be alive and so cannot possibly be
examined. We now adduce (10) to which, unlike all the others, even
those that stand, no objection whatsoever can be raised. This
clinches the argument.
Thus the sequence of thought is completely logical, but can only
be understood as an artificial arrangement of earlier material to
produce this literary pattern. The Amoraim themselves obviously
did not present their proofs in response to the difficulties they found
in each other's statements. Even if they all lived at the same time
and in the same place, which they did not, this would have involved
the preposterous notion that each presented his proof in such neat
order that (2) was able to take the matter up where (1) left off;
(3) where (2) left off; and so on in neat sequence. It must be as
clear as can be that the editors have re-shaped the material they
had so as to provide a dramatic sequence of thought. In other
words, we have here a contrived literary unit constructed out of
the raw materials. It is tempting, but would be hypercritical, to
suggest that the editors made it all up out of their own heads,
Probability 63
attaching names at random to the series of proofs. There is too much
circumstantial evidence for that, though the fact that there is a
round number of proofs, ten, a number of significance in Jewish
thought, may not be entirely coincidental. In any event, the basic
statements of the Amoraim have been elaborated on and a
framework of a literary nature provided to produce a literary unit,
the proofs being presented as z/'each Amora took up the matter
where the previous one has left off.
It remains only to be noted that the mnemonic at the beginning
of the sugya, in brackets, for the whole series of proofs, is undoubtedly
post-Talmudic and must have been composed some time after the
sugya was complete in its present form.
Davar she-lo ba le-'olam:
conveyance of a thing not yet in existence
The sugya described here is Kiddushin 62a-63a. The sugya is appended
to the Mishnah which rules as follows. If a man betroths a woman
and one of them is a Gentile but he declares that he wishes the
betrothal to take effect after his or her conversion to Judaism, the
betrothal is invalid. Similarly, if he is a slave but he declares that
he wishes the betrothal to take effect after he or she has been freed,
it is invalid. Two further cases are: a man betroths a married
woman, the betrothal to take effect when her husband dies; or a
man betroths his wife's sister, the betrothal to take effect when his
wife dies. Finally, there is the case of a man who declares to his
neighbour: ' If your wife will give birth to a girl she is betrothed
to me.' In all these cases the betrothal is invalid on the following
principle: that which cannot take effect now because of some major
legal objection does not take effect even when that obstacle is no
more.
Davar she-lo ba le-'olam, the term used in our sugya, means literally
' a thing that has not yet come into the world'. The question is
whether the conveyancing or transfer of something or of something
being effected is valid if the thing is not yet in existence. This is
formulated by the Amoraim as: ' a man can transfer [lit. 'make
another acquire'] a thing not yet in existence [adam makneh davar
she-lo ba le-'olam]' or ' a man cannot transfer a thing not yet in
existence [eyn adam makneh davar she-lo ba le-'olam].' The Amoraim
examine various Tannaitic statements from which the conclusion
is drawn that some Tannaim hold that a man can transfer
something not yet in existence while others hold that he cannot.
The reasons for the debate are not state explicitly in the Talmud
but the later commentators are in all probability correct when they
64
Conveyance of a thing not yet in existence 65
see it as depending on whether or not there is sufficient consent and
intention where the transfer is of a thing that is not yet in existence.
For an effect to be recognised as valid in law, when that effect
requires the intention and consent of a person, this intention and
consent must be given without equivocation. In normal instances
of transfer it can be assumed that it is being done with the full degree
of consent required. But where the thing to be transferred or
brought into effect is not yet in existence there may be sufficient
mental reservations, precisely because the thing is not in existence,
to interfere with the degree of consent and intention required.
Instances of this debate cover every type of acquisition, kinyan.
Thus, is the cases dealt with in the Mishnah, the concern is with
the 'acquisition' of a wife, i.e. the valid betrothal of a woman so
that she becomes a married woman in the eyes of the law and cannot
marry another without the bill of divorce, the get. Since the
Mishnah rules, in the cases recorded there, that the betrothal is
invalid, it is said that the Mishnah follows the opinion: eyn adam
makneh davar she-lo ba le-'olam, ' a man cannot cause a thing not yet
in the world to be acquired'. For instance, where the man is at
present a Gentile, who cannot effect a valid betrothal, the woman
he wishes to ' acquire' , that is to betroth, the woman he can
legitimately betroth in Jewish law, does not exist at the time of the
betrothal. He wishes, therefore, his present act to take effect with
regard to something not yet in existence. Since the reason why the
betrothal is ineffective in the Mishnah and the transfer is ineffective
in other cases of davar she-lo ba le-'olam is because the requisite degree
of consent and intention is lacking, it follows that in all such cases
where circumstances can easily be brought about for the thing to
come into existence, there is good reason to suppose that intention
and consent are not lacking. This helps us to appreciate why the
whole of our sugya implies that even according to those authorities
who hold that the transaction is invalid, there are circumstances
(where the effect can easily be produced), where it will be valid.
Conversely, it is implied, even those authorities who hold that
normally the transaction is valid, even if the thing is not yet in
existence, may agree if the circumstances are such that there are
difficulties of one kind or another which prevent the thing not yet
in existence from coming easily into existence.
The sugya opens with a discussion on a Mishnah (Terumot 1:5)
dealing with terumah, the tithe given to the priest. Untithed produce
66 The Talmudic argument
{level) must not be eaten. The effect of the separation of the terumah
and the other tithes is thus to permit the rest of the produce to be
eaten. Any produce separated as terumah may only be eaten by a
priest. A further rule is that terumah is only to be separated when
the produce has been harvested, detached from the soil. If a farmer
declares that a portion of his crops growing in the field should be
terumah his declaration is ineffective. The field remains in its
untithed state and, when it is harvested, the tithes will have to be
given. The Mishnah rules: 'Terumah from unattached produce
cannot be separated for attached produce and if it has been
separated it is not terumah.' The meaning is, if the farmer took some
of his unattached produce, produce that had been harvested, and
declared it to be terumah for the produce still attached to the soil
(i.e. to exempt him from giving further terumah from the attached
produce), this is invalid since there is no obligation to separate
terumah while the produce is still attached to the soil. Consequently,
the act of separation has no validity. The attached produce will still
require to be tithed when it is eventually harvested and the amount
of unattached produce that has been separated as terumah does not,
in fact, become terumah and even the priest may not eat it, since
it is still tevel.
A problem, arising out of the Mishnaic rule, is set by the
third-century Palestinian Amora, R. Assi, to his teacher, R.
Johanan. We have noted that terumah from unattached produce
cannot be separated for attached produce and the same would
apply if terumah is separated from attached produce for unattached.
But suppose the farmer declared at the time of separation that his
intention is not for the terumah to become such immediately but
rather when the produce will be detached from the soil. At present
it cannot be terumah since there is no tithing for produce attached
to the soil. Equally obvious is the fact that he can, of course,
separate the terumah later when it does become detached. R. Assi's
problem is whether or not the present separation, with the express
intention of its taking effect when the produce will become
detached, makes the produce separated become terumah retro-
spectively, when eventually the produce does become detached.
The detached produce is not yet in existence in its detached terumah-
liable stage so that the question would seem to depend on whether
or not we follow the opinion of those who hold that an effect can
be brought about even though its object is not yet in existence - adam
Conveyance of a thing not yet in existence 67
makneh davar she-lo ba le-'olam. R. Johanan replies that it would be
effective even according to the authorities who hold that there can
be no transfer of a thing not yet in existence eyn adam makneh davar
she-lo ba le-'olam. R. Johanan' s reason is that ' it is in his hand' , i.e.
it lies in the power of the farmer to detach the produce from the
soil. Consequently, although the detached produce is not yet in
existence it can easily come into existence, depending as it does
solely on the act of the person concerned, the farmer. R. Johanan' s
formulation is: ' Whatever is now in his hand [in his power to do]
is not treated as something which requires an act [to bring it into
being].' R. Assi seeks to refute R. Johanan' s argument from our
Mishnah. The Mishnah rules in all the cases it considers that the
betrothal is invalid because all the cases are those of davar she-lo ba
le-'olam. But one of the cases is that of the Gentile who wishes the
betrothal to take effect when he becomes converted to Judaism.
Surely, objects R. Assi, this lies entirely in his own power and
according to R. Johanan the betrothal ought to be valid. The reply
given on R. Johanan' s behalf is that R. Hiyya bar Abba has stated
in the name of R. Johanan himself that a Court composed of three
Jews is required for a conversion to be valid. Hence it does not lie
in the power of the Gentile to become converted. He requires the
cooperation of the three who will constitute the Court to accept him
and their cooperation does not depend on him but on their own
choice.
The comment (Soncino, p. 313, note 3) that 'the answer given here shews
that one encountered real difficulties before he could be converted, and
was often denied it altogether' is way off the mark. The Talmud says
nothing at all here about the difficulties of conversion to Judaism. All the
Talmud means is that since a Court has first to be constituted it does not
lie in the power of the Gentile alone to become converted to Judaism and
hence the case in the Mishnah is not analogous to that of separating
terumah, where it depends solely on the farmer and on no other to detach
the produce from the soil.
An objection is now raised by the Palestinian Amora, R. Abba
bar Memel (third to fourth century). R. Johanan has stated that
the act is effective, even where the thing is not yet in existence, if
it lies in the power of the person concerned to bring the thing into
existence and so produce the effect. In that case it should follow
that if the owner of a slave-girl betroths her, the betrothal to take
effect after he has granted her her manumission, the betrothal
68 The Talmudic argument
should be valid (i.e. and it is notorious that it is, in fact, invalid).
In other words, if R. Johanan is correct, the rule in the Mishnah
would only apply to the man who betrothed a slave-girl owned by
another. Where he himself was the owner the betrothal ought to
be effective since it depends entirely on him to grant her her
freedom and so change her status to one in which she can be
betrothed. To this the Talmud replies: ' In that case there is an
animal whereas now there is another's consent.' The meaning of
this is, the slave-girl's status, while she is still a slave is, vis-a-vis the
owner, that of an ' animal' , i.e. she has no powers of consent, she
is owned by her master, whereas after her manumission she
becomes in law a fully fledged human being with independent
powers of consent. Consequently, although it does lie in the power
of her owner to free her, it is still considered to be a case of davar
she-lo ba le-'olam, since her new status will be so entirely different
from her present status (i.e. unlike the status of the produce which
lacks only detachment from the soil and which, apart from this, has
the same status now as later).
A saying of the third-century Palestinian Amora, R. Oshea, is
now quoted. R. Oshea said that if a man betroths his own wife, the
betrothal to take effect after he has divorced her, the betrothal is
invalid. But according to R. Johanan's reasoning it ought to be
effective since the husband does have the power to divorce his wife
(i.e. and it is different from the case of the slave-girl in that the wife
now has the same powers of consent that she will have later so that
no violent change of status is involved as it is in the case of the
slave-girl). To this the reply is given that although the husband does
have it in his power to divorce his wife, he does not have it in his
power to make her consent to the betrothal once he has divorced
her. No woman can be betrothed without her consent. Here it is
a case of davar she-lo ba le-'olam, since the desired effect cannot be
produced by the husband alone, unlike the case of the produce
where it depends entirely on the farmer himself. (The Talmud
could have given this reply to the earlier question regarding the
slave-girl. It is not in his power to make her consent to the
subsequent betrothal. Evidently, at first the Talmud thought that,
since the act of manumission depends solely on the owner, the need
for her subsequent consent to the betrothal does not matter since
the non-existence of the thing has, as it were, been weakened. Now
that the Talmud does introduce this idea that the woman's consent
Conveyance of a thing not yet in existence 69
is required, the same reply can be given in the case of the slave-girl;
see the Tosafists s.v. nehi de-veyado.) In that case, the Talmud
continues, we can solve a problem set by the same R. Oshea. He
set the following problem. The smallest coin to effect a betrothal,
i.e. given as a token for the act of betrothal, is a perutah. Suppose
a man gives two perutot to a single woman, declaring that he
betroths her now with one of these and with the other after he will
have divorced her. Is the second betrothal valid? But we have just
explained that R. Johanan agrees that where the effect depends on
the woman's consent it is a case of davar she-lo ba le-'olam. Why, then,
is it treated by R. Oshea as problematical? The Talmud replies that
the reason why R. Oshea sees the matter as problematical is that
where a husband wishes to betroth his own wife, the betrothal to
take effect after her divorce, he is seeking to bring about now an
effect that is meaningless in law. He is seeking to have a betrothal,
albeit a subsequent one, take effect for a woman who is already
married to him and that is impossible. Here the wife's later status
is certainly a davar she-lo ba le-'olam. But in R. Oshea's case the
woman is single at the time of the betrothal and here it can possibly
be argued that, since she can be validly betrothed now, a betrothal
at this time, to take effect after the first betrothal and her divorce
from it, can be valid. We thus have three cases of davar she-lo ba
le-olam\
(1) R. Johanan: terumah: certainly valid, because it lies solely in his
power
(2) Betrothal of the man's own wife: certainly invalid, because she may
not consent later and cannot now be betrothed
(3) Betrothal of single woman: R. Oshea uncertain, because he can
betroth her now
A Baraita is now quoted in support of R. Johanan. This Baraita
reads:' Terumah must not be separated from produce detached from
the soil for produce attached to the soil and if it had been separated
it is not terumah. How so? If he declared: " Let the detached produce
of this furrow be terumah for the attached produce of that furrow",
or if he declares: " Let the attached produce of this furrow be terumah
for the detached produce of that furrow", he has said nothing at
all. But if he declares: [let it take effect] " When it will be detached "
and it was later detached, his declaration stands. R. Eliezer b.
Jacob went even further. Even if he declares: "Let the detached
produce of this furrow be terumah for the attached produce of that
70 The Talmudic argument
furrow" or: "Let the attached produce of this furrow be terumah for
the detached produce of that furrow [adding in both instances]
when the produce will have grown to a third of its full height" and
it did grow to a third and was then detached, his declaration
stands.' The first part of the Baraita thus supports R. Johanan,
stating quite explicitly that if the farmer declares that it should be
terumah when it is detached it is terumah, i.e. because it is in his power,
as R. Johanan says, to detach it. The second part of the Baraita,
the ruling of R. Eliezer b. Jacob, goes further. Produce is only liable
to tithing when it has grown to at least a third of its full maturity
and then detached from the soil. Now while the farmer does have
it in his power to detach the produce from the soil, it does not lie
in his power to make it grow to the requisite height. Why, then,
does R. Eliezer b. Jacob rule that even here it is terumah? It can
only be because R. Eliezer b. Jacob holds, as Rashi observes and
as can be seen from the discussion later on in our sugya, that a transfer
can be effected even if the thing is not yet in existence - adam makneh
davar she-lo ba le-'olam. Thus the first opinion in the Baraita is that
a man cannot transfer a davar she-lo ba le-olam except where, as in
the case of the Baraita and as R. Johanan explains it, it lies in his
power to produce the required effect. But R. Eliezer holds that in
all circumstances adam makneh davar she-lo ba le-'olam.
The puzzling phenomenon that R. Assi's and R. Johanan's formulation is
in virtually the same words as the Baraita has, in fact, numerous parallels
in the Babylonian Talmud, where an Amora makes a statement and it is
then said that the same rule is found in a Baraita and, so it generally
appears, with exactly the same wording. I. H. Weiss, Dor, vol. n, pp. 242-4,
goes so far as to suggest that these Baraitot are not authentic and have been
invented or, if they are, the Amoraim have plagiarised them. Abraham
Weiss, le-Heker ha-Talmud, pp. 35-63, takes strong issue with I. H. Weiss,
arguing that the phenomenon is best explained on the grounds of literary
transmission and its methods. Sayings of earlier teachers were repeated so
often that when the later Amoraim came to render their decisions they
would naturally do so in the form in which they had become accustomed
to hearing these teachings. See my article ' Are there fictitious Baraitot in
the Babylonian Talmud?' in HUCA, 42 (1971), 185-96. If this is correct,
R. Assi is not here presenting for R. Johanan's consideration an entirely
new problem he had made up out of his own head. His use of the
expressions found in the Baraita shows that R. Assi is really asking
R. Johanan whether he, R. Johanan, is aware of an earlier distinction that
can be made and R. Johanan replies that he is and gives the reason as
'because it lies in his power'. This does not rule out the possibility that
the formulation is not R. Assi's own but has been put into his mouth by
the editors on the basis of the Baraita.
Conveyance of a thing not yet in existence 71
There now follows a debate between the early-fourth-century
Babylonian teachers Rabbah and R. Joseph. There are two stages
in the growth of produce before it reaches the third-of-its-growth
stage. The first of these is called agam (interpreted as ' bending' ),
that is, the first sprouting of the crops when there is sufficient length
in the blade to bend it towards its root. The agam stage is merely
that of the earliest sprouting. The second stage is called shahat,
'fodder'. The blades are sufficiently long to be suitable when
removed for cattle, but are not yet at the stage of a third-
of-its-growth. Rabbah argues that even R. Eliezer b. Jacob, who
holds that the terumah separation is valid, only does so when it has
reached the shahat stage at least. Where it has only reached the agam
stage even R. Eliezer b. Jacob will agree that it is not terumah. But
R. Joseph argues that R. Eliezer b. Jacob holds that the terumah is
valid even at the agam stage. The point of this debate is that even
those authorities, such as R. Eliezer b. Jacob, who hold that adam
makneh davar she-lo ba le-'olam, only do so where the thing is, at least,
partially in existence. No one holds that something can take effect
on a totally non-existent thing. If, for instance, a farmer declared
that he wishes his terumah to be for produce he will not even have,
let alone harvest, until much later, it is obvious that the separation
has no validity. Thus there must be some stage of partial existence
even according to R. Eliezer b. Jacob and the debate between
Rabbah and R. Joseph is whether or not the agam stage is to be
considered partially existent for the purpose of the law, or totally
non-existent.
Now the Mishnah states that when a man declares to his
neighbour:' If your wife will give birth to a girl let her be betrothed
unto me' the betrothal is invalid, since if the neighbour's wife is
not even pregnant the girl who was later born did not enjoy even
partial existence at the time of the declaration. The second-century
Palestinian Amora, R. Hanina, explains that if the neighbour's wife
was pregnant at the time, the betrothal is valid (because this clause
of the Mishnah follows the opinion of R. Eliezer b. Jacob so that
where there is partial existence now it is valid). The Talmud
comments that according to Rabbah, R. Hanina's comment refers
only to where the neighbour's wife's pregnancy was discernible at
the time (just as only the shahat stage suffices in the case of terumah)
whereas according to R. Joseph the betrothal is valid, if the wife
is pregnant, even if her pregnancy was not discernible (which
corresponds to the agam stage).
72 The Talmudic argument
A different version of the debate between Rabbah and R. Joseph
is now quoted. According to this version all agree that the
separation of the terumah is only valid, even according to R. Eliezer
b. Jacob, if the produce has reached the shahat, not the agam, stage.
The debate between Rabbah and R.Joseph concerns another
matter. Rabbah holds that even if it has reached the shahat stage
it is only valid if the field is one watered by the rain and requiring
no artificial irrigation. The crops of such a field do not generally
fail and here, at the shahat stage, the produce can be said to enjoy
partial existence. But the crops of a field requiring artificial
irrigation do frequently fail and here it would be invalid even at
the shahat stage. R. Joseph disagrees, holding that no distinction
must be made between the two kinds of field. Consequently, the
Talmud concludes, R. Hanina's explanation of the Mishnah must
refer to where the wife's pregnancy is discernible (since both
Rabbah and R. Joseph are now said to require the shahat stage and
reject the agam stage), but then the betrothal is valid according to
both Rabbah and R. Joseph since a pregnancy takes its course by
normal means and is analogous to the field watered by the rain and
requiring no artificial methods of irrigation for its growth.
The sugya concludes with a statement of Abbaye and a discussion
on this. Abbaye observes that there are three Tannaim who hold
that a man can effect a transfer of something not yet in existence
[adam makneh davar she-lo ba le-'olam). The three Tannaim are: R.
Eliezer b. Jacob (as above); Rabbi (the name given to R. Judah
the Prince); and R. Meir. Rabbi's statement occurs in the following
context. The verse states: ' Thou shalt not deliver unto his master
the servant' (Deuteronomy 23: 15; The Torah: Deuteronomy 23:
16). Rabbi applies this verse to a man who buys a slave from his
master on the condition that he will set him free and R. Nahman
b. Isaac explains this to mean that the buyer states, at the time of
the purchase, 'When I buy you, you belong to yourself from now.'
Thus it follows that Rabbi holds adam makneh davar she-lo ba le-'olam,
since when the man buys the slave he is still a slave and yet his
manumission becomes effective from that time. As for R. Meir, a
Baraita is quoted in which it is said that R. Meir holds, in all the
cases referred to in our Mishnah (betrothal to take place after
conversion, after manumission, after the death of the husband, after
the death of the wife's sister), that the betrothal is valid. Thus
Conveyance of a thing not yet in existence 73
R. Meir, too, holds that adam makneh davar she-lo ba le-'olam. In the
same Baraita it is stated that R. Judah the Prince holds that in all
these cases the betrothal is valid according to Biblical law but that
in the case of the husband and wife's sister the Sages refused to
countenance the betrothal since to do so would lead to 'enmity',
i.e. the husband would become aware that his wife is looking
forward to his death and the sister that her husband wishes her to
die. The Talmud asks: In that case why does Abbaye refer only
to R. Meir in this Baraita and not to R. Judah the Prince? The
obvious answer is given, Abbaye has already referred to Rabbi, in
the other case, and he is R. Judah the Prince.
The Talmud now asks why Abbaye does not list the opinion of
R. Akiba, who also appears to hold the same opinion as the other
three in his list. R. Akiba's statement is found in a Mishnah
(Nedarim 11: 4). A husband can annul the vows of his wife
(Numbers 30). But, the Mishnah rules, if she vows that the work
of her hands is forbidden to him there is no need for him to annul
the vow since it is invalid in any event. She has no power to impose
a vow on the work of her hands since she is obliged to work for her
husband by the terms of her marriage contract. She cannot impose
a prohibition on something that is not hers. R. Akiba holds that
annulment is required since, otherwise, she may provide him by
her work with more than that to which he is entitled by law and
on this the vow can take effect. Hence, we see that according to
R. Akiba a vow can take effect on the work still to be done and this
is davar she-lo ba le-'olam. The reply is given in the name of R. Huna
son of R. Joshua that the Mishnah deals with a wife who declares
that her hands be dedicated to their Maker, i.e. it is not the work
still to be done upon which she imposes her vow but on the hands
themselves. Hence it is not at all a case of davar she-lo ba le-
l
olam and
this is why R. Akiba is not mentioned in Abbaye's list of authorities.
Rashi, s.v.ke-man, holds that the ruling follows R. Eliezer b. Jacob that adam
makneh davar she-lo ba le-'olam but the Tosqfists, s.v. ve-amar R. Hanina, hold
that the ruling is the opposite. In the case discussed in the Middle Ages
and later of a man selling his portion in Paradise, the sale is declared to
be invalid in law because, among other reasons, it is a sale of davar she-lo
ba le-'olam, see Responsa Meshiv Davar by N.Z.J.Berlin, part in, no. 14.
The relevant material on davar she-lo ba le-olam is given in ET
i
vol. vn,
pp. 30-67 and in Kesef Nivhar, no. 37, pp. 58a-63b. Kesef Nivhar observes
that the Talmud in general, when dealing with this subject, speaks of three
74 The Talmudic argument
cases: (i) Where the object to be transferred is not yet in existence; (2)
Where the one to whom it is to be transferred is not yet in existence; (3)
Where a condition is made to depend on something not yet in existence.
The main Talmudic passages, in addition to ours, are: Tevamot 92b93b;
Ketubot 58b~59b; Nedarim 85a-86b; Gittin 13b; 42b-43a; Kiddushin 78b;
Bava Metiz'a 16b; 66b; Bava Batra 63ab; 7ga-b; 141b-143a; I57a-b.
7
Kol she-eyno be-zeh ahar ieu ajilu be-vat ahat eyno:
whatever cannot be established in a
consecutive sequence cannot be established
even in a simultaneous sequence
This is a comparatively brief sugya, Kiddushin 50b51 a. The meaning
of the central term is: where the law is insistent that a certain
effect - B - cannot follow another effect - A - then it implies that
A and B cannot take effect even when they occur simultaneously.
Since the taking effect of A frustrates the taking effect of B and the
taking effect of B frustrates the taking effect of A, this frustration
operates not only when one follows the others but even when both
occur together, even though if each were to occur on its own it
would take effect.
The sugya is here appended to a Mishnah which reads: ' If a man
betroths a woman and her daughter or a woman and her sister
simultaneously, they are not betrothed' (i.e. the betrothal is invalid
and no get, 'bill of divorce', is required before either can marry
another). A man cannot betroth his wife's daughter, mother or
sister and if he did the act of betrothal is invalid (Leviticus 18:
17-18). This is stated explicitly in the Torah. But the Mishnah goes
further to rule that if he betroths a woman and her daughter or
two sisters, the betrothal to take effect simultaneously, it is still
invalid. The Talmud asks how do we know this law, to which the
fourth-century Babylonian Amora, Rmai bar Hama, replies, it is
because Scripture states: 'Neither shalt thou take a wife to her
sister, to vex her' (Leviticus 18: 18). Rami bar Hama understands
this to mean that if a man takes (i.e. betroths) a woman and her
sister simultaneously (the case in the Mishnah) it is invalid -
'neither shalt thou take'. As Rami bar Hama puts it: ' The Torah
said: When they become rivals to one another [i.e. in a simultaneous
betrothal] he can have no taking of any one of them.' The rule of
the Mishnah has thus Scriptural authority according to Rami bar
75
76 The Talmudic argument
Hama. Rava objects that the verse cannot possibly mean what
Rami bar Hama says it means. The verse does not refer at all to
a simultaneous betrothal but to a man who takes (i.e. has intercourse
with) the sister of a woman he had previously married. That this must
be correct Rava proves from the verse which occurs at the end of
the Levitical list of prohibitions: ' the souls that commit them shall
be cut off from among their people' (Leviticus 18: 29). This shows
that we are dealing with a man who has intercourse with his wife's
sister. For if we are dealing, as Rami bar Hama holds, with a
simultaneous betrothal, which the Torah declares to be invalid,
there would be no penalty at all if he cohabited with either or, for
that matter, with both, since neither is his wife's sister.
Rava, thus dissatisfied with Rami bar Hama's proof, proceeds
to advance a different reason for the rule in the Mishnah. Rava
argues that the Mishnah is best understood on the basis of his
master's principle, that is of Rabbah, who lays down the rule: kol
she-eyno be-zeh ahar zeh qfilu be-vat ahat eyno, ' What ever cannot be
established in a consecutive sequence cannot be established even
in a simultaneous sequence.' This is a general legal principle laid
down by Rabbah and is to be applied to the case in the Mishnah.
Since, according to the Biblical law, there is no validity to the
betrothal of a wife's sister, it must follow, on Rabbah's principle,
that, as the Mishnah states, there is no validity to the simultaneous
betrothal of two sisters. If a man first betroths sister A and subsequently
betroths sister B, the betrothal of sister B is invalid. Consequently,
on Rabbah's principle, the betrothal of A and B at the same time
has no validity.
The sugya now proceeds to examine Rabbah's principle on its
own. This is prefaced with the termguja,' the main statement', used
in the Talmud whenever a statement referred to incidentally in the
course of a discussion is then examined on its own. Abbaye raises
an objection from a Baraita (Tosefta, Demai 8: 13). This Baraita deals
with the laws of tithing produce. The tithes are: terumah, a 40th,
50th or 60th, given to the priest; ma'aser rishon, ' the first tithe', a
tenth of the remainder of the produce, given to a Levite; ma'aser
sheni, ' the second tithe', a tenth of the remainder, it or its value,
to be taken to Jerusalem and consumed there. Any produce from
which the tithes have not been separated is known as tevel,' untithed
produce', and is forbidden to be eaten. It is the separation of the tithes
which removes the prohibition from tevel, not the actual giving to
Relation of a simultaneous to a consecutive sequence 77
the priest or Levite. Terumah may only be eaten by a priest but
ma'aser rishon, though it must be given to the Levite, may be eaten
by a non-Levite. Once ma'aser rishon has been separated it is no
longer level, i.e. the Levite is not obliged to separate from it ma'aser
sheni, though he is obliged to separate a tenth for the priest, known
as terumat ma'aser, ' t he t erumah given from ma'ser\ The Baraita
deals with the case of a farmer who separated more than a tenth for
ma'aser rishon and the Baraita rules that if he did this the remainder
of the produce is considered to be tithed, and may be eaten once
ma'aser sheni has been separated, but the amount separated as
ma'aser rishon is not treated as ma'aser rishon. The Levite may not
eat of it, even after he has separated terumat ma'aser, since there is
more than a tenth of the farmer's produce and this surplus is level.
The Levite has now in his possession an admixture of ma'aser rishon,
which he may eat, and level, which he may not eat. The farmer (or,
possibly, the Levite) must now separate as ma'aser sheni an amount
proportionate to the surplus in the hands of the Levite, in addition
to the amount he separates as ma'aser sheni to free his own produce
from the prohibition of level. The same rule would apply if the
farmer separated more than a tenth for ma'aser sheni, there would
be an admixture of ma'aser sheni and ordinary food but here there
would be no practical difference since ordinary food may, of course,
be eaten in Jerusalem and, once ma'aser sheni has been separated,
there is no longer any prohibition of level.
See Meiri for a further discussion of what the Levite is to do with the
admixture and that some hold that there is no remedy at all for the
admixture, not as we have explained it here.
In any event, we see from the Baraita that the tithing is effective
so far as the remainder of the produce is concerned. But, Abbaye
objects, if Rabbah is correct, the tithing should be null and void.
If a man separates an exact tenth as ma'aser rishon and then
separates some more of the produce for the same ma'aser it is obvious
that the second separation has no validity. Once the ma'aser has
been separated there is no meaning to the separation of a further
quantity for the same purpose. Consequently, tithing is an effect
that cannot take place consecutively and, according to Rabbah,
it cannot therefore take place simultaneously. Since this is so why
is the amount separated, too large for ma'aser, treated as ma'aser at
all? To this the reply is given that ma'aser is in a totally different
78 The Talmudic argument
category. A man can separate his ma'aser by giving half of the grains.
Consequently, when the ma'aser has been separated together with
a surplus it is incorrect to see this as the simultaneous separation
of two amounts of ma'aser so that Rabbah's principle would come
into operation. All that is, in fact, separated is a single tithe cutting
across the grains, as it were, the remainder being the surplus. But
where a man betroths two sisters it is absurd to say that there is
only a single betrothal of two halves of the sisters. Here, indeed,
there is the simultaneous betrothal of two and here Rabbah's
principle comes into operation.
The Talmud objects that this reply seems to be contradicted by
Rabbah himself in a ruling of his regarding the tithing of cattle.
Cattle is tithed in the following way. Tha animals are passed one
by one through a narrow gate, the farmer counting from one to
ten. Each tenth animal to pass through the gate is marked and is
later offered as a sacrifice. If the farmer counted, in error, the tenth
animal as the ninth and the eleventh as the tenth, the law is that
the eleventh animal is not ma'aser ('animal tithe'), since it is not,
in fact, the tenth, but is still sacred and has to be offered as shelamin,
' a peace-offering'. Rabbah's ruling concerns two animals that
came through the gate together, after the farmer had counted nine,
and he declared them both to be ma'aser. One of these, rules
Rabbah, is ma'aser and the other shelamin. Since they both came out
together after the ninth, we cannot, in fact, determine which is
ma'aser and which shelamin, so the following procedure is adopted.
The animals are left alone until they become blemished and as
such no longer fit for sacrifice. A sum of money is then taken and
the sanctity of the blemished shelamin is redeemed with this sum,
the money being used to buy another animal to be offered as
shelamin. As for the ma'aser, no redemption is required, the rule being
that a blemished ma'aser animal loses its sanctity and may be eaten
without redemption. From Rabbah's ruling here we see that when
two animals both come through the gate after the farmer has
counted nine and where he declared both to be ma'aser, one of them
does become ma'aser. But since ma'aser cannot be effected
consecutively (i.e. if ten had been counted and the tenth animal
became ma'aser there would obviously be no significance to the act
if the farmer declares the eleventh one, too, to be ma'aser), it should
not take effect simultaneously (i.e. and here the previous reply is
inapplicable, since half animals cannot be brought as sacrifices).
Relation of a simultaneous to a consecutive sequence 79
The reply to this is given by quoting a Mishnah (Bekhorot 9: 8) in
which the rule is stated that if the farmer, in error, counted the tenth
animal as the ninth (i.e. imagining it to be the ninth when in reality
it was the tenth) it becomes ma'aser. We see from this that mctaser
can take effect even in error and hence it can take effect simul-
taneously. The meaning here would seem to be, since mctaser can so
easily be effected (even in error), that Rabbah' s principle does not
apply. Here even though it cannot take effect consecutively it can
still take effect simultaneously. Rabbah' s principle does not mean
that because the effect cannot take place consecutively therefore it
cannot take place even simultaneously. Rabbah' s principle only
means that since it cannot take place consecutively this is an
indication that it does not take place even simultaneously. Thus there
can be exceptions to Rabbah' s principle and ma'aser, which can take
effect so readily, even in error, is such an exception. But betrothal
cannot be effected in error and hence, in the case of the two sisters,
Rabbah' s principle does apply.
So far we have managed to cope with two objections to Rabbah' s
principle. The principle does not operate, the Talmud has suggested
in reply, in the case of ma'aser of produce, because half-grains can
become ma'aser and in the case of animal ma'aser because it can
become effective even in error. It follows that if it can be shown
that there is a case in which the effect cannot come about either
in error or by halves and there is, none the less, a ruling in such
a case that it can take place simultaneously, even though it cannot
take place consecutively, Rabbah will have been refuted. The
Talmud now proceeds to demonstrate that there is such a case.
The case is that of a thanksgiving offering. Together with the
animal sacrifice 40 loaves have to be brought. As soon as the animal
is slaughtered in the Temple the loaves become automatically
sanctified as thanksgiving loaves and can then only be eaten by the
priests. The sanctification of the loaves cannot take effect by halves,
i.e. 80 half-loaves cannot be sanctified, nor can it be in error, i.e.
if at the time of the sacrifice the intention was for 40 dark loaves
to be sanctified and the loaves there were, in fact, light coloured,
these loaves have no sanctity. There is a debate between the early
third-century Palestinian Amoraim, Hezekiah and R. Johanan,
regarding a thanksgiving offering that was sacrificed together with
80 loaves. Hezekiah holds that 40 of these do become sanctified but
R. Johanan holds that none of the loaves become sanctified. Now
80 The Talmudic argument
it is obvious that if 40 loaves had first been sanctified and it was
then declared that another 40 be sanctified, the second 40 have no
sanctification. How can they have any sanctity, since there is no
longer any thanksgiving offering to sanctify them? Since this is so,
Rabbah's principle ought to operate. Since the second 40 cannot
become sanctified after the first 40, there ought to be no sanctification
of any of the loaves where all 80 are brought together. In other
words it would appear from the debate between Hezekiah and
R. Johanan that only the latter agrees with Rabbah's principle and
this is unlikely. (Rashi adds that since Hezekiah is the teacher of
R. Johanan it is unlikely, if he rejects Rabbah's principle, that
Rabbah would still maintain it merely because of the opinion of
the disciple, R. Johanan.)
To this the reply is given that the debate between Hezekiah and
R. Johanan is not at all what it appears to be on the surface. The
third-century Palestinian Amora, R. Zera, has, in fact, explained,
in any event, that if the man intended only 40 of the 80 loaves to
be sanctified they are sanctified and on this there can be no
disagreement. Conversely, if the man's intention was to sanctify all
80 loaves there can be no disagreement that, as Rabbah's principle
states, there is no sanctification whatsoever, not even of 40 of the
loaves. The debate is only where the man brought 80 loaves but
did not reveal his intention in bringing 80 instead of 40. Hezekiah
holds in this case that the man's intention was to bring only 40
loaves to be sanctified; the reason why he brought, in fact, 80 was
solely in order to have 40 loaves to fall back on if the first 40 were
lost or destroyed. R. Johanan, on the other hand, holds that, since
the man brought 80 loaves, we are justified in assuming that his
intention was for all 80 to be sanctified and this is ineffective
because of Rabbah's principle.
The sugya now reverts to the original attempts at explaining the
Mishnah. Rava, disagreeing with Rami bar Hama's attempt,
explains the Mishnaic rule on the basis of Rabbah's principle. Now
in the following sugya (Kiddushin 51a) there is a debate between
Abbaye and Rava regarding the case of a man who betroths one
of two sisters, that is to say, he declares: ' One of you be betrothed
to me' without specifying any particular one. Abbaye holds that
in this case the betrothal is valid in the sense that neither of the
sisters can marry another without a get from the man who betrothed
one of them. Rava disagrees, holding the betrothal to be invalid
Relation of a simultaneous to a consecutive sequence 81
because it i s: ' a betrothal that cannot lead to intercourse' (kiddushin
she-eyn mesurin le-viah). The purpose of betrothal is to lead eventually
to intercourse so that, according to Rava, it is reasonable to
conclude that any betrothal which by its very nature precludes
intercourse (as here, where, if the betrothal is valid, he will not be
allowed to have intercourse with either since she may be his wife's
sister) is bound to be invalid. Since this is Rava's view, the Talmud
asks, why does he have to explain the ruling in the Mishnah on
the basis of Rabbah' s principle? The Mishnah is perfectly intelli-
gible on the basis of Rava's own principle. If, as Rava holds, the
betrothal of even one of two sisters is invalid, the betrothal of both
is invalid a fortiori. The reply given is that, indeed, Rava can explain
the Mishnah perfectly well on his own premiss. He only invokes
Rabbah' s principle for the benefit of Rami bar Hama who, by
seeking a Scriptural verse for the Mishnaic rule, evidently does not
agree with Rava's principle. Rava is saying, in so many words, ' I
personally have no difficulty in explaining the Mishnah but you
evidently do not agree with me and do have a difficulty. The only
way out of the difficulty for you is not to rely on the verse you quote,
which you fail to understand, but to rely on Rabbah' s principle.'
Rabbah's principle is referred to in only two other Talmudic passages:
'Eruvin 49a~5oa and JVedarim 69b. The 'Eruvin passage contains, with only
very minor variants, the whole discussion, as in our passage, from gufa to
the debate between Hezekiah and R. Johanan inclusive. This section is also
prefaced by gufa. This is important for the literary analysis of the Talmud
in that it clearly demonstrates that the editors had before them this whole
section as a complete unit, which they then inserted in Kiddushin and
'Eruvin, in both instances after there had been an incidental reference to
Rabbah's principle. The Nedarim is very brief and deals with a husband's
annulment of his wife's vow (Numbers 30). The annulment by a husband
of his wife's vow is effective if he performs the annulment within 24 hours
of hearing of the vow. This is known as hafarah, 'annulment'. If, however,
when he hears of the vow, the husband declares that he wishes the vow
to stand, he no longer has the power of annulment. This is known as
hakamah, 'establishment (of the vow)'. A problem set by Rabbah in
Nedarim is: what is the law if a husband declared that his intention was
for the vow both to be annulled and stand, i.e. he declared: 'Let there
be hakamah and hafarah simultaneously'? The Talmud states that the
problem can be solved on the basis of Rabbah's own principle of kol she-eyno
be-zeh ahar zeh afilu be-vat ahat eyno. There can be no hakamah after hafarah
(since the hafarah has cancelled out the vow, there is no longer any vow
to be established) and there can be no hafarah after hakamah (since that is
precisely what hakamah means, that the vow stands and can no longer be
82 The Talmudic argument
annulled). Consequently, Rabbah's principle applies. Since the two cannot
take effect consecutively they cannot take effect simultaneously. On the
face of it, and as the Ran ad loc. holds, this means that neither is effective
and it is as if the husband had said nothing at all. The vow has not been
annulled but neither has it been established and the husband, within 24
hours, has the right either to establish it or to annul it. But Maimonides
[Tad, Nedarim 13: 22) rules that the hakamah stands but not the hafarah.
The Ran and other commentators find Maimonides' ruling very puzzling.
Since neither hakamah nor hafarah can take effect after the other has taken
effect, then, on Rabbah's principle which the Talmud applies to the case,
neither should take effect when they occur simultaneously. R. Hayyim
Soloveitchik, Hiddushey Rabbenu Hayyim ha-Levi, Nedarim, pp. 37 ab, very
ingeniously suggests that, in Maimonides' understanding, the Talmudic
application of Rabbah's principle only refers to hafarah not to hakamah. The
argument is as follows. The reason why there can be no hafarah after
hakamah is because that is precisely what hakamah signifies, the rejection of
any further possibility of hafarah. Hence Rabbah's principle applies. Since
hafarah cannot take effect after hakamah it cannot take effect together with
hakamah. The reason why there can be hakamah after hafarah, on the other
hand, is not because in essence hafarah prevents hakamah from taking effect
but simply because once hafarah has taken effect there is no longer any vow.
