Studies in Talmudic Logic and Methodology [1 ed.] 0853035873, 9780853035879

Citation preview

I

STUDIES I TAL

UDIC LOGI . . . . AND

ETHOD""""'LOGY •

LOUIS JACOBS

LONDON

VALLENTINE, MITCHELL •

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

© Lo11is Jacobs 1961 •

First published in 1g61 by Vallentine, Mitchell e- Co. Lttl., 37 Furniva/, Street, London, EC4

Made and p1'inted in G,eat B1'itain by C. Tinling e- Co. Ltd. Liverpool, London and P,escot

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

BM -9J3 .

.b

J?,

l Cf b I

l'1,A,It{ Contents Page •• vu

I nt,oduction Part I - Logic 1 The AristotelP.an Syllogism and the Qal Wa-.{:lomer

3

•

2 The Talmudic Her111eneutical Principle of Binyan 'Abh and J. S. Mill's 'Method of Agreement'

9

3 The TalIIJ.udicSebha,a in its Relationship to Scripture

16

4 The Reductio ad Absu,dum in the Talmudic Literature

38

Part II - Methodology 5 Evidence of Literary Device in the Babylonian Talmud 6 Further Evidence of Literary Device in the Babylonian Talmud

53 60

7 The Literary Analysis of the Talmudic Sugya 8 The Literary Analysis of the First Sugya in Baba Qama, 2a-3b

70 132

9 The Literary Analysis of Be,akhoth 45a-b

152

10 The Literary Analysis of Baba Metzi'a 76a

159

699 Original from

UNIVERSITY OF CALIFORNIA

DEDICATED TO THE REVERED MEMORY OF RABBI JACOB KORNBERG AUTHOR OF THE COMMENTARY ••KOHELETH YAAKOB'' TO THE MISHNAH

,n,M ,M,~01' 1:i :ipr

11 ,

n~,:i, p,-m ,~,

r,,,~~0

;,v

11

:ip.v, r,;npn

,n,,,0 Q 11 Ml':l

BY HIS SON ISAAC ELCHANAN KORNBERG

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

-

Introduction IN the traditional Talmudic Methodologies the subjects of Talmudic logic and methodology are generally treated together. They share the unhappy fate of almost total neglect by presentday students of the Talmudic literature. Talmudists of the old school are content in their attempt to unravel the complexities of the Talmudic sugya and tend to look askance at any kind of systematic approach to Talmudic method. The modem schoo] of Talmudic learning has achieved wonders in delineating the background of the Talmud and describing the historical, sociological, economic and cultural forces which had their influence on the Talmudic teache1s. But, with a few notable exceptions, there has been hardly any attempt at a systematic presentation of such matters as the structure of the Talmudic discussions, the for1nulation of the Talmudic arguments, the division of the sugya into its component parts, the methods of Talmudic reasoning, and the literary style and character of the Amoraic debates. The present very small attempt to fill the gap does not offer a complete guide to the numerous problems in the field. Its aim is rather to note some of these problems and suggest some tentative solutions. It is hoped that the four chapters on Talmudic logic will shed some fresh light on this difficult subject. In Chapter I I. have tried to show that the conventional identificationoftheQalWa-/JomerwiththeAristoteleanSyllogism is erroneous. In Chapter 2 I have suggested that the Rabbis, in their investigation of Scripture, used a method that is almost identical with John Stuart Mill's classification of the 'Method of Agreement'. Taken in too bald a fashion this would be, of course, grossly anachronistic. My thesis is only this, that the reasoning of the Rabbis led them to discover (over a long process of time) a means of examining Scripture which accords with similar methods used in the investigation of the natural order by modern science. If this is correct the established notion that the ancients knew nothing of Induction will have to be revised. Chapter 3 deals with the Talmudic sebhara in its relationship •• vu

Original from

UNIVERSITY OF CALIFORNIA

•••

INTRODUCTION

Vlll

to Scripture and Chapter 4 with the reductio ad absurdum in the Talmudic literature, both subjects that have, till now, not received adequate treatment. The second part of the book treats of the literary analysis of the Amoraic literature. In Chapters 5 and 6 I have tried to provide the evidence for literary device in t~e Babylonian Talmud. Chapter 7, the main chapter in this part of the work, hence its disproportionate length, seeks to analyse three sugyoth and draw certain conclusions which, it is hoped, will prove valuable to the 11nderstanding of this branch of Talmudics. In the remaining chapters three further sugyoth are subjected to the same kind of treatment. In all these chapters the main thesis is that the Gemara in the form we have it now is very far from being a verbatim report of discussions which took place in the Babylonian schools but is rather a 'contrived' literary product of great skill, in which the older material used has been reshaped by methods bearing a close resemblance to those of literary artists throughout the ages. Our concern is chiefly with the Babylonian Talmud but the Palestinian is quoted, as well as other contemporary and modem literature~ _wherever these throw light on the argument. The somewhat lengthy accounts of various Talmudic passages are necessary for the development of the argument but, in addition, they may be of some assistance to the student with the more limited aim of grasping the meaning of the Talmudic sugya. Expert Talmudists readily admit that the only way to a complete appreciation of a given passage is to see it as a whole and to recog1iise clearly the place of its detailed parts in the general scheme. Although the work is presented in the first instance for the consideration of students of the Talmud it may be of interest to the layman who has a desire of gaining a closer acquaintance with the full flavour of Talmudic reasoning. There are few works indeed, even in Hebrew, and hardly any in English, to which such a reader can be referred. Learned theological works, for example, frequently quote from the Talmud, but the reader without first-hand knowledge of the subject would hardly guess that these convey nothing of the essential thrust and parry, argument and counter-argunient, thesis and refutation, which give the Talmudic dialectic its special form. Quotations from Hamlet are no doubt of value but 'the play's the thing'. If we

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

1

•

INTRODUCTION

lX

are to understand what Shakespeare is about the only way to do it is to read the play as a whole with the necessary aids. The Soncino English translation of the Babylonian Talmud is, of course, of great help here but students generally find the need for some kind of supplementary guide into the intricacies of the debates. . It has not been thought necessary to provide either a glossary of technical terms or a list of abbreviations. The more important of those used have been explained in the context, the others will be of little interest to laymen while the scholar will recognise them without aid. The works referred to in the notes will be found helpful for further study. The scheme followed for the transliteration of Hebrew and Aramaic is that of the Jewish Encyclopedia with some slight deviations. A word should, however, be said on the meaning of the term 'sugya' (plural sugyoth) which occurs frequently. The root of this Aramaic word means 'to walk', hence, the 'lay-out' of a particular Talmudic passage, a complete unit of the Talmudic debate or discussion, a sustained Talmudic argument in its entirety. My thanks are due to the Editor of the Journal of Jewish Studies in which periodical a part of this material appeared originally in article for1n. Since then I have had many opportunities of discussing the questions raised with colleagues and students, to all of whom my thanks are due.