Consequently, here Rabbah's principle does not operate since, in essence,
there can be hakamah after hafarah, i.e. there is no basic principle that
hafarah itself frustrates hakamah. Therefore, according to Maimonides, the
hakamah does stand since with regard to hakamah Rabbah's principle does
not apply. Rabbi A. I. Karelitz, in his notes to Hiddushey Rabbenu Hayyim
ha-Levi, entitled Hazon Ish (end of volume, pp. 8-9), finds R. Hayyim's
distinction contradicted by our sugya. In the case ofma'aser the reason the
second separation of the same ma'aser is ineffective is not because of the
first separation in itself but simply because once ma'aser has been separated
the produce is no longer subject to tithes! It is possible that Maimonides
is making a rather different distinction. In all the cases considered in our
sugya of A followed by B there is no difference whatsoever between A and
B, except that A happens to be first. For instance, in the case of the two
sisters, the betrothal of sister B cannot take place after the betrothal of sister
A but if the betrothal of sister B had taken place first exactly the same
obstacle would have been provided to the betrothal of sister A. Similarly,
in the case ofma'aser the first tenth prevents the second from taking effect,
but only because the first tenth is the first tenth and similarly with regard
to the animal tithe and the 40 loaves. Hence here B frustrates A as much
as A frustrates B and hence Rabbah's principle applies. The two frustrate
one another and so cannot take effect simultaneously. But in the Nedarim
case the difference between hakamah and hafarah is not only which came
first. The function of the two is different. The function of hakamah is to
frustrate hafarah whereas the function of hafarah is, as R. Hayyim notes,
simply to remove the vow. It follows that it is the hakamah that frustrates
the hafarah, not the other way round. Consequently, Rabbah's principle
does apply to hafarah but not to hakamah as Maimonides rules.
8
Tesh horesh telem ehad: a single act of
ploughing can result in a number of penalties
The main sugya is Makkot 2ib-22a, though there is a reference
also in Pesahim 47a-b. The general principle is that an offence
against a Biblical, negative precept renders the offender liable to
a flogging. Even if a single act is performed but the act offends
against a number of negative precepts, there is a separate penalty
of flogging for each offence.
In the Mishnah, to which the sugya is appended, it is stated that
it is possible for a man to plough a single furrow (yesh horesh telem
ehad) and yet be liable (to a flogging) for offending against eight
separate, negative precepts. The man ploughs with an ox and an
ass yoked together (Deuteronomy 22: 10) and they are sacred
animals dedicated to the Temple (Deuteronomy 15: 19). At the
time of ploughing he sows seeds in a vineyard (Deuteronomy 22:
9), i.e. as the Talmud explains, while ploughing he covers the seeds
with the earth the plough throws up, which falls under the
definition of'sowing'. It is, moreover, the Sabbatical year in which
it is forbidden to sow (Leviticus 25:4) and it is a Festival on which
work is forbidden (Leviticus 23). The plougher is in a field in which
corpses are buried and he thus contaminates himself through
contact with a grave and he is a priest (Leviticus 21: 13) and a
Nazirite (Numbers 6) to whom such contact is forbidden. The eight
prohibitions are thus: (1) ploughing with an ox and an ass; (2)
ploughing with a sacred ox; (3) ploughing with a sacred ass; (4)
sowing mixed seeds in a vineyard; (5) sowing on the Sabbatical
year; (6) sowing on the Festival; (7) a priest contaminating
himself; (8) a Nazirite contaminating himself. R. Hanania b.
Hakhinai adds a ninth prohibition: the plougher is wearing at the
time a garment of wool and linen (Deuteronomy 22: 11). The Sages
83
84 The Talmudic argument
refuse to count this since it is an offence totally unconnected with
the ploughing. But H. Hanania retorts that neither have the
offences of the priest and Nazirite anything to do with ploughing.
The distinction made, none the less, by the Sages is obvious. The
prohibition of wearing the garment has nothing whatsoever to do
with the act of ploughing. Even if the man wore no garment at all
his ploughing would be unaffected. But although the prohibition
of contamination in itself has nothing to do with his ploughing, yet
were it not for the ploughing he would not have entered the place
in which he suffers contamination. The sugya opens with a report
by the late-second, early-third-century Palestinian Amora,
R. Jannai, that a group of Sages had met and decided that there
is a penalty of flogging for one who covers mixed seeds in a vineyard
with earth, as this constitutes 'sowing'. R. Johanan objects that
there is no need for R. Jannai's statement in the name of the group.
The ruling can be inferred from the Mishnah, one of whose
prohibitions is that of ploughing in a vineyard, which can only be
on the grounds of sowing', as above. R. Jannai replies that, indeed,
this ruling can be inferred from the Mishnah but he claims that
it is his report that has provided R. Johanan with the clue. As
R. Jannai is said to have put it: ' If I had not raised the shard for
you, you would not have discovered the pearl beneath it.' Resh
Lakish demurs, that, were it not for the fact that ' a great man' ,
R. Jannai, had praised R.Johanan' s perspicacity in inferring it
from the Mishnah, it could have been argued that there is no proof
to the ruling from the Mishnah. A Baraita is quoted in which
R. Akiba holds that even to preserve mixed seeds in a vineyard,
e.g. by weeding the vineyard so that the seeds can grow there,
renders the perpetrator liable to a flogging. Consequently, the
Mishnah might follow the opinion of R. Akiba and the offence is
not because of 'sowing' but because of 'preserving'. If this is
correct, it can be argued, the Sages, who disagree with R. Akiba
and do not admit that there is any prohibition of' preserving', may
hold that neither is any prohibition of' sowing' incurred merely by
covering the seeds with earth. Hence, R. Jannai's report is required,
that even according to the Sages, who admit no prohibition of
'preserving', there is a prohibition on the grounds of'sowing'. The
Talmud rounds off this section by quoting a Baraita in which
R. Akiba supports his view by referring to a Biblical text.
Penalties resulting from single act 85
See Tosqfists Bava Metzi'a 17b s.v. i lav, that one does not normally find
pearls under a shard. The Tosqfists suggest that the reference is to a shard-like
stone at the bottom of the sea under which pearls are sometimes found and
refers to Bava Kama 91a:' You have dived into deep waters and have only
succeeded in bringing up a shard.' Possibly, however, the 'shard' is the
oyster shell and the meaning in Bava Kama is: ' You have dived into deep
waters to no avail because all you have brought up is a shell containing no
pearl' and the meaning here would, then, be: 'If I had not brought up
the shell you would not have discovered the pearl it contains.'
At this stage a discussion is introduced among fourth- and fifth-
century Amoraim. We have established that the act of covering the
seeds is considered to be 'sowing'. Now if it constitutes 'sowing'
with regard to seeds in a vineyard, it must also constitute 'sowing'
with regard to a Festival. In that case, Ulla asks R. Nahman, there
are nine prohibitions, not eight, since in addition to the offence of
ploughing on the Festival there is the offence of sowing on the
Festival. To this R. Nahman replies that the Tanna of the Mishnah
does not exhaust all the instances and could have included the
instance of sowing on the Festival but chose instead to omit this.
Ulla refuses to accept R. Nahman's reply. The Tanna in the
Mishnah states explicitly that there are eight prohibitions; eight,
surely, means eight, and not nine.
Rava (variant reading Rabbah) gives a different reply to the
question why the Mishnah does not count sowing on the Festival.
Rava argues that although with regard to the Sabbath there is a
separate penalty for each type of work, even when several of these
were carried out together, yet with regard to a Festival 'there is
no division of work', i.e. although two separate offences, that of
ploughing and of sowing have been committed, since these were
carried out at the same time, there is only one penalty, for working
on the Festival. To this Abbaye raises an objection from a Baraita.
The Baraita states that if a man on a Festival day boils the sciatic
sinew in milk and then eats it, he incurs five penalties of flogging:
for eating the sciatic sinew (Genesis 32: 32; The Torah: Genesis 32:
33); for cooking food unnecessarily on the Festival (i.e. only food
that can be eaten on the Festival may be cooked and the sinew may
not be eaten at all); for boiling meat and milk together (Exodus
23:19; 34:26; Deuteronomy 14: 21 are interpreted by the Rabbis
as prohibiting the boiling of any meat and milk together and the
eating of the result); for eating meat boiled in milk (as above); and
86 The Talmudic argument
for kindling a fire on the Festival. From this Baraita we see that there
is a separate penalty for cooking on the Festival and for kindling
a fire, which refutes Rava. To this the reply is given that the Baraita
must be emended. The fifth prohibition is not for kindling a fire
on the Festival but the Baraita should read instead that we are
dealing with one who cooks a sciatic sinew of carrion and the offence
is an additional one of eating, not of cooking, that of eating carrion
(Deuteronomy 14: 21). To this it is objected that R. Hiyya has
taught in a Baraita, commenting on the previous Baraita, that, of
the five prohibitions, three are for cooking and two for eating. But
if the Baraita is emended as suggested it emerges that there are only
two prohibitions for cooking and three for eating. In reply a different
emendation is suggested. Instead of kindling fire on a Festival the
Baraita should be emended to include kindling with wood from an
asherah (a tree sacred to idolatrous worship) of which Scripture
states: ' And there shall cleave nought of the cursed thing to thine
hand' (Deuteronomy 13: 17; The Torah: Deuteronomy 13: 18).
There are now, as R. Hiyya taught, three prohibitions for cooking
and only two for eating.
The Talmud continues with an objection raised by R. Aha son
of Rava to R. Ashi. Since at this stage of the argument we have
emended the Baraita so as to include the wood of an asherah, there
ought to be another penalty, that of' Neither shalt thou bring an
abomination into thine house' (Deuteronomy 7: 26) which makes
a total of six, not five. Hence still another emendation is suggested.
Instead of kindling fire on the Festival the Baraita deals with a fire
kindled with wood belonging to the Sanctuary, which it is forbidden
to destroy, since Scripture states: ' And burn their groves with fire
[but] Ye shall not do so unto the Lord your God' (Deuteronomy
12: 3-4).
It is perhaps worth noting that although the Amoraim mentioned
flourished over a very long period of time and must have presented their
arguments over this period, yet the whole passage hangs together as if it
had been compiled as a literary unit. First there is the objection, arising
out of the previous discussion regarding 'covering' as 'sowing', that there
ought, in that case, to be a further Festival prohibition. To this the reply
is given that the Tanna omits one. To which it is objected, if the Tanna
states that there are eight he must mean eight and no more than eight. To
which the reply is given, the Mishnah does not count Festival prohibitions
as two because there is no division of work on the Festivals. To this the
objection is raised, what of the Baraita of the sciatic sinew? (This is raised
Penalties resulting from single act 87
by Abbaye but only makes sense in response to Rava's interpretation
which, in turn, only makes sense in response to the earlier discussion.) To
this the reply is given, emend the Baraita to omit kindling fire on the
Festival and to include a further 'eating' prohibition, of carrion. To which
it is objected that R. Hiyya states that there are three prohibitions of
cooking not of eating. To this the reply is given, kindling is reinstated, but
not kindling fire on the Festival but kindling with wood of an asherah. At
a later date R. Aha son of Rava produces a fatal objection to this reply,
in that case there would be six, not five, prohibitions. To which the final
reply is given, that kindling is, indeed, reinstated, but not the kindling of
fire on the Festival nor the kindling of asherah wood but the kindling of
wood belonging to the Temple. It is possible, of course, that all this did
occur fortuitously over the generations but surely it is far more plausible
to see it all as an artificial, contrived re-working of earlier material by the
editors so that a clear pattern of argumentation is presented.
The sugya now continues with a list of seven Amoraim, each of
whom objects that since our Mishnah evidently intends to include
every possible case of a prohibition that can be incurred with this
single act of ploughing, at least one further prohibition can be
found, and why does the Mishnah fail to record this? First,
R. Oshia protests that the Mishnah might have included the
prohibition of one who sows in a 'rough valley', prohibited by
' which is neither eared nor sown' (Deuteronomy 21:4). Secondly,
R. Hanania protests that there could have been included one who
erases a divine name as he ploughs. This merits a flogging because,
as above, Scripture states: 'And burn their groves with fire [but]
Ye shall not do so unto the Lord your God' (Deuteronomy 12: 3-4).
R. Abbahu's objection is that the Mishnah could have included one
who cuts away a leprous sore as he ploughs. The prohibition is
found in the verse: ' Take heed in the plague of leprosy'
(Deuteronomy 24: 8), understood as meaning that the sore must be
taken care of i.e. preserved and not cut away. Abbaye finds two
further prohibitions the Mishnah could have included. These are:
removing, as he ploughs, the breastplate of the High Priest from
the ephod, of which Scripture says: ' And . . . the breastplate be not
loosed' (Exodus 28: 28) and removing, as he ploughs, the staves
of the ark from their rings, of which Scripture says: ' they shall not
be taken from it' (Exodus 25: 15). R. Ashi's objection is that the
Mishnah could have included one who ploughs with a plough made
of asherah wood, the prohibition being, as above, ' And there shall
cleave nought of the cursed thing to thine hand' (Deuteronomy
13: 17; The Torah: Deuteronomy 13: 18). Ravina objects that the
88 The Talmudic argument
Mishnah could have included one who cuts down goodly fruit trees
as he ploughs, the prohibition being: 'for thou mayest eat of them,
and thou shalt not cut them down' (Deuteronomy 20: 19).
Finally, the objection put by R. Zeira to R. Mani is recorded:
let the Mishnah include one who has taken an oath not to plough
on a Festival. When he does plough he incurs the further penalty
for breaking his oath. The reply is that such an oath has no validity
since no oath is binding when it is for the purpose of placing a
prohibition on something already forbidden by the Torah. Such
an oath is akin to taking an oath to refrain from some act and then
taking another oath to refrain from the same act. The second oath
has no effect and there is only a penalty for breaking the first oath.
Here, too, the man is already besworn not to plough on the Festival
since all Isreal has been besworn at Sinai to keep the precepts. But,
it is objected, the Mishnah could have included the man who takes
a general oath not to plough at all, neither on the Festival nor on
week-days. Here, since the oath does take effect so far as the
week-days are concerned, it takes effect, too, so far as the Festivals
are concerned. The analogy would be with a man who first took
an oath not to eat a particular loaf of bread and then took a second
oath not to eat any bread. Since the second oath is more embracing
than the first it embraces, as it were, that particular loaf of bread
as well. A new reply is attempted. The reason the Mishnah does
not refer to the case of the oath is because an oath, or, for that
matter, any verbal declaration, can be annulled by a Court of three
persons if there are good grounds for the annulment. The Mishnah
only lists prohibitions which remain permanently, not those that
can possibly be annulled. But, in that case, what of the sacred
animals mentioned in the Mishnah? The vow by means of which
these had been dedicated can be annulled and yet the Mishnah does
refer to this type of prohibition. The reply is that the Mishnah deals
with an ox that is a first born and is thus dedicated to the Temple
automatically. There is no possibility of annulment here since the
animal's dedication is not by means of a verbal declaration.
The difficulty here is that the Mishnah mentioned both an ox and an ass.
The ox can be a first-born but what of the ass? Riban's commentary 21a
s.v. ve-hen mukdashin, understands the reference to the ass to be in ac-
cordance with the view that there is a Biblical prohibition on working with
the first-born of an ass. This is very uncertain, see Tosafists s.v. ha-horesh,
that the Mishnah only refers in fact to an ox that has been dedicated and
Penalties resulting from single act 89
the total of eight prohibitions is arrived at not by counting the ox and ass
as two but by counting mixed seeds in a vineyard as two, one for sowing
mixed seeds on their own in any field and one for sowing seeds in a
vineyard. Gf. Albeck's Mishnah, Nezikin, to this Mishnah, Supplements, p.
467.
But, it is objected, what of the Nazirite prohibition which the
Mishnah mentions. As a verbal declaration, the Nazirite vow can
be annulled. To this the reply is given, the Mishnah refers to a
Samson type of Nazirite, i.e. a Nazirite who was ordered to become
one in a dream, like Samson. Such a Nazirite must observe the rules
and cannot hope for annulment since it is not the result of any
verbal declaration. But, it is objected, this cannot be, for a Samson
Nazirite is not prohibited from becoming contaminated through
contact with a corpse, just as we find that Samson himself did allow
himself to be contaminated. The final reply is that the reason the
Mishnah does not include the case of the man who took an oath
is because the Tanna of the Mishnah holds that such an oath is not
binding even where the week-days as well as the Festivals were
embraced by it. The oath would only take effect for the week-days
not for the Festivals. In the analogy mentioned above, if a man took
an oath not to eat a particular loaf of bread and then took another
vow not to eat any bread, his second oath would only take effect
on all other bread, not on that particular loaf which he may not
eat in any event because of his first oath. In the words of the
Talmud: Our Tanna does not accept the principle of a more-
embracing prohibition, i.e. the rule that a prohibition cannot take
effect where there is already a prohibition applies even where the
second prohibition is more embracing in scope than the first.
This list of objections is obviously late, featuring as it does those of R. Ashi
and Ravina. (The mnemonic at the beginning of this section, listing the
names of the Amoraim, is post-Talmudic and does not occur at all in some
texts, see DS.) It would appear that various Amoraim tried their hands
at discovering new possibilities to be included in the Mishnah. They are
all very remote and it would therefore seem that all this is a purely
academic exercise, not really to be taken seriously, which is why no reply
is given to the first six objections, see Tosqfists s.v. matkif lah. R. Zeira's is
less remote and to this a reply is, in fact, eventually given. The order in
which the Amoraim are listed is more or less chronological, except for
R. Zeira. This was put last because there is a discussion on it, unlike the
others, and a reply is eventually given. It is possible that R. Zeira's
objection and the discussion on it were originally quite independent of the
90 The Talmudic argument
rest of the list. It would seem, too, that the whole of this section was
originally an independent sugya, appended after the other by the editors,
though it does seem somewhat strange that two of the prohibitions, erasing
the divine name and ploughing with asherah wood, are virtually the same
prohibitions derived from the same proof-texts as those in the previous
section.
9
Simanim de-oraita o de-rabbanan: whether
reliance on distinguishing marks for the
purpose of identification is Biblical or
Rabbinic
The main sugya is Bava Metzi'a 27a (last line on page)28a, though
there is a cross reference to it earlier, on page 27a.
In the whole of the second chapter of Bava Metzi'a (cf. supra, p.
34) it is accepted as the rule that a lost article must be returned
by the finder to the one who claims that he lost it, if the latter
substantiates his claim either by producing two witnesses that it is
his or by declaring the distinguishing marks of the article. These
are known as simanin (or, in the Hebrew form, simanim, sing, siman).
Simanin are thus treated as a reliable means of identification. But
it is not clear whether reliance on simanin has Biblical warrant -
simanin de-oraita - or whether such reliance is by Rabbinic
enactment only - simanin de-rabbanan.
The sugya begins with the stock formula when a problem is set:
ibbaye le-hu, 'they [the anonymous scholars] set a problem'. The
problem is: 'Are simanin de-oraita or de-rabbanan?' The practical
difference, the Talmud continues, is whether or not a bill of divorce
(get) can be returned when its simanin have been stated. A husband
sends a get, which has to be written specifically for that husband
and wife (see supra p. 24), through his agent, to deliver it to the
wife and the agent loses the get but a get is later found. The agent
describes the simanin of the get he has lost and these coincide with
the simanin of the get that has been found. Can simanin be relied on
sufficiently to enable the agent to take the get and deliver it to the
wife? Is it the same get or is it another with, fortuitously, the same
names of the husband and wife, and therefore invalid? Now if
simanin are relied on by Biblical warrant, this reliance will not only
apply to monetary matters such as a lost article but also with regard
to religious law, e.g. so that the wife will be released from the
91
92 The Talmudic argument
marriage bond by that get. But if reliance on simanin in the case of
the lost article is only the result of a special Rabbinic enactment,
such an enactment is only effective in monetary matters, where the
Rabbis enjoy authority to change the law if the good of society
demands it, but it cannot be effective in religious law since if, in
Biblical law, the get is possibly invalid the Rabbis have no authority
to render it valid through reliance on simanin, and thus contradict
a Biblical law.
The Talmud, having stated the problem, now proceeds with
attempts to solve it one way or the other. The first attempted proof
is from the Mishnah (Bava Metzi'a 2: 5, 27a) to which the sugya is
appended. The Mishnah comments on the verse dealing with the
restoration of a lost article to the loser: And so shalt thou do with
his garment; and so shalt thou do with every lost thing of thy
brother's' (Deuteronomy 22: 3). Since the verse speaks of'every
lost thing of thy brother's' why does it specify, in addition, his
' garment' ? The Mishnah states: ' The garment was included in all
these, why, then, was it mentioned separately? To compare all
other articles to the garment. Just as a garment has simanin and has
someone who lays claim to it so, too, every lost article [cannot be
kept by the finder] if it has simanin and someone who claims it.'
That is to say, two conditions are required for the law of the return
of the lost article to apply. The first of these is that the article must
have distinguishing marks by means of which it can be identified.
The second condition is that the owner must claim it, i.e. he must
have retained his ownership. But if there is evidence that the owner
had abandoned the article, before the finder possessed it, the finder
may keep it. These two conditions are derived from the fact that
Scripture singles out the ' garment' . The ' garment' is thus singled
out so that it should act as a prototype of lost articles that have
to be returned. Generally speaking, a garment satisfies both these
conditions; it has distinguishing marks and the owner persists in
retaining his ownership. Now on the face of it, the Mishnah derives
the law of simanin from the Scriptural verse in which the ' garment'
is mentioned and this surely means that, according to the Mishnah,
simanin have Biblical warrant - simanin de-oraita. No, the reply is
given, there is no proof from the Mishnah. It is possible that the
Mishnah holds that simanin are de-rabbanan. The Biblical text quoted
in the Mishnah is not to teach us that simanin are to be relied on
but to teach us the other condition stated in the Mishnah, that the
Reliance on distinguishing marks 93
owner must retain his ownership. In that case why does the
Mishnah refer to simanin, which are Rabbinic and are not derived
from ' garment' ? It is simply because, since the Mishnah is quoting
the verse and since by Rabbinic law, at least, simanin are relied on,
the Mishnah sees no harm in referring to the simanin condition as
well as the other condition, though, in fact, only the latter has
Biblical warrant.
A further attempt to prove it is now made. The Mishnah has
explained why ' garment' is mentioned on its own in the verse. But
in the verse the word ' ass' also occurs. Why is the ass mentioned
on its own? The reply given in the previous sugya (Bava Metzi'a 27a)
is: we might have supposed that, if the claim is made that the ass
is his by a man who cannot describe the simanin of the ass itself but
only those of its saddle, this is insufficient means of identification.
That is why the Torah refers specifically to the ass. For the return
of the ass where this is itself identified no verse is required. It can
be derived from the general rule. So the verse intends to teach us
that the ass must be returned even to the one who only identifies
the saddle. In any event, we surely see from this that simanin are
relied on by Biblical law since the law regarding the simanin of the
saddle is derived from a word in Scripture, the word ' ass' . No, the
reply is given, there is no proof from the verse and the interpretation
of it as referring to the saddle. True the verse does teach us that
secondary identification, i.e. not of the article itself but of something
normally attached to it, like the saddle to the ass, is sufficient. But
this secondary identification might mean identification by the man
who makes the claim producing witnesses, i.e. he produces not the
simanin of the saddle but witnesses that he owns the saddle.
A third attempt is now made to prove it from the verse: ' and
it shall be with thee until thy brother seek after it' (Deuteronomy
22: 2). On this the Rabbis comment:' Would you imagine that you
have to give it to him before he seeks it? But the verse means: Seek
him to see whether or not he is a fraud.' The Rabbinic homily is
to understand derosh oto, 'seek it' , as 'seek him\ i.e. make inquiries
of him in order to establish whether or not he is telling the truth
when he claims that the article is his. Now, it is here suggested, how
can this inquiry be carried out if not by demanding of him that
he states the simanin and since this is based on a Biblical verse it
proves that simanin are de-oraita. To this the reply is given that the
inquiry can be conducted not by demanding simanin but by
94 The Talmudic argument
demanding that he produce witnesses who will testify that it is
his.
The Rabbinic homily quoted here resembles that found in a later Mishnah
(Bava Metzfa 2:10) but the quote here cannot be from this Mishnah since
according to the Mishnah the verse means that a man suspected of fraud
must produce witnesses, i.e. the verse deals, according to the Mishnah, with
a man known previously to engage in fraud. In all probability (see Tosqftsts
s. v. derdshehu) this is an independent Baraita and the attempted proof is from
'seek him\ and this would seem to mean by asking him to produce simanin,
since if he is required to produce witnesses it would be they (and not he)
who establish the claim. The progress of thought in this section is clear.
The weakest of the attempted proofs is from 'garment'. This is easily
refuted since the Mishnah derives two conditions from the verse (simanin
and retention of ownership) and it is possible, as the Talmud replies, that
the verse only refers to the second condition not to simanin. But then the
interpretation of'ass' is quoted. This is a much stronger proof since only
one derivation is mentioned. For all that, the reply is given, there is no
proof since the reference might be not to simanin but to witnesses. Finally,
the strongest proof of all is attempted; as if to say, but surely here where
the verse speaks of seeking him it must mean by demanding simanin. The
reply is, no, even here the reference is to witnesses.
The Talmud now attempts a proof from a Mishnah (Yevamot 16:
3) dealing with a completely different topic. A corpse is found, the
identity of which is unknown but which resembles a missing
husband. How is the corpse to be identified so as to enable the wife
to remarry? The Mishnah states that the face must be recognisable
and that, consequently, a headless corpse cannot be identified as
that of the missing husband even if there are simanin on the body
or the garments of the corpse which coincide with those of the
missing husband. But if simanin are de-oraita they would be relied
on to permit the woman to remarry. The statement in the Mishnah
seems to prove that the Mishnah holds that simanin are de-rabbanan.
The reply is given that the Mishnah may only mean that no
reliance is to be placed on vague simanin of the body, e.g. that it
is of a tall or short man. Since there are numerous men who are
tall and numerous men who are short, these are not really adequate
means of identification at all and would be unacceptable even if
simanin are de-oraita. As for the reference in the Mishnah to simanin
of the garments, these may, indeed, be adequate simanin and they
do establish that the garments are those of the missing husband,
yet the Mishnah still rules that the woman may not remarry
because the husband may have lent his garments to another man
Reliance on distinguishing marks 95
whose corpse this is. In that case, the Talmud asks, why do we allow
a man to claim an ass through his identification of the saddle; he
may have lent his saddle to another whose ass this is? To this the
reply is given that people do not normally borrow saddles, because
these do not fit animals for which they were not made and would
make the animal sore. Or, the Talmud continues, ' If you want I
can say' that the reference to the garments is, like that to the corpse,
not to proper simanin but to vague ones such as, the garments are
white and red. These are not simanin at all.
In the discussion the question has been raised whether or not
borrowing is to be taken into account. In that case, the Talmud
continues, why have we learnt in a Baraita that if an agent bearing
a get lost it and then found it, after a long period, tied to his
money-bag or purse or signet ring, this is sufficient means of
identification and the get can be used to divorce therewith the wife?
But if we take borrowing into account he may have lent these
objects to someone else and has forgotten that he has done so in
the long period that has elapsed. To this the reply is that people
do not lend these objects: money-bags and purses because it was
held to bring bad luck and signet rings because of the fear that they
may be used to forge documents.
There is considerable confusion in our texts. Once the ' if you want I can
say' reply has been given, what is the significance of the question
regarding the get, since at this stage of the argument it is held that we do
not take borrowing into account? The reply: 'If you want I can say' has
obviously been misplaced and should come at the end of the section (as
in the Munich codex, see DS, and in the parallel passage in Tevamot 120b)
after the question whether or not one takes borrowing into account has been
exhausted.
The Talmud now puts forward the suggestion that this question,
whether or not simanin are de-oraita, is debated by Tannaim. A
Baraita is quoted dealing with the previous question of identifying
a corpse. The Sages hold that a mole on the corpse on the same
part of the body on which the missing husband had a mole is
insufficient evidence that the corpse is that of the missing husband,
but R. Eleazar b. Mahabai holds that it is sufficient and that the
woman may remarry on the strength of the identification. Thus
it would appear that the Sages hold that simanin are de-rabbanan and
R. Eleazar b. Mahabai holds that simanin are de-oraita. No, suggests
Rava, it may be that the Sages, too, hold that simanin are de-oraita
96 The Talmudic argument
and they would rely on other simanin to allow the woman to remarry.
They only refuse to do so in the case of a mole because they hold
that it is likely for such a mole to be found on the body of one born
under the same star and at the same time as the missing husband
(ben gilo, 'his double'). This corpse may be that of the husband's
double. R. Eleazar holds that it is unlikely for a mole to be on the
same spot, even in a double. ' If you want I can say', continues the
Talmud, both authorities, the Sages and R. Eleazar b. Mahabai,
hold that simanin are de-oraita and both hold that it is unlikely for
a mole to be found on the husband's double and therefore a mole
is an adequate simanin. But the Sages hold that simanin tend to
change their colour after death, so that the mole found on the corpse
is not an adequate siman at all. ' If you want I can say', the Talmud
continues, all authorities hold that simanin do not tend to change
after death so that if simanin were de-oraita all would agree that the
women may remarry but, in fact, both hold that simanin are de-
rabbanan. In that case why does R. Eleazar b. Mahabai allow the
woman to remarry? It is because he holds that a mole is a far better
siman than any other since it is extremely unlikely that the corpse
is of another man who happened to have a mole on exactly the same
place. This is known as a siman muvhak, ' a perfect siman\ This is
effective even {{simanin are de-rabbanan. But the Sages hold that a
mole is not a siman muvhak. It has the same standing as other simanin
and since simanin are de-rabbanan the woman may not remarry.
See Maggid Mishneh to Gezelah Va-Avedah 13: 3 that from our sugya it
emerges that there are three types of simanin: (1) siman muvhak. This is so
conclusive as evidence that it is relied upon even if simanin are de-rabbanan,
and even to permit a woman to remarry. This is really far more than a siman.
It amounts to positive identification as if by witnesses. (2) A very weak
siman, e.g. that the garment is white or red. This is not relied upon even
if simanin are de-oraita and not even for the purpose of restoring a lost article.
It is not a siman at all. (3) Average simanin, i.e. a man states the weight of
the article he has lost or, say, the hands of the corpse have an unusual shape.
These simanin are definitely relied upon to restore a lost article and it is
in connection with these that there is the debate whether they are de-oraita
or de-rabbanan and so whether they can be relied upon even to permit a
woman to remarry. When it is said in our sugya that the testimony of
witnesses is accepted even if there are no simanin it means that the witnesses
identify the corpse by direct perception. They know that it is so. The other
Talmudic pasages dealing with reliance on simanin are: Yevamot i2oa-b;
Gittin 27b; Hullin 7ga-b; 95b-o,6a. The Yevamot sugya is virtually a duplicate
of ours. But a close examination of the way the material has been arranged
Reliance on distinguishing marks 97
in the two sugyot shows that the editors have arranged the Yevamot material
so as to accord with the main theme of corpse identification mentioned
in the Yevamot Mishnah, to which that sugya has been appended, whereas
in our sugya the material has been arranged so as to fit in with the more
general question discussed here of reliability on simanin and matters arising
out of this. The Yevamot arrangement is: (1) The Mishnah dealing with
the simanin of the corpse and its garments, from which it appears that simanin
are not de-oraita. (2) But this contradicts the Baraita regarding the get. (3)
Abbaye replies: this question, whether simanin are de-oraita or de-rabbanan,
depends on the debate between the Sages and R. Eleazar b. Mahabai, the
Mishnah following the view of the Sages, the Baraita the view of R. Eleazar
b. Mahabai. (4) Rava disagrees: the debate is rather whether a mole is
likely to be found on the husband's double. (5) 'Others say' the debate
is about whether there is change after death. (6) 'Others say' that Rava
says: all hold that simanin are de-rabbanan and the debate is whether a mole
is a siman muvhak. (7) According to the opinion that both authorities hold
that simanin are de-oraita, the Mishnah deals with ' tall' or ' short' and as
for 'garments' we take borrowing into account. (8) Continues, as in our
sugya at the beginning, with this question of whether borrowing is taken
into account. Our sugya embraces the material found in the Yevamot sugya
but contains much more material, i.e. that which precedes the discussion
around the question of the corpse and that which follows it. It is not possible
to determine whether or not the Yevamot sugya depends on ours or whether
the two sugyot are independent of one another but use the same ' raw mat-
erial ' in their different constructions. Cf. Abraham Weiss, The Babylonian
Talmud as a Literary Unit (New York, 1943), p. 208, note 152.
There now follows a saying of Rava: If you would say that
simanin are not de-oraita, why do we return a lost article on the
strength of simanin? That is to say, if Biblical law refuses to place
any reliance on simanin, what social purpose is served by Rabbinic
law accepting simanin? The reply is that the finder is quite willing
to give up the article he has found on the strength of the simanin
produced by the man who claims it, since he knows that if he loses
an article it will be returned to him when he produces simanin.
Strictly speaking, since simanin are not de-oraita he should be entitled
to keep the article. Yet he will be quite content to obey the new
Rabbinic enactment and surrender the article because he
appreciates that the new enactment is for the benefit of all losers
and he may himself be a loser one day. A kind of social contract
can here safely be assumed. But R. Safra demurs to the explanation
attributed to Rava:' Can a man benefit himself with his neighbour's
goods?' To be sure the finder may be content to surrender the
article, but it really belongs to the rightful owner and the man who
98 The Talmudic argument
lays claim to it by producing the simanin may not be the rightful
owner since simanin are not de-oraita. Any benefit that comes as a
result of the new Rabbinic legislation must be for the rightful
owners, for the losers not the finders, if it is to be acceptable.
Therefore, we should rather explain (Rava would rather say, Rashi)
that all losers are prepared to take the risk that if simanin are relied
on, their article may be surrendered to someone to whom it does
not belong. They are quite content to take the risk because the
alternative would be that they have always to produce witnesses
and these may not be forthcoming, whereas simanin are generally
present.
If that is the case, the Talmud now asks, how do we explain the
rule in the Mishnah (Bava Metzi'a 1: 8): ' Rabban Simeon b.
Gamaliel said: If one finds three bills of indebtedness on the same
debtor by three different creditors they are to be returned to the
debtor but if one finds bills of three different debtors but for the
same creditor they are to be returned to the creditor.' We have
argued that the effectiveness of the new legislation is because the
loser is prepared to take the risk since otherwise he might never
recover his property. In the case of the three bills made out against
three different debtors the reason, presumably, why they are
returned to the creditor is because he declares the siman, namely,
that there are three, otherwise how would he know how many
bonds there were? But here the reasoning that the loser is prepared
to take the risk in order to have the goods returned to him does
not apply. The debtors would be quite content for the bonds to
remain with the finder. All the debtor wants is that the bill should
not be given to the creditor. To this the reply is given that in the
case of the bonds a logical principle is at work. If the three bonds
are made out against the same debtor they are returned to him
because he must have lost them, otherwise how does it come about
that bonds made out against him by three different creditors are
found together? Consequently, the bills must have been paid and
he had had them returned to him and then lost them. Conversely,
if the three bonds are on three different debtors for the same
creditor it can safely be assumed that he lost them, otherwise why
would the three be found together? (And this is far more than an
ordinary siman; it is a siman muvhak, which, as above, is effective even
if simanin are de-rabbanan in the case of re-marriage and, conse-
quently, there is no need here for the special Rabbinic enactment.)
Reliance on distinguishing marks 99
But, the Talmud asks, another clause in the same Mishnah states
that if one finds a bundle of bills wrapped together, they are to be
returned to the creditor if he states as a siman that they are wrapped
together, even if all the bills are against the same debtor. This can
only be because of the Rabbinic enactment and here why should
the debtor, if they are really his, take the risk? Consequently, the
Talmud concludes, it is hard to defend the view that simanin are
de-rabbanan since the reason for the Rabbinic enactment if there
were such would not apply in all cases in which objects are returned
on the strength of simanin. Therefore, what Rava really said was
that simanin must be de-oraita and this is proven from the verse of
'seek him' and the earlier rejection of the proof (that it might mean
'seek him' by demanding witnesses) is itself rejected.
There now follows a series of problems set by Rava on the
assumption that simanin are de-oraita. The Talmud interjects, what
do you mean by ' on the assumption'? Rava has just stated that
simanin are quite definitely de-oraita! To this the reply is given that
the expression ' on the assumption' is not entirely out of place since
the proof from ' seek him' can, as above, be refuted ' as we have
explained it' .
This seems to show that this whole section has been added later, after the
previous sugya had been completed.' As we have explained it' can only refer
to the discussion in the previous section of the sugya.
Rava's series of problems is set if it is held that simanin are de-oraita
and hence enjoy full Biblical authority. Supposing two men lay
claim to the lost article and each states its simanin accurately. To
which of them should it be surrendered? It should be surrendered
to neither but must be left in the Court (until Elijah comes, Rashi).
Since each has as powerful a claim to the article as the other and
one of them is obviously a fraud, there is no valid reason for giving
it to either one. If A states the simanin but B has witnesses that it
belongs to him, it is given to B. Even though simanin are de-oraita
the testimony provided by witnesses is more reliable. If both A and
B have simanin but one of them has, in addition, a single witness
to support his claim, then since a single witness is not trusted to
testify in monetary matters the witness's testimony is disregarded
and the article has to be left in the Court as if it were simply a case
of both producing simanin alone. If A has witnesses that he had
woven the garment but B has witnesses that he lost it, it is given
ioo The Talmudic argument
to B since even if A had woven it this does not prove that he did
not sell it and it was purchased by B. If A states the garment's length
and B its width, it is given to A. The width of the garment can be
arrived at by seeing it on its owner but it is impossible to guess the
length of the garment since, in those days, the garment was wound
around the owner's person lengthwise and a fraud would have no
means of knowing how many times it had been wound. If A states
the length and width separately but B states the total of the length
and width only, it is given to A. His information is more precise.
If A states the length and breadth and B the weight, it is given to
B, since it is harder to guess the weight than the length and width.
The husband states the simanin of a get and the wife does the same;
it is given to her (i.e. it is assumed that she is telling the truth that
it had been delivered to her and she lost it so that the get was
effective and she may now remarry). To this the Talmud objects,
what kind ofsiman? If the meaning is the siman of the get's size, this
is no indication since the wife may have seen the get in her husband's
possession. The reply is that the siman is that of a hole at the side
of this or that letter. This is a siman muvhak, for how would she know
that the get has this peculiarity. If the husband states a siman of the
cord (to which the get is attached) and the wife does the same, it
is returned to her. To this the Talmud again asks, what kind of
siman? If it is a siman of colour, red or white, she may have seen
the cord while the get was in her husband's possession. The reply
is, it refers to the length of the cord, which she could not guess. If
the husband states, as a siman, that it was found in a valise and the
wife does the same, it is given to him. Since she knows that he always
keeps his documents in a valise she may simply be guessing. With
this the sugya is concluded.
To be noted in this last section is the logical order in which the problems
are presented, one leading more or less naturally from the other.
10
Devarim she-be-lev eynam devarim: mental
reservations in contracts are disregarded
The sugya is Kiddushin 4gb-5oa. Devarim she-be-lev eynam devarim
literally translated is: 'words in the heart (= the mind) are not
words'. As a technical, legal term the meaning is: if, in any
contractual arrangement, there are mental reservations contrary
to its verbal or practical implications, these reservations are
disregarded.
The sugya is appended to a Mishnah which reads: ' If a man be-
troths a woman, stating as a condition "that I am a priest" [i.e.
he assures her that she is being betrothed by a member of a priestly
family, the aristocracy] and he is found to be [only] a Levite
[belonging to a lower social class] or [he states that] he is a Levite
and he is found to be a priest [she may not wish to marry into the
higher social class]; a Nathin [a Gibeonite, a bastard according to
Rabbinic law] and he is found to be a mamzer [a bastard according
to Biblical law] or a mamzer and he is found to be [only] a Nathin;
a resident in a small town and he is found to be a resident of a large
town or a large town and he is found to be of a small town; "on
the condition that my house is near to the bath-house" and it is
found to be distant, or distant and it is found to be near [and she
may not want to live too near to the bath-house]; on the condition
that he has a daughter or an adult slave-girl and he has none, or
that he has none and is found to have; on the condition that he
has no children and he has; or on the condition that he has children
and he has none; in all these cases the betrothal is invalid even if
she [subsequently] stated: " It was my intention [lit. ' in my heart' ]
to be betrothed unto him" [regardless of his status or condition].
The same law applies where she misled him.'
The sugya opens with an actual case in which the fourth-century
IO2 The Talmudic argument
Babylonian Amora, Rava, gave a ruling. A man sold all his
property with the clear intention of emigrating from Babylon to the
Land of Israel but said nothing of this at the time of the sale.
Neither in the bill of sale nor in any verbal statement was any
condition attached to the sale, yet all the circumstances are such
that the man's intention was for the sale only to be valid if, in fact,
he did emigrate, otherwise he wished to retain his property. Now
if the man had stated explicitly at the time of the sale that it was
conditional on his emigrating, if eventually he remained in Babylon
the sale would be null and void, the condition being unfulfilled. In
the case under consideration, however, there was no explicit
condition, only an assumed one, although the assumption was
perfectly clear. Rava ruled that the sale was valid even if the man
did not eventually emigrate. ' It is a case of words', declares Rava,
' that are in the heart and words that are in the heart are not words.'
The mental reservation that would invalidate the sale was
disregarded by Rava. The point here is not that the man is
disbelieved about his true intention. All the circumstances go to
show that it was. Yet since he made no condition at the time of
the sale, the act of sale itself is without reservation and the man's
mental qualification is insufficiently strong to upset the categorical
nature of the sale. The reason for Rava's ruling would appear to
be that unless the condition has been verbalized it remains vague
and inchoate, without sufficient power to upset the sale. The man's
determination to sell only conditionally is weak, otherwise he would
have stated this explicitly. His mental reservation consists only of
'words in the heart' and these are not treated as 'words'.
The Talmud asks: 'How does Rava know this?', i.e. what is his
authority for the ruling. It is suggested that, possibly, Rava derives
his ruling from a Mishnah ('Arakhin 5: 6) . (Actually, although
di-tenan, 'for we have learnt in a Mishnah', is used, and the
reference is to this Mishnah in 'Arakhin, the full quote is, in fact,
an elaboration on the Mishnaic rule and is found in Sifra, va-yikra,
3, end.) Here there is a comment on the verse: ' he shall offer it
of his own voluntary will' (Leviticus 1:3). The Hebrew is: yakriv
oto li-retzono, which is rendered as: ' he shall offer it' but 'of his own
voluntary will', i.e. he must offer the animal he had set aside as a
burnt-offering, he is compelled to offer it, yet it is offered 'of his own
voluntary will'. How can this be, since if he is compelled to offer
it it cannot be of his own voluntary will? The reply is that he must,
Mental reservations in contracts are disregarded 103
indeed, be coerced but the coercion must take the form of
compelling him to say: ' I assent'. From this it follows that assent
given under coercion is treated in law as a valid assent. But even
though the man declares eventually that he gives his assent, since
it is only obtained through coercion, presumably he has strong
mental reservations and does not consent 'in his heart'. This shows
that the ' words in his heart' are disregarded, that devarim she-be-lev
eynam devarim. But, the Talmud continues, there is no support from
this for Rava's decision because 'we are witnesses that he wishes
to have atonement', i.e. the reason the assent is valid is not because
the mental reservation is disregarded but rather because there is
no mental reservation. Everyone knows ('we are witnesses') that
a man who sets an animal aside as a burnt-offering really wishes
the animal to be brought as such; he wishes to have the atonement
the sacrifice affords him, which is why, after all, he set it aside in
the first place. His reluctance actually to bring it has to be overcome
by coercion, but once the coercion has been exerted he can be
assumed really to mean it when he declares his assent. The assent
is given ' in the heart' so that there is no support from this for Rava's
ruling.
Possibly, the Talmud continues, Rava derives his ruling from the
second clause in the 'Arakhin Mishnah. After the statement that
assent given under coercion is valid in the case of sacrifices, the
Mishnah adds: ' And so, too, do you find it in connection with bills
of divorce and manumission of slaves: He is compelled until he
declares: "I assent." There are instances when a wife is entitled
to petition for a divorce and a slave to be free of his master. But
the assent of the husband and master is required. This can be
obtained by the Court exercising coercion until there is assent. This
shows that assent given under duress is valid and here the reply
given in the case of sacrifices, that the man wants to find atonement,
does not apply. Here, then, is a clear case of mental reservations
being disregarded and a support for Rava's ruling. No, the Talmud
continues, there is no proof since the principle obtains: 'it is
meritorious to obey the words of the Sages', i.e. the Sages who order
the husband to divorce his wife or the master to free his slave.
Consequently, here, too, it can be argued, the reason it is effective
is not because the mental reservation is disregarded but rather
because, as in the case of sacrifices, there is no mental reservation.
True it is under duress at first but when eventually the man gives
104 The Talmudic argument
his assent he gives it unreservedly because he knows 'deep down'
that it is right to obey the Sages and he does so wholeheartedly.
Having attempted two proofs for Rava's ruling and having
refuted both of these, the Talmud now quotes R.Joseph, who
sought to prove the case from a Mishnah (Kiddushin 3: 5). The
Mishnah reads:' If a man be troths a woman and he [subsequently]
said: "I thought that she belonged to a priestly family" but she
belongs, in fact, to a [lesser] Levite family or [he thought] she be-
longed to a Levite family and she belongs to a priestly family; that
she was poor and she is really rich, or that she was rich and is really
poor; she is betrothed because she did not mislead him.' But should
she be betrothed since, according to the man's testimony, which
we have no reason to doubt, he did not, in fact, intend to betroth
that woman? It can only be because it was a mental reservation
(he did not state explicitly that he only betroths her on the
understanding that she really is what he imagines her to be) and
this is disregarded. Abbaye objects that the Mishnah might merely
be stating a strict view, i.e. when the Mishnah rules that 'she is
betrothed' it might only mean that just to be on the safe side she
should not marry another without first receiving a get, as if the
betrothal had been valid. There is only proof that a mental
reservation is not taken into account when to disregard it results
in strictness. There is no proof that it is totally disregarded, e.g. if
someone else betrothed her afterwards, before she had received the
get, she might well require a get from the second man, too, since,
possibly, the mental reservation is taken into account so that, the
first betrothal being invalid, the second betrothal is valid.
Thus far we have advanced a little further on the road to proof.
We have, at least, established that where strictness is the result the
mental reservation is disregarded. Abbaye now proceeds to
demonstrate that it is disregarded totally, even when to do so results
in leniency. Abbaye's proof is from our Mishnah, the Mishnah to
which the sugya is appended. Here the rule is stated that where the
man betrothed the woman on the condition that he is a priest or
Levite etc. and it is later discovered that he is not what he had
claimed to be, she is not betrothed, even if she declares that she
was willing to be betrothed to him whatever his status. Now here
the Mishnah states 'she is not betrothed' and this expression can
only mean that there is no validity whatsoever to the betrothal and
she can marry another without first receiving a get. This
Mental reservations in contracts are disregarded 105
demonstrates that a mental reservation is totally disregarded, even
when this results in leniency.
The Talmud objects that there is still no proof. True, in the case
of the Mishnah the mental reservation is totally disregarded but
that might be because there has been an explicit condition that he only
betroths her if he really is what he says he is. Here if the mental
reservation is to be effective it has to cancel out the explicit
condition and this, the Mishnah holds, it cannot do. But where the
mental reservation does not have to cancel out an explicit condition
(as in the case of the property sale where, if the mental reservation
is to be accepted, all this had to do is to explain the conditions of
the sale) it may well be effective.
Thus far we have been unsuccessful in finding support for Rava's
ruling. At the most we have been able to establish that mental
reservations are not taken into account when to disregard these
results in strictness. We have not succeeded in proving, as Abbaye
sought to do, that they are disregarded totally, even where it results
in leniency. In other words, the matter is in doubt and,
consequently, where the prospective emigrant sold his real estate
the question is whether or not the sale is valid (in view of his mental
reservation) and since he has possession doubt cannot disturb it.
The Talmud now quotes R. Hiyya bar Avin who reported that a
case of this kind came before the Court of R. Hisda, which
presented it for consideration to the Court of R. Huna (first half
of third century) and this Court did succeed in proving conclusively
that mental reservations are totally disregarded.
Unlike those in many other Talmudic passages, the case considered is a
severely practical one. The reports are that an actual case came before the
Court of R. Hisda and R. Huna, before Rava, at a much later period and,
as at the end of the sugya, before R. Ashi at a later period still. No doubt
sales of property with the intention of emigrating from Babylon to the Land
of Israel were not infrequent in view of the highly meritorious act of settling
in the Land of Israel.
The proof is from a Mishnah (Me'ilah 6: 1) dealing with the
trespass-offering that has to be brought if profane use has been made,
inadvertently, of Temple property. A man instructs his agent
to bring him a money-bag that lies on a window ledge and the
money in the bag belongs to the Temple, the man overlooking this
fact. The Mishnah rules that as soon as the agent, acting on the
man's behalf, takes up the money-bag, profane use has been made
106 The Talmudic argument
of Temple property and the man issuing the instructions is liable
to a trespass-offering. This rule applies, the Mishnah continues,
even if the man later declared that his intention was for his agent
to bring him a different money-bag, one that did not belong to the
Temple. Thus it can be seen that the man's mental reservation is
disregarded. Now since a sacrifice can only be brought where there
is a certain obligation, not a doubtful one (for to bring a doubtful
sacrifice is to risk offering an unconsecrated animal, which is
forbidden), it must follow that the mental reservation is totally
disregarded as a certain ruling, not a doubtful one, proving the case
conclusively.
The Talmud is still dissatisfied. The ruling of the Mishnah may
apply not because mental reservations are disregarded but rather
because there is no mental reservation. The man's intention may
have been for the money-bag the agent actually took and he only
declares later that it was not because he wishes to spare himself the
expense of bringing a sacrifice. To this the reply is given that if he
simply wished to avoid bringing a sacrifice he could easily have
achieved this aim by later declaring that he knew the money-bag
contained Temple money. The rule is that a trespass-offering can
only be brought where there has been unintentional transgression.
No 'atonement' is available for one who intentionally makes
profane use of Temple property. No, the Talmud objects, this
option was not open to him since to choose it would mean that he
declares to all and sundry that he was a sinner in wishing
intentionally to use Temple property and 'people do not declare
themselves to be wicked'. Finally, the Talmud states that if his
intention had been simply to avoid bringing a sacrifice he could
have achieved his aim by later declaring that he had remembered
that the money belonged to the Temple before the agent took it
on his behalf. A Baraita is quoted in which it is stated that where
this happens it is the agent, not the man himself, who is liable to
the trespass-offering. Consequently, the man has the option of
avoiding the expense of the sacrifice by declaring that he remem-
bered. Since he did not avail himself of this option, it is clear that
he is an honest man and is telling the truth when he says that he
had the mental reservation. In that case, why is the mental
reservation disregarded? It can only be because mental reservations
are disregarded and this proof is conclusive.
It is now implied, though not stated explicitly, that Rava's ruling
Mental reservations in contracts are disregarded 107
was not, in fact, original to him. A similar case had been brought
for consideration to the Court of R. Hisda and eventually to the
Court of R. Huna, long before Rava's day. There the decision that
devarim she-be-lev eynam devarim found its conclusive proof in the
Mishnah from Me'ilah. This decision was handed down and Rava
evidently knew of it so that when the case came before Rava he
was able to decide it on the basis of his tradition. He had ample
precedence for his ruling. The final reply to the original question:
'How does Rava know this?' is thus that he knows it from
R. Huna's decision which, in turn, is based on the correct under-
standing of the Mishnah in Me'ilah. (It is not possible to determine
whether or not R. Huna actually used the formula: devarim she-be-lev
eynam devarim. He may simply have rendered the decision and the
actual formula is Rava's.) In any event this formula is a further
instance of Amoraic abstraction, a legal formulation of teachings
found in earlier, Tannaitic, sources.
The stages in the sugya are:
(1) Rava's ruling
(2) First attempted proof: case of sacrifice
refuted: wishes to have atonement
(3) Second attempted proof: case of divorce and manumission
refuted: wishes to obey the Sages
(4) R. Joseph's proof: case of betrothal
Abbaye: refuted: there it is strictness
(5) Abbaye's own proof: our Mishnah
refuted: there it is a condition, which is stronger
(6) R. Huna's proof: Mishnah MeHlah
refuted: wishes to avoid sacrifice
(7) Proof stands: could have said that it was intentional
refuted: people do not declare themselves wicked
(8) Proof finally stands: could have said that he remembered
A careful examination of this sugya demonstrates its completely
neat and logical progression of thought. The refutation offered to
(2) does not apply to (3); that offered to (3) does not apply to (4);
that offered to (4) does not apply to (5); that offered to (5) does
not apply to (6); (7) and (8) are elaborations of the final proof (6).
Clearly such an order is artificial though dramatically effective.
R. Huna gave his decision long before Rava. (The discussion on
R. Huna is probably the work of the later editors.) R.Joseph and
Abbaye are similarly independent of Rava. And even in the
discussion on Rava the material is not presented in the order we
108 The Talmudic argument
might have expected; note, especially, how our Mishnah, to which
the sugya is appended, is not produced immediately but only when
the other attempted proofs and their refutation have led up to it.
The only possible way of understanding all this is that the final
editors have used earlier material and have presented this with their
own framework so that a literary unit is the result.
The theme of devarim she-be-lev eynam devarim is discussed at length in Kesef
Nivhar, no. 45, pp. 78a-7ga and in ET, vol. 7, pp. 170-86. The actual
formula occurs in only two other Talmudic passages: Nedarim 28a and
Me'ilah 21a. In the Nedarim passages it is stated: 'Even though we have
established the rule that devarim she-be-lev eynam devarim yet ' The
Me'ilah passage is a comment on the Mishnah in MeHlah quoted in our
sugya and reads: ' What does this Mishnah teach us? It teaches that devarim
she-be-lev eynam devarim.' This would suggest that both the Nedarim and the
Me'ilah passages are later than our sugya, the conclusion of which they
accept.
The sugya finally discusses further cases of sale of property with
the intention of emigrating from Babylon to the Land of Israel.
Basically these cases have nothing to do with the theme of devarim
she-be-lev because in them the seller states explicitly that he sells his
property on the condition that he emigrates and if he does not he
cancels the sale. The cases are only recorded here because of the
association provided by the theme of selling property when the
intention is to emigrate.
A man sells his property, stating that he only does so because he
intends to emigrate to the Land of Israel. If he does not emigrate
and remains in Babylon the sale is null and void. The man did go
to live in the Land of Israel but, failing to find adequate accom-
modation there, he returned to Babylon. Is the sale null and void?
Do we say that since he did, in fact, emigrate to the Land of Israel
his condition has been satisfied and the sale is valid; or do we rather
say that the meaning of his condition was for him actually to reside
in the Land of Israel? In other words, does the clause in the
contract: ' On condition that I go up to the Land of Israel' mean
literally ' to go up' or does it mean ' to settle there'. Two versions
of Rava's ruling in this case are given. According to one version,
Rava ruled that to 'go up' means to settle there and the sale is null
and void but according to the other version, Rava ruled that to
'go up' means simply to go there and the sale stands.
Finally, a case is recorded that came before R. Ashi, of a man
Mental reservations in contracts are disregarded 109
who sold his property, stating explicitly that he only did so on the
condition that he emigrated to the Land of Israel, but eventually
he did not actually go there and then wished to cancel the sale. Two
versions of R. Ashi's ruling are given. According to the first version,
R. Ashi said: ' If he wants he can still go up' , and the sale is valid.
According to the other version, R. Ashi also ruled that the sale was
valid but he put it in this form: ' Can he not go up if he wants to?'
The Talmud asks, what is the difference between the two versions,
since according to both the sale stands? The practical difference
is where it became known after the sale that it was dangerous to
journey to the Land of Israel because of robbers and the like. If
R. Ashi simply said:' If he wishes he can still go up' , the sale is valid,
since though it is dangerous it is still possible. But according to the
second version: ' Can he not go up if he wants to?' , the sale is
invalid. This formulation suggests that there is nothing at all
preventing the man from realising his ambition to emigrate, but
where there is, the condition has not been fulfilled and the sale is
null and void.
Note the two versions of R. Ashi's ruling and the statement of the practical
differences between these, all of which must have been recorded by the
final editors some time later than the time of R. Ashi.
11
Hazakah: presumptive state
The principle of hazakah is found in numerous Talmudic
passages. It is probably the most ubiquitous of all Talmudic themes.
The term is from a root meaning ' to be strong', ' to persist in' , ' to
grasp', hence hazakah means' to follow the presumptive state' , ' that
which has been seized hold of. Where there is a doubt in law or
in the circumstances of a given case, the matter under consideration
is not treated as a mere doubt but as if it were a certainty. It remains
in that presumptive state which obtained when the doubt arose.
If, for instance, there is a doubt whether a given piece of meat is
kasher, it is forbidden but only as a doubtful prohibition. But if the
doubt arose whether an animal had been properly slaughtered it
is treated as forbidden as if it were a certainty, because here there
is a presumptive state, the animal while alive being forbidden.
Similarly, if a husband delivered to his wife a doubtful get she is
treated as a married woman not by doubt but by certainty, since
before the get had been delivered she had the presumptive status
of a married woman and she remains in this state until she has
definitely emerged from it. Thus hazakah determines the law in all
cases of doubt so that the matter is no longer treated as doubtful.
Naturally, hazakah does not tell us anything about the actuality
of the law or the circumstances. It does not tell us, for instance, that
where there are doubts regarding the slaughtering of the animal
the animal has not really been slaughtered correctly. How could
hazakah possibly do this? Hazakah is rather a matter of procedure.
Where the case is in doubt, the hazakah principle informs us, the
correct procedure to be adopted is to leave it in the state which
previously obtained. It is as if the law were saying, prove that the
circumstances have changed the status of the thing and if no proof
Presumptive state 111
is forthcoming leave it as it is. The procedure is not to demand proof
that it has not changed but to demand proof that it has. Our sugya,
Hullin i ob, seeks to discover Scriptural authority for the principle
oi hazakah.
We consider here only the sugya, this one, dealing with the question of the
authority for the hazakah principle. The Talmudic passages in which the
theme occurs are far too numerous to record. Moreover, the term hazakah,
in addition to its most frequent use, that of presumptive status, means, in
many passages, other quite different legal principles, e.g. sometimes the
term is used of a mode of acquisition and sometimes in the sense of rubba
where hazakah then means: ' It has been established as extremely likely.'
KesefNivhar, no. 60, part 11, pp. 6a-19a, gives a useful survey of the various
type. J. L. Kroch's Hazakah Rabbah (Leipzig-Jerusalem, 1927-63), is a
massive work of many volumes on the theme in all its ramifications. ET
devotes a large part of two whole volumes to the theme: vol. xm, pp.
453753; vol. xix, pp. 1-423. Cf. the fine introduction to the theme in vol.
xm, pp. 453-65 and the bibliography, p. 465.
The sugya opens with: ' How do we know that which the Rabbis
say: A thing is retained in its presumptive state (hazakah)?' The
reply is given by R. Samuel b. Nahmani in the name of the
third-century Palestinian Amora, R.Jonathan, who quotes a
Scriptural verse: ' Then the priest shall go out of the house to the
door of the house, and shut up the house seven days' (Leviticus 14:
38). The verse deals with a 'leprous house', a house in which a
plague-spot of discoloration had appeared on one of its walls. The
minimum size of the spot for the house to be unclean when it is shut
up is a geris, a bean. At the end of the seven days the priest has
to open the house to see whether or not the spot has increased in
size. During the seven days that the house is shut up anyone who
enters there becomes unclean (verse 48). If the house is in error shut
up for a spot that is less than a bean in size, the 'shutting up' is
invalid and the house does not contaminate. Now from the whole
tenor of these verses it appears that plague-spots were subject to
considerable fluctuations, increasing and decreasing in such a way
as to render it doubtful at any given moment whether the spot had
changed its size. Since the spot is on the walls of the house inside
and the priest is obliged to 'go out' of the house before he shuts
it up, then it is always possible that at the moment of his shutting
up of the house (the moment from which the house contaminates)
the spot has decreased and is no longer the size of a bean. And yet
the Torah states that whoever enters the house during the seven
112 The Talmudic argument
days after it had been shut up is unclean, and as of certainty not
as of doubt. This can only be because of the hazakah principle.
When the priest first observed the spot it did have the minimum
size, at least, of a bean and we, therefore, assume that it has
remained in that state. Even if at the end of the seven days, when
the house is opened up, it is seen that the spot has diminished to
less than the requisite size, or has even vanished entirely, the house
is seen, none the less, as having been unclean during the seven days.
This can only mean that until we know for certain that a thing has
emerged from its presumptive status it retains that status. Thus
R.Jonathan proves that the hazakah principle enjoys Scriptural
authority.
The formulation in Soncino, p. 46, note 6: 'And as the house has acquired
the status of being unclean, it is presumed to remain so, and requires to
be "shut up" is very imprecise. The house does not acquire the status of
being unclean until it is 'shut up'. The hazakah is not of the house but of
the plague-spot. Since the spot did have the size of a bean when the priest
saw it, we assume, through application of the hazakah principle, that it
had remained of this size at the moment of'shutting up'.
The Talmud states that the fourth-century Babylonian Amora,
R. Aha b. Jacob, raised an objection to R.Jonathan' s proof. It is
possible that the Torah refers to the priest going out of the house
backwards so that he can observe the spot all the time and, at the
moment he shuts the door, can see that it has not decreased in size
to less than that of a bean. Thus there is no proof that the Torah
relies on hazakah.
Abbaye remarks to R. Aha b. Jacob: 'There are two replies to
your objection.' First, the Torah cannot refer to the priest going
out backwards since Scripture would not call such an abnormal
manner of exit 'going out'. Secondly, the Torah does not specify
where the spot is found and this is as if the Torah had stated that
the house is unclean wherever the spot is found, even if it is found
on the inside of the door itself. But if the spot is on the inside of
the door how can the priest see it, even if he does go out backwards?
And if you will argue, continues Abbaye, that the priest can open
up a window in the door through which he can look to see if the
spot has retained the requisite size, this will not do. For there is a
law stated in the Mishnah (Nega'im 2: 3) and derived from
Scripture that for the house to be unclean the priest must be able
to see the spot by the ordinary light of the house. The house is not
Presumptive state 113
unclean if he can only see the spot by opening a window. Thus,
Abbaye argues, R. Aha's attempted refutation is itself refuted and
R. Jonathan's proof stands.
Rava now seeks to defend R. Aha b. Jacob against Abbaye's
strictures. Rava argues that there is no proof from Scripture that
the hazakah principle is accepted since, as R. Aha b. Jacob has said,
Scripture may be speaking of the priest going out backwards. That
going out in this fashion is still referred to by Scripture as 'going
out', contrary to Abbaye, can be seen from the verse ' Then the
priest shall go out' (Leviticus 14: 38) referring to the High Priest's
exit from the Holy of Holies on the Day of Atonement. The
Mishnah (Toma 5:1) informs us that he used to go out backwards
in order not to have his back to the sacred ark and yet the Torah
refers to this as 'going out'. As for Abbaye's second objection
regarding opening a window, this can easily be met. The objection
is only where the spot cannot be seen at all unless the window is
opened. In our case the priest has already seen the spot by the
ordinary light of the house. The sole purpose of opening the window
is to see whether or not the spot behind the door has decreased from
the requisite size and for this purpose a window may be opened.
That is to say, even if we do not rely on the hazakah principle, as
we do not at this stage of the argument, the principle can at least
be followed for the purpose of permitting the opening of the
window. For here, once the window has been opened, the house
is declared definitely unclean not through the operation of the
hazakah principle but through direct observation.
Rava has now defended R. Aha b. Jacob against Abbaye's
strictures. But the Talmud quotes a Baraita (Sifra, metzora\ 6: 6)
from the Scriptural interpretation of which we can see that R. Aha
b. Jacob's interpretation is untenable. The Baraita states that from
the words ' Then the priest shall go out of the house' we might have
held that it is in order for the priest to go to his own house before
shutting up the affected house, i.e. by means of a long rope. But
since the verse continues ' to the door of the house' this shows that
the priest must be adjacent to the door when he shuts up the house.
On the other hand, we might have held that the priest may shut
up the house while standing underneath the lintel of the affected
house, therefore Scripture states 'shall go out of the house', i.e. he
must leave the house entirely. Consequently, he must leave the
house entirely and yet be adjacent to it, which means that he must
114 The Talmudic argument
stand beside the lintel. Now from this Baraita we learn that there
is a special Scriptural injunction - ' to the door of the house' - that
the priest must not shut the door by means of a long rope at a
distance. But on R. Aha b. Jacob' s interpretation this would be
invalid in any event, since how could the priest then see the spot
on the inside of the door even if the window had been opened ?
R. Aha b. Jacob is thus refuted.
The Talmud replies that R. Aha b. Jacob will hold that it is still
possible for the priest to know that the spot has not decreased since
there may be a row of men who call out that the spot has retained
its size.
Thus at the conclusion of the sugya there is still a possibility that
R. Jonathan's proof is refuted, as R. Aha b. Jacob refutes it. According
to this, one would have to say that there is, in fact, no Scriptural proof
for the hazakah principle and this is derived, see Maharsha, from ' a law
given to Moses at Sinai'. Cf. Rashi, Hullin 12a, s.v. pesah, that the same
conclusion is arrived at with regard to rubba in the sugya which follows ours,
see supra p. 60.
The stages in the sugya are:
(1) Where is the hazakah principle found ?
(2) R.Jonathan: plague-spot; may have decreased
(3) R. Aha b. Jacob: perhaps priest goes out backwards
(4) Abbaye: (a) backwards not called 'going out'
(b) what if behind door
(5) Rava: (a) it is called 'going out' - High Priest
(b) it can be behind door and window opened
(6) Refutation of R. Aha b. Jacob: verse dealing with 'long rope'
(7) Defence of R. Aha b. Jacob: row of men
Note how the Baraita in (6) has been left to the end. It is only
produced after Rava has defended R. Aha b. Jacob. Note, too, that
the whole discussion is purely theoretical and academic, since no
one really disputes that the principle of hazakah is accepted.
12
Gadol kevod ha-beriot: the law and regard
for human dignity
The sugya is Berakhot i gb-2oa. The term gadol kevod ha-beriot, lit.
'great is the dignity of human beings', occurs in the sugya and is
the central point of the whole discussion. The problem considered
is, under which circumstances can there be a relaxation of the law
where the purpose is to safeguard human dignity?
The sugya opens with a statement by the early-third-century
Babylonian Amora, R. Judah, in the name of his master, Rav: ' If
one finds kilayim (a mixture of wool and flax, Leviticus 19: 19) in
his garment, it must be stripped off even in the market place.' The
reason is: because Scripture says: 'there is no wisdom nor
understanding nor counsel against the Lord' (Proverbs 21: 30),
which means: 'Wherever there is a profanation of God's name no
respect is to be paid to a teacher', i.e. where regard for the dignity
of a teacher, and a fortiori an ordinary person, is in conflict with
the requirements of the law, it is the former that must yield.
An objection is raised from a Baraita dealing with a mourner and
the congregation which accompanies him on the return from a
funeral. If the mourner decides to return by way of an unclean path,
i.e. a path that crosses a field in which there are graves, the
congregation should follow him in that path out of respect for him
(i.e. even if some of them are priests who are forbidden to become
contaminated by a corpse or a grave (Leviticus 21: 1)). Thus we
see that a prohibition is set aside out of regard for human dignity
and thus contradicts Rav's ruling.
This follows the reading in our editions. Some texts (see SM and DS) have
a different reading, namely, if the congregation returns by way of the
unclean path, the mourner, even if he is a priest, should follow in that path
out of respect for them. The point, however, is the same. We see that a
prohibition is set aside out of regard for human dignity.
116 The Talmudic argument
To this the reply is given that R. Abba (see Hyman, Toledot, s.v.
Abba, that there were a number of Amoraim of this name)
interpreted (tirgemah) the Baraita as referring to a bet ha-peras, the
prohibition of which is only Rabbinic. A bet ha-peras is a field
containing graves that has been ploughed up so that the bones are
scattered and are no longer in one place as in the Biblical law of
contamination through contact with a corpse. According to Biblical
law, such a field does not contaminate so that a priest may enter
there; but the Rabbis ruled that it does contaminate and the priest
may not enter there. That is why the prohibition is set aside,
because there is only a Rabbinic prohibition in any event and a
Rabbinic prohibition is set aside for the sake of human dignity.
R. Judah in the name of Samuel is quoted as saying that a man
can walk through a bet ha-peras by blowing in front of him, so that
any bones that may be there will be blown out of his path and he
does not then suffer contamination at all, even by Rabbinic law.
And R. Judah bar Ashi is quoted as saying that if a bet ha-peras has
been thoroughly trodden under it no longer contaminates. All this
demonstrates that the rule of bet ha-peras is a lenient one because
the contamination and hence the prohibition is only by Rabbinic
law. There is no proof that a Biblical law, as in the case of kilayim,
is set aside for the sake of human dignity.
A further proof is now attempted, introduced with the usual
formula: ta shema\ 'come and hear'. The early Tanna R. Eleazar
bar Zadok is quoted as saying: ' We used to leap upon the coffins
in order to go out to meet Jewish kings and the Sages did not only
permit it in order to meet Jewish kings but also to meet Gentile
kings so that if one has merit [to see the Messiah] he will notice
the difference between [the majesty of] the Jewish king and the
Gentile kings.' Now R. Eleazar bar Zadok was a priest and the
meaning of his testimony is that a priest may contaminate himself
through contact with a corpse for the sake of human dignity, i.e.
to pay respect to royalty. To this the reply is given that the matter
can be understood on the basis of a statement by the fourth-century
Babylonian Amora, Rava. Rava interprets the Biblical law of the
' tent' : that if a corpse lies in a tent, or under any other cover, it
is forbidden for a priest to go over that tent or cover since to do
so makes him subject to contamination. Rava observes, however,
that since the majority of coffins have a space of at least one
hand's-breadth between the corpse and the coffin lid, this space acts
The law and regard for human dignity 117
as a barrier and the contamination is, as it were, contained.
According to Biblical law, then, it was permitted in any event for
R. Eleazar bar Zadok to leap on the coffins. The only prohibition
is Rabbinic, the Rabbis ruling that all coffins contaminate because
of the minority which do not have the space of a hand's-breadth
between the corpse and the lid. Since the prohibition is only
Rabbinic it was set aside for the sake of human dignity, but there
is still no proof that a Biblical prohibition is set aside.
A further proof is now attempted. A Baraita is quoted which
states: ' Great is human dignity [gadol kevod ha-beriot) for it pushes
aside a negative precept of the Torah' , which shows that even a
Biblical prohibition is set aside and thus refutes Rav. To this the
reply is given that Rav bar Sheva has interpreted (tirgemah) the
Baraita in the presence of R. Kahana (Babylonian Amora, early
fourth century) that the negative precept referred to is that of thou
shalt not turn aside' (Deuteronomy 17: 11), i.e. the command not
to disobey the teachers of Israel. But, continues the Talmud, ' they
laughed at him' since this, too, is a Biblical, not a Rabbinic,
prohibition. To this R. Kahana retorted: ' If a great man says
something do not laugh at him. All Rabbinic laws are based on
"thou shalt not turn aside" and out of respect for human dignity
the Rabbis permitted it.' Thus, again, there is only proof that a
Rabbinic prohibition is set aside for the sake of human dignity and
there is no refutation of Rav's ruling.
Rashi seems to understand the discussion as follows. Those who laughed
at Rav bar Sheva understood him to mean by ' thou shalt not turn aside'
cases where the teachers of Israel explain the meaning of a Biblical law.
Consequently, they laughed at him since it is a Biblical law that is being
set aside. But R. Kahana explained to them that Rav bar Sheva only
meant to say that all Rabbinic laws are based on ' thou shalt not turn aside'
but are none the less Rabbinic. A somewhat different understanding of the
passage is possible. Those who laughed at Rav bar Sheva did so because
since 'thou shalt not turn aside' is a Biblical injunction to obey Rabbinic
law it is a Biblical law that is being set aside. R. Kahana replies: true
though it is that the Torah enjoins us to obey Rabbinic law yet it is Rabbinic,
not Biblical, law that is being obeyed. This passage is the locus classicus for
the status of Rabbinic law; see Maimonides, Sefer ha-Mitzvot, shoresh rishon
and commentaries. Maimonides holds that there is a Biblical injunction
to obey Rabbinic law whereas Nahmanides disagrees, holding that the
meaning of our passage is that the Rabbis ' pegged on' their teachings to
the verse 'thou shalt not turn aside'. Nahmanides argues that if Rabbinic
law really enjoys Biblical authority, why does the law make numerous
118 The Talmudic argument
distinctions between matters of Biblical law and matters of Rabbinic law,
e.g. in cases of doubt and the like? Cf. the subtle analysis of the whole
question by R. Meir Simhah Kagan, Meshekh Hokhmah to Deuteronomy
17: 11, pp. 281-2, that, according to Maimonides, the command 'thou
shalt not turn aside' is only a general command to obey the Rabbis, not
a command to obey the details of Rabbinic law. These latter only enjoy
Rabbinic authority. Cf. Rabbi J. L. Bloch, She'urey Halakhah (Tel-Aviv,
1958), no. 10, pp. 52-6, that, according to Maimonides, Biblical law is to
be compared to a father instructing his son to do this or that. Each thing
the father orders is his direct command. Rabbinic law is to be compared
to a father instructing a son to obey the son's teacher. Here the command
is general.
A Baraita is now quoted dealing with the law of finding a lost
animal. From the word ve-hit 'alamata, 'hide thyself (Deuteronomy
22:1 and 4), a Rabbinic understanding is that there are three cases
where it is permitted to refrain from returning the lost animal. The
Baraita lists these as: where the finder is a priest and the animal
is in the cemetery; where the finder is an elder and it is beneath
his dignity to lead an animal; where the loss to the finder in
expenses is greater than the loss to the owner if the animal is not
returned. In all three cases the Torah declares: 'hide thyself, i.e.
do not bother to return the animal to its owner. From this we see
that the elder is not obliged to return the lost animal because it is
beneath his dignity; from which we see that even a Biblical
prohibition - not to refrain from restoring a lost animal - is set
aside for the sake of human dignity. To this the reply is given that
there is a special verse - ve-hit'alamta - to permit it. But why not
take this, asks the Talmud, as a paradigm for all prohibitions to
be set aside where human dignity is at stake? To this the reply is
given: 'Religious law cannot be derived from monetary matters',
i.e. it by no means follows that because the Torah allows a man
to place his own dignity before his neighbour's monetary interests,
the loss of his animal, that the Torah will allow a religious
prohibition to be set aside.
Another Baraita is now quoted. Here there is a comment on the
verse ' Or for his sister' (Numbers 6:7) in which it is specified that
a Nazirite (and even if he is also a priest) may not contaminate
himself for his dead sister but may contaminate himself for a met
mitzvah, a corpse there is no other person to bury. And by an
elaborate process (see Rashi) this is extended even to the Nazirite
and priest who is on his way to offer his Pachal lamb or to
The law and regard for hum,an dignity 119
circumcise his son. He, too, must contaminate himself for a met
mitzvah. From this we see that the Biblical precepts of offering the
Paschal lamb and of circumcision are set aside for the met mitzvah
(an example of regard for human dignity) and this shows that,
contrary to Rav, human dignity does take precedence even over
a Biblical command. To this the reply is given that there is a special
verse - ' Or for his sister' - which makes it an exception. But why
not derive other cases from this, i.e. since here it is a religious law
that is being set aside? To this the reply is given: ' there it is, sit
down and do not do' , i.e. it is indirect. The offence of failing to
offer the Paschal lamb or circumcise the child is only an indirect
one, whereas the offence of kilayim is a direct one, the garment being
worn to avoid loss of dignity. It by no means follows from the
permission of the Torah to have an indirect offence set aside for
the sake of human dignity that it permits the setting aside of a direct
offence for the sake of human dignity. Thus according to the
conclusion of the sugya there are three instances: (1) A positive
direct Biblical offence - kilayim. This may not be set aside for the
sake of human dignity, hence Rav's ruling. (2) A direct but
Rabbinic offence - bet ha-peras and skipping over the tombstones.
(3) A Biblical but indirect offence - the Paschal lamb and circum-
cision. This can be set aside. In other words two conditions must
obtain for Rav's principle to come into operation. These are: (1)
it must be a Biblical offence; (2) it must be direct. An indirect
offence, even if it is Biblical, and a Rabbinic offence, even if it is
direct, may be set aside in order to safeguard human dignity.
See Rashi for the difficulty that when the Nazirite and priest contaminates
himself it is direct and Biblical. Rashi observed that the law of contamination
is not simply set aside for the Nazirite and priest. The law of contamination
does not apply here at all. Just as the Torah permits a priest, for instance,
to contaminate himself for his relatives it permits him to contaminate
himself for a met mitzvah. The man going to offer the Paschal lamb or to
circumcise his son, on the other hand, does not enjoy any special status.
His going for the purpose just happens to coincide with his discovery of
the met mitzvah. Here it is a case of a law being set aside for the sake of human
dignity and is only permitted because it is indirect, as the Talmud says.
From our sugya it would follow that if A saw B wearing kilayim there is no
need for A to do anything about it, since the offence he commits by doing
nothing is an indirect one. However, Maimonides (Tad, Kilayim 10: 29)
evidently did not have the reading 'in his garment' but simply: ' One who
finds kilayim' and he understands our sugya as referring not to the wearer
but to the one who observes it. This is called 'direct' because the wearer
120 The Talmudic argument
commits a direct offence. Maimonides also adds ' even if he [the wearer]
is his teacher', i.e. Maimonides understands this to be the reference to ' no
regard is to be paid to a teacher'. Cf. the famous Responsum of Ezekiel
Landau, Noda' Bi-Thudah, Kama, Orah Hayyim, no. 35, who discusses the
question whether a husband should be informed of his wife's infidelity
where to do so would offend human dignity. The Baraita: 'Gadol kevod
ha-berioV is quoted in Shabbat 81b; 94b; ' Eruvin 41b; Megillah 3b; Menahot
37b but our sugya is the only discussion on the whole theme of human
dignity. The saying: 'Whenever there is a profanation of God's name no
regard is paid to the dignity of a teacher' does not appear to be Rav's but
is an addition by the editors and is, in fact, a quote from elsewhere. In
fact, in the other instances where this occurs in the Talmud ('Eruvin 63a;
Sanhedrin 82a; Shevu'ot 30b) it appears on its own and in connection with
the dignity of scholars not in connection with the general theme of human
dignity. As above, this is no doubt why Maimonides adds: ' even if he is
his teacher'. On our reading it is difficult to know why a teacher should
be mentioned here at all unless, as we have suggested, this is really a quote
from an independent saying and has simply been tagged on to the verse
because it is a familiar comment on the verse. The real point of the sugya
is the concern with 'human dignity' (kevod ha-beriot) not with ' the dignity
of scholars' (kevod ha-rav). It should also be noted that in the instances given
in the three passages the question is with regard to the special dignity
reserved for scholars, e.g. whether or not a scholar should demean himself
in order to testify in a Court of Law; whereas in all three instances in our
sugya the question concerns general human dignity, e.g. the burial of a
corpse; being unclothed in public; respect for royalty. (The case of the elder
and the animal is not because of respect for scholarship but because since
he is the kind of person who does not lead animals in the street to expect
him to do so is an affront to general human dignity.) Thus the best way
of understanding the saying is to see it as a quote, not strictly germane
to our theme, that has become so closely associated with the verse in
Proverbs that it is quoted as if it were part of that verse. The pattern of
the sugya is:
(1) R. Judah in the name of Rav: Kilayim, must strip off
(2) Reason: verse in Proverbs and ' wherever. . . '
(3) Objection: mourner
(4) Reply: R. Abba; bet ha-peras; only Rabbinic
(5) Objection: R. Eleazar bar Zadok: Biblical
(6) Reply: no, also Rabbinic: Rava, space
(7) Objection: 'negative precept of the Torah\ Biblical
(8) Reply: 'thou shalt not turn aside'
(9) Objection: 'They laughed'; Biblical
(10) Reply: R. Kahana; also Rabbinic
(11) Objection: elder; Biblical
(12) Reply: special verse
(13) Objection: derive from it other Biblical offences
(14) Reply: religious law different
The law and regard for human dignity 121
(15) Objection: Paschal lamb
(16) Reply: special verse
(17) Objection: derive then other Biblical, religious offences
(18) Reply: indirect
One can see, even at a cursory glance, that the series of proofs
has been presented in a neat, artificial sequence. The reply in (4)
is not relevant to (5) since here the offence is Biblical, hence (6),
here, too, Rabbinic. But what of (7) where 'of the Torah' is
mentioned? Reply, as in (8) to (10). But what of (11) where it is
clearly Biblical? Replies (12) to (14), yes, but there it is a monetary
offence here a religious one. But, (15) to (17), religious offence and
Biblical? Reply, (18), indirect. It is obvious that the editors have
presented the material they had so as to provide this neat pattern
of proof, refutation and counter-refutation. Each item is only put
forward when its time has come, as it were, when it implies a further
refutation to which the previous reply is inoperative.