•

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

-

PART I

LOGIC

Digitized by

Go gle

Original from

UNIVERSITY OF CALIFORNIA

•

•

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

CHAPTER ONE

The Aristotelean Syllogism and the Qal Wa-Qomer in his well-known work, De, He,meneutische SyUogismus in de, Talmudischen Litte,atu,,1 suggests in the title, and develops in the work itself, the idea that the Talmudic hermeneutic mode of qal 1 wa-1,ome, is identical with the Aristotelean Syllogism. It will be shown here that not only is there no connection between the two fom1s of reasoning but that an analogy to the Syllogism is found in the Talmudic literature as something quite different from the qal wa-fw,ne,. The refutation of Schwarz's view is important because all too many scholars uncritically follow Schwarz in his identification. 1 Before commencing our investigation it is necessary to state that there are two kinds of qal wa-!Jome, - which I propose to call the simple and the complex. The simple qal wa-1,ome, is a plain argwnent de mino,e ad majus, in which the severity of the major over the minor is self-evident. The complex qal wa-fw,ne, is one in which this severity has to be proved by reference to external factors. The simple qal wa-1,ome, has a long history - its use is traced by the Mid,ash' to the Bible itself, as, for example, in the verse: 'Behold, the money, which we found in our ADOLF SCHWARZ,

1

Der Henneneutische Syllogismus in der Talmudischen Litte,atur, Ein Beitrag Zu, Geschichte De, Logik Im Morgenlande, Karlsruhe, 1901. 1 According to Schwarz, the correct reading is qol, a reading found in a number of important MSS. (op. cit., pp. 8-14). 1 In 'Otzar Yisrael, for example, Schwarz's view is quoted with apparent approval, s.v. qal wa-fJome,. ' Gen. R. 92. 7, in which it is said that there are ten qal wa-fJomerin in the Torah, cf. YalquJ, I Sam. 132. The ten qal /Jomerin are: Gen. :xliv. 8; Ex. vi. 12; Deut. xxxi. 27; Num. xii. 14 (not explicit in the verse, but see B.Q. 25a); Jer~ xii, 5; Jer. ibid.; I Sam. xxiii, 3; Prov. xi, 31; Esther ix, 12; Ezek. xv, 5. The commentaries state that far more than ten are to be found, see the learned discussion in H. Hirschensohn's Be,ut'e HaMiddoth, Jer., 1939, pp. 39-00, and the bibliography given in H. L. Strack's: 'Introduction to the Talmud and Midrash', Eng. Trans., Philadelphia, 1945, p. 285, n. 3. 3

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

4

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

sack's mouth, we brought again unto thee out of the land of Canaan: how then should we steal out of thy lord's house silver or gold?' (Gen. xliv,8) 1 • The complex qal wa-pC>me, is of later Halakhic origin and is found in a Ba,aitha,• for example, as follows: 8 'If priests who are not disqualified for service in the Temple by age are yet disqualified by bodily blemishes' then Levites who are disqualified by age 5 should certainly be disqualified by bodily blemishes'. There is no possibility here of a simple inference from priests to Levites since the 'severity' of the latter is not self-evident. But this 'severity' can be proved by referring to the factor of disqualification by age, which operates in the case of Levites and not in the case of priests, and which proves that Levites are more readily to be excluded from Temple service than priests. Hence, it is argued, where priests are excluded Levites should certainly be excluded. Symbolicallythe two types of qal wa-lwme, may be represented as follows:

Simple: If A has x then B certainly has x. Complex: If A, which lacks y, has x then B, which has y, certainly has x. Of the two types of qal wa-fiomer it is the simple one which has affinities with the Syllogism in that both the simple qal 1

,

A simple qal wa-/Jome,. of this kind occurs in the Mishnah, 'Abhoth, I, 5: 'They said this (and talk not much with womankind) in regard to a man's own wife: how much more (qal wa-/Jomer) in regard to his fellow's wife!' This may be the earliest reference in Rabbinic literature to the qal wa-/JomeY, occurring as it does in a saying of Jose b. Jol}.anan of Jerusalem (c. 16o B.C.E.), but the highly plausible suggestion has been made that the qal wa-fiomw does not belong to Jose's saying but is a later editorial addition, v. Schwarz, op. cit. p. 11. David Daube, in his article, 'Rabbinic Methods of I nte,-p,-etation and Hellenistic Rhetoric', in HUCA, Vol. XXII, 1949, p. 239f., points to the use of the qal wa-fiome,. in Matthew xii, 1of., Luke xiii, 14f., and Romans v, Sf., and remarks that they are of a more 'legal' nature than those found in the Q.T. They are, however, all examples of what I have called 'simple' qal wa-[iome,.. CJ. Foundations of Rabbinic Judaism, by Alexander Guttmann, in HUCA, Vol. XXIII, 1950-51, 75th Anniversary Publication, Part I, pp. 453-473. 1 A similar fo1m is found in the Mishnan, B.Q. II. 5. 1 l;lull, 24a. ' Lev. xii, 16f. I N . . .. um. 1V, 23; Vlll, 24.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

5

THE ARISTOTELEAN SYLLOGISM

wa-[wme, and the Syllogism draw their conclusion from a major and minor premise without having recourse to any external factors. The complex qal wa-[wme, is, as we have noted, of late origin and appears to be an halakhic-methodological development from the simple one found in the Bible. And this in itself is a valid refutation of Schwarz's view, for the Pentateuch and the other parts of the Bible in which this argument appears can hardly have been influenced by Aristotelean logic. But, apart from this, a careful examination of the two methods will convince us that the qal wa-fiome, and the Syllogi~m are not identical. .... For the purpose of comparison let us set ~side by side a Syllogism in Barbara with a simple qal wa-[wme, as used in the Mishnah: 1 '

•

SyUogism: All men are mortal; Socrates is a man; Therefore Socrates is mortal.

Qal Wa-Qome,: If then in regard to such a light precept (i.e. the law of sending away the dam - Deut. xxii. 6-7) which concerns a matter that is worth but an 'issa, the Law has said that it may be weU with thee and that thou mayest prolong thy days, how much more so in regard to the weightier precepts of the Law!

It can be seen at once that there are two fundamental differences between these two forms of reasoning, so fundamental, indeed, that they prevent us from claiming that the two are identical. First, there is the obvious point that the element of 'how much more so' is lacking in the Syllogism. Modern logicians have been at pains to criticise the Syllogism as being tautologous - we can only say 'all, men are mortal' if we know that Socrates, too, is mortal, and if we know this there is no need for the Syllogistic inference. 1 1bis objection cannot be levelled against the qal wa-fwme,. Observation informs us that the reward of long life is promised for the carrying out of the 1

]full. XII, 5.

V. L. S. Stebbing: A Modern Jnt,-oduction to Logic, London, 1945, p. 216f., and the attack on the traditional Logicians with regard to the a forlion argument, pp. 173-174. 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

6

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

light precept but observation has nothing to tell us about the reward for carrying out the weighty precept. This must be derived by using the mode of qal wa-fwme,. A weightier difference, and one which finally disposes of Schwarz's viewpoint, has been noted by Arnold K11nst in an important study in which the qal wa-fwme, is compared to an analogous form of Indian inference known as kimpuna,. 1 Kunst remarks:• The mistake (in Schwarz's reasoning) is that the Aristotelean Syllogism, which deals only with names, involves the reference between the subject and the predicate, whereby the relation between the minor and major premises is the relation of the species to the genus, both being nothing but names, whilst kimpuna, and qal wa-/wmef deal with sentences'. If I understand him correctly, what Kunst is saying is that in the Syllogism the inference concerns the relationship between genus and species; we are saying that since Socrates belongs to the class man then he must share the characteristics of that class. Whereas in the qal wa-fw,ne, inference we do not say that a weighty precept belongs to the class light p,ecepts,· it obviously does not. We say that what is true of light precepts is true of weighty precepts. 8 'An Overlooked Type of Inference' in Bulletin of IM School of Oriental and African Studies, Vol. X, Part 4, 1942, pp. 976--991. Ip. 987. • Kunst remarks (p. 991) with reference to the analogy to Indjan logic: 'Whether the similarity of this inferential procedure between the Jews and the Indians was a result of mutual influence, or whether it was only an expression of a common human tendency to eulogise great things by comparing them with smaller, or to raise the value of small things by juxtaposing them with greater-this problem may be left to further historical researches. The author would vote for the latter alternative'. And indeed the latter alternative does appear to be the more plausible one. It is, however, of interest in this connection that we do find, in quite another sphere, a remarkable parallel between Jewish and Indian law. All land in both Palestine and Babylon was divided, during the Talmudic period, into three grades., excellent, mediocre, and bad arable land, Mishnah, GilJ. V. 1, B.Q. 7b, 8a-b, and frequently elsewhere. Heic~elheim has noted that this grading existed in Hellenistic India, as evidenced by the Arthacastra of Kautilya, a voluminous Indian guide to state administration, discovered in 1905, and this unique division is not found in general Hellenistic administration, see his remarks in 'Economic History', 1938, p. 10, n. 27, and 'Cambridge Ancient History', Vol. VII, p. 897. 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