13
Hazmanah milta: whether the designation of an
object for a particular use is effective
The sugya is Sanhedrin 47b~48b. Hazmanah means ' to set aside',
' to designate'. The basic question considered is whether an object
that acquires a certain status in law when used for a particular
purpose acquires that status even when it has only been designated
for that purpose. The example given is the weaving of a garment
(a shroud) for a corpse. It is forbidden to enjoy any benefit from
a corpse, its grave or its shroud. Abbaye holds that the mere
designation for the purpose suffices to render the garment
forbidden - hazmanah milta hi, lit. 'hazmanah is a thing', i.e. it has
substance so that the garment is forbidden as if it had actually been
used. Rava holds that mere designation is insufficient to render the
garment forbidden - hazmanah lav milta hi, lit. ' hazmanah is not a
thing', i.e. has no substance, is ineffective, and the garment is not
forbidden until it has actually been used as a shroud.
The sugya begins with the debate between Abbaye and Rava: ' It
has been stated: If one weaves a garment for a corpse: Abbaye said:
It is forbidden. But Rava said: It is permitted. Abbaye said it is
forbidden; hazmanah is effective. But Rava said it is permitted;
hazmanah is ineffective.'
The term itmar, * It has been stated', the usual Talmudic term to introduce
a debate between Amoraim is obviously an editorial device. From the fact
that ' Abbaye said... But Rava said' is repeated, it would seem to be
equally certain that the explanation of Abbaye and Rava's views on
whether hazmanah is effective or ineffective is not Abbaye and Rava's own
but it similarly editorial. Thus the paragraph should be paraphrased as:
'Abbaye said: It is forbidden. But Rava said: It is permitted. The reason
why Abbaye said that it is forbidden is because Abbaye holds that hazmanah
is effective while the reason Rava said that it is permitted is because Rava
holds that hazmanah is ineffective.' When the Talmud continues with:
Effectiveness of mere designation 123
'What is the reason?' it introduces a new feature, namely, the support the
Amoraim find for their views in Scripture and why they disagree with each
other's derivation from Scripture. It is impossible to know how much of
all this is Abbaye's or Rava's own and how much the work of the editors,
but the actual wording appears to be that of the editors with the exception
of: ' It is forbidden' and ' It is permitted.' On this kind of analysis into the
original statements of the Amoraim and the editorial elaborations, see H.
Klein, articles on 'Gemara' and 'Sebara', JQR, new series, 38 (1947),
67-91; 43 (1953), 341-63; JSS, 3 (1958), 363-72.
The Talmud proceeds to ask: ' What is Abbaye's reason?' i.e.
on what authority does he base his ruling. The reply is that the word
sham, ' there' , occurs in a verse which speaks of the dead: ' And
Miriam died there, and there was she buried' (Numbers 20: 1; my
italics) as well as in the verse which speaks of the heifer whose neck
was broken: ' and shall break there the heifer's neck' (Deuteronomy
21:4, my italics). Now the rule is that the heifer becomes forbidden
for any use from the moment it is brought down there, i.e. to the
valley and so it follows, according to Abbaye, that the garment
woven for the corpse is forbidden by the mere act of designation
just as the heifer becomes forbidden even before it has actually been
beheaded. Rava relies on another verse in which there is a reference
to ' there' . This verse speaks of idolatry: ' Ye shall surely destroy
all the places where the nations ye are to dispossess serve there their
gods' (Deuteronomy 12: 2; my italics). Now if an object or an
animal is set aside to be offered to an idol, it does not become
forbidden until it has actually been used for the purpose. Conse-
quently, Rava derives from this his rule that the garment does not
become forbidden merely because of designation. We thus have two
Scriptural verses. In one - that of the heifer - the object does
become forbidden merely by designation. In the other - that of
idolatry - the object does not become forbidden merely by desig-
nation. The question, then, is whether the law of the garment is
to be derived from that of the heifer or from that of idolatry.
The Talmud now asks further why Rava fails to agree with
Abbaye, why, in other words, does Rava prefer to derive the law
from idolatry and not from the heifer. The reply is given, Rava
prefers to derive accessaries from accessaries (i.e. the garment is for
the corpse and the object for the idol) whereas the heifer is itself
a forbidden object and can, therefore, the more readily become
forbidden merely by designation. Abbaye, on the other hand,
prefers to derive the ' normal' from the ' normal' (i.e. laws to be
124 The Talmudic argument
observed as they come about in the normal course of human life)
whereas idolatry is ' abnormal' (forbidden even to Gentiles, Rashi).
Here mere designation is insufficient since the man so designating
for an' abnormal' purpose may change his mind and does not really
mean it until he actually does it. Thus according to Abbaye all cases
are derived from the law of the heifer that mere designation is
forbidden and the law of idolatry is the sole exception, since in that
case there is no real designation (because it is ' abnormal' and he
may change his mind). But according to Rava, all cases of
accessaries are derived from the law of idolatry and there are no
exceptions. The heifer is no exception since it is not an accessary
but the forbidden thing itself.
An attempt is now made to refute Rava from a Mishnah (Kelim
28: 5) dealing with the law ofmidras, 'resting'. If a person with an
issue (see Leviticus 15) sits or rests on any object that is sometimes
used for this purpose that object becomes unclean. But if the object is
used for other purposes, and not for sitting, the law ofmidras does not
apply, e.g. scales or axes because these
c
say' to the man sitting, as it
were:' Stand up and we will do our work.' The Mishnah states that
if a veil has been sat on it does become subject to midras, because
people sometimes fold their veils and sit on them. But once the veil
has been designated (lit. 'given') to be used as a mantle to wrap
around a book it loses its midras character, becoming an object that is
used entirely for a purpose other than sitting. (The point here is that
normally a veil is used for both purposes, as an item of dress and as
something on which to sit, when it is folded. But a mantle wherewith
a book is covered is never taken from it to be used for sitting on.)
From this we see that mere designation is effective and this refutes
Rava. Say rather, the Talmud replies, the veil has not only been
designated but actually used as a wrapper for the book, i.e., the
Mishnah refers to both designation and subsequent use. In that case,
asks the Talmud, since the veil has actually been used why refer to
designation at all? To this the reply is given that use on its own, with-
out prior designation for the purpose, does not suffice, just as desig-
nation on its own, according to Rava, does not suffice. A rule of R.
Hisda to this effect is quoted. R. Hisda ruled that if a cloth had been
designated as a wrapper for tefillin and had then actually been used
as a wrapper for tefillin it is forbidden to wrap money therein. But
if it had been designated but not actually used, or used without
Effectiveness of mere designation 125
prior designation, it is permitted to wrap money therein. But, con-
tinues the Tal mud, according to Abbaye, if it had been designated it
would be forbidden even if not actually used, whereas even if it
had been used it is permitted if there had been no prior designation.
A Baraita is now quoted. Here it is said that if a tomb had been
built for a living person (i.e. for a man now alive, to be used for
his burial when he dies) it is permitted to have use of it (since it
is not, in fact, for a corpse, the man being still alive). But if even
a single row of stones had been added for a corpse (i.e. the row was
added after the man had died) it is forbidden. This refutes Rava
since the addition of a row of stones is no more than a designation.
The reply is given that the Baraita means: it is forbidden if the corpse
is already there in the tomb. To this the Tal mud objects: if the
corpse is already there in the tomb, the tomb would be forbidden
even if no further stones were added since it is now a t omb that
contains a corpse. The Tal mud replies t hat the corpse was
subsequently removed from the tomb. Hence, when a row of stones
has been added there is both designation and use, but where no
stones have been added there is only use but no designation and
this is analogous to R. Hisda' s case of the tejillin. The Tal mud now
adds a ruling of Rafram bar Pappa in the name of R. Hisda, to
the effect that if the row that has been added can still be identified
that row can be removed and the rest of the tomb is then permitted
(once the corpse has been removed). The point here is, since,
according to R. Hisda, there has to be both use and designation
for it to be forbidden, then although the whole of the t omb has been
used it is only the row that has been added that has been
designated, so that once the row has been removed the rest of the
tomb is permitted, on the grounds that there has been use but no
designation. That row of stones, however, is forbidden even after
its removal since so far as that row is concerned there has been both
use and designation.,
A Baraita is now quot ed: ' If a man hews out a grave for his father
but then buries his father in another place he must never use that
grave to bury therein' , i.e. it is forbidden to use t hat grave to bury
someone else therein since to do so would be to have benefit from
a grave. This refutes Rava, since the grave has only been designated
for the father's corpse and has not actually been used. The reply
is given that the prohibition is not because o f benefit' from a corpse
126 The Talmudic argument
and if the grave had been designated for another corpse, not the
father, it would, indeed, be permitted. The reason it is forbidden
is out of respect for the father, i.e. once the son had set the grave
aside for his father it is disrespectful to his father's memory to use
it to bury therein someone else. And, the Talmud continues, this
explanation is in any event preferable since the Baraita continues:
' Rabban Simeon b. Gamaliel says: Even if a man merely hews out
stones for his father's grave and then buries his father in another
grave, he must never used that grave (into which the stones will
be placed) for burial.' Now if the reason is because of respect for
the father we can understand why Simeon b. Gamaliel forbids even
the stones to be used. He evidently holds that it is disrespectful for
the son to use even the stones. But if the reason is because of
hazmanah, 'designation', it would then follow that R. Simeon b.
Gamaliel forbids even the stones on the same grounds of hazmanah
and this is impossible since even Abbaye only forbids a garment that
has been woven for a corpse. Abbaye would not forbid the yarn
spun to be woven into the shroud. Abbaye only holds designation
to be effective where the object designated can be used for the
purpose that renders it forbidden. He does not forbid that which
is designated to be designated, namely, the yarn for the shroud for
the corpse and the stones for the tomb for the corpse.
Soncino translates: ' If one hews a grave for his father and buries him
elsewhere, he [himself] may never be buried therein.' This would seem to
be the interpretation of the Tur, Toreh De'ah 364. But the difficulty about
this interpretation is that it seems to place a prohibition on the son when
the son is dead, which seems absurd, unless the meaning is that the son
must never be buried there so as not to have the satisfaction while alive of
knowing that he will be buried there. Maimonides (Tad, Evel 14: 20)
formulates it as: 'he should never bury another corpse there', i.e. the son
must not use the tomb to bury therein another corpse because he would
then be benefiting from the tomb.
A further Baraita is then quoted, this time to refute Abbaye. The
Baraita reads: ' It is permitted to have benefit from a new grave
but if an abortion has been cast therein it is forbidden.' This shows
that hazmanah is ineffective. The new grave is not forbidden until
the abortion has been cast there. The reply is that a new grave is
in any event forbidden because it has been designated as a grave.
When the Baraita speaks of a new grave it means a grave not set
aside at all for a corpse but simply to be used in the future when
required. If it had been set aside for a corpse it is forbidden, as
Effectiveness of mere designation 12 7
Abbaye states. In that case why does the abortion have to be cast
therein? Why does not the mere designation for the abortion suffice
to render it forbidden? It is because we might have argued that,
as Rabban Simeon b. Gamaliel, who is quoted, actually holds, an
abortion does not render a grave forbidden. According to Rabban
Simeon b. Gamaliel a grave containing an abortion is only
forbidden while the abortion is there and not because of the grave
but because of the abortion itself. Once the abortion has been
removed from the grave the grave is permitted. That is why the
first Tanna speaks of the abortion being cast there. He personally
holds that the grave is forbidden even if there is only the designation
of the grave for the abortion. But he speaks of it being cast there
to show that Rabban Simeon b. Gamaliel holds that even if it has
been cast there the grave is permitted once it has been removed.
A Mishnah (Shekalim 2: 5) is now quoted to refute Abbaye. The
Mishnah rules that if money was collected for the purpose of
burying the dead poor any surplus must also be used for burying
the dead and for no other purpose. But if the money had been
collected to bury a particular person then any surplus belongs to
his heirs, it being assumed that the donors do not wish to have this
returned to them but are content that it be enjoyed by the dead
man' s family. But if Abbaye is correct, all the money had been
designated for the dead so why is the surplus permitted ? The reply
is, at first, given that the Mishnah refers to money collected while
the man was still alive. But this cannot be, the Tal mud observes,
since there is a Baraita in which it is stated explicitly t hat the
reference in the Mishnah is to money collected after the man had
died.
Yes, the Tal mud now implies, the Mishnah does seem to
contradict Abbaye' s ruling. But it has to be appreciated t hat the
Mishnah states in its conclusion that R. Meir and R. Nat han
disagree with the ruling of the first Tanna of the Mishnah. R. Meir
holds that the surplus must not be touched ' until Elijah comes' and
R. Nat han holds that the surplus must be used only for the dead
man, i.e. to build a monument over his grave or sprinkle aromatic
spices over it. Thus while it is true that the first Tanna seems to
contradict Abbaye, R. Nat han appears to contradict Rava.
Consequently, both Abbaye and Rava will be obliged to explain
the Mishnah, each according to his view. This is how Abbaye will
explain it. All three Tannai m hold, as Abbaye does, that hazmanah
128 The Talmudic argument
is effective. The question at issue is whether the original hazmanah
of the donors was of all the money or only of the actual expenses,
not of the surplus. The first Tanna holds that only the actual
expenses were designated but the surplus was never designated for
the corpse and hence the heirs may have it. R. Nathan holds that,
on the contrary, the whole of the money, including the surplus, is
designated, hence the surplus may only be used for a monument
and so forth. R. Meir is uncertain whether the surplus is designated,
in which case the heirs may not have it, or whether it was not
designated, in which case they may. Consequently, R. Meir holds
that we have to wait until Elijah comes and he will be able to inform
us whether or not the surplus was designated.
Rava, on the other hand, will hold that all three Tannaim will
agree with him that hazmanah is ineffective. In that case why does
R. Nathan hold that the surplus is forbidden to the heirs? It is not
because of hazmanah but for an entirely different reason. The man
has been humiliated by having to be the object of charity and while
he is prepared, while alive, to have the money collected for his
actual burial he does not wish to suffer this humiliation for the sake
of the surplus if that goes to his heirs. This is R. Nathan's view, i.e.
the man does not, in fact, bequeath this surplus to his heirs. The
first Tanna disagrees and holds that he does. R. Meir is uncertain
whether he does not and so he rules that it must await the coming
of Elijah who will be able to tell us whether he does or does not.
It appears from this passage that money designated for the dead, according
to Abbaye, is treated like the woven garment not the spun yarn. The reason
is probably because the designation of the money is more immediate in
that the money can now be used for the dead, unlike the yarn, see Tosqfists,
s.v. motor ha-met.
A Baraita is now quoted which seems to refute Abbaye. The Baraita
states that if a father and mother cast garments on to the bier of
their son (as if to say, nothing is worth anything to us now that
we have lost our son) it is meritorious for others to rescue these
garments and return them to the parents when their grief will have
subsided. This refutes Abbaye. To this the reply is given, the
parents only do it out of'bitterness', i.e. to express their grief but
do not really mean to designate the garments for the corpse. But,
the Talmud objects, the Baraita continues: ' Rabban Simeon b.
Gamaliel says: When is this [that the garments should be rescued],
only when the garments have not touched the bier? If they have
Effectiveness of mere designation 129
touched the bier they are forbidden.' But if the parents only do it
out of grief and do not really mean the garments to be designated
for the corpse they should be permitted even if they have touched
the bier. To this the reply is given, that, indeed, strictly speaking
they are permitted but the Sages placed a prohibition on them
when they have actually touched the bier because the reference is
to the bier which will be buried with the corpse. Here the garments,
if they have touched the bier, may be confused with the actual
shrouds and if people see that the Sages permit their removal they
might mistakenly conclude that even shrouds are permitted.
Another Baraita is now quoted dealing with tejillin. This reads:
' If a bag had been made to place tejillin therein it is forbidden to
place money therein. But if tejillin had been placed (temporarily)
in a bag it is permitted to place money therein.' This shows that
hazmanah is effective and refutes Rava. The reply is that the Baraita
means, the bag was not only made for the tejillin but the tejillin were
actually placed there as well. Thus the Baraita accords with the view
of R. Hisda that both designation and use are required for it to
be forbidden.
Another Baraita (Tosefta, Megillah 2) is now quoted. Here it is
stated that if a man instructed a craftsman to make him a sheath
for a scroll of the Law or a receptacle for tejillin it is permitted
to use these for secular purposes until they have actually been used
for their sacred purpose. This proves conclusively that hazmanah is
ineffective and Abbaye is refuted. The reply is given that, indeed,
the Baraita holds that hazmanah is ineffective but Abbaye can
produce another Baraita from which it is clear that the issue is
debated by Tannaim. The Baraita is then quoted: 'If tejillin were
overlaid with gold or covered with the hide of an unclean animal
they are unfit for use. If they were covered with the hide of a clean
animal they are permitted for use even if the hide had not been
dressed for the purpose. But Rabban Simeon b. Gamaliel holds that
they are unfit for use even if they had been covered with the hide
of a clean animal if the hide had not been dressed for that purpose.'
We see that according to the first Tanna there is no need for the
hide to be dressed specifically for the purpose of tejillin, while
Rabban Simeon b. Gamaliel holds that it has to be dressed
specifically for the purpose. It follows that according to the first
Tanna hazmanah is ineffective. Since tejillin do not require any
specific designation, any designation for the purpose of tejillin is
130 The Talmudic argument
irrelevant. It has no meaning, so that a receptacle made for tefillin
is not forbidden even according to Abbaye. But according to
Rabban Simeon b. Gamaliel tefillin do require designation and acco-
rding to him, hazmanah, being relevant, is effective (agreeing with
Abbaye). Thus all agree that where hazmanah is relevant it is
effective, as Abbaye holds. The only reason why it is ineffective in
the case of tefillin, according to one authority, is because there
hazmanah is irrelevant.
This is RashVs understanding of the passage. Tosqfists, s.v. af 'al pi,
understands it differently. The first Tanna who does not require the hide
to be dressed for the purpose holds that hazmanah is effective, hence the
designation of the hide for tefillin suffices and there is no need for the hide
to be dressed specifically for the purpose. Rabban Simeon b. Gamaliel
holds that the hide does have to be dressed specifically for the purpose
because mere hazmanah for the purpose is ineffective.
The Talmud, having brought the discussion to a successful
conclusion reports that Ravina asked Rava: ' Is there any place
where the dead lie unburied while the shrouds are being woven?',
i.e. the case debated by Abbaye and Rava is extremely unlikely.
The shrouds are always ready to hand. Rava replied: ' It can
happen in connection with the dead of Harpania' (where the
people are so poor that they wait until someone dies before
collecting the money to buy shrouds and these, therefore, have
sometimes to be woven while the corpse lies there: Rashi).
The sugya concludes with an exposition of the final ruling. This
reads: 'Meremar expounded (darash): "The law follows Abbaye."
But the Rabbis said: "The law follows Rava." And the law follows
Rava.' This small section appears to have been inserted after the
completion of the whole sugya, i.e. once the full debate between
Abbaye and Rava had been recorded, including the arguments for
and against, the editors inform us that the late-fifth-century Amora
'expounded' (the usual term for a clear, public pronouncement of
the actual ruling) that the ruling is in accordance with the opinion
of Abbaye. The other scholars (the Rabbis), however, disagreed
and 'said' (not 'expounded', because they, evidently, occupied no
official, authoritative position as Meremar did) that the ruling
follows the opinion of Rava. The final conclusion ' And the law
follows Rava' was almost certainly added even later (it is probably
a post-Talmudic addition), i.e. after the debate between Meremar
and the Rabbis had been recorded.
Effectiveness of mere designation 131
The mnemonic at the beginning of the series of proofs is also
almost certainly post-Talmudic, having been inserted when the
complete sugya as we now have it had been fully arranged and put
in its present order.
It is somewhat curious that there are no less than four references
to Rabban Simeon b. Gamaliel in this comparatively brief sugya and
they are from diverse sources. The best way of explaining this
phenomenon is that all four deal with laws having to do with
' setting aside' and the editors used this material in building up the
sugya of hazmanah, interpreting the various rulings of Rabban
Simeon b. Gamaliel accordingly. It is still odd that apart from the
law of the abortion the statements of Rabban Simeon b. Gamaliel
are not found in any other passage, Tannai t i c or otherwise. For all
that it would be hypercritical to suggest t hat the editors have simply
invented these statements, using the name of Rabban Simeon b.
Gamaliel because of the abortion law in which he features. The
most probable explanation is as we have suggested, that Rabban
Simeon b. Gamaliel was known to have stated these four laws but
not necessarily in the sense in which our sugya understands them.
The editors simply built on this material to further their own
dramatic aim.
The plan of the sugya, after the statements about the debate
between Abbaye and Rava, is as follows:
(1) Refutation of Rava (a): Mishnah of veil - mere hazmanah
(2) Reply: No - actual use
(3) Elaboration: R. Hisda - both hazmanah and use required
(4) Refutation of Rava (b): Baraita - row of stones - mere hazmanah
(5) Reply: corpse there - both hazmanah and use
(6) Elaboration: corpse was later removed
(7) Rafram's statement: independent of sugya but relevant
(8) Refutation of Rava (c): grave for father - mere hazmanah
(9) Reply: reason is out of respect for father
(1 o) Elaboration: proof that this must be the reason
(11) Refutation of Abbaye (a): new grave and abortion
(12) Reply: because of Rabban Simeon b. Gamaliel
(13) Refutation of Abbaye (b): Mishnah: surplus
(14) Reply: tentative: collected while alive
(15) Objection: contradicts Baraita
(16) Reply: in any event R. Meir and R. Nathan
(17) Consequently: both Abbaye and Rava explain it
(18) Refutation of Abbaye (c): father and mother casting garments
(19) Reply: because of' bitterness'
(20) Objection: then even if they have touched the bier
132 The Talmudic argument
(21) Reply: because then confused with shrouds
(22) Refutation of Rava (d): tefillin
(23) Reply: actually used - R. Hisda
(24) Refutation of Abbaye (d): receptacle of tefillin
(25) Reply: debate of Tannaim
(26) Ravina's question: how is the case at all possible?
(27) Rava's reply: dead of Harpania
(28) Final ruling: Meremar, the Rabbis and final ruling
Why has all this material been presented in precisely this way,
since the actual proofs have been taken from varied sources? The
answer is fairly obvious: because there is here a logical sequence of
thought. The reply in (2) does not apply in (4). Yes, it does, because
of elaboration in (6). But the reply does not work in (8), where the
father is not buried there and hence is not used. Turns now to
refutation of Abbaye ( n) . Reply as in (12). But what of (13) ? Then
(18), further refutation of Abbaye to which none of the earlier
replies is cogent. All of these have to do with corpse except (1),
which must come first in order to lead up to the others. Now (22),
dealing with a new subject, tefillin. Reply as in (23). Refutation of
Abbaye (24) and the strongest, since the only reply, as in (25), is
that the Tannaim debate it.
The theme ofhazmanah is found, too, in Berakhot 23b (the law of R. Hisda
and discussion on it as in our sugya); Megillah 23b (distinction as in our
sugya between yarn and woven garment); Menahot 34b. Cf. Nedarim ja.. For
a full treatment of the post-Talmudic discussions, see Kesef Nivhar, part 1,
no. 48, pp. 84a-86a and ET, vol. vm, s.v. hazmanah alef, pp. 623-45.
Mitzvat 'aseh she-ha-zeman geramah: positive
precepts dependent on time from which
women are exempt
The sugya is Kiddushin 33b~35a. Mitzvat 'aseh she-ha-zeman geramah
is literally: ' A positive precept that is caused by time', i.e. that
depends for its performance on a given time and is not a precept
that can be carried out at any time. The sugya is appended to the
Mishnah {Kiddushin 1:7): 'And every positive precept dependent
on time men are obliged to fulfil but women are exempt. And every
positive precept not dependent on time both men and women are
obliged to fulfil. And every negative precept, whether dependent
on time or not dependent on time, both men and women are
obliged to fulfil.'
The sugya opens with a Baraita: 'Which are positive precepts
dependent on time: sukkah (Leviticus 23: 42, only to be carried out
on Tabernacles); lulav ('palm branch, Leviticus 23: 40, only on
Tabernacles); shofar (ram's horn', Leviticus 23: 24; Numbers 29:
1, only on the New Year Festival); tzitzit ('fringes', Numbers 16:
38, only during the day, not at night); and tefillin (Deuteronomy
6: 8, only during the days and not the nights of the week and not
on Sabbath and Festivals). And which are positive precepts not
dependent on time: mezuzah (Deuteronomy 6: 9); ma'akeh ('battle-
ment', Deuteronomy 22: 8); returning lost property (Exodus 23:
4; Deuteronomy 22: 1-3); and sending away the mother-bird
(Deuteronomy 22: 67).'
The Talmud objects: but is this a general rule? There are: eating
unleavened bread on Passover (Exodus 12: 18, obligatory for
women, too, just as they are obliged to abstain from eating leaven);
rejoicing on the Festivals (Deuteronomy 16: 14); and assembly (on
the Festival of Tabernacles in the seventh year, Deuteronomy 31:
12). These are positive precepts dependent on time and yet women
134 The Talmudic argument
are obliged to fulfil them. Furthermore, there are the study of the
Torah, the duty of procreation and the redemption of the first-born.
These are positive precepts not dependent on time and yet women
are exempt from having to fulfil them. Thus although the Baraita
purports to give two general rules - one for precepts dependent on
time, the other for precepts not dependent on time - neither is
completely accurate, since there are exceptions to both. To this the
reply is given that R. Johanan says: ' One does not learn from [a
statement of a] general principle even when exceptions are stated',
i.e. so that it would appear that these are the only exceptions. There
are, in fact, or can be, other exceptions as well. In support a
Mishnah ('Eruvin 2: 1) is quoted and R. Johanan's comment
applied to it. The Mishnah deals with the law of 'eruv, the device
by means of which it is permitted for people with houses around
a common courtyard to carry therein on the Sabbath. Some food,
belonging to all the householders, is deposited in one of the houses
on the eve of the Sabbath, and, by a legal fiction, this renders the
courtyard as if it were a private domain in which it is permitted
to carry. The Mishnah states that an 'eruv can be of any food except
water and salt. But, the Talmud observed, there is also the
exception of mushrooms and truffles which cannot be used for the
l
eruv. To this R. Johanan is quoted as replying:' One does not learn
from [a statement of a] general principle even when exceptions are
stated.' Hence, even though the Baraita, quoted at the beginning,
lays down general rules, there are exceptions to these rules.
R. Johanan's statement occurs also in 'Eruvin 27a (on the Mishnah quoted
here) and 'Eruvin 29a, which is a cross-reference: 'As R. Johanan said'. A
comparison of the first 'Eruvin passsage with ours is revealing. There the
sugya opens with R. Johanan's statement. The Talmud then observes that
R. Johanan cannot have made his statement in the first instance as a
comment on the 'Eruvin Mishnah, since the clause 'even when exceptions
are stated' seems to imply that R. Johanan's original statement was with
reference to a Tannaitic formulation of a general rule in which no
exceptions are recorded and which he then elaborates on by observing that
his observation would apply even where exceptions are stated. This means
that he cannot have made his observation on the 'Eruvin Mishnah in which
exceptions are stated explicitly. To this the reply is given that R. Johanan's
original observation must have been made on our Baraita, which is then
quoted as here. We now have the full statement of R. Johanan, i.e. that
his observation applies even when exceptions are noted. There follows an
attempt by Abbaye or R. Jeremiah to prove the correctness of this. Finally,
Ravina or R. Nahman prove it from the 'Eruvin Mishnah. Thus the same
Exemption of women from some precepts 135
material has been used both in our sugya and in * Eruvin, but this has been
arranged differently in two tractates. The fuller discussion is in 'Eruvin. Cf.
Abraham Weiss, le-Heker ha-Talmud, pp. 370-2 and Halivni, Sources and
Traditions, p. 654, note 2.
There now begins a series of comments on the first rule of the
Mishnah, that women are exempt from carrying out positive
precepts dependent on time. How do we know this? The reply is
given: it is derived from tejillin. Just as women are exempt from
tejillin, which is a positive precept dependent on time, so, too, they
are exempt from all such precepts. And that women are, in fact,
exempt from tefillin is derived from the fact that in the Shema' tefillin
are compared to the study of the Torah: ' And thou shalt teach them
diligently unto thy sons, and shalt talk of them when thou sittest
in thy house. . . And thou shalt bind them for a sign' (Deuteronomy
6: 7-8). Now of the study of the Torah the verse states: ' unto thy
sons' and not ' thy daughters'. Women are thus exempt from the
study of the Torah and since tefillin are compared to the study of
the Torah (being placed adjacent to this in the verse) they are
exempt from tefillin. Once we have established in this way that
women are exempt from tefillin we can go on to derive other
precepts from tefillin. Just as women are exempt from tefillin they
are exempt from every positive precept dependent on time. The
Talmud goes on to point out, however, that there is a reference to
mezuzah in the next verse:' And thou shalt write them upon the posts
of thy house, and on thy gates' (Deuteronomy 6: 9). Thus while
it is true that tefillin are compared to the study of the Torah,
mentioned in the verse preceding that of tefillin, it is also true that
tefillin are compared to mezuzah, mentioned in the verse following
tefillin. Why, then, should we not rather compare tefillin to mezuzah
and deduce from it that women are obliged to carry out the duty
of tefillin? To this the reply is given that there are two separate
sections of the Shema
i
: Deuteronomy 6: 4-9 and 11: 13-21. In the
first section tefillin are, as above, placed between the study of the
Torah and mezuzah, but in the second section tefillin are in verse 18,
study of the Torah in verse 19, and mezuzah in verse 20. Thus tefillin
are adjacent to the study of the Torah in both sections but only
adjacent to mezuzah in the first section. Consequently, it is preferable
to compare tefillin to the study of the Torah and not to mezuzah.
Then, the Talmud asks, why not compare mezuzah to the study of
the Torah, to which it is adjacent in the second section, and women
136 The Talmudic argument
should be exempt from mezuzah? To this the reply is given that
Scripture continues, after stating the law of the mezuzah: ' that your
days may be multiplied' (Deuteronomy 11: 21). Is it only men who
require life and not women? I.e. since the result of carrying out this
precept is long life it must obviously apply to both men and women.
We have now established tefillin as the key-text. From it we derive
the rule of the Mishnah that women are exempt from positive
precepts dependent on time. The Talmud now turns to explore this
further. Of the sukkah Scripture says: 'ye shall dwell in booths
[sukkot] seven days' (Leviticus 23: 42). Thus sukkah is a positive
precept dependent on time and, according to the rule, women are
exempt from it. They are, indeed, exempt but the exemption is
derived from the second half of the verse: ' all that are home-born
in Israel shall dwell in booths'. A Baraita {Sukkah 28a) derives the
exemption of women from the word ha-ezrah ('home-born'). But
why should this verse be required to exempt women who are, in
any event, exempted by the general rule? To this Abbaye and Rava
each supply an answer. Abbaye suggests that ha-ezrah is required
because otherwise we might have said that women are not exempt,
i.e. that sukkah is an exception to the general rule. This is because
Scripture speaks of dwelling in sukkot and 'dwelling' usually
means as one normally 'dwells', husband and wife together. Rava
suggests a different reason why ha-ezrah is required and why,
otherwise, we might have supposed that sukkah is an exception to
the general rule. The same word fifteenth occurs both in connection
with sukkah: 'on the fifteenth day of the seventh month' (Leviticus
23: 34) and in connection with Passover: 'on the fifteenth day of
the same month is the feast of unleavened bread' (Leviticus 23: 6).
Now, as above, the precept of eating unleavened bread on Passover
is binding upon women so that we might have supposed that the
use of fifteenth in both verses is to teach us that just as women are
obliged to eat unleavened bread on Passover they are obliged to
dwell in the sukkah on Tabernacles. Sukkah, too, would, then, have
been an exception to the general rule, hence ha-ezrah is required
to teach that women are exempted.
A similar question to that of sukkah is now raised. The verse states
that on the three Festivals of Passover, Pentecost and Tabernacles
there is the duty of pilgrimage to the Temple: ' Three times in the
year all thy males shall appear before the Lord God' (Exodus 23:
17). Thus the precept of appearance is one dependent on time and
Exemption of women from some precepts 13 7
women would have been exempt from it, in any event, because of
the general rule. Why, then, does Scripture have to specify ' all thy
males' to imply ' and not thy females', since women would have been
exempted by the terms of the general rule? To this the Tal mud
replies that here, too, there is the phenomenon of the same word
in two separate verses. The word appear occurs in this verse and it
also occurs in the verse dealing with the assembly on the Festival of
Tabernacles in the seventh year : ' When all Israel is come to appear '
(Deuteronomy 31: 11). Now there it is stated explicitly that women
are obliged to take part in the assembly: ' Gat her the people
together, men, and women' (Deuteronomy 31: 12). Since the word
' appear ' is in both verses we might have supposed that j ust as
assembly is an exception to the general rule so, too, appearance is an
exception. That is why Scripture had to state explicitly ' thy males',
implying ' and not thy females'.
Thus far the following has been established: woman are exempt
from tejillin and from this we derive that they are exempt from
other positive precepts dependent on time. The duty of rejoicing on
the Festival is an exception, as above, since here Scripture states
explicitly: ' And thou shalt rejoice in thy feast, thou, and thy son,
and thy daughter, and thy manservant and thy maidservant, and
the Levite, the stranger, and the fatherless, and the widow, t hat
are within thy gates' (Deuteronomy 16: 14). In that case, the
Tal mud now asks, why derive other positive precepts dependent
on time from tejillin so as to conclude that women are exempt from
them? Let us rather derive from rejoicing that women are obliged
to carry out even positive precepts dependent on time, just as they
have an obligation to rejoice on the Festivals. To this a saying of
Abbaye is quot ed: a woman is made to rejoice by her husband, i.e.
a woman does not, in fact, have an obligation to rejoice. The
command is not addressed to the wife but to her husband, who has
a duty to make his wife happy. But, the Tal mud objects, the verse
speaks, too, of the ' wi dow' and here, since she does not have a
husband, the reference can only be to the widow herself, which
shows that the obligation is binding upon women. No, the Tal mud
replies, the verse refers to a widow who is looked after by someone
else, i.e. she enjoys hospitality in his home and the command is to
him, to make her happy, not to her.
The Tal mud now asks, why not derive from the precept of assembly
that women are obliged to carry out positive precepts dependent
138 The Talmudic argument
on time just as they are obliged to carry out the precept of assembly,
i.e. and why should we prefer to derive from tejillin that they are
not so obliged? To this the Talmud replies by referring to the well-
known hermeneutical principle: sheney ketuvim habaim ke-ehad eyn
melammedim, lit. 'two verses that come together [i.e. for the same
purpose] cannot serve as teachers', i.e. if two separate verses both
purport to teach the same matter neither of them can serve as a
basis of derivation for other similar matters. The logic of the
principle is that if Scripture had intended us to derive other matters
from one of these verses, Scripture would have recorded only one
and the other, too, could have been derived from that one. By giving
us two verses, Scripture seems to be warning us, as it were, against
deriving other matters from either. Now there are, in fact, two
verses, not one, in which it is stated that women are obliged to carry
out positive precepts dependent on time. These are: assembly and
eating unleavened bread on Passover. Consequently, we do not use
either of these as the basis for our derivation. Instead we use the
single instance of tejillin and so conclude that women are exempt,
as they are from tejillin.
We thus have two verses stating that women are obliged to carry
out precepts dependent on time assembly and unleavened bread and
one verse - tejillin - stating that they are exempt. Because of the
principle of' two verses' the two cannot serve as the basis of the
derivation of other precepts but the single verse can and does.
The Talmud now raises an objection. There are, in fact, two
verses stating that women are exempt - tejillin and appearance
('pilgrimage'). If the two verses of unleavened bread and assembly
cannot serve to teach that women are not exempt neither can the
two verses of tejillin and appearance serve to teach that they are
exempt. The reply is that the 'two verses' principle only applies
where, as above, one can be derived from the other. In that case
the Torah, by recording the other, can be said to be warning us
not to use these verses as a basis for derivation. But if one cannot
be derived from the other then neither is superfluous and we do not
then have two verses and then they can serve as a basis for derivation.
Appearance and tejillin cannot be derived one from the other so that
here the ' two verses' principle is not invoked. Appearance cannot be
derived from tejillin, for if the Torah had not stated explicitly that
women are exempt from appearance we would have argued from the
appear-appear words, as above, that just as women are obliged to
Exemption of women from some precepts 139
carry out the assembly precept so, too, they are obliged to carry out
the appearance precept. Conversely, if the Torah had not stated
explicitly that women are exempt from tefillin we would have
compared tefillin to mezuzah, as above, and concluded that just as
women are obliged to carry out the precept of mezuzah so, too, they
are obliged to carry out the precept of tefillin. Consequently, both
tefillin and appearance require to be stated explicitly. Neither can be
derived from the other and hence the ' two verses' principle cannot
be here invoked.
In that case, the Talmud continues, unleavened bread and assembly
may both be required so that here, too, the ' two verses' principle
cannot be invoked. No, the reply is given. It is true that had the
Torah stated that women are obliged to carry out the precept of
assembly we would not have known that they are obliged to eat
unleavened bread, for we would have relied on the fifteenth-fifteenth, as
above, to say that just as they have no obligation to dwell in the
sukkah they have no obligation to eat unleavened bread, i.e., the above
comparison in reverse. But the Torah need not have stated that
women are obliged to carry out the precept of assembly since here
the Torah states that even little children had to assemble, a fortiori
women. Consequently, the assembly statement is superfluous so that
when it is placed together with that of unleavened bread we can invoke
the 'two verses' principle.
The Talmud now advances two further difficulties. First, the
acceptance of the ' two verses' principle on which we have so far
relied is by no means unanimous. There are authorities who reject
it and according to them the question remains, why not derive other
precepts from unleavened bread and assembly? The second difficulty
is with regard to the second rule in the Mishnah, that women are
obliged to carry out positive precepts not dependent on time. This
is presumably derived from the law of fearing parents (Leviticus
19:3) which is binding upon women as well as men [Kiddushin 29a).
But why derive other precepts from fear; derive them rather from
the study of the Torah, from which, as above, women are exempt even
though it is a precept not dependent on time. The first difficulty
is shelved for the time being. The reply to the second difficulty is
that here, too, there is more than one verse. There is the study of
the Torah and there is procreation, as above. Neither of these is a
precept dependent on time and there are thus two verses in which
it is stated that women are exempt from precepts not dependent on
!4 The Talmudic argument
time. There are thus two verses to state that they are exempt and
only one -fear - that they are not exempt. With regard to the two
verses, the 'two verses' principle is invoked which leaves the
one fear from which we derive that women are not exempt.
The Talmud now quotes the opinion of R. Johanan b. Berokah
who holds that women are obliged to engage in procreation since
the command 'Be fruitful, and multiply' (Genesis i: 28) was given
to both Adam and Eve. According to R. Johanan b. Berokah we
have only one verse study that women are exempt, so why derive
from fear that they are not exempt? The reply is that we still have
two verses - study and redemption of the first-born, as above. A woman
is not obliged to redeem her son as a father is. But, the Talmud
continues, according to R. Johanan b. Berokah we also have two
verses saying that they are not exempt -fear and procreation. We thus
have two verses on either side - study and redemption that they are
exempt and procreation and fear that they are not exempt. The reply
is given thdit procreation a.nd fear cannot be derived from one another
and hence, as above, the 'two verses' principle cannot be here
invoked. For if the Torah had only stated that women are obliged
to fear we would not have concluded that they are obliged to
procreate since in the procreation verse it is stated: ' Be fruitful and
multiply, and replenish the earth, and subdue ie' (Genesis 1: 28)
and women do not engage in conquest. Conversely, if the Torah
had stated only procreation we would not have derived fear from it.
Since a married woman cannot always carry out this obligation,
because her husband may not allow her to, we would have argued
that no woman has the duty of fear. Consequently, both verses are
required and, as above, the ' two verses' principle cannot be here
invoked.
Now all this fits in very neatly with the view that 'two verses
which come together cannot act as teachers'. But, repeating now
the first difficulty, what of those authorities who do not accept this
principle at all? Therefore, Rava is reported as saying, the ' scholars
of Papunia', identified with R. Aha b. Jacob, resolved the whole
problem by quoting a different verse. This verse states:' And it shall
be for a sign unto thee upon thine hand, and for a memorial
between thine eyes, that the Torah of the Lord may be in thy
mouth' (Exodus 13: 9). In this verse ' the whole of the Torah' , i.e.
all the precepts, is compared to tefillin (which, in turn, as above, are
compared to the study of the Torah). Thus just as women are exempt
Exemption of women from some precepts 141
from tejillin they are exempt, too, by the actual comparison in the
verse, from every positive precept that, like tefillin, is dependent on
time. They are not exempt, it follows further, from positive precepts
not dependent on time since these are unlike tefillin.