THE ARISTOTELEAN SYLLOGISM

7

It is now our task to demonstrate that, in fact, an analogous mode to the Syllogism is used in the Talmudic literature but it is not a qal wa-/wmer. This mode is found frequently when the Talmud attempts to show that a given c.ase falls under the heading of a more general principle. A good example of this is found in the passage1 in which it is stated that all authorities agree that the benediction over the search for leaven before Passover must be recited before this religious duty is carried out, not afterwards. The Gema,a asks: 'How do we know this?' To which the answer is given: 'Because R. Judah said that Samuel said: In the performance of all religious ceremonies one must say the benediction over them before performing them' . 1 'Ibis is a clear example of Talmudic use of the Syllogism. To put the argument in Syllogistic for1n: All religious ceremonies req11ire their benediction to be said before them; The search for leaven is a religious ceremony; Therefore the search for leaven req11ires its benediction to be said before it. To sum up. The simple qal wa-fwm,e, is certainly, at least in its Biblical form, more ancient than the Aristotelean Syllogism. Apart from this, the two are different in form and content. The qal wa-fwm,e, is based on the proof de minore ad majus and this element is absent in the Syllogism. The Syllogism deals only with names, with propositions, the predicates of which are the genus of the species or subject. The qal wa-fwm,e,, on the other hand, deals with sentences and does not deal with classes and with mutual relations of those classes. Consequently, the qal wa-fwm,e, and the Syllogism are not identical. Finally, there is a form of Talmudic reasoning which has no connection Pes. 7b. 1 For further examples of this v. B.Q. 61a; Keth. 10b; ]:full. 59a-b; Qidd. 11a; B.M. 35a, 42a, and very freq. V. especially in this connection the frequent use of the word 'ha-kol' (better, the phrase of a Mishnah or Baraitha beginning with this word) e.g. in IJull. 2b-3b, 'Arak. 2a-3b, Tem. 2a, B.M. 77b and very freq. 1

B

Original from

UNIVERSITY OF CALIFORNIA

8

STUDIES IN TAIMUDIC LOGIC AND METHODOLOGY

with the qal wa-/,ome,1 but bears a remarkable affinity to the Syllogism. In support of the distinction between the 'ha,-kol' fo1m1.lla, which is identical with the Syllogism, and the qal wa-f,omer which has no connection with it, we may quote the following well-known Baraitla ('A.bholl,, VI.3): 'He who learns from his fellow a single chapter, a single rule, a single verse, a single expression, or even a single letter, ought to pay him honour, for so we find with King David, King of Israel.... Now is it not an argument from minor to major (qaJ wa-ltomer)? If David, the King of Israel ... how much more ought one who learns from his fellow ... to pay him honour!' Here the qal wa-ltomer method is used to establish the truth of the statement: 'He who learns from his fellow ... ' and this statement then becomes the major premise in a Syllogism so that the reader says (such are the implications of the Baraitha) something on these lines: 1

He who learns from his fellow has to pay him honour; I have learned from my fellow; Therefore I am obliged to pay him honour.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

CHAPTER TWO

The Talmudic He1·1neneutical Principle of Binyan 'Abh and J. S. Mill's 'Method of Agreement' THE idea of the influence of Aristotelean logic on the Talmudic her1neneutical principles (middoth), via Shemaiah and Abtalion, the teachers of Hille!, was first advanced as an hypothesis by Adolf Jellinek in the last century and has since won wide acceptance.1 As we have seen in the previous chapter, Adolf Schwarz in particular is influenced by this idea.• In the nature of things there can have been little correspondence in the Talmudic literature with modem inductive logic. Schwarz, in his work on the principle of bi,nyan 'abh, 8 remarks on this' and seeks to explain this principle as a for1n of analogy: · to use his own terminology, as a type of either 'Species Induction Reference' or 'Genus Induction Reference'. 5 It will be suggested here that a more fruitful way of explainin~ the principle is to •

V. I. H. Weiss: DOI' DOI' We-DOl'eshaw, 2nd ed., New York-Berlin, 1924, Vol. I, p. 164, n. 1. 1 C/. his Die Hemseneutische Antinomie in de,, Talmudischen Litte,-atuY, Vienna, 1913, and Die He,-meneutische Quantittitst'elation in de,- Talmudischen Litte,-atut'', Vienna, 1916. 1 Die He,-meneutische Induktion in de,- Talmudischen Littet'atut', Vienna, 1909. CJ. Encyclopedia Judaica, Berlin, 1929, Vol. VII, p. 1190, and Encyclopedia Talmudith, Vol. IV, Jer. 1952, pp. 2-11. ' Die Hemieneut. Induk., p. 158: 'Induktionsschliisse in diesem Sinne (der erkenntnistheoretischen Logik) hat weder das Altertum noch das Mittelalter gekannt; es ist darum selbstverstandlich, das die hermeneutische lnduktion in der Talmudischen Litteratur, welche den lnhalt dieser Arbeit bildet, nicht duf'ch die e,-kenntnisthe01'etische, sondern nut' du,-ch die At'istotelische Logik beleuchtet we,-den kann' (italics mine). (Inductive judgements in this sense (i.e. in accordance with the logic of theory of knowledge) were known neither in antiquity nor in the middle ages; it is obvious, therefore, that the hermeneutical induction in Talmudic literature, the subject of the present work, cannot be explained by the logic of theOYy of knowledge but only by the Aristotelean logic). 1 Hemien. Induktion, p. 17of. and p. 228f. 1

9

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

10

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

compare it with John Stuart Mill's method of agreement,1 to which it bears a striking resemblance. It goes without saying that we are not suggesting any kind of anachronistic 'anticipation' of Mill by the Rabbis. All we suggest is that the Rabbi!, in their attempt to discover general principles behind the laws of the Torah, used, apparently, a method similar in form to that classified by Mill as a means of discovering the laws of nature. The Mishnah in tractate Qiddushin• rules that a woman can be acquired in marriage in three ways: by the payment of money, keseph, by the delivery into her hands of a bond, she,a,, and by cohabitation, bi'ah. In Amoraic times, the M ishna.h was interpreted to mean that any one of the three ways suffices to bring about a valid marriage. The Gema,a 8 proceeds further to demonstrate the validity of yet a fourth way, that of the entrance of the man and woman under a canopy- fiuppah. Now this way argues the Gema,a-cannot be derived from keseph alone by simple analogy, for it may be objected that keseph has greater power than fiuppah, since it can be used for the purpose of redeeming consecrated things as well as the second tithe which, according to the Rabbinic interpretation of Deut. xiv. 22-26, has a certain degree of sanctity. Similarly, puppah cannot be derived from bi'ah alone, since the latter has greater power in that it may serve for the purpose of acq11isition in levirate mar1·iage. But by taking keseph and bi'ah in conj11nction we can derive !Juppah. Bi'ah by itself shows that the power of redeeming consecrated things and the second tithe is not the decisive factor in effecting a valid marriage, since bi'ah effects a valid marriage though it cannot be used for the purpose of redemption. Keseph by itself shows that the power of acquisition in levirate marriage is not the decisive factor since keseph effects a valid marriage even though it cannot be used for levirate marriage. If, therefore, the specific feature of either bi'ah or keseph fails to constitute the decisive factor in effecting a valid marriage, this factor must reside in some other feature which both bi'ah and keseph have in common. The common feature is that both keseph and bi'ah have the 'power of acquisition elsewhere' (outside betrothal). John Stuart Mill: 'A System of Logic', Book Ill, Chapters VIII and IX. I I. I. 1 Qidd. 5b. 1

Digitized by

Go ~ gle

Original from

UNIVERSITY OF CALIFORNIA

THE TALMUDIC HERMENEUTICAL PRINCIPLE

II

Now, since the deciding factor in effecting a valid marriage is due to the 'power of acquisition elsewhere' and /luj>pah also this power (i.e. where a betrothal has taken place a subsequent puppah acts as the means of acq11isition for the various monetary obligations between husband and wife), /luppah also can effect a valid marriage (i.e. a betrothal in the first place). Thus /luj>pah is derived from bi'ah and keseph. "Ibis can be represented symbolically as follows: Let A represent the 'power to acquire elsewhere', B the 'power to redeem consecrated things and the second tithe', C the 'power to acq11ire in levirate marriage' and a the power to acq11ire in ordinary marriage (betrothal). Let the sign---➔ mean 'results in' and the sign - - - mean 'is the cause of'. Then:

Keseph AB · ► a Bi'ah AC--➔ a Therefore A - - a Hence t,uppah, which has only A, ► a. The Gema,a then proceeds to show, by using the method of elimination, that the decisive factor in effecting a valid marriage is, in fact, the common feature of possessing the 'power to acquire elsewhere' and not any other feature which bi'ah and keseph have also in common. The other common feature in bi'ah and keseph is the 'degree of pleasure' accompanying the performance of the act. The Gema,a assumes first, by way of hypothesis, that it is this other common factor that constitutes the decisive factor in effecting a valid marriage. "Ibis, if the degree of pleasure is represented by D, would be symbolised thus: Keseph ABD ► a Bi'ah ACD ► a Therefore (possibly) D (not A) - - a The hypothetical assumption is then eliminated in the Gema,a by reference to sJiela,. IJuppah - it is argued - cannot be derived from she/a, alone for she/a, may effect a divorce, but not /luj>pah. Shqa,, however, does show that the decisive factor is not the 'degree of pleasure' since there is no pleasure in the case of shqa,. On the other hand, keseph and bi'ah show that the decisive factor is not the power to effect a divorce for they can effect a valid marriage though they do not possess this power. Consequently, the deciding factor must be the one that

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

12

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

keseph, she,a, and bi'ah have in common, namely the 'power to acq11ire elsewhere', and this feature is shared by !,uppah, too. If E represents the power of effecting a divorce we would have: Keseph ABD Bi'ah ACD

► ►

a

a

Shela, AE --➔ a •

Therefore A - - - a Hence !,uppah, which only has A,----+ a. It will be seen from an examination of the above (and practically every example of binyan 'abh in the Talmudic literature is developed on identical lines) 1 that we have here an exact analogy to what John Stuart Mill calls 'The Method of Agreement'. He defines it in the following way: 'If two or more instances of the phenomenon under investigation have only one circumstance in common, the circumstance in which alone all the instances agree is the cause (or the effect) of the phenomenon' . 1 Mill is careful to write 'have only one circumstance in common' for if there are more than one there is no means of knowing which of them is the antecedent of the consequence. Thus in the example of the binyan 'abh given above, !,uppah cannot be derived from keseph and bi'ah for these have two factors in common, only one of which is possessed by "l}uppah. Mill gives an example of the method: 'Instances in which bodies assume a crystalline structure are found to have been preceded by instances which have in common only one antecedent, namely, the process of solidification from a fluid state. This antecedent, therefore, is the cause of crystalline structure'. Another example, given by A. Wolf, a is: 'Brewster took impressions from a piece of mother-of-pearl in a cement of resin and beeswax, in balsam, in fusible metal, in lead, in gum arabic, in 1

Examples of the full employment of the method are to be found in Bet'. 35a; Yeb. 77a; Qidd. 5b, 21a; Naz. 40a; B .Q. 6a; B.M. 4a, 87b; Sanh. 66a; Makk. 4b; Zeb. 5a, 5b, 11a, 16b; Men. 6a, 6ob; ]:lull. 114a. These are given by I. Berlin in his notes to the Talmud, Vilna ed. Schwarz, Induktion, p. 61, n. 2, adds, further, Sabb. 28a and Qidd. 78a. There are also many examples of not fully developed binyan 'abh which imply the use of the method as described above. V., for example, Qidd. 5a and B.Q. 5b and Rashi ad loc. s.v. we-khulehu. 1 Op. cit., loc. cit.; cf. L. S. Stebbing: 'A Modem Introduction to Logic', London, 1948, pp. 333-334. 1 'Textbook of Logic', London, 1930, pp. 217-218.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

THE TAI.MUDIC HE'RMENEUTICAL PRINCIPLE

13

isinglass etc. In all cases the same iridescent colour appeared. But the only character which these substances had in common was the form of the surface produced by the impression of the piece of mother-of-pearl. Hence that form of surface must be a condition of the iridescent colour'. The principle underlying the method of binyan 'abh as developed in the Talmudic literature is the belief, very real to the Tannaim and Amoraim, of the unity of the Torah, just as the principle behind the method of agreement is the belief in the 11nity of nature. To the rational mind it appears inconceivable that the different characteristics of the instances examined should result in the same effect rather than the characteristic they have in common. In the same way it appeared impossible to the Rabbis that a principle found in a number of instances of the Torah laws should be attributed to diverse factors rather than to a common factor. In view of this the discussion of modem scholars as to whether the binyan 'abh is an exegetical or a logical principle1 is apt to be misleading. In a sense it is both. It is exegetical because its application is based on the exegetical principle of the unity of the Torah, but given that premise its application is logical. It must further be borne in mind that in many, perhaps in most, cases of its application the law derived was already known; the method being used not as a basis of derivation but as a means of 'discovering' a support in the Torah for the law already known.• Of course, the method of binyan 'abh, like the inductive method of agreement, is not infallible. There is always the possibility that a common factor has been overlooked. Generally speaking, as we have seen, the Talmud seeks to eliminate all factors but one in its attempt to establish the principle of causation. Sometimes, as will be seen from the following example, the common factor V. H. Hirschensohn's Be,-u,-e Ha-Middoth, Jer., 1939, pp. 184-185, and V. Aptowitzer in M.G.W.J., 1911, pp. 185-189 (review of Schwarz's book). 1 This is connected with the much-discussed question of whether the Mid,-ash method preceded that of the Mishnah or vice versa. For recent contributions to the elucidation of this problem, v. Jacob z. Lauterbach: 'Midrash and Mishnah' in 'Rabbinic Essays', Cincinatti, 1951, pp. 163-258, and H. Albeck in the Alexander Marx Jubilee Volume, New York, 1950 (Hebrew Section), pp. 1-8. At all events, it is certain that there are many de,-asholh which are later than the law they seek to support. 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