The Talmud now observes that all this is correct, on the
assumption that tefillin are precepts dependent on time. But this
itself is disputed (Shabbat 61 a) and some authorities hold that tejillin
are not dependent on time, i.e. they have to be worn at night as
well as by day and on Sabbaths and festivals. The Talmud neatly
replies that the authority who holds this view is R. Meir and it so
happens that R. Meir does accept the ' two verses' principle. True,
R. Meir cannot use the argument of the Papunians, since according
to him tejillin are not dependent on time. But then the only reason
we had to have recourse to the Papunians was because of those who
reject the ' two verses' principle and R. Meir does not, in fact, reject
it.
At this stage of the argument, if an authority will be discovered
who rejects the ' two verses' principle (and so cannot derive it from
unleavened bread and assembly) and also holds that tefillin are not
dependent on time (and so cannot derive it by the Papunian
derivation) the difficulty will indeed be great. And there is such
an authority, R. Judah, who holds that two verses can be used as
a basis of derivation (Sanhedrin 67b) and also holds that tefillin are
not precepts dependent on time (
l
Eruvin 96b). The reply is, there
are, in reality, three verses. These are: unleavened bread, assembly and
rejoicing (the Talmud now rejects Abbaye's view, above, that a
woman has no obligation to rejoice on the Festivals). Hence there
are three instances of a positive precept dependent on time, which
women are obliged to carry out. And even R. Judah who rejects
the ' two verses' principle, agrees that three verses cannot serve as
a basis of derivation. The logic of this appears to be that while
Scripture might state just one superfluous verse, perhaps for greater
emphasis, it would not repeat more than one unless the intention
is to emphasise that these three instances are exceptions and,
therefore, cannot serve as a basis of derivation.
The principle: sheney ketuvim habaim ke-ahad eyn melammedim is found, too, in:
Pesahim 26a; 45a; Toma 60a; Gittin 76a; Nazir 37b; Kiddushin 24a; our
passage; 34b; 35a; 37b; 42b; 43a; 58a; Sanhedrin 45b; 67b; 72b; Shevu'ot
26b; 'Avodah arah 54b; evahim 24a; 47a; 57a; Hullin 61 a; 113b; Bekhorot
49a; Keritot 6a; Me'ilah n b ; ' Arakhin 14b. See H. H. Medini, Sedey Hemed,
142 The Talmudic argument
vol. iv, p. 517 for two different interpretations of this principle: (a) that
the two verses cannot serve as a basis of derivation, so the matter is left
open; (b) the two verses teach the opposite, which fits in better with the
argument in our sugya. For other Talmudic passages dealing with mitzvat
'aseh she-ha-zeman geramah, see Berakhot 20b; Pesahim 4 3b; Sukkah 28a - b ;
Hagigah 4a. The theme is also referred to incidentally in numerous
passages. Gf. ET, vol. 11, pp. 244-7. There is a curious reference in Ramban's
list of precepts not dependent on time, in which one of the items is sefirat
ha-'omer, 'counting the
c
omer\ It has been suggested that this is counted
as ' not dependent on time' since the time limitation is not in the
performance of the precept but belongs to the nature of the precept, i.e.
the counting of this particular time, unlike, say, lulav which can physically
be carried out at any time and is yet limited to Tabernacles. However,
I would surmise a guess that sefirat ha-'omer is a copyist's error for sefirat
ha-niddah, ' the counting of the days of separation'. The first section of our
sugya, concerning R. Johanan's rule, is really independent of the rest of the
sugya and, as we have seen, is recorded separately in 'Eruvin. For all that,
it has its place here both because of the theme and because it leads up to
and prepares the ground for the second section of the sugya on the question
of derivation.
The pattern of the main sugya is:
(1) Rule derived from tefillin - which in turn derived from study
(2) Question: but let tefillin be rather derived from mezuzah?
(3) Reply: compared to study in both sections of Shema!"
(4) Question: let mezuzah be compared to study and women exempt?
(5) Reply: women also need 'life'
(6) Question: why is ha-ezrah needed for sukkah?
(7) Reply: Abbaye: husband and wife dwelling
(8) Reply: Km*:fifteenth-fifteenth
(9) Question: why 'all males' needed for appearance?
(1 o) Repl y: appear-appear from assembly
(11) Question: why not derive from rejoicing that women are obligated ?
(12) Abbaye's reply: husband's obligation not hers
(13) Question: But what of widow?
(14) Reply: refers to her host
(15) Question: why not derive from assembly that obligated?
(16) Reply: assembly and unleavened bread are ' two verses'
(17) Question: then say tefillin and appearance are 'two verses'?
(18) Reply: both required, one cannot be derived from the other
(19) Question: then assembly and unleavened bread both required ?
(20) Reply: no, assembly not required since even little children
(21) Question: but what of authorities who reject the 'two verses'
principle?
(22) Question (b): whence derive precepts not dependent on time?
(23) Reply to (b): derived from fear
(24) Question: but why not derive from study?
Exemption of women from some precepts 143
(25) Reply: study and procreation 'two verses'
(26) Question: but R. Johanan b. Berokah?
(27) Reply: study and redemption 'two verses'
(28) Question: but then procreation and fear 'two verses'?
(29) Reply: both required
(30) Question as in (21): but what of those who reject 'two verses'?
(31) Reply: Rava: ' Papunians' : verse compared tefillin to precepts
(32) Question: but what of authority who holds tefillin not dependent
on time?
(33) Reply: R. Meir and he does accept the 'two verses . . . ' principle
(34) Question: but what of R. Judah?
(35) Reply: there are three verses (12) rejected and (11) reinstated
There is sufficient evidence that the whole passage, complex though
it is, follows a carefully worked-out pattern of question and reply,
one point leading to another in neat sequence. Note especially how
Abbaye in (12) is rejected in (35) but not until the argument has
made it inevitable, so that early in the passage the way has been
paved for the climax.
Heyzek she-eyno-nikar: indiscernible damage to
property
The sugya is Gittin 53a~54b and is appended to the Mishnah (Gittin
5: 4) in which the rule is given that there are three instances of
damage to property in which the perpetrator is only liable if he did
the damage intentionally (mezid), not if he did it unintentionally
(shogeg). The three instances are: (1) contamination of terumah (the
tithe given to the priest, which cannot be eaten if unclean); (2)
mixing terumah with ordinary, non-sacred food (hullin), i.e. mixing
some terumah in a neighbour's hullin with the result that the mixture
can only be eaten by priests and hence loses a good deal of its market
value; (3) 'stirring', i.e. stirring a neighbour's wine as a libation
to idols, the wine thus being rendered unfit for use (according to
others, see Gittin 52b~53a, 'stirring' means mixing wine that has
been ' stirred' as a libation with the wine of a neighbour so as to
render the whole forbidden). These three are examples of indisc-
ernible damage. The property remains, to the eye of the beholder,
as it was before and yet damage has been done to it. The question
discussed is whether there is compensation in law for such indisc-
ernible damage, since it is possible to argue that the law only
demands compensation for damage that can be discerned as such.
The authorities who hold that there is compensation even for
indiscernible damage are said to hold that heyzek she-eyno nikar
shemah heyzak, literally ' damage that is indiscernible has the name
of damage', i.e. falls under the heading of damage and requires
therefore to be compensated for in law. The authorities who hold
that there is no compensation for such damage are said to hold that
heyzek she-eyno nikar lav shemah heyzak, lit. ' damage t hat is indiscern-
ible does not have the name of damage'. The debate is between
144
Indiscernible damage to property 145
the early-third-century Palestinian Amoraim, Hezekiah and
R. Johanan.
As the sugya immediately begins to discuss, the Mishnah present?
a difficulty according to both views. According to another Mishnah
(Bava Kama 2: 6) a man is responsible for any damage he does,
whether he does it intentionally (mezid) or unintentionally (shogeg).
Why, then, does our Mishnah make a distinction, in the three
instances of indiscernible damage, between damage done inten-
tionally and damage done unintentionally? If such damage is held
to fall under the general law of damages there ought to be liability
even if it was done unintentionally, as in the general law of
damages. If, on the other hand, it is not held to fall under the
general law of damages, there ought to be no liability even if done
intentionally. It is illogical to hold that it falls under the general
rule of damages where it was done intentionally and not where it
was done unintentionally. Intention has nothing to do with it, as
stated explicitly in the Mishnah in Bava Kama.
The sugya opens with the debate between Hezekiah and R.
Johanan. Hezekiah, commenting on the Mishnah, observes that
according to the law of the Torah (i.e. the original law before the
Rabbis introduced new legislation) the perpetrator of the kinds of
damage recorded in the Mishnah is liable whether he did it
intentionally or unintentionally. This is because, the Talmud adds,
Hezekiah holds that indiscernible damage does fall under the
heading of damage. In that case why did the Rabbis introduce the
new legislation to exonerate the perpetrator if he did it uninten-
tionally? The reason is, so that he will be encouraged to inform the
victim that his property is now forbidden, which he would not do
if as a result he were held liable. In that case, why did the Rabbis
not exonerate the perpetrator even if he did it intentionally? The
reply is that here, since his intention is malicious, he would not
inform the victim even if he were exonerated. There is no reason
here to exonerate him and here the original law stands and he is
liable. R. Johanan, on the other hand, observes that according to
the law of Torah he is not liable even if he did it intentionally. The
reason is, observes the Talmud, because indiscernible damage does
not fall under the heading of damage. In that case why did the
Rabbis introduce the new legislation to render him liable where
he did it intentionally? The reason is, in order to prevent men
146 The Talmudic argument
contaminating their neighbour's food and drink and getting offscot
free. The law cannot tolerate the opportunity of a man doing
intentional damage to his neighbour without his having to pay
compensation. This would be an open invitation to lawlessness.
Thus both according to Hezekiah and R. Johanan one of the two
rules in the Mishnah is due to Rabbinic legislation over and above
the original law. According to Hezekiah the liability for intentional
damage, even where indiscernible, is Biblical and the Rabbinic
legislation is for the purpose of exonerating the unintentional act.
According to R. Johanan the exoneration for the unintentional act
is the Biblical law in its original form. The Rabbinic legislation is
to make the perpetrator liable where he did it intentionally.
The next clause of our Mishnah is now quoted in an attempt to
refute Hezekiah. The Mishnah states that if priests, while offering
up a man's sacrifice in the Temple, do so with an untoward
intention (piggul, i.e. the intention of eating the meat outside the
proper place or after the proper time) they have to compensate the
man if they did it intentionally. On this a Baraita comments: ' This
rule is because of social benefit' (tikkun ha-'olam), i.e, it is a Rabbinic
enactment. This shows that it is the intentional act that is due to
Rabbinic legislation not the unintentional act and this supports
R. Johanan and refutes Hezekiah. The reply is that the Baraita does
not refer to the intentional act referred to explicitly in the Mishnah
but to the unintentional act implied by the Mishnah. Since the
Mishnah states that the priests are liable for the intentional act, the
clear implication, though this is not stated explicitly, is that they
are not liable for their unintentional act and it is on this that the
Baraita makes its comment, namely, that, as Hezekiah states, the
exoneration for the unintentional 'act is due to the new Rabbinic
legislation.
R. Eleazar, a contemporary of Hezekiah and R. Johanan, now
raises an objection to Hezekiah from a Baraita, dealing with the laws
of the red heifer and the purifying water into which the ashes of
the red heifer were cast (Numbers 19). If any manner of work was
done with either it is thereby rendered unfit for the purpose of
purification. The Baraita states that if a man did such work with
the red heifer or purifying water belonging to his neighbour and
thereby rendered them unfit he is not liable to compensate his neigh-
bour by the laws of man (i.e. the Courts cannot enforce payment)
but he is liable by the laws of Heaven (i.e. he has a moral obligation
Indiscernible damage to property 147
to pay, though this cannot be enforced). Since there is an obligation
by the laws of Heaven this cannot refer to unintentional damage
for then there would be total exoneration, as in our Mishnah. It
must therefore refer to intentional damage, i.e. the work was done
intentionally and maliciously in order to render the neighbour' s
property unfit. In that case, according to Hezekiah, who holds t hat
indiscernible damage is called damage, the perpetrator ought to
be liable by the laws of man.
See Maharsha, who raises the obvious difficulty that this Baraita is difficult
even according to R. Johanan, since even R. Johanan holds that there is
compensation, albeit only according to Rabbinic enactment, for intentional
damage. Probably the meaning of our passage is that R. Johanan can easily
defend his view by postulating that the special Rabbinic legislation was
only introduced, as stated, in order to prevent man getting away with
malicious damage and this would not apply to the red heifer which was
very rare and the case so unusual that no special legislation was required.
There is a need to introduce legislation to prevent people contaminating
their neighbour's terumah, for example, because many people do have
terumah. There was no need to introduce new legislation to meet the
contingency of a man rendering his neighbour's red heifer unfit since it
is extremely unlikely that his neighbour will ever have a red heifer. But
according to Hezekiah the law of liability for intentional damage is Biblical
and with regard to the actual law of damage there is obviously no
distinction to be made between rare and less rare forms of damage.
The Tal mud states that R. Eleazar, who raised this objection,
himself supplied the answer. The Baraita, when it speaks of doing
work with the heifer and the water, refers to a type of' work' that
only becomes such by intention. For instance, the man brought the
heifer into the stall of its mother where it could be suckled, but his
intention was for it to thresh corn at the same time and, as for the
water, he placed it on the balance of a pair of scales so that all he
did in essence was to place it there and it was his intention to use
it as a weight in the scales that renders it unfit. Since it is his
intention that renders it unfit this is considered to be an indirect
form of damage and even Hezekiah will agree that there is no
liability by the laws of man.
The distinction between this and the three instances in our Mishnah is
clear. There it is the act that does the damage, of contaminating, mixing
or stirring. But what of the case of the priests where it is their intention
that causes the damage? How does this case differ from the case of the
priests ? See Ritba and other commentaries which raise this difficulty. The
distinction appears to be this. Where there has been a piggul intention on
148 The Talmudic argument
the part of the priest it is not the intention that invalidates the sacrifice.
The piggul intention in fact only operates at the time of the actual carrying
out of one of the 'services' of the sacrifice, e.g. sprinkling the blood. Thus
the intention makes the act a. piggul act, as it were, and it is this that renders
the sacrifice invalid so that it is analogous to the three in the Mishnah.
Here, however, the act is innocent and does not render the animal or the
water unfit. The fact that there happens to be produce in the stall or an
object to be weighed in the balance is external to the act and here, since
the act relies entirely on the man's intention to turn it into 'work', it is
considered a form of indirect damage so that Hezekiah will agree that there
is no liability according to the laws of man.
The Talmud now objects that the fourth-century Babylonian
Amora, Rava, has said that if one weighs with the purifying water
it is, in fact, fit for use. The Baraita cannot, therefore, be speaking
of such a case for then the water would still be fit and there would
be no need for any compensation since there had been no damage.
To this the reply is given that Rava, when he declares the water
to be fit, refers to water used as a weight in a balance of scales; but
the Baraita does not, in fact, refer to this but to a different kind of
weighing, namely with the water itself, e.g., the water is placed in
a marked vessel with grooves, and then the object to be weighed is
placed in the water, the weight of the object being determined by
the rise of the water at the grooves. But in that case, the Talmud ob-
jects, the original reply falls to the ground since here it is an act that
takes place, and it is direct, not indirect. Therefore, the Talmud
concludes, both Rava and the Baraita deal with water weighed in
the balance of the scales. The reason the Baraita holds the water
to be unfit is because the Baraita, unlike Rava, is thinking of a man
who by using the water as a weight took his mind off the proper
use of the water and there is a rule that if such positive absence
of mind took place the water is thereby rendered unfit. But this
disqualification is certainly not the result of an act but of absence
of thought and this is so indirect that Hezekiah agrees that there
is no liability by the laws of man.
Rava's disciple, R. Pappa, now raises a fatal objection to
Hezekiah from a Mishnah [Bava Kama 9: 2). Here the Mishnah
rules that if a man stole a coin and this was later taken out of
circulation by the government, or he stole terumah and it later
became contaminated, or he stole leaven and kept it until the
Passover so that it was no longer allowed to be used, the thief can
still return that which he had stolen, saying: ' Here is your
Indiscernible damage to property 149
propert y' (i.e. which I now return to be absolved from my
responsibility). Now the law is that a thief can only be exonerated
if he returns the object he had stolen in an undamaged state. It
follows from the Mishnah t hat in these three cases the property is
considered as being restored in an undamaged state and this can
only be, contrary to Hezekiah, because indiscernible damage is not
held to be damage in the eyes of the law. The Tal mud concludes
with tiyuvta, ' It is, indeed a refutation.'
The Tal mud now asks: ' Shall we say t hat the point at issue
between Hezekiah and R. Johanan is debated by Tannai m?' and
a Baraita is quoted. Here it is stated that in the three instances
recorded in our Mishnah contaminating terumah, mixing terumah
with hullin and ' stirring' - R. Meir holds that the perpetrator is
liable even if he did it unintentionally, whereas R. J udah holds t hat
he is only liable if he did it intentionally, our Mishnah thus
following the view of R. J udah. Now why does R. Meir hold him
to be liable even if he did it unintentionally? It can only be because
he holds that indiscernible damage is held to be damage in the eyes
of the law, whereas R. J udah agrees with the opinion expressed by
R. J ohanan that according to Biblical law there is no liability and
the Rabbinic legislation, as above, covers only intentional damage.
To this the Babylonian Amora (fourth-century) R. Nahman b.
Isaac replies that both R. Meir and R. J udah agree with the
opinion expressed by R. J ohanan that indiscernible damage is not
damage. In t hat case why does R. Meir hold him to be liable even
if he did it unintentionally. It is because R. Meir holds, as a general
principle, that wherever the Rabbis imposed a penalty for an
intentional act they extended this penalty to cover even an
unintentional act and it is with this general principle that R. J udah
disagrees. The rest of the sugya is taken up with a discussion of this
general principle of whether ' t hey penalised the unintentional
because of the intentional act ' - kanisu shogeg atu mezid.
The pattern of this section of the sugya is:
(1) Hezekiah: 'named damage' ; R. Johanan: ' not named damage'
(2) Reasons
Hezekiah: he is exonerated where the damage is unintentional,
in order that he may inform the victim
R. Johanan: intentional, liable because of Rabbinic enactment
(3) Objection to Hezekiah: Mishnah of priests and Baraita on this
(4) Reply: Baraita refers to unintentional damage implied in Mishnah
150 The Talmudic argument
(5) R. Eleazar's objection to Hezekiah: Baraita of heifer and water
(6) R. Eleazar's reply: intention the cause and therefore indirect
(7) Objection to this: Rava's ruling regarding weighing
(8) Reply: Baraita refers to weighing with water itself
(9) Objection to this: but this is direct and hence original objection
remains
(10) Reply: indirect for different reason: taking out of mind
(11) R. Pappa's final objection: a case of thief in Mishnah
(12) Indeed this is a refutation
(13) But shall we say it is debated by Tannaim
(14) No, R. Nahman b. Isaac: debate concerns whether unintentional
act penalised.
Although the material is presented in more or less chronological
order so far as R. Eleazar, Rava, R. Pappa and R. Nahman b.
Isaac are concerned, the more definite pattern is logical, i.e. the
reply in (4) cannot be used for (5); replies in (6) to (10) cannot
be used for (11). And the framework of the sugya has been organised
with this progression of thought in mind.
Basically we have here, in fact, two separate sugyot: (a) the sugya of heyzek
she-eyno nikar; (b) the sugya of kanesu shogeg atu mezid. The ' Shal l we say'
in (13) appears, then, to be no more than a device in order to introduce
the second sugya (b), which meets the objection of the Tosqfists, s.v. tiyuvta,
of stating this after Hezekiah has in any event been refuted. The theme
of heyzek she-eyno nikar is found, in addition to our sugya in M(fed Katan 13a;
Gittin 41a; 44b (duplicate of Mo'ed Katan 13a); Bava Kama 5a. See Kesef
Nivhar, part 1, no. 49, pp. 86a-87b and ET, vol. vn, pp. 702-22.
Thus far we have established that R. Meir penalises even an
unintentional act because of an intentional one, while R. Judah
refuses to do this. The Talmud now quotes a Baraita which seems
to suggest that R. Meir and R. Judah hold, in fact, exactly opposite
views. The case dealt with in this Baraita is that of a man who cooks
food on the Sabbath. May the illegally cooked food be eaten? There
are three opinions on this in the Baraita. R. Meir holds that if the
cooking was done intentionally, i.e. the man knew that he was
cooking on the Sabbath and knew that it is forbidden to cook on
the Sabbath, the food may not be eaten; but if done unintentionally,
i.e. the man either did not know that it was the Sabbath or did
not know that it was forbidden to cook on the Sabbath, then the
food is permitted even on that Sabbath. R. Judah holds that if the
cooking was done intentionally it may never be eaten, but if done un-
intentionally it may not be eaten on that Sabbath but may be eaten
after the Sabbath is over. R. Johanan the sandal maker holds that
Indiscernible damage to property 151
if the cooking was done intentionally neither the man who did the
cooking nor anyone else may ever eat it, but if done unintentionally
then others may eat it after the Sabbath but never the man himself.
Thus here R. Meir is lenient, permitting the food cooked uninten-
tionally to be eaten even by the man himself and even on that
Sabbath, whereas R. Judah is strict, forbidding the food to be eaten
on the Sabbath itself even if the cooking was done unintentionally.
This would seem to contradict the statements in the previous de-
bate where it is R. Meir who is strict, penalising the unintentional
act because of the intentional, whereas it is R. Judah who is lenient,
not penalising the unintentional act because of the intentional. Or,
as the Talmud puts it: ' R. Meir contradicts R. Meir and R. Judah
contradicts R. Judah. '
The Talmud first proceeds to resolve the contradiction between
the two statements of R. Meir. R. Meir does, in fact, penalise even
the unintentional act because of the intentional but only where the
intentional act involves a Rabbinic prohibition as in the cases of
the first Baraita, contaminating terumah, mixing terumah and 'stir-
ring'. But cooking on the Sabbath is a Biblical prohibition and
here R. Meir is content to penalise the intentional act but sees no
reason to extend the penalty to the unintentional act. The logic of
this position is that Rabbinic law is treated lightly by people and
so requires the reinforcement provided by the additional penalising
of even an unintentional offence, whereas Biblical prohibitions are
viewed strictly in any event and can stand on their own without
any additional reinforcement except when the act is intentional.
But, the Talmud asks, one of the three instances recorded in the
first Baraita is the stirring of the libation wine. This is a Biblical
prohibition and yet R. Meir does penalise even the unintentional
act. To this the reply is given that here R. Meir penalises even the
unintentional act because of the great severity of the prohibition,
idolatry being the worst of offences. Thus R. Meir does not penalise
the unintentional act of cooking on the Sabbath because the offence
is Biblical and requires no reinforcement. He does penalise the
unintentional acts of contaminating terumah and mixing terumah
with hullin because the offence is Rabbinic and therefore requires
reinforcement. He also penalises the unintentional act of'stirring',
though the offence is Biblical, because this offence is so severe that
it does require reinforcement. In other words, R. Meir holds that
reinforcement is required in two instances: where the offence is
15
2
The Talmudic argument
Rabbinic, because it is so light as to require reinforcement; and
where the offence is that of idolatry, because it is so severe as to
demand reinforcement. A' medium' offence, that of cooking on the
Sabbath, being Biblical, is sufficiently strong not to require
reinforcement on grounds of 'lightness' and, at the same time,
sufficiently weak, as it were, not to require reinforcement on
grounds of' severity'.
As for the contradiction between the two statements of R. Judah,
this is easily resolved. Contrary to R. Meir, R. Judah holds that the
more severe offence, the Biblical offence, requires reinforcement
and the less severe, the Rabbinic, does not. Hence in the instances
of contaminating terumah and mixing terumah with hullin, where the
offence is Rabbinic, R. Judah does not penalise the unintentional
act, whereas in the instance of cooking on the Sabbath, where the
offence is Biblical, he does penalise the unintentional act. But here
again the Talmud objects that one of the three instances in the first
Baraita is 'stirring' and here the offence is Biblical, and yet
R. Judah does not penalise the unintentional act. To this the reply
is given that the prohibition of idolatry is so severe that people will
shun it of their own accord so that no reinforcement is required.
Thus according to R. Judah the general principle is that a Biblical
prohibition does require reinforcement, hence he penalises the
unintentional act of cooking on the Sabbath, whereas a Rabbinic
prohibition does not require reinforcement, hence he does not
penalise the unintentional acts of contaminating and mixing
terumah. The exception is stirring the wine. Here the offence is
Biblical and yet because it is so severe no extra reinforcement is re-
quired. In other words, neither a very severe offence (idolatry) nor
a very light one (Rabbinic) requires reinforcement, the one because
it is too severe in any event to need reinforcement, the other because
it is too light to warrant reinforcement. The ' medium' offence of
cooking on the Sabbath is sufficiently severe to warrant reinforce-
ment and, at the same time, sufficiently light to need it.
We have now suggested that R. Meir only penalises the uninten-
tional act where the offence is Rabbinic, not where it is Biblical.
A Baraita is now quoted. This deals with a Biblical prohibition and
yet R. Meir does penalise even the unintentional act. The Baraita
states that R. Meir holds that if one plants a new shoot on the
Sabbath, he may preserve the shoot if he planted it unintentionally
but not if he planted it intentionally, when he must uproot it. But
Indiscernible damage to property 153
if he planted the shoot in the Sabbatical year he must uproot it
whether he planted it intentionally (i.e. knowing that it was the
Sabbatical year and that it was forbidden to plant during this year)
or unintentionally. R. Judah holds exactly the opposite view. If he
planted the shoot on the Sabbath he must uproot it whether he
planted it intentionally or unintentionally; but if he planted it in
the Sabbatical year he is only required to uproot it if he planted
it intentionally and he may preserve it if he planted it
unintentionally.
Now planting in the Sabbatical year is a Biblical prohibition and
yet R. Meir penalises even the unintentional act. The Talmud
replies that the Baraita appears to be contradictory in any event,
since the Baraita itself appears to make a distinction between two
Biblical offences, that of the Sabbath and that of the Sabbatical
year. In fact, the reason for the distinction is stated by R. Meir
himself in the continuation of the Baraita. R. Meir here declares
that the distinction between the Sabbath and the Sabbatical year
is based on two factors. The first is:
e
Israel counts according to the
Sabbatical year but not according to the Sabbath.' This means that
one is obliged to count the years of every new shoot because the
fruit is forbidden during the first three years (
l
orlah) and even on
the fourth year can only be eaten after it has been redeemed (see
Leviticus 19: 23-4). Hence it is plausible to suggest that people will
count back from the fourth and third years and then they will know
that the shoot has been planted in the Sabbatical year. This is why
R. Meir forbids it even when planted unintentionally. It is not
because he generally penalises even an unintentional act but
because if it is here permitted people may conclude that it is really
permitted to plant shoots in the Sabbatical year. But with regard
to planting on the Sabbath, people do not count back the weeks
or days and no one will know that the shoot has been planted on
the Sabbath. And R. Meir adds a second reason for the distinction:
'Israelites are suspected of ignoring the laws of the Sabbatical year
but not of ignoring the Sabbath laws', i.e. since people treat the
Sabbatical year laws lightly, reinforcement is required and therefore
even the unintentional act is penalised as it is, according to R. Meir,
with regard to the light offences of Rabbinic law. The Talmud then
asks, why is this second reason required, since the first seems
entirely adequate? The reply is given that on occasion people do
count back to the Sabbath. This is because of the law which permits
154 The Talmudic argument
the planting of new shoots 30 days before the Sabbatical year. These
30 days are then counted as a whole year for the purpose of 'orlah,
i.e. the fruit may be eaten after two whole years and 30 days as
if three years had passed. Now when people see that the fruit is
permitted after only two years they will appreciate that this is
because of the 30 days' law. They will then count back these 30
days and discover that the shoot had been planted on the Sabbath.
Hence the first reason is not entirely adequate, which is why
R. Meir has to produce the second reason.
The contradiction between the two views of R. Meir has now
been resolved but there remains the contradiction between the two
views of R. Judah. According to R. Judah there is a penalty for
cooking on the Sabbath even where the act was done unintention-
ally because the offence is Biblical; this is consistent with his view
in the Baraita, which states that according to R. Judah the shoot
planted on the Sabbath must be uprooted even if planted unin-
tentionally. But why, then, does R. Judah permit the shoot planted
in the Sabbatical year to be preserved, if done unintentionally? The
reply is that in R. Judah' s district people were very strict with
regard to the laws of the Sabbatical year so that here the usual re-
inforcement of Biblical law is not required. In support the story is
told of a man who taunted his neighbour by calling him a proselyte,
the son of a proselyte, and the man retorted: ' As least I do not eat
the fruit of the Sabbatical year as you do' , which shows that for
some people the laws of the Sabbatical year were kept so strictly
that it was a serious insult to accuse anyone of disregarding these
laws.
We have thus far established that R. Meir does impose a penalty
on an unintentional act where the offence is Rabbinic. A Baraita
is now quoted which seems to state the opposite. The rule is that
if one eats terumah belonging to a priest he must give the priest in
return non-sacred food {hullin) to the value of the terumah he had
eaten and this becomes automatically terumah. The Baraita states
that if a non-priest had eaten contaminated terumah belonging to
a priest he must give the priest uncontaminated hullin which then,
as above, becomes terumah. But what if he gave the priest
contaminated hullin? There is a debate on this between Symmachus
and the Sages as to what R. Meir said. Symmachus states in the
name of R. Meir that if the man knew that his hullin was
Indiscernible damage to property 155
contaminated when he gave it in payment, his payment is
ineffective; but if he made this payment unintentionally, i.e. he was
unaware that his payment was made with contaminated hullin, his
payment is effective. The Sages say, however, t hat according to
R. Meir the payment is in any event effective. However, this only
means, the Baraita goes on to say, that the payment is ' effective'
in t hat the contaminated hullin then becomes contaminated terumah.
But in addition, as an extra penalty, he must pay the priest a further
amount of clean hullin. This, being no more t han a fine, does not
become terumah.
So far the Talmud has examined the Baraita as it stands. But,
the Talmud continues, 'we were puzzled' by this Baraita, which
is extremely difficult to accept as it stands. If the man ate
contaminated terumah why should he not be allowed to make
payment with contaminated hullin, which would then become
contaminated terumah, since this is, after all, what he had eaten?
In fact his intention is to go one better, since contaminated terumah
cannot be eaten at all, whereas contaminated hullin can be used for
food by the priest when he is unclean. (The hullin may not be eaten
by the priest once it has become terumah after the payment but the
man's intention, at least, is to improve matters.) In reply to ' our'
puzzle Rava, 'others say Kadi' , explains that the Baraita requires
to be emended and to read as follows. If he ate contaminated
terumah he can, in fact, repay any amount, i.e. he need not pay the
full value but simply set aside any amount of hullin, giving this to
the priest. On this there is no debate at all. The debate concerns
where he ate clean terumah and paid with unclean hullin. Here all agree
that if the payment was made intentionally, i.e. with full awareness
that the hullin was unclean, it is ineffective. The debate concerns
where the payment was made unintentionally, where he did not
know that he was paying with unclean hullin. The Sages hold that
even here the payment is ineffective but Symmachus states that
R. Meir holds that here the payment is effective. And, the Talmud
concludes, R. Aha son of R. Ika explained that this means that,
according to Symmachus, R. Meir does not penalise an uninten-
tional act because of an intentional one. Now here the prohibition
of paying with unclean hullin is Rabbinic, not Biblical, and this
contradicts the principle that where the offence is Rabbinic
R. Meir does penalise the unintentional act. The reply is that here
156 The Talmudic argument
the man's intention is good. He wishes to make repayment. Here
R. Meir would agree that it would be wrong to penalise him for
his unintentional act.
There now follows a series of further attempted refutations of the
view that R. Meir does penalise the unintentional act where the
offence is Rabbinic. The first of these is from a Baraita in which it
is stated that if the blood of a sacrifice has become contaminated
and is then sprinkled on the altar, then if the sprinkling was done
intentionally (i.e. the priest being aware that the blood was
unclean) the sacrifice is not acceptable (and its meat may not be
eaten) but if done unintentionally the sacrifice is acceptable. But
according to Biblical law it is acceptable even if the blood is unclean.
We have, therefore, a Rabbinic offence and yet the Baraita states
that there is no penalty for the unintentional act. There is, in fact,
no indication that the Baraita follows the opinion of R. Meir. As
Rashi comments, the Talmud could have replied that this follows
the opinion of R. Judah but the Talmud seeks to reply even if it
follows the opinion of R. Meir. The reply is similar to the previous
one. No penalty for an unintentional act is imposed where the
motives are good, as they are here, where the sacrifice is for the
purpose of finding acceptance in the eyes of God.
The second attempted refutation is from an anonymous Mishnah
{Terumot 2:3). An anonymous Mishnah does generally follow the
opinion of R. Meir. There is a Rabbinic offence of tithing on the
Sabbath and yet this Mishnah states that if the tithing was done
unintentionally it is valid. Here again the reply is given that the
motives were good, to free the produce from being untithed.
Finally, another clause of the same anonymous Mishnah (Terumot
2: 3) is quoted. It is a Rabbinic offence to immerse vessels in order
to purify them on the Sabbath, but if it was done unintentionally
the immersion is valid. Once again the reply is given that the man's
motives were good in that he wished to have clean vessels.
We now revert to a consideration of R. Judah' s views. We have
thus far established that R. Judah only imposes a penalty where
the act was done unintentionally, where the offence is Biblical not
where it is Rabbinic. A Baraita is quoted dealing with the law of
l
orlah, the fruit of a new shoot forbidden, as above, during the first
three years. According to Biblical law if'orlah fruit was mixed with
ordinary, permitted fruit and the forbidden fruit can no longer be
identified, the mixture is permitted if the ratio is two to one of
Indiscernible damage to property 157
permitted to forbidden fruit. The Rabbis, however, ordained that
for 'orlah fruit to become neutralised a ratio of one to two hundred
is required. Furthermore, things of great value and significance can
never become neutralised, even at a ratio of one to a thousand.
Thus the rule is that a certain species of nut, because of its
significance, can never become neutralised. Now the Baraita states
that if these nuts have been broken so as to lose their special
significance, they do become neutralised in a ratio of one to two
hundred according to R. Jose and R. Simeon, provided that they
have been broken, after they had become mixed, unintentionally,
and were not broken intentionally to deprive of them their signifi-
cance so that neutralisation can take effect. But R. Meir and
R. Judah hold that neutralisation cannot take effect, even if the
nuts were broken unintentionally. Now here the offence is only
Rabbinic since, as above, according to Biblical law there is always
neutralisation at a ratio of one to two and yet R. Judah does penalise
the unintentional act. The reply is that this belongs in an entirely
different category. The reason why R. Judah forbids it here, even
when the act was done unintentionally, is not because of a penalty
but simply in order to prevent fraud. For if the neutralisation were
held to be effective when done unintentionally the fraudulent
person would do it intentionally and pretend that he had done it
unintentionally.
Finally, the Talmud points to a contradiction between the view
of R. Jose stated in this Baraita, that where it is done intentionally
it is not permitted, and R.Jose' s view in a Mishnah ('Orlah 1:6).
The Mishnah rules that if an 'orlah shoot had been planted among
other shoots and cannot now be identified all the fruit of all the
shoots is forbidden to be gathered since a growing shoot is held to
be separate and the law of neutralisation does not come into
operation. Once the fruit had been gathered, however, the 'orlah
fruit does become neutralised at a ratio of one to two hundred
provided the gathering was not done intentionally, i.e. for the
purpose of neutralisation. But R. Jose holds that it does become
neutralised even if done intentionally, which contradicts R. Jose's
ruling in the other Baraita that the nuts do not become neutralised
if they had been broken with that intention. The reply is that we
have learnt with regard to this case that Rava explained: ' It is an
established fact (hazakah) that no man will allow his whole vineyard
to become forbidden because of a single shoot.' As Rashi understands
158 The Talmudic argument
the reply, normally the man will mark the forbidden shoot so as
to prevent his whole vineyard becoming prohibited. Consequently,
unlike the case of the nuts, there is no need for any special Rabbinic
rule to forbid intentional neutralisation since the Rabbis do not
impose their restrictions in very unusual cases. And, the Talmud
continues, when Ravin came to Babylon from Palestine he reported
that R. Johanan said the very same thing: ' It is an established fact
that no man will allow his whole vineyard to become forbidden
because of a single shoot.'
The phenomenon that a fourth-century Babylonian teacher, Rava, should
have given the same explanation as that given by the third-century
Palestinian teacher, R. Johanan, and, moreover, in exactly the same
words, is best understood on the grounds that comments of this kind
enjoyed a very wide circulation among the scholars of both Palestine and
Babylon, so that Rava, for instance, is not really stating his own original
opinion but rather quoting a well-known legal maxim with which he
happens to concur. The other Talmudic passages dealing with the theme of
kan&su shogeg atu mezid are: Shabbat 3b; Eruvin 68b and 100a; and Yevamot
90a. The Yevamot passage is a duplicate of the section in our sugya dealing
with the payment of terumah but the material is presented in a slightly
different order to fit in with the theme of that passage. In both passages
it is stated t hat ' we were puzzled', which suggests an earlier sugya on which
both sugyot rely. In the Yevamot passage, on the other hand, R. Aha son
of R. Ika is quoted as if he was making his original comment there, whereas
here our sugya records: 'and R. Aha son of R. Ika said' which appears to
suggest a cross-reference, so that our sugya depends on that of Yevamot. If
this is correct, there are three stages here: (1) the original ' we' who were
'puzzled', which means that there had been an early discussion of the
difficulties in the Baraita and the reply by emendation; (2) the sugya in
Yevamot, which relies on (1) but adds the comment of R. Aha son of R. Ika
and generally fits (1) into the new framework of that passage; (3) our sugya,
which relies on (2) and adapts it to the theme of our passage. For a full
discussion of the general theme of kanesu shogeg atu mezid see Kesef Nivhar,
part in, no. 143, pp. 12a-13b.
The pattern of this section of the sugya is:
(1) Debate of R. Meir and R. Judah: regarding contaminating
terumah etc.
R. Meir: penalises even unintentional act; R. Judah: penalises
only intentional act
(2) Contradiction of R. Meir I: cooking on the Sabbath
(3) Reply: R. Meir only penalises Rabbinic offences not Biblical
(4) Objection: 'stirring' is Biblical
(5) Reply: because of severity of offence of idolatry
(6) Contradiction of R. Judah I: cooking on the Sabbath
Indiscernible damage to property 159
(7) Reply: R. Judah only penalises Biblical offences not Rabbinic
(8) Objection: 'stirring' is Biblical
(9) Reply: no need for penalty because of severity of idolatry
(10) Contradiction of R. Meir II: Sabbatical year
(11) Reply: but in any event contradiction between this and Sabbath
(12) Therefore: R. Meir states two reasons: (a) ' counting' ;
(b) suspicion
(13) Question: why is (b) required, since (a) sufficient
(14) Reply: because sometimes there is 'counting' to Sabbath, hence
(a) insufficient
(15) Contradiction of R. Judah II: Sabbatical year
(16) Reply: in R. Judah' s district Sabbatical year treated very strictly
(17) Support: story of insult
(18) Contradiction of R. Meir III: payment of terumah
(19) Reply: no penalty if man wants to pay
(20) Contradiction of R. Meir IV: sprinkling of blood
(21) Reply: no penalty if man wishes to be accepted
(22) Contradiction of R. Meir V: tithing on the Sabbath
(23) Reply: no penalty if man wishes to tithe his produce
(24) Contradiction of R. Meir VI: immersion on the Sabbath
(25) Reply: no penalty if man wishes to purify his vessels
(26) Contradiction of R. Judah III: breaking nuts
(27) Reply: because of fear of fraud
(28) Contradiction of R.Jose: gathering fruit of plant
(29) Reply: because unusual
It is to be noted that (16) and (17) are found in Bekhorot 30a and
appear here to be quoted from there, i.e. this small section has been
appropriated by the editors in order to fit it in with the sequence
in our sugya. The sequence of thought in (18) to (25) appears to
be: in reply to (18) there is given (19), but this r e pl y- the man
wishes to pay - does not apply to (20). Reply: (21). But granted
that there is no penalty where he desires atonement, what of (22)
where there is no question of atonement? Reply (23): he wishes to
put right his untithed produce and this is meritorious since
otherwise he may come to eat that which is forbidden. But what
of (24), where there is no particular merit in immersing the vessels?