14

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

in the instances examined, which prevents the establishment of a general rule, is the very peculiarity of those instances. Thus, the Talmud1 seeks to prove that the negative injunction against cursing a parent is derived from the prohibition of cursing a prince, a judge, and a deaf man. The factors these instances have in common is that they are members of 'thy people', a feature shared also by the parent. The retort to this argument is that the former three instances are all exceptional the prince and the judge occupy a special position, and the deaf man stands alone in his affliction. There is, therefore, no evidence by which a prohibition against cursing a 'nor1nal' person (parent) can be derived from those instances. There are, however, examples in the Gema,a of an apparent rejection of the binyan 'abh inference. It is maintained in some cases that the instances examined, though they have no factor in common, have none the less separate characteristics not possessed by the new instance to be included under the general principle. At first glance this appears to question the validity of every binyan 'abh, but a proper understanding of the passages in question will reveal that this is not so. The suggestion is made in the Gema,a 1 that R. Judah (second century), who req11ires the penalty of flagellation for the transgression of a negative precept, even where no action is involved, derives this both from the case of the slander of a wife by her husband (Deut. xxii 13-19) and from that of false witnesses (Deut. xix 16-21). In each of these cases the penalty is flagellation even though no action is involved. From the case of slander by itself no inference could be drawn, because this case is distinguished by an unusual seveijty in that the husband has to pay a fine in addition to receiving flagellation. Also no inference could be drawn from the case of false witnesses by itself, for the witnesses are punished, unlike in other cases of transgression, without warning being given to them before the commission of the crime. But from both cases together it is possible to derive our law; for the case of the witnesses shows that the penalty of flogging is administered even where there is no fine, and the case of slander shows that this penalty is administered even where a prior war,iing is required. To this the 1 1

Sann. 66a. Makk. 4b.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

THE TALMUDIC HERM:ENEUTICAL PRINCIPLE

15

Gemara retorts: 'The common factor in the case of slander and in that of false witnesses is that they possess a severity of some kind' (she-yesh bahem tzad fiamur) i.e. in the case of slander there is a fine and in the case of the witnesses no warning is required. 1nis retort is met by the statement that R. Judah denies that there is a common factor in these two cases. It would thus seem that we have here an attempt to challenge the whole method of binyan 'abh ,· for if we accept the view that different characteristics are considered to be a common factor merely because the new instance has none of them, what becomes of the whole principle of binyan 'abh? "Ibis question is raised bythe Tosaphists 1 but their solution is an artificial one. One thing is clear. There can be no attempt here to challenge the general principle; for if this were so why is the challenge presented only here and nowhere else? On closer examination we can easily see why this case differs from others. For in this case there is no factor common to the instances examined and to the new instance that is to be derived from them. 1bis is a case of simple analogy and here it may well be that the different characteristics are the cause of the law. In other words, the method of binyan 'abh is based on the idea that rather than assume that different characteristics are the cause of an effect, we must assume that a common factor is the cause. But for this to be assumed there must, of course, be a common factor.• Symbolically: let F represent the case of false witnesses, S the case of slander, a the characteristic of punishment without warning, b the additional punishment of a fine, and x the punishment of flagellation. Then the premises would be:

Fax

Sbx

It is obvious that from these premises no new case possessing neither a nor b can be derived; whereas in the usual case of binyan 'abh there is yet another factor shared by the premises and the new instance. Thus, if N stands for the new case and c for the factor common to all three, then: if Facx and Sbcx then Ncx, that is, if F ac is followed by x and Sbc is followed by x, then the cause of x must be c: therefore N ex. 1

Makk. loc. cit., s.v. 'ela. 1 This correct solution of the difficulty is given in Betzalel Ashkenazi's ShiJJa M equbbetzeth to Keth. 32a.

Original from

UNIVERSITY OF CALIFORNIA

CHAPTER THREE

The Talmudic Sebhara in its Relationship to Scripture FoR the purpose of clarifying the Talmudic concept of sebhara 1 •

with the meaning of 'common sense' or 'reasonable argument•. in its relationship to the word of God as revealed in Sc1ipture the relevant Talmudic passages must be examined. The concept is Amoraic and, as we shall see, largely of a theoretical nature. None the less a study of the concept does throw light on late Rabbinic views on the relationship between reason and revelation, the problem which looms so large in Mediaeval philosophy. The-method we have adopted is to scrutinise all the passages in which the concept occurs and then to draw the necessary conclusions.

(a) B.Q. 46b . R. Samuel b. Na.J:imani said: 'How do we know that the onus of proof falls on the plaintiff (rather than the defendant, i.e. what is the Scriptural warrant for the established legal principle that possession cannot be disturbed without conclusive proof?)? It is said: ''If any man have any matters to do, let him come unto them'' (Ex. xxiv, 14) -let him bring proof before them.' R. Ashi raised an objection: 'Why is a Sc1iptural text required? Is it not common sense (sebhara)? One who is in pain visits the house of the doctor.' But the verse is required for the statement of R. Nal)man in the name of Rabbah bar Abahu (i.e. no verse is required for the principle that the onus of proof falls on the plaintiff) for he said: 'How do we know that the Court should first consider the claims of the plaintiff {and then any co11nterclaims the defendant may have)? It is said: ''If any man have any matters to do, let him come unto them'' - let him cause his matters to be brought before them' (i.e. the matters of the From root s1bha,-, 'to think', in Aramaic, v. Kohut, 'Arukh Compl,tu.m, s.v. and Levy's Wiwterbu.ch, s.v. · 1

16

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

17

THE TALMUDIC SEBHARA

plaintiff should be brought and a decision taken on them before any counter-claim is considered). Here the term sebhara is used to denote 'common sense', 'self-evident truth'. It would be just as unreasonable, suggests R. Ashi, to expect the defendant to prove his title to property in his possession as it would be to expect a doctor to go his rounds inquiring whether people are in pain and need his services. Two things should be noticed. First, that an illustration (of the doctor and his patients) is given to demonstrate the reasonableness of the sebha,a. Everyone, R. Ashi is suggesting, acts on this principle. Secondly, since it is self-evident, no Scriptural warrant is required for the principle and it must be accepted. It has, moreover, the full power of a law clearly adumbrated in Scripture.

(b) Pes. a5b Here it is stated that the Jew is obliged to forfeit his life rather than be guilty of idolatory, incest or adultery, or murder. The Scriptural source for the first is given and the warrant for the second is said to be derived by Rabbi Judah the Prince from the verse: 'For as when a man riseth against his neighbour, and slayeth him, even so is this matter' (Deut. xxii, 26). In the verse, states Rabbi Judah, the murderer and the ravisher of a betrothed maiden are compared. In accordance with the hermeneutical principle that where Scripture compares two matters the laws of one are to be applied to the other (heqqesh), this teaches that just as a man is obliged to give his life rather than commit murder he must give it rather than commit incest or adultery. The Gema,a then asks: 'How do we know, in fact, that is it forbidden to commit murder even to save one's life?' To this the reply is given that it is a sebha,a. The case of Raba is quoted in support of the sebha,a. A certain man came to Raba with the following problem. The local tyrant had ordered him to slay someone and if he failed to do this his own life would be forfeit. What should he do? Raba replied that he must allow himself to be killed rather than commit murder. The reason given (by Rab~?) is: 'How do you know that your blood is redder, perhaps his blood is redder'. Here, too, sebhara denotes 'common sense'. It is unreasonable to suggest that murder may be committed in order to save life

Original from

UNIVERSITY OF CALIFORNIA

18

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

since a life is lost in the very commission of the crime. 1 The only justification would be because the life of the murderer is of greater value than that of his victim and this no human can know. Thus no Scriptural warrant is req11ired for the law. Since it is self-evident it enjoys the full status of a Biblical law. Moreover. the Gemara suggests, Scripture itself is, as it were, so convinced of our power to reason in this way that it uses this axiomatic rule to serve as a basis of derivation for the rule obtaining in the case of the betrothed maiden. {Whether this is, in fact, the actual argument used by Rabbi Judah the Prince is a moot point. If it is, this means that the use of sebhara to deter1nine the Biblical status of a law was known as early as the second century. It is possible, however, indeed it is probable, that Rabbi Judah had some other basis of derivation. From the examples we shall quote it appears that the method was used only by the late Amoraim.)