Reply (25): nevertheless his motive is positive. If this is correct, it
will explain why (see Tosafists) (22) is quoted before (24) whereas
in the Mishnah (Terumot 2: 3) the order is reversed. The sugya thus
presents the material in a contrived way in order to allow each stage
of the argument to follow the previous stage and lead up to the next.
i 6
Kinyan hatzer: acquisition by means of a
domain
This comparatively brief sugya is Bava Metzi'a i ob- na. The
universally-accepted legal principle is that a man's domain can
acquire property for him, e.g. if a lost article happened to come
into his domain it belongs to him, or if A sells goods to B and
deposits them in B's domain, with the intention of both that it
should effect a transfer, it does so and neither can retract. This is
known as kinyan hatzer,' acquisition by means of a courtyard'. That
the hatzer acquires is held to be Biblical. But it is here assumed that
this mode of acquisition is not an entirely different one but rather
an extension of some other, Biblically-accepted model. There are
two possible candidates for the key-mode, as it were. The first is
acquisition by the person himself, e.g. by his lifting up the object
to be acquired. This is known asyad, lit. ' hand' . The second is
'agency', shelihut, i.e. A appoints B to be his agent to acquire the
object on his behalf.
Now it can be argued that hatzer is 'included' (the term used in
our sugya), i.e. introduced as a new mode of acquisition in the
Torah, as either an extension of yad or of shelihut. If as a mode of
yad, it means that the Torah provides each person, as it were, with
an extension of his own hand; his hatzer acquiring the object for
him as if that object had actually come into his hand. If as a mode
of shelihut, it means that the Torah provides him, as it were, with
a kind of automatic agent, his hatzer, which operates on the same
basis as that of his agent acquiring property on his behalf. This is
formulated as either hatzer mi-shumyad or hatzer mi-shum shelihut. The
debate in the sugya is between the early-third-century Amora, Resh
Lakish, speaking in the name of Abba Kohen Bardala, and his
contemporary, R. Johanan, speaking in the name of R. Jannai.
160
Acquisition by means of a domain 161
The reason why the sugya has been inserted here is because there
is a preceding statement of Resh Lakish in the name of Abba Kohen
Bardala (Bava Metzi'a ioa), that statement being relevant to the
Mishnah to which it has been appended and which is quoted in
the discussion on the statement (Bava Metzi'a iob). Our sugya
begins, in fact, with: ve-amar Resh Lakish mi-shum Abba Kohen Bardala,
'And Resh Lakish said further in the name of Abba Kohen Bardala.'
The actual formulation of the debate in these terms - mi-shumyad
or mi-shum shelihut - is not that of the protagonists themselves but
is editorial and is a typical, abstract formulation peculiar to the
Amoraic literature.
Two further points require to be made before proceeding to the
sugya itself. First, there is a Rabbinic extension of the law of hatzer,
known as 'four cubits', i.e. the four cubits at the centre of which
a person happens to be standing operate as his hatzer during the
time he stands there and acquire for him. The second point to be
noted is that if the father of a girl who is a minor (ketanah) betroths
her to a man, she has the full status of a married woman and
requires a get to release her from the marriage bond. But the get
has to be delivered into her hand. Unlike an adult woman, she
cannot appoint an agent to accept the get on her behalf since a
minor has no power to appoint an agent. The question discussed
in our sugya is whether a ketanah can acquire the get when it is placed
by the husband in her hatzer. If hatzer operates as yad she can, but
if as shelihut she cannot.
The sugya begins: 'And Resh Lakish said further in the name of
Abba Kohen Bardala: "A ketanah has neither hatzer nor four
cubits." But R. Johanan said in the name of R. Jannai: 'She does
have a hatzer and four cubits."' The Talmud asks: ' What is their
point of difference?' and replies: R. Johanan in the name of
R. Jannai holds that hatzer is an extension of yad. Therefore she has
hatzer just as she hasjW. But Resh Lakish, in the name of Abba
Kohen Bardala, holds that hatzer is an extension of shelihut. Since
she cannot appoint an agent she has no hatzer.
The Talmud now asks:' But is it at all possible for anyone to hold
that hatzer is because of shelihut?
9
The objection to anyone holding
such a view is stated. A Baraita, commenting on the verse: ' If the
theft be certainly found in his hand alive. . . he shall restore double'
(Exodus 22: 4; The Torah: Exodus 22: 3), observes that although
the verse speaks of' his hand' the thief is obliged to pay double even
162 The Talmudic argument
if he stole by means of his 'courtyard', i.e. he closed the gate on
an animal belonging to his neighbour that had strayed into his
domain. This the Baraita derives from the expression: 'found at all',
an inclusion expression. It follows that, as the Baraita states and
derives it from a Scriptural verse, theft by means of a hatzer
constitutes theft and the thief has to pay double, as if he had stolen
it himself by his own hand. But this would mean that agency is
effective even where a crime is committed and * we have established
the principle' (referring to Kiddushin 42b) that the principle of
agency does not operate where a crime is committed. This means
that if A instructs B to steal something on his behalf the act of
stealing is not transferred, as in other cases of agency, to the
principal, but the agent, not the principal, is the thief who is
responsible in law and he, the agent, not the principal, has to pay
double. Now if hatzer is an extension ofyad, all is well. But if hatzer
is an extension of shelihut, how can theft by means of hatzer render
the owner of the hatzer liable to pay double, since where a crime
is committed the agency principle does not operate?
To this a reply is given by the late-fifth-century Babylonian
Amora, Ravina, and a different reply is given by his contemporary,
R. Sama. Each of these advances a reason why the principle ' no
agency where a crime is committed' does not apply to the
particular kind of agency provided by hatzer. Hence the Baraita, and
by implication the verse on which it relies, affords no refutation of
the view that hatzer is because oishelihut. It may, indeed, be because
of shelihut and yet the thief, through his hatzer, is still liable despite
the ' no agency where a crime is committed' principle since, for the
reason each gives, that principle does not operate in the particular
kind of agency that is hatzer.
See Gilyon ha-Shas and DS, which indicate that some versions have
' R. Sama bar Rakta in the presence of Ravina in the name of R. Hiyya
bar R. Ivo' instead of 'R. Sama', as in our texts. In any event these two
are very-late-Babylonian Amoraim, sufficient evidence that our sugya in
the form we now have it must be very late and this is supported by the
1
but we have established the rule' which appears to be a quote from the
conclusion of the sugya in Kiddushin 42b.
Ravina argues that the principle ' no agency where a crime is
committed' only applies when the agent himself has an obligation
to keep the law and commits the crime if he does not. Here it is
held that the principal is not liable since the law against theft is
Acquisition by means of a domain 163
addressed as much to the agent as to the principal and it is, after
all, the agent who actually commits the crime, albeit on behalf of
the principal. The hatzer, on the other hand, even if it does operate
under the principle oishelihut, is an agent without obligations. The
hatzer has no obligation to keep the law and here the ' act' of the
' agent' is transferred to the principal, as in all cases of agency where
no crime is committed.
The Talmud now seeks to refute Ravina's argument by a reductio
ad absurdum. No one would dream of suggesting that if a man
instructed a married woman to steal or a slave to steal that he, and
not they, should be liable. And yet, since neither a married woman
nor a slave has property, if Ravina is correct he must be liable and
not they. The reply is obvious. The married woman and the slave
do have an obligation not to steal and they are, indeed, even
obliged to pay. The reason why payment cannot be made is simply
because they have no money with which to pay. And a Mishnah
(Bava Kama 8: 4) is quoted in which it is, indeed, ruled that when
the married woman is divorced and so has property of her own and
when the slave is freed and so has property of his own, they are
obliged to pay.
R. Sama argues for a slightly different distinction. The principle
of no agency where a crime is committed only applies where the
agent acts voluntarily, where he can please himself whether or not
to obey the instructions of the principal. If he then steals it is he
who commits the offence, not the principal, since no one compels
him to obey the principal's instructions. The hatzer, on the other
hand, even if it is an extension of shelihut, is involuntary. The hatzer
is, as it were, an agent who cannot help obeying his principal's
instructions and here the rule ' no agency where a crime has been
committed' does not apply.
We now have two reasons why ' no agency where a crime is
committed' does not apply to hatzer. Ravina's reason is because
hatzer has no obligation in law and R. Sama's because hatzer has no
choice. The Talmud asks: what is the difference between the two
reasons? The reply suggests that there is a practical difference
between the two reasons, not with regard to hatzer (where both
reasons apply, since hatzer has neither obligation nor choice), but
where an agent instructed to commit an offence has no obligation,
so far as that offence is concerned, but has free choice. The Talmud
discovers two such instances. The first is when a priest, who may
164 The Talmudic argument
not marry a divorcee (Leviticus 21: 7), instructs an Israelite, to
whom this prohibition does not apply, to act as his agent to betroth
a woman for him. Here there is no obligation on the part of the agent
but there is choice; no one compels the agent to carry out the priest's
instructions. Here according to Ravina the principle ' no agency
where an offence is committed' does not apply and the betrothal
is valid; but according to R. Sama, since there is choice even though
there is no obligation, the principle ' no agency where an offence is
committed' does apply and the betrothal is invalid. The second
instance is that of a man, who may not ' round the corners of his
head' nor have them ' rounded' for him (Leviticus 19: 27), who
instructs a woman, to whom these prohibitions do not apply, to
' round' the corners of the head of a minor (i.e. a male to whom
the prohibitions do apply but who, because of his youth, will not
prevent it). According to Ravina the principle ' no agency where
an offence is committed' does not apply since the woman has no
obligation. Hence the man who gave the instructions is liable, the
woman acting as his agent. But according to R. Sama, since the
woman has free choice whether or not to obey the man's instructions,
the principle of' no agency where an offence is committed' does
apply. Hence there is no agency here and the man who gave the
instructions is not liable.
The Talmud now asks:' But is it at all possible for anyone to hold
that hatzer is not an extension ofyad', i.e. the previous difficulty has
been how anyone could possibly hold that hatzer is an extension of
shelihut (in view of the principle ' no agency where a crime has been
committed') but now a new difficulty presents itself that even if,
theoretically, hatzer can be an extension of shelihut there is still ample
proof that it is, in fact, an extension not of shelihut but oiyad.
A Baraita is quoted, commenting on a verse which deals with the
get: 'give it in her hand' (Deuteronomy 24: 3). The Baraita, like
the previous one quoted, states that ' hand' is not to be taken
literally, but from the general form of the statement - ve-natan, ' he
shall give it' - it is derived that the husband can give the get to his
wife by placing it in her' courtyard.' This would suggest that hatzer
is an extension oiyad since the very fact that a get can be given by
means of hatzer is derived from a verse which speaks of giving it in
her hand. The Talmud, therefore, now suggests that both Resh
Lakish in the name of Abba Kohen Bardala and R. Johanan in the
name of R. Jannai agree that, in the case of get, hatzer is because
Acquisition by means of a domain 165
oiyad. Consequently, there is no debate on the question of whether
a minor can be given her get by means of her hatzer. All agree that
she can. When Resh Lakish in the name of Abba Kohen Bardala
observes that a ketanah has no hatzer he is not referring to get but
to acquiring a lost article. His argument is that it does not follow
at all that, because a ketanah can be divorced by means of hatzer ,
in all other instances, such as acquiring a lost article, hatzer, is be-
cause oiyad. In other instances it is, in fact, because oi shelihut and
hence a ketanah cannot acquire a lost article by means of her hatzer
and this is what Resh Lakish in the name of Abba Kohen Bardala
means when he states that a ketanah has no hatzer. R. Johanan in
the name of R. Jannai holds that it is feasible to deduce the law
of acquiring a lost article from that of get. Since the Torah has
informed us that a ketanah has a hatzer for the purpose oi get this
demonstrates that the Torah considers hatzer to be an extension of
yad, not of shelihut, and it would therefore be effective for a ketanah
even for the purpose of acquiring a lost article.
' If you want I can say', the Talmud continues, with regard to
a ketanah (a minor who is a female) all agree that just as she can
receive her get by means of hatzer so, too, she can acquire a lost
article by means oi hatzer. Both agree, according to this version, with
the view stated in the first version as that of R. Johanan, namely,
since the Torah has informed us that a ketanah has a hatzer for the
purpose of get it follows that the Torah considers hatzer to be because
oiyad and hence hatzer does acquire for a ketanah even a lost article.
But Resh Lakish in the name of Abba Kohen Bardala argues,
granted a ketanah has a hatzer and that this is because oiyad so that
she can even acquire a lost article by means of the hatzer, it does
not follow that a katan (a minor who is a male) can acquire a lost
article by means of his hatzer since, in all other instances than that
of ketanah, hatzer is not because oiyad but because of shelihut. The
argument is that since there are minor females who require to have
a hatzer for the purpose of divorce the Torah would not give them
this hatzer in part, but the Torah gives it to them categorically.
Hence with regard to ketanah we say, since she has a hatzer for the
purpose of acquiring her get she has it, too, for the purpose of
acquiring a lost article. But R. Johanan in the name of Abba Kohen
Bardala argues that since from the case of ketanah we learn that
hatzer is because oiyad it is illogical to treat it as because oi shelihut
in the case of a katan. Hence, according to this ' If you want I can
166 The Talmudic argument
say', Resh Lakish in the name of Abba Kohen Bardala did not refer
at all to a ketanah but to a katan.
This is one of numerous instances in the Talmud where it is stated that
Rabbi A said... and the Talmud itself later suggests that he did not in fact
say it or did not say it in these words. At first in our sugya Resh Lakish in
the name of Abba Kohen Bardala is quoted as saying:' A ketanah has neither
hatzer nor four cubits' whereas it now turns out that what he really said
was: ' A katan has neither hatzer nor four cubits' and the same applies to
the saying of R. Johanan in the name of R. Jannai. Cf. my article: ' How
much of the Babylonian Talmud is pseudepigraphic?' in JJS, 28, 1 (Spring
1977),
Finally, the Talmud concludes, ' If you want I can say' that in
fact there is no debate at all between Resh Lakish in the name of
Abba Kohen Bardala and R. Johanan in the name of R. Jannai.
They speak of two different cases. Thus Resh Lakish in the name
of Abba Kohen Bardala speaks ofhatzer for the purpose of acquiring
a lost article, while R. Johanan speaks of hatzer for the purpose of
acquiring a. get; or Resh Lakish in the name of Abba Kohen Bardala
speaks of a katan and R. Johanan in the name of R. Jannai of a
ketanah.
The pattern of the sugya is:
(1) Resh Lakish - Abba Kohen Bardala: ketanah has no hatzer
R. Johanan - R. Jannai: ketanah has hatzer
(2) Point at issue: whether hatzer because oiyad or shelihut
(3) Question: hatzer- theft - but ' no agency where crime
committed' ?
(4) Reply I: Ravina: only where agent has obligation
(5) Question: but what of married woman and slave?
(6) Reply: they do have obligation but no money
(7) Reply II: R. Sama: only where voluntary
(8) Question: what is practical difference between (4) and (7)?
(9) Reply: priest and divorcee - man who instructs woman to round
head
(10) Question: but hatzer in case of get derived from yad?
(11) Reply I: debate only with regard to lost article
(12) Reply II: debate only with regard to katan not ketanah
(13) Reply III: no debate - each deals with a different case
The artificial element in this sugya is particularly obvious. First
there is the strange phenomenon (see Tosqfists) that in the case of
get the proof that hatzer is because ofyad is from a verse that includes
hatzer in a verse dealing withyad. Why, then, does the Tal mud not
use the very same method to prove it from the original verse of theft
Acquisition by means of a domain 167
where, in exactly the same way, hatzer is included in a verse dealing
withy ad? Even after the difficulty of' no agency where a crime is
committed' has been disposed of, this same verse can still be used,
as the other one is, to prove that hatzer is because oiyad. It looks
as if this whole section has, in fact, only been introduced in order
to discuss not the question ofhatzer but the principle of' no agency
where a crime is committed' and when this applies. Secondly, the
final series of replies in the form of If you want I can say' effectively
contradicts the whole of the opening passage and has been kept to
the end so as to provide a startling denouement to the discussion.
Our sugya, in connection with the principle of' no agency where an offence
is committed' is discussed at length in a Responsum of Ezekiel Landau, Noda'
Bi-Thudah, Kama, Even ha-Ezer, no. 78. The theme of hatzer as a mode of
acquisition is found in numerous instances in the Talmud but the theme
of our sugya - whether hatzer is because oiyad or because of shelihut - is also
referred to in Gittin 21a; 77b; Bava Metzi'a 12a. See Kesef JVivhar, part 11,
no. 66, pp. 23a-25b.
Palginan be-dibbura: admission of part of a
testimony even though another part of the
same testimony is rejected
The sugya is Sanhedrin gb-ioa and is in the form of a discussion
around legal statements of the fourth-century Babylonian Amoraim
R.Joseph and Rava. The term palginan be-dibbura means literally:
'we (i.e. the Court or the Rabbis) divide up the word (e.g. of a
witness's testimony)'. According to the Rabbinic interpretation of
Deuteronomy 24: 16, the testimony of a man against his own near
relative, his karov, is unacceptable, i.e., if a father testifies that his
son has committed a crime that testimony is rejected. Again,
according to the Rabbinic interpretation of: ' put not thine hand
with the wicked to be an unrighteous witness' (Exodus 23: 1), a
'wicked' person, i.e. one who has committed a crime, cannot serve
as a witness. Now it is held that a man is a karov (a ' near relative')
to himself and consequently a man's confession that he has
committed a crime is also rejected. If, therefore, a man testified in
Court together with another (two witnesses being required in a
criminal charge) that he has committed a crime, his own testimony
against himself is rejected so that only one witness remains and he
cannot be convicted. Supposing, now, A and B testify that Chas
committed sodomy with A and with A's consent so that A's
testimony is, in fact, that he himself is 'wicked'. Obviously, as
above, A cannot be convicted on his own testimony, but can his
testimony together with that of B serve to convict C? Since his
testimony against himself is rejected he is not 'wicked' and,
therefore, his testimony that Chas committed sodomy is accepted.
It can be argued that when A testifies: ' Chas committed sodomy
with me and with my consent', since that part of it relating to
himself, 'with me and with my consent', is rejected the whole of
the testimony is rejected. On the other hand, it can be argued that
168
Acceptance of partially rejected testimony 169
while this part of A's testimony ('with me and with my consent')
is rejected, the rest of the testimony ('Chas committed sodomy')
is accepted so that Ccan be convicted on the strength of A and
B's testimony. If this second line of argument is followed the
principle of palginan be-dibbura is said to be accepted. The Court
rejects that part of the testimony relating to A's own guilt but
accepts that part relating to C's guilt.
The sugya opens with a statement of R. Joseph dealing with the
laws of evidence. This is inserted here because the previous
discussions in this section of tractate Sanhedrin have to do with
questions of evidence in a Court of Law. This first statement of
R. Joseph is then followed by a second statement of R. Joseph
regarding testimony and this second statement has to do with our
question of palginan be-dibbura.
R.Joseph's first statement is with regard to the following. It is
an accepted ruling that if a person commits an act which makes
him liable to the death penalty and to pay compensation, since
he is liable to suffer the more severe death penalty he is free from
the lesser penalty of compensation. In the technical terms used in
our sugya, where there are nefashot ('souls', 'lives', i.e. the death
penalty) there is no mamon. If a husband produces witnesses who
testify that his wife has committed adultery, their testimony seeks
to achieve two things: (a) her condemnation to death; (b) her
forfeiture of her ketubah (her marriage settlement). According to the
law of false witnesses (zomamim), these, when their falsehood has
been established, have imposed on them the penalty they sought
to impose on their innocent victim (Deuteronomy 19:19). Thus if
the father of the wife then produced two witnesses to testify that
the witnesses produced by the husband are false, these witnesses,
although they wished to inflict two penalties on the wife - the death
penalty and the forfeiture of her ketubah are only liable to the
death penalty and they do not have to pay the value of the ketubah.
As above, where there are nefashot there is no mamon. Now the
testimony of these second two witnesses - those produced by the
wife's father - intends to achieve two things: (a) the death penalty
for the first two witnesses - produced by the husband - and (b) the
obligation of the husband to pay his wife the ketubah, since
according to their testimony she is innocent and therefore entitled
to her ketubah. Now if the husband then produces a third pair of
witnesses to prove the second pair false, R. Joseph rules that this
17 The Talmudic argument
second pair have to suffer the death penalty and, in addition, to
pay the husband for the ketubah from which they sought to deprive
him. The reason why the principle 'where there are nefashot there
is no mamon' does not apply here is because that principle only
applies where the offence for which there is the death penalty and
the offence for which mamon is paid are offences against the same
person, as in the first case. Here, however, the death penalty is for
the witnesses' offence against the first pair of witnesses whereas the
mamon is for their offence against the husband. This is a case of
'nefashot to one and mamon to another' and R. Joseph declares that
in such cases both penalties are imposed.
There now follows R. Joseph's second statement. If A together
with B, as above, testifies that Chas committed sodomy with A,
then if A's testimony is that the act had been done against his will,
i.e. he had been raped by C, he is a valid witness and Cis convicted
on the strenth of A's and B's testimony. But if A's testimony is to
the effect that he, A, had been a willing partner to the act, he
thereby disqualifies himself as a 'wicked' person (rasha') from
acting as a witness and there remains only B's testimony, on the
strength of which alone Ccannot be convicted. Rava, however, as
above, holds that palginan be-dibbura. A's testimony regarding his
own willing participation in the act is disregarded but the rest of
his testimony - that Chas committed sodomy - is accepted and C
is convicted on the strength of A and B's testimony. Thus on the
question of palginan be-dibbura R.Joseph and Rava disagree.
R. Joseph rejects the palginan be-dibbura principle, hence since A's
testimony regarding his own participation is rejected the whole of
the testimony is rejected. Rava, on the other hand, does accept the
palginan be-dibbura principle, hence while A's testimony regarding
his own willing participation is rejected, the rest of the testimony
stands that Chas committed sodomy and Cis convicted on the
strength of A and B's testimony.
The question of when exactly the Jewish Courts lost the right to impose
the death penalty is complicated (see Excursus xxx:' The Abolition of Capi-
tal Punishment' in Sydney B. Hoenig, The Great Sanhedrin (Philadelphia,
1953), pp. 21113) but it was certainly centuries before the period of the
Babylonian Amoraim R. Joseph and Rava. This and all such discussions
in the Babylonian Talmud are purely academic. There is, in fact, no
evidence that at a time when the ancient Jewish Courts did execute
criminals the criminal's confession was rejected.
Acceptance of partially rejected testimony 17
J
A second statement of Rava is now recorded. A and B testify that
Chas committed adultery with A's wife. A's testimony against his
own wife is rejected. He is a karov to his wife. The wife cannot,
therefore, be convicted. But so far as Cis concerned there are still
two witnesses, A and B, that he has cohabited with a married
woman. Thus Rava states that Cis convicted on the strength of
A's and B's testimony, A's testimony being rejected in part - that
C's adultery was with A's wife - but accepted in part - that Chad
cohabited with a married woman.
The Talmud objects that this second case is, in fact, identical with
Rava's first case. Here, too, the reason Rava accepts A's testimony
is because ofpalginan be-dibbura. Why, then, does Rava repeat his
ruling. The reply is that if Rava had not made his second statement
we might have supposed that while ' we say a man is a karov to
himself we do not say a man is a karov to his wife'. Therefore, Rava
is obliged to repeat the case so as to show that it applies also where
A's testimony is not against himself but against his wife.
See Tosqfists, s.v. etzel ishto lo amrinan that, in fact, a husband is disqualified
from acting as a witness against his own wife. Tosqfists, therefore, understand
our passage to mean that here, and here alone, we might have supposed
that the husband is believed, even against his wife, since his statement that
Ccommitted adultery with her is believed.
There now follows a second statement of Rava. If witnesses testify
that a man had had intercourse with a betrothed maiden, without
specifying which maiden, and they are found to be false, they are
executed because they wished to impose the death penalty on the
man, but there is no payment of mamon for an attempted loss of the
ketubah since they had not named the maiden. If, however, they did
specify 'with the daughter of So-and-so' and are subsequently
found to be false, they surfer the death penalty and, in addition,
must pay the value of the ketubah to the girl's father, to whom the
ketubah is due. This, as above, is a case oVnefashot to this one and
mamon to another' . As Rashi points out, this is different from the
first case of R. Joseph (where a husband produces witnesses to his
wife's infidelity and they were proved false, where, as R.Joseph
states, there is no mamon) since there the woman was not merely
betrothed, but actually married, so that the ketubah is hers and it
is a case of nefashot to the same person.
A third statement of Rava is now quoted. Two witnesses testify
that a man had commited bestiality with an ox and they are
172 The Talmudic argument
subsequently found to be false; they are executed because of their
intention to have their victim executed, but there is no payment
of mamon since no particular ox was specified. If, however, they
testify that it was ' with the ox of So-and-so' and are found to be
false, then they have, in addition, to pay mamon to the owner of the
ox, the law being that the ox has to be stoned, and so the false
witnesses had intended to deprive the owner of his ox. This, too,
is a case of nefashot to one [the man against whom they testified]
and mamon to another [the owner of the ox]. But, the Talmud
objects, this is exactly the same as the case of the betrothed maiden.
Why, then, does Rava have to repeat it? The reply is that Rava
repeats it not because it is necessary to do so for itself but because
he wishes to lead into a problem he had set regarding testimony
against an ox.
This problem of Rava's is as follows. A and B testify that Chad
committed bestiality with A's ox. Cis certainly to be executed on
the evidence of A and B, but is the ox to be destroyed? Here it can
be argued that just as we say, in Rava's earlier case, ' a man is a
karov to himself so, too, we say ' a man is a karov as far as his own
ox is concerned
5
, i.e. A's testimony that his ox is to be destroyed
is not accepted since this, too, constitutes testimony against himself.
Or, possibly, we do accept his testimony regarding the ox and the
ox is destroyed. Rashi points out that, in fact, a man is not believed
normally to testify that his own ox is to be destoyed, since this is
a testimony against himself; but here it is possible that the
testimony is accepted for if it were not it would have to be because
of the palginan be-dibbura principle and it is possible that, while this
principle is applied so far as A himself is concerned, it is not applied
to his testimony regarding his ox. The Talmud concludes that after
Rava had set the problem he eventually concluded that A's
testimony would be accepted even to destroy his ox and here we
do not apply the palginan be-dibbura principle. As the Talmud puts
it: 'After he [Rava] had set the problem he solved it thus: We do
say that a man is a karov to himself but we do not say that a man
is karov so far as his mamon is concerned.'
The pattern of the sugya is:
(1) R. Joseph I: mamon to one and nefashot to another
(2) R.Joseph II: A and B testify that Gcommitted sodomy with A
and with A's consent; A disqualified, therefore Gacquitted
(3) Rava I: palginan be-dibbura: Cconvicted
Acceptance of partially rejected testimony 173
(4) Rava II: A and B testify that Ccommitted adultery with A's wife:
the wife acquitted but Cconvicted: because palginan be-dibbura
(5) Objection: (3) and (4) identical: why repeat?
(6) Reply: in (4) we might have supposed that, since A's testimony
is accepted vis-a-vis C, it is also accepted vis-a-vis the wife
(7) Rava III: mamon to one and nefaskot to another
(8) Rava IV: ox: mamon to one and nefashot to another
(9) Objection: (7) and (8) identical: why repeat?
(10) Reply: (8) required because of Rava's problem
(11) Rava V: problem: A and B testify that Ccommitted bestiality with
A's ox
(12) Rava's solution: ox stoned because A's testimony accepted
The artificial element is particularly obvious with regard to (9) to
(12), but it all fits in neatly with the logical sequence of the sugya.
To be noted especially is the progress of thought from ' a man is
a karov to himself to ' a man is a karov to his wife' to Rava' s problem
as to whether he is a karov to his mamon (his ox).
Other references to palginan be-dibbura are: Tevamot 25a-b (implied); Gittin
8b; Bava Batra 134b; Sanhedrin 25a (implied). See KesefNivhar, part m, no.
135, pp. 48b-5ob.
i8
Tadir u-mekuddash: which takes precedence:
the more constant or the more sacred?
The sugya is evahim 9ob-gia. The Mishnah (Zevahim 10: i) states
with regard to sacrifices: 'Whatever is more constant than another
takes precedence over the other.' Three examples are given in the
Mishnah. (i) The daily offerings (the 'perpetual' offerings) take
precedence over the special, additional Sabbath offerings, i.e. the
morning 'daily offering' of the Sabbath is offered before the
additional offerings of the Sabbath. The daily offerings are ' con-
stant ' whereas the additional offerings are limited to the Sabbaths,
Festivals and New Moons. (2) If the New Moon falls on the
Sabbath, the additional offerings of the Sabbath are offered before
the additional offerings of the New Moon, the Sabbath, falling
every week of the year, being more 'constant' than the New Moon.
(3) The New Year Festival falls on the New Moon of the seventh
month. The New Moon additional offerings are offered before the
additional New Year offerings, the New Moon being more
'constant'. The term used for 'constant' is tadir ('regular'). The
next Mishnah (evahim 10:2) continues: 'Whatever is more sacred
than another takes precedence over the other.' Thus the sprinkling
of the blood of a sin-offering has to be done before that of a
burnt-offering, the sin-offering being more ' sacred' in that it atones
for sin. The term used in the Mishnah for ' sacred' is mekuddash. The
Mishnah goes thus far. The problem discussed in our sugya is: which
takes precedence where A is tadir (more 'constant' than B) but B
is mekuddash (more 'sacred' than A) ? For example, the blood of the
daily offering and the blood of a sin-offering are to be sprinkled
on the altar. Which of them is to be sprinkled first? The blood of
the daily offering is tadir (i.e. is more 'constant' than that of the
More constant or more sacred: which takes precedence? 175
sin-offering) but the blood of the sin-offering is mekuddash (i.e. is
more ' sacred' than the blood of the daily offering).
The sugya opens with the problem set by the anonymous scholars,
with the usual formula for this: 'Ibbaya le-hu', ' They set a
problem'. The problem is formulated as: ' Tadir and mekuddash,
which of these takes precedence? Does tadir take precedence
because it is tadir or does mekuddash take precedence because it is
mekuddash?", i.e. which is the more significant, the tadir principle (so
that it overrides the mekuddash principle) or that of mekuddash (so
that it overrides the tadir principle)? A solution is attempted
(introduced with the usual formula ta shema\ 'come and hear')
from the first example given in the first Mishnah, that the daily
offerings take precedence over the additional offerings of the
Sabbath. Now the daily offerings, precisely because they are offered
on week-days as well as on the Sabbath, are less sacred than the
special, additional Sabbath offerings, which acquire, as it were, the
special sanctity of the Sabbath, and yet the daily offerings take
precedence. This demonstrates that tadir takes precedence over
mekuddash. The reply is given: ' Is the Sabbath only effective for the
additional offerings and not for the perpetual offerings?', i.e. the
daily offering brought on the Sabbath also has imparted to it the
special sanctity of the Sabbath and thus is different in terms of
sanctity from the daily offerings brought on week-days. Thus there
is no greater sanctity to the additional offerings brought on the
Sabbath than to the daily offerings brought on the Sabbath.
Consequently, here the question of mekuddash does not arise at all,
both possessing the same degree of sanctity. Therefore, the daily
offering, being tadir, takes precedence. The point at issue here
would seem to be whether the 'daily' offering brought on the
Sabbath is no more than a 'daily' offering (no different from those
brought on week-days and only brought on the Sabbath because
the Sabbath is also a day of the week) or whether the Sabbath
imparts its sanctity even to the daily offering so that, on the
Sabbath, this is, in fact, a special Sabbath offering.
The next attempted solution is from the second example given
in the Mishnah, that the additional sacrifices of the Sabbath take
precedence over the additional sacrifices of the New Moon. Now
the additional sacrifices of the New Moon are more ' sacred' than
those of the Sabbath since (Rashi) the New Moon is called a
176 The Talmudic argument
'Festival' and (SM) there is a special New Moon sin-offering. This
demonstrates that tadir takes precedence over mekuddash. A similar
reply is given: ' Is the New Moon only effective for its own
additional offerings and not for those of the Sabbath?' , i.e. the
sanctity of the New Moon is imparted also to the additional
Sabbath offerings. Both are thus equally sacred and tadir takes
precedence.
The next attempted solution is from the third example of tadir
given in the Mishnah, that the additional sacrifices of the New
Moon take precedence over those of the New Year, even though
the New Year, as a major festival, is obviously more 'sacred' than
the minor festival of the New Moon. This demonstrates that tadir
takes precedence over mekuddash. Again the same kind of reply is
given: ' Is the New Year only effective for its own additional
offerings and not for those of the New Moon?', i.e. the New Year
imparts its sanctity to the New Moon offerings as well as to its own
additional offerings. Both are equally sacred and hence tadir takes
precedence.
A Baraita (Tosefta, Berakhot 6: 1) is now quoted. Here it is
recorded that the sanctification benediction - kiddush - recited at
the Sabbath meal consists of two separate benedictions, the one over
the wine and one for the ' day' , i.e. praising God for giving Israel
the Sabbath, and the former must be recited first because it is tadir,
i.e. since it is recited not only on the Sabbath but whenever one
partakes of wine. Now the second benediction, as the special
Sabbath benediction, is more ' sacred' than the first and yet the first
takes precedence, showing that tadir takes precedence over
mekuddash. Again a similar reply is given: ' Is the Sabbath only
effective for the benediction of the day and not for the benediction
of the wine?', i.e. the Sabbath imparts its sanctity to the wine
benediction as well as to the benediction of the day; the wine
benediction becomes, as it were, not an ordinary wine benediction
but a special Sabbath benediction. Consequently, both are equally
'sacred' and tadir takes precedence.
A fifth solution is now attempted. The Sages and R. Judah
debate {Berakhot 28a) which takes precedence, the additional
Sabbath prayer or the Sabbath afternoon prayer. The early-
third-century Palestinian Amora, R. Johanan, states the rule that
the Sabbath afternoon prayer must be recited first since it is tadir,
this prayer being recited on every afternoon of the year whereas
More constant or more sacred: which takes precedence? 177
the additional prayer is only recited on the Sabbaths and Festivals.
But the additional prayer, as a special Sabbat h prayer, is more
' sacred' and yet tadir takes precedence. Again the reply is given:
' Is the Sabbat h only effective for the additional prayer and not for
the afternoon prayer?' , i.e. and the afternoon prayer is also a
special Sabbat h prayer so that both are equally sacred, hence tadir
takes precedence.
These five attempted solutions to our problem were to prove t hat
tadir takes precedence over mekuddash. They all failed to establish
this. At this stage a proof is attempted that, on the contrary,
mekuddash takes precedence over tadir. This is from a further
Mishnah (evahim 10: 6) in the same chapter; in fact, the Mishnah
to which our sugya is appended! Here the following case is debated.
Priests in the Temple have in front of them, for eating, the meat
of peace-offerings brought yesterday and the meat of sin-offerings
brought today. Which meat is to be eaten first? The Sages hold
that the meat of the sin-offerings is to be eaten first, because a
sin-offering is more ' sacred' than a peace-offering. But R. Meir
holds that the meat of the peace-offerings should be eaten first
because, the sacrifice having been brought yesterday, this meat may
only be eaten until nightfall, whereas the meat of the sin-offering,
the sacrifice having been brought today, may be eaten until
tomorrow morning. It is better, according to R. Meir, to ignore the
more sacred offering because the demands of the peace-offering are
more pressing. It follows that where this reason would not apply,
e.g. where both the peace-offerings and the sin-offerings had been
brought today, even R. Meir would agree that the sin-offerings take
precedence because they are more ' sacred' . But the peace-offerings
are more ' const ant ' since they are brought as free-will offerings,
whereas sin-offerings are only brought when sins have been com-
mitted. This demonstrates that mekuddash, the sin-offering, takes
precedence over tadir, the peace-offering.
Rava, the fourth-century-Babylonian Amora, retorts t hat there
is no proof at all from this Baraita. Rava argues t hat a clear
distinction must be made between that which is ' const ant ' , tadir,
and t hat which is merely ' frequent' , matzuy. Tadir refers to that
which occurs at regular intervals in a cycle, such as the daily
offerings or the Sabbat h prayers. It is true that peace-offerings are
more ' frequent' than sin-offerings but this does not render them
tadir. Our original problem, observes Rava, is whether tadir takes
178 The Talmudic argument
precedence over mekuddash but there is no problem regarding
whether matzuy takes precedence over mekuddash. It certainly does
not, which is why the Mishnah implies that the sin-offerings, which
are mekuddash, take precedence over the peace-offerings, which are
only matzuy', 'frequent', but not tadir, 'constant'.
A Baraita (Sifra, hova 1: 6) is now quoted by R. Huna bar
R. Judah, through which he challenges Rava's distinction between
tadir and matzuy. We see from this Baraita that matzuy is referred to
as tadir. The law is that a sin-offering is only brought for an offence,
committed unintentionally, the penalty for which, if committed
intentionally, is karet, 'extirpation'. But although a failure by
intention to bring the Paschal lamb or to be circumcised involves
the karet penalty, there is no sin-offering for unintentional failure
in these instances, the reason being that a sin-offering is only
brought for sins of commission and these are sins of omission. Thus
both the Paschal lamb and circumcision are excluded from the law
of the sin-offering, but in our Baraita there is a special verse to
exclude the Paschal lamb and another special verse to exclude
circumcision. The Baraita states that were it not for the special verse
to exclude circumcision we could not have derived the exclusion
from the verse excluding the Paschal lamb, since the Paschal lamb
is less severe than circumcision in that the Paschal lamb is not tadir
while circumcision is tadir. Now circumcision is not tadir in the sense
of'regular'. It is only more 'frequent' than the Paschal lamb since
circumcision happens all the time, whereas the Paschal lamb can
only be brought once a year. Thus circumcision is matzuy,
'frequent', and yet the Baraita refers to it as tadir. This refutes Rava
who has argued that matzuy does not qualify as tadir.
Two replies are given. The first is that when the Baraita speaks
of circumcision in relation to the Paschal lamb as tadir it does not
mean tadir in time but 'tadir in commandments', i.e. there are far
more injunctions in Scripture regarding circumcision than regard-
ing the Paschal lamb. It is in this sense alone that the term tadir
is used of circumcision. But when the term tadir is used in a time
sense it is never used of that which is only matzuy, 'frequent'. The
second reply (prefaced by: ' If you want I can say') is that
circumcision is referred to as tadir, even in the time sense, but only
when comparing circumcision to the Paschal lamb. That is to say,
generally speaking matzuy does not qualify as tadir, which is why
peace-offerings do not take precedence over sin-offerings, even if
More constant or more sacred: which takes precedence? 179
tadir does take precedence over mekuddash. Even so the matzuy
principle does have some significance, though not sufficient sig-
nificance to override the mekuddash principle. Hence, circumcision,
which is matzuy, cannot be derived from the Paschal lamb, which
is not. It is in this sense that the Baraita speaks of matzuy as tadir.
The above is a plausible way of understanding this part of the sugya. The
objection, however, is why does the Baraita refer to matzuy as tadir? This
is probably why Rashi offers a different explanation. According to Rashi
the meaning of this section is that, for something to qualify as tadir merely
because it is matzuy, it must be very frequent in relation to that to which
it is compared. Now circumcision is far more frequent than the Paschal
lamb, which is only brought once a year, whereas even though it is true
that peace-offerings are brought more frequently than sin-offerings they
are not sufficiently more frequent to enable their matzuy to qualify as tadir.
The most puzzling feature of our sugya is the repeated attempt
at a solution to the problem when the reply given as to why there
is no solution in the first instance would seem to apply to all the
attempts. (See Tosqfists s.v. ta shema\ who raise this question
without supplying an answer.) What is the point in attempting
what seems to be the same solution five times and in each instance
giving what seems to be the same reply in almost identical words?
Once the first reply had been given could we not have supplied it
ourselves to the other four instances? But the truth is that we have
here a beautiful example of how, without stating it explicitly, the
Talmudic sugya leads from one argument to another. A careful
examination of the five instances, demonstrates that the stock reply:
' Is A only effective?' becomes progressively weaker so that the sugya
is really saying: 'This reply is all very well in case (1) but what
of (2) ?' and so on. Thus while it may well be accepted, in the first
instance, that the daily offering of the Sabbath has imparted to it
the special Sabbath sanctity, it is more difficult to see how, in the
second instance, the New Moon sanctity is imparted to the Sabbath
offerings, since these have nothing to do with the New Moon. And
even when it is suggested that these do have imparted to them the
New Moon sanctity it is difficult to see how, in the third instance,
the New Year sanctity can be imparted to the New Moon sacrifices,
since, by definition, New Year is New Tear and not New Moon. And
even when this, too, is accepted, it is difficult to see how, in the
fourth instance, an ordinary wine benediction can be said to enjoy
the special Sabbath sanctity. And even when this is accepted, it is
180 The Talmudic argument
difficult to see how the Sabbath prayer, which is merely an ordinary
'afternoon' prayer, with no connection at all with the additional
Sabbath prayer (unlike the wine, which has some connection with
the second benediction, since they are always recited together), can
have imparted to it the sanctity of the additional prayer. (This
argument ignores the special nature of the Sabbath afternoon
prayer. Evidently, the argument follows the line that the Sabbath
afternoon prayer is merely the daily afternoon prayer with reference
to the Sabbath and with the omission of the petitionary prayers
which must not be recited on the Sabbath.) Thus the reply: ' Is A
only effective', given in all five instances, is not, in fact, the same
reply at all. This helps us to understand why the attempted proofs
are presented in this order. They are so presented because only in
this way can the reply given be seen to become progressively
weaker. Had the final instance been recorded first and the reply
given to it where, as above, it is weakest, all the other attempted
solutions would be unnecessary because we would have appreciated
that if the reply operates successfully in its weakest form it operates
a fortiori in its stronger forms.
We can now observe the pattern of the sugya:
(1) Problem: which takes precedence: tadir or mekuddash?
(2) Proof that tadir takes precedence: daily and Sabbath offerings
(3) Reply: no proof, there both mekuddash
(4) Proof II that tadir takes precedence: Sabbath and New Moon
offerings
(5) Reply: no proof, there both mekuddash
(6) Proof III that tadir takes precedence: New Moon and New Year
offerings
(7) Reply: no proof, there both mekuddash
(8) Proof IV that tadir takes precedence: benediction of wine and day
(9) Reply: no proof, there both mekuddash
(10) Proof V that tadir takes precedence: Sabbath afternoon and
additional prayer
(11) Reply: no proof, there both mekuddash
(12) Proof that mekuddash takes precedence: meat of peace-offering and
of sin-offering
(13) Reply: there not tadir only matzuy, Rava
(14) Objection to Rava: Paschal lamb and circumcision
(15) Reply I: tadir there means 'in commandments'
(16) Reply II: only in relation to Paschal lamb
We have already observed the neat sequence of thought in (1)
to (11). Further to be noted is that (12), the proof that mekuddash
More constant or more sacred: which takes precedence? 181
takes precedence, is on the Mishnah to which the whol e sugya has
been appended. Why, then, was this proof not adduced first? The
answer is obvious. This attempted proof that mekuddash takes
precedence could only have been produced after the other proofs,
from the Mishnah, had been refuted, otherwise the Mishnah woul d
be in flat contradiction, at first teaching that tadir takes precedence
over mekuddash and then impl ying here that mekuddash takes prece-
dence over tadir.
The other Talmudic passages dealing with our theme are: Berakhot 28a;
51b; Shabbat 23b; Pesafyim 114a; Sukkah 56a; Megillah 29b; Menahot 49a.
See Kesef Nivfyar, part in, no. 154, pp. 39a~4oa for a full discussion of the
theme. For the question of precedence with regard to the rights of persons
see Bava Metzi'a 33a-b; Bava Batra 9a; Horayot 13a-14a.
19
Palga nizka: the nature of the payment of
half-damages to which the owner of a goring
ox is liable
The sugya is Bava Kama i5a-b. Based on Exodus 21: 35-6, a
distinction is made between an ox that has only gored once or twice
and an ox that has gored three times. The former is known as tarn
('docile', ' tame' ), the latter as mu'ad ('warned', ' an habitual
gorer'). If a mu'ad gores another animal the owner of the goring
ox has to pay the full amount of the damage but the owner of a
tarn that gores only has to pay a half of the damage. A second
distinction between the two is that the owner of a mu'ad is liable
to make good the damage out of his own pocket, irrespective of the
value of his ox, but the owner of the tarn is only liable to pay out
of the value of his ox. If the goring tarn is not worth half the amount
of the damage the owner is not obliged to make good the
remainder out of his own pocket. A further law on which our sugya
is based makes a general distinction between payment that is for
the purpose of compensation and payment that is for the purpose
of imposing a fine. The former is known as mamona, ' money' , i.e.
an amount to which a victim is entitled because of his loss. The
latter is known as kenasa, ' a fine', the purpose of which is not to
compensate the victim but to penalise the victimiser. For example,
if a man steals another's property and is later found out, he has
to pay double the amount he had stolen. The payment of the
amount he had stolen is mamona, to which the victim of the theft
is obviously entitled. But the extra amount, to which the victim has
no title, since this was not stolen from him, is kenasa, a fine imposed
as a penalty on the thief. The practical difference between the two
kinds of payment is that a man's own admission of his guilt releases
him from payment of kenasa but not, of course, from the payment
otmamona. Thus if a thief stole 100 zuz and later admits to his guilt,
182
Payment of half-damages: compensation or fine? 183
he has to pay only 100 zuz and not the other 100 he would have
had to pay if he had not admitted his guilt but was found out.
The sugya opens with a debate between the late-fourth-century
Babylonian Amoraim, R. Pappa and R. Huna son of R. Joshua,
regarding the nature of the half-damages to which the owner of a
tarn is liable. R. Pappa holds that this payment is mamona. The
owner of the ox the tarn has gored is entitled to this amount as
compensation. R. Huna son of R. Joshua holds that it is kenasa, a
fine imposed on the owner of the tarn. Thus if the owner of a tarn
admits of his own accord that his tarn has gored another animal,
then according to R. Huna son of R. Joshua he does not have to
pay, since there is no payment of kenasa where there is an admission
of guilt.
The Talmud explains thus the reasoning of R. Pappa and
R. Huna son of R. Joshua. The suggestion is made that it all
depends on whether it can be assumed that oxen are ' in a state
of protection' (be-hezkat shimur) so far as goring is concerned, i.e.
whether it can be assumed that no special protection is required
to prevent a tarn, a docile ox, breaking out to gore. R. Pappa, it
is suggested, holds that even a docile ox cannot be assumed to
remain permanently docile and the owner is, therefore, obliged
constantly to be on his guard. Hence if the tarn does gore this
demonstrates that the owner has failed in his duty to guard it
adequately and he should, strictly speaking, be obliged to make
good the whole of the damage. Since, however, it is, after all, a tarn,
Scripture lets him off lightly and deducts from his payment half
the amount of the damages. R. Huna son of R. Joshua holds that
it can be assumed that a tarn will not gore, so that if it does this
betokens no evidence that the owner has failed in his duty to guard
it. Strictly speaking, therefore, he should not be liable to pay
anything at all; but Scripture does, none the less, impose a penalty
to make him take better care in the future, this penalty being the
fine of half the damages. In other words, according to R. Pappa
the half payment is for the damage done, while according to
R. Huna son of R. Joshua it is to prevent damage being done by
the ox in the future. Thus according to R. Pappa the payment is
mamona, compensation for the damage done, while according to
R. Huna son of R. Joshua it is kenasa, payment to prevent damage
being done in the future.
The significance of the debate as thus formulated appears to be
184 The Talmudic argument
this. On the face of it, payment of half the damage is a curious idea.
There cannot be any such thing in law as partial responsibility
for damages. Either the requisite degree of responsibility has been
attained or it has not. The person who is only half responsible is
not responsible at all in law. Consequently, in the case of the tarn
where the owner has to pay only half the damage, one is obliged
to say that, strictly speaking, the owner is either responsible, and
should, therefore, pay the full damage, or not responsible, and
therefore should pay no damages at all. And yet Scripture does
depart from what the strict law should be, stating that the payment
is half the damages. The question, then, is whether this departure
is to render the owner liable to pay the half imposed on him
(because, strictly speaking, he should not have been obliged to pay
any damages) or whether the departure is to free him from the other
half (because, strictly speaking, he should have been obliged to pay
in full). R. Pappa holds that the departure is to free him from
payment of the other half, the half he does pay being no departure
at all but the compensation demanded by law. R. Huna son of
R. Joshua holds that the departure is to impose the half-payment,
since strictly speaking, the victim is not entitled to compensation
at all.
The Mishnah (Bava Kama 1: 3), to which the sugya is appended,
is quoted. The man whose ox has been gored, the man suffering
the damage, is known as the nizzak. The owner of the goring ox,
the man responsible for the damage, is known as the mazzik. The
Mishnah states that in cases of payment for damages ' both the
nizzok and the mazzik have to pay' . This is a puzzling statement.
How can the nizzak be said to pay? It is to him that the mazzik pays.
According to R. Pappa we can well understand the meaning of this
clause in the Mishnah. According to him the nizzak, where the
goring ox is a tarn, is, strictly speaking, entitled to full compensation
so that when he receives only half it can be said that he, as well
as the mazzik, ' pays' . But according to R. Huna son of R.Joshua
the nizzak, strictly speaking is not entitled to anything at all. Even
the half he does receive is not in the nature of compensation to him
but only as a penalty for the mazzik. What sense, then, does it make
for the Mishnah to state that the nizzak also ' pays' ? The reply is
that the nizzak sometimes ' pays' in that he does not receive even
the half damages awarded to him by the Torah. This is what the
Mishnah means by the nizzak ' pays' . The circumstances are where
Payment of half-damages: compensation or fine? 185
the carcass of the nizzak's ox deteriorates after it has been killed so
that by the time the Court pronounces its verdict its value is
considerably less. It is the nizzak, not the mazzik, who has to bear
this loss. If, for example, the ox was worth 200 zuz before it was
gored and the carcass is now worth 40 zuz, the amount of the
damage is 160 zuz and, since the goring ox was a tarn, the owner,
the mazzik , has to pay 80 zuz- But supposing, before the verdict of
the Court is declared, the carcass so deteriorates that it is now
worthless. We might have supposed that the total damage is now
200 zuz and the mazzik should be obliged to pay 100 zuz> But the
law is not so. The damages are assessed as at the time of the killing
of the ox and any further deterioration is not taken into account,
the nizzak suffering the loss. This is the meaning of the nizzak, too,
having to ' pay' .
An objection is now raised that this law that the nizzak bears
the loss through deterioration of the carcass - is, in fact, derived
from an earlier Mishnah (Bava Kama 1: 1) as is stated in a Baraita
(Bava Kama 10b) commenting on this earlier Mishnah. Why, then,
repeat it here if this is what our Mishnah means? The reply is that
a repetition is, indeed, necessary to demonstrate that this law
applies both to tarn and mu'ad. It does not follow that because the
nizzak has to bear the loss through deterioration in the case of tarn
that he has to bear it in the case ofmu'ad, since, it might have been
argued, tarn is, in any event, treated more leniently than mu'ad.
Conversely, it does not follow that because the nizzak has to bear
the loss through deterioration in the case ofmu'ad, which is treated,
in any event, more severely, he should certainly have to bear the
loss in the case of tarn. For it may well be that precisely because
mu'ad is treated more severely, in that the nizzak receives the full
amount of damages, it is right that he, the nizzak, should bear the
loss through deterioration. In the case of tarn, on the other hand,
the mazzik is let off lightly in having to pay only half the damages
and here, it might be argued, he should at least be obliged to bear
the loss of deterioration, paying half of that as well. Consequently,
two statements are required to show that both in the case of tarn
and of mu'ad it is the nizzak who bears the loss through deterioration.
A further proof is now attempted from the next Mishnah (Bava
Kama 1: 4) which states: ' What difference is there between tarn and
mu'ad? Tarn pays half the damages from itself (i.e. from the value
of the ox, not from the owner's pocket, if the ox is not worth half
186 The Talmudic argument
the damage) whereas mu'ad pays full damages from the aliyah
('upper storey', i.e. from the owner's possessions kept in the upper
storey of his house)'. Thus the Mishnah only records two differences
between the law of tam and mu'ad. But according to R. Huna son
of R. Joshua there is yet another difference: that tam, as kenasa, does
not pay on the owner's admission whereas mu'ad, as mamona, does.
The reply is that the Mishnah could have, indeed, recorded this
further difference but chose simply to omit it, not because it is an
invalid distinction but simply because the Mishnah does not list all
the differences, only some of them. But, the Talmud objects, ' What
else has he omitted that he should omit this?', i.e. the argument that
the Mishnah simply omits a case it could have recorded is only valid
if there is, at least, one other case it omits by choice. If there is only
one omission, this argument cannot be accepted. The logic of this
is that, if the Mishnah omits two or more cases, it can be said that
the Mishnah is not bent on listing all the examples but simply
recording some of them. But if the Mishnah only leaves out a single
instance, listing all the others, this shows that the Mishnah is bent
on listing all the instances and it would then be arbitrary for the
Mishnah to omit a single instance. The reply is that there is, in fact,
yet another omission. If a mu'ad kills a human being the owner has
to pay 'ransom', kofer (see Exodus 21: 28-30), but if a tam kills a
human being the owner does not have to pay even half the kofer.
Consequently, since there is no law of kofer for tam, two further
differences between tam and mu'ad have been omitted: (1) that tam
is kenasa and mu'ad mamona; (2) that there is kofer where the ox is
mu'ad but not even half-kofer where the ox is tam. The Talmud
concludes, however, that this second omission is not really an
omission at all, since the Mishnah may agree with the opinion of
R. Jose the Galilean, who holds that the owner of a tam does have
to pay half-kofer.
See Rashi, who states that the distinction between the full kofer of mu'ad
and the half-kofer of tam, according to R. Jose the Galilean, is no omission,
since this distinction is embraced by the one made in the Mishnah, that
tam only pays half-damages whereas mu'ad pays in full. It is only according
to the opinion of the Sages, who disagree with R. Jose the Galilean and
hold that there is not even half-kofer in the case of tam, that this constitutes
a. further distinction. See Tosqfists, s.v. i mi-shum hatzj, kofer, that the Talmud
means, according to R. Pappa that there is only a single omission - that
of kofer - and this will not do; to which the reply is that the Mishnah follows
the opinion of R. Jose the Galilean. For further Talmudic examples of the
Payment of half-damages: compensation or fine? 187
idea of omission and that it is illogical to omit only a single instance, see
Td'anit 13b; 14a; Sukkah 54a; Tevamot 21 b; 73a; 84b; Ketubot 41a; Kiddushin
16a; 40a; Nazir 38b; Sotah 16a; Bava Kama 10a; 43b; 62b; Makkot 21b.
A further attempt at proving the case is made from a Mishnah
{Ketubot 3: 9) in which instances are given of where payment is
demanded even on the strength of one's own admission (because
it is mamona) and of where no payment is demanded (because it is
kenasa). The payment of thirty shekels when an ox gores and kills
a slave (Exodus 21: 32) is held by all to be kenasa. Thus the Mishnah
states: ' [If a man declares:] "My ox killed So-and-so", or "My
ox killed the ox of So-and-so", he pays on the strength of his own
admission. [But if he declares:] "My ox killed the slave of
So-and-so" he does not pay on the strength of his own admission.'
Now the Mishnah states that when a man declares:
c
My ox has
killed the ox of So-and-so' it refers even to where the ox is a tarn.
This show that the payment of half-damages in the case of tarn is
mamona. No, is the reply, the Mishnah does not refer to tarn but to
mu'ad. But in that case, the Talmud objects, why does the Mishnah
have to give the instances of the ox killing the slave as an example
of no payment because of the owner's admission; he could have
found an example, that of tarn, in the case of the killing of an ox?
The reply is given: ' The whole of the Mishnah deals with mu'ad,
i.e. the Mishnah wishes to inform us that it is kenasa where the ox
kills a slave, even if the ox is a mu'ad.
Another proof is now attempted from the conclusion of this same
Mishnah (Ketubot 3: 9). This reads: 'This is the rule: Wherever the
amount to be paid is greater than the damage done, there is no
payment on the strength of one's own admission', i.e. because it
cannot be compensation since the victim is not entitled to a greater
degree of compensation than the loss he has suffered. The
implication of this Mishnah is that the payment is kenasa where the
payment is greater than the damage done, but where it is less than
the damage done (i.e. in the case of the half-damages of tarn) it is
not kenasa but mamona and this supports R. Pappa. To this the reply
is given, the implication is not that where the amount is less it is
mamona but rather that where it is the same it is mamona. In that case,
objects the Talmud, why does the Mishnah formulate it in the
misleading way: 'Whenever the payment to be made is greater',
which seems to exclude where it is less? Let the Mishnah formulate
it as: ' Wherever the payment is not the same as* and this would only
188 The Talmudic argument
exclude where it is the same as but not where it is either greater or
less. It follow from the formulation in the Mishnah that only where
it is greater is it kenasa, but where it is not greater, i.e. where it is either
the same as or less, it is mamona.
The Talmud concludes: tiyuvta, 'This is, indeed, a refutation.'
But the Talmud continues immediately: 'And the law is in
accordance with the opinion that half-damage is kenasa.' To this
the Talmud objects: tiyuvta ve-hilkheta,'" It is, indeed a refutation",
and yet "And the law is"!' , i.e. how can the law be in accordance
with the view of R. Huna son of R.Joshua since he has just been
refuted? The reply is that the alleged refutation can itself be
refuted. This is because there is another case - that of tzerurot,
' pebbles' - where all agree that although only half the damage is
paid it is mamona not kenasa. If an animal, while walking along in
the normal way, causes 'pebbles' or some other object to spring
from under its feet and these break a vessel, there is a tradition,
said to go back to Moses at Sinai, that only half the damage is to
be paid even though it is mamona (since the reason given above why
the half-damage of tarn is kenasa does not apply here). Consequently,
the Mishnah speaks only of where the amount paid is greater than
the damage done. If the Mishnah were to state, wherever the
damage is not the same as it would have implied that wherever it
is less it is kenasa and while it is perfectly true that the half-payment
of tarn is kenasa it is not true that the half-payment oftzerurot is kenasa.
Thus R. Huna son of R. Joshua is not really refuted and it is
perfectly logical to declare that the law follows his opinion.
The pattern of the sugya is:
1i) Half-damage of tarn: R. Pappa: mamona; R. Huna son of R. Joshua:
kenasa
(2) Reasoning: whether oxen are assumed to gore
(3) Proof I for R. Pappa: Mishnah; the nizzak also pays
(4) Reply: refers to loss through deterioration
(5) Objection: this stated previously
(6) Reply: two statements required, one for tarn and another for mu'ad
(7) Proof II for R. Pappa: only two differences in Mishnah between
tarn and mu'ad
(8) Reply: there is a further difference but omitted
(9) Objection: what else omitted?
(10) Reply: half-kofer also omitted
(11) Elaboration on this: R. Jose the Galilean
(12) Proof III for R. Pappa: Mishnah Ketubot; no payment on strength
of admission where ox gores ox
(13) Reply: this refers to mu'ad not to tarn
Payment of half-damages: compensation or fine? 189
(14) Objection: then why state case of slave, state case of tarn
(15) Reply: all deals with mu'ad, ox and slave
(16) Proof IV for R. Pappa: Mishnah Ketubot: only where greater is it
kenasa
(17) Reply: does not exclude where it is less
(18) Objection: then should have stated: where it is not the same
(19) Tiyuvta, indeed refuted
(20) 'and the law is that it is kenasa'
(21) Objection: since refuted, how can law follow this opinion?
(22) No refutation because less of tzerurot
The sequence is clear. Proof after proof is attempted for R.
Pappa's view and against that of R. Huna son of R. Joshua until
the final, total refutation of R. Huna son of R. Joshua's opinion.
It is possible that (20) to (22) are the result of later editorial work,
but if we are correct that the sugya is so contrived that it leads to
the denouement in the total refutation of R. Huna son of R. Joshua
it is possible, even plausible, to suggest that this, too, is part of the
contrived sugya. That is to say, the impression is first created that
all the arguments favour R. Pappa and none R. Huna son of
R. Joshua, so much so that R. Huna son of R. Joshua is, in fact,
finally refuted conclusively and yet, when all seems lost, the surprise
conclusion is recorded that the law follows the opinion of R. Huna
son of R. Joshua and the refutation is itself refuted.
The next section of the sugya, beginning:' And now that you say' through
to the end, does appear to be a later editorial addition, supplied at a later
date after the whole of the sugya had been completed. This whole section,
in fact, consists of practical rules in the same vein as that used by the earliest
Godifiers in post-Talmudic times. On the phenomenon of'And now' as
a Saboraic addition, see Lewin, Rabbanan Savorai ve-Talmudan, pp. 154-6.
It should be noted that Ketubot 4ia-b is an exact duplicate of our sugya,
including the final section, i.e. the later editorial addition. It has been
inserted in Ketubot on the Mishnah quoted in the sugya. The theme of
half-damage is a
l
Bava Kama' theme, dealt with, in fact, in the previous
pages of this tractate. It seems certain therefore, that originally the sugya
belonged here and was later duplicated in Ketubot by still later editors.
There are thus at least three distinct stages of reduction to be observed:
(1) the original sugya up to (22) as above; (2) the addition of: 'and now
that you say' and the rest of the material until the sugya ends where it now
ends; (3) the appending by still later editors of the whole of (1) and (2)
to the Mishnah in Ketubot. There are cross-references to the debate in our
sugya in Sanhedrin 3a and Shevu'ot 32a.
The final section begins with:' And now that you say, half-damage
is kenasa, if a dog devours lambs or a cat devours chickens, it is
unusual and payment cannot be enforced in Babylon.' The
J
9 The Talmudic argument
meaning of this is that the half-damage of tarn does not only apply
to a goring ox but to any malicious or unusual damage done by
any animal. This is known as keren, ' horn' (see Bava Kama 2a~3b).
Now in Talmudic times only the Palestinian Amoraim had full
ordination, semikhah, and only they, not the Babylonian teachers,
were empowered to impose fines, though they could, of course,
render decisions in cases where compensation was called for
(mamona but not kenasa). If, therefore, the final ruling had been that
the half-damage of tarn is mamona, the Babylonian Courts could have
enforced payment by the owner of the dog or the cat. But ' now
that you have said' that half-damage is kenasa, then the owner of
the dog or cat cannot be compelled to pay by the Babylonian
Courts, who have no semikhah and cannot, therefore, impose fines.
This only applies, however, the Talmud continues, to large lambs
or chickens. If they are small it is ' usual' , so that it is not keren at
all and full payment would have to be made and this can be
enforced in Babylon. Furthermore, the Talmud adds, even where
it cannot be enforced, if the nizzak manages to obtain for himself
some of the mazzik's property to the value of half his damages, he
can hold on to this and the Courts are not empowered to take it
from him. Again, if the nizzak declared that he wished to be given
time to submit his case to a Palestinian Court, he is given the time
and if the mazzik then refuses to have the case submitted to the
Palestinian Court he is placed under a ban until he agrees to do
so. In any event he is placed under the ban until he gets rid of the
vicious dog or cat. In this connection the saying of R. Nathan is
quoted: ' How do we know that one should not keep a vicious dog
or an unsafe ladder in the house? Because Scripture says: "thou
bring not blood upon thine house" (Deuteronomy 22: 8).'
20
Patur mi-diney adam ve-hayyav be-diney shamayim:
cases where there is liability in the eyes of
God even though the human Courts cannot
enforce payment
The sugya is Bava Kama 55b-56a. There are instances of loss and
damage to another where, for various technical reasons, the law
cannot enforce payment but where, nevertheless, there is a moral
obligation to pay. This is called: patur mi-diney adam ve-hayyav be-diney
shamayim, literally, ' exempt from the laws of man but liable by the
laws of Heaven'. Actually the term denotes something more than
a moral obligation to pay. It is rather that there is a legal obligation
to pay, only it is one that cannot be enforced by the Courts.
The sugya begins with a Baraita which reads as follows:' R. Joshua
said: There are four things which, if a man does them, he is exempt
from the laws of man but is liable by the laws of Heaven and these
are they: He who breaks down a fence before his neighbour's
animal (and the animal escapes and is lost); he who bends his
neighbour's corn into the path of a fire; he who hires false witnesses
to testify; and he who can testify on behalf of his neighbour but
fails so to do.' In each of these cases the Courts cannot enforce
payment, since the loss is caused indirectly and the law only
demands compensation where there is a direct act of injury or
damage to property. Thus where a man breaks down a fence and
the animal escapes and is lost his act is indirect, but since otherwise
the neighbour's animal could not have escaped he is liable by the
laws of Heaven. Similarly where a man sees a fire spreading but
in such a way that it will pass by his neighbour's corn and he bends
the corn so that it will eventually catch fire, this, too, is indirect
since at the time he bends the corn the fire has not yet reached that
spot. In the third case, the man causes his neighbour a loss by hiring
false witnesses to testify that his neighbour is liable when, in fact,
he is not. Here, too, the loss is caused by the witnesses who are his
192 The Talmudic argument
indirect agents. Finally, in the fourth case, he does cause his
neighbour a loss by failing to testify on his behalf, but it is an
indirect cause.
See Rashi who refers to gerama bi-nezakin, 'causing damage', i.e. 'indirect
damage'. On this subject see Kesef Nivhar, part 1, no. 43, pp. 75b~77a and
ET, vol. vi, pp. 461-97. On the conceptpatur mi-diney adam ve-hayyav be-diney
shamayim see ET, vol. vn, pp. 382-96. The Baraita quoted in our sugya is
Tosefta, Shevu'ot 3: 1-3 but there the order is different and, instead oipatur
mi-diney adam ve-hayyav be-diney shamayim, reads: ' Four are not liable to pay
by law but Heaven does not forgive them until they pay.' It would seem
that our sugya has paraphrased the Tosefta using a form that fits in with
the needs of our sugya's argument, e.g. the cause of the animal is mentioned
first (unlike in the Tosefta, which mentions the case of testimony first
because this is the Shevu'ot context) because this is the previous theme
treated in this section of Bava Kama
The Talmud now examines each one of these instances in turn. First
the Talmud examines the case of the man who breaks down the
fence in front of his neighbour's animal. The Talmud observes that
this cannot refer to a sound fence, i.e. to one that otherwise would
not have fallen in, since then the man would be liable even by the
laws of man, at least for his destruction of the fence, even though
here, too, he would be exempt for the loss of the animal since this
is still indirect causation. The reply is that the Baraita deals with
a shaky fence that would, in any event, have fallen in.
This follows Rashi and Tosafists but see Meiri and ha-Gra who show that
other commentators understand the question to be, if the fence is sound
the man ought to be liable by the laws of man even for the loss of the animal,
i.e. because to break down a sound fence so that the animal will escape
is considered to be direct, not indirect, damage.
The Talmud now examines the second instance in the Baraita,
that of the man who bends his neighbour's corn in the path of a
fire. The Talmud asks, why should the man not be liable to pay
by the laws of man if the fire will reach the corn even if fanned by
a normal wind? The significance of the question is that according
to the law (Exodus 22: 6; The Torah: Exodus 22: 5) as interpreted
by the Rabbis, if a man lights a fire in his own field and the fire,
fanned by a normal wind, spreads to his neighbour's field and does
damage there, the man who lit the fire is liable to pay for the
damage. Now if the fire, in our case, can reach the corn when
fanned by a normal wind, the man who bent the corn so that the
fire will now reach the corn ought to be liable by the laws of man.
Liability to God even where no human enforcement 193
This is, in fact, the case mentioned in Scripture of esh, 'fire', since
there is surely no difference between putting the fire in the way of
the corn (the Scriptural case) and putting the corn in the way of
the fire (our case, see Tosafists). The reply is given that the Baraita
speaks of fire that cannot travel by a normal wind, only by an
abnormal wind. Consequently, there is no liability by the laws of
man but there is a liability by the laws of Heaven. R. Ashi gives
a different answer, interpreting the Baraita's reference to ' bending'
the corn in a novel way. According to R. Ashi the Baraita does not
mean that the man simply ' bent' the corn so that it came into the
path of the fire for then, indeed, he would be liable by the laws of
man, as above. ' Bending' the corn means that he covered the corn,
already in the path of a spreading fire, with a sheet, thus ' bending'
it under the weight of the sheet. Now the law is (Mishnah, Bava
Kama 6: 5) that a man who lights a fire, as above, which spreads
to do damage is only liable if the corn and so forth that is burnt
is uncovered. If it is 'covered', tamun, ' hidden' , there is no liability.
Consequently, if A kindled a fire that spread when fanned by a
normal wind he is liable for any damage it causes to B's corn. But
if, before the fire reached B's corn, Ccovered the corn with a sheet,
A is exempt from paying, since the corn is now tamun. Thus C's act
has the effect of causing B to suffer loss, since B cannot now claim
damages from A. C's offence is that he introduced a technicality
by means of which B loses his case. For this there is no liability by
the laws of man, since, after all, it was the fire kindled by A that
did the damage, but there is liability by the laws of Heaven and,
according to R. Ashi, it is to this that our Baraita refers.
The third example given in the Baraita is now examined, that
of the man who hires false witnesses. The Talmud objects that if
he hired these false witnesses to obtain money for himself then, when
they are found to be false, he must return the money he has
obtained through false pretences. Consequently, here he is liable
according to the laws of man. The reply is that he hires the witnesses
to testify falsely on behalf of another (i.e. who has died in the
meantime or has vanished, Tosafists). By hiring the false witnesses
he has caused his neighbour to lose money yet, since this is indirect,
there is no liability according to the laws of man but there is by
the laws of Heaven.
Finally, the fourth case in the Baraita is examined, that of the
man who can testify on behalf of his neighbour but fails so to do.
194 The Talmudic argument
Now according to the Rabbinic interpretation of: 'if he do not utter
it, then he shall bear his iniquity' (Leviticus 5:1), if two witnesses
refuse to testify they commit a sin, since in monetary matters the
evidence of two witnesses is required. But if one witness refuses to
testify, this Biblical sin is not incurred since the testimony of a single
witness can only succeed in imposing an oath on the person against
whom the testimony is given. The Talmud asks, when the Baraita
speaks of a man failing to testify does it refer, in fact, to two
witnesses, each of whom refuses to testify? If it does then ' It is
obvious' that each is liable by the laws of Heaven (Rashi) since this
is stated explicitly in the Scriptural verse: 'if he do not utter it, then
he shall bear his iniquity'. The reply is that the Baraita speaks of
a single witness. A has a claim on B and asks Cto testify on his
behalf, which Crefuses to do. Now even if Cdoes testify all his
testimony can succeed in doing is to compel B to take an oath and
it is possible that B will then swear falsely. On the other hand, B
may not wish to swear falsely so that C, by refusing to testify on
A's behalf, may be causing A a loss. Therefore, there is no liability
according to the laws of man (there is no liability according to the
laws of man even if two witnesses refuse to testify, as above) but
there is liability according to the laws of Heaven. This is not,
however obvious and requires to be stated, since this case, of a single
witness, is not covered by the Biblical text.
The Talmud now raises a new difficulty. R. Joshua in the Baraita
refers to only four cases which are 'exempt by the laws of man and
liable by the laws of Heaven'. But, the Talmud proceeds to ask,
there are, in fact, other examples of this rule. Why, then, does
R.Joshua only refer to these four? The Talmud proceeds to list
these further examples of cases ' exempt by the laws of man and
liable by the laws of Heaven'. All these are from Tannaitic sources,
Baraitot and in one case a Mishnah.
This is the list as given in the Talmud.
(a) If a man does work with the red heifer or the purifying water
(Numbers 19) and by so doing renders them invalid and thus causes
their owner a loss, he is exempt by the laws of man but liable by
the laws of Heaven (see supra, p. 148).
(b) If a man places poison in front of his neighbour's animal and the
animal eats it and dies, he is exempt from the laws of man but liable
by the laws of Heaven. This is indirect damage, since the animal
may not have eaten the poison but there is, none the less, complete
liability by the laws of Heaven.
(c) If a man gives a flaming torch to an imbecile or a minor and they
Liability to God even where no human enforcement i95
take it and set a haystack alight the man who gave them the torch
is exempt by the laws of man (because it is indirect) but liable by
the laws of Heaven (Mishnah, Bava Kama 6:4).
(d) If a man gives his neighbour a sudden fright, i.e. by shouting into
his ear and the like, he is exempt by the laws of man, since this,
too, is considered to be a case of indirect injury, but he is liable
by the laws of Heaven.
(e) A man's jar broke in the public domain or his camel fell down there
and he failed to remove them and they did damage. According to
R. Meir he is liable even by the laws of man, but the Sages hold
that he is exempt by the laws of man and liable by the laws of
Heaven.
Thus there are five further examples o^patur mi-diney adam* etc.,
so why does R. Joshua in our Baraita only record four and omits
these? The reply is that there are, indeed, more examples and
R. Joshua is not bent on listing all possible examples. What
R. Joshua means by his examples is that even with regard to these
four there is a liability by the laws of Heaven. R. Joshua has to state
this because there are sound arguments in all these four cases for
holding that in these there is no liability at all, not even by the laws
of Heaven. The Talmud then proceeds, case by case, to show what
these arguments are.
It hardly needs saying that, historically speaking, the problem raised by
the Talmud here is artificial. R. Joshua may have had his reasons for only
referring to these four. He may, indeed, have known only these four. One
of them is, in fact, debated by R. Meir and the Sages and, for all we know
to the contrary, R. Joshua may have held the same view as that held later
by R. Meir. The truth is that we have here an example of purely artificial
sugya in which the question, 'why these four only', is raised for the sole
purpose of developing the argument that in these instances a good case
can be made out for exempting the man even by the laws of Heaven. See
SM, who raises the question about R. Meir and suggests that the 'Sages'
who disagree with R. Meir are to be identified with R. Judah, the usual
anonymous protagonist of R. Meir, and that the reading in our Baraita is
not R. Joshua but R. Judah. This is forced and, in any event, in the Tosefta
texts the reading is R. Joshua.
As stated, the Talmud now proceeds to show why it could have
been argued in these four cases that the man is not even liable by
the laws of Heaven. In the case of breaking down the fence, since
we have established that it must be a case of a shaky fence, it could
have been argued that there is no liability at all, not even by the
laws of Heaven, since the fence is due to be demolished. In the
second case, that of the man who bends his neighbour's corn in the
196 The Talmudic argument
path of the fire, since we have established, according to the first
reply above, that the fire only travels by means of an abnormal
wind, the man might have argued: ' How could I have anticipated
that an abnormal wind would bring the fire to the corn?' And
according to R. Ashi's reply, that the man 'covered' the corn, it
might have been argued that his intention was not malicious in
spreading the sheet but rather was to prevent the fire getting at the
corn. Though, in fact, by covering the corn, he rendered it tamun, we
might have argued that he is not liable even by the laws of Heaven
(see Tosafists). In the case of a man who hires false witnesses, we
might have argued that the hirer can say to the witnesses: ' Where
there are the instructions of the master and the (opposite) instruc-
tions of the disciple, whose instructions should one obey?', i.e. God
instructs witnesses not to testify falsely and they should not have
heeded the opposite instructions to give false witness. If they did,
it is the witnesses who should bear the full load of the guilt and the
hirer, we might have said, is not even liable by the laws of Heaven.
Finally, in the case of the witness who refuses to testify, since we
have established that it is a case of a single witness, then there is
no guarantee that his testimony would have had any effect, since,
even if he did testify, the man against whom he testified might have
sworn falsely and have been acquitted. Hence we could well have
argued that through his refusal to testify he does not incur liability
even by the laws of Heaven. Thus in all four cases there is a strong
argument for exempting the man even by the laws of Heaven,
which is why R. Joshua sees fit to mention these four cases, to inform
us that he is liable by the laws of Heaven.
The plan of the sugya is:
(1) Joshua's statement:
(a) breaking the fence
(b) bending the corn
(c) hiring false witnesses
(d) refusing to testify
(2) In (a) let him be liable for the fence by laws of man?
No, (a) refers to shaky fence
(3) In (b) let him be liable as 'fire' by laws of man?
No, abnormal wind or (R. Ashi) 'bending' means tamun
(4) In (c) let him be liable for the money he has gained illegally?
No, he hires the witnesses not for himself but for another
(5) In (d) if there are two witnesses it is obvious since Biblical?
No, it speaks of a single witness
Liability to God even where no human enforcement 197
(6) Why does R.Joshua only list these four; there are also:
(i) red heifer
(ii) poison
(iii) torch sent by imbecile
(iv) giving another a fright
(v) the jar or camel in the public domain
(7) These four are necessary because we might have thought no
liability here at all, not even by laws of Heaven because:
(a) since the wall is weak, as above
(b) since the wind is abnormal, as above or (R. Ashi) since
he could have covered it with good intentions
(c) since it is not for himself and he does not have the money
the full guilt should be borne by the witnesses
(d) since a single witness, as above
As soon as one studies the construction of this sugya one sees
immediately that the replies given in (7) to the question in (6) only
make sense after R. Joshua's four cases have been qualified in the
way they have been in (2) to (5). In other words (2) to (5) pave
the way for the solutions in (7) and this can only mean that the
whole sugya has been constructed as a unit with a degree of
artificiality but one that gives power to the sequence of argument.
This is an excellent illustration of the Talmudic sugya as a literary
unit, contrived, but all the more effective for it. All this must be
the work of editors subsequent to R. Ashi since R. Ashi's inter-
pretation and the alternative one in (3) have been utilised as bricks
in the complete edifice. The mnemonic referring to the list in (6),
given in brackets in 56a, is probably post-Talmudic but this list may
originally have been independent of our sugya. It is perhaps not
without significance that all five items in the list are quoted
elsewhere in Bava Kama and it is just possible that they may have
formed originally a 'Bava Kama'' collection.
21
Mahal 'al kevodo mahul: renunciation of honour
by one to whom it is due
The sugya is Kiddushin 32a-b. It deals with the question, how far
does the honour and respect due to certain persons extend?
According to Rabbinic teaching there are four persons it is a
religious obligation to honour: a father (and mother); a teacher
(rav)\ the Prince (nasi), the leader of Palestinian Jewry, mainly of
the House of Rabban Gamaliel; and the king (melekh). Now
although to pay these honour is a religious injunction yet the
honour is paid to another human being. Consequently, it can be
argued that if that person is prepared to renounce the honour due
to him, there is no longer any obligation to pay him honour, or,
it can be argued, it does not lie in his power to renounce the honour
since, after all, it is a religious obligation. Renouncing the honour
is referred to as: mahal
l
al kevodo: ' he renounces his honour'. If he
can renounce it, then we say: kevodo mahul,' his honour is renounced'.
If he cannot, then we say: eyn kevodo mahul, ' his honour is not
renounced'. The sugya is based on a ruling by the third-century
Babylonian Amora, R. Hisda. With regard to the father and the
teacher the question was of practical significance in R. Hisda's day.
With regard to the Prince and the king the whole question was
purely academic. The sugya has been inserted here because this
section of tractate Kiddushin considers the rules of honouring
parents.
The sugya opens with a saying of R. Isaac bar Shila in the name
of R. Mattena in the name of R. Hisda: ' If a father renounces his
honour his honour is renounced but if a teacher renounces his
honour his honour is not renounced.' The reason for the distinction
is because the honour due to the teacher is really not because of
his own person at all but because of the Torah he teaches and it
198
Renunciation of honour by one to whom it is due 199
does not lie within his power to renounce the honour due to the
Torah. But R. Joseph disagrees with R. Hisda and holds that even
a teacher can renounce the honour due to him. R. Joseph quotes
in support: 'And the Lord went before them' (Exodus 13: 21), i.e.
God is Israel's Teacher and yet He renounced His honour by
leading them and, as it were, serving them. To this Rava,
R. Joseph's pupil, objects that there is no proof whatsoever from
this verse. The whole world belongs to God and the Torah belongs
to God. It is all for His glory and He can renounce His due. But the
Torah is not the teacher's, so that he should be able to renounce
it. Yet Rava, it is said, later changed his mind, arguing that a
teacher can renounce the honour due to him because the ' Torah
is his.' Rava quotes the verse: ' And in his Torah doth he meditate
day and night' (Psalm 1:2).
Rava's interpretation of the text is better understood by the parallel
reference in 'Avodah ^arah 19a where Rava says that at first it is the Torah
of the Lord and then (i.e. after a man has mastered his studies) it becomes
'his Torah'. Rava interprets the verse: 'But his delight is in the Torah of
the Lord; And in his Torah doth he meditate day and night', i.e. the
personal pronoun refers, not to God as in the plain meaning, but to the
student. The Torah becomes ' his Torah' and the teacher can, therefore,
renounce the honour as due to him.