(c) Ber. 35a The M ishnah states that a benediction should be recited before partaking of food. The Gemara asks how we know this and various attempts are made, without success, to discover a Scriptural text in support of the rule. The Gemaf'a concludes that there is no Scriptural warrant but the principle is derived by a sebhara. The sebha,a is that it is forbidden to have benefit from this world without reciting a benediction. We must first note that the for1nula: 'It is forbidden' etc. is Tannaitic and is quoted as such in the following sugya. 1 Furthermore, it is clear that the rule that one must recite a benediction before partaking of food is Rabbinic, not Biblical (de-rabbanan not de-'oraitha). 1"his is everywhere accepted as axiomatic in the Talmud. 1 The question arises, why should the law not enjoy Scriptural status, as do the laws in (a) and (b)? It is possible, of course, to see here a dissenting voice on the whole question of sebhara as a substitute for Scripture but this hardly seems likely in view of the widest acceptance of the principle in the V. Rashi, s.v. Other prohibitions may be set aside where life is at stake but this cannot apply to the prohibition of murder since to transgress here means that a life is in any event forfeit. 1 Ber. 35a, bot., but v. Tosephta, IV, I, where a different fo1m11la is used and cf. the quotation at the beginning of the sugya from Siph,a, Qedoshim, III, 9, ed. Weiss p. go. 1 V. Ber. 21a, Maimon. Yad, Hil. Ber., VIII. 12 and freq. 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

19

THE TALMUDIC SEBHARA

Amoraic literature, without the faintest suggestion that it is anything but unanimous. 1 It seems very plausible to suggest that a distinction must be made between two types of sebha,a. In (a) and (b) the sebha,a is one that is universally accepted. All rational men would give their assent to it. Consequently, it enjoys Scriptural status for, in the view of the Amoraim, no special Biblical text is req11ired for a law which can be derived by normal methods of reasoning. But the sebha,a that it is forbidden to partake of the world without reciting a benediction is undoubtedly acceptable to the believer but would not commend itself to the non-believer. It can hardly be considered, therefore, to be axiomatic. Such a principle belongs to the sphere of religious law and for this to be binding Scripturally (tle-'oraitha) a Biblical text is req11ired. The meaning of the passage is then that this sebha,a is acceptable to the religious mind, hence the Sages instituted the benediction before partaking of food, but such a sebha,a is not of the order which allows us to dispense with a Biblical text. In the absence of such a text it would enjoy only the authority of Rabbinic law (de- ,abbanan). (d) Some general observations Although the principle that a sebha,a enjoys Biblical authority is not stated explicitly until the late Amoraic period (as in (a) and (b) ) it is implicit in much of the earlier Rabbinic, even Tannaitic methods. Some of R. Ishmael's thirteen principles, particularly the qal wa-lwmer and the binyan 'abh, are based on human reasoning and yet enjoy Scriptural status. 1 It is still debatable whether the midtloth have been influenced to a large extent by Greek logic 3 but the human reasoning element cannot 1

J. Joshua of Frankfort in his Commentary 'Pene Yehoshua' ad loc. suggests that according to this sugya the benediction is, in fact, de'oraitha. But this is extremely unlikely and contra.diets the assumption of the Gemal'a everywhere. Ezekiel Landau in his 'Tzion Le-Nephesh l:layyah' ad loc. suggests that the principle that a sebhal'a can serve as a substitute for Scripture can only apply to the details of a precept, not to a new precept, otherwise there would be no need for Scripture to record any of the 'rational' precepts! 1 V. Chapter I, supra, p. 3. 1 On this question v. S. Liebe1man: 'Hellenism in Jewish Palestine', N.Y., pp. 47ft. and D. Daube: 'Rabbinic Methods of Interpretation and Hellenistic Rhetoric' in HUCA, Vol. XXII, pp. 239-264. CJ. sujwa, page 9, n.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

20

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

be gainsaid and was clearly recog11ised as such by the Tannaim. 1 Here, however, an important new development is recorded in two passages in the Babylonian Talmud, both belonging to the late Amoraic period.

(i) B.Q. 7ab It is stated here that two witnesses who have been caught out in a falsehood by two other witnesses ('ed. zomam), 1 become disqualified from further testimony, according to Raba, not, as we might have expected, from the time their false witness was given but from the time they were discovered to be false. One of the reasons• given in the Gema,a is that the law of 'ea zomam is a fiiddush ('a strange law', 'an anomaly') for 'why rely on the second pair of witnesses, rely rather on the first pair?' Consequently, we are not justified in applying normal methods of reasoning and we must not go beyond the actual law recorded even though it seems reasonable to do so, i.e. the second pair are believed to disqualify the first pair but only from that time, not retrospectively. (ii) Pes. 44-b The Gema,a discusses whether the mere taste of forbidden food (i.e. where food that is per 1nitted has been cooked together with forbidden food, the latter giving a taste to the former) is forbidden by Scriptural law (ta'am ke-'iqqa,). It is suggested that this can be derived from meat cooked in milk which Scripture forbids even though there is only a 'taste' of milk in the meat or meat in the milk. To this the reply is given that the prohibition of meat and milk is a /1,iddush and no new laws can be derived from a biddush by the usual methods of interpretation. The anomaly of meat and milk is said to be that even C/. Pes. 66a that a man may advance a qal wa-[aomer of his own accord even if he has no tradition from his teacher in contradistinction to a gezeYah shawah. CJ. Pes. 18b, Y om. 43a and Qidd. 4a that occasionally even where there is a qal wa-[iomer a verse is written but the expression used is /ara[i, which suggests that, in reality, the verse is not required. V. Qidd. 4b that if it is at all possible to do so this solution is avoided and cf. Z. Lauterbach's observation in J. E., Vol. XII, p. 32, col. 1, bot. 1 Deut. xix. 16-21. 1 According to the first version of Raba's opinion in the sugya. According to the second opinion, Raba agrees basically with Abaye. 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

21:

THE TALMUDIC SEBHARA

though meat soaked in milk for a long period absorbs as much milk as meat boiled in milk it is only the latter that is Scripturally forbidden. 1 In both (i) and (ii) human reasoning, even of the most convincing kind, cannot operate as a derivation for new laws since the premises are in any event 'irrational'. Since in the very premises the normal laws of reasoning have been suspended no rational conclusions can be drawn from them. Here the sebha,a would have us contradict the law stated in Scripture• and since, in spite of this, Scripture records the law it is obvious that here the principle of sebha,a, potent elsewhere, cannot operate. 'lnis is not to be confused with the Rabbinic idea of the IJ,oq, the law for which there is no reason. 8 The IJ,oq is not 'irrational' i.e. although it is not capable of discovery by human reasoning it affords no contradiction to human reasoning or the laws of reasoning. The dietary laws, for instance, do req11ire Scriptural texts for they would not have been arrived at by 11naided reason but they do not offend in themselves the normal canons of reason. Consequently, once the fwq has been recorded in Scripture other laws can be derived from it by means of the her1neneutical principles, including the 'rational' ones of Qal W a-Qome, and Binyan 'A bh.' But in the case of the f,iddush the law is 'irrational' i.e. it contradicts common sense, it offends against the laws of reasoning, it rejects the sebha,a, and it must therefore obey laws of its own and cannot be used as a basis of derivation for new laws. · 1