We have thus established that, eventually, Rava came to hold
that a teacher can renounce the honour due to him. But, the
Talmud objects, it once happened that at the wedding feast of
Rava's son, Rava handed a cup of wine to R. Pappa and R. Huna
son of R. Joshua and they rose before him (to pay him respect) but
when he handed the cup to R. Mari and R. Phinehas son of
R. Hisda they did not rise before him. Rava took offence and
declared: 'Are these Rabbis Rabbis and those Rabbis not
Rabbis?', i.e. R. Pappa and R. Huna son of R.Joshua are also
scholars and yet they paid Rava honour by rising out of respect;
why, then, did R. Mari and R. Phinehas remain seated? Now this
shows that Rava objected to the failure of the two Rabbis to pay
him the respect due to him as their teacher even though, by
handing them the cup of wine in the first place, he had demonstrated
that he had renounced the honour due to him and, according to
Rava's own eventual decision, if a teacher renounces his honour,
his honour is renounced. Furthermore, it once happened that
R. Pappa, Rava's pupil, who, presumably, agrees with Rava's
200 The Talmudic argument
decision, was serving drinks at the wedding feast of his son, Abba
Mar, and he handed a cup of wine to R. Isaac son of R. Judah and
when the latter did not rise before him he took offence. All this
shows that even when a teacher renounces his honour his honour
is not renounced. The reply is:' Nevertheless, they should have paid
him some token of honour', i.e. even though, it is true, a teacher
can renounce his honour, the disciple, while accepting this, should
pay his teacher some small token of respect, such as rising before
him.
There is now reported the saying of the fifth-century Babylonian
Amora, R. Ashi. R. Ashi said: ' Even according to the authority
who holds that if a teacher renounces his honour, his honour is
renounced, if a Prince renounces his honour his honour is not
renounced', i.e., because it is not his to renounce, unlike the teacher
whose Torah, as Rava has said, is his own. An objection is raised
to R. Ashi's statement. A Baraita is quoted where it is stated that
R. Eliezer and R. Joshua and R. Zadok were reclining at the
wedding feast of Rabban Gamaliel's son and Rabban Gamaliel
was handing round the drinks. When he handed the cup to
R. Eliezer he refused to accept it but when he handed the cup to
R. Joshua he did accept it. R. Eliezer said:' What is this, O Joshua!
Shall we recline and allow Rabban Gamaliel to serve us with
drink?' R.Joshua replied: ' We find that a greater one than he
served others. Abraham was the greatest one of his generation (i.e.
he was the equivalent in his day of the nasi, as Rabban Gamaliel
was) and yet of him it is written: "and he stood by them" (Genesis
18: 8). And perchance you might say that they appeared to him
as angels. They only appeared to him as Arabs. So, as for us, shall
not Rabban Gamaliel stand by us to serve us drinks?' R. Zadok
thereupon said to them: 'For how long will you leave aside the
honour of God and discuss the honour of His creatures? The Holy
One, blessed be He, causes the winds to blow. He makes the
vapours (nesiim) ascend. He brings down dew. He causes the soil
to yield its produce. And He prepares a table before every one (on
earth). And as for us, shall not Rabban Gamaliel stand by us and
serve us drinks?.' Thus we see that the Sages argued that Rabban
Gamaliel could renounce his honour and this refutes R. Ashi's
contention that when a nasi renounces his honour is not renounced.
The reply is given: 'But if it was stated thus was it stated', i.e.
R. Ashi did not say what he is reported to have said. The true
version of R. Ashi's saying is:' Even according to the authority who
Renunciation of honour by one to whom it is due 201
holds that a Prince who renounces his honour, his honour is
renounced [presumably referring to R. J os hua and R. Zadok], a
king who renounces his honour, his honour is not renounced. As
it is said: "Thou shalt in any wise set him king over t hee"
(Deuteronomy 17: 15).' His fear must be ' over t hee' , i.e. and,
therefore, a special injunction exists to the effect that a king cannot
renounce his honour.
It should be noted that: 'As it is said: Thou shalt in any wise set him king
over thee' is a quote from the Mishnah {Sanhedrin 2:5), cf. Abraham Weiss,
le-Heker ha-Talmud, p. 58. In our sugya, at first, R. Ashi refers to the nasi
and it is only after the objection from the Rabban Gamaliel narrative that
R. Ashi's statement is emended to read: ' Even according to the authority
who holds that a Prince who renounces his honour, his honour is
renounced, a king who renounces his honour, his honour is not renounced.
As it is said: thou shalt in any wise set him king over thee. His fear must be upon
thee'. Thus when in Ketubot 17a, Sotah 41b and Sanhedrin 19b the Talmud
quotes this version in the words: 'As R. Ashi said' it presumably means that
in those instances we have a quote from our sugyal In other words after
our sugya had been completed there are cross-references to it in the other
tractates. It is also just possible, however, that the cross-references are not
to our sugya but to the original and correct version of R. Ashi. But if this
is so then it means that in our sugya, at first, an incorrect version was
recorded in order for the sugya to lead on from father to teacher to nasi
and eventually to king! It is perhaps significant that the word for ' vapours'
(nesiim) is also the plural of' prince' so, possibly, there is a pun here: 'God
causes the princes to rise to greatness', i.e. if He can renounce His honour,
how much more so the nasi to whom He has given greatness!
The pattern of the sugya is:
(1) R. Hisda:/fl^r can renounce his honour; teacher cannot renounce
his honour
(2) R. Joseph: even teacher can renounce his honour: And the Lord went
before them
(3) Rava objects: is the Tor ah his?
(4) Rava's reply: yes, the Torah is his and he can renounce his honour
(5) Objection: but in that case why did Rava take offence and why
did R. Pappa take offence?
(6) Reply: because some small token of respect should have been paid
(7) R. Ashi: even if teacher can renounce his honour, nasi cannot
renounce his honour
(8) Objection: narrative of Rabban Gamaliel and the Sages
(9) True version of R. Ashi: even if nasi can renounce his honour, king
cannot renounce his honour
It is clear that the sugya has been so constructed that it leads from
the lesser to the greater - from father through teacher and nasi to king.
202 The Talmudic argument
For this purpose, as above, the Talmud first quotes the incorrect
version of R. Ashi, which it then refutes in order to lead to the
climax: that even if in the other three cases there can be renunciation
there cannot be in the case of the king. Since the whole sugya
depends on the two versions of R. Ashi, it must have been edited
some considerable time after R. Ashi's death.
On the face of it there appears to be a contradiction in the sugya. Thus
at the beginning Rava refutes the argument, from the fact that God led
the people, that a teacher can renounce his honour and yet in the narrative
later R. Zadok proves his case from the fact that God serves 'everyone'.
But the distinction is an obvious one. The proof from God leading them
was, if God as Teacher can renounce His honour, how much more so a
human teacher? To this Rava retorts that God can renounce His own
honour but the teacher cannot renounce God's honour. But in R. Zadok's
application it is not to God's honour that he refers but to the honour of
the nasi. To this it is perfectly reasonable to argue that if God can renounce
His honour how much more so can a human Prince renounce his honour?
22
Conclusions
We have examined a number of sugyot in some detail, each forming
a unit complete in itself. These have been chosen, however, as
altogether typical of the Babylonian Talmud as a whole so that it
is possible to draw from them conclusions regarding the literary
constructions and methods of argumentation of this apparently
unsystematic work.
With regard to literary style, it must first be noted how futile it
is to expect the kind of elegance usually associated with a work of
literature (not, at least, in the Halakhic portions constituting the
major part of the Talmud). The vocabulary is severely limited,
consisting, in the main, of technical terms; the presentation of the
material is cryptic, with hardly any spelling out of the meaning;
there is no conscious striving for rhyme, rhythm or alliteration; and
wherever punctuation is required it has to be supplied vocally by
teacher and student, who 'sing' rather than ' read' the Talmud.
Yet the Talmud does have a literary style of its own, provided
by the ordering of the material in a dramatic way. It is a style not
of words but of ideas. The arguments and debates are so arranged
that there is a building up of the discussion step by step until the
climax is attained. Information is withheld until it can be given at
the appropriate stage for drawing the threads of the argument
together, as if the editors are saying: We have been leading you,
the student, into tortuous paths and you have gone astray. Now
we will show you the true path and all will become clear. The
student is led to side first with one protagonist then with another
until all the pros and cons have been exhausted. Even the
digressions from the main argument have been planned so as to give
the impression of spontaneity, much as a dramatist will use
203
204 The Talmudic argument
interruptions of the main plot in order to prevent the stereotyping
of his play and its characters.
Repeatedly throughout the Talmud this element of contrivance
can easily be discerned once more than a surface reading is
attempted. Take the sugya on the Scriptural derivation of the
probability principle (pp. 50-63). Ten Amoraim, over a two-
hundred-year period, are recorded as endeavouring to prove, each
in his own way, that Scripture permits reliance on probability. But,
as we have seen, these proofs are not arranged in any chronological
order but in a sequence in which the objection which can be ad-
vanced against the first proof cannot be advanced against the
second; the further objection to this cannot be levelled against the
third, and so on. The words used for ' R. Y. said' are not in Hebrew
but in Aramaic, the language used by those responsible for
providing the framework of the sugya. We do not have, in other
words, the actual formulations of the original teachers who ad-
vanced the ten attempted proofs but a paraphrase by the editors.
It is as clear as can be that the final editors have collected the ten
proofs, originally independent of one another, and then arranged
these to provide a dramatic sequence. In a word, this sugya and all
the others are structured. (We are still in the dark as to the actual
process through which the material from different ages and places
came into the hands of the editors. Did the various generations act-
ually transmit the sayings or is there an element of artificiality in
it all?) It is not beyond the bounds of possibility that ten proofs,
neither more nor less, are given because ten is a 'sacred number',
as in the Decalogue and other Jewish institutions! In any event the
sugya is far more than a mere recording of differing opinions. It is
a literary unit in its own right.
Or take the sugya on the inability of mental reservations to
invalidate a transaction (pp. 101-9). The sugya first gives an
illustration of case law and then proceeds to a suggestion of what
might have been in the mind of the teacher when he rendered his
decision. The editors do not declare that they know the arguments
by which Rava arrived at his decision that mental reservations do
not count. The question,' How does Rava know this?', implies that
the editors have no tradition regarding the reasoning for Rava's
decision. All they had was the decision itself. Moreover, the
suggested reasoning is conveyed in the form of proof and refutation
and then further proof typical of every Talmudic sugya. The first
Conclusions 205
attempted proof is from the case of the man who has to be coerced
into bringing the offering he had promised to bring. This shows that
mental reservations are irrelevant, since otherwise there is a mental
reservation even though, under coercion, the man states his
willingness to bring the sacrifice. No, the argument continues, in
that case the man may actually give his unreserved consent, albeit
through coercion, because he really does wish to have the atonement
that the bringing of the sacrifice will afford him. The second
attempted proof is from the case of the man who is compelled to
give a bill of divorce to his wife. Here the idea of consent without
reservations does not apply, since here no atonement through the
bringing of a sacrifice is involved. No, but here, too, there are no
mental reservations since the man does wish to obey the law and
the law demands that his wife should obtain her release. The third
proof is from the Mishnah to which the sugya is appended. Here
mental reservations are undoubtedly present, since there is no ques-
tion of atonement or of obedience to the law and yet the reservations
are ignored, which proves Rava's case.
A less contrived presentation - but that would not be the way
of Talmud would have been simply to make the suggestion at
once that Rava derives his ruling from the Mishnah, the third
attempted proof being quoted directly, the others simply ignored,
since they are not conclusive in any event. But the Talmud is not
interested solely in how the teacher arrives at his decision. The
skilful working up of the material is the essence of the Talmudic
presentation. What we have in this, and in all the other sugyot we
have examined, is the telling of a story, as it were, in which, as in
any good story, the denouement comes in its right place. It is only
in the final pages of the detective story that all the false clues are
shown to be such and we learn t hat ' the butler did it' . In fact, the
sugya does not allow even this third proof to stand. The clinching
argument is from tractate Me'ilah and, in fact, as we have noted,
in the Talmud to Me'ilah there is what appears to be a cross-reference
to our sugya. Our sugya is appended to the third proof, which has
taken us at least three-fourths of the road we have to traverse in
order to obtain complete certainty.
Occasionally, as in the sugya on effects taking place simultaneously
(pp. 75-82), the editors state explicitly that the argument is a step-
by-step one. Here the steps in the argument from things that can
be done by halves, to things that can be done in error, to things
206 The Talmudic argument
that can be done neither by halves nor in error, is all stated by the
editors as part of their total construction. In the far more usual
scheme, however, it is left to the reader to supply the step-by-step
element, which he can easily do once he has grasped the nature of
the Talmudic style.
To turn now to the conclusions to be reached regarding the
Talmudic form of argumentation as a result of our investigation,
we have noted in the first chapter the various moves open to the
protagonists. In each of the sugyot described these moves are
employed as the need arises, contributing to the liveliness of the
debates. The student comes to feel that the Babylonian Amoraim
are there with him in his room. He hears their voices, at times raised
in vehement protest at a faulty conclusion, at other times haltingly
exploring the admissibility of an alternative view to their own, and,
again, at times giving assent, grudgingly or with brave acceptance,
to the refutation of their deeply-held propositions. With their skills
in presentation, the editors have brought about an uncanny sense
of student participation in the debates which took place so long ago.
It is no accident that the subsequent debates among the
commentators on what the Talmud means in this or that passage
are conducted in the Talmudic way as if the Talmud did not receive
its final editing at all, but continued down the ages in the Yeshivot.
Yet for all the formal structuralism of the Talmudic sugya, the
reasoning processes themselves are conveyed in no more than a few
words. The student is expected to elaborate on these in order that
their full meaning may be grasped. There is generally only a
skeleton, to be clothed with flesh by acute, penetrating commentary.
The danger is ever present, of course, of reading into these cryptic
texts meanings their authors never intended. It is really a matter
of becoming attuned to the mental processes of the Amoraim, of
putting oneself into their universe of discourse, of acquiring the
ability to anticipate their views. In the traditional Yeshivot this
capacity to think the thoughts of the Amoraim after them, to delve
into the profundities beneath the surface texts of the Talmud (the
Gemara, as the Talmud came to be called) is known as 'having
a Gemara head' or having a head for Gemara. It is partly art as
well as science. We repeat that this whole area of Talmudic
thinking has been largely ignored by modern scholarship, just as
the Talmudists of the old school have been indifferent, on the
whole, to the kind of question, raised by the moderns, of authorship,
Conclusions 207
historical background, form and textual criticism. In this book we
have tried to combine the two approaches, considering the literary
problems without losing sight of that which gives the Talmud its
force and appeal, the meaty argument in itself.
Let us examine some of the illustrations of legal maxims quoted
in the sugyot to see how the bare statements require to be filled out
if their meaning is to be grasped. In the sugya on probability (pp.
50-63) two different types are noted: (1) 'where the majority is
before us' ; (2) 'where the majority is not before us' . The rule that
the majority be followed in a Court decision or that, where nine
shops sell kosher meat and one forbidden meat, any meat found is
assumed to come from the nine shops and is kasher, is called a
'majority that is before us'. Where the meat of an animal whose
internal organs have not been examined is allowed to be eaten, it
is said to be on the basis of a 'majority that is not before us' . That
the Torah relies on a 'majority before us' does not entitle us to
conclude that it relies on a 'majority that is not before us' . On a
superficial reading the distinction is a purely formal one. In the one
case the 'majority' is actually present - the members of the Court
are here and the nine shops are here - whereas the majority of
healthy animals is not actually present when the meat of this
particular animal is being considered. But the Talmud gives no hint
of the logic behind the distinction. Why should it matter whether
the majority is present or absent? It all becomes abundantly clear
once it is appreciated that in the words they use the editors are
thinking of two kinds of probability, as we have noted. The
'majority that is before us' is a purely mathematical probability.
If the Torah relies on it that can only be a procedural matter. That
nine shops sell kasher meat does not mean that the Torah informs
us that any meat is actually kasher, only that, even if it is not, it
may be eaten by reliance, as a procedure, on the probability
principle. In the other case of 'majority not before us' the
probability that the animal is kasher is based on empirical invest-
igation. The majority of animals examined are found to be healthy
and not to suffer from a disease that renders their meat forbidden
so that in any individual case it can be assumed that the animal
is healthy and kasher. Here, where the 'majority is not before us' ,
if the Torah does permit us to rely on probability it is a different
type of probability on which we are to rely, that the animal is not
only procedurally kasher but actually kasher. In the one case the
208 The Talmudic argument
probability principle is only applied as a matter of law. In the other
case it is sought to apply it as a matter of fact. Obviously, the second
type cannot be derived from the first. The reasoning is clear, but
only after the precise but elusive formulations of the Amoraim or
the editors have been translated into the language of abstract
thought.
That mental reservations cannot be allowed to invalidate a
contract (pp. 101-9) is formulated as: 'words in the heart are not
words'. The word for heart (lev) denotes in Hebrew the whole inner
being of man. The Amoraic formulation means that while the
reservation has remained unexpressed in speech it is too intangible
to revoke a verbal or written contract. Only verbal explication is
sufficiently forceful, psychologically speaking, to denote real resolve.
Otherwise, the mental reservation is too inchoate, too vague and
uncertain, for it to be taken into consideration by the law, since
the person who has this unexpressed reservation is not sufficiently
definite about it when it contradicts his actual words or those of
his bond. Similarly, when the Talmud wishes to suggest that a legal
procedure is no test of veracity (so that, as on pp. 168-73,
t n e
testimony of a witness against himself is rejected by the law while
the same testimony is accepted in relation to others), the way the
Talmud expresses this is simply by saying 'we divide his words'.
Again, when the Talmud discusses whether or not mere designation
for a purpose is considered in law as if that purpose has been
achieved, the way this is put is whether designation is or is not ' a
thing' (pp. 122-32). That is to say, is the act of designation
sufficiently substantial to count in law or is it too insubstantial; is
it a ' thing' or is it not a ' thing' ? This is a general principle, which
is why the Talmud can consider in the same sugya such apparently
unconnected items as designation of a cloth for a shroud and the
designation of hide for tefillin. The question involved does not have
to do specifically with the law of a corpse or of tefillin but with the
general principle of whether mere designation counts. This
Talmudic method of comparing diverse cases because they conform
to a general principle can be observed in all the passages examined
in this book.
In the sugya on the conveyancing of a thing not yet in existence
(pp. 64-74), the term used for what we would call unrealised assets
is ' a thing not yet in the world'. A distinction is made in the sugya
between two types of unrealised assets. Where these, although not
Conclusions 209
yet realised, can be realised (literally, 'it is in his hand' ) by the
person seeking to transfer them, the transfer can more readily be
accepted in law than where he has no power to realise them
(literally 'they are not in his hand' ). No reason for the distinction
is given by the Talmud. If unrealised assets cannot be transferred
in law, why should his ability to realise them make any difference?
For that matter, the Talmud does not tell us why it is held that
realised assets cannot be transferred. The later commentators
explain it all in this way and what they have to say is very
convincing. The reason why a transfer of unrealised assets is
doubtful is because, while a man may have full resolve to sell or
otherwise transfer to another something he has, his resolve to sell
assets as yet unrealised is too weak for him to have the requisite
degree of assent the law requires for a valid transaction. But where
he himself has the power now to realise those assets his resolve is
sufficiently strong (since it depends solely on him) to make his
decision to sell or otherwise transfer them capable of being
recognised by the law.
A similar type of reasoning would seem to be implied in the sugya
on retrospective specification (pp. 2433). Where a man instructs
the scribe to write out a bill of divorce for the wife who will first
emerge through the door, the bill of divorce (the get) is valid, it
being considered as if the later specification was present at the time
of writing. But, it is suggested, there may be no retrospective
specification if the man's instruction to the scribe was to write the
get for whichever wife the man will later decide to divorce. The
wording of the distinction is, in the first case the later specification
'depends on others' (i.e. on the wife to emerge first through the
door), while in the second case it 'depends on him' (i.e. on his later
decision to divorce wife A rather than B). No reason is given for
this bald distinction. In both cases there is lack of specification at
the time the instructions were given and the get written by the
scribe. If in the one case retrospective specification operates why
should it not operate in the other? Rashi follows the standard
method of amplification in order to make the distinction clear.
Where the later specification depends on others, the husband has
done all the specification he wishes to do. He has decided to leave
it to others and this is in itself a sufficiently strong form of
specification. But where the lack of specification is due to the
husband's own indecision, there is so little specification at the time
21 o The Talmudic argument
when the instructions were given that no amount of later
specification by him can be allowed to operate retrospectively.
It can be seen from all the sugyot examined in this book that the
task the Amoraim set themselves was to give abstract expression
(although in very concrete formulation requiring, as we have just
noted, further elaboration in abstract terms) to the individual
formulations and cases of the Tannaim. The earlier, Tannaitic
method of discourse is to consider particular cases. The Amoraim
examined these particular cases in order to see whether or not a
general, abstract principle of the law can be discerned. Thus the
Tannaim speak of a man divorcing the wife who will first emerge
through the door and of a man who wishes to drink out of a jar
of wine from which he later gives the tithes. The Amoraim subsume
these otherwise distinct cases under the heading of retrospective
specification (pp. 24-33).
A feature of the Amoraic literature, insufficiently noted by many
scholars but which can be seen from our investigation, is the purely
academic and theoretical nature of a good deal of the material
discussed. It is a moot point, for example, whether polygamy was
widely practised during the Amoraic period but, even if it was, it
is inconceivable that the case of a man instructing the scribe to write
out a bill of divorce for whichever of his wives, both with the same
name, emerges first through the door can ever have come before the
Courts to be decided as a matter of practical law. Reliance on
probability is everywhere accepted, but the whole question of how
this principle is derived from Scripture has no practical conse-
quences. There was no capital punishment in the Amoraic period so
that all the passages dealing with this subject are about pure legal
theory, as are all the debates and discussions on the sacrificial
system, abolished long before the Amoraic period. Other than the
love of learning for its own sake, what practical advantage could
have resulted from the discussion of the man who betroths one of
two sisters without specifying which of them he betroths? Even
when the discussion centres around an actual case, as in the
instance of mental reservations, the discussion itself is for its own
sake, with no practical consequences. Unless this element of study
of the Torah for its own sake, of occupying the mind with the word
of God in much the same way as the pure scientist may be
indifferent to the practical use to which his work may be put, is
recognised, there is total failure to understand the Talmud. The
Conclusions 211
Talmud is not a Code of Law, although it became the basis and
the final court of appeal for all the later Codes. The Rabbis were,
of course, observant Jews and the Talmud does contain much
practical law. But the main thrust both of the Amoraim and of the
final editors of the Talmud is in the direction of pure legal
theory - naturally, the legal theory as contained in the divine word
they sought mightily to explicate.
Finally, it must be noted that all the passages considered in this
book belong to the Halakhah, the legal element in the Talmud,
which is after all the largest component in this gigantic work.
Considerations of space dictated that the Aggadic element - the
religion, history, folklore, medicine, science, legends and stories of
the saints - be omitted. In the Aggadic passages there is, naturally,
more poetry, less prose; more appeal to the emotions, less to the
intellect; more attention to literary composition, less to acute
analysis. Nevertheless, no hard and fast distinction between the
Halakhic and Aggadic passages can be drawn. It can be shown that
the type of argumentation exhibited in the Halakhah is not entirely
absent from the Aggadah and, as we have seen, a striving for
literary effect is present throughout the Halakhah.
In summary, then, the following are the conclusions we can be
said to have reached.
1i) The Talmudic sugya, in its present form, is a complete literary
unit. To affirm this does not preclude the existence of earlier strata
in the sugya as we now have it. Indeed, form criticism has succeeded
in detecting such strata in many a sugya. But there is abundant
evidence that the earlier material has not been left intact. Rather
it has been reshaped and adapted to the requirements of the sugya
as we now have it, so as to form a unit complete in itself.
(2) This reshaping of the sugya has not been done haphazardly
but with consummate skill. The argument is made to proceed step
by step leading, with dramatic effect, to a climax. The presentation
of the material in the form of question and answer, argument and
counter-argument, affirmation and refutation, suggestion and
rejection, all produce a vivid effect, as if the student is present while
the debates are taking place and encourage his own participation
in them.
(3) The Talmud is a literary work. Literary recasting of this kind
can only have been achieved by creative literary effort. If any
doubts remained about the literary form of the Talmud they have
212 The Talmudic argument
surely been removed by the kind of investigation undertaken in this
book.
(4) The teachers who provided the literary units, those whom
we have called, for want of a better word, the editors, must have
flourished at the end of the fifth century and the beginning of the
sixth. That this is so can be seen from the fact that the framework
for which they were responsible includes, frequently, elaborations
of teachings stemming from early and mid-fifth-century Amoraim.
Nor can it be maintained that the sections containing these
elaborations alone are late, the others early. The amount of later
material and the uniform style of the framework are sufficient
evidence for the view that if some of the material of the framework
is late all of it is late.
(5) The actual identity of the editors cannot now be determined.
Their number is unknown, as are the methods by which they worked.
Did they have something like a card-index? From the difference
in vocabulary it seems certain, as both mediaeval and modern
scholars have maintained, that some tractates were compiled in
different places and schools from others. Yet, despite the slight
differences, the literary form is virtually the same throughout the
Talmud, which seems to suggest that at this period there existed
a uniform Babylonian method of presentation. It remains puzzling
that the Talmud contains not a hint of how it was put together or
how it succeeds in giving the impression of having all come from
a single hand. The conventional term Saboraim for the post-
Amoraic teachers can still be accepted, provided the vast extent of
their work as the actual creators of the Talmud is acknowledged.
A further problem is posed by the existence of later work, differing
markedly in style from the rest of the framework material. There
was sound tradition in the middle ages that, for instance, the
opening passage of tractate Kiddushin is a later addition and there
are other such passages, as well as obvious later glosses on the
framework. It would seem, therefore, that at least two kinds of
Saboraic material are found in the Talmud. The first and by far
the most important of these is the creative framework itself- the
Talmud proper. The second consists of the comparatively small
amount of later Saboraic material added to sugyot already com-
pleted. It cannot be denied, nevertheless, that the problems around
this whole question remain unsolved.
(6) As for the Talmudic argument itself, this is an extremely
Conclusions 213
complex form in which various moves are open to the protagonists;
in the light of what has been said previously we should rather say,
the alleged protagonists, it all being the work of those who provided
the framework. In the course of a single sugya there may be found,
for instance, an appeal to authority; a rejection of the appeal on
the grounds that the two cases are different or the authority quoted
not followed because there is a rival authority to be relied on; a
request for further information; a statement of two possible opinions
and reasoning about which of them is correct; objections on logical
grounds; and other forms of argument sketched in the first chapter
of this book.
(7) A major portion of the Talmud consists of precise, legal
definitions. These are, however, expressed in the concise and
concrete form of Hebrew and Aramaic. It is the task of the
commentator and student to translate these into the terminology
of logical analysis, getting behind the actual formulations in order
to discover their true meaning. Perhaps it can be put this way: the
Talmud is a literary work but different from every other literary
work in that its verbal exposition by those who study the work is
indispensable to the form itself. Here it is true that the medium is
the message and the message the medium.
Glossary
Aggadah: the non-legal material in the Talmud
Amora, Amoraim: 'interpreter'; 'interpreters'; name given to the post-
Mishnaic teachers (from the beginning of the third century to the end
of the sixth century) in Palestine and Babylon
Baraita:' outside teaching'; a teaching from the period of the Tannaim but
not included in the official Mishnah
Gaon, Geonim: ' excellency'; ' excellencies'; the Jewish leaders and tea-
chers in Babylon who followed the Saboraim
Gemara: 'text', originally the text of a Talmudic passage, later used as
a synonym for the Talmud as a whole
get: bill of divorce
Halakhah: the legal material in the Talmud
kosher: 'fit'; the meat of an animal fit for food
lulav: 'palm branch', used in the Tabernacle rituals (Leviticus 23: 40)
mezuzah: piece of parchment on which the Shema' is inscribed, affixed to
the doorposts (mezuzot) of the house (Deuteronomy 6: 9)
Midrash: Rabbinic exposition of Scripture
Mishnah: the official digest of Judaism compiled by Rabbi Judah the
Prince at the end of the second century
mitzvah: 'commandment', a precept of the Torah
perutah: a copper coin, the smallest coin of the realm
piggul: 'abomination', a sacrifice rendered unfit by an unlawful intention
on the part of the priest
Saboraim: 'expounders', name given to the immediate post-Talmudic
teachers
sefirat ha-'omer: the counting of the days from Passover to Pentecost
(Leviticus 23: 15-16)
Shema': 'Hear, O Israel' (Deuteronomy 6: 4- 9; 11: 13-21), the Jewish
declaration of faith recited twice daily
shofar: ram's horn sounded on the festival of the New Year (Leviticus 23:
24; Numbers 29:1).
Sifra: a Rabbinic Midrash to Leviticus
Si/re: a Rabbinic Midrash to Numbers and Deuteronomy
214
Glossary 215
sugya: a complete Talmudic unit
sukkah: ' booth' , the tent-like structure in which Jews dwell during
Tabernacles (Leviticus 23: 42)
Tanna, Tannaim: 'teacher', 'teachers', who flourished in Palestine in the
first two centuries and whose views and opinions are found in the
Mishnah, Baraita and Tosefta
tejillin: ' phylacteries'
terefar: a diseased animal, the meat of which is forbidden
terumah: the tithe given to the priest
tevel: untithed produce
Tosefta: 'Supplement' to the Mishnah
tzitzit: 'fringes', worn on the corners of garments (Numbers 15: 37-41)
Yeshivah: school or college in which the Talmud is studied
zuz: a coin, the equivalent of a denar
Bibliography
Only works referred to in the book are listed.
GENERAL
Texts
Mishnah, ed. C. Albeck, Jerusalem-Tel-Aviv, 1957
Talmud Bavli, edn. Vilna, Romm, various printings
Talmud Yerushalmi, ed. Vilna, Romm, various printings; edn. Krotoschin,
1886
Tosefta, ed. M. S. Zuckermandel, Pasewalk, 1881
Sifra, ed. I. H. Weiss, Vienna, 1862
Sifre, ed. M. Friedmann, Vienna, 1864; ed. L. Finkelstein, Berlin, 1939
Midrash Rabbah, edn. Vilna, 1878
Avot de-Rabbi JVatan, ed. S. Schechter, Vienna, 1887
Midrash Tehillim, ed. S. Buber, Vilna, 1891
Midrash Tanhuma, ed. S. Buber, Vilna, 1913
Pesikta de-Rav Kahana, ed. S. Buber, Lyck, 1868
Yalkut Shim'eoni, Warsaw, 1877
The Mishnah, translated by H. Danby, Oxford, 1933
The Babylonian Talmud, translated into English under the editorship of I.
Epstein, Soncino Press, London, 1948
Iggeret de-Rav Sherira Gaon, ed. B. M. Lewin, Haifa, 1921
Maimonides, Tad ha-Hazakah, Amsterdam, 1702
Sefer ha-Mitzvot, Warsaw, 1883
Kovetz Teshuvot ha-Rambam, ed. A. Lichtenberg, Leipzig, 1859
Moreh Nevukhim, Lemberg, 1886, English translation: The Guide of the
Perplexed, by S. Pines, Chicago, 1963
Jacob ben Asher, Turim, Warsaw, 1886
Encyclopedias / Dictionary
Encyclopedia Judaica, editors in chief: Cecil Roth and Geoffrey Wigoder,
Jerusalem, 1972
216
Bibliography 217
Entziklopedia Talmudit, ed. S. J. Sevin, Jerusalem, 1947-
Jewish Encyclopedia, general editor: Isidore Singar, printed by Funk and
Wagnalls, New York-London, 1925
Eliezer ben Yehudah, Thesaurus: Complete Dictionary of Ancient and Modern
Hebrew, Jerusalem, 1959
Commentaries
Isaiah Horowitz, Sheney Luhot ha-Berit (Shelah), Amsterdam, 1649
Joshua ha-Levi of Tlemcen, Halikhot 'Olam, edn. Warsaw, 1883
Malachi ha-Kohen, Tad Melakhi, Jerusalem, 1976
Mevo ha-Talmud, attributed to Samuel ha-Naggid, in Vilna, Romm, edn.
of Talmud, after tractate Berakhot
Samson of Chinon, Sefer Keritut, ed. J. Z. Roth with commentary, New
York, 1961; ed. S. B. D. Sofer with commentary, Jerusalem, 1965
Introductions
Albeck, C. Mevo ha-Talmud ('Introduction to the Talmud' ), Tel-Aviv,
Goitein, Baruch Benedict, Kesej Nivhar, Jerusalem, 1964
Meilziner, M. Introduction to the Talmud, 4th edn. with new bibliography by
Alexander Guttmann, New York, 1968
Strack, H. Introduction to the Talmud and Midrash, Phil adel phia, 1945
Variant readings
Rabbinovicz, R. Dikdukey Soferim (Latin title: Variae Lectiones in Mishnam et
in Talmud Babylonicum), New York, 1960
SPECIAL WORKS
Ashkenazi, Bezalel. Shitah Mekubbetzet, Lemberg, 1837 and various editions
Bacher, W. Exegetische Terminologie derjiidischen Traditionsliteratur, translated
into Hebrew: ^Erkhey Midrash by A. Z. Rabbinowitz, Tel-Aviv, 1924;
Agada der Tannaiten, translated into Hebrew: Aggadot ha- Tannaim by
A. Z. Rabbinowitz, vol. 11, Berlin, 1922
Berlin, N. Z. J. Meshiv Davar, Warsaw, 1894
Bloch, J. L. She'urey Halakhah, Tel-Aviv, 1958
Daube, David. 'Rabbinic methods of interpretation and Hellenistic
rhetoric', HUCA, 22 (1949), 239-64
De Friess, B. Mehkarim be-Sifrut ha- Talmud, Jerusal em, 1968
Edels, Samuel. Hiddushey Maharsha, commentary in Vilna, Romm, editions
of the Talmud
Eger, Akiba. Gilyon ha-Shas, marginal notes in Vilna, Romm, editions of the
Talmud
Elijah Gaon of Vilna. ha-Gra, marginal notes in Vilna, Romm, editions
of the Talmud
218 Bibliography
Engel, J. Lekah Tov, Warsaw, 1892
Epstein, J. N. Mevuot le-Sifrut ha-Amoraim (Latin title: Prolegomena Ad
Litteras Amoraiticas), ed. E. Z. Melamed, Jerusalem, 1972
Feigensohn, S. Homat Terushalayim, in Vilna, Romm, editions of Talmud
Terushalmi, beginning
Friedman, Shamma. Perek ha-Ishah Rabbah be-Bavli, Jerusalem, 1978
Ginzberg, Louis. The Legends of the Jews, Philadelphia, 1925
Goodblatt, David. Rabbinic Instruction in Sassanian Babylon, Leiden, 1975
Guttman, M. 'Sheelot Akademiot ba-Talmud\ Dvir, 1 (1923), 38-87; 2,
(Berlin, 1924), 101-64
Halevy, I. Dorot ha-Rishonim, Berlin-Vienna, 1922
Halivni, David. Sources and Traditions (in Hebrew): vol. 1, Tel-Aviv, 1968;
vol. 11, Jerusalem, 1978
Heller, Aryeh Laib. Shev Shema'tata, Jerusalem, 1957
Higger, M. Otzar ha-Baraitot, vol. x, New York, 1948
Hirschensohn, C. Berurey ha-Middot, New York, 1929-31
Hoenig, Sidney B. The Great Sanhedrin, Philadelphia, 1953
Hoffman, David. Melammed le-Hoil, New York, 1954
Hyman, A. Toledot Tannaim ve-Amoraim, Jerusalem, 1964
Jacobs, Louis. Studies in Talmudic Logic and Methodology, London, 1961
TETKU: The Unsolved Problem in the Babylonian Talmud, London-New
York, 1981
'Are there fictitious Baraitot in the Babylonian Talmud?' , HUCA, 42
(1971), 185-96
'Hermeneutics', EJ, vol. vra, pp. 366-72
' The "qal va-fyomer" argument in the Old Testament', BSOAS, 35, 2
(1972), 221-7
' How much of the Babylonian Talmud is pseudepigraphic?', JJS, 28,
1 (Spring, 1977), 46-59
Kagan, Meir Simhah. Meshekh Hokhmah, Jerusalem, n.d.
Kaplan, Julius. The Redaction of the Babylonian Talmud, New York,
1933
Klein, H. Articles on' Gemara' and ' Sebara', JQR, New Series, 38 (1947),
6
7- 9
J
5 43 (i953)> 34
I
~
6
3; JSS, 3 (i95
8
)> 3
6
3~72 and printed, together
with other articles, as Collected Talmudic Scientific Writings of Hyman
Klein, introduction by A. Goldberg, Jerusalem, 1979
Kroch, J. L. Hazakah Rabbah, vols I-X I: ed. P.J. Kohn, 1927-61 (vols I-IX
publ. in Leipzig; vol. x publ. in London; vol. xi publ. in Jerusalem);
vol. XII : ed. A. Krauss, Jerusalem, 1963
Krochmal, Nahman. Moreh Nevukhey ha-^eman, ed. S. Rawidowicz, 2nd
edn., London-Waltham, Mass., 1961
Landau, Ezekiel. Noda' Bi-Thudah, New York, 1973
Lauterbach, J. Z. 'Talmud hermeneutics', JE, vol. xn, pp. 30-3
Lewin, B. M. Rabbanan Savorai ve-Talmudan, in memorial volume for Rabbi
A. I. Kook, Azkarah, vol. iv, Jerusalem, 1937, pp. 145-208
Lieberman, Saul. Hellenism in Jewish Palestine, New York, 1950
Maimon, J. L. Abbaye ve-Rava, Jerusalem, 1965
Bibliography 219
Medini, H. H. Sedey Hemed, ed. A. I. Friedmann, New York, 1962
Meiri, Menahem. Bet ha-Behirah: to Gittin, ed. K. Schlesinger, Jerusalem,
1974; to Kiddushin, ed. A. Sofer, Jerusalem, 1963; to Bava Kama, ed.
K. Schlesinger, Jerusalem, 1973; to Makkot, ed. S. Sterlitz,
Jerusalem, 1965
Neusner, Jacob, ed.The Formation of the Babylonian Talmud: Studies in the
Achievements of Late Nineteenth and Twentieth Century Historical and
Literary-Critical Research, Leiden, 1970
Ostrowsky, M. ha-Middot she-ha-Torah Nidreshet ba-Hem, Jerusal em, 1924
Radbad. Strictures of R. Abraham ben David of Posquires on
Maimonides' Tad, printed together with the work in most editions
Ramban. Commentary of R. Moses ben Nahman, Nahmanides to Talmudic
tractates, Jerusalem, 1928
Ran. Commentary of R. Nissim Gerondi on tractate Nedarim, in Vilna,
Romm, editions of the Talmud
Rashba. Commentary of R. Solomon Ibn Adret on Talmudic tractates,
Briinn, 1798, Warsaw, 1883 and various editions
Rashi. Standard commentary on the Talmud by R. Solomon Yitzhaki,
printed at the side of the text in practically all editions
Riban. Commentary on part of tractate Makkot in Vilna, ed. by R. Judah
ben Nathan
Ritba. Commentary on tractates of the Talmud by Yom Tov Ishbili,
various editions
Rosh. Digest of the Talmudic Arguments by R. Asher b. Yehiel (c.
1250-1327), printed in the Vilna, Romm, editions of the Talmud
Rubinstein, S. M. le-Heker Siddur ha-Talmud, Kovno, 1932
Schechter, S. Studies in Judaism, Philadelphia, 1945
Schwarz, A. Der Hermeneutische Syllogismus in der Talmudischen Litteratur,
Karlsruhe, 1901
Die Hermeneutische Antinomie in der Talmudischen Litteratur, Vi enna,
Die Hermeneutische Quantitdtsrelation in der Talmudischen Litteratur, Vienna,
Skopf, Simeon. Shtfarey Tosher, Warsaw, 1928
Soloveitchik, Hayyim. Hiddushey Rabbenu Hayyim ha-Levi, with notes by
Rabbi A. I. Karelitz, entitled Hazon Ish, n.p., n.d.
Talmage, Frank E. David Kimhi The Man and His Commentary, Harvard
University Press, 1975
Tavi ob, I. H. ' Talmudah shel Bavel ve-Talmudah shel Eretz TisraeV, in
his Collected Writings (in Hebrew), Berlin, 1923, pp. 73-88
Tenenblatt, M. A. The Formation of the Babylonian Talmud (in Hebrew),
Tel-Aviv, 1972
Tosafists. Standard glosses to the Talmud from the French and German
schools, printed together with the text in most editions of the Talmud
Urbach, E. E. Hazal, Jerusalem, 1969
Vidal Yom Tov of Tolosa. Maggid Mishneh, Commentary on
Maimonides' Tad, printed in most editions
220 Bibliography
Weiss, Abraham. Hithavut ha-Talmud Bishlemuto (English title: The
Babylonian Talmud as a Literary Unit), New York, 1943
le-Heker ha- Talmud, New York, 1954
Weiss, I. H. Dor Dor ve-Doreshav, New York-Berlin, 1924
'Survey of Talmudic methodologies' (in Hebrew) in Bet Talmud, vol.
I, Vienna, 1881, pp. 26-31, 53-60, 85-9, 105-12, 153-9, 181-4; vol.
II, Vienna, 1882, pp. 1-8