CJ. the sugya here where, at first, it is suggested that the fiiddush lies in the fact that milk and meat separately are permitted and yet when cooked together are forbidden. The Gemaf'a refutes this as no real anomaly since there are other examples of two permissible things combining to form something prohibited, e.g. forbidden mixtures of clothes, animals and seeds kil' aim. 1 It is of interest that in the case of 'ed zomam the illustration of the /Jiddush begins with the same form11la used in the illustration of the sebhara in (b): mai f,azithl • Siphf'a to Lev. xviii. 4, ed. Weiss p. 86, Yom. 67b, cf. I. Heinemann: Ta'ame Ha-Mitzwoth Be-Siphruth Yisrael, Part I, Jer. 1949, Chapter 3, p. 2If. ' This is assumed everywhere in the Talmudic literature, cf. Encyclopedia Talmudith, Vol. IV, Jer. 1952, pp. 1-11, s.v. Binyan 'Abh, the illustrations quoted here of the .application of Binyan 'Abh to the 'fioq' can easily be multiplied.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

22

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

(e) Gi#. 6b

•

R. Joseph said: How do we know that R. Abiathar is a reliable person? Did he not send a message to Rabb Judah: 'People who come here (to Palestine) from there (Babylon) fulfil in their own person the verse: ''They have given a boy for a harlot and sold a girl for wine and have drunk'' (Joel iv, 3)' (i.e. they neglect their families)? And he wrote the verse without underlining it whereas R. Isaac said that two words of Scripture may be written without lines but not three. In a Baraitha it was taught: Three may be written but not four. Abaye said to him: Is one who bas not heard the ruling of R. Isaac therefore not considered a great man? If it were a matter of sebhara well and good. But this •is gemara ('tradition') and it so happens that he had not heard of this gemara. Here, and in many other passages, sebhara is contrasted with gemara. Gemara, in this sense, is a 'ruling', an established law or text. Sebhara here is a logical exposition, a reasoned argument. Hence Abaye's point: lack of knowledge of a gemara is no evidence of mental incapacity or umeliability but failure to expound an idea in a reasoned way ~·sebhara - is sure evidence of scholarly inefficiency. Here the term se~hara is used in a different sense from (a) and (b). There it means 'self-evident truth', an argument that is simply the application of common sense. This in itself would hardly suffice to produce evidence of sound scholarship or scholarly capacity. The meaning of sebhara here, and in the other passages where it is contrasted with gemara, is an argument that is not at all self-evident but which, once propounded by a capable scholar, produces conviction.

(f) B.B. 77a Amemar said: The law is that letters (i.e. a bond, which is made up of 'letters') are acq11ired by the act of delivery (and it is not necessary to write a special bill of sale). R. Ashi said to Amemar: Is this a gemara or a sebhara? He replied: A gemara (i.e. he had it by tradition). R. Ashi said: It is a sebhara, too, for letters are words and words cannot be acquired by words (i.e. of what advantage would the bill of sale be since it, too, consists only of words like the original bond?). Here sebhara, in contradistinction to gemara, is used, as in e),

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

23

THE TALMUDIC SEBHARA

in the sense of 'logical deduction', 'reasoned argument'. It is not 'self-evident'. Amemar does not, in fact, appear to accept it but R. Ashi feels that the· argument is convincing. It is of interest to note that in the sugya this question of whether 'letters' are acquired by delivery is debated by Rabbi and the Sages. Amemar states that he has a tradition that the ruling is in accordance with Rabbi's view. R. Ashi, on the other hand, decides between Rabbi and the Sages on the grounds that Rabbi's view is the more plausible. 'Ibis method of ar1iving at a decision by means of a sebhara on a matter debated by Tannaim is very infrequent. Generally speaking the Amoraim do not decide such matters by having recourse to their own sebhara for this would imply that they were more gifted in matters of sebhara than some of the Tannaim. The probable meaning of the passage is that R. Ashi only relies on the sebhara to explain the gema,a recorded by Amemar ('R. Ashi said: It is a sebham, too'). (g) Be,. 4b R. Jol_tanan said: Who is a child of the World to Come? He who joins the Ge'ullah hymn (in praise of the Exodus) to the TephiUah (the 'standing prayer') in the evening (i.e. just as the Ge' ullah precedes the T ephiUah in the morning, on this all are agreed, so should it do in the evening). R. Joshua b. Levi said: They instituted the Tephilloth (the three 'standing prayers' of the day, morning, afternoon and evening) in the middle (i.e. the morning Shema and the Ge'ullah is read first, then at appropriate times of the day the three Tephilloth, and last of all the evening Shema and the Ge'uUah hymn. Consequently, according to R. Joshua b. Levi, the Ge'ullah hymn is not 'joined' to the TephiUah in the evening, but the Ge'ullah follows the Tephillah). What is the point of difference? (between R. Jol_tanan and R. Joshua b. Levi). If you want I can say that they debate the meaning of a Scriptural verse, if you want I can say that they debate a sebhara (i.e. it can either be said that their differences stem from different interpretations of the same verse or it can be said that they differ as to which is the more reasonable sebhara, and they have no recourse to Scripture). If you want I can say that they debate a sebha,a: for R. Joltanan holds (sabhar) that the redemption from Egypt began at night, even C

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

24

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

though the true deliverance did not take place until the morning (hence the night Ge' ullah is no different from the mor11ing Ge'ul/.ah). But R. Joshua b. Levi holds (sabha,) sinr,e the deliverance did not take place 11ntil mo1x;ipg the evening deliverance is not considered to be important. If you want I can say that they debate the meaning of a Scriptural verse and they both expound the same verse. For Scripture says: 'When thou liest down and when thou risest up' (Deut. vi, 7). R. Job.anan holds (sabha,) that 'lying down' is compared to 'rising up', just as at the time of rising the reading of the Shema (and the Ge'ullah which follows it) precedes the Tephi/J,a,h so, too, at the time of lying down the reading of the Shema precedes the Tephi/J,a,h. R. Joshua b. Levi holds (sabha,) that 'lying down' is compared to 'rising up', just as at the time of rising the reading of the Shema is near to the bed (i.e. it is nearer to rising from bed than Tephillah which is recited after it) so, too, at the time of lying down the reading of the Shema should be near to the bed (i.e. it should be recited later than the TephiUah and hence nearer to bed). Here to be noted first is that the tex m sabha, is used not alone for the suggested debate on the sebha,a but also for the debate on the Scriptural text. The meaning of the passage is undoubtedly that according to the first version R. Jo};lanan and R. Joshua b. Levi debate on a matter of pure sebha,a, i.e. a sebha,a quite independent of a Scriptural verse, whereas according to the second version their debate concerns the sebhara to be used in interpreting the Scriptural verse. (It is, of course, obvious that even according to the second version the debate is not really on a Scriptural law for the duty of 'joining' Ge'ullah to Tephillah is clearly Rabbinic 1 and the verse is 'asmakhta,• a mere 'support' from Scripture.) Here, as in (e) and (f), the sebhara is not 'self-evident truth' or 'common sense' but a 'reasoned argument'. Hence, there is no illustration to convey its self-evident nature (as in (a) and (b) ) . In any event the clear implication of the sugya is that if the sebhara is truly well-established and convincing there is no need Since Tephillah itself is de-f'abbanan, v. Bel'. 21a and freq. 1 The term generally used for a law that is obviously of Rabbinic origin but for which a 'support' is found in Scripture, e.g. in Betz. 15b and freq. 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

25

THE TALMUDIC SEBHARA

for a Scriptural text. The sugya suggests that if they do debate the meaning of the Scriptural text there is no sebhara (i.e. the sebhara is rejected as unconvincing and inconclusive) but if the sebha,a is accepted no text is required.

(h) Yeb. 35b It was stated: If a levir performed f,alitzah (Deut. xxv, 9) for a pregnant woman who subsequently miscarried: R. J ol}.anan said: She does not require l;uzlitzah from the brothers. Resh Laqish said: She does require f,alitzah from the brothers. R. Jol}.anan says that she does not req11ire l)alitzah from the brothers for the f,alitzah given to a pregnant woman is a valid 'fia,litzah, and · marital cohabitation with a pregnant woman is similarly valid (i.e. since the wQman eventually miscarried it becomes evident retrospectively that, in fact, she had no child at the time of f,alitzah or cohabitation, and the act is consequently valid like any other act of f,alitzah or cohabitation perfortned by a levir for his sister-in-law whose husband had died without issue). Resh Laqish says that the IJ,a,litzah of a pregnant woman and the act of cohabitation with a pregnant woman is invalid (i.e. since at the time of the act she was with child the act has no validity, for it is ordained in Scripture only where the husband died without issue). What is the point of difference? If you want I can say that it concerns the interpretation of a Scriptural verse and if you want I can say that it is a matter of sebhara. If you want I can say that it is a matter of sebhara, for R. Jol}.anan holds (sabhar) that if Elijah would come and tell us that she will miscarry she would surely be a person capable of perforn1ing a valid fia,litzah (since she would not really be with child) therefore, now, too, it becomes evident retrospectively that this is a valid f,alitzah. Resh Laqish holds (sabhar) that we do not apply the retrospective test. If you want I c.an say that they debate the interpretation of a Scriptural text. R. Jol}.anan holds (sabhar) that the All Merciful says: 'And have no child' (Deut. xxv, 5) and, in fact, he has none (since his widow miscar1ied). Resh Laqish holds (sabhar): 'And have no child' means 'investigate whether he has a child' (a play on the words, 'ayyin, 'to investigate' for 'en, 'has none'). This _is exactly the same as in (g) except that here the law debated is certainly Scriptural, not Rabbinic. It follows from

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

26

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

this, too, that if the sebhara is accepted no verse is req11ired. In both (g) and (h) it is clear that even the type of sebhara that is not self-evident enjoys Sctiptural authority if it is accepted (or, better, if it is accepted it is treated as if it were self-evident, but the fact that the sebhara can be debated demonstrates that it is not of the order of (a) and (b) which are recog1,jsed as self-evident by all menl).

(i) Qidd. 35b How do we know that the law of marring the beard (Lev. xix. 27) does not apply to women? If you want I can say it is a sebhara since they have no beard (i.e. even if a woman does grow a beard it is not ter1ned a 'beard') 1 If you want I can say it is derived from Scripture, for Scripture says: 'Ye shall not round the comer of your hea.ds, neither shalt thou mar the comer of thy beard' (Lev. xix, 27). Since Scripture varies its speech (using first the plural then the singular), for it should have written: 'The comer of your beards' (why: 'thy beard'?), this suggests 'thy beard' but not 'thy wife's beard'. Here sebhara is not of the self-evident type (as in (a) and (b)) for this type of sebhara, as we shall see later, must be so evident, in the Amoraic view, that there is not the slightest suggestion of its inadequacy. The very fact that here an alternative is suggested ('If you want I can say it is derived from Scripture') shows that the sebhara is of the 'reasoned argument' type, as in (e) - (h). In any event it is clear that if the sebhara is accepted no Scriptural text is req11ired, the . sebhara itself enjoying Scriptural status. (j) B.B. 8b Raba said to Rabbah b. Mari: What is the derivation for the maxim of the ·Rabbis that it is a great religious duty to redeem captives? He replied: Scripture says: 'And it shall come to pass when they say unto thee, Whither shall I go forth, then thou shalt tell them, Thus saith the Lord, Such as are for death, to death·, and such as are for the sword, to the sword, and such as are for famine, to the famine, and such as are for captivity, to captivity' (Jer. xv, .2), and R. Jo\}anan said: Each of the This appears to be the meaning of the passage, cf. Rashi s.v. we-lo zeqan 'ishtekha. - · 1

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

THE TALMUDIC SEBHARA

punishments mentioned later in the verse is worse than the one which precedes it. Sword is worse than death - if you want I can say that this can be demonstrated from Scripture, if you want I can say that it is a sebha,a. If you want I can say it is a sebha,a for the sword deforn1s but death does not deform. If you want I can say that this can be demonstrated from Scripture, as the verse says: 'Precious in the eyes of the Lord is the death of his saints' (Ps. cxvi, 15). Famine is worse than the sword. If you want I can say that it can be demonstrated from Scripture, if you want I can say it is a sebha,a. If you want I can say that it is a sebha,a, for in the one case there is prolonged suffering, in the other there is no prolonged suffering. If you want I can say that there is a demonstration from Scripture, for the verse says: 'They that be slain with the sword are better than they that be slain by hunger' (Lam. iv,· 9). Captivity is worst of all because it includes all of them. Here the sebha,a is of the 'reasoned argument' type since here, too, the possibility of an alternative means of derivation is suggested. Although the discussion appears to belong to the purely theoretical there are, in fact, important practical implications with regard to which type of suffering should be first alleviated.1 But a careful examination of the sugya shows that we have here a sebhara of the 'common sense' type as well, although the terxn sebhara is not actually tised to describe it. For the sugya concludes that captivity is worst of all (and therefore to redeem captives is a great religious duty) since it includes all of them. 'Since it includes all of them' is a statement of the sebha,a of the 'common sense' type. No verse is required to demonstrate that this is worse than the others since it includes the others.

(k) B.B. 9a R. Huna said: Applicants for food (from the charity chest) are examined (to see if they are gen11ine) but not applicants for clothes. If you want I can say that this can be demonstrated from Scripture, if you want I can say it is a sebhara. If you want I can say it is a sebhara, because one is exposed to contempt but not the other. If you want I can say that it can be demonstrated from Scripture: 'Is it not to deal (paros) thy 1

C/. the sugya as a whole.

Digitized by

Google

Original from

UNIVERSITY OF CALIFORNIA

28

STUDIES IN TALMUDIC LOGIC AND METHODOLOGY

bread to the hungry' (Is. lviii, 7). The word pa,os is written with a shin to yield perosh 1 ('to examine')-first examine him and then give him bread. But of the naked it is written: 'When thou seest the naked, that thou cover him' (Is. ibid.), which means, immediately. Rabb Judah said: Applicants for clothes are to be examined but not applicants for food. If you want I can say that it can be demonstrated from Scripture, if you want I can say that it is a sebha,a. If you want I can say it is a sebha,a, for the one is suffering, the other is not. If you want I can say that it can be demonstrated from Scripture, for Scripture says: 'Is it not to deal thy bread to the hungry' which means, at once, whereas later it says: 'When thou seest the naked' which means, when you have looked into the matter. It should first be noted that in all probability the for1nula: 'If you want' etc. is a later addition of the Gema,a and not part of the original argument of R. Huna and Rabb Judah. Consequently, there is no evidence for the view that the sebha,a was considered to enjoy Scriptural status as early as the period of the second generation of Amoraim. 1 But the for111ula is the one generally used, as we have seen, and it contains the implications we have noted. However, in this passage two contradictory sebharas are recorded, both of which are plausible, depending on how we look at the case. Since they are the 'reasoned argument' type of sebhara there is nothing surprising in R. Huna's refusal to see th~ sebhara as Rabh Judah sees it and vice versa. (1) A.Z. 34b R. Ahdaboi said in the name of Rabb: If one married a woman • '

Though not in our texts, v. M es